Quantifying CERES Angular Distribution Model (ADM) Scene Type Uncertainty with the CALIPSO, CloudSat, CERES, MODIS (C3M) Product

Angular distribution models (ADMs) are used to estimate top-of-atmosphere (TOA) longwave (LW) and shortwave (SW) fluxes from radiances directly measured by the Clouds and Earth's Radiant Energy System (CERES) instrument on board the Terra and Aqua spacecraft. Different ADMs are used for clear and cloudy scenes, and for different surface types. Cloud properties within each CERES satellite footprint are determined using simultaneous retrievals from the Moderate Resolution Imaging Spectroradiometer (MODIS). Since MODIS is a passive sensor, retrievals sometimes miss clouds that are optically thin, and retrievals at night are particularly challenging without information from the visible channels. Active sensors such as the CALIPSO lidar and CloudSat cloud profiling radar can detect cloud properties more accurately than passive sensors can. The CALIPSO, CloudSat, CERES, MODIS (C3M) product provides coincident CERES-MODIS cloud property retrievals for each CERES footprint that CloudSat and CALIPSO pass over, along with the cloud properties retrieved by CloudSat and CALIPSO. When the TOA fluxes determined using the C3M cloud properties are compared to fluxes determined using the CERES-MODIS cloud properties, the difference can be considered a measure of uncertainty due to errors in scene identification. Preliminary results show that LW flux estimates using the C3M cloud properties tend to range between 0.1 W m^-2 higher over daytime snow and ice scenes and 0.9 W m^-2 lower over daytime land scenes than flux estimates obtained using CERES-MODIS cloud properties. Uncertainties in SW fluxes due to scene identification error will also be explored.