Plymouth Marine Laboratory (abbreviated as PML) in the city of Plymouth, England, is a marine research organization and registered charity. It is a partner of the UK Research & Innovation's Natural Environment Research Council (NERC). PML's chair is Janice Timberlake, its chief executive is Prof. Icarus Allen and its patron is the film-maker James Cameron.
Marine biogeochemistry models are critical for forecasting, as well as estimating ecosystem responses to climate change and human activities. Data assimilation (DA) improves these models by aligning them with real-world observations, but marine biogeochemistry DA faces challenges due to model complexity, strong nonlinearity, and sparse, uncertain observations. Existing DA methods applied to marine biogeochemistry struggle to update unobserved variables effectively, while ensemble-based methods are computationally too expensive for high-complexity marine biogeochemistry models. This study demonstrates how machine learning (ML) can improve marine biogeochemistry DA by learning statistical relationships between observed and unobserved variables. We integrate ML-driven balancing schemes into a 1D prototype of a system used to forecast marine biogeochemistry in the North-West European Shelf seas. ML is applied to predict (i) state-dependent correlations from free-run ensembles and (ii), in an “end-to-end” fashion, analysis increments from an Ensemble Kalman Filter. Our results show that ML significantly enhances updates for previously not-updated variables when compared to univariate schemes akin to those used operationally. Furthermore, ML models exhibit moderate transferability to new locations, a crucial step toward scaling these methods to 3D operational systems. We conclude that ML offers a clear pathway to overcome current computational bottlenecks in marine biogeochemistry DA and that refining transferability, optimizing training data sampling, and evaluating scalability for large-scale marine forecasting, should be future research priorities.
The Antarctic Peninsula is warming rapidly, with more frequent extreme temperature and precipitation events, reduced sea ice, glacier retreat, ice shelf collapse, and ecological shifts. Here, we review its behaviour under present-day climate, and low (SSP 1–2.6), medium-high (SSP 3–7.0) and very high (SSP 5–8.5) future emissions scenarios, corresponding to global temperature increases of 1.8 °C, 3.6 °C and 4.4 °C by 2100. Higher emissions will bring more days above 0 °C, increased liquid precipitation, ocean warming, and more intense extreme weather events such as ocean heat waves and atmospheric rivers. Surface melt on ice shelves will increase, depleting firn air content and promoting meltwater ponding. Under the highest emission scenario, collapse of the Larsen C and Wilkins ice shelves is likely by 2100 CE, and loss of sea ice and ice shelves around the Peninsula will exacerbate the current trends of land-ice mass loss. Collapse of George VI Ice Shelf by 2300 under SSP 5–8.5 would substantially increase sea level contributions. Under this very high emissions scenario, sea level contributions from the Peninsula could reach 7.5 ± 14.1 mm by 2100 CE and 116.3 ± 66.9 mm by 2300 CE. Conversely, under the lower emissions scenarios, the Antarctic Peninsula’s sea ice remains similar to present, and land ice is predicted to undergo only minor grounding line recession and thinning. Changes in sea surface temperatures and the change from snow to rain will impact marine and terrestrial biota, altering species richness and enhancing colonisation by non-native species. Ranges of key species such as krill and salps are likely to contract to the south, impacting their marine vertebrate predators. These changing conditions will also influence Antarctic Peninsula research, fisheries, tourism, infrastructure and logistics. The future of the Peninsula depends on the choices made today. Limiting temperatures to below 2 °C, and as close as possible to 1.5 °C (by following the SSP 1–1.9 or 1–2.6 scenarios), combined with effective governance, will result in increased resilience and relatively modest changes. Any higher emissions scenarios will damage pristine systems, cause sustained, irreversible ice loss on human timescales, and spread to Antarctic regions beyond the Peninsula.
We demonstrate that assimilating neural network (NN) predicted surface nitrate leads to a major improvement in phytoplankton short-range (1-5 day) dynamical model forecasts for the Northwest European Shelf (NWES) seas. We show that assimilation of only ocean-colour chlorophyll- in the current Met Office NWES operational system can lead to excess surface nitrate concentrations in the post-spring bloom period and these are a major reason behind some known, fast-growing biases in NWES phytoplankton forecasts during late spring and summer. Assimilating observations of nitrate would potentially help to address this, but NWES nitrate data are typically not available in sufficient abundance to be assimilated effectively. We have therefore used a recently developed and validated NN model predicting surface nitrate concentrations from a range of observable variables and assimilated the NN-predicted nitrate within a research and development version of the Met Office's NWES operational forecasting system. As a result of nitrate assimilation, the phytoplankton five-day forecast skill improves by up to 30%. We show that, although much of this improvement can be achieved by using a weekly nitrate climatology predicted by the NN model, there is a clear advantage in using flow-dependent nitrate data. We discuss the impacts of this improvement on a range of additional eutrophication indicators, such as dissolved inorganic phosphorus and sea-bottom oxygen. We argue that it should be feasible to upgrade this approach to a fully hybrid machine-learning-data assimilation within the near-real-time NWES operational forecasting system.
Abstract Air‐sea gas exchange regulates the exchange of climatically important gases between the ocean and the atmosphere, shaping both climate and ocean biogeochemistry. Bubbles beneath the sea surface enhance this exchange by introducing an additional transfer pathway in parallel to the interfacial transfer route. Although the role of bubbles in gas flux has been debated since the 1980s, recent advances in laboratory experiments, field observations, and modeling have provided new insights. Bubble‐mediated gas transfer differs from interfacial transfer in three key ways: (a) it shows strong nonlinearity with wind speed due to its link with wave breaking; (b) it depends on gas solubility because of the finite volume and short lifetime of bubbles; and (c) it shifts the equilibrium toward slight oversaturation through the overpressure of submerged bubbles. These characteristics make bubble‐mediated gas transfer complicated to quantify, and existing observations and models indicate a wide range of bubble contributions to air‐sea carbon dioxide and oxygen exchange. Three critical knowledge gaps are identified: (a) limited understanding of near‐surface (0–1 m) bubble dynamics, including volume flux, size distribution, and evolution, which directly control the solubility and diffusivity dependence of bubble‐mediated gas exchange; (b) the absence of consistent field constraints spanning the full range of gas solubilities; and (c) the lack of knowledge to scale laboratory results to oceanic conditions. Addressing these gaps will require integrated efforts combining near‐surface bubble measurements and simulations, field observations of gas transfer across diverse solubilities using complementary techniques, and improved modeling frameworks.
Artificial Intelligence (AI) Foundation models (FMs), pre-trained on massive unlabelled datasets, have the potential to drastically change AI applications in ocean science, where labelled data are often sparse and expensive to collect. In this work, we describe a new foundation model using the Prithvi-EO Vision Transformer architecture which has been pre-trained to reconstruct data from the Sentinel-3 Ocean and Land Colour Instrument (OLCI). We evaluate the model by fine-tuning on two downstream marine earth observation tasks. We first assess model performance compared to current baseline models used to quantify chlorophyll concentration. We then evaluate the FMs ability to refine remote sensing-based estimates of ocean primary production. Our results demonstrate the utility of self-trained FMs for marine monitoring, in particular for making use of small amounts of high quality labelled data and in capturing detailed spatial patterns of ocean colour whilst matching point observations. We conclude that this new generation of geospatial AI models has the potential to provide more robust, data-driven insights into ocean ecosystems and their role in global climate processes.
