The second Large Scale Area (LSA) in
the GEWEX Continental-Scale International Project (GCIP) is the
Upper Mississippi River basin, also known as LSA-NC. As compared
with the first GCIP LSA (the Arkansas-Red River basin, LSA-SW)
the distinguishing features of LSA-NC are the dominance of a north-south
gradient in the mean temperature , the absence of strong precipitation
gradients, a semi-humid climate, and the importance of cold season
processes, including snow and frozen soils. The two-layer Variable
Infiltration Capacity Model, VIC-2L, is a macroscale water and
energy balance model that has been applied successfully over LSA-SW
to predict streamflow and surface energy fluxes at a seasonal
time scale.
This poster describes strategies for modifying the VIC-2L to model frozen soil hydrologic processes in the Upper Mississippi basin. These include refinements to the snow accumulation and ablation algorithm, and explicit modeling of the effect of frozen soils on infiltration and runoff response. The result will be a new subsurface temperature and infiltration algorithm, that will be tested using surface radiation and heat flux data, as well as soil temperature and moisture profiles from the Rosemount Agricultural Station in Minnesota. Preliminary runs at the Rosemount site have been made using the Simultaneous Heat and Water (SHAW) model, which simulates layers of frozen soils in a column, in a method similar to that selected for implementation with the VIC-2L. Following model evaluation, the revised version of VIC-2L will be applied to the entire LSA-NC at a daily timestep, representing the area with 202 one-half degree grid cells. Precipitation and temperature data were taken from NCDC cooperator stations, with other surface meteorological data interpolated from NCDC Surface Airways stations. Proposed modifications to VIC-2L are discussed, as well as preparations for model runs in the LAS-NC. Preliminary comparisons between data collected at the Rosemount site, and simulations using the SHAW model, are presented.
Frozen soils play an important role
in winter hydrologic processes in regions where the average daily
temperature stays below freezing for weeks or months.
Many approaches to frozen soils modeling
exist, including:
For the VIC-2L model, a numerical method was selected that solves for temperature at various levels within the soil.
The Variable Infiltration Capacity -
Two Layer (VIC2L) hydrologic model has been used
successfully to simulate water and energy balances in the Arkansas-Red
Basin (LSA-SW). The model applies a variable infiltration
capacity (Figure 1) over a grid cell. This allows the
model to represent the sub-grid spatial variation of infiltration
within a large grid cell.
The VIC-2L model solves water flow through
Frozen soils are incorporated into the
VIC-2L, by splitting the upper layer:
The freezing depth is calculated by
computing soil temperatures at various soil depths, see Figure
3. For the solution:
Soil and meteorological data were obtained
for the winter of 1994-95 for the Rosemount Agriculture
Experimental Station, in southeastern Minnesota. Data made available
include:
Using soil and meteorological information, the VIC model will be used to simulate the soil moisture levels and freezing depth. The model must be able to simulate the system dynamics seen in the measured data, before being applied to the LSA-NC.
Once the model has been tested for a
single site, it will be applied to the entire LSA-NC. The Upper
Mississippi has been divided into 202, half degree grid cells.
The derived routing scheme for runoff is shown in
Figure 4.
A modeling dataset for the Upper Mississippi is currently
being assembled. Types of data being collected include:
The location of the NCDC precipitation and surface airways stations can be seen in Figure 5.
The proposed solution technique for the VIC model is similar to that employed by the Simultaneous Heat and Water (SHAW) model. As a preliminary test of this solution technique, the SHAW model was setup and run for the Rosemount site. Results for layer temperatures are shown in Figure 6. Moisture profiles indicate that the SHAW model does not predict as much infiltration through the frozen soils as is observed at the Rosemount site.
Rosemount data was provided by John
Baker, and Egbert Spaans, from the Department of Soil, Water and
Climate, at the University of Minnesota.
Gridded soil property data was provided
by Doug Miller at Pennsylvania State University, Earth System
Science Center.
The SHAW model was provided by Gerald
Flerchinger from the USDA-Agriculture Research Service, Boise,
Idaho.
Flerchinger, G. N., and K. E. Saxton,
"Simultaneous Heat and Water Model of a Freezing Snow-Residue-Soil
System I. Theory and Development", Transactions of the ASAE,
Vol 32, No. 2, March-April, 1989.
Flerchinger, G. N., and K. E. Saxton,
"Simultaneous Heat and Water Model of a Freezing Snow-Residue-Soil
System II. Field Verification", Transactions of the ASAE,
Vol 32, No. 2, March-April, 1989.
Gel'fan, A. N., "Comparison of
Two Methods of Calculating Soil Freezing Depth", Meteorologiya
i Gidrologiya, No. 2, pp. 98-104, 1989.
Kane, Douglas L., and Edward F. Chacho,
"Frozen Ground Effects on Infiltration and Runoff",
Cold Regions Hydrology and Hydraulics, pp. 259-300.
Liang, Xu, Eric Wood, and Dennis Lettenmaier,
"Insights of Surface Heat Fluxes In Energy Balance Equation
From VIC-3L Land-Surface Scheme",
Woo, Ming-ko, "Permafrost Hydrology in North America", Atmosphere and Oceans, Vol 24., No 3, pp 201-234, 1986.
