Analysis of Critical Rainfall of Flood Disaster in Urumqi River Basin
The critical rainfall is an important index of the mountain torrents disaster forecast, which plays an important role in flood prevention. It determined the critical rainfall by different calculation methods, based on the presence or absence of hydrological and precipitation data By the XIAO-Quzi, Da-Xigou, BAI-Yanggou stations from January 1st, 2006 to December 31st, 2010. The HBV hydrological model and statistical model were used to calculate the critical rainfall caused disaster at Hero bridge station in Urumqi River Basin .Model and calculate process were analyzed and proven by the historical process of rainfall flood. HBV hydrological model set the rainfall runoff and snowmelt runoff simulation in one, through the generalized to consider the differences in elevation, vegetation and other surfaces to carry out the simulation of rainfall runoff. It has the reference significance to northwest rivers runoff snowmelt flood recharge simulation. Level 2nd and level 3rd disaster critical rainfall calculated by HBV hydrological model were 40.6 mm and 25.9 mm, which obtained from statistics model were 47.5 mm and 26.9 mm. Observed from the numerical value were reasonable and practical possible. To the fitting effect, correlation coefficient between the flow simulated by HBV model and actual flow simulation was 0.87. The multiple correlation coefficients between the statistical model and actual was 0.53. Contrasted with the correlation of simulation results, HBV model was better than statistical model. Contrasted with the results, the critical rainfall calculated by the HBV model and the statistical model is less than the actual minimum rainfall value of the flood process. By comparison, the critical rainfall calculated by the statistical model was more close to the real minimum rainfall value of flood process. In the calculated process, change trend of flow can be fitted well, but the peak value is always less than the actual value of the flow. This should be the main reason which critical rainfall final calculation is less than the actual minimum rainfall value of the flood process.