EGU23-12981
https://doi.org/10.5194/egusphere-egu23-12981
EGU General Assembly 2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Overall analysis of protective effects of forest on shallow landslides triggered by storm Vaia in Agordino (Italy)

Lorenzo Faes1, Andrea Rizzi1, Maximiliano Costa1,2, Paul Richter1, and Emanuele Lingua1
Lorenzo Faes et al.
  • 1Department of Land, Environment, Agriculture and Forestry, University of Padova, Legnaro (PD), Italy
  • 2Forest Ecology, Institute of Terrestrial Ecosystems, ETH Zurich, Universitätstrasse 16, Zurich 8092, Switzerland

Shallow landslides are one of the most frequent gravitative natural hazards in the Alpine region that could affect human infrastructures. Forests can play a direct protective function, preventing the triggering of such events thanks to the role they play in water regulation and mechanical effects, in particular with root reinforcement. Few studies, however, report empirical assessments, based on after-events shallow landslides inventory, of the protective effects given by the presence of the forest on this natural hazard. With this study, an attempt was made to assess the possible influence of the presence of the forest on the topographic triggering conditions and the magnitude of the landslides respect to those triggered in open lands. A comparison was then developed between the structural characteristics of forest stands, in which landslides were recorded, and the reference parameters of the protection forests guidelines. In addition, it has been evaluated how root reinforcement can have an influence at a local scale on the location of shallow landslide triggering. Finally, a subsample of the forest landslides was selected for field surveys, in order to analyze the influence of stand structure on the magnitude of landslides. The study area corresponds to the territory of upper Agordino valley (405 km2), in the Veneto Region (Italy), which was severely affected by storm Vaia in October 2018 that caused widely the trigger of numerous shallow landslides and large windthrows. Through the analysis of the orthophotos pre and post-event and the Dem of Difference of DTMs, overall 469 (116 triggered in the forest) shallow landslides were identified with median values of area of 177 m2 and volume of 163 m3. In terms of density, forest landslides are less frequent than those in open lands and are triggered on slopes with higher inclination. Forest stands where landslides were recorded show median values of coverage of 60%, gap area of 551 m2, gap length of 18 m, and gap width of 16 m. It turned out that comparing to the silvicultural guidelines on the management of protection forests, the most important parameter appears to be the gap length. Such gaps represent the weakest zone in terms of root reinforcement where the landslides can be triggered more easily. This has been confirmed by the application, at the plot scale, of the SOSlope model (Cohen and Schwarz, 2017) which results that most of the landslide scarps (42 out of 53) were located in the zones with the lowest lateral root reinforcement. A multivariate analysis carried out on data collected in the field on a subsample of 20 forest landslides highlights that landslides with higher volume and area were recorded mostly in young forests with high density. A stand with a good amount of large trees and an uneven-aged structure seems to be the most effective in these terms. These results emphasize the protective effects of forests against shallow landslides and suggest the need for their optimal silviculture management, taking also into account the increasing susceptibility to other natural disturbances which could compromise the protective function.

How to cite: Faes, L., Rizzi, A., Costa, M., Richter, P., and Lingua, E.: Overall analysis of protective effects of forest on shallow landslides triggered by storm Vaia in Agordino (Italy), EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-12981, https://doi.org/10.5194/egusphere-egu23-12981, 2023.