Development of a spatially distributed model of Arctic thermal and hydrologic processes (MATH)

Abstract

A process based, spatially distributed hydrologic model with the acronym MATH (Model of Arctic Thermal and Hydrologic Processes) is constructed to quantitatively simulate the energy and mass transfer processes and their interactions within arctic regions. The impetus for development of this model was the need to have spatially distributed soil moisture data for use in models of trace gas fluxes (carbon dioxide and methane) generated from the carbon-rich soils of this region. The model is applied against the data from the Imnavait watershed (2.2 $\rm km\sp2)$ and the Upper Kuparuk River basin (146 $\rm km\sp2)$ located on the North Slope of Alaska. Both point and spatially distributed data such as precipitation, radiation, air temperature, and other meteorological data have been used as model inputs. Based on the digital elevation data, one component of the model determines drainage area, channel networks, and the flow directions in a watershed that is divided into many triangular elements. Simulated physical processes include hydraulic routing of subsurface flow, overland flow and channel flow, evapotranspiration (ET), snow ablation, and active layer thawing and freezing. This hydrologic model simulates the dynamic interactions of each of these processes and can predict spatially distributed snowmelt, soil moisture, and ET over a watershed at each time step as well as discharge at any point(s) of interest along a channel. Modeled results of spatially distributed soil moisture content, discharge at gauging stations and other results yield very good agreement, both spatially and temporally, with independently derived data sets, such as Synthetic Aperture Radar (SAR) generated soil moisture data, field measurements of snow ablation, measured discharge data and water balance computations. The timing of simulated discharge results do not compare well to the measured data during snowmelt periods because the effect of snow damming on runoff generation is not considered in the model. It is concluded that this model can be used to simulate spatially distributed hydrologic processes within the arctic regions provided that suitable data sets for input are available. This physically based model also has the potential to be coupled with atmospheric and biochemical models.