Pakistan Getting More from Water William J. Young, Arif Anwar, Tousif Bhatti, Edoardo Borgomeo, Stephen Davies, William R. Garthwaite III, E. Michael Gilmont, Christina Leb, Lucy Lytton, Ian Makin, and Basharat Saeed About the Water Global Practice Launched in 2014, the World Bank Group’s Water Global Practice brings together financing, knowledge, and implementation in one platform. By combining the Bank’s global knowledge with country investments, this model generates more firepower for transformational solutions to help countries grow sustainably. Please visit us at http://www.worldbank.org/water or follow us on Twitter at @WorldBankWater. Pakistan Getting More from Water William J. Young, Arif Anwar, Tousif Bhatti, Edoardo Borgomeo, Stephen Davies, William R. Garthwaite III, E. Michael Gilmont, Christina Leb, Lucy Lytton, Ian Makin, and Basharat Saeed ©2019 The World Bank International Bank for Reconstruction and Development The World Bank Group 1818 H Street NW, Washington, DC 20433 USA DISCLAIMER This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. In the case of any discrepancies between this English version and any subsequent translations, the English version prevails. The report reflects information available up to September 30, 2018. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Please cite this work as follows: Young, William J., Arif Anwar, Tousif Bhatti, Edoardo Borgomeo, Stephen Davies, William R. Garthwaite III, E. Michael Gilmont, Christina Leb, Lucy Lytton, Ian Makin, and Basharat Saeed. 2019. “Pakistan: Getting More from Water.” Water Security Diagnostic. World Bank, Washington, DC. Cover and interior design: Francis Gagnon, Chez Voilà. iii Contents Foreword  ix Acknowledgments  xi Abbreviations  xiii Executive Summary  xv Is Pakistan ‘Water Secure’?  xv What Undermines Water Security in Pakistan?  xvi How Well Are Water Resources Understood?  xvii What Interventions Can Improve Water Security in Pakistan?  xviii Is Water Scarcity a Constraint to Reaching Upper-Middle-Income Status by 2047?  xix Twelve Recommendations for Improving Water Security  xx Next Steps  xxiii Chapter 1:  Setting the Scene  1 Economic, Demographic, and Geographic Context  1 Water Resources Overview  3 Pakistan’s Water Economy in the Global Context  7 Purpose and Structure of Report  9 Chapter 2:  What Pakistan Gets from Its Water  13 Key Messages  13 Economic Outcomes  14 Social Outcomes  19 Environmental Outcomes  21 Chapter 3:  Pakistan’s Water Endowment  27 Key Messages  27 Average Water Balances  28 River System Gains and Losses  29 Provincial Water Availability and Use  32 Temporal Patterns  32 Groundwater  35 iv PAKISTAN: GETTING MORE FROM WATER Chapter 4:  Pakistan’s Water Sector Architecture  41 Key Messages  41 Infrastructure  42 Water Governance  54 Political Economy Challenges  62 Financing  67 Chapter 5:  Pakistan’s Water Sector Performance  75 Water Resources Management  75 Water Service Delivery  95 Water-Related Risk Mitigation  102 Chapter 6:  Water Security Trajectories  115 Key Messages  115 Introduction  116 Structural Change in the Economy  116 Future Water Demand and Use  118 Changing Consumer Preferences  123 Policy Reform and Trade Shifts  124 Improved Environmental Management  126 Chapter 7:  Pathways to Water Security  129 Water Resources Management  131 Water Supply Service Delivery  135 Water-Related Risk Mitigation  136 Appendix A:  Pakistan Water Balance Data Sources  139 Appendix B:  Legal Framework for Water Resources  143 Appendix C:  Summary of WSTF Priority Actions  153 Appendix D:  CGE Modeling Approach and Assumptions  157 Figures ES.1 Indus Basin Average Annual Water Balance   xvii ES.2 Historical Water Availability, Withdrawals and Consumption per Capita; and Projected per Capita Availability and Demand   xviii ES.3 Complexity, Urgency, and Scale of Impact of Key Recommendations   xxii 1.1 GDP of Pakistan, 2000–16   2 1.2 Sector Contributions to GDP in Pakistan, 1960–2000   2 1.3 Sector and Agricultural Subsector Contributions to GDP in Pakistan, 2006–16   2 v 1.4 Share of Urban and Rural Population in Pakistan, 1950–2050   3 1.5 –2016), Withdrawals (1975–2016), and Consumption (1975–2016); Historical Water Availability (1960­ and Projected Availability and Demand to 2047   6 1.6 Global GDP per Capita and Total Renewable Water Resources per Capita   7 1.7 Global Structural Transformation Trajectory   8 1.8 Water Productivity and Surface Water Stress by Country   9 1.9 Water Security Diagnostic Framework   10 2.1 Share of Cropping and Livestock Contributions to Agricultural GDP in Pakistan, 2006–16   14 2.2 Share of Agricultural Water Use and Water-Dependent Agricultural to GDP in Pakistan, 2016   15 2.3 Irrigated Areas for the Four Major Irrigated Crops in Pakistan, Fiscal Years 2007–16    15 2.4 Average of Irrigated Areas by Province and Water Source in Pakistan, 2006–16   16 2.5 Yield Index for Major Irrigated Crops in Pakistan, 1987–2015   17 2.6 Value of Water in Irrigation for Punjab and Sindh Relative to 1980 Value for Punjab, 1980–2013   18 2.7 Share of Population Affected by Riverine Floods in Pakistan, 1973–2016   20 3.1 Average Annual Water Balance for the Three Hydrologic Units of Pakistan   30 3.2 Annual River Losses and Gains in Key Losing Reaches of Indus River, Pakistan, 1940–94   31 3.3 Average Annual Pattern of Average 10-Day Inflows and Canal Withdrawals, Pakistan   34 3.4 Annual Indus Basin Inflows, Canal Withdrawals, and Outflows, Pakistan, 1975–2015   34 3.5 Annual Inflows to Pakistan from Ravi and Sutlej Rivers, 1960–2015   35 3.6 Groundwater Levels at Khanewal and Sahiwal Divisions, Punjab, 1910–2010   36 4.1 Indus Basin Irrigation System of Pakistan   43 4.2 Timeline of Major Irrigation and Water Resources Infrastructure in Pakistan, 1870 to Present   44 4.3 Reservoir Storage Volume as Ratio of Mean Annual Flow Compared to Coefficient of Variation of Annual Flow for Selected Countries and Regions   47 4.4 Pakistan’s Dam Readiness Stage by Number and Generating Capacity   48 4.5 Projected Increases in Hydropower Capacity, Pakistan   49 4.6 Capacity of New Hydropower Projects by Readiness Stage and Province or Territory, Pakistan   49 4.7 Length of Levees and Number of Spurs by Province in Pakistan   50 4.8 Federally Operated Regular Flow Gauging Stations in the Upper Indus Basin of Pakistan and Average Period of Record, 1960–2016   51 4.9 Major Policy and Institutional Milestones before Partition and before and after the Indus Waters Treaty in Pakistan, 1940 to Present   56 4.10 Major Federal and Provincial Legal Instruments before Partition and before and after the Indus Water Treaty in Pakistan, 1860 to Present   58 4.11 Federal and Provincial Government Water Sector Funding Allocations and Percentage of Total Federal Budget in Pakistan, 2000–17   67 4.12 National Investment and Expenditure for First WSTF Action in Pakistan, 2013–17   68 4.13 National Investment and Expenditure for Second WSTF Action in Pakistan, 2013–17   69 4.14 National Investment and Expenditure for Third WSTF Action in Pakistan, 2013–17   69 4.15 National Investment and Expenditure for Fourth WSTF Action in Pakistan, 2013–17   70 4.16 National Investment and Expenditure for Fifth WSTF Action in Pakistan, 2013–17   70 5.1 Relationship between Annual Irrigation Shortfall and Total Annual Inflow in Pakistan, 1992–2015   82 vi PAKISTAN: GETTING MORE FROM WATER 5.2 Share among Provinces of Annual Shortfall of Canal Withdrawals Relative to Water Apportionment Accord Baseline Allocation Volume in Pakistan, 1992–2015   83 5.3 Annual Series of Total Canal Withdrawals, Outflow below Kotri Barrage, and Losses as Components of Indus Basin Inflows, Pakistan, 1975–2015   84 5.4 Annual Water Balance Closure and Mean Annual Temperature Anomaly for Karachi, Pakistan, 1975–2015    84 5.5 Indus River System Outflows from Kotri Barrage as Share of System Inflow, by Season, 1975–2015   87 5.6 Total Economic Productivity of Water in Selected Countries   90 5.7 Agricultural Water Productivity in Selected Countries   90 5.8 Economic Productivity of Major Crops for Selected Countries and Globally, 1961–2016   91 5.9 Estimated Average Blue Water (Irrigation) Footprints of Wheat, Raw Sugar, Rice, and Cotton for Pakistan   92 5.10 Key Water Footprint Metrics for Rice in Major Rice-Growing Countries    93 5.11 Share of Farm Sizes in Pakistan by Number and Aggregate Area, 1990, 2000, 2010   94 5.12 Share of Access to Improved Water Supply and Improved Sanitation in Rural and Urban Pakistan, 2015   96 5.13 Share of Access to Improved Drinking Water and Improved Sanitation in Pakistan, 2000–2015   96 5.14 Urban Water Access by Source in Pakistan, 2015   97 5.15 Share of Urban Water Supplies Unfit for Human Consumption by Province in Pakistan, 2005 and 2015   97 5.16 Operating Cost Recovery Ratio of Irrigation Departments by Province in Pakistan, 2000–09   101 5.17 Abiana Collection Efficiency by Province in Pakistan, 2000–09   101 5.18 Estimated Increases in Water Demand Attributable to Projected Warming in Pakistan, 2025 and 2050   103 5.19 Sector Shares of Water Demand Increase Attributable to Projected Warming in Pakistan, 2025 and 2050   103 5.20 Mean Sea Level, Karachi Coast, Pakistan, 1860–2000   105 6.1 Output per Laborer, by Sector, in Pakistan, 1991–2016   118 6.2 Changes in Sector Shares under BAU-Lo and RUMI-Hi by 2031 and 2047, and Differential between BAU-Lo and RUMI-Hi, in Pakistan, 2014 Baseline   118 6.3 Projected Water Demand Increases, Relative and Absolute, Attributable to Slow and Fast Climate Warming and Growth, by Sector, 2025 and 2050   119 6.4 Pakistan Total Water Demand in 2015 and Projected for 2025 and 2050   120 6.5 Modeled Annual Nonagricultural Water Demands, by Scenario, 2014–47   120 6.6 Modeled Annual Irrigation Water Consumption, by Scenario, 2014–47   121 6.7 Modeled Annual Groundwater Irrigation Consumption, by Scenario, 2014–47   121 6.8 Modeled Annual Crop Water Use in Pakistan under BAU-Lo, 2014–2047   122 6.9 Modeled Annual Crop Water Use in Pakistan Under RUMI-Hi, 2014–2047   122 6.10 Modeled Annual Crop Water Use in Pakistan under RUMI-Hi-Diet, 2014–2047   123 6.11 Monthly Household Expenditure on Food Groups in Pakistan by Income Quintile, 2015   124 7.1 Complexity, Urgency, and Scale of Impact of Key Recommendations for Pakistan   132 B.1 Hierarchy of Regulatory Objectives for Water Resources Management in Pakistan   144 B.2 Share of Completeness of Provincial Legal Frameworks in Pakistan   147 B.3 Comparison of Completeness of Legal Frameworks for Water Resources Management across Pakistani Provinces and South Asian Countries   148 B.4 Completeness of Legal Frameworks for Water Resources Management in Pakistani Provinces and Comparator Countries, and Level of Water Stress Given as Withdrawals as Share of Total Renewable Resource   148 vii B.5 Comparison of Completeness of Legal Frameworks for Water Resources Management in Pakistani Provinces, Countries for which Similar Assessments Exist, and GNI per Capita   149 B.6 Evolution of Provincial Legal Frameworks in Pakistan   149 D.1 Sequence of Modeling in the Linked CGE-W Framework   161 Maps 1.1 Pakistan   4 3.1 Oblique Aerial View of the Upper Indus Basin   33 3.2 Groundwater Depth across Upper Indus Plain, 2002 and 2014   37 3.3 Groundwater Salinity Levels across the Indus Basin of Pakistan   38 5.1 Percentage of District Population Vulnerable to Drought in Pakistan   80 Tables ES.1 Twelve Recommendations for Improving Water Security in Pakistan, Indicating the Required Legal, Policy, and Institutional Reforms and Necessary Infrastructure Investments   xx 1.1 Population Growth across Provinces in Pakistan   3 1.2 Estimated Contributions to Total Average Annual Renewable Water Resource, Pakistan   5 1.3 Average Annual Water Withdrawal and Consumption Volumes, Pakistan   5 2.1 Distribution by Irrigated Area across Provinces of Four Major Crops in Pakistan, 2016   16 2.2 Average Yields of Major Irrigated Crops Nationally and by Province, 2006–16   16 2.3 Average Land and Water Productivity in Punjab and Sindh at 1980 Prices, 2009–13   17 3.1 Average Annual Available Water Resources of Pakistan   28 3.2 Average Annual Water Balance for Pakistan’s Three Hydrologic Units    29 3.3 Average Annual Provincial Water Resource Availability in Pakistan   32 3.4 Provincial Withdrawals, Level of Use, and per Capita Availability and Use in Pakistan   32 3.5 Estimated Average Annual Groundwater Balances by Province in Pakistan   37 4.1 Current and Future Reservoir Capacity and Active Groundwater Storage Capacity in Pakistan   46 4.2 Priority Dam Projects in Pakistan   50 4.3 Hydrological and Meteorological Monitoring Infrastructure Maintained and Operated by Federal Authorities in Pakistan   52 4.4 Recommended Investments in Priority Actions Areas, Commitments, and Expenditures by Pakistan’s Water Sector Task Force, 2013–17   68 5.1 Droughts in Pakistan   79 5.2 Options to Achieve Government Pakistan Vision 2025 Energy Targets with Water-Energy Nexus Issues   107 6.1 Summary of the Scenarios for Pakistan Modeled and Analyzed using CGE-W   117 6.2 GDP per Capita and Average GDP Growth Rate for 1970–2016 in Pakistan and Comparator Countries, with Growth Required for Pakistan to Reach Comparator Growth Rate by 2047, and Share of Food Expenditure   117 6.3 Water Productivity in Pakistan by Crop for Baseline Year (2013/14) and Productivity Growth Rates under BAU and RUMI   123 6.4 Modeled Growth in Commodity Consumption by Scenario in Pakistan   124 6.5 Modelled Water Use by Major Crops in Pakistan under RUMI-Hi and Changes in Water Use for RUMI-Hi Variants, 2047   125 viii PAKISTAN: GETTING MORE FROM WATER 6.6 Modelled GDP under RUMI-Hi and Impacts of Policy Reforms, Changing Export Prices, and Consumer Preferences in Pakistan, 2047   126 6.7 Annual Value of Lost Production with Increased Annual Water Demand below Kotri Barrage, Pakistan, under RUMI-Hi   127 7.1 High-Level Recommendations and Finances Required by Performance Area in Pakistan   130 A.1 Indus Basin Water Balance   139 A.2 Makran Coast Water Balance   141 A.3 Kharan Desert Water Balance   141 B.1 Links to National Legislation Relevant to Water Resources Management in Pakistan   144 B.2 Presence or Absence of Key Legal Elements in Pakistan’s Provincial Legal Frameworks for Five Areas of Water Resources Management   145 B.3 Links to Provincial Legislation in Pakistan   150 D.1 Farm Household Expenditure Elasticities by Farm Size for RUMI Hi and RUMI Hi-Diet in Pakistan   158 D.2 Rural Nonfarm and Urban Expenditure Elasticities for Reaching Upper-Middle Income-Hi and Upper-Middle Income-Hi-Diet for Income Quartiles in Pakistan   159 D.3 Key Average Annual Water Balance Terms for the CGE-W Model in Pakistan   160 Foreword W ater security is an important and growing Basin water resources between Pakistan and India and challenge for Pakistan, and one that extends defined the basic water resource envelope for Pakistan. far beyond the traditional water sector. It Although blessed with a large water endowment, influences diverse aspects of economic and social and with extensive glacier storage that buffers supply development, as well as national and regional security. variations, the huge increase in population means This study takes a long-term view of water security— Pakistan is now challenged by relative (per person) water out to 2047 when Pakistan turns 100. The work has scarcity, and both the population and water demands been closely coordinated with the World Bank’s broader are projected to grow for several decades. The challenge economic policy work for Pakistan@100. This work of balancing supply and demand will be exacerbated thus contributes not only to the important water sector by climate change, which will increase the variability dialogue but also to the broader conversation on in supply and, because of higher temperatures, will Pakistan’s economic and social development. push water demands even higher. These challenges are further vexed by widespread pollution that is degrading The 4,000-year-old Indus civilization has its roots in the resource base and undermining both public and irrigated agriculture. Pakistan still relies heavily on environmental health. the Indus River for water supply to all sectors of the economy as well as for energy generation. Water for Pakistan cannot continue business as usual (BAU) water irrigation across the semi-arid Indus floodplains, which management. How Pakistan tackles these challenges, underpins national food security, is the dominant use. and the speed at which is does so, will have a major Nonetheless, for many Pakistanis the foremost water influence on the country’s rate of economic development security concern is that of inadequate domestic water and the quality of life for her people. While there are supply and sanitation services, which uses a very small major infrastructure and financing challenges to surmount, share of the available water. the fundamental challenges are ones of governance, in irrigation and urban water supply, at federal, provincial, Pakistan is home to nearly 210 million people—a near and local levels. The World Bank stands ready to support seven-fold increase since the formation of the country in Pakistan—in partnerships with governments, civil 1947. In 1960, after almost a decade of negotiations, the society, the private sector, and regional and international Indus Waters Treaty formalized a partitioning of the Indus organizations—to improve all facets of water security. Illango Patchamuthu Jennifer Sara Pakistan Country Director Senior Director, Water Global Practice The World Bank The World Bank Acknowledgments T his report was prepared by a World Bank team led Melinda Good (Operations Manager, Pakistan), by William Young with contributions from William Lixin Gu (Sustainable Development Program Leader, Garthwaite III, Michael Gilmont, Christina Leb, Pakistan), Alex Ferguson (Senior Manager, SAR External Lucy Lytton, and Basharat Saeed. The report is based Communications), Mariam Altaf (Communications upon work commissioned from the International Water Officer, SAR External Communications), Huma Zafar Management Institute (IWMI) with economic modeling (Operations Officer, Pakistan), and Fei Deng (Country undertaken by the International Food Policy Research Program Coordinator, Pakistan). Institute (IFPRI). Key contributors to the commissioned Valuable consultations—both formal and informal, work were Arif Anwar (IWMI), Stephen Davies (IFPRI), in-country and internationally—helped to frame this Edoardo Borgomeo (IWMI), Ian Makin (IWMI), and work and shape the ideas and messages that have Tousif Bhatti (IWMI). Important early guidance for emerged from the analysis. For positive influence and this work was provided by Claudia Sadoff while with wide-ranging insights, thanks are due to Ali Sheikh and the World Bank Water Global Practice (Water GP); Hina Lotia (LEAD Pakistan); William Doan (U.S. Army her support and oversight of the IWMI team in her Corps of Engineers); Uzma Khan and Rhiannon Bramer current role as IWMI Director General is also greatly (U.S. State Department); Sohail Naqvi and Hammad appreciated. Khan (World Wildlife Fund [WWF] Pakistan); Simi For technical guidance, including through formal peer Kamal and Zohair Ashir (Hisaar Foundation); Sonia Amir reviews and many engaging discussions, the team (The Asia Foundation); Zaigham Habib (hydrologist); thanks Greg Browder, Richard Damania, Maitreyi Das, Rafay Alam (environmental lawyer); Peter Wallbrink, Johannes Jansen, Winston Yu (IWMI), Ghazala Mansuri, Mobin Ahmed, and Mac Kirby (Commonwealth Hanan Jacoby, Maximillian Hirn, Rikard Liden, Shiva Scientific and Industrial Research Organisation [CSIRO] Maki, Toru Konishi, and Mohammad Farhan Sami. Australia); Undala Alam (Department for International For wise counsel, encouragement, and support, the Development [DFID] U.K.); and Brek Batley, David team thanks Michael Haney (SAR Practice Manager, Preston, and Hamza Khalid (Department of Foreign Water GP), Jennifer Sara (Senior Director, Water GP), Affairs and Trade [DFAT], Australia. In addition, Guang Zhe Chen (Former Senior Director, Water GP), interactions with many generous and insightful people Illango Patchamuthu (Country Director, Pakistan), across multiple government agencies and organizations xii PAKISTAN: GETTING MORE FROM WATER at federal, provincial, and municipal levels have Development (DFID). Within the World Bank the team positively influenced this work. thanks Maria Angelica Sotomayor, Joel Kolker, Craig Kullmann, and Ai Ju Huang for GWSP assistance. This work was financed through the Global Water Security and Sanitation Partnership (GWSP). GWSP is This work is part of a growing suite of water security a multidonor trust fund administered by the World studies across the World Bank Water GP. Many World Bank’s Water GP and supported by DFAT; the Bill and Bank colleagues have helped develop the concepts and Melinda Gates Foundation; The Netherlands’ Ministry of frameworks used in this study, including Greg Browder, Foreign Trade and Development Cooperation; Norway’s Anders Jägerskog, and Irene Rehberger Bescos. Efficient Ministry of Foreign Affairs; the Swedish International administrative support was provided by many, including Development Cooperation Agency; Switzerland’s State Lucson Pierre-Charles and Georgine Badou from the Secretariat for Economic Affairs; the Swiss Agency Water GP in Washington, D.C., and many dedicated for Development and Cooperation; the Rockefeller professionals in the Pakistan Country Management Unit Foundation; and the U.K. Department for International in Islamabad. Abbreviations ADB Asian Development Bank IMF International Monetary Fund ADP Annual Development Program IRSA Indus River System Authority AEDB Alternative Energy Development Board IRSM Indus River System Model AWB area water board IWMI International Water Management Institute AWS automatic weather stations IWRM Integrated Water Resources Management BAU business as usual KMC Karachi Metropolitan Corporation CGE computable general equilibrium KP Khyber Pakhtunkhwa CPEC China–Pakistan Economic Corridor  KWSB Karachi Water and Sewerage Board DPR delivery performance ratio LEP lower export price EPA environmental protection agency LGD local government department FATA Federally Administered Tribal Areas LIC low-income country FEWS flood early warning system MIC middle-income country FFC Federal Flood Commission MOWP Ministry of Water and Power GAMS General Algebraic Model System NDMA National Disaster Management Authority GDP gross domestic product NGO nongovernmental organization GLOF glacial lake outburst floods NPV net present value GoP Government of Pakistan NWP National Water Policy HEC Higher Education Commission O&M operation and maintenance HEP hydroelectric power OECD Organisation for Economic Co-operation and IBIS Indus Basin Irrigation System Development IBMR Indus Basin Model Revised PAMRA Punjab Agricultural Marketing Regulatory IFPRI International Food Policy Research Authority Institute PCIW Pakistan Commissioner for Indus Waters xiv PAKISTAN: GETTING MORE FROM WATER PCRWR Pakistan Council of Research in Water SAM Social Accounting Matrix Resources SCARP Salinity Control and Reclamation Projects PDMA provincial disaster management authority SRI System of Rice Intensification PHED public health engineering department SUPARCO Pakistan Space and Upper Atmosphere PID provincial irrigation department Research Commission PIDA provincial irrigation and drainage authority WAA Water Apportionment Accord PMD Pakistan Meteorological Department WAPDA Water and Power Development Authority PPP public-private partnership WASA water and sanitation agency PSDP Public-Sector Development Program WMO World Meteorological Organization RUMI reaching upper-middle-income WSTF Water Sector Task Force RWSM Regional Water System Model WUA water user association Executive Summary Is Pakistan ‘Water Secure’? GDP—around US$14 billion per year. Other economic contributions from water are difficult to accurately Water security describes the social, economic, and assess, but hydropower generation is economically environmental outcomes—beneficial and d ­ etrimental— significant, with a current market value of US$1 billion from how water is managed and used. Assessing to US$2 billion. these outcomes indicates that Pakistan is not water secure. Pakistan is well endowed with water—only The economic costs to Pakistan from poor water and 16 countries have more water—but because Pakistan sanitation, floods, and droughts are conservatively is the world’s sixth most populous country, water estimated to be 4 percent of GDP, or around availability per person is comparatively low. Fewer US$12 billion per year. These costs are dominated by than 10 percent of the global population lives in the costs of poor water supply and sanitation. The countries with less water per person. Water scarcity is economic costs of degradation of the Indus Delta are challenging but does not define a country’s economic estimated to be around US$2 billion per year, while the destiny. There are 32 countries with less water per costs of pollution and other environmental degradation person than Pakistan; across these countries the have not been assessed. These estimates of economic average per capita gross domestic product (GDP) is benefits and costs cannot be directly compared or 10 times that of Pakistan. Only six of these 32 water aggregated, but they demonstrate that Pakistan gets a scarce countries are poorer than Pakistan—all African poor economic return from its significant water resource. nations with little irrigation investment and a heavy However, for many Pakistanis, poor social outcomes reliance on traditional rainfed agriculture. from water best characterize water insecurity. Water- Pakistan does not make the best use of its water borne diseases are a leading cause of suffering and endowment. Water use is heavily dominated by death in Pakistan, reflecting widespread contamination agriculture, which contributes around one-fifth of of water supplies by sewage effluent. Poor water national GDP, but less than half of this is from irrigated supply, sanitation, and hygiene contribute to high cropping. Irrigation contributes around US$22 billion levels of childhood stunting, undermining human to annual GDP. The four major crops (wheat, rice, capital. Women and children are the most vulnerable, sugarcane, and cotton) that represent nearly 80 percent especially in rural areas where sanitation is particularly of all water use generate less than 5 percent of inadequate, and most water supplies are contaminated. xvi PAKISTAN: GETTING MORE FROM WATER Up to a quarter of the population may be at risk from recent decades have been achieved through increased arsenic contamination of drinking water. Floods and fertilizer use, additional labor, and a huge increase droughts also have significant social impacts, again in groundwater pumping. But there has been little affecting women and children the most. improvement in water use efficiency and very little intensification or transition toward higher-value crops. Scant attention is paid to the environmental o ­ utcomes Agricultural water productivity lags well behind that of from water in Pakistan, and water-dependent most other countries. ecosystems—rivers, lakes, wetlands, and the Indus ­ characterized Delta—are in rapid decline. This decline is ­ Irrigation service delivery is poor and contributes to low by biodiversity loss, greatly reduced stocks of ­freshwater productivity. Hydraulic efficiency of water distribution and estuarine fish stocks, and a loss of other ecosystem is very low, and water delivery across command areas services, including the storm protection afforded by is inequitable. Irrigation services are not financially coastal mangrove forests. Excessive water withdrawals sustainable and financial performance is declining. and widespread pollution are the main causes of Service tariffs are set too low and are decoupled decline, but river fragmentation by infrastructure and from service quality, and the operational costs of changed sediment regimes contribute. service providers are far too high. Poor operational performance in irrigation continues to exacerbate What Undermines Water waterlogging and salinization, especially in Sindh. Despite large-scale reclamation efforts, high water Security in Pakistan? withdrawals and poor drainage mean salt continues to Water security in Pakistan is undermined by poor water accumulate in soils and groundwater in the lower Indus resource management and poor water service delivery— Basin, affecting agricultural productivity. including irrigation and drainage services—and d­ omestic Domestic water supply coverage is high—especially for water supply and sanitation services. In addition, urban households, but coverage is declining because of some growing, long-term water-related risks are not rapid urbanization. And although coverage is high, the ­ adequately recognized and are poorly mitigated. quality of supply services is poor—especially in terms Water resource management is compromised by of water quality and reliability. Sanitation services are (i) poor water data, information, and analysis; (ii) weak variable: open defecation is increasingly uncommon processes for water resources planning and allocation; even in rural areas, but collection, treatment, and (iii) environmentally unsustainable levels of water disposal of sewage effluent are all grossly inadequate. withdrawal; (iv) widespread pollution; and (v) low Most water supplies are therefore contaminated. water productivity in agriculture. Inadequate monitoring Climate change is the biggest longer-term and currently and data management prevent robust water resource unmitigated external risk to Pakistan’s water sector. assessments and accounting to guide water planning Climate change is not expected to greatly alter average and management and prevents reliable flood and water availability over coming decades, but inflows drought forecasting. Water resources planning has will become more variable between and within years, historically focused on supply augmentation and increasing the severity of floods and droughts. Climate has not addressed sustainable resource use or been warming is expected to drive water demands up by linked adequately to broader economic planning. 5 percent to 15 percent by 2047, in addition to the Although provincial water shares have been formally demand increases from population and economic defined, they have been demonstrated to be growth. In the upper Indus Basin, accelerated glacial economically suboptimal, and there is insufficient melting will increase the risks of dangerous glacial lake clarity on risk sharing during times of acute scarcity. outburst floods. In the lower Indus Basin, sea level rise These deficiencies are expected to become starker and increases in the frequency and severity of coastal with increasing water demands and climate change. storms will exacerbate seawater intrusion into the delta Water resources management in Pakistan does little to and into coastal groundwater. In coastal Sindh, this will protect water-dependent ecosystems either by way of further degrade groundwater quality, ­ groundwater- environmental flows or pollution control. dependent ecosystems, and irrigation productivity. No formal mechanisms exist within provinces for A second overlooked risk is change in basin-scale reallocating water between sectors to match shifting river sediment dynamics. Sediment dynamics in demands or to cope with extreme drought. Irrigation the Indus—sourcing, transport, and deposition— water allocation is suboptimal in terms of efficiency, have been significantly altered by water resources equity, and transparency, contributing to the low development. Without greater attention, these changes productivity of irrigated agriculture and causing a will increasingly threaten the safety and operational lack of trust between farmers and service providers. performance of water infrastructure—and the health of Improvements in water productivity in agriculture in river and delta ecosystems. xvii How Well Are Water drawing on a range of data and past studies, suggests that Pakistan’s current total average annual renewable Resources Understood? resource is 229 billion cubic meters (BCM). Only Some of Pakistan’s water resources are well 4 percent of this is outside of the Indus Basin. qualified, while others are poorly assessed or simply Water availability per capita varies between years overlooked in most resource assessments. The surface because of climate fluctuations, but in recent decades water inflows to Pakistan from the Indus and its has declined because of population growth (see tributaries are measured sufficiently well to give high figure ES.2). There has also been a small but important confidence to average annual flows (see figure ES.1). reduction in inflow from the eastern tributaries of However, runoff generated within Pakistan— the Indus because of development in India, which is including in Balochistan outside the Indus Basin—is permitted under the Indus Waters Treaty. Currently, not well measured and is often ignored in resource average water availability is estimated to be around assessments. Groundwater has usually been quantified 1,100 cubic meters per capita, considering all renewable in terms of withdrawals, but this leads to a significant water resources. Withdrawals per capita have declined double counting in resource estimates: much of the with rising population, while actual consumption has groundwater is simply surface water withdrawals that remained a fairly constant proportion of withdrawals, seep from canals, distributaries, and fields into the given little improvement in water use efficiency. Water aquifers. A careful assessment of all water resources, demand is projected to rise, driven mostly by economic Figure ES.1  Indus Basin Average Annual Water Balance Indus, Kabul, Jhelum, Chenab inflows 170 Ravi, Sutlej, Beas inflows 3 Runoff inside Pakistan 32 Evaporation loss 41 Rainfall recharge 13 Crop water use and other Surface water beneficial consumption withdrawals 80 125 Canal Surface water flows leakage Basin inflows and outflows Irrigation Groundwater recharge recharge Saline Fresh withdrawals 27 Flow from surface water to 6 aquifer aquifer 62 ground water Water withdrawals Surface water Irrigation recharge 11 River and flood recharge 4 Groundwater Non-beneficial consumption Loss to the atmosphere Irrigation return flow 22 Loss to saline groundwater Return flows Outflows To surface water to delta Natural losses 68 To ground water 30 Sources: Ahmad and Rashida 2001; FAO 2011; Karimi et al. 2013; Laghari, Vanham, and Rauch 2012; MacDonald et al. 2016; van Steenbergen and Gohar 2005. Details are in appendix A, table A.1. Note: Flows are in billion cubic meters. xviii PAKISTAN: GETTING MORE FROM WATER Figure ES.2  Historical Water Availability, Withdrawals and Consumption per Capita; and Projected per Capita Availability and Demand 7,000 6,000 Cubic meters 5,000 4,000 3,000 2,000 1,000 0 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Historical consumption Historical withdrawals Future average demand Historical annual water Future average water resource resource Sources: WAPDA unpublished data and author calculations. and population growth, but also by climate warming. institutional strengthening; urban services; flood mitigation; Demand management will be critical to stay within and environmental management. the available resource envelope, as will efficiency The biggest challenges, however, are ones of improvements that can allow consumption to increase. governance, especially regarding irrigation and urban The converging supply and demand projections highlight water. The governance challenges relate to inadequate a key aspect of the water security challenge for Pakistan. legal frameworks for water at federal and provincial Water withdrawal in Pakistan—as a fraction of the levels, and the incompleteness of policy frameworks available resource—is high compared to that of most and the inadequacy of policy implementation. The other countries. However, adjusting for the double policy deficiencies stem from institutional problems counting of surface and groundwater withdrawals including unclear, incomplete, or overlapping reveals that water stress is less extreme than institutional mandates, and a lack of capacity in commonly quoted, although the stress on water water institutions at all levels. Behind these multiple ecosystems is still high. Severe groundwater depletion challenges in the formal governance arrangements are is evident in Lahore, Quetta, and parts of southern deeply embedded vested interests in the status quo Punjab. But depletion is a very small fraction of the that have proved resistant to reform. overall groundwater balance, and in any case, it The most important infrastructure gaps are associated follows decades of water-level rise caused by excessive with water supply and sanitation services and irrigation. Waterlogging remains a bigger problem, irrigation and drainage services. Wastewater treatment especially in Sindh, yet the greatest threat to long-term infrastructure is woefully inadequate for both urban and groundwater sustainability is contamination—both rural communities. Treatment capacity is inadequate, and salinization and other pollutants. existing infrastructure is poorly maintained and operated. Sewerage network coverage is very limited, and the What Interventions Can Improve current partial network is poorly maintained. Water distribution networks are often similarly inadequate. Water Security in Pakistan? Many rural areas lack both public water and sanitation There is no single simple solution to address water security infrastructure. While irrigation infrastructure is very in Pakistan. It will take concerted effort on many fronts by extensive following more than a century of incremental all governments and water users over many years. Large investment, the distribution network is outdated and infrastructure gaps must be addressed, which require poorly maintained. Despite considerable investment significant financial resources. Provincial-level water in drainage infrastructure, waterlogging continues sector financing has increased in recent years, but federal to worsen. Modernization of irrigation and drainage financing has declined significantly in proportional terms. infrastructure is required on a massive scale, including Collectively, sector financing is well below recommended upgrading flow control structures and installing real-time levels. This is the case for major infrastructure, reforms, and data acquisition systems for improved operations. xix Pakistan needs continued investment in flood Although population growth is slowing, projections protection infrastructure. The country has made suggest Pakistan’s population will exceed 300 million moderate progress in flood mitigation, given the by 2047, driving water demands much higher. Without significant damage and disruptions from floods over serious demand management and reform, and if the the last 50 years; however, climate change will climate warms rapidly, water demand could increase increase the risk of flood damage, meaning greater by nearly 60 percent by 2047. This would exceed water investment is required. Flood infrastructure should be availability, even if no environmental limits were placed complemented with “soft” measures such as floodplain on withdrawals. The largest increases in demand will zoning, improved flood forecasting, and early warnings. be for irrigation. Population and economic growth are Large storage reservoirs can help improve some the main drivers, but climate warming will contribute aspects of water security but do not address the significantly. The fastest rates of demand growth will be most pressing water security issues. New reservoirs for domestic and industrial supply. These changes require would deliver relatively modest additional yield, a major focus on demand management to improve and the water supply benefits would not justify the water use efficiency and water productivity. significant financial costs. It is only the benefits from For many years, adequate water availability and modest hydropower—either from storage or run-of-river urban demands have resulted in little urgency and few facilities—that justify new dams in economic terms. incentives to improve water use efficiency or to seriously New reservoirs will help mitigate floods and seasonal tackle demand management. Food production increased flow variations, both of which are expected to increase to keep pace with population growth, although food with climate change. Additional storage upstream of security was compromised by problems of affordability, Tarbela Dam will help to slow its incremental loss of access, and dietary diversity. Water planning and live storage caused by rapid sedimentation. investment was dominated by large supply-side projects The legal frameworks for water management need to that did not improve water productivity. be far more comprehensive. Out of 48 legal elements Water resource constraints mean a far stronger focus identified as important for sound water resources on demand management is required. Water losses management, only 16 to 19 are in the legal frameworks must be reduced, and water productivity growth must among the provinces. The 2018 National Water Policy be accelerated. It is commonly believed that Pakistan provides strong support for improving water resource has inadequate water storage, and that new reservoirs management, echoing other policy documents (e.g., the will dramatically enhance water supply. Planned new National Climate Change Policy). However, significant reservoirs will provide limited additional supply—and policy work is required at the provincial level, because of lower reliability. Reservoirs buffer inflow variations policy frameworks for irrigation and water resources to stabilize supply. Existing reservoirs adequately buffer management are partial, fragmented, or nonexistent, inflow variations between years, although supply and implementation has been inadequate. Provincial shortfalls in Rabi are common. New reservoirs would policy frameworks for urban water services should be improve the reliability of Rabi supply. But given the clarified and aligned with relevant legislation, including severe environmental degradation of the lower river that of local government. Institutional responsibilities for and delta, partly caused by high water withdrawals, several aspects of water resource management need to any increase in withdrawals, especially in drier years, be better delineated both national and provincial levels must be carefully assessed in terms of additional and between entities at these levels. The institutional environmental degradation. responsibilities for urban water need to be clarified and overlaps resolved. Changes in water allocation and use will be critical to driving economic growth. First, demand growth and Is Water Scarcity a Constraint changing demand patterns mean that meeting the to Reaching Upper-Middle- increasing and higher-value water demands outside of agriculture, will, within a few decades, limit the Income Status by 2047? growth in agricultural consumption of water. Until Because of sustained and rapid population growth, then, water consumption in agriculture can increase relative water availability has shrunk to less than a without increasing water withdrawals. This will require quarter of what of it was half a century ago. Municipal reforms and investments that dramatically reduce and industrial water demands are increasing, and water losses. Second, water will need to be secured environmental water allocations need to be agreed and and managed to protect water-related environmental implemented. attention. Economic modeling suggests services and benefits, especially those associated with however, that despite projected population increase and the Indus Delta. With additional storage this may be climate change, water scarcity will not prevent Pakistan possible without reducing withdrawals but confirming from reaching upper-middle-income status by 2047. this requires more detailed modeling. Third, the xx PAKISTAN: GETTING MORE FROM WATER use of water for irrigation needs to be dramatically Conversely, without major reforms, Pakistan would improved. To ensure continued food security and see only minor improvements in water productivity to contribute to accelerated economic growth, the and continued slow economic growth to reach only productivity of water in agriculture must be greatly US$2,200 GDP per capita by 2047. Urban water security increased. Reforming distorting agricultural policies would likely decline, and environmental degradation that support wheat and sugarcane will help move would worsen. A lack of resilience, especially to water toward higher-value crops. increasing drought severity, could lead increased conflict over water between provinces and sectors. Changes in diet—already apparent as incomes rise—will further change patterns of food demand. If production of low-value cereals declines in response to falling Twelve Recommendations for demand, more water may shift to growing cotton for Improving Water Security export. Cotton—and the associated textile industry— Twelve high-level recommendations emerge from generate considerable export income for Pakistan, and the analysis in this report: six for improved water should remain economically attractive over the long resource management, three for improved service term, especially if greater value-addition postharvest is delivery, and three for improved risk mitigation achieved. These benefits can only accrue, however, if (see table ES.1). These address the major areas of major reductions in water losses can be achieved. poor sector performance and would ensure water Assuming optimistic rates of economic growth, security is not a constraint to Pakistan’s economic modeling suggests Pakistan can reach upper-middle- development ambitions. The recommendations are income status (GDP of US$6,000 per capita) by 2047, qualitatively assessed in terms of complexity, urgency, ensure adequate food supply, improve environmental and scale of water security impact (bubble size) sustainability, and deliver better municipal and (see figure ES.3). For each recommendation, more industrial water security, even in the context of a actions are provided for reforming water governance rapidly warming climate. However, this will not (laws, policies, and institutions) and infrastructure be easy, and will require action on many fronts. investments. Table ES.1  Twelve Recommendations for Improving Water Security in Pakistan, Indicating the Required Legal, Policy, and Institutional Reforms and Necessary Infrastructure Investments Legal reforms Policy reforms Institutional reforms Infrastructure investments Water Resources Management Strengthen Water Data, Information, Mapping, Modeling, and Forecasting Clarify federal legal mandates for water information collation and sharing. Strengthen provincial legal frameworks for land-use planning that consider flood risks. Establish an implementation framework for the National Water Policy, with clear roles and responsibilities for water data and information. Develop standards and guidelines for flood risk mapping and a policy framework for floodplain zoning. Strengthen federal capacity for water data management, modeling, and forecasting, including the use of Earth Observations. Strengthen provincial capacity for monitoring and reporting water distribution and use. Strengthen federal capacity for flood risk mapping and flood forecasting. Build provincial capacity for floodplain zoning. Expand national and provincial hydromet networks, including for cryosphere and groundwater monitoring. Establish interoperable national and provincial water information systems. Establish a Multistakeholder Process of Basin-Scale Water Resources Planning Establish a sound legal mandate for federally led cooperative basin planning. Strengthen provincial legal frameworks for water resource planning. Establish an implementation framework for the National Water Policy that articulates roles, responsibilities, time frames, and processes for basin planning. Establish a National Water Council, as proposed in the National Water Policy, to provide strategic framing for cross-jurisdictional basin planning. Strengthen the federal government capacity for river basin management (either within the Indus River System Authority (IRSA), the Pakistan Water and Power Development Authority (WAPDA), or by establishing a new authority), in cooperation with provincial governments. Establish consultative processes for effective and broad stakeholder input. Establish Provincial Water Planning and Intersectoral Water Allocation Mechanisms Establish clear legal property rights (licenses) for water—separate from land—and the legal requirement to maintain public register of water licenses. Develop and implement provincial water policies to establish sectoral priorities and to define allocation processes. Incrementally transform provincial irrigation departments into water resources management agencies with broad responsibilities, including environmental management. table continues next page Establish robust participatory processes to guide water allocation planning. Accelerate Agricultural Water Productivity Increases Scope legal provisions to support pricing and trading of water rights. Strengthen the federal government capacity for river basin management (either within the Indus River System Authority (IRSA), the Pakistan Water and Power Development Authority (WAPDA), or by establishing a new authority), in cooperation with provincial governments. Establish consultative processes for effective and broad stakeholder input. Establish Provincial Water Planning and Intersectoral Water Allocation Mechanisms xxi Establish clear legal property rights (licenses) for water—separate from land—and the legal requirement to maintain public register of water licenses. Table ES.1  continued Develop and implement provincial water policies to establish sectoral priorities and to define allocation processes. Incrementally transform provincial irrigation departments into water resources management agencies with broad responsibilities, including environmental management. Establish robust participatory processes to guide water allocation planning. Accelerate Agricultural Water Productivity Increases Scope legal provisions to support pricing and trading of water rights. Phase out subsidies for wheat and sugarcane. Liberalize agricultural commodity markets. Support adoption of water efficiency technologies and diversification to higher-value crops. Strengthen capacity for economic modeling within federal and provincial governments. Improve on-farm water management through farmer training and awareness raising. Introduce methods of rice cultivation that require less water. Increase investment in agricultural research. Adopt Conjunctive Planning and Management of Surface and Groundwater Establish provincial-level regulatory frameworks for groundwater access and for management and regulation. Develop district-level conjunctive water management plans that focus on building drought resilience. Strengthen the capacity of provincial water resource management departments for groundwater management and conjunctive planning. Strengthen water user associations for local monitoring and management of groundwater resources in line with agreed conjunctive water management plans. Build federal capacity for basin-scale modeling and analysis of surface–groundwater interactions. Construct Limited New Storage and Review Reservoir Operations Review and revise reservoir standard operating procedures, based on detailed modeling and analysis. Strengthen federal capacity to enable periodic reviews of operating procedures and to support a multiobjective approach to operations. Secure finance for construction of Diamer Bhasha Dam and associated power generation and distribution infrastructure (if HEP justifies the expense). Water Services Delivery Modernize Irrigation and Drainage and Improve Operations Revise the Provincial Irrigation and Drainage Authorities Act to clarify roles and responsibilities in irrigation management between irrigation and drainage authorities and provincial government departments. Replace warabandi with new water sharing rules based on economic efficiency and farmer equity. Reform irrigation tariffs to reflect realistic operations and maintenance (O&M) costs. Strengthen the capacity with provincial government water resources management departments to oversee irrigation and drainage authorities and performance of water user associations and farmer organizations. Strengthen water user associations for improved system operation and improved abiana collection. Reform governance of water user associations and farmer organizations to prevent elite capture. Modernize irrigation systems, including new hydraulic control structures and lining of canals in waterlogged and saline areas. Automate control of hydraulic structures using real-time data acquisition systems. Systematically improve drainage infrastructure. Reform Urban Water Governance and Close the Infrastructure Gap Establish legal mandate for regulatory oversight of urban water service provider performance. Strengthen the regulatory framework for pollution discharges. Rationalize overlaps in the provincial policy frameworks and align with local government legislation. Develop and disseminate standards for urban water service delivery, and link service tariff increases to service quality. Strengthen and empower urban water service providers. Establish independent regulator to oversee service provider performance and to help reduce political interference. Establish an enabling environment for increasing private sector participation in urban water sector. Greatly increase the capacity and performance of wastewater treatment. Improve O&M of existing major distribution infrastructure. Increase the coverage and reliability of urban water meters. Improve Rural Sanitation Establish clear legal mandate for the provision of rural sanitation services. Establish provincial standards and targets for rural sanitation services. Strengthen the capacity and increase the financing of provincial government departments responsible for rural sanitation. Establish appropriate district-level institutional arrangements to engage with communities in infrastructure improvement. Establish appropriate mechanisms to ensure sustainable revenue base for O&M costs. Monitor and report progress toward rural sanitation targets. Invest in public infrastructure for rural sanitation services including wastewater collection and basic treatment and disposal at village level. Water-Related Risk Mitigation table continues next page Improve Understanding and Management of Climate Risks to the Lower Indus and Delta Develop long-term plans for sustainable management of the Indus Delta. Strengthen the technical capacity of water and environmental management agencies in Sindh for climate change impact assessments and mitigation planning. Resource relevant agencies for effective implementation of management plans. Establish provincial standards and targets for rural sanitation services. Strengthen the capacity and increase the financing of provincial government departments responsible for rural sanitation. Establish xxii appropriate district-level institutional arrangements to engage with communities in infrastructure improvement. PAKISTAN: GETTING MORE FROM WATER Establish appropriate mechanisms to ensure sustainable revenue base for O&M costs. Monitor and report progress toward rural sanitation targets. Invest ES.1  Table continued in public infrastructure for rural sanitation services including wastewater collection and basic treatment and disposal at village level. Water-Related Risk Mitigation Improve Understanding and Management of Climate Risks to the Lower Indus and Delta Develop long-term plans for sustainable management of the Indus Delta. Strengthen the technical capacity of water and environmental management agencies in Sindh for climate change impact assessments and mitigation planning. Resource relevant agencies for effective implementation of management plans. Assess the feasibility of barrier groundwater wells to slow sea water intrusion. Strengthen Planning and Management of Water–Energy Interactions Establish provincial-level regulatory frameworks for groundwater access and management. Analyze the synergies and antagonisms between current national energy and water policy frameworks to inform policy implementation. Increase coordination between government departments at federal and provincial levels. Strengthen capacity for joint energy–water analysis that considers economic and environmental outcomes. Expand solar and wind power investment where sensible. Explore feasibility for small-scale hydro on irrigation canals. Continue major HEP investment with run-of-river focus. Improve Understanding and Management of Basin-Scale Sediment Dynamics Develop a management plan to guide long-term, basin-scale sediment management. Strengthen capacity in relevant technical institutions for multiple aspects of sediment monitoring, modeling, and analysis. Ensure that new reservoir designs and barrage rehabilitation projects consider sediment-related risks to structural safety and operational performance. Figure ES.3  Complexity, Urgency, and Scale of Impact of Key Recommendations Water Services Delivery 2 3 1 Irrigation and drainage services 2 Urban governance and infrastructure More urgent 1 4 3 Rural sanitation 1 2 Water Resources Management 1 Data, information, and analysis 2 Participitory basin planning 3 Provincial planning and allocation 5 4 Agricultural water productivity 3 5 Conjunctive water management 6 New dams and revised operations Less urgent 6 1 3 Water-Related Risk Mitigation 1 Climate risks to the delta 2 Water-energy nexus 2 3 Basin sediment management Less complex More complex Note: Relative scale of impact is indicated by bubble sizes. xxiii Next Steps and transparent process for tracking and reporting implementation progress to demonstrate political Pakistan’s National Water Policy (2018) outlines many commitment and to ensure accountability. Given the of required reforms and investments to improve long history of significant interprovincial tensions water security. It can provide a platform for increased around water sharing, the establishment of a National sector dialog, especially between the provinces, Water Council as proposed in the National Water but also among diverse stakeholders within the Policy is fundamental. The National Water Council provinces. Establishing an implementation plan for should establish long-term social, environmental, and the National Water Policy that identifies agreed economic objectives for the management of the Indus priority actions with clear responsibilities is critical. Basin water resources in the national interest that Implementation will require realistic assessment and guide cooperative basin planning as well as provincial commitment to increased sector financing and a robust water management. C HAPT E R 1 Setting the Scene Economic, Demographic, been slower than in other Asian countries (Briones and Felipe 2013; Felipe 2007). This slower structural and Geographic Context transformation and slower movement of labor and Pakistan has experienced two decades of steady resources from low to high productivity sectors have economic growth (figure 1.1). By 2017, gross domestic constrained economic growth (Sanchez-Triana et al. product (GDP) exceeded US$300 billion (around 2014). The relative decline of the industry sector can US$1,500 per person) with an annual growth rate of be partly attributed to severe power shortages and 5.4 percent and strong performance in the agriculture, weak international competitiveness. Within agriculture services, and industry sectors (GoP 2017). This growth over the last decade, cropping has had much slower has delivered significant reductions in poverty, with productivity growth compared to that of livestock, such the headcount poverty rate falling from 64.3 percent that Pakistan’s major crops—that use most of the in 2001/02 to 24.3 percent in 2015/16. Demand-side water—now contribute less than 5 percent of GDP growth has been dominated by domestic consumption, (figure 1.3, panel b). and an increase in foreign investment from China Pakistan covers more than 880,000 square kilometers for China–Pakistan Economic Corridor projects has and comprises four provinces (Punjab, Khyber contributed to growth. Pakhtunkhwa, Sindh, and Balochistan, the Federally The structure of Pakistan’s economy changed Administered Tribal Areas (FATA), the Islamabad significantly between 1960 and 1990: the relative Capital Territory, and the Jammu and Kashmir region. contribution from agriculture to GDP fell from around The current population of Pakistan is estimated to 45 percent to around 25 percent between (figure 1.2). be nearly 208 million (table 1.1), making it the In the last decade structural change was very gradual. sixth most populous country in the world. Punjab The agricultural share in the economy declined and Sindh are home to more than three-quarters of slowly to be around 20 percent, similar to that of the national population. Population growth is high industry, while the services sector grew to be around (currently over 2 percent); however, the fertility rate 60 percent (figure 1.3, panel a). The decline in the has fallen from 6.5 percent to 3.5 percent over the relative contribution of agriculture to the economy last three decades. Further fertility decline is expected, (and in the employment share in agriculture) has and a medium population projection for 2050 is 307 2 PAKISTAN: GETTING MORE FROM WATER Figure 1.1  GDP of Pakistan, 2000–16 350,000 300,000 250,000 US$ (millions) 200,000 150,000 100,000 50,000 0 00 02 04 06 08 10 12 14 16 20 20 20 20 20 20 20 20 20 Source: GoP 2017. Note: GDP = gross domestic product. Figure 1.2  Sector Contributions to GDP in Pakistan, 1960–2000 60 50 40 Percent 30 20 10 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 Agriculture Industry Services Source: WDI https://datacatalog.worldbank.org/dataset/world-development-indicators. Figure 1.3  Sector and Agricultural Subsector Contributions to GDP in Pakistan, 2006–16 a. Sector contributions b. Agricultural subsector contributions 70 25 60 20 50 15 Percent Percent 40 30 10 20 5 10 0 0 6 7 8 9 0 1 12 13 14 15 16 6 7 8 9 0 1 2 3 4 5 6 0 0 0 0 1 1 0 0 0 0 1 1 1 1 1 1 1 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Agriculture Industry Services Agriculture Livestock All crops Major crops Other crops Source: GoP 2017. Note: GDP = gross domestic product. 3 Table 1.1  Population Growth across Provinces in the source of the Indus mainstream. To the west, the Pakistan Kabul, Kurram, and Gumal rivers all largely originate millions from Afghanistan, although key tributaries of the Kabul—the Kunar and Swat—rise in Pakistan, with the 1981 1998 2017 former tributary only then flowing into Afghanistan. Punjab 47.3 73.6 110.0 The extensive Indus floodplain is closely connected to Sindh 19.0 30.4 47.9 alluvial aquifers extending across 16 million hectares, of which 6 million hectares are fresh (mostly in Punjab) Khyber Pakhtunkhwa 11.1 17.7 30.5 and remainder saline (mostly in Sindh). The Indus Balochistan 4.3 6.6 12.3 exits through an extensive delta system to the Arabian FATA 2.2 3.2 5.0 Sea. The Balochistan Plateau, to the west of Pakistan, Islamabad Capital Territory 0.3 0.8 2.0 is a rugged, arid landscape characterized by several rivers, some flowing to the Arabian Sea, others into Pakistan 84.3 132.4 207.8 Afghanistan or the Islamic Republic of Iran. Only the Source: GoP 2018. Zhob and Kundar (tributaries of the Gumal) and the Note: FATA = Federally Administered Tribal Areas. Nari are within the Indus Basin. Pakistan has a semi-arid monsoonal climate, although physiographic diversity gives rise to very different Figure 1.4  Share of Urban and Rural Population in climates. Annual precipitation varies across Pakistan Pakistan, 1950–2050 from as much as 2,000 millimeters in the mountainous headwaters (mostly occurring as winter snowfall) to 100 less than 200 millimeters across most of the low-lying 90 and semi-arid expanse of the Indus plains and western 80 Balochistan. Across most of the country, around 70 60 percent of the precipitation falls between July 60 and September. The relative flow contributions from Percent 50 snowmelt, ice melt, and rainfall runoff vary among the 40 tributaries, reflecting catchment elevation. These flow 30 fractions have different seasonal patterns: snowmelt 20 peaks in June, and rainfall runoff and glacier melt peak 10 in August (Lutz et al. 2016). 0 1950 1970 1990 2010 2030 2050 Urban Rural Water Resources Overview Source: UN 2018. Pakistan comprises three hydrologic units: the Indus Basin, the Kharan Desert system, and the Makran coastal drainage. Most surface and the groundwater million (UN 2017). Pakistan is rapidly urbanizing and resources are in the Indus Basin. Pakistan’s geography is the most urbanized country in South Asia. By 2035 means that few interbasin transfers are economically around half the population is expected to be urban or technically feasible. Desalination of seawater or (figure 1.4). saline groundwater can help augment supply for high-value uses. Except for sparsely populated, semi-arid Balochistan, Pakistan is geographically and hydrologically defined Pakistan’s total water resource is somewhat uncertain, by the Indus Basin, which encompasses all of Punjab as data are limited (especially for Balochistan where and Khyber Pakhtunkhwa (KP), and most of Sindh the hydrology is highly variable), and a lack of (map 1.1). The Indus Basin extends across four robust water accounting means only approximate countries: Afghanistan (6 percent of the basin), resource estimates are available. Additionally, surface China (7 percent), India (34 percent), and Pakistan groundwater exchanges are not well quantified. The (53 percent). The Upper Indus has its headwaters in current total renewable water resource is estimated China; it then flows northwest through Jammu and to be 229 billion cubic meters or around 1,100 cubic Kashmir before turning sharply to exit the mountains meters per capita (table 1.2). This estimate includes through KP. Moving east, the Jhelum, Chenab, Ravi, and the water resources outside of the Indus Basin, as Sutlej (and its tributary the Beas) all flow from India well as the water within the Indus that is generated into the Pakistan province of Punjab. The headwaters within Pakistan. This estimate reflects the current level of the Sutlej are at high elevation in China, close to of flow to Pakistan from the eastern tributaries of the 4 PAKISTAN: GETTING MORE FROM WATER Map 1.1  Pakistan TAJIKISTAN UZBEKISTAN CHINA TURKM E N IS TA N TURKME Khunjrab Pass SH KU D U H IN Gilgit Indu Chitral s KHYBER Approximate PAKHTUNKHWA Line of Control Mansehra TARBELA Mansehra ISLAMABAD CAPITAL Jammu AFGHANISTAN Khyber Pass Mardan BARRAGE TERRITORY and Kashmir Jammu Peshawar Hasan Abdal ISLAMABAD and Kashmir Kohat FEDERALLY Rawalpindi BARRAGE MANGLA ADMINISTERED JINNAH NGE MARALA TRIBAL AREAS Bannu BARRAGE RA Dina BARRAGE LT BARRAGE RASUL m lu Gujrat Jammu Jhe SA CHASHMA lum KHANKI BARRAGE Jhe ab BARRAGE en Ch Gujranwala Sargodha D.I. Khan PUNJAB AN Lahore b BALLOKI Zho Faisalabad BARRAGE Zhob AIM TRIMMU BARRAGE FEROZEPUR vi Chaman SUL Ra BARRAGE TAUNSA SIDHNAI Sahlwal BARRAGE SULEIMANKE BARRAGE BARRAGE Quetta D.G. Khan Multan j tle BOLAN PASS Su us Lodhran ISLAM Ind BARRAGE Bahawalpur Nok Kundi Kalat Rajanpur Uch Sharif Kot Mithan BALOCHISTAN PANJNAD BARRAGE Surab GUDU BARRAGE Shikarpur I.R. OF Basima Khuzdar Ratodero Nag Larkana Ranipur IRAN KIRT SUKKUR RT Panjgur BARRAGE SE HAR DE N KRA Dadu AR MA Moro TH Bela SINDH INDIA Turbat Hoshab KOTRI Gwadar Pasni BARRAGE Hyderabad Irrigated areas Selected cities and towns Karachi Port Bin Oasam Thatta Badin Province capitals National capital us Ind Arabian Sea Province boundaries International boundaries 0 50 100 150 Kilometers 0 50 100 150 Miles IBRD 44137 | DECEMBER 2018 Indus (the Ravi, Sutlej, and Beas) allocated to India “rim stations”). The estimated fraction of the resource under the Indus Waters Treaty (1960). This estimate sourced from outside the country is high, at 74 percent is higher than other widely quoted estimates that (table 1.2), and yet the internally generated resource, consider only the major surface water inflows to the although poorly quantified, is important, and indeed Indus Basin irrigation system (measured at so-called critical for Balochistan. 5 Water withdrawals in Pakistan are high. The sum of surface water (diversions into irrigation canals) and annual surface water and groundwater withdrawals then leaked to groundwater from the canal system. is around 184 billion cubic meters, or 78 percent Of the groundwater withdrawals, around 70 percent of the total average annual resource (table 1.3). is supported by canal leakage and irrigation drainage, However, this total embodies a significant double the remainder being rainfall recharge and river counting error because much of the groundwater recharge (Laghari, Vanhamm, and Rauch 2012). withdrawal is water that was first withdrawn as Adjusting for this double counting error suggests a net annual withdrawal of around 136 billion cubic meters, or 59 percent of the total renewable water resource. Table 1.2  Estimated Contributions to Total An estimated 94 percent of withdrawals are for Average Annual Renewable Water Resource, agriculture, 5 percent for municipal use, and 1 percent Pakistan for industry (FAO 2011). BCM Actual water consumption is difficult to assess Indus Basin because it is not directly measured, and even indirect measurement is complex and rare. An estimated   External 90 percent of municipal and industrial withdrawals    Indus (including Kabul), Jhelum, Chenab 170.5 are not “consumed” but flow back to the rivers or    Ravi, Beas, Sutlej 3.3 to groundwater, albeit with much poorer quality. Of   Internal the water diverted into the major irrigation canals (122 billion cubic meters on average), a large fraction    Surface runoff 32.6 leaks from the distribution system (canals and    Groundwater rainfall recharge 12.7 distributaries) to groundwater. Of the water applied to   Total 219.1 fields, a considerable fraction is lost to evaporation, or Kharan Desert drains back to groundwater, rivers, or surface drains. Modeling suggests that actual water consumption by   Surface runoff 5.5 crops is around 80 billion cubic meters (table 1.3), or   Groundwater rainfall recharge 0.7 a little over 60 percent of the adjusted withdrawal   Total 6.2 volume. However, even this estimate includes some Makran Coast field-level evaporation. While some evaporation, especially from paddy rice, is unavoidable,   Surface runoff 2.9 improved agronomic practices can reduce field-level   Groundwater rainfall recharge 0.6 evaporation. Nearly 98 percent of total consumptive   Total 3.5 use is by irrigated crops. Grand total 228.8 The difference between water withdrawals and water Sources: FAO 2011; Halcrow 2007; Laghari, Vanhamm, and Rauch consumption in irrigation reflects both the high internal 2012; van Steenbergen et al. 2015; WAPDA unpublished data. leakage to groundwater and the high level of water Note: This resource estimate is based on data for different time lost to evaporation in irrigated areas. Water accounting periods, for different parts of the total resource, and quoted by different sources using differing assumptions. There is no complete, studies confirm that over half the water applied to consistent published total national resource estimate. BCM = billion fields for irrigation is lost to evaporation, with over cubic meters. 40 percent of this being associated with crop rotations involving paddy rice (Karimi et al. 2013). Total actual beneficial consumption is only around 36 percent of Table 1.3  Average Annual Water Withdrawal and the total average available resource. Consumption Volumes, Pakistan billion cubic meters Pakistan is commonly considered to be both water scarce (low water availability per capita) and water   Water   Water stressed (high water withdrawals high relative to withdrawal consumption water availability). However, in each case, important Canals 122 Irrigation 80 aspects of Pakistan’s water situation are commonly Groundwater 62 Livestock 1 overlooked. Most water scarcity assessments Total 184 Municipal 1 ignore the 24 percent of the total resource that is internally generated (including rainfall recharge to Double counting 48 Industrial <1 groundwater) (table 1.2). Average availability has Net withdrawal 136 Total 82 been declining with the rising population for many Sources: Amir and Habib 2015; FAO 2011; IFPRI CGE-W baseline decades, but also varies annually with climate model; Laghari, Vanhamm, and Rauch 2012. fluctuations (figure 1.5). Water stress is typically 6 PAKISTAN: GETTING MORE FROM WATER Figure 1.5  Historical Water Availability (1960­ –2016), Withdrawals (1975–2016), and Consumption (1975–2016); and Projected Availability and Demand to 2047 7,000 6,000 Cubic meters per capita 5,000 4,000 3,000 2,000 1,000 0 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Historical consumption Historical withdrawals Future average demand Historical annual water Future average water resource resource Sources: GoP 2017 and author calculations. noted as being very high. FAO (2011) indicates internal water resource varies between years in withdrawals are 74 percent of the total renewable proportion to the variation in measured inflows, and resource. The resource estimate in table 1.2 and that future average annual inflows will be unchanged total withdrawal estimate in table 1.3 suggest a from the recent past. The demand projection assumes stress level of 80 percent. Adjusting for the double a 1.3 percent annual growth in demand based on the counting inherent in the withdrawal total, however, projection of Amir and Habib (2015) for 3 degrees indicates a less alarming stress level of 59 percent. Celsius of global warming by 2050 (see chapters Pakistan is indeed “water stressed,” but perhaps 5 and 6). Adjustments to account for the double less so than typically assumed. This highlights the counting of surface and groundwater withdrawals importance of understanding the internal recycling of are made for both historical withdrawals and future water between the rivers, canals, and groundwater demand. The resource and demand projections system. During drought years, withdrawals are assume the population will grow to more than kept high despite low system inflows, and thus the 300 million by 2047. Consumption is not projected, level of water stress—and environmental impact— because this will depend on water management— rises considerably. Only during the worst drought especially improvements in water use efficiency. years (e.g., 2001/02) is water availability seen to The water resource as assessed is not completely significantly constrain water withdrawal (figure 1.5). available to meet consumptive demands: a fraction is unregulated peak monsoon flows. With current Average annual water withdrawal per capita is available storage, and without improved demand currently around 885 cubic meters, compared to around management, future water demand would exceed 600 cubic meters in India, 420 cubic meters in China, supply, even in the absence of a reasonable allowance and 560 cubic meters in Turkey. For Pakistan, this value for environmental water. includes a significant double counting error between surface and groundwater. Adjusting for this indicates These simple projections of steady total supply and net withdrawal per capita is around 655 cubic meters. increasing demand imply a gradual increase in the (Potential double counting in other national estimates average level of water stress. In addition, because has not been assessed). As noted previously, only interannual variability of water availability is expected around 60 percent of water withdrawal is actually to increase, supply limits will more frequently constrain consumed by crops. withdrawals and cause more frequent and severe instances of high environmental stress. As discussed Simple per capita projections of water resource and in chapter 3, active management of water storage— withdrawals highlight the macro water resource reservoirs and groundwater—can help buffer these challenge facing Pakistan (figure 1.5). The water variations. Other aspects of the future supply-demand resource projection assumes that the ungauged challenge are explored in chapter 6, both from water 7 resource management and economic perspectives. This and compared. Water scarcity does not preclude includes more granularity on demand projections by reaching upper- or middle-income status, and although sector and considering the role of climate warming on water is important in many economies, no upper- demand increase. It also includes consideration of the income countries and few middle-income economies crucial issues of water use efficiency and productivity. rely heavily on irrigated agriculture (figure 1.6). As highlighted in figure 1.5, actual consumption is far Pakistan is a lower-middle income agricultural lower than withdrawal; therefore, there is considerable economy—most of the lower economies are rainfed opportunity to increase water consumption for greater agriculture dominated—but it needs to transition production through enhanced efficiency. away from its reliance on agriculture as the engine of economic growth. Pakistan’s Water Economy This report explores the ability of Pakistan to transition to an upper-middle-income country by 2047 in the in the Global Context context of increasing relative water scarcity given a It is useful to briefly locate Pakistan’s water economy growing population. The trajectory of this transition in the global context. Using World Bank economic is indicated on figure 1.6, with Pakistan needing data and FAO agricultural and water resources data, to move to the approximate current position of metrics of economic productivity, water availability, South Africa. Pakistan’s position on a trajectory of water productivity, and water stress are calculated structural economic transformation suggests that to Figure 1.6  Global GDP per Capita and Total Renewable Water Resources per Capita More scarce Less scarce 100,000 CHE Upper income SGP DNK US GBR NLD AUT BEL DEU FRA ISR JPN ITA KOR PRI ESP TTO PRT CZE GRC EST OMN CY SVK LTU POL 10,000 KAZ MUS TU MEX GDP (2015 US$ per capita) LBN CHN BGR CUB TKM DOM BWA Middle income ZAF AZE THA BLR IRN JAM LBY JOR 2047 IRQ MKD DZA LKA SLV AGO BIH TU GTM EGY CP NGA ARM SWZ IDN PSE MAY PHL SDN UZB SYR UKR VNM DJI IND MDA KEN PAK GHA MRT CIV ZMB TLSBGD KGZ 1,000 LSO TJK SEN ZWE HTI COM TZA BEN TCD RWA SS MLI NPL UGA BFA ETH AFG SOM ERI TGO MOZ Lower income GMB MWI NER BDI 100 100 1,000 10,000 Total renewable resources (m3 per capita) Source: http://www.fao.org/nr/water/aquastat/main/index.stm. Note: Countries in blue have more than 15 percent of GDP in agriculture. Many countries—not shown—have more than 10,000 m3 per capita and a few have less than 100 m3 per capita. Country codes at http://www.fao.org/countryprofiles/iso3list/en/. 8 PAKISTAN: GETTING MORE FROM WATER Figure 1.7  Global Structural Transformation Trajectory 1,00,000 Pakistan 10,000 GDP per capita 1,000 100 70 60 50 40 30 20 10 0 Percent of GDP from agriculture Source: http://www.fao.org/nr/water/aquastat/main/index.stm. Note: Bubbles represent countries, with size scaled for water productivity (GDP per cubic meter of withdrawal). GDP = gross domestic product. reach middle-income status, it will have to reduce Pakistan is in the highest decile of countries in terms the share of agriculture in the economy, as others of water stress. Adjusting for the double counting in that status have done (figure 1.7). The latter error gives a lower and more accurate assessment stages of this transformation see a consequential of water stress but does not greatly shift Pakistan’s increase in the overall economic productivity of water. position on figure 1.8. This level of water stress is Pakistan’s water productivity is currently very low typically associated with considerable environmental even considering its current position on the structural degradation and makes managing water supply transformation curve. fluctuations difficult. An indicative trajectory for Pakistan (assuming GDP growth to reach upper- Water productivity is an indicator of the economic middle-income status by 2047) is shown on figure 1.8. output per unit of water withdrawn from the This trajectory reflects a reduction in overall water environment. Countries with high levels of water use through efficiency improvements to enable productivity ensure there is secure water for high-value reallocation of water to meet priority environmental sectors of the economy. Countries managing their needs. Dimensions of this trajectory, primarily relating water resources sustainably ensure water stress does to irrigated agriculture, are explored using modeling not creep too high (figure 1.8). Countries withdrawing in chapter 6. or using more than the renewable resources—that is, with water stress exceeding 100 percent—are either Water scarcity and water productivity are just two mining groundwater, significantly supplementing supply aspects of the much broader concept of water with desalination, or as in the case of the Arab Republic security that is explored and evaluated in this of Egypt, withdrawing the same water multiple times diagnostic. Assessing water security and identifying given internal recycling within the irrigation system. water sector priorities require going beyond single- Pakistan is in the lowest 5 percent of countries issue indicators of water or economic performance in terms of water productivity, indicating water is to consider the social and environmental outcomes comparative far less productive in economic terms from water management, service delivery, and risk and in most countries. Even considering productivity mitigation, and how these in turn are enabled or of water only in the agricultural sector, Pakistan is still constrained by water governance, infrastructure, in the lowest decile. and financing. 9 Figure 1.8  Water Productivity and Surface Water Stress by Country Less stressed More stressed 500 DNK GBR IRL CHE QA SWE SVK More productive ISR NOR CZE LVA AUT DEU HRV FRA CY LTU AU DJI TTO NLD B CP BHR BWA PA 50 BIH JPN COM LSO KOR POL Less productive RWACAN Water productivity (US$ per m3) UGA BLR SV NGA US NAM NZL FIN ITA ESP JOR GHA ROU OMN HUN PRI LBN PSE BR MYS PRT RUS CRI MNG GRC ZAF MKD KEN CHN DZA TGO CIV GTM JAM TU VE URY MUS ARG PE EST MEX TU SS SLV PRY HND BFA CUB MOZ BGR ZMB TCDECU SR ALB DOM NIC GEO KAZ MAR YEM SUR BDI MDA THA GMB CHL IDN UKR HTI LKA TZA ETH 2047 NER 5 BGD GNB SEN AZE ERI IRN EGY MWI ARM MRT ZWE LAO PHL SWZ SDN IND SYR GUY VNM NPL MMR MLI IRQ SOM PAK TKM TLS UZB KGZ AFG MDG TJK 0.5 0.5 5 50 500 Water stress (withdrawal as % of total renewable resource) Source: http://www.fao.org/nr/water/aquastat/main/index.stm. Note: Countries more than 15 percent of GDP in agriculture in green. Country codes at http://www.fao.org/countryprofiles/iso3list/en/. Purpose and Structure of Report In 2018, Pakistan’s federal and provincial governments jointly adopted a National Water Policy, acknowledging Pakistan has seen steady economic growth and the criticality of water to Pakistan’s economic prospects reductions in poverty levels over recent decades, and stability. It provides the first comprehensive but these will be difficult to sustain without greatly policy framework to guide coordinated water reform improved water security. There are many economic and investment across Pakistan. To facilitate faster growth trajectories Pakistan could follow, but growth, Pakistan should leverage the approval of these differ in the degree to which they improve the 2018 National Water Policy to act decisively on water security. Population growth, limited water water. Long-standing issues that have made water resource, and increasing climate change suggest security elusive need to be addressed—as well as that without careful attention, water issues could emerging issues. Slower or incomplete reform may disrupt development progress—economically, socially, lead to more frequent or more significant water-related environmentally, and politically. disruptions to economic growth—or disruptions to 10 PAKISTAN: GETTING MORE FROM WATER social and political stability—together with ever greater Figure 1.9  Water Security Diagnostic Framework environmental degradation. Much has been said and written about water in People Pakistan over the last few decades. The federal ices Mitig government, international organizations, and civil serv ati on d society have all provided assessments of Pakistan’s te rastructur ela of water sector. Two particularly influential reports are f wa r-r Pakistan’s Water Economy: Running Dry (Briscoe and n very of wate ter- e Qamar 2005) and A Productive and Water-Secure I related risks Pakistan (FoDP 2012). Briscoe and Qamar (2005) Water identify 14 sobering facts and five hopeful facts to endowment highlight critical water sector issues for Pakistan to Envi Deli address while working toward water security. FoDP ns my (2012) identifies five priority areas and key actions to titution I s ron address water sector challenges: (i) major infrastructure o an on M es and associated institutions, (ii) raising agricultural c m ag ur em so c e productivity, (iii) living with floods, (iv) sustainable er re E e nt of w at nt urban services, and (v) knowledge management. Many findings of these and other studies remain valid; however, changing demographics and economics, new information on climate change, emerging energy and agricultural technologies, and greater attention architecture and performance—and how these to the political economy are reframing the challenges determine outcome—lead to recommendations for and opportunities. improving aspects of sector performance and adjusting This report builds on prior work to provide a new, sector architecture for better outcomes. The analysis comprehensive, and balanced view of water security, of sector performance considers: (i) management of stressing the importance of the diverse social, the water resource, (ii) delivery of water services, and environmental, and economic outcomes from water. (iii) mitigation of water-related risks. The description The report highlights the complex water issues that of sector architecture considers water governance, Pakistan must tackle to improve water security infrastructure, and financing. and sheds new light on conventional assumptions The remaining chapters of the report are as follows: around water. It seeks to elevate water security as an issue critical for national development—not solely a • Chapter 2 describes the positive and negative challenge for the water sector. outcomes of water for Pakistan’s economy, people and society, and the environment. The report assesses current water security and • Chapter 3 describes the extent, distribution, identifies important water-related challenges that variability, and quality of Pakistan’s surface and may hinder progress in economic and human groundwater resources. development. It identifies unmitigated water-related risks, as well as opportunities in which water can • Chapter 4 describes Pakistan’s water sector architecture—infrastructure, water governance (legal contribute to economic growth and poverty reduction. framework, policy, and institutions), and financing. While some are well-known risks and opportunities, others are emerging. Some are the result of rapidly • Chapter 5 assesses Pakistan’s water sector growing environmental and demographic pressures, performance in terms of managing water and others have simply been overlooked. The report resources, delivering water services, and mitigating water-related risks, and highlights where sector analyzes how the performance and architecture of performance or architecture must improve to deliver the water sector relate to broader economic, social, better outcomes. and environmental outcomes. It models alternative economic trajectories to identify how intervention can • Chapter 6 considers the extent to which water lead to a more water secure future. may enable or constrain Pakistan reaching upper- middle-income status by 2047 and describes The report adopts a conceptual framework of water potential trajectories for social and environmental security (figure 1.9) that highlights the balance of outcomes from water. economic, social, and environmental outcomes (costs • Chapter 7 summarizes the report’s key findings and benefits) from water and the appropriateness and recommends priority areas for reform and of this balance. A consideration of water sector investment. 11 References ———. 2018. Pakistan Population Census 2017, 1998, 1981. Pakistan Bureau of Statistics, GoP. http:// Amir, P., and Z. Habib. 2015. Estimating the Impacts www.pbscensus.gov.pk/. of Climate Change on Sectoral Water Demand in Pakistan. Nottinghamshire, U.K.: ACT. https:// Halcrow Group. 2007. Supporting Public Resource cyphynets.lums.edu.pk/images/Readings​ Management in Balochistan. Basin-Wide Water _­concluding.pdf. (accessed June 2018) Resources Availability and Use. Irrigation and Power Department, Government of Balochistan, Briones, R., and J. Felipe. 2013. “Agriculture and Royal Netherlands Government. Halcrow Pakistan Structural Transformation in Developing Asia: (Pvt) in association with Cameos. Review and Outlook.” ADB Economics Working Paper 363. ADB, Manila, Philippines. Karimi, P., W. G. M. Bastiaanssen, D. Molden, and M. J. M. Cheema. 2013. “Basin-Wide Water Briscoe, J., and U. Qamar. 2005. Pakistan’s Water Accounting Based on Remote Sensing Data: An Economy: Running Dry. Washington, DC: Application for the Indus Basin.” Hydrology and World Bank. http://documents.worldbank.org​ Earth System Sciences 17: 2473–86. /curated/en/989891468059352743/Pakistans​ -water-economy-running-dry. (accessed June Laghari, A. N., D. Vanhamm, and W. Rauch. 2012. “The 2018) Indus Basin in the Framework of Current and Future Water Resources Management.” Hydrology FAO (Food and Agriculture Organization). 2011. and Earth System Sciences 16 (4): 1063–83. “Pakistan Aquastat Country Profile” (database). http://www.fao.org/nr/water/aquastat​ Lutz, A. F., W. W. Immerzeel, P. D. A. Kraaijenbrink, /countries_regions/PAK/PAK-CP_eng.pdf. A. B. Shrestha, and M. F. P. Bierkens. 2016. “Climate (accessed June 2018) Change Impacts on the Upper Indus Hydrology: Sources, Shifts and Extremes.” PLOS One 11(11): Felipe, J. 2007. “A Note on Competitiveness e0165630. doi:10.1371/journal.pone.0165630. and Structural Transformation in Pakistan.” ERD Working Paper 110, ADB, Manila, Sanchez-Triana, E., D. Biller, I. Nabi, L. Ortolano, Philippines. G. Dezfuli, J. Afzal, and E. Santiago. 2014. Revitalizing Industrial Growth in Pakistan: Trade, FoDP (Friends of Democratic Pakistan). 2012. Infrastructure and Environmental Performance. A Productive and Water-Secure Pakistan: Directions in Development—Private Sector Infrastructure, Institutions, Strategy. Islamabad, Development. Washington, DC: World Bank. Pakistan: Water Sector Task Force, FoDP. http:// metameta.nl/wp-content/uploads/2013/11​ UN (United Nations). 2018. World Population Prospects: /FoDP-WSTF-Report-Final-09-29-12.pdf. The 2018 Revision. New York: UN Department of Economic and Social Affairs. GoP (Government of Pakistan). 2017. Pakistan Statistical Yearbook 2016. Pakistan Bureau of van Steenbergen, F., A. B. Kaisarani, N. U. Khan, and Statistics, GoP. http://www.pbs.gov.pk/sites​ M. S. Gohar. 2015. “A Case of Groundwater /default/files//PAKISTAN%20STATISTICAL%20 Depletion in Balochistan, Pakistan: Enter into the YEAR%20BOOK%2C%202016.pdf (accessed Void.” Journal of Hydrology: Regional Studies June 2018). 4 (Part A): 36–47. C HAPT E R 2 What Pakistan Gets from Its Water Key Messages • Irrigation, predominantly in Punjab, contributes around US$22 billion to the economy annually. The four major crops (wheat, rice, sugarcane, and cotton) contribute US$14 billion (less than 5 percent of GDP) but are responsible for 80 percent of all water use. The full agricultural sector (including cropping, livestock, forestry, and fisheries) employs 43 percent of the labor force. • Despite improvements in recent decades, both land and water productivity are low. Given rapid population growth, food security remains a major challenge. Food production is sufficient, but deficiencies in food procurement, storage, and distribution undermine food security. • The potential to increase agricultural productivity by increasing inputs is now limited because groundwater is overexploited, land is nearly fully used, mechanized ploughing is widespread, and fertilizer use is high. Productivity improvements will require better control of water delivery, better on-farm water management, increased input quality (e.g., seeds), crop diversification, and better pest control. • Hydropower represents around 30 percent of national power generation, a much smaller share than in past decades, but a major contribution to the economy nonetheless. Pakistan has considerable untapped hydropower potential, but there are many complexities and challenges associated with exploiting this potential. • Inadequate water supply and sanitation, flood damage to property, and water scarcity for agriculture cost Pakistan an estimated 4 percent of GDP annually, with three-quarters of this associated with inadequate water supply and sanitation services. • Water-related diseases are a leading cause of suffering and death in Pakistan, and poor water supply, sanitation, and hygiene contribute to very high levels of childhood stunting. Domestic water supplies are generally unsafe, with contamination by sewage effluent, industrial effluent, and geogenic arsenic common, but poorly assessed, especially in rural areas. • Pakistan’s water-dependent ecosystems are under increasing stress from high levels of water withdrawal, widespread pollution, rapid urbanization, and agricultural expansion. Biodiversity loss, declining fish stocks, and degradation of the ecosystems of the Indus Delta, which offer valuable ecosystem services, are increasing, with little effort to monitor or mitigate this damage. 14 PAKISTAN: GETTING MORE FROM WATER T his chapter assesses the economic, social, and People environmental outcomes from water in Pakistan. Mitig ices Economic outcomes are considered in terms erv ati eds on t of the productive (benefit) and destructive (cost) of ru ela rast ctur wa nf r-r outcomes from water. The major economic benefits ter- ivery of wate e I are from irrigation and hydropower. The economic related risks Water benefits of improved domestic water supply and endowment sanitation are unquantified but are likely to be Envir Del ns y very significant. This chapter treats them as social titution om I s o outcomes. Major economic costs are associated M s nm na n ce ur a ge co with inadequate water supply and sanitation, me so nt of w ater re e E nt floods, droughts, poor water quality, and the loss of ecosystem services. Many economic outcomes from water are closely linked to social outcomes, given strong connections between the agricultural economy and social transformation—including rural and cropping represents 37 percent (figure 2.2). to urban migration. Human health and well-­ being The remaining value comes from cotton ginning, and social dynamics and conflict are the main fisheries, and forestry. Livestock production uses social outcomes. Environmental outcomes include very little water compared to irrigated cropping. freshwater ecosystems and their biodiversity, The relative economic contribution from livestock is pollution, eutrophication, and other water quality steadily increasing as the relative contribution from problems. cropping declines (figure 2.1). While a diverse mix of crops is grown in Pakistan, around three-­ quarters Economic Outcomes of the area and two-thirds of the value comes from two food crops (wheat and rice) and two cash Economic Benefits crops (sugarcane and cotton). The direct benefits to the economy from irrigation are the order of Agriculture has an important, although declining US$22 billion per year. role, in the Pakistan economy. It currently contributes a little under one-quarter of GDP. On an area The four major crops, which are responsible for around basis, cropping is by far the dominant agricultural 80 percent of agricultural water consumption, currently activity, but from an economic perspective, livestock contribute less than 5 percent of total GDP, and this production dominates. Livestock currently represents share is in decline. Figure 2.2 shows the fractions 58 percent of the agricultural GDP contribution, of water use and water-dependent agricultural GDP Figure 2.1  Share of Cropping and Livestock Contributions to Agricultural GDP in Pakistan, 2006–16 70 65 60 55 Percent 50 45 40 35 30 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Cropping Livestock Source: GoP 2017. Note: GDP = gross domestic product. 15 of the main agricultural subsectors, indicating major For the thirsty four major crops, the areas have differences in water productivity: livestock is well ­ remained reasonably stable over the last decade, with above the 1:1 line while major crops are well below over half the area dedicated to wheat (figure 2.3). this line. This contrasts with other Asian countries in which the composition of output has shifted toward higher- The agricultural sector employs around 43 percent of value crops driven by growing demand for fruits the labor force; but the fraction directly involved in and vegetables, contributing to raising the economic irrigated cropping is uncertain. However, most irrigation returns from water in agriculture; this has not been is still undertaken by with small farming households observed in Pakistan. that also own livestock. Irrigation in Pakistan is dominated by Punjab (73 ­percent of the total irrigated area), but with significant areas of all crops also grown in Sindh Figure 2.2  Share of Agricultural Water Use and (table 2.1). Punjab produces significantly more than Water-Dependent Agricultural to GDP in Pakistan, required to meet the provincial food demand and 2016 thus dominates exports, both to other provinces 100 and overseas. Khyber Pakhtunkhwa (KP) has a Share of water-dependent agricultural GDP (%) greater reliance on rainfed agriculture, but it imports from Punjab to meet the provincial demand 80 for food. The nature of irrigation also differs between the 60 Livestock provinces. In addition to the largest share of the canal water, Punjab has access to very significant groundwater resources, with 80 percent of the 40 irrigated area being at least partially dependent on groundwater (often in the rabi season) (figure 2.2). Major crops In Sindh, much of the groundwater resource is 20 saline—either naturally or because of poor irrigation Other crops management—and is thus not a useful agricultural resource. 0 20 40 60 80 100 Yields per hectare are very low by global standards. Share of agricultural water use (%) Average yields for the major food crops (table 2.2) Source: PBS 2016. are 1.5 to 4.2 times below field potential and 2.1 to Figure 2.3  Irrigated Areas for the Four Major Irrigated Crops in Pakistan, Fiscal Years 2007–16 18 16 14 Hectares (millions) 12 10 8 6 4 2 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Wheat Cotton Rice Sugarcane Source: PBS 2016. 16 PAKISTAN: GETTING MORE FROM WATER Table 2.1  Distribution by Irrigated Area across 5.6 times below international best practice (Aslam Provinces of Four Major Crops in Pakistan, 2016 2016). Although Punjab dominates production, yields of the four major crops are considerably higher in Punjab Sindh Khyber Balochistan Sindh. Yields in KP and Balochistan are well below Pakhtunkhwa national averages for all crops, other than rice in Wheat 75.2 12.0 8.4 4.3 Balochistan (table 2.2). If the current wheat and rice Rice 67.7 24.0 2.1 6.2 yields in Sindh could be achieved across Pakistan, Maize 55.1 0.3 44.2 0.5 total production would increase 28 percent and 46 percent, respectively. Sugarcane 66.2 24.1 9.7 0.1 Cotton 80.2 18.6 0 1.2 Yields have improved over the last three decades but yield growth has been slow for the four major irrigated Source: PBS 2016. crops (figure 2.5). Annual yield growth has been highest for wheat, averaging 2.6 percent (just above the population growth rate). Figure 2.4  Average of Irrigated Areas by Province While yield per unit area is an important metric for and Water Source in Pakistan, 2006–16 benchmarking performance, production per unit of irrigation water (or water productivity) is also a 16 critically important metric when water is scarce. Over the decade following the 2000–02 drought, water 14 withdrawals for irrigation (combining surface and groundwater) have not increased, and thus the small 12 gains in yield reflect small improvements in water productivity. These improvements have been largely Hectares (millions) 10 achieved by increasing inputs such as fertilizer and 8 mechanization. Water productivity can also be considered in the 6 economic value generated per unit volume of water withdrawn. The economic return per unit of 4 total water withdrawn (surface and groundwater) is significantly higher in Punjab than in Sindh 2 (table 2.3), even though yields per hectare are 0 generally much lower. The far lower economic Balochistan KP Punjab Sindh productivity of water in Sindh is because (a) in Canal area Sindh, the impacts of water logging and drainage are Canal and groundwater area greater; (b) in Punjab, groundwater provides greater Groundwater area irrigation control (especially during rabi season); (3) in Sindh, water losses are a greater fraction of Source: PBS 2016. withdrawals given higher temperatures and lower Note: KP = Khyber Pakhtunkhwa. humidity; and (d) in Sindh, a greater proportion of the irrigated area is devoted to rice, which has higher evaporative losses compared to other crops. Table 2.2  Average Yields of Major Irrigated Crops Nationally and by Province, 2006–16 tonnes per hectare Punjab Sindh Khyber Pakhtunkhwa Balochistan Pakistan Wheat 2.7 3.5 1.6 2.1 2.7 Rice 1.9 3.4 2.0 2.8 2.3 Sugarcane 54.5 57.9 45.7 48.2 54.3 Cotton 0.7 1.0 0.4 0.4 0.7 Source: PBS 2016. 17 Figure 2.5  Yield Index for Major Irrigated Crops in Pakistan, 1987–2015 2.0 1.8 Proportion of 1986 base year 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 1985 1990 1995 2000 2005 2010 2015 Wheat Cotton Rice Sugarcane Source: Federal and provincial statistics bureaus. Table 2.3  Average Land and Water Productivity in Punjab and Sindh at 1980 Prices, 2009–13 Irrigated area Annual water Mean annual irrigation Annual revenue Land productivity Water productivity (Mha) withdrawn (bcm) depth (mm) (US$, millions) (US$/ha) (US$/m3) Punjab 12.2 119 975 9,931 817 0.08 Sindh 2.5 50 2,000 2,944 1,169 0.06 Source: Federal and provincial statistics bureaus. These factors can be partly inferred from the average control, better on-farm water management, higher values for area-based water withdrawals (table 2.3), input quality (e.g., seeds), and better pest c ­ ontrol. which indicate average annual total irrigation depths Technology thus has a key role to play. of 0.96 meters in Punjab and 2.20 meters in Sindh Irrigated agriculture, especially wheat, is the foundation of (values are based on withdrawals). Given the high food security for Pakistan. Per capita wheat consumption losses in the distribution system, actual water is among the highest in the world (USDA 2017) and application at field level is much lower (estimated represents 72 percent of the daily caloric intake. Pakistan average crop water requirements for Pakistan are has significantly increased food supply, more than keeping around 0.4 meters for wheat and around 1.0 meters pace with population growth, thus improving food for rice (Linstead, Sayed, and Naqvi 2015). security (Kirby et al. 2017). Increases in rice production, The economic return from irrigation water has however, have not mirrored wheat, with a slowdown in doubled over the last three decades from around production and yield growth in the decade around 1980. US$0.03 to US$0.06 in Sindh and from around Given significant rice exports and rapid population growth, US$0.04 to US$0.08 in Punjab (figure 2.6). This has per capita availability of rice has declined significantly been achieved through expansion of the irrigated in recent decades (Kirby et al. 2017). Despite expansion area, increased groundwater use in Punjab, increased of the irrigated area and increases in yield, food security use of fertilizer and mechanization, and some remains a serious challenge. Food production exceeds improvements in water management. The potential demand, but deficiencies in the food procurement, to increase yields through additional inputs is now storage, and distribution systems undermine food limited: groundwater is overexploited, arable land security (Hussain and Routray 2012). Food access is thus is nearly fully used, 90 percent of ploughing is uneven, malnutrition is high among certain groups, and mechanized, and fertilizer use is often (although not there are widespread micronutrient deficiencies (Davies uniformly) high. Water productivity needs to improve et al. 2018). In 2014, 47 percent of the population were markedly if Pakistan is to revitalize economic assessed as being food insecure (WFP 2014), and Pakistan growth, and should come from better water delivery ranks among the bottom third of countries surveyed by 18 PAKISTAN: GETTING MORE FROM WATER Figure 2.6  Value of Water in Irrigation for Punjab and Sindh Relative to 1980 Value for Punjab, 1980–2013 2.5 2.0 US$, constant 1980 prices 1.5 1.0 0.5 0 1980 1985 1990 1995 2000 2005 2010 2015 Sindh Punjab Source: Federal and provincial statistics bureaus. the Global Food Security Index. By 2030, annual grain unexploited hydropower potential in the upper Indus demand may exceed 36 million tonnes (Hayat, Hussain, Basin, the complexities of extreme weather, difficult and Yousaf 2016)—more than 5 million tonnes above terrain, high sediment loads, territorial disputes, and current production levels. To avoid a supply shortfall, grain considerable social and environmental impacts mean production needs to increase by 16 percent, assuming expanding the portfolio of large hydropower facilities is no changes in foods imports and exports. Production difficult, slow, and expensive (see chapter 3). increases and diversification will require a portfolio of Beyond irrigation supply and hydropower, Pakistan’s options, many centered on water, supported by greatly major dams offer some flood control benefits that improved postharvest food management systems. are economically important. Tarbela Dam has Freshwater aquaculture—mostly for export—plays had an active role in flood control, although dam a minor (less than 1 percent of GDP) role in the management is prioritized heavily toward irrigation national economy (FAO 2016). Freshwater aquaculture and then hydropower, limiting the ability to mitigate production has increased threefold over the last flood peaks. During the 2010 flood, early reservoir 20 years but is sensitive to drought. Increased releases informed by forecasts mitigated damage and production has helped to improve food security and associated economic loss downstream. increase farmer incomes. In 2016, hydropower generation was 35 terawatt Economic Costs hours (IHA 2017); equivalent fossil fuel-based power The economic losses associated with water cost generation would have cost several billion U.S. Pakistan billions of U.S. dollars every year. Conservative dollars. The exact value is difficult to assess given estimates suggest average annual losses of about the opportunity costs of prioritizing water releases 4 percent of GDP (Sadoff et al. 2015), considering for energy generation and the cost differential of inadequate water supply and sanitation, flood damage building hydropower dams instead of thermal power to property, and water scarcity in agriculture. Other plants, which are partly offset by the value of reservoir water-related economic impacts, such as loss of storage to manage grid variability. Hydropower once ecosystem services and the indirect costs of water- underpinned the country’s power sector, accounting related disasters, are additional, suggesting the total for 60 percent of power generation in the late 1970s. economic costs of water insecurity are much higher. This share has dropped to around 30 percent because short-term planning has favored thermal power. More detailed assessments of the economic costs Given volatility in fossil fuel prices and an increasing of water insecurity aspects have been made. These, energy supply-demand gap, low-cost and indigenous however, are based on different methods and energy sources (such as hydropower) are becoming assumptions and cannot be simply aggregated. increasingly important. The proportion of hydropower Inadequate water supply and sanitation services have in the total electricity generation mix could increase to been estimated to cost the equivalent of 3.9 percent more than 40 percent by 2030. Despite considerable of GDP annually (World Bank 2012a). The costs are 19 associated with healthcare, lost work time due to outcomes, including the many foregone nonmarket water-related illnesses, lost work time due to a lack benefits. In addition, the indirect, unquantified of improved water supply and sanitation close to the economic impacts of water-related disasters, home, and premature mortality. These losses are about especially drought, are likely to be significant. At the three times higher than the estimated economic costs household level, these indirect impacts include costs of water scarcity for agriculture, salinity, and flood related to healthcare, lack of economic and labor damage combined, suggesting that inadequate water opportunities, and migration. At the business and services are the biggest water-related drag on the industry level, these indirect costs include reductions Pakistan economy. in inputs and labor productivity, as well as changes in consumption patterns that affect business revenue. Flooding causes direct financial loss because of The value placed on social and environmental infrastructure damage and temporary reductions in outcomes from water typically increases with agricultural and business productivity. The 2010 flood economic development, improvements in living caused losses estimated at US$10.5 billion, or 6 percent standards, and greater education; thus, the perceived of the year’s GDP. Post-flood reconstruction and balance of benefits and costs associated with how recovery can, however, stimulate economic growth, water is used and managed in Pakistan can be and estimates of the average annual economic losses expected to change significantly into the future. associated with floods range from US$800 million (Sadoff et al. 2015) to US$1.8 billion (World Bank 2015), or considerably less than 1 percent of GDP. Social Outcomes While floods cause short-term GDP impact, GDP growth reduces the impact of floods, because stronger growth Human Health and Well-Being allows greater investment in protection infrastructure, Water-borne diseases (cholera, typhoid, hepatitis, and mitigation systems, and response mechanisms (Sardar, diarrhea) are a leading cause of suffering and death Javed, and Amir-ud-Din 2016). in Pakistan and reflect widespread contamination of The economic costs of water scarcity to agriculture are water supplies by sewage effluent. The total health significant—estimated to be at least US$600 million burden from water-related diseases is difficult to annually, considering only the impact on irrigation assess because of a lack of hospital records and limited production and not rainfed production and ignoring reporting; however, the burden is disproportionally the likely significant indirect economic losses (Sadoff borne by poorer children and other vulnerable groups. et al. 2015). The cost of soil salinity to agriculture is An estimated 20 percent to 40 percent of hospital significant; estimates for 2004 alone suggest losses of admissions and a large proportion of infant deaths have US$250 million to US$700 million (World Bank 2006). been linked to water-related diseases (Azizullah et al. Soil salinity is worsening and poses a serious long-term 2011). It is estimated that, on average, 110 children threat to agriculture. die each day in Pakistan because of water-related diseases, poor sanitation, and hygiene (UNICEF 2016), Degradation of the Indus Delta has been estimated to which equates to 39,000 every year. The mortality cost over US$2 billion annually because of foregone rate attributable to poor water supply, sanitation, and ecosystem services. Environmental degradation in hygiene is 20 deaths per 100,000 individuals—well Sindh alone costs an estimated 4 percent to 6 percent above that of the global average of 15 (WHO 2012). of provincial GDP. Around half is agricultural loss caused by waterlogging and salinity, and half is loss of delta Poor water supply, sanitation, and hygiene contribute ecosystem services (including from mangrove forests to childhood stunting. The main determinants of and fisheries) (Sánchez-Triana et al. 2015). The national stunting are food insecurity, inadequate personal costs of water-related environmental degradation are care and feeding, an unhealthy environment, likely to be of a similar magnitude, given the economic and inadequate health care. Poor water services costs of groundwater depletion, land subsidence, influence all of these factors. Despite significant widespread water pollution, and inadequate reduction in poverty across Pakistan, stunting rates environment water allocations for rivers and lakes. remain high at 44 ­ percent nationally and over 50 percent in Balochistan and Federally Administered The economic benefits Pakistan derives from water Tribal Areas (FATA). Each U.S. dollar spent on and water-dependent ecosystems clearly far outweigh specific interventions to reduce stunting nutrition-­ the costs. The benefits are in the order of 10s of in Pakistan generates an estimated US$30 return billions of U.S. dollars, while the economic losses (Hoddinott et al. 2013). Water and sanitation are in the order of a few billions of U.S. dollars investments can strengthen nutritional outcomes by annually. However, this equation depends on the reducing food contamination and diarrhea (Shekar, economic value placed on environmental and social Dayton Eberwein, and Kakietek 2016). 20 PAKISTAN: GETTING MORE FROM WATER Many drinking water supplies across Pakistan are During periods of greater water scarcity, the time spent contaminated by geogenic pollutants and industrial collecting water can rise by as much as 60 percent in effluents. High arsenic concentration in groundwater rural Balochistan and 40 percent in rural Sindh (Hamid is widespread, and highest in Punjab and Sindh and Afzal 2013). Poor sanitation facilities in schools where 50–60 million people are at risk (Podgorski in Pakistan deter children, especially adolescent girls, et al. 2017). Arsenic is primarily geogenic in origin, from education, with up to 50 percent of girls not although anthropogenic sources contribute in some attending school during ­ menstruation (Aslam 2012). areas (Sanjrani et al. 2017). Prolonged exposure Lower school enrollment and retention rates for girls to elevated arsenic concentration in drinking water mean they ­ typically receive fewer years of schooling, can cause skin lesions, cancer, and cardiovascular with consequences for labor force participation and disease (Azizullah et al. 2011; Fatmi et al. 2009). ­ economic production. Although several local assessments have been made, the number of people using arsenic-contaminated Excluding women from water information perpetuates drinking water nationally has not been verified. gender inequality in Pakistan. Early warning systems Heavy metal contamination of drinking water supplies use a language and medium not accessible to women (especially cadmium and chromium) has been and other excluded groups, thus increasing their reported in many areas (e.g., Waseem et al. 2014). vulnerability to water-related disasters (Mustafa Although the health impacts of this contamination et al. 2015). Water–gender relationships influence have not been systematically quantified, they are social outcomes, including the limited presentation known to cause headaches, joint pains, hypertension, of women in formal water management institutions. renal disease, and increased cancer and diabetes Global evidence indicates gender differences risk (Rehman et al. 2017). Effluents from marble, in the perceptions of, and coping strategies, for steel, and aluminum factories are the main sources drought. Women are more proactive in adapting of cadmium, and effluents from leather tanneries are water management strategies to drought even the main source of chromium. Industrial leaching of when excluded from formal water management lead causes lead levels in surface and groundwater arrangements (Su et al. 2017). But women are across Pakistan to consistently exceed World Health not always marginalized in small-scale irrigation, Organization (WHO) guidelines (Ul-Haq et al. 2011; indicating water–gender relationships are complex and Waseem et al. 2014). multifaceted (Das 2017). Women in rural Pakistan are less water secure than Floods are the most frequent and damaging natural men, being commonly responsible for collecting hazard in Pakistan. Over the past 65 years Pakistan water for domestic use, and more vulnerable to has experienced more than 30 major floods affecting related disasters (Parker 2016). Where climate-­ significant fractions of the population (figure 2.7). The public water supply infrastructure is nonexistent or 2010 floods affected 20 million people—about 10 unreliable, women spend 15 percent of their time on percent of the country’s population. From 2010 to 2015, average collecting water (Ilahi and Grimard 2000). Pakistan’s population suffered a major flood at least once Figure 2.7  Share of Population Affected by Riverine Floods in Pakistan, 1973–2016 12 10 8 Percent 6 4 2 0 1973 1978 1983 1988 1993 1998 2003 2008 2013 Source: EM-DAT. 21 year, affecting at least 1 million people annually. From search of work or in response to water-related shocks, 1950 to 2016, around 15,000 fatalities were reported and women from higher socioeconomic groups don’t from riverine floods, with high numbers in the 1950, leave their villages unless it’s a drought year (Sattar 1992, and 2010 floods (Paulikas and Rahman 2013). 2014). Long-term migration because of water stress and climate change has received significant attention in Droughts have significant social impacts, especially the popular press; however, little quantitative evidence for children. During the extended drought in Sindh of exists. Heat stress appears to be a stronger predictor of 2014–17, more than 1,000 children died and 22,000 migration in rural Pakistan than rainfall shocks (Mueller, were hospitalized with drought-related diseases in the Gray, and Kosec 2014), but may partly reflect the Tharparkar District alone (ACAPS 2016). During droughts larger relief efforts made to counteract rainfall shocks. in rural Pakistan, girls are most at risk of malnutrition, Continued deterioration of the Indus Delta has led because they receive less food when resources are to drinking water shortages, seawater intrusion, and stretched given the common preference toward sons increased vulnerability to coastal storms, and these (Mansuri 2006). Short-term migration during droughts appear to influence migration (Sattar 2014). can alleviate resource constraints, but gender gaps in development outcomes are often exacerbated by drought. Environmental Outcomes Conflict and Migration Pakistan’s environment resources and ecosystems are under increasing stress from high levels of Some instances of civil unrest and violence in Pakistan water withdrawal, widespread water pollution, rapid have been linked to water. Protests over water urbanization, and agricultural expansion. Biodiversity shortages can turn deadly, as in Karachi in 2001, loss, declining fish stocks, and degradation of or lead to property damage and violent encounters internationally important ecosystems in the Indus with the police, as in Sindh in 2012 (Mustafa et al. Delta and other parts of the Indus Basin are key 2017). Evidence suggests that disputes over water consequences. allocation have led to deaths and injuries in KP and in FATA (Mustafa et al. 2017), and inequitable access The Indus Basin is home to more than 180 species to municipal water or irrigation water contributes to of freshwater fish, with distributions along a conflicts. In one instance, Perween Rahman, an activist longitudinal gradient from the headwaters to the working to reduce these inequities in Karachi, was delta (Mirza and Mirza 2014). Of these, 86 are of murdered in 2013. Despite the interprovincial Water special concern, 34 are endemic to Pakistan, 11 have Apportionment Accord, interprovincial disputes over special International Union for Conservation of Nature water sharing are common. These disputes have not (IUCN) status, 31 are commercially important, and yet turned violent, but with increasing water demand eight are very rare (Rafique and Khan 2012). Most and more frequent droughts, disputes may escalate. endemic fish species are restricted to mountainous In Pakistan and around the world, insurgent and terrorist and submountainous river reaches, which are now groups use access to water and water infrastructure to highly fragmented by dams and diversion structures pressure civilians or threaten opponents. In Pakistan, the and are characterized by modified flow regimes, such Taliban have threatened to contaminate water sources that the level of ecological alteration could lead to and reservoirs (Roul 2010), fought for control of urban extinction (Regnier, Fontaine, and Bouchet 2009). The water supplies in Karachi (Hamid 2015), and threatened only comprehensively assessed endemic species— to blow up Warsak Dam that supplies Peshawar Glyptothorax ­ kashmirensis—has been declared critically (Mustafa, Akhter, and Nasrallah 2013). endangered (using IUCN criteria). This species inhabits the regulated and fragmented Jhelum River. The IUCN The links between water and migration are complex (2011) predicted an abundance decline of more than because migration choices reflect many economic, 80 percent over five to 10 years, given the species’ political, and demographic issues. Pakistan has preference for fast-flowing habitat. Detailed studies seen short-term, temporary migration and long- are few, but Magurran (2009) suggests that for the term migration. The former is a common response restricted range endemic species, overexploitation, to droughts and floods, especially in Balochistan habitat loss, and degradation of breeding grounds are and Sindh (e.g., Ashraf, Routray, and Saeed 2014). likely to have led to unrecorded extinctions. In Tharparkar District in Sindh, recurrent seasonal migration is exacerbated by drought. During the The 31 commercially important fish species are vital 2014–17 drought, 35 percent to 45 percent of families to rural livelihoods, providing high-quality protein and migrated to barrage areas in search of labor and essential nutrients and minerals that are often difficult grazing for livestock (Alvarez-Quinones 2015). Women to obtain from other food (Rafique and Khan 2012). are less likely than men to migrate individually in The abundance of many of these species is declining, 22 PAKISTAN: GETTING MORE FROM WATER including Tor putitora, which has been declared Eutrophication leads to uncontrolled growth of algae critically endangered because of overfishing, river and depleted oxygen levels in the water, killing fish and fragmentation by water resource infrastructure, and causing a major decline in biodiversity. loss of habitat (including critical breeding grounds). The Indus Delta—the fifth largest delta in the world—is IUCN (2011) notes that Tor putitora abundance had characterized by rich biodiversity and valuable ecosystem declined by more than 50 percent, and that trends services, including productive fisheries and coastal storm suggested declines could reach 80 percent; no protection by mangrove forests. The area is estimated to more recent assessment is available. Several other be around 0.6 million hectares (Amjad, Kasawani, and commercially important species are also declining Kamaruzaman 2007), with mangrove forests originally because of habitat loss and degradation, water covering more than one-third of the total area. However, abstraction, wetland drainage, dam construction, reduced river flows and sediment loads—and sea level pollution, and eutrophication. Distributional ranges of rise—are driving a multifaceted environmental crisis for several species have shrunk greatly since the 1980s to the delta, including sea water intrusion, soil salinization, small, localized remnant populations; many are now mangrove forest loss, reduced freshwater supply, and on the verge of extinction (Rafique and Khan 2012). depleted fisheries. The 17 channels that once delivered In the lower Indus, barrages block fish migration routes, freshwater to the delta have been reduced to one and the fish ladders constructed at Muhammad and (Kidwai et al. 2016), and no freshwater reaches the delta Kotri barrages have proved ineffective. The iconic for 138 days per year on average (Renaud et al. 2013)— Indus dolphin is recognized as one of the world’s most and for much longer periods during drought years. Given endangered mammal species. By the early 1990s its the level of water resource development upstream, range had been reduced by 80 percent because of flows downstream of the Kotri Barrage are now usually major barrages that have fragmented its habitat into limited to August and September, allowing seawater 17 separate reaches (Braulik et al. 2014). In most to penetrate the delta for hundreds of kilometers for of these reaches, dolphins have disappeared within much of the year (Inam et al. 2007). Flow reduction is 50 years of barrage construction; dolphins are now discussed in more detail in chapter 3. found in only six reaches. Worsening water quality is also affecting riverine and lake ecosystems. Sediment delivery to the delta is just 4 percent of predevelopment level. Construction of dams and Pakistan has 19 Ramsar sites—wetlands of international barrages has reduced sediment delivery to the delta importance—covering a total of more than 1.3 million from an estimated 270 million tonnes per year to hectares, and over 225 important perennial or around 13 million tonnes per year (Syvitski et al. ephemeral wetlands. Many of these wetlands 2013). Flow reductions have led salinity in the delta are associated with rivers or are dependent on to increase significantly, leading to a reduction in groundwater and thus influenced by water resource plant diversity: four out of eight plant species that had management. From a water resources management thrived in the delta have disappeared in recent years perspective, the 12 most important of the Ramsar sites (Salik et al. 2015). are (1) the Indus Delta; (2) the Indus Dolphin Reserve— the reach between the Sukkur and Guddu barrages; (3) Degradation of the Indus Delta has affected the lives the Chashma and Taunsa barrages in Punjab; (4) four of at least half a million people. Shrimp production freshwater lakes and one costal lagoon in Sindh; (5) and the catch of the prized Palla fish have fallen by the Tanda Dam and the braided channels of Thanedar 90 percent (Amanullah, Ahmed, and Ali 2014; Renaud Wala in KP, which are important migratory bird et al. 2013). The flow, salinity, and sediment regimes wintering sites; and (6) the Miani Hor coastal lagoon at are the main causes of a drastic reduction in the extent the terminus of the Porali River in Balochistan. of mangrove forests from around 0.24 million hectares to 0.10 million hectares (Renaud et al. 2013). This In addition to the fish fauna discussed previously, contraction of the mangrove forests has had significant Pakistan’s Ramsar sites and other wetlands support 18 impacts on biodiversity, because they are an important threatened mammals, including the endemic Punjab urial wintering habitat for migratory birds on the central (Ovis vignei punjabiensis) and the Indus river dolphin Asian flyway (Khan 2006). The loss of mangrove forests (Plantanista minor), 20 threatened bird species, 12 has also compromised their ability to act as an active reptiles, and two endemic amphibians. Nutrients from barrier against tropical cyclones and storms, leaving the fertilizer in agricultural drainage, untreated municipal delta at greater risk of coastal erosion and flooding. wastewater, and industrial effluent (especially from the The mangrove forests support the livelihoods of more textiles industry) are widespread and polluting freshwater than 100,000 people, with an estimated direct value to ecosystems across Pakistan. Eutrophication is affecting households of US$1,300 per hectare (Adhikari, Baig, several water bodies, including the Manchar Lake in and Iftikhar 2010); further contraction would put these southern Sindh—the largest freshwater lake in Pakistan. livelihoods at risk. 23 The Miani Hor coastal lagoon in Balochistan is another Azizullah, A., M. N. K. Khattak, P. Richter, and productive estuarine fishery that provides important D. P. Hader. 2011. “Water Pollution in Pakistan nursery habitat for the juvenile of several marine fish and Its Impact on Public Health—A Review.” species. Its productivity is dependent on the freshwater Environment International 37: 479–97. flows of the Porali River. The lagoon has habitat and Braulik, G. T., M. Arshad, U. Noureen, and S. P. Northridge. rich faunal diversity with more than 300 finfish species 2014. “Habitat Fragmentation and Species reported (Shah and Jusoff 2007). Reports of fish landing Extirpation in Freshwater Ecosystems: Causes indicate as many as 350 commercially important of Range Decline of the Indus River Dolphin species (Hayat 2003), with several important fish (Platanista Gangetica Minor).” PLoS ONE 9 (7): stocks now considered overfished (Lindley 2008). Water e101657. https://doi.org/10.1371/journal​ resource development in the Porali Basin is, however, .pone.0101657. comparatively limited and is not known to have significantly impacted Miani Hor. Das, M. B. 2017. The Rising Tide: A New Look at Water and Gender. Washington, DC: World Bank. References Davies, S., S. Soofi, A. Turab, S. Zaidi, A. Ejaz, A. S. Tariq, ACAPS. 2016. Global Emergency Review: February 2016 W. Saeed, S. Malik, S. Moeen, and S. Saleem. (database). https://reliefweb.int/sites/reliefweb​ 2018. A Strategic Review of Food Security and .int/files/resources/geo_74.pdf. Nutrition in Pakistan. Islamabad, Pakistan: World Food Programme. Adhikari, B., S. P. Baig, and U. A. Iftikhar. 2010. “The Use and Management of Mangrove Ecosystems in FAO (Food and Agriculture Organization). 2016. Pakistan.” Journal of Environment & Development National Aquaculture Sector Overview: Pakistan. 19 (4): 446–67. Rome, Italy: FAO. Alvarez-Quinones, T. 2015. “Drought in Tharparkar: Fatmi, Z., I. Azam, F. Ahmed, A. Kazi, A. B. Gill, From Seasonal to Forced Migration.” In The M. M. Kadir, M. Ahmed, N. Ara, N. Z. Janjua, State of Environmental Migration 2015 – a and Core Group for Arsenic Mitigation in Pakistan. Review of 2014, edited by F. Gemenne, 2009. “Health Burden of Skin Lesions at Low C. Zickgraf, and D. Lonesco. Le Grand-Saconnex, Arsenic Exposure through Groundwater in Switzerland: IOM. https://www.­ sciencespo​ Pakistan: Is River the Source?” Environmental .fr/psia/sites/sciencespo.fr.psia/files​ Research 109 (5): 575–81. /StateofEnvironmentalMigration2015.pdf. GoP (Government of Pakistan). 2017. Pakistan Amanullah, M., A. Ahmed, and G. Ali. 2014. “Ecological Statistical Yearbook 2016. Statistics Impacts of Altered Environmental Flow on Division, Bureau of Statistics, Islamabad, Indus Deltaic Ecosystem, Pakistan: A Review.” Pakistan. http://www.pbs.gov.pk/content​ Journal of Environmental Professionals Sri Lanka /pakistan-statistical-year-book-2016. 3 (1): 11–21. Hamid, O. 2015. “In 2013, Two Factions of the Amjad, A. S., I. Kasawani, and J. Kamaruzaman. Pakistani Taliban Fought for Control of 2007. “Degradation of Indus Delta Mangroves Karachi’s Water Infrastructure.” CityMetric in Pakistan.” International Journal of Geology (blog), April 27. https://www.citymetric​ 3 (1): 27–34. .com/politics/2013-two-factions-pakistani​ -taliban-fought-control-karachis-water​ Ashraf, M., J. Routray, and M. Saeed. 2014. -infrastructure-978. “Determinants of Farmers’ Choice of Coping and Adaptation Measures to the Drought Hazard Hamid, Y. A. and J. Afzal. 2013. Gender, Water and in Northwest Balochistan, Pakistan.” Natural Climate Change: The Case of Pakistan. Islamabad, Hazards 73: 1451–73. Pakistan: PWP. Aslam, M. M. 2012. Menstrual Hygiene Management Hayat, M. 2003. “Fishing Capacity and Fisheries in for Schoolgirls in Pakistan. Thematic Session: Pakistan.” In Measuring Capacity in Fisheries, MHM during Emergencies. Islamabad, Pakistan: edited by S. Pascoe and D. Gréboval, 121–27. UNICEF Pakistan. Rome, Italy: FAO. Aslam, M. 2016. “Agricultural Productivity Current Hayat, N., A. Hussain, and H. Yousaf. 2016. Scenario, Constraints and Future Prospects in “Food  Demand in Pakistan: Analysis and Pakistan.” Sarhad Journal of Agriculture 32: Projections.” South Asia Economic Journal 289–303. 17 (1): 94–113. 24 PAKISTAN: GETTING MORE FROM WATER Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, Mirza, M. R., and Z. S. Mirza. 2014. “Longitudinal and S. Horton. 2013. “The Economic Rationale for Zonation in the Fish Fauna of the Indus River in Investing in Stunting Reduction.” Maternal and Pakistan.” Biologia 60: 149–52. Child Nutrition 9 (Suppl. 2): 69–82. Mueller, V., C. Gray, and K. Kosec. 2014. “Heat Stress Hussain, A., and J. Routray. 2012. “Status and Factors of Increases Long-Term Human Migration in Rural Food Security in Pakistan.” International Journal Pakistan.” Nature Climate Change 4 (3): 182–5. of Development Issues 11 (2): 164–85. Mustafa, D., M. Akhter, and N. Nasrallah. 2013. IHA (International Hydropower Association). 2017. IHA: Understanding Pakistan’s Water-Security Nexus. Pakistan (database). https://www.hydropower​ Washington, DC: USIP. .org/country-profiles/pakistan. Mustafa, D., G. Gioli, S. Qazi, R. Waraich, A. Rehman, Ilahi, N., and F. Grimard. 2000. “Public Infrastructure and R. Zahoor. 2015. “Gendering Flood Early and Private Costs: Water Supply and Time Warning Systems: The Case of Pakistan.” Allocation of Women in Rural Pakistan.” Economic Environmental Hazards 14 (4): 312–28. Development and Cultural Change 49 (1): 45–75. Mustafa, D., G. Gioli, M. Karner, and I. Khan. 2017. Inam, A., P. Clift, L. Giosan, A. R. Tabrez, M. Tahir, Contested Waters: Subnational Scale Water M. M. Rabbani, and M. Danish. 2007. Conflict in Pakistan. Washington, DC: USIP. “The Geographic, Geological and Oceanographic Parker, H. 2016. Three Ways to Build Water Security Setting of the Indus River.” In Large Rivers: for Rural Women. London, U.K.: ODI. https:// Geomorphology and Management, edited by A. www.odi.org/comment/10367-three-ways-build​ Gupta, 333–46. New York: John Wiley & Sons. -water-security-rural-women-resilience. IUCN (International Union for Conservation of Nature). Paulikas, M. J., and M. K. Rajman. 2013. “A Temporal 2011. IUCN Red List of Threatened Species Assessment of Flooding Fatalities in Pakistan database). http://www.iucnredlist.org. (­ (1950–2012).” Journal of Flood Risk Management Khan, Z. M. 2006. “Current Status and Biodiversity 8: 62–70. of Indus Dolphin Reserve and Indus Delta Podgorski, J. E., S. A. M. A. S. Eqani, T. Khanam, R. Ullah, Wetlands (Ramsar Sites).” Paper presented at H. Shen, and M. Berg. 2017. “Extensive Arsenic “9th International River Symposium,” Brisbane, Contamination in High-pH Unconfined Aquifers Australia. 17pp. in the Indus Valley.” Science Advances 3 (8): Kidwai, S., P. Fanning, W. Ahmed, M. Tabrez, J. Zhang, e1700935. and M. W. Khan. 2016. “Practicality of Marine Rafique, M., and N. U. H. Khan. 2012. “Distribution Protected Areas—Can There Be Solutions for the and Status of Significant Freshwater Fishes of River Indus Delta?” Estuarine, Coastal and Shelf Pakistan.” Records of the Zoological Survey of Science 183: 349–59. Pakistan 21: 90–5. Kirby, M., M.-D. Ahmad, M. Mainuddin, T. Khaliq, and Regnier, C., B. Fontaine, and P. Bouchet. 2009. “Not M. J. M. Cheema. 2017. “Agricultural Production, Knowing, Not Recording, Not Listing: Numerous Water Use and Food Availability in Pakistan: Unnoticed Mollusk Extinctions.” Conservation Historical Trends, and Projections to 2050.” Biology 23: 1214–21. Agricultural Water Management 179: 34–46. Rehman, K., F. Fatima, I. Waheed, and M. S. H. Akash. Lindley, R. H. 2008. Balochistan Fisheries Development 2017. “Prevalence of Exposure of Heavy Metals Study: Options for Balochistan Coastal Fisheries & and Their Impact on Health Consequences.” Aquaculture. Competitiveness Support Fund. 59pp. Journal of Cellular Biochemistry 119: 157–84. Linstead, C., A. H. Sayed, and S. A. Naqvi. 2015. Water Renaud, F. G., J. P. M. Syvitski, Z. Sebesvari, S. E. Footprint of Key Industrial Sectors in Punjab, Werners, H. Kremer, C. Keunzer, R. Ramesh, Pakistan. Lahore, Pakistan: WWF-Pakistan. A. Jeuken, and J. Friedrich. 2013. “Tipping Magurran, A. E. 2009. “Threats to Freshwater Fish.” from the Holocene to the Anthropocene: How Science 325: 1215–16. Threatened Are Major World Deltas?” Current Opinion in Environmental Sustainability 5 (6): Mansuri, G. 2006. “Migration, Sex Bias, and Child 644–54. Growth in Rural Pakistan.” World Bank Policy Research Working Paper 3946, World Bank, Roul, A. 2010. “Taliban and Weapons of Mass Disruption Washington, DC. Threat.” CBW Magazine 3. 25 Sadoff, C. W., J. W. Hall, D. Grey, J. C. J. H. Aerts, Su, Y., S. Bisht, A. Wilkes, N. S. Pradhan, Y. Zou, S. Liu, M. Ait-Kadi, C. Brown, A. Cox, S. Dadson, D. and K. Hyde. 2017. “Gendered Responses to Garrick, J. Kelman, P. McCornick, C. Ringler, M. Drought in Yunnan Province, China.” Mountain Rosegrant, D. Whittington, and D. Wiberg. 2015. Research and Development 37 (1): 24–34. Securing Water, Sustaining Growth: Report of Syvitski, J. P. M., A. J. Kettner, I. Overeem, L. Giosan, the GWP/OECD Task Force on Water Security and G. R. Brakenridge, M. Hannon, and R. Bilham. 2013. Sustainable Growth. Oxford, U.K.: University of “Anthropocene Metamorphosis of the Indus Delta Oxford. and Lower Floodplain.” Anthropocene 3: 24–35. Salik, K. M., S. Jahangir, W. Z. Zahdi, and S. Hasson. 2015. “Climate Change Vulnerability and Ul-Haq, N., M. A. Arain, N. Badar, M. Rasheed, and Adaptation Options for the Coastal Communities Z. Haque. 2011. “Drinking Water: A Major Source of Pakistan.” Ocean & Coastal Management of Lead Exposure in Karachi, Pakistan.” Eastern 112: 61–73. Mediterranean Health Journal 17 (11): 882–86. Sánchez-Triana, E., S. Enriquez, B. Larsen, P. Webster, UNICEF (United Nations Children’s Fund). 2016. Pakistan and J. Afzal. 2015. Sustainability and Poverty Annual Report. Islamabad, Pakistan: UNICEF. Alleviation: Confronting Environmental Threats USDA (U.S. Department of Agriculture) Foreign in Sindh, Pakistan. Directions in Development Agricultural Service. 2017. Pakistan: Grain and Series. Washington, DC: World Bank. Feed Annual. Islamabad, Pakistan: USDA Foreign Sanjrani, M. A., T. Mek, N. D. Sanjrani, S. J. Leghari, Agricultural Service. H. T. Moryani, and A. B. Shabnam. 2017. “Current Waseem, A, J. Arshad, F. Iqbal, A. Sajjad, Z. Mehmood, Situation of Aqueous Arsenic Contamination in and G. Murtaza. 2014. “Pollution Status of Pakistan, Focused on Sindh and Punjab Province, Pakistan: A Retrospective Review on Heavy Metal Pakistan: A Review.” Journal of Pollution Effects & Contamination of Water, Soil, and Vegetables.” Control 5: 207. BioMed Research International. http://dx.doi​ Sardar, A. S., A. Javed, and R. Amir-ud-Din. 2016. .org/10.1155/2014/813206. “Natural Disasters and Economic Growth in WFP (World Food Programme). 2014. Pakistan: Food Pakistan: An Enquiry into the Floods Related Security Bulletins, December 2014. Rome, Hazards’ Triad.” Working Paper 10, Pakistan Italy: WFP. https://www.wfp.org/content​ Institute of Development Economics, Islamabad, /pakistan-food-security-bulletins-2014. Pakistan. WHO (World Health Organization). 2012. “Mortality and Sattar, A. 2014. Climate Change and Migration: Burden of Disease from Water and Sanitation.” Exploring the Linkage and What Needs to Be Global Health Observatory (data). http://www​ Done in the Context of Pakistan. Series on .who.int/gho/phe/water_sanitation/burden/en/. Vulnerability and Resilience. Islamabad, Pakistan: LEAD Pakistan. World Bank. 2006. Pakistan: Strategic Country Environmental Assessment: Main Report. Shah, A., and A. K. Jusoff. 2007. “Mangrove Washington, DC: World Bank. Conservation through Community Participation in Pakistan: The Case of Sonmiani Bay.” Paper pre- ———. 2012a. “The Economic Impacts of Inadequate sented at “5th WSEAS International Conference Sanitation in Pakistan.” Water & Sanitation on Environment, Ecosystems and Development,” Discussion Paper 77839, World Bank, Tenerife, Spain, December 14–16. Washington, DC. Shekar, M., J. Dayton Eberwein, and J. Kakietek. 2016. ———. 2012b. “Fiscal Disaster Risk Assessment: “The Costs of Stunting in South Asia and the Options for Consideration. Pakistan.” World Bank Benefits of Public Investments in Nutrition.” and Global Facility for Disaster Risk Reduction, Maternal & Child Nutrition 12: 186–95. Washington, DC. C HAPT E R 3 Pakistan’s Water Endowment Key Messages • Most of Pakistan’s water is associated with the Indus Basin and flows into Pakistan from outside the country. The water generated within the country is an important fraction of the total water resource but is often overlooked and is less well measured. • Average inflows to Pakistan have remained reasonably stable over many decades, although development in India permitted under the Indus Waters Treaty has recently reduced the minor inflows to Pakistan from the eastern Indus tributaries. • Interannual variations in river flow are low compared to other large, arid zone rivers, because of comparably reliable glacier and snowmelt. Flows are strongly seasonal, reflecting the annual cycles of meltwater and monsoonal precipitation. • While Balochistan has access to some Indus water, the province is mostly outside the basin and is largely reliant on highly variable and often flashy rainfall to recharge sedimentary aquifers and supply arid zone rivers, many of which are ephemeral. • Groundwater is an important resource for Pakistan and in the Indus is tightly coupled to surface water. Groundwater pumping is very significant and is largely sustained by leakage from the surface water distribution system. Groundwater depletion is, however, a significant issue at some locations in Punjab and Balochistan. • Pakistan’s irrigation command areas are characterized by high levels of internal water recycling and considerable evaporative losses. The Indus Basin also loses a considerable fraction of its water naturally because of the hot, arid nature of the lower basin. These losses are likely to increase as the climate warms. • Given the complex, variable, and changing climate, and the history of transboundary agreements and development upstream of Pakistan, detailed river system modeling, analysis of Earth observation data, and comprehensive water accounting are required. This would improve water resources assessments and enhance the understanding of natural and induced water losses to guide irrigation efficiency investments, conjunctive water management, and drought planning for climate resilience. 28 PAKISTAN: GETTING MORE FROM WATER T his chapter reviews the extent, variability, and People quality of Pakistan’s water resources, considering ices Mitig surface water and groundwater and their erv ati eds on interactions. Pakistan’s total water endowment is t of ru ela rast ctur wa poorly quantified because of limited data and a lack nf r-r ter- ivery of wate e of robust water resource assessments. The common I related risks focus has been on the main river inflows to the Water endowment Indus Basin irrigation system (at the so-called “rim Envir Del stations”), which ignores all internally generated ns y titution om I s runoff and groundwater recharge, including outside o M s nm na n ce ur a of the Indus Basin in Balochistan. This chapter ge co me so nt of w ater re e E nt provides a detailed picture of the water resource based on multiple data sources and prior water accounting efforts. Average Water Balances direct rainfall recharge to groundwater. The major groundwater resources of Pakistan are the shallow Pakistan is comprised of three surface water alluvial aquifers of the Indus River plain, which hydrologic units: (i) the Indus Basin, (ii) the Makran are highly connected to the river. Under natural Coast, and (iii) the Kharan Desert. The Indus Basin conditions the groundwater recharges from the covers 65 percent of Pakistan and represents over river during high flow seasons, and groundwater 95 percent of its water resources (table 3.1). It discharges back to the river as baseflow during includes the mountainous areas of the north and the low flow seasons. Under contemporary conditions west, the Indus Plain, the Kacchi Plain, the desert groundwater recharge is dominated by leakage and areas of Bahawalpur and Sindh, and the Rann of drainage from the surface irrigation distribution Kutch. Balochistan is the only province not fully system. Given the internal recycling of water (from within the Indus Basin. Of the 18 river basins of canals to groundwater) and the limited extent Balochistan, seven are part of the Indus basin; the of groundwater monitoring and modeling, these Nari River terminates in Hamal Lake in Sindh—its aspects of the resource assessment are uncertain. waters never reaching the Indus; and the other Groundwater is an important water source, however, six rivers contribute small volumes to the Indus. even if most is ultimately sourced from surface There are seven rivers in the Makran Coast basin water withdrawals. of Balochistan (18 percent of the area of Pakistan) and four rivers in the endorheic Kharan Desert Average annual water balances for each of Pakistan’s (17 percent of Pakistan) that flow either into the three hydrologic units highlight the high natural water Islamic Republic of Iran or Afghanistan. These latter losses in these arid and semi-arid landscapes, and rivers are low-volume, intermittent rivers but of the high induced losses—nonbeneficial evaporation— considerable value to the sparse populations that associated with irrigation (table 3.2). In the Indus Basin live in these basins. Although the summary resource the high level of withdrawals means that the average assessment (table 3.1) suggests groundwater is of basin outflow is low, currently averaging 16 percent of minor importance, this portrayal reflects only the the total system resource. Less than one-third of the total resource goes to beneficial consumptive use. In the Indus Basin and the Kharan Desert the sum of use and losses slightly exceeds total inflows and internal Table 3.1  Average Annual Available Water contributions, indicating groundwater depletion. Resources of Pakistan billion cubic meters The Indus Basin dominates Pakistan’s water resources, with a high level of water recycling in Indus Basin Surface water Groundwater Total irrigation. Groundwater withdrawals are largely Indus Basin 205.7 12.7 218.4 wsupported by leakage from irrigation canals and Makran Coast 6.2 0.7 6.9 distributaries and irrigation drainage (figure 3.1). Two simplifications are made in figure 3.1, panels a–c: Kharan Desert 2.9 0.6 3.5 (i) canal leakage is shown separately, but watercourse Total 214.8 14.0 228.8 leakage and field-level drainage to groundwater are Sources: FAO 2011; Laghari, Vanham, and Rauch 2012; van combined; and (ii) irrigation returns to surface water Steenbergen, Basharat, and Lashari 2015. are shown as all returning to the river, while in reality 29 Table 3.2  Average Annual Water Balance for Pakistan’s Three Hydrologic Units billion cubic meters Indus Basin Makran Coastal Basin Kharan Desert Total Percent Inflows 174 0 0 174 76 Internal contributions 45 6.2 3.5 55 24 Beneficial consumption 80 1.2 0.7 82 36 Natural losses 68 2.0 1.2 68 31 Induced losses 41 0.7 0.5 39 18 Outflows 30 2.3 1.2 41 15 Sources: FAO 2011; Karimi et al. 2013; Laghari, Vanham, and Rauch 2012; van Steenbergen, Basharat, and Lashari 2015; WAPDA unpublished data. a fraction is saline drainage to the sea, to saline lakes, Natural land uses, including the delta mangrove forests, or to evaporation basins. The split between irrigation provide important ecosystem services that should be recharge to fresh versus saline groundwater is an protected. Water accounting for a single year for the estimate based on the relative proportions of area entire Indus Basin (Karimi et al. 2013) indicates that covered by fresh and saline shallow groundwater irrigated crops represent 69 percent of transpiration and (approximated as a one-third to two-thirds split). In 39 percent of the evaporative loss. The largest share appendix A, three separate balances are tabulated for of the evaporative loss (44 percent) is associated with each of these hydrologic units: (i) a river water balance, natural land uses (Karimi et al. 2013). It is important (ii) a groundwater balance, and (iii) a withdrawal to better understand these natural losses, especially balance. The withdrawal balances are dominated by because they are strongly temperature-driven and so irrigation but include small volumes of nonirrigation will increase with climate warming. Minimizing these withdrawals. losses should not be a management priority and would in any case be very difficult. The groundwater balances for the Indus Basin and the Makran Desert (figure 3.2, panels c and b, respectively, and tables A.2 and A.3) include groundwater depletion. River System Gains and Losses Groundwater depletion is discussed in chapter 5, but Managing the Indus River system requires depletion is the smallest term in these groundwater understanding the water losses from and gains to the balances. The outflow value in figure 3.1a is the river. Losses include evaporation and flow from the average gauged outflow from Kotri Barrage and so river to groundwater. Unmeasured water withdrawals includes water used downstream, including Karachi that contribute to flow differences between gauging supply. The outflow value is the average for 1975– points are sometimes treated as losses, too. Gains 2015; over the last 15 years of this record outflows include unmeasured contributions from minor have been about 15 billion cubic meters or half of this tributaries and drains, direct runoff to the river, and average. water movement from groundwater to the river. The high natural losses of water across the Indus Losses and gains vary considerably across the basin and are inferred in the mass balance and are indicated through time, both within and between years. During by basin-level landscape water accounting based kharif, high flows typically recharge groundwater, on remotely sensed data (Bastiaanssen, Ahmad, and the river is losing water overall. During rabi, as and Chemin 2002; Karimi et al. 2013). Natural river flows start to recede, the river gains typically losses include evaporation and nonagricultural plant water from landscape and aquifer storage. The gains transpiration. From a water resources management in September and early October are important for perspective, transpiration from irrigated and rainfed maturing kharif crops, while the gains from mid- crops and from pasture is considered beneficial, while October to March support rabi crops (Euroconsult 2011). transpiration from forests and savannah is considered nonbeneficial. Bastiaanssen, Ahmad, and Chemin The pattern of losses and gains varies strongly across (2002) show the mangrove forest of the Indus Delta the basin. Between 1940 and 1994 there was a have highest levels plant water use in the basin, at significant net loss in the Indus mainstem (11 billion around 1.3 meters of evapotranspiration annually. cubic meters per year on average) and a small net gain 30 PAKISTAN: GETTING MORE FROM WATER Figure 3.1  Average Annual Water Balance for the Three Hydrologic Units of Pakistan a. Indus Basin IBRD 44138 | DECEMBER 2018 CCH HIN NAA Indus, Kabul, Jhelum, Chenab A A F FG GHH A A N II SST TA A N N inflows a 170 P A K I S T A N Ravi, Sutlej, b IIN N D II AA Beas inflows I.R. OF I.R. OF 3 IRAN IRAN Hydrologic units Runoff inside c a Indus river basin Pakistan Closed basin of Kharan desert b c Makran coastal basins 32 Arab i an S e a Evaporation loss 41 Rainfall recharge 13 Crop water use and other Surface water beneficial consumption withdrawals 80 125 Canal Surface water flows leakage Irrigation recharge Basin inflows and outflows Groundwater recharge Saline Fresh withdrawals 27 Flow from surface water to 6 aquifer aquifer 62 ground water Water withdrawals Surface water Irrigation recharge 11 River and flood recharge 4 Groundwater Non-beneficial consumption Loss to the atmosphere Irrigation return flow 22 Loss to saline groundwater Return flows Outflows To surface water to delta Natural losses 68 30 To ground water b. Kharan Desert c. Makran Coastal Drainage Surface Surface water water inflows Natural Natural inflows 5.5 losses losses 2.9 2.0 1.2 Evaporation loss Evaporation loss 0.69 0.5 Crop water use and Crop transpiration and beneficial consumption Outflows Rainfall Surface water beneficial consumptive 1.2 1.2 Surface water recharge withdrawals 0.68 withdrawals 0.56 1.2 Environmental 0.5 use/loss Rainfall Outflows recharge Irrigation 0.05 Groundwater to the Groundwater 0.74 withdrawals recharge Arabian withdrawals 0.8 0.24 Sea 0.69 2.3 Sources: Ahmad and Rashida 2001; FAO 2011; Halcrow Group 2007; Karimi et al. 2013; Laghari, Vanham, and Rauch 2012; MacDonald et al. 2016; van Steenbergen and Gohar 2005; WAPDA unpublished data. Details in appendix A, tables A.1, A.2, and A.3. Note: Flows are in billion cubic meters. 31 Figure 3.2  Annual River Losses and Gains in Key Losing Reaches of Indus River, Pakistan, 1940–94 20 10 Cubic meters (billions) 0 –10 –20 –30 –40 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 Attock-Kalabagh Sukkur-Kotri Source: WAPDA unpublished data. in the Jhelum-Chenab zone (0.4 billion cubic meters on Punjab. These losses may be supporting increasing average). In the Indus zone, net losses are high in the groundwater use in the Thal Doab of Punjab, especially Attock-Kalabagh reach downstream of Tarbela Dam, the area of significant depletion (approximately 15 and even higher in the Sukkur-Kotri reach. These high meters) identified by Khan et al. (2016) in the Mianwali losses are partially offset by moderate gains in the District. The losses below Sukkur Barrage have also Kalabagh-Taunsa and Taunsa-Guddu reaches. A fraction been increasing, possibly reflecting the significant rate of the losses between Sukkur-Kotri may be to saline of climate warming in the lower basin (figure 3.2). groundwater; however, the high losses in this zone During the 1980 to 1990s, losses in the reach below keep the groundwater fresh near the river. Additionally, Tarbela averaged around 14 billion cubic meters and the losses to groundwater in the Sukkur-Kotri reach losses in the Sukkur-Kotri reach averaged around 12 support the high levels of natural evapotranspiration in billion cubic meters. These are very significant volumes the hot, dry lower parts of the basin. The losses in the of water—equivalent in aggregate to twice the national reach below Tarbela are likely to be a major fraction of municipal and industrial withdrawals. River losses and the groundwater recharge at least into the Thal Doab gains remain poorly accounted for in water allocation between the Indus and Jhelum rivers. In the Jhelum- and delivery operations, and the physical processes Chenab zone, while there is a small net gain overall, driving them are poorly understood. A detailed study of there are significant net losses along the Jhelum in basin-scale surface-groundwater interactions is required the Rasul-Trimmu and Trimmu-Panjnad reaches. These to inform improved resource assessment, planning, and losses are likely to important groundwater recharge operations. pathways for the lower Chaj, Rechna, and Dari doabs. They are partly offset by net gains to the Chenab and The water balance diagrams (figure 3.1, panels a–c) Ravi rivers. This suggests some lateral groundwater highlight the internal recycling of water in the irrigation flow and significant inflows from the hill torrents in system, especially for the Indus Basin. Water leaks these areas. from the canals and distributaries into the groundwater, and excess water applied to the fields flows to Between 1940 and 1994, river losses and gains varied drains and thence to the river, or seeps to underlying significantly, indicating complex surface-groundwater aquifers. Although some data exist to describe these dynamics that are poorly understood (figure 3.2). exchanges, full quantification would require both The increasing losses in the Indus zone, especially better measurement and detailed hydrologic modeling. in the more recent period, have previously been Leakage and drainage to fresh groundwater supports largely attributed to the presence and operation groundwater pumping, but leakage and drainage to of Tarbela Dam and the lower barrages (e.g., Khan saline groundwater (as is the case across much of 1999). However, these increasing losses also mirror Sindh) is nonrecoverable for irrigation use. Desalination the increase in tube wells, especially in Punjab, and of saline groundwater could potentially augment urban the significant depletion of groundwater in parts of supply. 32 PAKISTAN: GETTING MORE FROM WATER Provincial Water The relative level of resource use—often called “water stress”—indicates the level of stress on the water resource Availability and Use and on water-dependent ecosystems, not the level of Current water availability varies between the provinces water stress experienced by communities or economic because of differences in the natural hydrology and activities. Water availability per capita, while broadly extent of the provinces, and because of the water indicative of resource availability, is not a meaningful sharing arrangements enshrined in the 1991 Water indicator of water security overall. To illustrate: Balochistan Apportionment Accord (table 3.3). The Accord sharing and Sindh have the highest water availability per capita largely reflects historical patterns of use. The Accord (table 3.4) yet are the least water secure provinces of is discussed in more detail in chapters 4 and 5. The Pakistan. Although Khyber Pakhtunkhwa (KP) has the groundwater resource is solely direct rainfall recharge; lowest level of water availability per capita and the lowest river outflows are not allocated to any province but levels of water use, rainfall is much higher in KP, meaning are reflected in the total resource estimate (table 3.3). the reliance on streamflow and groundwater is lower. Per The distribution across the provinces of internally capita water use is highest in Sindh, probably because of generated water (runoff and recharge) is uncertain few opportunities to recover the irrigation leakage and but is estimated according to climatic and hydrologic drainage water from (saline) groundwater. conditions. The Accord-apportioned volume represents 61 percent Temporal Patterns of the total assessed resource. Punjab has access to Water availability varies through time, largely driven over 41 percent of the national resource (excluding by the temporal patterns in inflows. For users outflows) and has 53 percent of the national (including the environment) lower in the Indus Basin, population. Water withdrawals in Punjab are 63 percent flow variability also reflects the temporal patterns in of the national total, meaning the level of resource use withdrawals, consumption, and losses. These vary is highest here, with withdrawals exceeding availability seasonally, especially given seasonal temperature by 20 percent—reflecting unsustainable groundwater cycles that drive water demand. The temporal pattern abstraction and hence depletion (table 3.4). in Indus Basin inflows reflects the dominant inflow Table 3.3  Average Annual Provincial Water Resource Availability in Pakistan billion cubic meters Accord apportioned surface Internally generated Renewable fresh Total renewable resource water runoff groundwater Khyber Pakhtunkhwa 10.83 11 2 24 Punjab 69.00 19 9 97 Sindh 60.14 3 2 65 Balochistan 4.77 8 1 14 Pakistan 144.75 41 14 229 Source: Water Apportionment Accord and author calculations. Table 3.4  Provincial Withdrawals, Level of Use, and per Capita Availability and Use in Pakistan billion cubic meters (total withdrawals); cubic meters (per capita values) Total water Relative level of Water availability per Water withdrawal per withdrawalsa resource usea capita capita Khyber Pakhtunkhwa 7 0.29 781 230 Punjab 118 1.21 882 1,069 Sindh 55 0.85 1,360 1,158 Balochistan 4 0.29 1,120 325 Pakistan 184 0.80 1,102 885 Source: GoP 2017 and authors’ calculations. a. Not adjusted for doubled counting of surface water and groundwater withdrawals. 33 Map 3.1  Oblique Aerial View of the Upper Indus Basin 0 12.525 50 75 100 E N S Kilometers W Indus Basin Jhelum Basin Chenab Basin Rivers Glaciers Source: William Doan, USACE, personal communication. sources in the major tributaries. Map 3.1 highlights the released during rabi. The seasonal pattern of canal topographic differences between the main headwater withdrawals reveals the temporal mismatch between catchments of the western rivers, which strongly supply and demand (figure 3.3). The real mismatch determine their respective hydrological signatures. is starker, because rabi canal diversions are an underestimate of total demand and partially reflect the The lower altitude Sutlej, Beas, and Ravi rivers are supply constraint because of limited storage. Adequacy dominated by monsoon rainfall runoff, peaking in of water storage is discussed further in chapter 4. August and September (Lutz et al. 2016). For the Indus mainstem, and hence of Tarbela Dam inflows, glacier In Balochistan, precipitation has a less marked meltwater dominates, peaking from July to September. seasonal pattern; however, summers are very hot For the Jhelum, and thus for Mangal Dam inflows, and dry, and potential evapotranspiration far exceeds snowmelt dominates, peaking earlier from May to rainfall. High natural losses mean many of the rivers July. Combining these tributary signals illustrates a are intermittent. Some are ephemeral and only flow basin inflow pattern characterized by a wetter summer following intense and localized storms, meaning (karif) season (April to October) and a drier winter interannual flow variability is high. (rabi) season (October to April) (figure 3.3). Average Despite a strong seasonal flow pattern, flow karif inflows are four to five times that of average rabi variability in the Indus between years is low inflows, and hence most inflows occur in the three- compared to other major rivers in semi-arid regions. month period from June to August. This is largely because the headwater glaciers Kharif inflows far exceed irrigation demands; however, (and to a lesser degree the multiyear snowpack) rabi inflows are inadequate to meet winter irrigation ensure more stable inflows, even with considerable demand. Surplus karif inflows are therefore stored and variation in annual precipitation. Over the last 34 PAKISTAN: GETTING MORE FROM WATER Figure 3.3  Average Annual Pattern of Average 10-Day Inflows and Canal Withdrawals, Pakistan 15 14 13 Flow volume (billion cubic meters) 12 11 10 9 8 7 6 5 4 3 2 1 0 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Canal withdrawals System inflows Source: WAPDA unpublished data. Figure 3.4  Annual Indus Basin Inflows, Canal Withdrawals, and Outflows, Pakistan, 1975–2015 250 200 Billion cubic meters 150 100 50 0 1975 1980 1985 1990 1995 2000 2005 2010 2015 Outflows Withdrawals Inflows Source: WAPDA unpublished data. 40 years, annual Indus system inflows have varied is the significant reduction in the (albeit relatively from around 125 billion cubic meters to 220 billion small) inflows of the eastern tributaries (figure 3.5). cubic meters (figure 3.4). While inter-seasonal Following the 1960 Indus Water agreement, which storage is important for irrigation in Pakistan, allocated the waters of the eastern tributaries to interannual storage is seldom required. Only during India, permitted development in India on the Ravi the driest years on record have inflows constrained and Sutlej has resulted in progressing reduced annual canal withdrawals. inflows to Pakistan from these tributaries. This change accounts for less than one-quarter of the There is no statistically significant trend (at P = 0.05 observed reductions in inflows between the 16-year significance level) in total Indus Basin inflows over periods before and after 2000. the 55-year record from 1960 to 2015. However, the average annual inflow for the 16 years since 2000 There is no clear evidence of a contemporary flow (161.5 billion cubic meters) is significantly lower reduction related to climate change. A detailed statistical (by student t test) than the average annual inflow analysis of long-term trends in flows in the Upper Indus for the 16 years before 2000 (193.2 billion cubic Basin reveals falling trends in high-elevation glacial meters). One factor contributing to this reduction subcatchments balanced out by increasing trends in 35 Figure 3.5  Annual Inflows to Pakistan from Ravi and Sutlej Rivers, 1960–2015 30 Bhakra Dam 25 (9.3 BCM) Billion cubic meters 20 Ranjit Sagar Dam (3.3 BCM) 15 10 Nathpa Jhakri Dam (3.4 BCM) 5 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Ravi Sutlej Source: WAPDA unpublished data. Note: Figure shows commencement dates and storage volumes for three major upstream reservoirs. other subcatchments (Sharif et al. 2013). Reggiani and a steady rise in groundwater levels across Punjab and Rientjes (2014) find no statistically significant trends in Sindh, leading to widespread water logging, especially a 50-year record of combined Upper Basin output or in Sindh. Over the last few decades, rapid expansion of in 100-year records of Upper Basin precipitation. Rao groundwater pumping in Punjab has led to significant et al. (2018) provide a paleohydrologic reconstruction declines in water tables, almost to predevelopment of six centuries of flow at a number of Upper Indus levels (figure 3.6). Basin locations that shows greater flow variability Farther down the Indus Basin, the groundwater typology than captured in the instrumental record, but again is influenced by natural variations in alluvial sediments no long-term trend. They conclude that the observed and climate, and by the pattern of irrigation canals and higher flows of the 1980s and 1990s are unusual in the distributaries. The changing depositional environment context of the past six centuries. They also note that that led to the accumulation of the sediments of the sensitivity of streamflow to summer temperatures the Indus Plain has contributed to a heterogeneous suggests that expected future warming may increase sequence of sediment distribution, both vertically and basin inflow over coming decades, but any longer-term laterally. These sediment characteristics directly impact change will depend on long-term changes in snowfall the ability of the alluvium to retain, transmit, release, and glacial mass balance. and store water, but in the absence of detailed surveys all groundwater assessments are approximate. In all Groundwater but the marine-influenced deltaic areas, the irrigation Given the seasonal variability of streamflow across schemes are the dominant contributors to contemporary Pakistan, and strong interannual variability of groundwater recharge and exert a strong control on streamflow in Balochistan, groundwater is a critical modern groundwater levels. water resource. An extensive, unconfined aquifer of Renewable groundwater is defined by the recharge. unconsolidated alluvial sediment, covering 16 million However, with multiple recharge pathways—direct hectares, lies under the Indus Basin (Qureshi et al. rainfall recharge, river recharge, flood recharge, canal 2008). In Balochistan, the more limited but nonetheless leakage, and irrigation drainage (none of which important groundwater resources are found within older are directly measured)—assessing the renewable consolidated sedimentary landscapes. Groundwater is groundwater resource is difficult. The average annual both accessible and plentiful across much of Pakistan. renewable groundwater resource has been estimated It is important for urban water supply, particularly for as 63 billion cubic meters (Briscoe and Qamar 2006). Lahore and Quetta, and for irrigation supply, especially However, around three-quarters of this are sourced in Punjab. However, this was not always the case. Easily from surface irrigation leakage, and so cannot be accessible groundwater in the Indus Basin is largely simply added to the surface water resource. a result of decades of seepage from surface water irrigation, which caused groundwater levels to rise in FAO (2011) estimates an internally generated resource many areas by 10–15 meters over half a century. Thus, for Pakistan of 55 billion cubic meters, of which nearly in the early decades of irrigation, overirrigation led to 48 billion cubic meters is an overlap between surface 36 PAKISTAN: GETTING MORE FROM WATER Figure 3.6  Groundwater Levels at Khanewal and Sahiwal Divisions, Punjab, 1910–2010 0 IBRD 44139 | DECEMBER 2018 lum AFGHANISTAN Jhe ena b Ch Lahore b Zho Groundwater level (m bgl) vi Ra Khanewal Division Sahiwal Division Quetta j tle Su s u Ind 10 INDIA WATER-TABLE TREND 2000–12 (m/year) >+0.05 +0.05 – –0.05 –0.05 – –0.25 –0.25 – –0.75 <–0.75 Karachi PROVINCE CAPITALS 20 INTERNATIONAL BOUNDARIES 1900 1920 1940 1960 1980 2000 2020 us Ind Sahiwal Division Khanewal Division Source: Alan McDonald, British Geological Survey, personal communication. water and groundwater resources. In other words, this recharge, and severe groundwater depletion has volume is both river flow that recharge groundwater occurred in several parts of the province. and groundwater that discharges to rivers as baseflow, implying a mere 7 billion cubic meters of direct FoDP (2012) provides a provincewide groundwater rainfall recharge to groundwater. Van Steenbergen and balance indicating a combined average annual Gohar (2005) estimate rainfall recharge to be around groundwater recharge of 74 billion cubic meters 14 billion cubic meters: this value is used in the water (table 3.5). This estimate is higher than the FAO (2011) accounts and resource estimates in this report. Prior to and the Briscoe and Qamar (2005) estimates, and higher groundwater development, natural recharge would have than the total groundwater flux included in figure 3.1a–c. been balanced by discharge, including evapotranspiration The FoDP groundwater balance indicates 66 percent of of groundwater from floodplain wetlands and total recharge is from irrigation recharge and 18 percent groundwater discharge to rivers and the sea. from rainfall—similar percentage values to those in van Steenbergen and Gohar (2005). The provincewide Van Steenbergen and Gohar (2005) estimate that balances show recharge is dominated by irrigation under contemporary conditions groundwater recharge seepage, especially in Punjab and Sindh. In Punjab, comprises direct rainfall recharge (21 percent), canal discharge is nearly all abstraction, while in Sindh and and distributary leakage (45 percent), irrigation Balochistan, environmental evapotranspiration dominates. returns (26 percent), river recharge (6 percent), and other return flows (2 percent). This suggests FoDP (2012) suggests groundwater is in balance for that river recharge to groundwater is only 4 billion all provinces except KP. This is not consistent with cubic meters, which is difficult to reconcile with the several other assessments that indicate significant high river losses observed below Tarbela Dam. This groundwater depletion, including assessments based reinforces the importance of better spatially distributed on GRACE satellite data. Assessments based on GRACE water accounting, especially of surface-groundwater satellite data are at very coarse spatial scale and are exchanges. Despite the uncertainties, surface water and typically poorly validated. However, a robust regional- groundwater are tightly coupled in the Indus Basin, and scale assessment, based on in situ observations, shows need to be managed in a more integrated manner. that rising and falling groundwater levels exist across Pakistan (MacDonald et al. 2016). In Balochistan, groundwater is a smaller fraction of the total internal water resource. However, given the More local-scale assessments of the spatial pattern much greater interannual variability in river flows, and of depth to groundwater highlight the areas of the very low levels of built water storage, groundwater Punjab where groundwater depletion is concentrated dominates water use (Halcrow Group 2007). Indeed, (map 3.2). Data from 2002 and 2014 reveal a similar groundwater use in Balochistan exceeds average pattern but show an increase in the extent of depletion 37 Table 3.5  Estimated Average Annual Groundwater Balances by Province in Pakistan billion cubic meters Punjab Sindh Khyber Pakhtunkhwa Balochistan Total Recharge Rainfall recharge 8.1 2.4 1.3 1.5 13.3 Recharge from irrigation system 27.0 18.9 2.3 0.8 49.0 Return flow from groundwater abstraction 8.5 1.0 0.2 0.1 9.7 Recharge from the river system 1.4 0.4 0.1 0.2 2.2 Total 45.0 22.7 3.9 2.6 74.2 Discharge Groundwater abstraction 42.5 4.3 2.2 0.6 49.6 Nonbeneficial evapotranspiration losses 2.5 17.0 0.3 1.4 21.2 Base flow from rivers and subsurface 0 1.4 1.8 0.6 3.8 Total 45.0 22.7 4.3 2.6 74.6 Net balance 0 0 –0.4 0 –0.4 Source: FoDP 2012. Map 3.2  Groundwater Depth across Upper Indus Plain, 2002 and 2014 a. 2002 b. 2014 Jhelum BHIMBER ANNU CHAKWAL JHELUM MIANWALI GUJRAT KKI MARWAT Mianwali MANDI BAHAUDDIN SIALKOT Sargodah KHUSHAB NAROWAL Dera Ismail Khan GUJRANWALA SARGODHA HAFIZABAD Lahore IL KHAN Faisalabad SHEIKHUPURA BHAKKAR CHINIOT NANKANA SAHIB LAHORE PUNJAB FAISALABAD JHANG KASUR LAYYAH A Dera Ghazi Khan TOBA TEK SINGH I D OKARA Jalalpur N I SAHIWAL Meters bgl MUZAFFARGARH KHANEWAL Meters bgl 0–0.9 PAKPATTAN <0.9 1.0–1.4 1.0–1.4 VEHARI Bahawalpur 1.5–2.9 MULTAN 1.5–2.9 3.0–4.4 BAHAWALNAGAR 3.0–8.9 LODHRAN 4.5–5.9 9.0–12.9 6.0–11.9 13.0–17.9 >12.0 >18.0 Sources: Panel a: Basharat, Umair, and Azhar 2014; panel b: Khan et al. 2016. exceeding 12–13 meters in the lower Bari Doab. solute loads are typically high. Groundwater flow is Depletion is also significant in parts of Balochistan. slow because of naturally low gradients. Groundwater depletion can mobilize deeper (saline) groundwater, The resource value of groundwater is strongly and marine incursion occurs in the southern deltaic influenced by its quality. By area, well over half the region of Sindh. In areas distant from canal seepage alluvial aquifers of the Indus basin are saline and of recharge (parts of the Punjab and most of Sindh), limited value, especially in Sindh, although there are groundwater use is low, groundwater levels are high, pockets of fresh groundwater associated with the and waterlogging and high salinity are common. lower river (map 3.3). Groundwater salinity patterns Groundwater dynamics and sustainability are discussed reflect distance from freshwater recharge, groundwater further in chapter 5 under the assessment of water levels, and evaporation rates. In water logging areas, resources management performance. 38 PAKISTAN: GETTING MORE FROM WATER Map 3.3  Groundwater Salinity Levels across the Indus Basin of Pakistan TAJIKISTAN UZBEKISTAN CHINA TURK M EN IS T A N Indu s KHYBER Approximate PAKHTUNKHWA Line of Control ISLAMABAD CAPITAL Jammu Jammu AFGHANISTAN TERRITORY Kashmir andKashmir and Peshawar ISLAMABAD FEDERALLY ADMINISTERED TRIBAL AREAS lum Jhe nab he C b Zho Lahore PUNJAB vi Ra Quetta tlej Su us Ind BALOCHISTAN I.R. OF INDIA IRAN Groundwater salinity levels SINDH across the indus basin Fresh Relatively fresh Marginal Hazardous Karachi Selected cities and towns Province capitals us I nd National capital Province boundaries Arabian Sea International boundaries 0 50 100 150 Kilometers IBRD 44140 | 0 50 100 150 Miles DECEMBER 2018 Source: Qureshi et al. 2004. 39 References Khan, A. S. 1999. An Analysis of the Surface Water Resources and Water Delivery Systems in the Ahmad, S., and M. Rashida. 2001. “Indus Basin Indus Basin. Lahore, Pakistan: IWMI. Irrigation System Water Budget and Associated Problems.” Journal of Engineering and Applied Laghari, A. N., D. Vanham, and W. Rauch. 2012. Sciences 20 (1): 69–75. “The Indus Basin in the Framework of Current and Future Water Resources Management.” Hydrology Basharat, M., A. S. Umair, and A. H. Azhar. 2014. and Earth System Sciences 16: 1063–83. “Spatial Variation in Irrigation Demand and Supply across Canal Commands in Lutz, A. F., W. W. Immerzeel, P. D. A. Kraaijenbrink, Punjab: A Real Integrated Water Resources A. B. Shrestha, and M. F. P. Bierkens. 2016. “Climate Management Challenge.” Water Policy 16: Change Impacts on the Upper Indus Hydrology: 397–421. Sources, Shifts and Extremes.” PLoS One 11 (11): e0165630. doi:10.1371/journal.pone.0165630. Bastiaanssen, W. G. M., M. D. Ahmad, and Y. Chemin. 2002. “Satellite Surveillance of Evaporative MacDonald, A. M., H. C. Bonsor, K. M. Ahmed, Depletion across the Indus Basin.” Water W. D. Burgess, M. Basharat, R. C. Calow, Resources Research 38 (12): 1273. A. Dixit, S. S. D. Foster, K. Gopal, D. J. Lapworth, R. M. Lark, M. Moench, A. Mukherjee, M. S. Rao, Briscoe, J., and U. Qamar. 2005. Pakistan’s Water M. Shamsudduha, L. Smith, R. G. Taylor, Economy: Running Dry. Washington, DC: J. Tucker, F. van Steenbergen, and S. K. Yadav. World Bank. http://documents.worldbank​ 2016. “Groundwater Quality and Depletion in .org/curated/en/989891468059352743​ the Indo-Gangetic Basin Mapped from in situ /Pakistans-water-economy-running-dry. Observations.” Nature Geosciences 9: 762–68. Euroconsult. 2011. Handbook on Water Statistics of Qureshi, A. S., M. N. Asghar, S. Ahmad, and I. Masih. Pakistan. Report to the Water Sector Capacity 2004. “Sustaining Crop Production in Saline Building and Advisory Services Project of the Groundwater Areas: A Case Study from Pakistani Government of Pakistan. Unpublished; Lahore, Punjab.” Australian Journal of Agricultural Pakistan. Research 55: 421–31. FAO (Food and Agriculture Organization. 2011. AQUASTAT Pakistan (database). http://www.fao​ Qureshi, S., P. G. McCornick, M. Qadir, and Z. Aslam. .org/nr/water/aquastat/countries_regions/PAK​ 2008. “Managing Salinity and Waterlogging in /PAK-CP_eng.pdf. the Indus Basin of Pakistan.” Agricultural Water Management 95: 1–10. FoDP (Friends of Democratic Pakistan). 2012. A Productive and Water-Secure Pakistan: Rao, M. P., E. R. Cook, B. I. Cook, J. G. Palmer, Infrastructure, Institutions, Strategy. Islamabad, M. Uriart, N. Devineni, U. Lall, R. D. D’Arrigo, Pakistan: Water Sector Task Force, FoDP. http:// C. A. Woodhouse, M. Ahmed, M. U. Zafar, N. Khan, metameta.nl/wp-content/uploads/2013/11​ A. Khan, and M. Wahab. 2018. “Six Centuries of /FoDP-WSTF-Report-Final-09-29-12.pdf. Upper Indus Basin Streamflow Variability and Its Climatic Drivers.” Water Resources Research. doi: Halcrow Group. 2007. Supporting Public Resource 10.1029/2018WR023080. Management in Balochistan. Basin-Wide Water Resources Availability and Use. Irrigation Reggiani, P., and T. H. M. Rientjes. 2014. “A Reflection and Power Department, Government of on the Long-Term Water Balance of the Upper Balochistan, Royal Netherlands Government. Indus Basin.” Hydrology Research. doi: 10.2166​ Halcrow Pakistan (Pvt) in association with /nh.2014.060. Cameos. Sharif, M., D. R. Archer, H. J. Fowler, and N. Forsythe. Karimi, P., W. G. M. Bastiaanssen, D. Molden, and 2013. “Trends in Timing and Magnitude of Flow M. J. M. Cheema. 2013. “Basin-Wide Water in the Upper Indus Basin.” Hydrology and Earth Accounting Based on Remote Sensing Data: System Science 17: 1503–16. An Application for the Indus Basin.” Hydrology van Steenbergen, F., and S. Gohar. 2005. “Ground Water and Earth System Sciences 17: 2473–86. Development and Management.” Background Khan, A. D., N. Iqbal, M. Ashraf, and A. A. Sheik. 2016. Paper 12, World Bank, Washington, DC. Groundwater Investigations and Mapping in the van Steenbergen, F., M. Basharat, and B. K. Lashari. Upper Indus Plain. Islamabad: Pakistan Council 2015. “Key Challenges and Opportunities of Water Resources Research. http://www.pcrwr​ for Conjunctive Management of Surface and .gov.pk/Publications/Water%20Management​ Groundwater in Mega-Irrigation Systems, Lower /Groundwater%20Report-Final.pdf. Indus, Pakistan.” Resources 4: 831–56. C HAPT E R 4 Pakistan’s Water Sector Architecture Key Messages • Pakistan has very extensive irrigation infrastructure, but it is poorly maintained and outdated given the level of performance now required. Extensive modernization is required to support efficient and effective irrigation and drainage services. • It is widely believed that Pakistan has inadequate reservoir storage for reliable irrigation supply. But the low interannual flow variability of the Indus means interannual storage is not critical. In addition, new dams offer limited additional supply and comparatively low reliability. Nonetheless, Diamer Bhasha Dam can help mitigate the increases in flow variability anticipated with climate change and better match seasonal patterns of supply and demand, thus enhancing rabi supply reliability. Diamer Bhasha and proposed run-of-river dams are justified economically by hydropower, which can improve Pakistan’s energy security. • Pakistan has made major investments in flood protection infrastructure over recent decades. However, flood infrastructure needs significant additional investment because of expected increases in flood hazard (due to climate change) and increases in flood exposure (due to population growth and economic development). • Pakistan has grossly inadequate infrastructure for the collection, storage, sharing, and analysis of hydrometeorological data and information. Significant investment is required in hydromet infrastructure to support improved water resource assessments; water accounting; and the forecasting of water availability, droughts, and floods, especially given a changing climate. • None of Pakistan’s cities have adequate water supply and sanitation infrastructure. In some cases, supply infrastructure is reasonable but poor maintenance undermines service delivery. In many cases, however, supply infrastructure is not keeping pace with rapid urbanization. In all cases, wastewater treatment infrastructure is grossly inadequate, causing widespread pollution and serious environmental and public health impacts. • The national and provincial legal frameworks to support water policy implementation and to clearly define and assign legal mandates to relevant institutions are incomplete and require strengthening. • The 2018 National Water Policy provides strong support for improving water resources management, echoing other policy documents, including the National Climate Change Policy. Provincial policy frameworks for irrigation and water resources management are partial, fragmented, or nonexistent, and implementation has been inadequate. The policy frameworks for urban water services lack clarity and are not well aligned with relevant legislation, including local government legislation. • The institutional responsibilities for several aspects of water resources management are poorly delineated between national and provincial levels and between entities at these levels. Institutional responsibilities for urban water overlap or are unclear. • Provincial water sector financing has increased in recent years; however, federal financing has declined significantly in proportional terms. Collectively, sector financing is well below recommended levels. This is assessed to be the case for financing of major infrastructure, reforms and institutional strengthening, urban services, flood mitigation, and environmental management. 42 PAKISTAN: GETTING MORE FROM WATER T his chapter describes the architecture of Pakistan’s Guddu Barrage, 1962; and Chashma Barrage, 1971, water sector as a foundation for the subsequent increasing the canal command area by 35 percent to assessment of water sector performance. Sector 14 million hectares. architecture is the enabling environment for sector Fundamental to the design and operation of the IBIS performance. Sector architecture is described here in are a series of link canals that move water eastward terms of infrastructure, governance, and financing. from the mainstem Indus and the western tributaries Governance encompasses the legal frameworks, policy to the eastern tributaries. The earliest link canals—the settings, and institutional arrangements for water Upper Jhelum and the Upper Chenab links—expanded management. It is impossible to entirely separate irrigation on the western Rechna and Dari doabs, sector architecture and sector performance, and thus respectively. Following partition, further link canals this chapter, while primarily descriptive, includes some were built, and with the signing of the Indus Waters critique of the adequacy of sector architecture. Treaty in 1960 and the allocation of the waters of eastern rivers to India, the Trimmu-Sidhnai and Mailsi- People Bahawal canals were constructed to supply water Mitig from the western rivers to canal systems that were ices erv ati previously fed from Ravi and Sutlej rivers. Some of the eds on t multiple link canals, in addition to transferring water of ru ela rast ctur wa nf r-r from one river system to another, supply irrigation ter- ivery of wate e I water to one or more irrigation channels that off-take related risks Water endowment directly from the links. Envir IBIS water, which services 17.2 million hectares, is Del ns y titution om I s regulated through three major reservoirs, 16 barrages, o M s nm na n two headworks, two siphons across major rivers, and ce ur a ge co me so nt of w ater re e 12 interriver link canals. The irrigable area consists E nt of 44 canal commands: Punjab has 23; Sindh, 14; Khyber Pakhtunkhwa (KP), five; and Balochistan, two. The upper IBIS (11.3 million hectares) comprises 28 canal commands in KP and Punjab; the lower IBIS Infrastructure (5.9 million hectares) comprises 16 canal commands in Sindh and Balochistan below Guddu Barrage. The Water infrastructure—public and private— distributary network that services these command is necessary for all aspects of water sector areas is extensive, with an estimated 4,000 distributary performance. It is crucial for measuring water stocks channels divided into 107,000 watercourses. There and flows, storing and distributing water, generating is an estimated 44,000 kilometers of canals and hydropower, ensuring appropriate quality of water distributary channels and close to 20,000 kilometers supply, removal and treatment of wastewater of drainage channels (Euroconsult 2011). Euroconsult (domestic, industrial, and agricultural), and (2011) provides detailed descriptions of the irrigation protection from floods. This section summarizes the systems of each province. Figure 4.2 provides a brief evolution of Pakistan’s water infrastructure, describes summary of key infrastructure and the nature of the major new infrastructure being planned, and irrigation provided by province. highlights major infrastructure gaps. Chapter 5 provides additional discussion of the adequacy and The IBIS is supply-driven rather than demand-driven: performance of key infrastructure. demand usually exceeds supply, and available water is “pushed out” through the distribution system according Indus Basin Irrigation System to largely fixed rules. IBIS operation is almost fully manual. There is no internal reregulating storage, very The Indus Basin Irrigation System (IBIS) (figure 4.1) is rudimentary control for farm-level water delivery, a large, complex system of hydraulic infrastructure that and despite considerable and ongoing investment has been developed incrementally over many decades and improvement, most of the extensive distribution (figure 4.2). It represents an estimated US$300 billion network is unlined and leaky. in investment. Some of the irrigation infrastructure predates the formation of Pakistan but following Maintaining and operating the IBIS costs an estimated partition (when the total canal command area was US$102 per hectare per year (FoDP 2012); the around 10.4 million hectares), new irrigation systems largest annual costs per hectare are for the main were developed. Jinnah Barrage was completed in canals (US$38); distributaries (US$24); headwater 1947; Kotri Barrage, 1955; Taunsa Barrage, 1959; dams (US$20); and headworks, barrages, and link 43 Figure 4.1  Indus Basin Irrigation System of Pakistan Indus river Neelam river Jhelum river Kunhar river Tarbela reservoir Pak Kas ista hm Warsak dam n ir Chenab river Haro river Kabul river Mangla reservoir Soan river Marala Kalabagh reservoir Ravi river barrage (proposed) Jinnah Rasul power barrage Khanki Sutlej river channel Ind Kurram river barrage ia U-Jhelum B.R.B.D link Thal canal U-Chenab Chashma internal Qadirabad Chenab Pak internal B.R.B.D. ista reservoir barrage L.C.C. n internal CBCD L.C.C. U. Depalpur L-Jhelum feeder Sulemanki Thal reservoir L.C.C. west B.S. link barrage Chenab Balloki Gomal river (proposed) (Jhelum) B.S. link L.C.C. east barrage Trimmu barrage (Gujera) L. Depalpur Sidhnai barrage L. Pakpattan Haveli internal Ravi Islam Sidhnai Fordwah barrage Chenab Eastern L. Pakpattan Siddiqia Taunsa barrage Mailsi Qaim Muzaffargarh Dera ghazi khan Punjnad river U. Bahawal L. Bahawal Panjnad Abassia Guddu barrage N Pat feeder Desert feeder Ghotki feeder E Sukkur Beghari feeder barrage Northwest W Rice Dadu Nara Khairpur east Sehwan reservoir Rohri (proposed) Khairpur west S Kotri barrage Kalri Lined channel Proposed reservoir Pinyari fuleli Existing reservoir Arabian sea Barrage Source: FoDP 2012. canals (US$6). These costs are not significant relative rural domestic needs and livestock, yet they receive to the gross margin of wheat (approximately US$300 untreated wastewater. per hectare per year; e.g., Ishfaq et al. [2017]). IBIS’s condition is generally poor, and there is no asset management plan. The poor state of the irrigation Khyber Pakhtunkhwa infrastructure reflects deferred maintenance, low cost- Irrigation in KP is primarily in three geographic areas. recovery and collection efficiency, and a build–neglect– The largest is a contiguous area serviced by water rebuild cycle. The poor condition of IBIS is one cause from the Indus River (through the Pehur canals), of poor service delivery (see chapter 5). For example, the Swat River (through the Swat canals), and Kabul many irrigation canals also supply water for urban and River (through the Warsak and Kabul canals). To the 44 PAKISTAN: GETTING MORE FROM WATER Figure 4.2  Timeline of Major Irrigation and Water Resources Infrastructure in Pakistan, 1870 to Present SCARP Investment Bambawli-Ravi- Panjnad Barrage Bedian Link Canal Sidhani Barrage Rasul-Qadirabad Link Canal Upper Chenab Marala Link Canal Barrage Taunsa Barrage Kurramtangi Dam Sidhnai-Maisli Link Canal Balloki Sulemanki Link Canal Marla Barrage Mangla Mangla Dam Headworks Kotri Barrage (Jhelum) completed Thal Canal Islam Barrage Tausa-Punjnad Link Canal Khanki Headworks Sirhind Canal Qadirabad Barrage Trimmu Barrage Tarbela Dam (Indus) completed 1870 1890 1910 1930 1950 1970 1990 2010 Sidhnai weir Tarbela augmentation Marla-Ravi Link Canal Rasul Headworks Sukkur Chasma-Jhelum Link Canal Barrage Warsak Dam Chasma Barrage Kachi Canal Balloki Barrage Qadirabad-Balloki Link Canal Rasul Barrage Sulemanki Jinnah Barrage Malisi-Bhawal Link Canal Upper Jhelum Barrage Link Canal Trimmu-Sidhnai Link Canal Siphon Barrage Guddu Barrage Note: Shadings indicate pre-Pakistan, pre-IWT and post-IWT periods. southeast of this contiguous area is a smaller area supplies water to three headworks (Sulemanki, serviced by the Tanda Dam canals and the Marwat Islam, and Panjnad), which feed eight main canals. Canal system. The Chashma Right Bank canals service Across the thousands of kilometers of canals and a long narrow irrigation area parallel to the Indus distributaries, an estimated 58,000 outlets supply water mainstem downstream of Chashma Barrage. In addition to farms. Maintenance and upgrading is complex and to government-operated and -maintained canals, problematic and has been the focus of considerable there are a series of smaller private (or “civil”) canals, development finance in the past, including for lining recognized under the Water Apportionment Accord. thousands of kilometers of canals and distributaries. Groundwater pumping supplements a mere 4 percent In addition to this complex water distribution system, of the area irrigated by canals. an estimated 800,000 private tube wells across Punjab supplement surface water supply (Qureshi 2010). Punjab Around 21 percent of the irrigated area relies solely on groundwater, and an additional 55 percent of the Punjab has the largest and most complex irrigation area relies on supplementary groundwater irrigation, system of Pakistan. The Indus River supplies water from especially during rabi when canal water is insufficient the Tarbela Dam to the Jinnah and Chashma barrages to meet demand. The reliance on groundwater sets and the Taunsa Headworks, which connects to two Punjab apart from the other provinces; therefore, major link canals and four main canals. The Jhelum conjunctive management of surface water and River supplies water from the Mangla Dam to the Rasul groundwater is critical. Barrage, which connects to two major link canals and two main canals. The Chenab River supplies water Achieving adequate irrigation drainage in Punjab is to four major headworks (Marala, Khanki, Qadirabad, difficult. While some water drains back to the river, and Trimmu), which feed five major link canals and a fraction is too saline for safe disposal in the river. four main canals, and a number of smaller branch Trials of evaporation basins have not proved successful, canals. The Ravi River supplies water to two headworks given the risks of groundwater contamination and (Ballokai and Sindhnai), which feed two major link the large land areas required. Saline drainage from canals and two main canals, and the Sutlej River around 2 million hectares is moved from near the 45 Punjab–Sindh border in the Left Bank Outfall Drain (sailaba), which uses very simple diversion structures either directly to the Arabian Sea or to the shallow, to harvest short, flashy floods into bunded basins to saline Shakoor Lake that straddles the Pakistan–India pond and infiltrate. Water harvesting (khushkaba) border west of the Great Rann of Kutch. is a smaller version of flood irrigation that relies on capturing local unchanneled surface runoff in bunded Sindh basins; it serves around two-fifths of the irrigable area. The provincial government manages canal irrigation Irrigation in Sindh dates back several thousand and, to a lesser extent, small-scale irrigation schemes. years. Irrigation canal systems were extended and Farmer communities manage water harvesting and improved during the late 1800s, and a major program flood irrigation, although the government supports of irrigation expansion began under British rule in infrastructure construction. the latter half of the 19th century. Barrage irrigation commenced in 1932 when the Sukkur Barrage Major Reservoirs and Hydropower became operational, followed by Kotri (1955) and Guddu (1962). Over half of the Sindh command area Three large dams constructed in the 1960s and 1970s— is supplied from Sukkur Barrage through four left-bank the Tarbela on the Indus, the Mangla on the Jhelum, and three right-bank canals. Guddu supplies around and the Chashma on the Indus—account for most of one-quarter of the Sindh command area, and Kotri the built water storage in Pakistan (WAPDA 2016). supplies less than one-quarter. Saline groundwater lies Designed primarily to supply water for irrigation, the under at least 80 percent of the irrigated area, and original combined live storage capacity of these dams groundwater irrigates less than 20 percent of the total was 19.4 billion cubic meters (Tarbela, 12 billion cubic area. Sedimentation around the lower barrages has meters; Mangla, 7.3 billion cubic meters; and Chashma, been a major problem for effective barrage operation 0.87 billion cubic meters). Ongoing sedimentation has, and barrage safety—affecting the ability of barrages however, decreased capacity by around 1 percent per to pass major floods. There has been significant recent year to 15 billion cubic meters by 2007. Inspections and ongoing investment in restoring and modernizing of the Tarbela Dam have found it could have been the Guddu and Sukkur barrages. designed to remove sediment using drawdown flushing, but subsequent downstream development Thirteen surface drainage systems in Sindh service half (barrages and irrigation off-takes) would have the irrigated area, and two subsurface drainage systems precluded major sediment flushing. service 2 percent of the irrigated area. However, the Sindh drainage system is neither contiguous nor The average annual sediment discharge into Tarbela integrated, and waterlogging is widespread due to high Dam is about 181 million tons. The trap efficiency surface water delivery (van Steenbergen et al. 2015). of the reservoir, that is, the percentage of incoming For more than one-third of the command area, the sediment retained by the reservoir, is greater than water table is within 1 meter of the surface, and across 95 percent in most years. The live reservoir capacity another third it is between 1–1.5 meters. The root zone in 1974 was 11.94 billion cubic meters but declined to is thus waterlogged across 70 percent of the command 8.55 billion cubic meters by 2006—a reduction of more area for much of the time, only decreasing at the end than 28 percent in 32 years. The volume of sediment of rabi when canal supply dwindles (van Steenbergen accumulated in the reservoir is now too large for et al. 2015). practical removal. Construction of Diamer Bhasha Dam upstream of Tarbela will create a sediment trap, thus incrementally reducing Diamer Bhasha live storage Balochistan but significantly slowing the sedimentation rate of Canal irrigation in Balochistan is limited and supports Tarbela. Mangla Dam was enlarged between 2005 just 0.3 million hectares (around one-fifth of the and 2009 (at a cost of around US$1 billion) adding an irrigable area). Water from the Indus is supplied additional 3.6 billion cubic meters of live storage. Due by the recently completed Kacchi Canal from the to continued sedimentation, combined live storage is Taunsa Barrage, by the Pat Feeder and Desert canals estimated to be around 16 billion cubic meters. Diamer from the Guddu Barrage, and by the Khirthar Canal Bhasha Dam, at preliminary construction stage and from the Sukkur Barrage. Small-scale irrigation from with an estimated total cost of around US$14 billion, groundwater and the perennial rivers (of the Makran will add 7.9 billion cubic meters of live storage. At coastal and Indus drainage systems) support close to projected completion in 2023, total system storage will another 0.3 million hectares. Groundwater is accessed be around 21 billion cubic meters. The ongoing loss of through traditional karezes, shallow dug wells, and storage because of sedimentation costs tens of millions deep tube wells. An additional 0.3 million hectares, of U.S. dollars per year, which is very small relative to approximately, are serviced by flood, or spate, irrigation the total annual IBIS maintenance cost. 46 PAKISTAN: GETTING MORE FROM WATER Table 4.1  Current and Future Reservoir Capacity and Active Groundwater Storage Capacity in Pakistan Current reservoir Future reservoir Active groundwater capacity capacity storage capacity Days of average inflow 34 48 5,739 Days of kharif inflow 20 29 3,397 Days of rabi inflow 94 135 16,107 Days of average irrigation supply 46 66 7,863 Days of kharif irrigation supply 35 51 6,016 Days of rabi irrigation supply 68 98 11,612 Years of national urban demand 11 15 1,824 Years of Karachi demand 18 26 3,123 Source: Author calculations. An oft-cited reason for Pakistan’s lack of water security for water supply, but discussions and planning of is insufficient reservoir storage. Many comparisons storage reservoirs must recognize this important natural have been made to other major river systems based storage in the Indus basin because it ensures a far on storage volume per capita or days of storage in more naturally regulated flow than in most other large terms of the average flow. Table 4.1 indicates the days river basins of semi-arid regions. or years of storage in terms of average and seasonal Although comparisons between river systems on flows, and agricultural and urban demands, both for such metrics as “days of average flow storage” are the current and future storage volumes. Because interesting, they can be very misleading. The level of kharif demand is largely synchronous with natural storage required in a water supply system depends supply timing, it is storage for rabi demand that is on the variability of inflows, the temporal pattern of most important. Current storage can meet 68 days of demand, and the economically acceptable level of average rabi irrigation supply. Current storage is not a variation in meeting these demands. In many large constraint to urban water supply reliability. Although arid-zone rivers, inflows are far more variable between distribution infrastructure is not in place to supply years than in the Indus. The dominance of meltwater reservoir water to all major cities, Indus system storage (glaciers and snowpack) in the Indus means inflows is equivalent to 11 years of current total urban demand are far more stable between years than in most other or 18 years of Karachi demand. Urban supply reliability large irrigated basins. is an issue of intersectoral demand prioritization and storage operation. A global comparison of the level of system storage relative to the variability of annual flows indicates Storage comparisons can be made for active that while storage in the Indus is indeed very low groundwater storage and glacier water storage. Khan (less than 10 percent of the mean annual flow), et al. (2016) estimate Pakistan’s active groundwater this is commensurate with the low variability of storage to be 2,736 billion cubic meters—170 times annual flows (figure 4.3). The high relative storage current reservoir storage—notionally equivalent to over volumes commonly cited for other basins—two to 20 years of canal irrigation supply. Investigations of three times the mean annual flow—are not required the pragmatism of enhanced operational management for the Indus for managing the low interannual flow of the groundwater storage capacity are therefore variability. Nonetheless, although in the past system warranted. Current rates of groundwater depletion storage has enabled annual canal withdrawals to that have attracted much attention may turn out to be be reasonably reliable, storage is inadequate for reasonable in the context of a multiyear conjunctive mitigating the impacts of major drought sequences water use strategy, except in the cases of severe such as 1999–2001. Past modeling has indicated that localized depletion. the addition of Diamer Bhasha will help mitigate The volume of glacial ice between the entire Himalaya (but not remove) the impacts of major droughts and Karakoram ranges is estimated to be between (Robinson and Gueneau 2014). There are other 3,000 billion cubic meters and 5,000 billion cubic options besides additional storage to manage inflow meters. Around half the glaciated area is in the variability, including conjunctive surface water and Karakoram range of the Upper Indus (Azam et al. groundwater management and water markets 2018). Glacier “storage” cannot be actively managed (see chapter 5). 47 Figure 4.3  Reservoir Storage Volume as Ratio of Mean Annual Flow Compared to Coefficient of Variation of Annual Flow for Selected Countries and Regions 3.5 3.0 Southern Africa Storage volume/Mean annual flow 2.5 2.0 Australia 1.5 Northern Africa 1.0 Asia South Pacific South America Global average 0.5 Pakistan North America Europe 0 0.2 0.4 0.6 0.8 1.0 CV of annual flows Source: WAPDA unpublished data for Pakistan and data from McMahon et al. 2007. Note: Relative storage value is the median of quoted ranges, assuming 75 percent draft (yield) and 95 percent reliability. CV = coefficient of variation. Although the Indus is characterized by strong seasonal Low reliability yield is not without value but requires flow patterns, the timing of inflows is not entirely an irrigation sector that adapts well to yearly changes mismatched with the timing of water demand, unlike in water supply—this is not currently the case. Low many basins in temperate climates in which winter reliability yield can support opportunistic irrigation that inflows support summer irrigation. The performance generates intermittent economic returns. Farmers with of the IBIS is to some degree limited by available only low reliability supply need alternative sources of storage, but primarily because of an inability to fully income in drier years. meet rabi demands, for which groundwater pumping is critical. Additional storage would enhance the ability to Global analyzes and modeling of the Indus support the regulate flow and store water within the year between view that additional reservoir storage alone will be of seasons. Additional storage could increase total system limited value for addressing water supply shortages. yield. However, as established by Lieftinck, Sadove Gaupp, Hall, and Dadson (2015) consider the role of and Creyke (1968), once storage on the Indus reaches water storage infrastructure for managing intra- and around 22 billion cubic meters, additional storage interannual water supply variability for more than volume becomes increasing inefficient: for each unit of 400 important river basins around the world, based storage volume added the volume of additional yield on macro hydrologic modeling. For the Indus-Pakistan, is progressively less. The addition of Diamer Bhasha they find that while it ranks highest in the world in will bring storage on the mainstem Indus to around terms of scarcity (defined in terms of a temporal 16 billion cubic meters. Further, as shown by World analysis of supply-demand gaps), storage dependency Bank (1998), the additional yield from new storage is among the lowest of all water scarce areas, would be of lower reliability, quickly dropping below the because the scarcity index for the Indus in Pakistan is current 75 percent (proportion of years that storages fill very similar with or without storage. This is because totally). For example, the incremental additional storage demands are similar very high relative to supply and from the Mangla enlargement is estimated to have are not compromised by interannual supply variability. a reliability of 72 percent. At 25 billion cubic meters A simpler analysis by Brown and Lall (2006) reaches of system storage, overall yield reliability would drop similar conclusions. They assess demand in terms of the to 40 percent to 60 percent or even lower depending water required to grow sufficient food for the national on environmental flow targets (World Bank 1998). population and calculate annual and intra-annual water 48 PAKISTAN: GETTING MORE FROM WATER balances by analysis of rainfall variability—which for (see chapter 2). Total installed capacity is about the Indus undervalues meltwater. Their calculations 7.3 gigawatts, dominated by Tarbela (3.5 gigawatts), enable analysis of the storage required to match supply Ghazi Barotha (1.5 gigawatts), and Mangla to demand, and, consequently, the water efficiency (1.0 gigawatts). Pakistan has ambitious plans to needs (“soft” measures) to balance supply and demand increase hydropower capacity more than fivefold when storage is not a constraint. Even considering only through 55 new projects that are at various stages of rainfall variability, Brown and Lall (2006) find that for readiness, including 10 under construction (figure 4.4, the Indus, shortages should be managed entirely with panels a and b). Many of the proposed projects are “soft” measures and not with additional storage. run-of-the-river, with little storage. Run-of-the-river schemes rely directly on river flows, although most Hydro-economic modeling of the Indus system by have a small reservoir to dampen the short-term inflow Yu et al. (2013) includes a scenario with new storage variations and manage sediment impacts on gates and equivalent to more than twice the live capacity of the turbines. They are sometimes downstream of a major Diamer Bhasha. While the additional storage would storage scheme (e.g., Ghazi Barotha downstream of likely mitigate the economic impacts of drought, Tarbela). Generation by storage schemes (e.g., Tarbela the overall economic outcome would be negative and Mangla) is a function of reservoir water level if the significant value of additional hydropower (hydraulic head to turbines) and rate of flow release. were excluded, and both agricultural gross domestic product (GDP) and household income declined. This The sequence from a prefeasibility study to identify scenario is likely because the current irrigation system potential projects through construction and operation does not have the capacity to benefit from increased takes many years. Despite considerable uncertainties, average supply, and even improved reliability of rabi including questions of financing, a timeline of supply is of relatively minor economic value given the increasing capacity shows Pakistan could triple its dominance of wheat production and the large costs hydropower generation capacity within the next of major new infrastructure. In contrast, scenarios of few decades (figure 4.5. The proposed expansion improved irrigation system efficiency and improving is predominantly in KP and Jammu and Kashmir crop technologies and yield—either separately or (figure 4.6), including a string of 10 dams along the combined—delivered significant economic benefits. Upper Indus Basin commonly referred to as the “Indus These analyses did not consider scenarios of managed Cascade.” A small number of the total proposed environmental flows or the economic benefits of projects are considered priority projects by WAPDA, environmental flows. and of these, only Diamer Bhasha adds significant The hydropower benefits of existing and proposed additional storage (table 4.2). Continued development new dams are significant. Hydropower accounts for of this significant series of hydropower projects can about 35 percent of national electricity generation help improve Pakistan’s energy security, but the Figure 4.4  Pakistan’s Dam Readiness Stage by Number and Generating Capacity a. Number b. Generating capacity (percent) Under study Under study 7 15 In operation In operation 14 16 Under construction Desk studies/ 10 feasibility stage Under 24 Desk studies/ construction feasibility stage 10 24 Ready for construction Ready for construction to start to start Design/ 27 Design/ 6 procurement stage procurement stage 8 8 Source: WAPDA unpublished data. Note: Tarbela is counted as “in operation,” including extensions IV and V. 49 Figure 4.5  Projected Increases in Hydropower Capacity, Pakistan 60 50 40 Gigawatts 30 20 10 0 Now 5–10 10–20 15–20 >20 >25 Approximate time horizon (years) Under operation Under construction Construction ready Under design/procurement Feasibility studies Prefeasibility studies Source: WAPDA unpublished data. Note: Tarbela counted as “under operation,” including extensions IV and V. Figure 4.6  Capacity of New Hydropower Projects by Readiness Stage and Province or Territory, Pakistan 30 25 20 Gigawatts 15 10 5 0 a no nd ab so nd dh n w sta nj ir, a ir, u a Sin rth h kh i ut Pu m u ch un sh mm sh mm lo ht Ba Ka Ja Ka Ja ak m e rP yb Kh Under operation Under construction Feasibility study complete Under study Source: WAPDA unpublished data. cascade investments will have limited impact on water Indus mainstem and its major tributaries. About management further downstream (with the exception 6,800 kilometers of river are embanked with levees, of Diamer Bhasha, which will assist with drought and mostly in Punjab and Sindh (figure 4.7). In Punjab, there flood mitigation and improve reliability of rabi supply). are levees on the left and right banks of the main stem of the Indus River between Taunsa Barrage and the confluence with the Chenab River. The Chenab is leveed Flood Protection Infrastructure on both sides between Sidhnai Barrage and Panjnad Pakistan has an extensive system of flood protection, Barrage to protect the city of Multan and surrounding mainly comprising levees and spurs along the areas. The southern bank of the Sutlej River between 50 PAKISTAN: GETTING MORE FROM WATER Table 4.2  Priority Dam Projects in Pakistan Dam River Capacity (GW) Live storage (BCM) Estimated cost Status (US$, billions) Diamer Bhasha Indus 4.50 7.9 11.18 Construction ready Kurram Tangi Kurram 0.08 1.1 0.70 Under construction Tarbela 4th Extension Indus 1.35 — 0.83 Ready for construction Munda Swat 0.74 0.9 1.40 Under study Kohala Jhelum 1.10 run-of-river 2.40 Design/procurement Bunji Indus 7.10 run-of-river 6.84 Construction ready Dasu Indus 4.32 0.8 5.21 Under construction Total 19.19 10.7 28.55 Sources: FoDP 2012; WAPDA 2016. Note: — = not available. Figure 4.7  Length of Levees and Number of Spurs by Province in Pakistan 3,500 800 3,000 700 600 2,500 Spurs (number) 500 Levees (km) 2,000 400 1,500 300 1,000 200 500 100 0 0 Punjab Sindh Khyber Balochistan Pakhtunkhwa Levees Spurs Source: Ali 2013. Islam Barrage and Panjnad Barrage is also protected Flood management infrastructure faces considerable by levees. There are few levees, however, along the challenges from sedimentation. In response to Jhelum River. In Sindh, the left bank of the Indus is embankment construction, the Indus has been leveed along its full length (around 600 kilometers), aggrading rapidly over the last two decades, leading to and the right bank is leveed from Guddu Barrage to breaches upstream of barrages and inundation of large Manchar Lake. A total 1,410 flood spurs have been built areas (Gaurav et al. 2011). In addition, climate change since 1960 to protect riverbanks from erosion and direct is increasing flood frequencies in the Indus Basin flood flows. Nearly half of Pakistan’s flood spurs are in (Nepal and Shrestha 2015) suggesting flood standards Balochistan, where their flow-diverting function is an should be revised. Levee heights are usually an integral part of the many spate irrigation systems. arbitrary 1.8 meters (Ali 2013). But river morphology and climate changes suggest flood infrastructure The 18 barrages can be operated to divert flood flows should be upgraded and integrated with improved and reduce downstream flooding. Although constructed nonstructural flood protection (see chapter 5). primarily for water supply, Mangla and Tarbela dams provide some flood mitigation potential. Following the Hydrometeorological Infrastructure 1992 flood, reservoir operating protocols were revised to incorporate flood mitigation objectives (Ali 2013). Pakistan has inadequate hydrometeorological In the 2010 flood, Mangla Dam operations reduced infrastructure, and much of the monitoring network downstream flood heights by 35 percent, and Tarbela is in disrepair. Development finance has supported Dam operations reduced downstream flood heights by improvements to the monitoring network, but this 28 percent (Ali 2013). has been far from adequate, and the usefulness of 51 monitoring is compromised by antiquated infrastructure used in volumetric water accounting or to guide for data transmission, processing, and storage. operation of the reservoirs and barrages. Most of the other hydrologic stations provide data for feasibility For hydrologic monitoring, federal authorities operate studies to guide construction project implementation primary rim stations of the Indus Basin that gauge or for flood management (forecasting, early warnings, inflows to the IBIS. These are usually defined as the and operations). Most of these stations also measure Indus at Kalabagh (which includes the Kabul River), suspended sediment loads. Actual discharge the Jhelum (at Mangla), the Chenab (at Marala), the measurements for establishing and revising “ratings Ravi (at Balloki), and the Sutlej (at Sulamankai). In curves” (to convert water level measurements to flow some instances, the Indus (at Tarbela—upstream) estimates) are undertaken using a mixture of wading, and the Kabul (at Nowshera) are used instead of the cableways, bridges, and boats. Indus at Kalabahgh. Records began on the Jhelum in 1922; the Indus, in 1936; the Chenab, in 1940; and Flow gauging of irrigation withdrawals is undertaken eastern tributaries, in Pakistan in 1960. In addition, by provincial authorities. There remains considerable federal authorities have gauged flows at key barrages uncertainty in the hydrological monitoring and a (Taunsa, Panjnad, Guddu, Sukkur, and Kotri) since lack of trust among the governments in the flow their construction, and maintain 59 other regular flow measurements. Prior flow gauging telemetry systems gauging stations mostly in the Upper Indus Basin have failed due to deficiencies in both design (e.g., lack as well as for various major drains and hill torrents of adequate power backup supplies) and operation, (figure 4.8). For these stations, the earliest flow records with opportunities for accidental or deliberate human- began in 1960, and the average length of record for introduced errors. New telemetry systems are being these stations is just 35 years (ignoring several gaps in designed to improve the accuracy and reliability of the records). Federal authorities also maintain 59 flow measurements, as well as to improve the transparency gauging stations associated with specific water or of real-time data sharing. hydropower development projects (Daimer Bhasha There is very limited active hydrological monitoring Dam, Dasu Hydroelectric Power (HEP), Neelum-Jhelum outside the Indus Basin in Balochistan. Several prior HEP, and Kachhi Canal, among many others). The gauging stations have fallen into disrepair given project-related gauge stations have differing but mostly security, access, capacity, and resourcing challenges. relatively short periods of record. Lack of hydrological data compromises many aspects of water resources management and development. Around 35 of the regular monitoring stations take hourly measurements using automatic water level There is limited operational monitoring of sensors—either data loggers or data transmission groundwater in Pakistan. Some municipal authorities facilities. The remainder use manually read staff monitor groundwater in urban centers. In Punjab, gauges. Only a small number of the gauging stations— groundwater monitoring in irrigation command largely the rims stations and the barrage flows—are areas is more systematic than elsewhere due to Figure 4.8  Federally Operated Regular Flow Gauging Stations in the Upper Indus Basin of Pakistan and Average Period of Record, 1960–2016 60 40 35 Average length of record (years) 50 Cumulative no. of stations 30 40 25 30 20 15 20 10 10 5 0 0 1960 1970 1980 1990 2000 2010 2020 Average length of record Cumulative number of stations Source: WAPDA unpublished data. 52 PAKISTAN: GETTING MORE FROM WATER Table 4.3  Hydrological and Meteorological Monitoring Infrastructure Maintained and Operated by Federal Authorities in Pakistan Station type No. Remarks Hydrologic monitoring Flow gauging stations of 7 Data logging and intermitted discharge measurements. glacier melt Indus key sites—rim stations 24 Variable record lengths reflecting construction history. Some manual and some and barrages automatic water level sensors. New flow telemetry system planned for these key sites; hydraulic calibrations completed for seven key sites. Regular streamflow and 59 Variable record lengths, some intermittent records sediment gauge stations Project-related streamflow 59 Installed as part of dam or other infrastructure project (mostly mid-1990s onward). and sediment gauge stations 56 operating. Flood forecasting telemetry 49 Flood forecasting, river water level, precipitation, discharge/flow measurements stations during floods. All operating. Canal network flow gauge — Provincial monitoring of irrigation distribution network, including main, submain, and stations minor canals. Groundwater observation Multiple Monitoring of groundwater levels in irrigation command areas, primarily in Punjab; wells ad hoc elsewhere. Meteorological monitoring Rainfall radar 7 Two S-band and five C-band rainfall radar Synoptic stations 97 50 automatic weather stations (35 functioning) and 33 agrometeorological stations Upper air stations 6 None functional due to lack of consumables Manual climate stations 12 Measuring evaporation, temperature, humidity, rainfall. All operating. Regular climate stations 40 Variable lengths of record; none installed before 1960. Project-related climate 12 Installed as part of dam or other infrastructure project (mostly last decade onward). stations Nine operating. High-altitude automatic 20 Collection and transmission of hourly temperature, precipitation, relative humidity, weather stations wind, solar radiation, and snow water equivalent; 17 new stations planned. Source: WAPDA unpublished data. Note: — = not available. a network of piezometric wells. Even in Punjab, outdated, and poorly maintained. The systems and however, groundwater monitoring is inadequate for capacity for flood forecasting and early warnings are sound resource management. There is very limited inadequate and in need of major upgrade. groundwater monitoring in Sindh, KP, and Balochistan The infrastructure for managing (archiving and (Bhatti et al. 2017). In Punjab there are an estimated accessing) hydrometeorological data is rudimentary 1.1 observation wells per 1,000 square kilometers and not standardized. Other than for basic statistics (Government of Punjab 2012); manual readings are at key Indus sites, much of the hydrologic data from made twice per year on average. Data are stored in before the 1990s exist only in hard copy form. During digital form for ad hoc use (Bhatti et al. 2017). There is the 1990s, some hydrological data began to be stored no complete or systematic inventory of the estimated in DBHydro, a purpose-built software developed with 1 million tube wells accessing groundwater. German aid. More recently, HYSTRA software has been Weather, climate, flow, and flood forecasting are used (with support from Australia) to store, analyze, informed by meteorological measurements including and report hydrometeorological data. However, these from a few weather radars, a limited number of systems are not accessible online, and so outside of automatic weather stations, and numerous manual the custodian agencies, data access is very limited. The weather stations (table 4.3). Additional meteorological information technology (IT) infrastructure to support monitoring, including the use of around 500 basic hydrometeorological data management, analysis, rainfall gauges, is undertaken across the provinces by forecasting, and dissemination is hampered by low- federal and provincial governments. Overall, however, speed Internet access, a lack of forecast workstations, the meteorological observation network is inadequate, and outdated servers. Both federal and provincial 53 authorities share some real-time or near-real-time data to drains and canals. A recent survey indicates that online, but do not provide public access to historical most of Peshawar’s sewerage, including three pumping data. Protocols for public data access vary, and the stations, is unused because of poor maintenance (NDC capacity to manage requests is low. Pakistan should 2014). There are three wastewater treatment plants, but move to a modern open access system for managing none are functional. Untreated effluent is discharged to and sharing water data and information. rivers and canals or used to irrigate crops. Water Supply and Sanitation Punjab Pakistan’s domestic water supply infrastructure is in poor About 35 percent of Punjab’s publicly owned water state. Much of the existing water supply infrastructure, supply schemes are dysfunctional (PCRWR 2011). The including pipe networks, pumping stations, groundwater infrastructure is in poor condition, and its capacity is wells, and water treatment facilities, are not inadequate to meet quantity requirements and quality functioning and are inadequate for a rapidly urbanizing standards. Many households, therefore, rely on private population. Pipe networks are aging and in need of tube wells, making it difficult to monitor or control the replacement with very high levels of nonrevenue water quality of the water supplied or to prevent groundwater (NRW). Because of largely inadequate public supply depletion. infrastructure, alternative supplies such as private tankers and privately-owned groundwater wells are increasingly In Lahore, water is withdrawn from nearly 600 tube common, especially in Karachi and Quetta. wells, of which nearly 500 are publicly managed, before being pumped directly into the 3,200-kilometer Sanitation infrastructure is grossly inadequate. None piped distribution network (AIIB 2018). Given very of Pakistan’s major cities has adequate sewerage or high levels of groundwater contamination, the lack of wastewater treatment capacity, and in many cases filtration or other treatment for this supply represents existing wastewater treatment facilities are not well a major public health risk. The supply network is maintained. Wastewater treatment capacity is sufficient aging, with most pipes needing replacement (AIIB for only about 8 percent of the wastewater load 2018). The Lahore sewerage system includes about (Murtaza and Zia 2012). Inadequate maintenance, lack 4,000 kilometers of underground sewers and 14 major of funds, and lack of human resources mean much less drains. There are no wastewater treatment facilities, than 8 percent of wastewater is treated. Most urban and an estimated 2.4 cubic megameters of raw sewage and other wastewater is discharged untreated sewage from the drains are discharged into the Ravi into surface water bodies or used to irrigate crops. This River each day (AIIB 2018; Qureshi and Sayed 2014). pollutes waterways and contaminates food and water Water supply and sanitation infrastructure in other large supplies. The extent and state of urban water supply and cities of Punjab is in equally poor condition as Lahore’s. sanitation infrastructure for the major cities of Pakistan Nearly all of Faisalabad’s water supply is groundwater is summarized in the following sections by province. In pumped from tube wells along the Chenab River and rural areas, public infrastructure is very limited, other than the Jhang Branch Canal; public supply is augmented by dilapidated open drains that combine storm runoff and private tube wells (JICA 2016). Estimates of the total untreated wastewater. The rural situation is discussed in available water supply suggest a daily range of about 0.3 chapter 5 from a service delivery perspective. million cubic meters to 0.33 million cubic meters (JICA 2010; WASA-F 2018). Only about 0.02 million cubic per Khyber Pakhtunkhwa day undergo treatment and filtration before distribution The public water supply for Peshawar is mostly (WASA-F 2008). There is one wastewater treatment plant groundwater, with over 700 tube wells and 33 filtration in western Faisalabad with a capacity of 0.076 million plants (sand filters), of which 24 are functional (NDC cubic meters per day (WASA-F 2018). An estimated 1.1 2014). Around 23,000 cubic meters are withdrawn million cubic meters per day of wastewater (sewage each day from the Bara River and treated at the Bara and storm water) are discharged to rivers through Water Treatment Plant. The water distribution system drains without treatment (WASA-F 2018). In Multan, of approximately 1,670 kilometers of pipes—37 percent there are no water treatment facilities and only a single above ground and mostly of galvanized iron—has been wastewater treatment plant (Soncini et al. 2014). extended haphazardly to keep up with rapid urbanization. Much of the network is well beyond its 20-year design life Sindh and needs replacing. Across Sindh, 58 percent of the publicly owned water Sanitation infrastructure includes open and covered supply infrastructure (pumps, water treatment works) drains that convey domestic and industrial wastewater are dysfunctional (PCRWR 2010). Karachi’s water supply as well as surface runoff. Underground sewers serve only system comprises surface water storages and transfers a few areas, and these are often clogged and overflow (Hub and Haleji systems), groundwater (Dumlottee 54 PAKISTAN: GETTING MORE FROM WATER Wellfield) and a distribution network developed as valves. Asbestos cement leaches from the pipes, which Karachi has expanded. Around 2.4 million cubic meters contaminates the supply (IUCN 2015). The Islamabad per day are supplied to Karachi, of which two-thirds is sewerage system has not expanded as the city has treated at one of seven filtration plants (KWSB 2018). grown and is now overloaded (Manarvi and Ayub The supply network is in very poor condition with 2013). There is a single wastewater treatment plant extremely high leakage and other unaccounted losses; maintained and operated by the Capital Development widespread cross-connections with the sewerage Authority, although a recent audit indicated inadequate system contaminate the supply. performance, with treated effluent not meeting national standards. Karachi generates an estimated 1.8 million cubic meters per day of sewage (KWSB 2018). This amount indicates high system losses given that typically Water Governance less than 10 percent of urban water supplied is Many of Pakistan’s water-related challenges consumed. There are three wastewater treatment are governance challenges. Water governance plants (combined capacity of 0.6 million cubic meters encompasses “overarching policies, strategies, per day), of which two are functional (treating plans, finances and incentive structures that concern 0.2 million cubic meters per day). Thus, around or influence water resources; the relevant legal 1.6 million cubic meters per day of wastewater and regulatory frameworks and institutions; and is discharged untreated into the Arabian Sea or planning, decision-making and monitoring processes” freshwaters close to Karachi. The Karachi sewerage (FAO 2018). The focus here is on legal frameworks, system is extensive, with 5,670 kilometers of sewers, policies, and institutional arrangements; sector six major pumping stations, and 32 minor pumping financing is covered in a subsequent section. Effective stations (KWSB 2018). However, the system is in very water governance underpins water security by poor condition. Inadequate stormwater drains are sustainably, equitably, and transparently determining clogged with solid waste and overloaded with sewage “who gets what” and “who does what” in terms of and industrial effluent, undermining their capacity water resources and services, and mitigation of water- to drain stormwater from flood-prone areas (World related risks. For Pakistan, effective water governance Bank 2018). Hyderabad has a major water supply and must be tailored to the country’s unique biophysical wastewater infrastructure gap; water is withdrawn and geopolitical context and reflect its cultural and directly from the Indus River and about half is treated political traditions. before distribution. About a quarter of the 0.2 million cubic meters per day of sewage load is treated before Federal systems for water governance often have a discharge (PCRWR 2010). complex patchwork of institutions, policies, and legal provisions at provincial and national levels. This is the Balochistan case in India, Malaysia, Nigeria, Argentina, the United States, and Australia (Goldface-Irokalibe 2008), as Quetta’s water supply relies on regulated tube wells; well as in Pakistan. Although constitutionally, water is the groundwater source is heavily depleted and largely a provincial matter in Pakistan, relevant policies, severe water shortages are common (Ahmed 2013). institutions, and legal provisions are distributed across Sanitation infrastructure is inadequate and poorly the national and provincial levels. National institutions maintained. About 100 kilometers of sewers cover coexist with, and sometimes overlap with, provincial a small fraction of the city, and there is a single institutions, and the legal framework for each province dysfunctional wastewater treatment facility. Most of includes its own laws and regulations overlain by Quetta’s wastewater and urban runoff is discharged into relevant national provisions. open drains, which are often clogged by solid waste and prone to overflow. There is evidence of agriculture The formation of Pakistan in 1947 severed the majority use of untreated wastewater in urban orchards, posing of the Indus Basin’s irrigated land (in Pakistan) from significant health hazards (Khalil and Kakar 2011). the waters that had supplied it (in India) (Briscoe and Qamar 2005), meaning existing interprovincial and state agreements, resource governance, and water Islamabad supply patterns required significant reworking. This Most of the water for the Federal Area (Islamabad) occurred over several decades to establish a formal of Pakistan is supplied from Khanpur and Simly dams mechanism to define Pakistan and Indian shares of and is treated before distribution at the Sangjani and the basin water resource, and to move toward reliable Simly water treatment plants, respectively (Shabbir nationally managed irrigation systems. The former and Ahmad 2016). In addition, 180 groundwater tube was achieved in 1960 with the signing of the Indus wells augment the supply (IUCN 2015). The distribution Waters Treaty, which allocates waters from the three system is in poor condition, with leaking pipes and gate eastern rivers of the Indus Basin (Ravi, Beas, and 55 Sutlej) to India, and waters of the western rivers a deputy commissioner, who is engaged in receiving (Jhelum, Chenab, and Indus) to Pakistan. and forwarding river flow and gauging data from India. During the monsoon, a flood cell operates in this office The hierarchy of water governance arrangements is 24 hours per day to receive flood information and share summarized in the following subsections. Pakistan’s with the Flood Forecasting Division of the Pakistan legal framework is particularly complex because of Meteorological Department (PMD). PCIW also receives the interplay between scattered early legislation with data and information on Indus waters from the Pakistan multiple amendments and a wave of mostly irrigation- Water and Power Development Authority (WAPDA) and focused provincial enactments over the past few from the provincial irrigation departments. decades. Appendix B provides full details and citations of the legal provisions relating to water. The description The office of PCIW is the only dedicated entity in of institutional arrangements covers government Pakistan for transboundary water governance. Section 9 and nongovernment institutions, and the processes of the 2018 National Water Policy (NWP) recognizes for citizen engagement in setting water policy and the need to work out a mechanism for the sharing holding public institutions accountable for policy of “of trans-boundary aquifers and joint watershed implementation. management including sharing of composite real- time flow information especially relating to hydro- International Transboundary meteorological disasters/disaster-like situations Water Governance endangering Pakistan’s important infrastructure, communication network and economy.” This At the international level, water governance is usually mechanism should extend beyond the geography of determined by international arrangements (including the Indus Waters Treaty to consider other transboundary declarations, treaties, minutes of ministerial meetings, rivers and tributaries that Pakistan shares with or institutional mechanisms) between countries sharing its neighbors. It would be appropriate to consider transboundary water systems. National implementation establishing a new federal body to support both the of international arrangements is usually supported technical and diplomatic aspects of transboundary by an assigned ministry or through coordination water management with all of Pakistan’s riparian between relevant ministries and government entities. neighbors, which would include the office of PCIW. Sometimes a dedicated national entity supports implementation. Federal Role and an Evolving For the waters of the Indus Basin, Pakistan’s Policy Framework relations with India are regulated by the 1960 Indus Waters Treaty. Coordination of transboundary basin Water resources are not included in the enumerated management (planning, decision making, and federal list of the 1973 Constitution of Pakistan; water monitoring) with China and Afghanistan is not formally management is largely, therefore, the purview of the defined. The Indus Waters Treaty established the provinces. The 18th Amendment to the Constitution, Permanent Indus Commission as a joint coordination enacted in 2010, moves issues of environmental mechanism. Its purpose and functions are to “establish pollution and ecology—both relevant for water and maintain co-operative arrangements for the management—from the list of concurrent matters implementation of the Treaty, to promote co-operation (for both federal and provincial jurisdiction) to the list between the Parties in the development of the waters of solely provincial matters. of the Rivers….” (Indus Waters Treaty 1960). Two areas fall within federal jurisdiction: interstate The Permanent Indus Commission comprises one water disputes and policy setting for water and power commissioner from Pakistan and one from India. development, as originally covered by the Water and It is responsible for the exchange of information, Power Development Authority Act (1958). Article 155 notification of planned development projects, and of the Constitution includes a dedicated procedure responses. It is required to meet at least once per year in case of water allocation disputes. Disputes may and undertake a general tour of inspection of the rivers be referred to the Council of Common Interests every five years. The treaty provides a mechanism for decision, and it is the legal duty of federal and for resolving questions, differences, and disputes. provincial governments to honor the council’s decision. Questions are handled by the commission; differences This provision was used for the approval of the 1991 are referred to a neutral expert, and disputes are Water Apportionment Accord. referred to an independent court of arbitration. National level water policy was limited in Pakistan prior Within Pakistan, the Pakistan Commissioner for Indus to the 1990s, although 1959 and 1970 land reforms Waters (PCIW) is supported by a team of advisers. and the 1977 creation of the Federal Flood Commission The data cell in the office of the PCIW is headed by (FFC) influenced water management (figure 4.9). 56 PAKISTAN: GETTING MORE FROM WATER Figure 4.9  Major Policy and Institutional Milestones before Partition and before and after the Indus Waters Treaty in Pakistan, 1940 to Present National Disaster Risk Management Framework National Water Policy WAPDA Water Vision 2025 Irrigation and drainge authorities National Drainage Program Devolution Plan First private water utility Indus River System National Climate Change Policy Land reforms WAPDA Authority Balochistan IWRM Policy Sindh Punjab Agreement Pakistan Vision 2030 Federal Flood Commission National Sanitation Policy Indus Water Treaty National Environmental Quality Standards Land reforms Water boards and National Environment Policy farmer organizations National Drinking Water Policy Council of Common Interests Partition Water Apportionment Accord Vision 2025 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 Note: IWRM = Integrated Water Resources Management; WAPDA = Pakistan Water and Power Development Authority. The land reforms were only partially completed because institutional reforms, and capacity building within of political challenges and changes in government. WAPDA; provincial authorities; and farmer organizations The FFC moved responsibility for flood protection from (World Bank 1998). In 2004, WAPDA’s Water Vision 2025 the provincial to the federal level; however, floodplain was published—Pakistan’s first long-term national plan for management remained a provincial responsibility. Legal water resources. The Water Vision focuses on mitigating tools to support floodplain planning and development the impacts of climate change on the water sector; in Pakistan are limited, although the Punjab Flood protecting agriculture from drought; replacing storage lost Plain Regulation Act (2016) provides government with to sedimentation; and developing new hydropower, all the power to declare certain areas as floodplains, and supported by US$33 billion of investment. then by notification in the Gazette, to prohibit new In 2005, the National Environment Policy was adopted construction in declared floodplain areas. to provide a national framework for addressing all The Water Apportionment Accord was signed in 1991, types of environmental issues, including impacts on and in 1992 the Indus River System Authority (IRSA) water. It promotes action for improved drinking water was established by federal legislation to implement and treatment provision including through low-cost the accord. Although the accord was a major step technologies, improved water quality and ecosystem flow in interprovincial water sharing, little progress has monitoring, water metering, artificial recharge, rainwater been made since in resolving important ambiguities, harvesting, metering, and the rehabilitation of water particularly with reference to the initial conditions. bodies (GoP 2005). It also calls for the development For example, Punjab maintains that the volumes of a National Disaster Risk Management Framework, apportioned in clause 2 are contingent upon additional published in 2007. The framework lays out cooperation storage becoming available, whereas Sindh considers between national and provincial governments for disaster clause 2 as the baseline volume and “shortages and management to redress Pakistan’s reactive response to surpluses” are dealt with appropriately in the accord. natural disasters (GoP 2007a). In 1993, Pakistan adopted the first iteration of its National Pakistan’s Vision 2030 (2007) is a general national Environmental Quality Standards, including standards for economic policy document that highlights the industrial and domestic effluent. In 1998, the National country’s water insecurity. It stresses the growth Drainage Program was established to improve saline of industrial and municipal water needs while irrigation drainage through infrastructure investments, recognizing the continued importance of irrigation. 57 It proposes changes to institutional arrangements The NWP lists 33 objectives under an IWRM umbrella for the ownership and pricing of water, new storages that span improved water allocation across competing to capture peak flows, and incentivizing water demands, ensuring water for social and economic saving technology in irrigation and pollution control development, and supporting national food security technology in industry. It also recognizes the need objectives. The policy is wide-ranging, covering supply to improve agricultural management to ensure and demand management, regulation, and sectoral food security and land sustainability (GoP 2007b). resilience. It is organized into seven strategic priority Implementation progress toward this vision over the areas. Conservation and efficiency are highlighted in last decade has, however, been very limited, and addition to the familiar calls for storage rehabilitation the more difficult reforms and larger investments and augmentation. Technology is prioritized, have not occurred. including for improved irrigation management and hydrometeorological monitoring and data sharing. The 2012 National Climate Change Policy (GoP 2012) Renewable energy, including hydropower and solar gives much attention to water in the context of pumping, is embraced, emphasizing the need for multiple climate risks including increased climate such developments to be planned with careful extremes, glacier retreat, increasing agricultural consideration of nexus issues. Pakistan’s regulatory water demands, and coastal saline intrusion. The framework is highlighted for strengthening, especially 2013 implementation framework for the policy sets coordination between federal and provincial out goals, strategies, and adaptation actions for governments and agencies, and the coordination of several aspects of water management and its nexus policy and improved implementation across government with agriculture and energy generation. It highlights and economic sectors. Planning principles to underpin rehabilitation and augmentation of storages and water management include sustainability, financial new options for water desalination and recycling, as fairness, and knowledge and innovation. Research is well as demand management programs to reduce needed both to drive technological innovation and for system losses and improve irrigation efficiency. IWRM improved decision making. The inadequacy of sector is used to oversee intersectoral demands and the financing is clear, as is the importance of financial interrelationships between water sources and uses, as sustainability in subsectors. There are many references well as improve stakeholder engagement. It proposes to environmental sustainability, including a call to ensure legislative changes to facilitate IWRM, including for environmental flows and the need for detailed action environmental protection and water management. plans to improve surface and groundwater quality. It promotes improved hydrometeorological monitoring and information exchange to improve forecasting and Governance for Key Aspects of water resource assessments. For irrigation, the policy focuses on water input technologies and other farm Water Resources Management practices and mechanization. Improved forestry and This subsection describes the legal frameworks and land use practices are encouraged, including to reduce institutional frameworks arrangements for national erosion and sedimentation and to improve groundwater and provincial water resources management, with recharge and protection. Hydropower expansion is reference to the policy context outlined previously. encouraged to reduce greenhouse gas emissions. Four key areas of water resources management are Although the implementation framework identifies considered: data, information, and analysis; resource time-bound targets and institutional responsibilities, planning and allocation; system operations; and progress over the last five years appears very limited. environmental sustainability. In the absence of robust implementation monitoring, however, it is difficult to accurately assess progress. The legal framework for water management combines remnant colonial legislation, the Pakistan Constitution, In 2014, the Pakistan Vision 2025 was released and a small number of federal acts focused on (GoP 2014), reiterating many of the water goals of establishing key national institutions, and more recent the earlier Vision 2030, including a focus on additional but often piecemeal provincial legal instruments that storage and water harvesting, investment in technology affect aspects of water management (figure 4.10). to improve water efficiency and pricing, improvement Relatively minor differences are observed in the legal of water allocation to reflect the economic value frameworks that apply to the different provinces. of water, and institutional mechanisms to manage Sindh has the most comprehensive framework, and sectoral and regional allocations. Vision 2025 stresses Balochistan, the least. However, many basic legal the need to achieve baseline levels of personal water provisions for supporting water resources management and sanitation access and social education on water. found in other countries are absent in Pakistan’s The proposed NWP was agreed to by federal and all provinces. A fuller description and comparative analysis provincial governments in 2018. of the legal frameworks are provided in appendix B. 58 PAKISTAN: GETTING MORE FROM WATER Figure 4.10  Major Federal and Provincial Legal Instruments before Partition and before and after the Indus Water Treaty in Pakistan, 1860 to Present Penal Code Constitution of Pakistan Easements Act WAPDA Act Environmental Protection Act Indus River System Authority Act Council of Research in Water Resources Act Groundwater Ordinance Farmer Organization Regulations Canal and Drainage Ordinance Local Government Act WUA Ordinance Irrigation and Drainage Authority Act WASA Act Environment Protection Act Canal and Drainage Act IWRM Ordinance Environmental Protection Act Rivers Protection Ordinance Local Government Act Irrigation and Drainage Authority Act Canal and Drainage Act WUA Ordinance Local Government Act Environmental Protection Act Irrigation and Drainage Authority Act AWB/Farmer Organization Rules Irrigation Act Water Management Ordinance Local Government Act Environmental Protection Act 1860 1880 1900 1920 1940 1960 1980 2000 2020 2040 National Balochistan Khyber Pakhtunkhwa Punjab Sindh Note: AWB = area water board; IWRM = Integrated Water Resources Management; WUA = water user association; WASA = water and sanitation agency. Data, Information, and Analysis Hydrological monitoring is undertaken by WAPDA at the federal level and by irrigation departments at Multiple institutions have water information and the provincial level, at least for the irrigation supply analysis roles, guided by clear policy directions from system. WAPDA operates hydrological gauging stations recent policy documents. However, the roles lack (for water levels and aspects of water quality) in the clarity at the national level and between national and Upper Indus Basin, at rim stations, and at key project provincial levels, and there are gaps and duplication sites. WAPDA maintains meteorological stations in in effort. Institutions often either lack clear legal the Upper Basin, including cryosphere monitoring mandates for these roles, or mandates are diffuse and to support inflow forecasting. WAPDA is upgrading spread across many institutions. and expanding its network. WAPDA holds significant water data records, but its data systems and data The office of the PCIW is responsible for data exchange management procedures are mostly outdated and with India. Although not established by separate inadequate. IRSA does not have a clear legal mandate legislation, PMD within the Aviation Division of for collecting water data, but is working to establish the Cabinet Secretariat has a leading role in water a modern, robust system for monitoring flows at information and modeling. PMD maintains much of 27 key sites across the irrigation water delivery and the national meteorological and agrometeorological distribution system. These will be supported by modern monitoring network and associated data systems. data management infrastructure and procedures. PMD provides weather forecasting, flood forecasting, glacial lake outburst floods (GLOF) and early warning The PCRWR has an important role in national water and drought monitoring services. The PMD Flood research. Established in 1964 and reorganized as a Forecasting Division provides flood, streamflow, and separate corporate body under the Ministry of Science dam water level forecasts. and Technology by legislation in 2007, PCRWR has 59 a research mandate for all matters related to water, water resource inventory and sets a mandatory including irrigation, drainage, reclamation, navigation, timeline for periodic updating of this inventory. flooding, drinking water, industrial water, and sewage. It requires the creation of a publicly accessible water The establishing legislation tasks PCRWR with developing user registry and requires the government to monitor and maintaining a national water resources database water resources and publish monitoring results. The for use by planning and implementing agencies and Punjab legal framework is less comprehensive because the public, as well as initiating a national water quality it does not set a timeline for a periodic update of a monitoring program and advising government on water resource inventory and does not require the water quality and the development, management, government to publish the results of water resources conservation, and utilization of water. PCRWR has monitoring. The Balochistan legal framework is weaker established a water quality testing program but has not still, because it does not require either the water established a national water resources database. resource inventory or the water user registry to be FFC has an expansive institutional mandate for flood publicly accessible, although it does require monitoring information and modeling. Its functions include results to be published. The legal framework of KP development of flood forecasting and warning systems, mirrors Balochistan except it does not require water flood research, standardizing and recommending flood resources monitoring to be published. infrastructure designs, and evaluating progress under Apart from the PCRWR Act, legal mandates for water flood protection plans (GoP 2018). information and analysis are missing at the federal level At the provincial level, several institutions have and are variable but generally weak at the provincial mandates relevant to water information and analysis. level. The important water data and information Provincial government agencies collect data on functions of PMD lack a clear legal mandate. In many irrigation water distribution with differing efficacy, other countries the legal frameworks for water are and some efforts are being made in water data more comprehensive, requiring creation of a periodically management. The Punjab Irrigation Department updated water resource monitoring plan and the creation maintains a digital database of flows in each canal, of a pollution discharge information system. Water and the Sindh Irrigation Department is making flow conditions commonly change over time, and water data and information for irrigation canals publicly information is most useful if up-to-date and accessible. available. The Punjab Irrigation Department is exploring Clear legal mandates for these functions can help ensure automated or digital data acquisition as the next consistent institutional action over time. iteration of its data and information platform. Pakistan does not promote the role of citizens The NWP notes the need for improved water information in collecting and analyzing water data. Citizen and analysis for “improved asset management and to involvement in the planning, installation, and derive evidence- and data-driven decision making.” It management of hydrometeorological or water quality promotes “research on water resources issues of national monitoring networks could help improve water data importance and building capacity/delineating roles and community knowledge, as demonstrated in other and responsibilities of federal research institutions and parts of the world (Paul et al. 2018; Zemadim et al. promoting coordination among them.” The policy has 2013). For Pakistan, this could be especially powerful detailed sections on water information management for improved groundwater management. and water research, and highlights the need for a national water resources database and a national Resource Planning and Allocation water research agenda. The Climate Change Policy also At the national level, WAPDA has an institutional stresses improved information and analysis. It promotes responsibility for strategic resource planning, and set improved data for irrigation water use, remote sensing national directions in 2004 in its Water Vision 2025. of agricultural systems, and real-time meteorological Water Vision 2025 reflects WAPDA’s historical supply- information collection and exchange. It advocates for side development focus, within limited consideration of an increase in research, including of water resources basin-scale environmental sustainability, interprovincial and agricultural resilience, supported by enhanced sharing, or economic productivity. By approving the Water monitoring, with a view to improving forecasting of Apportionment Accord, the Council of Common Interests seasonal water availability. has adopted a procedural role in water resource planning. Pakistan’s legal frameworks contain elements to Within the framework established by the accord, IRSA has support water information and analysis. The PCRWR the primary responsibility for resource allocation guided Act (2007) outlines a clear national mandate for water by information from WAPDA on resource availability research and analysis. The Sindh legal framework and information from provincial irrigation departments requires the creation of a publicly accessible on irrigation demands. With growing but changing 60 PAKISTAN: GETTING MORE FROM WATER water demands, growing environmental concerns, promote women’s participation in water resources and increasing climate change driven flow variability, management institutions. Managing water resources a more robust allocation process and clear institutional typically requires managing how people interact with responsibilities for long-term strategic resource planning the resource, but water conditions, water uses, and are required. Within the provinces, water allocation is objectives all change through time, and a plan is useful the responsibility of provincial irrigation and drainage only if current. authorities (PIDAs), which issue and revoke water licenses and settle water allocation disputes. Pakistan has some legal foundations that could support the establishment of a modern water rights or permit The NWP outlines six principles to guide resource system for water allocation, but there are significant planning processes at federal and provincial levels: gaps. In Sindh, a permit or right is required before (i) equity and participatory decision making; (ii) water abstracting water, and in Balochistan, public notice of is a strategic resource and access to affordable and new water abstraction permit or right applications is safe drinking water is a fundamental human right; required before a decision is made. In KP and Punjab, the (iii) efficiency and conservation; (iv) environmental length of this public notice period is legally defined. In sustainability; (v) practicability and innovation; and other countries, legal frameworks sometimes go further, (vi) command area development is the responsibility requiring the establishment of a priority order for water of farmers with government support for small land allocation between types of water uses, prescribing the holdings. It emphasizes basin-level water resource procedure to acquire a new water abstraction permit or planning including improved allocation across right, setting a duration for water abstraction permits multiple sectors. Allocation changes are intimated for or rights, providing a shorter or simpler procedure for enhancing food security and climate adaptation and water abstraction permit or right renewals, defining resilience. IWRM principles are highlighted, especially the length of a public notice period prior to a decision public participation, enhancement of public-private on a new water allocation permit or right application, partnerships (PPPs), and improved institutional and setting out required means of giving public notice capacity. Priorities for institutional strengthening include of new water abstraction permit or right applications improved implementation of the accord, coordination before a decision is made. Priority orders can help to between provincial and national planning, and inclusion rationalize daily allocation decisions, aligning them of hydropower development in resource planning. with broader objectives for the water sector. Specific The National Climate Change Policy outlines key procedures help to increase clarity and legal certainty for water resource planning concepts including new and water users, so permit procedures don’t become a barrier rehabilitated storages, hydropower development, and to development. Clear permit durations help provide contingency planning for water shortages. investment security for water users and provide a defined The WAPDA Act (1958) provides legal support for federal period for water managers to periodically reevaluate water resource planning, although its provisions are very water allocations. While permit renewals provide an focused on water resource development. The legislation important review point for water managers, they should requires WAPDA to prepare “a comprehensive plan for not be so frequent as to create undue uncertainty. A clear the development and utilization of the water and power notice period helps to ensure a predictable, inclusive resources of Pakistan on a unified and multi-purpose process that meets minimum expectations for providing basis.” The provincial legal frameworks contain a few an opportunity to comment. A clear mechanism for public features that support inclusive planning. The Sindh notice helps to ensure a predictable and inclusive process. and KP legal frameworks require the creation of water resources management plans and require water users System Operations to be represented in water resources management The key institutions with operational functions for water institutions. Punjab establishes a mechanism to promote management at the federal level are WAPDA and IRSA. women’s participation in water resources management WAPDA operates the major headwater reservoirs and institutions, and Balochistan specifies the required hydropower facilities for water supply, flood mitigation, components of water resources management plans. and power generation. In operating headwater In other countries, however, legal frameworks are often reservoirs for flood mitigation, WAPDA is guided by more comprehensive and require public consultation flood forecasting by PMD, as well as their own modeling in the development of water resources management and analysis. IRSA specifies and reviews river and plans, mandatory timelines for periodic updates, water reservoir operations and communicates these to WAPDA allocation decisions consistent with water resource and provincial irrigation departments. IRSA’s role in management plans, the establishment or adoption of specifying operations is based on the compilation and water resource quality criteria, water quality objectives review of rolling provincial irrigation demand estimates for water bodies, and quotas or other mechanism to to determine required reservoir releases. 61 Historically, provincial irrigation departments were region in a province.” IRSA’s establishing legislation responsible for the operation and maintenance gives them specific legal mandates to “specify river (O&M) of barrages and irrigation canals for delivery and reservoir operation patterns” and to “issue of irrigation water services and flood mitigation. In consolidated operational directives to Water and Power 1997, as a part of broad irrigation reform, PIDAs were Development Authority for making such releases from established by legislation to manage and distribute reservoirs as the Authority may consider appropriate.” canal water, and to oversee the O&M of the main The legislation that established PIDAs, AWBs, and distributaries in association with area water boards farmer organizations focuses on administrative aspects (AWBs). AWBs devolve irrigation system management of the institutional setup, rather than prescribing legal to stakeholder-based institutions, and, thus, PIDAs mandates or operational powers and responsibilities. were intended to oversee irrigation and drainage systems within a largely decentralized governance Environmental Sustainability architecture. This is supposed to make provincial irrigation departments more agile and less burdened The primary responsibility for environmental with O&M tasks, and to improve equity in irrigation management rests with the federal and provincial service delivery. However, PIDAs typically operate in environmental protection agencies (EPAs). The parallel with the irrigation departments, increasing the federal EPA was established under the Pakistan complexity of water governance without increasing its Environmental Protection Act (1997) to enforce the effectiveness (WWF 2012). environmental rules and regulations contained in the act; conduct environmental impact assessments Below the level of AWBs, farmer organizations of and initial environmental examinations; establish elected farmers from the Khal Panchayats manage the National Environmental Quality Standards; and local supplies, maintain on-farm distributaries, collect promote environmental research. With the 18th abiana, and make payments to AWBs. Khal Panchayats Amendment to the Constitution, environment issues are watercourse-level water user associations created have become provincial responsibilities, each of which in conformity with irrigation management transfer has an independent EPA, but weak enforcement policies of the 1990s to promote the devolution of capacity undermines effective environmental authority and costs of irrigation systems to beneficiary- management. The regulatory responsibility of the led groups (Mekonnen et al. 2015). The need to Pakistan EPA now applies only to the Islamabad improve irrigation management was a key argument Capital Territory. For major federal water and power for the creation of water user associations (WUAs); projects, the environmental cell within WAPDA however, it is difficult to quantify actual improvements conducts environmental impact assessments, (World Bank 1994). Evidence suggests that on-farm monitors environmental impacts during construction water use efficiency is higher for farmers belonging and operation, and implements environmental to farmer organizations (Chaudhry 2018), but equity management plans. PCRWR researches water and is not necessarily improved in command areas agricultural environmental issues. managed by farmer organizations compared to those managed bureaucratically (Jacoby, Mansuri, and Fatima Pakistan has national and provincial environmental et al. 2018). This suggests that broader institutional policies and regulations for environmental factors, such as community characteristics and social management and pollution control. However, interactions, influence community-based water implementation is slow and incomplete, and regulatory governance and related improvements in water use enforcement is inadequate. Several NWP objectives efficiency (Chaudhry 2018). address environmental sustainability, including watershed management and restoring and maintaining The NWP has little to say about the operational the health of water-dependent ecosystems (including functions of WAPDA and IRSA, except to note the Ramsar and other wetland sites). Groundwater need to revitalize WAPDA and strengthen IRSA’s regulation is advocated to curb overabstraction and role in real-time monitoring. The policy highlights the enhance recharge, as well as to help prevent seawater importance of financial sustainability for provincial intrusion into coastal aquifers. Environmental flows irrigation operations and the role of technology are highlighted, and renewable energy is promoted. to improve operational efficiency and effectiveness. These objectives however, are unsupported by specific Pakistan’s legal frameworks include elements that or strong regulatory measures, which will hamper help guide federal and provincial water operations. implementation. Environmental sustainability features The Water and Power Development Authority Act strongly in the National Climate Change Policy, (1958) gives WAPDA the legal mandate for “control particularly the issue of environmental and resource over waters, power houses and grids,” including resilience. The policy stresses the importance of control over “underground water resources of any protecting environmental resources, including rational 62 PAKISTAN: GETTING MORE FROM WATER groundwater use, wetland and watershed protection, with PHEDs and urban services in the large cities are and enhanced environmental flows. The National Food delivered by WASAs. Security Policy notes the threats posed by resource Provincial planning and financing frameworks are degradation and key soil-water interactions, including relatively well developed, but planning is hampered soil-water retention, water pollution, and reservoir by inadequate data, lack of institutional cohesion, and sedimentation. Water pollution with agrochemicals is the absence of an independent regulator. While broad highlighted for attention to improve the environmental service goals and targets have been defined, there sustainability of the food system. is no planning process and no agreed timeframe for All four provinces recently enacted environmental meeting the agreed targets. The virtual absence of protection acts, which provide general frameworks regulation, the inability to raise tariffs to recover costs, for environmental sustainability, including for water and poor cost recoveries, force municipal entities to rely management. However, provisions to support heavily on large annual subsidies that are increasingly managing water pollution and water depletion are difficult to sustain. Sector monitoring is weak, with limited, and the new legislation is yet to be supported a lack of definitional consistency, clear targets, and with any detailed regulations. Water resources unified data sources. Standardized monitoring has protection is better supported in KP given provisions in been discussed for some time, but it has yet to be its Rivers Protection Ordinance (2002) and Integrated established at national or provincial levels. In the face Water Resources Management Board Ordinance (2002). of deteriorating service delivery, recent judicial inquiries Sindh has measures for managing water shortages in by apex courts in Sindh and Punjab have demonstrated its Water Management Ordinance (2002); however, the political will to enforce the basic constitutional right legal provisions for water quality management are to safe drinking water. lacking in all the provinces. Provincial approaches to water supply and sanitation governance differ, but most are characterized by Governance of Water Supply residual policy and institutional overlaps and unclear legal mandates. In Punjab, there is a lack of role and Sanitation delineation between the LGD, the Urban Unit, and In 2006 the federal Ministry of Environment published PHED. In Sindh, multiple policies for drinking water and a National Sanitation Policy focused on driving behavior for sanitation and solid waste have been produced change and ensuring safe waste disposal and universal by multiple departments without implementation. access to basic sanitation. This was followed by a Policy overlaps lead multiple agencies to seek to National Drinking Water Policy in 2009, focused on establish mandates to obtain additional resources. improving water access, treatment, and conservation Policy implementation varies according to departmental through enhanced community participation and public priorities, capacity, and operational norms, creating awareness, cost-effective infrastructure, research further confusion and conflict. The situation is often and development, and PPPs. Following the 18th exacerbated by political masters selectively delegating Amendment to the Constitution, however, water supply responsibilities and by the provision of donor funds to and sanitation responsibilities—including legislation, institutions lacking clear legal mandates. policy, planning, and service provision—moved fully to provincial governments. The earlier national policies Political Economy Challenges provide general guidance to the provinces, which have each developed policy frameworks. Informal governance—the political economy— significantly influences the water sector. The political Despite policy progress, provincial institutions have economy of water in Pakistan is discussed in terms of largely resisted reforms because of entrenched and the evolution of irrigation governance and Karachi’s contested interests, amplified by a lack of capacity. urban water sector. Pakistan faces serious irrigation and The policy frameworks do not adequately separate urban water governance challenges, and future reform institutional roles for water supply, asset ownership or progress will require tackling difficult political economy management, and service delivery, and the absence of issues based on an understanding of where and why an independent regulator further undermines progress. past reform efforts have failed. Service delivery is spread across many institutions with varying capacities, differing reporting lines, and limited Political Economy of Irrigation Governance coordination. Relevant institutions include public health engineering departments (PHEDs), local government The foundations of irrigation in Pakistan date back to departments (LGDs), and water and sanitation agencies the late 19th century when the British government of (WASAs). Although LGDs have broad service delivery the Indian subcontinent began construction of the canal responsibility, rural service delivery remains de facto network and passed the Canal and Drainage Act 1873). 63 Despite political, economic, and demographic change— tariff nonpayment has been removed, eliminating the and continued infrastructure investment—there has most powerful means for tariff enforcement. Water been continuity across a century and half in the public rights remain coupled to land ownership, preventing administration of irrigation by state and provincial water trading or any formal water market. The Canal governments, which adopted the colonial legislation and Drainage Act (1873) remains in force, retaining (with some amendments) as provincial acts. elements of the old centralized governance model. Although PIDA powers and responsibilities are clearly Scrutiny of irrigation performance (chapter 5) highlights described in the act, those for AWBs and farmer the lack of financial sustainability of irrigation and its organizations are vague, undermining decentralization. reliance on subsidies (Strosser 1997), poor performance of hydraulic infrastructure (Rinaudo and Tahir 2003), Some of these seemingly regressive revisions reflect low agricultural water productivity, widespread a degree of pragmatism, given the sheer scale and rent-seeking and corruption (Jacoby and Mansuri complexity of modernizing the low-technology, supply- 2018; Jacoby, Mansuri, and Fatima 2018), and poor driven irrigation system, which would be required to administration that has enabled illegal water trading enable these reforms. However, other revisions reflect and theft (Mustafa et al. 2017; Rinaudo, Strosser, and entrenched interests that successfully reframed the Thoyer 2000; Rinaudo and Tahir 2003;). Pressure reform agenda during consultations on the draft PIDA since the early 1990s to transform irrigation from a Act and the public debate by experts and opinion centralized bureaucracy to devolved, inclusive, service- leaders. Privatization, although not central to the oriented management was partly triggered by reforms reforms, was emphasized and characterized as a push advocated by the World Bank (1994). These reform for foreign control of Pakistan irrigation. The proposal proposals became a component of a World Bank loan to delink water rights from land rights was portrayed that financed the National Drainage Program (Rinaudo as land reform, which was sensitive given reform and Tahir 2003); it has three core pillars: efforts of the 1970s that failed partly because of the dominance of large landowners in federal parliament, • Restructuring PIDAs into decentralized public utilities as well as a verdict against land reform by the at the command area level, with the autonomy to Supreme Court. collect and spend water tariffs, enabling progressive withdrawal of subsidies and, potentially, eventual By 1996, the draft legislation, by then widely seen privatization. The government renamed these as a donor-driven attempt at privatization and utilities as AWBs and proposed PIDAs as regulators. land reform, was strongly rejected by nearly all The government neither explicitly ruled out stakeholders, including the Pakistan Kissan Board privatization nor overtly embraced the concept. (a small farmer lobby group), which had originally • Farmer-led management at the distributary been strongly supportive. Primarily large landowners level, including water tariff collection and and provincial irrigation departments would have had expenditure decisions. This was accepted by to concede power. Van der Velde and Tirmizi (2004) government in spirit and was pursued via farmer identify important overlaps in political, professional, organizations and WUAs. and informal authority positions of key individuals • Establishment of water markets, and, potentially, who reframed the reform discourse. These include water trading, which includes delinking water rights overlaps between large landowners and politicians, from land ownership. Again, this has not been ruled and between irrigation departments and irrigation out by government, but has not been explicitly engineering consulting firms. One large landowner endorsed. was both the head of a powerful lobby group and a The reform effort that began in 1994 culminated in the high-ranking state official. A former senior official of Provincial Irrigation and Drainage Act (1997). However, a provincial Irrigation Department also held a stake this legislation is markedly different in both spirit and in an engineering firm delivering irrigation projects. vision from what was originally proposed. It is far less In addition to individuals holding multiple positions effective at transforming irrigation to an equitable, with conflicting interests, there was collusion between sustainable, and participatory management model for large landowners, politicians, and irrigation department several reasons. Farmers have absolute majority in the officials, which negatively impacts system performance executive committees and general assemblies of PIDAs and small farmer welfare (e.g., Ali 2015; Gazdar 2009; and AWBs and have the final say on tariff increases. Hussain 2008; Malik 2008). The responsibility for water pricing has been transferred It is less clear why small farmers who stood to gain from provincial governments to PIDAs, meaning from these reforms objected. Most commentators pricing innovation critical for financial sustainability blame the absence of a government-framed narrative is unlikely given the control of PIDAs by large land- explaining the need for reform, which created an holding farmers. Supply disconnection as a penalty for information vacuum filled by misinformation. Central 64 PAKISTAN: GETTING MORE FROM WATER government was ineffective at communicating to ownership (Ali 1988; Gilmartin 2015). The colonial small farmers how they would be empowered by Punjab Irrigation Department introduced chakbandi— management decentralization. Instead, small farmers the assignment of fixed areas (chaks) around were swayed by local irrigation department officials remodeled water channels—that made access to water and patwaris (revenue officers), supported by a media contingent upon access to land. Once landholding size narrative dominated by the rural elite and large farmer became the determinant of an individual’s “water lobbies (Rinaudo and Tahir 2003). Van der Velde and right,” any market for water trading independent of Tirmizi (2004) cite “Privatization of Canal System to Be land was impossible, and inevitably land inequity Disastrous for Economy”—a 1996 article in the English fostered water inequity (Mustafa et al. 2017). newspaper The Muslim—for illustration. The following PIDs also claimed the reforms were ill-designed excerpt is a statement attributed to spokespersons of because system decline was so pervasive that even the Farmers Association of Pakistan and the Pakistan the technically competent and legally empowered Engineering Congress (p233): PIDs were struggling to operate and maintain it. What “What was being proposed was not even genuine chance would poor uneducated farmers have? Yet privatization. The plan is to sell irrigation channels PIDs’ struggles were, of course, key drivers for the to big landlords under the umbrella of this newly reforms. Suggesting that farmers were not competent created PIDA. It is a diabolical scheme against the to manage their own inputs and assets also implicitly rural masses and our agricultural economy. The undermined the existing warabandi system of water proposed law denies water rights to poor farmers distribution proportional to land holding, although by changing entitlement of irrigated lands to never led to any call to reform warabandi. water by making canal irrigation water freely and More concrete reasons underlay the defensive independently tradable to land owners with money stance of PIDs in the reform debate yet were and under their own authority.” scarcely discussed. First, devolution of irrigation Urdu dailies carried similar claims that the government management and O&M to farmers would have was handing over Pakistan’s canals and water resources made thousands of irrigation staff redundant. to external financers, while senior government officials There was also a fear that the composition of PID continued promising to prevent “foreign privatization” management—overwhelmingly from engineering of Pakistan’s water resources. The World Bank advised backgrounds—would be diluted by recruitment of the government in late 1995 that decentralization was staff trained in management sciences, economics, to facilitate participatory management and did not and social sciences, opening the door to private entail privatization. Nonetheless, negative framing of sector management consultants (van der Velde and the reforms contributed to their dilution and reinforced Tirmizi 2004). Second, decentralization would reduce other concerns. In mid-1996, in a joint meeting opportunities for rent-seeking by irrigation staff. with the president and prime minister of Pakistan, PIDs acknowledged the decrepit state of the irrigation both the Pakistan Kissan Board and the Farmers system they were tasked to manage, but neither Association of Pakistan rejected the draft act on two accepted responsibility for the situation, nor felt that grounds: first, because of inadequate representation institutional transformational was required. Van der of farmers in PIDA and AWB executive committees; Velde and Tirmizi (2004) conclude (p226): “Although and second, because of insufficient accountability to many PID functionaries were willing to concede ensure increased tariffs would be invested in irrigation that by the 1990s departmental discipline had (Rinaudo and Tahir 2003). In hindsight, these were deteriorated to levels that adversely impacted upon well-founded concerns. canal system O&M, few were willing to acknowledge any institutional responsibility for that condition.” Despite the wider stakeholder pushback, it was ultimately provincial irrigation departments that The PIDA Act reduces control of the irrigation prevented full implementation of the reform agenda. bureaucracy; however, the concessions mostly favor Ostensibly, the rejection of the draft legislation by the large farmers, not small or tenured farmers, and PIDs was to protect small farmers, who, it was claimed, the concessions are not conducive to broadening would struggle to compete in water markets given the expertise guiding irrigation performance their poor education, impoverishment, and traditional improvement. While the farmer lobbies eventually farming methods. This narrative helped convince small fully accepted the reforms, reluctance from PIDs farmers to join the protest. However, it overlooked persisted. The absence of a comprehensive and legislative and institutional developments in the late conducive legal framework, lack of cooperation and 19th century, which created or reinforced many of ownership of the partial institutional reforms by these inequalities and factionalisms, that were key irrigation departments, and rent-seeking behavior drivers of the reforms, especially those linked to land by irrigation officers have made the introduction 65 of participatory irrigation management extremely cannot enforce the accountability envisioned by the difficult and undermines WUAs and the larger act. Legislative amendments are required to define irrigation reform process in Sindh. This has been and establish the powers and responsibilities of illustrated by Starkloff’s (2001) review of the failure AWBs and farmer organizations. of a 1995–97 participatory irrigation management • Conflicts of interest of individuals engaged in higher- pilot, as well as many reflections on the AWB and level decision making on irrigation reform must be farmer organization pilots over the last two decades. avoided or identified and properly managed. The popular view is one of some, but limited, success • Overlaps and contradictions between the remits of of participatory irrigation management, but despite— the PIDs and PIDAs, which arise from dual legislation, rather than because of—the PIDA Act. Mustafa et al. must be resolved for role clarity and efficiency. (2017) (p33–34) assert that “implementation of • New technological opportunities for data-driven participatory water reforms reflects the deeper operational management will likely an important structural problems that persist within the Pakistani foundation for improved irrigation governance. water bureaucracy” and confirm the reluctance of PID staff to facilitate farmer-based management. Political Economy of Karachi The 1990s reform effort was focused mainly on Water Services governance and economic performance, paying less attention to infrastructure modernization. Two decades The political economy of Karachi’s water services is on, many of the governance and economic challenges a relevant case study due to the size and economic remain pertinent, but there is a new opportunity to invest importance of Pakistan’s largest city, and because many in improved hydraulic infrastructure at all levels of water of its local issues are emblematic for the challenges distribution, and to improve performance through data- of the country’s urban water sector. In Karachi, only driven control systems. Improved data and information about 55 percent of water demand is being met, and around water allocations and distribution would be NRW is estimated to be 58 percent. Only 25 percent critical to the establishment and effective operation of industrial and commercial customers have metered of a formal water trading system. Such a system may supply, and there is no metering for retail customers. not be fully devolved, but would require clear roles An average tariff of only US$0.13 per cubic meter and and accountability across multiple levels of irrigation a collection efficiency below 50 percent contribute governance. The principle of subsidiary is pertinent to the lack of cost recovery by the Karachi Water and here, and some increased decentralization is likely to Sewerage Board (KWSB) (World Bank 2018). be critical for reducing operational costs to improve Until recently, water theft was widespread: illegal financial sustainability. Moving forward with the irrigation water hydrants far outnumbered legal sources, and reforms outlined in the NWP will require recognizing the water was stolen from Keenjhar Lake and the Hub following lessons from earlier reform efforts: Dam before it even reached the city (Felbab-Brown • Irrigation reform requires genuine ownership 2017). Water theft supported illegal tanker operators, and clear willingness to share power by those in delivering water valued at over US$500 million per positions of authority. year, with those providing assistance and protection • Irrigation reform requires clear messaging on the within and outside KWSB also benefitting (Hashim purpose and process, agreed among federal and 2017). In 2017 however, a Supreme Court order led provincial governments, donors, development to the closing down of illegal hydrants, leaving just partners, and sector specialists. Social media can six government-approved hydrants that now provide quickly propagate misleading narratives, which can metered water through legal tankers. be hard to counter. Although rather dated now, the most recent • The extent and success of farmer participation will comprehensive assessment of water infrastructure and depend on farmers’ legal empowerment and the management needs for Karachi (JICA and KWSB 2008) degree of their integration into multilayered irrigation governance. Handing over watercourse management proposed a master plan to tackle wastage, theft, and is unlikely to succeed without early farmer participation NRW. Ten years later, the operational and financial and adequate capacity building for financial sustainability of water services remains elusive. There management, conflict resolution, and O&M—together is broad agreement on technical and financial solutions with initial external financial and technical support. (infrastructure rehabilitation, comprehensive metering • Rural land-based inequality and power asymmetry for retail supply, reducing NRW, and transforming KWSB persists and precludes small farmers acting against into a modern efficient utility) and consensus that the interests of large landowners or kinship-based Karachi’s water problems are a problem of governance authorities. The PIDA Act’s lack of clarity on AWB and and not water resource scarcity (ADB 2007; Mansuri farmer organization powers means these institutions et al. 2018; SBP 2017). 66 PAKISTAN: GETTING MORE FROM WATER Consumer and utility issues are the key political behavior. This has had a reputational cost for KWSB, economy factors that contribute to poor service delivery discouraging talented and hardworking professionals in Karachi. Consumer issues center on a lack of trust from joining the utility, further constraining and dissatisfaction with service quality, and hence a performance improvement. low willingness to pay and a readiness to use informal The financial dependence of Pakistan’s water utilities sources of water. JICA and KWSB (2008) report that only on provincial governments limits their autonomy 30 percent of water users across Karachi (and none (SBP 2017). KWSB relies on direct subsidies from the in the katchi abadis) trust KWSB. An estimated half of Sindh government as well as federal funds for payments registered KWSB customers pay their bills (World Bank to Karachi Electric, infrastructure expansion, and debt 2018), and many Karachi residents are not registered servicing (World Bank 2018). In line with the Sindh Local as customers at all. Utility issues, including inefficient Government Act 2013, the Sindh government retains administration, political interference, and corruption, influence by “approval of budgets, regulations and aggravate financial unsustainability and thus subsidy tariffs, hiring and postings, and provision or facilitation of dependency. The resulting lack of autonomy constrains locally mobilized funds or foreign loans or grants” (World the utility’s scope for reform and investments and thus Bank 2018). State and local control overlap, however, its ability to improve services and increase trust—a because the Karachi Metropolitan Corporation (KMC) vicious cycle. (of which KWSB was part prior to 1996) is represented For retail consumers, the distinction between ability and on the utility’s board. The incomplete devolution and willingness to pay is important. JICA and KWSB (2008) ambiguous institutional responsibilities have caused conclude that only the latter was a constraint, both for tension between provincial and municipal governments domestic and retail customers. Attempts to increase in Karachi, which has had a destabilizing influence on the cost recovery by offering concessions to bill defaulters governance of the urban water sector. were unsuccessful. Beyond poor water supply, a major These structural problems have aggravated financial reason for nonpayment was simply never receiving a and human resource management challenges at KWSB. bill, reflecting an inefficient billing and tariff collection In the absence of comprehensive (provincial and local) system. Briscoe and Qamar (2005) note that residents water supply and sanitation policy, financial support of poorer localities lacking access to reliable piped to KWSB has been ad hoc and most often directed supply pay more to tanker operators and informal toward relieving immediate financial constraints vendors than they would for piped supply, and that (e.g., payments for electricity) or financing urgent people would be willing to pay higher tariffs if there pump station repairs. Politically motivated hiring of were commensurate improvements in service delivery. staff—enabled by loopholes in the KWSB Employees’ Global experience—for example from Phnom Penh, Rules (1987) and the KWSB Act (1996)—has been Johannesburg, and Manila—indicate that financially widely reported. In the 1980s and 1990s, thousands sustainable water service delivery is possible provided of employees were hired based on political and ethnic incentives exist for service providers to improve service affiliations, laying the foundation for “ghost employees” quality (World Bank 2005). Improving cost recovery will at KWSB. Periodic acknowledgement of this problem require additional efforts to ensure payment of water has triggered mass employment terminations. However, bills by provincial government institutions and senior overstaffing remains a major issue, with 6.5 KWSB officials and politicians, which figure prominently in employees per 1,000 connections—more than three published lists of bill defaulters and an 2012/13 report times the benchmark staffing ratio for low-income from the Auditor General of Pakistan. countries (LICs). With around 13,500 employees, On the utility side, ADB (2007) identifies corruption salaries, benefits, and electricity charges represent over as the most notable governance issue in the urban 90 percent of KWSB expenditure (World Bank 2018). water sector and concludes there are inadequate Poor service, inefficient cost recovery, corruption, incentives for utility staff to implement technical and political interference on the one hand—and low solutions that reduce corruption. Mustafa et al. consumer trust and willingness to pay on the other— (2017) identify an insufficient revenue base and have led to a decline in the quality and coverage corruption amid lower-level staff among KWSB’s of urban water services in Karachi, with access to most pressing problems. Corruption is a consequence improved water sources falling from 90 percent to and a cause of KWSB’s financial problems and 86 percent over the past decade (World Bank 2018). constitutes a structural problem rather than just one of individual morality. JICA and KWSB (2008) There is no single simple entry point for water highlight low morale and a lack of motivation and governance reform in Karachi, given the interwoven enthusiasm among utility workers, resulting from and cascading causes and effects. However, a critical top-down imposition of rules and regulations that starting point is to address the structural deficiencies neither recognize good performance nor punish bad that enable political interference and undermine utility 67 autonomy. The KWSB Act (1996), that was supposed to corporations, and some special programs. Historically, ensure autonomy, has been managed by the provincial the relevant water sector ministerial division is the government through chairing the utility’s board. The water and power division. Provincial governments ambiguous and outdated legal framework should be use Annual Development Programs (ADPs) to allocate updated to support autonomous water management financial resources to support development visions. and regulation. This requires rationalizing the overlaps In addition to receiving a share of federal government in the provincial policy frameworks and aligning revenue, the water sector is financed by urban service these with Local Government Act (2013), as well as tariffs, irrigation tariffs, private sector investment, establishing a legally empowered independent sector and donor contributions. Urban service tariffs cover regulator for service providers. only 16 percent of the cost of urban water supply and sanitation services. Water supply and sanitation Financing budgets are not correlated to need or poverty level, and the largest share of available finance goes to Federal government revenue in Pakistan is mainly provincial capitals, rather than the rural poor. Irrigation from income tax, general sales tax, wealth tax, capital tariffs fund a small fraction of irrigation O&M costs. gains tax, and custom duties. The Pakistan National Finance Commission divides this revenue into federal At the federal level, allocations to the water sector and provincial shares. The provincial share is distributed have averaged 11 percent over the last 18 years and between provinces primarily based on population. Total only 4 percent over the last three years (figure 4.11). revenue has more than doubled in the last decade, In 2018–19, the total allocation to PSDP was around and the provincial share has increased from around PRe 2,000 billion. Given the significant increases in the 20 percent to more than 30 percent. total government revenue, the absolute allocations to the water sector in recent years are similar to the Expenditure is dominated by debt servicing (around long-term average (figure 4.11). At the provincial 40 percent) and other nondevelopment expenditure level, the allocations to the water sector have more (around 40 percent; half of which is defense spending). than doubled over the last five years (figure 4.11). Development expenditure—through the Public-Sector Development Program (PSDP)—is thus around one- To assess water sector financing, government fifth of the total budget and allocated approximately investment and expenditure are compared to equally at the federal and provincial levels. The PSDP investment recommendations of the Water Sector is the government’s primary mechanism for directing Task Force (FoDP 2012), which set priorities and public sector resources to development goals and describe interventions for a four-year period across five targets. Annual federal PSDP plans indicate the action areas: (i) major infrastructure and associated financial allocations for individual development institutions; (ii) raising agricultural productivity; projects, grouped under 41 ministerial divisions, two (iii) living better with floods; (iv) sustainable urban Figure 4.11  Federal and Provincial Government Water Sector Funding Allocations and Percentage of Total Federal Budget in Pakistan, 2000–17 120 100 80 60 40 20 0 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Share of federal budget (%) Federal allocations (PRe, billions) Provincial allocations (PRe, billions) Source: GoP 2018. 68 PAKISTAN: GETTING MORE FROM WATER services; and (v) knowledge management. For each actual expenditure has been only a little over half the action area multiple subactions are recommended, recommended level (table 4.4). with indicative costs and timelines. A summary is presented in appendix C (table C.1). A detailed analysis For the action area “major infrastructure and associated of the federal PSDPs and provincial ADPs has been institutions,” FoDP (2012) recommends investment in undertaken to assess allocations to, and expenditure large storage dams, rehabilitation of water infrastructure, against, priority projects that align to the Water Sector and strengthening institutional capacity, especially Task Force recommendations. regarding IRSA. These investments represent 83 percent of the total recommended by FoDP (2012). A major Sector financing over the four-year period 2013–17 proportion of the recommended investment was to was a little more than US$6 billion—one-fifth of the rehabilitate three barrages and construct new large dams. level of financing recommended by FoDP (2012). The Remaining funds were for IRSA reforms, developing underperformance is a mixture of undercommitment revenue sharing and resettlement frameworks, and (commitment is 57 percent of recommended) and determining and implementing environmental flows. underexpenditure (table 4.4). Performance has been Federal and provincial investment is around half of the poorest for the action area “major infrastructure level recommended by FoDP (2012). The main reason and associated institutions,” followed by “raising for the shortfall is that despite government prioritizing agricultural productivity.” Commitments against investment in dams, commitments for these were not “sustainable urban services” and “knowledge made during 2013–17. Actual expenditure has been management” have exceeded recommendations, but 35 percent of funds committed (figure 4.12). Table 4.4  Recommended Investments in Priority Actions Areas, Commitments, and Expenditures by Pakistan’s Water Sector Task Force, 2013–17 US$, millions Action area Recommended recommended recommended Commitment commitment commitment expenditure Expenditure Expenditure investment as share of as share of as share of Actual Actual Major infrastructure and associated institutions 26,556 12,605 0.47 4,411 0.35 0.17 Raising agricultural productivity 1,920 1,406 0.73 338 0.24 0.18 Living better with floods 1,120 1,047 0.93 269 0.26 0.24 Sustainable urban services 2,299 3,134 1.36 1,198 0.38 0.52 Knowledge management 115 164 1.43 61 0.37 0.53 Total 32,010 18,356 0.57 6,277 0.34 0.20 Source: Author calculations. Figure 4.12  National Investment and Expenditure for First WSTF Action in Pakistan, 2013–17 30,000 25,000 20,000 US$, millions 15,000 10,000 5,000 0 2013 2014 2015 2016 2017 National expenditure National investment FoDP five–year recommended investment Source: Author calculations. Note: The first action area is “major infrastructure and associated institutions.” FoDP = Friends of Democratic Pakistan; WSTF = Water Sector Task Force. 69 For the second action, “raising agricultural productivity,” commitment, reflecting, in part, the delays in federal and provincial governments have made reaching a federal-provincial funding agreement for significant financial commitments; however, this work (figure 4.14). actual expenditure has been only at one-quarter of the planned investment level (figure 4.13). The fourth action area, “sustainable urban services,” Cumulative investment and expenditure for some promotes productive, secure, and sustainable cities, projects are reported under PDSP and ADP as reducing addressing major social challenges of urbanization in over time, reflecting both adjustments in planning and Pakistan. Federal and provincial commitment have corrections to prior reporting. exceeded the level recommended by FoDP (2012), which only considers investments for one large city. For the third action, “living better with floods,” FoDP Expenditure progress has been steady (figure 4.15). (2012) recommends US$1.1 billion of investment to mitigate flood impacts and improve resilience. For the fifth action area, “knowledge management,” National investment commitment during the four- FoDP (2012) recognizes the inadequacies of water year period was close to the recommended level, research, modeling, and analysis in Pakistan. mainly for key elements of the National Flood Recommendations cover capacity building for Protection Program IV. Actual expenditure, however, management and research, development of water has been only one-quarter of the investment models, a groundwater knowledge base and Figure 4.13  National Investment and Expenditure for Second WSTF Action in Pakistan, 2013–17 2,000 1,800 1,600 1,400 US$, millions 1,200 1,000 800 600 400 200 0 2013 2014 2015 2016 2017 National expenditure National investment FoDP five–year recommended investment Source: Author calculations. Note: The second action area is “raising agricultural productivity.” FoDP = Friends of Democratic Pakistan; WSTF = Water Sector Task Force. Figure 4.14  National Investment and Expenditure for Third WSTF Action in Pakistan, 2013–17 1,200 1,000 800 US$, millions 600 400 200 0 2013 2014 2015 2016 2017 National expenditure National investment FoDP five–year recommended investment Source: Author calculations. Note: The third action area is “living better with floods.” FoDP = Friends of Democratic Pakistan; WSTF = Water Sector Task Force. 70 PAKISTAN: GETTING MORE FROM WATER Figure 4.15  National Investment and Expenditure for Fourth WSTF Action in Pakistan, 2013–17 3,500 3,000 2,500 US$, millions 2,000 1,500 1,000 500 0 2013 2014 2015 2016 2017 National expenditure National investment FoDP five–year recommended investment Source: Author calculations. Note: The fourth action area is “sustainable urban services.” FoDP = Friends of Democratic Pakistan; WSTF = Water Sector Task Force. Figure 4.16  National Investment and Expenditure for Fifth WSTF Action in Pakistan, 2013–17 180 160 140 120 US$, millions 100 80 60 40 20 0 2013 2014 2015 2016 2017 National expenditure National investment FoDP five–year recommended investment Source: Author calculations. Note: The fifth action area is “knowledge management.” FoDP = Friends of Democratic Pakistan; WSTF = Water Sector Task Force. decision support systems, and international agency References collaboration. The planned investment for this priority ADB (Asian Development Bank). 2007. Country action has been US$164 million, well above the Assistance Program Evaluation for Pakistan. recommended level (figure 4.16). The largest project https://www.adb.org/sites/default/files​ under this action is the World Bank–financed Water /evaluation-document/35031/files/cape​ -pak​ Sector Capacity Building and Advisory Services Project. -2007-0.pdf. Actual expenditure is low. Ahmed, S. N. 2013. Quetta Water and Sanitation The preceding analysis shows that sector financing Authority: Performance and Prospects. Quetta, is inadequate. The biggest problem is slow Pakistan: PHE Department. expenditure, with only around one-third of the committed funds spent over the last four years. AIIB (Asian Infrastructure Investment Bank). 2018. While expenditure remains low, it will be difficult Lahore Water and Wastewater Management to make the case for any additional funding for the Project: Project Summary Information. https:// sector. Slow expenditure reflects low institutional www.aiib.org/en/projects/proposed/2018​ capacity and a risk-averse public sector; these will be /_download/lahore/PSI-Lahore-Water-and​ slow to change. -Wastewater-Concept-Stage-4-May-2018.pdf. 71 Ali, A. 2013. Indus Basin Floods: Mechanisms, Impacts, Gazdar, H. 2009. The Fourth Round, and Why They Fight and Management. Manila, Philippines: ADB. On: An Essay on the History of Land and Reform https://www.adb.org/sites/default/files​ in Pakistan. PANOS South Asia. Collective for /publication/30431/indus-basin-floods.pdf. Social Science Research, Karachi, Pakistan. Ali, S. M. 2015. Development, Poverty, and Power Gilmartin, D. 2015. Blood and Water: The Indus River in Pakistan: The Impact of State and Donor Basin in Modern History. Berkeley, CA: University Interventions on Farmers. Routledge. London. of California Press. Ali, I. 1988. The Punjab under Imperialism 1885–1947. Goldface-Irokalibe, I. J. 2008. “Water Management in Princeton, NJ: Princeton University Press. Federal and Federal-Type Countries: Nigerian Perspectives.” https://www.africaportal.org​ Azam, M. F., P. Wagnon, E. Berthier, C. Vincent, K. Fujita, /publications/water-management-in-federal​ and J. S. Kargel. 2018. “Review of the status and -and-federal-type-countries-nigerian-perspectives/. mass changes of Himalayan-Karakoram glaciers.” Journal of Glaciology 64 (243): 61–74. GoP (Government of Pakistan). 2005. National Environmental Policy http://waterinfo.net.pk​ Bhatti, M. T., Anwar, A. A and Aslam, M. 2017. /sites/default/files/knowledge/National%20 “Groundwater monitoring and management: Environmental%20Policy%202005.pdf. Status and options in Pakistan”. Computers and Electronics in Agriculture 135: 143–153. ———. 2007a. National Disaster Risk Management Briscoe, J., and U. Qamar. 2005. “Pakistan’s Water Framework Pakistan. http://www.ndma.gov​ Economy: Running Dry.” Water P-Notes 17. World .pk/plans/National%20Disaster%20Risk%20 Bank, Washington, DC. https://openknowledge​ Management%20Framework-2007.pdf. .worldbank.org/handle/10986/11746. ———. 2007b. Pakistan in the 21st Century: Vision 2030. Planning Commission. http://www.national​ Brown, C., and U. Lall. 2006. “Water and Economic planningcycles.org/sites/default​ /planning​ /files​ Development: The Role of Variability and a _cycle_repository/pakistan​ /vision_2030​ -full.pdf. Framework for Resilience.” Natural Resources Forum 30: 306–17. ———. 2012. National Climate Change Policy. http:// www.gcisc.org.pk/National_Climate_Change​ Chaudhry, A. M. 2018. “Improving On-Farm Water Use _Policy_2012.pdf. Efficiency: Role of Collective Action in Irrigation Management.” Water Resources and Economics ———. 2014. Pakistan 2025: One Nation – One Vision. 22: 4–18. Planning Commission. http://www.ntb.gov.pk​ /ntb/userfiles1/file/Pakistan-Vision-2025.pdf. Euroconsult (2011) Handbook on Water Statistics of Pakistan. Report to the Water Sector Capacity ———. 2018. National Water Policy. Ministry of Water Building and Advisory Services Project of the Resources. http://mowr.gov.pk/wp-content​ /uploads​ Government of Pakistan. /2018/06/National-Water​ -policy-2018-2.pdf. FAO (Food and Agriculture Organization of the United Government of Punjab. 2012. Groundwater Monitoring Nations). 2018. Water Governance. http://www​ Report. Directorate of Land Reclamation, Punjab .fao.org/land-water/water/watergovernance​ /en/. Irrigation Department, Lahore, Pakistan. Felbab-Brown, V. 2017. Water Theft and Water Hashim, A. 2017. “Parched for a Price: Karachi’s Water Smuggling: Growing Problem or Tempest in a Crisis.” Al-Jazeera English. https://interactive​ Teapot? Washington, DC: Brookings Institute. .aljazeera.com/aje/2017/parched-for-price​ /index.html. FoDP (Friends of Democratic Pakistan). 2012. A Productive and Water-Secure Pakistan: Hussain, A. 2008. Power Dynamics, Institutional Infrastructure, Institutions, Strategy. Islamabad, Instability and Economic Growth: The Case of Pakistan: Water Sector Task Force, FoDP. http:// Pakistan. Islamabad, Pakistan: Asia Foundation. metameta.nl/wp-content/uploads/2013/11​ Ishfaq, M., M. Younas, M. Waqar, and Z. Muhammad. /FoDP-WSTF-Report-Final-09-29-12.pdf. 2017. “Comparative Analysis of Wheat and Gaupp, F., J. Hall, and S. Dadson. 2015. “The Role of Brassica in Terms of Relative Profitability: A Case Storage Capacity in Coping with Intra-and Inter- Study of District Chakwal, Punjab, Pakistan.” Annual Water Variability in Large River Basins.” Journal of Agricultural, Food, and Environmental Environmental Research Letters 10: 125001. Sciences 10 (71): 71–77. Gaurav, K., R. Sinha, and P. K. Panda. 2011. “The Indus flood IUCN (International Union for Conservation of Nature). of 2010 in Pakistan: a perspective analysis using 2015. Water Supply & Sanitation Analysis in remote sensing data.” Natural Hazards 59 (3): 1815. Sectors G & I Islamabad. Islamabad, Pakistan: IUCN. 72 PAKISTAN: GETTING MORE FROM WATER Jacoby, H., and G. Mansuri. 2018. “Governing the Sanitation and Poverty in Pakistan and Its Impact Commons? Water and Power in Pakistan’s Indus on Child Stunting. Washington, DC: World Bank. Basin.” Policy Research Working Paper, World McMahon, T. A., R. M. Vogel, G. G. S. Pegram, M. C. Peel, Bank, Washington, DC. and D. Etkin. 2007. “Global streamflows – Part 2: Jacoby, H., G. Mansuri, and F. Fatima. 2018. Reservoir storage–yield performance.” Journal of “Decentralization and Redistribution: Irrigation Hydrology 347: 260– 271. Reform in Pakistan’s Indus Basin.” Policy Research Mekonnen, D. K., H. Channa, and C. Ringler. 2015. Working Paper, World Bank, Washington, DC. “The Impact of Water Users’ Associations on the JICA (Japan International Cooperation Agency). 2010. Productivity of Irrigated Agriculture in Pakistani “Implementing Review Study Report on the Punjab.” Water International 40 (5–6): 733–47. Project for the Expansion of Water Supply System Murtaza, G., and M. H. Zia. 2012. “Wastewater in Faisalabad in Islamic Republic of Pakistan. JICA, Production, Treatment and Use in Pakistan.” Faisalabad, Pakistan. http://open_jicareport.jica​ Presented at the “Second Regional Workshop .go.jp/pdf/12020186_01.pdf. of the Project ‘Safe Use of Wastewater in ———. 2016. “Record of Discussions on the Project for Agriculture,’” New Delhi, India, May 16–18. Water Supply, Sewerage and Drainage Master Mustafa, D., G. Gioli, M. Karner, and I. Khan. 2017. Plan of Faisalabad in the Islamic Republic of Contested Waters: Subnational Scale Water Pakistan Agreed upon between Faisalabad Water Conflict in Pakistan. Washington, DC: USIP. and Sanitation Agency and Japan International Cooperation Agency.” JICA, Lahore, Pakistan. NDC (National Development Consultants). 2014. Planning https://www.jica.go.jp/english/our_work​ and Engineering Services for Master Plan in Peshawar /social_environmental/id/asia/south/pakistan​ Khyber Pakhtunkhwa: Drinking Water, Sanitation/ /c8h0vm00009t9g1t-att/c8h0vm00009t9ga5.pdf. Storm Water and Solid Waste Services. Executive Summary: Consolidation of Key Findings of Vols. 1–3. JICA and KWSB (Karachi Water and Sewerage Board). Prepared for USAID. Peshawar, Pakistan: NDC. 2008. The Study on Water Supply and Sewerage System in Karachi in The Islamic Republic of Nepal, S. and A. B. Shrestha. 2015. “Impact of climate Pakistan: Final Report. Vol. 2. Karachi, Pakistan. change on the hydrological regime of the Indus, Ganges and Brahmaputra river basins: a Khalil, S., and M. K. Kakar. 2011. “Agricultural Use of review of the literature.” International Journal Untreated Urban Wastewater in Pakistan.” Asian of Water Resources Development 1–18. doi: Journal of Agriculture and Rural Development 10.1080/07900627.2015.1030494. 1 (1): 21–26. Paul, J. D., W. Buytaert, S. Allen, J. A. Ballesteros‐Cánovas, Khan, A. D., N. Iqbal, M. Ashraf, and A. A. Sheik. 2016. J. Bhusal, K. Cieslik, J. Clark, S. Dugar, D. M. Hannah, “Groundwater investigations and mapping in the M. Stoffel, A. Dewulf, M. R. Dhital, W. Liu, J. L. Upper Indus Plain.” Pakistan Council of Water Nayaval, B. Neupane, A. Schiller, P. J. Smith, and Resources Research, Islamabad. 72pp. R. Supper. 2018. “Citizen Science for Hydrological KWSB. 2018. “Karachi Water and Sewerage Board” Risk Reduction and Resilience Building.” Wiley (website). http://www.kwsb.gos.pk/. Interdisciplinary Reviews: Water 5 (1): e1262. Lieftinck, P., A. R. Sadove, and T. C. Creyke. 1968. PCRWR (Pakistan Council of Research in Water “Water and Power Resources of West Pakistan: Resources). 2010. Technical Assessment A Study in Sector Planning.” Volume I, Main Survey Report of Water Supply Schemes. Report. World Bank, Washington DC. Islamabad, Pakistan: PCRWR. www.pcrwr.gov​ .pk/Publications/Water%20Quality%20Reports​ Malik, N. 2008. “Reply: What’s changed (since 1975)? /Water%20Suply%20Schemes%20Sindh.pdf. From good government to good governance.” Dialectical Anthropology 32 (1/2): 31–36. ––––––. 2011. Technical Assessment Survey of Water Supply Schemes: Provision of Safe Drinking Water Manarvi, I. A., and M. Ayub. 2013. “Issues and Project. Factsheet, PCRWR, Islamabad, Pakistan. Remedies of Sewage Treatment and Disposal in www.pcrwr.gov.pk/Publications/Water%20 Islamabad, Pakistan.” Chemistry and Material Quality%20Reports/Technical%20Assesment%20 Research 3 (9): 108–18. Survey%20(Punjab).pdf. Mansuri, G., F. Sami, T. T. D. Dung, P. Pandey, and Quetta District Government. 2011. Quetta-Integrated M. Ali. 2018. When Water Becomes a Hazard: District Development Vision. Quetta, Pakistan: A Diagnostic Report on the State of Water, Quetta and IUCN. 73 , A. S., McCornick, P. G., Sarwar, A., and Qureshi­ van der Velde, E. J., and J. Tirmizi. 2004. “Irrigation Sharma, B. R. 2010. “Challenges and prospects Policy Reforms in Pakistan: Who’s Getting the of sustainable groundwater management in Process Right?” In The Politics of Irrigation the Indus Basin, Pakistan”. Water Resources Reform: Contested Policy Formulation and Management 24 (8): 1551­ –1569. Implementation in Asia, Africa and Latin America, edited by P. Mollinga and A. Bolding. Aldershot, Qureshi, A., and A. H. Sayed. 2014. Situation U.K.: Ashgate. pp207–239. Analysis of the Water Resources of Lahore: Establishing a Case for Water Stewardship. van Steenbergen, F., M. Basharat, and B. K. Lashari. Islamabad, Pakistan: WWF. 2015. “Key challenges and opportunities for conjunctive management of surface and Rinaudo, J., and Z. Tahir. 2003. “The Political Economy groundwater in mega-irrigation systems: Lower of Institutional Reforms in Pakistan’s Irrigation Indus, Pakistan.” Resources 4 (4): 831–856. Sector.” In The Economics of Water Management in Developing Countries: Problems, Principles and WAPDA. 2016. Hydropower Development in Pakistan: Policies, edited by Koundouri, P., P. Pashardes, Location Map. Islamabad: Pakistan Water T. Swanson, and A. P. Xepapadeas. Cheltenham, and Power Development Authority. http:// U.K.: Edward Elgar. pp41–60. www.wapda.gov.pk/index.php/projects​ /hydropower-development-in-pakistan. Rinaudo, J. D., P. Strosser, and S. Thoyer. 2000. “Distributing Water or Rents? Examples from a WASA-F. 2018. WASA Faisalabad: An Overview. http:// Public Irrigation System in Pakistan.” Canadian wasafaisalabad.gop.pk/. Journal of Development Studies 21 (1): 113–39. WWF (2012) Development of Integrated River Basin Robinson, S., and A. Gueneau. 2014. Economic Management for Indus Basin. Challenges and Evaluation of the Diamer-Basha Dam: Analysis opportunities. WWF: Islamabad. https://www​ with an Integrated Economic/Water Simulation .wwfpak.org/publication/pdf/irbm_2013.pdf. Model of Pakistan. IFPRI Working Paper No.14. World Bank. 1994. ‘Pakistan Irrigation and Drainage: SBP (State Bank of Pakistan). 2017. Water Sustainability Issues and Options.’ Report No. 11884-PK. World in Pakistan—Key Issues and Challenges. State Bank, Washington, DC. Bank of Pakistan’s Annual Report 2016–17. ———. 1998. ‘Pakistan Public Expenditure Management: Karachi, Pakistan: SBP. Accelerated Development of Water Resources Shabbir, R., and S. Ahmad. 2016. “Water Resource and Irrigated Agriculture’. Volume II. Report No. Vulnerability Assessment in Rawalpindi and 25665-PK. World Bank, Washington, DC. Islamabad, Pakistan using Analytic Hierarchy ———. 2005. ‘Managing Karachi’s Water Supply and Process (AHP).” Journal of King Saud University— Sanitation Services: Lessons from a Workshop.” Science 28: 293–9. WSP Working Paper Series, World Bank, Siddiqui, K. M. 2014. State of Urban Water Supply and Washington, DC. https://openknowledge​ Sanitation Sector in Sindh. Karachi, Pakistan: PUF. .worldbank.org/handle/10986/17259. Soncini, A., D. Bocchiola, R. Rosso, S. Meucci, F. Pala, ———. 2018. Transforming Karachi into a Livable and G. Valé. 2014. “Water and Sanitation in and Competitive Megacity: A City Diagnostic Multan, Pakistan.” In Sustainable Social, Economic and Transformation Strategy. Directions in and Environmental Revitalization in Multan Development. Washington, DC: World Bank. City, edited by A. Del Bo and D. Bignami. Cham, Yu, W., Y. C. E. Yang, A. Savitsky, D. Alford, C. Brown, Switzerland: Springer. 149–162. J. L. Wescoat JR, D. Debowicz and S. Robinson. Starkloff, R. 2001. “Farmers’ Perception of the Social 2013. Indus Basin of Pakistan: Impacts of Climate Mobilization of Water User Organization in the Risks on Water and Agriculture. Directions Sindh, Pakistan.” Working Paper 33, IWMI, Lahore, in Development; Countries and Regions. Pakistan. Washington, DC: World Bank. Strosser, P. 1997. “Analysing Alternative Policy Zemadim, B., M. McCartney, S. Langan, and B. Instruments for the Irrigation Sector: An Sharma. 2013. A Participatory Approach for Assessment of the Potential for Water Markets Hydrometeorological Monitoring in the Blue Development in the Chishtian Sub-Division.” Nile River Basin of Ethiopia. IWMI Research Ph.D. Thesis, Wageningen Agriculture University, Report 155. Colombo, Sri Lanka: IWMI. doi: Wageningen, Netherlands. 10.5337/2014.200. C HAPT E R 5 Pakistan’s Water Sector Performance T his chapter assesses the performance of Pakistan’s People water sector in terms of the management of ices Mitig water resources, the delivery of water-related erv ati eds on services, and the mitigation of water-related risks. t of ru ela rast ctur wa Flood management and pollution control—although nf r-r ter- ivery of wate water-related risks—are covered under water resources e I related risks management. Risk mitigation covers exogenous and Water endowment longer-term risks, including intersectoral complexities Envir Del and climate change. ns y titution om I s o M s nm na n ce ur a ge co me so nt of w ater re e E nt Water Resources Management Management of water resources is central to water • System operations. Reservoir operations for security. To assess performance in managing water irrigation supply; flood management; hydropower; resources the following aspects are considered: and environmental flows. • Data, information, and analysis. For water resource • Environmental sustainability. Specification and assessments and accounting; drought and flood management of environmental flows; sustainable forecasting; and data and information management groundwater management; pollution and water and sharing. quality. • Resource planning and allocation. Strategic basin • Productivity. Economic productivity of planning; flood planning; drought planning; and water and land; water footprints; and land water allocation between and within provinces. fragmentation. 76 PAKISTAN: GETTING MORE FROM WATER Key Messages • Basin-level water resources management in Pakistan is constrained by insufficient data and analysis, and a lack of strategic basin planning to guide sustainable resource use and economic development. There is a lack of clarity in risk sharing between provinces and sectors in times of acute scarcity. • Within provinces there are no formal mechanisms for intersectoral reallocation of water in the face of changing demand patterns and changing climate. Water allocation processes are suboptimal in terms of efficiency, equity, and transparency, which contributes to the low productivity of irrigated agriculture and a lack of trust between farmers and service providers. • The economic productivity of water is very low, especially in agriculture. Increasing productivity requires improving clarity and accountability in water governance, modernization and improving operational performance of irrigation systems, agricultural policy reform, improving on-farm water management and diversification of crop types, and reversing the fragmentation of land holdings. • Groundwater is overexploited in parts of Punjab and Balochistan, but depletion is a small fraction of the annual groundwater balance, except in a few local cases. Severe depletion is problematic for urban water supply, especially in Lahore and Quetta. The greatest long-term risk to groundwater sustainability is pollution. • Pakistan does little to protect water-dependent ecosystems through either environmental flows or water quality management. Efforts to protect the quality of the water resource base are inadequate. Climate change and increasing water demand will intensify the challenge of improving the sustainability of water management. • Reservoir operations should be systematically reevaluated to assess their adequacy for meeting multiple benefits (including environmental benefits) across the Indus Basin in the face of a changing climate and changing demand patterns. Data, Information, and Analysis Farm-level water use is not monitored, leaving much uncertainty in how, when, and where it is used. Monitoring and Water Information Systems Pakistan—with its water scarcity issues—should be Although average Indus inflows to Pakistan and concerned about its inability to accurately describe outflows to the Arabian Sea are well known based what happens to around half of the total resource. Flow on reasonably long-term and reliable records, to the sea supports important environmental assets there is only partial monitoring of runoff within and functions, especially in the delta region, but is still Pakistan. Hydrological monitoring across the widely considered a waste in Pakistan. Unaccounted Makran Coast and Kharan Desert drainage areas losses, including beneficial consumptive landscape water is almost nonexistent. The data for internally use and large nonconsumptive losses in irrigation, are generated water are typically ignored in water around three times the magnitude of flow to the sea, resource assessments and planning. Gross canal and should a primary focus for improved water resources withdrawals are reasonably well monitored, management. Pakistan urgently needs to strengthen although there are concerns about the veracity and hydrological monitoring systems and develop robust transparency of these records given the tensions water accounting to guide improved water resource surrounding interprovincial water sharing. planning and allocation. This should combine ground- based observations with Earth observations in robust As shown in chapter 3, the internal recycling of water modeling and accounting frameworks. in the Indus Basin and the high level of water loss mean that monitoring of inflows, outflows, and canal Pakistan needs to adopt modern, integrated systems for withdrawals is insufficient to arrive at a fully accurate data storage, retrieval, and sharing. Numerous federal and reliable water balance that resolves internal and provincial agencies collect data but use different fluxes—even on a long-term average basis. A dynamic software platforms for data storage—with limited water account is required to inform seasonal water interoperability and manual sharing. Few data from allocation decisions, irrigation system management, before the 1990s have been captured in electronic form and conjunctive surface water and groundwater and so are difficult to access. This limits the value of management. This requires much better monitoring and these data to water planners and managers, as well as analysis. The significant river gains and losses, which as to researchers. While some real-time data are shared shown in chapter 3 are not in dynamic equilibrium, are by federal and provincial agencies, historical data are poorly quantified and poorly understood. not routinely placed in the public domain, and accessing 77 these data is difficult and slow. Donor-funded assistance (including more actionable forecasts and warnings) and to the Pakistan Water and Power Development Authority for medium- to long-term planning, particularly in the (WAPDA) is helping to establish a modern water data context of increased climate variability. Provincial irrigation management system, but much work is required to departments need improved hydrometeorological capture legacy data, encourage wider adoption of information to better manage irrigation water distribution, this platform (or compatible platforms) across other and provincial agriculture departments need monthly federal and provincial agencies, establish processes and weather outlooks tailored to 19 agricultural zones. Better protocols for interagency data exchange, and facilitate hydrologic forecasts, including of transboundary flows, are for online public data access. needed to guide reservoir management and hydropower operations. Improved river flow forecasts should inform Failure to develop and maintain transparent water interprovincial water allocation. information systems undermines water management efforts, creating uncertainty and controversy over the Flood forecasting is important for Pakistan, and while size of the water resource and the volumes allocated good progress has been made, capabilities are still for use. There remain ongoing controversies over relatively rudimentary. Data from weather radar, measuring flows at both the barrages and in distributary telemetered AWS, and a larger manual network are canals. Each province is responsible for measuring used mostly for manual analyses (e.g., hand-drawn water diversions. In principle, other provinces may hydrograph analysis and simple regress analysis for send officials to “check” observations; however, the snowmelt estimation) as well as for some computer arrangement does not work effectively. An Indus Basin modeling. Quantitative precipitation forecasts are telemetry system intended to provide confidence in generated using the flood early warning system (FEWS). measurement of water flows at key locations failed Hydrologic modeling is conducted with the U.S. NOAA/ (Bhatti, Anwar, and Aslam 2017; FoDP 2012). New NWS Sacramento Soil Moisture Accounting Model; and telemetry investments that will overcome past technical hydraulic simulation of flood wave routing and flood deficiencies, supported by improved governance inundation mapping is conducted using SOBEK. arrangements, are planned with World Bank support. The current forecasting system has many weaknesses. There have been many hydrologic studies of the Upper The sparse AWS network has limited telemetry and Indus Basin, including assessing current climate change inadequate radar coverage. The monitoring network impacts and projecting potential future changes. has operational problems, such as inadequate power However, all have been hampered by a paucity of supply backups and a lack of centralized data storage long-term, good-quality hydrometeorological data. and management. Forecasting is of river stage only (not In the Upper Indus there is one gauge for precipitation flow), there is no routine monitoring of the upper basin for approximately each 5,000 square kilometers snow pack or snow pack modeling for flow prediction, (UNDP 2017), well below the WMO (1994) standard and there is no reservoir simulation modeling. Routine of one gauge per 250 square kilometers. ADB (2010) data analysis is not automated and there is no objective recommends that at least 75 automatic weather process for forecast verification. Interagency data sharing stations (AWS) and 35 hydrological monitoring stations is very limited. The Indus Water Treaty provides for data should be installed at high elevation across the Upper sharing with India for flow forecasting, but there are no Indus Basin. A denser monitoring network will capture arrangements for data sharing with Afghanistan or China. seasonal variations across the basin and correct the Flood forecasting and early warning coverage are current bias due to a predominance of valley floor incomplete and should be extended to the remote monitoring stations. Better hydrometeorological data areas of the country, especially the hill torrents from the Upper Indus Basin will be especially important and Swat and Kabul rivers (Ali 2013). The steep for research into the changing hydrology and glaciology topography of these catchments means runoff is of the Indus. Increases in flow variability at different particularly rapid, generating dangerous flash floods, time scales are expected and understanding these is especially during the monsoon. The 2010 flood was critical for future water management. largely caused by exceptional rainfall in the Kabul and Swat basins of the Upper Indus, which were Forecasting not monitored. This delayed early warnings and flood responses (Tariq and van de Giesen 2012). Pakistan’s weather and flow forecasting are largely Coordination with Afghanistan on flood forecasting for insufficient to meet diverse stakeholder information the Kabul could help mitigate floods. needs. Currently, one- to two-day weather forecasts, three- to five-day outlooks, and 24-hour hydrological Existing hydrometeorological monitoring is inadequate forecasts are generated (World Bank 2018). These for drought forecasting and planning. Monitoring and are insufficient to meet the needs of stakeholders forecasting focus on rainfall, snowfall, and irrigation who require information for short-term operations system inflows, but ignore soil moisture and other 78 PAKISTAN: GETTING MORE FROM WATER water parameters needed to identify the onset of model of the national economy, is used here (see agricultural drought (Khan and Khan 2015). Drought chapter 6). The IBMR is an optimizing model, focused planning suffers from a lack of standardized risk on monthly water allocations to irrigation agriculture. assessments, and drought response suffers from While it represents the complex river and canal inadequate data sharing protocols (Khan and Khan network, it does not a model the hydrologic routing 2015). These shortcomings hinder drought preparation of flows through the system. More recently, an Indus and mitigation. River System Model (IRSM) has been developed for the Indus using the Source modeling platform (Stewart Groundwater and Water Quality et al. 2018). IRSM is a daily flow routing and allocation model that includes river gains and losses to and There is limited monitoring of groundwater levels from groundwater. This model has been developed and quality in Pakistan, despite the importance of to strengthen water accounting and flow allocation groundwater and the growing challenges of salinization processes at the Indus River System Authority (IRSA), and depletion. Data are collected and published but is also suitable for strategic basin planning. While on the growing number of private tube wells, but significant effort has been put to capacity building there are few operational programs for systematic in the use of IRSM, it has not been adopted to inform groundwater condition monitoring (Bhatti et al. 2017). the seasonal water allocation process, which continues Groundwater is monitored in some urban areas, but in to rely on daily manual updating of spreadsheets and agricultural areas, only Punjab has any semblance of an sharing allocation assessments with the provinces institutionalized and systematic monitoring program. by facsimile. There is an important opportunity to The Punjab program monitors levels and quality through modernize this process for more reliable water a network of piezometric wells and water quality accounting and more transparent and efficient data sampling points (Bhatti et al. 2017). Greater investment and information sharing. in groundwater monitoring and analysis is required to inform sustainable groundwater use, conjunctive surface water and groundwater management, and irrigation Resource Planning and Allocation drainage and salinity management. Strategic Basin Planning There is some, but far from adequate, monitoring of In Pakistan, water resources planning has focused water quality across Pakistan. Sediment concentrations strongly on supply-side infrastructure with much of are monitored routinely at the regular gauging stations the water resources development discourse revolving (table 4.3), mostly in the Upper Indus Basin, and around new dams. There is no established mechanism some chemical samples are collected approximately for strategic basin-scale planning—either for the monthly. Some provincial authorities monitor the Indus Basin or the minor basins of Balochistan—that quality of drinking water supplies—surface water comprehensively considers sustainable management and groundwater—but monitoring is infrequent and of existing infrastructure assets, surface water and inconsistent. National and international scientific groundwater interactions, interprovincial water sharing, organizations have monitored water quality to assess intersectoral water management, environmental the status and trend in environmental conditions and sustainability, or basin-scale management of sediment the risks to human health (e.g., Grigg et al. 2018). and salinity and other water quality issues. Flood But there is no consistent approach to data storage planning and interprovincial sharing have been or access, limited data quality assurance, and limited addressed with some success, and management of data sharing between agencies. Some sediment and some major system assets—especially the headwater contaminant concentration data have been collected, dams—has been the responsibility of WAPDA with but no comprehensive analysis has been undertaken development financing support. of loads or transport and how these have changed with agricultural expansion and urban and industrial The 2018 National Water Policy (NWP) and the development. Balochistan Integrated Water Resources Management (IWRM) policy espouse a desire to operationalize a more comprehensive and integrated approach to Modeling water resources management, but this has yet to be Numerous hydrologic and water resource simulation seriously tackled. As noted in chapter 4, this requires models have been developed for the Indus, including institutional reforms and a more comprehensive by the World Bank and several research organizations. legal framework. IRSA, as a key institution with WAPDA is the custodian of the latest version of the a basin-scale perspective, currently has a narrow Indus Basin Model Revised (IBMR) developed by the operational role in water distribution, and has neither World Bank and used in Yu et al. (2013). A variant, the mandate nor the capacity to embrace a more linked to a computable general equilibrium (CGE) strategic planning function. Nonetheless, establishing 79 a process of strategic basin planning is increasingly term, to strengthen or upgrade flood protection critical for Pakistan, given increasing water demands infrastructure, and over the medium term. NFPP IV and shifting sectoral balance, the growing challenges prioritizes improvements in flood forecasting and of climate change, and the increasing evidence of notes the importance of other nonstructural measures environmentally unsuitable water management with such as vulnerability and risk assessments, floodplain significant negative consequences. Strategic planning zoning, and land-use planning and enforcement. is recommended for the entire Indus Basin of Pakistan However, beyond the forecasting work, these critical for long-term environmentally sustainable economic nonstructural improvements are left to provincial development, as well as for subbasins such as the governments. because they are outside the current Kabul and key river basins in Balochistan. For the Kabul, financing envelop, they are not likely to progress. there are opportunities to establish transboundary The FFC should show stronger leadership on governance arrangements with Afghanistan to support nonstructural aspects of flood planning, because joint water resources planning and development. a basin-scale approach is required that addresses While many aspects of water resources planning can the trade-offs between flood risks and the benefits and should happen at the provincial level, issues of arising from floodplain development. These strategic environmental sustainability, sediment management, planning questions need to be addressed within the major asset management (dams and barrages), national flood management framework, because they interprovincial sharing, and transboundary water issues guide flood management investments. In evaluating require a suitably resourced (funding and capacity) nonstructural measures, flood management needs to and sufficiently empowered federal institution with more prominently include catchment management, mechanisms for effective provincial consultation. especially in the north and northwest of the country. Catchment management in the hill torrent areas, using Flood Planning community-based approaches, can help reduce flash Flood planning is one aspect of basin-scale flood risk and reduce river flood peaks while providing management that has received long-standing attention irrigation water in the Hindu Kush Himalaya (Saher in Pakistan, with the establishment in 1977 (following et al. 2015; Shrestha, Shah, and Karim 2008). While the 1976 floods) of a Federal Flood Commission and federal and provincial governments have agreed to a a sequence of strategic plans backed by significant cost-sharing arrangement to finance NFPP IV, the FFC investment. While flood planning has traditionally requires considerable additional capacity to effectively followed a largely infrastructure-based approach, there oversee implementation of this major program. has been a positive shift toward more integrated flood management solutions in recent years. This includes Drought Planning the work of the National Disaster Management Authority, established in 2010. Its mandate is to Droughts occur frequently in Pakistan (table 5.1) and implement vulnerability assessments, multihazard early can affect almost one-third of Pakistan (map 5.1). warning systems, and community-level vulnerability The most drone-prone areas include Cholistan in reduction programs; and to promote disaster Punjab, Thar in Sindh, and the Chagai-Kharan region preparedness planning. in Balochistan (Khan and Khan 2015). Balochistan is by far the most drought-prone province because Initial flood investment plans supported the of its arid to hyper-arid climate (van Gils and construction of the current system of spurs and Baig 1992). Balochistan’s agricultural sector has levees to train manage flood discharges and protect riverbanks from erosion. The latest iteration—the fourth Table 5.1  Droughts in Pakistan 10-year National Flood Protection Plan—describes an investment of around US$1.7 billion, of which Extent Year or period nearly 90 percent is related to infrastructure. Although Pakistan 1871,1881, 1899, 1902, 1920, 1931, the Federal Flood Commission (FFC) recognizes the 1935, 1947, 1951, 1998–2001 significant climate change risks confronting Pakistan, Balochistan no early data, 1967–69, 1971, the current investment plan has few specifics 1973–75, 1994, 1998–2002, on how to address these risks beyond undefined 2009–15 studies. Given the changing intensity and frequency Khyber Pakhtunkhwa 1902, 1951, 1986, 1999 of hydrological extremes under climate change, flood planning needs to make specific provisions Punjab 1899, 1920, 1935, 1999–2001 for floods that exceed the design criteria of existing Sindh 1871, 1881, 1899, 1931, 1947, infrastructure, and revise existing design standards, 1999, 2014, 2015 if necessary. Current inspection protocols for flood Sources: Ahmad et al. 2004; Ahmed et al. 2016; Ata-ur-Rahman and protection structures should be reviewed in the short Shaw 2015. 80 PAKISTAN: GETTING MORE FROM WATER Map 5.1  Percentage of District Population Vulnerable to Drought in Pakistan TAJIKISTAN UZBEKISTAN CHINA TU T RKMEN U RKM I S TAN ENI KHYBER Approximate PAKHTUNKHWA Line of Control ISLAMABAD CAPITAL Jammu AFGHANISTAN TERRITORY Jammu and Kashmir Peshawar ISLAMABAD and Kashmir FEDERALLY ADMINISTERED TRIBAL AREAS PUNJAB Lahore Quetta BALOCHISTAN I.R. OF INDIA IRAN Drought hazard risk ranking SINDH Acutely vulnerable Very vulnerable Vulnerable Moderately vulnerable Selected cities and towns Karachi Province capitals National capital Arabian Sea Province boundaries International boundaries 0 50 100 150 Kilometers 0 50 100 150 Miles IBRD 44141 | DECEMBER 2018 Source: Shariff 2015. 81 experienced major drought losses, especially from Interprovincial Water Allocation 1998 to 2002 when agricultural productivity halved Interprovincial water disputes, especially between (Ahmad et al. 2004). Nationally, this drought was the Punjab and Sindh, predate the creation of Pakistan most extreme since independence. It affected more by many years. Prepartition ambitions of Punjab to than 3.5 million people, caused hundreds of deaths, divert water for irrigation were opposed by Sindh, and increased migration from rural Balochistan and in 1945 the British imposed a solution on the two (GoP 2004). Three years of drought from 2012 to provinces, giving priority to Sindh’s right to access Indus 2015 had further serious impacts across the arid waters. This arrangement remained in place until 1970. zone (WFP 2015). Subsequently, the federal government began allocating The institutional arrangements for drought planning are water on an ad hoc basis, leading to ongoing disputes. well defined and managed by the National Disaster Numerous commissions failed to reach agreement, Management Authority (NDMA). It is hampered, until the four provinces agreed to the 1991 Water however, by a lack of capacity, and its functions do not Apportionment Accord. include basin-scale water sharing arrangements during The Accord is focused on the distribution of canal periods of extreme scarcity. The extent to which NDMA water entitlements between the provinces. Rather effectively advises decision makers on droughts is than explore hydraulic or economic optimality, it unclear. Given the far-reaching impacts of droughts on specifies and protects existing uses of canal water Pakistan’s economy and society, more attention needs for each province. It also recognizes the importance to be paid to drought planning, including development of an environmental flow allocation and provides and implementation of a drought forecasting system. guidance on how the balance of river supplies—above Sánchez-Triana et al. (2015) recommend a three- the baseline allocation volumes—should be shared. tiered drought forecasting system based on rainfall The Accord notes proposed environmental flows, predictions including (i) probabilistic one- to 15-day but the signatories did not agree on quantity or rules forecasts at 25-kilometer resolution for agriculture and and deferred the decision to further studies (Anwar and water resource management; (ii) probabilistic 15- to Bhatti 2018). 30-day forecasts using extended monthly forecasts statistically adjusted for intraseasonal variability; and An international panel of experts recommended a (iii) probabilistic one- to seven-month national drought continuous minimum environmental flow with an forecasts. occasional larger flow (Gonzalez et al. 2005). Gippel (2015) is critical of this recommendation because it Drought planning should include medium- and was not based on scientific analysis and notes the long-term measures that increase resilience to diverse environmental flow objectives and lack of drought and improve water management outcomes, robust scientific study to determine the flows required including intercropping for soil diversification and to meet specific environmental objectives. The Accord amelioration, drought-resistant crops, soil water thus makes no specific allowance for environmental accounting, and conservation agriculture. Farmers flows. Current flow to the delta is essentially a default in Sindh are more affected by drought than those unmanaged environmental flow of seemingly marginal in Punjab because of limited access to groundwater benefit. This remains a major shortcoming of water (FAO 2016). More equitable water sharing during basin-scale management in Pakistan. extreme scarcity is required to compensate for this disparity. The Accord shares a baseline volume of 144.749 billion cubic meters per year between the provinces in the following approximate shares: Punjab, Water Allocation 48 percent; Sindh, 42 percent; Khyber Pakhtunkhwa Water allocation refers to the rules and procedures (KP), 7 percent; and Balochistan, 3 percent. Anwar that define access to water in relation to availability. and Bhatti (2018) estimate that this volume has This section examines existing water allocation been available in 90 percent of the years on record. mechanisms between and within the provinces The Accord also indicates—in appendices agreed of Pakistan. Pakistan’s water allocation framework after signing—how these shares will be allocated provides a basis for water sharing and water access— across 21 separate irrigation systems. This sharing in the absence of any legal system of water property is summarized in Anwar and Bhatti (2018). In the rights. Despite providing some level of certainty to 10 percent of years when the available water is water users, the framework fails to ensure equitable below the baseline volume, the shortfalls need to be and efficient water delivery, is not sufficiently shared between provinces. While the Accord gives transparent, and fails to adequately address some guidance on this issue, the provinces have environmental sustainability. different interpretations of this guidance, and this 82 PAKISTAN: GETTING MORE FROM WATER remains an area of interprovincial dispute. As flow Figure 5.1  Relationship between Annual variability increases with climate change, and as the Irrigation Shortfall and Total Annual Inflow in intersectoral balance of water demands change, more Pakistan, 1992–2015 sophisticated and economically efficient approaches 50 to water sharing during periods of temporary scarcity will become increasingly urgent. Further, a key aspect of this improved drought planning will be clarity on 40 Annual shortfall (BCM) appropriate environmental objectives, environmental water allocations, and necessary protections for these 30 allocations. The Accord provides guidance for the sharing of water 20 above the baseline volume in the wettest 10 percent of years: Punjab and Sindh receive 37 percent each, KP, 10 14 percent; and Balochistan, 12 percent. This sharing is less contentious than that of shortfalls, because in 0 excess years, KP and Punjab, in particular, typically 100 150 200 250 receive higher than average rainfall, and thus demand Annual inflow (BCM) for additional irrigation supply in these provinces in these years can be low. Source: WAPDA unpublished data and author calculations. Note: Shortfalls are withdrawals are relative to baseline allocation Since the adoption of the Accord, actual total annual volume. canal withdrawals have averaged 19.7 billion cubic meters (or 13.6 percent) below the baseline allocation allocation; its annual shortfall has varied between volume. In low inflow years, withdrawals are of 25 percent and 53 percent. Until 2005, KP was unable course constrained by supply volume; however, in to use its full allocation, and its annual “shortfall” wetter years withdrawals are lower than full allocation varied from 42 percent to 50 percent. Since 2005, because rainfall in the command areas reduces when the Pehur High-Level Canal was commissioned— canal water demand, or because of canal capacity bringing an additional 5,000 hectares under irrigation— constraints. Anwar and Bhatti (2017) suggest that canal the KP annual shortfall has averaged 11 percent. capacity constrains affect withdrawals for Punjab and Sindh in years when system inflows exceed around An analysis of the annual shortfalls between provinces 162 billion cubic meters and 167 billion cubic meters, shows that while Punjab has the greatest allocation respectively. Additionally, withdrawal shortfalls reflect share, it has had a lower share of annual shortfalls the aggregate outcome of the incremental process because it is better able to use its full allocation through the year (on a sequential 10-day basis) of (figure 5.2). In 1998 and 2011, Punjab withdrew assessing irrigation demand, announcing allocations, more than its baseline allocation volume (reflecting and making reservoir releases, allowing for river gains allocation of excess water), while other provinces and losses. Operationally, a high priority is placed on withdrew less than their baseline allocations. In 1998, achieving a carry-over storage volume at the end of rainfall and inflows were above average, and KP was the rabi season to meet early kharif demands. This not equipped to use its full allocation. In 2011, inflows places another constraint on allocations and may lead were below average, and Sindh went over 90 percent to more conservative allocations earlier in the year. The of the total shortfall (equivalent to around 17 percent Accord does not specify detailed operating rules. These, of the basin baseline allocation volume). Although this however, have evolved over time, and the current is only a partial picture of interprovincial sharing, it (post-2003) approach is typically presented by IRSA as suggests there remain equity issues in implementation a “three-tier rule” (Anwar and Bhatti 2018) for low, of the Accord. medium, and high levels of annual water availability From an economic efficiency perspective, Yu et al. defined in terms of levels of historic withdrawal over a (2013), p7, conclude that the Accord is suboptimal, specified period. and that relaxing it and implementing an economically Canal withdrawal shortfalls relative to the Accord’s based water allocation mechanism would benefit both baseline allocation volume has been lowest in years Punjab and Sindh and enable the provinces “to better around the median annual flow (figure 5.1). At manage extreme events by more reliably meeting the provincial level, withdrawal shortfalls in some system-wide demands.” This view is consistent with cases simply reflect an inability to use the available other model-based assessments of the impacts allocation. Because of topography and limited of more flexible water allocation, which suggest infrastructure, Balochistan is unable to use its full increased flexibility would increase agricultural profits 83 Figure 5.2  Share among Provinces of Annual Shortfall of Canal Withdrawals Relative to Water Apportionment Accord Baseline Allocation Volume in Pakistan, 1992–2015 100 80 60 Percent 40 20 0 –20 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 19 19 19 19 19 19 19 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Punjab Sindh Khyber Pakhtunkhwa Balochistan Source: WAPDA unpublished data and author calculations. by 2.5 percent to 5.0 percent (Yang et al. 2014). Yang generally lower inflows of the last 15 years have et al. (2014) suggest, however, that the largest gains meant reduced outflows. Additionally, system losses in water allocation efficiency can be achieved through have increased significantly. Because these losses water transfers and reallocation within provinces. are calculated as a water balance closure term, they combine errors in measurements (including falsification Moving to a more equitable and more economically of withdrawal records), water theft, and natural system efficient approach to interprovincial sharing requires losses. One factor that may explain a fraction of the advanced analytical approaches, including detailed increasing losses over this period is climate warning modeling of water distribution, and ideally informed in the Lower Indus Basin. Figure 5.4 compares system by improved flow monitoring and inflow forecasting. losses as a percentage of the total balance to the mean This in turn requires technical capacity strengthening, annual temperature anomaly (departure from the long- especially of IRSA, but also of WAPDA and the term mean) at Karachi: both show a significant and Pakistan Meteorological Department (PMD) for inflow increasing trend. forecasting. These advances would help improve the trust of the provinces in the interprovincial water The basin water balance is clearly changing, which allocation process. means the sharing arrangements of the Accord will increasing be tested, and their suboptimality will The Accord does not constrain the provinces in how be increasingly challenged. Because the Accord is a they use the allocated water. It does, however, require consensus agreement between provinces and not a existing reservoirs to be operated to prioritize irrigation. federal legislative instrument, it is extremely difficult While it acknowledges industrial and urban water to negotiate changes to the existing procedures. demands, the lack of specific provisions leaves the For example, the political process required to get all issue of intersectoral allocation for provinces to address. four provinces to agree to forsake a fraction of their This is especially problematic for Sindh, given the size water apportionment to augment supplies to the twin and rate of growth of Karachi and its importance to the cities of Rawalpindi and Islamabad took many decades national economy, and the lack of viable alternative (Anwar and Bhatti 2017). Nonetheless, as systemwide water supplies beyond limited groundwater and demands grow, and climate change increases the internal runoff. variability of system inflows, the limitations of Both intersectoral and interprovincial tensions are The Accord and its inflexibility will increase the typically greatest during drought, as observed in vulnerability of the Indus Basin Irrigation System 2000–01 and early 2018. A basin-level view of (IBIS) and Pakistan’s water sector more broadly. The how annual inflows are partitioned between canal complexity of intergovernmental politics means this withdrawals, outflows below Kotri Barrage, and system situation will not be easy to tackle without new losses (figure 5.3) shows that while withdrawals are federal legislation to articulate improved water sharing remarkedly constant (suggesting adequate system mechanisms and to empower and strengthen IRSA to storage)—only falling during the worst droughts—the implement these. 84 PAKISTAN: GETTING MORE FROM WATER Figure 5.3  Annual Series of Total Canal Withdrawals, Outflow below Kotri Barrage, and Losses as Components of Indus Basin Inflows, Pakistan, 1975–2015 250 200 Cubic meters (billions) 150 100 50 0 1975 1980 1985 1990 1995 2000 2005 2010 2015 Canal withdrawals Outflows Losses Source: WAPDA unpublished data. Note: Losses calculated as water balance closure, and inflows refer only to rim station flows. Figure 5.4  Annual Water Balance Closure and Mean Annual Temperature Anomaly for Karachi, Pakistan, 1975–2015 30 2.0 1.5 20 Degrees (Celsius) 1.0 Percent 10 0.5 0 0 –0.5 –10 –1.0 1975 1980 1985 1990 1995 2000 2005 2010 2015 Water balance closure (%) Mean annual temperature anomaly, Karachi Source: WAPDA unpublished data; author calculations; and temperature data from http://berkeleyearth.lbl.gov/auto/Stations/TAVG/Text/157348​ -TAVG-Data.txt. Intraprovincial Water Allocation that increased flexibility in surface water allocation within provinces—both within agriculture and between Within provinces, there are no established processes sectors—can increase agricultural profits and improve for formally allocating water to sectors or reallocating outcomes for domestic, industrial, and environmental between sectors. Historically this has not been water users (Yang et al. 2014). Establishing processes required, because the nonirrigation water demands for intersectoral reallocation will be best achieved have been comparatively small and, in many cases, by restructuring provincial irrigation departments as are met largely by groundwater. However, as Pakistan’s agencies responsible and empowered to define and economy and population grow, competition for manage (through planning and operations) water for water between agriculture, industry, households, multiple outcomes. and the environment is increasing. Because options for supply augmentation are very limited, water will Provincial government departments manage need to be reallocated from agriculture to industrial irrigation water allocations, and implement these and domestic sectors, where its economic value is through operation of barrages, major canals, and the highest. Ensuring reliable urban supplies, especially distributary network. Water orders (or indents) are during periods of extreme scarcity, requires more communicated by provincial irrigation authorities to flexible and responsive mechanisms for intersectoral IRSA every 10 days, which inform the release of water water allocation. Model-based assessments indicate from headwater reservoirs and the distribution of water 85 through the link canals and major canals, according to could increase economic benefits (OECD 2017). For the detailed sharing arrangements of the appendices the Indus, modeling suggests multipurpose operations to the Accord. For significant periods of the year, could increase economic benefits by up to 20 percent however, water orders far exceed the available water (FoDP 2012). and the capacity of the irrigation system. The irrigation Revised reservoir operations could improve flood system is thus often operated at capacity. Given the mitigation. Following the 1992 flood the standard supply constraints, the IBIS is supply-driven rather operating procedures for Tarbela and Mangla were than demand-driven (Shah et al. 2016), and shortages revised for flood mitigation (GoP 2018a). This helped mean some land is often left fallow, and pressure on to mitigate the 2010 flood peak: Mangla operations groundwater continues to increase. reduced the flood peak in the Jhelum by 35 percent, The lowest level of water allocation follows the and Tarbela operations reduced the flood peak in traditional warabandi system. Warabandi is the the Indus by 28 percent (Ali 2013). However, these weekly schedule below the canal turnout (outlet) reductions did not significantly reduce flood damages whereby water is distributed sequentially among (Tariq and van de Giesen 2012). Noncompliance with land holdings, for durations proportional to the area standard operating procedures led to levee breaching of the land holding. Included in warabandi are canal near the Jinnah and Taunsa barrages with serious operations plans (or rotational plans) that are typically impacts in Punjab (Shah, Shakir, and Masood 2011). applied at the tertiary (distributary) canal level. For Recognizing the importance of fully capturing the any given week in a cropping season, these plans flood mitigation potential of large reservoirs, the determine which tertiary canals remain open and Pakistan Ministry of Water Resources has indicated the which remain shut. “strong need to review and improve Tarbela’s existing operating policy to provide more flood mitigation relief The warabandi system acts as a constraint to allocative to downstream areas” (GoP 2018a). Operating the efficiency. Economically efficient allocation means headwater dams for greater flood mitigation requires all farmers receive equal marginal net benefits from improved hydrological forecasts—better skill and longer irrigation water (Akram 2013). Significant gains lead times—and processes to integrate these forecasts in total agricultural profits—in both Punjab and into reservoir operation. Sindh—could be achieved by implementing water allocation mechanisms that move water to those canal Neither Tarbela’s nor Mangla’s standard operating commands and crops that are relatively more profitable procedures consider environmental flow management (Yang et al. 2014). While such a reallocation would be or reservoir sedimentation. Environmental degradation economically beneficial overall, there would of course of the lower river and delta, including salinity intrusion be individual winners and losers. Reallocating water (chapter 2), indicates current inadequate environmental either by financial incentives (water pricing or markets) flows especially during rabi, as well as sediment or by government-managed compensation schemes deprivation of the delta. Current flows to the sea are is complex and would require capable, trusted, and primarily unregulated monsoon flows. Revised reservoir independently audited water institutions, supported by operating protocols could incorporate managed releases robust and transparent water accounting. A sensible for environmental flows during rabi. With the addition first step would be to establish robust monitoring of of Diamer Bhasha, enhanced storage capacity would water delivery across command areas, with all data enable these flows to be met with monsoon inflows shared openly and in near-time on multiple platforms with limited impact on irrigation supply. including mobile phones. Operating protocols of large dams influence sediment trapping, and even partial drawdown during flood System Operations seasons can increase sediment transport (Roca 2012). Operation of the major multipurpose reservoirs (Tarbela Rashid, Shakir, and Khan (2014) show that sediment and Mangla) affects water distribution, hydropower flushing is more technically and economically generation, and flood mitigation. Although reservoir feasible for Tarbela than dredging or trucking. Khan operation is one of the least discussed aspects of and Tingsanchali (2009) demonstrate alterative Pakistan’s water resources management, optimizing operating rules for Tarbela that reduce sediment operations may offer important opportunities for trap efficiency from 93 percent to 80 percent with improving water outcomes with minimal investment. impact on the reliability of irrigation supply. However, Tarbela and Mangla have been operated to maximize development of the river downstream of the dam irrigation water supplies, with energy generation and and the level of sediment accumulated may preclude flood protection as secondary objectives (Yu et al. such operations (Annandale et al. 2016). A detailed 2013). Global experience suggests that adopting a analysis of alternative reservoir management options dynamic, multipurpose approach to reservoir operation and downstream consequences is required to assess 86 PAKISTAN: GETTING MORE FROM WATER the feasibility of flushing and any trade-offs with preparedness and response were inadequate, and irrigation supply. that deviating from standard operating procedures caused flood levees at the Jinnah and Taunsa Advances in computer science and multi-objective barrages to breach (Shah, Shakir, and Masood 2011). optimization can inform design of reservoir operation Flood response plans should be based on improved rules that are resilient to a wide range of future forecasting and warning services. Evidence suggests climatic uncertainties and that balance multiple that public response to flood warnings is very weak objectives (Giuliani et al. 2016). Trade-off analysis (Mustafa et al. 2015). Additional trials are required of multi-objective optimization of reservoir cascades of community-based early warning systems that can identify rules that can enable environmental flow integrate improved forecasting and maximize releases while incurring impacts only for irrigation citizen participation. Capacity building for agency water security (Konrad, Warner, and Higgins 2012; staff is required in aspects of disaster prevention Krchnak, Richter, and Thomas 2009). This analysis and forecasting. Few staff members have the skills requires an advanced system modeling platform such required for sound barrage operation during floods or as IRSM (Stewart et al. 2018), complemented by for assessment of structural weaknesses in the levee appropriate economic analyses. system. Professional training in the use of forecasting Barrage operations are critical to system performance. tools and in transforming medium-range climate Although the primary function of the barrages is forecasts into actionable probabilistic hydrological irrigation supply, flood operations are also important forecasts is needed. for barrage safety. Inadequate maintenance and significant sedimentation upstream of the barrages Environmental Sustainability have compromised the flood performance of key barrages including Sukkur, making rehabilitation and Pakistan does little to protect water-dependent upgrading for better operational control critical. None ecosystems (rivers, lakes, and wetlands—including of the barrages were designed with consideration of the Indus Delta) by either water quantity or water environmental issues. Barrages fragment river habitats quality management, and the efforts to protect the into a series of ecologically disconnected reaches. quality of the water resource base are inadequate. Fish ladders were retrofitted to Muhammad and Kotri No environmental flow regime has been agreed or barrages but poorly designed for endemic species and implemented for the Indus River. As noted in chapter 3, have proved ineffective. Improved fish passages at key system outflows from Kotri Barrage—an indication of barrages should be investigated to offset some of the flow reaching the Indus Delta—have declined markedly environmental impacts of flow regulation, and there in recent years. Since 2000, annual system outflows should be a detailed analysis of environmental flow have averaged 18 billion cubic meters—just 10 percent options. of system inflows—and outflows during rabi season have averaged just 3 percent of rabi system inflows Postflood disaster operations are important to reducing and were zero for half of the years in this period flood impacts. Disaster response is supposed to be (figure 5.5). As discussed in chapter 3, reduction from coordinated between the NDMA and the Provincial inflows to outflows is not solely because of water Disaster Management Authorities (PDMAs). While withdrawal for consumptive use. As a semi-arid zone the NDMA mainly has policy, planning, and guidance river, the Indus is characterized by high natural losses functions, it is also supposed to coordinate response (or consumptive environmental water use—natural efforts during events. PDMAs coordinate provincial evapotranspiration) in the lower reaches. In absence of relief, compensation, and rehabilitation efforts. either prewater resource development measurement Historically, postflood operations have been weak in of outflows or detailed hydrologic simulation of an Pakistan. Deficiencies include lack of flood response unimpaired river flow regime for comparison, the plans to guide overflow on the floodplain and actual outflow reduction as a reduction on irrigation identify embankment breaching sites; lack of citizen water use is uncertain. Suffice to say, the reductions involvement; and limited access to early warning are very large, especially during rabi, and are clearly information for marginalized groups, including women environmentally unsustainable. (Mustafa et al. 2015). Developing flood response The Accord notes the need for an environmental water plans will require addressing political economy issues, allocation; while noting that Sindh has proposed an including political influencing of flood peak diversion annual volume of around 15 billion cubic meters, and embankment breaching sites. it does not specify an agreed or required volume. During the 2010 floods, the NDMA failed to adequately Current system management does not seek to deliver coordinate responses across the country (Sánchez- environmental flows, and flows to the delta are Triana et al. 2015). A judicial enquiry concluded flood primarily a combination of poor quality irrigation return 87 Figure 5.5  Indus River System Outflows from Kotri Barrage as Share of System Inflow, by Season, 1975–2015 100 80 60 Percent 40 20 0 1975 1980 1985 1990 1995 2000 2005 2010 2015 Rabi Kharif Source: WAPDA unpublished data. flows during rabi and unregulated high flows during current conditions, less than 5 percent of annual kharif, especially in wetter years. Efforts have been outflows are recorded during this season. Once Diamer made to specify environmental flows, including three Bhasha Dam is operational, providing greater capacity studies completed in 2005 for the FFC to guide the to store and regulate Indus inflows, it is likely that flows envisaged by the Accord, and an international in the absence of an agreed environmental flow panel of experts (Gonzalez et al. 2005), which regime and the institutional capacity to deliver this reviewed these studies. More recently a comprehensive effectively, end-of-system flows will be further eroded review was undertaken for WWF-Pakistan by Gippel given an enhanced ability to meet rabi irrigation. The (2015). The international panel of experts (Gonzalez additional operational control that Diamer Bhasha’s et al. 2005) recommended around 31 billion cubic storage capacity will provide could enable the current meters per year, on average, with a constant flow end-of-system flow volume to be managed more equivalent to 4.4 billion cubic meters per year and effectively, with a fraction purposefully delivered to the periodic larger pulses, managed over a five-year delta during Rabi as an environmental base flow, with accounting period. However, Gippel (2015) notes there periodic environment pulses released and managed are widely divergent recommendations among the through the system. Even better, and as recommended studies for environmental flows, reflecting divergent or by Gippel (2015), a more comprehensive unclear environmental objectives, different methods environmental flow analysis is required that explicitly and assumptions, a lack of good data on which to links environmental flow regime options with base analyses, and many scientifically unsupported ecological health and ecosystem services outcomes, recommendations. Gippel (2015), p93, notes that using a process that engages all stakeholders to the work undertaken for the FFC was “impressive in ensure the trade-off involved in supplying water its breadth of coverage and extensive reporting, but for irrigation and other consumptive uses are well disappointing in the number of errors, inconsistency in understood, to avoid unrealistic expectations that the data, vaguely stated or non-existent environmental modest environmental flows will ever restore the flow objectives, and flimsy recommendations.” He entire delta to ecological health. While there are further notes (p93) that the subsequent international technical and scientific and challenges in undertaking panel of experts did not adopt these recommendations, such an analysis, the main barriers are institutional, but instead “recommended a flow regime that was not including not acknowledging the problem. Defining supported by any scientific analysis.” environmental flows needs to be a government-led process with inputs from all stakeholders, driven by The current annual average end-of-system volume is a shared recognition of the importance of improved similar to that proposed by Sindh. The environmental environmental sustainability in water resources benefits achieved by this flow, however, are likely to management. be minimal—with great environmental stress for much of the year—because it is primarily a short period of Seawater intrusion in the Lower Indus, compounded unregulated high flows during kharif, typically with by the lack of fresh water below the Kotri Barrage, several months of essentially zero flow during rabi. is degrading water-dependent ecosystems and Indicative of this flow regime change is that while agricultural productivity. Seawater intrusion has caused 18 percent of annual inflows occur during rabi, under vast areas of agricultural land to become unsuitable 88 PAKISTAN: GETTING MORE FROM WATER for farming and some has even disappeared into the rather than a crisis. However, this reflects the specific sea (Majeed et al. 2010). Coastal Sindh (especially the opportunities and social circumstances of this location Badin, Sujawal, and Thatta districts) is more vulnerable and is not a generic conclusion. to seawater intrusion than coastal Balochistan. In Balochistan, groundwater use exceeds recharge by Seawater has penetrated 30–50 kilometers inland in an estimated 22 percent (Halcrow Group 2007) with some coastal areas of Sindh (SCCDP 2012), which has overexploitation occurring in 10 of 19 subbasins. In affected groundwater quality, agricultural productivity, the Pishin Lora Basin, where abstraction is four times and the livelihoods of some of the poorest populations the recharge rate, pumping has entirely depleted the (Memon and Thapa 2011). In the coastal belt of shallow alluvial aquifer, and new deep wells with Makran in Balochistan, the Gwadar District is affected powerful electric pumps have been installed to access in parts by seawater intrusion, which is degrading the underlying fractured rock aquifer (van Steenbergen groundwater quality and exacerbating the already et al. 2015). In the Pishin Lora, overexploitation has led extreme water scarcity of one of the poorest and most to neither conflict nor cooperation for more strategic, underdeveloped regions of Pakistan. sustainable, and productive use. Rather, in the absence Pakistan is often considered a global hotspot of of any intervention, urbanization and changing groundwater depletion. Uncontrolled groundwater employment options simply saw a gradual shift away pumping, mostly via diesel-fueled private tube wells, from high-value, low-cost horticulture to a less lucrative has contributed to groundwater depletion in the production system (van Steenbergen et al. 2015). Indus Basin, particularly in the Punjab, home to about Groundwater depletion in urban areas is likely to 90 percent of the private pumps. Assessments based represent the greatest challenge, because it affects on satellite observations have sought to highlight this the largest number of people for whom access to issue; however, published estimates of groundwater alternative water is limited. Groundwater levels depletion vary widely. In a recent and high-profile around Lahore have been falling at by 0.5–0.8 meters global modeling-based analysis, Dalin et al. (2017) rank per year since the mid-1960s because of increased Pakistan third in the world in terms of groundwater abstraction in response to a loss of reliable supply depletion, citing an unbelievable depletion rate of from the Ravi River. Recharge has been reduced nearly 28 billion cubic meters per year based on because of flow reductions in the Ravi River (Mahmood coarse-scale global hydrologic modeling. This estimate et al. 2013). Ravi flow reductions reflect upstream is supposedly validated against coarse-scale Earth development in India as permitted under the Indus observations (400 kilometer by 400 kilometer NASA Waters Treaty. Mahmood et al. (2013) demonstrate GRACE satellite data). For Pakistan this validation has that the groundwater cone of depression under Lahore relied on an even coarser regional estimate for a much expanded from 2004 when depth to groundwater was larger region spanning three states in India. A more less than 38 meters everywhere. By 2011, depth to realistic regional assessment by MacDonald et al. groundwater exceeded 38 meters across 150 square (2016) based on in situ measurements suggests a kilometers. A daily average volumetric groundwater current net annual depletion rate for the entire Indo- balance for the Lahore aquifer suggests abstraction Gangetic basin of 8 billion cubic meters per year (plus exceeds recharge by around 10 percent (Qureshi or minus 3 billion cubic meters), with groundwater and Sayed 2014). The aggregate annual depletion levels stable or rising over 70 percent of the assessed volume is estimated to be around 0.25 billion cubic area and falling over 30 percent. As indicated in the meters, leading to a fall in average groundwater levels water balance analysis of chapter 3 (and appendix A), of 0.55 meters per year. Qureshi and Sayed (2014) annual groundwater depletion across the Indus Basin suggest that demand management (through education, of Pakistan, appears to be of the order of 1 billion regulation of groundwater pumping, and water pricing) cubic meters. This is a small fraction (about 2 percent) and supply enhancement (including managed aquifer of the annual groundwater balance. As highlighted recharge, rainfall harvesting, and canal water) could in map 3.2, depth to groundwater across most of the ensure water security and sustainable use for Lahore. basin is less than 1.2 meters, with waterlogging and salinization major concerns. Groundwater depletion Surface water and groundwater quality across Pakistan is a not basinwide concern, but is largely confined has deteriorated significantly because of point and to Punjab and Balochistan, in agricultural hotspots, nonpoint source pollution. Pollution sources include such as the Khanewal Division in Punjab (MacDonald untreated domestic effluent, agricultural drainage et al. 2016) and the Kuchlagh region of Balochistan contaminated with pesticides and fertilizers, and (van Steenbergen et al. 2015), and in urban areas, such unregulated industrial effluents containing toxic as Lahore and Quetta. While serious, the exhaustion chemicals. In addition to these anthropogenic sources, of the aquifer in the agricultural region of Kuchlagh is naturally occurring arsenic is increasingly contaminating reported as one of adaptation to suboptimal outcome, groundwater across much of Pakistan. 89 For Lahore, several outfall drains discharge untreated agrochemicals each year, of which an estimated 10,000 municipal and industrial effluent to the Ravi River die—many as a result of exposure to contaminated (Qureshi and Sayed 2014), which, combined with water (Shahid et al. 2016). The eastern tributaries streamflow reductions, contributes to the deterioration of Indus—the Ravi and Sutlej—provide very limited in measured water quality downstream from this point wastewater dilution capacity of wastewater (because and to the deteriorating quality of groundwater (Hassan these waters are allocated to India), and metal and et al. 2016). Many components of this effluent will not microbiological contamination in these two rivers (and deteriorate with passage through sediments before nearby groundwater) is ubiquitous (Grigg et al. 2018). reaching the water table and thus will contaminate the Groundwater quality has been deteriorating because groundwater for drinking water and irrigation for years of salinity and contamination by agriculture and to come. industry (MacDonald et al. 2016). The area affected Inadequate solid waste management, uncontrolled by salinization has increased because of surface wastewater discharge, and leakage of sewage water and shallow groundwater evaporation and result in microbial contamination of drinking water excessive pumping, which have mobilized older supplies in all major cities of Pakistan. In some saline groundwater, especially in Sindh (MacDonald cases, microbial contamination may even be linked et al. 2016). Improved water management will be to improper filtration at water treatment plants essential to prevent further groundwater salinization, (Azizullah et al. 2011). In rural areas, open dug wells in addition to changes in agricultural and industrial and low water tables mean that water supplies are practices to prevent contamination. Although often contaminated with fecal matter (Raza et al. widespread, degradation of groundwater quality is 2017). The surface water quality situation has been less well recognized, and yet is great concern for the deteriorating, and it is worst during dry months. Many sustainability of this important resource. of the major cities are located along the rivers, which Current water quality management is grossly directly receive untreated municipal and industrial inadequate. Unless prevention and control measures wastewater. An estimated 95 percent of shallow are taken, water pollution will increasingly affect groundwater supplies in Sindh are bacteriologically the health and productivity of people, especially the contaminated (PCRWR 2004). Discharge of untreated poorest households. In the short term, better regulation wastewater into irrigation canals is increasingly of fertilizer and pesticide use and industrial discharges common, and these canals are widely used for rural is required. This would build on the interim national drinking water supply. environmental quality standards that are realistic An estimated one in every six industries in Pakistan for most polluters to meet. In the medium to long are heavily polluting (Sial et al. 2006), the worst being term, more stringent water quality standards and textile and leather factories, agroprocessing factories improved monitoring should be adopted. Monitoring (including oil and sugar mills), and petrochemical is critical for targeting interventions for in areas of factories. These industries are located close to or in most concern. These efforts should be supported by major cities, which have had contamination episodes institutions capable of enforcing quality standards and (e.g., Ahmed 2015; Guriro 2016). An estimated able to work across sectors (especially agriculture and 1 percent of industrial wastewater is treated only industry) to implement and finance interventions. prior to discharge (Azizullah et al. 2011). Untreated Significant improvements in water quality are not industrial effluent seriously degrades surface water possible without better management of industrial and groundwater. Lead, chromium, and cyanide have discharges. This will require a major change for most been detected in groundwater near Karachi and in the manufacturers and agribusinesses, because very few Layari and Malir rivers, which flow through Karachi to have treatment facilities. Increasingly, international discharge into the Arabian Sea (PCRWR 2002). Chemical firms supplied by Pakistani manufacturers are oxygen demand of rivers exceeds the national demanding that environmental factors, including environmental quality standard, in some cases by more water quality, are considered. These pressures can be than 500 percent. expected to encourage investment in the wastewater The use of pesticides and agrochemicals is increasing, treatment facilities required to enable compliance and residues have been reported in waters in several with national regulations and to help firms remain parts of Pakistan. Currently, an estimated 5.6 million competitive in international markets (Sánchez- tonnes of fertilizer and 70,000 tonnes of pesticides Triana et al. 2015). Construction of common effluent are used in Pakistan annually (Daud et al. 2017), treatment plants to serve industrial clusters and cleaner and a significant fraction reaches surface water or production methods that minimize the generation groundwater (e.g., Ahad et al. 2006; Shahid et al. of wastewater can help improve performance. As 2016). Around 0.5 million Pakistanis are poisoned by institutional and monitoring capacity are strengthened, 90 PAKISTAN: GETTING MORE FROM WATER and as treatment infrastructure is put in place, pollution agrarian economies (figure 5.6). Pakistan’s water charge schemes could be introduced to incentivize productivity is 35 percent of the average for this cohort. pollution control at source and to generate revenues for (The double counting inherent in the withdrawal value provincial environmental protection agencies. [see chapter 1] does not affect Pakistan’s ranking in this cohort.) Water Productivity For water scare countries, the overall economic Total water productivity, that is, the economic output productivity is water is important. Additionally, per unit of water withdrawn from the environment, is the water productivity of agriculture is of interest, low in Pakistan compared to most other countries. On especially in countries with significant irrigation. On 2015 data, Pakistan ranks eighth lowest in the world, the same 2015 data, Pakistan’s agricultural water generating just US$1.38 per cubic meter of water productivity is US$0.37 per cubic meter of water withdrawn. Pakistan ranks third lowest within a cohort withdrawn—again ranking third in the selected cohort of countries with (i) more than 80 percent of water (figure 5.7). Adjusting for the double counting in the use in agriculture; (ii) agriculture more than 10 percent withdrawal value would improve Pakistan’s ranking of gross domestic product (GDP); (iii) less than 3,000 to close to the middle of this cohort—similar to India, cubic meters of water available per capita annually; Zimbabwe, and the Arab Republic of Egypt—with a and (iv) GDP per capita of between US$800 and productivity value equivalent to 62 percent of the US$4,000—low-income but not the poorest rain-fed cohort average. Agricultural water productivity in Figure 5.6  Total Economic Productivity of Water in Selected Countries 10 Water productivity (US$ per m3) 9 8 7 6 5 4 3 2 1 0 an n n ic a e p. l ia n . co iti ia ga ep di bw ta ta da bl an an Re Ha oc ist In ne ,R kis kis pu Su rit nz ba or jik ab Se en be Pa Re au Ta M Zim Ta Ar m Uz M ab Ye t, Ar yp Eg ria Sy Source: FAO 2015. Figure 5.7  Agricultural Water Productivity in Selected Countries 2.0 Water productivity (US$ per m3) 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 an an an ic a e p. al ia n . co iti ia ep di bw da bl an an g Re Ha oc ist t ist In ne ,R kis pu Su rit nz ba or jik k ab Se en be Pa Re au Ta M Zim Ta Ar m Uz M ab Ye t, Ar yp Eg ria Sy Source: FAO 2015. 91 Pakistan is thus very low, even within a cohort of Using crop production data (tonnes), estimates of low-income countries with low levels of investment in total crop water requirement from Linstead et al. high-technology irrigation. (2015) and modeled irrigation water use by crop (see chapter 6), allows green water (rainfall) and blue water In the absence of good data on water use by crops, (irrigation water) use for the major crops in Pakistan to robust comparisons of crop-by-crop water productivity be estimated. Blue water footprints (cubic meter per between countries is not possible. Some inferences tonne) can then be determined. Because sugarcane can be made on the basis of crop modeling and data has such a high moisture content at harvest, using on cropped areas, and these are explored in chapter harvested tonnage is misleading; therefore, the water 6 based on results from CGE modeling. However, footprint for raw sugar is estimated for comparison measures of agricultural productivity on an area basis with other major crops (figure 5.9). This reveals can be used to compare between countries (figure 5.8, that cotton, although the best performer on an area panels a–d). productivity basis, it is the most water thirsty of these Pakistan’s performance on an area basis is best for major crops, requiring around 2,500 cubic meters per cotton, for which productivity is equivalent to the tonne of crop produced. world average and markedly better than in India. For the other major crops Pakistan’s productivity is CGE modeling for Pakistan suggests that given the significantly below the world average, although its rate crop irrigation demands and areas typically grown, of productivity improvement over the last five decades rice consumes around 32 percent of the water used largely mirrors the global trend. Australian wheat by these four crops; wheat and cotton both consume productivity is comparatively low, because it is mostly around 25 percent; and sugarcane, 18 percent. Around a dryland crop rather than irrigated; the interannual half of the rice crop (and 5 percent to 10 percent of fluctuations reflect rainfall variability. Australian rice the sugarcane crop) is exported, thus presenting a crops are highly productive, being high quality for niche very significant virtual water export. Growing low export markets. High interannual variability reflects productivity paddy rice for export in an arid, water the flexible nature of the rice industry, which relies on scarce country does not make good economic sense. low reliability water licenses that do not yield water in Reforms and investment are required to move this dry years. water to higher-value crops (fruit and vegetables) for Figure 5.8  Economic Productivity of Major Crops for Selected Countries and Globally, 1961–2016 a. Wheat b. Rice 1.2 3.5 Wheat productivity (US$/ha) Rice productivity (US$/ha) 1.0 3.0 2.5 0.8 2.0 0.6 1.5 0.4 1.0 0.2 0.5 0 0 1960 1970 1980 1990 2000 2010 2020 1960 1970 1980 1990 2000 2010 2020 c. Sugarcane d. Cotton Sugarcane productivity (US$/ha) 4.5 7.0 Cotton productivity (US$/ha) 4.0 6.0 3.5 5.0 3.0 2.5 4.0 2.0 3.0 1.5 2.0 1.0 0.5 1.0 0 0 1960 1970 1980 1990 2000 2010 2020 1960 1970 1980 1990 2000 2010 2020 Australia China Egypt, Arab Rep. India Pakistan World Source: FAO 2016. 92 PAKISTAN: GETTING MORE FROM WATER Figure 5.9  Estimated Average Blue Water green water footprint for cotton grown in Sindh is (Irrigation) Footprints of Wheat, Raw Sugar, Rice, more than 20 percent higher than for Punjab, and and Cotton for Pakistan the blue water footprint is over 65 percent higher 3,000 in Sindh because of lower rainfall. In Punjab, where most of the cotton is grown, groundwater is key to irrigation—more than 20 percent of the irrigated area 2,500 Blue water footprint (m3/tonne) receives only groundwater, and 55 percent of the irrigated area receives canal water and groundwater. 2,000 Punjab groundwater use exceeds recharge, and thus groundwater levels are falling in parts of the province. 1,500 Around a quarter of the groundwater depletion in Pakistan is associated with agricultural exports, of 1,000 which cotton represents a significant fraction. Cotton growing and cotton textile production also have a gray water footprint—the dilution volume required to bring 500 irrigation drainage water (polluted with agricultural chemicals) or textile processing effluents to a quality 0 suitable for subsequent use. Close to 20 percent of the Wheat Raw sugar Rice Cotton estimated total water footprint of growing cotton in Sources: FAO 2015; Linstead et al. 2015. Pakistan is the gray water footprint. The wet processing and finishing of cotton yarn into textiles consumes a small volume of blue water, but has a large gray water export and to meet the growing domestic demand. footprint. Between one-third and a half of the total These issues are explored further with the modeling water footprint of producing cotton textiles is the gray work in chapter 6. water footprint of textile production from yarn. The Cotton—with the highest blue water footprint of gray water footprint of textile production would be the Pakistan’s major crops—is an important component of easiest fraction of the overall water footprint of cotton the Pakistani economy. Pakistan is the world’s fourth to reduce through cleaner production technologies and largest cotton producer. Cotton occupies 14 percent effluent treatment. Comparing water use in Pakistan of the cropped area, employs 20 percent of the cotton production to global averages, WWF (2015) agricultural workforce, and represents nearly 12 percent concludes that while the overall water footprint per of the agricultural value-add. Around 1.3 million tonne is close to the global average, because of the farmers grow cotton in Pakistan, mostly on small lower green water input, the blue water footprint is holdings less than 5 hectares. Around 80 percent 165 percent of the global average, and the gray water of the crop is grown in Punjab and 20 percent in footprint is 162 percent of the global average. Sindh. In recent decades, Balochistan has begun to Mekonnen and Hoekstra (2011) review agricultural grow cotton, but produces less than 1 percent of the water footprints by country. They show the blue national crop. Pakistan’s textile industry is supported by water footprints for wheat and sugarcane (raw sugar domestically grown and some imported cotton. Textile equivalent) in Pakistan are around four times the world production is the largest industrial sector in Pakistan, average, and for rice, more than six times the world employing 40 percent of the industrial labor force, and average. Pakistan ranks second highest in the world generating 25 percent of industrial GDP and 57 percent for the blue water footprints for wheat and sugarcane of exports by value. The cotton crop is separated (raw sugar equivalent), and seventh highest for rice. into cottonseed and cotton lint, with most of the Chapagain and Hoekstra (2010) review the water value in the lint, which is spun into yarn from which footprints of rice production across the top 13 rice textiles are produced. Annual lint production varies growing countries of the world. Pakistan grows just but has increased over the last 20 years from around 1.2 percent of the global rice crop. However, Pakistan 1.5 million tonnes to 2.4 million tonnes. produces two-thirds of the global crop of basmati rice, The blue water footprint of cotton in Pakistan is around and this is Pakistan second-largest export earner after double the global average; conversely its blue water cotton textiles. The Chapagain and Hoekstra (2010) productively (weight produced for given volume of water footprint analysis distinguishes between the irrigation water) is around half the global average. evaporative water loss associated with paddy rice and The blue water footprint of cotton varies by province, the water that percolates into the soil for crop use. partly because crop water requirements vary with Paddy rice in Pakistan uses 2.8 times the average climate, and because of the different levels of green irrigation water use across the major rice growing water (rainfall) availability. The combined blue plus countries, and the evaporative water loss from rice in 93 Pakistan is more than four times the average of the volume of water to cotton in support of the textiles major rice growing countries (figure 5.10). industry. The System of Rice Intensification (SRI) has been Farm size affects the productivity of the major crops used with some success in India to reduce water use. in Pakistan (Ahmad et al. 2014a). It influences It has been introduced to Pakistan but has not been the extent to which practices are adopted and the adopted widely. SRI is not a fixed package of technical system-scale effectiveness of these practices in terms specifications, but a system of production spanning soil of overall water use. Farm sizes in Pakistan are mostly fertility management, planting method, weed control, less than 5 hectares. In recent decades fragmentation and water (irrigation) management. Critically, SRI aims of land holdings has increased the proportion of very to keep the root zone kept moist, not submerged, small farms (less than 1 hectare in area) (figure 5.11, using intermittent water applications. Interesting recent panel a). By aggregate area, around half the total innovations in rice cultivation trialed in the United Arab farmed area comprises farms between 3 hectares Emirates that might hold some promise for a very and 20 hectares (figure 5.11, panel b). The fraction different rice industry in Pakistan include the use of of area associated with very small farms (less than hydroponics and salt-tolerant rice cultivars developed 3 hectares) is increasing, as is the aggregate area by Chinese scientists. from the largest farms (greater than 60 hectares). The aggregate area from farms between 3 hectares Ultimately, while both cotton and rice are major and 60 hectares is thus declining. export earners for Pakistan, the water performance of these crops is very poor compared to that of other Farmers of smaller holdings tend to have less access countries; combined, they account for well over half to machinery, and being poorer, are typically less the total irrigation water use of Pakistan. For a water likely to be able to invest in water efficient irrigation scarce country, directing over half of the water used technologies. However, adoption of water saving to water-intensive crops that are not essential for methods (such as zero tillage) may deliver water domestic food security and that deliver comparatively savings for farmers of smaller holdings, because they poor economic return is not a good long-term option. have less opportunity than farmers of larger holdings to The large volumes of water used in irrigation beyond increase cropping intensity or expand irrigated area to what is required for food security could deliver use any “saved” water. The use of conservation farming much greater economic return by securing water methods such zero-tillage wheat cultivation, laser land for cities and industry. However, modeling of future leveling, and crop residue retention can improve water scenarios (see chapter 6) indicates that improved management and crop productivity. In Pakistan, laser water management and water productivity would leveling and zero tillage wheat cultivation are used enable Pakistan to ensure food security for a growing only across around 0.9 million hectares and 0.5 million population, meet growing water demands outside hectares, respectively (Gill, Mujeeb-ur-Rehman, and of agriculture, and continue to allocate a significant Choudhary 2013). These methods, if closely monitored Figure 5.10  Key Water Footprint Metrics for Rice in Major Rice-Growing Countries 2,500 Water footprint (m3/tonne) 2,000 1,500 1,000 500 0 am nm he p. n s a sia d h il ia es an ne az in an pa es d Re at ist ya f t Ko ar ne In Ch io Rep etn Br pi lad Ja ail St M co k a, do ilip Pa Th Vi ng d re of li In ite Ph n ub Ba Un Un Blue water evaporation loss Irrigation water used by crop Source: Chapagain and Hoekstra 2010. 94 PAKISTAN: GETTING MORE FROM WATER Figure 5.11  Share of Farm Sizes in Pakistan by Number and Aggregate Area, 1990, 2000, 2010 a. Number 25 20 15 Percent 10 5 0 <0.5 0.5–1 1–2 2–3 3–5 5–10 10–20 20–60 >60 Hectares b. Aggregate area 25 20 15 Percent 10 5 0 <0.5 0.5–1 1–2 2–3 3–5 5–10 10–20 20–60 >60 Hectares 1990 2000 2010 Source: GoP 2017a. and accompanied by institutional mechanisms to control cropped area in Pakistan remains under the major water use, can use reduce the application of irrigation crops. with less than 10 percent dedicated to higher- water. Resource conservation technologies have likely value crops. The main reason is long-standing reduced irrigation water applied at the field level in government subsidies that support major low-value Pakistan by 25 percent and increased wheat yields by crops (especially wheat, for which per capita demand around 30 percent (Ahmad et al. 2014a). While these is falling), rather than high-value commodities for practices can improve water productivity, they do not which per capita demand is rising. These policies necessarily generate basin-level water savings, because keep the sector locked in a high-cost and low-return farmers may use the “saved” water for other on-farm mode that delivers low incomes for farmers and high activities or to expand their irrigated area. prices for consumers. These policies serve vested interests, and reforms will be politically challenging. While other countries in South and Southeast Asia Government subsidies far exceed the public resources have diversified away from rice (the main staple) allocated to productive investments in agriculture, toward high-value agriculture (horticulture, pulses, are often regressive in nature, and generate negative oilseeds, etc.), Pakistan has not diversified agriculture environmental externalities. Pakistan produces excess significantly (despite shifts in demand that favor wheat at a high cost. Sugarcane productivity is very high-value agriculture), thus limiting overall economic low and could be imported at far lower cost to the productivity of the sector. Over 90 percent of the consumer. Government needs to reform the agricultural 95 subsidies that drive farmer behavior and lock the sector and prior analyses, including Mansuri et al. (2018), into a low productivity mode. GoP (2012), and PCRWR (2016). For urban water supply and sanitation services, metrics of access, Beyond the farm, agricultural productivity can be reliability, financial sustainability, quality, and improved with better marketing. In Punjab, agricultural customer satisfaction are used. For irrigation and marketing (other than for cotton, sugarcane, and wheat) drainage services, hydraulic efficiency, equity is regulated by the Punjab Agricultural Produce Markets and affordability, and financial sustainability are Ordinance (1978). This gives government a monopoly reviewed. on the establishment of wholesale markets, and only a limited number of licensed dealers can operate in the People market. Market fees are high and incommensurate with the level of service provided. The Punjab Agricultural ices Mitig erv ati eds on Marketing Regulatory Authority (PAMRA) Act (2018) is t of ru ela expected to significantly liberalize agricultural marketing rast ctur wa nf r-r ter- ivery of wate by allowing any registered individual to establish a e I related risks market dealing in primary agriculture produce, and Water endowment by strengthening the regulations governing these En markets. For livestock products, occasional price caps Del ns my viro titution I s have little effect on prices paid by consumers, but act o M s n na on as a negotiating factor for intermediaries buying milk ce ur a m ge so c me nt of w ater re e from producers and as a rent extraction mechanism E nt for local officials. A similar situation prevails in the meat market. Price capping acts as a disincentive to producing better quality products. Discontinuation of notification of meat and milk prices would stimulate Water Supply and Sanitation Services production and marketing of larger quantities and better quality and safer livestock products, thus raising the Nationally, Pakistan has achieved a high level of access incomes of livestock farmers while enhancing supplies to improved drinking water. Of the 9 percent of people to urban areas. Thus, Pakistan has huge opportunities lacking access, two-thirds are in rural areas (figure 5.12, to improve agricultural productivity through improving panels a and b). However, rapid urbanization is water allocation mechanisms, improving water delivery contributing to a decline in access. Given rapid to farms, improving on-farm water management, population growth the number of people lacking access diversifying crop mix, reversing farm fragmentation, increased by nearly 6 million between 2000 and 2015 reforming agricultural policies, and improving the (figure 5.13). In Karachi, access fell from 90 percent marketing of agricultural products. to 86 percent between 2005 and 2015, while the city grew from 12 million to 17 million people, nearly doubling the number of people without access. Access Water Service Delivery to sanitation services improved steadily over the last In this section the reliability, affordability, and 15 years. But 13 percent—or over 26 million people— financial sustainability of agricultural and municipal still defecate in the open (GoP 2016a), mostly in water services are assessed using existing datasets rural areas. Key Messages • Although coverage of drinking water supply service is high, especially in urban areas, coverage is declining with rapid urbanization, and service quality is generally poor. Sanitation services are variable: open defecation is at low levels, but the collection, treatment, and disposal of sewage effluent are grossly inadequate. • Irrigation service delivery is poor and contributes to the low productivity of irrigated agriculture. Hydraulic efficiency of the distribution system is very low, and water delivery across command areas is inequitable. Irrigation services are not financially sustainable and financial performance is declining. Service tariffs are set too low and are decoupled from service quality. The operational costs of service providers are far too high. • Poor operational performance in irrigation water delivery continues to exacerbate waterlogging and salinization, especially across much of Sindh. Despite large-scale reclamation efforts, high water withdrawals and poor drainage mean excess salt continues to accumulate in irrigation areas in both soil and groundwater, impacting agricultural productivity. 96 PAKISTAN: GETTING MORE FROM WATER Figure 5.12  Share of Access to Improved Water Supply and Improved Sanitation in Rural and Urban Pakistan, 2015 a. Improved water supply b. Improved sanitation Unserved, rural 6% Unserved, urban 3% Unserved, rural 26% Unserved, urban 7% Served 67% Served 91% Source: WHO/UNICEF 2015. Figure 5.13  Share of Access to Improved Drinking Sindh (65 percent), KP (54 percent), and Punjab Water and Improved Sanitation in Pakistan, (46 percent) (Mansuri et al. 2018). The completeness 2000–2015 of piped supply coverage has fallen over the last 100 decade, with a greater proportion of households forced to rely on motorized groundwater pumps (KP, Punjab, 90 and Balochistan) and informal private vendors 80 (Sindh and Balochistan). 70 Piped urban water supplies are unreliable. Only 60 27 percent of households receive water for more Percent 50 than 6 hours per day. Reliability is highest in Punjab 40 (57 percent), but very low in Sindh and Balochistan where most households get water for only a few 30 hours per day. Low reliability reflects poor customer 20 orientation by water service providers. Intermittent 10 services discourage users from paying water tariffs, 0 impacting the financial sustainability of service 2000 2003 2006 2009 2012 2015 providers, which further undermines service quality. Improved sanitation Improved water supply In Sindh, supply reliability has decreased. Currently, 93 percent of households receive water for less than Source: WHO/UNICEF. 6 hours per day compared to 87 percent a decade ago (Mansuri et al. 2018). Urban Services Karachi residents experience severe water shortages Socioeconomic improvement in urban areas has during summer because of poorly maintained and lowered poverty rates and increased access to water outdated pumping stations, a leaky distribution supply and sanitation services. However, for many network, and theft from water mains. Bulk water urban dwellers, these services are low quality, supply for Karachi represents around 115 liters per unreliable, or unaffordable. Almost half of urban capita per day, similar to consumption levels in some households rely on piped water (figure 5.14), although modern European cities. Thus, with efficient delivery the percentage is decreasing in all provinces because and careful demand management, the current bulk of rapid unplanned urbanization. Access to piped water supply should be sufficient. However, the hot climate varies between provinces: Balochistan (68 percent), and associated high evaporative losses and the 97 Figure 5.14  Urban Water Access by Source in Pakistan, 2015 Covered well Filtration plant 1% 6% Unimproved 8% Hand pump 8% Piped water 48% Motorized pump 29% Source: Mansuri et al. 2018. Figure 5.15  Share of Urban Water Supplies Unfit for Human Consumption by Province in Pakistan, 2005 and 2015 100 90 80 70 60 Percent 50 40 30 20 10 0 Khyber Punjab Sindh Balochistan Pakhtunkhwa 2005 2015 Source: PCRWR 2016. inevitability of some leakage mean an increased bulk water mains and sewers (Haydar et al. 2009). Arsenic supply is required. The projected additional 3 million and iron levels exceed safety limits in 6 percent inhabitants 2047 and the expected increases in per to 10 percent of piped urban supplies nationally capita water demand with increasing wealth mean a (PCRWR 2016). 50 percent increase in the bulk water supply is likely to Cost recovery for urban water services is extremely low. be required for Karachi. Nationally, cost recovery is estimated to be 8 percent The quality of urban water supplies is very low, with (Danilenko et al. 2014). With insufficient finances, 80 percent being unsafe for consumption in Sindh utilities are unable to keep supply systems running and Balochistan (figure 5.15). Over the last decade continuously and lack the resources to expand services major improvements have been made in Punjab and to keep pace with growing urban populations. Low KP, while the already poor quality in Balochistan has cost recovery partly reflects low tariff levels, and partly worsened (figure 5.15). The most common problem is reflects high levels of leakage and theft (nonrevenue fecal contamination from cross connections between water [NRW]). Nationally, NRW averages 57 percent 98 PAKISTAN: GETTING MORE FROM WATER (Danilenko et al. 2014). In Karachi it is higher because 30 percent of the wastewater (Bashir 2012; Ensink et al. of numerous illegal connections and an old, poorly 2004). Karachi and Islamabad have secondary (biological) maintained pipe network. Nationally, 62 percent of treatment, but less than 8 percent of wastewater in households pay their water tariffs, ranging from 21 these cities is treated to this standard (Murtaza 2012). percent in Quetta to 98 percent in Lahore (SBP 2017). Rawalpindi, Multan, and Gujranwala have no wastewater Water utilities are subsidized by provincial governments treatment (World Bank 2016). Pakistan’s urban population for both operation and maintenance (O&M) costs and is expected to double over the next three decades to for debt servicing. 155 million, posing huge challenges for water supply and sanitation services (Ellis et al. 2018). Failure to maintain distribution systems has led to large-scale, systematic illegal connections that benefit private water vendors and disadvantage poor populations Rural Services (Rahman 2008). Tankers supply an estimated 20 percent Rural water services are far worse than urban services, of Karachi households, with monthly charges ranging reflecting the technical challenge of delivering water from 50 percent to 100 percent of the average household services over long distances and the financial challenge income (Mustafa et al. 2017). Anecdotal evidence from of higher costs and fewer customers. Water supply is Karachi’s informal settlements suggests water from mostly self-provided, and increasingly so. Groundwater is private tankers costs 30 times the government water the predominant source. In Punjab and Sindh, 90 percent tariff. This means water services are unaffordable for of rural households rely on groundwater (mostly most urban poor households in Karachi, which often have motor or handpumps). KP and Balochistan have more access only to contaminated water (Alamgir et al. 2015). diverse supplies, but around 29 percent and 21 percent, The common assumption is that residents of informal respectively, have access to piped supply (Mansuri et settlements have the poorest and least affordable service al. 2018). Few rural supplies are monitored for quality. and must use extrajudicial solutions to solve water supply Across much of Punjab and Sindh, arsenic in groundwater problems given low capacity to use established law or exceeds international standards for human consumption, administrative procedures. However, recent interviews exposing 50 million to 60 million people to serious health and focus group discussions across five of the largest risks (Naseem and McArthur 2018; Podgorski et al. 2017). informal settlements in Karachi reveal significant variation in water demand and access both across and within Installation of toilets connected to septic tanks has settlements, reflecting a long history of regularization of improved rural sanitation across parts of Punjab and slums and state-society relations. The common narrative KP. In Sindh and Balochistan, however, one-half and of urban poor households in slums being “caught up in two-thirds of the households, respectively, rely on webs of illegality” is thus simplistic, and the reality is unimproved toilets and pit latrines. Rural open defecation more nuanced than dichotomies of legal or illegal, formal is still common in Punjab (23 percent) and Balochistan or informal, or civil or political. The Karachi situation is (17 percent) (Mansuri et al. 2018). Sewerage is improving, as outlined in the political economy discussion almost nonexistent in rural Pakistan, although covered in chapter 4. or underground sewers serve a small percentage of households in Sindh and Punjab. The complete absence Urban sanitation services vary considerably across of public services for rural wastewater management Pakistan. Services are best in Punjab, in which poses a significant health hazard. Despite improvements 59 percent of the population have access to flush in rural sanitation, the lack of public water supplies toilets connected to sewer systems and 26 percent essentially negates the human health benefits. are serviced by septic tank. In Sindh, 63 percent use flush toilets connected to sewers, but as conditions are Irrigation Services generally unsuitable for septic tanks, more than one- third of the urban population is unserved by sewerage, The discussion of water resources management and effluent is discharged to open drains (Mansuri covered aspects of irrigation performance from an et al. 2018). Services are worst in Balochistan and have allocation perspective and introduced the equity-based worsened over the last decade. More than half of the warabandi system. A deeper assessment of irrigation urban population use toilets that flush to open drains, service delivery at the command area level is provided and 22 percent use simple latrines. here, considering operational performance (hydraulic efficiency and drainage), equity and affordability, and Across Pakistan, most wastewater is discharged untreated financial sustainability. into rivers and coastal waters. This degrades ecosystems and impacts human health. Only four of the 10 cities Operational Performance with more than 1 million inhabitants (Islamabad, Lahore, Karachi, Faisalabad) have any wastewater treatment Operational performance—in terms of the overall facilities. Existing facilities have capacity to treat less than efficiency of water delivery—is very low. Uncertainties 99 in measurement and incomplete water accounting The waterlogged area varies seasonally, with much mean that the estimates of water delivery from greater areas affected at the end of the annual barrage off-takes to the farm outlet vary widely from monsoon. Surveys in the 1980s have indicated 20 percent (SBP 2017) to more than 60 percent 6 million hectares of salt-affected land, but four (Raza et al. 2013). Much of the inefficiency comes decades of Salinity Control and Reclamation Projects from high levels of canal leakage, which is a major (SCARPs), costing US$2 billion, have enabled share of groundwater recharge. Some inefficiencies reclamation of significant areas. The extent of salinity were accepted as part of the design of the distribution varies strongly between provinces. Around half the system, at a time when water was less scarce in farmland in Sindh and Balochistan is affected, and relative terms. Other inefficiencies reflect poor about 10 percent across KP and Punjab (Zulfiqar and maintenance of the system and poor operation. Thapa 2017). Investments in canal lining have improved delivery efficiency relative to original designs in some areas. Seawater intrusion exacerbates salinization in coastal Sindh and Balochistan. Despite ongoing reclamation Another measure of operational performance is the efforts, shallow groundwater is increasingly saline delivery performance ratio (DPR). This compares the in coastal areas and 40,000 hectares are abandoned delivered flow to the delivery capacity (as a fraction annually because of secondary salinization (WAPDA or percentage), which is relevant because supply 2007). The current rate of salt imports to the basin limitations mean the system is usually operated at full by the river and salt mobilization from groundwater capacity. Jacoby et al. (2018) assess DPR for a nine-year through pumping far exceed the contemporary rate period (2006–14) in kharif in more than 1,000 channels of salt export in basin outflows. Salt is therefore across Punjab. They compare values between the head accumulating in soil and groundwater of the basin and tail of the different distributaries. They find that the (Butta and Smedema 2007). Not all salt accumulation DPR varies between 56 percent and 80 percent with an is necessarily harmful, and a much more detailed average of around 70 percent, indicating water delivery understanding and quantification of salt dynamics is consistently below capacity. Noting that design across the basin is required to guide management. capacity reduces with distance along the distributaries, Ultimately, managing soil salinity will require irrigation they find that the delivery shortfalls relative to design modernization and improved operation, as well as capacity are higher at the tail than at the head, adequate dry season environmental flows in the Lower by around 5 percent, on average. The reduction in Indus, to counter seawater intrusion. performance with distance along distributaries likely reflects the combined effects of inadequate channel maintenance and upstream water theft. Equity and Affordability As the irrigation and drainage system has fallen into The warabandi system was designed distribute water disrepair, the efficiency of delivery has declined. equitably; however, there is no agreed measure of Irrigation efficiency in Pakistan is now among the equity, which makes evaluation difficult. Equity is often lowest in the world. The deterioration of the irrigation assumed to be described by duration, prorated by area, infrastructure—including siltation of canals and degraded which for equal flow rates imply equal volumes per canal walls—and inadequate maintenance have resulted unit area, or equal irrigation depth. However, delivery in increased water losses through seepage, exacerbating flow rates vary considerably between irrigation areas. the problems of water logging and salinity. From a water KP has much higher delivery flow rates than Punjab balance point of view, most of the canal seepage in or Sindh, because it has relatively limited irrigable Punjab and KP is not lost because it recharges underlying land, and can thus “afford” to deliver more water per shallow freshwater aquifers and is accessible through unit area from its allocation under the Accord. These groundwater pumping. However, in Sindh, lost canal differences are sometimes interpreted are “inequity by water recharges saline aquifers and thus can no longer design” between irrigation areas. With command areas, be used for productive use. Across the Indus Basin, it some level of inequity was embedded in the original is estimated that about one-third of all canal seepage hydraulic designs (Van Halsema and Vincent 2006). is to saline aquifers (chapter 3). From an operational KP development investments (Mardan SCARP, Swabi performance point of view, any canal seepage is SCARP, Pehur High Level Canal, Chashma Right Bank undesirable (even if it recharges freshwater aquifers), Canal, Warsak Gravity Canal) have all focused on because farmers receive less of their allocated amounts improving existing irrigation systems rather than and experience lower levels of service. Groundwater extending the irrigated area, thus doubling or tripling pumping incurs additional costs. the delivery flow rates. In a context in which the water Waterlogging and salinity affect an estimated resource was available, and the existing irrigation 4.5 million hectares of irrigated land and reduce system had limited capacity, an obvious solution was agricultural production by 25 percent (Qureshi 2016). to upgrade the irrigation system and enhance the 100 PAKISTAN: GETTING MORE FROM WATER capacity. In contrast, when rehabilitation investments lining would affect access to groundwater, further are undertaken in the Sindh or Punjab, and supply exacerbating inequities. There is little agreement is limited, system capacity has not increased. Within on the institutional responsibilities for groundwater command areas, operational inequity can occur, and whether farmer organizations or water user particularly as a result of opening and closing tertiary associations (WUAs) could manage conjunctive water canals when insufficient water is available to operate use (Nagrah and Rosell 2012). The proposal in the NWP all canals at, or near, capacity. The arising inequity is an to establish groundwater authorities in each province unintended consequence of the “rotational program” could be counterproductive in terms of irrigation that guides canal operations and that has remained efficiency and equity if not closely coordinated with the largely unchanged since British colonial times. management and regulation of surface water delivery. Irrigation inequity can be measured by comparing flows through canal outlets in the top, middle, and bottom Financial Sustainability thirds of the canal (head-middle-tail). Measurements Irrigation management transfer reforms introduced indicate that in many cases head outlets draw more in 1997 (chapter 4) were partly in response to than their share (e.g., Ghumman et al. 2014). A more low financial sustainability. The reforms included detailed method uses discharge measurements at the decentralization of irrigation service delivery and head of a tertiary canal and at every outlet along its abiana (the system of irrigation tariffs) collection, but length across an entire season to calculate a Gini index with mixed results in terms of financial performance. (Shah et al. 2016). Shah et al. (2016) find no significant inequity at the tail end, contrary to widespread Abiana can be considered affordable given the belief. Jacoby et al. (2018) find small reductions willingness of farmers to pay for improvements in in performance (water delivery relative to design service delivery and their significant expenditure on capacity) toward the tail end of distributaries. Equity in groundwater pumping (Bell et al. 2016). The cost irrigation services appears to be worsening as a result of diesel for motorized pumps represent around of declining maintenance and the breakdown of the 20 percent of farmers’ incomes (GoP 2016b). Abiana arrangements for control at the main and distributary is not based on water consumption, but is levied on canal levels (Blackmore and Hasan 2005), as well as cropped area, in some cases differentiated by crop manipulation of farm outlets (Rinaudo 2002). type. This means that there is no incentive for water conservation and in many cases no incentive to shift to In command areas in which delivery flows are high more water productive crops. (as in KP), farmers often cease to irrigate earlier in the season, because crops have already received Abiana ranges from PRe 85 per hectare in Punjab to sufficient water. Anwar, Bhatti, and de Vries (2016) PRe 618 per hectare in KP (GoP 2012). These levels are report that in the Pehur High Level Canal system of KP, grossly inadequate to cover operating costs, let alone 92 percent of farmers did not irrigate in September. the costs of system upgrades. On average, only 20 Warabandi, however, supplies water at system capacity percent of the total operating cost of the distribution whenever possible, causing major spatial and temporal system is covered from abiana. This is dominated by mismatches between crop water requirements the low cost-recovery in Punjab and Sindh (figure 5.16). and water delivery. Attempts at more flexible Cost recovery in KP and Balochistan is higher because management—called “demand-based irrigation,” the limited extent of irrigation in these provinces “arranged demand-based irrigation,” or “crop-based means operating costs are lower. Low cost recovery is irrigation operations”—have been made in the not unique to Pakistan. Few Organisation for Economic Mardan SCARP, Pehur High Level Canal, and Chashma Co-operation and Development (OECD) countries Right Bank Canal systems. Unfortunately, none have achieve full cost recovery (OECD 2013), but Pakistan’s persisted, and all have reverted back to warabandi. financial performance for irrigation is among the lowest in the word (Bell et al. 2014). Originally, farmers were entitled to a water share equivalent to 70 percent of the design cropping Cost recovery is partly determined by collection intensity (Mustafa 2001). However, cropping efficiency. Collection efficiency is very low in intensities have risen to over 150 percent, supported Balochistan and declining, but 70 percent to 90 percent by groundwater pumping (Khan 2009). Access in other provinces and improving (figure 5.17). to groundwater is not determined by any formal Collection efficiency is often higher for tail enders who allocation mechanism, but simply by location. Quality are pressured into paying while sometimes receiving and depth determine groundwater value. Shallow poorer service. Those at the head of the distributaries and good quality groundwater is found closer to leaky are more likely to default on payments because canals. From a water services delivery perspective, they will receive water in any case. There is limited this raises the question of whether improved canal enforcement capacity and no legal basis for penalize 101 Figure 5.16  Operating Cost Recovery Ratio of Irrigation Departments by Province in Pakistan, 2000–09 1.6 1.4 Operating cost recovery ratio 1.2 1.0 0.8 0.6 0.4 0.2 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Punjab Sindh Balochistan Khyber Pakhtunkhwa Source: GoP 2012. Note: data not available for all years for all provinces. Figure 5.17  Abiana Collection Efficiency by Province in Pakistan, 2000–09 100 90 80 70 60 Percent 50 40 30 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Punjab Sindh Balochistan Khyber Pakhtunkhwa Source: GoP 2012. Note: data not available for all years for all provinces. defaulters (SBP 2017). Because abiana is not linked to pay scales and increase almost annually. To prevent service quality and nonpayment does not affect service, staff salaries from consuming an increasing proportion achieving higher collection efficiency without coercion of the operating budget, budgets need to be revised or stronger regulation will be difficult. and linked to inflation. Irrigation departments are often perceived to be In the 1970s, abiana fully covered O&M costs but a overstaffed and lacking the right balance of technical government decision to freeze abiana (Khan 2009) led expertise. A substantial proportion of their O&M to heavy subsidization. Subsidization was estimated budget is therefore used for staff costs, making to be US$44 million in 2012 (SBP 2017), covering financial sustainability more difficult. In Punjab, 75 percent of O&M costs. Government subsidies are salaries for more than 35,000 staff members equivalent to around 2 percent of GDP. Service delivery consumes 76 percent of the operating budget. costs need to be reduced and tariffs incrementally Operating budgets seldom increase with inflation, increased. Tariffs should to be linked to clear, published while staff salaries are tied to national or provincial measures of service quality. 102 PAKISTAN: GETTING MORE FROM WATER Water-Related Risk Mitigation People ices Mitig This section considers risks for which drivers are largely erv ati ds on beyond the control of the water sector. In some cases, te of ru ela these risks may be mitigated by water sector actions; rast ctur wa nf r-r ter- ivery of wate in other cases, adaptive responses are required. The e I related risks risks are climate change, the unintended consequences Water endowment of energy policies (water-energy nexus), and erosion Envir and sediment transport. A description of current and Del ns y titution om I s future consequences is provided for each risk and an o M s nm na n ce assessment of how well the risk is being recognized ur a ge co me so nt of w ater re e nt and mitigated. E Key Messages • Pakistan’s biggest water challenges are not externally imposed. However, climate change represents an additional challenge to improving water security. Adaptive responses are required. • Climate change is not expected to have major impacts of the average availability of water in the coming decades. However, water availability is expected to become more variable (and less predictable) between and within years. This is expected to mean more extreme floods and droughts. In the Upper Indus Basin, accelerated glacial melting will greatly increase the risks of glacial lake outburst floods (GLOFs) that are often devastating at the local level. Improved data, modeling, and forecasting to guide preparedness and response to extreme events will be increasingly important. • In the Lower Indus Basin, sea level rise and increases in the frequency and severity of coastal storms will exacerbate seawater intrusion into the delta and into coastal groundwater. In coastal Sindh, this will further degrade the groundwater resource, groundwater-dependent ecosystems, and the productivity of irrigation. • Potentially the greatest challenge from climate change will be the increases in water demand, especially for irrigated agriculture. Climate change alone is expected to increase water demand by 5 percent to 15 percent over the next three decades, depending on the level of warming. • Pakistan suffers chronic energy shortages, which have many connections with water management. Careful consideration the multiple cross-sectoral trade-offs between energy and water are required in the coming decades. • Basin-scale sediment sourcing, transport, and deposition have been significantly modified by water resources development. This has consequences for the safety and operational performance of water infrastructure as well as for river and delta ecosystems. A more integrated approach to sediment management and better monitoring of sediment sources, erosion, and sediment transport are required to guide intervention strategies and environmental management. Climate Change 1 degree Celsius to 2 degrees Celsius increase by 2050, with a sharp increase (4 degrees Celsius to 6 degrees Although Pakistan’s biggest water challenges are Celsius) by the end of the century (Chaudhry 2017). internal, climate change is a significant additional Warming will increase the frequency of deadly heat challenge to improving water security. Expected climate waves by the end of the century (Im et al. 2017). Heat change impacts for the water sector are summarized waves—with temperatures exceeding 40 degrees Celsius here. The investments in infrastructure, information, for 10 consecutive days—are expected to become more and institutions to build resilience and mitigate these common in Punjab, Sindh, and Balochistan (Zahid and climate change risks are discussed. Rasul 2012). Heat waves increase urban water demand and the use of untreated water (ACAPS 2017). Heat waves Climate Warming also affect energy security because the warmer water Warming of 0.23 degrees Celsius to 0.33 degrees Celsius used for thermal plant cooling reduces power output per decade has been observed in the Lower Indus by up to 0.5 percent (ADB 2012). Warming increases over the past 30 years (Ahmad et al. 2014). Unless the evapotranspiration. Thus, crop water requirements and targets of the Paris Agreement are achieved, Pakistan’s natural water losses through landscape evapotranspiration agricultural regions and coastal zones will experience a will increase. Without improved demand management 103 severe water shortages will increase (Adnan et al. 2017b; dominate the overall increase (figure 5.19), and alone Ahmad et al. 2014a). could increase water demand by 25 billion cubic meters. Although uncertain, there is evidence that Estimates of the impacts of warming on water demand while climate change would increase crop water use, suggest differing sensitivity by sector. Industrial water it may also enhance crop growth and thus increase demands are most sensitive to climate warming. yields, given longer growing seasons (Chaudhry 2017). Under a faster warming scenario, warming could cause increase industrial demand by more than 20 percent by Hydrologic Change 2050 (figure 5.18). Under a faster warming scenario, total water demand could increase by 30 billion cubic Annual precipitation averaged across Pakistan, meters (Amir and Habib 2015), or around 20 percent while varying yearly, has increased by 25 percent of current withdrawals. Because irrigation strongly (or 63 millimeters) over the past century dominates water use, increases in irrigation demand (Yu et al. 2013). But there is considerable spatial Figure 5.18  Estimated Increases in Water Demand Attributable to Projected Warming in Pakistan, 2025 and 2050 25 20 15 Percent 10 5 0 2025 2050 2025 2050 1°C warming 3°C warming Agriculture (crops and livestock) Domestic/urban (drinking, sanitation, and urban services) Industry Environment Source: Amir and Habib 2015. Figure 5.19  Sector Shares of Water Demand Increase Attributable to Projected Warming in Pakistan, 2025 and 2050 100 90 80 70 60 Percent 50 40 30 20 10 0 2025 2050 2025 2050 1°C warming 3°C warming Agriculture (crops and livestock) Domestic/urban (drinking, sanitation, and urban services) Industry Environment Source: Amir and Habib 2015. 104 PAKISTAN: GETTING MORE FROM WATER and seasonal variability in precipitation trends. The reduce runoff in these months (Lutz et al. 2016). Overall, strongest increasing trends are for the Greater Himalaya annual mean river flows are expected to increase by and the Northern Balochistan Plateau, in the monsoon around 10 percent by the end of the century. season. A significant reduction in monsoon rainfall has Climate change is expected to increase interannual been observed for the Coastal Belt over this period. flow variability. A greater proportion of precipitation Historical inflows to the Indus Basin of Pakistan reveal is expected to fall as rain instead of snow, eventually statistically significant trends. However, data records are reducing (but probably not eliminating) the glacier relatively short and there is little evidence to attribute meltwater contribution, which is the least variable these changes to anthropogenic climate change. As component of inflows. Yearly variations in precipitation indicated in chapter 3, the small but significant decrease are expected to increase and will increasingly dominate in total Indus inflows appears to be largely a result inflow variability. This will change the intensity and of the increased water use in India on the eastern frequency of extreme discharge events (Lutz et al. tributaries, as allowed under the Indus Waters Treaty. 2016). River flood risk may double at the subnational Yu et al. (2013) show a slight increase in the Chenab level within 25 years, with Sindh and Punjab most during rabi and a slight decrease in the Indus during affected (Willner et al. 2018). kharif. Pakistan has a complex hydrology, and thus explaining observed hydrologic change and projecting The climate change impacts on Indus Basin flows future change is difficult. This is especially true for the apply in general to the Kabul subbasin, but the Indus Basin, where rainfall runoff, snowmelt, and glacier Kabul will have its own unique climate changes and melt all contribute to river flow, in proportions that vary responses. These need to be understood to guide joint considerably between tributaries. development in this transboundary subbasin. While warming logically increases melting, there is scant GLOFs occur when the ice wall retaining the lake fails, temperature data from high elevations in the Indus sending the entire stored water volume downstream as a Basin where meltwater is generated to evaluate climatic flash flood. In the Upper Indus Basin, climate change will change. Changes in snowfall patterns and the dynamics of cause existing glacial lakes to get larger and cause new glaciers—including snow contributions to glacier volume, glacial lakes to form. Pakistan has around 2,420 glacial rates of glacier flow, the role of debris coverage, and black lakes, mostly in Jammu and Kashmir, of which 52 have carbon—are complex and only partially understood. Most been identified as GLOF risks (ICIMOD 2005). Since 2000, recent studies suggest a modest increase in streamflow the number of glacial lakes in the Hindu Kush-Karakoram- in the Indus Basin for the next several decades as result Himalaya of Pakistan has increased (Ashraf et al. 2017), of accelerated glacier melting (e.g., Lutz et al. 2016). and an increase in GLOFs is expected (Bajracharya et Longer-term projections are highly uncertain. Because a al. 2015). Several GLOF events have been recorded in significant fraction of the glaciated area in the Upper Indus Pakistan, and there is evidence the frequency is increasing Basin is at very high elevation, complete disappearance of (Rasul et al. 2011). More than 7 million people are glaciers is unlikely under expected warming trajectories. estimated to be at risk in Jammu and Kashmir and KP . However, the extent to which residual ice volumes will Since 1996 GLOFs have killed more than 600 people and generate meltwater is less clear and will depend on rates over 7,000 livestock, and more than 10,000 buildings have of snowfall addition to glaciers and the rates of ice flow been damaged or destroyed (UNDP 2015). The increasing down to lower, warmer elevations. A smaller fraction of GLOF risk requires additional investment in monitoring, future precipitation will occur as snow, and thus it is likely early warning systems, and, where appropriate, direct that the eventual residual glaciers will flow more slowly intervention to drain high-risk glacial lakes. and contribute less meltwater. In the Makran and Kharan basins, the hydrological The changing annual pattern of temperatures is impacts of climate change will be very different to the expected to alter the timing of inflows. With warm Indus Basin. Balochistan has experienced a general temperatures earlier in summer, the first changes in warming trend since 1980, with an increase in extreme river flow are expected to be gradual increases in rainfall events especially in the coastal area (Abbas et al. meltwater flows from May to September, peaking when 2018). It is likely to experience more variable rainfall glaciers still cover substantial areas (Lutz et al. 2014; and a reduction of snowfall at high altitudes (LEAD Mathison et al. 2015). In the Upper Indus, earlier onset 2017). Balochistan is very vulnerable to climate change of snowmelt and glacier melt—and likely increases in impacts, and the province’s capacity to adapt to climate winter precipitation—would increase in flows during change is very low. To buffer increasing variability, autumn and spring (Lutz et al. 2016). In lower altitude the province will need to manage groundwater more subbasins of the Indus, autumn and winter flows are strategically, scaling up managed aquifer recharge to likely to increase slightly because of increased winter capture high-intensity rainfall events. Balochistan should precipitation; decreases in precipitation during the also incorporate measures to enhance climate resilience monsoon and higher evapotranspiration are likely to into its legal and policy instruments (LEAD 2017). 105 Figure 5.20  Mean Sea Level, Karachi Coast, Pakistan, 1860–2000 1,800 1,750 Mean sea level (mm) 1,700 1,650 1,600 1,550 1,500 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 18 18 18 18 19 19 19 19 19 19 19 19 19 19 20 Source: Rabbani et al. 2008. Pakistan’s coastal areas are vulnerable to sea level rise. Water-Energy Nexus Observed rates of rise in Karachi average 1.1 millimeter per year (figure 5.20). Sea level rise will exacerbate land Pakistan is not energy secure. Chronic power shortage subsidence caused by overpumping of groundwater in costs the economy around 2 percent of GDP per annum urban areas (Rabbani et al. 2008). This will make irrigation (Aziz and Ahmad 2015). Pakistan Vision 2025 sets drainage in Sindh even more challenging, increase the risk ambitious targets for energy and water security. For of coastal flooding, and exacerbate seawater intrusion into energy, targets include closing the supply-demand the delta and coastal groundwater. gap and doubling generation capacity. For water, the primary target is ensuring all citizens have access to an adequate water supply, with improvements in Responding to Climate Change efficiency and storage as key enablers. Achieving these Pakistan’s climate change vulnerability is recognized in energy and water targets will require integrating these the National Climate Change Policy (2012), the Climate sectors’ planning to leverage synergies and avoid Change Act (2017), and the NWP (2018). These legislative unintended trade-offs. and policy documents establish climate resilience as a key objective for all development interventions. Water security Water is used in the energy sector for coal mining and has been recognized as a key concern under a changing processing and for electricity generation (UNDP 2017). climate, and policy measures, ranging from additional Pakistan’s expanding coal mines use large volumes storage to water conservation and awareness raising, are of water for dust suppression and processing, and highlighted in the NWP. discharge significant volumes of polluted water. Coal mining is focused in Sindh, in which water scarcity To support Pakistan’s water sector efforts toward and pollution challenges are large. The Thar coalfield climate resilience, policies could be complemented in Sindh is the largest in Pakistan and the sixth largest with adaptation targets and indicators that help in the world (Ali et al. 2015). Thermal and nuclear assess the effectiveness of alternative measures power plants account for 64 percent and 6 percent and help guide development finance investment. of national generation capacity, respectively, and International climate finance, technology require water for evaporative cooling (GoP 2017b). development, and transfer and capacity building can Cooling water supply needs to be high quality and all contribute to adaptation. Most investments in reliable. Although the energy sector accounts for only water security are also investments in climate change 1 percent of water withdrawals, water availability and adaptation, and there are significant opportunities variability already constrain electricity generation in to make water-related investments more climate Pakistan. Thirty percent of the electricity generation resilient, especially in Sindh and Balochistan. The is from hydropower (GoP 2017b). Hydropower does required annual investment for climate change not consume water, but energy demand patterns adaptation has been estimated to be US$7 billion drive reservoir releases that do not fully match to US$14 billion, including US$2.0 billion to US$3.8 irrigation demands. Modeling suggests that optimizing billion to reduce flood vulnerability (UNFCC 2015). dam operations (including Diamer Bhasha) solely International climate adaptation finance in recent for hydropower would increase energy production years has averaged US$500 million, well below what by 10 percent but reduce agricultural production is required (LEAD Pakistan 2013). by two-thirds (Yang et al. 2014). Careful tradeoff 106 PAKISTAN: GETTING MORE FROM WATER analyses are needed to identify solutions that balance energy sources may be most viable, especially for energy and agricultural benefits, while accounting for groundwater pumping. Solar groundwater pumping will environmental requirements (Zeng et al. 2017). need to be carefully managed to avoid exacerbating overexploitation of groundwater. Small-scale (less than In the water sector, energy used for groundwater 50 megawatts) hydropower potential is considerable pumping and distribution (urban pump stations), as and underdeveloped with only 128 megawatts in well as in water supply and wastewater treatment. operation (AEDB 2018). The potential is estimated to The agricultural sector uses only 1 percent to 2 percent be 3,100 megawatts (AEDB 2016; IRENA 2018), of of the national total, and its share has been generally which 20 percent is canal-based hydropower. declining as other energy hungry sectors of the economy grow (FAOSTAT 2016). Total energy use A water smart energy sector is critical to long-term in agriculture has more than doubled in the decade energy security for Pakistan. In a water scarce world, from the early 1980s with expansion of tube wells, the opportunity cost of water will increase, and the but has fluctuated since with no overall increase energy sector will have to compete for water with (FAOSTAT 2016). The largest energy use in agriculture other users. Water-intensive power generation will be is groundwater pumping. Punjab, where groundwater increasingly expensive. Options to assist Pakistan reach pumping is concentrated, accounts for most of the Vision 2025 targets for energy should be assessed agricultural energy use (Siddiqi and Wescoat 2013). in terms of both energy and water issues (table 5.2). Around three-quarters of groundwater abstraction in Pakistan relies on diesel pumps (Qureshi et al. 2003), Erosion and Sediment Transport but electricity use in agriculture has grown steadily for Erosion, sediment transport, and deposition affect several decades in response to government subsidies water security. Sediment damages hydropower for electricity (Khair, Mushtaq, and Reardon‐Smith turbines, reducing their performance and effective life. 2015). Subsidy reductions from 3 percent in 2011 to Sedimentation in reservoirs—upstream of barrages and in 0.8 percent in 2015 led to a reduction in agricultural irrigation canals—reduces hydraulic performance and can electricity use (IMF 2017). Falling groundwater levels lead to scouring and erosion of the lower river and delta. in parts of Punjab—partly attributable to electricity Sediment loads affect water quality, channel morphology, subsidies—have caused pumping costs to rise. This riverine habitat, and delta development. Sediment is mainly because as groundwater levels fall, diesel trapping by reservoirs and barrages has reduced pumps need to be replaced with more powerful sediments to the delta, causing large-scale geomorphic and energy-intensive electric pumps (Qureshi et al. changes and a loss of ecosystem services (see chapter 2). 2010). Khan et al. (2016) estimate that the cost of groundwater pumping in Punjab could rise by Erosion and sediment transport depend on climate 270 percent by 2030. Expanding wastewater treatment and catchment topography, geology, land use, and would require significant energy, potentially increasing management, as well as water resource development. total energy use by around 0.5 percent. Catchment disturbance has been linked to increased sediment loads in the Indus Basin (WWF 2012), but Doubling national electricity generation capacity in the absence of long-term monitoring the impact is in line with Vision 2025 would have significant unquantified. Naturally high erosion rates are expected water consequences. Failure to adequately consider with high relief, fast runoff, and low vegetation cover water issues into energy sector planning could (Ali and De Boer 2010). Reduced vegetation cover will have significant unintended consequences for other increase sediment loads. Deforestation rates are among water users. Optimizing the operation of storage the highest in Asia and are estimated at 0.4 percent reservoirs and increasing the use of run-of-the- (Qamer et al. 2016) to 2 percent (Ahmed et al. 2015) river hydropower can reduce cross-sectoral impacts. annually. In the Upper Indus Basin, however, snow and Solar and wind energy can help close the energy ice cover is probably the single most important factor supply-demand gap with minimal water impact. controlling sediment supply (Ali and De Boer 2007). Although solar and wind account for only 1 percent of the total energy mix (Wakeel, Chen, and Jahangir River sediment transport capacity is a function of the 2016), their potential to contribute to Pakistan’s flow regime, especially the flood regime (Lu et al. energy security is substantial. The World Bank and 2013; Walling 2009). Deforestation increases flood the Alternative Energy Development Board (AEDB) magnitude (Bradshaw et al. 2007), enhancing estimate a theoretical national wind energy potential sediment transport. Increasing flood magnitude is of 350 gigawatts (ESMAP 2015), and Pakistan’s occurring in some areas of Pakistan (Atta-Ur-Rahman unexploited solar energy potential is very significant and Khan 2013; Tariq and Aziz 2015; Webster, Toma, (IFC 2016). The low water footprint of these renewable and Kim 2011). Sediment loads from the Upper Indus energy sources makes them attractive options for are likely to increase with more intense rainfall, more a water scarce future. At the local scale, renewable frequent GLOFs, and glacial erosion (Lu et al. 2010). 107 Table 5.2  Options to Achieve Government Pakistan Vision 2025 Energy Targets with Water-Energy Nexus Issues Option Water issues Energy issues New large HEP dams Trade-offs with irrigation depending on Helps meet greenhouse gas emission targets. location, design, and operation. Impacts on Trade-offs with irrigation. High O&M cost for aquatic ecosystems. transmission lines in difficult terrain. Expand small-scale Negligible impacts. May reduce reliance on long-distance transmission hydropower lines. Increase wind and solar Negligible impacts, but increased risk of Cost-effective, scalable, and indigenous energy. energy groundwater depletion without regulation. Helps meet greenhouse gas targets. Increase biogas energy Negligible impact, but relies on agricultural Cost-effective, scalable, and indigenous energy. water supply. Helps meet greenhouse gas targets. New coal-fired power Water quality risks. Cooling water Increases greenhouse gas emissions and lowers stations (CPEC plansa) requirements. local air quality. Increase energy imports Minimized additional impact on local water Increased dependence on energy markets and resources. exposure to volatile energy prices. Note: CPEC = China–Pakistan Economic Corridor; HEP = hydroelectric power; O&M = operation and maintenance. a. IER 2017. Pakistan has implemented extensive watershed ACAPS. https://www.acaps.org/sites/acaps/files​ protection projects to reduce erosion. Community-led /products/files/170424_acaps_start_anticipatory​ tree planting projects have been implemented upstream _briefing_note_pakistan_heatwave.pdf. of Tarbela and Mangla reservoirs. These projects ADB (Asian Development Bank). 2010. Islamic Republic have been largely ineffective in reducing sediment of Pakistan: Glacial Melt and Downstream loads because they cover only a small fraction of the Impacts on Indus Dependent Water Resources catchment areas of the reservoirs (WWF 2012) and do and Energy. Manila, Philippines: ADB. http://lib​ not address areas exposed by snow and glacier retreat. .icimod.org/record/29300/files/FinalReport.pdf. A more integrated approach to sediment management is required that includes erosion control (through targeted ———. 2012. Adaptation to Climate Change: The Case of revegetation and slope stabilization) and inclusion of a Combined Cycle Power Plant. Philippines: ADB. sediment management in the design and rehabilitation Adnan, S., K. Ullah, A. H. Khan, and S. Gao. 2017b. plans of water infrastructure. To support a more “Meteorological Impacts on Evapotranspiration in integrated approach, better monitoring of sediment Different Climatic Zones of Pakistan.” Journal of sources, erosion, and sediment transport is required. Arid Land 9: 938. This will inform sediment budgeting and the modeling of sediment dynamics to guide intervention strategies AEDB (Alternative Energy Development Board). 2016. and environmental management. The Dasu hydropower Potential and Progress in Small Hydropower. project reflects an increased awareness of sediment Islamabad, Pakistan: AEDB. management, with nine 6.4-meter diameter low-level ———. 2018. Potential and Progress in Small dam outlets and two 9.4-meter diameter flushing Hydropower. Islamabad, Pakistan: AEDB. http:// tunnels in the right abutment to facilitate drawdown www.aedb.org/ae-technologies/small-hydro. flushing (Annandale, Morris, and Karki 2016). Ahad, K., A. Mohammad, F. Mehboob, A. Sattar, and I. Ahmad. 2006. “Pesticide residues in Rawal Lake, References Islamabad, Pakistan.” Bulletin of Environmental Abbas, F., N. Sarwar, M. Ibrahim, M. Adrees, and S. Ali. Contamination and Toxicology 76: 463. https:// 2018. “Patterns of Climate Extremes in the doi.org/10.1007/s00128-006-0944-8. Coastal and Highland Regions of Balochistan, Ahmad, M. D., I. Masih, and M. Giordano. 2014. Pakistan.” Earth Interactions. doi: 10.1175​ “Constraints and Opportunities for Water Savings /EI-D-16-0028.1. and Increasing Productivity through Resource ACAPS. 2017. “Pakistan: Sindh Heatwave April–May Conservation Technologies in Pakistan.” Agriculture, 2017.” ACAPS (website). Geneva, Switzerland: Ecosystems and Environment 187: 106–15. 108 PAKISTAN: GETTING MORE FROM WATER Ahmad, S., Z. Hussain, A. S. Qureshi, R. Majeed, Annandale, G. W., G. L. Morris, and P. Karki. 2016. and M. Saleem. 2004. Drought Mitigation in Extending the Life of Reservoirs: Sustainable Pakistan: Current Status and Options for Future Sediment Management for Dams and Run-of- Strategies. Working Paper 85. International Water River Hydropower. Washington, DC: World Bank Management Institute, Colombo. Group. Ahmad, W., A. Fatima, U. K. Awan, and A. Anwar. 2014. Anwar, A, and M. T. Bhatti. 2018. “Pakistan’s Water “Analysis of Long Term Meteorological Trends in Apportionment Accord of 1991: 25 Years and the Middle and Lower Indus Basin of Pakistan: Beyond”. Journal of Water Resources Planning A Non-Parametric Statistical Approach.” Global and Management 144 (1). and Planetary Change 122: 282–91. Anwar, A., M. T. Bhatti, and T. T. de Vries. 2016. “Canal Ahmed, K., M. Shahbaz, A. Qasim, and W. Long. 2015. Operations Planner I: Maximizing Delivery “The Linkages between Deforestation, Energy Performance Ratio.” Journal of Irrigation and and Growth for Environmental Degradation in Drainage Engineering 142 (12): 04016057. Pakistan.” Ecological Indicators 49: 95–103. Ashraf, A., R. Naz, and M. B. Iqbal. 2017. “Altitudinal Ahmed, S. I. 2015. “Lahore’s Killer Canal Awaits Dynamics of Glacial Lakes Under Changing Clean-up.” Thethirdpole.net. Available from: Climate in the Hindu Kush, Karakoram, and https://www.thethirdpole.net/en/2015/11/04​ Himalaya Ranges.” Geomorphology 283: 72–79. /lahores-killer-canal-awaits-clean-up/. Atta-Ur-Rahman, and A. N. Khan. 2013. “Analysis of Akram, A. A. 2013. “Is a Surface-Water Market 2010: Flood Causes, Nature and Magnitude in the Physically Feasible in Pakistan’s Indus Basin Khyber Pakhtunkhwa, Pakistan.” Natural Hazards Irrigation System?” Water International 38 (5): 66: 887. 552–70. Atta-Ur-Rahman, and Shaw, R. (2015). “Introduction and Alamgir, A., M. A. Khan, O. E. Hany, S. S. Shaukat, K. Approaches of Disaster Risk Reduction in Pakistan.” Mehmood, A. Ahmed, S. Ali, K. Riaz, H. Abidi, In Atta-Ur-Rahman, A. Khan, and R. Shaw (ed.s) S. Ahmed, and M. Ghori. 2015. “Public Health Disaster Risk Reduction Approaches in Pakistan. Quality of Drinking Water Supply in Orangi Town, Disaster Risk Reduction (Methods, Approaches and Karachi, Pakistan. Bulletin of Environment, Practices). Springer, Tokyo. Pharmacology and Life Sciences 4: 88–94. Aziz, R., and M. B. Ahmad. 2015. Pakistan’s Power Ali, A. 2013. Indus Basin Floods: Mechanisms, Impacts, Crisis. Special report. Washington, DC: USIP. and Management. Manila, Philippines: ADB. http://www.usip.org/sites/default/files/SR375​ https://www.adb.org/sites/default/files​ -Pakistans-Power-Crisis-The-Way-Forward.pdf. /publication/30431/indus-basin-floods.pdf. Azizullah, A., M. N. K. Khattak, P. Richter, and Ali, J., T. G. Kazi, J. A. Baig, H. I. H. I. Afridi, M. S. Arain, D. P. Hader. 2011. “Water Pollution in Pakistan K. D. Brahman, and A. H. Panhwar. 2015. “Arsenic and its Impact on Public Health: A Review.” in Coal of the Thar Coalfield, Pakistan, and Its Environmental International 37: 479–97. Behavior During Combustion.” Environmental Baig, I. A., M. Ashfaq, S. Hassan, K. Mushtaq, and Science and Pollution Research 22 (11): 8559–66. M. A. Raza. 2009. “An Analysis of Irrigation Ali, K. F., and D. H. De Boer. 2007. “Spatial Patterns Reforms and Their Impact on Rural Livelihood in and Variation of Suspended Sediment Yield in Punjab, Pakistan.” Journal of Agricultural Research the Upper Indus River Basin, Northern Pakistan.” 47 (3): 281–92. Journal of Hydrology 334: 368–87. Bajracharya, S. R., S. B. Maharjan, F. Shrestha, W. Guo, ———. 2010. “Spatially Distributed Erosion and S. Liu, W. Immerzeel, and B. Shrestha. 2015. “The Sediment Yield Modelling in the Upper Indus Glaciers of the Hindu Kush Himalayas: Current River Basin.” Water Resources Research 46: Status and Observed Changes from the 1980s to W08504. 2010.” International Journal of Water Resources Development 31 (2): 161–173. Amir, P., and Z. Habib. 2015. Estimating the Impacts of Climate Change on Sectoral Water Demand Bashir, B. H. 2012. Wastewater Treatment Update. in Pakistan. Nottinghamshire, U.K.: ACT. https:// Presented at the Global Methane Initiative. cyphynets.lums.edu.pk/images/Readings​ https://globalmethane.org/documents/events​ _concluding.pdf. (accessed June 2018) _land_120702_msw_pakistan.pdf. 109 Bell, A. R., M. A. A. Shah, and P. S. Ward. 2014. S. J. Zhu. 2017. “Drinking Water Quality Status “Reimagining Cost Recovery in Pakistan’s and Contamination in Pakistan: A Review.” Bio Irrigation System Through Willingness‐to‐pay Med Research International. Article ID 7908183. Estimates for Irrigation Water from a Discrete https://doi.org​ /10.1155/2017/7908183 Choice Experiment.” Water Resources Research Ellis, P. F., J. S. Fria, and J. K. Kaw. 2018. Transforming 50: 6679–6695. Karachi into a Livable and Competitive Megacity: Bell, A. R., P. S. Ward, and M. A. A. Shah. 2016. A City Diagnostic and Transformation Strategy. “Increased Water Charges Improve Efficiency World Bank: Washington, D.C. and Equity in an Irrigation System. Ecology and Ensink, J. H. J., T. Mahmood, W. van der Hoek, L. Society 21 (3): 23. http://dx.doi.org/10.5751​ Raschid-Sally, and F. P. Amerasinghe. 2004. /ES-08642-210323. “A Nationwide Assessment of Wastewater Use Bhatti, M. T., A. A. Anwar, and M. Aslam. 2017. in Pakistan: An Obscure Activity or a Vitally “Groundwater Monitoring and Management: Important One?” Water Policy 6: 197–206. Status and Options in Pakistan.” Computers and ESMAP (Energy Sector Management Assistance Electronics in Agriculture 13: 143–53. Programme). 2015. Renewable Energy Resource Blackmore, D., F. Hasan. 2005. “Water Rights and Mapping in Pakistan. Washington, DC: ESMAP. Entitlements.” In Pakistan’s Water Economy FAO (Food and Agriculture Organization. 2015. Running Dry; Briscoe, J., and U. Qamar Eds. AQUASTAT Pakistan (database). http://www.fao​ Oxford University Press: Oxford, UK. pp. 201–253. .org/nr/water/aquastat/main/index.stm http://www-wds.worldbank.org/external/ default/WDSContentServer/WDSP/IB/2010/02 ———. 2016. FAO Contributes to Better Planning of /01/000333037_20100201014523/Rendered​ Humanitarian Relief Operations in Pakistan. /PDF/529140WP0Box341University0Press2006.pdf. Islamabad, Pakistan: FAO. Butta, M. N., and L. K. Smedema. 2007. “One Hundred ———. 2016. Pakistan Country Profile: Energy Use. Years of Waterlogging and Salinity Control in FAOSTAT (database). http://www.fao.org/faostat​ the Indus Valley, Pakistan: A Historical Review.” /en/#data/EE. Irrigation and Drainage 56: S81–S90. FoDP (Friends of Democratic Pakistan). 2012. Bradshaw, C. J., N. S. Sodhi, K. S. Peh, and B. W. Brook. A Productive and Water-Secure Pakistan: 2007. “Global Evidence that Deforestation Infrastructure, Institutions, Strategy. Islamabad, Amplifies Flood Risk and Severity in the Pakistan: Water Sector Task Force, FoDP. http:// Developing World.” Global Change Biology 13: metameta.nl/wp-content/uploads/2013/11​ 2379–95. /FoDP-WSTF-Report-Final-09-29-12.pdf. Chapagain, A. K., and A. Y. Hoekstra. 2010. The Green, Ghumman, A. R., S. Ahmad, H. N. Hashmi, and Blue and Grey Water Footprint of Rice from both R. A. Khan. 2014. “Comparative Evaluation a Production and Consumption Perspective. of Implementing Participatory Irrigation Value of Water Research Report Series No. 40. Management in Punjab, Pakistan.” Irrigation UNESCO-IHE Institute for Water Education, Delft, and Drainage 63: 315–27. the Netherlands. Gill, M. A., H. Mujeeb-ur-Rehman, and A. Choudhary. Chaudhry, Q. U. Z. 2017. Climate Change Profile of 2013. “Enhancing Wheat Production and Pakistan. Manila, Philippines: ADB. Productivity through Resource Conservation in Pakistan.” In Proceedings of the Regional Dalin, C., Y. Wada, T. Kastner, and M. J. Puma. 2017. Consultation on Improving Wheat Productivity “Groundwater Depletion Embedded in International in Asia, edited by R. Paroda, S. Dasgupta, Food Trade.” Nature 543 (7647): 700–04. S. Mal, S. S. Singh, M. L. Jat, and S. Gyanendra, Danilenko, A., C. van den Berg, B. Macheve, and pp. 180–88. Bangkok, Thailand, April 26–27, 2012. L. J. Moffitt. 2014. The IBNET Water Supply and International Maize and Wheat Improvement Sanitation Blue Book 2014: The International Centre, CIMMYT. Benchmarking Network for Water and Sanitation Gippel, C. J. 2015. Lower Indus River: Review of Utilities Databook. World Bank, Washington, D.C. Environmental Flows to Maintain the Ecosystem Daud, M. K., M. Nafees, S. Ali, M. Rizwan, R. A. Downstream of Kotri Barrage. Climate Change Bajwa, M. B. Shakoor, M. U. Arshad, S. A. S. Adaptation Project CCAP/KHI/238. Fluvial Chatha, F. Deeba, W. Murad, I. Malook, and Systems Pty Ltd, Stockton, WWF Pakistan, Karachi. 110 PAKISTAN: GETTING MORE FROM WATER Giuliani, M., D. Anghileri, S. Castelletti, P. N. Vu, and R. Planning and Development Department. Soncini-Sessa. 2016. “Large Storage Operations http://metameta.nl/wp-content​ under Climate Change: Expanding Uncertainties /uploads/2017/11/170323_Fourth-Draft​ and Evolving Tradeoffs.” Environmental Research -Irrigation-Managment-Strategy-for-Irrigated​ Letters 11 (3). -Agriculture-of-Sindh-Province-Pakistan.pdf. Gonzales, F. J., T. Basson, and B. Schultz. 2005. Grigg, N., T. Foran, T. Darbas, M. Kirby, M. J. Colloff, Final Report of IPOE for Review Studies on M. -U. -D. Ahmad, and G. Podger. 2018. Water Escapages Below Kotri Barrage. Delft, “The Water–Food–Energy Nexus in Pakistan: Netherlands. A Biophysical and Socio-Economic Challenge.” Proceedings of the International Association of GoP (Government of Pakistan). 2004. Economic Survey Hydrological Sciences 376: 9–13. of Pakistan (2003-04). Ministry of Finance, Islamabad, Pakistan. Guriro, A. 2016. “Karachi’s Catastrophic Pollution of the Sea.” The Third Pole (blog). https:// ———. 2012. Canal Water Pricing for Irrigation in www.thethirdpole.net/en/2016/06/08​ Pakistan: Assessment, Issues and Options. /karachis-catastrophic-pollution-of-the-sea/. Planning Commission, Islamabad, Pakistan. Hassan, G. Z., F. R. Hassan, and S. Akhtar. 2016. ———. 2016a. Pakistan Social and Living Standards “Environmental Issues and Concerns of Measurement Survey 2014–15. Islamabad: GoP, Groundwater in Lahore.” Proceedings of the Statistics Division, Pakistan Bureau of Statistics. Pakistan Academy of Sciences: Pakistan http://www.pbs.gov.pk/sites/default/files​ Academy of Sciences B. Life and Environmental //pslm/publications/PSLM_2014-15_National​ Sciences 53(3): 163–178. -Provincial-District_report.pdf. Haydar, S., Arshad, M. and Aziz, J.A., 2016. Evaluation of ———. 2016b. Household Integrated Economic Survey drinking water quality in urban areas of Pakistan: 2015–16. Islamabad: GoP, Statistics Division, A case study of Southern Lahore. Pakistan Journal Pakistan Bureau of Statistics. http://www.pbs​ of Engineering and Applied Sciences 9, 16–23. .gov.pk/sites/default/files//pslm/publications​ /hies15-16/write%20up%2015-16-HIES-final_0​ Halcrow Group. 2007. Supporting Public Resource .pdf. Management in Balochistan. Basin-Wide Water Resources Availability and Use. Irrigation and ———. 2017a. Pakistan Statistical Yearbook 2016. Power Department, Government of Balochistan, Statistics Division, Bureau of Statistics, Islamabad, Royal Netherlands Government. Halcrow Pakistan Pakistan. http://www.pbs.gov.pk/content​ (Pvt) in association with Cameos. /pakistan-statistical-year-book-2016. ICIMOD (International Centre for Integrated Mountain ———. 2017b. Energy. Chapter 13: “Ministry of Finance Development). 2005. Inventory of Glaciers, Survey.” Islamabad, Pakistan: GoP, Ministry of Glacial Lakes and the Identification of Potential Finance. http://www.finance.gov.pk/survey​ Glacial Lake Outburst Floods (GLOFs) Affected /chapters_17/14-energy.pdf. by Global Warming in the Mountains of India, ———. 2018a. Development of National Flood Pakistan and China/Tibet Autonomous Region. Protection Plan-IV (NFPP-IV) and Related Studies Final report for APN Project 2004-03-CMY- to Enhance Capacity Building of Federal Flood Campbell. ICIMOD: Kathmandu, Nepal. Commission-FFC. Islamabad, Pakistan: GoP, IER (Institute for Energy Research). 2017. “Despite Ministry of Water Resources. the Paris Accord, India and Pakistan Will ———. 2018b. “Potential and Progress in Small Continue to Use Coal.” https://www​ Hydropower.” Alternative Energy Development .instituteforenergyresearch.org/analysis/despite​ Board (website). http://www.aedb.org/ae​ -paris-accord-india-pakistan-will-continue-use​ -technologies/small-hydro. -coal/. ———. 2018c. Water Supply and Sewerage. Faisalabad, IFC (International Finance Corporation). 2016. A Solar Pakistan: GoP, Water and Sanitation Agency, Developer’s Guide to Pakistan. Washington, DC: IFC. Faisalabad. http://wasafaisalabad.gop.pk/. Im, E. S., J. S. Pal, and E. A. B. Elatahir. 2017. “Deadly GoS (Government of Sindh). 2014. Project Coordination Heatwaves Projected in the Densely Populated and Monitoring Unit: Sindh Water Sector Agricultural Regions of South Asia.” Science Improvement Project. Karachi, Pakistan: GoS, Advances 3(8) e1603322. 111 IMF (International Monetary Fund). 2017. Pakistan: Balochistan. Islamabad, Pakistan: LEAD Pakistan. Selected Issues. Washington, DC: IMF. https:// http://www.lead.org.pk/lead/Publications​ www.imf.org/~/media/Files/Publications/CR​ /Sustainable%20Development%20Goal%20 /2017/cr17213.ashx. 13%20-%20A%20Legislative%20and%20 Policy%20Gap%20Analysis%20for%20 IRENA (International Renewable Energy Agency). 2018. Balochistan.pdf. Renewables Readiness Assessment Pakistan. Abu Dhabi, United Arab Emirates: IRENA. Linstead, C., A. H. Sayed, S. A. Naqvi. 2015. “Water Footprint of Key Industrial Sectors of Punjab, Jacoby, H., G. Mansuri, and F. Fatima. 2018. Pakistan”. WWF-Pakistan: Lahore, Pakistan. Decentralization and Redistribution: Irrigation Reform in Pakistan’s Indus Basin. Policy Research Lu, X. X., L. S. Ran, S. Liu, T. Jiang, S. R. Zhang, and Working Paper. World Bank, Washington DC. J. J. Wang. 2013. “Sediment Loads Response to Climate Change: A Preliminary Study of Eight Khair, S. M., S. Mushtaq, and K. Reardon-Smith. 2015. Large Chinese Rivers.” International Journal of “Groundwater Governance in a Water-Starved Sediment Research 28 (1): 1–14. Country: Public Policy, Farmers’ Perceptions, and Drivers of Tubewell Adoption in Balochistan, Lu, X., S. Zhang, and J. Xu. 2010. “Climate Change and Pakistan.” Groundwater 53: 626–37. Sediment Flux from the Roof of the World.” Earth Surface Processes and Landforms 35: 732–735. Khan, F. 2009. “Water, Governance, and Corruption in Pakistan.” In Running on Empty: Pakistan’s Lutz, A. F., W. W. Immerzeel, P. D. A. Kraaijenbrink, Water Crisis. Washington, DC: Woodrow Wilson A. B. Shrestha, and M. F. P. Bierkens. 2016 “Climate International Center for Scholars. Change Impacts on the Upper Indus Hydrology: Sources, Shifts and Extremes.” PLoS One 11(11): Khan, A. N., and S. N. Khan. 2015. Drought Risk e0165630. doi:10.1371/journal.pone.0165630. and Reduction Approaches in Pakistan. In: Atta-Ur-Rahman, A. Khan, and R. Shaw Lutz, A. F., W. W. Immerzeel, A. B. Shrestha, (ed.s) Disaster Risk Reduction Approaches in M. F. P. Bierkens. 2014. “Consistent Increase in Pakistan. Disaster Risk Reduction (Methods, High Asia’s Runoff due to Increasing Glacier Melt Approaches and Practices). Springer, Tokyo. and Precipitation.” Nature Climate Change 4 (7): 587–92. Khan, H. F., Y. E. Yang, C. Ringler, S. Wi, M. J. M. Cheema, and M. Basharat. 2016. “Guiding MacDonald, A. M., H. C. Bonsor, K. M. Ahmed, Groundwater Policy in the Indus Basin of Pakistan W. D. Burgess, M. Basharat, R. C. Calow, Using a Physically Based Groundwater Model.” A. Dixit, S. S. D. Foster, K. Gopal, D. J. Lapworth, Journal of Water Resources Planning and R. M. Lark, M. Moench, A. Mukherjee, M. S. Rao, Management 143 (3): 05016014. M. Shamsudduha, L. Smith, R. G. Taylor, J. Tucker, F. van Steenbergen, and S. K. Yadav. Khan, N. M., and T. Tingsanchali. 2009. “Optimization 2016. “Groundwater Quality and Depletion in and Simulation of Reservoir Operation with the Indo-Gangetic Basin Mapped from In Situ Sediment Evacuation: A Case Study of the Tarbela Observations.” Nature Geosciences 9: 762–8. Dam, Pakistan.” Hydrological Processes 23 (5): 730–47. Mahmood, K., A. D. Rana, S. Tariq, S. Kanwal, R. Ali, A. H. Ali, and T. Tahseen. 2013. “Groundwater Konrad, C. P., A. Warner, and J. V. Higgins. 2012. Levels Susceptibility to Degradation in Lahore “Evaluating Dam Re-Operation for Freshwater Metropolitan.” Science International (Lahore) Conservation in the Sustainable Rivers Project.” 25 (1): 123–126. River Research and Application 28: 777–92. Majeed, S., S. B. Zaman, I. Ali, and S. Ahmad. 2010. Krchnak, K., B. Richter, and G. Thomas. 2009. “Integrating “Situational Analysis of Sindh Coast Issues and Environmental Flows into Hydropower Dam Options.” Managing National Resources for Planning, Design, and Operations (English).” Water Future Agriculture, Research Briefings 2 (11): Working Notes 22, World Bank, Washington, DC. 23pp. LEAD. 2013. “How Can Pakistan Leverage Climate Mansuri, G., F. Sami, T. T. D. Dung, P. Pandey, and Finance?” LEAD (blog). http://www.lead.org.pk​ M. Ali. 2018. When Water Becomes a Hazard: /attachments/updates/318%20(2).pdf. A Diagnostic Report on the State of Water, ———. 2017. Sustainable Development Goal 13: Sanitation and Poverty in Pakistan and Its Impact A Legislative and Policy Gap Analysis for on Child Stunting. Washington, DC: World Bank. 112 PAKISTAN: GETTING MORE FROM WATER Mathison, C., J. Wiltshire, P. Falloon, and A. J. Challinor. ———. 2016. Water Quality Status of Major Cities of 2015. “South Asia River Flow Projections and Pakistan 2015-2016. Pakistan Council of Research their Implications for Water Resources.” Hydrology in Water Resources, Ministry of Science & and Earth System Science. 12: 5789–5840. doi: Technology, Government of Pakistan, Islamabad. 10.5194/hessd-12-5789-2015. Podgorski, J. E., S. A. E. A. S. Musstja, T. Khanam, R. Ullah, Mekonnen, M. M., A. Y. Hoekstra. 2010. “A Global and H. Shen, and M. Berg. 2017. “Extensive Arsenic High-resolution Assessment of the Green, Blue Contamination in High-pH Unconfined Aquifers in and Grey Water Footprint of Wheat. Hydrology the Indus Valley.” Science Advances 3: e1700935. and Earth System Science 14: 1259–1276. Qamer, F., M. K. Shehzad, S. Abbas, M. S. R. Murthy, Memon, J.A., and G. B. Thapa. 2011. “The Indus C. Xi, H. Gilani, and B. Bajracharya. 2016. Irrigation System, Natural Resources, and “Mapping Deforestation and Forest Degradation Community Occupational Quality in the Delta Patterns in Western Himalaya, Pakistan.” Remote Region of Pakistan.” Environmental Management Sensing 8 (5): 385; doi:10.3390/rs8050385. 47:173–187. Qureshi, A. S. 2016. “Perspectives for Bio- Murtaza, G. 2012. “Wastewater Production, Treatment Management of Salt-Affected and Waterlogged and Use in Pakistan.” Presented at the Second Soils in Pakistan.” In J.C. Dagar, P.S. Minhas Regional Workshop of the Project ‘Safe Use of (eds.), Agroforestry for the Management of Wastewater in Agriculture’, 16–18 May 2012, Waterlogged Saline Soils and Poor-quality New Delhi. Waters, Advances in Agroforestry 13. Springer: India. doi: 10.1007/978-81-322-2659-8_6. Mustafa, D. 2001. “Colonial Law, Contemporary Water Issues in Pakistan.” Political Geography 20 (7): Qureshi, A. S., T. Shah, and M. Akhtar. 2003. The 817–837. Groundwater Economy of Pakistan. IWMI Working Paper No. 64. International Water Management Mustafa, D., G. Gioli, M. Karner, and I. Khan. 2017. Institute, Colombo, Sri Lanka. Contested Waters: Subnational Scale Water Conflict in Pakistan. United States Institute of Qureshi, A. S., P. G. McCornick, A. Sarwar, and Peace: Washington D.C. B. R. Sharma. 2010. “Challenges and Prospects of Sustainable Groundwater Management in Mustafa, D., G. Gioli, S. Qazi, R. Waraich, A. Rehman, the Indus Basin, Pakistan.” Water Resources and R. Zahoor. 2015. “Gendering Flood Early Management 24 (8): 1551–69. Warning Systems: The Case of Pakistan.” Environmental Hazards 14 (4): 312–28. Qureshi, D. A. S., and A. H. Sayed. 2014. Situation Analysis of the Water Resources of Lahore: Nassem, S., and McArthur J. M. 2018. “Arsenic and Establishing a Case for Water Stewardship. Other Water-quality Issues Affecting Groundwater, WWF-Pakistan: Lahore, Pakistan. Indus Alluvial Plain, Pakistan.” Hydrological Processes. 2018: 1–19. Rabani, R. R., A. Inam, A. R. Tabrez, N. A. Sayed, and S. M. Tabrez. 2008. The Impact of Sea Level Rise OECD (Organisation of Economic Co-operation and on Pakistan’s Coastal Zone in a Climate Change Development). 2013. Water Resources in Scenario. National Institute of Oceanography. Agriculture: Outlook and Policy Issues. http:// Second International Maritime Conference. Bahria www.oecd.org/agriculture/48498988.pdf. University: Karachi, Pakistan. ———. 2017. Multi-Purpose Water Infrastructure. Rahman, P. 2008. “Water Supply in Karachi: Situation/ Recommendations to Maximize Economic Issues, Priority issues and Solutions.” Orangi Pilot Benefits. Policy perspectives. Paris, France: OECD. Project: Karachi, Pakistan. PCRWR (Pakistan Council of Research in Water Rashid, M. U., A. S. Shakir, and N. M. Khan. 2014. Resources). 2002. Water Quality Status in “Evaluation of Sediment Management Options Pakistan. Pakistan Council of Research in Water and Minimum Operation Levels for Tarbela Resources, Ministry of Science & Technology, Reservoir, Pakistan.” Arabian Journal for Science Government of Pakistan, Islamabad. and Engineering 39 (4): 2655–68. ———. 2004. Water Quality Status in Pakistan: Second Rasul, G., Q. Z. Chaudhry, and A. Mahmood. 2011. Report 2002–2003. Pakistan Council of Research “Glaciers and Glacial Lakes Under Changing in Water Resources, Ministry of Science & Climate in Pakistan.” Pakistan Journal of Technology, Government of Pakistan, Islamabad. Meteorology 8 (15): 1–8. 113 Raza, A., M. Latif, and A. S. Shakir. 2013. “Long-Term SBP (State Bank of Pakistan). 2017. Annual Report Effectiveness of Lining Tertiary Canals in the Indus 2016–2017 (State of the Economy). Islamabad, Basin of Pakistan.” Irrigation and Drainage doi. Pakistan: SBP. org/10.1002/ird.1714. SCCDP. 2012. “Baseline Survey of Coastal Areas.” Sindh Raza, M., F. Hussain, J. L. Lee, M. B. Bilal Shakoor, Coastal Community Development Project, Vol 1. and K. D. Kwon. 2017. “Groundwater Status in Sindh Coastal Development Authority. Pakistan: A Review of Contamination, Health Sial, R. A., M. F. Chaudhary, S. T. Abbas, M. I. Latif, and Risks, and Potential Needs.” Critical Reviews in A. G. Khan. 2006. Quality of Effluents from Hattar Environmental Science and Technology 47 (18): Industrial Estate. Journal of Zhejiang University 1713–62. SCIENCE B 7: 974–980. Rinaudo, J. -D. 2002. “Corruption and Allocation of Stewart, J. P., G. M. Podger, M. D. Ahmad, M. A. Shah, Water: The Case of Public Irrigation in Pakistan.” H. Bodla, Z. Khero, and M. K. I. Rana. 2018. Water Policy 4: 405–22. Indus River System Model (IRSM)—A Planning Roca, M. 2012. Tarbela Dam in Pakistan: Case Study Tool to Explore Water Management Options in of Reservoir Sedimentation. HR Wallingford. Pakistan: Model Conceptualisation, Configuration http://futuredirections.org.au/wpcontent​ and Calibration. August 2018. Technical report. /uploads/2016/01/HRPP533_Tarbela_Dam​ South Asia Sustainable Development Investment _in_Pakistan._Case_study_of_reservoir​ Portfolio (SDIP) Project. Australia: CSIRO. _sedimentation.pdf. Tariq, M., and R. Aziz. 2015. “An Overview of Saher, F., M. Nasly, T. Kadir, N. Yahaya, and Deforestation Causes and Its Environmental W. wan Ishak. 2015. “Potential Source for Hazards in Khyber Pukhtunkhwa.” Journal of Irrigation.” Journal of Flood Risk Management Natural Sciences Research 5 (1): 52–57. 8: 87–95. Tariq, M., and N. van de Giesen. 2012. “Floods and Flood Management in Pakistan.” Journal of Sánchez-Triana, E., S. Enriquez, B. Larsen, P. Webster, Physics and Chemistry of the Earth 47–48: 11–20. J. Afzal. 2015. Sustainability and Poverty Alleviation: Confronting Environmental Threats UNDP (United Nations Development Programme). in Sindh, Pakistan. Directions in Development: 2015. Scaling-up of Glacial Lake Outburst Flood Environment and Sustainable Development. (GLOF) Risk Reduction in Northern Pakistan. Washington, DC: World Bank. Project proposal submitted to Green Climate Fund. http://www.greenclimate.fund/-/scaling-up-of​ Shah, M. A., A. S. Shakir, and S. Masood. 2011. A Rude -lake-outburst-flood-glof-risk-reduction-in​ -glacial​ Awakening: Report of the Judicial Flood Enquiry -northern-pakistan [accessed 09 Sep 2017]. Tribunal, 2010. Lahore, Pakistan: Judicial Enquiry Commission. 2017. The Vulnerability of Pakistan’s Water Sector to the Impacts of Climate Change: Identification of Gaps Shah, M. A. A., A. A. Anwar, A. R. Bell, and Z. Ul Haq. and Recommendations for Action. Islamabad, 2016. “Equity in a Tertiary Canal of The Indus Pakistan: UNDP. http://www​ .pk.undp.org/content​ Basin Irrigation System (Ibis). Agricultural Water /dam/pakistan/docs​ /Environment%20&%20 Management 178: 201–14. Climate%20Change​ /Report.pdf. Shariff, K. 2015. “Drought Management and Risk UNFCC. 2015. Pakistan’s Intended Nationally Determined Reduction in Pakistan.” Presentation, LEAD Contribution (PAK-INDC). Islamabad: Ministry of Pakistan, Islamabad, Pakistan. http://www.lead​ Climate Change. .org.pk/lead/pages/img/presentations/D3%20 (Drought%20Management%20and%20Risk%20 van Gils, H., and M.S. Baig M.S. 1992. Environmental Reduction%20in%20Pakistan)-%20Brig.%20 Profile: Balochistan, Pakistan. Land Resource Kamran%20Shariff.pptx. and Urban Sciences Department, International Institute for Aerospace Survey and Earth Sciences: Shrestha, A. B., S. H. Shah, and R. Karim. 2008. Enschede, Netherlands. Resource Manual on Flash Flood Risk Management. Kathmandu, Nepal: ICIMOD. van Halsema, G. E., and L. Vincent. 2006. “Of Flumes, Modules and Barrels: The Failure of Irrigation Siddiqi, A., and J. L. Wescoat Jr. 2013. “Energy Use in Institutions and Technologies to Achieve Equitable Large-Scale Irrigated Agriculture in The Punjab Water Control in the Indus Basin.” In A History Province of Pakistan.” Water International 38 (5): of Water, pp. 55–91, edited by T. Tvedt and 571–86. E. Jakobsson. London, U.K.: I. B. Tauris & Co. 114 PAKISTAN: GETTING MORE FROM WATER van Steenbergen, F., A. B. Kaisarani, N. U. Khan, and ———. 2018. “Pakistan: Hydromet and Disaster Risk M. S. Gohar. 2015. “A Case of Groundwater Management Services Project.” Project Appraisal Depletion in Balochistan, Pakistan: Enter into the Document, World Bank, Washington, D.C. Void.” Journal of Hydrology: Regional Studies WMO (World Meteorological Organization). 1994. Guide 4 (A): 36–47. to Hydrological Practices. WMO-164, 5th edition. Wakeel, M., B. Chen, and S. Jahangir. 2016. “Overview Geneva: World Meteorological Organization. of Energy Portfolio in Pakistan.” Energy Procedia WWF (World Wildlife Fund). 2012. Development of 88: 71–75. Integrated River Basin Management for Indus Walling, D. E. 2009. The Impact of Global Change Basin. Islamabad, Pakistan: WWF. https://www​ on Erosion and Sediment Transport by Rivers: .wwfpak.org/publication/pdf/irbm_2013.pdf. Current Progress and Future Challenges. Paris, Yang, Y. C. E., C. Ringler, C. Brown, G. Alam, A. F. Khan, France: UNESCO. T. Akhtar, and K. Mehmood. 2014. “Intersectoral WAPDA. 2007. “Waterlogging, Salinity and Drainage Water Allocation in the Indus Basin under Situation.” SCARP Monitoring Organization. Water Different Management Policies.” Working Paper and Power Development Authority. Lahore, 024, IFPRI, Washington, DC. Pakistan. Yu, W., Y. C. E. Yang, A. Savitsky, D. Alford, C. Brown, Webster, P. J., V. E. Toma, and H. M. Kim. 2011. J. L. Wescoat, Jr., D. Debowicz, and S. Robinson. “Were the 2010 Pakistan Floods Predictable?” 2013. Indus Basin of Pakistan: Impacts of Climate Geophysical Research Letters 38: L04806. Risks on Water and Agriculture. Directions in Development; Countries and Regions. WFP (World Food Program). 2015. Pakistan Food Washington, DC: World Bank. Security Bulletin (August 2015) Issue 3. http:// documents.wfp.org/stellent/groups/public​ Zahid, M., and G. Rasul. 2012. “Changing Trends of Thermal /documents/ena/wfp278089.pdf. Extremes in Pakistan.” Climatic Change 113: 883. WHO/UNICEF. 2015. Joint Monitoring Program. https:// Zeng, R., X. Cai, C. Ringler, and T. Zhu. 2017. “Hydropower washdata.org/. Versus Irrigation—An Analysis of Global Patterns.” Environmental Research Letters 12 (3): 034006. World Bank. 2016. Punjab Service Delivery Assessment: A Decision-making Tool for Transforming Funds Zulfiqar, F., and G. B. Thapa. 2017. “Agricultural into Improved Services. Water and Sanitation Sustainability Assessment at Provincial Level in Program. World Bank, Washington, D.C. Pakistan.” Land Use Policy 68: 492–502. C HAPT E R 6 Water Security Trajectories Key Messages • Reaching upper-middle-income (RUMI) status by 2047 is an ambitious goal that will require a significant change in the structure of the economy. The services sector must increase, while the agricultural sector must shrink. In absolute terms, however, the agricultural sector must continue to grow to meet rising food demands. • Without significant reform and demand management, water demand could increase 50 percent by 2047 to significantly exceed supply. Population and economic growth will be the dominant drivers of demand increase, but climate warming will contribute significantly. The largest increases will be for irrigation, while the fastest rates of increase will be for domestic and industrial use. • Despite projected population increase and climate change, water scarcity will not prevent Pakistan from RUMI status. Water consumption in agriculture can increase provided major improvements in water use efficiency are achieved to reduce losses. Even so, within a few decades, increasing municipal and industrial demand will restrict any further increase in agricultural water use. • As incomes rise, diets will change to reflect more expensive but more nutritious choices. A falling demand for basic cereals will enable water to move to higher-value crops (either to meet changing domestic demands or new exports) or to other more economically profitable sectors. • Current subsides for wheat and sugarcane should be phased out. This will encourage diversification toward higher-value commodities to help meet changing consumer food preferences and to deliver significant trade dividends. • Pakistan’s major agricultural exports consume a large fraction of the water used, and profitability is sensitive to international prices. The sector needs to become more responsive to changing international prices and to variations in water availability. This will increase the economic returns from water, while prioritizing social water needs. • Increased flows below Kotri Barrage will become increasingly important in the future both to meet the increasing demand for Karachi and to restore and sustain the Indus Delta. An increase in end-of-system flows may reduce agricultural production slightly, but the value to Karachi and the environmental benefits would far exceed these losses. • If the required reforms and performance improvements are not achieved, the consequences will be significant. Water capture by agriculture would increasingly affect the ability to provide adequate water services to industrial and service sectors. Even a 5 percent impact on productivity in industry and services is equivalent to 70 percent of the value of the four major irrigated crops. • Recent gains in agricultural water productivity have relied on unsustainable groundwater exploitation. A continuation of business as usual and the current slow rate of economic growth would exacerbate groundwater depletion, fail to address declining health of the Indus Delta, and would most likely see urban water security decline. 116 PAKISTAN: GETTING MORE FROM WATER Introduction and vegetables. This shift can have significant impact on agricultural water use. This chapter explores a range of water security trajectories out to 2047. The primary analytical basis Two other RUMI variants are modeled (both at for this chapter is scenario modeling of the Pakistan the higher warming level): (i) one that simulates economy using a computable general equilibrium (CGE) agricultural policy reform and changes in international model of the Pakistan economy coupled to a water trade, and (ii) one that simulates increased system model of the Indus Basin. Appendix C provides environmental sustainability through provision of a description of the coupled CGE-W model and its additional environmental flows to the Indus Delta. The assumptions, plus comparisons to prior hydro-economic first assesses removal of current policies that artificially modeling for Pakistan. raise the internal prices for wheat and sugarcane and assesses changes in the trade outlook for rice and The macro drivers of change in the model are cotton. Although considered a single variant here economic growth, population growth, urbanization, for simplicity, these agricultural reforms and trade and climate change, with economic growth driven by outlook changes are considered both separately and in increases in productivity, labor, capital, and land and combination in the modeling. The RUMI environmental water resources. To explore the effects of these macro variant increases freshwater flow below the Kori drivers and alternative water policies on economic Barrage, thus reducing water for irrigation. The different and environmental outcomes from water, a set of scenarios are summarized in table 6.1. future scenarios is defined and modeled as 33-year simulations from the base year of 2013/14 (table 6.1). In recent years, Pakistan’s economic growth has Assumed rates of economic growth are a key aspect been the slowest of a cohort of comparator countries of scenario definition, while population growth and (table 6.2). By 2047, Pakistan could reach India’s current urbanization are represented in labor force changes, GDP per capita with just a 1.4 percent growth rate, less increasing demands for food and water, and sectoral than the longer-term rate of 1.9 percent in the country. shifts in water demands. Water availability projections However, if Pakistan and comparator countries maintain for Pakistan are uncertain, and so consideration current rates of growth, Pakistan’s GDP per capita would of climate change is limited to warming, which be only halfway to the average income of low- and significantly affects water demand in all sectors middle-income countries (LICs and MICs). (chapter 5). The scenarios explore the role of water policy and management in determining economic The key to faster growth is productivity increases in outcomes in the context of population growth, climate all sectors of the economy. One of the easiest ways to change, and changing consumer preferences, and thus increase productivity is increased output per laborer, describe a range of potential water security futures. which is also a necessary to raise household incomes. The best growth in output per laborer has been in the The baseline BAU scenario is a continuation of the services sector (at 1.1 percent per year), while industrial current rate of growth, both overall and by economic and agricultural sector growth have stalled or declined, sector, to reach gross domestic product (GDP) per partly reflecting a rapidly growing but largely unskilled capita of about US$2,200 by 2047. A RUMI scenario labor force (figure 6.1). Agricultural yields have been explores the plausibility of Pakistan attaining a per growing at 1.2 percent to 1.8 percent (figure 2.3), capita income of US$6,000 by 2047. This requires an but less than half of this growth is from increased annual GDP per capita growth rate of 4.9 percent— productivity, with the rest coming from increased inputs. higher than the comparator countries have achieved. Both productivity and input increases have been slowing This is a stretch goal for Pakistan but illustrates the in the last decade. In addition to output per laborer, the importance of water security for economic growth. contribution to employment is important. In the 1990s, Several variants of BAU and RUMI are explored. The agriculture fell from 50 percent to about 42 percent of base case for both BAU and RUMI includes moderate the economy. Output per laborer is highest in services: climate change, described simply as a 1 degree expansion of the services sector provided the potential Celsius rise in mean annual temperature by 2047, for higher incomes. Current output per laborer and consistent with the recent rates of warming. A climate employment levels were used to set productivity growth change variant of BAU and RUMI explores more rapid rates by sector for BAU and these were increased for warming—a 3 degrees Celsius increase in mean annual RUMI to simulate economic growth (table 6.1). temperature by 2047. For the faster warming variant of RUMI, another variant explores the impacts of changing consumer preferences. As incomes and education Structural Change in the Economy improve, dietary preferences typically move away from The modeling simulates structural transformation of the cereals, fats, and sugar to include more protein, fruit, economy, largely driven by changing labor productivity 117 Table 6.1  Summary of the Scenarios for Pakistan Modeled and Analyzed using CGE-W Scenario Description Overall annual Assumed Consumer Climate change GDP growth productivity preferences growth by sector BAU-Lo Business as usual with current rate 1.9% to reach Agriculture 0.65% Unchanged Baseline rate of climate of climate warming. US$2,200 per Industry 0.5% warming: 1°C increase in capita by 2047. Services 1.0% mean annual tempera- ture by 2047. BAU-Hi Business as usual with faster rate of 1.9% to reach Agriculture 0.65% Unchanged Faster rate of climate climate warming. US$2,200 per Industry 0.5% warming: 3°C increase in capita by 2047. Services 1.0% mean annual tempera- ture by 2047. RUMI-Lo Accelerated economic growth with 4.9% to reach Agriculture 1.68% Unchanged Baseline rate of climate current rate of climate warming. US$6,000 per Livestock 1.44% warming: 1°C increase in capita by 2047. Industry 1.32% mean annual tempera- Services 2.5% ture by 2047. RUMI-Hi Accelerated economic growth with 4.9% to reach Agriculture 1.68% Unchanged Faster rate of climate faster rate of climate warming. US$6,000 per Livestock 1.44% warming: 3°C increase in capita by 2047. Industry 1.32% mean annual tempera- Services 2.5% ture by 2047. RUMI-Hi- Accelerated economic growth, 4.9% to reach Agriculture 1.68% Dietary shift Faster rate of climate Diet current rate of climate warming and US$6,000 per Livestock 1.44% to more warming: 3°C increase in dietary shifts. capita by 2047. Industry 1.32% meat, dairy, mean annual tempera- Services 2.5% and fruit. ture by 2047. RUMI-Hi- Accelerated economic growth, 4.9% to reach Agriculture 1.68% Unchanged Faster rate of climate Reform current rate of climate warming, US$6,000 per Livestock 1.44% warming: 3°C increase in agricultural policy reforms (wheat capita by 2047. Industry 1.32% mean annual tempera- and sugarcane taxed) and trade Services 2.6% ture by 2047. shifts (international prices for rice and textiles reduced). RUMI-Hi- Accelerated economic growth, 4.9% to reach Agriculture 1.68% Unchanged Faster rate of climate Env current rate of climate warming and US$6,000 per Livestock 1.44% warming: 3°C increase in increased environmental flows to capita by 2047. Industry 1.32% mean annual tempera- the Indus Delta. Services 2.7% ture by 2047. Note: CGE-W = computable general equilibrium-water model; GDP = gross domestic product. Table 6.2  GDP per Capita and Average GDP Growth Rate for 1970–2016 in Pakistan and Comparator Countries, with Growth Required for Pakistan to Reach Comparator Growth Rate by 2047, and Share of Food Expenditure Country GDP per capita GDP growth GDP growth rate for Pakistan to reach Food expenditure (US$, 2010) rate (%) comparator growth rate (%) (% of total expenditure) Pakistan 1,200 1.9 n.a. 37 India 1,900 3.5 1.4 — Egypt, Arap Rep. 2,700 2.6 2.5 50 Indonesia 4,000 3.5 3.8 42 Turkey 11,100 2.6 7.0 — Malaysia 14,100 3.6 7.7 21 LICs and MICs average 4,400 2.6 4.1 — Source: World Bank data and calculations. Note: GDP = gross domestic product; LIC = low-income country; MIC = middle-income country; n.a. = not applicable; — = not available; low- income countries are those with gross national incomes (calculated using the World Bank Atlas method) of $1,025 or less in 2015; middle-income countries are those with gross national incomes per capita between $1,026 and $12,475. 118 PAKISTAN: GETTING MORE FROM WATER and hence an ability to pay higher wages. Higher its share (across both agricultural and other processing incomes slow the growth in demand for agricultural and manufacturing sectors), while the services sector and other basic goods, so the agricultural sector grows share grows. Thus, demand for health, education, and more slowly than other sectors. Rates of structural financial services expands more than the demand for transformation reflect the rates of economic growth manufactured goods, and relative productivity gains in and the productivity improvements of the major sectors services makes them more competitive. (figure 6.2, panels a–c). Under RUMI-Hi, the service sector share increases by Under BAU-Lo, the agricultural share declines by 4 percent to become 58.5 percent of the economy, 3.5 percent by 2047 as income per capita grows and the agriculture sector share declines by 5 percent, and demand for agricultural goods slows relative to other the industry share remains steady. The overall demand commodities. The industrial sector largely maintains structure is the same in BAU and RUMI, so emergent differences are the result of sectoral productivity differences, which are highest in services and lowest Figure 6.1  Output per Laborer, by Sector, in in industry. Nonagricultural industry, with slightly lower Pakistan, 1991–2016 productivity, but better demand prospects, increases 60 it share marginally. The higher productivity growth in agriculture under RUMI reduces key input costs for 50 agriculture-related industries, which partially offset the 40 effects of slowing, allowing this sector to maintain its proportional share in the economy. The differentials US$, 2010 30 between BAU and RUMI are shown in figure 6.2, panel c. 20 10 Future Water Demand and Use 0 The CGE-W water balance (appendix C) sets the 1991 1996 2001 2006 2011 2016 context for modeled water use. Agriculture dominates Agriculture Industry Services withdrawals, although less than 60 percent of water Source: GoP 2016a. Figure 6.2  Changes in Sector Shares under BAU-Lo and RUMI-Hi by 2031 and 2047, and Differential between BAU-Lo and RUMI-Hi, in Pakistan, 2014 Baseline a. BAU-Lo b. RUMI-Hi c. Differential between RUMI-Hi and BAU-Lo 5 5 5 4 4 4 3 3 3 2 2 2 1 1 1 Percent Percent Percent 0 0 0 –1 –1 –1 –2 –2 –2 –3 –3 –3 –4 –4 –4 –5 –5 –5 re try try es re try y es re try y es str str ltu ltu ltu ic ic ic s us s us du du u u rv rv rv u u icu nd nd d nd Se Se Se ric ric in in in r ri ri ri Ag Ag Ag ed ed ed he he he lat lat lat Ot Ot Ot -re -re -re Ag Ag Ag BAU-Lo: 2031–2014 RUMI-Hi: 2031–2014 2031: RUMI-Hi – BAU-Lo BAU-Lo: 2047–2014 RUMI-Hi: 2047–2014 2047: RUMI-Hi – BAU-Lo Source: World Bank data. Note: BAU-Lo = business as usual (with current rate of climate warming); RUMI-Hi = reaching upper-middle-income (accelerated economic growth with faster rate of climate warming). 119 withdrawn is consumed by crops. Of the water importance of a greatly increased emphasis on demand consumed by irrigation, 80.3 percent is for wheat, management. Dimensions of future water use— sugarcane, rice, and cotton. Wheat and sugarcane have total amounts, intersectoral shifts, and shifts within politically sensitive policies that generate artificially agriculture—are explored using CGE-W, in the context of high prices, while rice and cotton dominate current accelerated economic growth. exports. At the commencement of simulations, CGE-W correctly captures the very significant losses of evaporation and precipitation are roughly in balance. water in the distribution system; however, the modeled With rising temperatures in all scenarios, but no change flows downstream of Kotri Barrage are higher than in inflows, water demands rise relative to availability observed, indicating that field-scale water use is more (see chapter 5). efficient in the model than in reality. This means that In addition to climate change and increasing increasing water demand is met until 2038 in the population, urbanization and economic growth drive model—longer than would be expected in reality. In increases in water demand outside of agriculture. the model, water required by industry (including water Domestic and industrial demand will grow several- for livestock) is determined by the level of industrial fold by 2050 because of greater household incomes output: the faster the economy grows, the greater the and industrial activities (figure 6.3, panels a and b). industrial activity, and the more water that industry Growth—population and economic—is the biggest demands. Similarly, growth in domestic water demand driver of demand increases across all sectors. In the is driven by increasing household expenditure: as GDP absence of demand management, faster warming per capita increases, so does domestic water demand. would cause significant additional increases, with the Under RUMI-Hi, domestic water use becomes 7.2 billion maximum projected water demand 58 percent higher cubic meters higher than under BAU-Hi, and industrial than now. Agriculture will continue to dominate water and livestock water demand becomes 5.0 billion cubic demands (figure 6.4). meters higher (figure 6.5). Amir and Habib (2015) demand projections include CGE-W demand projections differ from those of Amir assumptions for irrigation distribution efficiency and and Habib (2015), partly because of differences in economic growth, which differ from the assumptions how livestock demand is categorized, and because embedded in CGE-W. Nonetheless, in the absence of in the modeling, domestic and industrial demands reform or major structural change to the economy, are met from groundwater, except for Karachi, for the patterns of relative increase and approximate which demand is met from flows below Kotri Barrage. magnitude are realistic. The total projected demand by CGE-W nonagricultural demand projections (figure 6.5) 2047 well exceeds the available water, highlighting the are thus underestimates because they exclude Figure 6.3  Projected Water Demand Increases, Relative and Absolute, Attributable to Slow and Fast Climate Warming and Growth, by Sector, 2025 and 2050 a. Relative increases b. Absolute increases 250 40 35 200 Cubic meters (billions) 30 150 25 Percent 20 100 15 10 50 5 0 0 2025 2050 2025 2050 2025 2050 2025 2050 2025 2050 2025 2050 Climate driven Climate driven Growth-driven Climate driven Climate driven Growth-driven increase increase increase increase increase increase (1ºC warming) (3ºC warming) (1ºC warming) (3ºC warming) Agriculture (crops and livestock) Domestic/urban Industry Environment Source: Amir and Habib 2015. 120 PAKISTAN: GETTING MORE FROM WATER Figure 6.4  Pakistan Total Water Demand in 2015 and Projected for 2025 and 2050 300 250 Cubic meters (billions) 200 150 100 50 0 2015 2025 2050 2025 2050 2025 2050 No climate change 1ºC warmer 3ºC warmer Agriculture Domestic/urban Industry Environment Source: Amir and Habib 2015. Figure 6.5  Modeled Annual Nonagricultural Water Demands, by Scenario, 2014–47 20 18 16 Cubic meters (billions) 14 12 10 8 6 4 2 0 14 18 42 44 46 16 20 22 24 26 28 30 32 34 36 38 40 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Industry (BAU) Industry (RUMI) Domestic (BAU) Domestic (RUMI) Source: CGE-W simulations. Note: RUMI = reaching upper-middle-income; BAU = business as usual. Karachi demand. Significant industrial and domestic economic growth reduces the relative contribution of demand growth will be a growing challenge for water agriculture to the economy. Faster climate warming resources management. The nonagricultural demand could cause rapid increases in irrigation water use growth will not of course end in 2047, so long-term (figure 6.6). Irrigation water use is similar under planning is required. Most industrial and domestic BAU and RUMI in most years; however, late in the use is nonconsumptive, so there will be increased simulation period under a faster warming climate, opportunities for wastewater reuse in agriculture. irrigation water use declines under RUMI (this Untreated wastewater is too polluted for safe use in trend continues beyond 2047), because growth in many agricultural applications, and detailed economic nonagricultural demands constrains availability of and technical analysis of wastewater treatment and water for irrigation. reuse options will be required. Groundwater consumption in irrigation changes and Water consumption in agriculture continues to is strongly influenced by the rate of climate warming increase to meet growing demand, even while faster and the level of economic growth (figure 6.7). 121 Figure 6.6  Modeled Annual Irrigation Water Consumption, by Scenario, 2014–47 110 105 100 Cubic meters (billions) 95 90 85 80 75 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 RUMI-Lo RUMI-Hi BAU-Lo BAU-Hi Source: CGE-W simulations. Note: BAU-Hi = business as usual with faster rate of climate warming; BAU-Lo = business as usual (with current rate of climate warming); RUMI-Hi = reaching upper-middle-income (accelerated economic growth with faster rate of climate warming); RUMI-Lo = reaching upper-middle-income (accelerated economic growth with current rate of climate warming). Figure 6.7  Modeled Annual Groundwater Irrigation Consumption, by Scenario, 2014–47 65 60 Cubic meters (billions) 55 50 45 40 35 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 RUMI-Lo RUMI-Hi BAU-Lo BAU-Hi Source: CGE-W simulations. Note: BAU-Hi = business as usual with faster rate of climate warming; BAU-Lo = business as usual (with current rate of climate warming); RUMI-Hi = reaching upper-middle-income (accelerated economic growth with faster rate of climate warming); RUMI-Lo = reaching upper-middle-income (accelerated economic growth with current rate of climate warming). Until 2030, groundwater use varies significantly Around four-fifths of the water used in agriculture across years in response to changing surface water irrigates four major crops—wheat, sugarcane, rice, availability. From 2030, groundwater use in irrigation and cotton. Under BAU-Lo, the volume of water begins to decline as nonagricultural demand for used by wheat, sugarcane, and cotton increases groundwater increases. Under a faster warming while water use for rice slowly declines (figure 6.8). climate, there is reduced variability between Water use by wheat varies more between years years after 2030, because the maximum available than for other crops, because rabi water supply is groundwater is used each year. less reliable. Nearly half of current rice production 122 PAKISTAN: GETTING MORE FROM WATER is exported. If prices remain steady, water for cotton constraint revealed here would be reached sooner, becomes an increasingly better option than for rice. unless current field-level inefficiencies were Cotton production supports exports of yarn, cloth, and reduced. A mix of policy reforms, improved water garments, whose higher value leads to a transfer of management, and infrastructure and technology water away from rice. investments will be required to that ensure that nonagricultural demands are met, and that Under RUMI-Hi, total irrigation water use grows agricultural productivity growth continues. faster than under BAU because of greater economic activity and the effects of faster warming CGE-W provides information on crop water productivity (figure 6.9). Total water use peaks around 2038, (table 6.3). Baseline productivity varies from after which nonagricultural demands constrain US$0.13 per cubic meter for rice to US$1.57 per growth in agricultural water use. Within irrigation, cubic meter for maize in 2013/14 prices. Of the major water moves away from cotton (and this trend crops, wheat has the highest water productivity. continues beyond 2047) given increasing domestic Productivity growth across all commodities is twice food demand. In reality, the irrigation supply as high under RUMI as under BAU, highlighting the Figure 6.8  Modeled Annual Crop Water Use in Pakistan under BAU-Lo, 2014–2047 30 Individual crop water use (m3, billions) 100 Total crop water use (m3, billions) 25 80 20 60 15 40 10 5 20 0 0 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Sugarcane Cotton Wheat Rice Total Source: CGE-W simulations. Note: BAU-Lo = business as usual (with current rate of climate warming). Figure 6.9  Modeled Annual Crop Water Use in Pakistan Under RUMI-Hi, 2014–2047 30 Individual crop water use (m3, billions) 100 Total crop water use (m3, billions) 25 80 20 60 15 40 10 5 20 0 0 4 6 8 0 2 4 6 8 0 2 4 6 8 40 42 44 46 1 1 1 2 2 2 2 2 3 3 3 3 3 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Sugarcane Cotton Wheat Rice Total Source: CGE-W simulations. Note: RUMI-Hi = reaching upper-middle-income (accelerated economic growth with faster rate of climate warming). 123 Table 6.3  Water Productivity in Pakistan by Crop Proportional expenditure on staples does not appear to for Baseline Year (2013/14) and Productivity fall rapidly with rising income across these countries. Growth Rates under BAU and RUMI In Pakistan, 17.8 percent of food expenditure is on cereals. In the Arab Republic of Egypt, with more than Baseline Water Water double the per capita income, the percentage is slightly water productivity productivity higher. In Indonesia, with triple the per capita income, productivity growth rate growth rate the level is 25 percent. (US$/m3) (BAU, %) (RUMI, %) Wheat 0.42 1.5 3.4 Changes in consumer preferences are expected to have a major impact on the patterns of irrigation Rice 0.13 1.3 2.7 water use (figure 6.10). While total irrigation use Cotton 0.24 1.6 3.2 is not affected, a reduction in demand for cereals Sugarcane 0.20 1.2 2.6 allows more cotton to be grown and the export Maize 1.57 1.7 3.3 share of textile production rises from 31 percent to 41 percent. Eventually, increasing water demand Potato 0.53 1.1 3.4 outside agriculture constrains cotton production, which Vegetables 0.16 1.4 2.9 peaks and stabilizes around 2047. Other agricultural Other crops 0.61 1.5 4.4 or nonagricultural commodities could replace cotton Fruit 0.29 1.9 3.7 given changing preferences and demands—cotton Average is simply the most profitable option in the model 0.34 1.5 3.2 given the current configuration. Limits on even the Source: CGE-W simulations. best agricultural options are therefore likely to be Note: BAU = business as usual; RUMI = reaching upper-middle-income. encountered in the next few decades under strong economic growth and significant climate change. A range of policy options, awareness campaigns, potential for higher incomes. RUMI will require rapid education, information, and promotion of healthy improvements in water productivity. lifestyles can support this transition. Comparison of the modeling results with Pakistan Changing Consumer Preferences household survey data informs interpretation. Survey Consumption patterns change with income level, even data indicate how the level and composition of food though Pakistan households spend less of their income expenditure changes with income and captures on food than those in wealthier comparator countries cultural and supply differences better than inter- (table 6.2). This partly reflects higher urban populations country comparisons. Figure 6.11 illustrates how and an ability to meet more of the food demand at less absolute expenditures on food categories change as than international prices in these comparator countries. Pakistani households get richer. If relative expenditure Figure 6.10  Modeled Annual Crop Water Use in Pakistan under RUMI-Hi-Diet, 2014–2047 45 Individual crop water use (m3, billions) 100 Total crop water use (m3, billions) 40 35 80 30 25 60 20 15 40 10 20 5 0 0 4 6 8 0 2 4 6 8 0 2 4 6 8 0 2 4 6 1 1 1 2 2 2 2 2 3 3 3 3 3 4 4 4 4 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Sugarcane Cotton Wheat Rice Total Source: CGE-W simulations. Note: RUMI-Hi-Diet = reaching upper-middle-income (accelerated economic growth, current rate of climate warming and dietary shifts). 124 PAKISTAN: GETTING MORE FROM WATER Figure 6.11  Monthly Household Expenditure on Table 6.4  Modeled Growth in Commodity Food Groups in Pakistan by Income Quintile, 2015 Consumption by Scenario in Pakistan percent Q5 BAU-Hi RUMI-Hi RUMI-Hi-Diet Q4 Wheat 1.3 1.9 0.7 Rice 2.4 3.8 2.9 Q3 Fruit and vegetables 1.5 3.1 3.0 Livestock 3.4 4.4 3.7 Q2 Dairy 4.3 5.1 3.6 Q1 Sugar 1.3 2.7 0.5 Source: CGE-W simulations. 0 20 40 60 80 100 120 140 160 Note: BAU-Hi = business as usual with faster rate of climate warming; US$ RUMI-Hi = reaching upper-middle-income (accelerated economic Wheat Rice Meat Edible oil Pulses growth with faster rate of climate warming); RUMI-Hi-Diet = reaching Fruit Vegetable Sugar Milk Others upper-middle-income (accelerated economic growth, current rate of climate warming and dietary shifts). Source: GoP HIES 2016b. on a commodity declines with rising income, demand of the wheat procurement program and reform increase will tend toward the rate of population of indicative sugarcane prices, which jointly growth, as average income increases through time. cause domestic prices for these water-hungry If expenditure on a commodity increases with income, commodities to be well above international prices. and if water availability to irrigation decreases Cotton and rice are the main agricultural exports, given inter-sectoral competition, a decrease in but their dominance will depend on international exports (or an increase in imports) will be required prices. Vested interests and spurious food security to meet demand. Wheat is the only commodity for arguments have prevented policy reform. Despite which absolute expenditures falls across quintiles. many attempts to move toward high-value Expenditure on fruit, meat and milk all rise by produce, the base of exports remains tied to major between US$8 to more than US$18 per month. The lower-value, high water use crops, especially largest increases are between the fourth and fifth cotton. To simulate the removal of support to quintiles. Consumption of sugar is very similar across wheat and sugarcane, equivalent taxes on these the lower four quintiles. commodities are introduced (17.8 percent for wheat and 18.4 percent for sugarcane). To explore Under RUMI, faster growth shifts the income the implications of international prices for rice and distribution upward. Introducing changes in consumer cotton on water use and economic outcomes, prices preferences causes further changes in demand, are reduced by 0.7 percent each year. consumption, and hence water use patterns. RUMI-Hi-Diet displays minimal growth in wheat Policy reforms for wheat and sugarcane reduce water consumption, while consumption of vegetables, use for these crops by 1.4 billion cubic meters and livestock, and sugar increase, as suggested by the 1.8 billion cubic meters, respectively (table 6.5). survey results (table 6.4). With economic growth, As no economic value is placed on environmental consumption will shift toward a more nutritious and water below Kotri Barrage in the model, water stays diversified diet that improves the well-being of the in agriculture, simply moving to other commodities. population, while at the same time improving water These policy reforms alone would not reduce security, provided that the necessary reforms and agricultural water use, but simply redistribute water investments are made. within agriculture. Falling export prices for rice and cotton (column 5, table 6.5) shifts production toward meeting food domestic demand, with a small decline Policy Reform and Trade Shifts in overall irrigation water use. Cotton production Significant government investment supports drops by 26 percent overall and the fraction exported wheat and sugarcane in Pakistan. For example, drops from 40 percent to 25 percent; water use the Punjab government spent PRe 35 billion on for cotton reduces by nearly 7 billion cubic meters. wheat procurement in 2017. Two oft-discussed Declining international prices for rice and rising mechanisms to rationalize water use are reform domestic demand cause a near complete exit 125 from the international market, but little change in Reducing subsidies on sugarcane and wheat reduces production. The reduction in water use by cotton GDP per capita by 0.9 percent (column 3, table 6.6). allows increases for other crops, especially fruit and Reducing wheat support decreases economic growth vegetables, reflecting changing domestic demand. in all economic sectors by 0.6 percent to 0.8 percent, Water reallocation in the model is constrained by the but taxing sugarcane has a disproportionate cost to baseline parameterization, which limits expansion of agricultural processing given sugar’s strong reliance on fruit and vegetables. processing. Wheat reform, with or without sugarcane reform, encourages production in other agricultural Combining policy reforms and trade shifts (column 6, products, and there is a net gain in the agricultural table 6.5) reduces irrigation water use by 1.4 billion trade position: exports rise and imports fall. The price cubic meters. Most of this reduction comes from lower increases caused by added taxes reduces consumers’ textile exports, which is not necessarily desirable. ability to purchase other goods and services, and thus Adding changes in dietary preferences to these reforms has a small negative economic impact relative to the (column 7, table 6.5) sees significant reductions in RUMI-Hi base case. This impact is, however, very small water use for wheat and sugarcane and increases for compared to the more than fourfold increase in GDP by other crops (including fruit and vegetables); overall, 2047 under RUMI compared to BAU and is more than irrigation water use falls by 3.2 billion cubic meters. offset by other benefits. If, in addition, environmental flows are included (column 8, table 6.5), irrigation water use declines by Larger impacts occur in the scenarios with trade a further 3.4 billion cubic meters. However, there is a shifts, overwhelmingly because of reduced textile clear win-win outcome: nutrition improves, and water exports. GDP drops by 2.3 percent in 2047 relative becomes available for other higher-value uses. These to the RUMI-Hi base case (column 5, table 6.6); results are for RUMI-Hi variants, but if climate warms however, agriculture expands because the release less quickly, lower irrigation demands for a given level of water from cotton supports additional higher- of production would make it easier to shift water to value cropping. Because textile production mainly other sectors. appears as agricultural processing, it is here that the main reductions are seen, as well as in lower The model retains water in agriculture because exports because of lower prices. Rising agricultural nonagricultural requirements are met first, and no GDP and water released from cotton allows value is placed on flows below Kotri Barrage. The industrial growth, which expands by 2.0 percent. various reforms (and dietary change) free sufficient Industry draws resources and demand from the water to meet about half the projected increase in services sector, which is impacted by reduced nonagricultural water demand. Fully meeting these exports of textiles and rice. A similar response nonagricultural demands should not be difficult but will occurs by adding wheat and sugarcane reforms to require appropriate regulatory measures and adequate international price changes. investment. Table 6.5  Modelled Water Use by Major Crops in Pakistan under RUMI-Hi and Changes in Water Use for RUMI-Hi Variants, 2047 cubic meters (billions) Wheat Rice Cotton Sugar-cane Other crops Total RUMI-Hi 23.3 19.3 28.3 17.9 17.2 106 No wheat support −1.4 0.2 0.7 0.2 0.5 0.2 No sugar support 0.4 0.4 0.5 −1.8 0.4 −0.2 No wheat, sugar support −1.1 0.6 1.3 −1.7 0.9 0 LEPs for rice, cotton 1.3 0.1 −6.9 0.8 3.3 −1.5 No wheat, sugar support + LEP 0.2 0.6 −5.8 −0.9 4.5 −1.4 No wheat, sugar support + LEP, diet −4.6 1.1 0.5 −5.5 5.3 −3.2 No wheat, sugar support + LEP, diet, environmental flows −4.7 0.5 0.3 −5.6 2.9 −6.6 Source: CGE-W simulations. Note: LEP = lower export price; RUMI-Hi = reaching upper-middle-income (accelerated economic growth with faster rate of climate warming). 126 PAKISTAN: GETTING MORE FROM WATER Table 6.6  Modelled GDP under RUMI-Hi and Impacts of Policy Reforms, Changing Export Prices, and Consumer Preferences in Pakistan, 2047 RUMI-Hi No No No wheat, LEPs No wheat, No wheat, No wheat, (US$, wheat sugar sugar for rice, sugar sugar sugar support billions) support support support cotton support + support + + LEP, Diet, (% (% (% (% LEP (% LEP, Diet (% E-flows (% change) change) change) change) change) change) change) Agriculture sector 382 −0.6 −0.2 −0.8 1.7 1.1 −14.3 −17.0 Agriculture processing 177 −0.8 −0.6 −1.4 −12.8 −13.9 −18.0 −16.8 Industry sector 215 −0.8 −0.2 −1.0 2.0 0.8 7.4 5.2 Service sector 1,083 −0.6 −0.2 −0.8 −2.8 −3.6 −0.9 −0.1 Total 1,857 −0.6 −0.3 −0.9 −2.3 −3.1 −4.3 −4.6 Agriculture exports 143 0.6 −0.6 0.1 −65.8 −65.0 −67.8 −63.7 Agriculture imports 113 −0.8 0.4 −0.5 −16.6 −17.6 −37.7 −37.9 Source: CGE-W simulations. Note: GDP = gross domestic product; LEP = lower export price; RUMI-Hi = reaching upper-middle-income (accelerated economic growth with faster rate of climate warming. Changes in consumer preferences shift demand away heavily water dependent and establishing new from agricultural commodities (column 7, table 6.6) so industries in Pakistan can be slow and costly. If agricultural GDP declines by over 14 percent, impacting industry and service sector water demands are not agricultural processing and exports. The changes met, this could conceivably reduce productivity by favor the industrial and services sectors. Enforcing 5 percent. By 2047 this represents an annual GDP loss environmental flows (column 8, table 6.6) has only of US$45 billion, equivalent to 2.4 percent of total minor additional impact on total GDP because further GDP, or 70 percent of the GDP from the four major decline in agriculture is largely offset by lower trade crops. This scale of potential loss is easily sufficient to deficits and growth in services. justify major investments in urban water supply and restricting increases in agricultural water use. Policies The economic losses from the combination of policy and investments that fail to restrict growth in irrigation reforms, trade, and dietary shifts are very small relative water use will ultimately impose large economic to the rapid economic growth under the RUM-Hi base (and social) costs on Pakistan, far outweighing the case. This combination of changes and interventions benefits in agriculture. Chapter 3 notes that the costs of would free water from agriculture to support greatly inadequate water supply and sanitation are 3.9 percent improved urban water security and environmental of GDP; these are in addition to the costs of reduced sustainability. The resulting social and environmental industrial and services productivity assessed here. benefits would far outweigh the minor reduction in Without intervention, inadequate water for industry and GDP per capita from US$6,000 to US$5,700. services and inadequate domestic water and sanitation Reforming wheat and sugarcane policies would services could cost Pakistan more than 6 percent of improve water security with minor potential economic GDP by 2047. impacts, which might be offset by trade improvements through reduced exports of low-value agricultural Improved Environmental commodities, but increased export of higher value Management products. Political economy issues suggest the necessary reforms will be challenging (see chapter 4). As noted in chapter 2, flows downstream of Kotri Barrage are critical for sustaining the ecosystems of the Policy makers should consider the economic costs Indus Delta, as well as for meeting much of Karachi’s of inadequate water services to the industrial and water supply. Although the 1991 Water Apportionment service sectors, and of course, the social costs of Accord recognizes the importance of environmental further decline in the quality of domestic supply and flows, none are specified and there is no agreement sanitation services. CGE-W does not offer insights among the provinces on appropriate environmental into social outcomes, but does suggest the economic flows. Although not scientifically robust, widely costs of inadequate water for industrial and service accepted or implemented, prior work has suggested an sector growth. Not all industries and services are environmental flow below Kotri Barrage of 5,000 cubic 127 Table 6.7  Annual Value of Lost Production with US$0.72 billion. A moderate increase in end-of-system Increased Annual Water Demand below Kotri demand, which includes the water supply level Barrage, Pakistan, under RUMI-Hi recommended by the Karachi Water and Sewerage Board (KWSB) and an environmental flow level slightly Demand (BCM) Annual costs higher than that suggested by Amir and Habib (2015), (US$, billions) yields more than a fivefold increase in the net present Environment Karachi Net present 2047 value (NPV) of costs or between a two- and threefold value value increase in 2047 value. A major increase in end- Current 11.5 4.4 0.61 0.72 of-­system demand further increases the economic Moderate 17.5 7.4 3.28 1.76 cost. The magnitude of these losses (in 2047 value) increase is, however, very small in relative terms, being only 0.2 ­percent of GDP. Major 23.6 10.3 7.57 4.19 increase The future economic benefits to Karachi alone would Source: CGE-W simulations. far outweigh these losses. Therefore, mechanisms to Note: BCM = billion cubic meters; RUMI-Hi = reaching upper-­ middle- achieve these end-of-system flow increases should income (accelerated economic growth with faster rate of climate be explored and implemented. As noted previously, warming). the costs of poor health from inadequate water supply and sanitation and losses in economic productivity could reach nearly 6 percent of GDP. Because Karachi meters per second (4.44 billion cubic meters annually represents perhaps 20 percent of national GDP and and 0.37 billion cubic meters per month). Even this 10 percent of national population, adequate water constant and minimal flow is rarely achieved under supply and services for Karachi could mitigate perhaps current operations (Amir and Habib 2015), especially 1 percent of the water-related GDP loss, which is around at a monthly level. During the 2000/01 drought, flows five times the economic cost of reduced agricultural below Karachi were insufficient to even meet Karachi’s production. There are also many other benefits to demands, let alone provide an environmental flow for the more than 1 million people dependent on delta the delta. resources, including the mangrove forests and productive Although Karachi’s current volumetric supply is fisheries, so the total benefits relative to the costs would adequate, Karachi’s demand will grow significantly, be far higher. increasing the pressure for increased supply. This would further reduce flow to the delta, especially in References dry years. Even in the absence of a fuller assessment Amir, P., and Z. Habib. 2015. Estimating the Impacts of environmental flow requirements, Amir and Habib of Climate Change on Sectoral Water Demand (2015) argue that environmental water demands will in Pakistan. Nottinghamshire, U.K.: ACT. https:// increase considerably because of climate warming cyphynets.lums.edu.pk/images/Readings​ and the need to counter seawater intrusion. To _concluding.pdf. (accessed June 2018) explore the implications of an increase in the end-of- system water requirements (combining consumptive GoP (Government of Pakistan). 2016a. Pakistan and environmental needs), RUMI variants with Statistical Yearbook 2015. Statistics different increases in flows below Kotri Barrage were Division, Bureau of Statistics, Islamabad, modeled. These flows approximate the foregone Pakistan. http://www.pbs.gov.pk/content​ agricultural production that results from taking /pakistan-statistical-year-book-2015. securing higher monthly environmental flows to the ———. 2016b. Household Integrated Economic Survey delta (table 6.7). 2015–16. Islamabad: GoP, Statistics Division, The net present value of the annual costs to 2047 of Pakistan Bureau of Statistics. http://www​ meeting the current demands below Kotri Barrage .pbs.gov.pk/content/household-integrated​ is US$0.61 billion; the 2047 value of these costs is -economic-survey-hies-2015-16. C HAPT E R 7 Pathways to Water Security T his final chapter provides high-level this business as usual (BAU) baseline, urban water recommendations for improving water security security would be expected to decline, environmental in Pakistan. The recommendations emerge degradation would worsen, and groundwater from linking areas of weak sector performance depletion would worsen. The resilience of the water (water resources management, service delivery, sector would not be improved, leaving it more and risk mitigation) to aspects of sector architecture vulnerable to shocks. (policy, institutions, legal framework, infrastructure, and financing) that appear to be deficient. The A far higher rate of economic growth—sufficient recommendations are categorized for short- (less than to reach upper-middle-income status by 2047—is five years), medium- (five to 15 years) and long-term achievable and is not precluded by increasing water (more than 15 years) action. Several recommendations scarcity. This will require multiple reforms and are not new, and hence key political economy factors investments over coming decades. A significant share that appear to have prevented progress in the past are of the water currently used for irrigation will need to flagged. Without attention, these political economy be reallocated from agriculture to other sectors and factors are likely to continue to impede progress. Given users, including the environment. Water productivity the complex series of overlapping, intersecting, and in agriculture will need to be greatly enhanced. poorly quantified cause-effect relationships among Between the BAU and high-growth pathways, many the many water security variables, it is not possible alternatives may emerge. The actual growth trajectory to precisely quantify the improvements in economic, will depend on how aggressively the necessary policy social, and environmental outcomes that would accrue and institutional reforms are tackled, how rapidly from the recommended actions. the climate warms, and whether unexpected shocks (e.g., major floods, droughts, security incidents, or The assessment of what Pakistan gets from its water political unrest) occur during the transition period. makes it clear that Pakistan is not water secure. Modeling indicates that with continued incremental The full scope of water security considered in this improvement in agricultural water productivity, the report was not captured in the modeling. No modeling current rate of slow economic growth could probably was undertaken of new or modernized infrastructure, be maintained, and the food demands of a growing changed reservoir operations, conjunctive surface- population could continue to be met. However, under groundwater management, improvements to urban 130 PAKISTAN: GETTING MORE FROM WATER and rural water supply and sanitation services, Twelve high-level recommendations and their key improved flood management, basin sediment objectives are summarized in table 7.1. These are management, or many of the governance aspects then discussed in more detail, outlining for each discussed. A mix of these intervention would be the required improvements in water governance required to achieve the rates of economic growth (legal, policy, and institutional) and the nature and assumed in the modeling, especially in agriculture. scale of infrastructure investment. There are six In addition to the evidence from the modeling, the recommendations for improved water resources recommendations are supported by the assessments management, three for improved service delivery, and and descriptions of current sector performance, three for improved risk mitigation. The recommendations governance, and infrastructure. have been qualitatively assessed in terms of complexity, Table 7.1  High-Level Recommendations and Finances Required by Performance Area in Pakistan Recommendation Strategic objectives Cost Water Strengthen water data, • Improve water resources planning and system US$1–10 million per year resources information, mapping, operations management modeling, and forecasting • Improve flood/drought risk assessment, planning, and mitigation • Increase transparency of, and access to, water information Establish a multistakeholder • Guide long-term sustainable economic < US$1 million per year process of basin-scale water development resources planning • Define agreed upon basin-level environmental flows • Improve interprovincial sharing, especially during droughts • Build climate resilience across all sectors, including the environment Establish provincial water • Support a smooth economic structural US$1–10 million per year planning and intersectoral transformation water allocation mechanisms • Better manage temporary water shortages, including risk sharing • Improve efficiency and equity of irrigation water distribution Accelerate increases • Ensure future food security, given water US$1–10 million per year in agricultural water availability constraints productivity • Increase farmer incomes • Facilitate labor movement to other sectors • Contribute to overall increase in economic benefits from water Adopt conjunctive planning • Maximize the use of aquifer storage for < US$1 million per year and management of surface drought resilience and groundwater • Ensure sustainable groundwater use • Improve equity in water access across command areas • Reduce water logging and salinization Construct limited new • Better support multiple water management > US$1 billion per year storage (when hydroelectric objectives power justifies the expense) • Manage changing flood risk and changing and review reservoir demand patterns operations • Improve reliability of rabi irrigation supply • Manage increasing variability of inflows, including flood mitigation • Mitigate sedimentation to improve storage longevity • Contribute to improved energy security table continues next page 131 Table 7.1  continued Recommendation Strategic objectives Cost Service Modernize irrigation and • Improve irrigation service delivery in terms of US$10 million to US$100 delivery drainage and improve efficiency and equity million per year operations • Increase agricultural productivity, including more high-value crops • Ensure food security for a growing population • Enable reallocation of some water to cities and environment Reform urban water • Improve quality, equity, and sustainability of US$10 million to US$100 governance and close the urban water supply services million per year infrastructure gap • Reduce environmental and public health impacts of poor sanitation • Keep pace with population growth and urbanization Improve rural sanitation • Improve quality and coverage of rural sanitation US$1–10 million per year services • Reduce environmental and public health impacts of poor sanitation Risk Improve understanding and • Maintain natural green infrastructure for coastal < US$1 million per year mitigation management of climate risks protection to the Lower Indus and Indus • Protect coastal groundwater resources Delta • Protect and restore delta ecosystems • Build climate resilience of lower basin agriculture Strengthen planning and • Manage the trade-offs and synergies between US$1–10 million per year management of water- energy and water development and planning energy interactions • Inform investment choices Improve understanding and • Protect and improve water infrastructure US$1–10 million per year management of basin-scale operations sediment dynamics • Mitigate environmental impacts of changed sediment dynamics in the delta and lower river urgency, and scale of water security impact. Figure 7.1 The following section presents each recommendation maps the recommendations on complexity and urgency more fully. The recommended legal, policy, and axes; bubble sizes indicate the scale of impact. institutional inventions are summarized, and the required infrastructure investments are presented. Figure 7.1 suggests that the highest impact areas The time frames for these interventions are also to tackle are generally the more complex. Progress indicated as follows: (i) short-term, less than five years; on these challenges has therefore been limited, and (ii) medium-term, five to 15 years; and (iii) long-term, they are becoming increasing urgent for Pakistan. greater than 15 years. These areas include improving irrigation and drainage services and establishing a long-term strategic basin planning process. Improving the quality of urban Water Resources Management water supply services, especially in Karachi, is both Strengthen Water Data, Information, urgent and complex, but would greatly enhance Mapping, Modeling, and Forecasting Pakistan’s water security. Both irrigation services and urban water supply services are complex in part Water resources management in Pakistan is constrained because of the political economy challenges explored by inadequate water data, information, modeling, and in chapter 4. Progress will require strong political analysis. For a country worried about growing water leadership to establish better governance. Areas scarcity and the increasing risk of climate change’s impact that might initially be considered as “lower hanging on water resources, there should be a greater emphasis fruit”—in that they are urgent but less complex— on and investment in sound hydrometeorological include strengthening water data and information monitoring, data management, and analysis, and open systems and improving rural sanitation. sharing of water data and information. 132 PAKISTAN: GETTING MORE FROM WATER Figure 7.1  Complexity, Urgency, and Scale of Impact of Key Recommendations for Pakistan Water Services Delivery 2 3 1 Irrigation and drainage services 2 Urban governance and infrastructure More urgent 1 4 3 Rural sanitation 1 2 Water Resources Management 1 Data, information, and analysis 2 Participitory basin planning 3 Provincial planning and allocation 5 4 Agricultural water productivity 3 5 Conjunctive water management 6 New dams and revised operations Less urgent 6 1 3 Water-Related Risk Mitigation 1 Climate risks to the delta 2 Water-energy nexus 2 3 Basin sediment management Less complex More complex Note: Relative scale of impact is indicated by bubble sizes. Water resource assessments and water accounting Policy Reform need to be improved based on enhanced Short term. Establish an implementation framework for hydrometeorological monitoring (including for the National Water Policy (NWP), with clear roles and groundwater) and use of Earth observations. This responsibilities for water data and information. Develop will improve the understanding of natural and standards and guidelines for flood risk mapping and a induced water losses and the recycling of water policy framework for floodplain zoning. in the Indus Basin. Combining Earth observations with traditional monitoring in hydrologic and Institutional Reform agro-economic models can guide improved water resource planning and operations at basin, provincial, Short term. Strengthen the technical capacity in the and subprovincial scales. Better data, analysis, Water and Power Development Authority and modeling are required for improved flood risk (WAPDA) and Indus River System Authority (IRSA) for assessment, planning, and forecasting. Improved water data management, modeling, and forecasting, data and analysis will be critical for basin planning. including the use of Earth observations. Strengthen Water data need to be openly shared among all technical capacity in provinces, especially for monitoring stakeholders to build trust between water users and and reporting water distribution and use. Strengthen water managers. the Federal Flood Commission (FFC) capacity for flood risk mapping and flood forecasting. Build capacity in Legal Reform provincial governments for floodplain zoning. Medium term. Clarify the legal mandates at federal Infrastructure Investment level for water information collation and sharing. Strengthen provincial legal frameworks for land-use Short term. Expand national and provincial planning that considers flood risks. hydromet networks, including for cryosphere and 133 groundwater monitoring. Establish interoperable Establish Provincial Water national and provincial water information systems. Planning and Intersectoral Water Allocation Mechanisms Establish a Multistakeholder Process of Under the umbrella of basin-level planning, there is a Basin-Scale Water Resources Planning need for provincial water resources planning. Planning The highest priority for long-term sustainable should consider all water using sectors, integrate across water resources management is establishing surface water and groundwater, and address economic a multistakeholder process for strategic basin performance and environmental sustainability. Key to planning. Current water sharing arrangements implementing provincial plans will be processes that provide stability, but are not economically optimal, facilitate intersectoral reallocation of water to meet are insufficiently flexible to cope with expected changing sectoral demands, and mechanisms to future changes in water demands, and do not improve the efficiency and equity of irrigation water adequately embrace environmental sustainability. distribution. This will require strengthening of provincial Basin planning is important for improving legal frameworks for water resources management interprovincial water sharing, especially to clarify and establishing sound provincial water policies. Over risk sharing arrangements during drought years. the longer term, clear legal property rights should Climate change will increase drought severity, be established for water, separate from land. This and without intervention interprovincial conflicts would facilitate water trading that can help manage will escalate. Basin planning helps improve a scarce resource and mitigate the economic impacts environmental sustainability. The Indus Delta and of drought. Provincial irrigation departments will need the health of the lower river system are in rapid to be incrementally transformed into water resources decline. Basin planning should prescribe appropriate management agencies. These reforms are not simple and environmental flows to be delivered and monitored will need to surmount difficult political economy issues. jointly by federal and provincial government They need to be coupled with large-scale modernization agencies. National water planning is focused on of the irrigation system to greatly enhance hydraulic major infrastructure and is dominated by federal control and data-driven instrumentation of flows and agencies. A multistakeholder process needs to be allocations. Given vulnerability to water-related risks, and established to inform basin planning. This should the complexity of the water challenges, these reforms are involve all provinces and a range of nongovernment most urgent for Sindh. organizations (NGOs) that represent diverse water users and interest groups. Legal Reform Long term. Establish clear legal property rights Legal Reform (licenses) for water, separate from land, and legal requirements to maintain public register of water Medium term. Establish a sound legal mandate for licenses. federally led cooperative basin planning. Strengthen provincial legal frameworks for water resource planning. Policy Reform Short term. Develop and implement provincial water Policy Reform policies to establish sectoral priorities and to define allocation processes. Short term. Establish an implementation framework for the NWP that articulates roles, responsibilities, time Institutional Reform frames, and process for basin planning. Medium term. Incrementally transform provincial irrigation departments into water resources management agencies Institutional Reform with broad responsibilities including environmental Establish a national water council, as proposed in the management. Establish robust participatory processes to NWP, to provide strategic framing for cross-jurisdictional guide water allocation planning. basin planning. Strengthen the federal government capacity for river basin management (either within Accelerate Increases in Agricultural IRSA, WAPDA, or by establishing a new authority), that cooperate with provincial governments. Establish Water Productivity consultative processes for effective and broad Accelerating improvements in water productivity is stakeholder input. important for overall economic growth and for growth 134 PAKISTAN: GETTING MORE FROM WATER in the agricultural sector. It can support dietary shifts Policy Reform that improve nutrition. Increasing productivity will Short term. Develop conjunctive water management require reforming agricultural policies that distort farmer plans at the district level that focus on building drought incentives. It should be encouraged by incentives for resilience. wider adoption of water efficiency technologies and for diversification toward high-value crops. Investments Institutional Reform for value-chain addition of agricultural products and liberalization of the markets for agricultural commodities Short term. Strengthen the capacity of provincial water will contribute to productivity improvements. resources management departments for groundwater management and conjunctive planning. Strengthen Legal Reform water user associations (WUAs) for local-scale monitoring and management of groundwater resources Long term. Scope legal provisions to support pricing in line with agreed conjunctive water management and trading of water rights. plans. Build capacity of the Pakistan Council of Research in Water Resources (PCRWR) for basin-scale modeling Policy Reform and analysis of surface water–groundwater interactions. Short term. Phase out subsidies for wheat and sugarcane. Liberalize agricultural commodity markets. Construct Limited New Storage and Support adoption of water efficiency technologies and Review Reservoir Operations diversification to higher-value crops. A widely held view is that water management in Pakistan is greatly constrained by inadequate storage. Institutional Reform The evidence does not support this view. Given the Short term. Strengthen capacity for economic modeling current low productivity of water in Pakistan, it is within federal and provincial governments. Improve not possible to justify expensive major storages on-farm water management through farmer training on the economics of irrigation alone. Nonetheless, and awareness raising. Introduce lower water use sedimentation reduces existing live storage, and methods of rice cultivation. Increase investment in climate change will increase flow variability, making agricultural research. it more difficult to match supply and demand. Multipurpose reservoirs, which can be economically Adopt Conjunctive Planning justified by hydropower, should be constructed. These and Management of Surface will help to manage increasing flood risks and help to improve the reliability of rabi irrigation supply. Diamer Water and Groundwater Bhasha, upstream of Tarbela, will reduce the sediment Given the growing problems of groundwater depletion load to Tarbela, thus extending its life. in some areas and waterlogging and salinization in The operating procedures for Tarbela and Mangla others, there is a big opportunity to adopt active should be subject to periodic review. Changing demand conjunctive planning and management of surface patterns and flood regimes, the opportunity to better water and groundwater. While this needs to be manage sediment loads, and the increasing important coordinated at the provincial level, planning and of delivering a manage environmental flow regime implementation should happen at the district level. mean revised operating procedures are probably Conjunctive use can improve climate resilience by required to better balance across these multiple using the storage capacity of aquifers. It can improve objectives. A detailed modeling and optimization equity of water access, water use efficiency, and analysis should be undertaken to explore alternative water productivity. This will require new regulatory operating procedures under historical and potential frameworks to clarify the legal basis for groundwater future inflow regimes. access, and the legal responsibility and authority for groundwater regulation. It will also require Policy Reform capacity building within provincial government agencies and support to water user associations and Short term. Review and revise reservoir standard farmer organizations to facilitate implementation of operating procedures, based on detailed modeling and conjunctive plans. analysis. Legal Reform Institutional Reform Short term. Establish provincial-level regulatory Short term. Strengthen capacity at WAPDA and IRSA to frameworks for groundwater access and for enable periodic reviews of operating procedures and to management and regulation. support a multi-objective approach to operations. 135 Infrastructure Investment Policy Reform Medium term. Secure financing for construction of Short term. Replace warabandi with new water Diamer Bhasha Dam and associated power generation sharing rules based on economic efficiency and farmer and distribution infrastructure. equity. Reform abiana to reflect realistic operation and maintenance (O&M) costs. Water Supply Service Delivery Institutional Reform Modernize Irrigation and Drainage Medium term. Strengthen the capacity of new and Improve Operations provincial government water resources management departments to oversee PIDAs and performance of The quality of irrigation and drainage services WUA and farmer organizations. Strengthen WUAs for across Pakistan is generally very low. This keeps improved system operation and improved abiana farmer incomes low and retards improvement in collection. Reform WUA and farmer organization water productivity. The hydraulic efficiency of water governance to prevent elite capture. distribution is very low, and service delivery is not equitable across command areas. Irrigation services are Infrastructure Investments not financially sustainable and financial performance is declining. Service tariffs are set too low and are Medium term. Modernize irrigation system, including decoupled from service quality. The operational costs of new hydraulic control structures and lining of canals service providers are far too high. in waterlogged and saline areas. Automate control of hydraulic structures using real-time data acquisition Improving irrigation service delivery will require systems. Systematically improve drainage infrastructure. multiple interventions, including infrastructure investment, policy, and institutional reform and Reform Urban Water Governance strengthening. These should be underpinned by more comprehensive provincial legal and Close Infrastructure Gap frameworks that clarify the hierarchy of roles Urban water supply service delivery is not keeping and responsibilities in irrigation service delivery. up with the pace of urbanization. Poor quality, Irrigation networks should be modernized, including declining coverage, and inequity in urban water supply rehabilitation of canals and distributaries, and service delivery signal an urgent need to reform the installation of improved hydraulic control structures sector and to increase investment. Current policy with flow monitoring and automation. Water frameworks should be rationalized and simplified allocation processes within command areas need to to clarify institutional roles and responsibilities. This be updated to improve economic efficiency and to will help improve efficiency and accountability in increase transparency and equity. Farmers should service delivery. Public utilities need to be continually have clarity and certainty about the reliability of strengthened across many aspects of performance. water supply. Provincial government agencies Tariff structures need to be reformed and collection should gradually reduce the number of low- enforced. An enabling environment should be created skilled field-level support staff. These low-level to involve private sector operators in infrastructure and functions should initially be devolved to WUAs and service provision. Stronger coordination is required farmer organizations, but increasingly replaced by between public land owning and service delivery increasing automation of water delivery. agencies to link service delivery to urban planning. These reforms will require confronting difficult political Major infrastructure investment is required, especially economy issues relating to land ownership, water for wastewater treatment. This is critical to address access, wealth and political power. Given the sale and widespread environmental pollution and public health complexity of the surface water irrigation in the Indus, impacts from untreated effluent. Greatly improved O&M this modernization and associated reform are daunting of bulk water delivery systems and sewerage networks and will need to be progressed incrementally. Realistic is required. Improving urban service delivery will require implementation plans should be established in each confronting difficult political economy constraints, because province, considering capacity and finance constraints. some powerful individuals benefit from the status quo. Legal Reform Legal Reform Medium term. Revise provincial irrigation and drainage Medium term. Establish legal mandate for regulatory authority (PIDA) legislation to clarify roles and oversight of urban water supply service provider responsibilities in irrigation management between performance. Strengthen the regulatory framework for PIDAs and provincial government departments. pollution discharges. 136 PAKISTAN: GETTING MORE FROM WATER Policy Reform Infrastructure Investment Short term. Rationalize overlaps in the provincial policy Short term. Invest in public infrastructure for rural frameworks and align with the Local Government Act sanitation services, including wastewater collection and (2015). Develop and disseminate standards for urban basic treatment and disposal at the village level. water supply service delivery and link service tariff increases to service quality. Water-Related Risk Mitigation Institutional Reform Improve Understanding and Medium term. Strengthen and empower urban water Management of Climate Risks to supply service providers. Establish independent the Lower Indus and Delta regulators to oversee service provider performance There are several large, growing and unmitigated risks and to help reduce political interference. Establish an to the sustainability of the Lower Indus and its delta. enabling environment for increasing private sector Sea level rise and more intense coastal storms will participation in the urban water supply sector. increasingly threaten coastal Sindh and Balochistan. Declining end-of-system flows in the Indus and Infrastructure Investment groundwater salinization are additional pressures. Medium term. Greatly increase the capacity and A better understanding of the multiple threats to performance of wastewater treatment. Improve O&M the sustainability and productivity of the delta and of existing major distribution infrastructure. Increase the lower basin, especially those associated with climate coverage and reliability of urban water meters. change, is urgently required. This can guide long- term cross-sectoral planning and mitigation and rehabilitation efforts. The technical and economic Improve Rural Sanitation feasibility of barrier wells to slow saltwater intrusion Rural sanitation services in Pakistan are inadequate. should be investigated. Poor rural sanitation contaminates water supplies with widespread public health and quality of life consequences, Policy Reform especially for women and children. Poor sanitation Medium term. Develop long-term plans for sustainable contributes to poor childhood development, cognition, management of the Indus Delta. education, and ultimately labor force productivity. Rural sanitation suffers from a huge public infrastructure gap, Institutional Reform inadequate financing, and an absence of reliable revenue streams to cover O&M costs. Improving rural sanitation Medium term. Strengthen the technical capacity will also require increased public awareness and behavior of water and environmental management change in rural communities. agencies in Sindh for climate change impact assessments and mitigation planning. Resource relevant agencies for effective implementation of Legal Reform management plans. Medium term. Establish clear legal mandate for the provision of rural sanitation services. Infrastructure Investment Medium term. Assess the feasibility of barrier Policy Reform groundwater wells to slow seawater intrusion. Short term. Establish provincial standards and targets for rural sanitation services. Strengthen Planning and Management of Water-Energy Interactions Institutional Reform The water-energy nexus presents important risks Medium term. Strengthen the capacity and increase to both sectors. This nexus is not well addressed in the financing of provincial government departments policy and planning. Many energy sector policies and responsible for rural sanitation. Establish appropriate investments have had, and will continue to have, district-level institutional arrangements to engage with impacts on the water sector. Careful consideration communities in infrastructure improvement. Establish of trade-offs and synergies is required through appropriate mechanisms to ensure sustainable revenue broader economic analyses and cooperation between base for O&M costs. Monitor and report progress water and energy policy agencies at federal and toward rural sanitation targets. provincial levels. 137 Legal Reform deposition are not well monitored or studied, but have been greatly modified by water resources Short term. Establish provincial-level regulatory development, affecting the safety and performance frameworks for groundwater access and management. of water infrastructure, and contributing to the decline of the lower river and delta. Better Policy Reform monitoring is required to guide the development of Short term. Analyze the synergies and antagonisms intervention strategies and the operation of water between current national energy and water policy infrastructure. Additional investment in sediment frameworks to inform policy implementation. control measures in the Upper Indus Basin is recommended. Institutional Reform Policy Reform Short term. Increase coordination between government departments at federal and provincial levels. Medium term. Develop a management plan Strengthen capacity for joint energy-water analysis that to guide long-term, basin-scale sediment considers economic and environmental outcomes. management. Infrastructure Investment Institutional Reform Medium term. Expand solar and wind power Short term. Strengthen capacity in relevant technical investment where sensible. Explore feasibility for institutions for multiple aspects of sediment small-scale hydro on irrigation canals. Continue major monitoring, modeling, and analysis. hydroelectric power investment with run-of-river focus. Infrastructure Investment Improve Understanding and Management Short term. Ensure that new reservoir designs and of Basin-Scale Sediment Dynamics barrage rehabilitation projects consider sediment- Basin-scale sediment management requires much related risks to structural safety and operational more attention. Sediment sourcing, transport, and performance. A P P END IX A Pakistan Water Balance Data Sources Table A.1  Indus Basin Water Balance Component Value (BCM) Uncertainty level Data source, reference, remarks River water balance   Inflows   Indus, Jhelum, Chenab 170 Low (±5%) Mean rim station inflows, 1922–2016 (WAPDA). Indus inflows at Kalabagh (including Kabul); Jhelum at Mangla; Chenab at Marala.    Ravi, Sutlej, Beas inflows 3 Moderate While no water is allocated to Pakistan under the Indus (±10–20%) Waters Treaty from these tributaries, this value is the combined mean annual average gauged inflow since 2000 when Ranjit Sagar Dam was completed in India.    Internal inflows 32 High (±>20%) FAO (2011) value for total internal national resource (runoff plus recharge) less (i) the internal resources for the Makran and Kharan drainage units, and (ii) rainfall recharge estimate for the Indus (Pakistan) from van Steenbergen and Gohar (2005).    Total 205   Outflows   Withdrawal minus 103 Moderate Gauged canal withdrawals, 1977–2016 value (125) returns (±10–20%) less the return flow estimate from Karimi et al. (2013). The return flow has not been not adjusted for the saline drainage fraction.   River and flood recharge 4 High (±>20%) Estimate from Laghari, Vanham, and Rauch (2012). to groundwater table continues next page 140 PAKISTAN: GETTING MORE FROM WATER Table A.1  continued Component Value (BCM) Uncertainty level Data source, reference, remarks   Natural losses 68 High (±>20%) Water balance closure term. Includes flood waters (evapotranspiration) escaping to floodplains and evapotranspiration of wetlands, delta, riparian vegetation and open water evaporation.    Kotri outflow 30 Low (±5%) Gauged average at Kotri Barrage, 1975–2016.    Total 205 Groundwater balance   Inflows    Rainfall recharge 13 High (±>20%) Estimate from van Steenbergen and Gohar (2005).    Canal recharge 27 Moderate Estimate from Ahmad and Rashida (2001). Equivalent (±10–20%) to ~22% of withdrawals.   Irrigation recharge 17 High (±>20%) Estimate from Karimi et al. (2013). This value includes (includes saline) losses at the watercourse level.    River and flood recharge 4 High (±>20%) Estimate from Laghari, Vanham, and Rauch (2012).    Groundwater depletion 1 Moderate Water balance closure term. Consistent with (±10–20%) MacDonald et al. (2016).    Total 62   Outflows    Withdrawal 62 Moderate FAO (2011). (±10–20%)    Total 62 Withdrawal balance   Inflows   Surface water 98 Moderate Gauged canal withdrawals, 1977–2016 value, less (withdrawal minus (±10–20%) leakage estimate from Laghari, Vanham, and Rauch leakage) (2012).    Groundwater withdrawal 62 Moderate FAO (2011). (±10–20%)    Total 160   Outflows    Consumptive use 80 Moderate Modeled crop water use consumption from IFPRI (±10–20%) CGE-W. Includes some field-level evaporation associ- ated with crop growth.    Evaporative loss 41 High (±>20%) Water balance closure term.   Irrigation recharge 17 High (±>20%) Estimate from Karimi et al. (2013). This value includes (includes saline) losses at the watercourse level.    Return flow 22 High (±>20%) Flow estimate from Karimi et al. (2013). Return flow has not been not adjusted for the saline drainage fraction.    Total 160 Note: BCM = billion cubic meters; CGE = computable general equilibrium; IFPRI = International Food Policy Research Institute. 141 Table A.2  Makran Coast Water Balance Component Value (BCM) Uncertainty level Data source, reference, remarks River water balance   Inflows    Internal inflows 5.50 Moderate (±10–20%) Halcrow Group (2007).    Total 5.50   Outflows    Withdrawals 1.20 High (±>20%) Halcrow Group (2007).    Natural losses 2.00 High (±>20%) Water balance closure estimate.    Outflow 2.30 High (±>20%) Water balance closure estimate.    Total 5.50 Groundwater balance   Inflows    Rainfall recharge 0.74 Moderate (±10–20%) Halcrow Group (2007).    Total 0.74   Outflows    Withdrawal 0.69 Moderate (±10–20%) Halcrow Group (2007).    Environmental use, loss 0.05 Moderate (±10–20%) Water balance closure term.    Total 0.74 Withdrawal balance   Inflows   Surface water (withdrawal minus leakage) 1.20 Moderate (±10–20%) Halcrow Group (2007).    Groundwater withdrawal 0.69 Moderate (±10–20%) Halcrow Group (2007).    Total 1.89   Outflows    Consumptive use 1.20 High (±>20%) Water balance closure estimate.    Evaporative loss 0.69 High (±>20%) Water balance closure estimate.    Total 1.89 Table A.3  Kharan Desert Water Balance Component Value (BCM) Uncertainty level Data source, reference, remarks River water balance   Inflows    Internal inflows 2.90 Moderate (±10–20%) Halcrow Group (2007).    Total 2.90   Outflows    Withdrawals 0.50 High (± >20%) Halcrow Group (2007).    Natural losses 1.20 High (±>20%) Water balance closure estimate.    Outflow 1.20 High (±>20%) Water balance closure estimate.    Total 2.90 Groundwater balance   Inflows    Rainfall recharge 0.56 High (±>20%) Halcrow Group (2007). table continues next page 142 PAKISTAN: GETTING MORE FROM WATER Table A.3  continued Component Value (BCM) Uncertainty level Data source, reference, remarks   Irrigation recharge, groundwater depletion 0.24 High (±>20%) Water balance closure term.    Total 0.80   Outflows    Withdrawal 0.80 Moderate (±10–20%) Halcrow Group (2007).    Total 0.80 Withdrawal balance   Inflows   Surface water (withdrawal minus leakage) 0.50 Moderate (±10–20%) Halcrow Group (2007).    Groundwater withdrawal 0.80 Moderate (±10–20%) Halcrow Group (2007).    Total 1.30   Outflows    Consumptive use 0.68 High (±>20%) Water balance closure estimate.    Evaporative loss 0.50 High (±>20%) Water balance closure estimate.    Irrigation recharge 0.12 High (±>20%) Estimated, assuming similar percentage of withdrawals as for Indus.    Total 1.30 Note: BCM = billion cubic meters. References An Application for the Indus Basin.” Hydrology and Earth System Sciences 17: 2473–86. Ahmad, S., and M. Rashida. 2001. “Indus Basin Irrigation System Water Budget and Associated Laghari, A. N., D. Vanham, and W. Rauch. 2012. “The Problems.” Journal of Engineering and Applied Indus Basin in the Framework of Current and Sciences 20 (1): 69–75. Future Water Resources Management.” Hydrology FAO (Food and Agriculture Organization). 2011. and Earth System Sciences 16 (4): 1063–83. AQUASTAT: Pakistan (database). http://www.fao​ MacDonald, A. M., H. C. Bonsor, K. M. Ahmed, .org/nr/water/aquastat/countries_regions/PAK​ W. D. Burgess, M. Basharat, R. C. Calow, /PAK-CP_eng.pdf. A. Dixit, S. S. D. Foster, K. Gopal, D. J. Lapworth, Halcrow Group. 2007. Supporting Public Resource R. M. Lark, M. Moench, A. Mukherjee, M. S. Rao, Management in Balochistan. Basin-Wide Water M. Shamsudduha, L. Smith, R. G. Taylor, Resources Availability and Use. Irrigation and J. Tucker, F. van Steenbergen, and S. K. Yadav. Power Department, Government of Balochistan, 2016. “Groundwater Quality and Depletion in Royal Netherlands Government. Halcrow Pakistan the Indo-Gangetic Basin Mapped from in situ (Pvt) in association with Cameos. Observations.” Nature Geosciences 9: 762–68. Karimi, P., W. G. M. Bastiaanssen, D. Molden and van Steenbergen, F., and S. Gohar. 2005. “Ground Water M. J. M. Cheema. 2013. “Basin-Wide Water Development and Management.” Background Accounting Based on Remote Sensing Data: Paper 12, World Bank, Washington, DC. A P P END IX B Legal Framework for Water Resources Introduction which highlight provinces that have adopted similar or different approaches. This appendix summarizes the current content of the national and provincial laws and regulations Five components of water resources management, in Pakistan, which can be used to support the and thus regulatory objectives, are used for assessing management of water resources. It is not a detailed the legal framework and for aggregation (figure B.1): legislative assessment informed by inputs from (i) information, or understanding water resources, local experts; rather, it assesses the presence or uses, and risks; (ii) planning inclusively for rational absence of basic “legal elements” for water resources water management; (iii) allocation of water according management, and compares the Pakistan legal to agreed priorities; (iv) protection of water resources framework to that of other countries. This summary from overexploitation and pollution; and (e) adaptation may provide a foundation for the critical review of the to enhance system resilience. As indicated in figure B.1, legal framework called for in sec. 27 of the Pakistan these five objectives represent a maturing sequence National Water Policy (2018). of water resources management. The assessment considers the presence or absence of 48 specific legal A legal element is a provision in a law or regulation elements relevant to these five components. that contributes to one or more policy objectives. For example, a legal element could establish a mandate The assessment commences with a consideration of for an organization to monitor the quality and quantity national functions and arrangements. It then considers of water resources. A legal element for this purpose the legal framework that supports the five components may look different from a legal element for a similar of water resources management, noting that in a purpose in a different country or province given federal system, relevant legal elements may be differing contexts. This assessment identifies only either national or provincial, depending on how the whether specific elements are in place, not whether Constitution specifies areas of competence. In a federal they are being applied, or if so, how effectively. The system, certain legal elements are particularly important assessments are aggregated into proxy indicators for policy coordination and for the management of river of the comprehensiveness of the legal framework, basins that cross provincial boundaries. 144 PAKISTAN: GETTING MORE FROM WATER Figure B.1  Hierarchy of Regulatory Objectives for Water Resources Management in Pakistan Legal framework comprehensiveness Adaptation Protection Improving system flexibility and Allocation Protecting water resilience resources from Planning Allocating water depletion and according to pollution Information Planning priorities inclusively for Understanding rational water water resources, management uses, and risks Regulatory objectives National Level Table B.1  Links to National Legislation Relevant to Water Resources Management in Pakistan Chapter 4 provides an overview of relevant constitutional matters and key national legislation; Legislation Enacted Last details of the legislation can be accessed using the amended website URLs listed in table B.1. Penal Code, Act XLV 1860 2016 In federal systems, legal provisions relevant for water Easements Act 1882 1960 resources management are typically distributed Water and Power Development 1958 1998 across the provincial and national levels, with wide Authority Act scope for customization. However, national attention Constitution of the Islamic 1973 2015 is important in these federal systems, among others: Republic of Pakistan national policy and planning for water resources; flood Water Apportionment Accord 1991 1991 management; and management of transboundary water (international and interprovincial). Indus River System Authority Act 1992 1992 Environmental Protection Act 1997 1997 A legal basis for national policy and planning exists in Pakistan but is scattered and often implicit in broad Indus River System Authority 1998 1998 (Amendment) Ordinance or indirect provisions. The 2018 National Water Policy (NWP) in 2018 derives from amendments of 2017 to Statutory Notification 1033 (1)98 1998 1998 the 1973 Rules of Business of the executive authority. Water and Power Development 1998 1998 Under the Rules of Business—which allocate functions Authority (Amendment) within entities of the executive authority—the Ministry Ordinance of Water Resources has the broad remit of “matters Indus River System Authority 2000 2000 related to the development of water resources of (Amendment) Ordinance the country.” This reasonably encompasses national Indus River System Authority 2000 2000 water policy and planning but is not a direct legal (Chairman and Members mandate. Similarly, under the Constitution, the Council Conditions of Service) Rules of Common Interests has competence for policy IRSA Regulations 2000 2000 making on groundwater and hydropower development, Council of Research in Water 2007 2007 matters covered by the Water and Power Development Resources Act Authority Act. One function of the Pakistan Council of Research in Water Resources (PCRWR) is to provide the IRSA Regulation for Issuance of 2010 2010 NOC and Water Utilization Cess government with policy advice on the development, for Hydel Power Projects/Power management, conservation, and utilization of Projects Requiring Use of Water water resources. The federal institutional and policy arrangements therefore provide adequate scope for Note: IRSA = Indus River System Authority; NOC = No Objection Certificate. national water policy and planning, but no clear legal is long-standing agency practice that developed in mandates are established for these functions. response to devastating floods. Flood management The legal basis for flood management nationally is is the responsibility of the Federal Flood Commission not clearly anchored in constitutional provisions or (FFC), an agency housed in the federal Ministry primary or secondary legislation. The primary basis of Water Resources. Following a series of major 145 floods, the FFC was established in 1977 based on an limited legal support for broader aspects of water interprovincial agreement confirmed by resolution. resources management that go beyond the question of FFC’s institutional functions are summarized by the interprovincial water shares. Ministry of Water Resources as national flood protection planning, “scrutiny” of flood control and protection For international transboundary water management, schemes funded by the federal government, review there is a reasonably clear yet limited legal foundation. of flood damage, improving flood forecasting and Under the Constitution, international treaties such as the warning systems, researching floods, standardizing and 1960 Indus Waters Treaty fall under the competence recommending designs, and evaluating progress under of the federal government. The 2017 amendments to plans. Research on flood mitigation has a stronger legal the 1973 Rules of Business of the executive authority basis because it has been allocated as a function of list the Indus Waters Treaty and general liaison with the PCRWR. Pakistan has a reasonably comprehensive international engineering organizations in the water set of functions for flood protection and management, sector within the scope of activities of the Ministry but the legal framework provides only partial support of Water Resources. Thus, transboundary water to these functions and does not define clear legal management has a clear federal scope, but direct legal mandates for many of these functions. mandates for these functions are limited. The legal basis for interprovincial water management Legal Frameworks Applicable is stronger and clearer. The Constitution provides a mechanism for resolution of disputes on water to the Provinces allocation between provinces, and water allocation The legal framework relevant for each province between provinces is supported by the powers and includes the laws and regulations of that province, duties allocated to Indus River System Authority (IRSA) overlain by relevant national provisions (table B.2). under the Indus River System Authority Act (1992). This Of the 48 specific legal elements examined, only 27 gives IRSA broad powers to “lay down the basis for the are present across all provinces, and only 16 to 19 of regulation and distribution of surface waters amongst the 48 are found for any one province. Hence there the Provinces according to the allocations and policies is significant room for strengthening of the legal spelt out in the Water Accord.” However, there is very frameworks. Table B.2  Presence or Absence of Key Legal Elements in Pakistan’s Provincial Legal Frameworks for Five Areas of Water Resources Management Balochistan Khyber Punjab Sindh Pakhtunkhwa A. Information: understanding water resources, uses, and risks Water resource inventory N N+P N+P N+P Inventory, updating a a a P Inventory, public availability a a P P Water user registry P P P P Registry, public availability a a P P Monitoring planning a a a a Monitoring plan updating a a a a Monitoring P P P P Monitoring results, public availability P a a P Pollutant discharge information a a a a B. Planning: planning inclusively for rational water management Water resource planning N N N+P N+P Plan components P a a a Plan public consultation a a a a Plan updating a a a a Binding nature of plans a a a a table continues next page 146 PAKISTAN: GETTING MORE FROM WATER Table B.2  continued Balochistan Khyber Punjab Sindh Pakhtunkhwa Water resource quality criteria a a a a Water quality objectives a a a a Water user representation N+P N+P N+P N+P Women representation a a P a C. Allocation: allocating water according to priorities Priority orders a a a a Water abstraction permits, rights P P P P Abstraction permits, rights procedures a a a a Abstraction permits, rights duration a a a P Abstraction permits, rights renewal a a a a Predecision public notice P P P a Predecision public notice duration a P P a Predecision public notice means a a a a D. Protection: protecting water resources from depletion and pollution Special measures a a a P Reserve flows a a a a User recordkeeping a a a a Setbacks a P a a Discharge restrictions P P P P Nonpoint source pollution a a a a Discharge permit procedures a a a a Inspection mandate a P P a Inspection powers P P P P Offenses N+P N+P N+P N+P Penalties N+P N+P N+P N+P E. Adaptation: improving system flexibility and resilience Conservation P P a a Efficiency P P P a Obligation to pay resource charges a a a a Mandate to set resource charges a a a a Calculation of resource charges a a a a Mandate to collect resource charges a a a a Permits, rights transfers a a a P Transfer separate from land a a a P Transfer notification a a a P Transfer procedure a a a a Note: a = absent; N = national laws only; N+P = national and provincial laws; P = provincial laws only. Provinces’ legal frameworks are similar, but there for water resource inventories and water user registries. are differing strengths and weaknesses (figure B.2). Both legal frameworks also contain provisions to The legal frameworks of Sindh and Punjab are support public access to inventory and registry more comprehensive in provisions to support water information. In contrast, Balochistan does not require information systems, because both include elements the creation of a water resource inventory. 147 Figure B.2  Share of Completeness of Provincial Legal Frameworks in Pakistan a. Information— understanding water resources, uses, and risks 100 90 80 70 60 e. Adaptation— 50 b. Planning— 40 planning inclusively for improving system 30 rational water flexibility and resilience 20 management 10 0 d. Protection— protecting water c. Allocation— resources from depletion allocating water and pollution according to priorities Percent Balochistan Khyber Pakhtunkhwa Punjab Sindh Source: Author analysis. All four provinces have relatively sparse legal more comprehensive than elsewhere in South Asia frameworks for water resources planning. The (figure B.3). They are less comprehensive than the Khyber Pakhtunkhwa (KP) legal framework contains global average, however, particularly for water resources no supporting provisions. The legal framework of planning, water quality management, water allocation, Sindh has an anchor provision in the Sindh Water and support for water pricing and water transfers. Management Ordinance to support water resources Pakistan is one of the world’s most irrigation-dependent planning, but supporting legal elements for planning and water-stressed countries. Global comparisons are missing. Similarly, in Punjab, there is a basic (figure B.4) reveal that many of the most water- provision in the Punjab Canal and Drainage Act for stressed countries also have weak legal frameworks, water resource planning, but no specific legal elements suggesting that the inadequate legal frameworks have to support this. contributed to reaching this level of water stress, but also suggesting that dealing with these challenges will None of the provinces have adequate legal be more difficult with legal reform. frameworks to support implementation of a modern permit system for water use (even for high-volume From a national wealth perspective, Pakistan legal water users), and none have the full legal foundations frameworks for water resources management are not to support advanced features such as water resource exceptional. Many countries have less complete legal pricing or formal transfers of water between water frameworks, but many have more comprehensive users. For water resources protection, the KP legal frameworks, particularly in Central Asia and Sub- framework is relatively comprehensive compared Saharan Africa (figure B.5). to other provinces, thanks to provisions in KP Rivers Protection Ordinance, the KP Integrated Water Resources Management Board Ordinance, and the KP Legal Framework for Canal and Drainage Act. Sindh has special measures Groundwater Management in the case of water shortage in the Sindh Water Groundwater management has received relatively little Management Ordinance; however, detailed legal attention in the development of Pakistan’s federal and provisions to support water quality management are provincial legal frameworks. Historically, groundwater largely lacking across in all provinces. use was left largely uncontrolled and unregulated The legal frameworks for water resources under the common law principle of capture. Similar management in Pakistan’s provinces are generally provisions from the colonial era are within the 148 PAKISTAN: GETTING MORE FROM WATER Figure B.3  Comparison of Completeness of Legal Frameworks for Water Resources Management across Pakistani Provinces and South Asian Countries 100 80 Percent 60 40 20 0 dh ab a an l ) h ka pa ra hw es an nj ist Sin ht Ne ad Pu nk ch iL as gl tu ar Sr lo n kh Ba ah Ba Pa (M er a di yb In Kh Information Planning Allocation Protection Adaptation Source: Author analysis. Figure B.4  Completeness of Legal Frameworks for Water Resources Management in Pakistani Provinces and Comparator Countries, and Level of Water Stress Given as Withdrawals as Share of Total Renewable Resource 100 80 60 Percent 40 20 0 EGY JOR BAL KHY PUN SIN SDN TJK KOR ARM MAR IND ESP 0 20 40 60 Percent 80 100 120 140 Information Planning Allocation Protection Adaptation Source: World Bank, eba.worldbank.org. Note: Pakistan province abbreviations: BAL = Balochistan; KHY = Khyber Pakhtunkhwa; PUN = Punjab; SIN = Sindh. For ISO country codes see https://unstats.un.org/unsd/tradekb/knowledgebase/country-code. 149 Figure B.5  Comparison of Completeness of Legal Frameworks for Water Resources Management in Pakistani Provinces, Countries for which Similar Assessments Exist, and GNI per Capita 100 80 60 Percent 40 20 0 CHL THA BIH BFA LAO IND TZA BEN NLD ESP POL KAZ RUS MEX COL SRB GEO ARM LKA GTM PHL BOL LBR MWI NGA VNM NIC SDN ZMB GHA SIN BAL CIV KEN TJK KGZ KHM SEN HTI MLI NPL RWA UGA ETH BDI DNK ITA KOR GRC URY MYS TUR ROM PER JOR EGY MAR UKR PUN KHY CMR MMR BGD ZWE MOZ NER 0 20,000 US$ 40,000 60,000 Information Planning Allocation Protection Adaptation Source: World Bank, eba.worldbank.org. Note: GNI = gross national income. Pakistan province abbreviations (box highlight): BAL = Balochistan; KHY = Khyber Pakhtunkhwa; PUN = Punjab; SIN = Sindh. For ISO country codes see https://unstats.un.org/unsd/tradekb/knowledgebase/country-code. Figure B.6  Evolution of Provincial Legal Frameworks in Pakistan a. Balochistan Local Government Act Environmental Protection Act Canal and Drainage Ordinance Groundwater Ordinance 1973 1978 1983 1988 1993 1998 2003 2008 2013 2018 Irrigation and Drainage Authority Act Water and Sanitation Authority Act Water Users’ Association Ordinance Farmer Organization Regulations b. Khyber Pakhtunkhwa Environmental Protection Act Canal and Drainage Act IWRM Board Ordinance 1858 1878 1898 1918 1938 1958 1978 1998 2018 Irrigation and Drainage Authority Act Rivers Protection Ordinance Local Government Act figure continues next page 150 PAKISTAN: GETTING MORE FROM WATER Figure B.6  continued c. Punjab Farmers Organization Rules Soil Reclamation Act Irrigation and Drainage Authority Act Environmental Protection Act Canal and Drainage Act 1858 1878 1898 1918 1938 1958 1978 1998 2018 On-Farm Water Management And Water Users’ Association Ordinance Local Government Act Minor Canals Act Area Water Board Rules d. Sindh Environmental Samples Rules Local Government Act Water Management (Amendment) Act Irrigation Act 1858 1878 1898 1918 1938 1958 1978 1998 2018 Water Management Ordinance Environmental Protection Act Environmental Quality Standarda Rules Note: IWRM = Integrated Water Resources Management. Table B.3  Links to Provincial Legislation in Pakistan Balochistan Enacted Last amended Balochistan Irrigation and Drainage Authority Act 1997 1997 Balochistan Ground Water Rights Administration Ordinance 1978 2001 Balochistan Canal and Drainage Ordinance (not currently online) 1980 2006 Balochistan Water Users Association Ordinance (not currently online) 1981 1981 Balochistan Water and Sanitation Authority Act 1989 1989 Balochistan Community Irrigation Farmer Organization Regulations 2000 2000 Balochistan Local Government Act 2010 2010 Balochistan Environment Protection Act 2012 2012 Khyber Pakhtunkhwa Khyber Pakhtunkhwa Canal and Drainage Act 1873 2015 Khyber Pakhtunkhwa Irrigation and Drainage Authority Act 1997 1997 Khyber Pakhtunkhwa Integrated Water Resources Management Board Ordinance 2002 2002 Khyber Pakhtunkhwa Rivers Protection Ordinance 2002 2002 Khyber Pakhtunkhwa Local Government Act 2013 2013 Khyber Pakhtunkhwa Environmental Protection Act 2014 2014 table continues next page 151 Table B.3  continued Punjab Enacted Last amended Punjab Canal and Drainage Act, Act VIII 1873 2016 Punjab Minor Canals Act 1905 2003 Punjab Soil Reclamation Act 1952 1977 Punjab On-Farm Water Management and Water Users’ Association Ordinance 1981 1981 Punjab Environmental Protection Act 1997 2012 Punjab Irrigation and Drainage Authority Act 1997 2014 Punjab Local Government Act 2013 2017 Punjab Irrigation and Drainage Authority (Area Water Boards) Rules 2010 2010 Punjab Irrigation and Drainage Authority (Farmers Organizations) Rules 2010 2010 Sindh Sindh Irrigation Act 1879 2012 Sindh Water Management Ordinance 2002 2006 Sindh Irrigation and Drainage Authority Financial Regulations and Powers 2003 2003 Sindh Local Government Act 2013 2015 Sindh Environmental Protection Act 2014 2014 Sindh Environmental Quality Standards (Self-Monitoring and Reporting by 2014 2014 Industry) Rules Sindh Environmental Samples Rules 2014 2014 Easements Act (1882), which allow landowners to including sec. 26 of the Punjab Soil Reclamation Act withdraw unlimited groundwater from below their (1952, as amended), sec. 62A of the Punjab Canal property, as long as there is no malice or waste. Since and Drainage Act (1873, as amended in 2006). In then, there have been limited attempts to introduce Balochistan, the legal foundations for groundwater legal provisions to support the active management of management and regulation include sec. 12(e) and groundwater, which are uneven across the provinces. sec. 14 of the Balochistan Water and Sanitation Authority Act (1989) and sec. 3 and sec. 4 of the The Pakistan Water and Power Development Authority Balochistan Groundwater Rights Administration (WAPDA) Act (1958) provides a limited legal foundation Ordinance (1979). Neither Sindh nor KP have explicit for groundwater control and management across the legal provisions for groundwater; in these provinces country, but its chapeau includes a hedge against other groundwater is subject to relevant common law and in-force provisions, and any efforts by WAPDA would the Easements Act (1882). require the provincial agreement. This provision does not appear to have been used by WAPDA, leaving Pakistan needs to establish a clear legal mandate groundwater management largely to the provinces. for the groundwater management and regulation, However, few provinces have introduced provisions to whether at the provincial or national level. This would replace the Easements Act guidance on groundwater sensibly be supported by legal provisions that establish and provide a robust foundation sustainable a mandate to develop a stronger information base management and regulation of groundwater resources. on available groundwater and current extractions, to incorporate groundwater in water resources planning In Punjab, the provisions of the Easements Act (as Punjab has introduced), and to establish powers have been partially supplanted by provisions that to introduce controls for at least major extractions empower public authorities to manage groundwater, (as Balochistan has introduced). A P P END IX C Summary of WSTF Priority Actions No. Action, subaction Objectives Primary responsibility Timeline Indicative financing (US$, millions) 1 Major infrastructure and associated institutions 1.1 Rehabilitation of three major System sustainability Provincial irrigation departments 2012–16 400 barrages (PIDs) 1.2 Bhasha Dam in Jammu and Hydropower and irrigation Water and Power Development 2011–20 12,000 Kashmir Authority (WAPDA) 1.3 Kurram Tangi, Munda, Dasu, Flood control and WAPDA 2011–20 14,000 Kohala, Golen Gol, Bunji hydropower 1.4 Indus River System Authority Increase transparency and IRSA 2012–13 3 (IRSA) reforms predictability, and reduce conflict 1.5 Revenue-sharing framework Enhance equity and Ministry of Water and Power 2012–13 1 project acceptance (MOWP) 1.6 Resettlement framework and Enhance equity and WAPDA 2012–13 2 capacity project acceptance 1.7 Environmental flows in the Sustainability and equity IRSA, Sindh Province 2012–13 150 delta table continues next page 154 PAKISTAN: GETTING MORE FROM WATER No. Action, subaction Objectives Primary responsibility Timeline Indicative financing (US$, millions) 2 Raising agricultural productivity 2.1 On-farm water management Increase agricultural Provincial agriculture 2012–16 560 productivity departments, Jammu and Kashmir, Federally Administered Tribal Areas (FATA) 2.2 Public-private partnerships Increase agricultural PID and agriculture departments, 2012–16 460 (PPPs) for small dams productivity Jammu and Kashmir, FATA 2.3 Improved management of Increase agricultural PIDs 2012–16 500 main canals productivity 2.4 Spate irrigation Increase agricultural Provincial agricultural 2012–16 300 productivity departments, FATA 2.5 Optimal but judicious use of Sustainable productivity Provincial agricultural 2012–16 100 groundwater departments, FATA 3 Living better with floods 3.1 Construction on new dams Reducing flood peaks WAPDA 2012–2020 Included (see priority area 1) in priority area 1 3.2 Long-term institutional Capacity building FFC (Federal Flood Commission) 2012–16 20 development by partnership and the provinces with a successful organization (e.g., Mississippi River Commission) 3.3 Key elements of the National Pre-, during, and FFC, Pakistan Meteorological 2012–16 500 Flood Protection Plan IV, postflood management Department (PMD), federal including floodplain zoning and and provincial disaster enforcement; early warning management agencies, Jammu systems; community-based and Kashmir, FATA and provincial disaster risk management; governments flood protection infrastructure 3.4 Some federal and provincial Rehabilitation and Provinces, Jammu and Kashmir, 2012–16 500–600 actions, including asset maintenance of flood FATA management plans, and protection schemes rehabilitation and maintenance (including spurs and of existing infrastructure and bunds), estimated at new construction US$500 million by the FFC 3.5 Watershed management in Reduce severity of hill Federal government, Jammu 2012–16 50 Jammu and Kashmir, and floods and reduce erosion and Kashmir and KP Khyber Pakhtunkhwa (KP) 4 Sustainable urban services 4.1 Automatic tariff revision Improve financial Provincial governments and 2012 No cost sustainability WASAs action 4.2 Start reducing nonrevenue Improve service quality WASAs 2012–16 5 water (NRW) in 20 utilities and financial sustainability 4.3 Defining groundwater Secure resource base Provincial governments 2012–16 10 entitlements and regulating groundwater abstraction table continues next page 155 No. Action, subaction Objectives Primary responsibility Timeline Indicative financing (US$, millions) 4.4 Punjab Municipal Water Act Model for urban water Provincial governments 2012–16 4 reform 4.5 Save Quetta Ground Water Help secure the future of Government of Balochistan 2012–16 40 Quetta 4.6 Finance “wedge” to get to Sustainable services Provincial governments 2012–16 35 (for one sustainability large city) 4.7 Infrastructure for quality water Service quality Provincial governments 2012–16 250–700 services if they reform (per large city) 4.8 Pilot industrial pollution control Environmental health Provincial governments 2012–16 50 (per projects city) 5 Knowledge management 5.1 Partnership with an institution Consistent knowledge MOWP, FFC, IRSA, WAPDA, PIDs 2012–16 30 (e.g., eWater) to develop the base for operations at architecture and culture which different levels produces integrated, demand driven knowledge product 5.2 An operational simulation Management and WAPDA with PIDs 2012–16 20 model for the Indus Basin investment decisions 5.3 Knowledge base for Sustainability and MOWP, PIDs, FATA, Space and 2012–16 20 groundwater management productivity Upper Atmosphere Research Commission (SUPARCO) 5.4 Other decision support systems Operation of the 1991 PMD, IRSA, WAPDA, SUPARCO, 2012–16 30 for data sharing, canal, assets Indus Water Accord and PIDAs (provincial irrigation and management, and managing infrastructure, improved drainage authorities), and PIDs climate change water productivity 5.5 Capacity building for Developing capacity Higher Education Commission 2012–16 15 management and research (HEC), MOWP, Ministry of Science and Technology, standing committees of the National Assembly and Senate on water and energy, universities and research institutions Source: FoDP 2012. Reference FoDP (Friends of Democratic Pakistan). 2012. A Productive and Water-Secure Pakistan: Infrastructure, Institutions, Strategy. Islamabad, Pakistan: Water Sector Task Force, FoDP. http://metameta.nl/wp-content/ uploads/2013/11​ /FoDP-WSTF-Report-Final-09-29-12.pdf. A P P END IX D CGE Modeling Approach and Assumptions T his appendix summarizes the model and modeling that allocates available water to crops based on the approach used for the consideration of water impact of water stress on crop yields and crop values security futures. Reference to and results from (water allocation and stress model [WASM]), and previous water modeling efforts in the Indus Basin a hydropower module (not used in this study). The (Robinson and Gueneau 2014; Yu et al. 2013) are water models all use a monthly time step. In this study made, with differences in analytical approaches historical monthly precipitation and river inflows are explained. used as hydrologic input to the water modules. All the component models are coded in the General Algebraic Modeling Framework Model System (GAMS), which allows for integrated solution of the suite of models. The simulations for future water security (chapter 7) were run using the International Food Policy Research Institute’s (IFPRI’s) Computable General Equilibrium– CGE Model Water (CGE-W) model (Robinson and Gueneau 2014), The IFPRI CGE for Pakistan links consumers, producers, based on the most recent version of the IFPRI standard and government entities through production, CGE model (Lofgren, Harris, and Robinson 2002). consumption, trade, and taxes. A Social Accounting Several CGEs exist for Pakistan including GEMPAK,1 Matrix (SAM) connects the financial flows between PEP,2 and GTAP3; these are reviewed by Robinson and these actors for the base year (2013–14). Households Gueneau (2014). These CGEs have been used for trade receive income from wages paid by producers, analyses, gender evaluations, and climate change from owned assets and remittances abroad, and impacts on agriculture, among other topics. The CGE-W through government transfers. Households buy however, is the only model that interfaces with a goods from domestic producers and international detailed water model that includes water demand, imports. Producers sell to domestic entities and water routing, and water stress modules. exports. Households and producers pay taxes to the The CGE-W model consists of an annual economywide government, which purchases goods and services, CGE, a water demand module, a water basin and makes transfers (including subsidies) to actors management model (the Regional Water System in the economy. The model includes agricultural Model for Pakistan [RWSM]), a water allocation model information to represent the effects of water shocks 158 PAKISTAN: GETTING MORE FROM WATER on the economy—as well as disaggregated labor are assumed to be less mobile. The model considers and household categories—needed to capture the agricultural wage workers and nonagricultural distributional impacts of policy choices. unskilled and skilled workers. The CGE includes 64 activities: 17 in agriculture, 34 in The model uses 18 household groups. Farm household industry, and 13 in services; for a detailed description cohorts are defined by size and location, and nonfarm see Saeed (2017). The agricultural sector in the households are split into income ­ quartiles. Farm models includes 12 crops (rainfed wheat, irrigated households are represented as small, medium, wheat, basmati rice, irri rice, cotton, sugarcane, and or large holders, for Punjab province or all other other field crops and vegetables or horticulture) in provinces (six cohorts). Nonfarm households are three regions (Sindh Province; Punjab Province; and represented as rural landless agricultural households, the rest of Pakistan). Rainfed agriculture is included rural nonfarm households, and urban households, only for Punjab. Industrial activities include eight food with income quartiles for each category (12 cohorts). processing activities (e.g., meat, dairy, oils and fats, Household demand is estimated using a linear grain milling of wheat and rice, and sugar refining) expenditure system that relates expenditure on a ­ among others. Raw cotton production is transformed commodity to total household expenditure for a into cotton lint, yarn, cloth, knitwear, garments, and household group. Expenditures grow with income other textiles, all of which are sold domestically from wages, investments, transfers from the and as exports (these are the country’s primary government, and remittances. Full formal model exports). These production activities use various specification is provided in Lofgren et al. (2001). inputs, two major ones being land and labor, with The model uses elasticity values that define the land considered as an input only for agriculture. Land percentage change in consumption of a commodity is categorized into small, medium, and large rainfed resulting from an increase in household expenditure. parcels growing only wheat, and small, medium, and The elasticity values have a large impact on scenario large irrigated farms. Labor is also disaggregated: outcomes. The elasticity values used for RUMI-Hi family workers are separately identified for small, and RUMI-Hi-Diet are given in table D.1 for farm medium, and large farms, with the labor in the households and in table D.2 for nonfarm rural and smaller farms separated across regions because they urban households. Table D.1  Farm Household Expenditure Elasticities by Farm Size for RUMI Hi and RUMI Hi-Diet in Pakistan RUMI-Hi RUMI-Hi-Diet Commodity Small farms Medium farms Large farms Small farms Medium farms Large farms Food consumption elasticities Wheat flour 0.69 0.52 0.43 0.62 0.44 0.47 Rice 1.02 0.69 0.69 0.62 0.44 0.47 Refined sugar 1.02 0.86 0.69 0.62 0.44 0.47 Vegetable oil 1.02 1.04 0.69 1.04 0.44 0.47 Potato 1.11 1.04 0.92 0.62 0.65 0.71 Meat 1.02 0.69 0.92 1.04 0.87 0.95 Dairy 1.02 0.69 0.92 1.04 0.87 0.95 Food away 1.02 1.30 1.38 1.56 1.63 1.78 Selected manufactured goods elasticities Garments 1.02 1.04 1.38 1.14 0.98 0.83 Appliances 1.02 1.30 1.38 1.35 1.41 1.54 Selected services elasticities Transport 1.02 1.04 1.38 1.25 1.31 1.42 Dwellings 1.02 1.04 1.38 1.56 1.63 1.78 Education 1.02 1.30 1.38 1.56 1.63 1.78 Health 1.02 1.04 1.38 1.56 1.63 1.78 Source: IFPRI data. Note: RUMI = Reaching upper-middle-income; Diet = changed consumption. 159 Table D.2  Rural Nonfarm and Urban Expenditure Elasticities for Reaching Upper-Middle Income-Hi and Upper-Middle Income-Hi-Diet for Income Quartiles in Pakistan RUMI-Hi RUMI-Hi-Diet Commodity RNF (Q1) RNF (Q4) Urban (Q2) Urban (Q4) RNF (Q1) RNF (Q4) Urban (Q2) Urban (Q4) Food consumption elasticities Wheat flour 0.71 0.39 0.53 0.39 0.65 0.41 0.62 0.40 Rice 1.05 0.63 0.71 0.62 0.65 0.41 0.62 0.40 Refined sugar 1.05 0.63 0.88 0.62 0.65 0.41 0.62 0.40 Vegetable oil 1.05 0.63 1.06 0.62 1.09 0.41 1.03 0.40 Potato 1.14 0.84 1.06 0.83 0.65 0.62 0.62 0.60 Meat 1.05 0.84 0.71 0.83 1.09 0.82 1.03 0.79 Dairy 1.05 0.84 0.71 0.83 1.09 0.82 1.03 0.79 Food away 1.05 1.26 1.33 1.24 1.63 1.54 1.55 1.49 Selected manufactured goods elasticities Garments 1.05 1.26 1.06 1.24 1.20 0.72 0.93 0.69 Appliances 1.05 1.26 1.33 1.24 1.41 1.33 1.34 1.29 Selected services elasticities Transport 1.05 1.26 1.06 1.24 1.31 1.23 1.24 1.19 Dwellings 1.05 1.26 1.06 1.24 1.63 1.54 1.55 1.49 Education 1.05 1.26 1.33 1.24 1.63 1.54 1.55 1.49 Health 1.05 1.26 1.06 1.24 1.63 1.54 1.55 1.49 Source: IFPRI data. Note: Q = income quartile; RUMI = Reaching upper-middle-income; RNF = rural nonfarm; Diet = changed consumption. RWSM The RWSM assumes nonirrigation water (other than Karachi) is drawn solely from groundwater. The IBMR The RWSM closely follows the Indus Basin Model maximizes the sum of producer and consumer surplus Revised (IBMR), most recently summarized by Yang in agriculture by zone and does not have trade, et al. (2013) and Yu et al (2013). It models the nine government, or nonagricultural sectors. The CGE-W, main rivers of the Indus Basin that flow through therefore, is more appropriate given the interest in Pakistan and provide irrigation water (from east nonagricultural water security in groundwater pumping to west: Sutlej, Ravi, Chenab, Jhelum, Soan, Indus, is allowed only in nonsaline groundwater areas (each Swat, Kabul, and Haro) as well as the main Indus zone is disaggregated into fresh and saline areas), and Basin dams (Tarbela, Mangla, Chasma, and Chotiari). an annual cap of 62 billion cubic meters is imposed on Water is routed through 47 nodes of the Indus abstractions (as per Briscoe and Qamar [2005] and Yu system in Pakistan, including reservoirs, link canals, et al. [2013]). The model does not consider the water and barrages. Inflows, precipitations, runoff, and resources of the Makran Coast or the Kharan Desert crop water need data are generated externally by a hydrological units in Balochistan. climate model downscaled to Pakistan using historic data. Routing takes into account river routing time, reservoir evaporation, and link canal capacity. The CGE-W Balance model disaggregates the 45 main irrigation canals of The coupled water system model considers the water the Pakistan Indus basin into 12 agro-economic areas, resources of the Indus Basin of Pakistan—both surface based on provinces and crops grown. Four of these water and groundwater. The CGE-W uses the 2013/14 zones are in Sindh; five, in Punjab; two, in Khyber value of 182.3 billion cubic meters as the average Pakhtunkhwa (KP); and one, in Balochistan. Three other annual inflows at the Indus rim stations; this is higher zones cover the rest of Pakistan, in Punjab, Balochistan, than the current annual average inflow of 173.8 billion and KP, respectively. The water balance is broadly cubic meters reported in chapter 2, which reflects the similar to that of the IBMR, although groundwater is reduced inflows in the eastern rivers in recent years. more complete in the latest version of the IBMR. Although the CGE-W value is thus arguably too high, 160 PAKISTAN: GETTING MORE FROM WATER it is appropriate for future scenarios over the next three of the previous three years, which creates harvest decades, given the expected small increase in inflows expectations and a resulting allocation of land to with increased glacial melt. Notably, however, water different crops. The model is dynamic in that it steps withdrawals, delivery efficiencies, and crop water use through time after being solved for the base year largely drive model performance. (2013–14). Each following year is solved independently after bringing lagged values forward (such as exchange Average annual canal withdrawals in the CGE-W are rate or international prices) and adjustments to 128.9 billion cubic meters, which is the value cited important parameters, such as productivity levels in by SBP (2017) for the period 1975–2015; the value different sectors. This permits an evaluation of trends of 122 billion cubic meters cited in chapter 2 is based coming from different assumptions on exogenous on Food and Agriculture Organization (FAO) AQUASTAT factors. reporting based on data to 2008. For groundwater, the CGE-W uses the base year groundwater demand The shock due to water stress is defined as the ratio of of 39.5 billion cubic meters for agriculture and an crop yields for the current year compared to the base estimated 11.7 billion cubic meters for nonagricultural year yield. The base year data define the equilibrium demand from Habib and Amir (2015). The CGE-W of the water system in 2013–14 under an average also includes green water, which provides a supply weather pattern. In the first CGE run for each year, of 46.9 billion cubic meters from precipitation to the external water shock anticipated by farmers is agriculture. assumed to be the average of the four previous years, Water availability at the field level is considerably less so farmers anticipate a short-term moving average than the aggregate withdrawals because of losses level of water stress; this allows for some adaptation. from the canal and watercourse, especially seepage to The CGE then solves for irrigated and rainfed crop areas groundwater. In addition, some losses of groundwater based on these expectations. reflecting tube well inefficiencies are captured in the model. The losses assumed in the CGE-W water After the first CGE run, the Water Demand module balances are summarized in table D.3. calculates water demand for crops, industry, households, and livestock. Industrial water demand, for a given agroecological zone and month, varies Solving CGE-W Model in proportion to the square root of industrial gross The CGE-W is solved dynamically in a two-step domestic product (GDP), livestock demand varies procedure each year (figure D.1). First, the economic with the square root of livestock GDP, and household model is solved for a given year assuming exogenous demand varies with the square root of aggregate trends on various parameters, which provides projected household expenditures. These demands, therefore, outputs by sector and allocation of land to various increase more slowly than economic output, reflecting crops. Expected water stress is set to the average some efficiencies and economies of scale. These three Table D.3  Key Average Annual Water Balance Terms for the CGE-W Model in Pakistan billion cubic meters Water balance term Value Sources and comments Canal withdrawals (a) 128.9 Source: SBP (2016) Surface water losses to field level (b) 70.4 Losses in canals and watercourses Surface water at field level (a−b) 58.5 Groundwater pumped for agricultural use (c) 39.5 Modeled groundwater irrigation demand for base year Groundwater pumped for nonagricultural use (d) 11.7 Habib and Amir (2016) Groundwater irrigation tube well losses (e) 5.5 Estimated at 14% Field-level evaporation and drainage losses (f) 12.9 Estimated at 14% Groundwater pumped (c+d) 55.2 75 percent from canal/watercourse seepage Groundwater at field level (c−e) 34.0 Field-level total water availability (a−b+c−e) 92.5 Field-level crop water consumption (a−b+c−e−f) 79.6 Note: CGE-W = Computable General Equilibrium–Water. 161 Figure D.1  Sequence of Modeling in the Linked CGE-W Framework Ecomomic policy options and trends, with land variable CGE model previous (or base) year water stress Industrial and domestic water demand Water demand agricultural demand for water by crops Optimizes water distribution over months in the year RWSM calculates water shortages per water region by month Allocates supply of available water to crops Water stress calculates the impact of water stress on yields Yield shocks affect agricultural production; land fixed by crop CGE model CGE model solves for final equilibrium for current year Source: Robinson and Gueneau 2014. Note: CGE-W = Computable General Equilibrium–Water; RWSM = Regional Water System Model. demands are combined and are met from available CGE-W Assumptions groundwater prior to meeting irrigation demands. The water demand for irrigation, summed across The objective of the economic modeling in this study all crop types, is a function of the areal irrigation is to examine how basic drivers of change (population demand for a crop minus the soil moisture in a given growth, changing consumption preferences, climate month, multiplied by the area of the crop for a given change) affect water-related economic outcomes. agroecological zone. Nonagricultural demands are met The focus is on broad changes across water and food first. Water is then allocated across canals, regions, demand patterns and productivity growth. Commodity, and crops using the routing model in RWSM and in a household, and taxation options are not considered, manner to minimize water stress in WASM. because these would require more in-depth examination of specific policies. RWSM uses the computed water demands, along historical inflows and climate parameters, to partition Several important assumptions are made in each water among crops and regions each month, given simulation. On the demand side, preferences are set the objective function to maximize the value of outside the model as elasticities that vary by household production in a risk-averse manner. WASM then type, but do not vary between years. Household allocates water among crops in an area, given the expenditures within a simulation are determined as economic value of the crop. Because optimizing the a function of prices of goods and family income. On total value of production given fixed prices leads to the supply side, production is affected by the external excessive specialization in high-value crops, a measure choice of productivity growth for each commodity, of risk aversion for farmers is included in the objective as are growth rates for key inputs (such as land and function, which preserves a diversified production labor). Population is not considered in the model structure even in case of drought. The stress model directly, but it is captured as growth in the labor produces a measure of yield stress for every crop, force. Household expenditures are made on behalf of irrigated and rainfed, in each agroecological zone; nonworking family members, so they are captured in these are aggregated to the provincial level to match the model’s economic outcomes. With these choices, the regions in the CGE model. the model solves for supply, wages are paid so households receive incomes, and profits are earned by Finally, new yield shocks are calculated and applied businesses that invest and save. to the CGE, which is solved a second time for the final equilibrium, assuming allocation of land to crops Domestic prices are determined by interactions of is fixed since farmers cannot change their cropping supply and demand, and prices guide many of the decisions after planting. This solution yields all outcomes in expenditures, production, and water use. economic variables, including quantities and prices of International export and import prices remain fixed outputs and inputs, and all income flows. The process in these simulations, except under one scenario of then moves to the next year, updates parameters on policy and trade change. However, model outcomes trends, and starts calculations again. are sensitive to changes in prices, especially when a 162 PAKISTAN: GETTING MORE FROM WATER commodity has a high proportion that is imported or can shift between sectors, which is constrained by the exported. Because labor needs to be pulled from other many interactions across the economy. industries, or consumers need to make substantial changes, the structure of the economy is somewhat Comparisons with Prior resistant to change. However, when products have large trade positions, their reactions can be large and Economic Modeling driven by changes in international prices relative to Yu et al. (2013), Yang et al. (2013), Robinson and domestic ones. In the base year for these simulations, Gueneau (2014), and Davies et al. (2016a, 2016b) textiles and rice have large trade positions. use either a CGE or the IBMR to investigate issues Critical to these simulations is the growth in household, relating to water resources given climate change industrial, and livestock water demands, which the and population growth, including food security model meets from groundwater supply before meeting implications and economic growth. To address irrigation demands. Specifically, the water required by these topics, the economic benefits and costs of the industry (including livestock) is affected by the level following interventions were investigated: (i) improved of industrial and services output, so the faster the watercourse efficiency, (ii) water trading between economy grows, with greater industrial and services provinces, (iii) additional storage, and (iv) improved activities, the more water is required. Likewise, timing of water delivery. There are differences in domestic demand is driven by household expenditures, emphasis, however, because the IBMR does not include so, with greater GDP per capita, more water is used hydropower, and Yu et al. (2013) do not integrate the for domestic purposes. As temperatures have been CGE with the hydrological model. The latter integration rising historically, the baseline model includes a has been done by Robinson and Gueneau (2014) and rise of 1 degree Celsius over the simulation period has been used in Davies et al. (2016a, 2016b), as well (an increase of 5 percent in evaporation levels), which as in the current analysis. is the low end of the range estimated by Amir and The results from all simulations show that Habib (2015). Volumetrically, this increase primarily improvements in watercourse efficiency add affects crop water demand. significant benefits to GDP (around 2 percent relative The water resources situation is fixed across all to a base simulation, and higher in water scarce simulations, using a sequence of inflows based on years) (Davies et al. 2016a). For other interventions, the historical pattern. However, to avoid implying there is less agreement from the modeling. The value an end-of-scenario decline, and to explore recovery of new storage is much lower in Yu et al. (2013) from severe drought, the historical sequence was than in the CGE modeling because of the inclusion of rearranged to place the sequence containing the worst hydropower. In Davies et al. (2016a), climate change drought on record (years 2000–14) in the middle of scenarios reduce GDP by up to 1.25 percent, but the simulation period. The sequence of historical flows additional storage under climate change increases used as input to the 34-year simulations modeling is GDP up to 0.5 percent; this almost 2 percent thus: 1975–88 (14 years), 1999–2008 (10 years), and differential is similar to the benefits from enhancing 1989–98 (10 years). watercourse efficiency. Additional storage reduces the economic costs of shortfalls in water supply Thus, while dynamic, the model should not be during drought years. Hence in Davies et al. (2016a) considered as providing forecasts, but rather as a forecasted drought year sees a GDP drop of nearly providing quantitative comparisons between scenarios 4 percent, but only 2 percent with the Diamer Bhasha with different assumptions about key variables such Dam in place. as temperature, productivity performance, and trade. Economic recovery from shocks is quicker in the The IBMR (Yu et al. 2013) gives a much higher value model than in reality, because changes in asset values for water trading across provinces than the CGE-W, and their impact on consumption and production possibly because of differences in model structures. are not included. Similarly, if the government enters Yield effects are much stronger in Yu et al. (2013), an International Monetary Fund (IMF) program, the adding 3.66 percent to GDP, but less than 1 percent in model does not account for debt consequences and Davies et al. (2016b). In principle, from the perspective exogenous effects. The model prevents dramatic of an economic model, yield improvements should transformations in the structure of the economy. have a large effect because outputs are increased Although urban household incomes rise faster than without additional inputs; however, the CGE-W rural ones, and thus urban household expenditures simulations only alter yields for some crops and rise faster to become a larger fraction of the economy, locations, while the IBMR simulations assume more economic structure can change only as fast as labor widespread increases in yield. 163 Past studies demonstrate potential economic gains Agricultural Productivity.” In Agriculture and the from investment in the irrigation system, and the Rural Economy in Pakistan: Issues, Outlooks, current study has not sought to replicate this finding. and Policy Priorities, edited by D. J. Spielman, Rather, this new modeling puts a greater focus on S. J. Malik, P. A. Dorosh, and N. Ahmad, 117–70. exploring the value for productivity increases across Philadelphia, PA: University of Pennsylvania Press sectors (agricultural yields, industry, and services) that on behalf of the IFPRI. enable targeted levels of GDP per capita to be reached. Davies, S., W. Saeed, O. Majeed, S. Moeen, S. In addition, prior work has put less focus on the Shahzad, S. Robinson, and Z. Habib. 2016b. demand side of water management and responding “Economic Analysis of DBDP using an integrated to quantity and quality demands from all water users CGE-W Model.” In Report 13: Draft Economic is key to water security. Currently four major crops Analysis for the Environmental and Social Impact consume most of the water, and agricultural policies Assessment for Diamer Bhasha Dam Project, and other support ensure their dominance of water edited by P. Hartel, B. Trouille, J. Bello, and use. The focus of the modeling for this study is to D. Chinbat. Chicago, IL: MWH Americas. examine the full spectrum of water demand, including industry and services, and to recognize the important Lofgren, H., R. L. Harris, and S. Robinson. 2001. environmental demands for water that have been “A Standard Computable General Equilibrium largely ignored in prior modeling. (CGE) Model in GAMS.” Microcomputers in Policy Research 5: 69. Notes Robinson, S., and A. Gueneau. 2014. “Analysis with an 1. See the GEMPAK website, www.copsmodels.com​ Integrated Economic/Water Simulation Model /gempack.htm. of Pakistan.” PSSP Working Paper 14, IFPRI, Washington, DC. http://ebrary.ifpri.org/cdm/ref​ 2. See the PEP website, www.pep-net.org/pep-standard​ /collection/p15738coll2/id/128002. -cge-models. Saeed, W. S. 2017. A 2010–11 Social Accounting Matrix 3. See the GTAP website, www.gtap.agecon.purdue.edu/. for Pakistan. Unpublished Manuscript. Islamabad Pakistan: International Food Policy Research References Institute (IFPRI). Washington, D.C. Amir, P., and Z. Habib. 2015. “Estimating the Impacts SBP (State Bank of Pakistan). 2017. Annual Report of Climate Change on Sectoral Water Demand in 2016–2017 (State of the Economy). Islamabad, Pakistan.” Islamabad, Pakistan: Action on Climate Pakistan: SBP. Today. https://cyphynets.lums.edu.pk/images​ Yang, Y. C. E., C. Brown, W. Yu, and A. Savitsky. 2013. /Readings_concluding.pdf. (accessed June 2018). “An Introduction to the IBMR: A Hydro-Economic Briscoe, J., and U. Qamar. 2005. “Pakistan’s Water Model for Climate Change Impact Assessment in Economy: Running Dry.” Water P-Notes 17. World Pakistan’s Indus River Basin.” Water International Bank, Washington, DC. https://openknowledge​ 38 (5): 632–650. .worldbank.org/handle/10986/11746. Yu, W., Y. C. E. Yang, A. Savitsky, D. Alford, C. Brown, Davies, S., A. Gueneau, D. K. Mekonnen, C. Ringler, J. L. Wescoat, Jr., and D. Debowicz, and and S. Robinson. 2016a. “Irrigation and Water S. Robinson. 2013. The Indus Basin of Pakistan: Management in the Indus Basin: Infrastructure The Impacts of Climate Risks on Water and and Management Strategies to Improve Agriculture. Washington, DC: World Bank. W18052