Accurate quantification of rainfall is essential for climate assessments and disaster risk reduction. This study compares the performance of seven precipitation products in rainfall estimation against a dense rain gauge network over Peninsular Malaysia during 2014–2019. Assessments were conducted at daily and seasonal timescales using both continuous and categorical error metrics to examine spatiotemporal accuracy. The ability of each product to represent rainfall climatology, including the annual cycle and interannual extremes, was also investigated. In point-to-grid comparisons, MSWEP and IMERG exhibit good point-scale accuracy, though substantial spatial inconsistencies persist across all products. IMERG and CMORPH are the most skillful in replicating daily rainfall dynamics based on the modified Kling–Gupta efficiency and demonstrate more balanced categorical performance in detecting rain events. However, weak correlation remains a key limitation across all products. Spatiotemporal analysis reveals higher skills in wetter regions but lower skills in drier regions across all timescales, suggesting a magnitude-dependent error structure. Most products fail to replicate seasonal shifts in rainfall distribution, especially for lower and moderate intensity classes. While rainfall climatology is misrepresented in terms of magnitude, temporal variability is reasonably tracked. Overall, IMERG emerges as the best-performing product. The collective findings shed light on the reliability of precipitation products in estimating tropical rainfall for water resources management purposes. Future work should quantify uncertainty propagation in rainfall–runoff processes and characterize cross-correlation structures between rainfall and atmospheric drivers.
Ng et al. (2026) studied this question.