CLIMATE RISK AND BUSINESS HYDROPOWER Kafue Gorge Lower Zambia Executive Summary Acknowledgements © 2011, International Finance Corporation Authored by Vladimir Stenek, International Finance Corporation Donna Boysen, Carla Buriks, William Bohn, Mark Evans, Tetra Tech, Inc. The authors wish to acknowledge the extensive technical support provided for the following sections: Appendix 1: Mariza Costa-Cabral, Hydrology Futures, LLC. Appendix 2: Murthy Bachu, Uttam Singh, Kishore Dhore; RMSI, Inc. Appendix 3: Richard Hayes, Tetra Tech, Inc. Appendix 4: Urban Ziegler, RETScreen The authors would like to thank the management and staff of ZESCO for their support and cooperation in this study, especially Romas Kamanga, Mellon Chinjila and Benny Sindowe. We also thank Israel Phiri and Clement Sasa from the Office for Promoting Private Power Investment (OPPI). The authors also wish to thank the following institutions, and colleagues working in these institutions, for their valuable contributions to the study: Department of Water Affairs of Zambia; Zambezi River Authority; Centre For Energy, Environment and Engineering Zambia (CEEEZ); Environmental Council of Zambia; Climate Change Facilitation Unit (Environment and Natural Resources Management Department, Zambia); Ministry of Agriculture of Zambia; Kafue District Health Department; Zambian Wildlife Authority; University of Zambia; World Bank and IFC (Zambia); International Union for Conservation of Nature (IUCN, Zambia); Imagen GSI Consulting; BRL Ingenierie / NIRAS; and UNDP (Zambia). The data and information collected, elaborated, and generously provided by Zambia Meteorological Department, was essential for the elaboration of this study. Reviewers We thank Vahid Alavian, Jane Ebinger, Ian Noble (World Bank); Mellon Chinjila (ZESCO); and Nico Saporiti (IFC) for their critical comments and suggestions. This work benefited from support provided by the Norwegian Trust Fund for Private Sector and Infrastructure (NTF-PSI). About Climate Risk and Business Starting in 2008, IFC initiated the Climate Risk and Adaptation Program, a series of pilot studies that analyzes climate risks and adaptation options for projects in various sectors and regions. The studies’ focus are private sector projects but with a significant emphasis on the cooperation and synergies with the public sector, research institutions and the civil society. To help understand and respond to the risks of climate change, IFC is developing best practices in assessing private sector risk and adaptation strategies. Published so far in the Climate Risk and Business series: Hydropower (Run of the River), Khimti 1, Nepal Agribusiness (Tropical Plantation and Refinery), GOPDC, Ghana Manufacturing, Packages, Pakistan Ports, Muelles el Bosque, Colombia Financial Institutions and Climate Risk For more information on the Program and to download the published studies, see www.ifc.org/climatechange. CLIMATE RISK AND BUSINESS HYDROPOWER Kafue Gorge Lower Zambia Executive Summary Climate Risk and Business: Hydropower, Kafue Gorge Lower 2 Climate Risk and Business: Hydropower, Kafue Gorge Lower Foreword Climate change is a reality, and the events and impacts associated with it are increasingly evident. Recent studies indicate that the levels of warming that are considered safety thresholds may be crossed as early as 2030 in several regions of the world, if not before is some cases. Climate warming is expected to have significant effect on precipitation patterns and events, including variations in the quantities of average yearly precipitation and distribution of rainfall. Those factors may in turn considerably affect the availability of freshwater and the livelihood of many who depend on this resource. Continuing to plan the future based only on historical records without accounting for potential shifts in climate patterns may slow down economic and business performance. It also could have serious adverse effects on society and the environment, particularly in those areas that are heavily dependent on water use. This study, part of IFC’s Climate Risk Program, provides an analysis of the implications of climate change, in terms of climate risks and adaptation options, related to the hydropower production of the planned Kafue Gorge Lower hydropower project in Zambia. As in previous Climate Risk studies, study authors recognize that projects and sectors should not be analyzed in isolation but rather in the context of a broader range of climate related economic, social and environmental factors to identify relevant and meaningful solutions. This is particularly true of water intensive sectors, where the importance of freshwater resource for development and in particular the nexus between water, food and energy; and projected global water deficits all mandate a more comprehensive perspective. This latest study looks to take into account not only changes in the supply - precipitation and runoff - caused by climate change, but also the likely shifts in demand due to climate change by various users. It also considers the likely impacts that directly or indirectly may affect resources for hydropower production, and how all these factors may influence the overall water availability and use. Through this work and as part of its larger Climate Risk Program, IFC shows its continued commitment to providing critical information and analytical tools that help identify options and support decision making in the face of climate challenges. I would personally like to thank Zambia Electricity Supply Corporation Limited, ZESCO and the Zambian institutions for their support and cooperation in this ground breaking work. Mohsen A. Khalil Global Head Climate Business Group International Finance Corporation 3 Climate Risk and Business: Hydropower, Kafue Gorge Lower Introduction Kafue Gorge Lower climate risk study is part of IFC’s Climate Indeed, one of the key findings of this study is that under some Risk and Adaptation Program, which explores the implications scenarios the frequency of annual flows projected to be lower of climate change impacts on private sector financial, social and than the average demand of different water will be on the environmental sustainability. rise. This possibility points to the need for the Integrated Water Resource Management approach that incorporates the effects A focus on long-lived infrastructure projects such as hydropower of climate change in the planning process. In addition to other is particularly appropriate. The design and projected useful life benefits, such an approach has the potential of identifying the of this type of large capital-intensive, physical assets typically most efficient adaptation solutions that would not necessarily spans many decades. And over this time considerable changes in be identified or undertaken if the components were analyzed in climate are expected to occur, with the potential of considerable isolation of the others. This is especially relevant for the project direct and indirect impacts on the life of these projects. assessed in this study: the findings about future climate variability, especially those related to drought, point to the possibility of Another factor to consider is that sectors and businesses that financial underperformance; this is due to lower flows even in are water intensive are particularly sensitive to the impacts earlier periods and occurs under some scenarios and assumptions of climate change because both the supply and the demand about required financial returns. side are affected by shifts in climate. Increased content of greenhouse gases in the atmosphere and consequent increased As always, it is necessary to interpret the results in the light of warming lead to higher evaporation, which in turn is likely to underlying assumptions and limitations; these are addressed in lead to prolonged droughts. With each degree of warming the depth throughout the study. While the best available information capacity of air to hold moisture significantly increases, resulting has been used, several relevant assumptions, such as demographic, in larger quantities of water vapor in the atmosphere and economic and policy decisions, including the global preference of precipitation events that often exceed historic values, which, as levels of greenhouse emissions, are dependent on current and witnessed during the past year in several regions of the world, future policy choices. Another uncertainty is the response of the often results in increased flooding. The exact effects of these systems to climate change and the effects of these responses on changes and their scale will depend on local and global factors, the study area. both climatic and non-climatic, and the possible magnitude of the impacts indicates the need to assess these risks in the project For example, how will the projected increase in evapotranspiration and location specific context. and droughts impact the miombo forest and other vegetation – this is important for controlling the water flows-, whether this As in previous climate risk studies, one of the focuses of this work will cause increased desiccation and number of wildfires, which in is understanding of the implications of climate related changes turn may affect soil structure, erosion levels, and frequency and not simply on the physical assets that are in direct control of intensity of floods. Although analyses of this type escaped the a project but also on a wider range of factors, including the scope of the present study, they would need to be incorporated supply chain, overall supply and demand, environmental and in further work around the Basin. social effects, and other characteristics relevant for a project’s operation. This perspective is relevant for many sectors, but One of several issues highlighted in the study is to identify what especially for water intensive businesses. Other things being data, tools and analysis is most appropriately collected and equal, rising temperatures result in increased consumption provided by public authorities or sectoral associations to enable of water for human use in both urban and rural areas, higher the more detailed site and project specific analysis needed by evapotranspiration leads to an increase of the use of water individual stakeholders. The approach in this study is probably in agribusiness and the need for higher releases of flows for too expensive and complex for the capacity of all but the largest biodiversity maintenance, industries will require more water for companies, pointing to the need for the development of freely their processes, etc. At the same time more water will be needed accessible platforms that would provide the information that for increased demand for electricity: for example, for cooling in would enable appropriate decision making process in the face of the industrial and urban context-, possibly at the same time there climate change. is diminished supply to meet the growing needs especially during increasingly frequent drought periods. 4 Climate Risk and Business: Hydropower, Kafue Gorge Lower Two objectives of this project are related to this issue and the or to perform new runs of the same model but with different use of its outputs. One was to develop and propose an initial assumptions – which will certainly be needed as new information approach to the assessment of climate risk and adaptation related to initial assumptions is produced. For this purpose, all analyses for hydropower and hydro-related projects that could results and numeric datasets used in the study are made available be replicated and improved by other users. For this purpose, all for download from IFC’s website, with the exception of the daily modeling software chosen in the study is freely available on the meteorological station data. internet, and the methodology, tools, and datasets are described in detail. Clearly this is not to be the final word on protecting the Kafue Basin from the ravages of climate change but through this The other objective was to provide the possibility to use the study’s analysis the authors do hope that the work presented here can outputs as a base for further work related to the Kafue basin, be an important step forward. This Executive Summary presents an abbreviated version of The major components of the study include: the full report of the approach and findings associated with • developing downscaled projections of temperature and the climate change risk assessment for the Kafue Gorge Lower precipitation for the study area across a base period and three (KGL) hydropower project. In addition to overall objectives of future consecutive time periods extending through the year 2100, the Climate Risk and Adaptation Program1, this study’s purposes • modeling hydrologic flow in the Kafue River based on these future include development of an approach for identifying and projections of temperature and precipitation, evaluating climate change impacts and potential effects on power • modeling the corresponding reservoir/energy outputs for KGL, production, financial flows, operational risks, and adaptation • analyzing the potential financial implications of the energy measures related to the hydropower projects, more specifically outputs for KGL, for KGL, and development datasets, models and tools that are • considering climate risks for natural hazards and other uses of publicly available and suitable for supporting climate change risk water in the study area (including agriculture, conservation, assessment, planning, and adaptation strategies. The availability urban, and industrial), and of this information in the public domain allows interested • identifying possible adaptation goals and strategies. stakeholders to apply this process to support further assessments as new information becomes available or local conditions evolve. The complete KGL Climate Change Risk Assessment Report and its appendices are available at the IFC’s website. 5 Climate Risk and Business: Hydropower, Kafue Gorge Lower PROJECT STUDY AREA Figure 1-1: Map of Kafue River Basin in Zambia The Kafue River Basin plays a central role in Zambia’s economy with most of the nation’s mining, industrial, and agricultural activities and approximately 50% of Zambia’s total population concentrated within the basin area (Mwelwa, 2004). A map of the Kafue River Basin is provided in Figure 1-1; the inset shows the location of the Basin within Zambia on the right and includes a representation of the major streams within the Basin in the left portion of the figure. The Kafue River is one of the major tributaries of the Zambezi River. The area of Kafue River Basin measures about 156,000 square kilometers (km2) and lies entirely within the borders of Zambia. The basin area occupies about 20% of Zambia’s total land area. The Kafue River Basin, which comprises the project study area, originates in the Copper Belt Province at an elevation of 1,456 meters (m) above sea level and terminates at an elevation of 366 m above sea level at its confluence with the Zambezi natural resource and planning considerations for hydropower are River. The total length of the Kafue River is about 1,500 km working to incorporate means to protect this conservation area. (Williams, 1977; Imagen Consulting Ltd, 2008). After originating at the Zambia-Congo divide, the Kafue River flows southwards or Along the Kafue River, there are currently three areas with south-westwards close to the Lukanga Swamps and then into the operating and/or planned hydropower projects: (1) the IT dam Itezhi-Tezhi (IT) reservoir. (completed in 1976) and hydropower power plant (planned in the short term); (2) the KGU dam and hydropower plant (completed The Kafue River turns eastwards after the IT reservoir and flows in 1972); and (3) the KGL dam and hydropower plant (planned). for about 350 km across the Kafue Flats and into the Kafue Gorge The planned KGL project, planned to be operated by Zambia Upper (KGU) reservoir (Figure 1-2). The Kafue Flats are a wide Electricity Supply Corporation Limited (ZESCO) , the country’s and flat area of the river, with natural water flow moving slowly primary domestic electricity provider, is the focus of this risk across the flats at a shallow depth. The Kafue Flats are a valuable assessment pilot project. Figure 1-2: Schematic of the Reservoir Network for the Kafue River Basin2,3 PHYSICAL PARAMETERS FOR DAMS, RESERVOIRS, AND POWER STATIONS Top of Generating Combined Dam dam elev. Length Storage (mcm) Capacity2 Efficiency Structural Features/Notes IT 1035m 1800m 6,008 @ 1030.5m 120 MW 88% 3 spillway gates (4,425 cms @1030.5m), low level outlet for power releases KGU 980m 300m 1178 @ 977m 990 MW 91% 4 spillway gates (3,660 cms @ 978m) 11 Km tailrace tunnel, 400 m head KGL 586m 300m 80 @ 580m 750 MW 88% 3 spillway gates (3,959 cms @ 582 m), 7 km tailrace tunnel, 200m head 1. See www.ifc.org/ifcext/climatebusiness.nsf/Content/AssessingClimateRisks 2. Turbine and generator 3. Physical parameters were developed for this analysis during 2008 using background information, research, input from ZESCO, and professional experience. Though the IT and KGL power generating units were not in place, they were modeled as operating units for this study. Expectations for the design of these units have varied over time; final construction may be different from these parameters. (Acronyms: mcm = million cubic meters; MW = Megawatt; and cms = cubic meters per second.) 6 Climate Risk and Business: Hydropower, Kafue Gorge Lower The KGL site lies about 65 km upstream of the confluence of the Climate Change Assessment Kafue River with the Zambezi River, and about 20 km downstream from the existing KGU hydropower plant. Power unit capacities Time Horizons and operational parameters are presented in greater detail in The project examines the potential effects of climate change the reservoir (energy) modeling section of the report (Section during the remaining 90 years of the century, and also reflects 2.2). For modeling purposes, all units (existing and planned) are estimated lifetimes of the KGL dam and hydropower time. For modeled as operating power stations in this study. GCM modeling, projections for approximately the next 50 years are typically more accurate than longer-term projections, because Flow into the KGU reservoir is regulated by the IT dam, which the next 50 years will reflect the impact of current emissions. creates the 6,008 million cubic meters (mcm) IT reservoir located The 90 years are considered in 30-year segments (early-, mid- on the Kafue River about 230 km upstream and west of the KGU and late-century) to facilitate consideration of investment issues hydropower project. Since the initiation of this climate change risk (early-century) and to allow for separate consideration of the late- assessment project for KGL, plans for a 120 megawatt (MW) turbine century period results, which can be expected to be less certain project at the IT dam have evolved rapidly. This new, planned than the earlier periods. The period of 1960 to 1999 is used as capacity has been incorporated into the project assessment. Further the base period. information is provided in the full report and Appendix A3. Time Horizons = 4 (base, early-, mid-, and late-century In addition to flows released from the IT dam, various local periods) intervening inflows exist between the IT reservoir and the KGU hydropower project site. Flows released from the IT reservoir Global Climate Models pass through the natural wetland area of the Kafue Flats. This The climate change projections for this 90-year period were obtained area contains substantial environmental and ecological assets. from six Global Climate Models (GCMs). Nearly two dozen GCMs Stakeholders including ZESCO, the World Wildlife Fund (WWF), are used in the international climate science community. Their and others are continuing to negotiate and plan for appropriate results are collected at the Coupled Model Intercomparison Project water release schemes from the IT reservoir to protect wildlife (CMIP) website for the benefit of researchers internationally. These and fauna in the Kafue Flats area (Schelle and Pittock, 2005). GCMs are used to project the impacts of greenhouse gas (GHG) emissions on the major components of our climate system: land surface, ocean, atmosphere, and sea ice. These models are both CLIMATE CHANGE ISSUES data- and computationally-intensive and continue to be improved as more detailed data and more powerful computing resources Zambia’s reliance on hydropower to meet current and future become available. Figure 1-5 illustrates how GCMs represent the electricity demand faces three types of challenges: interconnected elements of climate in a three-dimensional system of grid cells. In Zambia, there are 270 of these grid cells. • increased economic development leading to growing demand for water for other uses, • the potential for increased water needs to address conservation Figure 1-3: Conceptual Structure of GCM Grid Cells goals in light of the potential impact of climate change and climate variability on water supply and evaporation, and • increased power demands requiring additional water for hydropower. Conservation needs include non-consumptive, timed releases to support the Kafue Flats; other uses include water demands for irrigation, domestic use, mining, and industry. Many stakeholders within government and the public and private sectors appreciate the need to better understand the timing and extent of supply- demand tensions related to water, with meaningful financial and technical resources being dedicated to better evaluating the complexity of the challenges and identifying possible solutions. PROJECT APPROACH The project approach is presented schematically in Figure 1-3, which is referenced throughout this section while Figure 1-4 provides a simplified version of the approach, highlighting the Source: www.cru.uea.ac.uk/cru/info/modelcc/ five major stages of analysis and their key components. 7 8 Figure 1-4: Assessment Approach Figure 1-5    Orientation Diagram    CCA  ORA NHR&SA   AO   R                                   Climate Risk and Business: Hydropower, Kafue Gorge Lower                                                                              Climate Risk and Business: Hydropower, Kafue Gorge Lower Model limitations are becoming better understood and GCMs Operational Risk Assessment are being continually revised to improve performance. For example, recent analysis of GCM projections compared with Hydrologic Modeling observed precipitation data indicates that the models tend to Hydrodynamic models are used to represent the functioning underestimate climate change impacts on extreme precipitation of complex water systems, including detailed water flow events (Min, 2011). As such findings emerge, researchers work to patterns and sediment transport. A model of the watershed is refine and improve GCMs over time. constructed by separating the water cycle into component parts and constructing boundaries around the watershed of interest, as Research has shown that use of all 22 GCMs is not necessary illustrated in Figure 1-7. for most assessments of climate change (Pierce, et. al., 2009). As explained further in Appendix 1 of the full report, adding Of these 12 original GCM/SRES combinations, the results from many GCMs makes relatively little additional benefit for the three GCMs were selected for further analysis, as shown in Figure range of future climate projections, once the number of GCMs 1-3. Of the original six GCMs, these three GCMs yielded the reaches about five. Therefore, this approach uses downscaled upper, lower and median results for projected climate change. To climate data obtained for Zambia from six GCMs (see additional manage the computational demands of the analysis, the results of discussion and references in the full report). these three GCMs became the focus of the subsequent analyses. GCMs = 6 (GFDL, GISS, ECHAM5, IPSL MRI, UKMO)* The climate outputs (principally temperature and precipitation) from the six GCM/SRES combinations were used as inputs for the IPCC** Emission Scenarios hydrologic flow model. For this study, the model used is the U.S. Many emission scenarios have been developed by the Army Corps of Engineers (USACE) Hydrologic Engineering Center Intergovernmental Panel on Climate Change (IPCC), in the Special (HEC), Hydrologic Modeling System (HEC-HMS), a generalized Report on Emissions Scenarios (SRES) (Nakicenovic, et. al., 2000), modeling system capable of representing many different watersheds. with the goal of characterizing future possible “story lines” for HEC-HMS is designed to simulate the precipitation-run-off processes development globally. The SRES scenarios include projections of dendritic watershed systems. It is applicable across a wide range of future concentrations of carbon dioxide (a key GHG) in the of geographic areas for addressing a wide range of project goals. atmosphere based on a range of factors. The IPCC developed the SRES scenarios to provide a standard approach to characterize This analysis yields projected water flow volumes due to climate future variables whose true future states cannot be known, such change at locations along the Kafue River for each of the six as level of economic growth, extent of technological innovation, GCM/SRES combinations. Outputs were calculated for base the carbon intensity of energy sources, and political attitudes period (in the 20th century) and the three future time periods for toward climate change. They are not designed to reflect actual the 21st century. future outcomes, but rather are intended to capture a range of possible outcomes. These outcomes are shown in Figure 1-6. Water Use Scenarios and the Weather Generator Equivalent probabilities are assigned to these story lines. These levels of flow were then further modified to reflect potential changes in available flow that could result from other For this project, two scenarios were selected, A2 and B1, for their usage requirements for the water supply in the Kafue River. The ability to reflect a relatively carbon-intensive future (A2) and a potential future with relatively low carbon emissions (B1). These Figure 1-6: SRES CO2 Projections through 2100 scenarios were developed in 2000. In the dozen years since their development, actual global emissions have exceeded the A1F1 projection, the SRES scenario with the highest projected concentration of carbon dioxide in the atmosphere (USGCRP, 2009). SRES Scenarios = 2 (A2 and B1) The combination of six GCMs and two emission scenarios provides 12 different forecasts of temperatures and precipitation levels due to climate change for each of the time periods for Zambia (270 grid cells). This results in nearly 13,000 data points each for temperature and pressure. Further information about the modeling approach is provided in Section 6 of this summary report and Appendix A1 of the full report. These climate change results were then used as the basis for two sets of further analysis: the operational risk assessment, and the natural hazard risk assessment. Source: www.grida.no/publications/other/ipcc_sr/?src=/climate/ipcc/emission *Global Circulation Models elaborated and maintained by different research institutions **Intergovernmental Panel on Climate Change 9 Climate Risk and Business: Hydropower, Kafue Gorge Lower initial scenario represents a power maximization scenario (P-1); Figure 1-7: Illustration of the Hydrologic Cycle it includes assumptions of baseline water withdrawals along the river for irrigation and domestic, mining, and industrial uses (DMI). The alternative use scenarios are divided into two categories: development (the “I” scenarios), and conservation (the “C” scenarios). The “P”, “I,” and “C” scenarios are based on assumptions presented in the Strategic Environmental Impact Assessment (SEIA) of KGL (SWP, 2003). The eight scenarios are presented in Table 1-2 and are summarized below. • P-1 is the maximum power scenario (P-1), • C-1, C-2, and C-3 are the conservation scenarios, and • I-3, I-6, I-9, and I-10 are the development scenarios. The first seven of the eight water use scenarios correspond to the scenarios of the same names in the 2003 SEIA. In addition, this study added I-10 in order to characterize a larger potential level of development during the mid- and late-century periods. Each of these scenarios includes the projected effects of climate change on the available water supply in the Kafue River. Source: www.cfses.org/salmonid/html/water/cycle.htm The details of each of the scenarios presented in Table 1-2 are presented in the full report. In general, the C-1, C-2, and C-3 scenarios also include all of the same assumptions about the ClimDex has been adapted for use with R as the platform – an annual volume of available water in the river as does the P-1 environment that does not depend on a particular operating (maximum power) scenario, except they vary the timing and system. R is a free and yet very robust and powerful software for volumes of water discharged from the IT dam for conservation statistical analysis and graphics; it runs under both Windows and purposes. The I-3, I-6, I-9, and I-10 scenarios also include the Unix environments. same base assumptions as the P-1 scenario, but increase the estimated water abstractions for development, thereby, reducing The RClimDex tool was developed and is maintained at the the expected flow rates available for power generation. Climate Research Branch of Meteorological Service of Canada. Its initial development was funded by the Canadian International A final set of flow analysis was applied to the P-1 scenario through Development Agency through the Canada China Climate Change the use of a weather generator model (WXGEN). The flow Cooperation (C5) Project. scenarios selected include ECHAM5, A2, late-century (highest increase in flow from the A2 base period) and the ECHAM5, B1, Reservoir/Energy Modeling mid-century results (highest decrease in flow from the B1 base The flow results were then used as inputs for the reservoir/energy period). A weather generator uses a Monte Carlo approach to model which was used to estimate the level of power generation produce a synthetic time series of weather data of any desired that would be produced by the available flow, as shown in duration for a location based on the statistical characteristics Figure1-3. of observed weather at that location. While the GCMs each produce a projection of the future that typically extends about Reservoir/energy models support water resource studies by a century, a weather generator provides results over a much predicting the behavior of reservoirs under different scenarios. longer period, thus better capturing climate extremes, such as They are designed to model operations at one or more reservoirs prolonged periods of high and low rainfall, which is useful in whose operations are defined by a variety of operational goals climate risk assessment. and constraints. These tools help reservoir operators plan releases in real-time during day-to-day and emergency operations. This Additional Statistical Analysis: RClimDex study used the USACE HEC Reservoir Simulation (ResSim) ClimDex was developed as a Microsoft Excel program to assist model. It is designed to reflect reservoir operations at one or researchers in the analysis of climate change and detection. It more reservoirs whose operations are defined by a variety of uses a a four-step analysis process that consists of quality control, operational goals and constraints. It uses an original rule-based homogeneity testing, calculation of desired indices, and regional description of the operational goals and constraints that reservoir analysis. operators must consider when making release decisions. 10 Climate Risk and Business: Hydropower, Kafue Gorge Lower Financial Modeling Recommendations Annual electricity generation results for KGL for two representative Upon completion of the assessment steps described in Section scenarios were then used as inputs into the financial model to 1.4, conclusions and recommendations were presented, and characterize the financial results of KGL. In order to determine are discussed below, and at greater length in the full report. the most desirable investment targets, investors evaluate a range The major themes of the findings and recommendations are as of financial information. Investors will have target performance follows. goals that include the size of the financial return, payback period for the investment, and the risk associated with the project. The science of climate change risk assessment includes Projects are often evaluated using proprietary financial models. uncertainty but is continually improving; the projected impacts For this study, a publicly available financial model was used. of climate change are better understood for temperature than for Natural Resources Canada developed the Renewable Energy precipitation. (Precipitation projections are complicated by many Technology Screening Tool (RETScreen) with project partners factors, some of which are global in nature, such as large-scale to provide a free, public resource for assessing clean energy weather patterns and feedback mechanisms, others of which are projects. It captures key financial and operational inputs for a much more local, such as topography). project and generates resulting financial performance outputs for project evaluation. Despite the uncertainty surrounding projections of climate change and climate change variability, many areas are experiencing Natural Hazard Risk & Sector Assessments changes in temperature and precipitation that exceed historical patterns. Therefore, it is appropriate to plan for change, especially The temperature and precipitation results from the 12 GCM/SRES where such planning can yield other co-benefits, regardless of climate change scenarios also informed an assessment of the the extent of climate change. impact of climate change on natural hazards that are a concern for the area, both for the operation of KGL and for the Kafue River Climate change challenges elevate the importance of managing Basin generally. These key hazards include flood, drought, wildfire, water in an integrated manner so that adequate supplies are landslides, and disease. These analyses yield initial projections of available for power generation and for sector-based needs upon possible future annual adjusted losses for each of the hazards. To which the area economy and livelihoods depend. The hydropower provide further insight into the water demands that were modeled system operates within the larger Kafue River Basin and water- in the development scenarios, further evaluation was applied to related challenges will be exacerbated by climate change impacts select sectors that rely on water, including: agriculture, mining, on natural hazards and other sectors. The impacts of climate and domestic residential use. change, population growth, and development will impact ZESCO operations in the future. Adaptation Options Finally, the study’s findings were used to identify adaptation goals and strategies that ZESCO and stakeholders can consider in response to the climate change risks identified. The report concludes with recommendations for further action and analysis, as well as a discussion of uncertainty associated with the analysis. TABLE 1-2: WATER WITHDRAWALS FOR P-1 (MAXIMUM POWER), CONSERVATION AND DEVELOPMENT SCENARIOS Below IT Dam to KGU Dam Total Above IT Dam (Kafue Flats Area) Total Abstractions Conserva- tion Re- New Ag (ha/ Total Req New Ag (ha/ Total Req Total Ag Total DMI Total Req leases (cms Scenario cms) (cms) cms) (cms) (cms) (cms) (cms)* [Months]) P-1 0 6.6 0 16.8 14.3 9.1 23.4 0 C-1 0 6.6 0 16.8 14.3 9.1 23.4 300 [Mar] C-2 0 6.6 0 16.8 14.3 9.1 23.4 300[Mar-Apr] C-3 0 6.6 0 16.8 14.3 9.1 23.4 400 [Feb] 600[Mar-Apr] I-3 20,000/5.3 11.9 0 16.8 19.6 9.1 28.7 0 I-6 0 6.6 20,000/11.4 28.2 11.1 9.1 34.8 0 I-9 20,000/5.3 11.9 10,000/5.7 22.5 11.0 9.1 34.4 0 I-10 60,000/17.2 23.8 40,000/26.9 43.7 58.4 9.1 67.5 0 Notes: Based on information in the 2003 SEIA (SWP, 2003). Acronyms: cms = cubic meters per second; Ag = agriculture; DMI = domestic, mining, industry; ha = hectare; Req = required. 11 Climate Risk and Business: Hydropower, Kafue Gorge Lower Climate Change Assessment The key findings of the climate change assessment are the projected changes in temperature, precipitation, and their CCA The range of variability. The full report provides additional information about evaluated climate each of these issues. change scenarios ORA NHR&SA show a projected PROJECTED CHANGES IN TEMPERATURE 3°C - 5°C increase AO in the average Figure 2-1 shows the projected mean annual temperature for annual temperatures the two emission scenarios for the six GCMs studied, and their in Zambia by 2100, average (black line) for Zambia and for the Kafue River Basin. R The Kafue River Basin’s mean annual temperature is about 0.5°C and 3°C - 6°C in the cooler than the mean for Zambia. The variation in results among Kafue River Basin. the six models is reasonably small, with all projecting large temperature increases over this century. Figure 2-1: Simulated Annual Time Series of Temperature Spatially Averaged over Zambia (top panels) and the Kafue River Basin (bottom panels), for the A2 (left) and B1 (right) Emission Scenarios Zambia Zambia 28 28 27 27 26 26 Temperature (°C) 25 Temperature (°C) 25 24 24 23 23 22 22 21 21 20 20 19 1950 2000 2050 2100 19 Year 1950 2000 Year 2050 2100 GFDL GISS IPSL ECHAM5 MRI UKMO Model Average GFDL GISS IPSL ECHAM5 MRI UKMO Model Average Kafue River Basin Kafue River Basin 28 28 27 27 26 26 Temperature (°C) 25 25 Temperature (°C) 24 24 23 23 22 22 21 21 20 20 19 19 1950 2000 2050 2100 1950 2000 2050 2100 Year Year GFDL GISS IPSL ECHAM5 MRI UKMO Model Average GFDL GISS IPSL ECHAM5 MRI UKMO Model Average 12 Climate Risk and Business: Hydropower, Kafue Gorge Lower PRECIPITATION Figure 2-2 shows the mean annual precipitation per day projected The impacts of climate change are not projected to through the 21st century by the six GCMs, and the average of significantly change average annual precipitation the models (black line). None of the models projects significant changes in mean annual precipitation, for either emissions overall; model results range from -3% to +3% for the scenario B1 or A2. These results should be viewed in light of the A2 and B1 scenarios through 2100. uncertainty in GCM precipitation projections. These uncertainties are addressed further in the text and in the full report. Of the six GCMs studied, four indicate decreases in average annual precipitation and two indicate increases with a change The impacts of climate change are projected to increase of -0.09 to 0.06 millimeter per day (mm/d) or -3% to 2% across all three future time periods in the B1 emissions scenario and the variability of precipitation. By the late-century a change of -0.06 to 0.09 mm/d or -2% to 3% across all three period, maximum 1-day precipitation increases by over time periods for the A2 emissions scenario. While average 275% for some scenarios; this type of variability makes annual rainfall is not projected to change very much, climate dry days drier, and wet days wetter. change impacts on rainfall intensity are expected to be more significant. Figure 2-2: Simulated Annual Time Series of Precipitation Spatially Averaged over Zambia and the Kafue River Basin for Six GCMs and Two Emissions Scenarios (A2 and B1) Zambia Zambia 5.0 -1800 5.0 -1800 4.5 -1600 -1600 4.5 -1400 -1400 4.0 Precipitation (mm/yr) -1200 4.0 -1200 Precipitation (mm/yr) -1000 Precipitation (mm/day) 3.5 -1000 Precipitation (mm/day) -800 3.5 -800 3.0 -600 3.0 -600 -400 -400 2.5 2.5 -200 -200 2.0 -0 -0 2.0 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 1950 2000 2050 2100 Year 1950 2000 2050 2100 Year GFDL GISS IPSL ECHAM5 MRI UKMO Model Average GFDL GISS IPSL ECHAM5 MRI UKMO Model Average Kafue River Basin Kafue River Basin 5.0 -1800 5.0 -1800 -1600 -1600 4.5 4.5 -1400 -1400 Precipitation (mm/yr) Precipitation (mm/yr) 4.0 -1200 4.0 -1200 -1000 -1000 Precipitation (mm/day) Precipitation (mm/day) 3.5 -800 3.5 -800 3.0 -600 3.0 -600 -400 -400 2.5 -200 2.5 -200 2.0 -0 2.0 -0 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 1950 2000 2050 2100 1950 2000 2050 2100 Year Year GFDL GISS IPSL ECHAM5 MRI UKMO Model Average GFDL GISS IPSL ECHAM5 MRI UKMO Model Average 13 Climate Risk and Business: Hydropower, Kafue Gorge Lower VARIABILITY Figure 2-3: Temporal Variation of Annual Rainfall Consideration of projected average annual impacts of climate Compared to Base Period - ECHAM5 A2 change on temperature and precipitation must be augmented with evaluation of the intra-annual changes in variability. 2-3(A): ECHAM5 – BASE PERIOD AVERAGE ANNUAL RAINFALL 1400 To quantify the impact of precipitation variability, several 1300 precipitation indices were analyzed using RClimDex: consecutive 1200 wet days, consecutive dry days, maximum 1-day precipitation Rainfall (mm) 1100 (Rx1Day), maximum 5-day precipitation (Rx5Day), total annual 1000 precipitation (PRECPTOT), and the simple daily intensity index (SDII). 900 800 Figure 2-3 (a-c) shows the temporal variation in the average 700 annual rainfall for the base period, A2 emissions scenario, and 600 B1 emissions scenario for the ECHAM5 GCM. Average annual 500 1961 1966 1971 1976 1981 1986 rainfall shows a more or less constant trend for the A2 emissions Year scenario, with significant increases in annual precipitation (red line) in the last decade of the late 21st century period and an 2-3(B): ECHAM5 A2 – PROJECTED AVERAGE ANNUAL RAINFALL overall decreasing trend in average annual precipitation (black 1500 line) in the B1 emissions scenario. The long term average annual 1400 rainfall changes are about -5% (early 21st century) and -2 % 1300 (mid and late 21st century) for the A2 emissions scenario and are 1200 -3% (early), -10% (mid), and -7% (late) respectively, for the B1 Rainfall (mm) 1100 emissions scenario. 1000 900 During the late century, the -2% change in rainfall temporally 800 and the +1 to -13% change in rainfall spatially results in a positive 700 change of +11% in Kafue River flows. In this period, the maximum 600 1-day rainfall (Figure 2-4(a)) and annual rainfall (Figure 2-4 (b)) 500 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 (particularly for the last 8-10 years) are increasing compared to Year the base period. 2-3(C): ECHAM5 B1 – PROJECTED AVERAGE ANNUAL RAINFALL Table 2-1 shows the summary statistics for precipitation comparing the base period and the late-century period for the 1500 A2 scenario in ECHAM5. 1400 1300 1200 These results can also be considered in terms of “consecutive Rainfall (mm) 1100 wet days” and “consecutive dry days” as shown in Table 2-2. A 1000 “dry day” is defined as a day with less than 1 millimeter (mm) 900 of precipitation, while a “wet day” represents daily precipitation 800 amounts that are greater than or equal to 1 mm. The values in 700 Table 2-2 reflect the maximum number of consecutive dry or wet 600 days in a year, summarizing the shortest duration, the average 500 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 length, and longest duration of each of these conditions in the Year Base and the Late-Century periods. Table 2-1: Comparison of Precipitation for Base and Table 2-2: Comparison of Consecutive Wet Days and Dry Days Late-Century Periods for ECHAM5 A2 for the Base Late-Century Periods for ECHAM5 A2 Daily Average (mm) Annual Average (mm) Consecutive Dry Days Consecutive Wet Days Parameter Base Late Century Base Late Century Parameter Base Late Century Base Late Century Period Period Period Period Period Period Period Period Minimum 19 13 619 848 Minimum 104 126 26 27 Average 27 31 1,020 1,065 Average 148 170 75 77 Maximum 46 56 1,233 1,341 Maximum 186 214 113 132 St. Deviation 7.1 9.3 149 164 St. Deviation 19 17 25 29 Notes: mm = millimeter. St. Deviation = standard deviation. Notes: St. Deviation = standard deviation. 14 Climate Risk and Business: Hydropower, Kafue Gorge Lower The rainfall intensity (Figure 2-4(c)) is greater than in the baseline Figure 2-4: Variations in Extreme Rainfall Indices period (also showing a considerable increase in the last 8-10 years (ECHAM5 A2 Emissions Scenario) of the period). These combined factors, including the greater rainfall intensity, result in increased flows (a positive 11% change), 2-4(A): 1-DAY MAXIMUM RAINFALL (ECHAM5 A2) compared to the baseline period. There are several factors including (but not limited to) the evapotranspiration, temperature, 60 solar radiation, sunshine hours, relative humidity, and wind speed beside the spatial and temporal distribution of rainfall that govern 50 the hydrological processes of the watershed. Rainfall (mm) Baseline Late Century 40 Early Century To study the combined and interrelated impact of all of these 30 parameters, a detailed distributed hydrological model such as SWAT (Soil and Water Assessment Tool) can be employed; 20 however, use of a SWAT model was beyond the scope of this project. 10 1950 2000 2050 2100 In the absence of a SWAT model, researchers focus on further Year analysis of precipitation, which is the most influential factor affecting flow (Mutreja, 1986). Therefore, precipitation was further evaluated to estimate spatial and temporal variability and 2-4(B): ANNUAL RAINFALL (ECHAM5 A2) its impacts on the flow regime of the Kafue River Basin. 1400 1300 Late Century The analysis shows that rainfall intensity and variability have considerable impact on the changes in the flow regime of the 1200 Baseline Rainfall (mm) watershed over time for the projected climate change scenarios. 1100 1000 900 800 700 600 1950 2000 2050 2100 Year 2-4(C): RAINFALL INTENSITY (ECHAM5 A2) 9.0 8.5 8.0 Baseline Late Century 7.5 Early Century Rainfall (mm) 7.0 6.5 6.0 5.5 5.0 4.5 4.0 1950 2000 2050 2100 Year 15 Climate Risk and Business: Hydropower, Kafue Gorge Lower Figure 2-5: ECHAM5 A2 Comparison of Base Period and Late Century Average Monthly and Total Annual Rainfall 500 1500 450 400 350 1000 Monthly Rainfall (mm) Annual Rainfall (mm) 300 250 200 150 500 100 50 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Month Monthly average rainfall (mm), Baseline (1961-1990) Annual average rainfall (mm), Baseline (1961-1990) Monthly average rainfall (mm), End Century (2070-2099) Annual average rainfall (mm), End Century (2070-2099) Comparison of monthly average rainfall (mm) of ECHAM5-A2 Scenario at GCM Grid Point #46 (near IT Dam). The lines for each column show the high and low ranges of monthly values. These projected precipitation changes are an indication of the type By the late century period, maximum temperatures are of challenge that water-dependent sectors, such as agriculture, projected to exceed historical ranges for 8 months per will encounter over the next century. As shown in Figure 2-5, the small changes in annual average precipitation levels between year; of the 7 months that have traditionally received the base period and the late-century period are not equally rain, 3 are projected to become drier. These combined distributed within a year. The five months ranging from May to effects are likely to be disruptive to current life cycle September have historically received little to no rainfall. In the patterns for a number of plant and animal species. presence of climate change, the months of October, November and December are also projected to experience reductions in average precipitation levels. (16.9°C to 24.5°C) is divided equally into three relative temperature categories of “cool” (16.9°C to 19.4°C), “warm” (19.5°C to Comparison of monthly average rainfall (mm) of ECHAM5-A2 21.9°C), and “hot” (22°C to 24.5°C). For the future period, Scenario at GCM Grid Point #46 (near IT Dam). The lines for each values in excess of the base period range define a new, fourth column show the high and low ranges of monthly values. temperature category, labeled “exceeds historical experience.” The values and duration of this new category will likely require These changes will occur in combination with the previously adaptation responses for humans, plants, and animals. summarized temperature findings. Figure 2-6 illustrates the effect of these temperature shifts, compared to the base period. For the Temperature changes, combined with the changes in precipitation purpose of this example, the temperature range of the base period variability and intensity shown above, will likely have an impact 16 Climate Risk and Business: Hydropower, Kafue Gorge Lower Figure 2-6: ECHAM5 A2 Comparison of Base Period and Late Century Average Monthly and Total Annual Rainfall B1 Projected Temperatures Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Period Base 22.8 22.8 22.7 21.7 19.6 17.3 16.9 19.4 22.7 24.5 24.0 22.9 Early 23.8 23.8 23.9 23.0 21.0 18.4 18.2 20.6 24.1 25.8 25.5 24.0 Mid 24.5 24.5 24.7 24.0 21.9 19.4 19.2 21.3 24.6 26.8 26.4 24.6 Late 25.1 25.1 25.3 24.7 22.6 20.1 19.7 22.2 25.4 27.5 27.1 25.3 B1 Legend: Number of Months per Temperature Category Exceeds Historical “Cool” “Warm” “Hot” Experience Base 3 2 7 0 Early 2 2 6 2 Mid 2 1 4 5 Late 0 2 2 8 A2 Projected Temperatures Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Period Base 22.8 22.8 22.8 21.8 19.6 17.3 17.0 19.4 22.7 24.5 24.0 22.8 Early 24.0 24.0 23.9 23.1 21.0 18.7 18.5 20.8 24.0 25.9 25.6 24.0 Mid 25.0 24.9 25.2 24.2 22.1 19.9 19.5 21.9 25.4 27.5 26.8 24.9 Late 26.5 26.5 26.7 26.1 24.1 21.7 21.3 23.6 27.2 29.4 29.0 26.8 A2 Legend: Number of Months per Temperature Category Exceeds Historical “Cool” “Warm” “Hot” Experience Base 3 2 7 0 Early 2 2 6 2 Mid 0 3 2 7 Late 0 2 2 8 on traditional patterns of seasonality that will cause phenological (life cycle) changes for a number of species. This will be particularly relevant to agriculture, where planting and harvest dates will need to adjust to the shifting seasons, and impacts may be aggravated by a potential mismatch in the temporal combination of key climatic conditions, such as temperature and precipitation. In a larger environmental context, different phenological responses to the changing climate may disrupt coordination and interaction between species and their life cycles; for example, plants and their pollinators, predators, and prey; insects and their host plants; etc. These types of interactions and dependencies may cause cascading impacts on the food chain. Analyses of these interactions and climate impacts, and consequences for the Kafue River Basin context are beyond the scope of this study, but merit attention in the future. 17 Climate Risk and Business: Hydropower, Kafue Gorge Lower Operational Risk Assessment KAFUE RIVER FLOW CCA For the ECHAM5 A2 emissions scenario, the decrease in flows is much more significant than the decrease in average annual rainfall during the early-century period: an approximate -15% change in ORA NHR&SA flows compared to an estimated -5% change in average annual rainfall for the temporal variation. Based on a review of the data and various indices, this greater decrease in flow appears to be AO attributable to the significant decrease in the 1-day maximum rainfall during the early-century period. R During the mid-century period, there is no significant change in average annual rainfall for ECHAM5 A2 (a projected change of -2%) and similar results are observed for flows (a projected change of -1%). KGL POWER The changes in power generation relative to the base period (1961- Due to increased variability in precipitation, changes in 1990) ranged from -17.1 to +1.2% in the early-century period, average annual rainfall of -2% to -5% correspond to -22.1 to +4% in the mid-century period, and -8.8 to +11.7% in the changes in projected flows between -1% and -15%. late-century period across the GCM/SRES scenarios. Table 3-1 provides a summary of energy production at the KGL For the B1 emissions scenario, the temporal changes in long term power plant for the maximum power scenario, the two emissions average annual rainfall correspond to the change in the long term scenarios, and the four time periods for ECHAM5. Results for the average annual flows. During the early-century period, there is other GCMs are provided in Appendix 3 of the full report. no significant change in the average annual rainfall (an estimated -3% change) and similarly, there is no significant change in flows As shown in Table 3-1, energy production decreases in the early- (an estimated -1% change). During the mid-century period, period for both the A2 and B1 scenarios, but increases for A2 however, the decrease in flows is much more significant than the due to higher flows in the mid- and late-century periods. For decrease in average annual rainfall. Flows change by an estimated ECHAM5, B1, energy production at KGL decreases in the mid- -23% compared to a -10% change in rainfall (long term temporal and late-century, compared to the base period. variation) and a -9 to -16% change in rainfall (spatial variation). During the late century, the change in average annual rainfall of Under the P-1 scenario (maximum power), results -7 % temporally and -6.5 to -14 % spatially results in changes of -9% in Kafue River Basin flows. During this period, maximum indicate that reductions in water supply due to climate 1-day rainfall and average annual rainfall are increasing compared change could cause reductions in power generation to the mid-century period, but decreasing compared to baseline of up to 17% to 22% at the KGL plant in the early- period. and mid-century periods; however, increases in water supply in the late-century period could increase power generation by nearly 12%. TABLE 3-1: KGL ENERGY PRODUCTION FOR ECHAM5 SCENARIOS KGL 750MW Capacity Unit – Annual Energy Production (GWh/Yr) GCM SRES Develop- Base Early % Change Mid % Change Late % Change ment Period Century from Base Century from Base Century from Base Scenario Period Period Period ECHAM5 A2 P-1 2,227 1,847 -17.1% 2,182 -2.0% 2,487 11.7% ECHAM5 B1 P-1 2,160 2,099 -2.8% 1,682 -22.1% 1,970 -8.8% Notes: GWh/Yr = Gigawatt-hour per year; GCM = global climate model; SRES = Special Report on Emission Scenarios 18 Climate Risk and Business: Hydropower, Kafue Gorge Lower A WXGEN simulation was implemented to further evaluate select In the B1 emission scenario, all four development flow scenarios: the ECHAM5, A2, late-century (highest increase in scenarios (I-3, I-6, I-9, and I-10), result in decreases flow from the A2 base period) and the ECHAM5, B1, mid-century in average annual energy production compared to the results (highest decrease in flow from the B1 base period). Table P-1. This is true also for A2, except for the late-century 3-2 presents these results for the KGL power plant. Additional information on weather generators, their uses, and strengths and period which shows an increase for I-3, I-6 and I-9. limitations is included in the full report. reductions of up to 65% could occur at the IT generating station The study also evaluated various development (withdrawal) due to the implementation of C-3. These results would be further scenarios (“I” and “C” scenarios) compared to the baseline exacerbated with the effects of climate change on water flow and P-1 power level scenario. Energy production at KGL for the water demand. These conflicts between power generation goals development scenarios for the ECHAM5 A2 scenario is projected and environmental/conservation goals may require ZESCO and to be reduced between 2% and 16% in all four periods (base, other stakeholders to reconsider water management approaches, early-, mid-, and late-century) as compared to P-1. Energy particularly in light of climate change impacts. production at KGL for the development scenarios for the ECHAM5 B1 scenario has similar effects. The review of the impacts of climate change, conservation releases, and development scenarios indicate that the hydropower Appendix A3 of the full report provides additional information system at KGL operates with positive power outputs for the on the impacts at each power station across the various time P-1 scenario, which assumes increases of 30% in future water periods and development scenarios, for KGL and system power withdrawals from the river for agricultural, domestic, mining, generation. and industrial demands due to economic growth and without the impacts of climate change. Additional increases in water Conservation releases from the IT reservoir help recreate the demand due to climate change and development impacts are natural ecological systems in the Kafue Flats by approximating considered in the next section. The extra capacity added at the the river flooding conditions that existed during the rainy season IT reservoir during development (World Bank, 2009 and Beilfuss, before construction of the IT dam. Operational rules for the 2001), provides additional water storage capacity that supports reservoir model are based on historical ZESCO operating data productive management of the power system even given rainfall and prevent the reservoir level from falling below a minimum variability and potential precipitation decreases for some climate threshold. In some cases, this threshold prevents the model from change projections. However, negative power generation impacts evaluating full conservation release volumes for scenarios C-2 compared to the base period are observed for all future periods and C-3. Therefore, current operational rules may mask the full for the ECHAM B1 scenarios and for the early- and mid-century impacts of implementing these scenarios on power output. periods with ECHAM5 A2; for ECHAM5 A2, late-century, modeling results project power increases. For average annual power generation at KGL, the conservation scenarios do not appear to have a significant impact. However, When development scenarios are considered, average annual the modeling power outputs do not reflect the full impact of increases in water withdrawals rise to 59% for I-3. For I-6 and conservation releases because operating rules based on ZESCO I-9, the increase is 92% (with differences in the location of the information include minimum drawdown levels for water depth withdrawals). For I-10, a scenario that is only applied to the in the IT reservoir. These operating rules in the HEC ResSim late-century period, the increase in water withdrawals is 275%. model sometimes prevent the full volume of water for the C-2 All of these withdrawal levels yield negative impacts on power and C-3 scenarios from being released. Therefore, the results of generation compared to the P-1 scenario for various time periods. the modeling do not clearly reflect the impacts on power that would result from requiring the releases of water prescribed by These negative impacts highlight the need to view hydropower the C-2 and C-3 conservation scenarios. This is consistent with the operations in a systemic manner within the Kafue River Basin, results of the SEIA Report (SWP, 2003) which indicates that power where climate change impacts will occur over time, in combination TABLE 3-2: KGL ENERGY PRODUCTION FOR WXGEN SCENARIOS (GWH/YR) GCM SRES Use Base WXGEN % Change and Period Scenario Period Power Simulation Power from Base ECHAM5 A2-Late P-1 2,227 3,328 49.4% ECHAM5 B1-Mid P-1 2,160 2,189 1.3% Notes: GWh/Yr = Gigawatt-hour per year; GCM = global climate model; SRES = Special Report on Emission Scenarios 19 Climate Risk and Business: Hydropower, Kafue Gorge Lower with continued development and population growth. In addition, (maximum power) scenario, followed by the results that reflect it is important to note that reductions in flows are not only likely additional impacts of the assumed additional development needs to result in the above-mentioned reductions in power, but they (I-9) during this time period. Subsequent rows then provide could also increase water pollution by reducing the availability of corresponding results for key financial indicators. Electricity is water to dilute contaminants and adjust temperature gradients valued at $153.3/MWh and indexed to inflation for future years. in current and future industrial and municipal discharges (Stenek Additional financial assumptions are explained in the full report et al., 2010). and its Appendix A4 (Financial Analysis). Further, while not estimated in this report, increases in water These results demonstrate that the differences between the two temperatures in the Kafue Flats will result from the rise in average future emission scenarios have a significant impact on power air temperatures that are predicted to occur by the late-century generation, and therefore, on the financial viability of KGL. Even period. The potential future impacts of climate change on water with water withdrawals for development (I-9) in the B1 scenario, temperature, industrial discharges, and water-borne diseases KGL performs better than in the A2 “maximum power” scenario were not modeled as part of the study. (P-1). Of the four scenarios in Table 3-3, only the B1 maximum power scenario is projected to yield an internal rate of return (IRR) exceeding 20%, a common threshold for defining an acceptable FINANCIAL ASSESSMENT return for investors. Figure 3-1 shows the relationship between projected average annual power generation and the after-tax IRR Investors in the KGL power plant will expect to realize the for each of these four scenarios, as well as the corresponding finding expected return on their investments during the early-century from the original (without climate impacts) financial analysis. The period. Thereforehe impacts of climate change alone on the red line indicates the threshold IRR of 20%, suggesting that an “maximum power” (P-1) scenario will not interfere with required average annual generation of about 2,450 gigawatt-hours (GWhs) returns. However, the combination of climate change and other is the level needed to satisfy typical investor requirements. expected demands on the water supply may reduce power generation and the corresponding revenue stream below levels The combination of climate change and other expected that would be required by most investors. demands on the water supply is projected to reduce power generation – and thus KGL’s internal rate of The power results presented in the previous section represent the projected average annual generation for the respective 30-year return (IRR) – to levels below 20%, an IRR threshold future time periods. For the financial analysis of the early- that is not uncommonly required by investors. century period, the projected annual generation for KGL was used for each of the 30 years. The financial analysis focuses on For each of these scenarios, these results effectively consider a comparison of the P-1 (maximum power) and I-9 (maximum one possible outcome for the future time periods. A probability development forecast for the early-century period) scenarios. assessment provides a better characterization of the likelihood of The conservation scenarios (particularly C-2 and C-3) sometimes a range of outcomes for the same period, varying the results of produce low reservoir levels that violate KGL operating rules key inputs according to their distribution around their respective established by ZESCO; therefore, the ResSim Model prevents averages. In this case, a limited probability assessment was these releases from occurring in some years. Because the full performed (i.e., nine times) for the “maximum power” option in impact of the conservation scenarios on power and revenue the A2 emission scenario. The results for power generation and generation is not captured, they are excluded from the financial their corresponding IRRs are shown in Figures 3-2 (a) and (b). analysis. In light of the variability in the annual generation across the 30 Table 3-3 shows the results of the financial analysis for the early- years of the early-century period, these figures show that the century period for the ECHAM5 A2 and B1 scenarios. The first average annual generation of KGL may become 2,221 GWh rather row of each table shows projected energy generation for the P-1 than 2,090 GWhs, with an IRR of nearly 18% rather than 16%. TABLE 3-3: FINANCIAL PERFORMANCE FOR ECHAM5 A2 SCENARIO; BASE PERIOD AND EARLY CENTURY TIME PERIOD ECHAM5 A2, Early Scenarios ECHAM5 B1, Early Scenarios Performance Category P-1 I-9 P-1 I-9 Average Annual Generation (GWh/yr) 2,090 1,950 2,475 2,175 Internal Rate of Return (%) (after tax) 16.2 14.5 21.5 17.3 Net Present Value (USD Millions) 732 547 1,240 843 Payback Period (on Equity, in years) 8.6 10.7 5.5 7.6 Notes: GWh/yr = gigawatt hours per year; USD = U.S. dollars. 20 Climate Risk and Business: Hydropower, Kafue Gorge Lower A more robust probability analysis will deliver different results, As discussed further in Section 3.5 of the full report, there is which should better inform planning decisions. Again, these uncertainty associated with a number of factors that tie to the results are generated by varying the distribution of the 30 years financial analysis; in addition, the financial analysis uses the of annual electricity production to illustrate the possible impact outputs of the modeling effort, which also includes uncertainty for ROI. Further improvement would result with probability in regards to climate change projections, flow projections, and assessments of the available flow to the power plant. energy generation. For example, the IPCC, which established the range of future emissions scenarios that includes A2 and B1, states that no one scenario is a more likely outcome than any Figure 3-1: Projected Early Century Annual other (Nakicenovic, N., et. al., 2000). In addition to uncertainty Generation and After-Tax Internal Rate of Return associated with the modeling steps, there is also uncertainty for A2 and B1 Scenarios about the future demands for conservation releases and other uses of water in the basin. Irrigation and DMI water demand levels will be driven by the growth of the economy in general 3000 25 and of the agriculture sector in particular and will be authorized 2500 by the Department of Water Affairs; in addition, climate change 20 may drive the need for increased irrigation due to impacts on 2000 15 evapotranspiration. Other uses, such as DMI, are also anticipated Percent to increase – but the precise amount is uncertain and generally GWh 1500 10 considered less significant than changes to irrigation water 1000 demand for agriculture. 5 500 Given the significance of water flow on the financial 0 0 viability of hydropower projects, adaptation planning Original B1 P-1 B1 I-9 A2 P-1 A2 I-9 P-1 should include considerations such as climate change, conservation, and development that introduce variability Average Internal Rate Required Annual of Return Rate of into available water flow to the project. Generation (after tax) Return Figure 3-2: Probability Assessment of KGL Performance, Early Century 3-2(A): PROJECTED FREQUENCY CURVE, KGL ANNUAL GENERATION 0 500000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 4,000,000 Avg = 2,019,217 GWh SD = 379,854 N = 9 Notes: Avg. = average; MWh = Megawatt-hour; SD = standard deviation; N = number of evaluations of annual MWh variability. 3-2(B): PROJECTED FREQUENCY CURVE, KGL INTERNAL RATE OF RETURN 14 15 16 17 18 19 20 21 22 23 24 25 Avg = 19.69% SD = 1.19 N = 9 Notes: Avg. = average; IRR = internal rate of return; SD = standard deviation; N = number of evaluations of annual MWh variability. 21 Climate Risk and Business: Hydropower, Kafue Gorge Lower Natural Hazard Risk & Sector Assessment NATURAL HAZARD RISK CCA Climate change variability causes greater uncertainty in water supply and demand in the future. This can result in fewer, more extreme precipitation events, and longer, more severe periods of ORA NHR&SA drought between them. This would greatly impact water supply and management issues for ZESCO and other sectors. AO FLOODS: The flood hazard risk assessment shows that floods have been occurring approximately 0.42 times each year or once every 2.3 years (OFDA/CRED, 2007). To help understand the R conditions which caused each flood event, precipitation data were analyzed for each flood time period. This data was used to identify a rainfall deviation from normal seasonal rainfall (October to March) that constitutes a flood event; this “flood threshold” was identified as 15% in Lusaka. That is, a 15% deviation from the for some future time horizons, changes to spatial and temporal normal rainfall for the rainy season results in a flood. According to variability in the basin may result in more extreme precipitation several of the GCMs/emissions scenarios, events that exceed the events and greater run-off; this could increase the frequency or flood threshold will continue and in some cases, increase, in the severity of floods. region. Therefore, flood frequency is predicted to remain high in the future. Table 4-1 shows the probability of exceeding the KGU has been in operation since 1977 and was damaged by flood flood threshold across the three GCMs, two emission scenarios, waters in 2005. Of the nine floods identified in 31 years in the and three time horizons. These probabilities may be converted to ZESCO operations area, damage to ZESCO facilities was recorded return period events. For example, ECHAM5, A2, Early Century for only one event. The annual loss for a hazard is determined shows a flood event (exceeding the threshold) occurring every by plotting the probability of occurrence on the x-axis and the 4-years (100/25), then increasing in the mid century to ever 2.9 corresponding losses on the y-axis and then calculating the area years (100/35). If the dam was built to withstand a maximum under the curve. The average annual loss for flood is calculated probable flood of 1000-years based on historical events, it should to be USD 449,586. To put this in perspective, the total annual be understood that the future maximum probable flood may revenue for the Kafue Gorge power system (IT, KGU, and KGL) change based on the GCM and emission scenario. is USD 850 million. The average annual loss due to flooding for the MRI B1 late century results is estimated as USD 449,586. More historical loss data would provide a more representative Greater variability in precipitation could increase the annualized loss value, but since the facility has only been in likelihood of floods and droughts. operation since 1977 the historical record is not long (in terms of meteorological timeframes). Table A5-3 in Appendix A5 of the full report was used to help calculate the modeled future annual Based on a review of historical events and corresponding water losses for each GCM/emissions scenario across the future time depths and impacts, flood magnitude should also be rated as horizons. The modeled future annual losses are shown in Table high now and in the future. This study shows that even if average 4-2. This loss does not include any potential future losses for the annual precipitation decreases for some GCM/emission scenarios additional power generating units planned for KGL and IT. TABLE 4-1: PROBABILITIES OF EXCEEDING SEASONAL PRECIPITATION FLOOD THRESHOLD Emissions Scenario A2 Emissions Scenario B1 Early Century Mid Century Late Century Early Century Mid Century Late Century GCM (2010-2039) (2040-2069) (2070-2099) (2010-2039) (2040-2069) (2070-2099) ECHAM5 25% 35% 39% 29% 16% 23% IPSL 21% 36% 16% 43% 26% 29% MRI 22% 28% 18% 19% 14% 27% 22 Climate Risk and Business: Hydropower, Kafue Gorge Lower TABLE 4-2: PROJECTED FINANCIAL LOSSES DUE TO FLOOD (ANNUALIZED USD) Emissions Scenario A2 Emissions Scenario B1 Early Century Mid Century Late Century Early Century Mid Century Late Century GCM (2010-2039) (2040-2069) (2070-2099) (2010-2039) (2040-2069) (2070-2099) ECHAM5 416,286 582,795 649,404 482,886 266,418 382,977 IPSL 349,677 599,445 266,418 716,004 432,936 482,886 MRI 366,327 466,236 299,727 316,377 233,118 449,586 TABLE 4-3: PROJECTED FINANCIAL LOSSES DUE TO DROUGHT (ANNUALIZED USD) Emissions Scenario A2 Emissions Scenario B1 Early Century Mid Century Late Century Early Century Mid Century Late Century GCM (2010-2039) (2040-2069) (2070-2099) (2010-2039) (2040-2069) (2070-2099) ECHAM5 13% 9% 10% 14% 19% 17% IPSL 15% 14% 22% 5% 9% 10% MRI 19% 13% 22% 21% 26% 13% DROUGHTS: The drought hazard assessment shows that Looking at the events which impacted this system for ZESCO, two droughts are occurring frequently at approximately 0.19 times major droughts caused a reduction in power over 25 years, one each year or once every 5 years (OFDA/CRED, 2007). Similar to in 1995 and one in 2005. Detailed loss information is available the approach used to identify a flood threshold, historical analysis for the 2005 drought event, which is described in the paragraph was used to identify a drought threshold. This threshold was above. This loss was used to approximate an average annual loss established as a negative 12% change from the average annual for drought of USD 2,105,360. To put this in perspective, the precipitation for each documented drought event. For the total annual revenue for the Kafue Gorge system (IT, KGU, and ECHAM5 A2 late-century, the precipitation results show that the KGL) is USD 850 million. Table A6-3 in Appendix A6 of the full probability of reaching drought threshold conditions would be report was used to help calculate the estimated future annual lower at 10% (once in 10 years) than historical trends. loss shown in Table 4-4. This loss does not include any modeled future loss for the additional power generating unit. These results align with average annual rainfall trends and precipitation variability analysis discussed in Section 3.2 of the The financial risks to ZESCO associated with drought events are full report for this GCM/emissions scenario in the late century by far the most potentially expensive compared to the other period (compared to its respective baseline). For the mid-century evaluated hazards, with costs in the millions of dollars (USD), ECHAM5 B1 results, the probability of droughts is 19% (about depending upon the scenario and timeframe. Therefore, ZESCO’s once in 5 years). priority for climate change adaptation planning would likely need to focus on the drought hazard. With reduced average annual precipitation projected during this time horizon for the same GCM/emissions scenario (compared The financial risks to hydropower productiion to its respective baseline period) and the precipitation variability discussed in Appendix A5 of the full report, it is expected that associated with drought events are by far the most droughts could occur more frequently for several of the GCM/ potentially expensive compared to the other evaluated emission scenarios, as shown in Table 4-3. Using the mid-century natural hazards. MRI B1 results, the analysis shows there is a higher probability of droughts of 26%, or once in 3.8 years compared to historic data, which aligns with the decrease in average annual precipitation The risk assessments presented in the full report have been and precipitation variability for this GCM/emissions scenario/time summarized in tables to facilitate comparison and ranking. Table horizon. 4-3 in the full report shows the hazard evaluation results with the elements of frequency and magnitude for current and future conditions for each natural hazard. Each component has been 23 Climate Risk and Business: Hydropower, Kafue Gorge Lower TABLE 4-4: PROJECTED FINANCIAL LOSSES DUE TO DROUGHT (ANNUALIZED USD) Emissions Scenario A2 Emissions Scenario B1 Early Century Mid Century Late Century Early Century Mid Century Late Century GCM (2010-2039) (2040-2069) (2070-2099) (2010-2039) (2040-2069) (2070-2099) ECHAM5 3,421,210 2,368,530 2,631,700 3,684,380 5,000,231 4,473,891 IPSL 3,947,550 3,684,380 5,789,741 1,315,850 2,368,530 2,631,700 MRI 5,000,231 3,421,210 5,789,741 5,526,571 6,842,421 3,421,210 ranked high, medium, or low. Table 4-4 in the full report shows 3) Develop better topographic data to support the hydrological the vulnerability assessment results with elements of exposure modeling and help to reduce uncertainty associated with flow and sensitivity for current and future conditions and adaptive estimates and provide a more accurate floodplain delineation capacity. The adaptive capacity is a positive trait so a high to assist in evaluating the flood hazard. adaptive capacity will decrease vulnerability, while high exposure 4) Document changes in hydropower operation associated with and sensitivity will increase vulnerability. Table 4-5 in the full periods of drought. This would provide better information report shows the overall risk assessment results, incorporating concerning potential losses in power generation associated the hazard and vulnerability findings. with drought conditions and help to determine if changing the operational procedures could be used in more dire The results of this analysis clarify that the financial risks to ZESCO circumstances. associated with drought events are by far the most potentially expensive, ranging from 1.3 million USD to 6.8 million USD depending upon the scenario and timeframe (all results are SECTOR ASSESSMENT annualized). The potential annualized cost of future loss due to floods yields results in the range of 233,118 to 449,586 USD. Figure 4-1 summarizes the combined effects of future The potential annualized cost of future damage to ZESCO due to development, with and without the additional impacts of climate wildfire yields results in the range of 115,000 to 170,000 USD. change, on water demand. For each future period, the first three Landslides are projected to be far less damaging, registering columns show current and projected demand for water from the annualized losses between 25,000 USD and 80,000 USD. The agricultural, domestic, mining, and industrial sectors. These are increased risk of disease, in the form of malaria, produces a low level the major categories of demand; however, because no other type of direct financial risk to ZESCO based on the assumptions used in of water demand is captured by these bottom three lines, they this assessment. Although assumptions made in the assessment represent an understatement of total potential demand. These of each of these risks include various levels of uncertainty, the estimated demands were developed from resources independent magnitude of the differences among the results suggests that the of the estimated water demands that were incorporated into the ranking of the potential costs of these risks to ZESCO is likely to be modeling analysis (based on the 2003 SEIA Report), and further correct. Therefore, ZESCO’s priority for climate change adaptation corroborate the values used as withdrawals to support flow and planning needs to be the drought hazard. power modeling. The hazards identified as high risk are considered further as part The final two bars for each future period in Figure 4-1 show of Adaptation Options in this report (and Section 5.0 of the full projected water supply in the Kafue River based on the outputs report). of the A2 and B1 emissions scenarios for the ECHAM5 GCM. To enhance risk modeling efforts in the future, several steps would be useful, including: Temperature increases are projected to contribute to increases in future water demands, due to increased 1) Collect detailed precipitation and temperature across the basin. evapotranspiration, with annual average increases of 6% Add additional rain and temperature gauge instrumentation in the early- and mid-century periods and 13% in the late across the basin to better understand spatial variability. Staff using the instrumentation would need to have proper training century period. These averages mask the much larger and funds for maintenance and data collection in order to effects of intra-annual variability in evapotranspiration support a consistent and accurate approach. that will drive demand higher among humans, animals 2) Capture detailed loss associated with hazards that are impacted and plants. by precipitation and temperature variability. If these losses can be tied to a return period interval, they can be used to better estimate average annual loss for specific hazards. 24 Climate Risk and Business: Hydropower, Kafue Gorge Lower These are the same flow rates used as inputs for electricity generation projections in the Figure 4-1: Comparison of Projected Average Annual Demand and HEC ResSim (energy) modeling. The height Supply of Water in the Kafue River Basin of the column shows the 30-year average supply for the two scenarios; the boxed 406 301 355 area surrounding the top of these columns 294 300 illustrates the extent of one standard Proj deviation of the 30 years’ of projected supply levels around the 30-year average. 250 Proj Finally, the lines bisecting the supply columns indicate the further extent of the 200 Proj maximum and minimum values for the 30-year period. The distance between the heights of the supply and demand columns 150 Proj represents instantaneous flow available in the Kafue River for hydropower generation. 100 Curr It is important to note, however, that flow to the hydropower plants is managed with reservoirs, where the available water 50 accumulates over time. As with economic demands for water, the water requirements of hydropower plants can be represented in 0 Early Mid Late units of cubic meters per second, but the power plants draw upon the stored volume Projected Supply Projected Needs in the reservoirs, not the instantaneous (ECHAM5 A2) due to Growth available flow in the river. Projected Supply (ECHAM5 B1) Current Levels The graph shows that for the ECHAM5 B1 scenario, the instantaneous supply for Projected Growth Needs hydropower generation in the early-century with Climate Change period is, on average, about twice as large (90 cms) as it is projected to be in the late period (45 cms); however the alternative Note: cms = cubic meters per second projection from ECHAM5 shows supply Projected supply columns show average results for the period, as well as the range of +/- 1 standard deviation (about 66% of the results), and the extent of the maximum and minimum projected values increasing at a rate that is nearly equal to for the 30-year period. Additional detail for supply results are included in the full report. the projected increase in demand (about 80 “Projected Demand” = projected combined annual water demand from the agricultural, domestic, mining, and industrial sectors. Projected values for early-, mid-, and late-century reflect average cms in the early period vs. about 75 cms in values for each 30 year period. Agricultural projections reflect scenarios I-3, I-9, and I-10. Other the late period). sectors incorporate input from other sources, as explained in the text. Because conservation releases for biodiversity are not consumptive uses, they are not reflected. “Base Year Demand” = 2008 Water Demand (World Bank, 2009). 25 Climate Risk and Business: Hydropower, Kafue Gorge Lower Adaptation Options Based on climate change projections for flow and energy impacts, other water uses, and the hazards risk assessment and sector CCA analyses, three adaptation goals were identified for ZESCO. These were based on high risk concerns to ZESCO operations as summarized below. ORA NHR&SA Although drought is not currently impacting power generation, this study indicates that drought and AO precipitation variability (or decreases in precipitation shown for some models), combined with other water uses and conservation needs, may negatively impact water available R for power generation in the future. This could decrease power generation in the future, compared to the present plans, during a future when power demand will be higher than the present (through population growth, increasing sector needs, and other causes). Therefore, the potential for drought and precipitation Given food security needs and the economic importance of variability warrant planning efforts to prevent losses to ZESCO. agriculture in the Kafue River Basin, the potential increases in local agriculture (family and commercial), agriculture’s large percent of A review of the risk assessment results shows that damage from water use (75% of water demand), and the potential impacts of flood-induced mudslides/landslides discussed in the full report climate change (increased temperatures, precipitation variability, has already impacted ZESCO facilities. It is likely this hazard will and changes in precipitation peaks), water demands for irrigation be exacerbated by climate change and further development of in the area can be expected to draw water away from ZESCO additional hydropower systems along highly sloped gorges (e.g., operations in the future. While the primary water demand is from KGL). Addressing landslide/mudslides was selected as adaptation agriculture, water needs for conservation, public uses, and industry goal, since adaptation will reduce losses that could occur from such uses also may increase water abstractions and impact the timing events either with or without additional exacerbation from climate of releases. Working with area stakeholders on water issues, with change impacts. a strong focus on agriculture, should help to reduce unplanned or negative impacts on ZESCO operations. Potential adaptation strategies have been identified and evaluated for each of the three adaptation goals. These strategies were first categorized using the “POSE” approach developed by USAID (USAID, 2007): physical, operational, social, and economic (POSE). The adaptation strategies were then screened using criteria developed by the U.S. Federal Emergency Management Agency (FEMA). The seven criteria are: social, technical, administrative, political, legal, economic, and environmental (STAPLEE). FEMA uses these criteria to evaluate potential hazard mitigation options; they are described in detail in FEMA publication 386-3 (FEMA, 2003). The STAPLEE evaluations for the three adaptation goals represent preliminary assessments which serve as examples for the process. They should be modified further based on ZESCO and appropriate stakeholder input at the local level. In these preliminary evaluations, some strategies are scored with a downward arrow to indicate possible negative impact. For example, conventional (fossil) energy source implementation would have an adverse impact on the environment as compared to renewable energy sources; other strategies may be relatively costly or technically challenging. An upward arrow represents likely benefits or positive traits that the strategy may have for the criterion being evaluated. For example, implementing some of the strategies may be beneficial to the 26 Climate Risk and Business: Hydropower, Kafue Gorge Lower TABLE 5-1: ADAPTATION GOAL #1 STRATEGY EVALUATION Adaptation Strategy (Category) S T A P L E E Total Integrate Water Management Approaches h h U U U U h + and Develop Water Use Regulations (P, O, S, E) Implement New Land Use Planning Approaches h h U U U h h + and Regulations (O, E) Expand Reservoir Capacity (P, O, S, E) U h U U U U $ U Conserve Energy (P, O, S, E) h U U U g U h + Diversify Energy Sources (P, E) g h $ U U $ $ U Implement Pumped Storage Electricity Generation (P) U $ U U U U $ U Water Efficiency, including a Drought Management Plan g h U U U h U + (S, O) Implement Training and Public Involvement (S) h h U U g h h + Financial Strategies U U U U U U h + Notes: For adaptation strategy: P = physical; O = organizational; S = social; E = economic. For evaluation criteria: S = social; T = technical; A = administrative; P = political; L = legal; E = economic; E = environmental; For ranking: $ = adverse; g = insignificant; h = beneficial; U = unknown; + consider further TABLE 5-2: ADAPTATION GOAL #2 STRATEGY EVALUATION Adaptation Strategy (Category) S T A P L E E Total Stabilize Slopes (P) g h U g g h g + Improve Elevation Mapping (P) g h U g g U g + Implement Real-Time Monitoring (P) g g U g g U g U Develop an Excavation and Fill Ordinance (O) g g U U U h h + Develop Land Use/Building Restrictions (O) g g U U U h h + Develop Training and Response Plans (P, O, S, E) g g U U U h h + Notes: For adaptation strategy: P = physical; O = organizational; S = social; E = economic. For evaluation criteria: S = social; T = technical; A = administrative; P = political; L = legal; E = economic; E = environmental; For ranking: $ = adverse; g = insignificant; h = beneficial; U = unknown; + consider further environment, be low cost or yield social benefits. A “U” indicates The six strategies for Adaptation Goal #2 include physical, that the directional effect for the strategy on a criterion is unknown organizational, social, and economic actions as indicated in Table at this time; each of these preliminary rankings may change with 5-2. The strategies were evaluated using the STAPLEE criteria and additional information. In the final column, “Total”, the net impact the preliminary ranking results are shown in Table 5-2. Again, of the individual rankings is represented and can be positive, input from ZESCO and appropriate stakeholders at the local level negative or unknown/incomplete. Strategies with a positive are recommended to refine the evaluation of the adaptation ranking in the “Total” column should be retained as a potential strategies. recommendation pending further consideration, as needed. The four adaptation strategies for Adaptation Goal #3 include Nine adaptation strategies were identified and evaluated to support organizational, social, and economic actions as indicated in Table Adaptation Goal #1: preventing losses to ZESCO from drought and 5-3. The strategies were evaluated using the STAPLEE criteria precipitation variability. The results of this preliminary evaluation and the preliminary ranking results are shown in Table 4-7. Here are presented in Table 5-1. too, input from ZESCO and appropriate stakeholders at the local level are recommended to refine the evaluation of the strategies. 27 Climate Risk and Business: Hydropower, Kafue Gorge Lower TABLE 5-3: GOAL #3 STRATEGY EVALUATION Adaptation Strategy (Category) S T A P L E E Total Assist the Government of Zambia in Implementing the h g g h g U h + Agricultural Adaptations in the NAPA Report (O, S, E) Support Comprehensive Study of Actual Water Usage in h g g h g U h + the Kafue River Basin (O, S, E) Forge/ Maintain Active Partnerships with Water Users/ h g g h g U h + Stakeholders in the Kafue River Basin (O, S, E) Assist in Public Education and Outreach on h g g h g U h + Implementation of Efficient Irrigation Methods (S) Notes: For adaptation strategy: P = physical; O = organizational; S = social; E = economic. For evaluation criteria: S = social; T = technical; A = administrative; P = political; L = legal; E = economic; E = environmental; For ranking: $ = adverse; g = insignificant; h = beneficial; U = unknown; + consider further As discussed above, stakeholder input would be valuable to To minimize reduced flows, four adaptation strategies were help check the rankings indicated above to ensure all promising identified and evaluated to reduce water demand from strategies have been identified. Based on the preliminary rankings shown in Tables 5-1 through 5-3, some promising options were development through improved agricultural methods. explored further. These include strategies that run across the three adaptation goals, such as: integrated water management Adaptation goals, descriptions and evaluations of the strategies, (ties to Adaptation Goal #1), stabilization of slopes (ties to and other information are included in Section 5.0 of the full Adaptation Goal #2), and implementation of efficient irrigation report. Recommendations regarding the promising adaptation methods (ties to Adaptation Goal #3). strategies identified from this study’s evaluation are presented in the recommendations of the full report and are summarized below. Recommendations This section presents the major recommendations of this study. Because different stakeholders may prioritize CCA these recommendations differently, the numbering of the recommendations is for ease of reference, and is not intended to indicate a ranking of importance. ORA NHR&SA Climate change challenges highlight the importance of managing water in an integrated manner so that adequate AO supplies are available for power generation and for sector- based needs upon which the area economy and food security/ livelihoods depend. While, in isolation, the hydropower system R may be able to cope with climate change, the system operates within the larger Kafue River Basin and water-related challenges will be exacerbated by climate change impacts on natural hazards and other sectors. The impacts of climate change, population concerns merit ZESCO’s involvement in climate change planning growth, and development will affect ZESCO operations in and basin-specific approaches to water management and the future. Therefore, combined water demand for power, development planning. conservation, domestic use, agriculture, industry, and other needs must be addressed through integrated planning efforts. Improved data collection is needed to evaluate climate change impacts, especially for hydropower planning. It may be in ZESCO’s interest to become proactively ZESCO’s involvement in national and basin-wide planning efforts involved with national adaptation planning efforts. would help identify opportunities to improve local meteorological, Climate change will likely exacerbate any existing water demand land use, hydrologic flow, and stream cross section data to and development tensions and challenges in the future. These support refined modeling and analysis over time. To support 28 Climate Risk and Business: Hydropower, Kafue Gorge Lower daily modeling, increased consistency in the collection of daily Climate changes that impact the flow of the Kafue River data would be useful; in addition, a number of studies have cited will have impacts on the financial performance of its that a greater density of meteorological stations would help to hydropower plants. These impacts highlight the importance refine water modeling in developing countries. of considering changes in water supply due to climate change when implementing financial analyses for hydropower projects. To improve the effectiveness of climate change adaptation Investors can be better informed about climate change risks to efforts, it is needed to expand engagement with other future hydropower projects by including projected changes in Basin users and relevant stakeholders. These are generally available water flow and power generation, rather than assuming lower cost strategies that can achieve significant positive results, constant flows and power generation rates. and develop good-will with customers and other stakeholders. The complexity of evaluating potential climate change For physical adaptation strategies, feasibility studies and benefit/ impacts suggests that probabilistic assessments will cost analyses, capital project planning would need to be provide substantial benefits over deterministic evaluations. completed to determine which options would the most In this study, multiple simulations were completed at several positive financial and operational impacts for ZESCO. stages of analysis. Increased use of probabilistic techniques for Preliminary financial data has been included in the analysis, but future assessments will better inform planning for climate change. additional stakeholder impact on priorities and costs are needed for detailed analysis and to support management decisions. Climate change and climate risk assessments are evolving fields. Therefore, this report presents findings and recommendations, as The significant withdrawals at the IT dam observed with C-2 and well as strengths and limitations of the data and methodologies C-3 indicate that these higher levels of conservation releases applied and uncertainties associated with the modeling and conflict with current operating rules for power generation. data inputs and outputs. The project approach recognizes that The resulting inability to support conservation releases may be while there are uncertainties associated with quantifying inconsistent with goals for preservation of the ecology of the Kafue climate change impacts, there is a pressing need to plan for Flats. Further study is required to identify and evaluate options for these impacts. Approaches and findings should be reviewed to reconciling power generation and environmental goals. identify an over-arching framework that can be used to integrate climate change considerations into existing environmental or In a larger environmental context, different phenological responses overall project considerations. It may be useful to convene a to the changing climate may cause disruption of coordination and workshop or working session with a number of agencies, donors, interaction between species and their life cycles (e.g. spawning countries, and technical specialists to review approaches and cycles of fish, or interactions between plants and their pollinators,.), identify appropriate, cost-effective means to integrate climate and have cascading impacts on the food chain. Analyses of this change considerations into project evaluation methods. type of climate change impacts are beyond the scope of this study but merit attention given the possible effects on the needs for different regimes of water releases. Sources: Data and Models TEMPERATURE AND PRECIPITATION PROJECTIONS HYDROLOGIC FLOW MODELING Temperature and precipitation analyses were performed for the To estimate the projected climate change impacts on flow in the statistically downscaled results of six GCMs, across a base period Kafue River Basin study area, the Hydrologic Modeling System and three future time periods (early-, mid- and late-century) (HMS) from the Hydraulic Engineering Center (HEC) of the U.S. of the 21st century, and two emissions scenarios (A2 and B1) Army Corps of Engineers (USACE) was used. USACE HEC-HMS developed by the IPCC. Outputs were provided as gridded data modeling software is available at no charge for download at: sets for both Zambia as a whole and for the Kafue River Basin http://www.hec.usace.army.mil/software/hec-hms/index.html. study area. Downscaled datasets are available free of charge from The data inputs and outputs for this step are provided in Appendix the University of Santa Clara, California at: http://www.engr.scu. B of the full report. edu/~emaurer/global_data/. The data inputs and outputs for this step are provided in Appendix B of the full report. 29 Climate Risk and Business: Hydropower, Kafue Gorge Lower Of the original six GCMs, the three that produced high, low, and tool was developed by Natural Resources Canada, with input median climate change results were selected for further analysis. from stakeholders such as the U.S. National Aeronautics and Scenarios developed from the three GCMs and two emissions Space Agency. The financial analysis also considered the impacts scenarios were used to model flow for a base period and for of the conservation and development scenarios on power the three future time periods. From these results, the scenario generation. RETScreen is available for download at no charge combinations that yielded the greatest increase and decrease in from Natural Resources Canada at: http://www.retscreen. flow from their respective base periods were selected for further net. The data inputs and outputs for this step are provided in analysis using a U.S. Department of Agriculture WXGEN tool. Appendix B of the full report. The WXGEN tool can be downloaded at no charge from: http:// epicapex.brc.tamus.edu/downloads/ model-executables/wxgn- v3020. The data inputs and outputs for this step are provided in NATURAL HAZARD RISK Appendix B of the full report. AND SECTOR ASSESSMENTS This study also considers the indirect risks of climate change to RESERVOIR (ENERGY) MODELING ZESCO through impacts on natural hazards (flood, droughts, landsides, wildfires, and disease) and potential impacts on select A hydrodynamic and energy model (also available from the HEC economic and environmental sectors that use water (agriculture, at USACE), the HEC ResSim model, was used to evaluate power mining, urban demand, and conservation). The data inputs and impacts for the flow results from the HEC-HMS and WXGEN outputs for this step are provided in Appendix B of the full report. modeling. In addition to the climate change impacts on flow, HEC ResSim was used to consider three conservation (freshet releases) scenarios (C-1, C-2, and C-3) and four development (consumptive ADAPTATION OPTIONS water withdrawal) scenarios (I-3, I-6, I-9, and I-10), as shown in Figure ES-1. HEC ResSim modeling software is available for Based on modeling and analysis and review of recent literature, download at no charge from USACE HEC at: http://www.hec. adaptation goals were identified. These goals and climate change usace.army.mil/software/HECRessim/. The data inputs and outputs risks were then evaluated to identify adaptation strategies for for this step are provided in Appendix B of the full report. achieving two primary objectives: (1) address the most threatening and urgent concerns as promptly and effectively as possible and (2) identify those responses that have multiple benefits and/or FINANCIAL ASSESSMENT are most cost-effective in order to minimize overall adaptation costs. The adaptation section identifies three primary adaptation A financial assessment was completed using the Renewable goals, associated strategies, and promising adaptation options Energy Technology Screening (RETScreen) tool to consider the that are recommended for further consideration and potential impacts of climate change on the financial viability of KGL. This implementation by ZESCO. Uncertainty Information to support assumptions and analysis in this study investments given current and future conditions without climate include previous reports prepared by (or for) the IFC, the World change impacts and become even more warranted given the Bank Group or other entities, such as the European Commission climate change risks identified in this study. and the United Nations Food and Agriculture Organization, as well as operational data shared by ZESCO, and financial assumptions shared by the IFC. In addition, a range of data sets were used TEMPERATURE AND PRECIPITATION PROJECTIONS to support modeling. Uncertainty associated with project steps is address below. While considerable uncertainty is associated The uncertainties associated with GCM projections of climate with modeling steps and climate change projections, the study change have been categorized as: (1) unknown future emissions indicates that the risks are sufficient to warrant adaptation of GHG gases; (2) uncertain response of the global climate planning, even in the presence of uncertainty. Because there system to increases in GHG concentrations; and (3) incomplete is considerable uncertainty associated with climate change understanding of regional climatic changes and their impacts, risk assessment, adaptation goals and strategies in this report that will result from global changes (IPCC, 2001). Of these, the focus on those strategies that appear warranted given current first two are the largest sources of uncertainty in GCM projections conditions (past hazard events, projected water demand and and are estimated to be of roughly equal magnitude on a global electricity needs, etc.). The strategies proposed would be sound scale (Karl and Trenberth, 2003; Wigley and Raper, 2001). The 30 Climate Risk and Business: Hydropower, Kafue Gorge Lower degree of uncertainty resulting from the combination of these Development of the basic model for the hydrologic system three sources is unquantifiable. Confidence bounds cannot be incorporated available background information, including land placed on any GCM projections. use, hydrology, precipitation, hydrologic flow, rivers, basin designations, and other data. Background documents and other Overall, the projected average annual precipitation changes are publically available data were used to develop the model. Data relatively small, but are associated with large uncertainty; the limitations introduce uncertainty associated with the model; for simulation of precipitation remains one of the principal challenges example, land use data are from the United Nations Food and in global climate modeling. (Additional information is provided in Agriculture Organization (FAO) and may not be current for the Appendix A1 of the full report.) In particular, Africa has been entire basin. Where assumptions or input data differ from actual noted as an area that presents significant challenges in terms conditions, limitations and uncertainty are introduced into the of climate change projections for precipitation based on factors modeling effort. Additional information on the data sources, such as limited meteorological data and historical variability in approaches, and assumptions included in model establishment weather patterns (IISD, 2009). are provided in Appendix A2 of the full report. Adaptation strategies and recommendations included in this report identify In addition, the El Niño/La Niña-Southern Oscillation (ENSO) options for improving meteorological, land use, and river cross- climate phenomena can impact weather patterns in Zambia. section data in the future; modeling can then be refined as ENSO causes periodic extreme weather such as floods, droughts, improved data is obtained. and other weather disturbances, and these impacts are not necessarily captured by current GCMs. Studies are ongoing Seven meteorological stations with observed, daily historic regarding how ENSO is impacted by climate changes and how precipitation and temperature data were available for model ENSO and climate change might impact climate and weather calibration and validation. These seven stations are generally patterns in particular countries over time. However, a study of located near the boundary of the study area, rather than within various GCM models and their ability to capture ENSO phenomena the interior of the basin. Based on U.N. World Meteorological indicated that ECHAM5 is one of the models that reflect ENSO Organization criteria, between 50 to 150 stations would reasonably well in sea surface temperature (SST) variability (van be desirable in a basin of study area’s size, with hilly areas Oldenborgh, et. al., 2005). benefitting from a higher density of stations. The findings and conclusions from this study note that similar to other studies in There is considerable uncertainty with projecting temperature and Africa, modeling and analysis of climate change would benefit precipitation changes associated with climate change, particularly from better local data. “The climate observing system in Africa for time frames beyond 2050. Current concentrations of GHGs in is in a far worse state than that of any other continent, and is the atmosphere will affect climate through 2050; however the deteriorating” (Elasha, et. al., 2006). There are eight times actual GHG emissions that will occur between now and 2050 are fewer weather stations on the continent than the minimum unknown; therefore, the corresponding impact on temperature recommended level and “vast parts of central Africa remain and precipitation is more uncertain. unmonitored” (IISD, 2009).” Limitations to GCM models are becoming better understood, allowing The spatial distribution of the meteorological data set includes for further modification to improve performance. For example, recent significant averaging of precipitation across the basin that is analysis of GCM projections compared with observed precipitation likely not completely representative of the actual precipitation data indicates that the models tend to underestimate climate change variability in the basin area. Despite the limitations, this data impacts on extreme precipitation events. For example, “[c]hanges set was used for model calibration and validation because it is in extreme precipitation projected by models, and thus the impacts a source of actual, observed data, and because the temporal of future changes in extreme precipitation, may be underestimated resolution is daily, thus providing important modeling benefits because models seem to underestimate the observed increase in over monthly data for informing decisions related to hydropower heavy precipitation with warming” (Min, 2011). projects. Analysis of climate change risk requires frequent updates to ensure A stochastic weather generator (WXGEN) produces synthetic that it continues to reflect the most recent data inputs and outputs time series of weather data of infinite length for a location that are available from the international research community. based on the statistical characteristics of observed weather at that location. This long term data captures climate extremes prolonged high and low rainfall periods which is useful in risk HYDROLOGIC FLOW PROJECTIONS assessment for hydrological or agricultural purposes. Some of the limitations in the application of WXGEN as observed by The hydrologic flow projections build on the temperature and Carney et. al., 2008 are (1) the WXGEN does not accurately precipitation projections from the selected GCMs, emissions reproduce the temporal auto-correlation of the annual scenarios, and time horizons. Therefore, the uncertainties precipitation and (2) the WXGEN cannot generate multiple associated with those outputs (discussed above) are carried correlated precipitation inputs. forward through the hydrologic flow modeling effort. 31 Climate Risk and Business: Hydropower, Kafue Gorge Lower The hydrologic model incorporates the impact of increased FINANCIAL ASSESSMENT temperatures in calculating water losses due to evapotranspiration in the runoff generation process. However, because rainfall is the The energy outputs used as inputs to the financial analysis build on most crucial factor in the runoff generation process, it is used the temperature and precipitation, hydrologic flow, and reservoir in the pilot study to evaluate the variability of flows among the (energy) modeling outputs that are associated with limitations and GCMs (see Section 3.3.2, Rainfall Variability and Impacts on Flow uncertainty discussed above. Therefore the financial assessment Projections). In the future, additional analysis of temperature carries forward any limitations and uncertainty associated with (spatial and temporal distribution of minimum and maximum those combined efforts. temperature) and other parameters (such as relative humidity and vapor pressure) could be evaluated. However, such analysis was beyond the scope of this project’s hydrologic modeling effort. NATURAL HAZARD RISK AND SECTOR ASSESSMENTS Additional uncertainty arises from the unknown hydrologic The average annual loss estimates were developed with limited response of the Kafue River Basin to future climate changes. loss data across a limited time period. These limited sets of data While the physics of hydrologic modeling are relatively well do not allow the development of a very comprehensive view of understood and reasonably represented in physically-based potential loss and should be reviewed and refined over time, models, even a well validated model may not faithfully represent either with new loss events or additional modeling. the natural system’s response to future climatic conditions when those conditions differ considerably from historical conditions The hazard frequencies were developed using national datasets which were used for model calibration. In the case of the Kafue and not local, Kafue River Basin data. Local loss data would have River Basin, the GCM projections take the mean seasonal and to be collected in the future to better represent flood frequency mean annual temperatures well outside of the range seen in potential. Also, a detailed hydraulic analysis should be conducted historical observations. for the entire basin to support estimates of potential flood frequency. RESERVOIR (ENERGY) PROJECTIONS ADAPTATION OPTIONS As discussed above, the hydrologic flow projections build on the temperature and precipitation and hydrologic flow modeling that The adaptation goals and strategies are based on the literature, are associated with limitations and uncertainty (as discussed in an understanding of climate change risks based on this study, and the two sections above). The reservoir (energy) modeling uses implementation of an evaluation framework (POSE and STAPLEE). the HEC ResSim model, which builds on the flow projection The adaptation goals and strategies focus on “no regrets” options outputs associated with the hydrologic modeling; therefore, the that appear merited based on current and anticipated conditions uncertainties associated with those outputs are carried forward with or without the exacerbating impacts of climate change. through the HEC ResSim modeling effort. While the study team used available information and professional knowledge to identify and rank appropriate goals and strategies, The use of the HEC ResSim model for evaluation of the Kafue uncertainty remains regarding actual local conditions and needs. River hydropower system is informed by available background Therefore, local stakeholder input would be valuable to ground documents, ZESCO operation data, and other publically available truth the goals, strategies and rankings to ensure they address information regarding power unit locations, sizes, efficiencies local priorities, goals, and needs. This information also would and operational rules. Numerous assumptions were made based assist in reviewing and refining preliminary loss data and cost/ on the available data to establish the model. Where assumptions benefit analysis approaches and data included in this report. differ from actual operational parameters and conditions, limitations and uncertainty are introduced into the modeling effort. Additional information on the data sources, approaches, and assumptions included in model establishment are provided in Appendix A3 of the full report. 32 International Finance Corporation 2121 Pennsylvania Ave. NW Washington, DC 20433 Tel. 1-202-473-1000 www.ifc.org/climatechange The material in this publication is copyrighted. IFC encourages the dissemination of the content for educational purposes. Content from this publication may be used freely without prior permission, provided that clear attribution is given to IFC and that content is not used for commercial purposes. The findings, interpretations, views, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the Executive Directors of the International Finance Corporation or of the International Bank for Reconstruction and Development (the World Bank) or the governments they represent, or those of ZESCO.