Slope displacement monitoring and prediction in the Qinghai-Tibet plateau based on AdaBoost algorithm

Abstract Due to the unique climatic conditions in the Qinghai-Tibet Plateau region, the relationship between rainfall and landslides is different from other regions. In order to explore the characteristics of slope displacement in Chengduo County, which is located in the Qinghai Tibet Plateau region, a new idea based on AdaBoost algorithm was adopted to analyze the relationship among slope cumulative displacement, rainfall and soil moisture content at different depths, and to predict slope displacement. The results show that the overall cumulative displacement of the monitoring area shows an increasing trend. Due to the melting of surface snow and ice in February and March each year, the soil moisture content at different depths is not directly related to rainfall. The cumulative displacement of slopes is directly affected by soil moisture content and shows a positive correlation, but is weakly affected by daily rainfall. The AdaBoost regression model trained on soil moisture content at different depths underground has achieved good results in predicting the cumulative displacement of slopes, and can predict the cumulative displacement for a period of time in the future. The research results can provide reference for slope displacement prediction and safety assessment in the Chengduo County area.