Highlights
- The study employed FFNN, RF, and TabNet ML models to estimate Ωcal using in-situ measurements and satellite-derived data.
- The TabNet model outperformed with significantly low errors (RMSE=0.39, MNB=0.0058, MRE=0.019) and high accuracy (R2=0.96).
- SST strongly correlated with Ωcal (r=0.98), SSS moderately (r=0.49), and Chla showed a weak negative correlation (r=-0.27).
- The study observed seasonal Ωcal variability, with higher values in summer months, notably in temperate and polar regions.
- Emphasized a declining trend of Ωcal from 2012 to 2022, likely influenced by changing oceanic and atmospheric conditions.
Abstract
The accurate estimation of surface ocean calcium carbonate saturation (Ωcal) is crucial for understanding the impacts of ocean acidification (OA) on marine ecosystems, particularly for calcifying organisms. This study investigates the estimation of global surface ocean Ωcal using machine learning (ML) models and satellite-derived data. Three ML models such as feed-forward neural networks (FFNN), random forests (RF), and Tabularnet (TabNet) were employed to estimate Ωcal, utilizing in-situ and satellite measurements of sea surface temperature (SST), sea surface salinity (SSS), and Chlorophyll-a concentration (Chla). Among these, the TabNet model exhibited superior performance, with a root-mean-square error (RMSE) of 0.39, mean relative error (MRE) of 0.019, mean normalized bias (MNB) of 0.0058 and coefficient of determination (R2) of 0.96. SST showed a strong positive correlation with Ωcal (r = 0.95), while SSS and Chla exhibited moderate positive (r = 0.49) and weak negative (r = −0.27) correlations, respectively. The study revealed significant spatiotemporal variability in Ωcal, driven by seasonal changes and ocean circulation patterns. Sensitivity analysis highlighted the robustness of the TabNet model, maintaining high predictive capability despite variations in SST, SSS, and Chla. The TabNet model high accuracy provides a valuable tool for monitoring and forecasting changes in ocean chemistry, informing conservation efforts and policy-making. This study emphasizes the importance of advanced ML models in marine science and their potential for enhancing our understanding of global oceanic processes.
Shaik I., Nagamani P. V., Yadav S., Manmode Y. & Rao G. S., 2025. Advanced deep learning technique for estimating global surface ocean calcium carbonate saturation (Ωcal). Marine Chemistry 268: 104483. doi: 10.1016/j.marchem.2024.104483. Article (subscription required).
