Photosynthesis-irradiance (PI) relationships are important for phytoplankton ecology and quantifying carbon fixation rates in the environment. However, the parameters of PI relationships are typically unknown across space and time. Here we use machine learning, satellite remote-sensing, and a global database of in-situ PI relationships to build models that predict PI parameters globally as a function of satellite-observed variables. Using only surface light, temperature, and chlorophyll, we achieve an R2 of 76% for predicting photosynthesis rates at saturating light (𝑃Bmax) and an R2 of 58% for predicting the light saturation parameter (𝐸k). Predictability is maximized when averaging covariates over monthly timescales, indicating that environmental history and community turnover are important for predicting PI relationships. These results will help improve satellite-based primary production models, quantify emergent timescales in photosynthetic communities, and provide global estimates of photosynthetic parameters that can be used in plankton ecology and biogeochemistry models.

The Transition Zone Chlorophyll Front (TZCF) is a dynamic region of elevated chlorophyll concentrations in the Northeast Pacific that migrates from a southern winter (February) extent of approximately 30°N to a northern summer (August) extent of approximately 40°N. The transition zone has been highlighted as important habitat for marine animals and fisheries. We re-examine the physical and biological drivers of seasonal TZCF variability using a variety of remote sensing, reanalysis, and in-situ datasets. Satellite-based remote sensing estimates of chlorophyll and carbon concentrations show that seasonal TZCF migration primarily reflects a seasonal increase in the chlorophyll to carbon ratio, rather than changes in phytoplankton carbon. We use our data compilation to demonstrate how the seasonality of light and nutrient fluxes decouple chlorophyll and carbon seasonality at the transition zone latitudes. Seasonal mixed-layer-averaged light availability is positively correlated with carbon and negatively correlated with chlorophyll through the transition zone, while climatological nitrate profiles show that chlorophyll to carbon ratios are facilitated by wintertime nitrate entrainment. These empirical results are consistent with physiological data and models describing elevated chlorophyll to carbon ratios in low light, nutrient-replete environments, demonstrating the importance of latitudinal structure in interpreting seasonal chlorophyll dynamics at the basin scale.

The North Atlantic phytoplankton bloom depends on a confluence of environmental factors that drive transient periods of exponential phytoplankton growth and interannual variability in bloom magnitude. I analyze interannual bloom variability in the North Atlantic via extreme value theory where the Generalized Extreme Value Distribution (GEVD) is fitted spatially to annual maxima of satellite-measured surface chlorophyll. I find excellent agreement between the observed distribution of interannual bloom maxima and those predicted from the GEVD. The spatial distribution of fitted GEVD parameters closely follows basin bathymetry where the largest extremes and heaviest distribution tails are found on the continental shelves and slopes. Trend analyses suggest weak evidence for changes in GEVD parameters, despite regional trends in mean chlorophyll levels and sea surface temperature. These results provide a framework to quantify interannual bloom variability and call for further work examining how extreme blooms propagate through food webs and contribute to carbon export.

Carbonyl sulfide (COS) was measured in firn air collected during seven different field campaigns carried out at four different sites in Greenland and Antarctica between 2001 and 2015. A Bayesian probabilistic statistical model is used to conduct multi-site inversions and to reconstruct separate atmospheric histories for Greenland and Antarctica. The firn air inversions cover most of the 20 century over Greenland and extend back to the 19 century over Antarctica. The derived atmospheric histories are consistent with independent surface air time-series data from the corresponding sites and the Antarctic ice core COS records during periods of overlap. Atmospheric COS levels began to increase over preindustrial levels starting in the 19 century and the increase continued for much of the 20 century. Atmospheric COS peaked at higher than present-day levels around 1975 CE over Greenland and around 1987 CE over Antarctica. An atmosphere/surface ocean box model is used to investigate the possible causes of observed variability. The results suggest that changes in the magnitude and location of anthropogenic sources have had a strong influence on the observed atmospheric COS variability.