Utilizing the TTOP model to understand spatial permafrost temperature variability in a High Arctic landscape, Cape Bounty, Nunavut, Canada

Abstract

Ground surface and permafrost temperatures in the High Arctic are often considered homogeneous especially when viewed at the scale of climate and environmental models. However, this is generally incorrect due to highly variable, topographically redistributed snow cover, which generates a substantial degree of ground thermal heterogeneity. The objective of this study is to describe and spatially model the variability in the ground thermal regime within the Cape Bounty Arctic Watershed Observatory (CBAWO), Nunavut, Canada, using the TTOP model, for current conditions in addition to a series of future climate change scenarios. While observed air temperature was mostly uniform, annual mean ground surface and permafrost temperatures across the paired watersheds were estimated to range between −3.8 to −13.8°C and −3.9 to −14°C, respectively, similar to the −5 to −15°C magnitude and range identified from boreholes across the High Arctic. The spatial models showed higher ground surface temperatures in topographic hollows (slope bases and stream channels), and lower temperatures in areas of topographic prominence (hilltops and plateaus) following the spatial pattern of snow accumulation and redistribution. Under projected climate change, the models predicted areas with the coldest permafrost to have the largest magnitude of warming (about 9°C), while areas of warm permafrost became closer to 0°C (warming 4–7°C). This thermal heterogeneity may have implications for ground instability such as permafrost‐related mass movements, hydrological connectivity, biogeochemical cycling, and microbial activity, which influence water quality and contaminant mobility.

Landscape‐scale variations in near‐surface soil temperature and active‐layer thickness: Implications for high‐resolution permafrost mapping

“…Our analysis of variation in Tnss indicates that snow is the key determinant of variation in permafrost temperature. This result is in agreement with other studies in this region 3,14,27,30,34,35 and in other areas 12,14,36 . This result is also supported by contrasting observations under nearly snow‐free conditions in a relatively flat area on the Qinghai‐Tibet Plateau, where variation in annual mean Tnss tends to be small, and the differences of annual mean Tnss among ecotypes are not statistically different 37 .…”

“…This range is similar to the variations observed in the outer Mackenzie Delta, 27 in Svalbard and southern Norway, 12 and in model estimates of ground temperatures in different ecological settings in Alaska 28 . However, the range of variation we reported here is lower than the range observed in Melville Island in Canada (10.0°C), 14 and in the North Slave Region near Lac de Gras (11.4°C) 13,29 . Our observed range of variation is larger than the multisite observations near Trail Valley Creek 15 (3.8 and 4.5°C in two years, respectively) and the five‐site observations in the Mackenzie Delta (4.2°C) 30 because our study sampled sites across a greater range of soil and snow conditions.…”

“…Our observed ALT was not correlated with , and the relative difference among sites in (varied about one‐fold) was much smaller than that of ALT (varied about four‐fold). This is similar to the observations near Trail Valley Creek 14 . According to the simplified Stephan equation (Equation ), the variation of ALT must therefore be due to differences in the edaphic factor (E).…”

“…Their analysis focussed on thawing season Tnss and relations with ALT in plot averages. Other examples of Tnss measurements at multiple sites include observations in the eastern part of the Swiss Alps, 11 in Svalbard and southern Norway, 12 in the North Slave Region near Lac de Gras in Canada, 13 in the Cape Bounty Arctic Watershed Observatory on the south‐central coast of Melville Island in Canada, 14 and near Trail Valley Creek (about 45 km north of Inuvik) in the Northwest Territories of Canada 15 . These studies reveal large variations of Tnss at the landscape scale.…”

Increased mean annual temperatures in 2014–2019 indicate permafrost thaw in Alaskan national parks

“…The n f for many natural surfaces in cold climates is 0.5 or less due to the effect of snow, whereas the value of n t is often close to one (Lunardini 1978;Way and Lewkowicz 2016). The mean annual temperature at the top of the permafrost T p can be modeled from air temperature data, n-factors, and the thermal offset (T o ) as follows (Romanovsky and Osterkamp 1995;Way and Lewkowicz 2016;Garibaldi, Bonnaventure, and Lamoureux 2020):…”