Principal Investigator (PI): William Rossow, The City College of New York

Investigations of land-atmosphere interactions have usually tended to rely, in part, on a variety of models to describe either atmospheric conditions near the land surface (global or regional weather analyses or forecast models) or the variations of land surface properties with changing weather conditions (land surface models) because some aspects of the near-surface meteorology or land properties were not readily available from observations. However, the detailed behavior of these models, usually run with the atmosphere and surface uncoupled, has not been verified. Earlier projects to pull together global, combined data products, namely NASA's International Satellite Land Surface Climatology Project (ISLSCP) and Global Soil Wetness Project (GSWP), which used some of the ISLSCP product, were focused primarily on assembling a comprehensive set of the atmospheric forcing for land surface models (LSMs).

There are three reasons to revisit this activity. The first reason is that there are a number of conventional and satellite products that were available at the time of the previous studies but not used: in particular, diurnally resolved surface observations of air temperature, humidity and winds, boundary layer cloud properties and the diurnal variations of land surface skin temperatures. Although there were problems with these data products, including sparse spatial coverage and accuracy, the second reason is that there are now available new versions or new data products from new sensors that improve the quality of information available for land surface albedo, skin temperatures, higher-time-resolution precipitation, new soil moisture products, high-vertical-resolution profiles of temperature, humidity and clouds (also land surface properties from Making Earth System Data Records for Use in Research Environments (MEaSUREs) products). The third reason is that in the previous products, many of the key quantities specific to the interaction of the atmosphere and land, namely fluxes of energy and water (even carbon) between them, were represented only by model outputs, not observation-based analyses, because the latter were not available.

Now there are new observation-based products quantifying the diurnal variations of all aspects of the surface radiative, sensible heat and latent heat fluxes soon to be available from Global Energy and Water Exchanges (GEWEX). However, the variety of data products and observations now available still represents a wide range of space-time sampling intervals and space-time coverage; moreover, there are often many alternate products available but without knowledge of their relative merits. To assemble this collection of information into a coherent and physically consistent set that accurately quantifies the global variations of the near-surface atmospheric and the surface properties from diurnal to decadal scales requires a lot of work to reconcile the resolution and coverage differences, to evaluate quality of the multiple products and to project these products into a common space-time framework.

To support and foster investigations of the interaction of the atmosphere and land surface, we propose to construct a multi-variate data product that reconciles the variation scales of these measurements, merges and maps them into a comprehensive description of the near-surface atmospheric properties together with the land surface property variations on diurnal-to-decadal time scales. The atmosphere would be represented by near-surface windspeeds, temperature, humidity, downwelling radiation, boundary layer cloudiness and type. The land surface would be represented by its albedo, infrared and microwave emissivities, skin temperature, upwelling radiation, soil moisture, topography, land type, roughness length and other relevant vegetation properties. We will also explore how much more information about the atmospheric boundary layer can be obtained from available observations to augment the surface meteorology and boundary layer cloudiness information.