The Mediterranean Sea, a biodiversity hotspot and one of the most heavily exploited marine regions worldwide, falls short of effective protected areas. To address this gap, we developed an integrated framework combining a spatially explicit multi-species ecosystem model with regional climate and biogeochemical models to assess the ecological and fisheries outcomes of six fully protected marine protected area (MPA) network scenarios, expanded to 30% coverage. Scenarios included current MPA expansions, random placements and science-informed designs based on ecological, social and economic criteria. Science-informed networks consistently outperformed others, yielding higher biomass recovery, greater spillover benefits and, and lower fisheries losses. They also triggered top-down trophic cascades, reversing the “fishing down the food web” trend and enhancing high trophic level biomass. Large, aggregated offshore MPAs offered the best outcomes in balancing biodiversity conservation and sustainable fisheries, underscoring the importance of science-driven planning for achieving global conservation targets in the Mediterranean and beyond.

Abstract A high proportion of the world’s population relies on marine fisheries as a source of food and employment, highlighting the need for sustainable exploitation strategies. However, fisheries management commonly relies on single-species models that overlook ecological interactions and economic trade-offs, which may lead to stocks being exploited above sustainable levels. To address this gap, we developed a novel size-structured, multispecies bioeconomic module integrated within the spatially explicit, individual-based OSMOSE ecosystem model. OSMOSE represents exploited fish communities in which individuals interact through opportunistic, size-dependent predator–prey interactions which are explicitly incorporated into the bioeconomic framework. The bioeconomic model accounts for multiple species and size classes, allowing for size-dependent market prices. Fishing costs and profits are represented using an extension of the Gordon-Schaefer model to multiple interacting species, while consumer demand is modelled using a nested Dixit-Stiglitz utility function with three levels of constant elasticity of substitution for fish commodity, species and size classes. We present a method for estimating parameters for the bioeconomic module when empirical estimates are unavailable, using the North Sea OSMOSE configuration, which comprises 15 species. Our method successfully estimated the six bioeconomic parameters required to operationalise the model. Results indicate that differences in cost parameters were primarily associated with variability in species biomass rather than fishing gear, while fish prices were more strongly influenced by consumer’s demand than by availability. This framework provides a basis for assessing the economic consequences of alternative climate change scenarios and supporting sustainable fisheries management.

Diel vertical migrations in the ocean play a key role in predator-prey dynamics and the functioning of the biological carbon pump. However, changes in ocean conditions including warming and deoxygenation threaten to significantly perturb vertical migration patterns over the twenty-first century. Specifically, vertical migrations over regions of critically low oxygen, known as oxygen minimum zones (OMZs), are likely to be most sensitive to changes in temperature and oxygen. In this study, we apply a simplified prognostic ecosystem model (APECOSM-1D) to changing conditions in the Pacific Ocean OMZ as simulated by 13 Earth System Models from the Coupled Model Intercomparison Project Phase 6 (CMIP6). We find that modeled fish migration depths at a given location in the region may deepen or shoal by over 100 m by the end of the century; however, there are large uncertainties across the CMIP6 ensemble for the geographic pattern of migration depth changes. To reconcile this, we adopt a water mass based approach which aggregates changes into regions defined by their vertical oxygen minimum value. In this framework, we find that fish migration depths over the lowest oxygen core of the OMZ remain stable due to compensating changes in temperature and oxygen. Meanwhile, away from the OMZ core, ocean warming and deoxygenation together drive shallower migration depths in projected conditions.

The ecosystem approach to fisheries is widely recognised as a key management goal, yet its definition and implementation remain debated. Most fisheries management relies on single-species strategies with technical measures to reduce bycatch. However, selective removals disrupt species composition, affecting ecosystem dynamics and resilience. We present a proof-of-concept model based on balanced harvesting that allocates fishing pressure proportionally across three tuna stocks—yellowfin (Thunnus albacares), skipjack (Katsuwonus pelamis), and bigeye tuna (Thunnus obesus)—in the Indian Ocean according to their size-biomass ratios. The model optimises fishing effort by gear using a predefined objective function based on length-based population dynamics, ensuring a balanced harvest while maintaining each tuna species' biomass within its maximum sustainable yield (BMSY) limit. By assigning fishing mortality (F-multiplier) to each fleet, the model aims to maintain, within the bounds of BMSY for each stock, the ecosystem structure (based on size-abundance relationships) over a 20-year simulation. Results indicate significant reductions in fishing mortality across gears relative to 2020 levels. While some gears, such as purse seine free-school, show increased catches and revenues (146%), others, like purse seine log-school, experience declines (-22%). Overall, fishing at BMSY improves total revenues and catches by 51% and 34%, respectively, compared to 2020. This work demonstrates that it is possible to maintain each tuna stock within BMSY bounds by managing fishing fleets while preserving ecosystem structure, a significant goal of the ecosystem approach to fisheries.

The strong advective circulation governing western boundary current systems poses challenges to coastal organism to retain their planktonic larvae and assembly in local communities. However, instabilities around the predominant current, generating features such as eddies, may increase retention and thereby enhance local diversity. Here, we have investigated this hypothesis through high-resolution (1/36°) community-based Lagrangian dispersal modelling experiments conducted along the Pernambuco Plateau and adjacent open ocean off Brazil, which is part of the strong Atlantic Meridional Overturning Circulation. In the highly advective north portion of the plateau, we found that retention is reduced and advection from open ocean is intense. Under such circumstances, the assembly of local communities is likely driven by mass effect, which results in homogeneous communities with reduced biodiversity. In contrast, in the south portion of the plateau, eddies and meanders associated with the interaction of the currents with the plateau topography and the seasonal variability of the South Equatorial Current bifurcation increase local retention and reduce advection from open ocean. In such conditions, both, species sorting and mass effect assembling archetypes, likely drive the distinct and rich biodiversity inhabiting the region to the south of the plateau. The dispersal patterns obtained from the Lagrangian experiments align with the known spatial patterns of biodiversity distribution along the Tropical Southwestern Atlantic and provide important insights for regions and taxa for which knowledge is limited or absent. These results also provide elements for the proper definition of biodiversity conservation and management strategies and for a better understanding of the processes regulating community assembly in highly advective systems around the globe.

The variability and predictability of the Tropical Atlantic primary productivity remains little explored on interannual-to-decadal time scales. Here, we present the results of two studies, in which find a decadal scale variability in phytoplankton abundance that can be predicted three years ahead. The predictions are made with NorCPM, which is a fully coupled climate prediction model with ocean biogeochemistry that assimilates temperature and salinity to reconstruct past variability. From these reconstructions, predictions are initialized that are run freely ten years ahead. We find that the predictability is a result of nutrient pulses that are advected with the southern branch of the South Equatorial Current from the most southern part of the Atlantic, and that then get caught in the Equatorial undercurrent before they reach the surface in the Tropical Atlantic Ocean. A more detailed analysis is being done in order to pinpoint the underlying mechanisms in a forced ocean model, where we find a link to the Pan-Atlantic decadal oscillation.

Systematic analysis of uncertainty is critical for consolidating ecosystem model projections. Sensitivity analysis is an essential step in understanding model uncertainty, but there are many challenges when dealing with complex models. One of these is related to the quantification of uncertainty in model input parameters, which is especially problematic when limited data leads to the use of arbitrary fixed ranges of variation to define parameter uncertainty. We show the drawbacks of this practice and propose an alternative approach based on the parameter reliability criterion. This criterion helps to classify the model parameters according to the source of information used to estimate their values and calculate the ranges of variability for each parameter used in a sensitivity analysis. Our proposed approach presented in this protocol is illustrated by implementing a sensitivity analysis of the OSMOSE marine ecosystem modelling platform applied to the northern Peru Current ecosystem. We compare the results from the sensitivity analysis based on the parameter reliability criterion with those obtained using fixed ranges of variation. We find that using arbitrary uncertainty ranges can produce different conclusions compared to alternative approaches, such as the one based on the reliability of the parameters. The parameter reliability criterion can be helpful in situations where direct quantification of uncertainty in model inputs is not available.

The Mediterranean Sea, a biodiversity hotspot and one of the most heavily exploited marine regions worldwide, falls short of efficient protected areas. To address this gap, we developed an integrated framework combining a spatially explicit multi-species ecosystem model with regional climate and biogeochemical models to assess the ecological and fisheries outcomes of six fully protected marine protected area (MPA) network scenarios, expanded to 30% coverage. Scenarios included current MPA expansions, random placements and science-informed designs based on ecological, social and economic criteria. Science-informed networks consistently outperformed others, yielding higher biomass recovery, greater spillover benefits and, and lower fisheries losses. They also triggered top-down trophic cascades, reversing the “fishing down the food web” trend and enhancing high trophic level biomass. Large, aggregated offshore MPAs offered the best outcomes in balancing biodiversity conservation and sustainable fisheries, underscoring the importance of science-driven planning for achieving global conservation targets in the Mediterranean and beyond.

Abstract Climate change is anticipated to considerably reduce global marine fish biomass, driving marine ecosystems into unprecedented states with no historical analogs. The Time of Emergence (ToE) marks the pivotal moment when climate conditions (i.e., signal) deviate from pre‐industrial norms (i.e., noise). Leveraging ensemble climate‐to‐fish simulations from one Earth System Model (IPSL‐CM6A‐LR) and one Marine Ecosystem Model (APECOSM), this study examines the ToE of epipelagic, migratory and mesopelagic fish biomass alongside their main environmental drivers for two contrasted climate‐change scenarios. Globally averaged biomass signals emerge over the historical period. Epipelagic biomass decline emerged earlier (1950) than mesozooplankton decline (2017) due to a stronger signal in the early 20th century, possibly related to trophic amplification induced by an early emerging surface warming (1915). Trophic amplification is delayed for mesopelagic biomass due to postponed warming in the mesopelagic zone, resulting in a later emergence (2017). ToE also displays strong size class dependence, with epipelagic medium sizes (20 cm) experiencing delayed emergence compared to the largest (1 m) and smallest (1 cm) categories. For the epipelagic and mesopelagic communities, the regional signal emergence lags behind the global average, with median ToE estimates of 2030 and 2034, respectively. This is due to stronger noise in regional time‐series than in global averages. The regional ToEs are also spatially heterogeneous, driven predominantly by the signal pattern akin to mesozooplankton. Additionally, our findings underscore that mitigation efforts (i.e., transitioning from SSP5‐8.5 to SSP1‐2.6 scenario) can potentially curtail emerging ocean surface signals by 30%.

Pelagic ecosystems are distributed throughout the world’s seas and oceans. They are characterised by strong vertical structuring, horizontal heterogeneity and temporal variability, which pose significant challenges for modelling them on a global scale. In this paper, we use the mechanistic high trophic level model APECOSM (Apex Predators ECOSystem Model) to assess how the physical and biogeochemical environment constrains the structure and trophic functioning of pelagic ecosystems worldwide.To this end, we configure the model to represent the three-dimensional and size-structured dynamics of six generic pelagic communities: small and medium epipelagics, tropical tunas, mesopelagic feeding tunas, small coastal pelagics, mesopelagic residents and mesopelagic migrants. We analyse their emergent three-dimensional spatial structuring on a global scale.We first show that the modelled horizontal and vertical distributions are consistent with the observed data. We then analyse the role of key environmental drivers, such as temperature, light, primary production, currents and oxygen on the response of the communities. Finally, we explore the trophic functioning of pelagic ecosystems, focusing on the emergent diets of communities and their variation with organism size.This study demonstrates the ability of a mechanistic ecosystem model to represent the multidimensional structural heterogeneity of marine ecosystems globally (encompassing three-dimensional distribution, size variations, and community composition) from a small set of universal principles and well-defined hypotheses. This approach helps to understand how the various processes at stake act and interact to shape the structure of global pelagic ecosystems, and eventually elucidate the heterogeneity of their trophic functioning.

The wedge clam, Donax hanleyanus, inhabits sandy beaches in the subtropical and temperate regions of the Atlantic coast of South America. Its distribution spans over 20 degrees of latitude from Brazil to Argentina, with the southernmost part of its range being influenced by the Rio de la Plata (RdlP) estuary, which limits the southward larval expansion. We used an individual-based model (IBM) to assess the larval connectivity patterns of the wedge clam during the period 2000-2012. The IBM combines a 3D hydrodynamic model with a biological sub-model that considers larval mortality due to low salinity (<7, and <9) and sea surface temperature range (high >30 degrees C or low <9 degrees C). The main larval connectivity patterns were observed near the release/recruitment areas, suggesting a high potential for self-recruitment. Based on the IBM and adult abundance data, we also identified the likely source and sink areas within this metapopulation. Source beach areas were Navegantes and Cassino in Brazil (from 26.3 degrees S to 34.34 degrees S), Arachania in Uruguay (34.56 degrees S), and Santa Teresita in Argentina (37.15 degrees S). A low probability of larval transport towards the poleward limit of the species' distribution was observed, supporting an irregular recruitment pattern typical of sink populations located at the edge of the distribution range of metapopulations. Larval mortality due to warm or cold waters did not affect connectivity patterns for this subtropical species. Southward larval transport across the RdlP estuary (from Uruguayan to Argentine beaches) only occurred for larvae released on early January 2011, concurrently with the strongest La Ni & ntilde;a year observed during the study period. In light of a changing climate, marked by potential increases in extreme La Ni & ntilde;a events and a poleward shift of atmospheric circulation patterns over the South Atlantic, we anticipate a strengthening of larval transport across the RdlP and a subsequent poleward expansion of the species' distribution range.

There is growing interest in developing and using ecosystem simulation models to advise fisheries management in the Southern Ocean. However, poor understanding of the impacts of uncertainty in ecosystem model parameters slows down progress towards operational ecosystem models. To address this issue, we explored uncertainty in the parameters estimated during the calibration of an OSMOSE ecosystem model for the Cooperation Sea ("OSMOSE-CooperationSea") and the impacts of this uncertainty. Our investigations pertained to four types of calibrated parameters: (1) Plank.access, the proportion of the biomass of background species groups available to focal species groups; (2) Bioflux, the parameter controlling the flux of migratory species group biomasses across the modelled domain boundaries; (3) Mlarval, the instantaneous larval mortality of the focal species groups; and (4) Mnatural, the additional natural mortality of the focal species groups. Results with the Morris method suggested that the community in the Cooperation Sea was most sensitive to changes in the Mlarval parameter of mesopelagic fishes. The biomasses of large-size, long-lived species such as toothfishes, Adelie penguin (Pygoscelis adeliae), seals, and whales were most sensitive to the parameters specific to these species groups. By contrast, the biomasses of small-sized, short-lived species such as mesopelagic fishes and krill species were most sensitive to changes in the parameters specific to the predators of these species groups. Monte Carlo simulations indicated that community dynamics were more sensitive to the Mlarvaland Mnatural parameters than to the Plank.access and Bioflux parameters. After gradually increasing the Mlarvalor Mnatural parameter, the biomasses of Adelie penguin, seals and whales decreased, while the biomasses of mesopelagic fishes and Antarctic krill increased. By providing a comprehensive analysis of uncertainty in the parameters estimated during the calibration process, the present study represents an important step towards an operational ecosystem model for supporting ecosystem-based management in the Cooperation Sea. The present study will serve as a valuable basis for similar ecosystem modelling efforts in the Southern Ocean.

The Fisheries and Marine Ecosystems Model Intercomparison Project (FishMIP) has dedicated a decade to unraveling the future impacts of climate change on marine animal biomass. FishMIP is now preparing a new simulation protocol to assess the combined effects of both climate and socio-economic changes on marine fisheries and ecosystems. This protocol will be based on the Ocean System Pathways (OSPs), a new set of socio-economic scenarios derived from the Shared Socioeconomic Pathways (SSPs) widely used by the Intergovernmental Panel on Climate Change (IPCC). The OSPs extend the SSPs to the economic, governance, management and socio-cultural contexts of large pelagic, small pelagic, benthic-demersal and emerging fisheries, as well as mariculture. Comprising qualitative storylines, quantitative model driver pathways and a "plug-in-model" framework, the OSPs will enable a heterogeneous suite of ecosystem models to simulate fisheries dynamics in a standardised way. This paper introduces this OSP framework and the simulation protocol that FishMIP will implement to explore future ocean social-ecological systems holistically, with a focus on critical issues such as climate justice, global food security, equitable fisheries, aquaculture development, fisheries management, and biodiversity conservation. Ultimately, the OSP framework is tailored to contribute to the synthesis work of the IPCC. It also aims to inform ongoing policy processes within the United Nations Food and Agriculture Organization (FAO). Finally, it seeks to support the synthesis work of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), with a particular focus on studying pathways relevant for the United Nations Convention on Biological Diversity.

The ocean provides both socioeconomic and climate benefits, but these benefits can potentially conflict when fishing disrupts the carbon sequestration potential of marine macrofauna and disturbs carbon-rich sediments. A lack of understanding about the spatial overlap between fishing activity and key carbon sequestration areas hinders efforts to incorporate climate considerations into fisheries management. Here, we identify potentially conflicting oceanic carbon areas with pelagic (COCA-P) and bottom-trawling fisheries (COCA-BT) separately. We show that COCA-P cover 11.2% of the ocean, with 60% occurring in the high seas, while COCA-BT cover 3.7% of the ocean, mainly in coastal areas where 56.4% of global trawling catches occur. Only 1% of COCAs overlap both fishery types, suggesting that targeted adjustments in fishing practices could mitigate conflicts. These findings provide critical insights for fisheries management, informing policies that balance food security, economic interests, and carbon sequestration in the ocean.

Pelagic ecosystems exhibit a strong regional heterogeneity, driven by physical and biogeochemical characteristics. Using the global 3D marine ecosystem model APECOSM, we simulate six high-trophic-level communities, capturing their size structure, spatial distribution, and trophic interactions up to 1,000 meters depth. We examine how different environments shape their contrasting organisation and interactions in three Pacific Ocean regions: the productive Humboldt Current System, the oligotrophic South Pacific Gyre, and the thermally stratified Pacific Warm Pool.Simulations reveal strong regional contrasts in ecosystem responses. In the Humboldt, high primary production supports important biomass of small coastal pelagic fish. Seasonal warming enables tuna to forage in these productive waters, while low-oxygen conditions restrict the vertical range and abundance of mesopelagic organisms and concentrate epipelagic organisms close to the surface. In the Warm Pool, apex predators remain abundant despite low primary production, thanks to efficient trophic transfer and biomass import from neighbouring regions. Seamounts concentrate mesopelagic organisms into shallow layers, making them accessible to epipelagic predators. In contrast, the South Pacific Gyre supports sparse, imported high-trophic-levels with limited trophic coupling and strong intra-community predation. We quantify regional differences in trophic transfer efficiency and network complexity, identifying thresholds below which high-trophic-levels collapse.These findings illustrate the emergent plasticity of pelagic ecosystems and the importance of bottom-up control of high-trophic-level biomass. They emphasise the importance of temperature, transport, light and oxygen in modulating horizontal and vertical distributions, controlling the co-occurrence of predators and prey, and influencing the formation of schools, ultimately impacting trophic interactions and community assemblages.

Although the role of marine macrofauna in the ocean carbon cycle is increasingly understood, the cumulative impacts of fisheries and climate change on this pathway remain overlooked. Here, using a marine ecosystem model, we estimate that each degree of warming reduces macrofauna biomass and carbon export by 4.2% and 2.46%, respectively. Under a high emission scenario (SSP 5-8.5), this translates to a 13.5% ± 6.6% decline in export by 2100, relative to the 1990s. Fishing further amplifies this reduction by up to 56.7% ± 16.3%, creating a sequestration deficit of 14.6 ± 10.3 GtC by 2100. On average, a 1% biomass loss from fishing results in a 0.8% decline in carbon export. However, sequestration durability (~600 years) remains unaffected. While measures restoring commercial macrofaunal biomass could yield carbon benefits comparable to mangrove restoration, multiple uncertainties limit their inclusion in the Nature-based Climate Solution portfolio, highlighting the need for further research.

The proliferation of Sargassum in the Tropical Atlantic has occurred almost every year since 2011, but a strong variability of biomass is observed among years. Essential knowledge about the drivers of Sargassum growth and decay is still lacking to explain this interannual variability. Benefiting from accurate basin scale Sargassum detections provided by remote sensing, and from physical and biogeochemical ocean simulations, we developed a Lagrangian drift-growth model to simulate Sargassum distribution over the period 2016–2020. The resulting trajectories and biomass time series of Sargassum aggregates were analyzed to highlight the main limiting factors of growth and decay. The nitrogen and phosphorous concentrations are found to be weakly restrictive compared to physical limiting factors, especially the temperature. In particular, the warm waters found off northern Brazil appear to be instrumental in triggering the end of seasonal growth of Sargassum. The timing of the seasonal warming of this region strongly impacts the quantities of Sargassum simulated each year. This suggests that this region should be monitored to anticipate the development of Sargassum and resulting strandings.

The recruitment of marine species in isolated oceanic island systems can be challenged by prevailing currents, as exemplified by the Tropical Southwestern Atlantic. In this region, the Fernando de Noronha ridge hosts several seamounts, the Rocas Atoll and the Fernando de Noronha Archipelago, which are home to great marine biodiversity. However, along the ridge, the central branch of the South Equatorial Current (cSEC), flowing westward, poses a challenge to the recruitment of organisms toward Fernando de Noronha. To unveil critical insights into the intricate processes shaping biodiversity in these insular ecosystems, we use a dispersal Lagrangian tool to explore the role of diel vertical migration (DVM) to depth strata influenced by the South Equatorial Undercurrent (SEUC), which flows eastward bellow the cSEC, in shaping species dispersal and metacommunity dynamics. Our results show that while not a direct journey, the DVM into SEUC-influenced strata increases the possibility that the seamounts and the Rocas Atoll act as stepping stones between the continental shelf and Fernando de Noronha. Propagules of organisms originating primarily from the continental shelf are transported to the western seamounts of the ridge. Upon reaching the western seamounts, organisms can find suitable habitats to recruit. The progeny of these communities that migrate to SEUC-influenced strata have the opportunity to reach suitable habitats at the Rocas Atoll and the Eastern seamounts, ultimately connecting to the Fernando de Noronha archipelago. These results provide scientific fundaments for the development of a functional network of marine protected areas in the Tropical Southwestern Atlantic.

Systematic analyses that examine uncertainty in models are essential for assessing their credibility. In this study, we implemented an uncertainty analysis that quantifies the effect of parameter uncertainty on a set of ecological indicators in outputs of the marine ecosystem OSMOSE model applied to the northern Peru Current ecosystem (NPCE OSMOSE). We worked under simple uncertainty assumptions corresponding to ranges of 10, 20, and 30% variability around the reference values of the parameters describing the dynamics of the species modelled in NPCE OSMOSE. The results based on nearly 1.5 million simulations help to identify the main sources of uncertainty that could be of use to focus future research and point to the most reliable indicators in the face of uncertainty. First, uncertainty in the parameters of some species, in particular a key zooplankton species and Humboldt squid, have far-reaching impacts on the modelled biomass of other key species. Second, a set of ecological indicators appear to be relatively insensitive to input uncertainty and may therefore be useful in supporting ecosystem-based management. Furthermore, our findings underline the need for better species representation in terms of data quality but also bottom-up and top-down processes in trophic models. We highlight the difficulties of studying uncertainty in complex models while presenting an approach that can serve as a template for addressing uncertainty analysis in other ecosystem models. Finally, although this approach focuses on parameter uncertainty, it could also serve as a guide to address structural, initial conditions and model forcing uncertainties.