Mean root depth estimation at landslide slopes

Plant rooting systems affect slope stability through the soil reinforcement given by the root network. The vertical root distribution in particular is crucial for the assessment of the critical slip surface in slope stability analyses. We propose here an expeditious way to assess the major characteristics of the root system at landslide slopes. More in detail, we extend and validate an ecohydrological model proposed for flat terrain and water-dependent ecosystems by Laio et al. (2006). This model has the merit to use readily available climatic and pedologic descriptors to predict the mean root depth, but its extension to hillslopes in semi-humid environments requires validation. The model has been improved and then tested on a case study in northern Tuscany (Italy) which considers 17 landslide sites where the tree rooting systems have been measured. The results show a quite good match between observed and modeled mean root depths. The accuracy of the results largely depends on the improvements brought to the model in the parameters estimation phase, in particular through the application of the Curve Number method and through the refinement of the definition of the growing season. The results show that in 14 cases out of 17 the error is lower than 30%. Furthermore, the error becomes lower than 30% at all sites if we take into account differences among soils in the estimation of the portion of precipitation which infiltrates into the soil. These results prove the potential of the proposed method: using few and quite readily available parameters, it allows one to determine the mean root depths of vegetation with good accuracy: an important parameter for stability assessment of vegetated slopes on a large scale. (C) 2014 Elsevier B. V. All rights reserved.