科研之友
微信
新浪微博
Facebook
分享链接
摘要
High-resolution satellite precipitation products(SPPs)provide forcing inputs for hydrologic applications. Complex mountain terrains have a significant effect on the occurrence and intensity of precipitation. In this study, ground-based observations were adopted as the benchmark to evaluate accuracy and precipitation detection capability of three SPPs(bias-corrected Climate Prediction Center morphing method, CMORPH CRT;Tropical Rainfall Measuring Mission, TRMM 3B42V7;and Integrated Multi-satellite Retrievals for Global Precipitation Measure, GPM IMERG)was validated by 104 rain gauges from 1 January 2016 to 31 December 2017 in different time scales(1 h, 3 h, 6 h, 12 h and 24 h)based on different precipitation intensity(light rain, moderate rain, heavy rain, very heavy rain and extreme rain)over the Taihang Mountains. Eight statistical metrics were utilized to quantitative analysis in this research. These indices include Correlation Coefficient(R), Root Mean Square Error(RMSE), Bias ratio(β), Variability ratio(γ), Kling-Gupta efficiency(KGE'), Probability of Detection(POD), False Alarm Ratio(FAR)and Critical Success Index(CSI). Results show that:(1)The accuracy performance of each SPP gradually stabilizes after the precipitation accumulate to 3 h, and the KGE' basically remains unchanged after 3 h time scale. In terms of the precipitation detection performance of SPPs, POD of three SPPs increase with the accumulation of precipitation time(POD range from 0.5 to 0.8), and FAR of CRT and 3B42 products decrease with the accumulation of precipitation time(FAR range from 0.5 to 0.75).(2)In terms of the accuracy analysis of different precipitation intensity ranges, there is a weak correlation between precipitation observations by SPPs and precipitation observations on the ground(R<0.4). RMSE increase with the increase of precipitation intensity for all SPPs, POD of all SPPs decrease with the increase of precipitation intensity over the Taihang Mountains, ranging from 0.4 to 0.65, and FAR of all SPPs increases with the increase of precipitation intensity, ranging from 0.6 to 0.85.(3)In terms of Probability density distribution(PDF), IMERG product underestimates the rain event of P<0.1 mm in all time scales. To varying degree, IMERG product overestimates the rain event of P≥0.1 mm over the Taihang Mountain. CRT product shows an overestimation of heavy rain and very heavy rain events at 1 h, 6 h, 12 h and 24 h. The PDF of 3B42 product in different precipitation intensity ranges are closer to the actual observed precipitation on the ground in study areas, but the observation of moderate rain events is underestimated by this product in all time scales. Our results not only demonstrate the superiority of different products at different time scales and precipitation intensity ranges, but also provide suggestions for further improvement of the SPPs especially for complex terrains.
- 单位
