Performance of five high resolution satellite-based precipitation products in arid region of Egypt: an evaluation

Abstract

High-resolution daily precipitation estimation is very important in climatological and meteorological studies in the arid regions of the world, as precipitation events in these areas can be sporadic, localized and of very high intensity. In this study, the daily performance of five, ARC2, CHIRPS v2.0, GSMaP (v. 6), TAMSAT (v. 3) and PERSIANN-CCS high resolution satellite-based gauge-corrected precipitation products were compared, and the individual performances validated against rain gauge station records in the arid region of Egypt. Seven statistical metrics (three continuous and four categorical), and selected intensity categories, were employed in the modelling of rainfall totals for the 2003 to 2018 period. In general, the results indicated poor outcomes for all the satellite-based products. CHIRPS was best at estimating rainfall of <1 mm/day; this represented 30% of wet days during the study period. ARC and GSMaP performed better in estimating rainfall events with an intensity category of ≥1 mm/day, however both produced a high number of false detections. Despite continuous improvement of TAMSAT, it recorded the worst performance among the products evaluated. The study concluded that GSMaP appeared to be the “best” to use for supporting research activities over the arid Egyptian domain given its performance relative to the other satellite-based precipitation products.