Machine learning model selection for predicting bathymetry. (July 2022)

• Record Type: Journal Article
• Authors: Moran, Nicholas; Stringer, Ben; Lin, Bruce; Hoque, Md Tamjidul
• Is Part Of: Deep sea research. Volume 185(2022)
• Publication Date: 2022-07
• Subjects: Machine learning -- Predicting bathymetry -- Earth gravitational models -- Classification -- Genetic algorithms; Oceanography -- Periodicals; Océanographie -- Périodiques; 551.4605
• DOI: 10.1016/j.dsr.2022.103788
• Languages: English
• ISSNs: 0967-0637

Abstract

Abstract: This research investigates the viability of using Machine Learning (ML) for predicting bathymetry. We built and trained several models using ocean features aggregated from multiple sources and predicted bathymetry from the ETOPO dataset at a 2-min resolution. Each model was evaluated to identify a global best fit, however we found that none performed well on a global scale. When training on subsets of the world, we observed that some models performed significantly better, which led to developing a novel model selection technique that identifies the best performing model and most relevant features for a given geospatial coverage. This leads to improved predictions and more reliable results. This model selection technique can be generalized to be applied to any set of models. Highlights: Viability of Machine Learning for predicting bathymetry. Feature-based selective model outperforms global model in bathymetry prediction. Compared a plethora of Machine Learning approaches. Divided the world into a set of coverages based on computational expenses.