Using imaging spectroscopy (hyperspectral imaging), we sought to assess the effects of image pixel resolution, size of mapping windows composed of pixels, and number of spectral species assigned to pixels on the capacity to map plant beta diversity using the biodivMapR algorithm, in support of the planned NASA Surface Biology and Geology (SBG) satellite remote sensing mission. BiodivMapR classifies pixels as spectral species, then calculates beta diversity as dissimilarity of spectral species among mapping windows each composed of multiple pixels. We used NEON airborne 1 m resolution hyperspectral images collected at three sites representing native longleaf pine ecosystems in the southeastern U.S. and aggregated pixels to sizes ranging from 1-90 m for comparative analyses. Plant community composition was groundtruthed. Results show that the capacity to detect plant beta diversity decreases with fewer pixels per mapping window, such that pixel resolution limits the size of mapping windows effective for representing beta diversity. Mapping window size in turn limits the spatial resolution of beta diversity maps composed of mapping windows. Assigning too few pixels per window, as well as assigning too many spectral species per image, results in overestimation of dissimilarity among locations that have plant species in common. This overestimation undermines the capacity to contrast mapping window dissimilarity within versus among community types and reduces the information content of beta diversity maps. These results demonstrate the advantage of maximizing spatial resolution of hyperspectral imaging instruments on the anticipated NASA SBG satellite mission and similar remote sensing projects.

Seagrass meadows are effective carbon sinks due to their high primary production and sequestration in sediments. However, methane (CH4) fluxes can partially counteract their carbon sink capacity. Here, we measured diffusive sediment-water and air-sea CO2 and CH4 fluxes in a coastal embayment dominated by Posidonia oceanica in the Mediterranean Sea. High resolution timeseries observations revealed large spatial and temporal variability in CH4 concentrations (2 to 36 nM). Higher emissions were observed in an area with dense seagrass meadows. A 6 − 40% decrease of CH4 concentration in the surface water around noon indicates that photosynthesis likely limits CH4 fluxes. Sediments were the major CH4 source as implied from radon (a natural porewater tracer) observations and evidence for methanogenesis in deeper sediments. CH4 sediment-water fluxes (0.1 ± 0.1 − 0.4 ± 0.1 µmol m-2 d-1) were higher than average water-air CH4 emissions (0.12 ± 0.10 µmol m-2 d-1), suggesting that dilution and CH4 oxidation in the water column could reduce net CH4 fluxes into the atmosphere. Overall, relatively low air-sea CH4 fluxes at this likely represent net emissions from subtidal seagrass habitats sites, which are not influenced by nearby allochthonous CH4 sources. The local CH4 emissions in P. oceanica offset less than 1% of the carbon burial in sediments (142 ± 69 g CO2eq m-2 yr-1). Combining our results with earlier observations in other seagrass meadows worldwide reveals that global CH4 emissions within seagrass meadows only offset a small fraction (<2%) of carbon sequestration in sediments.

Developing actionable early detection and warning systems for agricultural stakeholders is crucial to reduce the annual \$200B USD losses and environmental impacts associated with crop diseases. Agricultural stakeholders primarily rely on labor-intensive, expensive scouting and molecular testing to detect disease. Spectroscopic imagery (SI) can improve plant disease management by offering decision-makers accurate risk maps derived from Machine Learning (ML) models. However, training and deploying ML requires significant computation and storage capabilities. This challenge will become even greater as global scale data from the forthcoming Surface Biology \& Geology (SBG) satellite becomes available. This work presents a cloud-hosted architecture to streamline plant disease detection with SI from NASA’s AVIRIS-NG platform, using grapevine leafroll associated virus complex 3 (GLRaV-3) as a model system. Here, we showcase a pipeline for processing SI to produce plant disease detection models and demonstrate that the underlying principles of a cloud-based disease detection system easily accommodate model improvements and shifting data modalities. Our goal is to make the insights derived from SI available to agricultural stakeholders via a platform designed with their needs and values in mind. The key outcome of this work is an innovative, responsive system foundation that can empower agricultural stakeholders to make data-driven plant disease management decisions, while serving as a framework for others pursuing use-inspired application development for agriculture to follow that ensures social impact and reproducibility while preserving stakeholder privacy.