Last updated: October 17, 2022
Article
New Methods to Measure Glacier Change in Alaska
Glaciers are decreasing as the planet warms. They are crucial components of the global cryosphere and integral to the ecosystems of their local landscapes. Understanding how they are changing and how quickly can help us anticipate the resulting changes on the landscape and in the oceans. Until now, the commonly used methods for mapping glacier change were labor intensive and thus limited in temporal and spatial scope. We addressed some of those limitations by developing a novel deep learning-based method called GlacierCoverNet, a deep neural network that relies on an extensive, purpose-built training dataset. Using this model, we mapped glaciers at a temporal cadence of every two years for the state of Alaska.
The resulting maps show changes in debris free glaciers and supraglacial debris (rocks and soil on top of clean glacier ice) over the last 35 years at two-year time steps. There are 18 maps showing the mean area coverage for two years of clean glacier ice, supraglacial debris, and the total glacier-covered area. Overall, the glacier-covered area in Alaska retreated by 8,425 km2 (a loss of 13%) between 1985 and 2020 and supraglacial debris expanded by 2,799 km2 (a gain of 64%) during the same period. While there was some variability in mean annual air temperatures since 1985, temperatures continued a long-term warming trend seen across Alaska and Earth’s high latitudes. Alaska’s largest net glacier loss was seen in the most heavily glaciated areas of southcentral Alaska, while little change was seen in the relatively small and difficult to map glaciers of the Brooks Range. The largest changes in overall glacier covered area occurred at the lowest and mid-elevations of 800-2,200 m, and in the last decade, as mean annual temperatures have increased more rapidly.
Changes over the Last 35 Years in Alaska’s Glaciated Landscape: A Novel Deep Learning Approach to Mapping Glaciers at Fine Temporal Granularity
Abstract
Glaciers are important sentinels of a changing climate, crucial components of the global cryosphere and integral to their local landscapes. However, many of the commonly used methods for mapping glacier change are labor-intensive and limit the temporal and spatial scope of existing research. This study addresses some of the limitations of prior approaches by developing a novel deep-learning-based method called GlacierCoverNet. GlacierCoverNet is a deep neural network that relies on an extensive, purpose-built training dataset. Using this model, we created a record of over three decades long at a fine temporal cadence (every two years) for the state of Alaska. We conducted a robust error analysis of this dataset and then used the dataset to characterize changes in debris-free glaciers and supraglacial debris over the last ~35 years. We found that our deep learning model could produce maps comparable to existing approaches in the capture of areal extent, but without manual editing required. The model captured the area covered with glaciers that was ~97% of the Randolph Glacier Inventory 6.0 with ~6% and ~9% omission and commission rates in the southern portion of Alaska, respectively. The overall model area capture was lower and omission and commission rates were significantly higher in the northern Brooks Range. Overall, the glacier-covered area retreated by 8425 km2 (−13%) between 1985 and 2020, and supraglacial debris expanded by 2799 km2 (64%) during the same period across the state of Alaska.
Roberts-Pierel, B. M., P. B. Kirchner, J. B. Kilbride, and R. E. Kennedy. 2022. Changes over the last 35 years in Alaska’s glaciated landscape: A novel deep learning approach to mapping glaciers at fine temporal granularity. Remote Sensing 14(18): 4582.