Evaluation of hourly summer precipitation products over the Tibetan Plateau: A comparative analysis of IMERG, CMORPH, and TPHiPr

Abstract

The Tibetan Plateau (TP) plays a crucial role in regional and global climate dynamics, making accurate precipitation data essential for meteorological studies. However, owing to the complex terrain and sparse observations, precipitation data are inadequate, particularly at the hourly scale, which hinders precise climate modeling and forecasting. To address this, we evaluate three precipitation products—IMERG, CMORPH, and TPHiPr—against observations from 84 rain gauges during summer from 2007 to 2018. Using traditional evaluation metrics and event-based error decomposition, we quantify each error component's contribution to the total bias. Our results show that at the hourly scale, IMERG outperforms CMORPH and TPHiPr, exhibiting the highest correlation with rain gauge data (CC = 0.43) and strong detection ability (POD = 0.58), although TPHiPr performs better at the daily scale. IMERG and CMORPH capture the diurnal cycle of precipitation frequency, but all three products significantly overestimate the precipitation frequency at noon. Additionally, all datasets tend to overestimate light rain and underestimate heavy rain, although IMERG and CMORPH demonstrate stronger detection of heavy rain (>2.6 mm/h) than light rain (0.1–0.2 mm/h). Notably, IMERG tends to detect precipitation events that start and end earlier than observed, with errors in the 0.1–0.2 mm/h intensity range contributing 31.8 % to the total bias. While IMERG performs better than CMORPH and TPHiPr at the hourly scale over the TP, challenges remain in detecting weak precipitation. By highlighting the limitations and strengths of these products, our study provides valuable insights to improve satellite-based precipitation estimates and support better climate modeling and forecasting in this vital region.