Towards a TTOP-Model of Permafrost Distribution for Three Areas in Yukon and Northern British Columbia

Abstract

Air, ground surface and top of permafrost temperatures (TTOP) were measured at 58 sites in three areas of Yukon and northern British Columbia in order to: (1) explore relationships between climate-permafrost transfer functions and environmental variables, (2) assess and validate the TTOP-model, and (3) attempt the first implementation of the TTOP-model (Smith and Riseborough, 1996, 2002) for these regions with complex terrain. The strongest factors controlling climate-permafrost transfer functions are elevation and land cover, though slope, aspect, topographic position and surficial geology were also investigated. In 1000 iterations of the model using random equally possible scenarios, 64% of the TTOP-model predictions were within ±1°C of measured values, a result that is 6% better than applying a simple 3°C “total offset” to the mean annual air temperature. A sensitivity analysis confirmed that the TTOP-model is most sensitive to changes in snow, thermal conductivity of the ground and summer air temperatures. A land cover driven TTOP-model was then developed and implemented. The model correctly predicts high likelihoods of permafrost (> 0.8) for sites with permafrost present and low likelihoods (< 0.4) for non-permafrost sites.