Strengthening Agro-Met Services in Bhutan Page|2 Disclaimer: All analyses presented in this report is based on data obtained during the course of this assignment. The accuracy of results is limited to the quality of original data. With the exception of some basic data integrity checks, no systematic quality control of data has been carried out. All climate data was obtained from the Department of Hydro Met Services. The annual agricultural survey data was obtained from the Department of Agriculture in Bhutan. The agricultural statistics presented in this reports for each year is the production and area of each crop for the previous year. As most farmers do not keep actual records, the data gathered is based on a person’s memory and should be treated as best estimate only. Copyright © WB 2016 All rights reserved Page|3 LIST OF ABBREVIATIONS AND ACRONYMS ....................................................................................................... 7 ACKNOWLEDGMENTS ...................................................................................................................................... 9 EXECUTIVE SUMMARY ................................................................................................................................... 10 1 BACKGROUND............................................................................................................................................. 14 1.1 INTRODUCTION .......................................................................................................................................... 14 1.2 OBJECTIVE ................................................................................................................................................... 14 1.3 METHODOLOGY .......................................................................................................................................... 14 1.4 REPORT STRUCTURE ................................................................................................................................... 15 2 AGRICULTURE IN BHUTAN .......................................................................................................................... 16 2.1 GEOGRAPHY ............................................................................................................................................... 16 2.1.1 AGRO-ECOLOGICAL ZONES ................................................................................................................. 16 2.1.2 SOILS AND TOPOGRAPHY .................................................................................................................... 18 2.2 AGRICULTURE ............................................................................................................................................. 19 2.3 MAIN CULTIVATED CROPS .......................................................................................................................... 21 2.4 PRODUCTION RISK ...................................................................................................................................... 30 2.5 KEY MESSAGES ............................................................................................................................................ 31 3 WEATHER AND CLIMATE RISK ..................................................................................................................... 32 3.1 CLIMATOLOGY ............................................................................................................................................ 32 3.2 FUTURE CLIMATE RISKS .............................................................................................................................. 44 3.3 CLIMATE DATA COLLECTION AND RAINFALL NETWORK ............................................................................. 46 3.4 KEY MESSAGES ............................................................................................................................................ 49 4 VULNERABILITY OF AGRICULTURAL PRODUCTION TO CLIMATE .................................................................. 50 4.1 MAJOR CLIMATIC RISKS IN AGRICULTURAL PRODUCTION .......................................................................... 50 4.1.1 PEST AND DISEASES............................................................................................................................. 50 4.1.2 DROUGHT ............................................................................................................................................ 52 4.1.1 FLOODS AND EXTREME EVENTS .......................................................................................................... 52 4.2 SENSITIVITY OF AGRICULTURAL PRODUCTION TO CLIMATE. ...................................................................... 53 4.3 CLIMATE RISK MANAGEMENT: STRATEGIES TO DEAL WITH CLIMATE RISK. ............................................................ 58 4.4 VULNERABILITY ASSESSMENTS ................................................................................................................... 60 4.5 KEY MESSAGES ............................................................................................................................................ 61 5 DEVELOPING A NATIONAL FRAMEWORK FOR CLIMATE SERVICES .............................................................. 63 5.1 INSTITUTIONAL ARRANGEMENTS ........................................................................................................................ 63 5.2.1 Department of Hydro Met Services (DHMS) ....................................................................................... 63 5.2.2 Department of Agriculture .................................................................................................................. 65 5.3 FROM CLIMATE OBSERVATION TO CLIMATE SERVICE DELIVERY ................................................................................. 68 5.3 DEVELOPMENT OF AGRO-MET SERVICES IN BHUTAN .............................................................................................. 69 6 FARMER’S WEATHER AND CLIMATE RELATED INFORMATION NEEDS ......................................................... 73 6.1 METHODOLOGY .......................................................................................................................................... 75 6.2 RESULTS ...................................................................................................................................................... 75 6.2.1 SECTION A - Demographic Information ............................................................................................... 75 Page|4 6.2.2 SECTION B- Information currently used by farmers in agricultural decision making. ......................... 80 6.2.3 SECTION C- Climate and Weather Information Currently Received from Various Sources ................. 81 6.2.4 SECTION D- Climate Information and Services .................................................................................... 82 6.2.5 SECTION E- Constraints in Agricultural Production ............................................................................. 86 6.3 KEY MESSAGES ............................................................................................................................................ 89 7 RECOMMENDATIONS FOR DEVELOPMENT OF AGRO-MET ADVISORY SERVICES IN BHUTAN ...................... 90 8 REFERENCES ................................................................................................................................................ 92 9 APPENDIX – A CLIMATE SURVEY QUESTIONNAIRE ...................................................................................... 97 Page|5 FIGURE 2-1 AGRO-ECOLOGICAL ZONES OF BHUTAN, AND THE DZONGKHAGS. ................................................................17 FIGURE 2-2. CULTIVATED AREA OF MAJOR CEREAL CROPS IN BHUTAN. ..........................................................................24 FIGURE 2-3 PRODUCTION (MT) OF MAJOR CEREALS IN BHUTAN DURING 2004. .............................................................24 FIGURE 2-4. AVERAGE DISTRICT YIELD (KG/ACRE) OF MAJOR CEREALS IN BHUTAN...........................................................25 FIGURE 2-5. AREA GROWN TO VEGETABLES CROPS ACROSS BHUTAN. ...........................................................................26 FIGURE 2-6. PRODUCTION OF MAJOR VEGETABLE CROPS ACROSS BHUTAN.....................................................................27 FIGURE 2-7. YIELD OF VEGETABLES CROPS ACROSS BHUTAN........................................................................................27 FIGURE 2-8. NUMBER OF BEARING FRUIT TREES ACROSS BHUTAN. ...............................................................................29 FIGURE 2-9.PRODUCTION OF FRUIT CROPS ACROSS BHUTAN.......................................................................................29 FIGURE 2-10 YIELD PER BEARING FRUIT TREES ACROSS BHUTAN. .................................................................................30 FIGURE 3-1. AVERAGE MONTHLY DISTRIBUTION OF RAINFALL ACROSS DIFFERENT AGRO-ECOLOGICAL ZONES IN BHUTAN. ALL DATA ARE SHOWN ON SAME SCALE FOR COMPARISON. ......................................................................................33 FIGURE 3-2.THE SEASONAL RAINFALL VARIABILITY FOR KEY LOCATIONS ACROSS BHUTAN. .................................................36 FIGURE 3-3. RELATIONSHIP BETWEEN MONSOON DURATION AND CLIMATE DRIVERS. THE AMOUNT OF RAINFALL (MM) IS INDICATED BY THE SIZE OF THE CIRCLES AND IS POSITIVELY CORRELATED TO MONSOON DURATION (R = 0.0483 D +71.62 2 ; R = 0.43). DAILY RAINFALL DATA FOR PUNAKA 1990-2014.IOD TIME SERIES FROM WWW.JAMSTEC.GO.JP/FRGC/RESEARCH/D1/IOD/DATA/DMI.MONTHLY.TXT. ......................................................40 O FIGURE 3-4 MONTHLY DISTRIBUTION OF MINIMUM AND MAXIMUM TEMPERATURE ( C) FOR PUNAKHA IN CENTRAL BHUTAN .42 FIGURE 3-5 TIME SERIES OF MONTHLY MINIMUM AND MAXIMUM TEMPERATE (1990-2014) IN DIFFERENT AGRO-CLIMATIC ZONES OF BHUTAN. THE STATISTICS OF THE DOTTED TREND LINES ARE IN TABLE 3-4................................................45 FIGURE 3-6 AVAILABILITY OF RAINFALL DATA. MISSING DATA ARE SHOWN AS WHITE (OR YELLOW) COLOURS. ......................47 FIGURE 4-1 DROUGHT ANALYSIS FOR SIX AGRO-CLIMATIC REGIONS IN BHUTAN. .............................................................53 FIGURE 6-1. DISTRIBUTION OF HOUSEHOLD MEMBERS WHO COMPLETED THE QUESTIONNAIRE. .........................................76 FIGURE 6-2. EDUCATIONAL ATTAINMENT OF HOUSEHOLDS. ........................................................................................76 FIGURE 6-3. AGE DISTRIBUTION BY GENDER PER HOUSEHOLD. .....................................................................................77 FIGURE 6-4. INFORMATION CURRENTLY USED TO INFORM DECISION MAKING. 1- SEASONAL CONDITION AND OR INDIGENOUS KNOWLEDGE; 2- TRADITIONAL CROPPING CALENDAR; 3- PERSONAL EXPERIENCE; 4- FOLLOW OTHER SUCCESSFUL FARMERS; 5- ADVICE FROM DEPARTMENT OF AGRICULTURE; 6- CLIMATE AND WEATHER FORECAST; 7 VILLAGE OFFICIALS; 8 ADVICE FROM FARMER’S GROUP. ...............................................................................................................81 FIGURE 6-5.PERCENTAGE OF RESPONDENTS RECEIVING OR NOT RECEIVING VARIOUS CLIMATE INFORMATION. ......................82 FIGURE 6-6. CONSTRAINTS IN AGRICULTURAL PRODUCTION BASED ON CROP TYPE. .........................................................87 TABLE 2-1 CHARACTERISTICS OF VARIOUS AGRO-ECOLOGICAL ZONES OF BHUTAN IN THE CONTEXT OF AGRICULTURAL PRODUCTION. SOURCE: DRAFT NATIONAL BIODIVERSITY STRATEGIES AND ACTION PLAN OF BHUTAN, 2014. ..............18 TABLE 2-2 . A CROPPING CALENDAR OF MAJOR CROPS IN DIFFERENT AGRO-CLIMATIC ZONES. SOURCE (MR CHHIMI RINZIN, MOA). ....................................................................................................................................................22 TABLE 2-3. ANNUAL PRODUCTION, AREA AND YIELD OF MAJOR CEREALS GROWN IN BHUTAN. DATA COMPILED FROM VARIOUS MOAF ANNUAL STATISTIC REPORTS. ..............................................................................................................23 TABLE 2-4. ANNUAL PRODUCTION, AREA AND YIELD OF MAJOR VEGETABLES GROWN IN BHUTAN DATA COMPILED FROM VARIOUS MOAF ANNUAL STATISTIC REPORTS. .................................................................................................26 Page|6 TABLE 2-5. MAJOR FRUIT PRODUCTION IN BHUTAN. DATA COMPILED FROM VARIOUS MOAF ANNUAL STATISTIC REPORTS. ...28 TABLE 3-1 MONTHLY MEAN, MINIMUM AND MAXIMUM RAINFALL (MM) FOR DIFFERENT AGRO-CLIMATIC ZONES OF BHUTAN. 34 TABLE 3-2 AVERAGE ANNUAL RAINFALL (MM) AND RAINFALL VARIABILITY ACROSS DIFFERENT AGRO-CLIMATE ZONES OF BHUTAN. .................................................................................................................. ERROR! BOOKMARK NOT DEFINED. O TABLE 3-3 MINIMUM, AVERAGE AND MAXIMUM TEMPERATURE ( C) DISTRIBUTION ACROSS DIFFERENT AGRO-CLIMATIC ZONES OF BHUTAN. .............................................................................................................................................43 TABLE 3-4 ANNUAL COEFFICIENT OF LINEAR TREND IN MINIMUM AND MAXIMUM TEMPERATURES AND THE CORRESPONDING CHANGE IN TEMPERATURE OC IN DIFFERENT AGRO-CLIMATIC ZONES SINCE 1990. ..................................................43 TABLE 3-5 SUMMARY OF CLIMATE DATA FOR BHUTAN. .............................................................................................48 TABLE 4-1. A DROUGHT HISTORY REPORT FOR BHUR (TOP), DAGANA DZONG (MIDDLE), AND PUNAKHA (BOTTOM). .............54 TABLE 4-2 GENERAL SENSITIVITIES OF BHUTAN'S AGRICULTURAL SECTOR. SOURCE: (NEC 2000) ......................................56 TABLE 4-3 THE CLIMATE SENSITIVITY OF BHUTAN'S KEY AGRICULTURAL PRODUCE............................................................57 TABLE 4-4. A SELECTION OF STRATEGIES THAT FARMERS AND INDUSTRY USE TO ADDRESS CLIMATE INFLUENCES ON BHUTAN'S KEY AGRICULTURAL PRODUCTION. .......................................................................................................................59 TABLE 6-1. DISTRIBUTION OF CLIMATE SURVEY IN VARIOUS DISTRICTS (DZONGKHAGS) AND SUB-DISTRICTS (GEWOGS) OF BHUTAN...................................................................................................................................................74 TABLE 6-2. HOUSEHOLD POSSESSIONS AND FARM CHARACTERISTICS. ...........................................................................78 TABLE 6-3. FARM SIZE BY DIFFERENT LAND UTILIZATION. ............................................................................................78 TABLE 6-4. TYPE OF CLIMATE INFORMATION CURRENTLY RECEIVED BY USERS, FREQUENCY, SOURCE, LEAD TIME AND HOW THIS INFORMATION HAS BEEN USED IN DECISION MAKING. THE PERCENTAGE REPOSES ARE THOSE WHO INDICATED THEY HAVE RECEIVED INFORMATION, NOT THE WHOLE SAMPLE. RESULTS ARE COLOR CODED FOR EASIER CROSS REFERENCING........84 TABLE 6-5 TYPE OF CLIMATE INFORMATION CURRENTLY RECEIVED BY USERS, FREQUENCY, SOURCE, LEAD TIME AND HOW THIS INFORMATION HAS BEEN USED IN DECISION MAKING. THE PERCENTAGE REPOSES ARE THOSE WHO INDICATED THEY HAVE RECEIVED INFORMATION, NOT THE WHOLE SAMPLE. . RESULTS ARE COLOR CODED FOR EASIER CROSS REFERENCING. .....85 TABLE 6-6. MAJOR CLIMATE RELATED EVENTS EXPERIENCE BY HOUSEHOLD IN THE PAST 5 YEARS. .......................................88 List of abbreviations and acronyms ADB Asian Development Bank AWS Automatic Weather Station CPT Climate Prediction Tools CV Coefficient of Variation DEM Digital Elevation Model DHMS Department of Hydro Met Services DOA Department of Agriculture DAO District Agricultural Officer DOI Department of Irrigation DDM Department of Disaster Management DPNet Disaster Preparedness Network DRR Disaster Risk Reduction ECHAM European Centre/Hamburg Model ENSO El Niño Southern Oscillation ECMWF European Centre for Medium-Range Weather Forecast EPC Environment Protection Council EU European Union FAO Food and Agriculture Organization GCM General Circulation Model GDP Gross Domestic Products GEF Global Environment Fund GLOF Glacial Lake Outburst Flood GTS Global Telecommunication System FGD Focus Group Discussion ITCZ Inter Tropical Convergence Zone ICACS International Centre For Applied Climate Sciences IOD Indian Ocean Dipole IRI International Research Institute for Climate and Society ISM Indian Summer Monsoon MOEA Ministry Of Economic Affairs MOU Memorandum of understanding MOAF Ministry Of Agriculture And Forest NAPA National Adaptation Program of Action NPPC National Plant Protection Centre SMS Short Message Service SD Standard Deviation SOI Southern Oscillation Index SST Sea Surface Temperature SSTA Sea Surface Temperature Anomaly USQ University Of Southern Queensland Page|8 UNDP United Nations Development Program WB World Bank WMO World Meteorological Organisation Page|9 ACKNOWLEDGMENTS The report Strengthening Agro-Met Services in Bhutan was prepared in collaboration between the Departments of Agriculture and Hydromet Services of the Royal Government of Bhutan and the World Bank. This activity is a part of a broader regional program “Hydromet Modernization, Disaster Risk Management and Climate Resilience” for which the overall objective is to strengthen disaster preparedness and hydromet services in South Asia. We are sincerely grateful to Mr Karma Tsering, Director; Mr Phuntsho Namgyal, Chief; and Mr Tayba Buddha Tamang, Senior Meteorologist from the Department of Hydro Met Services (MoEA) for helpful discussions and the provision of climate data; Mr Chhimi Rinzin, Chief Agriculture Officer, Department of Agriculture, MoAF; Ms Yeshey Dema, Program Director, NPPC (National Plant Protection Centre); Dr Thinlay (Plant Protection Specialist), and Ms Tenzin Wangmo of the National Environment Commission Secretariat. Many other staff from various agencies gave their generous time to provide information, advice and references and their support and cooperation is greatly appreciated. (Names of district officials and other staff to be added). Special thanks to Mr Peter Davis, Research Fellow at ICACS for helping with data preparation and producing GIS outputs and Dr Allyson Williams with editing and review of the report. This report was prepared by a team including Poonam Pillai, Senior Environment Specialist (Task Team Leader); Dechen Tshering, Disaster Risk Management Specialist (co-Task Team Leader) of the Disaster Risk & Climate Change Unit, South Asia region, Yahya Abawi (Lead consultant) and Dr Sonam Wang and his team (Wang Research and Consultancy). Peer reviewers include…(names to be added) We are grateful to GFDRR including the Government of Japan and the European Union for their generous funding. P a g e | 10 EXECUTIVE SUMMARY (will be revised) Agriculture is the dominant sector in Bhutan, providing livelihood, income and employment to approximately 55% of the population. The majority of famers are subsistence farmers with average land holdings ranging from 1 to 4 acres. The farming community is the most vulnerable group to climate impacts as farm productions is predominately dryland and influenced by increasing climate variability and climate extremes. Despite the vulnerability of the agricultural sector to climate risks, there is no systematic assessment of how delivery of agro-weather information or services to farmers can help mitigate climate related risks. Understanding farmer’s needs, knowledge and practices is essential to ensuring effective agro-climate services and that products and research outputs meet user’s requirements. Objective: Main objective of this report is to ……add provide recommendations that will result in improved farmer and sector-wide resilience to climate variability and change, and improved agro-meteorological services in Bhutan. The specific objectives of this report are to;  undertake a baseline assessment of climate related risks facing farmers in different agro ecological zones in Bhutan;  assess institutional and organizational processes at the national and sub-national level for delivering climate and weather related information to farmers;  assess farmer’s information needs and priorities in different agro-climatic zones;  provide recommendations on how agro-weather services can be strengthened and institutionalized in Bhutan. Methodology (to be added) Process of Preparation (to be added) P a g e | 11 KEY FINDINGS  Although climate service delivery is a priority within the national agenda, Bhutan does not have a National Framework for Climate Services. Current capability of DHMS is limited to observation, storage and maintenance of hydro-meteorological data and does not extend to the provision of climate services. According to the WMO Global Framework for Climate Services classification Bhutan meets most but not all of the requirements for Basic Level of services (Category I).  Currently there is no capacity within DHMS to provide climate forecasts. DHMS only issues 24-hour weather forecast and annual forecast of monsoon sourced from international agencies (IRI). Developing an operational seasonal climate forecast is a high priority for agro-met service delivery. However, this is compounded by a lack of research capacity, limited climate data, complex topography and the lack of an identifiable climate driver for Bhutan.  An extensive literature review revealed only a few studies assessing drivers of rainfall variability in Bhutan due to poor data availability. Indeed, the sparse and short observational climate datasets do not meet the standards set by WMO for statistically meaningful analyses. While there are similarities in climatology and geography with other Himalayan regions and north east India, the differences are great enough to warrant caution when extending climate risk analyses from other regions to Bhutan. However, the regional extent and strength of ENSO relationships with rainfall across north-east India indicate significant seasonal and intra-seasonal forecasting potential in Bhutan for many aspects of identified climate risks. Further research is required using gridded or proxy data.  Currently in Bhutan due to lack of observational network and the complicated topography of the Himalayas, full climate prediction capability is yet to be developed and may take time to do so. However given the priority for agricultural climate service delivery it is strongly recommended that climate products and forecasts from Global Producing Centres and Regional Climate Centres be used to fill the gap until a specific seasonal climate forecasting system is developed for Bhutan. For that to happen there needs to be strong capacity building for staff in DMHS to understand forecasts and limitations and be able to communicate these to the public.  A number of agro-climatic products such as drought analysis, extreme event, frost prediction, heat stress index, trend analysis and general climatology information can be developed using the current data. This will be extremely useful for agricultural applications. However, with the lack of a clear mandate for agro-met service delivery it P a g e | 12 is unclear which organization (DHMS or DoA) is responsible for these developments. A clear mandate and regulatory framework is required to clarify the role of these departments in the delivery of agro-met services.  Effective climate services need strong involvement by stakeholders from various disciplines. The user needs to engage information providers to understand what climate information is available, how to interpret it correctly as well as understanding its underlying assumptions and limitations. Current partnerships are non-existent or very weak in Bhutan. Strong partnerships need to be pursued through a formal MoU between DHMS and DoA and other sectors where needed.  Analysis of agricultural production data highlighted the significant spatial and temporal variation in the production of all crops. The risk of agricultural production varies between agro-ecological zones. However there is little information available on the specific climate sensitivities of the various production systems in Bhutan and how this varies across agro-ecological zones. A vulnerability assessment of the agricultural sector to climate variability and climate change is a high priority.  Research activities in the department of agriculture are mainly carried out through the four Research and Development Centres located in strategic locations with specific foci. However most research is predominately related to field trials and crop improvement programs. There is little analytical research (crop modelling or Decision Support Modelling) to examine the effect of climate (climate variability and climate change) on production risk of crops across different agro-ecological zones. This capability needs to be developed in partnership with international researchers. RECOMMENDATIONS The primary objective of providing agro-meteorological services is to facilitate the Government’s capacity to manage climate risk and increase agricultural productivity. This goal is currently challenged by the overarching restriction of limited resources and institutional capacity for seasonal and long term climate risk assessments and subsequent development of climate risk management policies and programs. To achieve this goal the following recommendations are made to strengthen agro-meteorological services in Bhutan: DEVELOPING A NATIONAL FRAMEWORK FOR CLIMATE SERVICES IN BHUTAN The climate survey conducted as part of this study clearly highlights a significant demand from a range of sectors for climate services. Whilst observations and monitoring networks are in place and being developed, there is an urgent need through collaboration with other sector agencies for developing research capacity for climate services, development of a historical climate database and real time observation network, developing climate services information systems, and a user interface platform. P a g e | 13 DEVELOPING PARTNERSHIPS BETWEEN DHMS AND DOA AND OTHER SECTORS Effective climate services need strong involvement by stakeholders from various disciplines. A Memorandum of Understanding will provide a formal channel for engagement to identify climate needs of various sectors and to develop products specific to the needs of these sectors (climate service delivery). To achieve this it is recommended that an externally led visioning process should be conducted to ensure that there is mutual understanding and alignment of objectives across organizations (DoA and DHMS). This process would pull together key personnel across the board to ensure that issues raised in this report are understood by all and a collective goal is established whilst at the same time maintaining the integrity and value of each of the parts within the whole. This will ensure efficiency and establish a shared working vision as the basis for effective collaboration and communication. STRENGTHEN CAPABILITY AND HUMAN RESOURCE CAPACITY through capacity building (recruitment of staff) and capacity development (training of existing staff) in both DHMS and Department of Agriculture. In DHMS capacity needs to be improved In the areas of weather and climate forecasting, numerical weather prediction, interpretation of climate products including limitations and uncertainties, identification of user needs, development of sector specific products, and partnership creation and communication. A comprehensive capacity building initiative is also needed within the DoA in the following key areas:  Development and training in the use of climate concepts for the agricultural sector  Development of analytical and crop simulation capability  Vulnerability assessments of agricultural production due to impacts from climate variability and climate change. Before initiating any capacity building exercise, a training needs assessment should be undertaken to ensure high priority areas are targeted first. P a g e | 14 1 BACKGROUND 1.1 INTRODUCTION Agriculture is the dominant sector in Bhutan, providing livelihood, income and employment to approximately 55% of the population. The majority of famers are subsistence farmers with average land holdings ranging from 1 to 4 acres. The farming community is the most vulnerable group to climate impacts as farm productions are predominately dryland and influenced by increasing climate variability and climate extremes. Despite the vulnerability of the agricultural sector to climate risks, there is no systematic assessment of how delivery of agro-weather information or services to farmers can help mitigate climate related risks. Understanding farmer’s needs, knowledge and practices is essential to ensuring agro-climate services and that the products and research outputs meet user’s requirements. 1.2 OBJECTIVE The goal of the TA is to provide recommendations that will assist to improve farmer’s resilience to climate variability and climate change, and assist to strengthen the Government’s capacity for delivering agro-meteorological services in Bhutan. The specific objectives of this study are to:  undertake a baseline assessment of climate related risks facing farmers in different agro ecological zones in Bhutan;  assess institutional and organizational processes at the national and sub-national level for delivering climate and weather related information to farmers;  assess farmer’s information needs and priorities in different agro-climatic zones;  provide recommendations on how agro-weather services can be strengthened and institutionalized in Bhutan. 1.3 METHODOLOGY The following tasks were carried out using the following approach;  an extensive literature review of scientific literature and technical documents,  meetings with government and other relevant stakeholders such as key informants, farmer group leaders, community representatives and development partners,  design and analyze the results of a survey assessing the climate information needs of farmers. The main counterpart agencies contacted were the Department of Hydro-Met Services (DHMS) and the Ministry of Agriculture and Forests (MoAF). In addition several agencies including the Department of Geology and Mines, National Soil Services Centre (MoAF), the Department of P a g e | 15 Disaster Management, Plant Protection Centre (MoAF) and the National Environment Commission (NEC) were consulted. 1.4 REPORT STRUCTURE This report is organized into seven chapters: Chapter 1 introduces the study and its objectives. Chapter 2 provides a background to agriculture in Bhutan: the climate, soil, topography and agro-climatic characteristics. The agriculture sector is summarized, identifying the main crops and the level of productivity in each Dzongkhag. This includes a brief review of major constraints to agricultural production. Chapter 3 details the climate risk to Bhutan in both a temporal and spatial context. Drivers of climate variability are discussed with the intent of identifying possible forecasting skill. Chapter 4 reviews the vulnerability of groups across the agricultural sector who are most vulnerable to weather/climate events in Bhutan. Coping and adaptation strategies that farmers and other stakeholders use to manage weather and climate related risks are identified. Chapter 5 discusses institutional and policy arrangements for effective delivery of agro-met services including recommendation for capacity development and research. Chapter 6 presents a climate survey to assess farmer’s information needs and priorities including the analysis of results and recommendations. Chapter 7 summarizes the key recommendation from this report. P a g e | 16 2 AGRICULTURE IN BHUTAN This chapter provides an overview of the agricultural sector of Bhutan. This will offer the basis for climate risk analysis of the agricultural sector in later Chapters. Bhutan is a small landlocked country situated in the south-eastern Himalayas between latitudes 260 40’ and 280 20’ N and longitude 880 45’ and 920 7’ E and is surrounded by the Tibetan Plateau in the north, and by the alluvial plain of the Brahmaputra to the south, Arunachal Pradesh in the east and the Darjeeling and the Sikkim Himalaya in the west. It covers a total area of 47000 square kilometers with an estimated population of 735,000. The terrain is among the most rugged and mountainous in the world. The topography varies from an elevation of about 100 meters above sea level in the south to more than 7,500 meters above sea level in the north. The main features of the Bhutan landscape are aligned roughly north-south (in contrast to the east-west alignment of the Central Himalayan topography) and the country is divided into three geographic regions: eastern, central and western, with six, seven and seven administrative districts (Dzongkhag) in each region respectively. Agriculture in Bhutan has a dominant role in the economy of the country. Approximately 70% of the population of Bhutan are involved in the agricultural sector and 56% are farmers. Bhutan’s GDP per Capita has grown from $US2337 in 2000 to $US7075 in 2012 (World Bank 2013). Although the contribution of the agricultural sector to GDP has declined from about 35.9% in 2000 to about 17.1% in 2013, agriculture remains the primary source of livelihood for the majority of the population. Bhutan’s dependence on the climate-sensitive agricultural production systems makes it vulnerable to climate risk; a threat from both climate variability and climate change. 2.1 GEOGRAPHY 2.1.1 AGRO-ECOLOGICAL ZONES The climate is dominated by the Indian Summer Monsoon (ISM) which moves north from the Bay of Bengal. This brings approximately 75% of annual rainfall to Bhutan between June and September. In winter, the climate is cool and dry as it is significantly influenced by the Tibetan high pressure system and not as affected by the westerlies that drive winter rain in the Western Himalayas (Mani 2003, Norbu et al.2003a). The range in altitude and topography produces a wide range of climatic conditions. The climate ranges from subtropical, through temperate and alpine, to arctic, all within 100 km. Annual mean temperatures vary from above 20oC in the piedmont in the south to below zero in the High Himalaya. Temperature is most influenced by altitude with a decrease of 0.5-0.6oC per 100m of altitude (Eguchi 1991). A detailed discussion of spatial and temporal variations in rainfall and temperature is given in Chapter 3. P a g e | 17 Figure 2-1 Agro-ecological zones of Bhutan, and the Dzongkhags. Bhutan is divided into six major agro-climatic zones ( alpine, cool temperate, warm temperate, dry sub-tropical, humid sub-tropical and wet sub-tropical), each with similar climatic conditions that are influenced by latitude, elevation, temperature , seasonality and rainfall amount that determine their ability to support various form of agriculture as shown in Figure 2-1 and Table 2- 1. Annual rainfall varies from less than 650 mm in the Alpine region to more than 5500 mm in the Wet Sub-tropical region. P a g e | 18 Table 2-1 Characteristics of various Agro-ecological zones of Bhutan in the context of agricultural production. Source: Draft National Biodiversity Strategies and Action Plan of Bhutan, 2014. Altitude Annual Rainfall Farming Systems, major crops and Agro-Ecological Zone (masl) (mm) agricultural produce. Alpine 3600-4600 <650 Semi-nomadic people, yak herding, dairy products, barley, buckwheat, mustard and vegetables. Cool Temperate 2600-3600 650-850 Yaks, cattle, sheep & horses, dairy products, barley, wheat & potatoes on dryland, buckwheat & mustard under shifting cultivation. Warm Temperate 1800-2600 650-850 Rice on irrigated land, double cropped with wheat and mustard, barley and potatoes on dryland, temperate fruit trees, vegetables, cattle for draft and manure, some machinery and fertilizers used. Dry Sub-tropical 1200-1800 850-1,200 Maize, rice, millet, pulses, fruit trees and vegetables, wild lemon grass, cattle, pigs and poultry. Humid Sub-tropical 600-1200 1200-2500 Irrigated rice rotated with mustard, wheat, pulses and vegetables, tropical fruit trees. Wet Sub-tropical 150-600 2500-5500 As for the humid zones - irrigated rice rotated with mustard, wheat, pulses and vegetables, tropical fruit trees. 2.1.2 SOILS AND TOPOGRAPHY Agricultural production is significantly constrained by the difficult terrain that reduces the amount of available land for mechanized farming. Approximately 13% of Bhutan’s land is arable (FAO-2015). Apart from steep terrain, other factors that affect agricultural productivity are poor soil quality and associated low levels of nitrogen and phosphorous content. P a g e | 19 Information on Bhutan’s soil is very limited despite significant demand for soil information (especially inherent fertility) from planners, farmers, NGO’s and agricultural specialists. The only published information is the Land Cover Atlas of Bhutan (1:250,000, Ministry of Agriculture and Forests, Royal Government of Bhutan, 2011), prepared by the National Soil Services Centre (NSSC) and Policy and Planning Division, Ministry of Agriculture and Forests. Geologically, most of Bhutan consists of crystalline sheets with large masses of tertiary granite intrusions in the north. Approximately 27% of Bhutan is classified as either cambisols or fluvisols. Cambisols are most common in the medium-altitude zone, while fluvisols mostly occur in the southern belt. Less fertile acrisols, ferrasols and podzols are estimated to cover 45% of the country. About 21% of the soil-covered area suffers from shallow depth with mostly lithosol occurring on steep slopes. An overall typical feature of Bhutanese soils is their high levels of variability over a short distance and regolith heterogeneity. The most extensive arable areas in Bhutan are the moderately graded lower hill slopes (the “Inner Valleys”) at about 2500masl. These valleys are upstream of major ‘Knick points’ in large rivers where there are concave sections with floors up to 1km wide of alluvial terraces and soils (Baillie and Norbu 2000). These regions have been settled for long periods of time and cultivated, and hence are considered the demographic, economic and cultural centre of Bhutan (Baillie and Norbu 2004). The level of land degradation is generally low. Approximately 10% of Bhutan’s arable land is subject to some degradation (Young 1994). Norbu et al.(2003b) provide the first reliable account of the different types of land degradation within the country with special attention to their occurrence, causes and interactions. In situ degradation due to soil organic matter depletion is identified as the main degradation process. Generally the soil fertility in Bhutan is on the decline, due to inadequate input of organic matter and fertilizer, soil erosion and limited adoption of crop rotation systems such as legumes. Some conservation measures such as contouring have helped reduce the declining soil fertility. There are no records of famine in Bhutan (Banskota et al.1992). This most likely reflects high levels of indigenous knowledge that have enabled effective small-scale subsistence farming. The techniques utilised in indigenous land use strategies (for example shifting cultivation, crop rotation, intercropping, contour ploughing, preparation of manure and its regular application, and low plant population densities) are less effective as the population grows because the significant fallow periods of 15-20 years that are required to maintain its sustainability are no longer achievable. 2.2 AGRICULTURE Among the agricultural lands in Bhutan, an estimated 28% are wetland or irrigated (Chhuzhing), 60% are dryland (Kamzhing) 6.5% are used for orchards, and 3.7% are areca nut and cardamom plantations (source: Statistical Yearbook of Bhutan 2015, National Statistics Bureau). In 1996 shifting cultivation (tsheri) accounted for 28% of agricultural, however this practice is now P a g e | 20 discouraged by the government (source: Statistical Yearbook of Bhutan 2005, National Statistics Bureau). Most farming is subsistence with some integration of crops, forests and livestock. Until the mid- 1980s most agriculture was subsistence, however, more recently there has been a trend to commercial farm cash crops, some of which are exported such as oranges, areca nut and cardamom in the subtropical south of the country, apples in the more temperate southern regions. Despite the high proportion of the population being involved in agriculture, Bhutan is not self- sufficient in terms of food production. Food insecurity persists mostly in rural areas, especially in the eastern and southern parts of the country. Current coping mechanisms against food insecurity include, off farm activities, sale of vegetables, fruits and nuts, labor exchange, sale of livestock products, cash remittance from employed family members and borrowing from neighbors. In addition, Bhutan has been a net food importer, particularly of grains. The agricultural sector in Bhutan faces significant challenges including: 1. Limited agricultural land. Only 2.9% of arable land is currently used for agriculture of which 31% are on slopes with a gradient of more than 50%. This is further compounded by the strong policy of environmental conservation with 70% of the land under permanent forest cover, thus limiting agricultural expansion. 2. High vulnerability to climate hazards and natural disasters. Bhutan is highly vulnerable to natural hazards such as floods, landslides, cyclone and droughts. Despite high rainfall and an abundance of water resources, most agricultural production is rain-fed making it vulnerable to climate variability and climate extremes. 3. Crop damage from wild animals. In a survey conducted by the Department of Agriculture in 2012, about 67% of farmers identified damage by wildlife as a major constraint to agricultural productivity. In the climate survey of the 1205 households conducted in this study (Chapter 6) damage from wide life was ranked 4.6 on a scale of 1 to 5 ( 1- least important to 5- most important), well ahead of climate, soil, water, labor and other constraints. 4. Shortage of labour. The increasing shortage of farm labour, due to rural to urban migration, coupled with competition from growing imports of cheaper food is a constant constraint impacting on internal food production and agricultural development. Almost 61% of households surveyed, identified labour shortage as a major constraint (MoAF, 2012) to agricultural productivity. P a g e | 21 5. Access to Markets. Poor road network and rugged terrain, increases the cost of commodities and is a significant barrier to producing competitive products particularly for the international markets. 6. Lack of Irrigation Infrastructure. The main rivers provide water mainly for hydropower generation, tourism, recreation and ecology, with sparse use for irrigation. Tributaries and streams are the main source for most water users, with headwater streams used for irrigation and water supply. Agriculture accounts for around 90 percent of consumptive water demand, which is mostly through traditional irrigation conveyance systems that are small and gravity-fed with few properly engineered headwork or feeder canals. 7. Climate Change. Climate projections for Bhutan suggest that the mean annual temperature in Bhutan will likely increase by ~0.8oC by 2039 and the mean annual precipitation will likely increase by ~6% over the same period (Second National Communication to the UNFCCC - National Environmental Commission). Agriculture is already vulnerable due to increases in temperature and extreme events (e.g. cyclones Aila, and Phailin), flash floods, hailstorms, windstorms, droughts, pests and diseases. Although increasing temperature and CO2 due to climate change may have some positive impacts on some crops (e.g. potatoes), others like rice and maize will see a decline in yield in the short term. A 1 oC increase in minimum temperature is likely to reduce rice yield by some 10%. The positive impact of climate change may be offset by the increasing incidence of pest and diseases and extreme variability in rainfall. 2.3 MAIN CULTIVATED CROPS The major crops grown in Bhutan are Rice, Maize, Wheat, Millet, Barley, Potatoes, Mustard and Legumes. These crops are grown in the different agro-climatic zones of Bhutan often in rotation through the year. A cropping calendar of major crops in different agro-climatic zones is shown in Table 2-2. P a g e | 22 Table 2-2 . Cropping calendar of major crops in different agro-climatic zones. Source (Chhimi Rinzin, DoA) Crops Cropping Sequence Months Jan Feb March April May Jun July August Sep Oct Nov Dec Cool temperate (2600 m to 3600 m asl) Rice Maize Maize- Fallow Sowing Harvesting Wheat Potato-Wheat Harvesting Sowing Winter Wheat- Harvesting Sowing Buckwheat Barley Potato-Wheat Harvesting Sowing Millet Mustard Potato-Mustard Sowing Harvetsing Legumes Potato Potao- Fallow Planting Harvesting Marketing Marketing Marketing Warm temperate (1800 m to 2600 m amsl) Rice Rice- Fallow Nursery Nursery T/planting T/planting Harvesting Rice-Wheat Fodder Nursery Nursery T/planting T/planting Harvesting Harvesting Rice- Oat Nursery Nursery T/planting T/planting Harvesting Harvesting Maize Maize- Fallow Sowing Harvesting Harvesting Maize+bean (relay) Sowing Harvesting Harvesting Wheat Maize -Wheat Harvesting Sowing Barley Maize- Barley Harvesting Sowing Millet Mustard Potato-Mustard Harvesting Harvesting Sowing Legumes Potato Potato-Mustard Planting Planting Weeding Harvesting Marketing Marketing Marketing Potato- Turnip Planting Planting Weeding Harvesting Marketing Marketing Marketing Dry Sub-tropical (1200 m to 1800 m asl) Rice Rice -Wheat Nursery Nursery Nursery T/planting T/planting T/planting Harvesting Harvesting Rice- Mustard Nursery Nursery Nursery T/planting T/planting T/planting Harvesting Harvesting Rice- Vegetables Nursery Nursery Nursery T/planting T/planting T/planting Harvesting Harvesting Maize Maize - legumes Sowing Sowing Weeding Harvesting Harvesting Wheat Rice-Wheat Harvesting Sowing Barley Maize - Barley Harvesting Sowing Millet Millet- Fallow Nurserry June Harvesting Mustard Maize-Mustard Harvesting Harvesting Sowing Sowing Weeding Legumes Maize - Legumes Sowing Harvesting Potato Potato- Wheat Planting Planting Weeding Harvesting Harvesting Marketing Marketing Marketing Potato- Buckwheat Planting Planting Weeding Harvesting Harvesting Marketing Marketing Marketing Potato-Rice (wetland) Planting Harvesting Humid sub-tropical (600 m to 1200 m asl) Rice--Fallow Nursery Nursery& Harvesting Harvesting Rice T/planting T/planting Rice (first crop) Rice-Rice Nursery Nursery T/planting Harvesting Harvesting Maize Maize - Millet Sowing Sowing Weeding Harvesting Harvesting Maize second crop ( Maize - Maize Sowing Harvesting dryland) Maize in wetland Maize -Rice (Spring maize) Sowing Sowing Weeding Harvesting Wheat Rice-Wheat Harvesting Sowing Barley Maize- Barley Harvesting Sowing Millet Maize- Millet Nursery T/planting Harvesting Harvesting Mustard Maize- Mustard (dryland) Harvesting Harvesting Sowing Sowing Rice - Mustard (wetland) Harvesting Harvesting Sowing Sowing Legumes Maize- Rajma beans Sowing Harvesting Potato Potato- Maize Harvesting Marketing Planting Wet sub-tropical (150 m to 600 m asl) Rice Rice- Fallow Nurserry T/planting Harvesting Harvesting Rice-Rice Nursery Nursery & T/planting Harvesting T/planting Rice (first crop) Maize (dryland) Maize - Millet Sowing Sowing Weeding Harvesting Harvesting Maize in wetland Maize -Rice (Spring maize) Sowing Sowing Weeding Harvesting Wheat Rice-Wheat Harvesting Sowing Barley Millet Maize- Millet Nursery T/planting Harvesting Harvesting Mustard Maize-Mustard ( dryland) Rice - Mustard (wetland) Harvesting Harvesting Sowing Sowing Legumes Maize- Urd beans Sowing Harvesting Harvesting Potato Harvesting Potato- Rice (wetland) Planting P a g e | 23 Cereals The major cereals grown in Bhutan are Rice (Paddy), Maize, Wheat, Buckwheat and Millet. The staple food of the Bhutanese people is rice, followed by maize. Total rice production is approximately 71,630 t grown over an area of 56677 acres. The average rice yield in Bhutan is 1282 kg/acre with significant variation across the region (Figure 2-4). Rice is grown mainly in the western region (Thimphu, Paro, Punakha and Wangdue districts) and the southern region (Sarpang, Tsirang and Samtse districts). The highest yielding districts are Paro, Thimphu Punakha and Wangdue averaging (1992 Kg/ ha) and the lowest yield are in Pemagatshel, Samdrupjongkhar, Zhemgang and Gasa averaging 1132 kg/acre. Rice production has been relatively constant over the past decade with an annual coefficient of variation of 10% (Table 2-3). While yield has increased over this period, this has been offset by a slight reduction in the cultivated area. Rice production on a commercial scale is limited largely due to a shortage of arable land and farm labour, low cropping intensity, inadequate irrigation and crop losses to pests, especially wild animals. Domestic production and supply is less than the rising demand. At present self- sufficiency in rice is about 48%. From the total supply of about 75,229 t of rice in the country in 2013, a significant proportion of domestic demand was met by imports from India. Table 2-3. Annual production, area and yield of major cereals grown in Bhutan. Data compiled from various MoAF annual statistic reports. Major Cereals Production in Bhutan Year 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Mean SD CV Paddy Area (Acres) 46586.0 62458.0 67568.0 67565.0 47834.7 58607.0 56375.3 59167.0 52252.1 48361.0 56677.4 7812.7 0.1 Production (Metric Tonnes) 54326.0 67607.0 74380.0 74439.0 77313.0 65766.0 71636.6 77576.0 78014.0 75229.0 71628.7 7354.6 0.1 Yield (Kg/acre) 1166.1 1082.4 1100.8 1101.7 1616.3 1122.2 1270.7 1311.1 1493.0 1555.6 1282.0 204.4 0.2 Maize Area (Acres) 53939.0 75859.0 75413.0 71002.1 67278.9 70603.0 61475.9 66021.0 63488.2 58338.1 66341.8 7175.8 0.1 Production (Metric Tonnes) 90568.0 93969.0 71062.0 61792.1 66779.8 61161.0 57666.3 73643.0 73024.4 75716.7 72538.2 12006.2 0.2 Yield (Kg/acre) 1679.1 1238.7 942.3 870.3 992.6 866.3 938.0 1115.4 1150.2 1297.9 1109.1 250.7 0.2 Wheat Area (Acres) 7585.0 21826.0 17515.0 16981.0 7877.7 7708.0 5566.6 5447.0 5539.6 5440.0 10148.6 6160.0 0.6 Production (Metric Tonnes) 4192.0 11248.0 9586.0 8879.0 5648.0 3678.0 4874.3 5816.0 5038.1 4285.0 6324.4 2616.0 0.4 Yield (Kg/acre) 552.7 515.3 547.3 522.9 717.0 477.2 875.6 1067.7 909.5 787.7 697.3 204.9 0.3 Millet Area (Acres) 7326.0 16986.0 20977.0 21485.0 8696.3 10124.0 8454.0 9162.0 6462.1 5054.5 11472.7 6041.0 0.5 Production (Metric Tonnes) 2367.0 6953.0 8787.0 8776.0 5024.0 4231.0 4066.3 5881.0 3965.1 2949.9 5300.0 2259.8 0.4 Yield (Kg/acre) 323.1 409.3 418.9 408.5 577.7 417.9 481.0 641.9 613.6 583.6 487.6 108.6 0.2 Buckwheat Area (Acres) 6288.0 16062.0 21413.0 21701.0 8493.0 9524.0 7503.8 9184.0 6851.5 6590.0 11361.0 6056.3 0.5 Production (Metric Tonnes) 2510.0 7001.0 9353.0 8104.0 5138.0 3858.0 3950.5 6037.0 4303.1 3641.0 5389.6 2188.4 0.4 Yield (Kg/acre) 399.2 435.9 436.8 373.4 605.0 405.1 526.5 657.3 628.1 552.5 502.0 104.8 0.2 P a g e | 24 Figure 2-2. Cultivated area of major cereal crops in Bhutan. Figure 2-3 Production (Mt) of major cereals in Bhutan during 2004. P a g e | 25 Figure 2-4. Average District yield (kg/acre) of major cereals in Bhutan. Vegetables A variety of vegetables are cultivated in Bhutan. Most are produced for household consumption on a subsistence basis but a few are also sold in the market. The major vegetables crops are Potatoes, Chillies, Cabbage, Turnip and Radish. Annual production, area and yield of major vegetables for the past ten years in Bhutan is shown in Table 2-4. Major vegetable growing regions by production and area are shown in Figure 2-5, Figure 2-6, and Figure 2-7 respectively. Potatoes are the most important cash crop in Bhutan. The major potato production regions of Bhutan are in Wangdi, Mongar, Tasigan, Chhukha and Samdrup districts. Around 30000 households, are dependent on potato production for a significant portion of their livelihood. Most potato farmers meet their demand of annual supply of rice and household needs using the cash earned from selling potatoes. The cash crop thus has one of the important influences on the socio-economic conditions of the poor rural households of the country. Production of vegetable crops remain highly variable with annual coefficient of variation ranging from 20 percent (potatoes) to 50% (Turnip). P a g e | 26 Table 2-4. Annual production, area and yield of major vegetables grown in Bhutan. Data compiled from various MOAF annual statistic reports. Major Vegetable Production in Bhutan Year 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Mean SD CV Chili Area (Acres) 3060.0 5639.0 5971.0 5288.0 9454.4 5684.0 6984.6 6044.0 7797.3 5170.0 6109.2 1702.5 0.3 Production (Metric Tonnes) 4454.0 10444.0 11606.0 8368.0 7312.5 8886.0 6696.4 5668.0 7725.5 8320.0 7948.0 2114.3 0.3 Yield (Kg/acre) 1455.6 1852.1 1943.7 1582.5 773.5 1563.3 958.7 937.8 990.8 1609.3 1366.7 416.9 0.3 Cabbage Area (Acres) 704.0 1575.0 2024.0 2151.4 1189.0 863.4 1472.0 2676.9 2427.0 1675.9 689.8 0.4 Production (Metric Tonnes) 1889.0 3345.0 4298.0 4485.3 1552.9 1775.0 1299.1 2998.0 3413.0 3962.0 2901.7 1189.4 0.4 Yield (Kg/acre) 2683.2 2123.8 2123.5 2084.9 1492.9 1504.6 2036.7 1275.0 1632.5 1884.1 439.9 0.2 Radish Area (Acres) 2381.0 4711.0 4014.0 3651.2 3166.0 2739.7 2934.0 3371.0 806.1 0.2 Production (Metric Tonnes) 5628.0 12658.0 10218.0 10539.3 5977.1 5674.0 3882.1 5244.5 4534.0 7150.6 3127.7 0.4 Yield (Kg/acre) 2363.7 2686.9 2545.6 2886.5 1792.2 1417.0 1545.3 2176.7 585.8 0.3 Turnip Area (Acres) 962.0 2028.0 2112.0 2729.0 2139.0 1080.1 1616.2 1809.5 630.0 0.3 Production (Metric Tonnes) 4139.0 8469.0 12915.0 15104.0 5079.9 9366.0 2638.2 7993.6 9758.7 8384.8 4041.2 0.5 Yield (Kg/acre) 4302.5 4176.0 6115.1 5534.6 4378.7 2442.7 6038.2 4712.5 1298.7 0.3 Potato Area (Acres) 8455.0 14481.0 17632.0 14780.0 13738.5 12154.0 9266.0 11390.0 12548.0 13391.0 12783.6 2690.7 0.2 Production (Metric Tonnes) 47405.0 53595.0 68049.0 61134.0 52959.4 46161.0 44014.2 49419.0 43000.0 50390.0 51612.7 7833.0 0.2 Yield (Kg/acre) 5606.7 3701.1 3859.4 4136.3 3854.8 3798.0 4750.1 4338.8 3426.8 3763.0 4123.5 639.5 0.2 Figure 2-5. Area grown to vegetables crops across Bhutan. P a g e | 27 Figure 2-6. Production of major vegetable crops across Bhutan. Figure 2-7. Yield of vegetables crops across Bhutan. P a g e | 28 Fruit The major fruits grown in Bhutan are citrus (mandarins, oranges), apples, pears, banana and areca nuts. Most fruit produced is exported to India and Bangladesh. Agricultural exports to countries, other than India, constitute 50 to 70 percent of total exports. In 2010 oranges were the top ranked export commodity (by value), followed by cardamom, and potatoes; in 2014 oranges were ranked second after potatoes (Bhutan RNR Statistics 2015, MoAF). Table 2-5. Major Fruit Production in Bhutan. Data compiled from various MOAF annual statistic reports. Major Fruit Production in Bhutan Year 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Mean SD CV Apple Total Trees 773.9 425.7 477.1 372.8 321.2 395.4 659.4 698.3 306.2 322.1 475.2 172.6 0.4 Bearing Trees 492.2 338.5 379.0 278.9 245.5 315.9 570.4 557.6 244.0 236.1 365.8 129.8 0.4 Production (Metric Tonne) 11834.4 10420.0 7407.0 7074.2 5410.4 15085.1 17371.4 20751.0 7666.1 8031.5 11105.1 5085.7 0.5 Yield (Kg/Bearing Tree) 24.0 30.8 19.5 25.4 22.0 47.8 30.5 37.2 31.4 34.0 30.3 8.3 0.3 Arecanut Total Trees x 1000 1410.9 863.1 989.7 1597.2 1736.9 1703.2 1801.1 1751.3 1128.4 1234.8 1421.7 346.7 0.2 Bearing Trees x 1000 33.6 32.0 31.7 49.0 48.4 58.6 71.9 87.3 67.3 62.4 54.2 18.7 0.3 Production (Metric Tonne) 6838.0 6616.0 6400.0 6568.0 4035.8 6373.0 7280.3 9963.0 7788.4 6249.7 6811.2 1475.2 0.2 Yield (Kg/Bearing Tree) 20.3 20.7 20.2 13.4 8.3 10.9 10.1 11.4 11.6 10.0 13.7 4.8 0.4 Mandarin Total Trees x 1000 3662.6 1969.3 2297.4 3252.2 1971.5 2804.3 2939.7 3167.7 2076.4 2087.4 2622.8 619.1 0.2 Bearing Trees x 1000 1966.8 1310.1 1353.1 2015.4 1121.0 1570.4 1554.1 1765.5 1104.9 1046.9 1480.8 352.9 0.2 Production (Metric Tonne) 63830.0 48369.0 55557.0 72071.0 38163.5 44177.0 52624.3 60414.0 49500.8 33470.4 51817.7 11746.0 0.2 Yield (Kg/Bearing Tree) 32.5 36.9 41.1 35.8 34.0 28.1 33.9 34.2 44.8 32.0 35.3 4.7 0.1 Banana Total Trees x 1000 290.9 214.8 436.2 719.1 594.8 558.4 328.7 449.0 182.9 0.4 Bearing Trees x 1000 132.5 124.2 201.8 268.9 165.8 162.9 116.6 167.5 53.5 0.3 Production (Metric Tonne) 2870.2 2376.0 3412.0 3974.0 2181.0 2208.4 2432.0 4.0 1493.0 2327.8 1136.1 0.5 Yield (Kg/Bearing Tree) 21.7 19.1 16.9 14.8 13.2 13.6 12.8 16.0 3.4 0.2 Pear Total Trees x 1000 27.3 36.5 58.7 75.2 21.6 32.5 31.3 50.8 39.8 49.1 42.3 16.3 0.4 Bearing Trees x 1000 18.9 34.5 37.4 48.8 18.6 17.3 16.3 26.4 19.8 20.7 25.9 10.9 0.4 Production (Metric Tonne) 585.9 1402.0 1422.0 2202.5 1204.8 1110.0 758.4 1151.0 2.2 1698.1 1153.7 609.1 0.5 Yield (Kg/Bearing Tree) 31.0 40.6 38.0 45.2 64.7 64.0 46.6 43.6 45.2 81.9 50.1 15.4 0.3 P a g e | 29 Figure 2-8. Number of bearing fruit trees across Bhutan. Figure 2-9.Production of fruit crops across Bhutan. P a g e | 30 Figure 2-10 Yield per bearing fruit trees across Bhutan. 2.4 PRODUCTION RISK The preceding section has presented maps and tables showing the mean statistics including the variability of agricultural production (yield and area panted) over the past decade in Bhutan. This variation is reflective of the risk faced each year by farmers. This production risk is a function of many factors, including extreme weather events, drought, pests, diseases, fire, management (including planting and harvesting), labour amongst others. These risks all have the potential to affect the quantity and quality of the crop production. A simple tool to assess the relative levels of risk in agricultural production is the coefficient of variation (CV). The CV is defined as the ratio of the standard deviation to the mean. The relative variability of each of the productivity measures considered (area planted and yield) can be examined by the CV, and ranges from 10% to 60% (Tables 2.3-2.5). The higher the CV, the greater the variability relative to the mean. For example, of the cereal yields considered, wheat yield is most risky. It has a CV of 30% which means that about 67% of the time, yield varies +/- 30% from its long term average. Rice yield varies by slightly smaller amount (20%). The area planted of rice and maize varies little, however the area of wheat, millet, and buckwheat have CVs of between 50-60%. There is also a range of riskiness across the difference fruit species. Mandarin yield has a CV of only 10%; this is the least varying of all fruit yields. Areca nut yield in the most risky (CV is 40%) P a g e | 31 and apples, one of the largest export earners (Bhutan RNR Statistics 2015, MoAF), along with pears, they have a high level of variability (50%). 2.5 KEY MESSAGES  Agriculture in Bhutan has a dominant role in the economy of the country with over 70% of the population involved in the agricultural sector and 56% are farmers. Agriculture contributes to about 17% of GDP in Bhutan.  Despite abundant water supplies, most agriculture is rain-fed due to steep topography and lack of irrigation infrastructure. This makes agricultural production highly vulnerable to the impact of climate variability and climate change.  The major cereals grown in Bhutan are Rice (Paddy), Maize, Wheat, Buckwheat and Millet. The average rice yield in Bhutan is 1282 kg/acre with significant variation across the region. The highest yielding districts are in the western region and the lowest yielding districts are in the southern region. Paddy production is relatively stable with an annual coefficient of variation of 10% followed by Wheat (20%). However area and production of Wheat, Buckwheat and Millet varies significantly from year to year with coefficient of variation between 40-60%. This variability is most likely due to the exposure of these rain-fed crops to climate variability in Bhutan particularly in the warm temperature and humid sub-tropical agro-climatic zones where the annual coefficient of variation of rainfall is between 40-60%.  A vulnerability assessment of the agricultural sector to both climate variability and climate change needs to be carried out to assess the exposure of different crops to climate and to assist in developing adaptive capacity for the sector. P a g e | 32 3 WEATHER AND CLIMATE RISK This chapter provides an overview of the climate of Bhutan including variability across different agro-ecological zones and identifies key drivers of this variability. The climatology of Bhutan is summarized and challenges for the development of an effective climate forecasting system which is essential for the delivery of agro-met services in Bhutan are discussed. 3.1 CLIMATOLOGY Using daily meteorological data from the Department of Hydro-Met Services (DHMS), weather and climate profiles for the nation are produced. Of the 92 weather stations, there are 20 “Class A” stations (Table 3-5). Most of these extend from 1996 to present and are used in analyses in this report. The climate data collection and rainfall network are summarised at the end of this Chapter and provide details on data quality. It must be noted that the data used in this report are “raw data” and not corrected for homogeneity or continuity. Therefore, the results presented in chapter should be treated with a degree of caution. The mean annual rainfall in Bhutan is 2000 mm, with more than 75% of annual rainfall occurring between June and September (Figure 3-1). The climate is dominated by the June-September monsoon season with a dry winter and autumn. The climatic conditions are influenced by topography, elevation and rainfall patterns. There is year round snow in the north. Southwards, closer to India, the weather is hot and humid in summer and cool in winter. Annual average rainfall ranges from 650 mm in the alpine and cool temperate region to more than 5500 mm in the wet subtropical region. Monthly distribution of rainfall for several districts (Dzongkhag) and agro-ecological zones are shown in Figure 3-1. The data show a strong seasonal cycle with the greatest amounts occurring in June and July. In general there is a strong relationship between altitude and climate (Table 3-1). It gets colder and drier with height and wetter and warmer as altitude decreases. However there are important differences to this relationship due to Bhutan’s physiography (Norbu et al.2003). The inner valleys of central Bhutan are dry valleys in which cloud cover are suppressed by strong up- valley winds (Whiteman 2000). Poor data coverage and quality lends itself to some uncertainties regarding the precise spatial distribution of rainfall in Bhutan. Although it is clear that the southern foothills of Bhutan both in the west and the east have a wet climate (annual rainfall up to 7000 mm), there is some debate as to the degree of rainfall decrease from west to east. Many researchers consider that the rain shadow effect of the Meghalaya Plateau in north east India is highly significant for the eastern regions of southern Bhutan (e.g. Biswas et al. 2007; Adlakha et al.2012, Bookhagen and Burbank 2010); others consider the effect to be limited and characterise all of Southern Bhutan as having a wet monsoonal climate (Baillie and Norbu 2004). River flow data indicate that the main rivers generally all have similar flow regimes and that Eastern Bhutan is not significantly drier than western Bhutan (Baillie and Norbu 2004). Some geomorphological indicators suggest that part of the eastern valleys are somewhat wetter than the rest (e.g. Van der Poel and Tshering 2003). This is a good example of the importance of P a g e | 33 high-quality climate data so that a proper assessment of rainfall variability across different regions can be made. Figure 3-1. Average monthly distribution of rainfall across different agro-ecological zones in Bhutan. All data are shown on same scale for comparison. P a g e | 34 Table 3-1 Monthly mean, minimum and maximum rainfall (mm) for different agro-climatic zones of Bhutan. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Bhur- Wet Subtropical Longitude 90.4 E Latitude 26.9 N Elevation 375 m asl Maximum 78.2 199 310 998 803 1670 2422 1882 1009 590 85.3 77.2 Median 10 17.2 60.9 221 378 942 1267 964 569 166 7.7 2 Mean 17.5 29.8 79.1 254 441 1021 1298 964 633 213 18 10.4 Minimum 0 0 6.6 0 224 566 425 429 169 16.8 0 0 Chamkhar - Cool Temeprate Longitude 90.8 E Latitude 27.5 N Elevation 2470 m asl Maximum 38.2 34.7 81.5 146 176 157 277 238 179 191 81.4 50 Median 0.7 8.6 27.9 57.3 85.6 114 149 146 80.2 42.8 1.4 0 Mean 6.5 11 31.1 63.7 87.6 102 147 138 85.9 56.1 6.7 4.1 Minimum 0 0 0 31 35.9 38.5 40.9 24.3 34.4 5.6 0 0 Dagana dzong - Warm Temperate Longitude 89.9 E Latitude 27.1 N Elevation 1460 m asl Maximum 60.2 196 102 334 570 839 705 908 864 352 85.4 55 Median 10 21.9 40 59.6 126 316 351 291 256 90 1.1 0 Mean 17.2 38.9 41.5 104 191 332 360 321 275 115 8.4 5.1 Minimum 0 0 5.9 10.8 12.9 85.4 66.3 11.5 39 7 0 0 Mongar- Humid Sub-tropical Longitude 91.2 E Latitude 27.3 N Elevation 1600 m asl Maximum 34 82.3 109 151 291 488 425 303 329 254 40.2 20 Median 1.4 9.3 25.8 82.4 90 135 206 200 86.4 64 2.6 0 Mean 7.3 13.5 34.5 74.7 104 158 210 179 104 73.4 5.6 3.4 Minimum 0 0 0 9.8 0 73.6 85.6 6.4 24.6 0 0 0 Punakha - Dry Subtropical Longitude 88.9 E Latitude 27.6 N Elevation 1236 m asl Maximum 134 304 72.2 150 317 769 410 296 202 130 18.5 20.1 Median 3.4 7.7 9.6 35.1 69.4 116 161 149 110 43.3 1.4 0 Mean 15.3 26.1 15.4 44.1 89.1 152 161 153 103 48.1 4.4 4 Minimum 0 0 0 8.8 7.9 11.8 63.4 6 20 0 0 0 Sipsu - Wet Subtropical Longitude 88.9 E Latitude 27.0 N Elevation 550 m asl Maximum 54.6 172 242 912 1055 2028 1914 1806 1059 364 164 64.8 Median 20.3 28.6 61.2 289 585 1024 1277 912 650 212 17.8 5.2 Mean 21.2 43.3 95.3 339 609 1062 1297 1065 662 217 28.6 16.9 Minimum 0 0 1.2 24.2 252 395 830 699 297 63.8 0 0 EXTREME EVENTS Heavy rainfall, 1-day extreme events, are particularly important from a risk management perspective as it is often these events which lead to severe floods, landslides and infrastructure damage (Nandargi and Dhar 2012). They are most likely to occur in July, August, and September and are associated with particular meteorological patterns, namely:  monsoon depressions: these move north from their origin in the Bay of Bengal causing heavy rainfall in the Himalayan foothills ( (Dhar and Nandargi 2000),  monsoon breaks within the monsoon season (Gadgil and Joseph, 2003, Wang et al.2005): intra-seasonal oscillations of the break and active periods of the monsoon that are primarily driven by the Madden Julian Oscillation (MJO) usually occur in the mid-late monsoon months of July and August, and occasionally in September. A break in the ISM refers to a cessation of the monsoon over much of India as the monsoon trough shifts to the Himalayan foothills. While the monsoon breaks are the driving force of below average rainfall anomalies across much of India, it is during these breaks that extreme P a g e | 35 precipitation occurs in the Himalayan foothills of Bhutan and northern India, including flooding of rivers. This period of enhanced rainfall during the monsoon breaks appears to only happen during the first few days of the break period and produces a threefold increase in precipitation compared with active monsoon conditions. INTER-ANNUAL VARIABILITY The following section discusses the inter-annual variability of rainfall in Bhutan and the influence of remote drivers (such as the El Niño Southern Oscillation, ENSO, the Indian Ocean Dipole, IOD, and the Pacific Decadal Oscillation, PDO) with a view to building capacity in seasonal or decadal climate forecasting. The inter-annual variability of rainfall is high across Bhutan as shown in Fig. 3-2 and Table 3-2. This variability is across all seasons, but is especially evident in the months of high monsoon rainfall. This ‘riskiness’ of rainfall varies across Bhutan: coefficient of variation of annual rainfall ranges from 13% in the cool temperate region to about 42% in the humid sub- tropical and 50% in the warm temperate region (Table 3-2). . Table 3-2 Average annual rainfall (mm) and rainfall variability across different agro-climate zones of Bhutan As mentioned previously the seasonality of rainfall is primarily driven by the Indian Summer Monsoon (ISM) originating from the Bay of Bengal. The variability of the amount of monsoon seasonal rainfall is driven by the variability of the onset and retreat of monsoon, the intensity, and the frequency of active and break periods. The monsoon onset is associated with the changes in the direction of seasonal winds and the shift in the position of the Inter Tropical Conversion Zone (ITCZ). P a g e | 36 Figure 3-2.The seasonal rainfall variability for key locations across Bhutan. ENSO has been established in the literature as a primary driver of the ISM (Webster et al.1998, Kumar et al.2006). In general, for all-India rainfall, ENSO accounts for approximately 30% of inter-annual Indian monsoon rainfall variability: rainfall decreases in a warm ENSO phase (an El Niño where sea surface temperatures in the central Pacific Ocean are warmer than the long term average) and increases in a cool ENSO phase, the La Nina where SSTs are cooler than average (e.g. Shukla and Paolino 1983, Yasunari 1990). This relationship has provided the foundations for seasonal climate risk management in many parts of India. However the literature is less clear on the role of ENSO as a driving force of inter-annual rainfall variability in Bhutan where climatological observations rarely extend beyond 30 years. The relationships discussed below are developed from tree ring analyses (Sano et al. 2013) and gridded rainfall data sets (Prevez and Henebry 2015), both of which have restricting caveats. There is evidence that annual area-averaged rainfall of Bhutan has the same relationship with ENSO as North East India (Yadav 2012), so conclusions may be drawn from studies in that region. However, caution must be used when examining regional scale effects or extrapolating results from other areas of India or the Himalayas. P a g e | 37 More explicitly, the monsoonal rainfall (June-October) in Bhutan shows an ENSO signal but it is quite regionalized using 0.5 x 0.5 degree gridded rainfall (Pervez and Henebry 2015). During a La Nina event there are positive rainfall anomalies in the south west of the country and no signal elsewhere. During an El Niño event there is decreased rainfall in the southwest and a small region in the north east of positive rainfall anomalies. This is corroborated by tree ring evidence (Sano et al.2013) from a single location which also shows that El Niño events are generally linked to dry conditions in central Bhutan, and La Nina events are related to wet conditions. It is important to note the spatial nature of relationship: the correlations between rainfall and ENSO indicators weaken moving east across the western Himalayas and northern India through to the eastern Himalayas and Bangladesh. The rainfall-ENSO relationship in Bhutan is not strong, nor is it linear or stationary. It is modulated by the IOD (Pervez and Henbry 2015). A positive IOD (warmer sea surface temperatures in the western Indian Ocean relative to the east; an event often associated with El Niño) is associated with lower than normal rainfall in eastern Bhutan. A negative IOD has no effect in Bhutan or over much of the Himalayas. The strongest signal in Bhutan is when a negative IOD event occurs simultaneously with a La Nina event: above average rainfall occurs across the country. There is little effect when an El Niño occurring with a positive IOD (Pervez and Henbry 2015). The La Nina-negative IOD events modify the rainfall in the months of June-July-August the most, and most importantly increase the frequency of extreme rainfall events (Pervez and Henebry 2015). There is also evidence of other drivers of multi-decadal modulations of the ENSO-ISM rainfall relationships in Bhutan, such as the PDO (Sano et al.2013). Tree-ring studies suggest that the meteorological extremes of flooding and drought in NE India have multi-decadal shifts and these may be associated with the PDO: a warm phase of the PDO is associated with a weak ISM (Krishnan and Sugi 2003). It has been shown that Indian monsoon rainfall more often tends to be below (above) normal when El Niño (La Niña) events occur during positive (negative) phases of the PDO [Berelhammer , Krishnan and Sugi 2003). The ENSO-rainfall relationship appears to be consistent only during positive phases of the PDO (Sano et al.2013). In terms of stationarity, the ENSO ~ monsoon relationship in Bhutan appears to have remained stable during the 1971-2011 period (Sano et al.2013), however there are reported changes over recent decades in the strength of ISM~ENSO~rainfall relationships in nearby north-east India and elsewhere in India (Kumar et al.1999, Torrence and Webster 1999, Clark et al.2000). There is debate over the nature of the change however, as other studies using different proxies of the P a g e | 38 Indian monsoon suggest the relationship with ENSO is still significant (Gershunov et al.2001, Goswami and Xavier 2005, Van Oldenborgh and Burgers 2005). ENSO appears to affect the ISM rainfall totals through its effect on the duration of the monsoon, and more specifically through the lengths of active and break spells (Dwivedi et al.2015). In India, during El Niño years the frequency of longer breaks of the ISM and shorter active spells increase significantly. This affects the extreme rainfall events as mentioned in the previous section. There is lack of clarity in the literature as many researchers mistakenly compare results of ENSO~ rainfall relationships from various regions, or use an all-Indian rainfall total, or compare relationships from different time periods, or do not distinguish between the Indian Summer Monsoon and the North East Monsoon. From an agricultural risk management perspective it is critical to use climate data for the region in which the farming system is located, and be aware of possible fluctuations in relationships through time. Although the rainfall records of Bhutan are not long enough for a statistically significant analysis, it may still be possible to identify some indications of the above relationships. Here we analyze 20 years of daily rainfall data for Punakha, to determine any patterns in the onset and duration of monsoon from 1990-2014. After a thorough analysis assessing many possible criteria, the onset of the monsoon was defined as the first date after 1 June when more than 40 mm of rain occurred over a 10 day period. Similarly end of monsoon was defined as the first date after 1 September when accumulated rainfall over 10 days was less than 20 mm. The results are presented in Table 3-3. P a g e | 39 Table 3-3 Onset and duration of monsoon (1990-2013) using daily rainfall data for Punakha in Bhutan. Start date End date Duration (days) Difference from mean (days) 1/06/1990 4/10/1990 125 28 10/06/1991 21/09/1991 103 6 26/06/1992 27/09/1992 93 -4 1/06/1993 3/09/1993 94 -3 3/06/1994 16/09/1994 105 8 9/06/1995 18/09/1995 101 4 27/06/1996 15/09/1996 80 -17 29/06/1997 1/09/1997 64 -33 10/06/1998 12/09/1998 94 -3 1/06/1999 14/10/1999 135 38 18/06/2000 14/09/2000 88 -9 1/06/2001 6/09/2001 97 0 11/06/2002 24/09/2002 105 8 1/06/2003 11/09/2003 102 5 1/06/2004 12/09/2004 103 6 30/06/2005 1/09/2005 63 -34 1/06/2006 10/10/2006 131 34 3/06/2007 21/09/2007 110 13 12/06/2008 21/09/2008 101 4 1/06/2009 8/09/2009 99 2 29/06/2010 29/09/2010 92 -5 20/06/2011 1/09/2011 73 -24 24/07/2012 24/09/2012 62 -35 1/06/2013 13/09/2013 104 7 Average Duration (days) 96.8 La Nina El Nino Neutral The mean duration of monsoon over the 1990-2014 period was 96.8 days ranging from 62 to 135 days. The average start date of monsoon was 16 June and the average end date of monsoon was 15 September. There appears to be increased variability in the onset and duration of monsoon particularly since 2005 with the shortest duration occurring in 2005 and 2012. The effect of ENSO on the monsoonal onset, duration and rainfall amount in Punakha are shown in Figure 3.4. The strongest relationship is between monsoon rainfall amount (duration) and ENSO phase. There appears to be more rainfall in El Niño phases and a slight trend towards longer duration. Interestingly, across Bhutan as a whole, the 2006 El Niño was a drought year (Khandu 2015); however as shown in the data from Punakha, 2006 was an anomalously wet year. Our analysis shows that the drought actually occurred in 2005 with the shortest monsoon duration (see also section 4.2.1). Perhaps 2006 is incorrectly attributed as a drought year using crop statistics in 2006 which actually reflects the 2005 production data. P a g e | 40 Relationship between monsoon duration, ENSO and IOD 50 40 Monsoon duration (anomalous days) 30 20 10 El Niño 0 La Niña -0.60 -0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1.00 Neutral -10 -20 -30 -40 IOD Figure 3-3. Relationship between monsoon duration and climate drivers. The amount of rainfall (mm) is indicated by the size of the circles and is positively correlated to monsoon duration (R = 0.0483 D +71.62 ; R2= 0.43). Daily rainfall data for Punaka 1990-2014.IOD time series from www.jamstec.go.jp/frgc/research/d1/iod/DATA/dmi.monthly.txt. The influence of the IOD on the onset and duration at Punakha was more difficult to assess as there have only been 2 negative IOD events since 1990 (Figure 3.4): 1992 which was a neutral ENSO phase and 1996 a La Niña. These 2 years have slightly shorter and drier monsoons. It is the La Nina events in positive IOD years that have the shorter monsoons, and the El Niño and neutral ENSO events in positive IOD years that have the longer wetter monsoons (Figure 3.4). This is different to the Pervez and Henbry (2015) study which suggests that there is little effect when an El Niño occurring with a positive IOD (using gridded rainfall for the period 1901-2010). In summary, southern Bhutan has greater inter-annual rainfall variability than central/northern regions. There is variability in different aspects of the monsoon rainfall including the total amount of rain, the onset and duration of the monsoon period, and also the frequency and length of active and break periods within the monsoon period. The relationship of these monsoon parameters with ENSO provides confidence in the prospects of seasonal predictions and risk assessments. LONG TERM TRENDS AND CHANGES IN RAINFALL RISK P a g e | 41 Although the rainfall data from Bhutan is not long enough to draw significant long-term trends, trends have been identified in regional neighbours such as Nepal and NE India. In NE India annual and seasonal rainfall between 1971-2007 has either decreased or has had no trend, depending on the specific location (Patle and Libang 2014). There have been no increasing trends in annual or seasonal rainfall. For longer time periods decreasing trends in monsoon rainfall have been observed (Krishnakumar et al.2009, Choudhury et al.2012, Kothyari and Singh 1996). Most studies find post-monsoon rainfall has increased but the level of statistical significance varies. For rain-fed agricultural systems the post-monsoon rainfall is important for crop intensification and also the fruit and vegetable production of the following winter season (Choudhury et al.2012). There has been an increase in extreme rainfall frequency from 1951 onwards. From 1971-2000 extreme 1-day rainfall events in Nepal has increased (Shrestha 2005). However there is a notable decrease in extreme events during the 2000-2007, a period when the monsoon was weak. TEMPERATURE RISK Temperature varies significantly across different agro-climatic zones and time of the year ranging from a minimum of -13oC to a maximum of 38.8oC. July and August are the warmest months and January is the coolest month. In the south and eastern districts temperatures are significantly warmer than the northern regions of Bhutan. Average monthly minimum, maximum and mean temperatures for several agro-climatic zones of Bhutan are shown in Table 3-4. The seasonal temperature risk can be further examined by, for example, assessing the seasonal distribution of monthly minimum and maximum temperatures for Punakha (Dry Subtropical, Figure 3-4). In addition to the clear seasonal cycles, there is clearly a much stronger seasonal cycle in the minimum temperatures; there is nearly a 20oC range between summer and winter. Although there is not a long enough time series to assess the long-term changes in temperature, we can still assess changes since 1990, the starting point of our key stations. The analysis of minimum and maximum temperature data (1990-2014) for six stations in different agro-climatic zones show a difference in trend between minimum and maximum temperatures. At all locations minimum temperature has decreased during this period; the largest decline is at Mongar (-0.23oC or -0.01 oC year-1). P a g e | 42 Figure 3-4 Monthly distribution of minimum and maximum temperature (oC) for Punakha in central Bhutan In general there is an increase in maximum temperatures in most agro-climatic zones with the largest increase being in Mongar (0.25oC), Punakha (0.2oC) and Sipsu (0.14oC) over this period (Table 3-4 and Figure 3-5). Bhur and Dagana Dzong , on the other hand show a decrease in maximum temperatures of -0.26 oC and -0.022 oC respectively. Of interest is that the trends in maximum temperature of the two locations in the Wet Subtropical zone (Bhur and Sipsu) are different in direction; Bhur is getting cooler and Sipsu is getting warmer. This demonstrated the necessity of having an adequate observational network. Associated with this divergent trend between minimum and maximum temperatures is an increase in variability as shown in Figure 3-5. In general temperatures are increasing however there is large spatial and temporal temperature variability within Bhutan (Hoy et al.2015). This trend is observed in the climate change literature for nearby regions. There are estimates that snow cover in the Greater Himalayas has decreased by one-third and duration by 23 days at altitudes of 4000-6000m between 1966-2001 (Rikiishi and Nakasato 2006). A decrease in snow cover in the Greater Himalayas represents a serious threat to sustainable agricultural water supply, and water for other uses (Rohrer et al.2013). The magnitude of the rate of decrease is not certain due to data limitations and also decadal climate influences (Zhang et al.2004). P a g e | 43 Table 3-3 Minimum, average and maximum temperature (oC) distribution across different agro- climatic zones of Bhutan. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Bhur - Wet Subtropical Longitude 90.4 E Latitude 26.9 N Elevation 375 m asl Maximum 30.0 35.0 33.2 36.4 37.0 36.0 38.5 38.8 39.0 37.0 35.0 32.0 Minimum 1.0 1.5 10.0 13.0 15.5 15.0 13.5 14.5 14.5 15.0 10.5 7.0 Mean 17.4 19.5 22.3 24.0 25.5 26.2 26.5 27.0 26.5 25.0 22.2 19.1 Chamkhar - Cool Temperate Longitude 90.8 E Latitude 27.5 N Elevation 2470 m asl Maximum 18.5 20.0 26.0 24.0 28.0 28.0 27.0 27.0 27.0 25.0 22.0 19.0 Minimum -13.0 -11.0 -10.0 -2.5 0.0 3.0 10.0 10.0 4.5 -5.0 -10.0 -13.5 Mean 4.0 5.8 8.7 11.7 14.6 17.4 18.5 18.3 16.8 12.8 8.2 4.9 Dagana Dzong - Warm Temperate Longitude 89.9 E Latitude 27.1 N Elevation 1460 m asl Maximum 20.0 22.0 26.0 32.0 32.0 33.0 33.5 31.0 31.0 30.0 29.0 27.0 Minimum -1.0 0.0 1.5 6.0 10.0 9.0 8.0 10.0 11.0 6.0 1.0 0.0 Mean 9.1 10.3 13.6 16.6 18.7 20.3 20.6 20.4 19.7 17.6 13.8 10.8 Mongar - Humid Subtropical Longitude 91.2 E Latitude 27.3 N Elevation 1600 m asl Maximum 22.0 25.5 30.0 30.0 34.0 33.0 33.0 34.0 32.0 31.5 27.0 22.0 Minimum -1.0 -1.0 5.0 6.0 4.0 13.0 14.0 15.0 13.0 8.0 4.0 2.0 Mean 11.3 13.1 16.2 18.6 20.5 22.1 22.4 22.6 21.7 18.9 15.3 12.3 Punakha - Dry Subtropical Longitude 88.9 E Latitude 27.6 N Elevation 1236 m asl Maximum 26.0 30.0 32.0 35.5 36.0 37.0 36.0 35.5 35.0 35.0 32.5 28.0 Minimum -4.0 -2.0 1.0 4.0 10.0 12.0 15.0 13.5 5.5 7.0 2.5 1.0 Mean 12.1 14.0 17.2 19.8 22.6 24.6 25.3 25.0 23.7 21.5 17.2 13.1 Sipsu - Wet Subtropical Longitude 88.9 E Latitude 27.0 N Elevation 550 m asl Maximum 29.0 30.0 32.0 34.0 33.2 34.0 34.5 34.5 34.0 34.0 32.0 30.2 Minimum 6.0 10.0 12.5 15.0 17.0 18.5 17.5 20.0 19.0 16.5 13.0 8.0 Mean 16.9 19.2 22.4 24.3 25.3 26.1 26.1 26.5 26.2 24.7 21.6 18.7 Table 3-4 Annual coefficient of linear trend in Minimum and Maximum Temperatures and the corresponding change in Temperature oC in different agro-climatic zones since 1990. Coefficient of Linear Trend Change from 1990-2014 o C Min T Max T Min T Max T Bhur - Wet Subtropical -0.007 -0.011 -0.156 -0.257 Chamkar - Cool Temperate -0.008 0.003 -0.194 0.077 Dagana Dzong - Warm Temperate -0.006 -0.001 -0.132 -0.022 Mongar - Humid Subtropical -0.010 0.011 -0.230 0.252 Punakha - Dry Subtropical -0.005 0.009 -0.125 0.204 Sipsu - Wet Subtropical -0.004 0.006 -0.103 0.137 P a g e | 44 Previous research has suggested that in general, globally, temperature variability and trend are much greater at higher altitudes (Ohmura 2012). However there is no consistent evidence for a relationship between air temperature trends and altitude for the Greater Himalayas (Bhutiyani et al.2007), yet a relationship was found for the Tibetan Plateau and at altitudes of up to 4800masl (Qin et al.2009). 3.2 FUTURE CLIMATE RISKS The IPCC AR5 assessments of the South Asian region (IPCC 2013, IPCC 2014) often exclude Bhutan due to poor data coverage, however general conclusions can be made. Ensemble-mean changes in mean annual temperature exceed 3°C above the late-20th-century baseline over South Asia in the mid-21st century under RCP8.5, are less than 2oC in both the mid and late-21st century under RCP2.6. There is more uncertainty surrounding assessments of changes to precipitation risk. Precipitation increases are very likely over South Asia by the late-21st century. Under the RCP2.6 scenario, it is likely that changes at low latitudes will not substantially exceed natural variability. Future increases in precipitation extremes related to the monsoon are very likely in South Asia. More than 95% of models project an increase in heavy precipitation events. All models and all scenarios project an increase in both the mean and extreme precipitation in the Indian summer monsoon. Monsoon onset dates are likely to become earlier or not to change much and monsoon retreat dates are very likely to delay, resulting in lengthening of the monsoon season. Future climate change is suggested to cause an increase in the frequency of breaks in the ISM, such that the frequency of extreme rainfall events could increase. The interannual standard deviation of seasonal mean precipitation also increases. P a g e | 45 Dry Subtropical Wet Subtropical Wet Subtropical Warm Temperate HumidSub-tropical Cool Temperate Figure 3-5 Time series of monthly minimum and maximum temperate (1990-2014) in different agro-climatic zones of Bhutan. The statistics of the dotted trend lines are in Table 3-4. P a g e | 46 3.3 CLIMATE DATA COLLECTION AND RAINFALL NETWORK The Department of Hydro-Met Services (DHMS) has the mandate for the collection, storage and dissemination of climate and hydrological data across Bhutan. Information collected by DHMS is disseminated to potential users in the hydropower, agriculture, health and other sectors as required. The Department operates some 92 weather stations (20 Class A, 11 AWS and 61 Class C) across the country with over 90 percent being at elevations of less than 3000 meters in the Warm Temperate, Dry Sub-tropical, Humid Sub-tropical and Wet Sub-tropical agro-climatic zones. Weather stations in the Alpine and Cool Temperate zones are sparse. Length of climate data is relatively short with most stations operating from 1996 onwards giving a maximum of 20 years of data. Class A stations generally measure temperature, precipitation, relative humidity, wind speed, wind direction, solar radiation, and atmospheric pressure. Class A stations are manually operated with two readings per day which are reported to head office by phone. This information is used to generate the forecasts of daily temperature and rainfall. Monthly reports are sent via mail to head office. Class A stations data are used to produce weather charts shared through the Global Telecommunication System (GTS). At the time of this assignment Bhutan did not have access to GTS, but preliminary testing of the service was being carried out in mid-2015. Class C stations are manned by local staff in different Dzongkhags with observations recorded on a daily basis and sent to head office by mail on a monthly basis. Observations are usually limited to daily minimum and maximum temperatures and daily rainfall. Eight Class C stations (Chazam, Drukguel, Dungmein, Gaselo, Gidakom, Khaling, Samtengang and Samtse NIE) have daily records of rainfall, minimum and maximum temperature starting in 1986. However, the data from these stations (1986-1996) are not quality checked and hard copies are not available. These stations have been excluded from the analyses in this report. A summary of Class A and Class C stations is shown in Table 3-3 . The data for stations used in this report is shown in Figure 3-6. No stations yet meet the international standard of 30 years of continuous records, and the data runs are too short and interrupted for detailed statistical analysis. This sparse network of gauge observations and topographical complexity has resulted in a relatively low level of knowledge regarding the spatial and temporal variability of rainfall across Bhutan compared with many other mountains of the world (Anders et al.2006); (Bookhagen and Burbank 2010). Using high resolution (10km grids), the spatial resolution of anomalous circulation and precipitation events can be resolved. Alternate data sets should be investigated for possible use in lieu of observed date. For example, the large size of the Himalayan region and typically difficult access to high mountain areas P a g e | 47 especially during winter resulting in poor information on snow cover. However, large scale snowcover dynamics has mainly been assessed by remote sensing (eg (Rohrer et al.2013). Figure 3-6 Availability of rainfall data. Missing data are shown as white (or yellow) colours. However, it must be noted that it is highly beneficial to use observed climate data rather than gridded data (Hulme 1992) such as utilized by the World Bank for its online analysis tools (http://sdwebx.worldbank.org/climateportal/). These data come with their own set of limitations (Hulme and New 1997). P a g e | 48 Table 3-5 Summary of Climate Data for Bhutan. Station Name Class Start Year End Year Note Bhur A 1994 2014 Missing the first third of 1995 and last half of 2011 Chamkar A 1990 2014 Missing the occassional month during 1990-1993 Daga Dzong A 1990 2014 Missing most of 1991 Damphu A 1990 2014 Missing first half of 1994 and 1995 Deothang A 1993 2014 Missing second half of 2011 Gasakhatey A 1996 2014 Missing most of 2000 and 2 months in 2006 Kanglung A 1990 2014 Missing most of 1993 and second half of 2011 Lhuentse Dzong A 2006 2014 Missing several months throughout the time period Mongar A 1994 2014 Namjeyling A 1996 2014 Paro A 1995 2014 Pemagatshel A 1991 2014 Missing the second half of 2011 Phuentsholing A 1996 2014 Punakha A 1990 2014 Simtokha A 1995 2014 Sipsu A 1995 2014 Tashiyangtse A 1990 2014 Trongsa A 1990 2014 Wangdue A 1990 2014 Zhemgang A 1994 2014 Missing the first half of 1995 Chazam C 1987 2014 Drukgyel C 1985 2014 Dungmein C 1986 2014 Gaselo C 1986 2014 Missing Significant data from 1992 - 1995 Gidakom C 1986 2014 Khaling C 1986 2014 Samtengang C 1986 2014 Samtse NIE C 1986 2014 Autsho C 2005 2014 Begana C 2008 2014 Missing parts of 2014 Betikha C 1996 2014 Missing most of 2008 and the first quarter of 2014 Bijzam C 1995 2014 Missing first half of 2013 Buli C 2003 2014 Missing several months throughout 2009 - 2014 Chendebji C 1996 2013 Damji C 2005 2014 Missing various months throughout the time period Degala C 2012 2014 Dungkar C 2006 2014 Missing most of 2011 and several months in 2014 Gunitsawa C 2011 2014 Missing most of 2011 Gyetsa C 1996 2014 Kamichu C 1996 2014 Missing second half of 2001 Kuengarabten C 1996 2011 Laya C 2014 2014 Only 5 months of data Langthel C 1996 2014 Missing most of 1998 and all of 2013 Lingmethang C 1996 2014 Lingshi C 2004 2010 Nobding C 1996 2013 Missing most of 2010, 2011 and 2012 Panbang C 1990 2014 Missing data from 2001-2013 Pelela C 2003 2014 Phobjikha C 1996 2014 Missing most of 2014 Samdingkha C 2007 2014 Shelgana C 1996 2014 Missing most of 2011 and 2014 Sherichu C 2005 2014 Shingkar C 2010 2013 Thinleygang C 2004 2014 Ura C 1996 2014 Missing data from 1999-2003 Yabilaptsa C 1996 2012 Yadi C 1996 2014 Missing data for all 1997 and most of 2012 Yotongla C 2003 2014 Missing data for all 2010 and most of 2012 P a g e | 49 3.4 KEY MESSAGES  The geography of Bhutan lends itself to a high level of spatial variation in its climatology. While there are similarities with other Himalayan regions and north east India, the differences are great enough to warrant caution when extending climate risk analyses from other regions to Bhutan.  The sparse observational climate datasets do not meet the standards set by WMO for statistically meaningful analyses. There are other complementary dataset such as gridded data, tree ring data, remote sensing, the can be used to support analyses of observational data. However, significant capacity development is required to make use of such information.  The Agro-ecological Zones of Bhutan are a reflection of both altitude and latitude, especially in central Bhutan. The heterogeneity means that it is not possible to generalise about a typical Bhutanese location, and it is important to use the station closest to the location of interest. This emphasizes the need for improved data network and observations throughout the country  The greatest variability in rainfall is in the warm temperate and humid sub-tropical region with implication for rain-fed crop production. A higher density of observation network in these regions would assist in the development of climate prediction models in the future.  The primary driver of seasonal and inter-annual climate risk is the Indian Summer Monsoon (onset, duration, monsoon “breaks”).  Using limited data, research indicates that ENSO and the IOD are drivers of risk. There may be others on a decadal time scale. Further research is needed to assess the impact of these drivers on the climate of Bhutan with the aim of developing seasonal and longer time climate prediction models.  From a risk management perspective, a significant feature of the ISM is the intra- seasonal variability of the wet (active) and dry (break) spells of the monsoonal rainfall. Intra-seasonal variations of monsoon drive floods and droughts. There is forecasting potential for many aspects of these climate risks.  Long term climate risks from climate change: Temperature increases particularly during winter. Rainfall projections indicate an increase in the intensity of rainfall and a decrease in winter precipitation, potentially increasing the risks of landslides and floods.  In general maximum temperatures show an increasing and minimum temperatures show a decreasing trends which has implications for agricultural production ( heat stress and frost risks). P a g e | 50 4 VULNERABILITY OF AGRICULTURAL PRODUCTION TO CLIMATE This chapter identifies the vulnerability of Bhutan’s agricultural production to climate risk and how it may be reduced through various strategies. The major climate risks facing Bhutan are twofold: those experienced historically through climate variability, and those resulting from climate change. New risks resulting from climate change include changes in the frequency and intensity of rainfall, uncertainty in the onset and duration of monsoon, increases in temperature, evaporation, floods and droughts. Weather and climate play a prominent role in agricultural production. Critical decisions such as when to plant, what to plant, when to harvest, fertilizing and pesticide applications require advance information on climate. Pest infestations and crop disease are often associated with changes in weather. Climate risks are more challenging in horticulture crops compared to annual food crops. The issues of climate risk and strategies to these problems requires thorough analysis, advance planning and improved management. The crop productivity is subjected to number of stresses and potential yields are rarely achieved without stress (Datta 2013). The forecasting and detection of these events, development of weather bulletins, and targeted agro advisory services can help better inform extension workers and farmers and support decision making at the farm level. Increases in temperature, variability in onset and the duration of monsoon are discussed elsewhere in this report. Further risks are detailed below. 4.1 MAJOR CLIMATIC RISKS IN AGRICULTURAL PRODUCTION 4.1.1 PEST AND DISEASES Pest and diseases have a major impact on agricultural productivity in Bhutan. Despite the significance of pest and diseases on agricultural production, the extent of financial losses are not well documented. Most outbreak of pests and diseases in major crops of Bhutan (e.g. rice, maize, wheat, potatoes and chilies) are influenced by climate conditions. An understanding of the cause and impact of these diseases is critical to identifying critical periods in the crop cycle (See Table 2-2 crop calendar) where provision of agro-met services and climate forecasts can reduce such impacts. Generally most damage occurs during an early start to the monsoon season (May-October) and occurrence of these outbreaks are less evident when the onset of monsoon is delayed. Some of the major diseases affecting agricultural production in Bhutan are discussed below. P a g e | 51 Grey Leaf Spot (GLS) and Turcicum Leaf Blight (TLB). This fungal disease affects maize crops causing defoliation and a significant reduction in yield. Maize is the largest crop by production in Bhutan (75716 Mt in 2013). About 45% of the total maize production comes from the six eastern districts of Trashigang, Samdrupjhongkhar, Pemagatshel, Trashiyangtse, Mongar and Lhuentse. Maize accounts for 49 percent of the national food basket and represents 42 percent of the cultivated area. GLS is a major problem in higher altitudes (2200 masl) and is most prevalent in wet monsoon seasons where a combination of moisture and temperature (15-25oC) is conducive to the outbreak. Favorable environmental conditions are more than 12 hours of leaf wetness (dew and rain), or relative humidity higher than 90%. TLB, also known as northern corn leaf blight, is a fungal disease (Exserohilum turcicum) and is a major constraint to maize production with a growing season characterized by high humidity and moderate temperatures (17 to 27°C). Yield losses as high as 70% have been attributed to TLB. In Bhutan TLB outbreaks mainly occur as 1200 masl where temperatures are usually higher. In 2006 and 2007 over 80% of the maize crop was affected or destroyed due to TLB. GLS and TLB resistant varieties are being developed in collaboration with CIMMYT in Mexico and currently being trialed in eastern districts. A limited quantity of seed is being propagated and distributed to farmers. Potatoes Tuber Moth (PTM). Potatoes Tuber Moth (PTM) usually occurs in mid-lowland areas (1200-1800 masl) but is now being seen in high altitude areas (>1800 masl), possibly due to warming effects. PTM usually affect the crop in storage after harvest when the ambient temperature is high. Rice Blast. The major disease of rice is Rice Blast, a fungal disease that can cause serious loss of yield. Rice Blast usually occurs from seedling stage to two months. In 1995, 50% of rice crop in Bhutan was destroyed due to Rice Blast. Environmental conditions favoring this disease include long periods of high humidity and a temperature in the range of 25-28oC. Excessive use of nitrogen fertilization as well as drought stress increase rice susceptibility to this pathogen, as the plant is placed in a weakened state and its defenses are low. In the context of climate change the following scenarios of pest dynamics are likely:  Temperature increases will lead to changes in the composition of pests. The high population growth rate of many species will ensure changes in pest distribution.  Increasing temperatures may increase the area where pests can live by increasing temperatures of areas currently too cold for some species.  Increases in winter temperatures will decrease the duration of hibernation of pest species thereby increasing their activity. Risks to agriculture from pests and diseases are critical especially in the context of the Government of Bhutan’s aspirations for organic agriculture. P a g e | 52 4.1.2 DROUGHT Drought is a major factor in crop production and despite frequent occurrences of drought in Bhutan, there is limited data on the extent of crop losses that can be attributed to drought. Currently there are no drought monitoring and forecasting capabilities in Bhutan. The Department of Agriculture does not collect information on drought impacts. For example in 2006 most of the agricultural production in Bhutan was affected by drought, however the extent of damage and financial losses are unknown. In years with higher temperatures and decreased rainfall, there is often an increase in evapotranspiration leading to severe crop water stress conditions. An analysis of rainfall percentiles (six month accumulated values) was carried out to identify periods of drought in three of the agro-climatic zones to assess the extent and severity of droughts that have occurred in Bhutan since the availability of rainfall data (1990-2014). A drought period is defined when the six-month percentile values are in the lowest 10 percentile. A warning of drought was defined when the six-month percentile values falls below 40 %. The six-month percentile values were selected to reflect “agricultural droughts”. The time series of drought analysis for the six regions in different ago-climatic zones is shown in Figure 4-1. A drought history report for three of the stations (Bhur, Punakha and Dagana Dzong ) is shown in Table 4-1 . The data clearly show the occurrence of more than 10 drought events between 1990-2014. The most severe being the drought of May 2005-Jan 2006 (Punakha), July 1994-Jan 1995 (Bhur) and June 1998- Mar 1999 (Dagana Dzong). Other periods of significant drought are shown in Table 4-1. There appears to be no relationship with El Niño or La Nina events. However, as discussed in Chapter 3, there is a relationship between the Indian Summer Monsoon rainfall and ENSO. In India since 1980, all droughts occurred in El Niño years (but not all El Niño years resulted in drought). 4.1.1 FLOODS AND EXTREME EVENTS Flash floods, hailstorms, and extreme weather events cause considerable damage to crops, and during harvest season can result in the loss of an entire crop. For example in 2013 cyclone Phailin caused massive losses to rice crops in Paro when it hit the region in October when most farmers had harvested their crop and left it in the field to dry. As discussed in Chapter 3 the period of greatest risk for extreme rainfall events is in the summer monsoonal months. In the south western regions this risk is greatest in June, whereas in the south eastern regions the greatest risk is in July. The monsoon rains and those from cyclonic events are the main trigger for flooding events especially in southern regions of Bhutan where the landscape P a g e | 53 Figure 4-1 Drought analysis for six agro-climatic regions in Bhutan. is conducive to flash flooding: that is, deeply eroded, steep and closely spaced gullies, gorges and river valleys, combined with low permeability or saturated soils. 4.2 SENSITIVITY OF AGRICULTURAL PRODUCTION TO CLIMATE. There are few studies in Bhutan that focus on the sensitivity of various crops to climate. However there are studies available from elsewhere in south Asia that may provide some useful insights. These include studies assessing the sensitivity of agriculture and agronomy to climate variability and also effects of changes in climate including increased levels of atmospheric CO2. The historical sensitivities that farmers have dealt with are summarized in Table 4-3. For example, chilli’s are sensitive to particular temperatures: there is a high percentage of chilli seed germination (90%) when minimum temperatures are at 20oC and germination stops completely when minimum temperature reach 10oC. Therefore the decrease in minimum temperatures that has been observed will increase the risk of seed germination. P a g e | 54 Table 4-1. A drought history report for Bhur (top), Dagana Dzong (middle), and Punakha (bottom). P a g e | 55 Potatoes are sensitive to temperature. An increase in temperature risk will favour production in Bhutan’s high altitude growing regions as the growing season will be extended. In the subtropical plains, the growing period will be shortened during the winter season. Potatoes need low temperatures for flowering. They also have high temperature sensitivity for tuber formation. Optimum tuber formation takes place at 20oC. An increase in temperature of above 21oC causes a decrease in the potato tuber yield, and at 30oC there is complete inhibition of tuber formation. Apple farming in Bhutan is very sensitive to winter temperature and precipitation in the form of snow as they are key determinants of the induction of dormancy, bud break and also ensure proper flowering in apples. They require 1200-1500 hours of chilling depending on the specific variety or else there is poor fruit formation. Apples are also a good example of the different climate sensitivities in different parts of the growing calendar. For example, a delay in the onset of cold conditions in December and January will affect the chilling requirements. There also needs to be uninterrupted winter rest. However, poor apple yield may result from freeze injury from extreme low winter temperatures. In summer, temperature impacts apples farming throughout the season from immediately after blooming period in the apple orchards to the fruit size at harvest. Increases in temperatures especially in low altitudes see flower buds producing few fruit clusters. Apart from temperature, the decline in apple farming at low hills is also attributed to hailstorms, decrease in snowfall and inferior quality of fruit production due to pest attack. All insect species have a temperature range within which they remain active from egg stage to adult stage. Within this favorable range there is also an optimum temperature for development, and if there is exposure to temperatures either side of this value there is an adverse impact such as slowing down the speed of development. In general, the activity and population of pests such as aphids increase with temperature. For example, the climate conditions creating a high risk for pest outbreaks include for mustard aphid (Lipaphis erysimi): maximum temperature ranges from 19 to 24oC with a mean of 12–15oC; for gram pod borer maximum temperature is 23 – 27oC; for the rice stink bug maximum temperature is between 26.9 and 28.2◦ C with a relative humidity of 80.6–82.1%; for rice green leaf hopper, temperature from 20 to 28oC. These pests are the vectors of various viral diseases of vegetable crops mainly chilli, potato, and legumes causing severe loss in yield of these crops. P a g e | 56 Table 4-2 General sensitivities of Bhutan's agricultural sector. Source: (NEC 2000) AGRICULTURAL CLIMATE SENSITIVITIES Crop production  Loss of production and quality (due to variable rainfall, temperature, etc.).  Decreased water availability for crop production.  Increased risk of extinction of already threatened crop species (traditional crop varieties) Changes in runoff  Loss of soil fertility due to erosion of top soil and runoff.  Loss of fields due to flash floods, landslides and rill & gully formations.  Soil nutrient loss through seepage Impact of changes  Crop yield loss (flowers & fruit drop) to hailstorms. in hailstorms  Deteriorated produce quality (fruit & vegetables) by untimely incessant heavy rains and hailstorms Agronomy  Delayed sowing (late rainfall).  Damage to crops by sudden early (paddy) and late spring (potato) frost (ref. seasons shifting) Pests and diseases  Outbreak of pests and diseases in the fields and during storage where they were previously unknown Infrastructure  Damages to road infrastructures (food security) Studies are available from North East India that utilize crop models to assess the climate sensitivity of various crops. However, such studies are extremely sensitive to the choice of crop model and its parameterizations such as soil type, the choice of location, and the climate model (if one is used). In terms of climate change, there is potential for severe impacts to agriculture. In general, if agriculture is practiced in marginal production regions then they are likely to have relatively high levels of climate sensitivity. Climate change impacts are likely to result in production moving to high altitudes to steep slopes that are usually unsuitable for agriculture. There is a likely risk of increased severity and frequency of monsoonal storms which could aggravate the occurrence of landslides and flooding. This may well contribute to deterioration in crop production and in the quality of agricultural land (NEC 2000). P a g e | 57 Table 4-3 Climate sensitivity of Bhutan's key agricultural produce. CROP CROP- SPECIFIC CLIMATE SENSITIVITY Cereal  o Rice/Paddy Rice Blast (fungal): high humidity, temperature 25-28 C, drought stress.  Low minimum temperatures during the reproductive stage can be catastrophic to crops.  Higher radiation: higher grain yields.  Higher temperatures during grain formation and filling results in fewer grains.  o Maize Grey Leaf Spot: Temperatures: 15-25 C, > 12 hours of leaf wetness (dew and rain).  Turcicum Leaf Blight: High humidity and moderate temperatures (17 to 27°C). Vegetable Potato  Potatoes Tuber Moth. High ambient temperatures in storage.  Yields temperature sensitive: mean daily temperatures 18 to 20°C.  Minimum temperature <15°C is required for tuber initiation.  Soil temperature for normal tuber growth is 15 to 18°C, with growth limited when o o T<10 C and T>30 C. Sensitive to soil water deficits < 50% Chilli  Tolerant of drought and floods.  o o Minimum temperatures: seed germination stops at 10 C, 90% success at 20 C.  o Fruit Leaf production increases by one leaf per month for every 3.3 to 3.7 C rise in o o minimum or mean temperature from 10-20 C or 13.5 to 25 C respectively.  o Higher temperature (31-32 C), in general, increases the rate of plant maturity in annual species, thus shortening the growth stages, during which developing fruits o and suckers absorb photosynthetic products. The temperature below 10 C leads to impedance of inflorescence and malformations of bunches  o o Apple High (26 C) or low temperatures (15 C) during the flowering phase reduce the apple crop; Sensitive to frost, hail, and pests. Mandarin  Growth ceases at high temperatures.  Frequency, timing and duration of frosts.  Wind: strong or cold winds may decrease yield and scar fruit P a g e | 58 4.3 CLIMATE RISK MANAGEMENT: strategies to deal with climate risk. Climate risk management has taken many forms in Bhutan. Some strategies used to manage the agricultural climate sensitivities discussed in the preceding section are detailed in Table 4-4. For example the increasing pattern of chill unit at 2700 m above mean sea level reports that the area is conducive for apple cultivation and hence there has been a shift in apple farming from low hills to middle and high hills. The harvesting period of apple is also delayed for a week to a fortnight. In many such areas, apple farming is being replaced by raising coarse grains and seasonal vegetables. Paddy farmers have used difference varieties of rice in addition to changes in agronomic practices to deal with climate risk. Paddy and other Irrigated farming has managed risk through improvements to water distribution systems such as leakages and evaporation, crop choice to increase return per liter, water application methods, irrigation scheduling and moisture monitoring. There are a range of strategies to reduce climate risk that are not crop specific and may be modified slightly depending on the region or cropping activity (Commission 2000). These include:  Diversify crops or enterprise. This may involve the introduction of new varieties that may be pest/disease resistant and more suited to extremes in temperature.  Change production methods and farming patterns.  Maintain farm infrastructure and equipment. Improve or upgrade storage facilities to store and have access to food grains as an insurance against crop loss or damage or bad yields  Use protected cultivation and other production tools such as terracing and contouring.  Site selection. Some areas may no longer be suitable for growing historical crops, and new areas that were previously unsuitable may become highly suitable.  Develop a contingency plan. Create more off-farm or cash earning job opportunities (weaving, constructions, road labor, etc.) for farmers who are affected by crop loss due to climate change effects. Climate risk affects all sectors and therefore risk management and adaptation need an integrated approach across agriculture, soil conservation, forest management, water resources management, disaster risk reduction, and community ‘development’. In terms of adapting to longer term risk factors introduced by climate change, several adaptation strategies have been identified that address the needs of reducing agricultural vulnerability (Lhendup 2012). Some of these are extensions of the climate risk management strategies above. They include:  Increase climate risk awareness  Diversifying livelihood resources  Diversifying agricultural crops P a g e | 59  Sustainable land and soil management  Livestock intensification program  Promotion of fruit and fodder trees in the villages  Local capacity-building  Linking with relevant institutions  Mainstreaming adaptation into development plans  Building ecosystem resilience Table 4-4. A selection of strategies that farmers and industry use to address climate influences on Bhutan's key agricultural production. CROP MECHANISMS TO COPE WITH CLIMATE RISKS Cereal Rice/Paddy  Rice Blast: adjust application rates of nitrogen fertilizer (high loads increase incidence).  Cropping area modification.  Nutrient management.  Water management.  Stubble retention. Maize GLS and TLB resistant varieties are being developed in collaboration with CIMMYT in Mexico and currently being trialed in eastern districts. Vegetable Potato There are a choice of varieties: early (90 to 120 days), medium (120 to 150 days) and late varieties (150 to 180 days) to suit different climatic conditions. Cool conditions at planting lead to slow emergence which may extend the growing period. Early varieties bred for temperate climates require a day length of 15 to 17 hours, while the late varieties produce good yields under both long and short day conditions. For tropical climates, varieties which tolerate short days are required for local adaptation. Chilli Germination in greenhouses. Fruit Apple Replace with coarse grains and seasonal vegetables. shift in apple farming from low hills to middle and high hills P a g e | 60 4.4 VULNERABILITY ASSESSMENTS Vulnerability is defined by the IPCC as “ a function of the character, magnitude and rate of climate variation to which a system is exposed; its sensitivity; and adaptive capacity (IPCC, 2001)”. The scope of this project did not extend to a vulnerability assessment. It is critical that this be a priority as is enables:  The identification and prioritization of the most vulnerable sectors and districts  The identification of adaption options  The inclusion of adaption in policy across all levels of government. Countries that are the most vulnerable to climate risk are those with degraded ecosystems and natural resources, and have significant proportions of the population who are unable to implement strategies to decrease their exposure to high levels of climate risk. As already described in Chapters 2 and 3, Bhutan is characterised by a large rural population, low population density, a large forested area, a large proportion of the population dependent on natural resources, and poor infrastructure development. In addition Bhutan has diverse climate regimes which are highly dependent on monsoon, a high proportion of crop area in rain-fed agriculture and is therefore vulnerable to climate variability. The GAIN (Global Adaptation Institute -2011) vulnerability index suggest that Bhutan has a high level of vulnerability to climate change and a low level of ‘readiness’ (Asian Development Back 2014). These two measures are essentially a function of economic development, institutional capacity, and social factors. Although there is knowledge of the climate sensitivity of some crops in Bhutan, there are only limited vulnerability assessments. However there are Vulnerability Asssement available for North East India (Ravindranath et al.2011). There are some similarities between the two regions which may enable limited conclusions. Indeed much of the literature on climate vulnerability is concerned with vulnerability to climate change, and does not directly address historical climate risk. A regional, sector specific vulnerability assessment will require relevant scientific research, stakeholder engagement, policy and institutional analysis as well as economic and development projections (Lhendup 2012). P a g e | 61 4.5 KEY MESSAGES  Agriculture in Bhutan is highly sensitive to climate risk, and the low levels of adaptive capacity makes it vulnerable. Through the effective use of climate and weather information, climate risk management has the potential to maximize opportunities presented by climate risk as well as reducing losses. However these analyses are not currently available.  A climate-resilient agricultural sector will require information at a range of institutional levels and for a wide range of user-groups. The critical information requirements and pathways are detailed in Chapter 5.  In terms of climate sensitivity and vulnerability, the following data and knowledge is required: 1. Meteorological observational data from stations meeting all the standards set by the WMO. 2. Historical climate risk assessments on a range a temporal and spatial scales. Presently the knowledge of historical climate risk in Bhutan is limited, however there are data and techniques available that can enable more comprehensive analyses. 3. Develop and utilise existing agricultural Decision Support Systems (DSS). DSS have been recognised globally as a framework in which agricultural systems can be simplified and structured in a way that enables insights, and utilised for sensitivity studies or predictive purposes. 4. Assess the climate sensitivity of agricultural systems: cereal, fruit, and vegetable crops. While there is general information available on the climate sensitivity of key crops, some of which is presented here, this needs to be targeted for specific regions. In addition, there are indirect influence of climate on agriculture such as the sensitivity of water resources, relevant infrastructure, land at risk from flooding, sediment transport and areas at risk from erosion. 5. Vulnerability assessments. Information gaps and data constraints to vulnerability assessments in Bhutan were identified by the Bhutan Government in 2000 (NEC 2000) and these are still highly pertinent. This includes cross-sectoral interactions and the possible effect of climate variability and change on these interactions. 6. Future climate change scenarios for Bhutan. These scenarios are most effective when presented in a risk management format, rather than as a single scenario. This includes future climate variability and extremes in Bhutan. Vulnerability needs to incorporate environmental and social consequences, practicability, and effectiveness of adaptation opportunities. P a g e | 62  Addressing these information gaps while implementing specific programs to increase the adaptive capacity of agricultural sectors, ecosystems, and communities.  As detailed in Chapter 5, a key requirement of providing improved information for agricultural sustainability is institutional capacity building. This will enable strengthening of education, training, data, and institutional capabilities. P a g e | 63 5 DEVELOPING A NATIONAL FRAMEWORK FOR CLIMATE SERVICES Although the provision of agro-met services is clearly a priority in the national agenda of Bhutan and is identified as one of the top six projects for implementation (Bhutan NAPA-II 2013), under the National Adaptation Program of Action (NAPA) Bhutan at present lacks the resources, skilled workforce and institutional arrangements to provide comprehensive agro-met services. Many sectors are sensitive to climate impacts including agriculture, water resources, energy, health, tourism and ecosystems. However, given that the focus of this report is on improving the agro-met services in Bhutan, this chapter focuses on climate service delivery in the agriculture sector by identifying gaps and opportunities as well as the institutional arrangements necessary for strengthening climate service delivery in this sector. 5.1 Institutional Arrangements The two main departments with responsibility for the delivery of agro-met services in Bhutan are the Department of Hydro Met service (DHMS) under the Ministry of Economic Affairs (MoEA) and the Department of Agriculture (DoA) under the Ministry of Agriculture and Forest (MoAF), Royal Government of Bhutan (RGoB). The section below reviews the structure of DHMS and DoA at the national and sub-national levels and identifies areas for capacity strengthening and institutional arrangements that could help in the effective delivery of climate services. 5.2.1 Department of Hydro Met Services (DHMS) The Department of Hydro Met Services is one of the eight departments under the Ministry of Economic Affairs. Established in 2011, its mandate is to provide reliable and timely hydro- meteorological information and services to other agencies and the public. Specifically the functions of DHMS are to; • develop policies, legislations, and regulations related to hydro-met services • establish and operate the national hydro-meteorological network to monitor and collect weather, climate, water, and environmental data • set up and operate telecommunication for data acquisition, information dissemination, and delivery of services • advance weather, climate, and water science and technology through research and development • participate in the development and operation of the national multi-hazard early warning system • fulfil relevant international and regional commitments and further national interest through participation in programs and activities under such agreements and conventions • set up and operate the data archival, processing, and forecasting system to develop hydro-met services for protection of life and properties P a g e | 64 • develop capacity in the field of hydrology and meteorology The total number of staff in DHMS as approved in the 10th five year plan is 155. Currently 26 are classified as professional with an academic qualification (Engineers, Hydrologists and Meteorologists), 117 as technicians (a person who has completed a Basic Instruction Package for Meteorological Technician; BIP-MT) and 5 Administrative staff. There are currently several vacancies mainly in the professional stream. The DHMS has four divisions (Planning Coordination and Research Division; Meteorology Division; Hydrology Division; and Snow and Glacier Division). The Planning, Coordination and Research Division has responsibilities for human resources, staffing, training and research planning. The Hydrology Division is responsible for monitoring and maintaining the hydrology network and related operational hydrological activities; The Snow and Glacier Division is responsible for monitoring Bhutan’s snowpack and glaciers. The Meteorology Division is responsible for operation and management of surface weather stations and provision of 24 hour weather forecasts. An organogram of DHMS is shown in Fig.5-1. Fig. 5.1 : Organogram of the Department of Hydrology and Meteorology Service (DHMS) P a g e | 65 A detailed description of all four divisions is beyond the scope of this report. However, given the importance of the Meteorology Division (MD) in the collection of weather and climate data and provision of climate information and products to different sectors, a detailed account of the MD, its functions and identified gaps follows. The main functions of the Meteorology Division are to;  provide daily weather forecasts information to the public  prepare and disseminate seasonal climate forecasts including monsoon outlook  study/observe extreme weather events such as cyclones and heavy rainfall for press releases and to disseminate timely information to the public  conduct analysis and research on climatic data for specific purposes  operate and maintain existing meteorological network (11 real time AWS, 20 Class A, 61 Class C stations) across the country  collect meteorological data and ensure quality control, archiving and database management providing climatic data and services to the government for planning of developmental activities for specific studies and projects, including in the private sector  process data in coordination with PCRD for archive in a central database for statistical analysis, dissemination, and publication  maintain agro-meteorology and related information database required for RNR sector  collect and provide aviation forecast services The Meteorology Division has 36 approved staff positions at Thimphu of which 16 are filled and 20 vacant. Of the 16 staff currently in service, 4 are professional and 12 technical staff. Within the Meteorology Division there are two sections (Agromet/Climatology and Aviation Meteorology) with 1 and Nil staff (respectively) assigned to these sections at present. The lack of resources clearly impacts the ability of DHMS to provide routine weather and climate services to different sectors. In fact DHMS at present only provides a 24 hour weather forecast which is not verified or validated. Moreover, the division’s activities are constrained by limited internet connectivity and a lack of access to an automated observation network, models, infrastructure, and computer resources. Although the Meteorology Division have recently gained access to the Global Telecommunication Network (GTS) and now accesses global and regional data, including climate products from Global Producing Centers (GPC) and Regional Climate Centers (RCC), without proper training and resources it will be difficult to take advantage of this service and interpret and disseminate the information from these sources to the public and other sectors. 5.2.2 Department of Agriculture Agriculture is the most important sector of the Bhutanese economy. It contributes up to nearly 16.8% of GDP, accounts for 4.3% of exports, and provides a large proportion of the raw materials for industries. The sector directly employs about 59.4% of the total population and will continue to be the key determinant in the country’s efforts to reduce poverty in the immediate years ahead. Agriculture is also very vulnerable to the impact of climate variability and climate change as most agricultural production is rain-fed. P a g e | 66 The Department of Agriculture is one of four Departments under the Ministry of Agriculture and Forest (MoAF- often called as Renewable Natural Resources or RNR Sector) with a total staff of more than 442 personnel (Bhutan RNR Statistics 2012) – Fig.5-2. The department delivers agricultural services to the farming community through a network of Extension, Research and Central program offices established at strategic locations throughout Bhutan. The Department of Agriculture is comprised of three divisions (Agriculture, Horticulture and Engineering) including Extension, Communication and Information Management. P a g e | 67 Fig. 5.2- Organogram of the Ministry of Agriculture and Forest Most research activities related to Agriculture are carried out through four Research and Development Centres (RDC’s) spread across the country. These include;  RNR-RDC at Wengkhar, – with a specific focus on Horticulture Research and Development.  RNR-RDC at Bajo, – with a focus on Field Crops Research and Development.  RNR-RDC at Bhur, – with a focus on Sub-tropical Crop Research and Development. And;  RNR-RDC at Yusipang – with a focus on Forestry Research and Development. In addition to the four RDC’s, seven Research Sub Centres are located in various parts of Bhutan conducting research in cropping systems in different agro-climate zones. Agricultural research in Bhutan is supported by a network of 40 Agricultural Extension Centres throughout the country and is well positioned to communicate climate services to the rural community. The Department has the highest number of Professional Staff in the RNR sector by position. About 173 of the 442 positions are classified as professionals with many staff holding a Masters or PhD degree from recognised overseas institutions. However, staff with a higher P a g e | 68 degree are naturally promoted to administration responsibility which limits the research capacity within the organisation. Research activities are predominately related to field trials and crop improvement programs. There is little analytical research (crop modelling or Decision Support Modelling) to examine the effect of climate (climate variability and climate change) on production risk of crops across different agro-ecological zones. 5.3 From Climate Observation to Climate Service Delivery Climate Services refers to transforming climate data into climate information in a way that responds to user needs and assists decision-making to reduce the impacts of climate-related hazards and increase benefits from favourable climatic conditions (WMO- Global Framework for Climate Services). The National Hydromet Services has a clear mandate for collection of climate data and developing generic climate products for both public and private sectors. However, this mandate does not extend to formal engagement with the user community to identify climate needs and various sectors and to develop products specific to the needs of these sectors (climate service delivery). The user community need to engage with information providers to understand what climate information is available, how to interpret it correctly, as well as understanding its underlying assumptions and limitations. Currently partnerships are non-existent or very weak in Bhutan. Strong partnerships needs to be developed to create effective multi-disciplinary environments. Within the context of a Global Framework for Climate Services, the provision of climate services is a collective responsibility of several institutions involving a range of skilled personnel across different disciplines to ensure the climate services are authoritative, credible and dependable, and used to better inform decision-making by the end users. Although each institution designates their own particular job roles, the institutions through the collective skill of their staff must be able to demonstrate the following competencies:  The creation and management of climate data sets.  The ability to derive products from climate data;  The creation and/or interpretation of climate forecasts and model outputs;  The ability to verify the quality of climate information and services;  The skills and avenues to communicate climatological information to users. The Global Framework for Climate Services (GFCS) classifies the climate service providers within a country according to four categories. Each category includes the functions described in the previous category plus additional requirements as stated; P a g e | 69 Category I: Basic capacity - includes design, operation and maintenance of national observing systems; data management including quality assurance; development and maintenance of data archives; climate monitoring; climate diagnostics and climate analysis; climate assessment; dissemination via a variety of media of climate products; and, participation in regional climate outlook forums and some interaction with users. Category 2: Essential climate services – includes the capacity to develop and/or provide monthly and longer climate predictions including seasonal climate outlooks; conduct or participate in regional and national climate outlook forums; interact with users in various sectors to identify their requirements f,; and, provide advice on climate information and products Category 3: Full climate services – includes developing specialized climate products to meet the needs of major sectors which should be able to downscale long-term climate projections as well as develop and/or interpret decadal climate prediction. Category 4: Advanced climate services – includes running Global and Regional Climate Models; working with sector-based research teams to assist them in developing applications models (e.g. to combine climate and agriculture information and produce food security products); developing software and product suites for customized climate products; modelling and utilising statistical expertise in a multi-disciplinary context; and, downscaling global scale information to regional and national levels. Bhutan at present meets most but not all of the requirements at Category I level of competency. Whilst observations and monitoring networks are already being developed, there is an urgent need for the development of:  research capacity for climate services (as distinct from their capacity to collect climate data)  a historical climate database and real time observation network,  climate services information systems, and a user interface platform through collaboration with other sector agencies. 5.3 Development of Agro-met Services in Bhutan The provision of agro-met services requires close collaboration between the DHMS and Department of Agriculture to enable the integration of user needs into the development of services and to facilitate feedback for their improvement. Currently no organisation in Bhutan has a clear mandate for agro-met service delivery, albeit the importance of agriculture in the economy of the country and its vulnerability to climate. A comprehensive capacity building initiative is needed to strengthen DHMS’s capability and human resource capacity in the areas of weather and climate forecasting, numerical weather prediction, interpretation of climate products including limitations and uncertainties; identification of user needs, development of sector specific products, in conjunction with partnership creation and inter-agency communication. P a g e | 70 Similarly, a comprehensive capacity building initiative is needed within the DoA to conduct a comprehensive vulnerability assessment of agricultural production due to the impacts of climate variability and climate change. Furthermore, analytical and crop simulation capability within the DoA is lacking and needs to be developed, preferably in partnership with leading international institutions. Based on an initial assessment of institutional capacity for the effective delivery of agro-met services in Bhutan, a list of competencies has been identified as a priority for development (Table 5.1). Although not comprehensive it provides a starting point towards urgently needed capacity building and establishing a dialogue between climate providers and users of the information, thus transforming the Department of Hydro-met services from a “National Climate Centre” to a credible Climate Service Provider. Achieving this goal will require a formal partnership either with a legal mandate or a Memorandum of Understanding (MoU) between the DHMS and DoA. To achieve this it is recommended that an externally led visioning process should be conducted to ensure that there is mutual understanding and project alignment across the two organisations. This process would pull together key personnel across the board to ensure that issues raised in this report are understood by all and a collective goal is established, whilst at the same time maintaining the integrity and value of each of the parts within the whole. This will ensure efficiency and establish a shared working vision as the basis for effective collaboration and communication. P a g e | 71 Table 5.1 – Identified competencies to be achieved for climate service delivery with a focus on the agricultural sector – with agencies with prime responsibility for implementation Identified Gaps DHMS DoA External Recruitment and training of professional staff in Climate Science and related fields l l Apply quality control processes to climate data including creation of homogenious data l Create sub-seasonal and seasonal forecast products l Identify and retrieve adequate climate data from different sources to generate climate products l Access and interpret climate forecasts issued by regional and global agencies including measures of uncertainty l Conduct routine verification and validation of weather and climate forecasts l Compute basic climate products, such as normals, or anomalies defined relative to a reference period l Compute Climate Indices for the monitoring of climate change and climate extremes l Create value-added products such as graphics, maps and reports to explain climate characteristics and evolution l Prioritize the communication of climatological information to different sectors l Establish effective communication channels with climate services users l l Conduct and evaluate user satisfaction surveys on a regular basis l l Revise climate services and their communication based on user feedback l Recruitment and training of professional staff in agricultural systems modelling l l Conduct a vulnerability assessment for the Agricultural sector l Develop analytic and simulation capacility to assess th impact of current and future climate on production risk l Identify critical cycles in the production system where climate information is useful l Develop specialised products such as climatic indices, heating and cooling degree days, and heat index l l Communication Climate Information to the rural farmers l Conduct regular survey's to identifty user needs and priorites l Develop drought monitoring and frost prediction tools l l KEY MESSAGES  At present, Bhutan does not have a national framework for climate services. The Climate Survey conducted as part of this report clearly highlights the significant demand from a range of sectors for accurate and effective climate services. Whilst observations and monitoring networks are being developed, there is an urgent need for developing research capacity in climate services, the development of a historical climate database and a real time observation network, developing climate services information systems, and the creation of a user interface platform through collaboration with other sector agencies.  Effective climate services need strong involvement by stakeholders from various disciplines. The user community need to engage information providers to understand what climate information is available, how to interpret it correctly as well as understanding its underlying assumptions and limitations. Current partnerships are non-existent or very weak in Bhutan. Strong partnerships need to be pursued through a formal MoU between DHMS and DoA and other sectors where needed.  A well trained workforce is essential to improving quality of services. There is an urgent need for capacity building (recruitment of staff) and capacity development (training of eisting staff) in both DHMS and the Department of Agriculture. However, before P a g e | 72 initiating any capacity building exercise, a Training Needs Assessment should be undertaken to ensure that high priority areas are targeted first. P a g e | 73 6 FARMER’S WEATHER AND CLIMATE RELATED INFORMATION NEEDS A key aim of the Technical Assistance (TA) project is to improve service delivery of agro-climatic information to farmers in Bhutan. To achieve this, a survey questionnaire “Farmer’s weather and climate related information needs” was designed and distributed to approximately 1200 random households in 5 districts (Dzongkhag). The samples represent different target communities, demographics, gender, education, agro-climatic zones and socio-economic conditions selected from a database of households used by the Department of Agriculture to conduct annual agricultural statistics survey. Total rural households in Bhutan is approximately 64,283. Therefore the sample size of 1200 used in this survey represents approximately 2 % of total households. The number of samples conducted in each district is shown in Table 6-1. The main objectives of the Climate Survey were to;  Assess the needs and priorities of farmer groups and households on accessing and using climate information in their decision-making.  Identify the frequency and mode of delivery of climate information to farmers.  Inform design of new climate services and projects.  Identify current gaps in climate service delivery; and,  Assess the impact of any climate related services and products provided to farmers to date. An outline of the survey questions and responses sought is given in Appendix A. The survey was arranged in five sections: Section A- was designed to obtain general demographic information on farmer age, education, size of farm and other farm characteristics. Section B- was designed to obtain information currently used by farmers in agricultural decision making. Section C- assessed the type of climate information that farmers have received from various sources and how it had been applied in farm decisions Section D- was designed to obtain what type of climate information is most useful, how it should be obtained, how often it should be delivered, and from which source(s) Section E- was designed to identify current constraints in agricultural production. P a g e | 74 Table 6-1. Distribution of climate survey in various districts (Dzongkhags) and sub-districts (Gewogs) of Bhutan. Dzongkhags Gewogs Altitude Range (masl) Total population Sample size Samtse 600 - 1400 434 53 Norboogang 600 - 3000 386 51 Samtse Namgaychhoeling 600 - 3400 375 51 Tashichhoeling 600 - 2200 296 47 Pemaling 600 - 1800 516 55 Drongtoed 1000 - 3000 236 44 Total 2243 300 Chhubu 1800 - 3400 354 37 Kabjisa 1800 - 3800 427 58 Punakha Shenga Bjime 1800 - 3000 153 27 Talo 1800 - 2600 186 28 Total 1120 150 Gaseshog Wom 1800 - 3800 80 22 Wangdue Phojikha 3000 - 3800 458 91 Phagul 1800 - 2600 85 37 Total 623 150 Bhur/Samtenling 400 - 1800 156 49 Shompangkha 200 - 400 420 45 Gelephu Gelephu 600 - 1000 319 46 Hilley/gakiling 1000 -1800 307 52 Jigmechhoeling 1000 - 1800 513 59 Umling 200 - 2600 287 50 Total 2002 300 Bartsham 1800 - 2200 291 45 Kanglung 1000 - 3000 454 54 Khaling 1200 - 3400 454 51 Trashigang Lumang 1400 - 2600 558 54 Merak 2200 - 4200 296 45 Samkhar 1400 - 2200 443 51 Total 2496 300 P a g e | 75 6.1 METHODOLOGY Data Collection A local consultant group (Wang Research and Consultancy) was engaged to conduct face-to-face interviews with each household. For the proper management and assessment of the data collection, consultative meetings were organized with Ministry of Agriculture and the local consulting group prior to the conduct of the survey. A Focus Group Discussions (FGDs) preceded each interview where communities were briefed on the objectives of the interview and also information on the content of the survey, such as understanding of climate change and variability, coping strategies for climate vulnerability, needs and gaps, community perceptions of climate adaptation and clarification of terms and terminologies used in the survey. Household interviews followed the focus group discussion, which aside from completing the questionnaire also collected metadata information relevant to the survey. Other related issues shared apart from the questionnaire were separately noted and discussed in this report. Data Management The data compiled from the field was encoded into the Excel database by the two data encoders. Data entered by one encoder was re-checked by a second encoder, to ensure high data quality. 6.2 RESULTS The survey results were analyzed using custom written software. Prior to analysis the data sets were cleaned to remove obvious errors in data entry or recording. Despite this the data set may contain erroneous outliers. To minimize this error, the use of median values instead of average is recommended for interpretation of data. 6.2.1 SECTION A - Demographic Information The purpose of this section was to obtain general demographic information on farmer age, education, size of farm and other farm characteristics. Of the 1200 households surveyed, 57 % were male and 43 % female. The average age of the respondent was 46.5 years ranging from 17 to 86 years with a median age of 47 years. Forty six % of the survey was completed by the head of household, 18 % by a child, 17 % by a relative and 14 % by a spouse (Figure 6-1). Seventy six % of the household surveyed have never been to school; 15.2 % have completed less than six years of schooling; 3.5% less than 10 years of schooling; 1.6% completed grade 12; and, only 1.2% had received a diploma or University degree (Figure 6-2). P a g e | 76 Figure 6-1. Distribution of household members who completed the questionnaire. 80 70 Percentage of Respondents 60 50 40 30 20 10 0 Never been to school Less than 6 years Completed year 10 Completed year 12 Diploma or University Level of Education Figure 6-2. Educational attainment of households. P a g e | 77 Figure 6-3. Age distribution by gender per household. On average the number of people living and working on the farm was 3.14 (1.53 male and 1.61 female). The age distribution per household was 2.2% of individuals between the age of 0 and 6; 5.8% of individuals between the age of 7 to 14; 83% of individuals between the age of 15 and 60; and, 9% older than 60 years old (Figure 6-3). Technology and Communication Access Almost half of the population (53%) have a television and (55%) a radio. Landline telephone and access to a computer with internet is very low with only 2.3% of the households surveyed having these facilities. However, almost 94.5% of the respondents possessed a mobile phone. Most farmers kept farm animals to supplement farm income with 86.3% having cows (average of 5.4 per household), and 19.3% possessing small livestock (average of 5.6) per household (Table 6-2). Average farm size is 0.8 acres of wetland, 0.3 acre of dryland, 0.4 acre of orchard and 1.2 acres of plantation (spices and nuts). On average 0.2 acres of land is left as fallow (Table 6-3). P a g e | 78 Table 6-2. Household possessions and farm characteristics. Yes No Maximum Average Median Radio 44.8 55.2 Television 53 47 Mobile Phone 94.5 5.5 Computer 2.3 97.7 Computer (with Internet and email access) 2.3 97.7 Telephone (land Line) 0.3 99.7 Pump (electric) 0.2 99.8 Pump (petrol or diesel) 0.4 99.6 Farm Animals Cows 86.3 13.7 60 5.4 4 Small livestock 19.3 80.7 70 5.6 4 Other 4.4 95.6 400 19.8 3 Access to Electricity 99 1 Access to Irrigation Water 57.2 42.8 Access to sealed road < 2 hours of walking 83.5 16.5 Table 6-3. Farm size by different land utilization. Land Utilisation Own Land Cultivated Land Left Fallow Leased-Out Leased-In (Acres) (Acres) (Acres) (Acres) Average 0.8 0.2 0.0 0.1 Chhuzhing (Wetland) Minimum 0.0 0.0 0.0 0.0 Maximum 19.0 7.7 4.0 4.0 Median 0.3 0.0 0.0 0.0 Average 0.3 0.0 0.0 0.0 Kamzhing (Dryland) Minimum 0.0 0.0 0.0 0.0 Maximum 10.0 3.0 2.5 1.3 Median 0.0 0.0 0.0 0.0 Average 0.4 0.0 0.0 0.0 Orchard Minimum 0.0 0.0 0.0 0.0 Maximum 7.0 8.0 3.3 1.0 Median 0.0 0.0 0.0 0.0 Average 1.2 0.3 0.0 0.0 Plantation (Spices and Nuts) Minimum 0.0 0.0 0.0 0.0 Maximum 19.0 19.0 3.0 1.5 Median 0.7 0.0 0.0 0.0 P a g e | 79 P a g e | 80 6.2.2 SECTION B- Information currently used by farmers in agricultural decision making. This section of the survey explored current sources of information that landholders use in decision making related to the following activities;  Land preparation  Timing of planting  Crop type  Fertilizer application  Timing of fertilizer or pesticide application  Timing of harvest  Quantity to plant A range of sources of information were presented and the respondents were asked to select the most relevant answers for each of type of above activities. The sources of information included; 1- Seasonal condition and or indigenous knowledge; 2- Traditional Cropping Calendar; 3- Personal experience; 4- Following other successful farmers; 5- Advice from Department of Department of Agriculture and extension staff; 6- Climate and weather forecast; 7- Village officials; 8- Advice from farmer group. For most activities the farmers relied on seasonal condition, traditional cropping calendar and personal experience for decision making. Advice from the Department of Agriculture and extension staff was sought by about 15% of farmers in relation to application of fertilizer and pesticide. Climate forecasts and advice from farmer groups recorded less than 1% of responses, reflecting a lack of service in these areas at present (Figure 6-4). P a g e | 81 Figure 6-4. Information currently used to inform decision making. 1- Seasonal condition and or indigenous knowledge; 2- Traditional Cropping Calendar; 3- Personal experience; 4- Follow other successful farmers; 5- Advice from Department of Agriculture; 6- Climate and weather forecast; 7 Village officials; 8 Advice from farmer’s group. 6.2.3 SECTION C- Climate and Weather Information Currently Received from Various Sources This section assessed what types of climate and weather information landholders have received in the past 5 years, from whom and delivery frequency, and whether or not farmers had changed any decision based on this information. When designing this question it was obvious that the majority of responses would be in the negative given that currently there is no climate forecasting service in operation in Bhutan with the exception of 1-2 days of temperature and precipitation forecasts issued by the Department of Meteorology and Hydrology. Nonetheless this question was included to establish a benchmark for future climate surveys in Bhutan. As expected, with the exception of 1-2 day forecasts where 65.7% of responses were in the affirmative, the responses for the remaining climate services were overwhelmingly (~ 90%) in the negative (Figure 6-5). P a g e | 82 Figure 6-5.Percentage of respondents receiving or not receiving various climate information. Of the small percentage (~10%) who responded “YES” to the questions, they were asked to specify the frequency and source of information they received. They were also asked if the information came with advice on how to use it and if they made a decision based on the information. Most of the respondents changed crop or variety based on the information they received. The results are summarized in Table 6-4. 6.2.4 SECTION D- Climate Information and Services In this section the respondent were asked what type of climate information was most important for their decision making. The type of climate service and information listed included;  Seasonal rainfall outlook (3 month forecast of rain)  Forecast of onset of monsoon  Forecast of end of monsoon  Temperature forecast (average or maximum)  Forecast of drought/dry spell  Weather forecasts (rain and temperature for 7 days ahead) P a g e | 83  Daily weather forecasts (rainfall and temperature)  Frost forecasts  Forecast of extreme events (cloudburst, heavy rain, cyclone, strong winds) They were asked to rate the level of importance of each climate information on a scale of 1 to 3 (1- not important and 3 - very important), the lead time by which they would like to receive the information, the preferred format and the source of information (for detail see the attached questionnaire). The responses are given in Table 6-5. The overall average score for the level of importance was 2.72/3. The highest level of importance with an average score of 2.94/3 was for forecast of seasonal rainfall outlooks, forecast of the onset of monsoon, weather forecast (rain and temperature for seven days ahead) and daily weather forecasts (rainfall and temperature). The lowest rating was for frost forecast with an average score of 2.3/3. The preferred format for information delivery was Television as stated by approximately 50% of respondents, followed by Radio (29%), Government Extension Officers and Village Elders (8%). Despite the high ownership of mobile phones in Bhutan, less than 1% of the respondent preferred receiving climate information my SMS. The preferred lead time was a week (50%) followed by the beginning of cropping calendar (30%) and a month (20%). The preferred source of information was a central weather agency and local extension officers (Table 6-4). Table 6-4. Type of climate information currently received by users, frequency, source, lead time and how this information has been used in decision making. The percentage reposes are those who indicated they have received information, not the whole sample. Results are color coded for easier cross referencing. Type of Climate Information Received Percentage of Responses % % % % % % Forecast of extreme events 92.2 DW 77.6 TRF 77.9 W D 60.1 No 69.5 No 62.2 Yes CV Forecast of monsoon onset 80.5 DS 71.0 TRFA 80.8 D M 77.7 No 85.6 No 25.4 Yes CV Forecast of monsoon duration 84.8 DS 60.1 RTAF 98.8 D W 85.9 No 87.6 No 12.5 Yes CV Forecast of rain for the following 2-3 months 76.7 DW 73.3 TRM C 93.0 D W 90.3 No 95.4 No 11.1 Yes CV Forecast of weather for 1-2 days 97.7 DW 72.7 TRFA 81.4 D W 64.3 No 71.0 No 23.3 Yes CV Forecasts of weather for 7-10 days 97.3 DW 75.0 TRFA 94.5 D W 98.6 No 98.6 No 2.2 Yes CV Warning of pest and crop diseases 63.4 SW 66.6 TRFA 76.9 W D 36.8 No 28.9 No 60.5 Yes PCV How often do you receive this informatipon? D-Daily; W-Weekly; S- Seasonaly From which source? T-Television;R Radio; F- Friends and Relatives; A- Agricultural Ext Staff; M- Met Office; C- Community Leader; N- NGO Lead time D- Days; W-Weeks Did the information come with advise on how to use it? Yes; No Did you use this information? Yes; No Did you change any decision? Yes; No What decision(s) you changed? C- Changed Crop; V- Changed variety; P- Used pesticide; M-Planted more; L- Planted less; E- Early harvest; F- More fertilizer P a g e | 85 Table 6-5 Type of climate information currently received by users, frequency, source, lead time and how this information has been used in decision making. The percentage reposes are those who indicated they have received information, not the whole sample. . Results are color coded for easier cross referencing. Climate Information Level imporatance Prefered Format in oder of imporatnce Prefered Lead time in oder of imporatnce Prefered Source in oder of imporatnce (weighted average) % followed by source in ( ) % followed by source in ( ) % followed by source in ( ) Seasonal rainfall outlook (3 month forecast of rain) 2.94 49.3(2) 30.1(1) 8.0 (4,5) 50.1 (2) 30.2 (3) 18.2 (1) 48.9 (2)30.1 (1) Forecast of onset of monsoon 2.71 48.6(2) 30.5(1) 8.0 (4,5) 50.4(2) 29.6(3) 18.9 (1) 73.5 (1) 25.4 (2) Forecast of end of monsoon 2.94 48.9(2) 30.1(1) 8.0 (4,5) 50.1 (2) 30.3 (3) 18.2 (1) 48.9 (2) 30.1 (1) Temperature forecast (average or maximum) 2.91 49.5(2) 29.5(1) 8.0 (4,5) 50.9(2) 30.2 (3) 17.6 (1) 49.5(2) 29.5 (1) Forecast of drought/dry spell 2.71 48.2(2) 29.1(1) 8.0 (4,5) 51.1 (2) 28.3 (3) 18.9 (1) 70.1 (1) 28.0 (2) Weather forecasts (rain and temperature for 7 days ahead) 2.94 49.1(2) 29.3(1) 8.0 (4,5) 50.1 (2) 30.2 (3) 18.2 (1) 48.9 (2)30.1 (1) Daily weather forecasts (rainfall and temperature) 2.94 49.5(2) 29.1 (1) 8.0 (4,5) 59.6(2) 28.3 (3) 17.1 (1) 49.7(2) 29.0 (1) Frost forecasts 2.34 48.0 (2) 28.7 (1) 8.0 (4,5) 52.5(2) 27.3 (3) 18.5 (1) 48.0(2) 28.7 (1) Forecast of extreme events (cloudburst, heavy rain, cyclone, strong winds) 2.92 49.5(2) 29.5 (1) 8.0 (4,5) 50.9(2) 30.3 (3) 16.9 (1) 49.1(2) 29.5 (1) 1- Not important 1- Radio 6- SMS 1 - Beginning of crop calender 1- Central Weather Authority 2- Of some importance 2- Television 7- NGO 2- A week before 2- Local Extension agency 3- Very important 3-Newspaper 8- Workshop 3- A month before 3- NGO 4- Village Elder 9- News letter 4- Expert farmers 5- Government Extension Office 5- Friends 6.2.5 SECTION E- Constraints in Agricultural Production This section was design to elicit information on the main risks (climate and non-climate) to agricultural production including the followings;  Farm labour  Excessive rain  Lack of input (fertilizer and pesticide)  Soil fertility  Drought  Land availability  Agricultural Extension Support  Occasional extreme temperatures  Market access  Commodity price  Access to water  Pests and diseases  Timely climate information  Timely weather information  Damage from wildlife The respondents were given a list of constraints and were asked to rate the level of importance for each on a scale of 1 to 5 (1-unimportant; 2-Slightly important; 3-Important; 4-Very important and 5- Extremely important). They were asked to provide this information in relation to Cereals, Vegetables, Fruits and Nuts and Horticultural crops. The data for Fruits and Nuts and Horticultural crops were aggregated together due to some difficulties in differentiation. The results shown in Figure 6-6 clearly show that the major constraint to agricultural production in Bhutan is damage from wildlife followed by the lack of timely climate information and weather forecasts particularly for cereal crops and access to water. Farm labour although important scored an average of (3/5) being the lowest of the list of constraints. Finally the households were asked if they have experienced any of the following major weather related disasters in the past 6 months, 12 months and 5 years and to rank the scale of damage on scale of 1 to 3 (1- no damage; 2- minor damage; 3- major damage);  Drought  Cyclone  Strong wind  Heatwave  Frost  Flood  Heavy rain P a g e | 87  Land slide The results of their responses in presented in Table 6-6. Most notable is heavy rain experienced by almost half the respondents (48.8%) with an average damage level of 1.4; followed by drought experienced by about 6.8% with an average damage level of 2.2. Figure 6-6. Constraints in agricultural production based on crop type. P a g e | 88 Table 6-6. Major climate related events experience by household in the past 5 years. Last 6 months Last 12 months Last 5 years % Damage % Damage % Damage Drought Yes 3.3 1.5 6.7 1.9 6.8 2.2 No 96.3 91.7 92.0 Cyclone Yes 1.7 2.0 2.2 1.9 1.9 2.0 No 98.1 96.5 96.8 Major Climate Related Events Strong wind Yes 1.7 2.0 2.2 1.9 1.9 2.0 No 98.1 96.5 96.8 Heatwave Yes 1.7 2.0 2.2 1.9 1.9 2.0 No 98.1 96.5 96.8 Frost Yes 9.0 1.2 8.5 1.2 9.4 1.3 No 90.8 90.7 89.9 Flood Yes 1.4 1.6 1.9 1.5 2.7 1.5 No 98.3 96.8 96.1 Heavy rain Yes 48.8 1.4 45.2 1.4 35.1 1.4 No 51.1 54.7 64.5 Land slide Yes 1.7 2.0 2.2 1.9 1.9 2.0 No 98.1 96.5 96.8 P a g e | 89 6.3 KEY MESSAGES The survey has clearly identified a strong need amongst farmers for climate services. At present the only climate information provided to the public is a 24 hour weather forecast issued by the Department of Hydro Met Services distributed through the local TV stations. Specifically the survey has identified a high level of demand for forecast of seasonal rainfall outlooks, forecast of the onset of monsoon, weather forecast (rain and temperature for seven days ahead) and daily weather forecasts (rainfall and temperature). The preferred channel for information delivery was through Television (50% of respondents), Radio (29%) and Government Extension Officers and Village Elders (8%). Despite the high ownership of mobile phones in Bhutan, less than 1% of the respondent preferred receiving climate information by SMS. The preferred source of information was central weather agency (DHMS) and the local extension officers. The survey was a first of its kind to be conducted in Bhutan and provides a benchmark for the development and communication of climate services in the country. P a g e | 90 7 RECOMMENDATIONS FOR DEVELOPMENT OF AGRO-MET ADVISORY SERVICES IN BHUTAN The primary objective of providing agro-meteorological services is to facilitate the Government’s capacity to manage climate risk and increase agricultural productivity. This goal is currently challenged by the overarching restriction of limited resources and institutional capacity for seasonal and long term climate risk assessments and subsequent development of climate risk management policies and programs. To achieve this goal the following recommendations are made to strengthen agro-meteorological services in Bhutan: The primary objective of providing agro-meteorological services is to facilitate the Government’s capacity to manage climate risk and increase agricultural productivity. This goal is currently challenged by the overarching restriction of limited resources and institutional capacity for seasonal and long term climate risk assessments and subsequent development of climate risk management policies and programs. To achieve this goal the following recommendations are made to strengthen agro-meteorological services in Bhutan: DEVELOPING A NATIONAL FRAMEWORK FOR CLIMATE SERVICES IN BHUTAN The climate survey conducted as part of this study clearly highlights a significant demand from a range of sectors for climate services. Whilst observations and monitoring networks are in place and being developed, there is an urgent need through collaboration with other sector agencies for developing research capacity for climate services, development of a historical climate database and real time observation network, developing climate services information systems, and a user interface platform. DEVELOPING PARTNERSHIPS BETWEEN DHMS AND DOA AND OTHER SECTORS Effective climate services need strong involvement by stakeholders from various disciplines. A Memorandum of Understanding will provide a formal channel for engagement to identify climate needs of various sectors and to develop products specific to the needs of these sectors (climate service delivery). To achieve this it is recommended that an externally led visioning process should be conducted to ensure that there is mutual understanding and alignment of objectives across organizations (DoA and DHMS). This process would pull together key personnel across the board to ensure that issues raised in this report are understood by all and a collective goal is established whilst at the same time maintaining the integrity and value of each of the parts within the whole. This will ensure efficiency and establish a shared working vision as the basis for effective collaboration and communication. STRENGTHEN CAPABILITY AND HUMAN RESOURCE CAPACITY through capacity building (recruitment of staff) and capacity development (training of existing staff) in both DHMS and Department of Agriculture. In DHMS capacity needs to be improved In the areas of weather and climate forecasting, numerical weather prediction, interpretation of climate products including P a g e | 91 limitations and uncertainties, identification of user needs, development of sector specific products, and partnership creation and communication. A comprehensive capacity building initiative is also needed within the DoA in the following key areas:  Development and training in the use of climate concepts for the agricultural sector  Development of analytical and crop simulation capability  Vulnerability assessments of agricultural production due to impacts from climate variability and climate change. Before initiating any capacity building exercise, a training needs assessment should be undertaken to ensure high priority areas are targeted first. P a g e | 92 8 REFERENCES Anders AM, Roe GH, Hallet B, Montgomery DR, Finnegan NJ, Putkonen J (2006) Spatial patterns of precipitation and topography in the Himalaya. Geological Society of America Special Papers 398:39-53 Baillie I, Norbu C (2004) Climate and other factors in the development of river and interfluve profiles in Bhutan, Eastern Himalayas. Journal of Asian Earth Sciences 22 (5):539-553 Banskota M, Jodha N, Gupta A, Partap T, Ura K (1992) Indigenous Farming Technologies and Environment: Experiences in Bhutan. Sustainable Mountain Agriculture 2:539-568 Bhutiyani M, Kale VS, Pawar N (2007) Long-term trends in maximum, minimum and mean annual air temperatures across the Northwestern Himalaya during the twentieth century. Climatic Change 85 (1-2):159-177 Bookhagen B, Burbank DW (2010) Toward a complete Himalayan hydrological budget: Spatiotemporal distribution of snowmelt and rainfall and their impact on river discharge. Journal of Geophysical Research: Earth Surface 115 (F3) Cai W, Cowan T and Sullivan A 2009 Recent unprecedentedskewness towards positive Indian Ocean dipole occurrence0 sand their impact on Australian rainfall Geophys. Res. Lett.36 L11705 Choudhury B, Das A, Ngachan S, Slong A, Bordoloi L, Chowdhury P (2012) Trend analysis of long term weather variables in mid altitude Meghalaya, North-East India. Journal of Agricultural Physics 12 (1):12-22 Clark CO, Cole JE, Webster PJ (2000) Indian Ocean SST and Indian summer rainfall: Predictive relationships and their decadal variability. Journal of Climate 13 (14):2503-2519 Commission NE (2000) First Greenhouse Gas Inventory. Thimpu. Bhutan. Datta S (2013) Impact of climate change in Indian horticulture-a review. International J of Sci, Environ and Tech 2 (4):661-671 Dhar O, Nandargi S (2000) A study of floods in the Brahmaputra basin in India. International journal of climatology 20 (7):771-781 Dwivedi S, Goswami B, Kucharski F (2015) Unraveling the missing link of ENSO control over the Indian monsoon rainfall. Geophysical Research Letters 42 (19):8201-8207 Eguchi T. (1991): Regional and seasonal change in precipitation in Bhutan – analysis of daily precipitation in 1988. In: Life Zone Ecology of the Bhutan Himalaya II (M.Ohsawa ed.), Laboratory of Ecology, Chiba University, Japan; pp. 1-20. P a g e | 93 Gadgil S, Joseph P (2003) On breaks of the Indian monsoon. Journal of Earth System Science 112 (4):529-558 Gershunov A, Schneider N, Barnett T (2001) Low-frequency modulation of the ENSO-Indian monsoon rainfall relationship: Signal or noise? Journal of Climate 14 (11):2486-2492 Goswami B, Xavier PK (2005) Dynamics of “internal” interannual variability of the Indian summer monsoon in a GCM. Journal of Geophysical Research: Atmospheres 110 (D24) Gupta A.K. & Ura K. (1992): Indigenous Farming Technologies and Environment: Experiences in Bhutan. In: Sustainable Mountain Agriculture (N.S. Jodha, M. Hoy A, Katel O, Thapa P, Dendup N, Matschullat J (2015) Climatic changes and their impact on socio-economic sectors in the Bhutan Himalayas: an implementation strategy. Regional Environmental Change:1-15 Hulme M (1992) A 1951–80 global land precipitation climatology for the evaluation of general circulation models. Climate Dynamics 7 (2):57-72 Hulme M, New M (1997) Dependence of large-scale precipitation climatologies on temporal and spatial sampling. Journal of Climate 10 (5):1099-1113 IPCC, 2013: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp, doi:10.1017/CBO9781107415324 IPCC, 2014: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Regional Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Barros, V.R., C.B. Field, D.J. Dokken, M.D. Mastrandrea, K.J. Mach, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 688 pp Kothyari U, Singh V (1996) Rainfall and temperature trends in India. Hydrological processes 10 (3):357-372 Krishnakumar K, Rao GP, Gopakumar C (2009) Rainfall trends in twentieth century over Kerala, India. Atmospheric environment 43 (11):1940-1944 Krishnan R, Sugi M (2003) Pacific decadal oscillation and variability of the Indian summer monsoon rainfall. Climate Dynamics 21 (3-4):233-242 P a g e | 94 Kumar KK, Rajagopalan B, Cane MA (1999) On the weakening relationship between the Indian monsoon and ENSO. Science 284 (5423):2156-2159 Kumar KK, Rajagopalan B, Hoerling M, Bates G, Cane M (2006) Unravelling the mystery of Indian monsoon failure during El Niño. Science 314 (5796):115-119 Lhendup, P. 2012. Integration of Climate Adaptation into Development and Conservation Planning in Bhutan: Issue Identification and Recommendations. Adaptation Knowledge Platform, Partner Report Series No. 1,. Stockholm Environment Institute, Bangkok. Miehe G, Winiger M, Böhner J, Yili Z (2001) The Climatic Diagram Map of High Asia: Purpose and Concepts (Klimadiagramm-Karte von Hochasien. Konzept und Anwendung). Erdkunde:94-97 Nandargi S, Dhar O (2012) Extreme rainstorm events over the Northwest Himalayas during 1875-2010. Journal of Hydrometeorology 13 (4):1383-1388 NEC (National Environment Commission). 2000. Initial national communication under the United Nations Framework Convention on Climate Change. Thimphu: NEC, Royal Government of Bhutan. http://unfccc.int/resource/docs/natc/bhunc1.pdf Neuhoff, Daniel; Tashi, Sonam; Rahmann, Gerold; Denich, Manfred). Organic agriculture in Bhutan: potential and challenges. Norbu C., Baillie I.C., Dorji Tshering, Dorji Tsheten, Tamang H.B., Tshering K. and Hutcheon A.A. (2003a): A provisional physiographic zonation of Bhutan. Journal of Bhutan Studies 8: 54-87. Norbu C., Baillie I.C., Dema K., Jamyang, Dema Y., Tamang H.B., Tshering K., Turkelboom F. and Norbu S. (2003b): Types of land degradation in Bhutan. Journal of Bhutan Studies 8: 88-114. Ohmura A (2012) Enhanced temperature variability in high-altitude climate change. Theoretical and Applied Climatology 110 (4):499-508 Patle G, Libang A (2014) Trend analysis of annual and seasonal rainfall to climate variability in North-East region of India. Journal of Applied and Natural Science 6 (2):480-483 Pervez MS, Henebry GM (2015) Spatial and seasonal responses of precipitation in the Ganges and Brahmaputra river basins to ENSO and Indian Ocean dipole modes: implications for flooding and drought. Natural Hazards and Earth System Science 15 (1):147-162 Qin J, Yang K, Liang S, Guo X (2009) The altitudinal dependence of recent rapid warming over the Tibetan Plateau. Climatic Change 97 (1-2):321-327 Rao, Y.P., 1981. The climate of the Indian subcontinent. Climates of Southern and Western Asia, World Survey of Climatology, Smithsonian Institute, Washington, DC. P a g e | 95 Rikiishi K, Nakasato H (2006) Height dependence of the tendency for reduction in seasonal snow cover in the Himalaya and the Tibetan Plateau region, 1966–2001. Annals of Glaciology 43 (1):369-377 Rohrer M, Salzmann N, Stoffel M, Kulkarni AV (2013) Missing (in-situ) snow cover data hampers climate change and runoff studies in the Greater Himalayas. Science of the Total Environment 468:S60-S70 Sano M, Tshering P, Komori J, Fujita K, Xu C, Nakatsuka T (2013) May–September precipitation in the Bhutan Himalaya since 1743 as reconstructed from tree ring cellulose δ18O. Journal of Geophysical Research: Atmospheres 118 (15):8399-8410 Shrestha KL (2005) Global change impact assessment for Himalayan mountain regions for environmental management and sustainable development. GLOBAL ENVIRONMENTAL RESEARCH-ENGLISH EDITION- 9 (1):69 Shukla J, Paolino DA (1983) The Southern Oscillation and long-range forecasting of the summer monsoon rainfall over India. Monthly Weather Review 111 (9):1830-1837 Torrence C, Webster PJ (1999) Interdecadal changes in the ENSO-monsoon system. Journal of Climate 12 (8):2679-2690 Van Oldenborgh GJ, Burgers G (2005) Searching for decadal variations in ENSO precipitation teleconnections. Geophysical Research Letters 32 (15) Van der Poel, P., Tshering, T., 2003. Debris flows and floods in Bumdeling and Yangtse. In: Jamtsho, K., Hansen, P.K. (Eds.), Consultation on Land Management and Sustainable Water Resources, Ministry of Agriculture, Thimphu. Wang P, Clemens S, Beaufort L, Braconnot P, Ganssen G, Jian Z, Kershaw P, Sarnthein M (2005) Evolution and variability of the Asian monsoon system: state of the art and outstanding issues. Quaternary Science Reviews 24 (5):595-629 Webster PJ, Magana VO, Palmer T, Shukla J, Tomas R, Yanai M, Yasunari T (1998) Monsoons: Processes, predictability, and the prospects for prediction. Journal of Geophysical Research: Oceans 103 (C7):14451-14510 Whiteman CD (2000) Mountain meteorology: fundamentals and applications. Oxford University Press, Yasunari T (1990) Impact of Indian monsoon on the coupled atmosphere/ocean system in the tropical Pacific. Meteorology and Atmospheric Physics 44 (1-4):29-41 Young A. (1994): Land degradation in South Asia: its severity, causes, and effects upon the people. World Soil Resources Report 78, UNDP/UNEP/FAO, Rome, Italy P a g e | 96 Zhang Y, Li T, Wang B (2004) Decadal change of the spring snow depth over the Tibetan Plateau: the associated circulation and influence on the east Asian summer monsoon*. Journal of Climate 17 (14):2780-2793 P a g e | 97 9 APPENDIX – A Climate Survey Questionnaire To be inserted P a g e | 98