Forecasting Vertical Profiles of Ocean Currents from Surface Characteristics: A Multivariate Multi-Head Convolutional Neural Network–Long Short-Term Memory Approach

• Published in: Journal of marine science and engineering 2023-10, Vol.11 (10), p.1964
• Main Authors: Kar, Soumyashree, McKenna, Jason R., Anglada, Glenn, Sunkara, Vishwamithra, Coniglione, Robert, Stanic, Steve, Bernard, Landry
• Format: Article
• Language: English
• Subjects: Algorithms; Artificial neural networks; chained multivariate multi-output regression; Deep learning; Depth; Design; encoder–decoder model; Forecasting; Long short-term memory; Machine learning; Missing data; Modelling; Multivariate analysis; nearshore monitoring; Neural networks; Ocean circulation; Ocean currents; Ocean dynamics; Ocean models; Oceanic analysis; Reconstruction; Regions; Remote sensing; Salinity; Sea state; Sea states; Sensors; State variable; Support vector machines; Surface properties; Time series; time series data reconstruction; vertical profile modeling; Vertical profiles
• ISSN: 2077-1312; 2077-1312
• DOI: 10.3390/jmse11101964

While study of ocean dynamics usually involves modeling deep ocean variables, monitoring and accurate forecasting of nearshore environments is also critical. However, sensor observations often contain artifacts like long stretches of missing data and noise, typically after an extreme event occurrence or some accidental damage to the sensors. Such data artifacts, if not handled diligently prior to modeling, can significantly impact the reliability of any further predictive analysis. Therefore, we present a framework that integrates data reconstruction of key sea state variables and multi-step-ahead forecasting of current speed from the reconstructed time series for 19 depth levels simultaneously. Using multivariate chained regressions, the reconstruction algorithm rigorously tests from an ensemble of tree-based models (fed only with surface characteristics) to impute gaps in the vertical profiles of the sea state variables down to 20 m deep. Subsequently, a deep encoder–decoder model, comprising multi-head convolutional networks, extracts high-level features from each depth level’s multivariate (reconstructed) input and feeds them to a deep long short-term memory network for 24 h ahead forecasts of current speed profiles. In this work, we utilized Viking buoy data, and demonstrated that with limited training data, we could explain an overall 80% variation in the current speed profiles across the forecast period and the depth levels.