Thermokarst lakes are prevalent in Arctic coastal lowland regions and sublake permafrost degradation and talik development contributes to greenhouse gas emissions by tapping the large permafrost carbon pool. Whereas lateral thermokarst lake expansion is readily apparent through remote sensing and shoreline measurements, sublake thawed sediment conditions and talik growth are difficult to measure. Here we combine transient electromagnetic surveys with thermal modeling, backed up by measured permafrost properties and radiocarbon ages, to reveal closed‐talik geometry associated with a thermokarst lake in continuous permafrost. To improve access to talik geometry data, we conducted surveys along three transient electromagnetic transects perpendicular to lakeshores with different decadal‐scale expansion rates of 0.16, 0.38, and 0.58 m/year. We modeled thermal development of the talik using boundary conditions based on field data from the lake, surrounding permafrost and a borehole, independent of the transient electromagnetics. A talik depth of 91 m was determined from analysis of the transient electromagnetic surveys. Using a lake initiation age of 1400 years before present and available subsurface properties the results from thermal modeling of the lake center arrived at a best estimate talk depth of 80 m, which is on the same order of magnitude as the results from the transient electromagnetic survey. Our approach has provided a noninvasive estimate of talik geometry suitable for comparable settings throughout circum‐Arctic coastal lowland regions.

Lakes in Arctic systems contribute to hydrologic storage, biogeochemical cycling, and permafrost thaw. Here, we have used surface nuclear magnetic resonance (NMR) measurements on lakes of Alaska’s North Slope to investigate the extent of permafrost thaw below lakes with different annual ice conditions. Our purpose is to understand if annual lake ice conditions are related to development of thawed permafrost below lakes. We investigated 10 lakes and two terrestrial permafrost control sites using surface NMR and direct measurement under spring conditions when lake ice is nearly at its thickest. We did not observe unfrozen water below our surveyed bedfast ice lakes, whereas unfrozen water (indicating permafrost thaw) was measured below floating ice lakes. We found that transitional ice lakes, ones that alternate between floating and bedfast ice conditions over multiyear timescales depending on winter ice growth and lake level conditions, have complex vertical unfrozen water content profiles attributed to sporadic periods of thaw. Based on that finding, we speculate that predicting the presence of talik based on remotely sensed lake ice conditions is unreliable. We applied a scheme to subtract the lake water signal from the NMR data and found the resulting inversions to be improved.

Strong winter warming has dominated recent patterns of climate change along the Arctic Coastal Plain (ACP) of northern Alaska. The full impact of arctic winters may be best manifest by freshwater ice growth and the extent to which abundant shallow ACP lakes freeze solid with bedfast ice by the end of winter. For example, winter conditions of 2016-17 produced record low extents of bedfast ice across the ACP. In addition to high air temperatures, the causes varied from deep snow accumulation on the Barrow Peninsula to high late season rainfall and lake levels farther east on the ACP. In contrast, the previous winter of 2015-16 was also warm, but low snowpack and high winds caused relatively thick lake ice to develop and corresponding high extents of bedfast ice on the ACP. This recent comparison of extreme variation in lake ice responses between two adjacent regions and years in the context of long-term climate and ice records highlights the complexity associated with weather conditions and climate change in the Arctic. Recent observations of maximum ice thickness (MIT) compared to simulated MIT from Weather Research and Forcing (Polar-WRF) model output show greater departure toward thinner ice than predicted by models, underscoring this uncertainty and the need for sustained observations. Lake ice thickness and the extent of bedfast ice not only indicate the impact of arctic winters, but also directly affect sublake permafrost, winter water supply for industry, and overwinter habitat availability. Therefore, tracking freshwater ice responses provides a comprehensive picture of winter, as well as summer, weather conditions and climate change with implications to broader landscape, ecosystem, and resource responses in the Arctic.

Lakes and drained lake basins (DLBs) combined are estimated to cover up to ∼80% of the western Arctic Coastal Plain of Alaska (∼30,000 km 2) (Grosse et al., 2013; Hinkel et al., 2005; Jones & Arp, 2015). There are a variety of lake types in the Arctic, but the most common are thermokarst lakes in lowland regions with ice-rich permafrost (Grosse et al., 2013; Kling, 2009) that form due to permafrost thaw and surface subsidence. Deeper lakes developed in permafrost terrain are often underlain by layers or bodies of perennially unfrozen ground below the lake bed known as a talik (van Everdingen, 1998). Arctic lakes can persist for thousands of years, but, due to ongoing margin expansion and other landscape changes, they eventually drain laterally to create a mosaic of extant lakes and DLBs (Hinkel et al. 2007; Mackay, 1992). Arctic lake drainage can occur through a variety of processes, and where and when lake drainage occurs influences landscape succession and permafrost aggradation (refreezing of the talik). Remote-sensing analysis of historical imagery of the western Arctic Coastal Plain of Alaska identified that 1-2 lakes larger than 10 ha have partially (>25% area reduction) or completely drained per year between

No abstract