The Uncertainty in MODIS Aerosol Optical Depth Retrievals in the Vicinity of Clouds
Abstract
The dark-target MODIS aerosol product has been validated extensively against ground based AERONET data. About 68% of the ocean retrievals match AERONET data to within 0.03± 10%. Since the MODIS retrievals follow a LUT approach, there are several sources of uncertainty associated with the retrievals. Data filtering is another source of uncertainty and it includes filtering out clouds from the retrieval area. Any residual cloud contamination or cloud adjacency effects can result in biased retrievals. A meaningful use of AOD uncertainties (e.g in data assimilation, aerosol direct and indirect effect studies) requires information on these effects on AOD retrievals. We explore the possibilities of quantifying the uncertainty in AOD in the vicinity of clouds by performing multiple retrievals at varying distances from clouds. We use our cloud distance product for this. For example, when we limit retrievals to pixels that are > 20 pixels from nearest cloud and compare them against standard retrievals, our preliminary results show that the differences can range from 0.01 to 0.05. We continue to explore the differences over global oceans and will present the findings from this study.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.A51G2155P
- Keywords:
-
- 0305 Aerosols and particles;
- ATMOSPHERIC COMPOSITION AND STRUCTURE;
- 0320 Cloud physics and chemistry;
- ATMOSPHERIC COMPOSITION AND STRUCTURE;
- 0365 Troposphere: composition and chemistry;
- ATMOSPHERIC COMPOSITION AND STRUCTURE;
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSES