A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence
Abstract Satellite altimeters provide global observations of sea surface height (SSH) and present a unique data set for advancing our theoretical understanding of upper‐ocean dynamics and monitoring its variability. Considering that mesoscale SSH patterns can evolve on timescales comparable to or sh...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Geophysical Union (AGU)
2021-01-01
|
| Series: | Journal of Advances in Modeling Earth Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2019MS001965 |
| _version_ | 1827787360411058176 |
|---|---|
| author | Georgy E. Manucharyan Lia Siegelman Patrice Klein |
| author_facet | Georgy E. Manucharyan Lia Siegelman Patrice Klein |
| author_sort | Georgy E. Manucharyan |
| collection | DOAJ |
| description | Abstract Satellite altimeters provide global observations of sea surface height (SSH) and present a unique data set for advancing our theoretical understanding of upper‐ocean dynamics and monitoring its variability. Considering that mesoscale SSH patterns can evolve on timescales comparable to or shorter than satellite return periods, it is challenging to accurately reconstruct the continuous SSH evolution as currently available altimetry observations are still spatially and temporally sparse. Here we explore the possibility of SSH interpolation via Deep Learning by using synthetic observations from an idealized quasigeostrophic model of baroclinic ocean turbulence. We demonstrate that Convolutional Neural Networks with Residual Learning are superior in SSH reconstruction to linear and recently developed dynamical interpolation techniques. Also, the deep neural networks can provide a skillful state estimate of unobserved deep ocean currents at mesoscales. These conspicuous results suggest that SSH patterns of eddies might contain substantial information about the underlying deep ocean currents that are necessary for SSH prediction. Our training data are focused on highly idealized physics and diversification of processes needs to be considered to more accurately represent the real ocean. In addition, methodological improvements such as transfer learning and implementation of dynamically aware loss functions might be necessary to consider before its ultimate use with real satellite observations. Nonetheless, by providing a proof of concept based on synthetic data, our results point to deep learning as a viable alternative to existing interpolation and, more generally, state estimation methods for satellite observations of eddying currents. |
| first_indexed | 2024-03-11T16:49:11Z |
| format | Article |
| id | doaj.art-63bacb4338e34e6ebb851f7876e64c82 |
| institution | Directory Open Access Journal |
| issn | 1942-2466 |
| language | English |
| last_indexed | 2024-03-11T16:49:11Z |
| publishDate | 2021-01-01 |
| publisher | American Geophysical Union (AGU) |
| record_format | Article |
| series | Journal of Advances in Modeling Earth Systems |
| spelling | doaj.art-63bacb4338e34e6ebb851f7876e64c822023-10-21T14:51:47ZengAmerican Geophysical Union (AGU)Journal of Advances in Modeling Earth Systems1942-24662021-01-01131n/an/a10.1029/2019MS001965A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean TurbulenceGeorgy E. Manucharyan0Lia Siegelman1Patrice Klein2School of Oceanography University of Washington Seattle WA USAJet Propulsion Laboratory California Institute of Technology Pasadena CA USAJet Propulsion Laboratory California Institute of Technology Pasadena CA USAAbstract Satellite altimeters provide global observations of sea surface height (SSH) and present a unique data set for advancing our theoretical understanding of upper‐ocean dynamics and monitoring its variability. Considering that mesoscale SSH patterns can evolve on timescales comparable to or shorter than satellite return periods, it is challenging to accurately reconstruct the continuous SSH evolution as currently available altimetry observations are still spatially and temporally sparse. Here we explore the possibility of SSH interpolation via Deep Learning by using synthetic observations from an idealized quasigeostrophic model of baroclinic ocean turbulence. We demonstrate that Convolutional Neural Networks with Residual Learning are superior in SSH reconstruction to linear and recently developed dynamical interpolation techniques. Also, the deep neural networks can provide a skillful state estimate of unobserved deep ocean currents at mesoscales. These conspicuous results suggest that SSH patterns of eddies might contain substantial information about the underlying deep ocean currents that are necessary for SSH prediction. Our training data are focused on highly idealized physics and diversification of processes needs to be considered to more accurately represent the real ocean. In addition, methodological improvements such as transfer learning and implementation of dynamically aware loss functions might be necessary to consider before its ultimate use with real satellite observations. Nonetheless, by providing a proof of concept based on synthetic data, our results point to deep learning as a viable alternative to existing interpolation and, more generally, state estimation methods for satellite observations of eddying currents.https://doi.org/10.1029/2019MS001965baroclinic instabilityDeep Learningdeep ocean flowsmesoscale eddiessea surface height interpolationstate estimation |
| spellingShingle | Georgy E. Manucharyan Lia Siegelman Patrice Klein A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence Journal of Advances in Modeling Earth Systems baroclinic instability Deep Learning deep ocean flows mesoscale eddies sea surface height interpolation state estimation |
| title | A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence |
| title_full | A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence |
| title_fullStr | A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence |
| title_full_unstemmed | A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence |
| title_short | A Deep Learning Approach to Spatiotemporal Sea Surface Height Interpolation and Estimation of Deep Currents in Geostrophic Ocean Turbulence |
| title_sort | deep learning approach to spatiotemporal sea surface height interpolation and estimation of deep currents in geostrophic ocean turbulence |
| topic | baroclinic instability Deep Learning deep ocean flows mesoscale eddies sea surface height interpolation state estimation |
| url | https://doi.org/10.1029/2019MS001965 |
| work_keys_str_mv | AT georgyemanucharyan adeeplearningapproachtospatiotemporalseasurfaceheightinterpolationandestimationofdeepcurrentsingeostrophicoceanturbulence AT liasiegelman adeeplearningapproachtospatiotemporalseasurfaceheightinterpolationandestimationofdeepcurrentsingeostrophicoceanturbulence AT patriceklein adeeplearningapproachtospatiotemporalseasurfaceheightinterpolationandestimationofdeepcurrentsingeostrophicoceanturbulence AT georgyemanucharyan deeplearningapproachtospatiotemporalseasurfaceheightinterpolationandestimationofdeepcurrentsingeostrophicoceanturbulence AT liasiegelman deeplearningapproachtospatiotemporalseasurfaceheightinterpolationandestimationofdeepcurrentsingeostrophicoceanturbulence AT patriceklein deeplearningapproachtospatiotemporalseasurfaceheightinterpolationandestimationofdeepcurrentsingeostrophicoceanturbulence |