Modelling the impact of past and future climate scenarios on streamflow in a highly mountainous watershed: A case study in the West Seti River Basin, Nepal

Hydrological model parameters are important during representation of the hydrological characteristics of a watershed. The West Seti River Basin (WSRB), a prominent Himalayan Basin of Nepal, is a major source of fresh water in the western region of the country. We used the Soil and Water Assessment Tool (SWAT) for hydrological modelling and identified the most sensitive hydrological parameters, while the Sequential Uncertainty Fitting (SUFI-2) technique was employed for model calibration. The model was calibrated for the study period (1999-2005) with a three-year warm-up period (1996-1998). Subsequently, it was validated for three years (2006-2008). The results show that the large number of Hydrological Response Units (HRUs) for model simulation took a considerable time, without improving the performance statistics. Importantly, significant improvements were observed during both calibration and validation periods when elevation bands (EBs) were taken into consideration. The p-factor, r-factor, coefficient of determination (R-2), Nash-Sutcliffe efficiency (NSE), percent bias (PBIAS), Root mean square error (RMSE)-observations, and standard deviation (STDEV) ratio (RSR) were used to measure the performance between observed and simulated values. The values of p-factor, r-factor, R-2, NSE, PBIAS, and RSR during the calibration were 0.82, 0.80, 0.84, 0.82, 7.2, and 0.42, respectively, whereas during validation they were 0.79, 0.72, 0.83, 0.82, 11.8, and 0.42, respectively. The calibrated model was then used to assess the anticipated river discharge. This study used four regional climate models (RCMs) for precipitation and six for temperature, together with their arithmetical average as multi-model ensembles (MMES) under two representative concentration pathways (RCPs). We analysed the changes in precipitation, temperature, and river discharge for three future time frames: Futurel (F1: 2020-2044), Future2 (F2: 2045-2069), and Future3 (F3: 2075-2099) with respect to the baseline (1996-2005). The magnitude of changes varied according to the different climate models and warming scenarios. In general, the MMEs showed slightly increasing precipitation (higher during the F2 period), significantly increasing temperature (continuous rising trend), and moderately increasing river discharge (higher during the F2 period). Information such as the anticipated shift in the flow duration curve may be helpful to stakeholders across different water sectors for effective water resource management in the future. From the modelling perspective, the results show greater significance for EBs than HRUs during the modelling of high mountain basins with SWAT. This take-home message would be useful to hydrologists and other stakeholders in evaluating different scenarios over a short duration, without iteratively spending higher computational time. (C) 2020 Published by Elsevier B.V.