RESUMEN
According to the fifth report of the Intergovernmental Panel on Climate Change, an increase in the frequency and the intensity of extreme rainfall is expected in the Mediterranean area. Among different impacts, this increase might result in a variation in the frequency and the spatial distribution of rainfall-induced landslides, and in an increase in the size of the population exposed to landslide risk. We propose a method for the regional-scale evaluation of future variations in the occurrence of rainfall-induced landslides, in response to changes in rainfall regimes. We exploit information on the occurrence of 603 rainfall-induced landslides in Calabria, southern Italy, in the period 1981-2010, and daily rainfall data recorded in the same period in the region. Furthermore, we use high-resolution climate projections based on RCP4.5 and RCP8.5 scenarios. In particular, we consider the mean variations between a 30-year future period (2036-2065) and the reference period 1981-2010 in three variables assumed as proxy for landslide activity: annual rainfall, seasonal cumulated rainfall, and annual maxima of daily rainfall. Based on reliable correlations between landslide occurrence and weather variables estimated in the reference period, we assess future variations in rainfall-induced landslide occurrence for all the municipalities of Calabria. A +45.7% and +21.2% average regional variation in rainfall-induced landslide occurrence is expected in the region for the period 2036-2065, under the RCP4.5 and RCP8.5 scenario, respectively. We also investigate the future variations in the impact of rainfall-induced landslides on the population of Calabria. We find a +80.2% and +54.5% increase in the impact on the population for the period 2036-2065, under the RCP4.5 and RCP8.5 scenario, respectively. The proposed method is quantitative and reproducible, thus it can be applied in similar regions, where adequate landslide and rainfall information is available.
RESUMEN
/ The preparation of landslide maps is an important step in any landslide hazard assessment. Landslides maps are prepared around the world, but little effort is made to assess their reliability, outline their main characteristics, and pinpoint their limitations. In order to redress this imbalance, the results of a long-term research project in the Upper Tiber River basin in central Italy are used to compare reconnaissance and detailed landslide inventory maps, statistical and geomorphologically based density maps, and landslide hazard maps obtained by multivariate statistical modeling. An attempt is made to discuss advantages and limitations of the available maps, outlining possible applications for decision-makers, land developers, and environmental and civil defence agencies. The Tiber experiment has confirmed that landslides can be cost-effectively mapped by interpreting aerial photographs coupled with field surveys and that errors and uncertainties associated with the inventory can be quantified. The experiment has shown that GIS makes it easy to prepare landslide density maps and facilitates the production of statistically based landslide hazard models. The former supply an overview of the distribution of landslides that is easily comprehended but do not provide insight on the causes of instability. The latter, giving insight into the causes of instability, are diagnostically powerful, but are difficult to prepare and exploit.