Lee, Zhongping (Ping): University of Massachusetts Boston (Project Lead)
Project Funding:
2015 - 2018
NRA: 2013 NASA: PACE Science Team
Funded by NASA
Abstract:
Inherent optical properties (IOPs) play a key role in modulating the color of oceanic and coastal waters, and provide the critical link to infer the concentrations of constituents in the upper water column. In the recent decade, various algorithms, both empirical and semi-analytical, have been developed for the retrieval of IOPs from ocean color, which is measured by the spectrum of remote-sensing reflectance (Rrs, sr-1). These algorithms, in particular the algebraic algorithm (QAA) and the spectral optimization algorithms (e.g., GSM, GIOP), have been implemented to retrieve various IOPs from Rrs measured by SeaWiFS and MODIS, thus providing prototype IOP products at a few bands for the global oceans. The quality of these products, however, depends on the validity of the spectral shapes of the IOPs (SSIOP) used in these semi-analytical algorithms, but the determination of the SSIOP from remote sensing is far from mature. More importantly, the PACE mission will provide hyperspectral Rrs of the global oceans, thus the derivation of hyperspectral IOP products will demand accurate estimation of hyperspectral SSIOP. The improvement of the SSIOP estimation and the determination of hyperspectral IOP algorithms for PACE will depend critically on a robust hyperspectral Rrs-IOPs dataset, but there is no such a dataset yet for the community to use. To fill this void, with an ultimate goal to maximize the IOP products from the PACE hyperspectral measurements, we propose to 1) compile a hyperspectral Rrs-IOPs dataset from field measurements; 2) improve the estimation of SSIOP from ocean color; 3) revise the QAA and HOPE (a hyperspectral optimization algorithm) to take advantage of the hyperspectral and UV measurements offered by PACE, with a goal to expand the current IOP products to include information beyond chlorophyll-a (e.g., the absorption coefficients of chlorophyll-b,-c, and phycocyanin); and 4) test and evaluate these semi-analytical algorithms with HICO measurements.
Outcomes from this effort will be fourfold: 1) a hyperspectral Rrs-IOPs dataset with closure for the community to use, 2) improved estimation of SSIOP from ocean color to benefit all semi-analytical algorithms, 3) revised QAA and HOPE to derive hyperspectral IOP products, and 4) experience with HICO in processing and storing hyperspectral image products. These results will provide desired tools and knowledge for processing hyperspectral measurements by PACE, and contribute to 'consensus and community- endorsed paths forward for the PACE sensor(s)'.
More details may be found in the following project profile(s):
