AEROSOL OPTICAL DEPTH OVER OCEANS - HIGH SPACE-RESOLUTION AND TIME-RESOLUTION RETRIEVAL AND ERROR BUDGET FROM SATELLITE RADIOMETRY

Citation
R. Wagener et al., AEROSOL OPTICAL DEPTH OVER OCEANS - HIGH SPACE-RESOLUTION AND TIME-RESOLUTION RETRIEVAL AND ERROR BUDGET FROM SATELLITE RADIOMETRY, Journal of atmospheric and oceanic technology, 14(3), 1997, pp. 577-590
Citations number
39
Categorie Soggetti
Metereology & Atmospheric Sciences","Engineering, Marine
ISSN journal
07390572
Volume
14
Issue
3
Year of publication
1997
Part
2
Pages
577 - 590
Database
ISI
SICI code
0739-0572(1997)14:3<577:AODOO->2.0.ZU;2-U
Abstract
A method to retrieve aerosol vertical optical depth at 0.64 mu m from satellite observations of cloud-free scenes over oceans with high spat ial resolution (similar to 1 degrees) and instantaneous temporal resol ution is described and evaluated. The observed radiance is treated as the linear sum of contributions to path radiance by different scatteri ng processes in the atmosphere-ocean system. This treatment allows exa mination of errors in the retrieved vertical aerosol optical depth con tributed by each process and approximation. Random error in retrieved aerosol optical depth is typically 0.03. The systematic error due to a bsolute calibration uncertainty in the measured radiance is 0.01. The largest errors and biases are due to radiative transfer approximations (+22%) and assumptions regarding aerosol microphysical and optical pr operties (-20%). The latter errors, which are due to the optical prope rties (e.g., phase function), vary systematically with latitude and se ason because of the variation of the mean observing geometry. This met hod is applied to Advanced Very High Resolution Radiometer global area coverage data, and example maps of aerosol optical depth are presente d for specific dates in July and October 1986. The aerosol optical dep th derived from the satellite data is suitable for examining large aer osol signatures by instantaneous comparison of the amplitude and locat ion of aerosol plumes with model predictions based on meteorological c onditions at and preceding the time of observation.