Aerosols are a major atmospheric variable influencing both the transfer of radiative energy, and the conversion of water vapor into cloud droplets and raindrops. As such, most GCMs are now incorporating aerosol parameters and physical processes linking aerosols with the energy and water cycles so that aerosol direct and indirect effects on climate can be computed. Over the years, numerous global aerosol products have been generated from past (e.g. AVHRR, TOMS, POLDER) and current satellite sensors (e.g. MODIS, MISR, CALIPSO), and more are expected from future sensors (e.g. APS, OMPS, and VIIRS). As the number of aerosol products increases, discrepancies among the products also increase. Data users, especially modelers, are confronted with an ever-increasing challenge to decide which product to use and how much uncertainty exists in the different products.
Under the auspices of the Global Energy and Water Cycle Experiment (GEWEX) and the GEWEX Radiation Panel (GRP), working groups have been established to assess several key products: clouds, precipitation, radiation and aerosols. The GEWEX Aerosol Product Assessment (GAPA) was recently established and is led by Z. Li and X. Zhao from the University of Maryland (UMD), with team members representing all major groups producing global aerosol products.
The photo shows a partial list of the members attending the first kick-off meeting held from September 14-15, 2006, in College Park Maryland, USA.
First row from left: A. Higurashi (NIES, Japan), C. Hsu (GSFC/NASA), L. Remer (GSFC/NASA), M. Chin (GSFC/NASA), O. Torres (UMD/NASA); Centre row from left: R. Kahn (JPL/NASA), D. Diner (JPL/NASA), M. Wang (NOAA), I. Laszlo (NOAA), D. Winker (LaRC/NASA), Z. Obradovic (Temple University); last row: H. Maring (HQ/NASA), B. Holben (GSFC/NASA), S. Vucetic (Temple University), Z. Li (UMD), S. Tsay (GSFC/NASA), C. Ichoku (UMD/NASA), T. Zhao (UMD/NOAA), M. Schulz (CEA, France), M. Mishchenko (GISS/NASA)
The objectives of the working group are to:
The workshop was charged to:
During the two-day meeting, aerosol product development teams provided updates of various products (AVHRR-GACP, AVHRR-PATMOS, AVHRR-NIES, TOMS, SeaWiFS, MODIS, MODIS-Deep Blue, MISR, CALIPSO) and detailed descriptions of the retrieval procedures. Such information is instrumental in understanding the causes for any discrepancies. A couple of comparative studies/analyses were presented that helped gain some insight into the impact of different assumptions made in the retrieval algorithms (e.g. aerosol size distribution, refractive index, surface spectral albedo ratio, etc.) on the differences in the retrieved aerosol quantities. It was noted that the MODIS aerosol product is upgraded in its latest release (version 5) to correct some errors in the AOD retrieval over land. New advances in aerosol remote sensing technology were introduced, such as the CALIPSO mission and the deep-blue method. An upcoming newer version (version 2) of the widely used ground-based AERONET product is expected to produce more realistic aerosol size distributions and single scattering albedos. Issues associated with spatial and temporal matching between satellite and ground aerosol retrievals were discussed. A new computationally more efficient retrieval approach based on machine-learning techniques appears to offer some guidance toward the development of a highly integrated algorithm that can incorporate all available satellite data. In parallel with satellite remote sensing of aerosols, model simulations of global aerosol distributions have made remarkable progress. Over land, the quality of the AODs simulated by some models is at par with that of remotely sensed AODs, while the latter is superior over oceans. This suggests the importance of exploiting the synergy between model simulations and remote sensing. Joint experimental studies are planned to quantify and eventually remove discrepancies between the various aerosol products, which will lead to a coherent aerosol product derived from all available sensors that is compatible with the long-term historical product of the GACP.