Process Based Prediction of Flow Distribution at Ungauged Himalayan Catchments using Satellite Data

In-line hydropower (IHP), whereby micro hydropower turbines are integrated in gravity fed rural water supply systems, is a promising yet underutilized infrastructure synergy in rural mountainous areas of developing countries. In Nepal, although institutional and economic considerations are critical, the available local hydropower potential - in particular the flow availability to drive turbines - is the key feasibility criterion. Yet flow projections in remote areas remain a significant challenge largely due to difficult access and data scarcity. We recently showed that satellite data could be successfully adjusted to generate parameters describing the frequency and intensity of daily rainfall. The flow predictions needed for hydropower siting relate primarily to a frequency domain description, specifically the fluw duration curves, which are closely controlled by these rainfall parameters. This project combines bias-corrected satellite rainfall parameters with remotely sensed stationary watershed data to predict the flow duration curves at ungauged Himalayan watersheds. In particular, we compare the performance of several approaches with different levels of process representation, applied to sparsely gauged, snow-fed and topographically complex catchments with highly seasonal rainfall.