Climate Change and Sustainable Water Management in Central Asia

Project Number: 44066-012 (RETA 7532) June 2012

Knowledge Product of RETA 7532: Water and Adaptation Interventions in Central and West Asia

Prepared by FCG Finnish Consulting Group Ltd

This consultant’s report does not necessarily reflect the views of ADB or the Government concerned, and ADB and the Government cannot be held liable for its contents. (For project preparatory technical assistance: All the views expressed herein may not be incorporated into the proposed project’s design.

Climate Change and Sustainable Water Management in Central Asia

FCG International

Introduction

Climate change has well been documented all over the world in vast number of scientific investigations and global climate model simulations. The field observations in Central Asia indicate that the climate has been warming several decades and the consequences of this trend have already been observed in the everyday life of the population. However, there is still substantial lack of knowledge on the phenomena related to the climate change and its impacts on environment and human life.

New climate and hydrological models show that river water must be seen partly as a non- renewable resource in Central Asia. Today about one third of the water in rivers originates from the mountain glaciers that are quickly losing their volume due to global climate warming. During the past decades the rivers have received significant amount of excess water from the melting glaciers, but in the future this source will increasingly be lost as a consequence of vanishing glaciers.

Climate change will also make the plains hotter and drier. In the future, water shortage will be a serious problem for national economy and environment. The need of water will increase at the same time when the river discharges will diminiish. This situation may generate water management disputes and conflicts between people living in the mountains and plains. Therefore, the decision-makers of the countries should urgently enhance regional co-operation and launch programs to increase resilience to negative climate change impacts as well as plan and implement adaptation interventions.

Increasing temperatures in the mountains will also result in thawing permafrost which again may mobilize massive landslides and mudflows. Every year these disasters destroy settlements, agricultural lands and infrastructure. In river basins where snow-melt is the main source of water, spring floods may become more frequent. The Central Asian mountains have unique landscapes and nature. The overall climate change will degenerate mountain ecosystems which maintain biodiversity - rare animals and plants.

The Central Asian countries are signatory members of the United Nations Framework Convention on Climate Change (UNFCCC). The objective of the Convention is to achieve stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner. The countries should promote sustainable development. Policies and measures to protect the climate system against human-induced change should be appropriate for the specific conditions of each country and should be integrated with national development programs, taking into account that economic development is essential for adopting measures to address climate change. Capacity building is needed to increase understanding of needs for policy and strategy formulation and consequent investments for adaptation and resilience to climate change.

It is important that all the countries have strategies to mitigate climate change by e.g. controlling GHG emissions and protecting vegetation resources. Especially in Central Asia the countries should also have strategies to develop resilience and adaptation to climate change. Most existing strategies underestimate the problems caused by climate change and also wrong conclusions have been presented. For example, some documents predict that water discharges in rivers will increase, because glaciers will start to melt. However, glaciers have already been 2

melting a century, their volume has decreased and therefore discharges will diminish all the time.

It is self evident that the whole population of Central Asia will suffer from the climate change impacts in their daily life. Many national strategy papers on climate change ask questions, what the changes really are, how vulnerable the region is and what can be done to increase adaptation and resilience against the likely impacts. The purpose of this knowledge product is to find answers to these questions based on new scientific information and an ADB funded project. This information is needed when investment and action plans are drafted and when institutional management capacities are developed.

In this booklet, results of research activities conducted by an ADB funded project in Central Asia are described. The project "Water and Adaptation Interventions in Central and West Asia" (RETA 7532) combined field observations with sophisticated satellite based data and created models to demonstrate impacts of climate change on the hydrology of the Aral Sea Basin.

The project collected past temperature and precipitation data into a 3-dimensional map matrix of the Aral Sea Basin. This baseline served as a reconstruction of past and present climate of the basin. The climate change models used were based on the extensive international research on physics of climate systems (Global Circulation Models). The future climate change scenarios have been drafted by IPCC (Intergovernmental Panel on Climate Change) and several research institutes have made projections for future climate. In the project, this information was used to model changes of the climate parameters in the map matrix.

The hydrological model SPHY utilized the past and predicted climate data to analyze rainfall, snow formation and melt as well as mass balance of glaciers. Outputs from the model included water discharges in all the rivers in detailed scale. Such hydrographs describe changes of water quantity over the annual water cycle and future trends can clearly be seen.

Water allocation model analyzes the changing needs for water for different purposes in each river section were undertaken using the WEAP framework. It demonstrates well that at the same time when water is more and more needed, the reality is that river discharges will radically go down.

ASSESSING THE CLIMATE CHANGE

Earth’s climate varies and changes due natural reason but during the most recent decades the anthropogenic climate change has become more and more evident; the emissions of the greenhouse gases have kept on increasing and this has led to a very rapid increase of global temperatures. This rapid change has many impacts on the earth’s natural environment that are further reflected to human societies. Some of these iimpacts may be positive but most of them have so far been found to be negative and in some areas of the globe the future living conditions may be worsened very seriously.

The foreseen positive impacts are related to better growing conditions at cold and cool climate regions where the warming makes growing season longer and increases temperature sums. As well, heating energy demand is predicted to become smaller and there is a possibility that e.g. hydropower potential will increase. The most seriously negatively affected areas of the globe are the areas that already now suffering of the cllimate related hazards. These areas are typically experiencing shortage of water leading to drought or their climate may be characterized 3

by extreme weather events like floods. The sea level rise is one of the future risks the coastal areas must be prepared for (IPCC, 2012).

Climate change is linked with the increase of extreme weather events, i.e. increase of extreme winds, heat waves and torrential precipitation. These extreme events can lead to disastrous impacts at different sectors of society. In Central Asia there has been a likely increase in the number of warm days and decrease in the number of cold days. The change in the number of heavy precipitation cases is not clear (Alexander et al., 2006). According to the climate projections the maximum temperatures experienced roughly once in 20 years in the past climate will be met once in 2-5 years at the middle of current century and almost annually at the end of this century (IPCC, 2012).

The impact of extreme events on society is not dependent only on the frequency and magnitude of extreme weather events but also on the vulnerabiility of society to extreme events and also the exposure of the society. Vulnerability and exposure depend on the country’s level infrastructure including the capacity to operate efficiient early warning systems. For example, countries having habitation in areas exposed frequently to the severe weather events suffer more of the negative impacts of climate change. The increasing population may also lead to non-resilient development of social structure of a society (IPCC, 2012).

Climate change requires adaptation to the future conditions. This need to adaptation has been recognized almost with one accord by the various stakeholders of the globe. The adaptation to climate change is not only adaptation to something that will realize after decades. Climate change is already existing phenomena influencing the conditions. As well, being prepared to future extreme conditions contributes to the preparedness of current extreme weather events. These strategies offer economic benefits almost immediately and same time reduces vulnerability of society on long run (IPCC, 2012).

Efficient adaptation to climate change requires wide and multidisciplinary information; starting from the basic meteorological observations ending to very sophisticated socioeconomic analyses. The National Meteorological and Hydrological Institutes (NMHI) have an important role as the provider of the climate and hydrological observation data series and climate statistics and analyses. Typically NMHIs also are receiving and processing remote sensed environmental earth observations, i.e., satellite measurements. As well, NMHIs possess typically a comprehensive knowledge on the modeling of future climate. NMHIs role is also becoming wider as the role of climate change communication is growing larger and larger.

When the future climate conditions are been estimated the most important scientifically based tool for that purpose are the climate models (Fig. 1). These very sophisticated models rely on the basic laws of physics. Present models have modules for atmospheric, ocean and earth surface processes. The models include the interaction between the different modules like the exchange of heat between surface and atmosphere, evaporation, surface friction, flow of water from the continental areas to the ocean etc. The model calculations are done in a grid covering the whole globe at several levels from the deep ocean up to top of the atmosphere. Running the models require very large computer resources typiically called as super computers. Global climate model simulations are done in several research center having needed computer and human resources. These centers have agreed on sharing the climate projections based on their model enabling also countries with no needed modeling resources to get the needed climate information (Meehl et al., 2000). The Coupled Model Intercomparison Project (CMIP) was established under the World Climate Research Programme (WCRP), the Working Group on 4

Coupled Modelling (WGCM) as a standard experimental protocol for studying the output of coupled atmosphere-ocean general circulation models (AOGCMs). CMIP provides infrastructure in support of climate model diagnosis, validation, intercomparison, documentation and data access. The Program for Climate Model Diagnosis and Intercomparison (PCMDI) has archived much of the CMIP data and provides other support for CMIP. Phase three of CMIP (CMIP3) included "realistic" scenariios for both past and present climate forcing. The research based on this dataset provided much of the material used in the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) and dataset provided basis for numerous scientific research articles and projects. The same way Phase five of the CMIP (CMIP5) is the key source of climate model simulations used in the Fifth Assessment Report (AR5). Altogether more than 20 Modelling Centers are making climate simulations and providing data for the use of climate research community (see CLIVAR Exchanges Newsletter, 2011).

There are some differences when CMIP5 archived simulations are compared wiith the previous CMIP3 simulations. CMIP5 models can be regarded more comprehensive, there is a broader set of experiments and wider variety of scientific questions, the models have higher spatial resolution as 50% models have latitudinal resolution finer that 1.3 ° (only one CMIP3 model is as detailed), there is a richer set of output fields and the documentation is more detailed (Taylor et al., 2012).

Construction of climate change projections; Required steps to estimate future climate External factors affecting climate (e.g., greenhouse gas concentrations) as a function of time

Climate model: laws of nature as a computer program, as well as the Atmos. Land Ocean Ice current knowledge allows

“Long weather forecast” Temperature

Time Climate = statistical

Temperature properties of weather

Time

Figure 1. Construction of Climate Change Projections

Climate projections stored in the CMIP database are simulations made using models having global coverage (GCM). Though the spatial resolution of these models has improved it is still relatively coarse for the small scale studies. The spatial variation of e.g. terrain elevation has large influence on the spatial variation of temperature and precipitation. To be able to assess the local features of future climate the prediction made to the GCM grid must first be downscaled to a denser grid or event to point values (Fig. 2). This downscaling can be done by applying either dynamical or statistical methods. In case of dynamical downscaling a Regional Climate Model (RCM) with dense grid is run for the study area. RCM needs as input the large scale features from the global model. In case of statistical downscaling first the relationship between large scale features of climate and local observations is defined and this statistically defined dependence is applied for the projected future climate available from the global models (e.g. Benestad et al., 2008).

Dynamical downscaling is a physically justified approach and e.g. includes many of the feedbacks of the climate system. However, this method is very dependent on the boundary conditions defined by the GCM and e.g. the magnitude of the change is roughly the same as in the GCM. The output is also sensitive to parameterizations used in the RCM and method requires a lot of computing resources and also human know how. This method can seldom be applied to assess the climate change as predicted by a large number of GCMs and/or various emission scenarios. Statistical downscaling requires less computational effort than dynamical downscaling and it offers the opportunity for testing scenarios for many decades or even centuries. It is possible to use large number Global model results and different emission scenarios. The drawback is that the method requires long time series of observations needed. One can also question can present day climate depict future conditions?

The principle of downscaling with a statistical model

-15 -20 Climate of the -10 a) search for statistical global model relationships between the observed climate and broad-scale statistical relationships -> circulation features -5 -10 Observations -15 b) b) Use the broad-scale -15 of climate -10 circulation features as -10 projected in GCMs to -10 -5 develop projections of

local climate -5 -15 -15 -15 Climate of the -10 statistical model -5 -10 -10 -5

Figure 2: The principle of Downscaling with a Statistical Model. Source: Finnish Meteorological Institute.

Processing and analyzing the large meteorological and hydrological dataset require advanced analyzing tools. More and more the R environment (http://www.r-project.org/) has been applied in this kind of applications1.

In the project ADB TA-7532 “Water and Adaptation Interventions in Central and West Asia” detailed climate projections were needed for the future decades. Projections of future climate were based on GCM simulations reported in the 4th assessment report of the IPCC (Randall et al. 2007). Model simulations of five different models (Table 1) in the intermediate emission scenario were used. The criteria for the selection of these models was that their spatial resolution was 1.9° or higher and the models originated from different countries. This ensured that the models are genuinely separate models and this way the results obtained by employing these models depicted the scale of variation different climate projections possess. As well, in a comparison made among 19 different GCMs the ability of these four models to simulate the past climate was as good as the accuracy of any other model.

Table 1. Global climate models used in the study. Model Institute Country Resolution CGCM3(T63) Canadian Centre for Climate Modeling and Canada 1.9˚ x 1.9˚ Analysis CNRM-CM3 Météo-France France 1.9˚ x 1.9˚ ECHAM5/MPI-OM Max Planck Institute for Meteorology Germany 1.5˚ x 1.5˚

MIROC3.2(HIRES) Centre for Climate System Research Japan 1.1˚ x 1.1˚ (University of Tokyo) NCAR-CCSM3 National Center for Atmospheric Research USA 1.4˚ x 1.4˚

Source: Randall et al. (2007)

The climate scenarios produced by the selected five GCMs were generated to daily temperature and precipitation data sets for the period 2011-2050 by a method known as the delta change method (e.g., Arnell, 1996). The use of delta change method for the estimation of the change in near future is well justified (Räisänen and Räty 2012). First the differences between the simulated current and future climates were computed and then these changes were added to the high resolution (0.2° x 0.2°) daily temperature and precipitation data sets for the period 2001-2010. The gridded daily mean temperature data to be used as a base of generating scenarios for the future, were produced by kriging interpolation (e.g. Krige, D. G., 1951) from daily temperature observations during the period 2001-2010. The gridded daily precipitation data for the period 2001-2010 originates from satellite based data of the Tropical Rainfall Measuring Mission (TRMM; Huffman et al. 2007; Huffman et al. 2012) and Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN; Sorooshian et al. 2000). The processing of climate data; interpolation and downscaling in the project was done using the R software environment.

During the next 40 years, the mean temperatures were projected to rise in the Central Asian region everywhere around the year with an annual mean temperature rise of about three

1 R is a free software environment for statistical computing and graphics. The use of R has expanded recently and is has become almost a standard in statistical computing and data processing. R is available as Free Software under the terms of the Free Software Foundation's GNU General Public License in source code form. This means that there are no installation or maintenance costs related to R. 7

degrees. The warming was projected to be strongest in the mountains and in the northern parts of the area (Fig. 3).

Figure 3. Average change of annual mean temperature vs. simulations for 2045-2065.

Average change of annual mean temperature (°C) between the control simulations (simulation period: 1971-2000) and the simulations into the future (five simulation period: 2045-2065).

3.4 5 3.4

3.2 Kazakhstan 4

45 3

3 Uzbekistan

Kyrgyzstan °C

TITUDES 2.8 2

LA 40 3.2 Tajikistan Turkmenistan 1 3

3.4 3.4 35 3.2 0

60 65 70 75

LONGITUDES

Figure 4. Average change of annual precipitation vs. simulations for 2045-2065. Average change of annual precipitation (%) between the control simulations (simulation period: 1971-2000) and the simulations into the future (five simulation period: 2045-2065).

Kazakhstan 10 45 10 5 Uzbekistan

0 %

TITUDES Kyrgyzstan LA

5 −10 40 0 Tajikistan Turkmenistan 5

−5

0 35

60 65 70 75 LONGITUDES

The projected changes in annual precipitation were relatively small during the coming 40 years and varied from model to model. The already dry south-western parts were projected to become even drier especially during summer time. In contrast to this, in the northern parts and in the mountains the annual totall precipitation was projected to increase by 5-10% (Fig. 4).

The predicted climate changes in Central Asia can lead to more favorable conditions in the parts of the area in Kazakhstan (Lioubimtseva and Hennebry, 2012). The growing season is projected to become longer and availability of water may improve. However, in the regions suffering already now on water shortage, the conditions will become even more difficult as temperature and evaporation will raise and precipitation decrease. Parts of the region are predicted to become more and more arid like the western parts of Turkmenistan, Uzbekistan and Kazakhstan. For example in Uzbekistan, the economy is very dependent on irrigated agriculture which is using substantial amount of water resources of Amu Darya and thus the foreseen decrease of these resources can have severe impact on the economy of the country (Schlüter et al., 2010).

In Kazakhstan, the drought of 1997-1998 destroyed nearly half of the harvest in most grain- producing provinces and led to a deterioration of the financial security of farms. In Tajikistan, drought causes the largest amount of economic damage, estimated on the average of 1.7 million US dollars during a 10 year period. The country faced severe droughts in 2000-2001 and 3 million people were at risk of famine. Hot and dry weather prevailed and the country lost a considerable part of its cereal crop, with the livestock sector being severely affected. In the same years also Uzbekistan and Turkmenistan faced serious problems in food production. In February 2005, Tajikistan was hit by heavy snowfall in the Rasht Valley where two meters of 11

snow had fallen in two days (ADRC 2006). Exceptional heat wave events and torrential rains have been dangerous in mountainous areas where sudden snow melt has generated destructive avalanches, floods and mudflows. Such calamities require several casualties every year and economic losses are substantial.

Understanding of climate change and its impacts is essentially important in Central Asia where climate risks are high. The adaptation to the predicted large risks requires coordination among the Central Asian countries, like in case of Aral Sea water system water management involves Kyrgyz Republic, Tajikistan, Uzbekistan, Kazakhstan and Turkmenistan (Ibatullin, et al. 2009).

IMPACTS OF CLIMATE CHANGE

Scientists and policy makers allied in the United Nations Intergovernmental Panel on Climate Change (IPCC) state that “it is very likely that most of the observed increase in global average temperatures is due to increase in anthropogenic greenhouse gasses concentrations”. The models applied by the IPCC members strongly indicate that those changes will intensify over the coming century. Moreover these models provide abundant evidence that freshwater resources are vulnerable and have the potential to be strongly impacted by climate change, with wide-ranging consequences for human societies and ecosystems. Observed warming over the last decades has altered the hydrological cycle already. Typical examples include changing precipitation patterns, intensity and extremes, changes in snow and ice cover and changes in runoff.

The science community is starting to understand better how the climate system works and hydrologist are able to evaluate the impact of changes in climate on the water resources. Obviously not only the change in climate is affecting our water resources, but climate change is expected to exacerbate current stresses on water resources from population growth and economic and land-use change. For Central Asia, mountain snow pack, glaciers and small ice caps play a crucial role in freshwater availability. Retreating glaciers and reductions in snow cover as observed over recent decades are projected to accelerate throughout the 21st century. Consequences will be a reduction in overall water availability, lower hydropower potential, and changing seasonality of flows in regions supplied by melt water from Tien Shan and Pamir mountains.

Besides the projected changes in precipitation, the increase in temperature is an important factor for Central Asia. The large scale irrigation systems are nowadays already suffering from water shortage, and higher temperatures will increase the water required by the irrigated crops. Moreover higher temperatures will also have an impact on the natural vegetation and evaporation from these areas will increase so that less water becomes available to flow into the streams and rivers.

Glaciers

Glaciers cover 18,128 km2 from the Aral Sea Basin and they have an important role in hydrology as they release melt water especially during the dry summer months. Mountain glaciers include: 1) Small cirque glaciers resting on rather steep mountain slopes; 2) Large ice caps covering mountain tops associated with valley glaciers, narrow and long ice tongues flowing down in U- shaped valleys.

The percentage of glaciated area of the two catchments differs significantly. In the Amu Darya Basin glaciers cover 15,500 km2 (2% from the area) and in the Syr Darya Basin 1,800 km2 12

(0.15% from the area). The biggest glaciers are located in the Pamir Mountains in Tajikistan (Fig. 5).

Figure 5. Occurrence of glaciers in the upper watersheds of the Aral Sea Basin in 2010 and 2050.

It is a well known fact that after the ice-age (Holocene) the glaciers reached their maximum latest in 1850s. In most glaciers the terminal moraine representing the maximum extent can clearly be seen in the valleys especially in satellite images. Glaciers respond relatively quickly to changes in climate (temperature, precipitation, humidity and cloudiness). If the snowfall (accumulation) sustaining a glacier declines, or if there is an increase in ice loss (ablation), these will result in recession of the glacier or an overall thinning of the ice mass (or both). Melting of glaciers has accelerated since the Little Ice Age (c. 1650 – 1850) due to the gradual climate warming (Oerlemans 2005; Seversky 2006; Kutuzov & Shahgedanova 2009; Kelly 2013).

All the glaciers have their own dynamics and it is common that sometimes their margins are retreating and some other times advancing. Some glaciers have periods of very rapid advancement called surges. These glaciers exhibit normal movement until suddenly they accelerate, then return to their previous state. During these surges, the glacier may reach velocities far greater than normal speed and the margin may advance. For example, the Medvejiy glacier in Tajikistan moves periodically downhill every 12-15 years.

As temperatures have risen almost everywhere, the retreat of glaciers from mountain valleys is one of the most visible symbols of global warming. The overall picture of widespread recession is unequivocal and reflects the well-known record of global warming. The situation in Asia is 14

more complex. Although there seems to be a general loss of ice in this region, some regions with higher altitudes such as the Karakorum show an increase, and uncertainties about rates of change are considerable (Immerzeel et al. 2010). Some glaciers located in higher altitudes may grow as they are well above the ablation (snowline) altitude. Increase of precipitation may result in growth of glaciers. However, the general trend in Asia is that the glaciers are melting in an accelerating mode.

Vanishing Glaciers

The Fedchenko glacier is a large glacier covering over 649 km2 in the Pamiir mountains in central Tajikistan at altitudes of 2900 - 6300 m (Fig. 6). The associated valley glacier is 77 km long and narrow ice tongue. It is the biggest glacier in the world outside of the polar regions. The maximum thickness of ice is 1000 m, and the volume of the glacier including its tributaries is estimated at 144 km3. The edge of this glacier has retreated 1 km in 70 years. During the recent decades, the valley glacier has thinned 1 m/yr while its surface area decreased by 11 2 3 km and it lost about 2 km of ice (Aizen & Aizen 2010). As a consequence of climate warming, more than 40 million m3 of melt water has annually been fed to the Naryn River from the contraction of Fedchenko glacier only. If we consider all the glaciers of the basin, the total volume of such non-renewable water resources is immense. 3 th The glaciers of Tajikistan have lost more than 20 km of ice in the 20 century. During the last decades, the glaciers in the mountains of south-east Kazakhstan was annually reduced by 0.85% as regards glacier area, and 1.0% in ice volume (Alamanov et al. 2006). The glaciers in the Tien Shan and Pamir are retreating and the rates of retreat vary between regions and time periods. The largest retreat rates have been observed in the northern Tien Shan where glaciated area has declined by 30-40% during the second half of the 20th century. Acceleration of glacier retreat has been noted in the eastern Pamir from 7.8% over 1978–1990 to 11.6% over 1990–2001 periods. Glaciers have lost 12.6% (0.33% /yr) of their 1965 area in the 1965-2003 period. Small glaciers have diminished more than the average (Kutuzov & Shahgedanova 2009). It has been estimated that glaciers in the Syr Darya basin have lost 14% of their total volume over the last 60 years and that 15-40% of the volume will be lost in the coming 40 years (Siegfried et al., 2010). The glacial area in the Amudarya basin has shrunk 13.1% from 1957 to 1980, i.e., from 7144 to 6205 km² (Agaltseva 2005).

Figure 6. The Fedchenko glacier in the Pamir mountains of central Tajikistan is the largest mountain glacier in the World. The width of the glacier is c. 2 km. (3D visualization, Google Earth).

In this study, the future melting rate of glaciers in the Aral Sea Basin was modeled more precisely (see Box: Modeliing Glacier Processes). The modeling work revealed what will happen to the glaciers and runoff in the coming 40 years (Fig. 7). The present records show that small ice caps especially from Tien Shan will disappear and also the glacial melt water will dramatically be reduced (Fig. 8). It is important to realize that climate warming and melting of glaciers have already been going on especially after 1940. It is evident that there is an accelerating trend in the melting of glaciers.

Figure 7. Change in glacier extent in Central Asia according to various GCM projections.

Usually aerosols, small particles or droplets suspended in the atmosphere, have cooling effect on climate. Such particles originate from regional biomass burning, industrial pollution and dust storms caused by desertification. However, whenever the black carbon or dust will accumulate onto the glacier surfaces, melting will be accentuated. Thick gravel (debris) beds on top of a glacier will slow down its melting. To protect glaciers, it is important that Central Asian countries will control their own pollution emissions, desertification and wildfires.

Figure 8. Geomorphological evidence, old terminal moraines representing earlier and more extensive glaciers, show that glaciers have already receded substantially during the last 150 years. In this case only 1/3 of the maximum extent is left (Kazakh-Kyrgyz Republic border, Almaty; 3D visualization, Google Earth).

Modeling Glacier Processes

The modeling of processes involving glaciers is based on the so-called Degree Day Factor. The use of temperature index or degree day models is widespread in cryospheric models to estimate ice and snow melt. In these models an empirical relationship between melt and air temperature based on a frequently observed correlation between these two quantities.

“Melt from clean ice glaciers” is defined as the air temperature (if above 0 °C) multiplied by the 17

degree day factor for clean ice, multiplied by the clean ice fraction of the glacier cover and the cell fraction with glacier cover. For the melt from debris covered glaciers the calculation is similar, although a different degree day factor for debris covered glaciers is specified. Melt rates for debris covered glaciers are lower, since incoming radiation and other heat flows are blocked by the (thick) debris cover.

The total glacier melt is then calculated by summing the two components from clean ice glacier melt and debris covered glacier melt. A part of glacial melt also refreezes in the glacier when it percolates the ice.

For each cell, the model determines if precipitation fallls as snow or rain by comparing the actual air temperature to a critical temperature. When air temperature is below or equal to the critical temperature, precipitation will fall as snow. When air temperature is above the critical temperature, precipitation will fall as rain.

The potential snow melt is defined as the air temperature (if above 0 °C) multiplied by a degree day factor for snow multiplied by the cell fraction covered with snow. The actual snow melt however, is limited by the thickness of the snow pack. No more snow can be melted than the amount of snow which is available at the considered time step. The snow storage is then updated, to be used for the next time step. The updated snow storage is the ‘old’ snow storage with the fresh snow added and the actual snow melt subtracted.

The water resulting from snow melt will partially refreeze as it infiltrates the underlying snow pack. The maximum of water that can refreeze is defined by the water storage capacity of the snow pack which depends on the thickness of the snow pack present and the storage capacity of snow (e.g. the total millimeters of melt water that can refreeze per millimeter of snow). The actual amount of water that is stored in the snow pack is defined as the water stored in the snow pack during the previous time step summed by the actual snow melt. Snow melt will become actual snow melt when the amount of snow melt exceeds the water storage capacity of the snow pack.

Glacial Lakes

Because of the rapid meltiing and retreat of the margin of glaciers, new proglacial lakes may be generated in places. It is common that older ice-cored moraines create dams in front of fast receding valley glaciers (Fig. 9). Also glacier itself may act as a dam for waters of an adjacent valley. Ice dam may be formed as a result of fast advancing, surging glacier. Such lakes may have extensive amount of water and whenever the dam collapses, a catastrophic flood may occur.

There are hundreds of proglacial lakes in Central Asia and many of them have been classified to be dangerous. The recent ice surges, outbursts of glacier-dammed lakes and floods of glacier rivers have caused major disasters. In the Kyrgyz Republic, there are several lakes with unstable natural dams and there is a permanent threat of outburst. Of more than 1,000 high mountain lakes, 199 have been identified as being dangerous. Since 1952 to 2007 about 70 cases of dangerous outbursts with human victims occurred on the territory of the Kyrgyz Republic. In Tajikistan the situation is even worse (ADRC 2006).

In 1963 and 1973, the surges of the 15 km long Medvezhi glacier in the Pamir mountains, Tajikistan, have caused lake formation, outburst and subsequent floods into the Vanch River. The 1-2 km long glacier advance created 100 m high ice dam which dammed a lake of over 20 18

million m3 of water and debris. The outbursts of that lake have generated a series of large flood waves. Due to early warning and monitoring, there were no victims, although infrastructural damage was significant (Novikov 2002). Recently 2011, the surge took place again.

Increasing glacial lake outbursts can be related to climate warming as receding glaciers generate proglacial lakes. Thawing of an ice-core from terminal moraines damming a lake may breach causing sudden mudflow down to the valley. Central Asian countries have governmental institutions mapping and managing emergency situations. Flood protection interventions and early warning systems are needed in areas identified to be vulnerable to glacier instability and lake outbursts.

Figure 9. When an ice-core of a terminal moraine damming a proglacial lake will melt, a catastrophic flood and mudflow may occur. The lake is 2 km in diameter. (Petrov Lake, Kyrgyz Republic. 3D visualization Google Earth).

Water Resources

The main rivers in Central Asia, Syr Darya and Amu Darya, play a prominent role in the region. People depend on the water for their domestic use, farmers cannot exist without irrigation, the environment alters if water resources are changing, and hydropower supplies the necessary energy in the region. Many initiatives have been started to ensure a proper and sustainable use of the water resources in the region, where difficult decisions regarding sharing benefits have to be faced. Besides ongoing economic development with associated changes in water requirements, it is clear that climate change puts an additional challenge to appropriate planning and managing of the water source.

Currently most water in the region is generated in the upstream mountainous areas which will flow from smaller streams into bigger streams and finally into the two main rivers: Syr Darya and Amu Darya. The origin of this water in the rivers exits of four components:  Rainfall can flow overland directly into the streams;  Rainfall can infiltrate into the soil and will flow through the soil into the streams, or becomes available as groundwater resource;  Precipitation can fall as snow and will flow into the streams when it melts;  Precipitation can feed glaciers and will after longer times melt and flow into the streams.

Knowledge about those four different components is very relevant for understanding current and future water resources. Rainfall flowing directly into a stream can result in rivers filling up quickly and can cause flooding. While rainfall that enters into the soils is buffered and will generate a much more constant inflow into the streams (this is why fighting land degradation is so important). Precipitation falling as snow can be seen as a very useful natural buffer, since only when temperatures are increasing during spring and summer this snow will melt and will become available as liquid water for downstream users; just at the moment when water is mostly needed. Finally water stored in glaciers is released very slowly and gradually providing a stable flow during spring and summers even if rains have been low.

In Figure 10 the relative contribution from glacier melt compared to total flow is presented. It is clear that in the Amu Darya glacier melt is an important contributor to the entire flow, especially in the smaller streams at higher elevations. Total water resources generated in the upstream parts of the Aral Sea Basin, as shown in Table 2, indicate that for the upstream Amu Darya almost 40% of the total flow is generated by glacier melt, while for the upstream Syr Darya this figure is just above 10%. It is clear that by receding glaciers caused by climate change this will have a major impact on total flow as well as timing of flows.

Figure 10. The relative contribution of glacier melt to the total flows in streams and rivers. The upper reaches of the Amu Darya river are mostly fed by meltwaters from the glaciers of the Pamir Mountains.

Table 2. The relative flow contribution of the four components for the upstream Syr Darya and upstream Amu Darya over the period 2001-2010.

Syr Darya Amu Darya Direct runoff 31% 16% Base flow 23% 19% Snow melt 35% 27% Glacier melt 11% 38%

Future Water Availability in Syr Darya and Amu Darya

To project future water resources in order to support appropriate planning, scientists have developed advanced so-called assessment tools. These assessment tools have been developed to mimic the world we are living in and can also be used to explore our future world if the climate changes. Assessment tools have been used over various scale levels from the entire world down to individual farmer plots. Using these impact assessment tools the future of the water resources in Central Asia has been evaluated.

The impact assessment was based on two modeling approaches. Current and future water resources were assessed using the SPHY model, which is a revised high-resolution version of the PCR-GLOBWB hydrological model. SPHY is a conceptual, dynamic and distributed model. SPHY runs on a daily basis on a 6 arc-minute grid and was setup for the period 2000-2050. Changes in irrigation water requirements are based on: (i) changes in irrigated area and (ii) changes in crop water demand as a consequence of a changing climate. Projected changes in domestic and industrial demands are based on the relationships between Gross Domestic Product (GDP) and Gross Domestic Product per Person (GDPP). A water allocation model was used to link water supply and water demand and to explore adaptation strategies. This model, referred to as ARAL-WEAP, was developed using the WEAP package which considers the following features: streams, reservoirs, groundwater, irrigation demands, domestic demands and industrial demands. ARAL-WEAP was run on a monthly base for the period 2000-2050. The cost-effectiveness of various adaptation measures to close the supply-demand gap was assessed by means of the “water-marginal cost curve”. Such cost curves show the cost and water saving potential of a range of different strategies - spanning productivity improvements, demand reduction and supply expansion – to close the water supply-demand gap.

The relative contribution of these four different sources of water in the rivers is most likely going to change in the future. Rainfall patterns will be shifting and temperatures will increase resulting in a lower contribution of snow melt and especially of glacier melt. Figure 11 indicates that total annual runoff and changes in flow contribution can be substantially. Though the figure indicates that glacier melt will be reduced substantially, still over 50% of the glaciers will remain but at higher altitudes so that melt will be lower. Recent scientific developments indicate that some specific glaciers will melt faster in the short run so an increase in inflow might be expected. The current analysis show that inflow into the downstream areas will decrease by 22-28% for the Syr Darya and 26-35% for the Amu Darya by the year 2050. This range in projected decreases reflects the uncertainty in the climate projections. The relatively small range in these projections 21

is caused by the fact that only one climate scenario (A1B) was used; the uncertainty reflects therefore only the limitation within our current scientific knowledge as reflected in the range of GCM used. Strongest decreases in stream flow are expected for the late summer months (August, September, October), where inflow into downstream areas decreases around 45% for both river basins.

The major user of fresh water in the region is irrigated agriculture. About 97% (93,800 Mm3 per year) is consumed by irrigated agriculture, while other sectors such as domestic and industry consume about 2,700 Mm3 per year. Climate change will not only reduce available water resources as explained in the previous sections, it will also increase the demand by crops as higher temperatures will result in elevated evaporation rates. Based on the assessment tools as developed for the regions it is projected that total water demand in the Syr Darya basin increases by about 3 to 4% in 2050. Annual water demand in the Amu Darya basin increases by about 4 to 5% (Fig. 12).

The combined effect of higher demand and lower inflow will amplify the current water shortage in the two River Basins. For the Syr Darya the assessments show that the total water shortage will increase to a level of 13,700 Mm3 per year in 2050; this is about 35% of total demand. For the Amu Darya annual unmet demand increases to 29,400 Mm3 per year in 2050 (about 50% of total demand). It is clear that such a reduction will put substantial challenges for the region to cope with (Fig. 13).

Figure 11. Typical examples of changes in total flow and flow composition of two main reservoirs in the Syr Darya (top) and Amu Darya (bottom). Note that after 2050 about 50 % of 22

the glacier extent has been lost and remaining glaciers are above the 0oC level and they will hardly contribute to streamflow anymore.

Figure 12. Changes in annual water demand and unmet demand (water shortage) under climate change for the two main basins in Central Asia.

Figure 13. Satellite image from the Isfara River, Fergana Valley (Tajik-Uzbek border). During the irrigation season, the rivers flowing down from the mountains receive major part of their water from melting glaciers. All the river water is used for irriigation (3D visualization, Google Earth).

Floods

Floods have caused major disasters in Central Asia destroying roads, bridges, villages and there have also been big number of human casualties. An important question is, if in the future floods will become less frequent as the overall volume of water in the rivers will decrease. The modeling results show that in most rivers Spring time floods will remain the same as in the recent past. However, there will be major differences in flooding between the river basins and these should be understood whenever flood management is developed.

In basins which have extensive middle-altitude uplands and lots of snow, have biggest flood risk in May-June. This peak is the highest in the annual hydrographs and the situation will remain the same also in the future. A typical example is the Naryn River Basin (the upper Syr Darya Basin above the Toktogul reservoir) (Fig. 14a). A torrential rain or a sudden warm period in Spring or Summer time may cause major flooding disasters. This was the case e.g. 2005 in the Zarafshan and Pyanj rivers, Tajikistan, where more than 11,000 people had to be evacuated from flooded areas. 24

In the areas of high-altitude uplands, high mountains and major glaciers, the Spring time flooding is not so common. Glacial melt is a slow process and it cannot generate floods. Highest discharges are measured in the late Summer and melting of glaciers is a main contributor of water. In such basins flood risk will decrease as the volume of glaciers will continue to diminish. There is no any reason to develop flood protection in these basins. A typical example is the Vakhsh River basin (the upper Amu Darya Basin above the Nurek reservoir) (Fig. 14b).

In the areas of lower mountains and plains (in central parts of the Aral Sera Basin) thin snow and small cirque glaciers (ice caps) are typical. Here snow accumulation will decrease and the small glaciers will disappear during the next few decades. In the future, water shortage will be severe and flooding will not occur at all. The Dushanbe, Samarkand, Tashkent, Chimkent, Bishkek and Almaty regions can be mentioned as example areas.

Figure 13 a and b. Annual hydrographs for the Nurek reservoir (Tajikistan, Amu Darya Basin) and Toktogul reservoir (Kyrgyz Republic, Syr Darya Basin) in the past (field observations) and in the future modeled using five different climate change projections. In January-May the discharges (Q) will remain similar as today, but major reduction will take place in June-November discharges because of diminishing glaciers.

Climate hazards in the Kyrgyz Republic

The regions most frequently affected by climate related natural disasters are the Jalal-Abad and Osh oblasts, followed by the Chui and Issyk-Kul oblasts. Approximately 80% of all the disasters occur between May and August. The most destructiive natural disasters of the past 10 years include:

• Torrential rain and an earthquake in Osh and Jalal-Abad in 1992 destroyed 51,440 hectares of agricultural land and affected 20,000 people; direct economic damage was estimated at US$ 31 million; • Heavy rainfalls, snowfalls and frosts in spring 1993 caused economic losses estimated at US$ 21 million; • Large-scale landslides and mudflows in 1994 in the Osh and Jalal-Abad oblasts killed 115 people and made 27,000 homeless; economic damage: US$ 36 million; • A glacial lake outburst flood in 1998 killed over 100 people and caused damage over an areas stretching to Uzbekistan; • Severe and widespread floods in Jalal Abad in 1998, caused by torrential rains, damaged or destroyed an estimated 1,200 houses and public buildings; direct economic damage was estimated at US$ 240 million. (UN 2000)

Permafrost and slope instability

As a result of climate warming, the snowline will raise 200 - 300 m in average until 2050. Extensive parts of hill slopes which have always been covered by snow or have had frozen ground (permafrost) will melt. Thawing of permafrost in the higher mountains will make the slopes unstable and this will generate landslides and mudflows. At first, only the surface layers of the ground will thaw and the underlying icy ground will form a slide surface. The thawed surface layers may slide down fast to the valley and destroy forest, infrastructure and settlements on the way (Fig. 15).

Landslides and mudflows have been common in Central Asia and every year settlements, infrastructure, agricultural lands and natural areas are destroyed. Also human casualties are common in major disasters. Mass movements are muddying water resources and filling reservoirs with sediments. In the Kyrgyz Republic, more than 200 settlements and communication structures are located in landslide-prone zones. About 2,500 landslides have been registered in the south since the mid-1950s (UN 2000). In average, landslides cause 46 deaths in 10 years. In Tajikistan, in worst years more than one thousand houses have been destroyed in mudflows. Some 85% of Tajikistan’s area is threatened by mudflows (mountainous areas, hill slopes and river valleys) and 32% of the area is situated in the high mudflow risk zone (ADRC 2006).

In permafrost areas the depth of frozen ground can be several hundred meters. Consequently permafrost prevents groundwater flow as it is frozen. Melting of permafrost again will result in increased groundwater flow which will end up to lakes and rivers. It is difficult to estimate the volume of water released from the thawing permafrost, but certainly it is significant.

It is possible to map the future areas subject to landslides and mudflows by using similar SPHY hydrologic model that was used here to estimate river discharges. The model is able to define altitude zone where thawing of permafrost will take place now and in the future. Such risk assessments could be important whenever infrastructures and settlements are planned and emergency preparedness developed.

Figure 15. Thawing of permafrost in the higher mountains will make the slopes unstable and this will generate landslides and mudflows (Mailuu Suu, Kyrgyz Republic. M. Punkari).

(maybe this does not need caption – figure is related to permafrost chapter – shows how the rising temperature will thaw permafrost in wide areas)

Drying environment

The lower parts of the Aral Sea Basin is mostly arid area where precipitation is 40-200 mm/yr. Desertification is one of the most severe problems in Central Asia where huge areas of fertile land is lost every year. Land degradation from overgrazing, soil erosion, salt damage to irrigated land, and desertification is directly affecting the livelihood of nearly 20 million rural inhabitants.

Farm yields are reported to have dropped 20 - 30% across the Central Asian region since the 1990s. About 70% of the total area of Turkmenistan has become desert, while salinized irrigated areas account for 50% in Uzbekistan and 37% in Turkmenistan (CASILM). In the future, the continuous increase of temperature and evaporation, slight decrease of precipitation in the most arid lowlands and the radical reduction of river discharges will worsen the situation.

It is likely that wildfires in the plains will increase due to climate change and this again will result in soil degradation, erosion and desertification (U.S. Forest Service 2013). Human activities as overgrazing and fuel wood collection will contribute desertification especially in populated areas.

The Aral Sea was once the fourth largest lake in the world, but is now nearing extinction. Long- lasting unsustainable water management has caused the Aral Sea to shrink, which will be made worse by climate change. It has decreased over the last 50 years from 68,000 to about 12,000 km2. Where once 178 species inhabited the Aral region, there are now fewer than 40 (Alamanov et al. 2006). Salt air pollutiion from the open sea bottom is dangerous for agriculture and human and animal health. Warming temperatures are only making it worse - for example, by increasing evaporation over the widespread irrigated fields and their canals, including the 1300 km man‐ made Karakum canal.

Environmental changes may have several feed-back effects on climate the final results of which are not yet fully understood. For example, increasing dust storms and wildfires increase aerosols in the atmosphere which may cool climate. However, these particles deposited onto snow and glaciers will decrease their albedo and increase melting. Especially the Tajik environmental experts have raised questions about the protection of glaciers. One of the very few feasible options is to minimize aerosol emissions by controlling deforestation, erosion and wildfires as well as by reducing air pollution.

Climate Change in Development Programs

It is clear that the need to adapt to climate change in Central Asia is felt by everybody. International efforts to limit greenhouse gas emissions will not be sufficient and fast enough to prevent the harmful effects of changes in precipitation, increase in temperatures and increased frequency and severity of extreme weather events. Climate change can also create opportunities, particularly in the agricultural sector. Increased temperatures can lengthen growing seasons, and higher carbon dioxide concentrations can enhance plant growth. However, these positive opportunities will not be sufficient to compensate for the negative effects of climate change as a whole.

The risks of climate change cannot be effectively dealt with, and the opportunities cannot be effectively exploited, without a clear plan for aligning policies with climate change. Developing 28

such planning involves a combination of high-quality quantitative analysis and consultation of key stakeholders. It has been well accepted that the most effective plans for adapting to climate change will involve both human capital and physical capital enhancements. Moreover, it is well- accepted that the capacity to adapt to changes in climate is in part dependent on financial resources – in emerging economies, among small-holder farmers with limited financial resources, adaptive capacity is particularly low. As a result, the donor community will continue to be a key stakeholder in developing climate change policies and implementation measures (Ibatullin et al. 2009).

Adaptation to climate change

There is no silver-bullet approach that can be used as the ultimate adaptation strategy. Two different types of actions are essential to tackle the climate change challenge. First of all, “development as usual” without consideration of climate risks and opportunities, will not allow us to face these challenges. Although a range of development activities contribute to reducing vulnerability to many climate change impacts, in some cases, development initiatives may increase vulnerability to climatic changes. This integrating of climate change in existing development planning is sometimes referred to as “mainstreaming” or using a “climate lens” in existing development planning.

Secondly, separate adaptation planning and implementation is required to overcome the negative impacts of climate change. It has been advocated that this adaptation planning and implementation has various dimensions. An important dimension is that some adaptation will take place autonomous and other adaptation requires actual planning. A typical example of autonomous adaptation is farmers changing the planting date of their crops as response to temperature shifts. An example of actual planning is that irrigation water should be delivered earlier and proper irrigation water requirement monitoring systems should be in place.

A second dimension of adaptation is the timing of response, being the short run or the long run. The long run adaptation includes issues like building capacity, changing institutions, and large infrastructural development, amongst others. Typical examples relevant to short run adaptation in Central Asia are related to water allocation and reservoir operations. Finally the third dimension to consider in adaptation is the scale-level to consider. In general one should consider the following scales: farm, community, national, and regional. Each of these scales has their specific needs and opportunities.

Concrete adaptation options in Central Asia

It is clear that adaptation strategies are very scale dependent and local specific. The assessment tools as presented in the previous sections have been used to analyze a range of adaptation actions that might be suitable for the Central Asia region. Based on such a broad range of adaptation options, actions can be subsequently fine-tuned to national and local conditions and preference. The options explored can be summarized into three broad categories: (i) expanding supply of future water availability, (ii) increasing productivity of water, and (iii) reducing future demand. Within each of these three categories typical options can be chosen such as: increased reservoir capacity, improved agricultural practice, increased reuse of water in irrigated agriculture, increased reuse of water for domestic use, reduction of irrigated areas, reduction of domestic demand, and deficit irrigation.

The ARAL-WEAP models for the Amu Darya and Syr Darya basins were used to evaluate the impact and effectiveness of these adaptation measures. The WEAP-model was run for five 29

different climate projections, based on five different Global Circulation Models. The impact of adaptation measures is evaluated for the MIROC GCM, for which the climatic impact for water availability is closest to the mean of the five outputs. No distinction is made between the Amu Darya and Syr Darya basin, the effectiveness of the adaptation measures is evaluated for the total Amu and Syr Darya basin. The effectiveness of the adaptation measures is evaluated for 2041-2050.

For the two main rivers in Central Asia it is expected that total water shortage in 2050 will be 43,000 Mm3 per year. Moreover, changes in monthly flow regimes will change quite substantially, especially for the Amu Darya resulting from retreat of glaciers and reduction of snowfall (Fig. 16). By analyzing the various adaptation strategies the effectiveness of each measure can be assessed. By ranking the adaptation options by their unit costs the so-called “water-marginal-cost-curve” can be obtained. The water availability cost curve’s use is limited to comparing measures’ financial costs to close the gap. It is important to note that these might be different from the economic costs for society as a whole. The cost curve should be therefore considered as a guide for comparing the financial costs of measures for decision-making (Fig. 17).

It is clear that for Central Asia the most cost-effective adaptation measures are improving agricultural practice, deficit irrigation, increasing the reuse of water in agriculture and the reduction of irrigated areas. In general, the measures applied to agriculture are much more effective compared to the ones related to domestic water use. Applying the most cost-effective adaptation measures will close the water gap and costs US$ 1,730 million per year in 2050 (net present value). Closing the water gap caused by climate change only will cost US$ 550 million per year in 2050.

Figure 15. Current and future monthly flows in the two main rivers in Central Asia if no actions are taken.

Figure 17. Water marginal cost curve Amu and Syr Darya basin. Red arrow indicates total expected water shortage in 2050; green arrow indicates the hypothetical water shortage in case no climate change would occur. Note: Cost-axis has been cut off at US$ 0.30. Cost for decreasing domestic demand is 2.00 $/m3.

RFERENCES

ADB (2009). Understanding and responding to climate change in developing Asia. Philippines. 75 pp. ADRC (2006). Country Report for Asian Disaster Reduction Center. ADRC country report, Tajikistan. 23 pp. Agaltseva, N. (2005). Climate changes impact to water resources within Amudarya River basin. Report for the NeWATER Amudarya Case Study, NIGMI, Tashkent, Uzbekistan. Aizen, V. & E. Aizen (2010). Diagnosis of changes iin alpine water storages and land surface degradation in Pamir mountains and Amu Dar’ya River basin. Annual, second year progress report. University of Idaho, USA. 20 pp. Alamanov, S. K., Leveckin V., Podrezov O.A., Podrezov A.O. (2006). Climate change and water problems in Central Asia. Moscow-Bishkek: WWF Russia and UNEP. Arnell, N. W., (1996). Global Warming, River Flows and Water Resources. Wiley, Chichester, United Kingdom. Benestad, R., Hanssen-Bauer, I. and Chen, D., (2008). Empirical-statistical downscaling. World Scientific Publishing, 215 p. CLIVAR Exchanges, Special Issue, (2011). WCRP Coupled Model Intercomparison Project - Phase 5- CMIP5 - No. 56, Vol. 16, No.2, May 2011. Hewitt, C., Mason, S. and Walland, D., (2012). The Global Framework for Climate Services. Nature Climate Change, 2, 831–832. Huffman et al. (2007). The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-Global, Multiyear, Combined-Sensor Precipitation Estimates at Fine Scales. Journal of Hydrometeorology, 8, 38-55. DOI:10.1175/JHM560.1. Huffman, G.J., E.F. Stocker, D.T. Bolvin, E.J. Nelkin, R.F. Adler, (2012). TRMM Version 7 3B42 and 3B43 Data Sets. NASA/GSFC, Greenbelt, MD. Data set accessed 28th September 2011 at http://mirador.gsfc.nasa.gov/cgibin/mirador/present Navigation.pl?tree=project&project=TRMM&dataGroup=Gridded&CGISESSID =5d12e2ffa38ca2aac6262202a79d882a. Immerzeel, W. W., L. P. H. van Beek, and M. F. P. Bierkens (2010). Climate change will affect the Asian water towers., Science, 328(5984), 1382-1385, doi:10.1126/science.1183188. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20538947 (Accessed 20 July 2011). Ibatullin, S., V. Yasinsky, A. Mironenkov (2009). The impact of climate change on water resources in Central Asia. Sector report 6. Eurasian Development Bank. 43 pp. 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 [Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D. Mastrandrea, K.J. Mach, G.-K. Plattner, S.K. Allen, M. Tignor, and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, UK, and New York, NY, USA, 582 pp. Kerr, R. A., (2013). Melting glaciers, not just ice sheets, stoking sea-level rise. Science Vol. 340:6134, p. 798. Krige, D. G., (1951). A statistical approach to some basic mine valuation problems on the Witwatersrand, J. Chem. Metall. Min. Soc. S. Afr., 52, 119–139. Kutuzov, S & M. Shahgedanova (2009). Glacier retreat and climatic variability in the eastern Terskey-Alatoo, inner Tien Shan between the middle of the 19th century and beginning of the 21st century. Global and Planetary Change, 69, 59-70. 2

Lioubimtseva, E. and Hennebry, G., (2012). Grain production trends in Russia, Ukraine and Kazakhstan: New opportunities in an increasingly unstable world? Front. Earth Sci. 2012, 6(2): 157–166, DOI 10.1007/s11707-012-0318-y Meehl, G. A., Boer, G. J., Covey, C., Latif, M. and Stouffer, R. J., (2000). The Coupled Model Intercomparison Project (CMIP). Bull. Amer. Meteor. Soc., 81, 313-318. Novikov, V. (2002) Severe Hydrometeorological Events and their Fluctuation. World Meteorological Organization, CBS Teschnical Conference poster. Oerlemans J. (2005): Extracting a climate signal from 169 glacier records. Science 308, 675- 677. Räisänen, J. and Räty, O., (2012). Projections of daily mean temperature variability in the future: cross-validation tests with ENSEMBLES regional climate simulations. Clim Dyn, DOI 10.1007/s00382-012-1515-9. Randall, D.A., Wood, R.A., Bony, S., Colman, R., Fichefet, T., Fyfe, J., Kattsov, V., Pitman, A., Shukla, J., Srinivasan, J., Stouffer, R.J., Sumi, A. and Taylor, K.E., (2007). Climate Models and Their Evaluation. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Schlüter, M., Hirsch, D. and Pahl-Wostl, C., (2010). Coping with the change: responses of the Uzbek water management regime to socio-economic transition and global change. Environmental Scienced and Policy, 13(7), 620-636. Seversky, I.V. (2006). Present and predictable changes of snow and glaciation of zone of flow formation and their possible influence on water resources of Central Asia. Almaty. Taylor, K.E., Stouffer, R.J., Meehl, G.A., (2012). An Overview of CMIP5 and the experiment design.” Bull. Amer. Meteor. Soc., 93, 485-498, doi:10.1175/BAMS-D-11-00094.1, 2012. UN (2000). Kyrgyz Republic. Environmental Performance Reviews Series No. 9. 155 pp. U.S. Forest Service (2013). Feds project climate change will double wildfire risk in forests - The Denver Post. http://www.denverpost.com/breakingnews/ci_22943189/feds-project-climate- change-will-double-wildfire-risk#ixzz2Su9dleqE WCRP Coupled Model Intercomparison Project – Phase 5, (2011). Special Issue of the CLIVAR Exchanges Newsletter, No. 56, Vol. 15, No. 2.