Mannino, Antonio: NASA GSFC (Project Lead)
Project Funding:
2010 - 2013
NRA: 2009 NASA: The Science of Terra and Aqua
Funded by NASA
Abstract:
The primary objective of the proposed work is to conduct algorithm development to introduce new MODIS and SeaWiFS data products for Colored Dissolved Organic Matter (CDOM) absorption coefficient (a_CDOM), CDOM spectral slope (S_CDOM) and Dissolved Organic Carbon (DOC) that will yield new Algorithm Theoretical Basis Documents (ATBD) for peer-review. For the a_CDOM and S_CDOM algorithms, we will expand our existing empirical band-ratio algorithms validated for the coastal ocean along the northeastern U.S. (Mannino et al. 2008) to the global ocean by applying our own datasets and measurements from other sources such as the NOMAD SeaBASS database. We will also apply several machine learning approaches, including neural network analysis, support vector machines and gaussian process models, to develop global algorithms for DOC, a_CDOM and S_CDOM using our datasets, the Hansell/Carlson DOM data collection, the NOMAD SeaBASS database and other data sources. The machine learning approaches will be trained using currently available field data and coincident MODIS and SeaWiFS remote sensing reflectances (Rrs) and MODIS sea-surface temperatures. Rigorous quantitative assessment of the machine learning and band-ratio algorithms will be applied to determine the best performing algorithms for our study. The global MODIS-Aqua, -Terra and SeaWiFS ocean color time series for each product will be computed at monthly intervals from 1997-2012 to examine regional and global long-term trends in DOC, a_CDOM, and S_CDOM.
2011 NASA Carbon Cycle & Ecosystems Joint Science Workshop Poster(s)
- Satellite-derived distributions of CDOM, DOC and particulate organic matter along the northeastern U.S. continental margin
-- (Antonio Mannino, Stanford B. Hooker, Kimberly Hyde, Michael Novak)
[abstract]
More details may be found in the following project profile(s):
