Rainfall-runoff simulation and flood dynamic monitoring based on CHIRPS and MODIS-ET. Issue 11 (2nd June 2020)

• Record Type: Journal Article
• Authors: Yu, Chen; Hu, Deyong; Duan, Xin; Zhang, Yani; Liu, Manqing; Wang, Shasha
• Is Part Of: International journal of remote sensing. Volume 41:Issue 11(2020)
• Page Start: 4206
• Page End: 4225
• Publication Date: 2020-06-02
• Subjects: Remote sensing -- Periodicals; Télédétection -- Périodiques; 621.3678
• DOI: 10.1080/01431161.2020.1714779
• Languages: English
• ISSNs: 0143-1161

Abstract

ABSTRACT: In this study, satellite meteorological products were used as important data sources to achieve rainfall-runoff simulation and flood dynamic monitoring in the Songhua River basin (below the Sancha estuary) in northeastern China. Satellite precipitation products CHIRPS (the Climate Hazards Group Infrared Precipitation with Stations) and satellite evapotranspiration products MODIS-ET (the Moderate Resolution Imaging Spectroradiometer-Evapotranspiration) were collected for the study area during 2010–2015. The accuracy of satellite meteorological products was validated by ground observation data in the same period. Then, combined with satellite meteorological products, a daily-scale rainfall-runoff model suitable for the study area was established. The daily river runoff (RR) and surface runoff depth (SRD) of the study area were obtained through this model. It is found that the overall performance of satellite meteorological products is good through verification. The annual average deviation of MODIS-ET and observation data is less than 1 mm. In tertiary basins of the study area, the correlation coefficient (CC) between CHIRPS and observation data is in the range from 0.59 to 0.71, and the root-mean-square error (RMSE) varies from 3.75 to 4.50 mm. The rainfall-runoff model based on satellite meteorological products can well simulate the variation characteristics of observed RR: the CC reach 0.83, and the Nash–Sutcliffe efficiency coefficient (NSE) reach 0.72. The established model was used to analyse a flood event in 2013. And the simulation results showed the changing process of the RR accurately. The dynamic changing process of the SRD was monitored by the established model during the flooding period. The model can be used to reflect the spatial distribution of potential inundation areas. This study validates the performance and deviation of CHIRPS and MODIS-ET, and proves that these satellite meteorological products have good applicability in rainfall-runoff simulation and flood dynamic monitoring. The output results, daily RR and SRD data with 0.05 ° grid unit, can provide the decision support for hydrological information monitoring and flood forecasting.