Modelling approaches to evaluate the role of climate change in landslide activity in Europe

Landslides are natural phenomena of different types, with high randomness and variability. They are triggered or influenced by multiple natural phenomena, among which rainfall plays by far the main role in most areas of the world. A relevant complication in the evaluation of landslide activity is global warming, in particular the related ongoing and expected changes in rainfall and temperature patterns. Indeed, due to larger atmospheric retention capability, an increase in the frequency and magnitude of intense rainfall events was already observed in many areas and more significant changes are expected due to climate change. Given the high spatial and temporal variability of the landslides, climate change can affect them in multiple ways and at different temporal and geographical scales.

One of the main approaches used to study the impact of climate change on landslides relies on the adoption of landslide models forced with climate projections generated by physically based, data-driven or hybrid simulation chains. Overall, these studies are based on a similar framework, with a climate modelling chain and a landslide model. The climate chain involves the choice of the Earth System models and concentration scenarios, of a downscaling technique for the production of local assessments, and, in most cases, a bias correction technique with the aim of removing the errors (assumed as systematic) in the assessment of the key climatic variables. The landslide models fed with these variables can be either physically-based (geotechnical, hydrological) or statistical (including empirical analyses or susceptibility assessment), and can operate at different spatial scales, from the slope to the regional/national scale.

In the scientific literature, most of the studies that aimed at evaluating the impact of expected climate and environmental changes to landslides considered case studies in the European continent.

We analyze fifty articles, book chapters and proceedings that proposed modelling approaches in the study of climate-change-landslide relationships, published between 1999 and 2023. We study the spatial scale, the investigated period (both the control and future periods), and the considered climate and landslide variables. Moreover, we study all the components of the climate modelling chain and the features of the landslide models.

We observe an increase in the number of basin-, regional- and national-scale works over the years. In addition, we observe that most of the works focusing on the slope scale are related to the hydro-geotechnical modelling of deep-seated landslides, while most of the basin-scale analyses consider shallow landslides and debris flows and use statistical analyses to model the landslide-climate relationships.