Evaluation of satellite based Global Precipitation Measurement (GPM) over a tropical urban region

Abstract

Satellite precipitation products are improving rapidly, and are becoming an irreplaceable part of our everyday life including for use by the research community. For use by the latter, these products for example, are able to estimate precipitation in areas where ground measurements are impossible, such as the ocean and mountainous areas. They are used to forecast weather, and aid in predicting natural disasters such as floods. A critical evaluation of a satellite precipitation product such as the Global Precipitation Mission (GPM) Integrated Multi-Satellite Retrievals (IMERG) is essential for the end users and product developers. This project provides an assessment of IMERG rainfall estimates over Singapore at a daily scale for the period of 22 months from April 2014 to January 2016. The IMERG data is compared with local gauge-based observations averaged for each of four IMERG Grids. Several continuous and binary metrics are employed to evaluate the IMERG estimates and ground measurements on annual and monthly basis. Both annual and monthly results show that IMERG data show reasonable capabilities in estimating the occurrence of rainfall events, however it has difficulties in estimating the precipitation amount. In addition, IMERG tend to overestimate the amount of rainfall in the Eastern part of Singapore.