ORCAst: Operational High-Resolution Current Forecasts
This provides improved operational forecasts for oceanography and related fields, but is incremental as it builds on existing methods with architectural and training refinements.
The paper tackles the problem of forecasting global ocean surface currents over one week using satellite and in situ data, achieving stronger nowcast and forecast performance than state-of-the-art methods.
We present ORCAst, a multi-stage, multi-arm network for Operational high-Resolution Current forecAsts over one week. Producing real-time nowcasts and forecasts of ocean surface currents is a challenging problem due to indirect or incomplete information from satellite remote sensing data. Entirely trained on real satellite data and in situ measurements from drifters, our model learns to forecast global ocean surface currents using various sources of ground truth observations in a multi-stage learning procedure. Our multi-arm encoder-decoder model architecture allows us to first predict sea surface height and geostrophic currents from larger quantities of nadir and SWOT altimetry data, before learning to predict ocean surface currents from much more sparse in situ measurements from drifters. Training our model on specific regions improves performance. Our model achieves stronger nowcast and forecast performance in predicting ocean surface currents than various state-of-the-art methods.