Modern Earth system models (ESMs) disagree on the impacts of anthropogenic global warming on the distribution of oxygen and associated low‐oxygen waters. A sensitivity study using the GFDL CM2Mc model points to the representation of lateral mesoscale eddy transport as a potentially important factor in such disagreement. Because mesoscale eddies are smaller than the spatial scale of ESM ocean grids, their impact must be parameterized using a lateral mixing coefficientAREDI. The value ofAREDIvaries across modern ESMs and nonlinearly impacts oxygen distributions. This study shows that an increase in atmospheric CO2results in a decline in productivity and a decrease in ventilation age in the tropics, increasing oxygen concentrations in the upper thermocline. In high latitudes global warming causes shallowing of deep convection, reducing the supply of oxygen to the deep. The net impact of these processes depends onAREDI, with an increase in hypoxic volume yet smaller total deoxygenation in the low‐mixing models, but a decrease in hypoxic volume yet larger total deoxygenation in the high‐mixing models. All models show decreases in suboxic volume, which are largest in the low‐mixing models. A subset of Coupled Model Intercomparison Project Phase 5 models exhibits a similar range of responses to global warming and similar decoupling between total deoxygenation and change in hypoxic volume. Uncertainty in lateral mixing remains an important contributor to uncertainty in projecting ocean deoxygenation.
An important technique used by climate modelers to isolate the impacts of increasing greenhouse gasses on Earth System processes is to simulate the impact of an abrupt increase in carbon dioxide. The spatial pattern of change provides a “fingerprint” that is generally much larger than natural variability. Insofar as the response to radiative forcing is linear (the impact of quadrupling CO2 is twice the impact of doubling CO2) this fingerprint can then be used to estimate the impact of historical greenhouse gas forcing. However, the degree to which biogeochemical cycles respond linearly to radiative forcing has rarely been tested. In this paper, we evaluate which ocean biogeochemical fields are likely to respond linearly to changing radiative forcing, which ones do not, and where linearity breaks down. We also demonstrate that the representation of lateral mixing by mesoscale eddies, which varies significantly across climate models, plays an important role in modulating the breakdown of linearity. Globally integrated surface rates of biogeochemical cycling (primary productivity, particulate export) respond in a relatively linear fashion and are only moderately sensitive to mixing. By contrast, the habitability of the interior ocean (as determined by hypoxia and calcite supersaturation) behaves non-linearly and is very sensitive to mixing. This is because the deep ocean, as well as certain regions in the surface ocean, are very sensitive to the magnitude of deep wintertime convection. The cessation of convection under global warming is strongly modulated by the representation of eddy mixing.
Pelagic tunicates (appendicularians, doliolids, pyrosomes, salps) are cosmopolitan members of open-ocean food webs that serve as a link to the microbial loop and play a disproportional role in vertical carbon flux. Soestia zonaria is an oceanic salp species studied for more than a century, but little information exists on its ecology. Specimens of Soestia collected between 2008 and 2021 during four research expeditions (three to the Chatham Rise, New Zealand, and one to the Northeast Atlantic) using MOCNESS-1 net and large midwater trawl were analyzed for reproductive biology, stoichiometry and vertical distribution. Populations at the Chatham Rise sampled in early winter were more developed than in spring/early summer and dominated by small and medium-sized sexually reproducing blastozooids. Whole Soestia specimens had high organic content (mean ± SD = 32.8 ± 7.5%) and carbon-to-nitrogen values (6.8 ± 0.9) compared with other salp species, indicating a stronger dependency on carbon to meet its nutritional needs. Depth-stratified sampling showed that Soestia is primarily a low-amplitude diel vertical migrator occurring in the top 150 m of the water column, but also found at depths exceeding 500 m. Soestia is primarily an epipelagic salp species adapted to living in warm-temperate nutrient-depleted and more productive ocean regions.
Antarctic krill (Euphausia superba) are integral to Southern Ocean pelagic ecosystems. Winters with extensive sea ice have been linked to high post-larval krill recruitment the following spring, suggesting that sea ice plays a critical role in larval overwinter survival. As the ocean warms and sea ice declines under climate change, understanding the mechanisms linking sea ice and krill recruitment is increasingly urgent. To address this, we developed a qualitative network model (QNM) that integrates evidence-based and hypothesized interactions to explore larval overwinter survival and growth under future climate scenarios in the southwest Atlantic sector. Our model highlights habitat-specific impacts, with substantial declines predicted for the North Antarctic Peninsula continental shelf due to reduced autumn primary productivity and warming. In contrast, survival may improve in open-ocean habitats under cooler scenarios that enhance sea-ice-associated processes, such as food availability and refuge. The inclusion of hypothesized mechanisms, such as sea-ice terraces providing refuge from predation, strengthened these conclusions and highlighted critical uncertainties, including the influence of glacial melt on food web dynamics. These findings demonstrate the value of QNMs in complementing quantitative approaches, offering a framework for identifying critical mechanisms, addressing knowledge gaps, and guiding future field and laboratory studies to improve predictions of krill responses to climate change.