降雨型滑坡的机理及其启示

Mechanism of Rainfall-Induced Landslides and Its Implications to Landslide Prediction

按照有效应力原理,降雨引发的规模较小的浅层滑坡,主要受岩土体的内聚力控制,而规模较大的深层滑坡则主要受摩擦力控制。对于大多数的降雨型滑坡易发地区,就滑坡预报而言,预报日前期的降雨量和土体含水量观测是非常重要的。影响滑坡的临界孔隙压力的因素在空间上是变化的。因此,在降雨条件相同时,在不同地点降雨引发滑坡的概率不同。根据目前物理学对颗粒物质的认识,可以推测:降雨引发的斜坡破坏是复杂的非线性系统。如果把降雨型滑坡的发生过程看作一个系统,降雨和环境因素作为它的输入,滑坡发生概率作为其输出,则输入与输出之间的关系是非线性的。由此,可以认为在降雨型滑坡预报中采用非线性的方法将比确定性的模型或一般线性统计方法更为恰当。

According to the principle of effective stress, for rainfall-induced landslides, the small and shallow landslides are mostly controlled by the cohesive strength; the large and deep landslides are controlled by friction. For the prediction of rainfall-triggered landslides, the measuring of antecedent precipitation and soil moisture is a highly important issue for most landslide-prone areas. Because under a given rainfall condition, the various environmental factors affecting critical pore-water pressure needed to trigger landslides, such as geology, topography, soil deposits, land cover and so on, vary spatially in a region, the probability of rainfall-induced landslide occurrence varies from location to location. From the understanding of granular flow in current physics, it is suggested that landslides induced by the rainfall are involved in a complex non-linear system. If rainfall and the environmental factors are taken as input to the system, landslides are considered as the output variables of the system, and then the relationship between the input and the output is non-linear. Thus, the forecast issues for rainfall-induced landslides should be treated with a non-linear method rather than traditional statistical procedures and process-based deterministic models.