Search

Summary: The objective of this study is to quantitatively evaluate the successive Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA) products and further to explore the improvements and error propagation of the latest 3B42V7 algorithm relative to its predecessor 3B42V6 using the Coupled Routing and Excess Storage (CREST) hydrologic model in the mountainous Wangchu Basin of Bhutan. First, the comparison to a decade-long (2001–2010) daily rain gauge dataset reveals that: (1) 3B42V7 generally improves upon 3B42V6’s underestimation both for the whole basin (bias from −41.15% to −8.38%) and for a 0.25°×0.25° grid cell with high-density gauges (bias from −40.25% to 0.04%), though with modest enhancement of correlation coefficients (CC) (from 0.36 to 0.40 for basin-wide and from 0.37 to 0.41 for grid); and (2) 3B42V7 also improves its occurrence frequency across the rain intensity spectrum. Using the CREST model that has been calibrated with rain gauge inputs, the 3B42V6-based simulation shows limited hydrologic prediction NSCE skill (0.23 in daily scale and 0.25 in monthly scale) while 3B42V7 performs fairly well (0.66 in daily scale and 0.77 in monthly scale), a comparable skill score with the gauge rainfall simulations. After recalibrating the model with the respective TMPA data, significant improvements are observed for 3B42V6 across all categories, but not as much enhancement for the already-well-performing 3B42V7 except for a reduction in bias (from −26.98% to −4.81%). In summary, the latest 3B42V7 algorithm reveals a significant upgrade from 3B42V6 both in precipitation accuracy (i.e., correcting the underestimation) thus improving its potential hydrological utility. Forcing the model with 3B42V7 rainfall yields comparable skill scores with in situ gauges even without recalibration of the hydrological model by the satellite precipitation, a compensating approach often used but not favored by the hydrology community, particularly in ungauged basins. [ABSTRACT FROM AUTHOR]