OD52A:
Data Science for Modern Oceanography: Statistics, Machine Learning, Visualization, and More II

Session ID#: 92496

Session Description:
Oceanographic research in today's world increasingly relies on analyzing multiple datasets, including ship-based measurements, profiles from autonomous instruments such as floats and gliders, satellite remote sensing data, as well as output from models and state estimates. These datasets are growing larger and more complex every day, and future advances in ocean observing and modeling will add significantly to the quantity and variety of "Big Data" across all disciplines of oceanography. Innovative statistical methods, computational techniques, and data visualizations will be needed in the coming decades to distill these data and to extract maximum scientific understanding. New developments in statistics and data science have the potential to transform our knowledge of the ocean across many spatial and temporal scales and can help address various emerging challenges in oceanographic data analysis. This session solicits studies on using the latest techniques from statistics, machine learning, and visualization to analyze datasets in oceanography and related areas of climate science, both those currently existing as well as those that will be available in the near future. Presentation topics may include computational methods for large datasets; software platforms and tools; model diagnostics, validation, and parameterization; spatio-temporal interpolation; uncertainty quantification; classification and regression techniques; pattern recognition; as well as other advanced data science topics.
Co-Sponsor(s):
  • OB - Ocean Biology and Biogeochemistry
  • OM - Ocean Modeling
  • PL - Physical Oceanography: Mesoscale and Larger
Primary Chair:  Alison R Gray, University of Washington, School of Oceanography, Seattle, WA, United States
Co-chairs:  Mikael Kuusela, Carnegie Mellon University, Department of Statistics and Data Science, Pittsburgh, PA, United States and Donata Giglio, University of Colorado Boulder, Department of Atmospheric and Oceanic Sciences, Boulder, CO, United States
Primary Liaison:  Alison R Gray, University of Washington, School of Oceanography, Seattle, WA, United States
Moderators:  Alison R Gray, University of Washington, School of Oceanography, Seattle, WA, United States and Mikael Kuusela, Carnegie Mellon University, Department of Statistics and Data Science, Pittsburgh, PA, United States
Student Paper Review Liaisons:  Alison R Gray, University of Washington, School of Oceanography, Seattle, WA, United States and Mikael Kuusela, Carnegie Mellon University, Department of Statistics and Data Science, Pittsburgh, PA, United States

Abstracts Submitted to this Session:

Machine learning for inference and parametrization of ocean turbulence (657279)
Laure Zanna, University of Oxford, Dept of Physics, Oxford, United Kingdom and Thomas Bolton, University of Oxford, Department of Physics, Oxford, United Kingdom
Long-term trends in ocean chlorophyll: update from a Bayesian hierarchical space-time model (650393)
Matthew Hammond, National Oceangraphy Centre, Southampton, United Kingdom, Claudie Beaulieu, University of California Santa Cruz, Department of Ocean Sciences, Santa Cruz, United States, Stephanie Henson, National Oceanography Centre, Southampton, United Kingdom and Sujit K Sahu, University of Southampton, Mathematical Sciences, Southampton, United Kingdom
Cluster-based ocean model evaluation on the Antarctic continental shelf (655055)
Qiang Sun, Atmospheric and Environmental Research, Lexington, MA, United States, Christopher M Little, Atmospheric and Environmental Research Lexington, Lexington, MA, United States and Alice Barthel, Los Alamos National Laboratory, Los Alamos, United States
Novel characterization and correction of ocean model biases using sparse observations: application to surface ocean radiocarbon (650931)
Heather D Graven1, Gerald Lim2, Kei Yeung2, Joanna Lester2 and Peer Johannes Nowack3, (1)Imperial College London, Physics, London, SW7, United Kingdom, (2)Imperial College London, Physics, London, United Kingdom, (3)Imperial College London, Faculty of Natural Sciences, Department of Physics, Grantham Institute and the Data Science Institute, London, United Kingdom
Reconstruction of incomplete spatial data with feature preserving information transport (656772)
Siavash Ameli, University of California, Berkeley, CA, United States and Shawn Shadden, University of California, Berkeley, Berkeley, CA, United States
Echopype: Interoperable and Scalable Processing of Ocean Sonar Data (649395)
Wu-Jung Lee, University of Washington, Applied Physics Laboratory, Seattle, WA, United States, Valentina Staneva, University of Washington, eScience Institute, Seattle, WA, United States and Kavin Nguyen, University of Washington, Department of Physics, United States
On the use of the Ocean Virtual Laboratory open tools to prepare training sets for AI deep learning of ocean surface signatures (655019)
Fabrice Collard, OceanDataLab, Locmaria-Plouzané, France, Lucile Gaultier, OceanDataLab, Brest, France, Benjamin Holt, NASA Jet Propulsion Laboratory, Pasadena, CA, United States, Sylvain Herlédan, Oceandatalab, Locmaria Plouzané, France, Ziad El Khoury Hanna, Oceandatalab, France and Gilles Guitton, OceanDataLab, Plouzané, France
Data Science and Signal Processing for Drifter Data (651389)
Adam Sykulski, Lancaster University, Lancaster, United Kingdom, Jeffrey J Early, NorthWest Research Associates, Redmond, WA, United States, Jonathan M Lilly, Theiss Research, La Jolla, CA, United States, Sofia Olhede, Ecole polytechnique fédérale de Lausanne (EPFL), Switzerland and Arthur P. Guillaumin, University College London, London, United Kingdom