|Kustas, William - Bill|
Submitted to: International Association of Hydrological Science
Publication Type: Proceedings
Publication Acceptance Date: 6/8/2000
Publication Date: 9/25/2001
Citation: Interpretive Summary: Regional-scale calculations of energy exchange between the land and atmosphere inherently entail spatial aggregation of remotely sensed data, whether at the measurement level (e.g., sensor resolution) or at the computational level (e.g., numerical model grid size). The spatial and temporal variability of soil moisture and surface temperature found within typical remote sensing footprints demonstrates the fact that averaged values taken from remote sensors mask a range of scales of variability that may be important for determining energy exchange across the land-atmosphere boundary. Surface energy balance models applied at various scales using simple averaging rules have shown that errors resulting from scaling up can be relatively minor, but a more comprehensive problem is the way in which the heterogeneity induces non-local atmospheric circulation. In this study, we apply a large-eddy simulation model which simulates atmospheric circulation to remotely sensed boundary conditions in order to examine the effects of scale and modeled boundary layer dynamics on energy exchange. Results suggest that, under certain conditions, variability in surface conditions does not significantly affect atmospheric flow.
Technical Abstract: Remotely sensed surface temperatures from the Monsoon 1990 experiment in Arizona were used as boundary conditions for a large-eddy simulation (LES) model in order to compute surface sensible heat fluxes (H). The LES model takes into account the dynamic effects of surface heterogeneity. Model runs were conducted over boundary condition resolutions of 200m, 400m, and 800m. Regional averages of H determined by the LES model did not vary significantly over the differing scales of surface resolution and were similar to values derived by an "offline" method in which average meteorological variables of air temperature and wind speed were used as inputs. However, localized differences in H, calculated by the two methods over the surface, were as high as 37 W/m^2, reflecting the surface heterogeneity-induced feedbacks present in a dynamic atmospheric boundary layer as captured by the LES model.