Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico
<p>Deep learning models have demonstrated remarkable success in fields such as language processing and computer vision, routinely employed for tasks like language translation, image classification, and anomaly detection. Recent advancements in ocean sciences, particularly in data assimilation...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2025-01-01
|
| Series: | Ocean Science |
| Online Access: | https://os.copernicus.org/articles/21/113/2025/os-21-113-2025.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589874944802816 |
|---|---|
| author | O. Zavala-Romero O. Zavala-Romero A. Bozec E. P. Chassignet J. R. Miranda J. R. Miranda |
| author_facet | O. Zavala-Romero O. Zavala-Romero A. Bozec E. P. Chassignet J. R. Miranda J. R. Miranda |
| author_sort | O. Zavala-Romero |
| collection | DOAJ |
| description | <p>Deep learning models have demonstrated remarkable success in fields such as language processing and computer vision, routinely employed for tasks like language translation, image classification, and anomaly detection. Recent advancements in ocean sciences, particularly in data assimilation (DA), suggest that machine learning can emulate dynamical models, replace traditional DA steps to expedite processes, or serve as hybrid surrogate models to enhance forecasts. However, these studies often rely on ocean models of intermediate complexity, which involve significant simplifications that present challenges when transitioning to full-scale operational ocean models. This work explores the application of convolutional neural networks (CNNs) in data assimilation within the context of the HYbrid Coordinate Ocean Model (HYCOM) in the Gulf of Mexico. The CNNs are trained to correct model errors from a 2-year, high-resolution (<span class="inline-formula"><math xmlns="http://www.w3.org/1998/Math/MathML" id="M1" display="inline" overflow="scroll" dspmath="mathml"><mrow><mn mathvariant="normal">1</mn><mo>/</mo><mn mathvariant="normal">25</mn><mi mathvariant="italic">°</mi></mrow></math><span><svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="31pt" height="14pt" class="svg-formula" dspmath="mathimg" md5hash="0fa693bbf5c78df25ed18818c53372c4"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="os-21-113-2025-ie00001.svg" width="31pt" height="14pt" src="os-21-113-2025-ie00001.png"/></svg:svg></span></span>) HYCOM dataset, assimilated using the Tendral Statistical Interpolation System (T-SIS). The CNNs are trained to replicate the increments generated by the T-SIS data assimilation package, aiming to correct model forecasts of sea surface temperature (SST) and sea surface height (SSH). The inputs to the CNNs include real satellite observations of SST from the Group for High Resolution Sea Surface Temperature (GHRSST), along-track altimeter SSH observations (ADT), the model background state (previous forecast), and the innovations (differences between observations and background). We assess the performance of the CNNs across five controlled experiments, designed to provide insights into their application in environments governed by full primitive equations, real observations, and complex topographies. The experiments focus on evaluating (1) the architecture and complexity of the CNNs, (2) the type and quantity of observations, (3) the type and number of assimilated fields, (4) the impact of training window size, and (5) the influence of coastal boundaries. Our findings reveal significant correlations between the chosen training window size – a factor not commonly examined – and the CNNs' ability to assimilate observations effectively. We also establish a clear link between the CNNs' architecture and complexity and their overall performance.</p>
<p>This research uses artificial intelligence to enhance ocean forecasting in the Gulf of Mexico. By using convolutional neural networks, the study improves predictions of sea temperatures and heights by integrating real satellite data with existing models. Through five comprehensive experiments, the team found that the amount of training data and the design of the neural networks significantly affect accuracy. These insights pave the way for faster, more reliable ocean models, benefiting environmental monitoring and maritime operations.</p> |
| format | Article |
| id | doaj-art-dbd5339c104d4138bfffc777fae008e1 |
| institution | Kabale University |
| issn | 1812-0784 1812-0792 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | Ocean Science |
| spelling | doaj-art-dbd5339c104d4138bfffc777fae008e12025-01-24T07:49:16ZengCopernicus PublicationsOcean Science1812-07841812-07922025-01-012111313210.5194/os-21-113-2025Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of MexicoO. Zavala-Romero0O. Zavala-Romero1A. Bozec2E. P. Chassignet3J. R. Miranda4J. R. Miranda5Department of Scientific Computing, Florida State University, Tallahassee, FL 32306, USACenter for Ocean–Atmospheric Prediction Studies, Florida State University, Tallahassee, FL 32306, USACenter for Ocean–Atmospheric Prediction Studies, Florida State University, Tallahassee, FL 32306, USACenter for Ocean–Atmospheric Prediction Studies, Florida State University, Tallahassee, FL 32306, USADepartment of Scientific Computing, Florida State University, Tallahassee, FL 32306, USACenter for Ocean–Atmospheric Prediction Studies, Florida State University, Tallahassee, FL 32306, USA<p>Deep learning models have demonstrated remarkable success in fields such as language processing and computer vision, routinely employed for tasks like language translation, image classification, and anomaly detection. Recent advancements in ocean sciences, particularly in data assimilation (DA), suggest that machine learning can emulate dynamical models, replace traditional DA steps to expedite processes, or serve as hybrid surrogate models to enhance forecasts. However, these studies often rely on ocean models of intermediate complexity, which involve significant simplifications that present challenges when transitioning to full-scale operational ocean models. This work explores the application of convolutional neural networks (CNNs) in data assimilation within the context of the HYbrid Coordinate Ocean Model (HYCOM) in the Gulf of Mexico. The CNNs are trained to correct model errors from a 2-year, high-resolution (<span class="inline-formula"><math xmlns="http://www.w3.org/1998/Math/MathML" id="M1" display="inline" overflow="scroll" dspmath="mathml"><mrow><mn mathvariant="normal">1</mn><mo>/</mo><mn mathvariant="normal">25</mn><mi mathvariant="italic">°</mi></mrow></math><span><svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="31pt" height="14pt" class="svg-formula" dspmath="mathimg" md5hash="0fa693bbf5c78df25ed18818c53372c4"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="os-21-113-2025-ie00001.svg" width="31pt" height="14pt" src="os-21-113-2025-ie00001.png"/></svg:svg></span></span>) HYCOM dataset, assimilated using the Tendral Statistical Interpolation System (T-SIS). The CNNs are trained to replicate the increments generated by the T-SIS data assimilation package, aiming to correct model forecasts of sea surface temperature (SST) and sea surface height (SSH). The inputs to the CNNs include real satellite observations of SST from the Group for High Resolution Sea Surface Temperature (GHRSST), along-track altimeter SSH observations (ADT), the model background state (previous forecast), and the innovations (differences between observations and background). We assess the performance of the CNNs across five controlled experiments, designed to provide insights into their application in environments governed by full primitive equations, real observations, and complex topographies. The experiments focus on evaluating (1) the architecture and complexity of the CNNs, (2) the type and quantity of observations, (3) the type and number of assimilated fields, (4) the impact of training window size, and (5) the influence of coastal boundaries. Our findings reveal significant correlations between the chosen training window size – a factor not commonly examined – and the CNNs' ability to assimilate observations effectively. We also establish a clear link between the CNNs' architecture and complexity and their overall performance.</p> <p>This research uses artificial intelligence to enhance ocean forecasting in the Gulf of Mexico. By using convolutional neural networks, the study improves predictions of sea temperatures and heights by integrating real satellite data with existing models. Through five comprehensive experiments, the team found that the amount of training data and the design of the neural networks significantly affect accuracy. These insights pave the way for faster, more reliable ocean models, benefiting environmental monitoring and maritime operations.</p>https://os.copernicus.org/articles/21/113/2025/os-21-113-2025.pdf |
| spellingShingle | O. Zavala-Romero O. Zavala-Romero A. Bozec E. P. Chassignet J. R. Miranda J. R. Miranda Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico Ocean Science |
| title | Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico |
| title_full | Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico |
| title_fullStr | Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico |
| title_full_unstemmed | Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico |
| title_short | Convolutional neural networks for sea surface data assimilation in operational ocean models: test case in the Gulf of Mexico |
| title_sort | convolutional neural networks for sea surface data assimilation in operational ocean models test case in the gulf of mexico |
| url | https://os.copernicus.org/articles/21/113/2025/os-21-113-2025.pdf |
| work_keys_str_mv | AT ozavalaromero convolutionalneuralnetworksforseasurfacedataassimilationinoperationaloceanmodelstestcaseinthegulfofmexico AT ozavalaromero convolutionalneuralnetworksforseasurfacedataassimilationinoperationaloceanmodelstestcaseinthegulfofmexico AT abozec convolutionalneuralnetworksforseasurfacedataassimilationinoperationaloceanmodelstestcaseinthegulfofmexico AT epchassignet convolutionalneuralnetworksforseasurfacedataassimilationinoperationaloceanmodelstestcaseinthegulfofmexico AT jrmiranda convolutionalneuralnetworksforseasurfacedataassimilationinoperationaloceanmodelstestcaseinthegulfofmexico AT jrmiranda convolutionalneuralnetworksforseasurfacedataassimilationinoperationaloceanmodelstestcaseinthegulfofmexico |