Use of scenario ensembles for deriving seismic risk.

High death tolls from recent earthquakes show that seismic risk remains high globally. While there has been much focus on seismic hazard, large uncertainties associated with exposure and vulnerability have led to more limited analyses of the potential impacts of future earthquakes. We argue that as both exposure and vulnerability are reducible factors of risk, assessing their importance and variability allows for prioritization of the most effective disaster risk-reduction (DRR) actions. We address this through earthquake ensemble modeling, using the example of Nepal. We model fatalities from 90 different scenario earthquakes and establish whether impacts are specific to certain scenario earthquakes or occur irrespective of the scenario. Our results show that for most districts in Nepal impacts are not specific to the particular characteristics of a single earthquake, and that total modeled impacts are skewed toward the minimum estimate. These results suggest that planning for the worst-case scenario in Nepal may place an unnecessarily large burden on the limited resources available for DRR. We also show that the most at-risk districts are predominantly in rural western Nepal, with ∼9.5 million Nepalis inhabiting districts with higher seismic risk than Kathmandu. Our proposed approach provides a holistic consideration of seismic risk for informing contingency planning and allows the relative importance of the reducible components of risk (exposure and vulnerability) to be estimated, highlighting factors that can be targeted most effectively. We propose this approach for informing contingency planning, especially in locations where information on the likelihood of future earthquakes is inadequate.

Rapid sequestration of rock avalanche deposits within glaciers.

Topographic development in mountainous landscapes is a complex interplay between tectonics, climate and denudation. Glaciers erode valleys to generate headwall relief, and hillslope processes control the height and retreat of the peaks. The magnitude-frequency of these landslides and their long-term ability to lower mountains above glaciers is poorly understood; however, small, frequent rockfalls are currently thought to dominate. The preservation of rarer, larger, landslide deposits is exceptionally short-lived, as they are rapidly reworked. The 2013 Mount Haast rock avalanche that failed from the slopes of Aoraki/Mount Cook, New Zealand, onto the glacier accumulation zone below was invisible to conventional remote sensing after just 3 months. Here we use sub-surface data to reveal the now-buried landslide deposit, and suggest that large landslides are the primary hillslope erosion mechanism here. These data show how past large landslides can be identified in accumulation zones, providing an untapped archive of erosive events in mountainous landscapes.