Investigation of the Suess effect in the South-West Indian Ocean over the last two decades – A model-data study

Abstract

Measurements of dissolved inorganic carbon (DIC) concentration and its isotopic composition (δ13CDIC) are essential to study chemical and biological processes involved in the ocean carbon cycle, including photosynthesis, respiration, and air-sea CO2 fluxes. Anthropogenic CO2 emissions from fossil fuel combustion have caused an increase in DIC accompanied by a decline in δ13CDIC (called the Suess effect). δ13CDIC is thus a useful tracer to assess the oceanic uptake of anthropogenic CO2. Annual assessments of the Global Carbon Budget (e.g. Friedlingstein et al., 2023) have revealed a growing deviation over the last 10 to 15 years between the estimates of the ocean carbon sink based on observations and models, with the growth of the observation-based ocean CO2 sink being larger compared to the models. Discrepancies in the multi-decadal trend originate from all latitudes but are greatest in the Southern Ocean. Here, we present DIC and δ13CDIC measurements from surface and water column samples collected in the South-West Indian Ocean during repeated summer cruises over the last two decades (1998-2021) conducted on board the RV Marion Dufresne within the French monitoring program OISO (Océan Indien Service d’Observation). We compare these measurements with the DIC and δ13CDIC simulated over the same period by the δ13C-enabled version of the NEMO-PISCES ocean-biogeochemical model. We use different methods to separate the natural and anthropogenic signals over the last 20 years. Our analysis reveals some inconsistencies between simulated and observed DIC and δ13CDIC, as well as between other simulated and observed biogeochemical parameters, whereas physical parameters are generally well reproduced by the model. Identifying the cause for this mismatch bears the potential to explain all or part of the divergence between the observation-based and model-based estimates of oceanic carbon uptake.