Static and auto-regressive neural emulation of phytoplankton biomass dynamics from physical predictors in the global ocean

TL;DR

This study demonstrates that deep learning models, especially UNet and its auto-regressive variant, can effectively predict phytoplankton biomass dynamics in the global ocean using physical environmental data, aiding ocean monitoring.

Contribution

It introduces the use of UNet and auto-regressive UNet architectures for phytoplankton prediction, showing improved accuracy over other models and highlighting their potential for ocean health monitoring.

Findings

UNet outperforms CNNs, ConvLSTM, and 4CastNet in phytoplankton prediction.

Auto-regressive UNet improves short-term forecast accuracy.

Models effectively reconstruct and predict phytoplankton dynamics using physical predictors.

Abstract

Phytoplankton is the basis of marine food webs, driving both ecological processes and global biogeochemical cycles. Despite their ecological and climatic significance, accurately simulating phytoplankton dynamics remains a major challenge for biogeochemical numerical models due to limited parameterizations, sparse observational data, and the complexity of oceanic processes. Here, we explore how deep learning models can be used to address these limitations predicting the spatio-temporal distribution of phytoplankton biomass in the global ocean based on satellite observations and environmental conditions. First, we investigate several deep learning architectures. Among the tested models, the UNet architecture stands out for its ability to reproduce the seasonal and interannual patterns of phytoplankton biomass more accurately than other models like CNNs, ConvLSTM, and 4CastNet. When using one to two months of environmental data as input, UNet performs better, although it tends to underestimate the amplitude of low-frequency changes in phytoplankton biomass. Thus, to improve predictions over time, an auto-regressive version of UNet was also tested, where the model uses its own previous predictions to forecast future conditions. This approach works well for short-term forecasts (up to five months), though its performance decreases for longer time scales. Overall, our study shows that combining ocean physical predictors with deep learning allows for reconstruction and short-term prediction of phytoplankton dynamics. These models could become powerful tools for monitoring ocean health and supporting marine ecosystem management, especially in the context of climate change.