Michele Calvello

Definition and performance of a threshold-based regional early warning model for rainfall-induced landslides

A process chain for the definition and the performance assessment of an operational regional warning model for rainfall-induced landslides, based on rainfall thresholds, is proposed and tested in a landslide-prone area in the Campania region, southern Italy. A database of 96 shallow landslides triggered by rainfall in the period 2003–2010 and rainfall data gathered from 58 rain gauges are used. First, a set of rainfall threshold equations are defined applying a well-known frequentist method to all the reconstructed rainfall conditions responsible for the documented landslides in the area of analysis. Several thresholds at different exceedance probabilities (percentiles) are evaluated, and nine different percentile combinations are selected for the activation of three warning levels. Subsequently, for each combination, the issuing of warning levels is computed by comparing, over time, the measured rainfall with the pre-defined warning level thresholds. Finally, the optimal percentile combination to be employed in the regional early warning system, i.e. the one providing the best model performance in terms of success and error indicators, is selected employing the “event, duration matrix, performance” (EDuMaP) method.

Groundwater Modeling for the Analysis of Active Slow-Moving Landslides

Active slow-moving landslides in clayey soils exhibit continuous movements generally controlled both in the accelerating and decelerating phases by the pore-water pressure regime that, in turn, is strictly correlated to the net rainfall regime. The paper stresses the importance of a reliable groundwater model to predict these types of movements. To this aim a procedure is proposed to define the transient groundwater regime in the slope on the basis of recorded rainfall and monitoring data; the model is then used to derive the time-dependent shear strength along the main slip surfaces. The displacements at selected points along the slip surface are computed using a phenomenological (i.e., empirical) relationship between the local factor of safety and the displacement rate at those points. The procedure is employed for the analysis of a well-documented case history: the Porta Cassia landslide (central Italy).

Combined use of statistical and DInSAR data analyses to define the state of activity of slow-moving landslides

Statistical analyses have been often used for landslide susceptibility zoning at small to medium scale when relevant base and thematic maps are available. Since the beginning of the last decade, images remotely acquired by spaceborne Synthetic Aperture Radar (SAR) and processed via Differential SAR Interferometry (DInSAR) proved extremely useful for non-invasive and non-destructive monitoring of displacements of the topographic surface. The present paper proposes an original procedure for the definition of the state of activity of slow-moving landslides via the combined use of multivariate statistical analyses and DInSAR data. The procedure is based on the following essential elements: distinction between terrain units used for computational purposes and the final zoning units; independent statistical and DInSAR analyses and activity models leading to first-level state of activity zoning maps; a consistency model between statistical and DInSAR analyses; two confidence and combination models leading, respectively, to second- or third-level state of activity zoning maps. The application in a test area including 19 municipalities in southern Italy, where slow-moving landslides are widespread and accurately mapped by using geomorphological criteria, allowed the generation of the three above-mentioned levels of zoning maps. The results were successfully crosschecked by exploiting a different DInSAR dataset and the results of previous works based on the use of slow-moving landslide-induced damage to facilities surveys.

Landslide risk perception: a case study in Southern Italy

Perceptions of risk are a key issue when seeking to develop systems, practices and policies to protect local populations. This is particularly evident when risk mitigation strategies involve non-structural measures such as relocation and warning systems which presuppose the active involvement of the communities in question. This study adopts an interdisciplinary approach to studying the perceptions, knowledge and opinions on landslide risk amongst residents in Sarno, a small town in Southern Italy which was hit by disastrous landslides on 5–6 May 1998. The paper presents the results of a survey conducted in the months of March, April and May 2013 using a purpose-designed questionnaire. The survey was conducted using face-to-face interviews with 100 residents, 60 of whom live inside the so-called “red zone”, a territory declared at high residual risk soon after the events of 1998. The questionnaire included questions relating to perceived risk exposure, trust in institutions responsible for risk management, evaluations of risk mitigation measures and the early-warning strategy. The results of the study clearly emerges, amongst other issues, that the organisms which are responsible for risk management in Sarno need to develop more effective communication strategies in order to transmit knowledge about the actions implemented to reduce landslide risk in the area.

Displacement trends of slow-moving landslides: Classification and forecasting

A framework is proposed to characterize and forecast the displacement trends of slow-moving landslides, defined as the reactivation stage of phenomena in rocks or fine-grained soils, with movements localized along one or several existing shear surfaces. The framework is developed based on a thorough analysis of the scientific literature and with reference to significant reported case studies for which a consistent dataset of continuous displacement measurements is available. Three distinct trends of movement are defined to characterize the kinematic behavior of the active stages of slow-moving landslides in a velocity-time plot: a linear trend-type I, which is appropriate for stationary phenomena; a convex shaped trend-type II, which is associated with rapid increases in pore water pressure due to rainfall, followed by a slow decrease in the groundwater level with time; and a concave shaped trend-type III, which denotes a non-stationary process related to the presence of new boundary conditions such as those associated with the development of a newly formed local slip surface that connects with the main existing slip surface. Within the proposed framework, a model is developed to forecast future displacements for active stages of trend-type II based on displacement measurements at the beginning of the stage. The proposed model is validated by application to two case studies.

Early warning strategies to cope with landslide risk

Technicians who want to mitigate landslide risk have a variety of options available to them, from active measures addressing the reduction of the probability of occurrence of landslides, to structural engineering works designed to decrease the vulnerability of elements at risk, to early warning systems to adopt in areas where, in specific circumstances, the risk to life increases above tolerable levels. This paper presents the key elements of numerous early warning strategies implemented to cope with weather-induced landslide risk around the world, defines a coherent framework to classify and analyse landslide early warning systems, highlights the Author’s most recent research experiences on these systems, and provides comments on a series of issues relevant for directing future multidisciplinary research endeavours on this topic. A pivotal role for the discussion is played by an original framework proposed to identify the main components of an early warning system for weather-induced landslides. The framework is based on a clear distinction among three different modules: the landslide model; the warning model, which includes the landslide model; and the warning system, which include the warning model.

Inverse Analysis for Rheology Calibration in SPH Analysis of Landslide Run-Out

Landslide run-out modeling is an important issue in engineering-based procedures for landslide hazard assessment. Nowadays, several numerical techniques are available among which Smooth Particle Hydrodynamic (SPH) is a promising tool which ensures reasonable computational times to achieve an accurate description of the main kinematic quantities such as heights and velocities of the mobilized volume(s). The reliability of run-out modeling is mostly based on the comparison of numerical results and field evidences, from which the selection of the most appropriate rheological model and the calibration of the rheological parameters are obtained. Rheology calibration is usually based on trial/error procedures which, in turn, rely on expert-judgment and/or previous acquired experiences for the same type of phenomenon. In the paper, an inverse analysis procedure is tested to calibrate a simple frictional rheological model within a SPH analysis of a case study for which a detailed data-set of in situ evidences is available.

Structural safety of existing buildings near deep excavations

Structural safety of existing buildings near deep excavations is evaluated by computing exceeding probabilities of different damage criteria within a simplified probabilistic methodology based on monovariate or multivariate probabilistic analyses employing the results of a numerical model of the boundary value problem. Different limit domains, defined on one or more deformation parameters and associated to limit states, are used to contemplate: the type of the structural system (i.e., reinforced concrete or masonry buildings); the foundation typology (i.e., strip/raft or pad foundations). The sensitivity analysis is developed considering the design of a new underground station in Naples (Italy).

The Community-Based Alert and Alarm System for Rainfall Induced Landslides in Rio de Janeiro, Brazil

The community-based alert and alarm system for rainfall induced landslides in Rio de Janeiro, locally called A2C2 (Sistema de Alerta e Alarme Comunitario para Chuvas Fortes), may be considered the second tier of the citywide landslide early warning strategy initiated, between the late 1980s and the early 1990s, with the SIGRA and the—still operational—Alerta-Rio projects. The A2C2 landslide early warning system was deployed between 2011 and the first few months of 2012; it is currently employed in 103 informal communities of the city. Herein, the main characteristics of the A2C2 system are presented together with a preliminary analysis on its behaviour during this initial phase of deployment.

Displacement scenarios of a rainfall-controlled slow moving active slide in stiff clays

A model is proposed to compute displacement scenarios from rainfall scenarios of a rainfall-controlled slow moving active slide along a pre-existing slip surface. The model is calibrated and validated with reference to a well-monitored landslide in stiff clays in Central Italy. The model comprises a physically based transient groundwater analysis of the slope and a kinematic analysis relating the rate of movement and the inverse of the local factor of safety in a critical point along the slip surface. Once calibrated and validated, the model is used as a predictive tool to evaluate displacement scenarios from three rainfall scenarios. Although the reliability of the predicted displacement scenarios is not probabilistically evaluated, the scenarios may be considered as defining reasonable upper and lower bounds of future displacements.