Much of the Arctic is underlain by permafrost, defined as an Earth-material substrate that has continuously remained below zero degrees Celsius for a period of at least two consecutive years, and can remain frozen for thousands of years. As the Arctic is experiences significantly more rapid warming than the rest of the planet, permafrost thaw is accelerating. This has been proposed to be the primary driver of environmental change in fluvial and coastal Arctic landscapes, with proposed effects ranging from rapid changes in sediment and nutrient flux delivered to the Arctic Ocean, expansion of river channel networks, and increased rates of riverbank erosion. This increased riverbank erosion and channelization of rivers through thawing permafrost has led to the displacement of communities in the Arctic and has resulted in an accelerated release of previously sequestered organic material, greenhouse gasses, and heavy metals with yet to be understood impacts on regional water quality as well as the regional and global carbon cycle. Indeed, 87% of the 213 Alaska native communities are currently being affected by accelerating erosion, with 33% of the communities being located in the Yukon River Basin. Large rivers, such as the Yukon, incise into the permafrost and, as such, the bank erosion process of these rivers with depths exceeding the depth of the active layer is governed by the thermomechanical interactions between the eroding flow in the channel and the resistance to erosion of the frozen sediment that comprises the bank.

Canonically, rates of permafrost riverbank erosion have been thought to be set by the rate at which pore-ice in frozen riverbanks melts, thus liberating the sediment for instantaneous erosion by the river's flow. Under this thermomechanical erosion model, riverbank erosion rate is dependent upon the temperature of the water and the permafrost, the material properties of the permafrost and water, and the velocity of the flow. However, implementation of this model on an annual basis predicts river channel lateral migration rates multiple order of magnitude greater than is observed by remotely-sensed migration rates for permafrost rivers. As such, more extensive field-based validation to understand the thermo-mechanical drivers of riverbank erosion and lateral migration in cold environments.

To address this challenge, we conducted a field investigation in the Yukon River watershed to monitor the properties of the river flow and the erosion of riverbanks over the course of the summer. Two field sites were selected: one on the main branch of the Yukon near Beaver, Alaska, and the other on the Koyukuk River, a tributary of the Yukon River near Huslia, Alaska. Two field campaigns were conducted in the spring and fall in order to characterize actively migrating permafrost riverbanks and install and collect monitoring equipment. Stratigraphic surveys, core samples, and drone surveys were conducted to characterize the composition and structures of the banks. Bank erosion was monitored through camera stations positioned on the point bar side of migrating bends that captured stereoimagery of the eroding point bar side of the bend in ten-minute intervals throughout the summer. The flow depth and temperature of the river flow was concurrently monitored throughout the summer.

Results show two bank failure events at one bank that occurred 25 days after ice breakup. Historically, this bend has laterally migrated at 7.4m/yr. These failure events produced blocks approximately 10m wide and 5m deep. During the four weeks after ice breakup, water depth increased rapidly by 1.6m over approximately 20 days and then was relatively stable for 5 days preceding bank collapse. Water temperature increased from near 0 to 12°C over the same time period. The bank was undercut by many meters at the waterline prior to failure, suggesting the occurrence of at least some thermomechanical erosion. However, as water temperatures continued to warm throughout the summer up to approximately 20°C, rapid undercutting of the bank and large-scale cantilever failures did not persist, but rather ceased and/or continued through small, localized failures that did not result in change in bank position. This reduction in bank erosion rate correlates with a reduction in the fluid shear stress of the flow. In summation, during periods of high discharge, high fluid shear stress, and low water temperature corresponded to periods of rapid bank erosion, while periods of lower discharge, lower fluid shear stress, and higher temperature corresponded to periods of low to negligible rates of bank erosion. Therefore, it appears that changing water temperatures in Arctic and sub-Arctic rivers will not have a significant impact on the rates of bank erosion into the future. Rather, the changing hydroclimate of the region (e.g. earlier ice breakups, longer periods of high flow, higher discharges, etc.) will be the dominant driver of fluvial geomorphic change and generator of geomorphic hazard in cold regions under the warming climate.

Last Modified: 12/30/2023
Modified by: Kieran B Dunne