2024-B28 Timing is critical: improving our understanding of global landslide risks

Overview

Landslides produce an estimated €6Bn in damage to global infrastructure and cause 14000 deaths per year worldwide. These figures are likely to be underestimates. Improving our understanding of landslide risk is therefore of critical economic and societal importance. Our current understanding of landslide risk has largely focused on defining and monitoring locations considered at risk from landslides, whether slow-moving or catastrophic. This has greatly improved our understanding of where landslides are likely to occur, however, we lack comparable understanding and progress on when these landslides are likely to occur. This critical knowledge gap places a significant uncertainty in landslide risk management, which this project seeks to help address.  

Our preliminary database1–3 compilation and analysis of global landslide timing reveals a profound diurnal pattern, namely that the vast majority of landslides occur between 15:00 and 23:00, regardless of their global location, time of year, or climate (Figure 1). This yet unacknowledged strong constraint on the daily timing of landslides suggests an important role of daily temperature variations on e.g., pore water pressure and water viscosity in triggering landslides. This project will seek to explore this important constraint in landslide timing and determine the likely mechanisms and their variation, across the globe. In this context, the project will consider how landslide characteristics (size, materials, failure mechanism and, potentially, history) and environmental variables (terrain and landscape characteristics, as well as meteorological parameters) influence the observed failure timings. This is key to enhance landslide predictability. Specifically, this project will ask 1) how strong are the global signals on landslide occurrence at sub-daily, daily, seasonal, and annual timescales, 2) what are the key drivers, and how identification varies across temporal scales, and 3) how can this new understanding on temporal constraints better inform our understanding and modelling of landslide risk. This project’s unique focus on landslide timing will provide much-needed progress on timing controls on landslide risk, directly relevant to industry partners, especially within engineering and insurance sectors. 

Figure 1: Stacked bar chart showing preferred times of landslide occurrence. Preferred times: between 15 and 23. Sub-bars represent continents and show no evidence for large differences across world’s regions. (Sources: Froude and Petley, 2018; Kirschbaum et al., 2010; Kirschbaum et al., 2015).