HR: 08:15h
AN: H11A-02 [PDF]
TI: An Scheme for Application of Non-Uniform Grid Scales for Land Surface Modeling
AU: * Bastidas, L A
EM: luis.bastidas@usu.edu
AF: Civil and Environmental Engineering, Utah State University, 4110 Old Main Hill, Logan, UT 84322-4110
United States
AU: Li, S
EM: shujunli@cc.usu.edu
AF: Civil and Environmental Engineering, Utah State University, 4110 Old Main Hill, Logan, UT 84322-4110
United States
AU: Miller, N L
EM: nlmiller@lbl.gov
AF: Earth Sciences Division, Lawrence Berkeley National Laboratory, 90-1116 One Cyclotron Road, Berkeley,
CA 94720 United States
AB:
Vegetation, topographic, and hydrologic characteristics of the semi-arid Southwest U.S. suggest that, for those environments,
it will be more meaningful to link the inherent heterogeneity and scale of the terrain properties and hydrological processes
with the grid cell size of the numerical representation in land surface models, rather than pursue the traditional
mosaic-type or effective aggregation approach currently used by the state of the art models. We present a multi-scale
procedure for coupling a Land Surface Model (LSM) with atmospheric and subsurface processes that takes into account the
different sensitivity of the models to the process scales and the availability of information for characterizing the
parameter space. This Non-uniform Grid Scheme (NGS) utilizes hierarchical sub-grids with a degree of characterization that
are nested within a grid-matched coarser grids preserving spatial location. Conservation of mass and momentum is maintained
across the grids and aggregated fluxes are computed for the larger scale, which will be bi-directionally coupled to a
fine-scale regional atmospheric model.
A multiple resolution grid structure (500 m to 4 km) defined by the degree of land surface and sub-surface characteristics
over the San Pedro River Basin in Arizona has been setup for the NCEP Noah LSM The model is driven in offline fashion with
outputs from the NCAR Mesoscale Model (MM5). The results are compared to those from uniform grids with 1 and 4 km
resolution. The influence of the parameter values is also assessed using the default values for semi-arid conditions
(uniform for the entire domain) as a benchmark. The grid-dependent non-uniform parameterization is based on a multi-criteria
approach using the newly developed MOSCEM optimization algorithm. The parameter values are derived from remote sensing, and
values obtained from a combination of remote sensing and parameter optimization techniques.
DE: 1833 Hydroclimatology
DE: 1836 Hydrologic budget (1655)
DE: 1878 Water/energy interactions
SC: Hydrology [H]
MN: 2003 Fall Meeting