Flood Response To The Spatial Variability Of Storm In The Jinjiang Xixi Watershed

Rainfall is the primary input to most hydrological systems, and its change characteristics such as amount and temporal and spatial distribution would directly attribute to compositional diversities and complexities of runoff. The temporal and spatial variation of rainfall is an important factor affecting flood generation, and it is meaningful to study the flood response to temporal and spatial variation of rainfall. Jinjiang watershed lying in southeast of Fujian Province is one of the most developed regions in Fujian Province, where convective rainstorm and typhoon storm occur frequently in summer, and the loss caused by the flood is great. The flood response of Jinjiang watershed to temporal and spatial variation of rainfall is representative in southeast coast regions of China, and it is significant and imperative to study the influences of the spatial and temporal variability of storm on flood in Jinjiang xixi watershed. The study would provide reference value to flood forecasting, water resource management, and soil and water conservation.Xixi basin was selected as the study area, and qualitative and quantitative methods were combined to discuss the hydrological response to the spatial variability of storm utilizing GIS, a multiple linear regression model, and storm-flood model. Based on the30observed rainstorm flood events on the hourly scale occurring in1970s, several storm indices describing the spatial and temporal heterogeneity were calculated, and the relationship between these storm indices and flood peak and flood volume was established by a multiple linear regression analysis. Moreover, a spatially distributed hydrologic model, HEC-HMS, was calibrated and validated to quantitatively discover flood response to the spatial variability of storm. The results indicated that:①If the indices which describing spatial and temporal variation of storm events was appropriate selected, a better empirical relationship between flow and the storm characteristics could be established.②The total storm volume, storm centroid distance, temporal variation coefficient of storm and storm duration seemed to have significant influence on peak flood.③Storm centroid distance was a good index to investigate the integrated response of the flood peak to spatial variability of the storm and characteristics of runoff generation and convergence.④Moreover, temporal variation coefficient of storm and storm duration had a common affect on peak flow.⑤The relationship between flood volume and total areal storm volume and the initial discharge was very significant, and the flood volume increased with larger storm volume and higher initial discharge.⑥The simulation of hydrological model show that:spatial variability of storm seemed to have significant influence on peak flood, and the influence increased with the increase of flood return period.⑦A100-year flood may change into a200-year flood due to the spatial variability of storm.