降雨作用下三峡库区秭归谭家湾滑坡监测预警研究

Monitoring and Early Warning of Tanjiawan Landslide in Zigui,Three Gorges Reservoir Region under Rainfall Effect

三峡库区秭归谭家湾滑坡是一座典型的具有“阶跃型”位移特征的降雨诱发型堆积层滑坡,它的变形破坏严重威胁着公路行车和周围居民的生命及财产安全。为研究谭家湾滑坡位移-降雨间的响应关系,首先利用皮尔逊相关分析法,探讨了谭家湾滑坡累计位移与累计降雨量之间的相关性,得到滑坡中前缘区域累计位移与累计降雨量的相关性最强,滑坡中后缘及左侧冲沟附近累计位移与累计降雨量的相关性次之;然后,定义了非相似度用于研究滑坡变形各个监测点与G11监测点位移序列间的相似性,从空间上将滑坡变形划分为3个主要变形区,并依据滑坡所处的变形状态,从时间上将2020年滑坡变形过程划分为3个变形时段;最后根据滑坡变形特征和各个变形区域的差异,确定降雨诱发型滑坡进入加速变形阶段的前期降雨量与当次降雨量阈值,得到了滑坡不同变形区域各个演化阶段的位移速率阈值,并结合现场巡查的滑坡宏观变形建立了基于滑坡位移-降雨响应的4级综合动态预警模型,再利用2021年谭家湾滑坡变形监测数据对该模型进行了验证。

In the Three Gorges Reservoir Region,Tanjiawan landslide is a typical accumulation landslide with rainfall-induced“step-type”displacement characteristics,which seriously threatens the safety of road traffic and the surrounding residents.This paper mainly investigates the rainfall-displacement response relationship and thus develops the monitoring and early warning model of the Tanjiawan landslide.Firstly,the correlation between the accumulated rainfall and the regional accumulated displacement of Tanjiawan landslide is analyzed by Pearson correlation analysis,and the result shows that there is a strong correlation between the middle front of the landslide and rainfall,with middle back and left gully second.And then,the paper defines the dissimilarity to study the relation between the displacement sequence of G11 monitoring point and others to divide the deformation area of the landslide into three parts spatially,and also partitions the deformation progress into three stage by the deformation state of each area in terms of time.Finally,the deformation characteristics of Tanjiawan landslide and the difference of each area are dedicated to determine the initial rainfall and current rainfall thresholds for the accelerated deformation stage of the rainfall-induced landslide,which yields the displacement rate threshold of each evolution stage in different deformation area of landslide.Besides,a 4-Level comprehensive dynamic early warning model is developed by the on-site inspection of macro-deformation and later the verification of the model was accomplished by the monitoring data of 2020.