Evaluation of shallow landslide-triggering scenarios through a physically based approach: an example of application in the southern Messina area (northeastern Sicily, Italy)

Abstract

Abstract. Rainfall-induced shallow landslides are a widespread phenomenon that frequently causes substantial damage to property, as well as numerous casualties. In recent years a wide range of physically based models have been developed to analyze the triggering process of these events. Specifically, in this paper we propose an approach for the evaluation of different shallow landslide-triggering scenarios by means of the TRIGRS (transient rainfall infiltration and grid-based slope stability) numerical model. For the validation of the model, a back analysis of the landslide event that occurred in the study area (located SW of Messina, northeastern Sicily, Italy) on 1 October 2009 was performed, by using different methods and techniques for the definition of the input parameters. After evaluating the reliability of the model through comparison with the 2009 landslide inventory, different triggering scenarios were defined using rainfall values derived from the rainfall probability curves, reconstructed on the basis of daily and hourly historical rainfall data. The results emphasize how these phenomena are likely to occur in the area, given that even short-duration (1-3 h) rainfall events with a relatively low return period (e.g., 10-20 years) can trigger numerous slope failures. Furthermore, for the same rainfall amount, the daily simulations underestimate the instability conditions. The high susceptibility of this area to shallow landslides is testified by the high number of landslide/flood events that have occurred in the past and are summarized in this paper by means of archival research. Considering the main features of the proposed approach, the authors suggest that this methodology could be applied to different areas, even for the development of landslide early warning systems.

Shallow landslide initiation on terraced slopes: inferences from a physically based approach

“…In fact, considering the relatively low hydraulic conductivity of the soil together with the unsaturated conditions, the partial absorption of water infiltrating at the surface results in damping and smoothing of the rainfall input, which cause the triggering of fairly few cells in response to short rainfall events. In this respect, if we compare these results with those obtained for the case study occurred in Giampilieri in 2009 (see Section 3.2), it can be noted that -in that instance -the wetter initial soil moisture conditions resulted in a high number of unstable cells even in response to short rainfall events (Schiliro et al 2015a). From this point of view those initial soil moisture conditions have certainly enhanced the triggering of shallow landslides, also considering that the 2009 rainfall event was far from exceptional, at least if compared with that occurred in the Monterosso catchment in 2011.…”

“…This approach has been already tested (Schiliro et al 2015a; in a similar case study occurred in Southern Italy on 1 October 2009. On that day, a heavy rainstorm triggered several hundreds of shallow landslides in the southern Messina area (north-eastern Sicily), causing 37 fatalities and severe damage to buildings and infrastructure, mostly near the village of Giampilieri (Schiliro and Esposito 2013).…”

“…Nonetheless, also in this analysis the mean error is always higher for the maps obtained with the Saulnier's method, whereas there does not seem to be any particular influence of the vegetation cover, considering that there is not any remarkable difference of the MAE and RMSE by varying the terrace type, both in the presence and absence of the underlying debris cover. For the estimation of the initial soil moisture conditions during the 25 October 2011 event, we employed the same approach used in the Giampilieri study case (Schiliro et al 2015a): therefore, we considered the 1-month antecedent rainfall fallen in the study area by using the HYDRUS 1D model ( Simu nek et al 1998), which can simulate the water flow into unsaturated porous media resulting from a rainfall event. Daily rainfall data have been used as input for the model, whereas evapotranspiration is accounted for by inserting the maximum and minimum temperature values recorded during the investigated period into the Hargreaves equation (Jensen et al 1997).…”

“…In particular, the model was applied for the reconstruction via back-analysis of the landslide event occurred on 25 October 2011 in a coastal basin located in Eastern Liguria (Northern Italy), for which a comprehensive event-based landslide inventory has been specifically prepared. For the parameterization of the numerical model, we used a multi-methodological approach recently applied to a similar case study in Southern Italy (Schiliro et al 2015a(Schiliro et al , 2015b; however, we properly modified this approach in order to take into account the specific landuse of the study area, characterized by the widespread presence of agricultural terraces. Specifically, we performed several numerical simulations considering two different methods for the evaluation of the spatial variation of the soil thickness and four different spatial resolutions of the digital terrain model.…”

Reconstruction of a Storm Map and New Approach in the Definition of Categories of the Extreme Rainfall, Northeastern Sicily

“…Particularly affected by recent and intense storms has been the northwestern sector, where a high and continuous ridge, the Peloritani Mountains, extends between the Ionian and Tyrrhenian seas. During recent years, the most catastrophic event occurred on 1 October 2009 along the Ionian side, with more than 200 mm recorded in a few hours, and hit the village of Giampilieri, where landslides caused 36 deaths [17,18]. After 13 November 1855, the event of 2009 was the most catastrophic to have occurred in the Messina province, and follows other catastrophic events that have occurred recently.…”