Gifty Attiah, Homa Kheyrollah Pour, Andrea Scott (2022). Spatial trends and distribution of Surface Temperature and Ice thickness on Sub-artic lakes using remote sensing and modelling . Proceedings of the GWF Annual Open Science Meeting, May 16-18, 2022.

Sub-artic lakes are crucial to the physical, biological and climate processes of a region. Ice covered for majority of the year these lakes are sensitive to regional climate and their freeze-up, break-up and ice duration are good indicators of changes in climate. Additionally, lake ice is a valuable resource to northern communities because ice roads are constructed (e.g., the ice longest road in Yellowknife, (NWT), spreading over 80 lakes) during winter to haul goods to and from industrial establishments (e.g., mines) and for travel within and between communities. The shorter ice duration and decrease in thickness, however, are a major detriment to the ongoing use of ice roads due to climate warming. Studies show that a one-degree increase or decrease in air temperature leads to a 6-day almost linear change in ice cover duration making it an essential climate variable to monitor. Crucial knowledge on lake ice processes and temperature is however limited in availability especially in monitoring sub-artic lakes due to logistical difficulties in collecting direct measurements. To address this limitation, this study uses remote sensing data coupled with a thermodynamic lake ice model to monitor the spatial distribution of surface temperature, duration, and thickness of lake ice.

Over 500 small to medium lakes in Northwest Territories which is a lake-rich region with several small lakes, are monitored from 1984 to 2021. Monitoring on such a large scale for small to medium lakes has not been conducted in this region previously, hence this study provides a novel approach to demonstrating the temporal and spatial trends of lake ice cover in this region. To effectively simulate lake ice thickness, an algorithm-based lake surface temperature (LST) for each lake was derived from the thermal bands of Landsat, which showed good agreement with in-situ data (1.88°C > RMSE >1.54(°C)). Increasing temperature trends were observed for both ice cover and open water periods for lakes studied. The derived LST are used as an input in spatially distributed thermodynamic model in addition to other input variables such as (wind speed (m s-1), mean air temperature(°C), relative humidity (%), snow depth(m) and cloud cover (0-1)) derived from reanalysis (ERA5). Daily spatial distribution of ice thickness was simulated for each lake on a 100m spatial resolution from 1984 to present. To evaluate model estimations of ice thickness field work was conducted to collect ice thickness measurements across 10 lake sites in Northwest West Territories.

AOSM2022 Remotely Sensed Monitoring of Northern lake Ice Using RADARSAT Constellation Mission and Cloud Computing Processing First Author: Gifty Attiah, Wilfrid Laurier University Additional Authors: Homa Kheyrollah Pour, Wilfrid Laurier University Andrea Scott, University of Waterloo