Author(s): G. Formetta; G. Bertoldi; E. Bortoli

Linked Author(s):

Keywords: Hydrological effects; Vegetation; Hillslope stability; Landslide susceptibility

Abstract: We present here a modelling sensitivity analysis with a hydrological model coupled with a landslide triggering module to evaluate the relative impacts of the hydrological effects of vegetation on landslide susceptibility. The triggering of shallow landslides in sloped landscape is influenced by the distribution of vegetation. Plants stabilize the hillslope through mechanical and hydrological effects. The former is due to the plant roots that anchor the top soil to the deeper layers and increase the tensile strength of near surface soils. The latter is due to: i) the rainfall interception of the foliage system and ii) the root network water uptake/evapotranspiration. This results in higher suction pressure-head values and this increases the soil shear strength. Although several studies in the last decades have focused on the mechanical contribution of root reinforcement, only few recent papers have focused on understanding the hydrological effects of vegetation on the slope stability. The regulation of water fluxes in saturated/unsaturated soils, the correct simulation of evapotranspiration processes, the heterogeneity of vegetation types play an important role on slope stability at catchment scale, especially for short intense rainfall events. Understanding the combined effects of these processes on slope hydrology and stability at the catchment scale is the primary objective of this paper. The study combines long-term field measurement campaigns with advanced numerical simulations in the Mazia Valley Long Term (Socio) Ecological Research Area (LTSER) (South Tyrol, Northern Italy). The landscape is very complex from a hydrological perspective due to the strong interaction between the extremely heterogeneous biotic/abiotic system and the high topographic gradient (between 900 and 2200 m a.s.l.). The latter makes part of the catchment highly susceptible to shallow landslide. We focused our analysis on a 5 km² hillslope, which includes two sites with meteorological data, soil moisture at 2, 5, and 20 cm depth, eddy-covariance evapotranspiration observations, and one site with suction pressure-head data. Moreover, spatially distributed field campaigns for root and soil depth measurements have been carried out. Finally, distributed, ground, and remote sensing observations of surface soil moisture are available over the area for model validation. Measured data have been used to feed a 3-D, physically-based, distributed hydrological model, coupled with a module for computation of the probability of failure, based on the infinite slope assumption. The model solves the coupled Richards and surface energy balance equations (considering soil freezing and vegetation influence) to describe the subsurface flow in variably saturated soils, evapotranspiration, and snow melting. Finally, the stability model simulates the temporal variation of the probability of failure in the study area. Model results in terms of soil moisture/pressure head at different depths have been validated against field measurements and used to estimate the probability of failure within the analysed catchment. Preliminary results for a 2 years simulation period indicate that the model was able to capture the overall soil moisture dynamic measured in the stations. The simulated spatial soil water content distribution (Figure 1) strongly reflects land cover patterns, which are indeed controlled by the root depth distribution (assumed here 15 cm for meadow areas, 40 cm for pasture and 1 m for forest). The framework allows us to investigate the combined effect of land cover/use, vegetation types, and freezing soil on soil moisture and suction pressure-head dynamic. Subsequently, we will be able to evaluate the relative impacts of the hydrological effects of vegetation on landslide susceptibility. Once the model has been validated, we will perform a series of numerical experiments with different vegetation types and root depth and density. Future applications will focus on quantifying the potential effects of climate/land cover changes on the study area.

DOI: https://doi.org/10.3850/978-981-11-2731-1_357-cd

Year: 2018