No abstract

This paper provides the results of semi-distributed positive degree-day (PDD) modelling for a glacierized river basin in Nepal. The main objective is to estimate the present and future discharge from the glacierized Langtang River basin using a PDD model (PDDM). The PDDM is calibrated for the period 1993-98 and is validated for the period 1999-2006 with Nash-Sutcliffe values of 0.85 and 0.80, respectively. Furthermore, the projected precipitation and temperature data from 2010 to 2050 are obtained from the Bjerknes Centre for Climate Research, Norway, for the representative concentration pathway 4.5 (RCP4.5) scenario. The Weather Research and Forecasting regional climate model is used to downscale the data from the Norwegian Earth System Model general circulation model. Projected discharge shows no significant trend, but in the future during the pre-monsoon period, discharge will be high and the peak discharge will be in July whereas it is in August at present. The contribution of snow and ice melt from glaciers and snowmelt from rocks and vegetation will decrease in the future: in 2040-50 it will be just 50% of the total discharge. The PDDM is sensitive to monthly average temperature, as a 28C temperature increase will increase the discharge by 31.9%. Changes in glacier area are less sensitive, as glacier area decreases of 25% and 50% result in a change in the total discharge of -5.7% and -11.4%, respectively.

Discharge over the Narayani river catchment of Nepal was simulated using Statkraft's Hydrologic Forecasting Toolbox (Shyft) forced with observations and three global forcing datasets: (i) ERA-Interim (ERA-I), (ii) Water and Global Change (WATCH) Forcing Data ERA-I (WFDEI), and (iii) Coordinated Regional Climate Downscaling Experiment with the contributing institute Rossy Centre, Swedish Meteorological and Hydrological Institute (CORDEX-SMHI). Not only does this provide an opportunity to evaluate discharge variability and uncertainty resulting from different forcing data but also it demonstrates the capability and potential of using these global datasets in data-sparse regions. The fidelity of discharge simulation is the greatest when using observations combined with the WFDEI forcing dataset (hybrid datasets). These results demonstrate the successful application of global forcing datasets for regional catchment-scale modeling in remote regions. The results were also promising to provide insight of the interannual variability in discharge. This study showed that while large biases in precipitation can be reduced by applying a precipitation correction factor (p_corr_factor), the best result is obtained using bias-corrected forcing data as input, i.e. the WFDEI outperformed other forcing datasets. Accordingly, the WFDEI forcing dataset holds great potential for improving our understanding of the hydrology of data-sparse Himalayan regions and providing the potential for prediction. The use of CORDEX-SMHI- and ERA-I-derived data requires further validation and bias correction, particularly over the high mountain regions.

In the Himalayan region, aerosols received much attention because they affect the regional as well as local climate. Aerosol Optical Depth (AOD) observation from satellite are limited in the Himalayan region mainly due to high surface reflectance. To overcome this limitation, we have conducted a multivariate regression analysis to predict the AOD over the cryospheric portion of Nepalese Himalaya. Prediction using three meteorological variables from ERA-Interim: relative humidity, wind velocity components (U10 and V10) were taken into account for model development as independent variables, while the longest time series AOD observation at Pokhara station is used as dependent variable. Model coefficients were found significant at 95 percent level with 0.53 coefficients of determination for daily values. Correlation coefficients between model output and AERONET observations were found to be 0.68, 0.73, 0.75, 0.83, and 0.82 at Lumbini, Kathmandu Bode (KTM-BO), Kathmandu University (KTM-UN), Jomson, and Pyramid laboratory/observatory (EVK2CNR) AERONET stations, respectively. Model overestimate AOD at Jomsom, and EVK2CNR AERONET stations while slightly underestimates AOD in Lumbini, KTM-UN, and KTM-BO AERONET station, respectively. Both model output and MODIS observation showed that the highest AOD over Nepal is observed during winter and pre-monsoon season. While lowest AOD is observed during monsoon, and post-monsoon season. The result of this research supports that the use of linear regression model yields good estimation for daily average AOD in Nepal. The model that we have presented could possibly be used in other mountain regions for climate research.

Observation and model-based studies suggest substantial hydrological flow pattern changes in mountain watershed where hydrology is dominated by cryospheric processes (IPCC 2007