Research On Early Warning And Forecast Method Of Rainfall Shallow Landslide

Rainfall shallow landslide was one of the main types of geological disasters in China. It was simultaneity, bulk, harmful and explosive. Researchments of monomer landslide on risk assessment and early warning had been made currently, but the development of examination on regional early warning was slow. And landslide was a kind of secondary disasters which happened frequently, much attention had been paid owing to its huge disaster strength. Therefore, it had an important practical significance to study the rainfall shallow landslide warning in Sichuan province where has a major earthquake and seek out an early warning method which was scientific, integral and practical. Effective technical support had been provided for early warning methods by the development of GIS and RS.Sichuan Province was selected as the study area where was called "Land of Abundance". Based on rich data material, various factors that affected shallow landslide were studyed, distribution characteristics of shallow landslide and the correlation of factors were analyzed, and the geographic information database of shallow landslide of Sichuan Province was set up. Explored the critical value of rainfall that induced the shallow landslide, combined with TRMM data, the accuracy of rainfall threshold was lifted. The model of rainfall shallow landslide early warning was established preliminary. Based on the data of2009of study area, applicated the model, results showed that the model was actual and practicl. As the5.12earthquake-stricken areas, secondary disasters were more likely to happen in Sichuan. This paper had practical significance to disaster relief and reduce losses to some extent. The main contents and conclusions were as follows:1. This paper based on DEM data, remote sensing data and geological data, set up the geographic information database of shallow landslide of the study area which included landslide distribution database, impact factors database and the actual landslide catalog. It provided basic data material for the construction of shallow landslide warning system.2. The problem of overlay and operate different data layer was solved by using GIS software, based on logistic regression model, with the help of SPSS software, logical regression coefficient of factors which impact shallow landslide were confirmed, and the spatial sensitivit index of shallow landslide of Sichuan Province was calculated. The problem of data combinatio and weights in analysis of shallow landslide space sensitivity was solved, improved the accurac of the spatial sensitivity index. Through the classification of the sensitivity index, realized th sensitivity of shallow landslide space of quantitative evaluation.3. The I-D model introduced early warning system, TRMM as the basic data, the model o early warning of rainfall shallow landslide was set up on the basis of the study area shallow landslide spatial sensitivity analysis and shallow landslide rainfall threshold. The study propose to determine the correlation-induced shallow landslide occurred between the duration of rainfal and maximum rainfall intensity rainfall events experienced in the past shallow landslide actually occurred based on the use of the experience threshold. The model of rainfall shallov landslide early warning was established preliminary.Through the shallow landslide space sensitivity analysis, to make up for in the experience o regional rainfall critical threshold value as a warning of the shortcomings of the shallov landslide bring, set up sichuan province rain type landslide warning forecast system shallow, and obtained a certain result. It had guiding significance for the forecasting work and early warning of secondary disasters in Sichuan Province.