Towards consistent imaging spectroscopy-based global aboveground biomass retrievals in coastal wetlands across atmospheric states

NASA's Surface Biology and Geology (SBG) mission will provide spectral measurements that can be used to monitor the health of coastal wetlands—among other ecosystems—globally. Aboveground biomass (AGB) plays a critical functional role in coastal wetland stability, as high biomass vegetation contributes to organic matter production, sediment accretion potential, and carbon storage. Here we use data from NASA's Delta-X mission in coastal Louisiana to investigate the impact of atmospheric effects on AGB retrievals. Delta-X employed the Airborne Visible/Infrared Imaging Spectrometer—Next Generation (AVIRIS-NG) instrument to collect spectral data over the active Atchafalaya Basin and the inactive Terrebonne Basin, where AGB varies with hydrogeomorphic dynamics and seasonality. These remotely sensed images were obtained with a coincident herbaceous vegetation survey to estimate AGB with Partial Least Squares (PLS) regression modeling that leverages visible-shortwave infrared spectral reflectances. We use these models to quantify the following:

Total estimated AGB and aboveground carbon stocks across the Atchafalaya and Terrebonne Basins' herbaceous wetlands.

Identification of spectral features and characteristics that are closely associated with vegetation AGB and enable accurate AGB and carbon mapping using imaging spectroscopy.

The impact of atmospheric correction artifacts, particularly water vapor interference, on the quality, robustness, and scalability of these models.

With future imaging spectroscopy missions such as SBG, even tiny spatially-correlated errors from atmospheric artifacts make it difficult to construct consistent global ecosystem trait maps. Our analysis highlights the need for additional corrections over the ~900-980 and ~1110-1190 nm wavelengths where water vapor presence interferes with the characterization of water content within plant canopies. These experiments and corrections demonstrate the potential of AVIRIS-NG data to estimate AGB in wetland ecosystems, and enable consistent analyses in humid atmospheric states often encountered in coastal regions. They also show the tight accuracy tolerances for atmospheric correction in humid atmospheres, and the corresponding need for continual improvements in these protocols.