Department of Agricultural and Environmental Sciences (DiSAA), University of Milan, Via Celoria 2, 20133 Milan, Italy; Centre of Applied Studies for the Sustainable Management and Protection of Mountain Areas (Ge.S.Di.Mont), University of Milan, Via Morino 8, Edolo, 25048 Brescia, Italy.
Department of Agricultural and Environmental Sciences (DiSAA), University of Milan, Via Celoria 2, 20133 Milan, Italy; Centre of Applied Studies for the Sustainable Management and Protection of Mountain Areas (Ge.S.Di.Mont), University of Milan, Via Morino 8, Edolo, 25048 Brescia, Italy.
Sci Total Environ. 2019 Feb 20;652:1168-1186. doi: 10.1016/j.scitotenv.2018.10.318. Epub 2018 Oct 25.
In mountainous-forested landscape, quantifying the materials produced at hillslope scale that effectively reach the channel network with a given probability is currently challenging, due to the uncertainties in modelling the frequency-magnitude distribution of failures and in determining the sediment connectivity between unstable areas and channel network. The purpose of this study is to develop a modular approach to assess the sediment source areas and the probability of mobilization from hillslope, and to estimate the probability of sediment input to the streams proposing a new connectivity index. The first goal was faced adopting a 3D probabilistic slope stability method that includes the spatially distributed characteristics of forest coverage. The second aim was tackled by comparing sediment travel distance and the minimum-topographic distance to reach the nearest stream. A simple deposition model was applied to estimate the percentage of the sediment entering into the stream network. The methodology was tested on three headwater catchments in northern Italian Alps. The outputs were landslide susceptibility maps, which showed robust performances when compared to the available landslide inventories (AUC > 0.726), and maps of the probability that sediment reaches the channel network. In this way, it was possible to identify which areas are the most susceptible to landsliding, how many sediment materials can be mobilised with a given probability, and which is the degree of sediment connectivity with the channel system. Results obtained for the tested catchments, compared with data available from the literature, showed that the proposed methodology is of general validity, especially for those territories characterized by rainfall-triggered landslides and forest coverage. This study, then, provides a robust framework to improve debris-flow risk management and to implement watershed management strategies, such as planning forestry operations or positioning retention structures addressed to increase slope stability and to reduce sediment delivery.
在山区森林景观中,由于难以确定失败的频率-幅度分布以及不稳定区域与河道网络之间的泥沙连通性,因此很难量化具有一定概率到达河道网络的坡面尺度物质。本研究旨在开发一种评估山坡源区和移动概率的模块化方法,并通过提出一种新的连通性指数来估计泥沙输入溪流的概率。采用包含森林覆盖空间分布特征的 3D 概率边坡稳定性方法来实现第一个目标。通过比较泥沙运移距离和到达最近溪流的最小地形距离来实现第二个目标。应用一个简单的沉积模型来估算进入河道网络的泥沙百分比。该方法在意大利北部阿尔卑斯山的三个源头流域进行了测试。结果是滑坡敏感性图,与可用的滑坡清单(AUC > 0.726)相比,其表现稳健,以及泥沙到达河道网络的概率图。通过这种方式,可以确定哪些区域最容易发生滑坡,以一定概率可以移动多少泥沙材料,以及与河道系统的泥沙连通程度。与文献中可用数据相比,对测试流域的结果表明,所提出的方法具有普遍的有效性,特别是对于那些以降雨触发的滑坡和森林覆盖为特征的地区。因此,本研究为改进泥石流风险治理和实施流域管理策略提供了一个稳健的框架,例如规划林业作业或定位保留结构,以提高边坡稳定性并减少泥沙输送。