Prediction method of debris flow by logistic model with two types of rainfall: a case study in the Sichuan, China

Debris flow is a serious disaster that frequently happens in mountainous area. This study presents an effective method for forecasting it by rainfall, which is one of the important components for prediction. The Sichuan Province is taken as an example. The geographic information system (GIS) is chosen as a tool to estimate the precipitation of hazard point, and use of statistical technique is made to calculate attenuation coefficient of effective antecedent precipitation. With such methodologies, the logistic regression model is used to comparatively establish the prediction model of two forms rainfall combination: (1) intraday rainfall and 10-day previous rainfall, (2) intraday rainfall and two types of effective antecedent rainfall which are short-time-heavy rainfall and long-time-light rainfall. The results indicate that the location of debris flows and the distribution of rainfall are factors interrelated. Secondly, the contribution rate of intraday rainfall is the highest. Thirdly, the second form rainfall combination has a higher prediction accuracy, 2.3% for short-time-heavy rainfall and 2.1% for long-time-light rainfall, which suggests that a moderate improvement is achieved by the rainfall classification.