A Multisensor Perspective on the Radiative Impacts of Clouds and Aerosols

Abstract

AbstractThe launch of CloudSat and Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) in 2006 provided the first opportunity to incorporate information about the vertical distribution of cloud and aerosols directly into global estimates of atmospheric radiative heating. Vertical profiles of radar and lidar backscatter from CloudSat’s Cloud Profiling Radar (CPR) and the Cloud–Aerosol Lidar with Orthogonal Polarization (CALIOP) aboard CALIPSO naturally complement Moderate Resolution Imaging Spectroradiometer (MODIS) radiance measurements, providing a nearly complete depiction of the cloud and aerosol properties that are essential for deriving high-vertical-resolution profiles of longwave (LW) and shortwave (SW) radiative fluxes and heating rates throughout the atmosphere. This study describes a new approach for combining vertical cloud and aerosol information from CloudSat and CALIPSO with MODIS data to assess impacts of clouds and aerosols on top-of-atmosphere (TOA) and surface radiative fluxes. The resulting multisensor cloud–aerosol product is used to document seasonal and annual mean distributions of cloud and aerosol forcing globally from June 2006 through April 2011. Direct comparisons with Clouds and the Earth’s Radiant Energy System (CERES) TOA fluxes exhibit a close correlation, with improved errors relative to CloudSat-only products. Sensitivity studies suggest that remaining uncertainties in SW fluxes are dominated by uncertainties in CloudSat liquid water content estimates and that the largest sources of LW flux uncertainty are prescribed surface temperature and lower-tropospheric humidity. Globally and annually averaged net TOA cloud radiative effect is found to be −18.1 W m−2. The global, annual mean aerosol direct radiative effect is found to be −1.6 ± 0.5 W m−2 (−2.5 ± 0.8 W m−2 if only clear skies over the ocean are considered), which, surprisingly, is more consistent with past modeling studies than with observational estimates that were based on passive sensors.