A precise knowledge of radiation budget is essential in our understanding of Earth's climate and weather. Accurate measurements of fluxes at the top of the atmosphere (TOA) are critical for obtaining that knowledge. Generally, satellite TOA longwave (LW) radiances are converted into fluxes using limb-darkening models that account for the variation of radiance with viewing zenith angle (VZA). The most recent limb-darkening models are dependent on geotype and cloud amount, phase, and optical depth as well as precipitable water, lapse rate and surface emissivity. Using limited matched satellite imager data, Minnis and Khaiyer (2000) showed that shadowing effects cause azimuthal variations of infrared brightness temperatures over land that may impact estimates of surface temperature and perhaps the TOA LW flux.They concluded that local topography either as landforms or vegetation can cause these azimuthal variations and that they vary with solar zenith angle or time of day. The Clouds and the Earth's Radiant Energy System (CERES) instrument launched aboard the Tropical Rainfall Measuring Mission (TRMM) has provided a dataset, the Single Scanner Footprint TOA/Surface and Clouds (SSF) data, for studying this azimuthal phenomenon and perhaps providing a means for correcting fluxes for its impact. TRMM is in a 35° precessing orbit, limiting the coverage to 37°S - 37°N. The CERES SSF data include scanner data taken out a VZA of 70°. During every third day, the CERES scanner operated in the Rotating Azimuthal Plane Scan (RAPS) mode. The RAPS data provide views of a given region from multiple angles during a given overpass. This study uses the 89 days of RAPS data available from January to August 1998 to derive radiance patterns as a means of providing an initial correction of the azimuthal variation after removing the limb-darkening effect. Because there is insufficient angular sampling on a regional basis, the CERES RAPS LW (5-100 µm) and infrared window (8-12 µm) radiances are averaged in angular bins as a function of time of day, surface type, topography, and cloud fraction.. The solar day is divided into 4 equal spans to accommodate the variability of solar zenith angle with latitude. Initial results show no significant azimuthal dependence over ocean and maximum impacts over rough land and for partly cloidy skies over land.