|
|
Assessment of accuracy and runoff simulation of Stage IV, GPM, TRMM precipitation products in small watershed |
FENG Kepeng1, 2, 3,TIAN Juncang1, 2, 3*,HONG Yang4, TANG Guoqiang5,KAN Guangyuan5,LUO Xiangyu4 |
1. School of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, Ningxia 750021, China; 2. Ningxia Research Center of Technology on Water-saving Irrigation and Water Resources Regulation, Yinchuan, Ningxia 750021, China; 3. Engineering Research Center for Efficient Utilization of Water Resources in Modern Agriculture in Arid Regions, Yinchuan, Ningxia 750021, China; 4. School of Civil Engineering and Environmental Science, University of Oklahoma, Norman, OK 73072, USA; 5. State Key Laboratory of Hydroscience and Engineering, Tsinghua University, Beijing 100084, China |
|
|
Guide |
|
Abstract The accuracy of TRMM 3B42 V7, GPM-IMERG satellite precipitation, and Stage IV radar precipitation was compared and analyzed in the Little Washita River Basin, Oklahoma, the USA, using ground observation precipitation reference. The CREST, a distributed hydrological model, was driven by these three precipitation products and evaluated its hydrological simulation utility. The research shows that the correlation coefficient between the three kinds of precipitation products and ground precipitation is more significant than 0.5, passed the 0.05 confidence level test. GPM-IMERG and TRMM3B42 V7 precipitation have the underestimation of no rain and light rain, and the overestimation of heavy rain and violent rain; in general, Stage IV has the best precision, next, GPM precipitation accuracy is better than TRMM 3B42 V7. In the hydrological simulation utility evaluation, the same ca-libration period was set and the CREST model parameters were calibrated using three kinds of precipitation products. After the set of parameters is obtained, the basin's daily runoff process is simulated in the same validation period. The comparison results show that Stage IV radar precipitation has the best effect in small watershed hydrological simulation, followed by GPM-IMERG, and TRMM 3B42 V7 simulation effect is not ideal.
|
Received: 06 July 2019
|
Fund: |
|
|
|
|
|
|