Multi-Scale Modeling of Permafrost Thermal Hydrology
A multi-scale modeling framework being developed under the NGEE Arctic project aims to model the eco-hydrologic processes in the complex Arctic landscape. Micro-topography of patterned ground and drained thaw lake basins plays an important role in regional hydrology and biogeochemistry. High resolution unstructured meshes have been developed using 0.25m resolution LIDAR-derived Digital Elevation Model (DEM) at NGEE Arctic sites near Barrow, AK. Figure 1(a) shows a computational mesh developed for a watershed (4327.5 m2) near NGEE Arctic Site C consisting of 13,479,552 grid cells. While fine scale model mesh resolves the topography at a high resolution, its a large computational challenge. The intermediate scale model in the multi-scale modeling framework is focused on resolving all eco-hydrologic processes at coarser resolution while still explicitly capturing geomorphologically important landscape features like centers, rims, and troughs of the high-centered/low-centered/transitional polygons. Figure 1(b) shows a coarse resolution mesh which resolves the centers, rims and troughs of the polygons using multi-spectral remote sensing based landscape classifications developed by Chandana Gangodagamage from Los Alamos National Laboratory. The intermediate scale mesh for the watershed near Site C consists of 22,518 grid cells reducing the complexity of the problem by almost three orders of magnitude as compared to fine scale model.
|(a) Fine scale model
||(b) Intermediate scale model
|Figure 1. PFLOTRAN unstructured grid meshes for fine and intermediate scale models.
||Figure 2: Permafrost simulation using PFLOTRAN conducted over a watershed scale domain near NGEE Site C near Barrrow, AK. Figure shows the fraction of frozen water content (fully frozen soil in red vs thawed soil in blue).
Permafrost hydrology simulations were conducted for the watershed at fine and intermediate scales using these grids. Models at both scales were parameterized and forced with similar data sets. Observed meteorological data from Circumpolar Active Layer Monitoring (CALM) sites near Barrow, AK were used to derive the forcing data set for the simulations. While still un-calibrated, simulations at fine and intermediate scales produced qualitatively similar predictions of active layer developments. Figure 2 shows a snapshot of result from the two models that was presented at the recent American Geophysical Union Fall Meeting 2013 by Kumar et. al. Efforts are underway to calibrate the parameters for fine scale model to reproduce observed active layer dynamics, and to develop effective parameterizations for intermediate scale model. Calibrated intermediate scale model will be applied for regional scale simulations around Barrow Environmental Observatory.
Modeling Active Layer and Permafrost Dynamics of Ice Wedge Polygon Dominated Arctic Ecosystems, Jitendra Kumar; Gautam Bisht; Anna Liljedahl; Richard T. Mills; Satish Karra; Scott L. Painter; Peter E. Thornton, AGU Fall Meeting 2013 (C53A-0535).