Quantifying how small-scale, short-lived, advective and biologically driven processes alter the carbon uptake capacity in a shelf sea

Marine uptake of carbon dioxide limits the atmospheric concentration growth. Continental shelf seas are important areas for this uptake, but are also highly variable environments, with indications that their sink capacity is weakening. A way to reduce uncertainty of budgeting is to increase our observational capacity, such as through FerryBox installations on Ships-of-Opportunity. Here, we compare FerryBox observations in the North Sea for periods of interest in the autumn seasons of 2019 and 2020. We show that short-lived and small to medium-scale events can be identified when the sampling resolution is adequately high, and that these events cause changes in some essential environmental variables on the same magnitude as seasonal cycles. In particular, these events rapidly lowered seawater pCO₂ by 8-10%. In September 2019, an advectively-driven event caused a previously carbon source area (flux of 1.3 ± 0.6 mmol m^-2 day^-1) to be in relative balance with the atmosphere (new flux of −0.04 ± 0.34 mmol m^-2 day^-1). In November 2020, a late autumn bloom caused another previously carbon source area (flux of 2.7 ± 2.1 mmol m^-2 day^-1) to potentially become a carbon sink (flux of −0.6 ± 1.4 mmol m^-2 day^-1 during the bloom). We demonstrate the importance of including such events in regional carbon budget assessments and advocate for the tuning of models in order to capture this small-scale variability.