Detailed Climate Change Assessment

Climate Change and Disaster Resilient Water Resources Sector Project (RRP KGZ 51081-002)

DETAILED CLIMATE CHANGE ASSESSMENT

Table of contents

EXECUTIVE SUMMARY 1 I. COUNTRY BACKGROUND AND ITS CLIMATE 3

A. BACKGROUND 3 B. THE CLIMATE OF KYRGYZSTAN 4 C. CURRENT CLIMATE VARIABILITY AND CLIMATE CHANGE 6 II. BRIEF INTRODUCTION OF GLOBAL CLIMATE MODELS USED IN THIS CVRA STUDY 16

A. MODEL GENERATED CLIMATE DATA 16 B. DATA ANALYSIS TOOLS 20 C. DATA AND CHOICE OF METRICS 21 D. VALIDATION OF GLOBAL CLIMATE MODELS TO BASELINE OBSERVED (CRU, UK) CLIMATOLOGY 22 E. KYRGYZSTAN CLIMATE CHANGE ISSUES, POLICIES AND INVESTMENTS 28 III. CLIMATE CHANGE PROJECTIONS AND EXTREMES 29

A. KYRGYZSTAN’S MEAN CLIMATE & SEASONAL VARIABILITY –KEY ISSUES 29 B. SIMULATION OF FUTURE CLIMATE OF KYRGYZSTAN & SELECTED OBLASTS 31 C. UNCERTAINTIES IN FUTURE PROJECTIONS 69 D. SUMMARY - KEY FINDINGS ON CLIMATE CHANGE SCENARIOS 69 IV. IMPLICATIONS OF PROJECTED CHANGES IN CLIMATE ON THE RISKS AT PROPOSED SUB-PROJECT LEVEL AND ASSOCIATED VULNERABILITIES 72

A. THE SHORT-LISTED SUB-PROJECT: INTRODUCTION 72 B. CLIMATE RISK SCREENING AND VULNERABILITY ASSESSMENT OF THE SHORT-LISTED SUB-PROJECT 73 V. CLIMATE CHANGE & DISASTER RISK FRAMEWORK FOR CLIMATE RESILIENCE 89

A. MAINSTREAMING FRAMEWORK FOR CCA AND DRR 89 B. POTENTIAL ADAPTATION RESPONSES / STRATEGIES 90 C. REHABILITATION OF EXISTING HYDROMET MONITORING NETWORK AND REQUIREMENTS OF ADDITIONAL WEATHER STATIONS IN KYRGYZ REPUBLIC 97

Tables

TABLE 1: AVERAGE MONTHLY TEMPERATURES AND PRECIPITATION - BISHKEK 5 TABLE 2: AVERAGE MONTHLY TEMPERATURES AND PRECIPITATION - NARYN 5 TABLE 3: THE CMIP5 EARTH SYSTEM MODELS WITH A FAIRLY HIGH DEGREE OF SIMULATION SKILL OVER CENTRAL ASIA 19 TABLE 4: LIST OF CORE CLIMATE INDICATORS 22 TABLE 5: BASELINE CLIMATOLOGY OF AVERAGE MONTHLY MEAN MAXIMUM TEMPERATURE (DEGREE C) OVER KYRGYZSTAN AS OBSERVED (CRU CLIMATOLOGY) AND SIMULATED BY THE MODELS (RMSE IS THE ROOT MEAN SQUARE ERROR BETWEEN THE OBSERVED AND SIMULATED SURFACE AIR TEMPERATURES) 26 TABLE 6: BASELINE CLIMATOLOGY OF AVERAGE MONTHLY MEAN MINIMUM TEMPERATURE (DEGREE C) OVER KYRGYZSTAN AS OBSERVED (CRU CLIMATOLOGY) AND SIMULATED BY THE MODELS (RMSE IS THE ROOT MEAN SQUARE ERROR BETWEEN THE OBSERVED AND SIMULATED MINIMUM SURFACE AIR TEMPERATURES 27 TABLE 7: BASELINE CLIMATOLOGY OF AVERAGE MONTHLY PRECIPITATION (MM) OVER KYRGYZSTAN AS OBSERVED (CRU CLIMATOLOGY) AND SIMULATED BY THE MODELS (RMSE IS THE ROOT MEAN SQUARE ERROR BETWEEN THE OBSERVED AND SIMULATED PRECIPITATION) 27 TABLE 8: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AREA-AVERAGED OVER KYRGYZSTAN AT THREE TIME SLICES UNDER RCP 8.5 AND RCP4.5 PATHWAYS 34 TABLE 9: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER THE SEVEN OBLASTS OF KYRGYZSTAN DURING 2020S FOR RCP 8.5 PATHWAY 38 TABLE 10: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER THE SEVEN OBLASTS OF KYRGYZSTAN DURING 2050S FOR RCP 8.5 PATHWAY 38 TABLE 11: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER THE SEVEN OBLASTS OF KYRGYZSTAN DURING 2080S FOR RCP 8.5 PATHWAY 39 TABLE 12: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER THE SEVEN OBLASTS OF KYRGYZSTAN DURING 2020S FOR RCP 4.5 PATHWAY 40 TABLE 13: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER SELECTED SEVEN OBLASTS OF KYRGYZSTAN DURING 2050S FOR RCP 4.5 PATHWAY 41 TABLE 14: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED WARMING AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER SELECTED SEVEN OBLASTS OF KYRGYZSTAN DURING 2080S FOR RCP 4.5 PATHWAY 41 TABLE 15: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED DIURNAL TEMPERATURE RANGE AS INFERRED FROM FOUR MODEL ENSEMBLES AREA AVERAGED OVER KYRGYZSTAN DURING THE THREE TIME SLICES FOR RCP 8.5 AND 4.5 PATHWAYS 42 TABLE 16: MONTHLY, SEASONAL AND ANNUAL MEANS OF PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (ΔP, PERCENT CHANGE) OVER KYRGYZSTAN AT THREE TIME SLICES FOR RCP 8.5 AND RCP 4.5 PATHWAYS 43 TABLE 17: MONTHLY, SEASONAL AND ANNUAL MEAN PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (ΔP, PERCENT CHANGE) OVER SELECTED OBLASTS DURING 2020S UNDER RCP 8.5 PATHWAY 47 TABLE 18: MONTHLY, SEASONAL AND ANNUAL MEAN PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (ΔP, PERCENT CHANGE) OVER SELECTED OBLASTS DURING 2050S UNDER RCP 8.5 PATHWAY 48 TABLE 19: MONTHLY, SEASONAL AND ANNUAL MEAN PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (ΔP, PERCENT CHANGE) OVER SELECTED OBLASTS DURING 2080S UNDER RCP 8.5 PATHWAY 48 TABLE 20: MONTHLY, SEASONAL AND ANNUAL MEAN PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (ΔP, PERCENT CHANGE) OVER SELECTED OBLASTS DURING 2020S UNDER RCP 4.5 PATHWAY 49 TABLE 21: MONTHLY, SEASONAL AND ANNUAL MEAN PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (ΔP, PERCENT CHANGE) OVER SELECTED OBLASTS DURING 2050S UNDER RCP 4.5 PATHWAY 49 TABLE 22: MONTHLY, SEASONAL AND ANNUAL MEAN PROJECTED CHANGE IN AREA AVERAGED PRECIPITATION (Δ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

Figures

FIGURE 1: GEOGRAPHICAL LOCATION OF THE REPUBLIC OF KYRGYZSTAN 4 FIGURE 2: MAJOR METEOROLOGICAL MONITORING NETWORK EXISTING IN KYRGYZSTAN (AS OF FEB, 2018) 7 FIGURE 3: MONTHLY MEAN TEMPERATURES OVER KYRGYZSTAN IN 2015 ARE WARMER RELATIVE TO 1961- 1990 MEAN CLIMATOLOGY (AS INFERRED FROM CRU CLIMATOLOGICAL DATA SET) 8 FIGURE 4: A COMPARISON OF MONTHLY MEAN PRECIPITATION OVER KYRGYZSTAN IN 2015 RELATIVE TO 1961- 1990 MEAN CLIMATOLOGY (AS INFERRED FROM CRU CLIMATOLOGICAL DATA SET) 8 FIGURE 5: HISTORICAL TRENDS IN ANNUAL AND SEASONAL MEAN SURFACE AIR TEMPERATURE ANOMALIES AVERAGED ACROSS KYRGYZSTAN DURING THE 1951-2015 (BASED ON CRU CLIMATOLOGICAL DATA SETS) 9 FIGURE 6: HISTORICAL TRENDS AND INTER-ANNUAL VARIABILITY IN ANNUAL MEAN TEMPERATURE ANOMALIES OBSERVED OVER SEVEN OBLASTS OF KYRGYZSTAN 11 FIGURE 7: HISTORICAL TRENDS AND INTER-ANNUAL VARIABILITY IN ANNUAL AND SEASONAL PRECIPITATION OBSERVED OVER KYRGYZSTAN 12 FIGURE 8: HISTORICAL TRENDS AND INTER-ANNUAL VARIABILITY IN OBSERVED ANNUAL MEAN PRECIPITATION ANOMALIES IN EACH OF THE SEVEN OBLASTS OF KYRGYZSTAN 13 FIGURE 9: AVERAGE ANNUAL DISTRIBUTION (%) OF REPORTED DISASTERS IN KYRGYZSTAN (SOURCE: INTERNATIONAL DISASTER DATABASE - WWW.EMDAT.BE, 2015) 15 FIGURE 10: A COMPARISON OF MODEL SIMULATED AND OBSERVED CLIMATOLOGICAL (1961-1990) AVERAGE MONTHLY MEAN SURFACE AIR TEMPERATURES AVERAGED OVER THE SEVEN OBLASTS IN KYRGYZSTAN 23 FIGURE 11: A COMPARISON OF MODEL SIMULATED AND OBSERVED CLIMATOLOGICAL (1961-1990) AVERAGE MONTHLY MEAN PRECIPITATION AVERAGED OVER KYRGYZSTAN AND IN EACH OF ITS SEVEN OBLASTS 26 FIGURE 12: GEOGRAPHICAL LOCATION OF KYRGYZSTAN AND ITS SEVEN OBLASTS WITH ELEVATIONS ABOVE SEA LEVEL 31 FIGURE 13: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL MEAN SURFACE AIR TEMPERATURE WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2020S UNDER RCP 8.5 PATHWAY 35 FIGURE 14: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL MEAN SURFACE AIR TEMPERATURE WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2050S UNDER RCP 8.5 PATHWAY 35 FIGURE 15: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL MEAN SURFACE AIR TEMPERATURE WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2080S UNDER RCP 8.5 PATHWAY 36 FIGURE 16: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL MEAN SURFACE AIR TEMPERATURE WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2020S UNDER RCP 4.5 PATHWAY 37 FIGURE 17: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL MEAN SURFACE AIR TEMPERATURE WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2050S UNDER RCP 4.5 PATHWAY 37 FIGURE 18: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL MEAN SURFACE AIR TEMPERATURE WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2080S UNDER RCP 4.5 PATHWAY 37 FIGURE 19: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL PRECIPITATION (PERCENT) WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2020S UNDER RCP 8.5 PATHWAY 44 FIGURE 20: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL PRECIPITATION (PERCENT) WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2050S UNDER RCP 8.5 PATHWAY 44 FIGURE 21: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL PRECIPITATION (PERCENT) WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2080S UNDER RCP 8.5 PATHWAY 45 FIGURE 22: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL PRECIPITATION (PERCENT) WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2020S UNDER RCP 4.5 PATHWAY 45 FIGURE 23: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL PRECIPITATION (PERCENT) WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2050S UNDER RCP 4.5 PATHWAY 46 FIGURE 24: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN ANNUAL PRECIPITATION (PERCENT) WITH RESPECT TO BASELINE PERIOD (1961-90) OVER KYRGYZSTAN BY 2080S UNDER RCP 4.5 PATHWAY 46 FIGURE 25: SPATIAL DISTRIBUTION OF PROJECTED NUMBER OF HOTTER DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 8.5 PATHWAY 52 FIGURE 26: SPATIAL DISTRIBUTION OF PROJECTED NUMBER OF HOTTER DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 4.5 PATHWAY 53 FIGURE 27: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN NUMBER OF FROST DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 8.5 PATHWAY 59 FIGURE 28: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN NUMBER OF FROST DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 4.5 PATHWAY 59 FIGURE 29: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN NUMBER OF DRY DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 8.5 PATHWAY 62 FIGURE 30: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN NUMBER OF WET DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 8.5 PATHWAY 63 FIGURE 31: SPATIAL DISTRIBUTION OF PROJECTED CHANGE IN NUMBER OF VERY WET DAYS OVER KYRGYZSTAN LIKELY BY 2080S UNDER RCP 8.5 PATHWAY 63 FIGURE 32: OVERVIEW OF PILOT AREA SHOWING IRRIGATED AREA, HYDROPOST AND METEOPOST LOCATIONS 79 FIGURE 33: SYSTEM OVERVIEW OF CATCHMENT AREA 80 FIGURE 34: TIME SERIES OF MONTHLY FLOWS (M3/S) AT TENTJAK CHARVAK FROM 1933-1991 82 FIGURE 35: OVERVIEW OF CATCHMENT AREA SHOWING "R. TENTEK-SAI - KISCHLAK TSCHARBAK" HYDROPOST (HP) AS WELL AS THE RELATED CATCHMENT OUTLINE IN GREEN, PRAVAYA VETKA OFF-TAKE LOCATION WITH THE ADDITIONAL CATCHMENT IN RED, AS WELL AS THE RIVER NETWORK IN BLUE. 82 FIGURE 36: MONTHLY TEMPERATURE (°C) AND RAINFALL (MM) AT JALALABAD BETWEEN 1947 AND 2005 ON THE LEFT AND RIGHT AXIS RESPECTIVELY 84 FIGURE 37: LIKELY CHANGES IN SURFACE RUNOFF CONTRIBUTED BY WATER FLOW DUE TO CHANGES IN PRECIPITATION FOR RCP 8.5 AND RCP 4.5 PATHWAYS 89 FIGURE 38: A CONCEPTUAL FRAMEWORK FOR MAINSTREAMING CCA AND DRR IN WATER AND FOOD SECURITY WITHIN THE KYRGYZ REPUBLIC 90 FIGURE 39: THE HYDROLOGICAL POSTS REQUIRING REPAIR AND RESTORATION 99

List of Acronyms and Abbreviations

ADB Asian Development Bank AR5 Fifth Assessment Report of IPCC BCSD Bias-Corrected Spatial Disaggregation method CDF Cumulative Distribution Function CDI Climate Data Interface CDO Climate Data Operators Tool CMIP5 Coupled Model Inter-comparison Project Phase 5 CRU Climate Research Unit, University of East Anglia, Norwich (UK) CRVA Climate Risk and Vulnerability Assessment COP Conference of Parties CSA Climate Smart Agriculture DNA Designated National Authority DTR Diurnal Temperature Range DWIS Development of a Disaster Risk and Water Resources Information System ESM Earth System Model FAO-AQUAstat Food and Agriculture Organization - global information system on water and agriculture GCM Global Climate Model GEF Global Environmental Facility GFDRR Global Facility for Disaster Reduction and Recovery GHG Greenhouse gases GIS Geographical Information System INDC Intended Nationally Determined Contribution IPCC Intergovernmental Panel on Climate Change KYRGYZ-HYDROMET National Agency of hydro-meteorology under the Ministry of Emergency Situations of the Kyrgyz Republic MDG Millennium Development Goals MEA Millennium Ecosystem Assessment NASA-NEX NASA Earth Exchange Global Daily Downscaled Datasets NC National Communication to UNFCCC NDCs Nationally Determined Contributions ODA Official Development Assistance RCPs Representative Concentration Pathways RSE Republican state-owned Enterprise SDGs Sustainable Development Goals SMEs Small and Medium Enterprises SRES Special Report on Emissions Scenarios TRTA TRansaction Technical Assistance UEA University of East Anglia, Norwich, UK UNCSD United Nations Commission on Sustainable Development UNDP United Nation Development Programme UNFCCC United Nations Framework Convention on Climate Change WEAP Water Evaluation and Planning Tool WMO World Meteorological Organization

Executive Summary 1. This report covers the findings of the analysis on climate change scenarios within the seven oblasts of Kyrgyzstan and its implications on the sub-project identified for rehabilitation of irrigation infrastructures and mitigation of hydrological disasters and assessment of risks and vulnerability as outputs under of the project. 3. The Kyrgyz Republic is highly vulnerable to shocks associated with climate change. Recent years here have been marked by changing rainfall patterns, heavy snowfall events, and landslides – all of which have had a negative impact on the livelihoods and food security of vulnerable people throughout the country. Higher temperatures eventually reduce yields of desirable crops while encouraging weed and pest proliferation. Changes in precipitation patterns increase the likelihood of short-run crop failures and long-run production declines. Rising temperatures and changes in rainfall patterns have direct effects on crop yields, as well as indirect effects through changes in irrigation water availability. Given the current uncertainty about location-specific effects of climate change, good development policies and programs are the best climate change adaptation investments. Extension services that specifically address climate change adaptation include disseminating local cultivars of drought-resistant crop varieties, improved crop management systems, and sharing information to facilitate implementing cost effective measures to combat the effects of climate change. Climate change also places new and more challenging demands on additional investments in irrigation infrastructure specially to improve the efficiency of water use. 4. Salient features of current climate and its inter-annual variability and the projected annual mean climate change over Kyrgyzstan and its selected oblasts for the future based on our analysis suggest: (a) An increase in annual mean temperatures of 1.2°C with respect to the baseline period of 1961-1990 at an average rate of 0.26 to 0.37°C per decade. (b) The observed increase in annual mean temperatures was most pronounced in Chuy, Naryn and Talas oblasts. (c) More frequent heat waves (when the daily maximum temperature was above 90th percentile for at least six consecutive days) have also been observed in parts of Kyrgyzstan during the past three decades. (d) The frequency of frost days with daily minimum temperature below 0ºC is reported to be decreasing in Kyrgyzstan. (e) Precipitation in Kyrgyzstan and each of the seven oblasts is characterized by high variability on inter‐annual and inter‐decadal time scales. (f) No statistically significant decreases or increases in annual mean precipitation have been observed between 1951 and 2015 in Kyrgyzstan but an increase in winter and a marginal decrease in summer precipitation during summer are noted. (g) Some unusually high intensity precipitation spells have occurred in the dry season in recent years (2000‐2015), but no consistent trend is discernible. 5. Beyond the middle of this century, temperature changes will be determined primarily by the rate and amount of various emissions released by human activities, as well as by the response of the Earth’s climate system to those emissions. Efforts to rapidly and significantly reduce emissions of heat-trapping gases (e.g., RCP 4.5 pathway) might still limit mean warming to 2ºC relative to the 1901-1960 average. However, significantly greater temperature increases are

expected if emissions follow higher scenarios associated with continuing growth in the use of fossil fuels (e.g., RCP 8.5 pathway). Key findings include: (a) Mean annual temperatures are projected to increase by 2.5°C by the 2020s, 4.5°C by the 2050s and 7.0°C by the 2080s under RCP 8.5 concentration pathway. Under RCP 4.5 concentration pathway, the projected rise in surface temperature is expected to be 2.2°C by the 2020s, 3.7°C by the 2050s and 4.4°C by the 2080s. (b) The projected average warming over Kyrgyzstan at all time slices during this century is higher during the summer season (least during spring season). The peak warming averaged over Kyrgyzstan is expected to occur in August month by the middle of the 21st century and beyond. (c) Within the four seasons, autumn is projected to be warmest around 2020s in Batken and Osh oblasts, but the peak warming occurs during summer season in the remaining five oblasts. (d) The peak warming over the seven oblasts shifts to summer season with annual mean warming of about 4.5°C in Batken and Osh oblasts around 2050s. During the 2080s, annual mean warming of about 7.3°C in Osh oblast and 7.2°C in Batken oblast is expected. (e) Peak warming continues to occur during the summer season in all the seven oblasts in 2080s. (f) Kyrgyzstan could experience as many as 7 additional hot days by 2020s, 16 days by 2050s and over 25 days in a year by 2080s under the RCP 8.5 pathway. (g) The hotter days would likely only be marginally more frequent during the summer season as compared to the winter season (peak in number of hot days would be in July/August). (h) The spring season could experience relatively fewer hot days than any other season. Kyrgyzstan is likely to experience around 15 hotter days in a year by 2080s under the RCP 4.5 pathway. (i) The frequency of dry days over the seven oblasts in Kyrgyzstan could increase by between 19 to 23 days per year by the 2050s and between 33 to 40 days by end of this century under RCP 8.5 pathway. This suggests that the projected warming could accelerate land surface drying with increases in the potential incidence and severity of droughts. (j) On an annual mean basis, the projected decrease in precipitation averaged over the country is 3% of the baseline value by the 2020s and 2050s and a decrease in precipitation of 5% by the 2080s (under the RCP 8.5 pathway). (k) Within the four seasons, the model results suggest a modest increase in precipitation of about 19% during the spring season by the end of this century. However, during the summer season, a significant decline of 38% in precipitation is projected which could lead to drought conditions in some of the drought prone areas of Kyrgyzstan. (l) Under RCP 4.5 pathway, this decline in summer precipitation is restricted to only 16% while on annual mean basis, an increase in precipitation by 9% could be anticipated. During autumn season, this increase is most significant at about 23% by 2080s. (m) On an annual basis, an increase in precipitation of about 9% could occur by the end of the 21st century under the RCP 4.5 pathway. A decline in summer precipitation is restricted to only 16%. During spring season, the increase is most significant at about 23% by 2080s. 6. In such a future scenario, recurrence of floods in flood prone areas of Kyrgyzstan is expected in spring when snow melt begins. Following the observed trends over recent decades,

more precipitation is expected to fall as heavier precipitation events in future. In the Kyrgyz Republic, there could be fewer days with precipitation, but the wettest days could be wetter. A 16% increase in maximum precipitation intensity on an annual basis may occur under RCP 8.5 pathway and about 21% under RCP 4.5 pathway by the end of this century. Hence, the projected increases in heavy precipitation are expected to be greater under lower emission scenarios. Changes in peak precipitation intensity are significantly different in each oblast for each of the two RCP pathways thus highlighting the spatial as well as temporal variability in the trends of precipitation-related extremes. 7. A decrease in the number of frost days in Kyrgyzstan is projected for the future. The largest reduction in the frequency of frost days is expected in summer season under both the RCP 8.5 and 4.5 pathways. The maximum change in number of frost free days in a year by the end of this century is projected to be in Osh oblast (130 days) while the least could be in Talas oblast (90 days) under RCP 8.5 pathway. The average duration of the frost-free days in a year is expected to be about 133 days longer across Kyrgyzstan (75 days under RCP 4.5 pathway) by the end of this century than it was in the early 20th century. In the future, a longer cropping season is therefore likely in some of the oblasts.

I. Country Background and its Climate

A. Background

8. The Republic of Kyrgyzstan is a mountainous landlocked country in the north-east of Central Asia. The Kyrgyz Republic’s territory covers an area of more than 0.199 million km². The length of the territory from west to east is 900 km, and from the north to south is 450 km. Kyrgyzstan has common borders with the Republic of Kazakhstan, the People's Republic China, the Republic of Tajikistan, and the Republic of Uzbekistan1 (Figure 1). Kyrgyzstan has mostly mountainous landscape. The average altitude of the country’ territory is 2,750 m above the sea level with the highest point at 7,439 m, and the lowest is 394 m. Mountain ranges in Kyrgyzstan cover about a quarter of the territory and extend by parallel chains mainly in the latitudinal direction. In the north-west of the country lies the Fergana Mountains, in the south-west the foothills of Pamir and Alay, and in the center and east the various branches of the Tien Shan, whose highest point is Jengish Chokusu (formerly Pik Pobedy), on the border with China, 7,439 meters high2. There are vast glaciers at altitudes above 4,000 meters that feed rivers, which in turn flow in deep valleys between the mountains. About 94% of the country is located at the marks higher than 1,000 m above the sea level and only 20% of territories are the areas with relatively comfortable living conditions. In this zone resides the most part of population, as well as the economic activity is mainly concentrated. According to the National Statistics Committee3, the population of Kyrgyzstan is 5,863 million people. The population density is about 29 people / km². Approximately 66% of the population resides in rural areas. Economically active population in rural areas is 1.6 million people, from which 1.5 million or 94% employed. Roads are the main means of transport from north to the south of the country, accounting for 60% of freight-haulage and 80% of passenger transport. Administratively, the country is organized in 7 provinces (Oblasts) and two cities, the capital Bishkek and Osh city. The provinces are further divided into districts (Rayon) and municipalities (Ayil Okmotu) which typically comprises several villages and hamlets. The three southern Oblasts (Osh, Jalalabad and Batken) are the most densely populated

1 http://www.thegef.org/gef/sites/thegef.org/files/documents/document/ncsa-kyrgyzstan-cc-ap.pdf. 2 CKDN Report on Kyrgyzstan Climate Risk Profile, August 2013, 68 pp 3 NSC KR, 2014: National Statistical Yearbook of the Kyrgyz Republic

and account for 43% of the total population. Competition over natural resources such as land, water and pastures are higher in the South.

B. The Climate of Kyrgyzstan

9. The country’s remoteness from the oceans and seas, as well as the neighborhood of deserts predetermines formation of continental arid climate with clearly defined seasons of the year. The climate in Kyrgyz Republic is continental, with cold winters, often frosty, and warm and sunny summers, sometimes scorching hot at low altitudes, but cooler in the mountains. Most of the territory of the Republic is located in a temperate climate zone, and only the southern areas are in a subtropical climate zone. Precipitation is moderate in the west, while the center-east is arid, and also desert at lower elevations. From the vast deserts of neighboring countries, winds are able to bring blowing sandstorms within the Kyrgyz Republic4.

Figure 1: Geographical location of the Republic of Kyrgyzstan5

10. The national capital Bishkek is located in the north, in the valley of the river Chuy, at an altitude ranging between 650 and 950 meters. Here the climate is continental, with cold winters and hot summers with day time temperatures around 35/36°C. Precipitation is moderate, the annual total being around 450 millimeters (Table 1), with a summer minimum (when only rare showers occur to interrupt the sunny days), and a maximum in spring, between March and May.

4 Korshunov, A. 2008: Findings of Technical Mission to Kyrgyzstan. Report prepared for the World Bank as a part of TA on Improving Weather and Climate Service Delivery in Central Asia, July 13-17. 5 Source: UN cartographical section, 2009

Table 1: Average monthly temperatures and precipitation - Bishkek6

Bishkek Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Min Temp (°C) -9 -7 0 6 11 15 18 16 11 5 -1 -5 Max Temp (°C) 3 3 10 18 23 28 31 30 25 17 10 5

Bishkek Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Prec. (mm) 25 30 45 75 65 35 20 10 15 45 45 30 440 Number of Rainy 6 6 9 9 8 4 3 2 3 6 7 6 68 Days

11. Osh, Kyrgyzstan's second largest city is located in the southwest at 1,000 meters of altitude, and Jalalabad, the third, at 750 meters. In these cities the temperature is similar to that of Bishkek, with slight differences due to the altitude, but the more southern location and especially the presence of mountain ranges to the north make them less vulnerable to cold spells, however, glacial days are possible here as well.

12. In the vast mountainous areas7, winters can get really cold. In Naryn, located at 2,000 meters above sea level, the average temperature in January is -15.5°C, while in July it rises to 17°C. Annually only about 300 mm of rain or snowfall occurs (Table 2), so the climate is still arid, but the pattern is different from the western cities, because there is a relative minimum in winter and a maximum in late spring, with some rains possible even in summer, reflecting the fact that the central- eastern part of Kyrgyzstan is somewhat affected by the Asian monsoon.

Table 28: Average monthly temperatures and precipitation - Naryn Naryn Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Min Temp. (°C) -21 -18 -7 1 5 8 10 9 5 0 -8 -17 Max Temp. (°C) -10 -6 2 13 17 21 24 24 20 13 1 -7

Naryn Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Total Prec. (mm) 10 12 20 32 48 60 36 21 17 15 12 10 293 Number of Rainy 9 9 10 11 12 10 9 7 5 8 7 9 106 Days 13. Karakul city is located at an elevation of 1,600 meters above sea level, on the shores of Lake Issyk-Kul, which is large enough to create a micro-climate, slightly milder than other areas of equal altitude of the country. Winter is still cold, with an average in January of -10°C, though the lake never freezes being salty, while summers are pleasantly warm, with highs around 23/25°C and cooler nights.

6 Source: https://en.climate-data.org › Asia › Kyrgyzstan › Chui Province › Bishkek

7 Sh. Ilyasov, O. Zabenko, N. Gaydamak, A. Kirilenko, N. Myrsaliev, V. Shevchenko, L. Penkina, 2013: Climate profile of the Kyrgyz Republic, 99 pages 8 Source: https://en.climate-data.org › Asia › Kyrgyzstan › Naryn › Naryn

C. Current Climate Variability and Climate Change

1. A Review of historical climate trends in Kyrgyzstan based on literature surveys 14. The official date for the formation of the Hydro-meteorological Service of Kyrgyzstan is August 26, 1926, when a meteorological office was established under the People's Commissariat of the Kyrgyz ASSR. However, systematic meteorological and hydrological observations were started in the eighties of the XIX century9. The first meteorological station on the territory of Kyrgyzstan was opened in 1856 by the Russian traveler AN Severtsev in the village of Ak-Suu on the coast of the lake Issyk-Kul. During the period 1881-1895, the weather began to be observed in Osh, Karakol, Balykchi, Tokmok, Gulcha, Irkeshtam, and gauging stations were opened on the rivers Khadzhabakirgan and Akbuura. In 1896, a weather station was set up in Bishkek. By the year 1913, there were 13 weather stations in the territory of Kyrgyzstan and about 30 gauging stations. During the period from 1926 to 1932, more than 20 hydro-meteorological stations and 40 posts were established. 15. After the end of the war in 1945, work began on the opening of new stations and posts, especially in the highland zone, as well as along the air and highway routes. Along the major roads, snow-avalanche stations were organized. Systematic balloon observations were started in 1936, and with the development of civil aviation in 1957 the first aerological station was opened. In the last 20 years, however, the number of meteorological stations in Kyrgyzstan has decreased by 42%. In the 1980s, they had 79 weather stations, 140 hydro and agro-posts, and now the hydro-meteorological network consists of only 33 weather stations, 4 of which are automatic, 3 are snow-avalanche, 77 river posts and 5 posts on the Issyk-Kul Lake and the Kirov reservoir (see Figure 2 – obtained from Kyrgyz Hydromet Office). Over the years10, the volume of observations has decreased due to abandoning the monitoring in hard-to-reach mountainous areas. 16. On the basis of the perennial data of 19 meteorological stations of Kyrgyz Hydromet, located at altitude 0.76 to 3.64 km above sea level, the fluctuations of the air temperature and the precipitation on the territory of Kyrgyzstan in the 20th century have been analyzed. On average for the entire territory of Kyrgyzstan, the average annual temperature in the 20th century in terms of 100 years is reported to have increased by 1.6°C, which is significantly higher than global warming 0.6°С. The greatest warming was observed in winter (2.6°C), and the lowest in summer (1.2°C). In northern and northwestern Kyrgyzstan, the warming range for the year in 100 years was 0.8 to 2°C, in the southwestern - 0.6 to 2.4°C, in the Issyk-Kul basin the warming was about 2.4°С, in the Inner Tien Shan -1.2°C (the same at three stations). At most stations in winter, warming was observed to be more significant than in summer. For Naryn (2.04 km altitude), in January it reached 5.2°С for 100 years. The annual temperature trend for 7 out of 9 stations is statistically significant at the confidence level of 0.99. 17. Precipitation, in general, across the territory of Kyrgyzstan in the twentieth century has increased slightly (by 23 mm or 6%). In the Inner Tien Shan, which occupies a significant part of the territory of Kyrgyzstan, they either remained practically unchanged (Naryn trend was 11 mm / 100 years), or significantly decreased, by 126 to 167 mm over 100 years, which is 41-47% from the norm. It is thus possible that the precipitation in the Inner Tien Shan at altitudes of more than 2 km will be significantly reduced in future, which, with simultaneous warming, will greatly enhance the aridity of these areas, lead to an increase in the height of the snow line and, consequently, further reduction of the glaciations.

9 Ilyasov et al, ‘Climate Profile of the Kyrgyz Republic’, supra note 13, at 99. 10 National Statistical Committee of the Kyrgyz Republic, Bishkek, 2011 - 435 pp.

Figure 2: Major Meteorological monitoring network existing in Kyrgyzstan (as of Feb, 2018)

2. The Characteristics of Current Climate Variability in Kyrgyzstan as inferred from CRU Data Sets

18. The earth’s average surface temperature has risen about 1.0 degree Celsius since the late-19th century, a change largely driven by increased carbon dioxide and other human-made emissions into the atmosphere. The average annual mean air temperature over Kyrgyz Republic has also increased by 1.1oC per decade during 1991-2000. Figure 3 depicts the Kyrgyzstan’s climate during the years 1951-2015. The annual mean temperature anomaly average across the Republic in 2015 was higher by 1.2ºC than normal baseline period (based on CRU data records as the observed and digitized historical inventory from all available meteorological stations in Kyrgyz Republic identified in Figure 2 were not readily available from KYRGYZ-HYDROMET and most of these documents are in Russian language). Figure 3 suggests that with the exception of September month and in some local regions, temperature increases are recorded in all seasons of the year. The annual mean warming across the Republic has become more rapid in the first decade of 21st Century and beyond (1.8°C per decade during the years 2001-2015)11. The warming is moderately more pronounced in winter months than in any other season.

11 Third National Communication of the Kyrgyz Republic to UN Framework Convention on Climate Change, (2017), has been developed within the project «Kyrgyzstan: Enabling Activities for the Preparation of the Third National Communication under the United Nations Framework Convention on Climate Change», implemented by the State Agency for Environment Protection and Forestry under the KR Government and the United Nations Environment Programme with financial support from the Global Environment Facility, 2016. – 265 pp.; ISBN 978-9967-11-577-4.

Figure 3: Monthly mean temperatures over Kyrgyzstan in 2015 are warmer relative to 1961-1990 mean climatology (as inferred from CRU Climatological data set) 19. A comparison of baseline (1961-1990) monthly total precipitation and that recorded in the year 2015 reveals that at country scale no definite trend can be found (Figure 4). During January- February months, a modest increase can be seen while during the peak rainy season (March- May) and in autumn, a moderate decline is discernible. However, these changes in precipitation could be a manifestation of intra-seasonal and inter-annual variability in precipitation observed in recent years over the Republic of Kyrgyz.

Figure 4: A comparison of monthly mean precipitation over Kyrgyzstan in 2015 relative to 1961-1990 mean climatology (as inferred from CRU Climatological data set)

3. Inter-annual Climate Variability in Kyrgyzstan and Selected Oblasts 20. Kyrgyzstan has a high exposure to climate variability and extremes, which are expected to increase in frequency and intensity in the future. An assessment of the historical and current seasonal climate variability in Kyrgyzstan has been attempted based on the global gridded CRU

observed data series for the period 1951-2015 [Climate Research Unit (CRU) of the University of East Anglia (UEA), Norwich, UK.]. This data series facilitated us to infer the observed changes in air temperature and precipitation during the period from 1951 until 2015. Data for all grid points within the Kyrgyz Republic for the 30-year period 1961 to 1990 were used to establish the climatological normal. The time series of mean surface air temperatures and precipitation were also analyzed to assess trends. The CRU global climatological data set provides annual and seasonal anomalies in surface air temperatures over Kyrgyzstan with respect to the baseline period to 1961-1990 as depicted in Figure 5.

Figure 5: Historical trends in annual and seasonal mean surface air temperature anomalies averaged across Kyrgyzstan during the 1951-2015 (based on CRU Climatological data sets)

21. Figure 5 depicts the observed inter-annual variability in annual and seasonal surface air temperatures averaged over all the grid points lying over Kyrgyzstan. The trend analysis of mean annual temperature over land areas in Kyrgyzstan shows increase in annual mean temperatures during 1951-2015 with respect to the baseline period of 1961-90. In winter, the highest rate of increase in mean air temperature over the country is observed in winter. The average annual

temperature in Kyrgyzstan has been rising by 0.2ºC/decade. The rate of warming is more pronounced during winter season (0.30ºС/decade) than in the summer (0.17ºC/decade) or spring (0.18 ºC/decade) seasons12. Moreover, this increase is observed in both daytime maximum and nighttime minimum temperatures with minimum temperatures warming more than the maximum temperatures with accelerated warming since the early 1990s. 22. Figure 6 depicts the observed inter-annual variability in annual mean surface air temperature anomalies over the seven oblasts in Kyrgyzstan. The trend analysis of mean annual mean temperature anomalies over land areas in these oblasts of Kyrgyzstan shows that the increase in annual mean temperatures is most pronounced in Chuy, Naryn and Talas oblasts between 1951 and 2015. The trend in annual mean surface air temperature rise in Osh and Batken oblasts is the least of the seven oblasts considered here. Statistically significant trends of increase in total duration of heat waves (heat wave is recorded when the daily maximum temperature was above 90th percentile on at least six consecutive days) have been observed over Kyrgyzstan. The frequency of frost days when the daily minimum temperature is below 0ºC is also found to be decreasing in Kyrgyzstan.

12 Government of the Kyrgyz Republic, 2009 and 2017: Second and Third National Communications of the Kyrgyz Republic to the United Nations Framework Convention on Climate Change, Bishkek.

Figure 6: Historical trends and inter-annual variability in annual mean temperature anomalies observed over seven oblasts of Kyrgyzstan 23. It is evident from Figure 6 that the annual and seasonal mean observed surface air temperatures suggest a warming trend across all oblasts in Kyrgyzstan.

24. Precipitation in Kyrgyzstan is strongly controlled by topography and lack of marine influence. Long term annual mean precipitation trends over Kyrgyzstan do not exhibit any significant increasing or decreasing trend in general (Figure 7). Figure 7 illustrates that the annual mean precipitation over Kyrgyzstan has a discernible but only marginal increasing trend during

1951-2015 on a countrywide basis. In Kyrgyzstan, precipitation tends to slightly decrease by 0.7 mm per decade in spring or autumn seasons but in winter precipitation tends to increase by 1.8 mm per decade and is significant during the period from 1951-2015 (Figure 7).

Figure 7: Historical trends and inter-annual variability in annual and seasonal precipitation observed over Kyrgyzstan 25. Similar trends are noted in annual total precipitation over each of the seven oblasts (Figure 8). A marginal (statistically insignificant) increase in annual precipitation (by 0.5 – 5.0 mm/decade) has been observed in Batken and Osh oblasts. In other oblasts considered here, precipitation has a statistically insignificant decreasing trend by 0.6 – 4 mm per decade during the period 1951- 2015. The CDD index, which represents the maximum length of time when precipitation was less than 1 mm (duration of consecutive dry days), is very important in climate of Kyrgyzstan. During 2015, the CDD lasted for more than a month at almost all the oblasts. The longest dry period of 116 days was recorded at Chuy during the year 2015.

Figure 8: Historical trends and inter-annual variability in observed annual mean precipitation anomalies in each of the seven oblasts of Kyrgyzstan 26. More detailed analysis of daily surface air temperature and precipitation data over selected oblasts (obtained from CRU gridded climatological datasets) provided the following key climatological features: • Mean annual temperature has increased by 1.2°C in 2015 with respect to the baseline period of 1961-1990 at an average rate of 0.26 to 0.37°C per decade. • Available data indicates that the average number of ‘hot’ nights per year averaged over Kyrgyzstan has increased between 1951 and 2015.

• The trend analysis of mean annual temperature over land areas in seven oblasts of Kyrgyz Republic shows that the increase in annual mean temperatures is most pronounced in Chuy, Naryn and Talas oblasts between 1951 and 2015. The trend in annual mean surface air temperature in Osh and Batken oblasts is the weakest of the seven oblasts considered here. • Precipitation in Kyrgyzstan and each of the seven oblasts is characterized by high variability on inter‐annual and inter‐decadal time scales, which can make long‐term trends difficult to identify. • No statistically significant decreases or increases in annual mean precipitation have been observed between 1951 and 2015 in Kyrgyzstan but in winter, an increase and a marginal decrease in seasonal total precipitation during summer is noted. • Some unusually high intensity precipitation spells have occurred in the dry season in recent years (2000‐2015), but no consistent trend is discernible. • The Chuy oblast has experienced a statistically insignificant decrease in summer season precipitation between 1951 and 2015.

4. Hydro-meteorological Disasters in Kyrgyzstan 27. Kyrgyzstan is prone to natural hazards including floods, mudflows, landslides, steppe winds, and earthquakes13 (Figure 9). Close to 400 types of natural hazards occur every year here, largely due to its geographical location in an active seismic zone and its mountainous landscape. During the past 20 years, the number of hazards here has increased by at least 6 times14. Interestingly, about 70% of natural hazards occur in the south of the country. Kyrgyzstan is prone to avalanches - between 1990 and 2009 the country had over 330 avalanches. Mudflows, floods and highland lakes outbursts threaten almost the entire territory of the republic as more than 95% of settlements are located in close proximity to water sources, mostly along the riverbeds of rivers. More than 90% of the lakes are highland lakes, 200 of which are under burst danger every year. Floods in the country are the result of river overflows - a combination of heavy rains and insufficient drainage infrastructure – leading to water intrusion into agricultural lands. At the same time, rapid urbanization has caused an expansion of buildings and roads that have reduced infiltration and exacerbated the impacts of floods. Droughts are common particularly in southern parts of the country as a result of climate variability that manifests itself by a late onset of the rainy season, irregular spatial distribution of rains, and an early end to the rainy season. Droughts can cause significant reductions in crop yield and can contribute to annual revenue drops from millet/sorghum. The country has evidenced in recent decades an increasing frequency of natural hazards, devastating droughts and progressive desertification (www.emdat.be, 2015). In the plains, spring floods fed by rain and snowmelt occur and mountainous regions suffer mudflows triggered by rainfall or breaches of glacial lakes. About 13% of the country’s area containing over 26% of its population is prone to mudflows. Climate change is expected to exacerbate hydro- meteorological disasters in Kyrgyz Republic, adversely affecting farmers. The region is also expected to experience an increase in extreme weather events in the coming decades.

13 Emergency Events Data Base (EM-DAT) - 2015. http://www.emdat.be/ 14 Global Facility for Disaster Reduction and Recovery, 2011: Vulnerability, Risk Reduction, and Adaptation to Climate Change: Kyrgyz Republic Climate Risk and Adaptation Country Profile. http://www.gfdrr.org

Figure 9: Average annual distribution (%) of reported disasters in Kyrgyzstan (Source: International Disaster Database - www.emdat.be, 2015) 28. Over the last few decades Kyrgyz Republic has experienced an increase in climate induced disasters and this trend is likely to continue as the consequences of climate change— particularly increases in temperature and the reduction of snowfall—will likely increase the frequency and severity of floods and droughts15. Temperatures are rising fastest at the highest altitudes, affecting glaciers, snow and ice, and threatening the communities that depend upon them. By the end of the century, temperatures in the Central Asia region could rise by up to 6°C, leading to the disappearance of more than one third of glaciers from Central Asian mountains by 205016, putting nearby communities at greater risk and rolling back hard-won development gains. An increase in intensity of precipitation or accelerated snow melt will aggravate mudflows, landslides, and avalanches in Kyrgyz Republic. Retreating glaciers and changes in seasonal snow fall and melt will lead to greater uncertainty about water discharge patterns and may threaten hydropower generation, domestic water supply, agriculture production and infrastructure17. The very low capacity of institutions and individuals to anticipate and manage these risks prevents the Kyrgyz Republic from implementing measures to increase its resilience. To achieve the country’s goal of reducing poverty and accelerating economic growth, the country will need to chart a climate resilient development path. 29. In terms of current vulnerabilities, more than 95% of settlements in Kyrgyzstan are located in the direct vicinity of water sources, mainly along river beds that are extremely sensitive to landslides, mudflow and floods. Landslides destroy houses and the infrastructure of settlements located nearby, and remote landslides can block riverbeds and their inflows. Mudflows turn catastrophic due to mountain lakes and man-made reservoir breaches typically caused by the melting of snow in the high mountainous zone; every summer a catastrophic breach of the

15 http://ipcc-wg2.gov/SREX - IPCC Special Report - Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, 2011 – 594pp. 16 Abishev et al. 2015: Committee on Water Resources, Kyrgyz Republic 17 Kotlyakov, V.M. and Severskiy, I. 2009: Glaciers of Central Asia: current situation, changes, and possible impact on water resources. In: Braun, L.N., Hagg, W., Severskiy, I.V., and Young, G. (eds). Assessment of Snow, Glacier and Water Resources in Asia, Koblenz: IHP/HWRP.

Mertsbaher lake glacial dam occurs and water is released into the Sary-Jaz River. The number of households moved out of landslide zones since 1992 has reached 7,873 (i.e., 656 houses annually). Kyrgyzstan’s First, Second and the Third National Communications (NCs) submitted to UNFCCC in 2002, 2009 and 2017 report that the country will experience increases in surface air temperatures over the 21st century18. The second NC also suggests that increased precipitation would be experienced during winter months and decreased precipitation during the summer. The country, in its Third National Communication to UNFCCC submitted in 2017 has identified water, health, agriculture and extreme weather events as the sectors most vulnerable to climate change. 30. While Kyrgyzstan has a rapidly growing economy, rural population, farmers and pastoralists outside of the main urban centers face significant climate change risks to their livelihoods stemming from increased aridity, water management challenges and extreme weather events. Shrinkage of glaciers has been recorded - Glaciers in the largest mountain range of Tien Shan have lost over a quarter of their mass in the last 50 years, and nearly a fifth of their area19,20. Climate change-induced effects in Kyrgyzstan are expected to have a far-reaching regional impact on fresh water resources. The agricultural and water resource sectors are extremely vulnerable to climate change here. This is mainly associated with redistribution of precipitation and increasing severity and frequency of floods in some oblasts and drought in others. The international experience of countries in addressing and adapting to climate change should be useful for Kyrgyzstan. It indicates a need to elaborate a National Strategy (Program) on climate change adaptation, an action plan and the creation of an institutional structure for the coordination of this process nationally. In addition to good development practice, incremental measures and actions will need to be considered to address specific risks and reduce vulnerability to the impacts of climate change21. Simultaneously, a better understanding of climate change and its effects on water, as well as glacier melting associated hazards and risks will contribute towards improved knowledge and understanding of climate risks and the need for targeted actions at both national and regional level. II. Brief Introduction of global climate models used in this CVRA Study

A. Model Generated Climate Data

1. General 31. The continued emission of greenhouse gas (GHG) emissions in the atmosphere is likely to cause increased weather volatility with the impacts increasing in intensity of the weather extremes in coming years22. Anthropogenic warming of the climate system will increase the energy of the atmosphere, which will lead to more storminess and thus more extreme behavior. The negative impacts of climate change will disproportionally exacerbate problems such as the degradation of ecosystems, floods and associated disasters, food security, increased disease outbreaks, and access to fresh water.23

18 Government of the Kyrgyz Republic, 2002, 2009 and 2017: First, Second and Third National Communications of The Kyrgyz Republic Under the Un Framework Convention on Climate Change, Ministry of Ecology and Emergencies of the Kyrgyz Republic, Bishkek. 19 Aizen, V. B. et al. 2007: Glacier changes in the Tien Shan as determined from topographic and remotely sensed data, Global and Planetary Change. 56(3-4), pp. 328-340. 20 World Bank, 2009: Glaciers of the Tien Shan Mountains are reported to have been shrinking over the past 50 years, and at an accelerated pace over the past two decades (Adapting to Climate Change in Europe and Central Asia. 21 Mayatskaya, Inna. 2005: “Kyrgyz Republic – Weather-Climate Services.” Report prepared for the World Bank. 22 WMO (World Meteorological Organization), 2016: The Global Climate in 2011–2015, WMO-No. 1179, CH-1211 Geneva 2, Switzerland, 32 pp. 23 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. et al., (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp, doi:10.1017/CBO9781107415324.

32. The scientific assessment reports of the Intergovernmental Panel on Climate Change project an increase by up to 3 to 5% in winter precipitation across Central Asia by the end of the 21st century24. This increase may be associated with a shift in precipitation patterns; heavy rains could be concentrated in shorter and more violent spells. No significant trends are, however, projected for annual precipitation in Central Asia. 33. Before any downscaled data can be ingested into estimating the specific impacts of climate change over a country or region, we need to validate the performance of the models in simulating the historical climatology to attain a degree of confidence in simulations for the future (please note that this is a necessary but insufficient test of climate model skill)25. The performance of a set of 9 Earth System Models (ESMs) out of 22 ESMs, for which downscaled data are available from NASA26 were evaluated over Central Asia region by RMSI Private Limited as part of an internal evaluation of model skills in simulation of regional historical climatology. In this study, we have updated the future projections based on four of the best performing state-of-the- art ESMs (over Central Asia region). The dataset used for inferring these projections include those generated under RCP 8.5 and RCP 4.5 scenarios for which daily model simulated gridded surface maximum and minimum air temperature and precipitation data were produced under CMIP527 and used in IPCC 5th Assessment Report. 34. In the present scenario of uncertainty about a global agreement on actions for restricting greenhouse gas emissions (target of stabilizing the emissions at 2°C could still be far away), the RCP 8.5 and RCP 4.5 pathways28 represent the most plausible global concentration pathway range for the future. Moreover, since policy makers and decision makers at country level and at municipal level in any developing country are interested in higher and lower ends of the possibilities in the vulnerability due to future extremes, we shall opt for considering the best choice of the RCP 8.5 and RCP 4.5 pathways in our vulnerability assessment. 35. Although GCMs are valuable predictive tools, they cannot account for fine-scale heterogeneity of climate variability and change due to their coarse resolution. Numerous landscape features such as mountains, water bodies, infrastructure, land-cover characteristics, and components of the climate system such as convective clouds and coastal breezes, have scales that are much finer than 100–500 km. Such heterogeneities are important for decision makers who require information on potential impacts on crop production, hydrology, species distribution, etc. at scales of 10–50 km. 36. Various methods have been developed to bridge the gap between what GCMs can deliver and what society/businesses/stakeholders require for decision-making29. The derivation of fine- scale climate information is based on the assumption that the local climate is conditioned by interactions between large-scale atmospheric characteristics (circulation, temperature, moisture, etc.) and local features (water bodies, mountain ranges, land surface properties, etc.). It is possible to model these interactions and establish relationships between present-day local climate and atmospheric conditions through the downscaling process. The downscaling process

24 IPCC, 2007: Climate change: Impacts, adaptation and vulnerability, Contribution of WGII to IPCC Fourth Assessment Report, Parry et al. (eds), Cambridge University Press. 25 Pielke, R.A. Sr. and Wilby, R.L. 2012: Regional climate downscaling – what’s the point? Eos, 93, 52-53. 26 Thrasher, B., E. P. Maurer, C. McKellar, and P. B. Duffy, 2012: Technical Note on Bias correcting climate model simulated daily temperature extremes with quantile mapping, Hydrol. Earth Syst. Sci., 16, 3309–3314, www.hydrol- earth-syst-sci.net/16/3309/2012/ doi:10.5194/hess-16-3309-2012. 27 Taylor, Karl E., Ronald J. Stouffer, Gerald A. Meehl, 2012: An Overview of CMIP5 and the Experiment Design. Bull. Amer. Meteor. Soc., 93, 485–498. 28 Meinshausen, M. S.J. Smith, K. Calvin, J.S. Daniel, M.L.T. Kainuma, and et al., 2011: The RCP greenhouse gas concentrations and their extensions from 1765 to 2300, Climatic Change, 109, 213-241. 29 TGCIA (Task Group on Data and Scenario Support for Impact and Climate Assessment), 2007: General Guidelines on the Use of Scenario Data for Climate Impact and Adaptation Assessment, IPCC-DDC website, 71 pp.

adds information to the coarse resolution GCM output so that information is more realistic at a finer scale, capturing sub-grid scale contrasts and in-homogeneities. 2. Bias Corrections to the Model Generated Datasets 37. The Bias-Correction step “corrects” the bias of the GCM data through comparisons with historical data30. For each climate variable in a given day, the algorithm generates the cumulative distribution function (CDF) for the observed data and for the retrospective GCM simulations, respectively, by pooling and sorting the corresponding source values (day of year ± 15 days) over the period from 1950 through 2005. It then compares the two CDFs at various probability thresholds to establish a quantile map between the GCM data and the historical climate data. Based on this map, GCM values in any CDF quantile (e.g., p=90%) can be translated to corresponding Global Meteorological Forcing Data (GMFD31) values in the same CDF quantile. Assuming that the CDF of the GCM simulations is stable across the retrospective and the prospective periods, to “correct” the projected future climate variations the algorithm simply looks up the probability quantile associated with the predicted climate values from the estimated GCM CDF, identifies the corresponding observed climate values at the same probability quantile in the observed CDF, and then accepts the latter as the adjusted climate predictions. The downscaling algorithm compares the GCM outputs with corresponding climate observations over a common period and uses information derived from the comparison to adjust future climate projections so that they are (progressively) more consistent with the historical climate records and, presumably, more realistic for the spatial domain of interest. The climate projections adjusted in this way shall have the same CDF as the GMFD data; therefore, the possible biases in the statistical structure (the variance, in particular) of the original GCM outputs are removed by this procedure. 38. The Spatial-Disaggregation step spatially interpolates the Adjusted GCM data to the finer resolution grid of the 0.25-degree observed gridded data. Other than simple linear spatial interpolation, multiple steps are adopted in the SD algorithm to preserve spatial details of the observational data. First, the multi-decade daily climatologies of the observed data variables (temperature and precipitation) are generated at both native and GCM resolutions. The climatology for the SD step is the average for each day of the year calculated over the reference period, 1950- 2005. Second, for each time step, the algorithm compares the Adjusted GCM variables with the corresponding observed climatology to calculate “scaling factors”. In particular, scaling factors can be calculated as the differences between the bias-corrected GCM and the observed data for temperature, but as the quotients (between the two datasets) for precipitation to avoid negative values for the latter. Third, the coarse-resolution scaling factors are bi-linearly interpolated to the fine-resolution observed data grid. Finally, the scaling factors are applied, by addition or “shifting” for temperatures and by multiplication for precipitation, on the fine-resolution observed climatologies to obtain the desired downscaled climate fields. Thus, in principle, the algorithm essentially merges the observed historical spatial climatology with the relative changes at each time step simulated by the GCMs to produce the final results.

39. The Bias-Corrected Spatial Disaggregation (BCSD) method has been applied to data sets of downscaled CMIP5 climate projections for its use at the country, province or local level

30 Thrasher, B., E. P. Maurer, C. McKellar, and P. B. Duffy, 2012: Technical Note on Bias correcting climate model simulated daily temperature extremes with quantile mapping, Hydrol. Earth Syst. Sci., 16, 3309–3314, www.hydrol- earth-syst-sci.net/16/3309/2012/ doi:10.5194/hess-16-3309-2012. 31 For development of the NEX-GDDP dataset, 0.25° X 0.25° historical data for daily maximum temperature, daily minimum temperature, and daily precipitation from 1950 to 2005 has been used. Climate dataset used are from the Global Meteorological Forcing Dataset (GMFD) for Land Surface Modeling, available from the Terrestrial Hydrology Research Group at Princeton University (Sheffield et al. 2006).

assessment of climate change impacts32. This approach used in generating the downscaled dataset inherently assumes that the relative spatial patterns in temperature and precipitation observed from 1950 through 2005 will remain constant under future climate change. Other than the higher spatial resolution and bias correction, this dataset does not add information beyond what is contained in the original CMIP5 scenarios, and thus preserves the frequency of periods of anomalously high and low temperature or precipitation (i.e., extreme events) within each individual CMIP5 scenario.

3. The Earth System Model Datasets 40. The downscaled climate scenarios for the globe that are derived from the ESM runs conducted under the Coupled Model Inter-comparison Project Phase 5 (CMIP5) are available as NASA Earth Exchange Global Daily Downscaled Projections33 dataset and have now been archived at RMSI data base for some of the ESMs. The dataset available with RMSI includes projections for RCP 8.5 and RCP 4.5 scenarios for which daily scenarios were produced under CMIP5. Each of the climate model data sets includes maximum surface air temperature, minimum surface air temperature, and precipitation for the period from 1951 through to 2100. This dataset facilitates in conducting climate change vulnerability assessment and sector specific impacts at local to regional scales, and thus to enhance our understanding of possible future global climate patterns at the country, province or local spatial scale. Each of these climate projections is downscaled at a spatial resolution of 0.25 degrees x 0.25 degrees (approximately 25 km x 25 km). The key climate variables include daily maximum temperature, minimum temperature, and precipitation for the periods from 1950 through 2005 (“Retrospective Run”) and from 2006 to 2100 (“Prospective Run”). During the downscaling process, the retrospective simulations serve as the baseline data, and are compared against the observational climate records. We have extensively used the downscaled data sets in this study. 41. We have identified the first four out of nine better performing global climate models (Table 3) over Central Asia (stretching from the Caspian Sea in the west to China in the east and from Afghanistan in the south to Russia in the north) and evaluated the model-simulated near surface air temperatures and precipitation for the region under study and compared these with the average monthly observed climatology (CRU, UK) for the baseline period of 1961-1990. The standard deviation of model simulated surface air temperature and precipitation over Kyrgyzstan (averaged over land points only) are compared with CRU-based precipitation and temperature climatology (Climate Research Unit, Norwich-UK) for the period 1961-1990 (Harris et al., 2014) to examine the skill performance of these models in simulating the seasonal and annual cycle of surface air temperature. For precipitation, RMS errors in the model simulations of annual/seasonal cycle and correlation with observations have been analyzed to obtain the skill score. The simulations of daily intensity of precipitation is also analyzed to finalize the choice of the selected ESMs to be adopted in this study as most appropriate for obtaining climate variability and climate change scenarios at a time scale of a century or shorter. Table 3: The CMIP5 Earth System Models with a fairly high degree of simulation skill over Central Asia

ESM Abbreviations Institutions where the Earth System Model (ESM) was developed MPI-ESM-MR Max Planck Institute for Meteorology, Hamburg (Germany) Model MIROC-ESM National Institute for Environmental Studies Japan ESM GFDL-CM3 Geophysical Fluid Dynamics Laboratory, Princeton (USA) Model ACCESS1

32 Maurer, E. P. and Hidalgo, H. G., 2008: Utility of daily vs. monthly large-scale climate data: an inter-comparison of two statistical downscaling methods, Hydrology and Earth System Sciences, 12, 551-563. 33 NEX - GDDP (NASA Earth Exchange Global Daily Downscaled Projections), 2015: Data Access URL - https://cds.nccs.nasa.gov/nex-gddp/.

Australian Community Climate and Earth System Simulator coupled model, CSIRO-mk3 (CAWCR), CSIRO and BOM Commonwealth Scientific and Industrial Research Organization Australia CCSM4 Model CanESM2 NCAR Community Climate System Model, Boulder, COLORADO MRI-CGCM3 Canadian Earth System Climate Model BNU-ESM Meteorological Research Institute Japan Coupled Global Climate Model Beijing Normal University Earth System Model

42. Each of the four climate projection data sets includes daily maximum surface air temperature, minimum surface air temperature, and precipitation for the periods from 1950 through 2100. The spatial resolution of the climate model-based bias-corrected dataset used for scenario development in this study is 0.25 degrees x 0.25 degrees (approximately 25 km x 25 km). This high-resolution dataset facilitates in conducting climate change vulnerability assessment and sector specific impacts at local to regional scales, and thus to enhance our understanding of plausible future global climate patterns at the country, province or local spatial scale. These downscaled CMIP5 climate projections preserve the frequency of periods of anomalously high and low temperature or precipitation (i.e., extreme events) within each emission scenario and are now being widely used at the country, province or local level assessment of climate change impacts. We do not have the choice to use regional climate model (RCM) generated high resolution data sets in this study as only one34 RCM experiment has been performed for Central Asia region (Region 8) under CORDEX35 as of this date [Region 8 of the CORDEX domains basically covers the Central Asia with corners of the domain at (11.05°E; 54.76°N), (139.13°E; 56.48°N), (42.41°E; 18.34°N) and (108.44°E; 19.39°N) with a horizontal resolution of 50 km] for which daily data were not available. Through RCM nested within ESM can facilitate us to evaluate dynamical responses in regions with complex terrain and visualize relevant, fine-resolution climate features; a tradeoff is that uncertainty and errors due to dynamical inconsistencies in two models are difficult to quantify.

B. Data Analysis Tools

43. The Climate Data Operators (CDO)36 software (available as open source in public domain) is a collection of many operators (more than 600 operators available) for standard processing of climate and forecast model output (gridded observational and climate model data on daily and monthly basis including time series data or horizontal gridded data). The operators include simple statistical and arithmetic functions, data selection and sub-sampling tools, and spatial interpolation. CDO was developed to have the same set of processing functions for GRIB and netCDF datasets in one package. The tool was developed by the Max Planck Institute for Meteorology in Hamburg. CDO has very small memory requirements and can process files larger than the physical memory. CDO is open source and released under the terms of the GNU General Public License v2 (GPL). CDO is distributed as source code - it has to be compiled and installed by the user. CDO should easily compile on every Posix compatible operating system like IBM's

34 Tuğ ba Ö ZTÜ RK, Hamza ALTINSOY, Murat TÜ RKEġ and M. Levent KURNAZ, 2011: SIMULATION OF EXTREME EVENTS FOR THE CENTRAL ASIA CORDEX DOMAIN BY USING THE REGCM 4.0, International Journal of Climatology, pp 475-484. 35 CORDEX is a World Climate Research Programme (WCRP)-sponsored program to organize an international coordinated framework to produce an improved generation of regional climate change projections world-wide for input into impact and adaptation studies within the AR5 timeline and beyond. It was initiated by the WCRP in 2009 in response to the need for a coordinated framework for evaluating and improving regional climate downscaling (RCD) techniques and producing a new generation of RCD-based fine-scale climate projections for identified regions worldwide. 36 CDO User’s Guide, 2016: Climate Data Operators (CDO) software, Developed by Uwe Schulzweida, MPI for Meteorology, 200pp.

AIX, HP-UX, Sun's Solaris as well as on most Linux distributions, BSD variants and Cygwin. Thus, it is possible to use CDO similarly on general purpose PCs and Unix-based high-performance clusters. In case of HPC, it is quite common to install software via source code compilation. Special libraries, special compilers, special directories make binary software delivery simply useless even if operating systems support package management systems like rpm (e.g. AIX). So CDO uses a customizable building process with autoconf and automake. CDO offers a scripting interface for Ruby and Python. 44. The Climate Data Interface (CDI) is used for quick and file format independent access to GRIB and netCDF datasets. The local data formats SERVICE, EXTRA and IEG are also supported. CDO is a multi-threaded application. When we use chaining operators then all these operators are running in parallel on different threads. Therefore, all external libraries are compiled thread safe. Using non-thread safe libraries could cause unexpected errors. NetCDF4 (HDF5) in combination with operator chaining can cause problems if the HDF5 library is not compiled thread safe. CDO processes only the data variables and the corresponding coordinate variables of a netCDF file. All coordinate variables and dimensions, which are not assigned to a data variable, will be lost after processing with CDO. 45. Supported data formats are GRIB 1/2, netCDF 3/4, SERVICE, EXTRA and IEG. There are more than 600 operators available. There are some limitations for GRIB and netCDF datasets. A GRIB dataset has to be consistent, similar to netCDF. That means all time steps need to have the same variables, and within a time step each variable may occur only once. NetCDF datasets are supported only with 2-dimensional, 3-dimensional and 4-dimensional variables and the attributes should follow the GDT, COARDS or CF Conventions. The user interface and some operators are similar to the PINGO package. There are also some operators with the same name as in PINGO but with a different meaning. To order to cut specific areas such as basins or countries from the data, there is a CDO command called if-then. This command needs a mask with zeros/ones at the same grid as the data. When comparing two data sets with different spatial resolutions, we always remap the data with higher grid resolution to the lower one. To remap precipitation data, use remapcon (first order conservative remapping) in order to be sure that no precipitation amount will get lost. Further information on CDO is available from https://code.zmaw.de/projects/cdo. Documentation on CDO is available at: https://code.zmaw.de/embedded/cdo/1.6.1/cdo.html. Reference card is available at: https://code.zmaw.de/files/cdo/html/1.6.1/cdo_refcard.pdf. 46. The CDO data selection, sub-sampling, statistical and arithmetic functions have been deployed over CMIP5 model data available in netcdf format (post-script files). Additional operators were used to compute climate indices of daily temperature and precipitation extremes. The definition of these climate indices is extracted from the European Climate and Assessment (ECA) project. All computed climatology was finally converted to ASCII format, which were plotted in ArcGIS 10.2 (http://www.esri.com/).

C. Data and Choice of Metrics

47. The climate change data analysis for inferring projections of future climate change for Kyrgyzstan and the selected seven oblasts has been undertaken on four timescales. These time slices are baseline (1961-1990), near term (2020s i.e., 2010 to 2039), medium term (2050s i.e., 2040 to 2069) and long term (2080s i.e., 2070 to 2099).

48. A set of core climatic indicators (https://www.ncdc.noaa.gov/indicators/) was also selected to calculate from the bias corrected data sets for baseline and future time slices at each model grid-point before averaging over the total grid points within the region of interest (Kyrgyzstan and selected seven oblasts). The focus was to infer vital information required for country (or selected

oblasts) level decision-making in structuring future development plans in relevant sectors. A select list of the key core Indicators out of those listed in Table 4 were derived for this assignment. Table 4: List of core climate indicators

Core Climate Indicators

Based on temperature Based on Precipitation (rainfall + snowfall)

Warm nights (TN90p; % of days when Wet days (R3mm; annual count of days when Rtot Tmin > 90th percentile) ≥ 3 mm)

Hot days (TX90p; % of days when Tmax > 90th Very wet days (R5mm; annual count of days percentile) when Rtot ≥ 5 mm) Heat wave spell duration (HSDI; annual count of days with at least 6 consecutive days Wet day Rainfall/Snowfall (R95pTOT; annual when Tmax > 90th percentile) total rainfall on days when Rtot > 95th percentile) Cold nights (TN10p; % of days when Very wet day Rainfall/Snowfall (R99pTOT; Tmin < 10th percentile) annual total rainfall on days when Rtot > 99th percentile) Cold days (TX10p; % of days when Tmax < 10th percentile) Rainfall/Snowfall intensity (SDII; average ≥ 1 mm) Cold spell duration (CSDI; annual count of rainfall/Snowfall on days when Rtot days with at least 6 consecutive days when Longest wet spell (CWD; annual maximum Tmin < 10th percentile) number of consecutive days when Rtot ≥ 1 mm) Frost days (FD; annual count of days when Tmin < 0 °C) Dry days (Rnnmm; annual count of days when Rtot = 0 mm or another pre-selected threshold) Longest dry spell (CDD; annual maximum number of consecutive days when Rtot < 1 mm)

TN (Tmin) refers to daily minimum surface air temperature; TX (Tmax) refers to daily maximum surface air temperature, Rtot refers to total daily rainfall. Also, total rainfall when Rtot > 95th percentile is counted as total snowfall for days with sub-zero temperatures.

D. Validation of Global Climate Models to Baseline Observed (CRU, UK) Climatology

49. Confidence in ESM projections is based on well-understood physical processes and laws, the ability of ESMs to accurately simulate past climate, and the agreement in results across models. Multiple model comparisons unanimously project warming of globally averaged near-surface temperature over the next few decades in response to increased greenhouse gas emissions. However, at regional scales, the magnitude of this increase varies from one model to another. Additionally, in certain regions, different models project opposite changes in precipitation amount, which highlights the uncertainty of future climate change projections even when sophisticated state- of-the art ESM tools are used. We have undertaken a validation exercise to ascertain that the national and oblast level climatology simulated by a select list of ESMs for the baseline period compares well with the baseline (1961-1990) observed (CRU, UK) climatology37. The outcome of these exercises is presented and discussed below.

37 Harris, I., P.D. Jones, T.J. Osborn and D.H. Lister, 2014: Updated high-resolution grids of monthly climatic observations – the CRU TS3.10 Dataset, Int. J. of Climatology, Vol. 34, 623-634, First published: 21 May 2013, DOI: 10.1002/joc.3711.

50. Figure 10 below depicts a comparison of model simulated and observed climatological (1961-1990) average monthly mean surface air temperatures averaged over the seven oblasts in Kyrgyzstan. It is encouraging to find that the simulated maximum as well as minimum temperature ensembles for each month are within ≤ 0.8°C of the observed climatological means. The models are also able to realistically simulate the observed annual cycle of surface air temperatures. While each of the four models reproduces the annual cycle of the maximum and minimum surface temperatures realistically, the simulated monthly mean climatology is within ±0.8°C or less for each month when compared with observed CRU climatology. The deviation of simulated maximum and minimum surface temperatures for each of the four models considered here are insignificant which provides support for our choice of ESMs to generate ensemble future projections for Kyrgyzstan.

Figure 10: A comparison of model simulated and observed climatological (1961-1990) average monthly mean surface air temperatures averaged over the seven oblasts in Kyrgyzstan 51. The model-simulated ensemble of monthly mean precipitation averaged over all grid points over Kyrgyzstan and in each of its seven oblasts for the baseline period also compares well with the observed climatology (Figure 11). The seasonality in the observed precipitation climatology over Kyrgyzstan is very well captured in the simulations of baseline climatology in each of the four models. The maximum deviation between the simulated ensemble and observed precipitation is observed in the month of October where the model simulates monthly total precipitation lower by about 8 mm compared to observed climatology. The model-simulated precipitation in the seven oblasts within Kyrgyzstan for the baseline period is marginally lower than the observed climatological precipitation during August to October. The maximum inter- model differences in simulated monthly precipitation in any of the seven oblasts are within ±10 mm.

52. Table 5 and Table 6 summarize the comparative skills in ensemble of selected ESMs in simulating the observed monthly mean maximum and minimum surface air temperatures. In

general, each of the model simulations of monthly mean maximum or minimum surface air temperatures are fairly well reproduced for baseline period when compared with observed CRU climatology for Kyrgyzstan.

53. Table 7 summarizes the comparative skill in ensemble of selected ESMs in simulating the observed monthly precipitation climatology and the seasonal cycle. It is rather difficult to match the model simulated monthly precipitation with the observed climatology individually during the baseline as out of four models - some underestimate the precipitation in a particular month while others simulate a marginal excess precipitation. In general, each of the four models simulate

marginally lower than observed precipitation during August to October, although the peak difference between the observed and simulated monthly precipitation totals is ±10 mm.

Figure 11: A comparison of model simulated and observed climatological (1961-1990) average monthly mean precipitation averaged over Kyrgyzstan and in each of its seven oblasts

Table 5: Baseline climatology of average monthly mean maximum temperature (degree C) over Kyrgyzstan as observed (CRU Climatology) and simulated by the models (RMSE is the root mean square error between the observed and simulated surface air temperatures)

Table 6: Baseline climatology of average monthly mean minimum temperature (degree C) over Kyrgyzstan as observed (CRU Climatology) and simulated by the models (RMSE is the root mean square error between the observed and simulated minimum surface air temperatures

Table 7: Baseline climatology of average monthly precipitation (mm) over Kyrgyzstan as observed (CRU Climatology) and simulated by the models (RMSE is the root mean square error between the observed and simulated precipitation)

54. The findings summarized in

Table 7 suggest that during September-October months (autumn season) when the precipitation is relatively low compared with spring season, the skill of the simulated ensemble is somewhat limited (for both Kyrgyzstan as a country as shown in the table and also in each of the oblasts in Kyrgyzstan), when, in general, models are simulating marginally lower than the observed monthly precipitation.

55. In this assessment, we have presented and deployed the projected climate change for Kyrgyzstan and its seven oblasts as inferred from ensembles of the four ESMs selected for time slices centered around 2020s (i.e., 2010 to 2039 - near term), 2050s (i.e., 2040 to 2069 - medium term) and 2080s (i.e., 2070 to 2099 - long term) relative to the baseline period of 1961-1990 under Chapter III of this report.

E. Kyrgyzstan Climate Change Issues, Policies and Investments

56. The Kyrgyz Republic belongs to the furthermost vulnerable countries to climate change, understands the importance of addressing the challenge and is making every effort to ensure that these initiatives are successful. Actions on climate change are reflected in the "National Sustainable Development Strategy of the Kyrgyz Republic for 2013-2017" and the "Program of the Kyrgyz Republic on Transition to Sustainable Development for 2013-2017." 57. Kyrgyzstan is considered as a Non-Annex I party according to the UN-FCCC (United Nations Framework Convention on Climate Change. Non-Annex I Parties are required to submit their first National Communication within three years of entering the Convention, and every four years thereafter. Kyrgyzstan submitted the First National Communication, Second National Communication and the Third National Communication to UNFCCC in 2003, 2009 and 2017 respectively. The creation of these National Communications paved the way for Kyrgyzstan a unique opportunity to contribute with technically sound studies and information that can be used for designing mitigation and adaptation measures, and project proposals that can and will help increase their resilience to the impacts of climate change. The ultimate goal is the integration of climate change considerations into relevant social, economic and environmental policies and actions. 58. The Climate Change Coordination Commission (CCCC), headed by the First Vice Prime Minister of the Kyrgyz Republic, coordinates all the activities in the Kyrgyz Republic related to climate change. Actions for adaptation to climate change are developed and included in the "Priorities for Adaptation to Climate Change in the Kyrgyz Republic till 2017". The Kyrgyz Republic has developed the sectoral plans and programs for adaptation in all vulnerable sectors. For the Kyrgyz Republic, a mountainous country that has a high vulnerability to the impacts of climate change, the implementation of adaptation actions is vital. 59. Kyrgyzstan has more recently identified national-level policy actions specifically focused on adaptation to climate change. The country also created an Inter-agency Group tasked with the development of a National Strategy and Climate Change Adaptation Plan. National strategies and action plans in the Third National Communication under the UNFCCC have been approved for the sectors including emergency situations; forest and biodiversity; agriculture and water management; human health; and energy. Land, Water, and Forest Codes have also been amended and supplemented so as to ensure the implementation of effective adaptation measures. In addition, local adaptation initiatives are being actively developed and enjoy support from the government, on the one hand, and from international organizations, on the other. 60. On 29th September 2015 Kyrgyzstan submitted its new climate action plan to the UNFCCC (prepared in accordance with decisions of the Conference of the Parties to the UN Framework Convention on Climate Change (UNFCCC) 1 / CP.19 and 1 / CP.20). This plan, referred to as Intended Nationally Determined Contribution (INDC), was prepared in the context of the UN climate conference in Paris held in December 2015 (The Paris Agreement entered into force on

4 November 2016, thirty days after the date on which at least 55 Parties to the Convention accounting in total for at least an estimated 55% of the total global greenhouse gas emissions have deposited their instruments of ratification, acceptance, approval or accession with the Depositary). For mitigation action plan, the country recognizes that it has developed only 10% of its hydropower, and its energy development strategy calls for the construction of multiple small hydroelectric plants by 2025 potential (The Republic’s hydropower potential is expected to decrease beyond 2030s in view of impacts of climate change on water sources). Other renewable energy development options include heat supply through solar energy and biogas, and electricity from wind and solar. The Kyrgyz INDC recognizes the damages already inflicted across the country’s most vulnerable sectors, particularly to water resources, and adopt an adaptation target of preventing further climate change damage and losses. III. Climate Change Projections and Extremes

A. Kyrgyzstan’s Mean Climate & Seasonal Variability –Key Issues

61. According to the Second and Third National Communications of the Kyrgyz Republic to UNFCCC submitted in 2009 and 2017 respectively38, average annual temperature in the Kyrgyz Republic is reported to have risen by 0.8°C since the 20th century, and the average annual precipitation has increased by 6% over the same time period. In many instances, rainfall intensity has also increased. However, in the high-altitude zone, precipitation is reported to have decreased leading to increased aridity. Kyrgyzstan has about 1.4 million hectares of arable land, which is only about 7% of the nation's total area (only about 0.3% is used for planting permanent crops). The remainder is mountains, glaciers, and high-altitude steppe that are used for grazing. Wheat, potatoes, sugar beets, and tobacco are the most important crops. Cotton and silkworms for silk production also are grown. More than 70% of the arable area depends on irrigation for its productivity. The main agricultural regions are in the Fergana Valley (Osh and Jalalabad oblasts), in the northern Chu and Talas valleys, and in the Issyk-Kul basin in the northeast. Rising air temperatures impact on climate aridity, which is expected to increase, especially in the lowlands. Higher surface temperatures result in increased evaporation and reduced soil moisture content, especially during the dry summer months, thereby amplifying the risk of droughts.

62. More than four percent - 8,400 square kilometers - of Kyrgyzstan's territory consists of glaciers. These glaciers are crucial to the agricultural economy of the region. Kyrgyzstan's 6,500 distinct glaciers are estimated to hold about 650 cubic kilometers of water and cover 8,048 square kilometers or 4.2% of the land area of Kyrgyzstan (only around the Chuy, Talas, and Fergana valleys is there relatively flat land suitable for large-scale agriculture). At an altitude of 1,600 meters in the Tien-Shan mountain range in the north of Kyrgyzstan, Issyk-Kul Lake is the world's second largest high mountain lake. There has been a series of glacial outburst floods in the mountains of Kyrgyzstan during the past few years. With glaciers melting, glacial lakes produce water in the hottest and driest period of the year in summer and compensate for low precipitation. Some of them grow significantly and, if contained by unstable moraines, they occasionally burst releasing large amounts of water in destructive flashfloods. During the 21st Century, one third of glacier volume has already been lost. The area occupied by glaciers has decreased by 20% lately and there are concerns that glaciers in the country can disappear by 2100. The Syr Darya and the Amu Darya, the two largest and most important rivers in Central Asia originate in Kyrgyzstan, feed by glaciers, snow melt and steams in the Tien Shan and Pamirs mountains. The Syr Darya flows through Uzbekistan to Kazakhstan and then into the Aral Sea. The Amu Darya flows through Tajikistan and then runs along the border of Uzbekistan and Turkmenistan then winds into

38 Government of the Kyrgyz Republic, 2009 and 2017: Second and Third National Communications of the Kyrgyz Republic to the United Nations Framework Convention on Climate Change, Bishkek.

Uzbekistan and empties into the Aral Sea. Among the principal rivers, the Chu River arises in the mountains of northern Kyrgyzstan and flows northwest into Kazakhstan. The Naryn River arises in the Tien Shan Mountains of eastern Kyrgyzstan and crosses central Kyrgyzstan before meeting the Kara Darya to form the Syr Darya in the Uzbek part of the Fergana Valley.

63. Future projections in the Kyrgyz Republic, as reported in a Kyrgyz Food Security Report39, indicate a 2°C increase in annual temperature by 2050 compared to the baseline period of 1961– 1990. Changes in precipitation show a less clear trend, with likely increase in precipitation in the northern part of the country and decrease in the southern part of the country. As a result of the increase in temperature, more precipitation will fall as rain and glaciers and snowfields will shrink, with annual river run-off peaking approximately in 2020 and then dropping significantly to levels as low as one half of current levels by the end of the century. The seasonal distribution of water flows will change considerably, affecting hydropower generation potential, water supply for human consumption and irrigation, and crop yields.

64. Over the last few decades the Kyrgyz Republic has experienced an increase in climate induced disasters and this trend is likely to continue as the consequences of climate change— particularly increases in temperature and the reduction of snowfall—will likely increase the frequency and severity of floods and droughts40. Temperatures are rising fastest at the highest altitudes, affecting glaciers, snow and ice, and threatening the communities that depend upon them. Overgrazing and deforestation of steep mountain slopes have already increased the occurrence of mudslides and avalanches. Retreating glaciers and changes in seasonal snow fall and melt will lead to greater uncertainty about water discharge patterns and may threaten hydropower generation, domestic water supply, agriculture production and infrastructure. The very low capacity of institutions and individuals to anticipate and manage these risks prevents the Kyrgyz Republic from implementing measures to increase its resilience. 65. The negative effects of climate change in the Kyrgyz Republic will have implications for water resource management in the downstream regions. The discharge in the Syr Darya river basin, one of the two main rivers in the region, will decrease as a result of snow and glacial melt and lead to diminished flows in downstream regions. Increase in surface temperature and evapotranspiration will result in higher water demand for irrigation, affect the agro-climatic zones and result in a decrease in arable land. Intensity and frequency of landslides, mudslides, floods and outbursts of highland lakes will cause damage to transport and other infrastructure. These adverse effects will be compounded by underlying socioeconomic and environmental constraints, such as land degradation, crumbling infrastructure, and limited institutional and financial capacity. 66. While monitoring systems are still inadequate for predicting the likelihood of extreme events in mountainous remote areas and assessing possible changes in weather patterns, data and information on current climate variability, future climate change, and its impacts on economic growth and human development are also insufficient to inform decision making. Limited resources for acquiring and maintaining equipment restrain the ability of responsible agencies to generate, store, and analyze climate data, and produce information for decision makers. However, incorporating climate change considerations in planning and design of water-related infrastructure, as well as repair and rehabilitation of existing infrastructure has become crucial for large scale infrastructure projects. Therefore, we have attempted to produce downscaled climate change projections for the

39 Climate Risk and Food Security in the Kyrgyz Republic, An overview of climate trends and the impact on Food Security, State Agency on the Environment al Protection and Forestry, May 2014, 46 pp. 40 World Bank, 2012: World Development Indicators and Country Partnership Strategy for the Kyrgyz Republic for the period FY12–FY17.

Republic and its seven oblasts from the high resolution CMIP5 state-of-the-art climate models and present a climate vulnerability assessment here. 67. This Climate Risk and Vulnerability Assessment (CRVA) report has been prepared under the ADB Project (TRTA 9390-KGZ: Climate Resilience and Disaster Risk Reduction in Water Resources Management) comprising of all the seven oblasts of Kyrgyzstan (Figure 12). The report will also assess the expected changes in the pattern and level of precipitation and drought / flood risks for Core Pravaya-Vetka Irrigated Agriculture and Drought (IAD) subproject site in Jalalabad oblast. The site-specific assessment presented here are targeted to provide value added guidance for the sub-project design, including (i) selection of target beneficiaries, (ii) offtake points and flood defenses, (iii) canal alignment and distribution options, (iv) alternative water sources for irrigation, and (v) command area development.

68. The Republic of Kyrgyzstan has been divided into a total of seven administrative regions (oblasts). This climate change scenario assessed as part of this study focuses on all the seven oblasts so that it could be used for CRVA at any or all the sub-project sites taken up in near future for more focused risk assessment within the selected oblasts.

Figure 12: Geographical location of Kyrgyzstan and its seven oblasts with elevations above sea level

B. Simulation of Future Climate of Kyrgyzstan & Selected Oblasts

1. Future Climate for Kyrgyzstan and Selected Oblasts 69. The future climate scenarios for Kyrgyzstan and its seven oblasts are derived using the data generated by Earth System Model (ESM) runs conducted under the Coupled Model Inter- comparison Project Phase 5 (CMIP5) and available from NASA across two of the four greenhouse gas emission scenarios. RMSI’s database has archived this data set for nine regionally best

performing ESMs (see Table 3). The performances of these ESMs were evaluated over Central Asia region by RMSI Private Limited as part of an internal evaluation of model skills in simulation of regional historical climatology. We identified the four best performing global climate models (namely, MPI-ESM-MR model from Germany, MIROC-ESM from Japan, GFDL-CM3 model from USA and ACCESS1 model from Australia) over Central Asia (stretching from the Caspian Sea in the west to China in the east and from Afghanistan in the south to Russia in the north) and evaluated the model-simulated near surface air temperatures and precipitation for the region under study and compare these with the average monthly observed climatology (CRU, UK) for the baseline period of 1961-1990. The standard deviation of model simulated surface air temperature and precipitation over Kyrgyzstan (averaged over land points only) were compared with CRU-based precipitation and temperature climatology (Climate Research Unit, Norwich-UK) for the period 1961-1990 (Harris et al., 2014) to examine the skill performance of these models in simulating the seasonal and annual cycle of surface air temperature. For precipitation, RMS errors in the model simulations of annual/seasonal cycle and correlation with observations were analyzed to obtain the skill score (details of this analysis are provided in Chapter I of this report). The simulations of daily intensity of precipitation was also analyzed to finalize the choice of the selected ESMs to be adopted in this study as most appropriate for obtaining climate variability and climate change scenarios at a time scale of a century or shorter.

70. In any attempt to predict future climate, emissions scenarios are used to explore how much humans could contribute to future climate change given uncertainties in factors such as population growth, economic development, and development of new technologies. Representative Concentration Pathways (RCPs) are time and space dependent trajectories of concentrations of greenhouse gases and pollutants resulting from human activities, including changes in land use which influence the Earth’s climate. Four RCPs, namely, RCP2.6, RCP4.5, RCP6, and RCP8.5, are identified based on a possible range of radiative forcing values in the year 2100 relative to pre- industrial values (+2.6, +4.5, +6.0, and +8.5 Wm-2, respectively). The word “representative” signifies that each RCP provides one of many possible scenarios that would lead to the specific radiative forcing pathway. These four greenhouse gas concentration (not emissions) trajectories adopted by the IPCC for its fifth Assessment Report (AR5) in 2014 superseded the Special Report on Emissions Scenarios (SRES) projections published in 2000. In RCP 8.5, emissions continue to rise throughout the 21st century (Business-As-Usual scenario). In RCP 4.5, emissions peak around 2040 and then decline to achieve specific climate change targets (such as limiting global mean warning to 2°C) towards a “risk management” approach. This combines reduction in emissions (one of the strategies to mitigate climate change) and adaptation to reduce climate change damages. 71. Parties to the U.N. Framework Convention on Climate Change (UNFCCC) reached a landmark agreement on December 12, 2015 in Paris, charting a fundamentally new course in the two-decade-old global climate effort. The agreement, which came into force on November 4th, 2016, reaffirms the goal of keeping average warming below 2°C, while also urging parties to “pursue efforts” to limit it to 1.5°C, a top priority for developing countries highly vulnerable to climate impacts. With the Paris Agreement’s entry into force, the focus has now shifted towards implementation—in particular, countries, including Kyrgyzstan are looking at how to meet the mitigation commitments they set out in their Nationally Determined Contributions (NDCs) while strengthening Integrated Adaptation Policies, Plans and Strategies for adaptation (to combat climate change and limit future climate risks thus reducing the vulnerability to climate change) and resilience. 72. Whether climate change will be addressed effectively and the goals of the Paris Agreement will be met will depend on other factors which govern present and future policies. These include energy and commodity markets, and migration and economic policies. There are also important links to the implementation of the United Nations Sustainable Development Goals (SDGs) and the Sendai Framework for Disaster Risk Reduction. In view of this, the RCP 8.5 and RCP 4.5 represent a plausible global concentration pathway range for the future and various possibilities in terms of future extremes. The climate change data analysis for inferring projections of future climate change in Kyrgyzstan and its seven oblasts are undertaken on four time scales, namely, baseline (1961-1990), near term (2020s i.e., 2010 to 2039), medium term (2050s i.e., 2040 to 2069) and long term (2080s i.e., 2070 to 2099). a. Changes in Surface Air Temperature 73. Table 8 provides a snapshot of annual and seasonal means of projected warming area averaged over Kyrgyzstan for RCP 8.5 and RCP 4.5 pathways at the three selected time slices with respect to the baseline (based on CMIP5 ensemble model simulations). Mean annual temperatures are projected to increase by 2.1°C by the 2020s, 4.2°C by the 2050s and 6.8°C by the 2080s under RCP 8.5 concentration pathway. Under the RCP 4.5 concentration pathway, the projected rise in surface temperature is restricted to 2.0°C by the 2020s, 3.3°C by the 2050s and 4.1°C by the 2080s41. The projected average warming in Kyrgyzstan at all time-slices during this century is higher

41 The projected warming for Kyrgyz Republic inferred from our analysis is higher than those reported in Third National Communication of the Kyrgyz Republic submitted to UNFCCC in 2017 (2.0°C by 2060s and 3.7°C by 2100 relative to the average of 1986-2005 under RCP8.5 pathway as quoted on page 18, Table P.3). This could be due to the choice of GCMs considered as ensembles therein and the baseline reference period.

during summer than during the spring, autumn or winter seasons (least during spring). This is in contradiction to IPCC findings in that the projected global mean warming is likely to be pronounced in winters. A clearer view on the future projections appears from Table 8 which suggests that the peak warming averaged over Kyrgyzstan should be expected to occur in August by the middle of this century and beyond. Table 8: Monthly, seasonal and annual means of projected warming area-averaged over Kyrgyzstan at three time slices under RCP 8.5 and RCP4.5 pathways

74. Future emissions of greenhouse gases may not be the Business-as-Usual (RCP 8.5 pathway). If all nations follow stringent mitigation actions and take concerted efforts to reduce their emissions (as per the commitments given under their NDCs) the warming projections for Kyrgyzstan may be lower than those projected above. The warming projections for Kyrgyzstan for the three time slices, namely 2020s, 2050s and 2080s, are likely to be relatively lower under RCP 4.5 pathway (peaking to just about 4.4°C by 2080s as given in Table 8 above). 75. The spatial distribution of the projected rise in annual mean surface air temperatures over Kyrgyzstan (for RCP 8.5 pathways) at the three time slices are depicted in Figure 12 to Figure 13 below. It is evident from these maps that the annual mean rise in surface air temperature could be higher in Batken and Osh oblasts (towards the south west side of Kyrgyzstan) relative to Naryn and Chuy oblasts (northern Kyrgyzstan) during all time slices. During the period from 2050s to 2080s, the gradient in rise in surface air temperature over Kyrgyzstan from southwest to north is projected to have a significantly high range from 6.8°C to 7.7°C (Figure 14 and Figure 15).

Figure 13: Spatial distribution of projected change in annual mean surface air temperature with respect to baseline period (1961-90) over Kyrgyzstan by 2020s under RCP 8.5 pathway

Figure 14: Spatial distribution of projected change in annual mean surface air temperature with respect to baseline period (1961-90) over Kyrgyzstan by 2050s under RCP 8.5 pathway

Figure 15: Spatial distribution of projected change in annual mean surface air temperature with respect to baseline period (1961-90) over Kyrgyzstan by 2080s under RCP 8.5 pathway

76. For RCP 4.5 pathway, the spatial distribution of projected rise in annual mean surface air temperatures over Kyrgyzstan at the three time slices are depicted in Figure 16 to Figure 18 below. The annual mean rise in surface air temperature would be higher towards the west Kyrgyzstan relative to east Kyrgyzstan during all time slices. During the period from 2050s to 2080s, the gradient in rise in surface air temperature over Kyrgyzstan from southwest to northeast is projected to have a range from 0.5°C to 0.8°C (Figure 17 to Figure 18).

Figure 16: Spatial distribution of projected change in annual mean surface air temperature with respect to baseline period (1961-90) over Kyrgyzstan by 2020s under RCP 4.5 pathway

Figure 17: Spatial distribution of projected change in annual mean surface air temperature with respect to baseline period (1961-90) over Kyrgyzstan by 2050s under RCP 4.5 pathway

Figure 18: Spatial distribution of projected change in annual mean surface air temperature with respect to baseline period (1961-90) over Kyrgyzstan by 2080s under RCP 4.5 pathway

77. The monthly, seasonal and annual changes in surface air temperature projected over the seven oblasts in Kyrgyzstan under RCP 8.5 pathway are summarized in Table 9 to Table 11 for the three future time slices. On an annual mean basis, the projected warming of 2.1°C is likely to occur in each of the seven oblasts around 2020s (August/September would likely to be warmest months in Batken oblast). Within the four seasons, summer season is projected to be warmest around 2020s in each of the seven oblasts but the peak warming occurs during August in the Issyk-Kul and Chuy oblasts.

Table 9: Monthly, seasonal and annual means of projected warming as inferred from four model ensembles area averaged over the seven oblasts of Kyrgyzstan during 2020s for RCP 8.5 pathway

78. The peak warming over the seven oblasts continues during the summer season with mean summer warming of about 5.3°C in Batken oblast and 4.8°C in Osh oblast around 2050s (August would likely to be warmest month). During the 2080s, annual mean warming of about 7.2°C in Batken oblast and 7.3°C in Osh oblast is expected. Peak warming continues to occur during the summer season in all the seven oblasts in 2080s (August would continue to be warmest month). These seasonal patterns of projected warming over the selected oblasts are consistent with that projected over the country. Table 10: Monthly, seasonal and annual means of projected warming as inferred from four model ensembles area averaged over the seven oblasts of Kyrgyzstan during 2050s for RCP 8.5 pathway

Table 11: Monthly, seasonal and annual means of projected warming as inferred from four model ensembles area averaged over the seven oblasts of Kyrgyzstan during 2080s for RCP 8.5 pathway

79. The monthly, seasonal and annual changes in surface air temperature projected over the seven oblasts in Kyrgyzstan under RCP 4.5 pathway are summarized in Table 12 to Table 14 for the three future time slices. On an annual mean basis, a peak projected warming of 2.1°C is likely to occur in Batken and Osh oblasts around 2020s (August would likely be the warmest month).

The least annual mean warming of 1.9°C could be expected in Chuy and Issyk-Kul oblasts by 2020s. Within the four seasons, summer is projected to be warmest around 2020s in all the seven oblasts. Table 12: Monthly, seasonal and annual means of projected warming as inferred from four model ensembles area averaged over the seven oblasts of Kyrgyzstan during 2020s for RCP 4.5 pathway

Table 13: Monthly, seasonal and annual means of projected warming as inferred from four model ensembles area averaged over selected seven oblasts of Kyrgyzstan during 2050s for RCP 4.5 pathway

Table 14: Monthly, seasonal and annual means of projected warming as inferred from four model ensembles area averaged over selected seven oblasts of Kyrgyzstan during 2080s for RCP 4.5 pathway

80. The projected warming in this study is marginally on the higher side than those projected in the earlier studies reported elsewhere (Second National Communication of the Kyrgyz Republic to the UNFCCC, 2009). This could be attributed to improved earth system models and the emission scenarios used in this study. Given the model validation exercises undertaken before

generating the projected scenarios (see Chapter 2) and the skills exhibited by the models individually and collectively as ensembles, the warming projections reported here could be regarded as most plausible with the inherent uncertainties still existing in climate science. b. Change in Diurnal Temperature Range 81. In order to examine the potential changes in diurnal temperature range (DTR) with time over Kyrgyzstan, we have also analyzed the warming projections in diurnal temperature range (difference between maximum and minimum surface air temperatures at each of the three timescales) for the two greenhouse gas concentration pathways. The simulated diurnal temperature range at the baseline and that projected over Kyrgyzstan at the three time slices are listed in Table 15. It is evident from this Table that no significant change in the annual mean DTR is expected in the future at all time scales. However, during winter months, a marginal decline in DTR is simulated in the future suggesting that rise in maximum surface air temperature is likely to be lower than the minimum surface air temperature in winter during all time slices in the future. During summer months, a marginal increase in DTR is simulated in the future suggesting that the rise in maximum (day time) surface air temperature is likely to be higher than the minimum (night time) surface air temperature in summer during all time slices in the future. Table 15: Monthly, seasonal and annual means of projected diurnal temperature range as inferred from four model ensembles area averaged over Kyrgyzstan during the three time slices for RCP 8.5 and 4.5 pathways

c. Changes in Precipitation 82. The monthly, seasonal and annual changes in precipitation projected over Kyrgyzstan at the three future time slices are summarized in Table 16 for RCP 8.5 and RCP 4.5 pathways. On an annual mean basis, the projected change in precipitation averaged over the country is a decline in precipitation by about -3% of the baseline value around 2020s and 2050s, and a decrease in

precipitation of 5% of the baseline value around 2080s (under RCP 8.5 pathway). Within the four seasons, the model results suggest an insignificant increase in precipitation of about 3% during winter but a significant decline of about 38% in summer precipitation by the end of this century. Similar findings of increases in winter precipitation over central Asia have been reported elsewhere42. A modest increase of as much as about 19% may be expected in spring season precipitation by the end of this century. On a monthly basis, August is likely to be the month when precipitation could be expected to decline significantly (-20% in 2020s to -45% in 2080s) throughout during this century leading to drought conditions in some of the drought prone areas of Kyrgyzstan. The largest increase in precipitation (48%) is projected for March by the end of this century. Table 16: Monthly, seasonal and annual means of projected change in area averaged precipitation (ΔP, percent change) over Kyrgyzstan at three time slices for RCP 8.5 and RCP 4.5 pathways

83. It is interesting to note from Table 16 that, under the RCP 4.5 pathway, the model projections indicate a moderate increase in precipitation in all seasons except summer. An increase in precipitation of as much as 23% could occur during spring season by the end of this century. In such a future scenario, recurrence of floods in the flood prone areas of Kyrgyzstan is likely in early spring when the snow melt starts. 84. Figure 19 to Figure 21 suggest that on an annual basis the projected decline in precipitation over Kyrgyzstan at each time slice under RCP 8.5 pathway is likely to be rather insignificant across the country. However, as stated above, decline in precipitation during summer season would become most pronounced with time.

42 Song, Shikai and Jie Bai, 2016: Increasing Winter Precipitation over Arid Central Asia under Global Warming, Atmosphere, 7, 139; doi:10.3390/atmos7100139.

Figure 19: Spatial distribution of projected change in annual precipitation (percent) with respect to baseline period (1961-90) over Kyrgyzstan by 2020s under RCP 8.5 pathway

Figure 20: Spatial distribution of projected change in annual precipitation (percent) with respect to baseline period (1961-90) over Kyrgyzstan by 2050s under RCP 8.5 pathway

Figure 21: Spatial distribution of projected change in annual precipitation (percent) with respect to baseline period (1961-90) over Kyrgyzstan by 2080s under RCP 8.5 pathway 85. Figure 22 to Figure 24 depicts the spatial distribution of changes in annual precipitation in (percent) at the selected three-time slices of the 21st century under RCP 4.5 pathway. These maps suggest that the annual mean precipitation over Kyrgyzstan is likely to increase across the country at each time slice under RCP 4.5 pathway and is likely to be most pronounced in the southern regions of the country.

Figure 22: Spatial distribution of projected change in annual precipitation (percent) with respect to baseline period (1961-90) over Kyrgyzstan by 2020s under RCP 4.5 pathway

Figure 23: Spatial distribution of projected change in annual precipitation (percent) with respect to baseline period (1961-90) over Kyrgyzstan by 2050s under RCP 4.5 pathway

Figure 24: Spatial distribution of projected change in annual precipitation (percent) with respect to baseline period (1961-90) over Kyrgyzstan by 2080s under RCP 4.5 pathway

86. The monthly, seasonal and annual changes in precipitation projected for the seven oblasts in Kyrgyzstan at the three future time slices are summarized in Table 17 to Table 19. On an annual mean basis, the projected decline in precipitation over the oblasts (except in Batken oblast where a modest increase could be expected) is likely to be -5% (in Jalalabad and Talas oblasts) with respect to the baseline value around 2020s under RCP 8.5 pathway (Table 17). This decline in precipitation is most pronounced during summer (particularly in August month) in all the oblasts except in Batken oblast where the peak decline is in June month. Around 2050s, on an annual basis, an increase in annual mean precipitation is likely in all oblasts (except in Jalalabad oblast where -3% decline in precipitation could be expected) with maximum increase of 14% in Batken oblast (Table 17). The winter precipitation around this decade would increase between 9% (in Talas and Jalalabad oblasts) to 25% (in Issyk-Kul oblast). During the spring season, a modest increase in precipitation of 19% at Batken oblast and 13% in Naryn oblast is expected by the 2050s. However, in the summer, a decrease in precipitation of as much as -35% in Jalalabad and Talas oblasts) is likely. During autumn Talas oblast only could experience a marginal decline of - 8% in precipitation around 2050s.

87. The change in annual mean precipitation in selected oblasts around 2080s is projected to be within -6% (Talas oblast) to +9% (Issyk-Kul oblast) of the baseline value (Table 19). Within the four seasons, the model results suggest a significant decline of between -16% (in Batken oblast) to -48% (in Jalalabad and Talas oblasts) in summer precipitation. An increase in winter precipitation of +46% is projected in Issyk-Kul oblast (In Talas oblast, it could be restricted to only about +23%). On a monthly basis, June and September months are expected to be the months when precipitation could decline by the end of 21st century in each of the seven oblasts.

Table 17: Monthly, seasonal and annual mean projected change in area averaged precipitation (ΔP, percent change) over selected oblasts during 2020s under RCP 8.5 pathway

Table 18: Monthly, seasonal and annual mean projected change in area averaged precipitation (ΔP, percent change) over selected oblasts during 2050s under RCP 8.5 pathway

Table 19: Monthly, seasonal and annual mean projected change in area averaged precipitation (ΔP, percent change) over selected oblasts during 2080s under RCP 8.5 pathway

88. The projected monthly, seasonal and annual changes in precipitation over the seven selected oblasts for three time slices under RCP 4.5 pathway are detailed in Table 20 to Table 22 below.

Table 20: Monthly, seasonal and annual mean projected change in area averaged precipitation (ΔP, percent change) over selected oblasts during 2020s under RCP 4.5 pathway

Table 21: Monthly, seasonal and annual mean projected change in area averaged precipitation (ΔP, percent change) over selected oblasts during 2050s under RCP 4.5 pathway

Table 22: Monthly, seasonal and annual mean projected change in area averaged precipitation (ΔP, percent change) over selected oblasts during 2080s under RCP 4.5 pathway

89. The findings in Table 20 to suggest that, under RCP 4.5 pathway, the projected annual mean increases in precipitation in the seven oblasts of Kyrgyzstan are expected to range between 13% and 20% by the end of this century. Winter precipitation increases are projected to be most pronounced (+41% in Osh oblast and +40% in Issyk-Kul oblast) by 2080s. However, during summer all the oblasts would experience a decline in precipitation by the end of 21st century.

90. It may be interesting to point out here that even a 5% increase or decrease in precipitation could contribute significantly to the surface runoff flow in all the river basins of Kyrgyz Republic due to change in surface temperatures. For the most unfavorable scenario (RCP 8.5 Scenario and annual precipitation reduction by 5%), the runoff may be reduced by as much as 40% which could lead to serious implications for available water resources in most river basins of Kyrgyz Republic.

2. Projected Changes in Climate Extremes as Inferred from Selected Core Climatic Indicators 91. Climate extremes play an important role in influencing society and natural systems. The objective of this analysis was to assess the spatial distribution and temporal trends of extreme precipitation and temperature events for the selected oblasts in Kyrgyzstan. The analysis exhibited that the temperature extremes have a significant increasing trend, consistent with global warming. Warming trends in autumn and winter were greater than in spring and summer. Besides, precipitation extremes also exhibited statistically significant increasing trends, such as peak precipitation intensity and number of wet/very wet days, etc. In the following sub-sections, we shall present and discuss the key findings on the projected changes in selected core climatic indicators (measures of extremes) based on our analysis for Kyrgyzstan as a whole and the seven selected oblasts.

a. Change in Number of Hot Days

92. As we go forward in time and the impacts of climate change become more significant, the colder temperatures are expected to warm more rapidly than the mean temperatures. This means that climate change may be most apparent in the rising frequency of warmer winter nights and hotter summer days to come.

Table 23 lists the outcome of our analysis of number of days with hotter temperatures at different future time slices (as evident form daily maximum temperature data for the four time slices namely, near term (2010 to 2039), medium term (2040 to 2069) and long term (2070 to 2099) averaged over Kyrgyzstan under RCP 8.5 and 4.5 pathways.

Table 23: Monthly, seasonal and annual mean projected number of hot days each month over Kyrgyzstan at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

93. This analysis suggests that, broadly speaking, Kyrgyzstan could experience as many as 7 more hot days by 2020s, 16 days by 2050s and over 25 days in a year by 2080s under the RCP 8.5 pathway. The spatial distribution of additional number of hot days by 2080s suggests that the hotter days are likely to be significantly more over Batken and Osh oblasts than the other five oblasts under RCP 8.5 pathway (Figure 25). The hotter days would likely only be marginally more frequent during the summer season as compared to the winter season (peak in number of hot days would be in July and August). The autumn season could experience relatively fewer hot

days than any other season. Kyrgyzstan is likely to experience around 15 hotter days in a year by 2080s under the RCP 4.5 pathway. The spatial distribution of additional number of hot days under RCP 4.5 pathway also suggests that Batken and Osh oblasts would experience more hotter days by 2080s (Figure 26).

Figure 25: Spatial distribution of projected number of hotter days over Kyrgyzstan likely by 2080s under RCP 8.5 pathway

Figure 26: Spatial distribution of projected number of hotter days over Kyrgyzstan likely by 2080s under RCP 4.5 pathway

94. Table 24 and Table 25 present the monthly, seasonal and annual mean changes in projected hotter days in the seven oblasts of Kyrgyz Republic in 2050s and 2080s under RCP 8.5 pathway. Further, Table 26 and Table 27 present the monthly, seasonal and annual mean changes in projected hotter days in the seven oblasts of Kyrgyz Republic in 2050s and 2080s under RCP 4.5 pathway. It is noted that the projected changes in hotter days in each oblast are likely to be more in 2080s than in 2050s and in RCP 8.5 pathway than in RCP 4.5 pathway.

Table 24: Monthly, seasonal and annual mean projected number of hot days each month over the seven oblasts of Kyrgyzstan at 2050s under RCP 8.5 pathway

Table 25: Monthly, seasonal and annual mean projected number of hot days each month over the seven oblasts of Kyrgyzstan at 2080s under RCP 8.5 pathway

Table 26: Monthly, seasonal and annual mean projected number of hot days each month over the seven oblasts of Kyrgyzstan at 2050s under RCP 4.5 pathway

Table 27: Monthly, seasonal and annual mean projected number of hot days each month over the seven oblasts of Kyrgyzstan at 2080s under RCP 4.5 pathway

b. Changes in Number of Frost Days

95. The length of the frost-free season (and the corresponding growing season) has been increasing nationally since the 1980s and is projected to continue to lengthen. The model simulations agree with the observed pattern for late twentieth century of a greater decrease in number of frost days in Kyrgyzstan in the future. The peak in declining number of frost days are likely in the spring and the summer season of the year under both the RCP 8.5 and 4.5 pathways (Table 28). This would mean that the snow melt could start early and be more pronounced in a warmer atmosphere.

Table 28: Monthly, seasonal and annual mean projected change in number of frost days over Kyrgyzstan at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

96. Table 28 suggests that the total number of days with frost could be reduced by 133 days under RCP 8.5 emission pathway by the end of this century. Under RCP 4.5 emission pathway, the decline in number of days with frost over the Kyrgyz Republic would be expected to be about 75 days.

97. With respect to the total number of frost days in each oblast, it is seen from and Table 30 that under RCP 8.5 emission pathway, the decline in number of days with frost could be most in Osh oblast in 2050s and beyond. The maximum change in the number of frost free days in a year by the end of this century is projected to be in Osh oblast (130 days) while the least would be in Talas oblast (90 days) under RCP 8.5 pathway.

98. Under RCP 4.5 emission pathway, and Table 32 suggest that the decline in number of days with frost could be most in Batken oblast in 2050s but the maximum number of frost free days in 2080s are likely to be in Chuy oblast.

Table 29: Monthly, seasonal and annual mean projected change in number of frost days over the seven oblasts at 2050s under RCP 8.5 pathway

Table 30: Monthly, seasonal and annual mean projected change in number of frost days over the seven oblasts at 2080s under RCP 8.5 pathway

Table 31: Monthly, seasonal and annual mean projected change in number of frost days over the seven oblasts at 2050s under RCP 4.5 pathway

Table 32: Monthly, seasonal and annual mean projected change in number of frost days over the seven oblasts at 2080s under RCP 4.5 pathway

Figure 27: Spatial distribution of projected change in number of frost days over Kyrgyzstan likely by 2080s under RCP 8.5 pathway

Figure 28: Spatial distribution of projected change in number of frost days over Kyrgyzstan likely by 2080s under RCP 4.5 pathway

99. Figure 27 and Figure 28 depict the spatial distribution of projected change in number of frost days over Kyrgyzstan likely by 2080s under RCP 8.5 and RCP 4.5 pathways respectively. The numbers of frost days are most consistently related to sea level pressure, with more frost days occurring when high pressure dominates on the monthly time scale in association with clearer skies and lower nighttime minimum temperatures. Spatial patterns of relative changes of frost days are, thus, indicative of regional scale atmospheric circulation changes that affect nighttime minimum temperatures. The average duration of the frost-free season is expected to be about 133 days longer across Kyrgyzstan (75 days under RCP 4.5 pathway) by the end of this century than it was in the early 20th century. While there would continue to be variations in the amount of frost-free days from year to year, climate change would contribute to an overall increase in the number of days without frost in Kyrgyzstan.

100. The longer the time without a frost, the longer would be the growing season. While this may seem good — more time should lead to a larger crop yield — it could possibly have detrimental effects on the crops we grow. Warmer (winter) weather helps pests survive longer which can wreak havoc on crops. Rising temperatures are also expected to contribute to a shift in which areas are most agriculturally productive and what crops grow there.

c. Annual Mean Changes in Number of Dry Days

101. The frequency of dry days affects regional hydrology and ecosystems and potentially influences agriculture and transportation. Future changes in the number of dry days per year can either reinforce or counteract projected increases in daily precipitation intensity as the climate warms. The analysis of dry tails of regional precipitation distributions is normally focused on understanding the probability of long-term drought, dry spells, or dry months. Looking at changes in the number of dry days per year facilitates our understanding as to how climate change might affect us beyond just annual or seasonal mean precipitation changes and allow us to better adapt to and mitigate the impacts of local hydrological changes. The length of dry spells has increased over conterminous Kyrgyzstan concurrently with observed increasing trends in precipitation intensity. The climate model projections suggest that the frequency of dry days over the selected oblasts in Kyrgyzstan could increase by between 19 to 23 days per year by the 2050s and between 33 to 40 days by end of this century under RCP 8.5 pathway (and Table 34).

Table 33: Monthly, seasonal and annual mean projected change in number of Dry days over Kyrgyzstan and its seven oblasts at 2050s under RCP 8.5 pathway

Table 34: Monthly, seasonal and annual mean projected change in number of Dry days over Kyrgyzstan and its seven oblasts at 2080s under RCP 8.5 pathways

102. Figure 29 depicts the spatial distribution of annual mean change in number of dry days over Kyrgyz Republic by 2080s. This analysis suggests that the projected warming could accelerate land surface drying with increases in the potential incidence and severity of droughts.

Figure 29: Spatial distribution of projected change in number of dry days over Kyrgyzstan likely by 2080s under RCP 8.5 pathway

d. Changes in Number of Wet / Very Wey Days 103. Observations show that changes are occurring in the amount, intensity, frequency and type of precipitation over Kyrgyzstan. These aspects of precipitation generally exhibit large natural variability and are associated with increased water vapor in the atmosphere arising from the warming of the world’s oceans. There are also increases in some regions in the occurrences of both droughts and floods. The warmer climate increases risks of both drought − where it is not raining − and floods − where it is − but at different times and/or places. The spatial distribution of changes in the annual count of wet days and very wet days over Kyrgyzstan by the end of this century under RCP 8.5 pathways are illustrated in Figure 30 and Figure 31 respectively.

Figure 30: Spatial distribution of projected change in number of wet days over Kyrgyzstan likely by 2080s under RCP 8.5 pathway

Figure 31: Spatial distribution of projected change in number of very wet days over Kyrgyzstan likely by 2080s under RCP 8.5 pathway

104. The analysis, as presented above, suggests that significant changes are expected in the number of wet / very wet days (with decreases of between 7 to 21 days in wet /very wet meaning

thereby lesser number of rain days in a year) in a year over Batken, Talas, and Osh oblasts in Kyrgyzstan by the end of the 21st century under RCP 8.5 pathway.

e. Changes in Maximum Rainfall Intensity 105. Theoretically speaking, future increases in precipitation intensity are expected to be an important aspect of climate change, since warming will tend to accelerate the overall hydrological cycle, intensifying the wet extremes (IPCC, 2012)43. A well-established physical law (the Clausius - Clapeyron relation) determines that the water-holding capacity of the atmosphere increases by about 7% for every 1°C rise in temperature. Increased water vapor is expected to lead to more intense precipitation events even when the total annual precipitation is reduced slightly and with prospects for even stronger events when the overall precipitation amounts increase. An understanding of the ongoing changes in climate and climate-related extremes is, therefore, of central importance for adapting to the negative impacts of climate change (World Bank, 2009)44.

Table 35: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Kyrgyzstan at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

106. As is depicted in Table 35, the peak precipitation intensity across Kyrgyzstan is projected to increase with time albeit at a different rate under both the RCP pathways. A 16% increase in maximum precipitation intensity is expected on an annual basis under the RCP 8.5 pathway but

43 IPCC, 2012: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change; https://www.ipcc.ch/report/srex/docs/srex_citations.pdf 44 World Bank, 2009: Adapting to climate change in Europe and Central Asia (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/127181468024643244/Adapting-to-climate-change-in-Europe-and- Central-Asia

may exceed to about 21% by the end of this century under RCP 4.5 pathway. Hence, we can conclude that more precipitation is expected to fall as heavier precipitation events across Kyrgyzstan in the future. Also, the projected increases in heavy precipitation are expected to be marginally smaller under higher emission scenarios as compared to lower emission scenarios. 107. The area averaged changes in maximum precipitation intensity in each of the seven oblasts are given in Table 36 to Table 42 below. It is seen that the changes in peak precipitation intensity are significantly different in each oblast for each of the two RCP pathways thus highlighting the large spatial as well as temporal variability in the trends of precipitation-related extremes. The maximum change in peak precipitation intensity is expected in Batken oblast at all time-slices under RCP 8.5 pathway. Under RCP 4.5 pathway, this peak in precipitation extremes is expected to shift to Osh oblast.

Table 36: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Batken oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

Table 37: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Chuy oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

Table 38: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Jalalabad oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

Table 39: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Naryn oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

Table 40: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Osh oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

Table 41: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Talas oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

Table 42: Monthly, seasonal and annual mean projected change in peak precipitation intensity over Issyk-Kul oblast at 2020s, 2050s and 2080s under RCP 8.5 and RCP 4.5 pathways

C. Uncertainties in Future Projections

108. Confidence in climate projections is a measure of how plausible the projected range of change is for a given emission scenario. Confidence comes from multiple lines of evidence including physical theory, past climate changes, and climate model simulations. Three major sources of uncertainty exist, which should be considered while using the future projections presented in this report. These include the choice of climate model, the choice of emissions scenario, and the internal variability of the modeled climate system. The relative contribution to uncertainty depends on the climate variable considered, as well as the region and time horizon. Internal variability is roughly constant through time, whereas the other uncertainties grow with time, but at different rates45. Internal variability and inter-model uncertainty are more important for the near-term. The uncertainty due to the choice of emissions scenario becomes more important toward the end of the 21st century. For precipitation projections across the globe, the RCP scenario uncertainty beyond 2050s is relatively small when compared to the other sources of uncertainty. For mid-century precipitation projections, model internal variability is more important, and this persists in some regions into the late century. 109. Dealing with uncertainty is a fundamental task when using climate projections in a decision-making context. Even where there is high confidence in the response of the climate system to forcing, there is uncertainty from natural variability46 and our incomplete knowledge of how climate changes affect the context of the decision. Also, there will always be some inevitable uncertainty in the processes involved and their impacts. In cases where the direction of change is clear and we have High Confidence in the approximate ranges of change for a given scenario, such as for temperature, then the main risk management decisions may be about emission scenarios. In cases where there is considerable uncertainty in the response of the climate system to forcing, such as regional precipitation change, projections can still be useful but the decision- making framework must be adaptive and put more emphasis on sampling the range of uncertainty in response. 110. We have provided here the future projections of the climate change in Kyrgyz Republic and its seven oblasts that would result from a given set of assumptions or future emission scenarios regarding human activities (including changes in population, technology, economics, energy, and policy). Future changes also have some uncertainty due to natural variability, particularly over shorter time scales and limitations in scientific understanding of exactly how the climate system will respond to human activities. The relative importance of these three sources of uncertainty changes over time. Which type of uncertainty is most important also depends on what type of change is being projected: whether, for example, it is for average conditions or extremes, or for temperature or precipitation trends. In view of this, the projections of key climate variables and core indicators reported here should be used with caution.

D. Summary - Key Findings on Climate Change Scenarios

111. Salient features of projected annual mean climate change over Kyrgyzstan and its selected Oblasts for the future based on our analysis are as highlighted below: • The trend analysis of mean annual temperature over land areas in Kyrgyzstan shows increase in annual mean temperatures between by 1.2°C with respect to the baseline period of 1961- 1990 at an average rate of 0.26 to 0.37°C per decade.

45 Knutti, R, Allen, MR, Friedlingstein, P, Gregory, JM, Hegerl, GC, Meehl, GA, Meinshausen, M, Murphy, JM, Plattner GK, Raper, SCB, Stocker, TF, Stott, PA, Teng, .H, Wigley, TLM, 2008: A review of uncertainties in global temperature projections over the twenty-first century, J. Clim. 21: 2651–2663. doi:10.1175/2007JCLI2119.1. 46 Deser, C., Phillips, A.S., Alexander, M.A. and Smoliak, B. 2014: Projecting North American climate over the next 50 years: Uncertainty due to internal variability, Journal of Climate, 27, 2271-2296.

• The average number of ‘hot’ nights per year averaged over Kyrgyzstan has increased between 1951 and 2015. • The observed increase in annual mean temperatures was most pronounced in Chuy, Naryn and Talas oblasts. The trend in annual mean surface air temperature in Osh and Batken oblasts was the least of the seven oblasts considered here. • More frequent heat waves (when the daily maximum temperature was above 90th percentile for at least six consecutive days) have also been observed in parts of Kyrgyzstan during the past three decades. • During the past three decades, the frequency of frost days with daily minimum temperature below 0ºC is reported to be decreasing in Kyrgyzstan. • Precipitation in Kyrgyzstan and each of the seven oblasts is characterized by high variability on inter‐annual and inter‐decadal time scales, which can make long‐term trends difficult to identify. • No statistically significant decreases or increases in annual mean precipitation have been observed between 1951 and 2015 in Kyrgyzstan but an increase in winter and a marginal decrease in seasonal total precipitation during summer are noted. • On average, annual precipitation has a marginally declining trend on a countrywide basis. In Kyrgyzstan precipitation tends to slightly decrease in all seasons except winter when precipitation tends to increase and are significant during the period from 1951-2015. • The length of dry spells has increased over conterminous Kyrgyzstan concurrently with observed increasing trends in precipitation intensity. • Some unusually high intensity precipitation spells have occurred in the dry season in recent years (2000‐2015), but no consistent trend is discernible. • A marginal (statistically insignificant) increase in annual precipitation has been observed in Batken and Osh oblasts. In other oblasts, precipitation has a statistically insignificant decreasing trend during the period 1951-2015. • Beyond the middle of this century, global and regional temperature changes will be determined primarily by the rate and amount of various emissions released by human activities, as well as by the response of the Earth’s climate system to those emissions. Efforts to rapidly and significantly reduce emissions of heat-trapping gases (e.g., RCP 4.5 pathway) might still limit global warming to 2ºC relative to the 1901-1960 time period. However, significantly greater temperature increases are expected if emissions follow higher scenarios associated with continuing growth in the use of fossil fuels (e.g., RCP 8.5 pathway). • Following the observed trends over recent decades, more precipitation is expected to fall as heavier precipitation events. In many mid-latitude regions, including the Kyrgyz Republic, there could be fewer days with precipitation, but the wettest days could be wetter. • The future projections in key climate variables reported here are based on ensembles obtained from four best performing climate models (namely, MPI-ESM-MR model from Germany, MIROC-ESM from Japan, GFDL-CM3 model from USA and ACCESS1 model from Australia) over Kyrgyzstan. • Mean annual temperatures are projected to increase by 2.5°C by the 2020s, 4.5°C by the 2050s and 7.0°C by the 2080s under RCP 8.5 concentration pathway. Under RCP 4.5 concentration pathway, the projected rise in surface temperature is expected to be 2.2°C by the 2020s, 3.7°C by the 2050s and 4.4°C by the 2080s. • The projected average warming over Kyrgyzstan at all time slices during this century is higher during the summer season than during the winter, spring or autumn seasons (least during spring season). The peak warming averaged over Kyrgyzstan is expected to occur in August month by the middle of the 21st century and beyond. • The annual mean rise in surface air temperature could be greater towards the west Kyrgyzstan relative to east Kyrgyzstan during all time slices. The warming extends from 2.0°C over the east Kyrgyzstan to 2.3°C over the west of the country during 2020s under RCP 8.5 pathway.

During the period from 2050s to 2080s, the gradient in rise in surface air temperature over Kyrgyzstan from north to south is projected to have a significantly high range of over 1.0°C. • On an annual mean basis, the projected warming of 2.1°C under RCP 8.5 pathway is likely to occur uniformly in all the seven oblasts around 2020s (August month is expected to have the greatest warming). Within the four seasons, autumn is projected to be warmest around 2020s in Batken and Osh oblasts, but the peak warming occurs during summer season in the remaining five oblasts. The peak warming over the seven oblasts shifts to summer season with annual mean warming of about 4.5°C in Batken and Osh oblasts around 2050s. During the 2080s, annual mean warming of about 7.3°C in Osh oblast and 7.2°C in Batken oblast is expected. Peak warming continues to occur during the summer season in all the seven oblasts in 2080s. • During summer months, a marginal increase in diurnal temperature range is projected for the future suggesting that rise in maximum surface air temperature is expected to be more than the minimum surface air temperature in summer during all time slices in the future. No significant change in DTR is projected on annual mean basis in all the oblasts. • On an annual mean basis, the projected decrease in precipitation averaged over the country is 3% of the baseline value by the 2020s and 2050s and a decrease in precipitation of 5% by the 2080s (under the RCP 8.5 pathway). Within the four seasons, the model results suggest a modest increase in precipitation of about 19% during the spring season by the end of this century. However, during the summer season, a significant decline of 38% in precipitation is projected which could lead to drought conditions in some of the drought prone areas of Kyrgyzstan. Under RCP 4.5 pathway, this decline in summer precipitation is restricted to only 16% while on annual mean basis, an increase in precipitation by 9% could be anticipated. During autumn season, this increase is most significant at about 23% by 2080s under the RCP4.5 pathway. • Under the RCP 4.5 pathway, on an annual basis, an increase in precipitation of about 9% could occur by the end of the 21st century. A decline in summer precipitation is restricted to only 16%. During spring season, the increase is most significant at about 23% by 2080s. In such a future scenario, recurrence of floods in flood prone areas of Kyrgyzstan is expected in spring when snow melt begins. • Kyrgyzstan could experience as many as 7 additional hot days by 2020s, 16 days by 2050s and over 25 days in a year by 2080s under the RCP 8.5 pathway. The hotter days would likely only be marginally more frequent during the summer season as compared to the winter season (peak in number of hot days would be in July/August). The spring season could experience relatively fewer hot days than any other season. Kyrgyzstan is likely to experience around 15 hotter days in a year by 2080s under the RCP 4.5 pathway. • A decrease in the number of frost days in Kyrgyzstan is projected for the future. The largest reduction in the frequency of frost days is expected in summer season under both the RCP 8.5 and 4.5 pathways. The maximum change in number of frost free days in a year by the end of this century is projected to be in Osh oblast (130 days) while the least could be in Talas oblast (90 days) under RCP 8.5 pathway. The average duration of the frost-free days in a year is expected to be about 133 days longer across Kyrgyzstan (75 days under RCP 4.5 pathway) by the end of this century than it was in the early 20th century. • The frequency of dry days over the seven oblasts in Kyrgyzstan could increase by between 19 to 23 days per year by the 2050s and between 33 to 40 days by end of this century under RCP 8.5 pathway. This suggests that the projected warming could accelerate land surface drying with increases in the potential incidence and severity of droughts. • No significant changes are expected in the number of wet / very wet days (increases of between only 1 to 3 days are projected) over Kyrgyzstan by the end of this century under RCP 8.5 pathway. • More precipitation is expected to fall as heavier precipitation events across Kyrgyzstan in the future. A 16% increase in maximum precipitation intensity on an annual basis may occur under

RCP 8.5 pathway and about 21% under RCP 4.5 pathway by the end of this century. Hence, the projected increases in heavy precipitation are expected to be greater under lower emission scenarios. Changes in peak precipitation intensity are significantly different in each oblast for each of the two RCP pathways thus highlighting the spatial as well as temporal variability in the trends of precipitation-related extremes.

112. To summarize, floods and flash floods with mudflow are currently the most significant climate related disasters in Kyrgyzstan. The most affected Oblasts in total monetary terms are Jalalabad and Osh47. Batken has the greatest per capital damage and also scores highest in terms of livelihoods impacts. The expected impacts of a changing climate include: an increase in mean annual temperature of over 4°С by 2050. Changes in rainfall patterns are likely with an average increase during winter but a decrease during summer by 2050 and beyond. An increase in the intensity and frequency of extreme weather events, including heavy rainfall days are projected for future which would contribute to an increase in climate-related disasters, including floods, mudflows, droughts and landslides. Kyrgyzstan has identified agriculture and water sector48 as being among the most vulnerable to the impacts of climate change. Kyrgyzstan, with approximately 30% of the region’s water resources, is also one of the main suppliers of water in Central Asia. Rapid glacial melting is adversely affecting water supply and quality in Kyrgyzstan. Glacier surface area in Kyrgyzstan has already decreased 19.8% during the period 1970-2000 and observations of the Tien-Shan Mountains indicate a likelihood of permanent shrinking of glacier areas49. Continued rapid glacial melting and increased evaporation due to increasing temperatures is likely to lead to a decline in water supply in certain parts of the country.

IV. Implications of Projected Changes in Climate on the risks at proposed sub-project level and associated Vulnerabilities

A. The Short-listed Sub-project: Introduction

113. The Climate Change Scenarios presented in Chapter III above as part of the TRTA for Climate Resilience and Disaster Risk Reduction in Water Resources Management in the Kyrgyz Republic details the risk assessment due to climate change downscaled to the oblast level. Eventually these scenarios are to be made use of at the subproject level to identify the risks and vulnerability assessments. The sub-projects have been identified by the respective Implementing Agency with the clear focus on building climate resilience and enabling disaster risk reduction in the agriculture and water resource sectors to ease the burden on poor and vulnerable rural communities who are exposed to potentially significant impacts on water resource availability and damage to food productivity as well as critical infrastructure (homes, roads, canals, land) from climate and natural hazards (particularly droughts and floods). 114. This TRTA was mandated to ‘develop and prepare a proposed investment project to strengthen the resilience of the water resources sector to droughts in the Kyrgyz Republic’. Consultants from Egis Eau (EE from France) and Keremet Institute (KI from Kyrgyzstan) formed a joint venture to develop and prepare the project in accordance with Asian Development Bank (ADB) requirements. Ministry of Agriculture (MoA) nominated four priority subprojects, one each in Talas

47 CKDN Report on Kyrgyzstan Climate Risk Profile, August 2013, 68 pp. 48 Third National Communication of The Kyrgyz Republic Under the United National Framework Convention on Climate Change (2017), Government of the Kyrgyz Republic. 49 Glaciers of the Tien Shan Mountains are reported to have been shrinking over the past 50 years, and at an accelerated pace over the past two decades (Adapting to Climate Change in Europe and Central Asia, 2009), World Bank.

and Jalalabad oblasts, and two in Osh oblast. Out of these, the project team has, subsequent to several site visits and following the eligibility criteria set up by ADB, identified one sub-project namely the Core Pravaya-Vetka (Kooken Rayon) Irrigated Agriculture and Drought (IAD) management subproject within Jalalabad Oblast under the current TRTA investment project. Attention will be given to modernization of the irrigation systems and on-farm water management. For Ministry of Emergency Situations (MoES), assessment will be made of the current hydro- meteorological network in the subproject areas. There is also a separate program, overseen by the TRTA and coordinated with MoES, to develop a Disaster Risk and Water Resources Information System (DWIS), using a graphical information system platform, as a tool to assist disaster risk management-related decision making. It will utilize extensive knowledge and development experience of building similar systems elsewhere in the region and Asia. If the preliminary work shows promise, then a more comprehensive version will be progressively developed through the implementation stage of the Project. 115. A climate risk screening and vulnerability assessment for the Core Pravaya-Vetka (Kooken Rayon) Irrigated Agriculture and Drought (IAD) management subproject (Jalalabad Oblast) has been undertaken which are presented in this chapter. Also presented are the proposed adaptation responses in the next chapter to mitigate the adverse impacts and risks associated due to climate change. Chapter 5 shall also include the assessment of the current hydro-meteorological network and its proposed expansion plan to be implemented by Kyrgyz Hydromet Office of the MoES.

B. Climate Risk screening and vulnerability assessment of the short-listed sub-project

116. ADB, Ministry of Agriculture and Ministry of Emergency Situations in Kyrgyz Republic have agreed to take up Pravaya-Vetka site for irrigated agriculture and drought management sub- project. Likewise, ADB and Ministry of Emergency Situations in the Republic have agreed to develop a Disaster Risk and Water Resources Information System (DWIS) using a graphical information system platform as a tool to assist disaster risk management-related decision making. We have undertaken a climate risk screening for the irrigated agriculture and drought management sub-project at Pravaya-Vetka site in this report. This screening is based on detailed projections of future climate change and an assessment of the climate change impacts as detailed in Chapter III. 1. The baseline key climatic variables affecting the droughts 117. Surface air temperature and precipitation are two fundamental variables for describing the climate and can have wide-ranging effects on human life and ecosystems. For example, increases in air temperature can lead to more intense heat waves, which can curtail the crop growth due to thermal stress and hence productivity. Rainfall, snowfall, and the timing of snowmelt can all affect the amount of water available for irrigation, and industry. Annual and seasonal temperature and precipitation patterns also determine the types of crops) that can survive in particular locations. Changes in temperature and precipitation can disrupt a wide range of natural processes, particularly if these changes occur more quickly than plant species can adapt. Average temperatures at the Earth’s surface are rising and are expected to continue rising due to anthropogenic emissions of greenhouse gases in the earth’s atmosphere. Increased evaporation due to warmer temperature can produce more intense precipitation events—for example, heavier rain and snow storms that can damage crops and increase flood risk—even if the total amount of precipitation in an area does not increase. 118. For the assessment of baseline climate of the sub-project site, we have inferred monthly mean maximum and minimum surface air temperatures, monthly total precipitation and number of wet days in a month averaged over an area of 625 km2 from the center of the sub-project site from the 30 years daily mean climatological data for the period from 1961 to 1990 for temperature

and from 1981 to 2010 for precipitation50. The number of wet days in a month has been derived from the daily precipitation data over the sub-project site. The first four rows in the Table 43 given below provide this baseline climatology for the core IAD site. 2. Climate change projections for the sub-project site 119. Climate change is expected to impact extreme weather events or hydrometeorological hazards such as floods and droughts in complex and significant ways. There is a meaningful consensus that the increased energy in the global climate system caused by enhanced greenhouse gas effects will also have an impact on weather variability. It is anticipated that the proposed site would experience more frequent and intense hot days during summer months and an increase in heavy rain events during rainy season (and consequently more intense floods) along with widespread issues related to water shortage during the late summer and autumn months thus hampering the irrigation requirements at critical stages of the crop growth. 120. Over the last few decades Kyrgyz Republic has experienced an increase in climate induced disasters and this trend is likely to continue as the consequences of climate change— particularly increases in temperature and the reduction of snowfall—will likely increase the frequency and severity of floods and droughts51. Temperatures are rising fastest at the highest altitudes, affecting glaciers, snow and ice, and threatening the communities that depend upon them. A recent study52 suggests that during the past one and a half decade there is a general trend of significantly earlier snowmelt below 3400 m and the area of earlier snowmelt is 15 times greater in eastern than western districts of Kyrgyz Republic. By the end of the century, temperatures in the Central Asia region could rise by up to 6°C, leading to the disappearance of more than one third of glaciers from Central Asian mountains by 205053, putting nearby communities at greater risk and rolling back hard-won development gains. An increase in intensity of precipitation or accelerated snow melt will aggravate landslides and avalanches in Kyrgyz Republic. Retreating glaciers and changes in seasonal snow fall and melt will lead to greater uncertainty about water discharge patterns and may threaten hydropower generation, domestic water supply, agriculture production and infrastructure54. For the Kyrgyz Republic, a mountainous country that has a high vulnerability to the impacts of climate change, the implementation of appropriate adaptation actions is vital. 121. To date, global climate models (GCMs) have been the only science-based tools to provide internally consistent projections of future climate change. Downscaling of these model outputs55, either statistically or dynamically, is needed to derive climate change projections suitable for (often local-level) natural hazard risk analyses56. Efforts have been made in this assessment towards downscaling of GCM outputs to facilitate climate risk analyses at regional and local scales. In this study, we have used the daily model simulated gridded surface maximum and minimum air

50 CRU-UK climatological data has been deployed for obtaining the 30-year mean monthly climatology of maximum and minimum surface air temperatures. For precipitation, we have used the high resolution 30-year climatological monthly mean precipitation data available from Climate Hazard Group InfraRed Precipitation (CHIRP) data archive (CHRIPS2.0 – developed by Climate Hazards Group of the University of California with support from NASA, NOAA and USGS) at 0.25 and 0.05 degree resolution available from daily global data sets from January 1981 to December 2010. 51 http://ipcc-wg2.gov/SREX - IPCC Special Report - Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, 2011 – 594pp. 52 Tomaszewska M. A. and Henebry, G. M., 2018: Changing snow seasonality in the highlands of Kyrgyzstan, Environ. Res. Lett. 13 (2018) 065006, https://doi.org/10.1088/1748-9326/aabd6f 53 Abishev et al. 2015: Committee on Water Resources, Kyrgyz Republic 54 Kotlyakov, V.M. and Severskiy, I. 2009: Glaciers of Central Asia: current situation, changes, and possible impact on water resources. In: Braun, L.N., Hagg, W., Severskiy, I.V., and Young, G. (eds). Assessment of Snow, Glacier and Water Resources in Asia, Koblenz: IHP/HWRP. 55 TGICA (Task Group on Data and Scenario Support for Impact and Climate Assessment), 2007: General Guidelines on the Use of Scenario Data for Climate Impact and Adaptation Assessment, IPCC-DDC website, 71 pp. 56 Pielke, R.A. Sr. and Wilby, R.L. 2012: Regional climate downscaling – what’s the point? Eos, 93, 52-53.

temperatures and precipitation data available as NASA Earth Exchange Global Daily Downscaled 57 58 Projections dataset produced under CMIP5 and used in IPCC 5th Assessment Report. The future projections for Kyrgyz Republic and its seven Oblasts are based on four of the best performing state-of-the-art ESMs ((namely, MPI-ESM-MR model from Germany, MIROC-ESM from Japan, GFDL-CM3 model from USA and ACCESS1 model from Australia) over Central Asia region as detailed in Chapter 3 above. The time slices used for future projections are near term (2020s i.e., 2010 to 2039), medium term (2050s i.e., 2040 to 2069) and long term (2080s i.e., 2070 to 2099). For local scale projections, we restrict our projections to medium term only considering that the life time of rehabilitation projects can at best be 30 to 50 years and also as the uncertainty in future projections could be larger on longer time scales.

57 NEX - GDDP (NASA Earth Exchange Global Daily Downscaled Projections), 2015: Data Access URL - https://cds.nccs.nasa.gov/nex-gddp/. 58 Taylor, Karl E., Ronald J. Stouffer, Gerald A. Meehl, 2012: An Overview of CMIP5 and the Experiment Design. Bull. Amer. Meteor. Soc., 93, 485–498.

Pravaya Vetka Kara Unkur Say Channel (Jalalabad Oblast) (Long: 72°44'15.109"E, Lat: 40°58'52.414"N)

Table 43: Baseline climatology and projection of key climate elements for Core Pravaya-Vetka Irrigated Agriculture and Drought (IAD) Subproject at time slice centered on 2050s

Selected January February March April May June July August September October November December Climate Variables Maximum Temperature, oC -3.4 -1.6 5.7 14.5 19.5 24.6 27.4 26.2 21.3 13.4 5.3 -0.7 Minimum Temperature, oC -13.1 -11.3 -4.3 3.0 6.9 10.5 12.6 11.0 6.1 0.4 -5.3 -9.7 Precipitation (mm/month) 36.6 41.2 62.5 74.0 67.0 36.0 16.2 7.0 9.0 45.0 42.3 37.0 Number of wet days 12 13 17 18 19 15 8 2 0 10 12 12 Projected change in temperature 4.0 (1.2) 4.1 (0.7) 4.2 (0.1) 3.7 (1.0) 3.7 (0.8) 3.8 (0.1) 4.2 (0.8) 5.0 (2.3) 4.9 (1.9) 4.0 (1.2) 3.5 (0.2) 4.2 (0.7) Projected change in 40% 56% -6% -5% -10% -4% 61% 42% precipitation (%) (76%) 8% (44%) (34%) 7% (33%) (14%) (22%) (27%) 1% (77%) 13% (16%) (24%) (61%) (63%) Change in hot days 1 (1) 1 (1) 2 (1) 1 (1) 1 (1) 1 (1) 1 (1) 2 (2) 1 (1) 1 (1) 2 (1) 1 (1) Change in dry days 2 (0) 3 (0) 1 (0) 1 (0) 3 (1) 2 (0) 2 (1) 0 (1) 0 (0) 3 (2) 2 (0) 2 (-1) Change in frost days -3 (-2) -6 (-3) -10 (-7) -1 (-1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) -2 (-2) -9 (-8) -9 (-6) Change in peak precipitation 10% 30% 5% 21% 26% 55% -34% -12% 83% 52% intensity (%) 6% (10%) (32%) (-2%) (-11%) (36%) (17%) (18%) (-21%) (16%) (68%) (50%) 8% (20%) Change in number of wet days 1 (3) 0 (2) 2 (3) 0 (2) -1 (1) -1 (2) 0 (2) 0 (0) 0 (0) 0 (1) 1 (3) 1 (4) Note: The top four rows are present day values averaged over 625 square kilometers area from the center of the site on google earth. The numbers from fifth row until all rows below are changes by 2050s for RCP 8.5 pathway (RCP 4.5 pathway within the bracket).

122. In the present scenario of uncertainty about a global agreement on actions for restricting greenhouse gas emissions (target of stabilizing the emissions at 2°C as envisaged in The Paris Agreement could still be far away), the RCP 8.5 and RCP 4.5 pathways59 represent the most plausible global concentration pathway range for the future. Moreover, since policy makers and decision makers at country level and at municipal level in any developing country are interested in higher and lower ends of the possibilities in the vulnerability due to future extremes, we shall opt for considering the best choice of the RCP 8.5 and RCP 4.5 pathways in our vulnerability assessment. The rows 5 to 11 in the Table 43 provide the future projections of some of these key climatic variables and indicators for the two sites at a time slice centered on 2050s. 123. Future projection for the Pravaya-Vetka Irrigated Agriculture and Drought (IAD) Subproject site as given in Table 43 suggest an annual mean rise in temperature of 4°C by 2050s compared to the baseline period of 1961–1990 under the RCP 8.5 pathway (0.92°C under RCP 4.5 Pathway). The magnitude of the change projected in this study is significantly higher (a 2°C increase in annual temperature averaged for the Kyrgyz Republic by 2050) than those reported in Second National Communication of the Kyrgyz Republic to the UNFCCC60. Peak warming of close to 5°C is projected in the August and September months (close to 2°C in August and September months under RCP 4.5 pathway). On an annual mean basis, a 17% increase in precipitation is likely (41% under RCP 4.5 pathway) with peak precipitation increase in the month of December. However, a decline in summer precipitation (-10% during July month) combined with air temperature rise of 5°C can have serious implications for a rapid soil moisture depletion thus leading to drought conditions. However, precipitation is projected to increase in all months under RCP 4.5 pathway which could offset the impact of modest surface warming on soil moisture. No significant changes are projected in the number of hot (~15 days in a year) or dry (~21 days in a year) days at this site by 2050s. The frequency of dry days affects regional hydrology and ecosystems and potentially influences agriculture. Future changes in the number of dry days per year can either reinforce or counteract projected increases in daily precipitation intensity as the climate warms. However, on an annual basis, additional 40 days are likely as frost fee days mostly during late autumn and winter months by 2050s. This would mean that the snow melt could start early and be more pronounced in a warmer atmosphere. The longer the time without frost, the longer would be the growing season. While this may seem good — more time should lead to a larger crop yield — it could possibly have detrimental effects on the crops as warmer weather helps pests survive longer which can wreak havoc on crops. No significant change (an increase of just 3 days under RCP 8.5 pathway but could be as many as 23 days in a year under RCP 4.5 pathway) is likely in the number of wet days for this site by 2050s. Future increases in peak precipitation intensity are expected to be an important aspect of climate change, since warming will tend to accelerate the overall hydrological cycle, intensifying the wet extremes61. A decline of 34% in peak precipitation intensity during August but an increase in peak precipitation intensity of 55% in July and 83% in October month is plausible by 2050s at this site under RCP 8.5 pathway meaning thereby that more intense spells of rainfall are projected at this site. Under RCP 4.5 pathway, May, October and November months may experience heavy spells of daily rainfall (36% to 68% increase in peak intensity) by 2050s.

59 Meinshausen, M. S.J. Smith, K. Calvin, J.S. Daniel, M.L.T. Kainuma, and et al., 2011: The RCP greenhouse gas concentrations and their extensions from 1765 to 2300, Climatic Change, 109, 213-241. 60 Government of the Kyrgyz Republic, 2009: Second National Communication of the Kyrgyz Republic to the United Nations Framework Convention on Climate Change, Bishkek. 61 IPCC, 2012: Managing the Risks of Extreme Events & Disasters to Advance Climate Change Adaptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change; https://www.ipcc.ch/report/srex/docs/srex_citations.pdf

3. Assessment of disaster risks and its likely changes due to climate change for Pravaya Vetka Kara Unkur Say Channel sub-project site 124. Climate change is expected to impact extreme weather events or hydrometeorological hazards such as floods and drought in complex and significant ways. There is a meaningful consensus that the increased energy in the global climate system caused by enhanced greenhouse gas effects will also have an impact on weather variability. It is anticipated that the selected site at Pravaya Vetka Kara Unkur Say Channel (in Jalalabad Oblast) would experience more frequent and intense heat waves and an increase in heavy rain events along with widespread issues related to water shortage thus hampering the irrigation requirements at critical stages of the crop growth.

125. Drought, a complex and poorly understood phenomenon, can be categorized as meteorological, hydrological, agricultural, or socioeconomic. The first three categories focus on drought as a physical phenomenon, while the latter considers it in terms of supply and demand, tracking the effects of water shortfall as it ripples through socioeconomic systems. There are a number of measures that are commonly used to monitor drought conditions. The Standardized Precipitation Index (SPI) compares present and historical precipitation levels to identify areas where rainfall is below historical averages. The Palmer drought severity index takes into account not only precipitation but also temperature and estimates of soil moisture. a. Baseline Runoff and Water Balance of River Basin around Pravaya Vetka Kara Unkur Say Project Site 126. The main impact of climate change being considered for the selected sub-project site is on the frequency and/or severity of drought. Therefore, consistent with ADB disaster risk assessment (DRA) guidelines,62 the required CRVA will be integrated into an irrigated agriculture drought risk assessment (DRA), without and with climate change. However, there is no widely accepted DRA method. Therefore, the method summarized below,63 that is broadly consistent with the CRVA method adopted elsewhere, will be applied. Table 44: Integrated DDRA and CRVA methodology adopted in this study Integrated DDRA and CRVA Interpretation Design irrigated agriculture water balance: • 20% reliable monthly stream flow Hazard p (hi) • 20% reliable monthly precipitation • Mean monthly potential evapo-transpiration (ET0) Therefore, drought occurs once in every five years, on average Exposure e Ideally, the irrigation system and its irrigable service area Vulnerability v The annual cropping pattern, without and with climate change 64 Drought Risk f (p (hi), e, v) Loss of agricultural production in drought years Adaptation NA Practical drought risk management (DRM) interventions

127. The Pravaya-Vetka irrigation system consists of three irrigated agriculture sub-systems that are supplied by the existing Vetka Weir (improved by the World Bank) on the Kara-Ungur-Sai (KUS) River. The Pravaya-Vetka core IAD subproject sub-system is located on the right bank of the river. Levaya-Vetka main canal, on the left bank, supplies the inner Levaya-Vetka subsystem (improved by the World Bank), and the outer Jany-Jogorku Akman proposed non-core sub-

62 https://www.adb.org/sites/default/files/institutional-document/388451/drm-project-preparation-practicalguide.pdf. 63 Steley. unpublished. Drought Risk Assessment Method for WB Vietnam Managing Natural Hazards Project. 64 https://www.adb.org/sites/default/files/institutional-document/387631/natural-hazard-data-practical-guide-main.pdf.

system. There may be other diversion weirs also, supplying other irrigation sub-systems, located upstream and/or downstream of the Vetka Weir that are not accounted for as no data is available. An overview is shown in Figure 32. A system overview of the catchment area is provided in Figure 33.

Figure 32: Overview of Pilot area showing irrigated area, hydropost and meteopost locations

Figure 33: System overview of catchment area 28. Hydrologic conditions for the Pravaya Vetka pilot area have been assessed based on available hydro-meteorological data, obtained from Kyrgyz Hydromet, as laid out in the following sections. The approach followed for the analysis consists of: • Analysis of water availability based on historic runoff measurements at Tentyak Charvak station • Analysis of rainfall on the pilot irrigation area based on weighted interpolation between Jalalabad and Ak Terek meteoposts • Analysis of potential evapotranspiration ET0 in the pilot irrigation area and estimation of resulting water demands. 129. Stations that have been considered are shown in Table 45. Data for these stations has been obtained from Kyrgyz Hydromet as well as from the Global Runoff Data Center in Germany. Table 45: Hydropost and meteopost stations considered in the analysis Type Object Post name Location X Y Z Years ID Hydropost 37 R.Tentek Basar- 73.00 41.25 989.84 1925 Sai - Korgonsk -1966 Kischlak 1998 Tscharbak -2003 2010 2016

Meteopost 35 Ak-Terek Zapadnyy 72.8167 41.2667 1748 1947 (Ak-Terek- Sklon -2017 Gava) Ferganskogo Khr., R. Ak- Terek

Type Object Post name Location X Y Z Years ID Meteopost 28 Jalal-Abad Ferganskaya 73 40.95 763 1947 - Dolina 2017

130. Model results for hindcasted (current) conditions are shown in Table 46. The numbers differ from measured values which had been expected given the broad spatial coverage of the baseling global climate model datasets that has been used to derive the data. Relative numbers will anyhow be valid to derive projected future changes based on the measured datasets.

Table 46: Hindcasted (current) meteorological data based on global climate model results Current Max Current Min Current Current no. of Parameter Temp, oC Temp, oC Precip, mm wet days Jan -3.4 -13.1 36.6 12 Feb -1.6 -11.3 41.2 13 Mar 5.7 -4.3 62.5 17 Apr 14.5 3 74 18 May 19.5 6.9 67 19 Jun 24.6 10.5 36 15 Jul 27.4 12.6 16.2 8 Aug 26.2 11 7 2 Sep 21.3 6.1 9 0 Oct 13.4 0.4 45 10 Nov 5.3 -5.3 42.3 12 Dec -0.7 -9.7 37 12 131. The assessment of the present stationary climate / hydro-meteorological water balance of the critical sub-basin will consider: (i) sub-basin stream flow and total irrigation service area (ii) precipitation and (iii) potential (reference crop) evapotranspiration (ET0). 132. Historic stream flow data is available for "R. Tentek-Sai - Kischlak Tscharbak" station in the catchment, upstream the area of interest. Data is available with Kyrgyz Hydromet and has been requested for the project. For Tentek Sai (Tentjak Charvak) data is, in addition, available from the Global Runoff Data Center in Germany. Based on this data, 20% reliable for the monthly stream flows have been extracted, depicting conditions under the present stationary climate situation. The time series for Tentjak Charvak is shown in Figure 34. Catchment area is 1272 km2 for "R. Tentek- Sai - Kischlak Tscharbak". In comparison, the catchment area above the offtake at Pravaya Vetka has an area of 1576 km2, i.e. is 24% larger than the area above the gauge. An overview is provided in Figure IV-4, river network and catchment outlines have been derived from SRTM DEM data (https://www2.jpl.nasa.gov/srtm/). 133. Linear extrapolation has been used to adjust flows measured at the gauge site to reflect flows at Pravaya Vetka. This approach was found suitable as the additional catchment area is spread over a representative altitude range with conditions similar to what is found in the overall catchment area and respectively assumed similar runoff characteristics.

y = -0.0057x + 28.089 R² = 0.0015

Figure 34: Time series of monthly flows (m3/s) at Tentjak Charvak from 1933-1991

Figure 35: Overview of catchment area showing "R. Tentek-Sai - Kischlak Tscharbak" hydropost (HP) as well as the related catchment outline in green, Pravaya Vetka off-take location with the additional catchment in red, as well as the river network in blue. 134. This data is also used to calculate statistics including 20% reliable monthly stream flows as shown in Table 47.

Table 47: Statistics for monthly discharges at Pravaya Vetka site, scaled to 124% based on Tentyak Charvak values in m3/s

Statistics: min max mean median 20% quantile Average 6 243 36 17 11 Jan 6 29 11 11 9 Feb 6 26 11 10 9 Mar 10 53 21 20 15 Apr 26 166 73 71 47 May 26 243 104 104 68 Jun 24 179 81 74 56 Jul 12 113 43 38 29 Aug 9 68 24 21 16 Sep 6 37 16 15 11 Oct 6 41 15 14 11 Nov 6 53 16 14 10 Dec 6 33 12 11 10 135. Future conditions with climate change are dealt with in later sub-sections here. 136. For the present stationary climate, two stations, Ak-Terek, 18 km to the north at 1,748 m altitude and Jalalabad, 23 km towards the southwest at 763 m altitude are available. The pilot irrigation area that is being analyzed in this report is located below 830m altitude. Times of data availability differ as shown in Table 48. Table 48: Time periods of meteorological data availability Station Rainfall (mm) Temperature (°C) Ak-Terek 1947-2017 2007-2016 Jalalabad 1947-2017 1947-2000

137. Due to the same elevation of station and pilot area as well as the better data availability, Jalalabad data has been used for the analysis. The time series data is shown in Figure 36.

Figure 36: Monthly temperature (°C) and rainfall (mm) at Jalalabad between 1947 and 2005 on the left and right axis respectively

138. Statistics for Jalalabad station, including 20% reliable monthly rainfall, are shown in Table 49. The numbers are representative for the pilot area as elevation and geographic conditions are similar and the distance between station and pilot area is only about 23 km.

Table 49: Statistics for Jalalabad monthly temperature and rainfall data considered representative for the Pravaya Vetka pilot site Temperature Statistics: min max mean median 20% quantile unit Overall -9.8 29.1 12.6 13.8 2.0 °C Jan -8.5 2.3 -2.8 -2.6 -5.5 °C Feb -9.6 5.9 -0.1 -0.1 -3.0 °C Mar 2.4 11.7 6.9 7.1 5.4 °C Apr 12.0 17.2 14.4 14.5 13.0 °C May 16.3 22.3 19.1 19.3 18.2 °C Jun 20.7 26.2 23.7 23.7 23.0 °C Jul 23.3 29.1 26.1 26.0 25.2 °C Aug 22.4 26.8 24.6 24.6 23.9 °C Sep 17.4 27.6 20.1 20.0 19.2 °C Oct 9.1 17.1 13.0 13.0 11.8 °C Nov 1.0 10.1 5.8 5.7 3.7 °C Dec -9.8 4.2 0.3 0.5 -1.5 °C

Rainfall Statistics: min max mean median 20% quantile unit Overall 0 242 43 33 8 mm Jan 6 151 47 45 25 mm Feb 16 112 57 56 35 mm Mar 6 203 80 78 40 mm Apr 16 176 73 66 45 mm May 0 148 53 47 26 mm Jun 0 115 28 21 8 mm Jul 0 68 13 7 2 mm Aug 0 43 6 2 0 mm Sep 0 45 9 6 2 mm Oct 0 242 45 30 11 mm Nov 1 207 55 50 23 mm Dec 4 125 47 41 19 mm

139. For the present stationary climate, mean monthly potential reference crop evapotranspiration (ET0) has been estimated for the irrigation area using CROPWAT, i.e. using Penman-Monteith equation that is embedded in the software. The calculations are based on an FAO publication of the Irrigation and Drainage Series, namely, No. 56 "Crop Evapotranspiration - Guidelines for computing crop water requirements” using local meteorological station data from UZGEN station as input. The station is located 30km southeast of Jalalabad at an altitude of 1100m, i.e. showing conditions similar to the Pravaya Vetka scheme. Results are shown in Table 50 below.

Table 50: Current monthly potential evapotranspiration for the pilot area Pravaya Vetka

Annual Annual

Months Total

May Aug

Nov

Sep Dec

Feb Mar Jun

Jan Apr

Oct

Jul

ET0 20 24 48 91 137 175 193 174 127 76 37 22 1124 (mm/month)

140. For the present-day climate, the assessment of the (i) design cropping pattern and (ii) its vulnerability to drought (defined as when the 20% water balance proves unreliable) has been undertaken.

b. Future Water Balance for River Basin around Pravaya Vetka Kara Unkur Say Sub- Project Site 141. For Future Hazard Assessment, we have assessed the future, with climate change, hydro- meteorological water balance of the critical sub-basin. It provides projections of (i) sub-basin

stream flow and total irrigation service area (ii) precipitation and (iii) potential (reference crop) evapo-transpiration (ET0). 142. Future stream flow under climate change conditions was calculated using future monthly rainfall trends as well as evapotranspiration trends considering their increasing or decreasing effects on runoff. The results are an indication of trends considering the methodological uncertainties. Results are shown in Table 51. Table 51: Projected (for 2050s Time slice) monthly discharges at Pravaya Vetka site in m3/s

Statistics: Current Mean for RCP4.5 Mean for RCP8.5 Average 36 45 33 Jan 11 17 11 Feb 11 13 5 Mar 21 28 26 Apr 73 92 69 May 104 117 91 Jun 81 97 70 Jul 43 54 36 Aug 24 41 23 Sep 16 17 16 Oct 15 18 13 Nov 16 25 22 Dec 12 19 12

143. Mean Monthly Temperature and Potential Evapo-transpiration: Climate change is expected to have an impact on the temperature regime of the project area. Projections for the area of interest for the end of the century (Table 52) show a significant increase in temperature and precipitation increasing in winter and decreasing in summer months. In addition, the number of days considered to be hot as well as the number of dry days are increasing. The temperature changes are expected to have an effect on the snow line leading to earlier snowmelt and respectively earlier runoff, while the reduced number of dry days hints that intensities, and with that peak flow runoff events will increase. Table 52: Climate conditions at Pravaya Vetka Kara Unkur Say Channel, showing projected (for 2050s) conditions for RCP 8.5 (RCP 4.5) pathway projections Para- Δ temp, Δ precip, Δ hot Δ dry Δ frost Δ peak precip Δ wet meter °C % days days days intens., % days Jan 4.0 (1.2) 40% (76%) 1 (1) 2 (0) -3 (-2) 6% (10%) 1 (3) Feb 4.1 (0.7) 8% (44%) 1 (1) 3 (0) -6 (-3) 10% (32%) 0 (2) Mar 4.2 (0.1) 56% (34%) 2 (1) 1 (0) -10 (-7) 30% (-2%) 2 (3) Apr 3.7 (1.0) 7% (33%) 1 (1) 1 (0) -1 (-1) 5% (-11%) 0 (2) May 3.7 (0.8) -6% (14%) 1 (1) 3 (1) 0 (0) 21% (36%) -1 (1) Jun 3.8 (0.1) -5% (22%) 1 (1) 2 (0) 0 (0) 26% (17%) -1 (2) Jul 4.2 (0.8) -10% (27%) 1 (1) 2 (1) 0 (0) 55% (18%) 0 (2)

Aug 5.0 (2.3) 1% (77%) 2 (2) 0 (1) 0 (0) -0.34 (21%) 0 (0) Sep 4.9 (1.9) 13% (16%) 1 (1) 0 (0) 0 (0) -12% (16%) 0 (0) Oct 4.0 (1.2) -4% (24%) 1 (1) 3 (2) -2 (-2) 83% (68%) 0 (1) Nov 3.5 (0.2) 61% (61%) 2 (1) 2 (0) -9 (-8) 52% (50%) 1 (3) Dec 4.2 (0.7) 42% (63%) 1 (1) 2 (-1) -9 (-6) 8% (20%) 1 (4) Note: changes by 2050s for RCP 8.5 pathways (RCP 4.5 pathways within the bracket). 144. Potential evapo-transpiration is, next to other parameters, directly linked to the temperature regime of a location. With climate change, the mean temperature is increasing while the range between minimum and maximum temperatures is assumed to be maintained. Calculations for potential evapo-transpiration have respectively been adjusted yielding results as shown in Table 53. Table 53: Projected (for 2050s) monthly potential evapo-transpiration for the pilot area Pravaya Vetka, increased based on higher temperatures as compared to the baseline

Month ET0 (mm/month) ET0 (mm/month) for ET0 (mm/month) for current (baseline) RCP 4.5 pathway RCP 8.5 pathway Jan 20 22 26 Feb 24 29 37 Mar 48 48 58 Apr 91 96 102 May 137 139 147 Jun 175 179 190 Jul 193 195 205 Aug 174 179 185 Sep 127 136 144 Oct 76 78 83 Nov 37 38 43 Dec 22 23 28 Annual total 1124 1163 1248

145. A summary of hydro-meteorological projections for the period 2050s at the project site is provided in Table 54.

Table 54: Summary of hydro-meteorological projections for 2050s at the project site Hydro-meteorological J F M A M J J A S O N D parameter Mean monthly temp, RCP 1.6 0.6 7.0 15.4 19.9 23.8 26.9 26.9 22.0 14.2 6.0 1.0 4.5 pathway (°C) Mean monthly temp, RCP 1.2 4.0 11.1 18.1 22.8 27.5 30.3 29.6 25.0 17.0 9.3 4.5 8.5 pathway (°C) Mean monthly ET0 22 29 48 96 139 179 195 179 136 78 38 23 RCP 4.5 pathway (mm) Mean monthly ET0 26 37 58 102 147 190 205 185 144 83 43 28 RCP 8.5 pathway (mm) 20% reliable monthly precipitation, RCP 4.5 14 36 47 53 51 32 10 4 0 2 18 37 pathway (mm) 20% reliable monthly precipitation, RCP 8.5 11 27 55 68 42 25 7 2 0 2 18 33 pathway (mm) 20% reliable monthly runoff / stream flow, RCP 4.5 14 11 20 59 77 67 36 27 12 13 16 16 pathway (m3/s) 20% reliable monthly runoff / stream flow, RCP 8.5 9 4 19 44 60 48 24 15 11 10 14 10 pathway (m3/s)

146. For Future Vulnerability Assessment due to climate change, the vulnerability of the design cropping pattern to drought has been assessed (cropping pattern changes without and with climate change).

c. Capacity Requirements for Proper Management of Water Resources in the Project Area 147. Strengthening disaster risk reduction and response capacity building activities among water user associations, extension workers and farmers and other stakeholders should include: (i) Monitoring the weather (e.g., gale force winds) / flood / drought forecasts and early warnings issued by Regional and Local Kyrgyz Hydromet Offices; (ii) Preparing probabilistic risk analyses; (iii) Designing disaster risk reduction measures through both structural and non-structural solutions; (iv) Training national/local agencies and staff members in understanding and using hazard information; and (v) Developing a national/oblast level platform for data on natural hazard risks (floods and droughts).

d. Future Water Balance for River Basin around Kara Darya Sub-project Site and Implications of Climate Change on Flood Potential

148. The estimate of possible change in the surface runoff by all basins considering the glacial fluid loss is illustrated in the Figure 37 below – taken from 3rd NC of KYG to UNFCCC (2017). This

curve indicates a significant runoff reduction under both the RCP scenarios and options for precipitation changes. The decrease in the surface runoff will be the greater, the greater are the expected the surface temperature increase and precipitation decrease. For the most unfavorable scenario (RCP 8.5 Scenario and annual precipitation reduction by 5%), the runoff may be reduced by approximately 40%. This probabilistic estimate can serve as a basis for identification of the required adaptation measures.

Figure 37: Likely changes in surface runoff contributed by water flow due to changes in precipitation for RCP 8.5 and RCP 4.5 pathways V. Climate Change & Disaster Risk Framework for Climate Resilience

A. Mainstreaming Framework for CCA and DRR

149. Addressing the drought related food security issues and water related disaster risks require: • Increase knowledge-sharing and understanding with respect to communities at risk to water-related disasters, especially in a changing climate; • Adopt and implement monitoring and people-centered early warning systems for communities at risk to water-related disasters; • Manage extremes such as thermal stress, flood / drought events etc. through a participatory approach to mitigate damages and losses of life and property (enhance the capacity to anticipate, cope with, resist and recover from hazard impacts); • Adopt integrated disaster risk management approaches, including an appropriate mix of structural and non-structural approaches, to reduce mortality and economic losses (mitigate damages and losses of life and property) for water-related disasters; and • Apply a preparedness approach to water-related disaster management that sees the needs of communities being met, down to the last mile (past trends in hydro-climatic events at any locality is important for creating and enhancing community resilience). 150. The points listed above need to be practiced within the framework as illustrated in Figure 38 below. The recommended options/choices for adaptation strategies in the agriculture and water sectors in Kyrgyz Republic are discussed in the following sub-sections.

Figure 38: A conceptual framework for mainstreaming CCA and DRR in water and food security within the Kyrgyz Republic

B. Potential Adaptation Responses / Strategies

151. As the impacts of climate change become more clearly understood, so too does the need for people to effectively respond and adapt to these changes. The mountainous regions of Kyrgyzstan are particularly vulnerable to climate-related impacts and risks such as snow avalanches, landslides and flash floods. The Pravaya Vetka Kara Unkur Say sub-project site in Jalalabad oblast is characterized by both drought and flood hazards (in different months of the year – e.g., floods during April-May and droughts during August-September) and associated food insecurity problems and decreases in the volume of irrigation water.

152. As a mountainous country, the Kyrgyz Republic is particularly vulnerable to numerous disasters triggered by natural hazards. Of all the natural disasters, avalanches, floods and mudflows, droughts and glacier melt are most important in the country affecting water sector and bring about significant damages to the population, economic activities and infrastructure. The prevention measures for these disasters can bring the significant economic benefits and minimize threats to the ecosystems, economic development, property and infrastructure. The Government of the Kyrgyz Republic approved a set of priority directions for Adaptation to Climate Change in the Kyrgyz Republic till 2017 on October 2, 2013 #549 [3.5]65. These Priorities were aimed to establish a national policy to mobilize resources aimed to minimize the negative risks and to use the potential of climate change for the sustainable development through the adaptation measures implementation in the sectors most vulnerable to climate change.

65 Priority Direction for Adaptation to Climate Change in the Kyrgyz Republic till 2017. http://climatechange.kg/wpcontent/uploads/2014/12/CCA_Priorities_Rus_Kyrg_Eng1.pdf

153. It’s expected that in the longer-term, rising surface temperatures and decreasing glacier areas would lead to a decline in water resource availability. A fall in water supply due to diminished river flows will affect agriculture, the Kyrgyz Republic’s main economic sector. The country joined the Pilot Programme on Climate Resilience (PPCR) in 2015 and since then is actively seeking to address these issues by developing a climate finance coordination system in order to effectively mainstream climate change considerations into sustainable development planning. The Climate Finance Centre acts as a bridge with major climate donors, Multilateral Development Banks (MDBs) and development partners active in the country. This Centre is a central unit coordinating the Kyrgyz Republic’s efforts in accessing climate funds and channeling them into transformative investments supporting national development priorities. The Climate Finance Centre has a well-defined scope of work including establishing favorable investment environment for implementation of adaptation and mitigation projects for climate resilience in agriculture, irrigation, hydropower, renewable energy sources; for promotion of energy efficient and low-carbon technologies in transport and healthcare; for prevention of emergencies, sustainable land use, and changes in land and forest management within the process of achieving target indicators of the Development Strategy of the Kyrgyz Republic for the period up to 2040.

154. To date, most adaptation efforts have focused on reactive, short-term and site-specific solutions to climate-related vulnerabilities. Although important, these responses often fail to address the root causes of neither vulnerability, nor shed light on how to proactively spur larger- scale and longer-term adaptation that has positive effects on socio-economic development. The capacity and resource deficits at the sub-national and local levels also limit support for water and adaptation governance. In the follow-up development strategies, the economic sector activities should be harmonized with adaptation to climate change. The adaptation measures should be developed based on the analysis of the climate change risks, vulnerability assessment of the economic sectors, environment and population. Attention also needs to be focused on improving information tools to provide monitoring of the climate change process and climate risk assessment, involvement of civil society in the climate change adaptation process, increasing the scientific capacity to adapt to climate change, and organization and promotion of cross-border cooperation on climate change adaptation.

155. Climate proofing needs to be considered at the project design stage. Although climate proofing could increase the upfront costs of the infrastructure projects (such as water reservoirs, raising the sides of rivers and primary channels through protective walls etc.), such higher costs could be economically justified by lower total life cycle costs over the long lifetime of most infrastructure and by the high probability of climate-related damage. A robust adaptation pathway is one that allows for flexibility in dealing with future climate vulnerability in the face of newly available evidence. It requires continued monitoring and review of ongoing climate change measures to avoid locking in long-term investments in potentially inefficient undertakings. For example, to ensure that the irrigation rehabilitation project will be climate proofed so that reactive actions are not needed. We shall build a common understanding of current adaptation needs at the selected project levels. Prioritizing the co-production of adaptation responses would be undertaken that can yield appropriate, tangible and lasting benefits for water availability and would ensure that project impacts across the selected sites are sustainable in a changing climate.

1. Potential Adaptation Measures for Agriculture Sector (Irrigation Management) 156. A broad range of adaptation measures to overcome negative impact of climate change on agriculture sector has been described in various published publications, reports and scientific

reports. The second and third national communications of Kyrgyzstan to the UNFCCC (2009, 2017) identify some of the impact and adaptation strategies for the agricultural sector. 157. Climate change contemplates both positive and negative consequences for plant-growing of the Kyrgyz Republic. The following main negative consequences of expected climate changes are plausible: • increase in number of days with higher air temperature; • enhancement of climate dryness and increase in drought repetition during late summer season; • increase in share of rain-shower precipitation (more rainfall than snowfall in early spring); • reduction in period with snow cover (winter season); • increase in inter-annual and intra-seasonal variability of weather patterns; • migration of agro-climatic wetting zones to the north; • decrease in yield capacity of wheat and cotton crops; and • development of infectious diseases and insects during cropping season.

158. The development and use of seasonal climate forecasts also must be considered to identify the key adaptation measures that can be implemented such as: • no-till farming • plant-growing diversification (e.g., shifts in cropping calendar) • implementation of effective irrigation systems • optimization (adjustment) of terms for performance of agro-technical activities to weather pattern • refurbishment of agricultural vehicle and equipment fleet • preparation and professional improvement of agricultural specialists • improvement of plant-growing insurance system.

159. A recent publication regarding climate change adaptation66 is based on an extensive inventory of existing and well-proven measures in the agricultural sector ( Table 55). This overview is very relevant for the proposed irrigation rehabilitation project in Kyrgyz Republic. Table 55: An inventory of existing and well-proven measures for climate change adaptation in the agricultural sector Adaptation needs Measure Mechanism to overcome the impacts of climate change I. Improve (1) Implement regional Enhances effectiveness of resiliency and adaptation plans adaptation measures adaptive capacity (2) Improved monitoring and Mitigates consequences of adverse early warning events (3) Improve coordination Enhances effectiveness of planning adaptation measures (4) Innovation and Improves effectiveness of adaptation technology measures and reduces costs

66 Iglesias, A., L. Garrote, 2015: Adaptation strategies for agricultural water management under climate change in Europe - Agricultural water management, 155: 113-124.

Adaptation needs Measure Mechanism to overcome the impacts of climate change II. Response to (5) Innovation: water use Increases water availability changes in water efficiency(6) Improve soil moisture Increases water use efficiency availability retention capacity (7) Small-scale water Increases water management reservoirs on farmland flexibility at the local level (8) Improve the reservoir Increases management flexibility and capacity water availability at regional level (9) Water reutilization Increases water availability (10) Improve water charging Decreases inefficient use of water and trade (11) Re-negotiation of Improves water use efficiency allocation agreements (12) Set clear water use Improves water use efficiency priorities (13) Integrate demands in Increases management flexibility conjunctive systems and water availability III. Response to (14) Create/restore wetlands Reduces flood peaks floods and droughts (15) Enhance flood plain Reduces flood vulnerability management (16) Improve drainage Reduces extent and duration of systems flooding (17) Farmers as ‘custodians’ Decreases risk of flood damages of flood plains (18) Hard defenses Decreases risk of flood damages (19) Increase rainfall Reduces flood peaks at the local interception capacity level (20) Introduce drought Improves agronomic water use resistant crops efficiency (21) Insurance to flood or Decreases economic losses to the drought farmer IV. Response to (22) Change in crops and Decreases economic risk to farmers increased irrigation cropping patterns requirements (23) Improve practices to Decreases the need for additional retain soil moisture water to crops (24) Develop climate change Mitigates impacts of climate change resilient crops V. Response to (25) Relocation of farm Maintains industrial activity changes in processing industry agricultural land (26) Addition of organic Recovers soil functions use material into soils (27) Introduce new irrigation Develops new agricultural land areas

Adaptation needs Measure Mechanism to overcome the impacts of climate change VI. Response to (28) Improve nitrogen Reduces agricultural diffuse deterioration of fertilization efficiency pollution

Adaptation needs Measure Mechanism to overcome the impacts of climate change water and soil (29) Soil carbon Reduces soil erosion and improves quality management and zero tillage soil water retention capacity (30) Protect against soil Reduces land degradation erosion VII. Response to (31) Increase water allocation Improves ecosystem services, loss of biodiversity for ecosystems effective at the global level (32) Maintain ecological Improves biodiversity with positive corridors global consequences (33) Improve crop Improves biodiversity diversiﬁcation

160. Seasonal runoff flow of rivers is another critical factor leading to droughts in parts of Kyrgyzstan. Climate change can lead to lower river flows once 'peak water' (from glacial snow melt) has passed through particularly during late summer and autumn seasons. 161. Any strategy to adapt agriculture and food systems to a changing climate must exploit the diversified means of climate resilient strategies67, including irrigation for agriculture. Variability and extreme rainfall events have the potential to transform agricultural production systems (rain-fed or irrigated) and sub-sectoral diversifications of agricultural production (including crops, live-stock, forestry and fisheries) as well as downstream production activities (like processing, marketing and off-farm activities) which could help to smooth agricultural consumption and production along the value chain68,69. The ability to circumvent the negative impact of climate and weather variability in agricultural production is also an important consideration for climate-smart agriculture (CSA) and for maximizing its benefits of enhancing agricultural livelihoods and economic development. 162. CSA is an emerging concept and practice that seeks to adapt agricultural production to climate change and weather variability, while maintaining agricultural productivity, biodiversity and the ecosystem that sustains food security, livelihoods and economic development. CSA seeks to enhance productivity, water conservation, livelihoods, biodiversity, resilience to climate stress, and environmental quality70,71. The overall outcome of CSA is to improve agricultural livelihood incomes, promote food security and sustainable agricultural development by ensuring that agricultural production systems are best suited to respond to the challenges of climate change and variability. The resilience of agriculture to climatic changes and variability could boost agricultural production and broadly contribute to sustainable development.72 163. Technologies and practices do exist or have been developed in different parts of the world, to facilitate adaptation to climate change by the agriculture sector. These range from improved

67 Vermeulen, S.J., Aggarwal, P.K., Ainslie, A., Angelone, C., Campbell, B.M., Challinor, A.J., Hansen, J.W., Ingram, J.S.I., Jarvis, A., Kristjanson, P., Lau, C., Nelson, G. C., Thornton, P.K., Wollenberg, E., 2012: Options for support to agriculture and food security under climate change. Environ. Sci. Policy 15, 136–144. 68 Liverman, D.M., Kapadia, K., 2010: Food systems and the global environment: A Review. In: Ingram, J.S.I., Ericksen, P.J., Liverman, D. (Eds.), Food Security and Global Environmental Change. Earthscan, London. 69 Nelson, G.C., Rosegrant, M.W., Koo, J., Robertson, R., Sulser, T., Zhu, T., Ringler, C., Msangi, S., Palazzo, A., Batka, M., Magalhaes, M., Valmonte-Santos, R., Ewing, M., Lee, D., 2009: Climate Change: Impact on Agriculture and Costs of Adaptation. Food Policy Report #19, IFPRI, Washington, DC. 70FAO, 2013: Climate-Smart Agriculture Sourcebook, Food and Agriculture Organization of the United Nations, Rome. 71 Neufeldt, H., Kristjanson, P., Thorlakson, T., Gassner, A., Norton-Griffiths, M., Place, F., Langford, K., 2011: Making Climate-Smart Agriculture Work for the Poor. World Agroforestry Centre (ICRAF), Policy Brief 12, Nairobi, Kenya. 72 Nwanze, K.F., Fan, S., 2016: Climate change and agriculture: strengthening the role of smallholders. In: Global Food Policy Report. International Food Policy Research Institute, Washington, DC, pp. 13–21.

weather forecasts to water conservation, drip irrigation, sustainable soil management, better livestock management, and changes in crop types and planting, among others. Some of these measures may need investment while others primarily require improving awareness and building capacity to deal with new practices. The water saving and farm mechanization technologies that allow for and promote diversity are more likely to provide a strategy, which strengthens agricultural production in the face of uncertain future climate change scenarios. It is important here to understand local contexts – especially social and cultural norms – when working with sub-national and local stakeholders to make informed decisions about appropriate technology options. Some of these are: • Planning for climate variability and change, • Sustainable water use and management, • Soil management, • Sustainable crop management, • Sustainable livestock management, • Sustainable farming systems, and • Capacity building on water and soil management and stakeholder organizations (WUAs etc.).

2. Specific Adaptations and Investment Actions in Agriculture sector for Mitigating the Adverse Impacts of Droughts 164. Agriculture is the largest consumer of water in the country and employs roughly 30% of the workforce in the Kyrgyz Republic. Changes in water availability will impact mainly on the rural societies that depend on agriculture. During the design phase of the water sector rehabilitation project at Pravaya-Vetka Kara Unkur Say Channel sub-project site in Jalalabad Oblast, specific adaptation and investment actions for climate change should be considered. The ADB supported irrigation rehabilitation work at this sub-project site would, however, primarily comprise of cleaning of primary / secondary channels (removal of sediments and boulders as sections of the primary and secondary channels as well as main rivers and its tributaries have a huge load of sediments and large stone bounders carried into it during floods; some sections might require dredging as well) and re-sectioning, concrete lining of selected earthen channels, repair of concrete lining, repairing and/or construction of water metering structures, construction of outlets, repair of regulating structures, and construction of concrete flumes spread over the main river and its tributaries within the sub-project area. A portion of the investment could also go for the small- scale community driven irrigation infrastructure project activities (e.g., micro-drip irrigation for horticulture in backyard gardens which would primarily support female gender in their livelihood earnings) within the tertiary channels to be undertaken / supervised by WUAs. 165. The main conclusions of the projected climate change at the sub-project site as described above are: • Higher temperatures are to be expected throughout the year with a peak during summer. This increase is most pronounced during summer season and least during spring season. Jalalabad shows a slightly lower increase in temperature than in Osh and Batken oblasts during 2050s. • A -3% decline in annual mean precipitation around 2050s could be expected in Jalalabad oblast. The winter precipitation around this decade would increase by around 9%. During the summer season, a decrease in precipitation of as much as -35% in Jalalabad is likely. Reductions are expected to be more pronounced during summer (-26%

in September at the sub-project site). At the same time, higher daily rainfall intensities are projected in June, September and November months.

166. Using these climate change projections, the key impacts at the sub-project sites are: • Higher crop water requirements of about 15% to 20% by 2050s compared to the reference period. • Seasonal shift with earlier planting opportunities and therefore irrigation demands earlier in the season. • Substantial reduction in runoff to streams and rivers as a result of higher evaporation rates and precipitation reductions during summer. • Higher flash flood risks due to projected extremes in daily rainfall, especially during the early spring season.

167. Appropriate adaptation and investment actions for the sub-project in general have been identified as under: • Design crop water requirements at 10-15% higher compared to the baseline climate data. Design criteria for canals should also be 10-15% higher. • Design of storage capacity should be substantially higher to compensate for the projected decrease in base flows and increase in demand. Exact numbers should be determined at the project level. • In general, flooding is not considered as the most threatening consequence of climate change in this sub-project area. However, flood resilience through appropriate site-specific measures (mudflow passovers) should receive additional attention during early spring. • Training of water managers of WUA on ways to minimize water loss and maximize water productivity through effective deployment of sustainable water conservation technology is essential to ensure implementation of these measures and be prepared during management of the system over the coming decades. • Awareness raising amongst farmers is required to alert them that changes in climate will influence water delivery and gradually lead to changes in cropping pattern and farm water management. • Water pricing may in most cases not lead to reduction in water consumption but will enable the water resource manager to cover incremental costs to make water services more reliable.

168. Options for agricultural water management and adaptation to climate change include rainwater storage for surface and underground reservoirs, various irrigation schemes to optimize water and crop productivities. They also include dry-land farming, improved agronomics practices, livestock water productivity, seed genotype improvements, and improved policy and institutional frameworks. We would need to embrace a range of practices, including in situ moisture conservation and ex situ water management, and identify the best option of adapting to climate variability and change and facilitating water-centered development to increase food security. The following recommendations address some of problems related to water needs in the agricultural sector: • Adoption of crop varieties and forage with increased resistance to heat stress, shock, and drought is critical for minimizing the adverse impacts of climate change. Similarly, grassland management, which includes erosion control, grazing control, making strategic

watering points available for livestock, and different forms of water harvesting are key adaptation strategies to minimize effects of climate change and variability. • Reliable small-scale water control systems should be developed, soil fertility and the moisture holding capacity of agricultural soils should be built up together with rehabilitation of irrigation infrastructure. • The capacity of communities to use technologies that minimize water loss and maximize water productivity should be strengthened, and these technologies should be disseminated more widely. • Local WUAs should be supported in developing more integrated, climate proof (process that makes projects, strategies, policies and measures resilient to climate change, including climate variability) crop-water livestock practices. • Coordinated monitoring and evaluation of the impact of water policies, as well as institutional arrangements at basin levels, should be pursued in addressing transboundary water issues. • Water pricing policies should be re-considered. Pricing will improve irrigation efficiency and institutional performance at local and regional scales and will create a sense of community ownership of investments in water.

C. Rehabilitation of existing hydromet monitoring network and requirements of additional weather stations in Kyrgyz Republic

169. Knowledge on climate risk is essential for development of strategies and plans ensuring climate resilience, disasters risks mitigation and climate-oriented management of country’s economic sectors for the medium— and long–term perspectives. Significant complexity of Kyrgyzstan’s terrain — deep roughness, various expositions of mountain slopes for sun and airflows cause exclusive variety of climate characteristics and determine well—defined vertical climate zoning. There are four notably different climatic zones/ belts, namely, the valley—piedmont zone/belt, mid—mountain zone/belt, high—mountain zone/belt, and nival zone/belt. 170. All-round rise of average annual air temperature is registered on the territory of Kyrgyzstan. The most notable local areas with temperature rise during all seasons are the most populated districts of Kyrgyzstan – the central part of Chui valley (Bishkek, Kara-Balta, Jany-Jer Meteorological Stations) and valley zone of Jalalabad and Osh Oblasts (Jalalabad and Uzgen Meteorological Stations).

171. In general, across the territory of Kyrgyzstan for the period 1976-2014, every 10 years an increase in annual precipitation by 2.9% is observed, with the largest growth registered in summer and winter periods - by 7.0 and 5.2% for 10 years, respectively. In spring, the reduction of precipitation is registered on 60% of the territory. The most intensive reduction is observed in piedmont and high-mountain areas of Chui Oblast. The character of changes in precipitation in autumn is almost the opposite of that in summer — the reduction of precipitation is observed in the west of the territory, and the rise – in the east. The most frequency of heavy precipitations – 10 cases of 18 – has been registered in Jalalabad Oblast. 172. Snow avalanches are the most dangerous hydrometeorological disasters that pose significant threat for people, constructions, transportation lanes, energy bridges and communication lines. There are a lot of cases associated with cattle death and forest damage caused by avalanches. 105 thousand square kilometers — 53% of the entire territory of Kyrgyzstan is subjected to avalanche influence. Within 779 areas of avalanche formation, there are more than 30,000 avalanche site, 3% of which constitute specific threats to people.

173. USAID facilitated upgrading data monitoring systems for water allocation decisions and providing training to use in transboundary and national water management decisions. Successful installation of a demonstration automated meteorological station led to the procurement of five additional stations that are being used to support improved snowmelt and hydro-meteorological forecasting. Training of hydro-meteorological agency (hydro-met) staff resulted in the successful installation of two World Bank-financed transboundary stream monitoring stations. Training of Kyrgyz staff at the hydrological forecast center has led to improved cooperation and hydrologic forecasts for water resources shared by the Kyrgyz Republic and Uzbekistan. A USAID-funded system for communication of this data between all hydrological stations in the region is also in place and facilitates to receive digital data from the automated stations and to re-distribute it to local and regional users in near real-time. USAID's on-farm water management activities have made substantial progress with the design, procurement, manufacture and installation of fifteen water control and measure structures in the Kyrgyz Republic. Irrigation water measuring structures constructed in Osh Oblast are used in a World Bank Project supported water user associations in the Kyrgyz Republic. Adoption of new technology for a landslide monitoring and forecasting system within Kyrgyz Republic has been supported by United Nations Development Programme through funding from the World Bank during the past decade. In addition, the World Bank has also facilitated the Republic towards Improvements in achieving the accuracy and timeliness of hydro- meteorological services as part of the Central Asia Hydrometeorology Modernization Project. 174. However, a number of rehabilitation measures are still required within the Kyrgyz Hydrometeorological Offices to upgrade their observational hydro-meteorological /snow melt observational network as of this date. The Head of the International Cooperation Division in Kyrgyz Hydromet, Bishkek with input support from Head of the Meteo-forecast Division and Head of Hydro-forecast Division has proposed construction work at some of its monitoring sites, re- fabrication of their computing system for data handling and processing and an expansion of snow monitoring sites and other instruments within their hydro-meteorological network in Kyrgyzstan (in hot-spot regions for weather and climate related disasters within the country) (see Figure 39). The list of these stations with their coordinates is given in the Table 56 below. This project proposal with an estimate of costs is attached as Annexure-2 here. The support requested for rehabilitation measures proposed by Kyrgyz Hydromet are very crucial for water resource assessment as well as planning for mitigating the disasters due to natural hazards.

Figure 39: The draft hydrological posts requiring construction or restoration

Table 56: The draft list of hydrological posts requiring construction or restoration Chui river basin: 1.River Kara-Kudzhur – village Sary-Bulak 2.River Sujek – mouth of the stream Ichke-Sai 3.River Karakol – mouth of the river Irisu 4.River Kochkor – village Kochkor 5.River Chon-Kemin - mouth of the river Karagaili-Bulak 6.River Chu – settlement/village Nizhne-Chuiskij 7.River Chu – village Cholok 8.River Isyk-Ata – above mouth of the river Tujuk (village Jur’jevka) 9.River Ala-Archa – mouth of the river Adygine 10.River Chu – village Miljanfan Batken oblast: 11.River Abshir-Sai – village Uch-Terek Jalal-Abad oblast: 12.River Shaidan-Sai – village Shaidan 13.River Kegart – village Mikhailovka 14.River Chatkal –mouth of the river Ters 15.River Gava-Sai –mouth of the river Ters 16.River Padysha-Ata-mouth of the river Tostu Osh oblast: 17.River Ak-Buura –mouth of the river Min-Teke 18.River Ak-Buura –mouth of the river Papan 19.River Kyrgyz-Ata –village Kyrgyz-Ata 20.River Karadar’ja- river Uzgen