The frequency of large supraglacial landslides (rock avalanches) occurring in glacial environments is thought to be increasing due to feedbacks with climate warming and permafrost degradation. However, it is difficult to (i) test this; (ii) establish cause-effect relationships; and (iii) determine associated lag-times, due to both temporal and spatial biases in detection rates. Here we applied the Google Earth Engine supraglacial debris input detector (GERALDINE) to Glacier Bay National Park & Preserve (GLBA), Alaska. We find that the number of rock ava-lanches (RAs) has previously been underestimated by 53 %, with a bias in past detections towards large area RAs. In total, GLBA experienced 69 RAs during 1984-2020, with the highest frequency in the last three years. Of these, 58 % were deposited into the accumulation zone and then sequestered into the ice within two years. RA sources clustered spatially at high elevations and around certain peaks and ridges, predominantly at the boundary of modelled permafrost likelihood. They also clustered temporally, occurring mainly between May and September when air temperatures were high enough to initiate rock-permafrost degradation mechanisms. There was a chronic background debris supply from RAs, with at least one RA occurring in all but nine years; however, a debris rich period during 2012-2016 was driven by three large RAs delivering 44 % of all (1984-2020) debris (by area). Comparable investigation of slope-failures in other remote currently glaciated regions is lacking. If RA rates are similar elsewhere, especially the bias towards emplacement onto/into accumulation zones, their contribution to glacial sediment budgets has been globally underestimated.

An inventory of supraglacially deposited rock avalanches that occurred in Glacier Bay National Park, Alaska, between 1984 and 2020. Reference: Smith, W.D., Dunning, S.A., Ross, N., Telling, J., Jensen, E.K., Shugar, D.H., Coe, J.A. and Geertsema, M. (2023) Revising rock avalanche magnitudes and frequencies in Glacier Bay National Park, Alaska, Geomorphology, https://doi.org/10.1016/j.geomorph.2023.108591

Landslides in glacial environments are high-magnitude, long-runout events, believed to be increasing in frequency as a paraglacial response to ice retreat and thinning and, arguably, due to warming temperatures and degrading permafrost above current glaciers. However, our ability to test these assumptions by quantifying the temporal sequencing of debris inputs over large spatial and temporal extents is limited in areas with glacier ice. Discrete landslide debris inputs, particularly in accumulation areas, are rapidly “lost”, being reworked by motion and icefalls and/or covered by snowfall. Although large landslides can be detected and located using their seismic signature, smaller (M ≤ 5.0) landslides frequently go undetected because their seismic signature is less than the noise floor, particularly supraglacially deposited landslides, which feature a “quiet” runout over snow. Here, we present GERALDINE (Google Earth Engine supRaglAciaL Debris INput dEtector): a new free-to-use tool leveraging Landsat 4–8 satellite imagery and Google Earth Engine. GERALDINE outputs maps of new supraglacial debris additions within user-defined areas and time ranges, providing a user with a reference map, from which large debris inputs such as supraglacial landslides (>0.05 km2) can be rapidly identified. We validate the effectiveness of GERALDINE outputs using published supraglacial rock avalanche inventories, and then demonstrate its potential by identifying two previously unknown, large (>2 km2) landslide-derived supraglacial debris inputs onto glaciers in the Hayes Range, Alaska, one of which was not detected seismically. GERALDINE is a first step towards a complete global magnitude–frequency of landslide inputs onto glaciers over the 38 years of Landsat Thematic Mapper imagery.