Meteorological applications of surface bidirectional reflectance distribution functions retrieved from satellite data

Bidirectional reflectance distribution functions (BRDFs) quantify the manner in which terrestrial surfaces reflect incident solar radiation as a function of the view and illumination geometries. BRDFs allow for directional correction (normalization) of satellite radiance data via anisotropic correction factors to a standard view and illumination geometry. These clear-scene anisotropic correction factors are especially valuable in automated cloud detection models. After using BRDFs to predict clear-scene radiances for a given satellite overpass, actual scene radiance measurements can be compared with predictions. Pixels with reflectance observations that exceed predicted clear-scene values are identified as cloudy. Cloud detection results using this method repeatedly demonstrate high accuracies using near-infrared data collected by the NOAA-14 polar-orbiting satellite during August and September 1995 over New England. However, during this same period BRDF-predicted land-surface reflectances are of insufficient accuracy to estimate the atmospheric aerosol optical properties of cloud-free pixels. This is due primarily to a requirement for separating with high accuracy the atmospheric and surface contributions in top-of-atmosphere (TOA) satellite reflectance observations, a separation that depends strongly and sensitively on the absolute magnitude of the surface reflectance. BRDFs also provide for location-dependent albedos of the earth's surface on synoptic and global scales. Accurate mapping of albedo is required to model energy and mass exchange processes in atmospheric fluid dynamics and global climate models.; The present constellation of meteorological satellites measures upwelling radiances over a large variety of illumination and viewing geometries, and is suitable for estimating BRDFs on global scales. Results demonstrate successfully the ability of contemporary "kernel-driven" BRDF models to quantify the observed directional dependence of solar energy reflected from the earth's surface as observed by the NOAA AVHRR and GOES-Next geostationary satellite imagers. This research is unique in its use of satellite bidirectional reflectance data, especially geostationary data, that have been adjusted for atmospheric scattering effects over large areas and under diverse illumination and viewing geometries. An immediate application of these research results is in the assessment of how well the current generation of land-surface BRDF models can meet the requirements of the NASA Earth Observing System Moderate Resolution Imaging Spectroradiometer (MODIS) sensor for accurate geometric normalization of directional radiance observations and the retrieval of visible and near-infrared albedos on a global scale.