Abstract. SEAPODYM-LMTL is a global advection-diffusion-reaction model that simulates age-structured zooplankton and micronekton populations driven by physical and biogeochemical forcing. This study introduces SeapoPym, a simplified version of this model that decouples biological dynamics from physical transport and incorporates a Genetic Algorithm (GA) for stochastic parameter estimation within the Python scientific ecosystem. Comparisons with SEAPODYM-LMTL show that omitting transport produces notable discrepancies in highly dynamic warm regions and cold environments with long zooplankton life cycles. However, SeapoPym remains suitable for simulating mesozooplankton across most warm- and temperate-ocean regions. Sobol sensitivity analysis identifies mortality parameters as key drivers of biomass magnitude and variability, with strong parameter interactions. Twin experiments highlight challenges in estimating recruitment-timing parameters and emphasize the importance of data from cold, contrasting environments. SeapoPym provides a flexible, low-cost framework for exploring parameter estimation, designing observational strategies, and addressing challenges in zooplankton model assessment, with the potential to integrate with circulation models or machine-learning emulators.

Abstract. The predictive accuracy of marine biogeochemical models is fundamentally limited by uncertainty in their parameter values. We present a parameter optimization framework using iterative Importance Sampling (iIS) to constrain the PISCES model by leveraging the rich, multi-variable dataset provided by Biogeochemical-Argo (BGC-Argo) floats. Using data from a BGC-Argo float in the North Atlantic, we assimilate a comprehensive suite of 20 biogeochemical metrics to constrain all 95 parameters of the PISCES model within a 1D vertical configuration. Our global sensitivity analysis (GSA) identifies parameters controlling zooplankton dynamics as the dominant source of model sensitivity for this specific site. We compare three strategies: (1) optimizing a subset of parameters for their strong direct influence (Main effects); (2) optimizing a larger subset that also includes parameters influential through non-linear interactions (Total effects); and (3) simultaneously optimizing all 95 parameters. All three approaches achieve a statistically indistinguishable and significant improvement in model skill, reducing Normalized Root Mean Square Error (NRMSE) by 54–56 %. The rich, multi-variable dataset provides sufficient orthogonal constraints to yield posterior parameter distributions with negligible inter-correlation, shifting the long-standing challenge of correlated equifinality to uncorrelated equifinality, where a range of optimal parameter sets can be found independently. Parameter uncertainty is reduced by 16–41 %, and the optimized ensembles demonstrate strong portability. While all strategies produce a similar, tightly constrained predictive spread for the assimilated variables, they differ significantly in computational cost and in their estimation of uncertainty for unobserved parts of the model. The prerequisite GSA was ~40 times more computationally expensive than the optimization, while the All-parameters strategy, by exploring the full parameter space, provides a more comprehensive and robust quantification of the model's uncertainty in unassimilated variables. We therefore conclude that directly optimizing all model parameters is the recommended strategy. This work delivers a validated, parameter set for the North Atlantic and demonstrates a scalable framework to advance biogeochemical modeling from using static, globally-uniform parametrization to developing a map of regionally-tuned parameters.

Numerical models of ocean biogeochemistry serve as critical tools for detecting and predicting the impacts of climate change on marine resources, and for monitoring ocean health. Recent research by Rodgers et al. (2023) has identified significant limitations in current CMIP6 class models, particularly in their representation of seasonal partial pressure of CO2 (pCO2) temporal phasing and magnitude in key ocean regions such as the North Atlantic. These limitations arise primarily from parameter uncertainty, as model parameters are typically derived from laboratory experiments using a limited range of species that do not represent the diversity of marine organisms. In addition, certain parameters remain experimentally indeterminate, resulting in wide plausible ranges that introduce considerable uncertainty into model predictions.Our research addresses these challenges through a comprehensive approach to parameter optimisation using ensemble-based data assimilation techniques. In particular, we focus on reducing the systematic bias in the PISCES marine biogeochemical model distributed with NEMO v4.2, simulation of the North Atlantic seasonal pCO2 cycle, while generating robust uncertainty estimates through ensemble methods.The optimization process began with an extensive sensitivity analysis using SOBOL indices to identify the parameters most influential in controlling seasonal pCO2 dynamics. We then implemented a particle filter algorithm to optimize these key parameters in a NEMO-PISCES 1D configuration using data from a North Atlantic BGC Argo float. The particle filter generated an ensemble of thousands of state variable solutions, each representing a different PISCES parameterization with reference values varying between 0.01 and 2 times their nominal range. From this ensemble, we identified the ten parameter combinations that most effectively minimized the model-data discrepancy. These optimized parameter sets were then used to generate a 3D regional NEMO-PISCES ensemble in the North Atlantic - the ensemble approach providing a robust framework for uncertainty quantification.Our results show significant improvements in model performance, with the optimized PISCES parameter set in the 1D configuration achieving a 40% reduction in RMSE for seasonal cycle predictions of surface nutrients, chlorophyll, and carbon components compared to the standard PISCES configuration. Most importantly, all ensemble members successfully reproduce seasonal pCO2 phasing and magnitude in agreement with observation-based data, addressing a critical limitation of the reference model while providing uncertainty estimates consistent with observational uncertainties.This research demonstrates the effectiveness of ensemble-based data assimilation techniques in optimizing biogeochemical model parameters, thereby enhancing the accuracy and reliability of ocean simulations. These improvements significantly strengthen our capacity to monitor ocean health, forecast climate change impacts on marine ecosystems, and provide robust scientific guidance for marine resource management decisions.

Ocean primary production is a key process that regulates marine ecosystems and the global climate, but its estimation is still affected by multiple uncertainties. Typically, the chlorophyll-a concentration (CHL) is used to characterise this process, as it is considered as a proxy of phytoplankton biomass. To date, the most common observing systems for studying CHL are ocean colour satellites and Biogeochemical-Argo (BGC-Argo) floats. These are complementary systems: satellite observations provide global coverage but are limited to the ocean surface, while BGC-Argo floats provide punctual observations along the whole water column. Quantitative matching of these two observing systems has been obtained only at regional or single-float scales, while at a global scale the relatively low and irregular BGC-Argo coverage results in large uncertainties. Here, we propose a different method, by comparing satellite and BGC-Argo climatological annual time series within seven different bioregions, each characterised by a homogeneous phytoplankton phenology, allowing us to smooth the uncertainties. By comparing the mean values, amplitudes, and shapes of the two time series, we identify regions (a) where they agree (58%-61% of the ocean surface area); (b) regions undersampled by the BGC-Argo float network (particularly in the Arabian Sea and near the Amazon delta); (c) where the discrepancy may stem from satellite or (d) BGC-Argo performance (mainly found at subtropical and high latitudes, respectively). Caution is required when using BGC-Argo and satellite data in regions b-d, and, for each region, we provide suggestions on which system could be affected by the largest uncertainties.

AbstractMarine heatwaves (MHWs) represent anomalously warm temperature conditions of seawater that may affect marine life and ocean biogeochemistry. Under such conditions, phytoplankton communities may modify their structure and functions, and their resilience is not assured. This study characterizes the impact of MHWs on the phytoplankton spring bloom in the North‐Western Mediterranean Sea. Here, we synergistically combine autonomous observations from BioGeoChemical‐Argo floats, satellite‐based and marine ecosystem model data, and show that MHW events occurring during winter drastically inhibit phytoplankton carbon biomass in spring by up to 70%. Such reduction is related to the enhanced stratification of the water column under MHWs which hinders the renewal of nutrients from deep‐ocean reservoirs, thus preventing surface phytoplankton from blooming. This process negatively impacts particulate organic carbon stocks within the mixed layer, while severe events cause an earlier shift of phytoplankton phenology that provokes changes in zooplankton biomass distribution.

Quantification of oxygen uptake at the ocean surface and its surface-to-interior pathways is crucial for understanding oxygen concentration change in a warming ocean. We investigate the mean meridional global oxygen transport between 1950 and 2009 using coupled physical-biogeochemical model output. We introduce a streamfunction in latitude-oxygen coordinates to reduce complexity in the description of the mean meridional oxygen pathways. The meridional oxygen transport occurs in two main cells: (a) the Northern Cell, dominated by the Atlantic Meridional Overturning Circulation, is nearly adiabatic in the Northern Hemisphere, and transports well oxygenated waters equatorward; (b) The Southern Cell, strongly diabatic, is shaped by the circulation in the Indo-Pacific basin, and combines the subtropical and abyssal meridional circulation cells when represented in depth-latitude coordinates. Analysis of isopycnal meridional oxygen transport shows that the northward flow from the Southern Ocean transports well oxygenated waters within intermediate and bottom layers, while oxygen-rich waters reach the Southern Ocean within deep layers (27.6 < sigma(0) < 27.9 kg m(-3)), carried by the North Atlantic Deep Water (NADW). This oxygenated NADW loses around 25% of its oxygen concentration along its meridional journey from the North Atlantic (at 55 degrees N) to the Southern Ocean. These insights into the oxygen dynamics as driven by the meridional overturning circulation provide a new framework for future studies on ocean deoxygenation.

The biological carbon pump (BCP) contributes to the oceanic CO2 sink by transferring particulate organic carbon (POC) into the deep ocean. The magnitude and efficiency of the BCP is likely to vary on timescales of days to seasons, however characterising this variability from shipboard observations is challenging. High resolution, sustained observations of primary production and particle fluxes by autonomous vehicles offer the potential to fill this knowledge gap. Here we present a 4 month, daily, 1 m vertical resolution glider dataset, collected in the high productivity bloom, downstream of South Georgia, Southern Ocean. The dataset reveals substantial temporal variability in primary production, POC flux and attenuation. During the pre-bloom peak phase we find high export efficiency, implying minimal heterotrophic POC consumption, i.e. productivity is decoupled from upper ocean remineralisation processes. As the bloom progresses from its peak through its declining phase, export flux decreases, but transfer efficiency within the upper 100 m of the mesopelagic increases. Conversely, transfer efficiency in the lower mesopelagic decreases in the post-bloom phase, implying that the flux attenuation processes operating in the upper and lower mesopelagic are effectively decoupled. This finding underscores an important limitation of using a single parameter, such as Martin's ‘b’, to characterise POC flux attenuation in a given location or season. Frequent pulses of export flux are observed throughout the deployment, indicating decoupling between primary production and the processes driving export of material from the upper ocean. The mechanisms underlying the observed seasonal changes in BCP magnitude and efficiency are unclear, as temperature and oxygen concentration changed minimally, although the nature of the sinking particles changed substantially as the bloom progressed. Our results highlight the difficulty of capturing temporal variability and episodic flux events with traditional shipboard observations, which affects our conceptual understanding of the BCP. The increasing use of autonomous vehicles to observe particle fluxes will be essential to characterising the temporal variability in magnitude and functioning of the BCP.

Numerical models of ocean biogeochemistry are becoming the major tools used to detect and predict the impact of climate change on marine resources and to monitor ocean health. However, with the continuous improvement of model structure and spatial resolution, incorporation of these additional degrees of freedom into fidelity assessment has become increasingly challenging. Here, we propose a new method to provide information on the model predictive skill in a concise way. The method is based on the conjoint use of a k-means clustering technique, assessment metrics, and Biogeochemical-Argo (BGC-Argo) observations. The k-means algorithm and the assessment metrics reduce the number of model data points to be evaluated. The metrics evaluate either the model state accuracy or the skill of the model with respect to capturing emergent properties, such as the deep chlorophyll maximums and oxygen minimum zones. The use of BGC-Argo observations as the sole evaluation data set ensures the accuracy of the data, as it is a homogenous data set with strict sampling methodologies and data quality control procedures. The method is applied to the Global Ocean Biogeochemistry Analysis and Forecast system of the Copernicus Marine Service. The model performance is evaluated using the model efficiency statistical score, which compares the model–observation misfit with the variability in the observations and, thus, objectively quantifies whether the model outperforms the BGC-Argo climatology. We show that, overall, the model surpasses the BGC-Argo climatology in predicting pH, dissolved inorganic carbon, alkalinity, oxygen, nitrate, and phosphate in the mesopelagic and the mixed layers as well as silicate in the mesopelagic layer. However, there are still areas for improvement with respect to reducing the model–data misfit for certain variables such as silicate, pH, and the partial pressure of CO2 in the mixed layer as well as chlorophyll-a-related, oxygen-minimum-zone-related, and particulate-organic-carbon-related metrics. The method proposed here can also aid in refining the design of the BGC-Argo network, in particular regarding the regions in which BGC-Argo observations should be enhanced to improve the model accuracy via the assimilation of BGC-Argo data or process-oriented assessment studies. We strongly recommend increasing the number of observations in the Arctic region while maintaining the existing high-density of observations in the Southern Oceans. The model error in these regions is only slightly less than the variability observed in BGC-Argo measurements. Our study illustrates how the synergic use of modeling and BGC-Argo data can both provide information about the performance of models and improve the design of observing systems.

Analysis of global numerical experiments with physical and BGC forecasting model to estimate the impact of new observing system design.

The accuracy of the Copernicus Marine Environment and Monitoring Service (CMEMS) ocean analysis and forecasts highly depend on the availability and quality of observations to be assimilated. In situ observations are complementary to satellite observations that are restricted to the ocean surface. Higher resolution model forecasts are required by users of the CMEMS global and regional ocean analysis and forecasts. To support this with an efficient observational constrain of the model forecast via data assimilation, an increase observation coverage is needed, associated with an improved usage of the available ocean observations. This work exploits the capabilities of operational systems to provide comprehensive information for the evolution of the GOOS. In this report, we analyse the use and the efficiency of the in-situ observations to constrain regional and global Mercator Ocean systems. Physical and biogeochemical variables are considered. The in-situ observations are used either to estimate physical ocean state at global and regional scale via data assimilation or to estimate BGC model parameters. The impact of the physical in situ observations assimilated in open ocean and coastal areas is assessed with numerical data assimilation experiments. The experiments are conducted with the regional 1/36° resolution and global 1/12° resolution systems operated by Mercator Ocean for the Copernicus Marine Service. For the global physical ocean, the focus is on the tropical ocean to better understand how the tropical mooring observations constrain the intraseasonal to daily variability and the complementarity with satellite observations and the deep ocean. The tropical moorings provide unique high frequency observations at different depth, but they are far away from each other, so part of the signal in the observation are decorrelated from one mooring to the others. It is only via an integrated approach, as data assimilation into a dynamical model and complementarity with other observing networks that those observations can efficiently constrain the different scales of variability of the tropical ocean circulation. As the satellite observations brings higher spatial resolution between the tropical moorings but for the ocean surface, we show that the tropical mooring and Argo profile data assimilation constrain the larger scale ocean thermohaline vertical structure (EuroSea D2.2; Gasparin et al., 2023). The representation of the high frequency signals observed at mooring location is also significantly improved in the model analysis compared to a non-assimilative simulation. The ocean below 2000 m depth is still largely under constrained as very few observations exist. Some deep ocean basins, as the Antarctic deep ocean, shows significant trend over the past decade but they are still not accurately monitored. Based on the spread of four deep ocean reanalysis estimates, large uncertainties were estimated in representing local heat and freshwater content in the deep ocean. Additionally, temperature and salinity field comparison with deep Argo observations demonstrates that reanalysis errors in the deep ocean are of the same size as or even stronger than the observed deep ocean signal. OSSE already suggested that the deployment of a global deep Argo array will significantly constrain the deep ocean in reanalysis to be closer to the observations (Gasparin et al., 2020). At regional and coastal scales, the physical ocean circulation is dominated by higher frequency, smaller scale processes than the open ocean which requires different observation strategy to be well monitor. The impact of assimilating high frequency and high-resolution observations provided by gliders on European shelves is analysed with the regional Iberic Biscay and Irish (IBI) system. It was found that repetitive glider sections can efficiently help to constrain the transport of water masses flowing across those sections. BGC ocean models are less mature than physical ocean models and some variable dependencies are still based on empirical functions. In this task, Argo BGC profile observations were used to optimize the parameters of the global CMEMS biogeochemical model, PISCES. A particle filter algorithm was chosen to optimize a 1D configuration of PISCES in the North Atlantic. The optimization of the PISCES 1D model significantly improves the model's ability to reproduce the North Atlantic bloom Recommendations on the in-situ network extensions for real time ocean monitoring are given based on those results, and the one also obtained in the WP2, Task 2.2 where data assimilation experiments but with simulated observations where conducted. Argo extension and the complementarity with satellite altimetry was also extensively studied. (EuroSea Deliverable ; D4.8)

The Rhodes Gyre is a cyclonic persistent feature of the general circulation of the Levantine Basin in the eastern Mediterranean Sea. Although it is located in the most oligotrophic basin of the Mediterranean Sea, it is a relatively high primary production area due to strong winter nutrient supply associated with the formation of Levantine Intermediate Water. In this study, a 3D coupled hydrodynamic- biogeochemical model (SYMPHONIE/Eco3M-S) was used to characterize the seasonal and interannual variability of the Rhodes Gyre's ecosystem and to estimate an annual organic carbon budget over the 2013-2020 period. Comparisons of model outputs with satellite data and compiled in situ data from cruises and Biogeochemical-Argo floats revealed the ability of the model to reconstruct the main seasonal and spatial biogeochemical dynamics of the Levantine Basin. The model results indicated that during the winter mixing period, phytoplankton first progressively grow sustained by nutrient supply. Then, short episodes of convection driven by heat loss and wind events, favoring nutrient injections, organic carbon export, and inducing light limitation on primary production, alternate with short episodes of phytoplankton growth. The estimate of the annual organic carbon budget indicated that the Rhodes Gyre is an autotrophic area, with a positive net community production in the upper layer (0- 150 m) amounting to 31.2 +/- 6.9 g Cm-2 yr(-1). Net community production in the upper layer is almost balanced over the 7-year period by physical transfers, (1) via downward export (16.8 +/- 6.2 g C m(-2) yr(-1)) and (2) through lateral transport towards the surrounding regions (14.1 +/- 2.1 g C m(-2) yr(-1)). The intermediate layer (150-400 m) also appears to be a source of organic carbon for the surrounding Levantine Sea (7.5 +/- 2.8 g Cm-2 yr(-1)) mostly through the subduction of Levantine Intermediate Water following winter mixing. The Rhodes Gyre shows high interannual variability with enhanced primary production, net community production, and exports during years marked by intense heat losses and deep mixed layers. However, annual primary production appears to be only partially driven by winter vertical mixing. Based on our results, we can speculate that future increase of tem perature and stratification could strongly impact the carbon fluxes in this region.

Regional processes play a key role in the global carbon budget. Major ocean carbon uptake at mid-latitudes counteracts carbon release in the tropics, which is modulated by episodes of marine heatwaves (MHWs). Yet, we lack essential knowledge on persistent MHWs (PMHWs), and their effect on the carbon sensitive areas. Here, based on a 1985-2017 joint analysis of reconstructions, ocean reanalysis, in situ and satellite data, we show that PMHWs occur in major carbon uptake and release areas. Air-sea CO2 flux changes from PMHWs are strongest in the Pacific Ocean with a 35 +/- 2 % reduction in CO2 release in the tropics linked to ENSO, and a reduction in CO2 uptake of 28 +/- 9 % in the North Pacific. These results provide new insights into the interplay of extreme variability and a critical regulating ocean ecosystem service, and pave the way for future investigations on its evolution under climate change.

Abstract. This study assesses marine biological production of organic carbon based on the diel variability of bio-optical properties monitored by two BioGeoChemical-Argo (BGC-Argo) floats. Experiments were conducted in two distinct Mediterranean systems, the Northwestern Ligurian Sea and the Central Ionian Sea during summer months. We derived particulate organic carbon (POC) stock and gross community production integrated within the surface, euphotic and subsurface chlorophyll maximum (SCM) layers, using an existing approach applied to diel cycle measurements of the particulate beam attenuation (cp) and backscattering (bbp) coefficients. The diel cycle of cp provided a robust proxy for quantifying biological production in both systems; that of bbp was comparatively less robust. Derived primary production estimates vary by a factor of 2 depending upon the choice of the bio-optical relationship that converts the measured optical coefficient to POC, which is thus a critical step to constrain. Our results indicate a substantial, yet variable, contribution to the water column production of the SCM layer (16–42%). In the Ligurian Sea, the SCM is a seasonal feature that behaves as a subsurface biomass maximum (SBM) with the ability to respond to episodic abiotic forcing by increasing production. In contrast, in the Ionian Sea, the SCM is permanent, induced by phytoplankton photoacclimation and contributes moderately to water column production. These results emphasize the strong potential for transmissometers deployed on BGC-Argo profiling floats to quantify non-intrusively in situ biological production of organic carbon in the water column of stratified oligotrophic systems with recurring or permanent SCMs, which are widespread features in the global ocean.

The Mediterranean Sea has been identified as a hotspot for climate change. Furthermore, its very diverse trophic regimes, in such a little area, make it an extremely interesting region from a biogeochemical perspective. Numerous studies aim at better understanding and representing the Mediterrenean dynamics and biogeochemistry through modeling. This is a crucial step in order to predict the future anthropogenic impacts on the Mediterranean Sea and their possible effects on its biogeochemistry, and all what depends on it. The number of models that simulate the Mediterranean biogeochemistry, and the data available to compare with are now sufficient to draw an overall picture of the Mediterranean Sea biogeochemical models state of the art. In this study, we gathered 10 biogeochemical simulations of the Mediterranean Sea, including 8 regional and 2 high-resolution global configurations. The simulations are compared with surface chlorophyll estimates derived from satellite observations; chlorophyll, nitrate, oxygen, and particulate organic carbon concentrations derived from BGC-Argo floats, and phytoplankton group-specific primary production estimated from ocean color satellite observations. Our first aim is to describe and compare all known Mediterranean biogeochemical models, and to highlight their specificity. This should give an insight into the current achievements, and expose what biogeochemical model products are hence available for further ecological analysis. Furthermore, a specific attention is given to how well each model performs in selected regions of the Mediterranean Sea, in order to understand which specific process is needed to adequately represent the different trophic regimes of the Mediterranean Sea.

The MALINA oceanographic campaign was conducted during summer 2009 to investigate the carbon stocks and the processes controlling the carbon fluxes in the Mackenzie River estuary and the Beaufort Sea. During the campaign, an extensive suite of physical, chemical and biological variables were measured across seven shelf–basin transects (south–north) to capture the meridional gradient between the estuary and the open ocean. Key variables such as temperature, absolute salinity, radiance, irradiance, nutrient concentrations, chlorophyll a concentration, bacteria, phytoplankton and zooplankton abundance and taxonomy, and carbon stocks and fluxes were routinely measured onboard the Canadian research icebreaker CCGS Amundsen and from a barge in shallow coastal areas or for sampling within broken ice fields. Here, we present the results of a joint effort to compile and standardize the collected data sets that will facilitate their reuse in further studies of the changing Arctic Ocean. The data set is available at https://doi.org/10.17882/75345 (Massicotte et al., 2020).

Recent observations have shown that phytoplankton biomass increases in the North Atlantic during winter, even when the mixed layer is deepening and light is limited. Current theories suggest that this is due to a release from grazing pressure. Here we demonstrate that the often-used grazing models that are linear at low phytoplankton concentration do not allow for a wintertime increase in phytoplankton biomass. However, mathematical formulations of grazing as a function of phytoplankton concentration that are quadratic at low concentrations (or more generally decrease faster than linearly as phytoplankton concentration decreases) can reproduce the fall to spring transition in phytoplankton, including wintertime biomass accumulation. We illustrate this point with a minimal model for the annual cycle of North Atlantic phytoplankton designed to simulate phytoplankton concentration as observed by BioGeoChemical-Argo (BGC-Argo) floats in the North Atlantic. This analysis provides a mathematical framework for assessing hypotheses of phytoplankton bloom formation.

Stratified oceanic systems are characterized by the presence of a so-called Deep Chlorophyll a Maximum (DCM) not detectable by ocean color satellites. A DCM can either be a phytoplankton (carbon) biomass maximum (Deep Biomass Maximum, DBM), or the consequence of photoacclimation processes (Deep photoAcclimation Maximum, DAM) resulting in the increase of chlorophyll a per phytoplankton carbon. Even though these DCM (further qualified as either DBMs or DAMs) have long been studied, no global-scale assessment has yet been undertaken and large knowledge gaps still remain in relation to the environmental drivers responsible for their formation and maintenance. In order to investigate their spatial and temporal variability in the open ocean, we use a global data set acquired by more than 500 Biogeochemical-Argo floats given that DCMs can be detected from the comparative vertical distribution of chlorophyll a concentrations and particulate backscattering coefficients. Our findings show that the seasonal dynamics of the DCMs are clearly region-dependent. High-latitude environments are characterized by a low occurrence of intense DBMs, restricted to summer. Meanwhile, oligotrophic regions host permanent DAMs, occasionally replaced by DBMs in summer, while subequatorial waters are characterized by permanent DBMs benefiting from favorable conditions in terms of both light and nutrients. Overall, the appearance and depth of DCMs are primarily driven by light attenuation in the upper layer. Our present assessment of DCM occurrence and of environmental conditions prevailing in their development lay the basis for a better understanding and quantification of their role in carbon budgets (primary production and export).