Extensive evaluation of IMERG precipitation for both liquid and solid in Yellow River source region

The Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG) produces high-resolution precipitation estimates and serves as an invaluable data source data-scarce ungauged basins. This study comprehensively evaluates two near-real time products (IMERG-E IMERG-L) a post-real dataset (IMERG-F) of IMERG V06B from 2014 to 2019 on basin grid scales in the Yellow River region, data-sparse northeastern Tibetan Plateau. Results show that three can well depict spatial distribution area. Compared with IMERG-E IMERG-L, gauge-adjusted IMERG-F shows better agreement event detection both scale, yet is limited improving accuracy. products, underestimate across (~−37.82% ~−38.55%, respectively). At same time, product exhibits negative errors drier areas positive wetter regions are roughly separated by basin's 3900 m contour. correlation, error metrics, detecting ability upgrade northwest southeast climate becomes wet. Temporal properties exhibit dependence meteorological conditions. Specifically, performance wet season superior dry season. For instance, normalized root mean square was reduced 55.10% products' capability detect affected temporal intensity, phase. reliability these satellite increases when interval extends daily monthly. estimate moderate precipitation, especially range 5–10 mm/d. generally perform worse solid than liquid one. However, interactions phases intensities lead decrease accuracy light (<5 mm/d) but increase events (≥5 phase transforms liquid. Uncertainty reference results underestimation quality. These findings provide useful application cold mountainous reveal needs exhaustive calibration, precipitation.