Simulating Arctic Alaska Snowdrifts
Using a Numerical Snow-Transport Model
Glen E. Liston1 and Matthew Sturm2
A physically based, three-dimensional numerical snow-transport model (SnowTran-3D) is used to
simulate the snow-depth evolution over complex terrain in arctic Alaska. Included in the domain
are gently rolling topographic features, as well as several relatively sharp ridges that produce large
snow-accumulation traps. The model includes snow transport resulting from saltation and suspen-
sion, snow accumulation and erosion, and sublimation of the blowing and drifting snow. It is driven
by a wind model that computes the flow field over the complex topography. The snow-transport
model requires static inputs of vegetation type and topography, and temporally evolving atmo-
spheric forcings of air temperature, humidity, precipitation, and wind speed and direction. The
vegetation type is used to define a vegetation snow-holding capacity that determines the snow
depth that must be exceeded before any additional snow is available to be transported by the wind.
Model outputs include the spatial and temporal evolution of snow depth resulting from spatial and
temporal variations in precipitation, saltation and suspension transport, and sublimation. In these
simulations the model is driven using a one-day time step, and covers a two- by three-kilometer
domain using a grid increment of twenty meters.
Using four years of meteorological and snow-depth-distribution observations from Imnavait Creek
Basin in the foothills north of the Brooks Range in arctic Alaska, the model is found to closely
simulate the observed snow-depth distribution and the interannual variability. In addition to suc-
cessfully simulating the snow distribution over the gently rolling topography, the model also repro-
duces the cross-sectional profiles of the large drift traps within the domain. Because the snowcover
evolves as the winter progresses, the model methodology allows identification and analyses of the
individual precipitation and wind events that produce the snow distributions. Thus, the model simu-
lations, in conjunction with the observations, can be used to quantify the snow transport occurring
as part of each storm event leading to the end-of-winter drift profiles and the storm-event
stratigraphy.
1
Department of Atmospheric Science, Colorado State University, Fort Collins, Colorado 80523-1371 USA
2
U.S. Army Cold Regions Research and Engineering Laboratory, P.O. Box 35170, Fort Wainwright, Alaska
99703-0170 USA
11
Previous Page
