Fate of Glaciers. Both catchments respond with a clear decreasing trend in glacier area to the changes in climate. Multimodel median results indicate 53% (RCP45) or 70% (RCP85) glacier area loss in the Juncal region between 2001–2010 and 2091–2100 (Fig. 3). For the same period, the simulations indicate a decrease in glacierized area by 35% (RCP45) or 55% (RCP85) for the Langtang region (Fig. 3). Debris-covered glacier area (representing 27% of the total glacier area in Langtang) is less sensitive to the changes in climate and de- creases only by 25% (RCP45) or 33% (RCP85) until the end of the century. It is typical of heavily debris-covered glaciers with stagnant low-gradient termini that fronts are more stable (25). High air tem- peratures prevailing on the low-reaching tongues and enhanced melting on exposed ice cliffs and beneath supraglacial lakes can substantially mitigate the shielding effect of supraglacial debris (26– 29). However, in the Langtang region, melt rates of debris-covered ice are much lower than of non–debris-covered ice (12). In the long run, this leads to glacier tongues that are disconnected from the ac- cumulation areas (Fig. 1 iii and iv and SI Appendix, Fig. S12 ). Glacier Contributions to River Flow. Ice melt from glaciers represents roughly one-third of total simulated streamflow during the reference period (2001–2010) in Langtang, and one-fifth in Juncal (Fig. 2). We show that total ice melt is on a rising limb in Langtang at least until 2041–2050 and starts to decrease again after 2051–2060 (Fig. 3). These results confirm the findings by a previous modeling study (30). In Juncal, however, total ice melt was already beyond its tipping point at the beginning of the 21st century according to our simulations. This contrasting response to climate warming can be explained by differ- ences in the elevation distribution of the glaciers in the two regions. In Juncal, many glaciers are melting up to the highest elevations
Glaciers in Pakistan cover about 13,680 km2 which is 13% of the mountainous areas of the Upper Indus Basin (UIB) [5, 6]. Glacial and snowmelt water from these glaciers adds considerably to the flows of UIB rivers [2, 13]. According to a report of International Commission for Snow and Ice (ICSI), “Glaciers in Himalayas are retreating at a rate higher than in any other part of the world and, if the current rate continues, the possibility of their departure by the year 2035 is very high” [5, 6]. On the other hand, the Intergovernmental Panel on ClimateChange (IPCC) has accepted the errors in the estimation of glacier melting in Himalayan region . The IPCC stated that, “Widespread mass losses from glaciers and reductions in snow cover are projected to accelerate throughout the 21st century, reducing water availability, hydropower potential and changing seasonality of flows in regions supplied by melt water from major mountain ranges like Hindu-Kush, Himalaya, Andes” . In 2005, Hewitt described a broad evidence of glacier development in the late 1990s in the Central Karakoram, in contradiction of a worldwide reduction of mountain glaciers . Moreover, based on surveys between 1997 and 2002, Hewitt also described that some of the large Karakoram glaciers, 40 to 70 km in length, depicted 5 to 15 m thickening over considerable ablation zone areas, locally more than 20 m [5, 10]. These conflicting findings by different research studies make the impact of climatechange on Karakoram glaciers and Indus Riverflows quite vague. 2.2 Changes in Temperature
The central Himalaya lies between approximately 78 0 -85 0 E and 30 0 -28 0 N, covering the Uttarakhand region of India and Nepal to the east. It is hard to define where the western Himalaya ends and central Himalaya starts, and also which far eastern districts of Nepal can be defined under eastern Himalaya rather than central. Highmountains range between 3,000-7,000 m and the foothills measure from 600-1,200 m (Basistha et al., 2009). The Ganges flows through Uttarakhand in the west central Himalaya and then heads south-east towards Bangladesh. The Bhagirathi, Karnali, Gandaki, Kosi and Yamuna Rivers are the main central Himalayan tributaries of the Ganga. The upper Ganges catchment is a transitional zone as summer monsoon strengthens from west to east, but it is still influenced by winter westerlies in the western central Himalaya. Summer monsoon amount increases during August, but discharge dips during this month. Due to the transient snow line lowering as a result of snowfall at high elevation, runoff decreases with the higher albedo of snow coupled with cloud cover. Average precipitation in Nepal is 1768mm (Shrestha et al., 2000), but this will greatly differ regionally and with changes in elevation and aspect. The Annapurna range receives around 5 m of annual precipitation compared to 1948 mm at Mukkim at 1900 m a.s.l. (Shrestha et al., 2011). Across the central Himalaya, runoff variations occur as summer monsoon declines from east to west, reducing the percentage contribution of precipitation to runoff. Bookhagen & Burbank (2010) highlighted that snowmelt only contributes 20% of river flow in the central Himalaya by comparison to 30% for the Ganga at Devprayag (Singh et al., 1997).
circulation systems) is to be expected given that the spatial resolution of these datasets is larger than the spa- tial variability of precipitation. Consequently, it was shown that the bias‐corrected WRF precipitation output was more accurate than the gridded datasets, suggesting that it is more suitable for use in impact studies for forcing hydrological and water system models and has further implications on the potential of regional cli- mate modeling approaches to predict future changes in water resources in the HKKH region. Note that the basin‐wide precipitation patterns estimated by the bias‐corrected WRF output are broadly consistent with suggestions from Singh and Jain (2003) and Jain et al. (2010) that approximately 40–60% of the annual accu- mulation of these basins falls in the winter at high‐elevations (in the form of frozen precipitation). It is speculated that the cause of the dry bias over the Sutlej basin in the raw WRF model output is related to a failure to represent an early ‐morning maximum in precipitation during the monsoon period. Such a peak is apparent in the simulated output for the Beas basin and also reported over other locations focused on south facing slopes of the HKKH (e.g., Bhatt & Nakamura, 2005; Orr et al., 2017; Prasad, 1974). However, con ﬁr- mation of its existence was hampered by the unavailability of subdaily station data. It was further speculated that this is related to excessive precipitation falling upstream (over the Beas), resulting in a moisture ‐deﬁcit as the air passes over the Sutlej. It is also likely that the bias is related to the representation of localized diur- nal mountain ‐valley winds, which play a key role in inducing convection. Adequate representation of these winds in regional climate models is a long ‐standing issue, with a number of previous studies highlighting the dif ﬁculty of representing them in the HKKH region (e.g., Bhatt & Nakamura, 2006; Norris et al., 2017; Orr et al., 2017; Sato, 2013; Shrestha & Deshar, 2014).
The data required for the SWAT modeling are Digital Elevation Model (DEM), land use, soil and climate data, as well as reservoir and point source information. A DEM contains all the information about watershed terrain and streams networks. A 30 m DEM resolution for the study area was obtained from the USGS National Elevation Dataset  . The DEM was utilized to delineate the watershed and create 144 sub-watersheds. Similarly, the most recently available land use dataset (2011), which has a spatial resolution of 30 m, was acquired from the National Land Cover Dataset  . The distribution of land use in the watershed is presented in Table 1. State Soil Geographic (STATSGO) and Soil Survey Geographic (SSURGO) are the two most commonly used soil databases  . The STATSGO data was used in this study due to relatively large size of the watershed and the extremely detailed characteristics of the SSURGO data. As the hydrological processes are quantified in a small entity known as hydrologic response unit (HRU), the threshold values of 5%, 15% and 15% for land use, soil and slope respectively were used to create 2676 HRUs to improve computational efficiency of simulations.
A substantial number of large-scale climatechange im- pact studies that have been performed recently examine the future hydrological state analysing projections of runoff or river flow. Fung et al. (2011) compared the projected future water availability under + 2 and + 4 ◦ C of global warming, forcing the MacPDM Global Hydrological Model (GHM) with 22 Global Climate Models (GCMs) from the CMIP3 experiment. Arnell and Gosling (2013) performed a global assessment of the climate driven changes in runoff-based hy- drologic indicators in the mid-21st century, using multiple scenarios derived from the CMIP3 experiment. Schneider et al. (2013) focused on the impacts of climatechange for the European riverflows, using data from three bias corrected GCM scenarios. Van Vliet et al. (2013) performed a global assessment of future river discharge and temperature under two climatechange scenarios, forcing a GHM with an en- semble of bias corrected GCM output. They found that the combination of lower low flows with increased river water temperature can lead to water quality and ecosystem degra- dation in the south-eastern US, Europe, eastern China, south- ern Africa and southern Australia. An investigation of the fu- ture trends in flood risk at the global scale was performed by Dankers et al. (2014) and for the European region by Alfieri et al. (2015). Betts et al. (2015) performed a global assessment of the impact posed on riverflows and terres- trial ecosystems by climate and land use changes described by four RCPs (Representative Concentration Pathways). Var- ious multi-model hydrological simulations have also been performed, in an attempt to quantify the climatechange anal- ysis’ uncertainty resulting from the impact model (Hage- mann et al., 2013; van Huijgevoort et al., 2013; Dankers et al., 2014).
This research demonstrates the importance of coupling hydrologic and species habitat modelling to quantify habitat recovery potential from channel restoration efforts or changes to ﬂow conditions. In this analysis, historical and future climatechange scenario ﬂows, as well as physical restoration alterations, in ﬂuence ﬂoodplain habitat advantages for native ﬁsh. A suite of species life histories that span a range of necessary ﬂow requirements allow us to examine an array of impacts associated with four ﬂow scenarios for the San Joaquin river system. Notably, the modelled results project signi ﬁcant declines in the avail- ability of required ﬂow-related habitat conditions for splittail spawning and rearing and Chinook salmon rearing in the future under two climatechange scenarios. Under historical ﬂows, splittail and Chinook salmon thresholds for ecological bene ﬁts were lower than those estimated from the recent ﬂow record. Our modelled results for the two periods did not reveal differences in estimated produc- tion of phytoplankton, which require frequent lower ﬂood pulses. This is expected given that the river had regulated ﬂows during both of these periods. Currently, the operation of over 80 dams within the San Joaquin River watershed reduces and eliminates most ﬂow peaks (Cain et al., 2003), limiting phytoplankton and zooplankton produc- tion. The duration of high magnitude maintenance ﬂows varies greatly year to year and in many cases does not last long enough to provide suf ﬁcient habitat for splittail and salmon.
Abstract. Rivers are essential to aquatic ecosystem and so- cietal sustainability, but are increasingly impacted by water withdrawals, land-use change, and climatechange. The rela- tive and cumulative effects of these stressors on continental riverflows are relatively unknown. In this study, we used an integrated water balance and flow routing model to evalu- ate the impacts of impervious cover and water withdrawal on river flow across the conterminous US at the 8-digit Hy- drologic Unit Code (HUC) watershed scale. We then esti- mated the impacts of projected change in withdrawals, im- pervious cover, and climate under the B1 “Low” and A2 “High” emission scenarios on riverflows by 2060. Our re- sults suggest that compared to no impervious cover, 2010 levels of impervious cover increased riverflows by 9.9 % on average with larger impacts in and downstream of major metropolitan areas. In contrast, compared to no water with- drawals, 2005 withdrawals decreased riverflows by 1.4 % on average with larger impacts in heavily irrigated arid re- gions of Western US. By 2060, impacts of climatechange were predicted to overwhelm the potential gain in river flow due to future changes in impervious cover and add to the potential reduction in riverflowsfrom withdrawals, decreas- ing mean annual riverflowsfrom 2010 levels by 16 % on average. However, increases in impervious cover by 2060 may offset the impact of climatechange during the grow- ing season in some watersheds. Large water withdrawals will aggravate the predicted impact of climatechange on riverflows, particularly in the Western US. Predicted ecohydro- logical impacts of land cover, water withdrawal, and climatechange will likely include alteration of the terrestrial water balance, stream channel habitat, riparian and aquatic commu- nity structure in snow-dominated basins, and fish and mussel extirpations in heavily impacted watersheds. These changes
The accuracy of endmember hydrograph separation models is limited by the uncertainties of the estimated values of each endmember component, the uncertainty of the cubic spline interpolation at each data point and the uncertainty of δ 18 O in the river. While the uncertainty of δ 18 O in the river is likely to be relatively small, the uncertainties of each endmember component must be kept in mind (e.g. Cable et al., 2011; Arendt et al., 2015). The assumption of discrete values of each endmember component is unlikely to reflect the spatial and temporal changes in bulk δ 18 O of snowmelt, ice melt and rainwater. For instance, Raben and Theakstone (1998) found a seasonal increase in mean δ 18 O in snow pits on Aus- tre Okstindbreen, Norway, and episodic events such as pas- sages of storms (e.g. McDonnell et al., 1990; Theakstone, 2008) or melting of fresh snow in the late ablation season may cause temporal changes in one component. Also, snow- packs have a non-uniform layered structure with heteroge- neous δ 18 O composition, and isotopic fractionation is likely to occur as melting progresses and the snowpack is mixed with rainwater (e.g. Raben and Theakstone, 1998; Lee et al., 2010). It is also difficult to assess how representative snow pits and ice transects are for the bulk δ 18 O value of each component. Spatial differences in δ 18 O may exist within and between snow pits, but the overall effect on the isotopic com- position of the water leaving the melting snowpack at a given time is unknown.
The analysis of observed precipitation records revealed signi ﬁcant altitude dependency of precipitation in all the sub-basins ( Fig. 3 ), which supports earlier studies (e.g. Pang et al., 2014; Winiger et al., 2005; Hewitt, 2011; Weiers, 1995; Wake, 1989; Dhar and Rakhecha, 1981; BIG, 1979; Decheng, 1978 ). However, there is substantial differ- ence in the rate and magnitude of variation from one basin to another due to signi ﬁcant directional bias (spatial autocorrelation) and inﬂu- ence of highly diversi ﬁed orography (topography and exposure) interacting with multiple weather systems. Therefore, the complex alti- tudinal variation of precipitation in the high-altitude Indus basin cannot be represented by a single relation. Such an elusive behaviour of precip- itation gradient was also found by Immerzeel et al. (2014) in Nepalese Himalayas, where a uniform valley wide precipitation gradient could not be established due to in ﬂuence of several scale-dependent mecha- nisms. Although, we attempted a separate analysis for each sub-hydro- logical basin, yet the spatial variability in each sub hydrological basin is so high that the number of available observations is inadequate to infer an accurate distribution of altitudinal precipitation. Rather complex and nonlinear trend of precipitation increase with altitude is evident in most sub-basins. The south-west TP and eastern Karakoram regions display an elusive trend mainly due to higher variability and very less number of observation points. Astore and Chitral basins depict mixed trend, while Shigar, Hunza and Gilgit basins infer relatively strong positive ver-
Validation is evaluation of the model outputs with an independent data set without making further adjustments. The process is to confirm that the simulation is good enough that the validation was carried out using the calibrated parameters . For model validation the reaming observed stream flow of Kulfo hydrological station was used. In the validation process the model was run with a parameter set without any change of the parameter during the calibration process. The four performance indicators during the calibration were again used in the validation period to evaluate the performance of the model. The Validation result showed that the coefficient of determination (R 2 ) and the Nash-Sutcliff efficient (NSE) are 0.92 and 0.78 respectively. These are according to Moriasi suggestion the performance rating is Very good for both R 2 and NSE . The Standard Deviation Ratio (RSR) and Percent Bias (PBIAS) is 0.46 and 21% which is under Very good and satisfactory range respectively. The time series data of the observed and simulated flows on monthly basis were plotted for comparison (Figure 5). The scatter plot value of the measured and simulated flow has shown a linear correlation between the data sets (figure 6)
The most important part of calibration, particularly when a hydrological model is used to assess the impact of climatechange on a flow regime, is confidence in the modelled relationship between rainfall and runoff which may, or may not, be reflected by good objective function values. If modelled riverflows are what would have actually occurred if the given inputs of rainfall and PE had been spatially uniform over the modelling area (catchment or grid) and temporally uniform over the model time step then the model is a good representation of hydrological processes in the catchment. In reality, spatial and temporal uniformity of rainfall is unlikely to occur (except over very small areas and time steps), hence this is one of the main reasons why modelled flows always differ from observed. Care is therefore required during the process of calibration in the interpretation of objective function values. Poor final objective function values should not be disregarded nor the calibration automatically discarded but all values must be evaluated with discretion. A visual comparison of observed and simulated hydrographs should be made to ensure the calibrated
The river flow is a result of runoff from the catchment, which is affected by climate, as studied in this paper, but also other factors, such as land use. We expect that land-use changes will result in changes to the runoff regime. Land use has significantly changed in some sub-basins, notably in reduced forest cover. This study used a static 2003 land-use map for the whole simulation period, which would perforce a decrease in the accuracy of simulations, but they serve to isolate the differential effect of climatechange on runoff. The relative paucity of the available ground truth data, and the rapid rate of land-use change makes it extremely difficult to forecast landuse with much certainty , but the potential effects of land-use change on runoff should be explored further. In addition, water withdrawals for irrigation and urban water supply were not included in the model, and these are increasing, so that they will likely have a much greater effect in the future. There are, of course, a number of other caveats with the analysis. The effects of hydrological model structure and parameter uncertainties were not considered. However, the effects of model parameter uncertainty have been shown  to be small, relative to the differences between climate scenarios. There are insufficient long-term rain gauges in the headwaters of the Tonle Sap tributaries, thus SWAT-assigned rainfall from the nearest stations, in some cases in Thailand. There is also considerable variability the between climate model scenarios, particularly at the regional scale.
 assessed the likely impact of climate changes on catchment hydrology and water resources.  used sta- tistically downscaled out put from Global Climate Mod- els as forcing into a lumped conceptual rainfall-runoff model, to analyzed changes on ground storage, steam flow and extreme events. Future simulations using the rainfall-runoff models suggest the reductions in soil mois- ture storage throughout the summer and autumn months are likely for catchment across the globe. The decrease in storage is likely dependent on the storage potential of the individual catchments. The lower the capacity of a catch- ment to store water the greater the sensitivity of climatechange. Reductions in ground water storage during the recharge period according to  will increase the risk of severe drought because of failure of winter or spring pre- cipitation may results in prolong drought period where the ground water system is unable to recover.
The Sundarbans, an UNESCO Heritage site has a large rural population which depends on natural resources for sustenance. The present paper deals with the management of the salt water intrusion of the Piyali River a tributary of the Matla River which empties into the Bay of Bengal. The study also delves into the population affected by the effects of the perennially saline river and their dependence on it for their livelihood. A look into the soil texture, seasonal variation in chloride content of soil along with pH and Electrical Conductivity (EC) levels of water sampled at different time and locations is analyzed in order to improve management options. With (EC) values of 17,000 mS and pH 8.94, sustaining the inhabitants in this area is quite chal- lenging. Under the threat of climatechange, increased levels of salinity arising from sea level rise and coastal flooding will pose a serious problem to the rural inhabitants of the Sundarbans. The predicted negative im- pacts of climatechange are likely to bring new challenges in addition to magnifying existing problems, par- ticularly in the Sundarbans community that already has limited capacity to adapt to these changes.
Adaptation to environmental change is not a new phenomenon (Tompkins et al., 2010). Societies have adapted to their environments to mitigate risks associ- ated with climate variability throughout human history. This interaction is represented in cultural landscapes which constitute a testimony of the past and present- day relationships between society and its environment (Rescia et al., 2008). Indeed, indigenous people and farming communities are a life example of it. They are facing di ﬀ erent aspects of climatechange depending in where and how they live. People are not only keen ob- servers of climatechange, but also actively try to adapt to the changing conditions (Byg & Salick, 2009; Turner & Clifton, 2009). The coherence in this interaction is described by the theory of the co-evolution of ecologi- cal and social systems (Gual & Norgaard, 2010; Kallis & Norgaard, 2010), which contributes on the assessment of adaptation strategies.
the control climate. Daily snowfall rates are aggregated in 5 ◦ C bins with centers from -22.5 ◦ C to 12.5 ◦ C according to the climatological monthly surface air temperature in the control climate for each grid box and day. Snowfall extremes are calculated as high percentiles of the daily snowfall rates in each temperature bin including days with no snowfall. Both mean snowfall and snow- fall extremes in the different temperature bins are in good agreement with observational estimates (Extended Data Fig. 3). The response to climatechange is first presented for surface elevations below 1000m (Extended Data Fig. 4). Fractional decreases are greater for mean snowfall as com- pared to snowfall extremes for much of the temperature range considered here (Fig. 2a), which demonstrates the contrasting responses of mean and extreme snowfall even when monthly varia- tions in climatological temperature are controlled for. For the temperature bin centered at -2.5 ◦ C, mean snowfall decreases by 65% whereas the 99.99th percentile of snowfall decreases by only 8%. Changes in snowfall extremes transition from positive to negative at control-climate tempera- tures as high as -9 ◦ C in the multimodel median, whereas the corresponding temperature for mean snowfall is -14 ◦ C. Furthermore, the difference in behavior between mean and extremes is greater the higher the percentile of snowfall considered (Fig. 2a), and it is robust across different climate models (Extended Data Fig. 5).
moisture parameter set are FC (maximum soil moisture stor- age in millimeter), LP (fraction of FC above which potential evapotranspiration occurs and below which evapotranspira- tion will be reduced) and the coefficient BETA (determining the relative contribution to runoff from a millimeter of pre- cipitation at a given soil moisture deficit). These parameters are dependent on the properties of the catchment, such as the land use type, the wilting point and soil porosity. They will affect the simulated discharge volume. The other parame- ter set includes runoff parameters such as ALFA (measure of the non-linearity for runoff), HQ (the higher flow level at which the recession rate KHQ is assumed) and KHQ (re- cession coefficient at HQ). These parameters influences the shape of the hydrograph (SMHI, 2004). Because of the un- certainty of the parameters, the Monte Carlo Random Sam- pling (MCRS) method is popularly used for parameter esti- mation (Lamb, 1999; Liden and Harlin, 2000) in the cali- bration of the model. However, because the program source code is not available, the above method is difficult to apply in our case. Therefore, quasi-stratified sampling in the form of Latin Hypercube Sampling (McKay et al., 1979) is used. The limited sampling numbers of this method can produce similar results to the Monte Carlo approach (Yu et al., 2001; Murphy et al., 2004). In the previous HBV studies, much experience has been gained in the parameter estimation, which is used to acquire the reasonable ranges of the main parameters in our study (Uhlenbrook et al.,1999; Seibert, 1999; Krysanova et al., 1999; Diermanse, 2001; SMHI, 2004; Booij, 2005). FC ranges from 200 to 500, LP from 0.6 to 1.0, BETA from 1.0 to 5.0, ALFA from 0.8 to 1.1, KHQ from 0.08 to 0.14 and HQ is fixed to 3.0. According to Murphy et al. (2004) and the range of the different parameters, 50–100 sampling numbers are used in the calibration.
Ducks were captured and studied in July and August of 2014 and 2015. Five species were captured and tested at highaltitude (3812 m above sea level) at the Lake Titicaca National Reserve (Puno, Peru) in August 2014: speckled teal (Anas flavirostris oxyptera Vieillot 1816; n=12, 4 males and 8 females), Andean ruddy duck (Oxyura jamaicensis ferruginea; n=12, 5 males and 7 females), yellow- billed pintail (Anas georgica Gmelin 1789; n=13, 10 males and 3 females), Andean cinnamon teal (Anas cyanoptera orinomus Vieillot 1816; n=12, 8 males and 4 females) and puna teal [Anas puna (Tschudi 1844); n=12, 7 males and 5 females; body mass of 404±11 g]. Four species, representing closely related populations of the same species or sister species of four of these high-altitude taxa, were captured at low altitude in Oregon, USA (at either Summer Lake Wildlife Management Area at 1260 m or Malheur National Wildlife Refuge at 1256 m) in July 2015, and were tested at Summer Lake: green-winged teal (Anas crecca Linnaeus 1758; n=10, 5 males and 5 females), ruddy duck [Oxyura jamaicensis jamaicensis (Gmelin 1789); n=8, 4 males and 4 females], northern pintail (Anas acuta Linnaeus 1758; n=10, 7 males and 3 females) and northern cinnamon teal (Anas cyanoptera septentrionalium Vieillot 1816; n=11, 6 males and 5 females). Torrent ducks (Merganetta armata Gould 1841) were also captured and tested in August 2015, both at highaltitude (3000 – 4086 m above sea level; n=8, all males) on the Chancay River Valley near Vichaycocha, Lima, Perú, and at low altitudes (1092 – 1665 m above sea level; n=14, all males) on the Chillón River in Santa Rosa de Quives, Lima, Perú. Ducks were allowed to recover overnight from capture for at least 6 – 12 h, with unlimited access to water, before responses to acute hypoxia were measured. During this time, birds were held in large animal kennels with dry bedding. All experiments were performed within 2 days of capture, and birds were tube fed commercial duck chow if held for longer than 1 day in captivity, but food was always withheld for 6 – 12 h before measurements took place. Ducks were collected in accordance with permits issued by the Ministerio del Ambiente del Peru (004-2014-SERNANP-DGANP-RNT/J), the Ministerio de List of symbols and abbreviations
General Circulation Models (GCMs) are an important tool for assessing the impact of climatechange on a range of human and natural systems. Simulations at these inner scales are of considerable interest to hydrologists in assessing the possible impact of climatechange on water resources. Different climate models have been used worldwide for climateimpact assessment studies. Climate models, particularly the GCMs, currently provide the most important source of information for constructing scenarios of climatechange, which provide climate information at a higher spatial resolution, gradually becoming available. GCMs are based on physical laws and physical-based empirical relationships and are mathematical representations of the atmosphere, ocean, cryosphereand land surface processes.