Landslides and synoptic weather trends in the European Alps

Abstract

Landslides present a substantial geomorphological hazard in Alpine regions and there are expectations that their frequency and magnitude will change as weather and climate change in the future. Understanding the spatial extent and timing of landslides is therefore important if we are to assess their future behaviour. Using a regional landslide inventory for the European Alps (1970-2002) we analyse the influence of synoptic weather types on landsliding. Brier Skill Scores are used to assess the predictive ability of over 5,000 different synoptic weather classifications (COST733 dataset) with the landslide inventory. Monte Carlo permutation tests show a strong seasonal influence of weather types on landsliding, and that weather types associated with high precipitation are consistent with more landslides. Over the duration of the COST733 dataset (1957-2002), summer months have seen a significant increase in the number of days with weather types associated with high frequencies of landslides, whilst the converse is true of winter months; this is not reflected in the inventory landslide frequency. We finally discuss the applications of this analysis, and point to future avenues of research.