14:00–14:15 | ABSTRACT 1361
QUANTIFYING PIGMENTS AND DIATOM BIOMASS FROM PACE HYPERSPECTRAL REFLECTANCE: A GLOBAL-SCALE EVALUATION
Broad-scale views of phytoplankton community structure are essential for detecting how the ocean is responding to climate change. Because phytoplankton groups differ in physiology and ecological roles, regional-to-global scale estimates of accessory pigments and diatom carbon biomass can reveal shifts in community composition, food-web support, and carbon export potential. Here we present a global evaluation of three surface-ocean algorithms designed for next-generation ocean color observations: a new model to quantify diatom carbon biomass, and two algorithms based on prior work to estimate phytoplankton accessory pigment concentrations. Model development draws on a decade of ship-based observations spanning diverse ocean conditions. Independent model assessment uses a separate, recent dataset of coincident ship measurements of pigments and diatom carbon biomass collected during the first two years following the launch of NASA’s PACE mission (2024-2026), enabling direct evaluation against PACE hyperspectral reflectances. This independent validation yields errors and accuracy metrics that are either similar to, or an improvement upon, previous model development and baseline model evaluation. We show application of the pigment and diatom biomass models to PACE, which reveals coherent, previously unresolvable broad-scale patterns in phytoplankton community structure, and demonstrates the novel value of hyperspectral ocean color for tracking ecosystem change.
Alison Chase, University of Washington Applied Physics Laboratory, [email protected], https://orcid.org/0000-0002-1862-9822
Sasha Kramer, Boston University, [email protected]
Max Danenhower, University of Washington Applied Physics Laboratory, [email protected]
Ivona Cetinic, NASA Goddard Space Flight Center, [email protected]
Peter Gaube, University of Washington Applied Physics Laboratory, [email protected]
Guillaume Bourdin, University of Maine, [email protected]
Nils Haentjens, University of Maine, [email protected]
Emmanuel Boss, University of Maine, [email protected]
Lee Karp-Boss, University of Maine, [email protected]
