BG4Sea: Biogeochemical Seasonal Forecastability via Progressive Information Scaling
Machine Learning
Summary
The authors developed BG4Sea, a new computer model that predicts important ocean chemistry and biology changes up to six months ahead. It uses a smart system to simplify complex ocean data, learns patterns from past data, and includes information from nearby areas and physical ocean conditions. They tested it with real ocean data and found it predicts better than simpler methods that assume things stay the same or just follow average conditions. The authors see BG4Sea as a useful starting point for building even better ocean forecasting tools in the future. They also studied which parts of the model help most with making good predictions and noted some limitations in its current design.
Authors
Gabriela Martinez Balbontin, Anastase Charantonis, Dominique Bereziat, Stefano Ciavatta
Abstract
Marine biogeochemical forecasting is increasingly important for managing marine ecosystems and the carbon cycle, yet global, seasonal forecast products lag far behind physical oceanography, held back by the complexity of the processes involved and by data scarcity. We introduce BG4Sea, which to our knowledge is the first global, data-driven system to produce multivariate seasonal forecasts of the marine biogeochemical state. BG4Sea is a modular architecture with a column autoencoder that compresses the vertical column into a low-dimensional latent space, a latent forecaster propagates this representation forward in time, a surface-forcing conditioner that injects physical boundary information via Feature-wise Linear Modulation (FiLM), and a horizontal-coupling module that incorporates neighboring-column context through cross-attention. The model is trained and evaluated on the global ocean reanalysis BIORYS4 (NEMO/PISCES), and produces six-month forecasts at 1/4 degree, monthly resolution for dissolved chemistry, biology, and carbon-pool variables, outperforming persistence and climatology across most variables and lead times. We position BG4Sea as an interpretable baseline for future, more expressive approaches, and discuss predictability attribution to each component, alongside the model's structural limitations.