Evaluation of precipitation elasticity using precipitation data from ground and satellite-based estimates and watershed modeling in Western Nepal
Rocky Talchabhadel, Anil Aryal, Kenji Kawaike, Kazuki YAMANOI, Hajime Nakagawa, Binod Bhatta, Saroj Karki, Bhesh Raj Thapa
Abstract
West Rapti River (WRR) basin, Western Nepal. Hydrologic modeling requires an accurate precipitation data at a high spatial resolution, which is often limited in many regions of the globe. As a complement to the ground (gauge) precipitation data, satellite-based precipitation estimates (SPEs) appear useful. At first, this study evaluated performance of three different SPEs, namely i) CHIRPS, ii) PERSIANN-CCS, and iii) IMERG, with respect to gauge data using different event detection and quantification indices. Soil Water Assessment Tool (SWAT), a semi-distributed hydrologic model, was used to simulate the river discharge. We then analysed precipitation elasticity, as a first kind of such study in Nepalese river basin, by scaling the precipitation input in both positive and negative directions (ranging from -20 % to +20 %) in order to explore basin response on likely alteration of precipitation. A non-parametric precipitation elasticity was finally computed for three different cases: 1) observed river discharge, 2) gauge-based simulated river discharge, and 3) SPEs-based simulated river discharge. IMERG proved to be superior among three SPEs. All SPEs showed improved results after implementation of different levels of bias-correction where daily precipitation data were corrected using linear correction factors computed at a mean monthly scale. Computed correction factors are replicable to nearby basins. Precipitation elasticity of the study area ranged from +1.3 to +2.0 (approximately +1.5) which indicates that a 1.0 % change in precipitation will result in 1.5 % change in river discharge.