Skip to main content
U.S. flag

An official website of the United States government

Exploring a Novel Model Conceptualization for Watershed-scale Permafrost Simulation

Presentation Date
Wednesday, December 14, 2022 at 11:20am - Wednesday, December 14, 2022 at 11:30am
Location
McCormick Place - S504d
Authors

Author

Abstract

Better prediction of the hydrologic cycle in permafrost-rich Arctic watersheds requires the representation of complex permafrost physics and high spatial-temporal resolution of model inputs. Though integrated surface-subsurface, coupled mass-energy transport models have been successfully applied to simulate permafrost dynamics at local polygon and hillslope scales, these models have not been used at the scale of watersheds or full river basins. Largely this is because three-dimensional simulations have been found computationally infeasible. Therefore, in this study, we propose a novel model conceptualization to scale up permafrost simulations to watershed or full-river basin scales, allowing the projection of riverine fluxes under future conditions. Briefly, the core idea is to utilize model splitting strategy combined with parameterization in place of full three-dimensional simulation on the watershed. To validate this approach, we selected a HUC12 catchment within the Sagavanirktok River Basin and simulated the freeze-thaw cycle over multiple years. We split and delineated the HUC12 catchment into 36 sub-catchments based on the contributing area and other geomorphic features. Each sub-catchment was parameterized into an equivalent 2D hillslope. Both 2D and 3D simulations were conducted for each sub-catchment, and the resulting product is routed by the Model for Scale Adaptive River Transport (MOSART), a physically based 1D river routing model. We compared the 2D and 3D simulated results among sub-catchments, and compared the aggregated results of sub-catchments with the site-based benchmark simulation. Results demonstrate that the simulated results based on parameterized 2D sub-catchments match well with the full 3D benchmark results, and in the meantime, reduce the runtime significantly.

Funding Program Area(s)
Additional Resources:
NERSC (National Energy Research Scientific Computing Center)