Radar-Rainfall Estimation from S-band Radar and its Impact on the Runoff Simulation of a Heavy Rainfall Event in the Huaihe River Basin

抄録

 From July 1 to 13, 2007, a widespread heavy rainfall event occurred in the Huaihe River Basin (HB) in China, with an average rainfall of nearly 465 mm in the area. The main purpose of this study is to integrate a rainfall estimate by the China New Generation Weather Radar S-band radar (CINRAD-SB) into the Hydrologic Engineering Center’s Hydrologic Modeling System (HEC-HMS) and analyze the CINRAD-SB rainfall estimation and its impact on the runoff simulation of this type of rare flood event in a region with complex terrain.
 For the CINRAD-SB rainfall estimation four methods are considered: (1) Z = 300R^1.4 (Z: radar reflectivity, R: rainfall intensity); (2) a rainfall estimation error adjustment by using a Kalman Filter (KF); (3) Optimal Interpolation (OI); and (4) the Union method, which is composed of KF and OI. The HEC-HMS is used to investigate the spatial and temporal distribution of the CINRAD-SB rainfall and its impact on the hydrological simulation of the event.
 Rainfall estimations from the four methods are compared with rain gauge observations. The four methods underestimate the precipitation amounts, while for the Union method the values of the relative bias are closer to zero. The relative bias values of the four methods vary with different rainfall intensity, those of the Union method vary the least among the four methods. This evaluation indicates that runoff simulations based on radar-rainfall could reproduce similar overall patterns to the observed streamflow. The peak discharge contains obvious improvements - for instance, the skill score is 0.6 - in model runs with forcing that is provided by the Union method vs. rain gauge data. These results might guide the improvement of hydrological predictions that are driven by radar rainfall.