Study On The GMD Early Warning And Forecasting Model Of The Kualiangzi Landslide

In China, a type of landslide which has characteristic of near horizontal strata is widely developed, this type of landslide has controversies in formation mechanism and disaster mode. Due to the dip of rock usually vary from 2 degree to 5 degree and the slope surface is generally flat, so the reason of this type of landslide is difficult to be explained. Therefore it is difficult to reduce the disaster of this type of the landslide, the early warning and forecasting work is also more difficult. The Kualiangzi landslide is a typical near horizontal layered landslide in red rock area of eastern Sichuan, and is located in Zhongjiang County, Deyang, Sichuan. Historically, the landslide underwent two large-scale movements successively in flood season of 1949 and 1989, which caused varying degrees of damage to the residents. However, according to real-time monitoring found that the landslide is still in a state of deformation, especially in rainy season. The displacement rate of the landslide can reach 25 mm/d. Therefore, it is very important to carry out the early warning and forecasting work of the landslide.(1) This article used the engineering geological investigation, the test of ERT and the groundwater tracer test to get the engineering geological conditions, the material composition characteristics and the hydrogeology features of the landslide. According to the result of ERT, the resistivity of sliding zone is low perennially, this shows the sliding zone is full of water all the time which reduced the mechanical strength of sliding zone. According to the result of hydrogeological test, the migration of the groundwater is in the fracture which direction is nearly along the north and south direction, the average migration rate of the groundwater is 0.75 mm/h. The groundwater is also migrating to the front along the sliding zone, the average migration rate of the groundwater is 3.3 m/h. The maximum migration rate of the groundwater in parts of sliding zone is 13 m/h.(2) Through the monitoring data of rainfall, displacement, groundwater from 2013 to 2015.(1) It is founded that the precipitation of the landslide changed very little every flood season, however, according to the division rule of rainfall intensity which was promulgated by weather bureau, heavy rain accounted for the main part in flood season of 2013,moderate rain and light rain accounted for the main part in flood season of 2014 and heavy rain accounted for the main part in flood season of 2015.(2) The cumulative displacement of the landslide is 1000 mm from 2013 to 2015, there was a significant linear positive correlation between the cumulative amount of rainfall and the amount of accumulated displacement, through the analysis of curve regression obtained the fitted equation: 0.63 84.2.(3) The groundwater variation had the spatial asymmetry for the response of rainfall, the front of landslide with a thin layer of clay and gravelly soil is strong to the response of rainfall. Due to the thick sliding body, for the middle and back parts of the landslide, the rainfall recharge amount is few. Through the typical rainfall events from 2013 to 2015, the average infiltration coefficient at the front of the landslide is 0.27. Through fitting analysis,the minimum accumulated rainfall of the groundwater fluctuation is 12 mm.(3) According to using Seep/W module in Geo-studio to simulating the landslide seepage field, it is founded that the Kualiangzi landslide seepage field is steady in non-flood season which is consistent with the change of monitoring groundwater level. The seepage field of the slope deposit on the front of the landslide and the colluvial sediments in the back tension crack changed significantly under heavy rainfall. According to simulating the typical rainfall event in 2013,it is founded that the water level in the tension crack changed dramatically, which impacted the landslide seepage field. By using limit equilibrium theory and monitoring data to analysis the stability variation of the landslide.The early warning level of the Kualiangzi landslide was divided into four levels: I level, Fs?1.0, the slope is in high risk;? level, 1.0< Fs?1.05, the slope is in moderate risk; ? level, 1.05< Fs?1.1, the slope is in low risk; ? level, Fs >1.1, the slope is safe.Through the regression analysis, both the water level in the tension crack(Hw) and the daily displacement rate(v) demonstrate a good negative correlation with the slope stability(Fs), through the analysis of curve regression obtained the two fitted equation:FS?1.195-0.013 HW FS?1.082-0.008v(4) The water level in the tension crack was determined as a warning indicator through the monitoring data and stability analysis.By using UDEC to simulating the deformation condition of the landslide under the different water level in the tension crack,which was used to build the GMD early warning model of the landslide.It is founded that the numerical model can converge when the water level in the tension crack was 7m,however,when the water level in the tension crack was higher than 7m, the numerical model would be in a state of continuous deformation. With the increase of the water level in the tension crack, the displacement and the displacement rate of the numerical model were also increased. A good rule of quadratic function is obtained between the growth multiple of deformation rate of the GMD model and the water level in the tension crack. Through the analysis of curve regression obtained the fitted equation: nv=0.085Hw2-0.65 Hw. Eventually, we can use this early warning model and the result of the analysis of the stability to forecasting the deformation and stability of the Kualiangzi landslide.