An early warning system can effectively reduce the damage caused by landslides. To study the change law of hydrological indexes with the evolution process, a rainfall-induced landslide model test was carried out. The changes of volumetric moisture content, matrix suction, and pore water pressure with time were obtained. It was shown that the volumetric moisture content maintained the maximum value before failure, and the pore water pressure increased; at the same time, the failure was accompanied by a sudden change of volumetric moisture content. Pore water pressure and volumetric moisture content are usually used as indicators for landslide early warnings. In many cases, a slope stability model based on pore water pressure is better than that based on volumetric moisture content. However, heavy rainfall sometimes causes failure after a small amount of positive pore water pressure forms. It is difficult to explain this instability using a traditional slope stability model; therefore, we established a model that focused on volumetric moisture content and its variation over time and early warning thresholds for two landslides with different permeabilities. The early warning model proposed in this study can be applied reliably for landslides not triggered by positive pore water pressure and provide early warning for slope instabilities triggered in saturated conditions or with low pore water pressure.

预警系统可以有效减少滑坡造成的损失。为研究水文指标随演化过程的变化规律，开展了降雨诱发滑坡模型试验。得到了体积含水率、基质吸力和孔隙水压力随时间的变化。结果表明，体积含水率保持在破坏前的最大值，孔隙水压力增加；同时，故障伴随着体积含水量的突然变化。孔隙水压力和体积含水率通常用作滑坡预警的指标。在许多情况下，基于孔隙水压力的边坡稳定性模型优于基于体积含水量的边坡稳定性模型。然而，强降雨有时会在少量正孔隙水压力形成后引起破坏。使用传统的边坡稳定性模型很难解释这种不稳定性；因此，我们建立了一个模型，该模型专注于体积含水量及其随时间的变化以及两个具有不同渗透率的滑坡的预警阈值。本研究提出的预警模型可以可靠地应用于非正孔隙水压力触发的滑坡，并为饱和条件或低孔隙水压力触发的斜坡失稳提供预警。我们建立了一个模型，该模型专注于体积水分含量及其随时间的变化以及两个具有不同渗透率的滑坡的预警阈值。本研究提出的预警模型可以可靠地应用于非正孔隙水压力触发的滑坡，并为饱和条件或低孔隙水压力触发的斜坡失稳提供预警。我们建立了一个模型，该模型专注于体积水分含量及其随时间的变化以及两个具有不同渗透率的滑坡的预警阈值。本研究提出的预警模型可以可靠地应用于非正孔隙水压力触发的滑坡，并为饱和条件或低孔隙水压力触发的斜坡失稳提供预警。