River runoff estimation with satellite rainfall in Morocco

Authors:

Abstract
In most African countries, the lack of observed rainfall data is a major obstacle for efficient water resources management. The objective of this study is to evaluate satellite rainfall products’ ability to estimate river runoff over 12 basins in Morocco using four hydrological models: IHACRES, MISDc, GR4J, and CREST. Satellite products available with a short latency are compared: the H03, H05, H64, and H67 products from EUMETSAT H SAF, SM2RAIN-ASCAT, and IMERG. The best results to reproduce river runoff were obtained with SM2RAIN-ASCAT in combination with the IHACRES hydrological model, with the highest Kling Gupta Efficiency criterion and probability of detection for runoff events above the median or the 95th percentile. Since hydrological model performances differed across catchments and satellite rainfall products, it highlights the need to carefully select hydrological models for a given application. Thus, it is advisable to evaluate satellite rainfall products with different types of hydrological models. This first evaluation over Moroccan basins suggests that SM2RAIN-ASCAT could be a reliable alternative to observed rainfall for hydrological modelling.
Funding Information
  • EUMETSAT H SAF (H_AVS20_02)

This publication has 77 references indexed in Scilit: