Search for: All records

Abstract Basal melting of Antarctic ice shelves is primarily driven by heat delivery from warm Circumpolar Deep Water. Here we classify near-shelf water masses in an eddy-resolving numerical model of the Southern Ocean to develop a unified view of warm water intrusion onto the Antarctic continental shelf. We identify four regimes on seasonal timescales. In regime 1 (East Antarctica), heat intrusions are driven by easterly winds via Ekman dynamics. In regime 2 (West Antarctica), intrusion is primarily determined by the strength of a shelf-break undercurrent. In regime 3, the warm water cycle on the shelf is in antiphase with dense shelf water production (Adélie Coast). Finally, in regime 4 (Weddell and Ross seas), shelf-ward warm water inflow occurs along the western edge of canyons during periods of dense shelf water outflow. Our results advocate for a reformulation of the traditional annual-mean regime classification of the Antarctic continental shelf. 

Abstract Assessing the biological characteristics of high-latitude winter habitats of migratory marine predators is necessary for conservation and management in Antarctica. Tracking data from chinstrap penguins (Pygoscelis antarcticus) and southern elephant seals (Mirounga leonina), key Antarctic predators with different diets and foraging habits, indicate that some individuals undertake long-distance winter migrations to remote regions south of 55°S and west of 120°W. There, localized hotspots of increased use, with general reductions in mean swimming speed are evident. Presumably, these predators migrate to areas with higher productivity, however the marine productivity in this remote region during winter is unknown. Light limitation during winter precludes the use of optical satellite data to characterize marine productivity here, but biogeochemical-Argo floats can provide year-round chlorophyll data. These data inform the Biogeochemical Southern Ocean State Estimate (B-SOSE), which provides year-round estimates of marine productivity. The predator hotspots overlap with two areas with year-round elevated surface chlorophyll levels predicted by B-SOSE, consistent with previous studies indicating enhanced mixing in those areas. Our results suggest that persistent areas of elevated chlorophyll centered near 160°W and 120°W near the boundaries of the Ross Gyre and the southern boundary of the Antarctic Circumpolar Current support a productive food web capable of supporting the diverse foraging niches of pelagic species during winter. 

Abstract The Southern Ocean is a region of high surface nutrient content, reflecting an inefficient biological carbon pump. The variability, predictability, and causes of changes in these nutrient levels on interannual to decadal time scales remain unclear. We employ a deep learning approach, specifically a Temporal Convolution Attention Neural Network (TCANN), to conduct multi‐year forecasting of surface based on oceanic physical drivers. The TCANN successfully replicates testing data with a prediction skill extending to at least 4 years with the GFDL‐ESM4‐driven model and 1 year with the observation‐driven model. To benchmark the results, we compare the prediction skill of TCANN with a simple persistence model and two regression methods, a linear regression and a ridge regression. The TCANN model was able to predict variability with a higher skill than persistence and the two regression methods indicating that non‐linearities present in the system become too high to predict inter‐annual variability with traditional regression methods. To enhance the interpretability of the predictions, we explore three explainable AI techniques: occlusion analysis, integrated gradients, and Gradient Shap. The outcomes suggest a crucial role played by salinity processes and buoyancy/potential density fluxes on the prediction of on annual time scales. The deep learning tools' ability to provide skillful forecasts well into the future presents a promising avenue for gaining insights into how the Southern Ocean's surface nutrients respond to climate change based on physical quantities. 

Abstract Recent advances in explainable artificial intelligence (XAI) methods show promise for understanding predictions made by machine learning (ML) models. XAI explains how the input features are relevant or important for the model predictions. We train linear regression (LR) and convolutional neural network (CNN) models to make 1-day predictions of sea ice velocity in the Arctic from inputs of present-day wind velocity and previous-day ice velocity and concentration. We apply XAI methods to the CNN and compare explanations to variance explained by LR. We confirm the feasibility of using a novel XAI method [i.e., global layerwise relevance propagation (LRP)] to understand ML model predictions of sea ice motion by comparing it to established techniques. We investigate a suite of linear, perturbation-based, and propagation-based XAI methods in both local and global forms. Outputs from different explainability methods are generally consistent in showing that wind speed is the input feature with the highest contribution to ML predictions of ice motion, and we discuss inconsistencies in the spatial variability of the explanations. Additionally, we show that the CNN relies on both linear and nonlinear relationships between the inputs and uses nonlocal information to make predictions. LRP shows that wind speed over land is highly relevant for predicting ice motion offshore. This provides a framework to show how knowledge of environmental variables (i.e., wind) on land could be useful for predicting other properties (i.e., sea ice velocity) elsewhere. Significance StatementExplainable artificial intelligence (XAI) is useful for understanding predictions made by machine learning models. Our research establishes trustability in a novel implementation of an explainable AI method known as layerwise relevance propagation for Earth science applications. To do this, we provide a comparative evaluation of a suite of explainable AI methods applied to machine learning models that make 1-day predictions of Arctic sea ice velocity. We use explainable AI outputs to understand how the input features are used by the machine learning to predict ice motion. Additionally, we show that a convolutional neural network uses nonlinear and nonlocal information in making its predictions. We take advantage of the nonlocality to investigate the extent to which knowledge of wind on land is useful for predicting sea ice velocity elsewhere. 

Abstract West Antarctic Ice Sheet mass loss is a major source of uncertainty in sea level projections. The primary driver of this melting is oceanic heat from Circumpolar Deep Water originating offshore in the Antarctic Circumpolar Current. Yet, in assessing melt variability, open ocean processes have received considerably less attention than those governing cross-shelf exchange. Here, we use Lagrangian particle release experiments in an ocean model to investigate the pathways by which Circumpolar Deep Water moves toward the continental shelf across the Pacific sector of the Southern Ocean. We show that Ross Gyre expansion, linked to wind and sea ice variability, increases poleward heat transport along the gyre’s eastern limb and the relative fraction of transport toward the Amundsen Sea. Ross Gyre variability, therefore, influences oceanic heat supply toward the West Antarctic continental slope. Understanding remote controls on basal melt is necessary to predict the ice sheet response to anthropogenic forcing. 

Abstract Carbon export driven by submesoscale, eddy‐associated vertical velocities (“eddy subduction”), and particularly its seasonality, remains understudied, leaving a gap in our understanding of ocean carbon sequestration. Here, we assess mechanisms controlling eddy subduction's spatial and seasonal patterns using 15 years of observations from BGC‐Argo floats in the Southern Ocean. We identify signatures of eddy subduction as subsurface anomalies in temperature‐salinity and oxygen. The anomalies are spatially concentrated near weakly stratified areas and regions with strong lateral buoyancy gradients diagnosed from satellite altimetry, particularly in the Antarctic Circumpolar Current's standing meanders. We use bio‐optical ratios, specifically the chlorophyllato particulate backscatter ratio (Chl/b_bp) to find that eddy subduction is most active in the spring and early summer, with freshly exported material associated with seasonally weak vertical stratification and increasing surface biomass. Climate change is increasing ocean stratification globally, which may weaken eddy subduction's carbon export potential. 

Abstract The Southern Ocean is a region of intense air–sea exchange that plays a critical role for ocean circulation, global carbon cycling, and climate. Subsurface chlorophyll‐a maxima, annually recurrent features throughout the Southern Ocean, may increase the energy flux to higher trophic levels and facilitate downward carbon export. It is important that model parameterizations appropriately represent the chlorophyll vertical structure in the Southern Ocean. Using BGC‐Argo chlorophyll profiles and the Biogeochemical Southern Ocean State Estimate (B‐SOSE), we investigate the sensitivity of chlorophyll vertical structure to model parameters. Based on the sensitivity analysis results, we estimate optimized parameters, which efficiently improve the model consistency with observations. We characterize chlorophyll vertical structure in terms of Empirical Orthogonal Functions and define metrics to compare model results and observations in a series of parameter perturbation experiments. We show that chlorophyll magnitudes are likely to respond quasi‐symmetrically to perturbations in the analyzed parameters, while depth and thickness of the subsurface chlorophyll maximum show an asymmetric response. Perturbing the phytoplankton growth tends to generate more symmetric responses than perturbations in the grazing rate. We identify parameters that affect chlorophyll magnitude, subsurface chlorophyll or both and discuss insights into the processes that determine chlorophyll vertical structure in B‐SOSE. We highlight turbulence, differences in phytoplankton traits, and grazing parameterizations as key areas for improvement in models of the Southern Ocean. 

Abstract The strength and variability of the Southern Ocean carbon sink is a significant source of uncertainty in the global carbon budget. One barrier to reconciling observations and models is understanding how synoptic weather patterns modulate air-sea carbon exchange. Here, we identify and track storms using atmospheric sea level pressure fields from reanalysis data to assess the role that storms play in driving air-sea CO₂exchange. We examine the main drivers of CO₂fluxes under storm forcing and quantify their contribution to Southern Ocean annual air-sea CO₂fluxes. Our analysis relies on a forced ocean-ice simulation from the Community Earth System Model, as well as CO₂fluxes estimated from Biogeochemical Argo floats. We find that extratropical storms in the Southern Hemisphere induce CO₂outgassing, driven by CO₂disequilibrium. However, this effect is an order of magnitude larger in observations compared to the model and caused by different reasons. Despite large uncertainties in CO₂fluxes and storm statistics, observations suggest a pivotal role of storms in driving Southern Ocean air-sea CO₂outgassing that remains to be well represented in climate models, and needs to be further investigated in observations. 

Abstract The Southern Ocean (SO) plays a crucial role in the process of sequestering heat and carbon dioxide from the atmosphere and transferring them to the deep ocean. This process is intricately linked to the formation of Antarctic Intermediate Water (AAIW) and Subantarctic Mode Water (SAMW), which are pivotal components of the Meridional Overturning Circulation (MOC) and have a substantial impact on the global climate balance. AAIW and SAMW take shape in specific regions of the Southern Ocean due to the influence of strong winds, buoyancy fluxes, and their effects, such as convection, the development of thick mixed layers, and wind‐driven subduction. These water masses subsequently flow northward, contributing to the ventilation of the intermediate layers within the subtropical gyres. In this study, our focus lies on investigating the regional aspects of AAIW and SAMW transformation in CMIP6 models. We accomplish this by analyzing the relationship between the meridional transport of these water masses and air‐sea fluxes, particularly Ekman pumping, freshwater fluxes, and heat fluxes. Our findings reveal that the highest transformation rates occur in the Indian sector of the Southern Ocean, with notable values also observed in the southeast Pacific and south of Africa. Additionally, we assess the potential changes in these formation regions under future scenarios projected for the end of the 21st century. Although the patterns of formation regions remain consistent, there is a significant decrease in the transformation process. 

Abstract Global climate change has impacted ocean biogeochemistry and physical dynamics, causing increases in acidity and temperature, among other phenomena. These changes can lead to deleterious effects on marine ecosystems and communities that rely on these ecosystems for their livelihoods. To better quantify these changes, an array of floats fitted with biogeochemical sensors (BGC‐Argo) is being deployed throughout the ocean. This paper presents an algorithm for deriving a deployment strategy that maximizes the information captured by each float. The process involves using a model solution as a proxy for the true ocean state and carrying out an iterative process to identify optimal float deployment locations for constraining the model variance. As an example, we use the algorithm to optimize the array for observing ocean surface dissolved carbon dioxide concentrations (pCO₂) in a region of strong air–sea gas exchange currently being targeted for BGC‐Argo float deployment. We conclude that 54% of the pCO₂variability in the analysis region could be sampled by an array of 50 Argo floats deployed in specified locations. This implies a relatively coarse average spacing, though we find the optimal spacing is nonuniform, with a denser sampling being required in the eastern equatorial Pacific. We also show that this method could be applied to determine the optimal float deployment along ship tracks, matching the logistics of real float deployment. We envision this software package to be a helpful resource in ocean observational design anywhere in the global oceans.