CCE banner
 
Funded Research

Algorithm Development in Support of SGLI and GOCI Ocean-Color Missions

Frouin, Robert: Scripps Institution of Oceanography, UCSD (Project Lead)

Project Funding: 2011 - 2016

NRA: 2010 NASA: Earth Science U.S. Participating Investigator   

Funded by NASA

Abstract:
We seek support from NASA to conduct two separate, yet related investigations selected by the Japan Aerospace Exploration Agency (JAXA) and the Korean Ocean Research and Development Institute (KORDI) to support the SGLI/GCOM and GOCI/COMS ocean-color missions. The objective of the SGLI investigation is to improve the atmospheric correction of SGLI ocean-color imagery by (1) correcting the top-of-atmosphere imagery for adjacency effects, (2) extending the atmospheric correction to situations of thin clouds, and (3) developing a scheme to deal with situations of absorbing aerosols. The algorithms will be designed to operate with standard algorithms, i.e., in general they will constitute extensions, not alternatives, allowing one to deal efficiently with important issues not resolved by the standard algorithms. The objective of the GOCI investigation is to develop and evaluate an algorithm to estimate photo-synthetically available radiation (PAR) at the ocean surface from GOCI data. The product will cover the Northwestern Pacific Ocean and East Asia seas (GOCI area), with a spatial resolution of 0.5x0.5km and a temporal resolution of one day. The SGLI investigation will provide a de-convolution scheme/processor to correct the SGLI Level 1b imagery for adjacency effects, resulting in improved accuracy and quality of SGLI aerosol and ocean-color products where spatial variability is large (e.g., coastal zone, near sea-ice and clouds). It will also provide effective ways to perform atmospheric correction in the presence of thin or broken clouds and absorbing aerosols, increasing the daily coverage and operational utility of the SGLI ocean-color data. This includes a technique to estimate aerosol altitude by exploiting the coupling between aerosol scattering and oxygen absorption in the oxygen A-band. The GOCI investigation will provide an algorithm to compute daily PAR at the ocean surface that, thanks to the hourly GOCI observations, fully accounts for diurnal variability of clouds. The algorithms and the resulting aerosol ocean-color products, available without restriction and free of charge for research purposes, will benefit NASA and the U.S. Earth Science community in several ways. First, they will complement data sets  generated by current and planned U.S. Earth observation satellites. The increase in coverage obtained by combining data sets will allow a better description of biological phenomena and could lead to new information about temporal variability of biological processes. Second, they will offer the opportunity to compare and evaluate products from U.S. satellite missions. For example, comparisons of PAR estimates from SeaWiFS and MODIS, which do not account for diurnal changes in cloudiness, and GOCI, which do, will allow quantification of biases and definition/implementation of remedial actions. Third, some of the algorithms are applicable to U.S. ocean-color sensors, i.e., to deal with thin clouds and adjacency effects. The algorithm to correct adjacency effects, in particular, could be used systematically to pre-process SeaWiFS and MODIS (and later VIIRS) Level 1b imagery before generating Level 2 products with standard schemes that assume no adjacency effects. Finally, the investigation, by developing and testing on actual satellite imagery concepts/algorithms relevant to the NASA Decadal Survey Tier 2 ACE and GEOCAPE missions, in particular the use of polarization to determine aerosol scattering, measurements in the ultraviolet and oxygen A-band to constrain spectral matching for aerosol absorption, and hourly observations to account for cloud variability in the estimation of PAR, will contribute to the preparation of these missions.

Publications:

Tan, J., Frouin, R., Ramon, D., Steinmetz, F. 2019. On the Adequacy of Representing Water Reflectance by Semi-Analytical Models in Ocean Color Remote Sensing. Remote Sensing. 11(23), 2820. DOI: 10.3390/rs11232820


More details may be found in the following project profile(s):