Simulation of surface temperature and ice cover of large northern lakes with 1-D models: a comparison with MODIS satellite data and in situ measurements

• Published in: Tellus A: Dynamic Meteorology and Oceanography
• Main Authors: H. Kheyrollah Pour, C. R. Duguay, A. Martynov, L. C. Brown
• Format: Article in Journal/Newspaper
• Language: English
• Published: Stockholm University Press 2012
• Subjects: lake surface temperature; lake ice; lake modelling; MODIS; Great Slave Lake; Great Bear Lake; Oceanography; GC1-1581; Meteorology. Climatology; QC851-999; Northwest Territories; Yellowknife; Mackenzie River; Canada; Hay River; Deline; Mackenzie river
• Online Access: https://doi.org/10.3402/tellusa.v64i0.17614; https://doaj.org/article/61f5317f9fb04083b8132b699f3ad545

Lake surface temperature (LST) and ice phenology were simulated for various points differing in depth on Great Slave Lake and Great Bear Lake, two large lakes located in the Mackenzie River Basin in Canada's Northwest Territories, using the 1-D Freshwater Lake model (FLake) and the Canadian Lake Ice Model (CLIMo) over the 2002–2010 period, forced with data from three weather stations (Yellowknife, Hay River and Deline). LST model results were compared to those derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the Earth Observing System Terra and Aqua satellite platforms. Simulated ice thickness and freeze-up/break-up dates were also compared to in situ observations. Both models showed a good agreement with daily average MODIS LSTs on an annual basis (0.935  ≤  relative index of agreement  ≤  0.984 and 0.94  ≤  mean bias error  ≤  4.83). The absence of consideration of snow on lake ice in FLake was found to have a large impact on estimated ice thicknesses (25 cm thicker on average by the end of winter compared to in situ measurements; 9 cm thicker for CLIMo) and break-up dates (6 d earlier in comparison with in situ measurements; 3 d later for CLIMo). The overall agreement between the two models and MODIS LST products during both the open water and ice seasons was good. Remotely sensed data are a promising data source for assimilation into numerical weather prediction models, as they provide the spatial coverage that is not captured by in situ data.