Intercomparison of improved satellite rainfall estimation with CHIRPS gridded product and rain gauge data over Venezuela

Abstract

Satellite-derived rainfall products are useful for both drought and environmental monitoring, and they also allow for tackling the problem of sparse, unevenly distributed and erratic rain gauge observations provided their accuracy is well known. Venezuela is a country highly vulnerable to extreme weather events such as extensive droughts and flash floods; therefore, an understanding of the strengths and weaknesses of satellite-based rainfall products is useful for the planning of water resources. Using numerical metrics in order to evaluate performance, monthly rainfall estimates, from the Climate Hazards Group InfraRed Precipitation and Stations (CHIRPS v.2) product, are compared to gauge data from the 1981-2007 interval and categorical metrics for assessing rain-detection skills. The analysis was performed considering different rainfall categories, seasonality, and spatial context. The results show that the satellite product CHIRPS v.2 overestimates (underestimates) low (high) monthly rainfall values; although on the majority of numerical metrics of skill shows a good performance. This product, on the other hand, achieves better performance during the rainy season (April-September), significantly overestimating, however, the rainfall-events frequency. The product also shows best overall performance over flat and open regions (e.g., Los Llanos), where precipitation is influenced by the Intertropical Convergence Zone activity and local convective systems.