Abstract: Background, aim, and scope Climate variability is shown to be an important driver of spatial and temporal changes in hydro-meteorological system. And the influence of climate variability on flood damage has received more attention, especially for larger rivers, in China. The improvement of small and medium-sized river is still an especially important topic in the nation. Flood-prevention project have already implemented in the main stream and their tributaries for the Yellow River. However, extreme flood only occurred in small rivers with less convergence time and lack of flood control works. Upper and middle reaches of the Weihe River is mainly loess hilly-gully region, this can result in large economic losses when the storm has happened. Critical areal rainfall is a crucial indicator for the flood monitoring and forecasting, and also can benefit for river management. Materials and methods Calculation critical areal rainfall of flood are evaluated based on the hourly observed data of hydro-climatic time series for the period of 2006—2013, by using Thiessen Polygons method, power function with three parameters, non-liner regression model and probability distribution simulation, in a typical tributary of the Weihe River Basin. Results The results showed that power function with three parameters can simulate the statistic relationship between water level and discharge. The model can also display strong performance in well demonstrate the process of large typical flood in the Qianhe River. Cumulative precipitation before nine hours has great impact on discharge of monitoring cross section. Critical storm duration rainfall for Qianyang hydrological station increased up to 50 mm when the water level rise from 901m to flood exceeding the designed elevation, and the peak discharge also raised by 1418.51 m3/s. Log-Pearson 3 revealed high performance in simulate probability distribution of peak discharge in the study region. The corresponding critical areal rainfall for different return periods were computed, and water level raised more than 2 m when return period change from 10 years to 100 years. Discussion The power function with three parameters and non-liner regression model showed well performance in this study, however, advanced verification also should be taken when these methods used in other different weather conditions and complex terrains. Cumulative nine hours rainfall corresponding to the peak discharge was adopted in the article, and this time period should be dynamic change under local geographic conditions. Conclusions Disaster early warning information should be given to the public when the cumulative rainfall rose to 16 mm and the corresponding water level for the control hydrological station is 901 mm. Recommendations and perspectives Good knowledge of critical storm duration rainfall under the changing climate can provide great scientific and practical merits for flood simulation and forecast, and also benefit to in the water resource management in the basin scale for the government. More importantly, it can prevent disaster caused great damage to the society.
Keywords: flood; critical storm duration rainfall; stage-discharge relation curve; Weihe River Basin