Precipitation pattern in the Western Himalayas revealed by four datasets

Data scarcity is the biggest problem for scientific research related to hydrology and climate studies in the Great Himalayas region. High-quality precipitation data are difficult to obtain due to a sparse network, cold climate and high heterogeneity in topography. In this paper, we examine four datasets in northern India of the Western Himalayas: interpolated gridded data based on gauge observations (IMD, 1 degrees x 1 degrees, and APHRODITE, 0.25 degrees x 0.25 degrees), reanalysis data (ERA-Interim, 0.75 degrees x 0.75 degrees) and high-resolution simulation by a regional climate model (WRF, 0.15 degrees x 0.15 degrees). The four datasets show a similar spatial pattern and temporal variation during the period 1981-2007, though the absolute values vary significantly (497-819 mm year(-1)). The differences are particularly large in July and August at the windward slopes and high-elevation areas. Overall, the datasets show that the summer is getting wetter and the winter is getting drier, though most of the trends in monthly precipitation are not significant. Trend analysis of summer and winter precipitation at every grids confirms the changes. Wetter summers will result in more and bigger floods in the downstream areas. Warmer and drier winters will result in less glacier accumulation. All the datasets show consistency in the period 1981-2007 and can give a spatial overview of the precipitation in the region. Comparing with the Bhuntar gauge data, the WRF dataset gives the best estimates of extreme precipitation. To conclude, we recommend the APHRODITE dataset and the WRF dataset for hydrological studies for their improved spatial variation which match the scale of hydrological processes as well as accuracy in extreme precipitation for flood simulation.