Colloquium: Dr. M. Twardowski | Harbor Branch Oceanographic
In-Person PHYS 401
TITLE: Exploring physics-based algorithms for the NASA PACE mission
ABSTRACT: Semi-analytical algorithms (SAAs) for ocean color remote sensing derive in-water inherent optical properties (IOPs) such as absorption and backscattering from remote sensing reflectance. These IOPs can then be used as proxies for many important biogeochemical constituents in the ocean such as phytoplankton biomass, suspended particle mass, and particulate organic carbon. These constituents then are used as inputs into global models to understand processes such as oceanic productivity, global carbon flux, global warming, and climate change.
Past global ocean color imagers have been limited to only a few spectral bands. SAAs work by minimizing errors in physics-based spectral reconstruction of reflectance, so the spectral band limitations have constrained performance of SAAs. The NASA PACE mission will have a hyperspectral global ocean color imager for the first time with ~80 wavelengths through the visible domain, providing richer spectra with the potential to increase accuracy for physics-based algorithms.
Our algorithm development work for PACE has revolved around a new physics-based remote sensing relationship we published in 2018 called ZTT. Unlike previous algorithms, the full bidirectional geometry of the problem, i.e., the specific solar zenith and viewing geometry, is integrated in ZTT, which is expected to improve performance. Uncertainties assessed with existing multi-spectra data sets are at least as good as current state-of-the-art algorithms. We are currently preparing to apply ZTT to future in-water hyperspectral data sets and to data from the PACE imager when launched in Jan 2024.