Modeling pCO2,sw in the Canary Islands region based on satellite measurements

The improvement of remote sensing systems together with the emergence of new model fitting algorithms based on sophisticated methods, such as machine-learning techniques, have allowed the determination of the partial pressure of carbon dioxide (pCO_2,sw) in the Canary Islands waters based on mathematical modeling. Among all the fitted models, the most powerful one seems to be the bootstrap aggregation (bagging), giving an RMSE < 6 µatm (R² > 0.95), although the multilinear regression (MLR), neural network (NN) and categorical boosting (CatBoost) also have a good predictive performance, with RMSE ranging from 9 to 13 µatm for 360 < pCO_2,sw < 481 µatm. Using the most reliable model that uses sea surface temperature (SST), Chlorophyll a (Chla), and mixed layer depth (MLD), it was determined that during the period comprised between 2019 and 2024, the Canary basin behaved as a slight net sink of atmospheric CO₂, with an average daily flux of -1.45 ± 0.08 mmol m^-2 d^-1, resulting in the sequestration of -2.59 ± 0.15 Tg CO₂ yr^-1.