Assessing the performance of near real-time rainfall products to represent spatiotemporal characteristics of extreme events: case study of a subtropical catchment in south-eastern Brazil

ABSTRACT

This study evaluates the performance of four Near Real-Time (NRT) satellite rainfall products in estimating the spatiotemporal characteristics of different extreme rainfall events in a subtropical catchment in south-eastern Brazil. The Climate Prediction Centre Morphing algorithm (CMORPH), Tropical Rainfall Measuring Mission, Multisatellite Precipitation Analysis in real time (TMPA-RT), the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Global Cloud Classification System (PERSIANN-GCCS), and the Hydro-Estimator are evaluated for monsoon seasons, based on their capability to represent four types of rainfall events distinguished for: (1) local and short duration, (2) long-lasting event, (3) short and spatial extent, and (4) spatial extent and long lasting. Since the events are defined relative to a percentile, the relative performance variation at different threshold levels (75th, 90th, and 95th) is also evaluated. The data from the 13 Automatic Weather Stations (AWSs) for the period from 2007 to 2014 are used as the reference. The results show that the product performance highly depends on the spatiotemporal characteristics of rainfall events. All four products tend to overestimate intense rainfall in the study area, especially in high altitude zones. CMORPH had the best overall performance to estimate different types of extreme spatiotemporal events. The results allow for developing a better understanding of the accuracy of the NRT products for the estimation of different types of rainfall events.

Acknowledgments

This work is part of a PhD study of the first author, patially funded by the Colombian Administrative department of Science, Technology and Innovation (COLCIENCIAS), under Grand number 646. The authors would like to acknowledge Ms Maria Clara Fava from the University of São Paulo and Ms Denise Silva and Mr Ricardo Aguilera from Agronomic Intitute CIIAGRO – FUNDAG São Paulo State Government for providing the hourly data from the automatic weather stations and the agencies responsible for satellite database used in this research.

Disclosure statement

No potential conflict of interest was reported by the authors.

Author contributions

M.L. and G.C. designed the experiments. M.L. extracted and processed the information. M.L. G.C., D.S., and J.D. analysed the information. M.L., G.C., D.S., and J.D. wrote the article.

Additional information

Funding

This work is part of a PhD study of the first author, patially funded by the Colombian Administrative department of Science, Technology and Innovation (COLCIENCIAS), under [Grant number 646].