Dataset Information

Probabilistic Machine Learning Estimation of Ocean Mixed Layer Depth From Dense Satellite and Sparse In Situ Observations.

ABSTRACT: The ocean mixed layer plays an important role in the coupling between the upper ocean and atmosphere across a wide range of time scales. Estimation of the variability of the ocean mixed layer is therefore important for atmosphere-ocean prediction and analysis. The increasing coverage of in situ Argo profile data allows for an increasingly accurate analysis of the mixed layer depth (MLD) variability associated with deviations from the seasonal climatology. However, sampling rates are not sufficient to fully resolve subseasonal ( <90 day) MLD variability. Yet, many multivariate observations-based analyses include implicit modeled subseasonal MLD variability. One analysis method is optimal interpolation of in situ data, but the interior analysis can be improved by leveraging surface data with regression or variational approaches. Here, we demonstrate how machine learning methods and satellite sea surface temperature, salinity, and height facilitate MLD estimation in a pilot study of two regions: the mid-latitude southern Indian and the eastern equatorial Pacific Oceans. We construct multiple machine learning architectures to produce weekly 1/2° gridded MLD anomaly fields (relative to a monthly climatology) with uncertainty estimates. We test multiple traditional and probabilistic machine learning techniques to compare both accuracy and probabilistic calibration. We validate our methodology by applying it to ocean model simulations. We find that incorporating sea surface data through a machine learning model improves the performance of spatiotemporal MLD variability estimation compared to optimal interpolation of Argo observations alone. These preliminary results are a promising first step for the application of machine learning to MLD prediction.

SUBMITTER: Foster D

PROVIDER: S-EPMC9286844 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The surface mixed layer of the world ocean regulates global climate by controlling heat and carbon exchange between the atmosphere and the oceanic interior1-3. The mixed layer also shapes marine ecosystems by hosting most of the ocean's primary production4 and providing the conduit for oxygenation of deep oceanic layers. Despite these important climatic and life-supporting roles, possible changes in the mixed layer during an era of global climate change remain uncertain. Here we use oceanographic observations to show that from 1970 to 2018 the density contrast across the base of the mixed layer increased and that the mixed layer itself became deeper. Using a physically based definition of upper-ocean stability that follows different dynamical regimes across the global ocean, we find that the summertime density contrast increased by 8.9 ± 2.7 per cent per decade (10-6-10-5 per second squared per decade, depending on region), more than six times greater than previous estimates. Whereas prior work has suggested that a thinner mixed layer should accompany a more stratified upper ocean5-7, we find instead that the summertime mixed layer deepened by 2.9 ± 0.5 per cent per decade, or several metres per decade (typically 5-10 metres per decade, depending on region). A detailed mechanistic interpretation is challenging, but the concurrent stratification and deepening of the mixed layer are related to an increase in stability associated with surface warming and high-latitude surface freshening8,9, accompanied by a wind-driven intensification of upper-ocean turbulence10,11. Our findings are based on a complex dataset with incomplete coverage of a vast area. Although our results are robust within a wide range of sensitivity analyses, important uncertainties remain, such as those related to sparse coverage in the early years of the 1970-2018 period. Nonetheless, our work calls for reconsideration of the drivers of ongoing shifts in marine primary production, and reveals stark changes in the world's upper ocean over the past five decades.

Dataset Information

Probabilistic Machine Learning Estimation of Ocean Mixed Layer Depth From Dense Satellite and Sparse In Situ Observations.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets