Land Surface Hydrological Modelling of the Mackenzie River Basin: Parametrization to Simulate Streamflow and Permafrost Dynamics Concurrently
Abstract
Earth system models predict that the current global warming trends will continue. Continental high latitudes have been warming at higher rates than the global average, resulting in permafrost thaw with implications for soil moisture, hydraulic connectivity, streamflow seasonality, land subsidence, and vegetation. Feedbacks are generally complex and depend on a multitude of factors including changes to precipitation intensity, timing, and phase as well as soil composition and hydraulic and thermal properties. The Mackenzie River Basin (MRB) is the largest in Canada (1.8x106 km2) and is underlain by permafrost of various classes (continuous, discontinues, and sporadic) for most of its extent (70-80% by area). The MESH land surface hydrology model has been parameterized for the MRB with a deep soil profile (50 m) and organic soils to be able to simulate permafrost dynamics. Hydrological modelling of the MRB is complicated by the dominance of cold regions processes that require coupling of the energy and water balances. Permafrost regime simulation requires long spin-up periods to initialize the deep soil profile.
First, a few large sub-basins were calibrated against streamflow to estimate a limited number of influential parameters. Then, sensitivity and identifiability analyses at a few well-instrumented sites were used to pinpoint the most important parameters for permafrost. Lastly, those influential parameters were fine-tuned to reproduce the spatial distribution of permafrost occurrence probability provided by several available datasets that were produced by different methods. Additionally, site-based validations were performed where active layer depth or soil temperature data were available. The sheer size of the basin and the uncertainty and sparsity of permafrost data hinders usage of such point-based datasets for model parameter transferability across the entire Mackenzie Basin. The resulting model has high fidelity in simulating both the hydrology and permafrost dynamics in the basin to allow using it for predicting climate and land cover change impacts on both aspects and their interactions for different land cover classes. This can be used to inform adaptations to permafrost and hydrological change resulting from climate change.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.H25C..07E