OD43A:
Accelerating Ocean Science: Using Artificial Intelligence and Machine Learning to Gain New Insights I

Session ID#: 92483

Session Description:
For many research areas collecting and producing data no longer limits our scientific pursuits. Data volumes have increased dramatically and the main bottleneck to generate meaningful results is now the time to analyze vast datasets. Manual analysis or limited automation methods are quickly becoming unfeasible for big data and complex multi-disciplinary approaches. The recent advancement of accessible deep learning tools has opened up new possibilities for the application of artificial intelligence and machine learning models, facilitated by cloud infrastructure and scalable compute engines, to complex scientific challenges. The use of (un)supervised machine learning methods like artificial neural networks, clustering, random forests, and decision trees, in ocean sciences is an emerging and rapidly evolving field of research. The versatility of machine learning allows for novel applications and revisiting old ones such as pattern recognition, data mining and compression, spatial interpolation, parameter and error estimation, time series analysis and filtering, ocean model and forcing surrogates, observation inversion, forecasting, parameterization and many more. This session aims to examine the current practices, tools, and future direction of artificial intelligence and machine learning in processing all types of oceanographic data and invites contributions tackling key problems of ocean sciences using novel approaches from machine learning. The session is also open to common drawbacks and limitations of machine learning and how to solve them. Contributions that connect or compare machine learning to neighboring statistical or mathematical algorithms as data assimilation, MSSA, AR processes, recurrence networks or Markov chains are highly welcome too.
Co-Sponsor(s):
  • ME - Marine Ecology and Biodiversity
Primary Chair:  Carrie Wall, University of Colorado at Boulder, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, United States
Co-chairs:  Jan Saynisch Wagner, GFZ - Potsdam, Potsdam, Germany and Christopher Irrgang, Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Geodesy and Remote Sensing, Potsdam, Germany
Primary Liaison:  Carrie Wall, University of Colorado at Boulder, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, United States
Moderators:  Carrie Wall, University of Colorado at Boulder, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, United States and Jan Saynisch Wagner, GFZ - Potsdam, Potsdam, Germany
Student Paper Review Liaison:  Carrie Wall, University of Colorado at Boulder, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, United States

Abstracts Submitted to this Session:

A 2D Fully Convolutional Neural Network for nearshore and surf-zone bathymetry inversion from synthetic imagery of the surf-zone using the model Celeris (642325)
Adam Collins1, Katherine L Brodie2, Spicer Bak2, Ty Hesser3, Matthew Farthing3, Douglas W Gamble1 and Joseph W Long4, (1)University of North Carolina at Wilmington, Earth and Ocean Sciences, Wilmington, NC, United States, (2)U.S. Army Engineer Research and Development Center, Coastal and Hydraulics Laboratory, Duck, NC, United States, (3)US Army Corps of Engineers, Coastal and Hydraulics Laboratory, Vicksburg, United States, (4)University of North Carolina at Wilmington, Physics and Physical Oceanography, Wilmington, United States
VIIRS High Spatial Resolution Ocean Color Data Derived Using the Deep Convolutional Networks (643535)
Xiaoming Liu, NOAA College Park, College Park, MD, United States and Menghua Wang, NOAA/NESDIS/STAR, College Park, MD, United States
Deep Learning for Detecting Gulf Stream and Eddies from Satellite Images (646482)
Deepak Subramani, Indian Institute of Science, Computational and Data Sciences, Bangalore, India, Raghav Sharma, Indian Institute of Science, Bangalore, India and Avijit Gangopadhyay, Professor of Oceanography, School for Marine Science and Technology University of Massachusetts Dartmouth 836, S. Rodney French Blvd. New Bedford, MA, New Bedford, MA, United States
OrcaCNN: Detecting and classifying killer whales from passive acoustic data (646602)
Jesse Lopez, Axiom Data Science, Portland, OR, United States and Abhishek Singh, NIT Durgapur, Computer Science and Engineering, Durgapur, India
Using Machine Learning to Find Relationships in Oceanographic Datasets (647542)
Christopher Holder, Johns Hopkins University, Department of Earth and Planetary Sciences, Baltimore, MD, United States and Anand Gnanadesikan, Johns Hopkins Univ-EPS, Baltimore, MD, United States
Extracting submesoscales from drifter observations using adaptive Gaussian Process regression techniques (651995)
Rafael Carvalho Gonçalves1, Mohamed Iskandarani2, George R Halliwell Jr3, Matthieu Le Henaff1 and Gustavo Jorge Goni4, (1)CIMAS/University of Miami, Miami, FL, United States, (2)University of Miami - RSMAS, Miami, FL, United States, (3)NOAA Miami, Miami, FL, United States, (4)NOAA/AOML, Miami, FL, United States
Oh What a Drag! Using Artificial Intelligence to Catalyze Efficient and Cost-Effective Video Review for At-Sea Electronic Monitoring Systems (656779)
Benjamin Woodward, CVision AI, Medford, United States and Craig Heberer, The Nature Conservancy, United States
First Drain, Then Train: Unleash the Full Potential of AI on Large Underwater Datasets (After Removing the Water from Them) (657005)
Derya Akkaynak, Harbor Branch Oceanographic Institution, Florida Atlantic University, Ft. Pierce, United States and Tali Treibitz, University of Haifa, Charney School Of Marine Sciences, Department of Marine Technologies, Haifa, Israel