Benthic foraminifera are single-celled organisms inhabiting all marine environments. Despite their high tolerance to oxygen depletion, the prevailing hypothesis anticipates a reduction in their diversity in permanently oxygen-depleted environments, including oxygen minimum zones. Here we re-evaluate diversity and study the endemism of benthic foraminifera in the eastern Pacific, an oceanic area hosting the largest permanently oxygen-depleted waters of the world. We focus our analysis on the oxygen-depleted bottom waters and study how they compare with well-oxygenated waters. By utilizing extensive datasets of quantitative information on benthic foraminifera assemblages obtained from morphological traits, we present evidence that challenge traditional viewpoints. Contrary to prior inferences primarily derived from regional studies, our findings reveal that the median diversity (species richness and the Shannon index) calculated on both, living and dead assemblages does not decrease in the most oxygen-depleted bottom-waters. The analysis of unique (endemic) and shared species shows a divide between the neritic-bathyal oxygen-depleted bottom waters with low number of endemic species, and the well-oxygenated abyss hosting high number of unique species. These patterns could be explained by the long-term species exchange in the upper ocean and the isolation of the lower ocean.

To investigate the spatial distributions and determinants of nutrient concentrations, we measured NO3−+NO2−, PO43−, and Si(OH)4 concentrations in the eastern Indian Ocean sector of the Antarctic Ocean (80 − 150°E, south of 60°S) between December 2018 and February 2019. In the region influenced by the Antarctic Circumpolar Current, nutrient concentrations were increased by nutrients supplied from the deep layer and by organic matter decomposition and remineralization within the seasonal pycnocline after the development of strong stratification. Strong stratification also enhanced phytoplankton growth and nutrient consumption by photosynthesis. In contrast, in the subpolar region, nutrient concentrations were increased by nutrients supplied by brine discharged during sea ice formation and decreased by dilution with sea ice meltwater. Although high salinity in the surface and subsurface layers corresponded well to upwelling areas around subpolar subgyres, high salinity was not necessarily correlated with nutrient concentrations. We estimated primary production both from in situ nutrient data and from satellite-acquired chlorophyll-a data. According to both estimation methods, primary production was high in the subpolar region, especially around 120 − 130°E. However, nutrient-based estimation also showed high production in coastal areas where, because of sea ice and cloud cover, estimation based on satellite data was not possible. To understand primary production in seasonal ice areas, the best estimation method should be selected for the research goals or multiple methods should be used in combination.

Operational analysis and forecast products of shelf sea biogeochemistry often lack reliable information on uncertainty. This is problematic, as good quality uncertainty information is both requested by the product end-users and essential for data assimilation. To address this problem we developed a quality-assessed ensemble representation of many leading sources of uncertainty in a coupled marine physical-biogeochemical model of the North-West European Shelf. Based on these ensembles we have estimated the uncertainty of several marine ecosystem health indicators (MEHIs), acting as proxies for biological productivity, phytoplankton community structure, trophic fluxes, deoxygenation, acidification and carbon export. We have also evaluated how observable these MEHIs are from the most widely available observations of total chlorophyll (mostly from the surface), highlighting those MEHIs and locations that need to be better monitored. Our results show that the most uncertain and the least observable MEHI is the phytoplankton community composition, highlighting the value of its observations (and their assimilation) particularly in the UK regional waters. We demonstrate that daily operational estimates of the other MEHIs, produced by the Met Office, are fairly well constrained. We also quantify how much MEHI uncertainties are reduced when one substantially coarsens the MEHI spatial and temporal resolution, as in the global and/or climate applications.

The invasive European green crab (Carcinus maenas) was first detected on the US west coast around 1989 and has expanded its range northward from central California to southern Alaska. The eastern Salish Sea was initially thought to be protected from invasion by the dominant seaward surface current in the Strait of Juan de Fuca (SJdF). However, this “oceanographic barrier” has been breached as established green crab populations have been detected in the eastern Salish Sea in recent years. Here we carried out particle-tracking simulations to understand possible natural pathways of green crab larvae invading the eastern Salish Sea. Both diel vertical migration and temperature-dependent mortality were considered in these simulations. Our results suggest that green crab larvae from the outer coast (outside the Salish Sea) and Sooke Basin (in SJdF) could be carried into the eastern Salish Sea in a narrow time window during the later cold season (esp. in March) when frequent flow reversals in SJdF occur and the seasonally rising water temperature becomes relatively favorable for green crab larvae. The major pathway for larvae to reach the eastern Salish Sea is along the southern coast of SJdF. The probability of live larvae reaching the eastern Salish Sea is highly sensitive to water temperature. Sensitivity simulations indicate that a temperature increase of 0.5–1 °C would double or quadruple the probability of successful arrival in the eastern Salish Sea. This suggests that invading green crabs might have taken advantage of the mild winter conditions in recent warm years. Our results also suggest that the warming climate in the near future may facilitate green crab larval exchange across the Salish Sea.

The deep sea is the largest biome on earth, but one of the least studied despite its critical role in global carbon cycling and climate buffering. Deep-sea organisms largely rely on particulate organic matter from the surface ocean for energy – these organisms in turn play critical roles in energy transport, transformation, storage, and sequestration of carbon. Within the deep sea, submarine canyons are amongst the most complex and dynamic environments in our oceans, where varied morphology, powerful currents, and variable nutrient conditions influence the distribution of species and transport of organic material throughout the water column and the seafloor. Significant habitat heterogeneity provides ideal substrates for cold-water corals, making submarine canyons of interest to conservation and management. However, how these and other topographic features in the deep ocean influence energy flow and trophic pathways is poorly known. Thus, submarine canyons serve as model systems to track variability in organic material flux and consequential utilization and assimilation by the benthos. In this study, we used an extensive stable isotope dataset to examine food-web structure in Baltimore and Norfolk submarine canyons and compared them to their adjacent slopes located along the U.S. Atlantic margin. Linear models were used to construct geospatially-explicit consumer isoscapes that predicted variation in carbon and nitrogen isotopes across the canyon-slope seascape, providing a predictive map from which to test hypotheses on the distribution and flow of energy resources, relevant to understanding whole community function. Communities were composed of isotopically diverse feeding groups with photosynthetically-derived organic carbon providing the basal food resource. Canyon communities were distinct from the slope, with canyon consumers significantly 13C-depleted, indicating a greater supply and/or utilization of fresh organic matter compared to the slope. Isoscapes for benthic and suspension feeders were distinct, possibly due to the consumption of different quality organic matter sources (fresh = suspension feeders, old = benthic feeders), each with distinct isotope composition. To our knowledge, our modeled isoscapes represent the first spatially extensive isotopic maps of deep-sea consumers, providing insights into regional-scale variation in stable carbon and nitrogen isotopes for different consumer groups. They provide a baseline for tracking climate-change induced fluctuations in the quality and availability of surface primary production and the consequential impact to benthic communities, which play critical roles in carbon cycling in our world’s oceans.

Seabed cover of organisms is an established metric for assessing the status of many vulnerable marine ecosystems. When deriving cover estimates from seafloor imagery, a source of uncertainty in capturing the true distribution of organisms is introduced by the inherent variability and bias of the annotation method used to extract ecological data. We investigated variability and bias in two common annotation methods for estimating organism cover, and the role of size selectivity in this variability. Eleven annotators estimated sparse cold-water coral cover in the same 96 images with both grid-based and manual segmentation annotation methods. The standard deviation between annotators was three times greater in the grid-based method compared to segmentation, and grid-based estimates from annotators tended to overestimate coral cover. Size selectivity biased the manual segmentation; the minimum size of colonies segmented varied between annotators fivefold. Two modelling techniques (based on Richard’s selection curves and Gaussian processes) were used to impute areas where annotators identified colonies too small for segmentation. By imputing small coral sizes in segmentation estimates, the coefficient of variation between annotators was reduced by approximately 10%, and method bias (compared to a reference dataset) was reduced by up to 23%. Therefore, for sparse, low cover organisms, manual segmentation of images is recommended to minimise annotator variability and bias. Uncertainty in cover estimates may be further reduced by addressing size selectivity bias when annotating small organisms in images using a data-driven modelling technique.