Analysis Of Cause And Sliding Distance On Loess Landslide In Heifangtai Area

Lots of loess landslides have developed in Heifangtai of Gansu Province,which have seriously affected local people’s normal production, life and local economy as for high frequency and destructive.Detailed causes analysis of loess landslide and the sliding distance could ascertain the endanger scope and put forward suggestion for the prevention.According to the landslide disasters around the HeiFangtai platform as the research object, Based on the analysis of geological environment, and by means of field geological survey, measuring, spot and laboratory test, do a systematic and in-depth research the cause of landslide and sliding distance.The main conclusions are as follows:Firstly, loess landslide and loess-bedrock landslide developed around the platform, which account for60%and40%respectively, and majority are large and medium-sized ones. In addition, the amount of loess landslide in the east are more than that in the west.Secondly, the physico-mechanical propertise of the different rock-soil are atypism.The cohesion of the bottom silty clay is significant stronger than the upper loess and silt. Moreover, with the increase of depth, the density and shear strength tend to increase, permeability tends to decrease. The weathering of cretaceous mudstone, which is in the lower pebble bed is serious. Under the effect of irrigation water, combining with the physical property, slope structure, gradient, slope height, aspect and the bedrock attitude of rock-earth masses, they provide the favorable conditions for the happening of landslide.The emergence of spring is related to the happening of landslide. There are intensive springs in the eastern of platform,so it developed many landslides. And the more springs emerge, the more scale of landslide is. The water quality of the irrigation water is evidently different from that of the water exudated from slope mass, it reflects the interaction between the irrigation water and rock with the process of infiltration, which can influence the form of landslide.Thirdly, applying the method of single factor analysis and the effect conjecture can find out the major factors affecting the sliding distance, Applying Support Vector Machine to the analysis of sliding distance, as well as with the predictive results of Multiple Linear Regression and Fuzzy Inference System, consider that Support Vector Machine could acquire the well predictive effect in the analysis of sliding distance.Finally, according to the reason of loess landslide’s formation, together with the distribution of landslide which has great damage and the predicted results of sliding distance, we propose it should prevent the happening of the loess landslide along the roads as soon as possible, according to the predictive extent of the sliding distance, after that, it can reduce the loss as the result of disaster.