Landslide size matters: A new data-driven, spatial prototype
A landslide is defined as the movement of a mass of rock, debris, or earth down a slope. They are common in mountains, hills and along the coast, and they can cause serious human injury, loss of life, damage to public and private properties and economic devastation. To cope with the landslide problem, the ability to measure and predict them is crucial.
The first digital document we can find where the prediction of landslide occurrences is addressed dates back to 1972. At that time, geomorphologists mapped and labelled the landscape according to their experience to recognize slopes prone to fail.
Much has changed since then. The geomorphological community focused on estimating landslide occurrences over catchments, regional and national scale territories has now embraced data-driven models to generate reliable predictions and support master planning. Until today, these data-driven models have consisted of binary classifiers, where the aim is to recognize whether a slope will likely be stable or unstable in the near future. This concept is referred to as landslide susceptibility.
However, the susceptibility neglects a piece of equally important information other than where landslides may occur. It overlooks the actual size of the landslides, which intuitively represents a crucial parameter to assess the threat posed by a given slope failure.
This work presents the first statistically-based model in the literature able to address this issue by proposing a complementary model to the susceptibility one. We propose to estimate how large a landslide or multiple landslides together may be within a specific areal unit. We tested our model using the most complete global dataset of landslides induced by earthquakes showing that the model is able to suitably generate maps of expected landslide sizes. As a result, we expect this model to become an important support for decision-makers around the globe.
The article is available here.