Evaluation of Accuracy and Streamflow Simulation of TRMM Satellite Precipitation Data

2016 
The accuracy of satellite precipitation estimates has become a focus of attention, having a promising prospect in the hydrological researches and applications. In this study, the new precipitation product (3B42 V7) of Tropical Rainfall Measuring Mission (TRMM) was evaluated via comparison with the rain gauge precipitation data in Xiangjiang River Basin. The evaluations were conducted at different spatial and temporal scales. Simultaneously, the laws of data accuracy under different periods, rainfall intensity levels, and elevations were analyzed. These two precipitation data were then adopted to drive the Xinanjiang Model to perform streamflow simulation. The results showed that TRMM satellite precipitation data had a good correlation with rain gauge precipitation data; it had better accuracy under monthly scale than daily scale and showed better performance at watershed scale than it did at grid scale. The area precipitation of TRMM data showed better accuracy in humid season than in arid season. From an overall estimate on the amount of precipitation, areal precipitation from TRMM data on watershed scale was underestimated in contrast with areal precipitation interpolated from rain gauge data. While at grid scale, TRMM data from higher elevations inclined toward underestimation, while the contrary inclined toward overestimation. The frequency statistical result of different rainfall intensity levels indicated that TRMM tended to underestimate the occurrence of light rainfall intensity but overestimate the occurrence of heavy rainfall intensity and no-rain events. The TRMM-driven streamflow simulation performed better at monthly scale relative to the daily scale, but all were within acceptable range. However, the effects of streamflow simulation driven by TRMM data were not as accurate as those driven by rain gauge data displayed in the aspects of flood peak, flood volume and flood duration. This study assesses the accuracy of the latest product TRMM 3B42 V7, revealing the influential factors regarding accuracy for future reference and analyzing the effectiveness and possibility of streamflow simulation.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    17
    References
    3
    Citations
    NaN
    KQI
    []