HR: 0830h
AN: A21A-14    [Abstracts]
TI: Determining Small Scale Albedos Using High Resolution Multiangle Satellite Measurements
AU: * Markowski, G R
EM: gregm@jord.jpl.nasa.gov
AF: Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109 United States
AU: Davies, R
EM: roger@mail-misr.jpl.nasa.gov
AF: Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109 United States
AB: Current satellite short-wave (SW) albedo measurements, such as CERES's, have only a broad spatial resolution and cannot by themselves accurately measure reflectance (roughly solar "forcing") on small space and time scales. The major difficulty is that earth's surface reflectivity, including the atmosphere and clouds, is substantially anisotropic. However, accurate regional and time-dependent albedos are needed for studying causes of climate variability and change, and improving models from global to at least cloud resolving scales. A first step to obtain these albedos, for which we show results, is to accurately relate (and verify) the high resolution spatial and angular surface narrow-band MISR (Multi-Angle Imaging Spectroradiometer) radiance measurements aboard the Terra satellite to coincident total shortwave broadband (SWB) low resolution measurements from the onboard CERES instrument. Because MISR measures radiance of the same points along an orbital swath, it becomes possible to check and improve Angular (reflection) Distribution Models (ADMs) at small scales (< 1 km). The ADMs can later be used to invert a measured angular radiance to a local albedo. The difficulty lies in obtaining accurate ADMs for earth's highly varied surface and lighting conditions. We show prediction accuracy examples of CERES SWB vs. single and multiple band MISR data regressions. We include view angle dependence (9 angles: nadir plus 26, 46, 60, and 70 degrees fore and aft) and show improved accuracy when surface data, e.g., solar zenith and scattering angle, and surface type are included. In many cases, we predict angular (bidirectional) reflectance to ~ 0.01, or about 10 watts/sq m in irradiance. We also show examples of "difficult" scene types, such as varying levels of broken clouds, where accuracy degrades by a factor of ~2.
DE: 3399 General or miscellaneous
SC: Atmospheric Sciences [A]
MN: 2005 Joint Assembly