The Soil Moisture-Atmosphere Coupling Experiment (SMACEX) was conducted June 15-July 13, 2002 in the Walnut Creek Watershed near Ames, Iowa -- an agricultural region dominated by corn and soybean production. A primary goal in SMACEX was to study the role of heterogeneity in soil moisture and vegetation cover in influencing land-atmosphere exchanges of energy, water and carbon over a range in spatial scales. During the experiment, fluxes were collected continuously at 14 eddy covariance towers distributed across the watershed, and periodically along tracks over flown by the NRC Twin Otter atmospheric research aircraft. Together, tower and aircraft observations sampled flux footprints on the order of 100 to 1000 m. Surface vegetation and temperature observations were collected in-situ, at the 1m pixel resolution with the Utah State University Piper Seneca remote sensing aircraft, and at 30 to 5000 m pixel resolutions with the Landsat, MODIS, AVHRR and GOES satellites.

This multi-scale dataset, representing a mixture of point, linear, and gridded coverages, will be synthesized and aggregated and used to evaluate spatial scaling techniques and assumptions inherent in turbulent transport modeling. This paper will compare flux measurements aggregated to the watershed scale with predictions from the regional-scale Atmosphere-Land Exchange Inverse (ALEXI) model -- a coupled land surface-atmospheric boundary layer model based on satellite estimates of vegetation cover and surface radiometric temperature change. During the SMACEX experiment, ALEXI was run daily on a 5 km grid including Iowa and environs, with the study region occupying approximately 20 grids cells within the modeling domain. Modeled flux "snapshots" will be compared with large scale spatial heterogeneity detected in the aggregated ground, tower and aircraft observations, and with patterns in antecedent precipitation events.