Science-informed marine protected area networks outperform random and opportunistic designs

Science-informed marine protected area networks outperform random and opportunistic designs | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Science-informed marine protected area networks outperform random and opportunistic designs Clea Abello, Bruno Ernande, Fabien Moullec, Nicolas Barrier, Ignacio Pita-Vaca, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8653042/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 11 You are reading this latest preprint version Abstract 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. Biological sciences/Ecology Earth and environmental sciences/Ecology Earth and environmental sciences/Environmental sciences Earth and environmental sciences/Ocean sciences Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Although the effectiveness of large-scale networks of marine reserves has been demonstrated in several regions of the world 1 – 4 , the current global MPA estate remains largely ineffective and costly 5 . However, with sufficient dedicated financial and human resources for managing, monitoring and enforcing MPAs, the conservation benefits can be high 6 – 8 . These benefits are further enhanced by the implementation of fully or highly protected MPAs 9 , 10 , defined by the MPA Guide 11 as areas with no or minimal impact from destructive or extractive activities, particularly in overexploited areas where ecosystem structure and function are degraded 12 . The Mediterranean Sea is a particularly stark example of poor MPA implementation, with 95% of MPAs showing no regulatory difference from unprotected areas 13 and the majority not encompassing more biodiversity than expected by chance 14 . Moreover, apart from the fact that highly protected areas are almost non-existent (0.23% of the Mediterranean Sea area), the current network is highly fragmented, with the majority of protected areas concentrated in the western basin, resulting in poor network connectivity 15 – 18 . The rich biodiversity of the Mediterranean Sea 19 is threatened by multiple anthropogenic pressures, including the overexploitation of marine resources 20 . Although the percentage of overexploited stocks has fallen under 60% for the first time since decades, fishing pressure is still twice above the level considered sustainable 20 . This alarming situation is partly explained by the socio-political complexity of the region (surrounded by 22 countries), the lack of adherence to scientific advice, and the inadequacy of existing national management plans, which do not take sufficient account of the mixed nature of Mediterranean fisheries and the ecological interactions between fished stocks and with the environment 21 . Furthermore, the majority of Mediterranean fished stocks remain unassessed 22 and therefore cannot be managed conventionally, e.g. by setting quotas. However, studies have shown that the existence of scientific assessments and recommendations combined with well-developed management tools can positively correlate with improvements in stock status and harvest rates 23 – 26 . The implementation of management actions at the basin scale is greatly complicated by the lack of effective rebuilding plans to date 27 . The western Mediterranean multiannual plan (2019–2025) is the first Europe-wide restoration plan for demersal stocks, but it focuses on a small part of the Mediterranean Sea, and involves only three countries (France, Italy, Spain). In this context, the development of a network of large-scale connected marine reserves stands out as an option to explore for biodiversity conservation and ecosystem-based fisheries management in the Mediterranean Sea 28 . With conservation plans on the table to increase the extent of MPAs worldwide, such as the international thirty by thirty initiative (30 x 30), aiming at protecting 30% of the ocean by 2030 29,30 , it is crucial that we seize this opportunity to implement effective MPAs, rather than more ‘paper parks’ 31 . Current approaches in conservation planning increasingly emphasize the importance of maximizing ecological connectivity between reserves 32 , 33 . In the MPA modeling field, ecological connectivity is mostly addressed through larval dispersal or gene flow (see 34 , 35 for a review), using biophysical models 14 , 15 , 30 , graph theory 15 , 18 , and, less frequently, genetic markers 37 , with the majority of studies focusing on a single emblematic species (e.g., 11, 13, 26–28). Another widely used approach in conservation planning involves identifying biodiversity hotspots through species distribution models (see 29 for a review). Systematic conservation planning tools, such as Marxan 42 or Zonation 43 , are also commonly used to optimize the spatial design of reserves, aiming to achieve biodiversity targets while minimizing cost for human activities or maximizing biodiversity benefits under certain budget or area constraints 44 . While all these methods can be applied at fine spatial resolutions and provide science-based planning solutions, they commonly overlook demographic and ecosystem responses to changes in anthropogenic pressures such as fishing, especially those caused by the implementation of MPA networks. This is even more essential in a context of mixed fisheries, where biological (species predator-prey relationships) and technical (gears targeting multiple stocks and in competition for the same stock) interactions can result in non-linear and unexpected responses of species abundance and fisheries productivity (e.g., 45 ). End-to-end ecosystem models, which explicitly represent the dynamics of ecosystem physical and biogeochemical properties, trophic networks, species habitats and human activities, can overcome these limitations and be used to explore the strategic implementation of MPAs 46 – 48 . Yet, they are still rarely used for management purposes due to their complexity and inherent uncertainties that make them unfamiliar to decision makers 49 . Here, the conservation and fisheries outcomes of expanding the network of fully protected areas (FPAs) was investigated at the scale of the whole Mediterranean Sea. We hypothesize that carefully designed and strategically placed (i.e., science-informed) MPA networks will perform better in protecting biodiversity and rebuilding marine resources than the expansion of the current MPA network and/or randomly placed MPAs 14 . To address this, we propose a novel framework using the spatially-explicit multi-species individual-based ecosystem model OSMOSE (see Supplementary Note 1 ) coupled to a high-resolution regional climate model and a regional biogeochemical model. We fine-scale modeled the full life cycle of nearly 100 fish and macro-invertebrate species, representing ca. 85% of the total reported catches in the Mediterranean Sea 20 (see Supplementary Table 6. 1 ). We seek to inform policy recommendations for large-scale conservation planning in the Mediterranean Sea by comparing six types of FPA networks: one resulting from the expansion of current MPAs with existing fishing regulations (i.e., National MPAs, Marine Natura 2000 areas and Fisheries Restricted Areas) (Scenario “S1-Current”), two in which MPAs are randomly distributed, either over the entire Mediterranean Sea (Scenario “S2-Random basin”) or in existing or theoretical Economic Exclusive Zones (EEZ) in proportion to the size of each country’s maritime area (Scenario “S3-Random EEZ”), and three networks proposed in the scientific literature based on a combination of ecological, economic, social, political and cultural criteria, we hereafter name ‘science-informed’ networks 50 , 51 (Scenarios S4 to S6) (see Supplementary Note 4 ). Among the three science-informed FPA network scenarios, the first (Scenario “S4-Micheli”) results from the identification of areas with the strongest consensus on conservation priorities identified across more than five proposed whole-basin conservation initiatives. These initiatives are mainly biodiversity-driven, considering various criteria such as habitat types and species distributions, but also incorporate cultural and geological factors, as well as human threat intensity 50 . In contrast, the second and third scenarios 51 are designed to protect 10% of the distribution area of 77 threatened marine species while minimizing the impact on fisheries using the systematic planning tool Marxan 42 . The two latter differ in the origin of the data used to estimate commercial fishing opportunity cost (i.e., lost revenues because of protection): either from the General Fisheries Commission for the Mediterranean (Scenario “S5-Mazor GFCM”) or from the Sea Around Us project (Scenario “S6-Mazor SAUP”). For each FPA scenario, we compared the impacts of different levels of coverage of the Mediterranean Sea on fish and fisheries by increasing all networks from 1 to 30% coverage in steps of 1%. Additionally, we considered three possible strategies of fishing effort redistribution after FPA establishment: (1) fishing-the-line redistribution, i.e., reallocation of the displaced effort from each FPA at its boundary, (2) uniform redistribution by GFCM Geographical Sub-Area (GSA), and (3) redistribution by GSA proportionally to the spatial distribution of fishing effort prior to FPA establishment (see Supplementary Note 3 for further details). Results and Discussion Science-informed MPA networks better balance biomass gains and catch losses As expected, increasing FPA coverage resulted in an increase in total biomass of fish and macro-invertebrate species, regardless of network configuration (Fig. 1 A). However, we found that for a given level of coverage, the simulation of science-informed MPA networks (S4-S6) led to higher biomass gains than random networks (S2-S3) and even than a network resulting from the expansion of current MPAs, Natura 2000 sites and Fisheries Restricted Areas in the Mediterranean assuming they become no-take areas (S1). For instance, at 10% coverage, the biomass increase ranged from 1 to 2.9% for the science-informed FPA networks (S4-S6), from 0.6 to 1.8% for the random networks (S2-S3), and from 0.09 to 2.1% for the current network expansion (S1), depending on the fishing effort redistribution strategy. All scenarios resulted in a reduction in catch as FPA coverage increased, but the decrease in catch was proportionally smaller than the increase in MPA coverage. Besides the expansion of the current MPA network (S1) which has the most negative impacts on catch, catch losses ranged between ca. -2% and − 6% at 10% coverage and between ca. -8% and − 19% at 30% coverage for the other MPA scenarios (Fig. 1 B). Furthermore, the percent catch loss always exceeded the percent biomass gains in all MPA scenarios. Expanding current MPAs (S1) appeared to be the least optimal solution, with biomass gains similar to random networks (S2-S3), but with two to three times significant greater negative impact on fisheries catch (between − 10.4 and − 8.7% decrease for S1 vs. -1.8 to -3.5% for S2-S3 random networks at 10% coverage; Wilcoxon test p-value < 0.05). Among the science-informed networks (S4-S6), those that use fisheries data and minimize fishing opportunity cost when selecting areas to protect (S5-S6) achieved the best balance between biomass gains and catch losses: for example, 10% MPA coverage resulted in an increase in biomass ranging from 1 to 2.6% for S5-S6 vs. 1.5 and 2.9% for S4 and a decrease in catch between − 3.3 and − 1.4% for S5-S6 vs. -6 to -4.5% for S4. The changes in biomass and catch in the “S3-Random EEZ” scenario were very similar to the random placement of reserves throughout the Mediterranean Sea (“S2-Random basin”), suggesting that an even distribution of reserve coverage across Mediterranean countries may not be sufficient to achieve an effective MPA network and that concerted action among all Mediterranean countries is necessary. Benefits of MPA networks extend beyond their boundaries The conservation benefits of marine reserve networks are not confined within their boundaries, as biomass also increased outside MPAs (Fig. 2 ). For the most effective FPA network tested (“S5-Mazor GFCM”) at 10% coverage, the increase of biomass outside MPAs (distance > 20 km from the edge of the MPA) ranged between 0.8% and 1.9% according to the fishing redistribution strategy considered. This is linked to a spillover effect 52 (i.e., the export of biomass from the protected area to neighboring areas through a density-dependent process), as evidenced by the gradual decrease in biomass from inside to outside MPAs (Fig. 2 , left panel). There is a notable opposite response of fish biomass for the “fishing-the-line” effort redistribution strategy at reserve boundaries (Fig. 2 , left panel) where catches increased dramatically (Fig. 2 , right panel) due to high fishing effort concentration. We also found evidence of an edge effect 53 (i.e., the undermining of the effective size of protected areas, caused by human-related stressors in their surroundings) as biomass gains at the inner edge (i.e., < 20 km from the edge of FPAs) were less than half of those in the core, especially in the case of the “fishing-the-line” strategy. Such decline in the “effective size” of protected areas was highlighted in a recent meta-analysis combining data from 27 no-take areas around the world. The study found significant reduction in population size within 1 km inside MPAs 53 and highlighted fishing as a major driver, reinforcing the importance of establishing buffer zones around no-take reserves, where fishing activities are sustainably managed 8 . However, other human-induced environmental stressors, such as pollution and runoff, can also hinder MPA success 54 . Indirect effects of MPAs through top-down trophic cascades result in winners and losers between functional groups Examining the changes in biomass and catch by major species group revealed more nuanced results (Fig. 3 A, see Supplementary Figs. 6 . 2–6. 5 for other scenarios). Focusing on the MPA scenario with the greatest potential to increase conservation gains while minimizing catch losses (“S5-Mazor GFCM”), we observed that the decrease in total catch was mostly driven by a decrease in the biomass and catch of small pelagic species, followed by crustaceans and cephalopods. At 10% coverage, with fishing effort redistributed proportionally to pre-MPA levels (by GSA), these groups experienced biomass declines of -3.6%, -3.3% and − 14.7%, respectively, with corresponding catch losses of -11.7%, -15.4% and − 19.6%. In contrast, the biomass and catch of high trophic level species, i.e. large and medium-sized demersal and pelagic species, increased substantially with MPA coverage, with biomass gains of + 12.8% and + 9.4%, and catch increases of + 12.1% and + 8%, respectively, at 10% coverage. This was observed for nearly all of the MPA scenarios regardless of the fishing redistribution strategy considered, except for the current (S1) and random-basin (S2) scenarios under a uniform fishing redistribution strategy for MPA coverage below 10% (see Supplementary Figs. 6. 2–6. 5 ). Among small pelagic species, the European anchovy ( Engraulis encrasicolus ) and the European pilchard ( Sardina pilchardus ) experienced the steepest declines under the “S5-Mazor GFCM scenario”. At 10% coverage, their biomass dropped by -6.6% and − 3.7%, respectively, and at 30% coverage, these losses intensified to -22.1% and − 10.9% (see Supplementary Fig. 6. 1 ). The loss of biomass of low trophic level species, while the biomass of high trophic level species increased, is due to a higher predation pressure exerted by the latter on the former due to increased MPA coverage. Specifically, predation pressure on main low-trophic level groups (small pelagic fish, cephalopods and crustaceans) increased by 3.45%, 3.32% and 1.88%, respectively, under a 10% FPA network (see Supplementary Fig. 6. 2 ). This rise in predation was primarily driven by increased ingestion by large demersal predators, which became more abundant with MPAs. Such top-down trophic cascade effects due to increased predation pressure within marine reserves are common 55 , but not always the case 56 . Here, the decline in the biomass of small pelagic species within reserves started at different coverage levels depending on the MPA scenario (Fig. 3 B). The decline started at a much lower coverage level for two of the three science-informed networks (MPA coverage between 2 and 3% for S5 and S6) than for the current network and the other science-informed network (MPA coverage between 10% and 12% for S1 and S4) or random networks (MPA coverage between 5 and 20% for S2-S3). This is consistent with the fact that scenarios S5 and S6 were primarily designed to protect threatened and vulnerable species, which are mostly high trophic level species that are targeted by fisheries and exert predation pressure on low trophic levels. Although the implementation of reserve networks would most likely have a negative impact on low trophic level species, this is not necessarily alarming from both a conservation and fisheries economic perspective. From a conservation perspective, the Mediterranean Sea has experienced a "fishing-down the food web" effect 57 over the last 70 years with a decrease in the abundance and catch of overexploited high trophic level fish species and an increase in macroinvertebrates (ca. 23%) 58 . Thus, our results suggest that an increase in FPA coverage could partially reverse this trend, restoring the biomass pyramid to more closely resemble that of pristine areas 59 . From a fisheries economic perspective, the negative impact on fisheries due to the loss of fishing grounds could be economically offset despite the reduction in yields, as high trophic level species, such as European hake or bluefin tuna, are of high commercial value. Science-informed MPA networks are effective in restoring large fish even if biodiversity benefits remain modest The recovery of large-bodied fish with increasing MPA coverage was evidenced by the Large Fish Indicator (LFI), i.e., the proportion of the biomass of large fish (≥ 20 cm in length) 60 (Fig. 4 A). We estimated that if a 10% FPA network were implemented, the LFI within reserves would shift from 29% to between 30 and 38%, depending on the network and fishing redistribution strategy (Fig. 4 A, left panel). The benefits of FPAs for the recovery of large fish were projected to extend beyond their boundaries, as the proportion of large fish in unprotected areas also increased for a coverage ≥ 5% (Fig. 4 A, right panel). Science-informed FPA networks (S4-S6) were also more effective than random networks (S2-S3) for this indicator, but equivalent to expanding the current MPA network (S1). To assess the representativeness of biodiversity within the six FPA networks, we calculated the Hill-Shannon alpha-taxonomic diversity index 61 for the 95 species modelled (Fig. 4 B). We found that although across all scenarios, community structure within MPAs was extremely uneven and dominated by a few very abundant species, the current MPA network (S1) and consensual priority conservation areas 50 (S4) encompassed more biodiversity than areas protected randomly (S2-S3) (Fig. 4 B, left panel). However, this was not the case for the other two science-informed networks (S5-S6) (Fig. 4 B, left panel). Furthermore, in scenarios S1 and S4, biodiversity within FPAs was, on average, 1.5 times higher compared to unprotected areas, consistent with the findings of a recent meta-analysis 62 . In contrast, the difference between inside and outside FPAs was less pronounced in the other scenarios, with some random scenarios (S2-S3) even showing higher biodiversity outside FPAs (Fig. 4 B). These differences can be attributed to differences in network configuration, as diversity outside FPAs was positively correlated with the average distance of FPAs from the coastline (Fig. 5 ). Consequently, more coastal networks, such as scenarios S1 and S4, captured more biodiversity than more offshore networks (see section below). While our modelling framework enables comparison of diversity across MPA networks under a common baseline —specifically, the distributions of 95 fish and macroinvertebrate species— it cannot assess the effectiveness of MPAs aimed at protecting biodiversity components absent from our model. Therefore, we advise caution when interpreting this index, as it does not encompass all facets of the rich biodiversity of the Mediterranean Sea. Even if scenarios S5 and S6 appeared to capture less fish diversity than S1 and S4 (Fig. 4 B, left), more than half of the species they were designed to protect (primarily, marine mammals, turtles, seabirds and elasmobranchs) were either absent or underrepresented in our model. We further explored the potential role of FPAs in ecosystem recovery by calculating the Hill-Shannon diversity index before and after MPA implementation. Our findings showed that nearly all FPA networks experienced increases in alpha diversity following the ban on fishing activities (see Supplementary Fig. 7. 1 ), in line with previous studies 62 – 64 . However, the positive effect on biodiversity remained modest and was particularly noticeable when MPA coverage exceeded 10%. Among the different networks, FPAs in S1 and S4 proved most effective in enhancing local taxonomic diversity, even though we observed that beyond a certain threshold of MPA coverage, conservation gains plateaued, suggesting diminishing returns. These patterns align with previous research indicating that biodiversity gains within marine reserves may be limited by ecological feedbacks, such as increased predation pressure within reserves, which can reduce the diversity of lower trophic level species 65 , 66 . Although a global study assessing the ecological and economic costs and benefits of expanding MPAs put forward that protecting pristine areas overall yielded greater net benefits than targeting low biodiversity areas or heavily impacted ecosystems 67 , the high level of human activities in the Mediterranean Sea calls for a careful evaluation of socio-ecological trade-offs at a more local scale. Given the Mediterranean’s complex socio-political landscape, protecting high biodiversity areas might not always be feasible nor the first and best solution. Strategical prioritization that engages local stakeholders and decision makers 44 , and a stronger focus on actions rather than areas, as emphasized by 44,68 , is needed to inform decisions on most adapted conservation actions for each area in order to maximize conservation outcomes. Inclusive governance is particularly critical in areas where communities heavily depend on the sea and its resources 69 . Empirical evidence 8 , 70 has demonstrated that high levels of local participation, capacity building and transparent decision-making are significantly more likely to achieve ecological success and reduce illegal fishing activities. Moreover, perceived legitimacy and equity are essential predictors of compliance in marine resource management 71 . MPA network design is crucial to achieve effective outcomes The size, shape, spacing and location of MPAs have a direct influence on the ecological effectiveness of MPA networks 72 – 75 . To better understand the observed differences between our six FPA network scenarios, we computed several shape and aggregation metrics and used redundancy analysis (RDA) to explore their relationship with absolute changes in biomass and catch, the Large Fish Indicator (LFI) inside and outside protected areas and the Hill-Shannon diversity indicator inside and outside protected areas following MPA establishment using 10% coverage as an example (Fig. 5 ). We found that catch declined less when reserves were further from shore (Fig. 5 , Distance to coast red arrow), which was the case for nearly all of the random networks (Fig. 5 , S2-S3, gray and black dots). This is consistent with the spatial distribution of fishing effort in the Mediterranean Sea (see Supplementary Fig. 2. 4 ), which is mainly concentrated close to the coast. Catch also decreased less (even if less strongly) as the average compactness of MPAs in the network decreased (Fig. 5 , Mean shape index red arrow). Compactness is defined as the ratio between an MPA’s actual perimeter and the minimum possible perimeter it would have if it were perfectly compact. Biomass and the proportion of large fish both inside and outside FPAs increased more when the networks were more aggregated (Fig. 5 , PLADJ red arrow) and MPA patches larger (i.e., fewer patches; Fig. 5 , Number of patches red arrow). As the current MPA network consists of smaller reserves closer to shore (see Supplementary Fig. 4. 1 ), it is not surprising that this network (Fig. 5 , S1, yellow dot) appeared suboptimal compared to the others in achieving the best balance between biomass gains and catch losses, despite encompassing areas with above-average biodiversity. The latter was related to the fact that taxonomic α-diversity inside protected areas was higher when reserves were more coastal (Fig. 5 , Diversity inside MPA blue arrow versus Distance to coast red arrow and Supplementary Fig. 7. 2 ). This aligns with the findings of 76 , who reported fish biodiversity to be primarily concentrated in coastal regions, including Spain, France and Italy, the north-western coast of Africa, the Ionian and Aegean Sea, and the Adriatic Sea. The RDA also highlighted the structural differences between the three science-informed networks, the random networks and the current one (see Supplementary Fig. 8. 3 for a RDA triplot using scaling 1 under which the distances between scenario projections approximate their Euclidian distances). Our analysis indicates that compact networks of large reserves should be preferred over patchy ones to maximize conservation outcomes. To minimize the impact on fisheries, reserves should be placed further offshore. One of the downsides of MPA establishment in areas where fishing takes place is the emerging trade-off between fisheries management goals and conservation objectives that must be resolved 77 . In this study, for example, all tested FPA networks, although yielding benefits for conservation, projected a decrease in catch. However, our spatial analysis suggests that a win-win strategy, balancing both conservation and fisheries benefits is possible. Notably, conservation indicators such as biomass and the Large Fish Indicator showed relative independence from fishing metrics, as evidenced by the 90 degree angle between biomass/LFI and catch indicators (Fig. 5 , blue arrows). This is supported by real-world examples where MPAs have delivered tangible benefits to fisheries 78 . Moreover, the conservation and fisheries indicators responded to spatial metrics that were relatively independent as reflected by the nearly 90 degree angle between the number of patches or PLADJ and the distance to the coast (Fig. 5 ). Hence our analysis shows that a more optimal network configuration in terms of both conservation and fisheries benefits could be achieved by establishing a few large offshore reserves that are relatively aggregated at the network level (Fig. 5 , bottom right quadrant). This would be done without sacrificing the protection of biodiversity hotspots, as although diversity would decrease inside MPAs, it would increase outside (Fig. 5 , Diversity blue arrows). Indeed, some offshore areas exhibit high α-diversity (see Supplementary Fig. 7. 2 for the distribution of α-diversity across the Mediterranean Sea). Such recommendations are conditional on multinational cooperation, which is not an easy task in the Mediterranean Sea, given the region’s geopolitical complexity. In certain areas, like in the Eastern Mediterranean Sea, ongoing disputes over maritime boundaries 79 exacerbated by resource competition, are hampering any conservation action. Yet, existing ties between countries, supported by collaborative legislation frameworks, such as the European Union or the Arab Maghreb Union, have facilitated coordinated conservation initiatives, such as the Natura 2000 marine network or the Pelagos Sanctuary. Research has shown that coordinated actions could be more cost-efficient for most Mediterranean countries, reducing costs by about two-thirds compared to fully independent national efforts 80 . Encouragingly, many Mediterranean countries are actively declaring or negotiating their EEZs, offering hope that clearer jurisdictional delineations will further catalyze coordinated conservation actions in the region in the near future 81 . Among the 24 different networks we tested (incl. the expansion of the existing network (S1), 10 configurations for each random scenario (S2-S3) and 3 science-informed networks (S4-S6)), the three science-informed networks (S4-S6, Fig. 5 ) came closest to this optimal configuration. Nevertheless, our results suggest that there is still significant room for improvement in MPA placement and overall network design, as most of the ensemble of optimal configurations remain unexplored (Fig. 5 , bottom right quadrant). In addition, national and local considerations, as well as refined socio-economic criteria have not been prioritized in the scenarios found in the literature 50 , 51 , nor in their simulated expansion, despite the fact that socio-economic factors appear to be critical to identify suitable areas for MPA establishment 82 . This leads to some unrealistic MPA placement and size in all scenarios. Fishing redistribution strategies can alter the effectiveness of MPAs In addition to the spatial design of the network, our study highlights the importance of the fishing effort redistribution strategy modeled. We found that the conservation and fishery benefits were sensitive to the strategy deployed. We modeled three simple fishing redistribution strategies and found that the higher effectiveness of one network over the other in terms of biomass and catch could be reduced or reversed depending on the fishing strategy considered. As an example, total biomass gains were more important in the current MPA network (S1) than in random networks (S2-S3) or in the “Mazor SAUP” network (S6) up to a 20% coverage with a fishing-the-line redistribution strategy, whereas this was not the case with a uniform redistribution strategy (Fig. 1 ). Redistributing fishing effort proportionally to pre-closure effort density resulted in higher conservation gains in terms of total biomass than either the fishing-the-line or uniform redistribution strategy (Fig. 1 and Fig. 2 ). In contrast, the uniform redistribution strategy had the least impact on fishing activities in terms of catch loss, significantly lower than both the fishing-the-line and the proportional redistribution strategy (Fig. 1 and see Supplementary Fig. 5. 3 for map of catch change outside protected areas at 10% coverage). Although our archetypal strategies of fishing redistribution do not capture the complexity of fishing behavior following a fishery closure, the proportional redistribution strategy is commonly observed and could be considered as the default option 83 . Several fisheries around the world exhibit such site-fidelity behavior, resulting in increased fishing effort in areas of already high fishing pressure following spatial closures 84 – 86 . Concentration of fishing effort around reserves, commonly referred to as “fishing-the-line”, is another frequently observed strategy 87 , 88 . This strategy adopted by fishers assumes that the net export of biomass from the reserve should increase catch rates in adjacent unprotected waters. However, it has been shown to be highly fleet and resource-dependent 89 , 90 . Furthermore, boats tend to concentrate in the first few kilometers beyond the reserve boundary 88 , 91 . As our model’s spatial resolution (20 x 20 km) does not allow us to account for the exponential decrease in effort from the reserve boundary, we may underestimate fishing competition at the edge of the reserve when simulating this strategy (Fig. 2 and Supplementary Fig. 5. 3 ). Our results show that, although simplifying assumptions are sometimes needed when using already complex modeling tools, there are still opportunities for cost-effective improvements. For example, it may be more realistic to model a proportional fishing redistribution strategy rather than a uniform redistribution, or even the collapse of part of the fleets, as has been done in recent modeling studies 92 , 93 . A maximum redistribution distance within a sub-area of the system should also be considered in models to avoid unrealistic scenarios where fishers can travel implausible distances in a single day. Indeed, distance travelled has been identified as a key factor influencing the selection of new fishing grounds by artisanal fisheries 94 . This criterion is especially relevant in regions such as the Mediterranean Sea, where artisanal fisheries make up over 80% of the fishing fleet 20 . Within this context, and in the interest of transparency, we highlight below key methodological considerations that both frame the interpretation of our results and point toward promising avenues for future model development. First, as mentioned above, the spatial resolution of our model (20 x 20 km) prevents us from representing scenarios with smaller MPAs, as we are limited to an MPA minimum size of 400 km², while the mean size of current Mediterranean MPAs is 143 km². While MPAs with enhanced conservation benefits are often larger than 100 km² 9 , smaller MPAs have also been shown to yield positive outcomes 8 , in particular when protection aligns with species home ranges 94 . However, since the objective is to simulate an increase in MPA coverage in the Mediterranean, the lack of fine spatial resolution may not prevent the model from capturing the general patterns across the different expansion scenarios, Adopting a finer resolution (e.g., 10 x 10 km), as recommended by the European Union’s Directive 2007/2/EC for spatial planning, would enhance direct tactical support to management. Here, our contribution is rather strategic in nature: the different scenarios help illuminate key ecological and fishing processes involved and inform the optimal design–both in terms of location and configuration–of MPAs to support conservation and fisheries objectives at the basin scale. Second, the model assumes random movement of fish —represented as super-individuals or “schools” of biologically identical individuals of the same species, born at the same time— across each species' distribution area. While this approach allows to account for climate niche of the species, it does not account for fine behavioral processes such as territoriality or ontogenetic migrations, nor for functional differences between species linked to mobility (whether active for adults or passive for larval pelagic stages) or habitat preferences. Previous studies have demonstrated that adult movement is a critical determinant of the effectiveness of reserve networks 96 , 97 . In our model, schools have an equal probability of remaining in their current cell or moving to one of the eight adjacent cells at every 15-day time step. However, 24 out of the 95 species modeled are primarily sedentary, and 60 have relatively low mobility (see Supplementary Table 6. 1 for list of species). Consequently, population persistence within reserve networks, particularly for sedentary species, is most likely underestimated in our study. Constraining movement within home ranges is complex and data-intensive; a simpler approach could involve differentiating movement capacity by species or groups of species. Another limitation of our modeling approach lies in the simplified representation of larval dispersal, which does not incorporate physical oceanographic processes. Instead, newborn larval schools are randomly distributed across the species’ distribution map, potentially leading to an overestimation of connectivity both among MPAs and between MPAs and unprotected areas within the distribution map 98 . This random distribution indeed assumes that larvae have an equal probability of reaching any cell within the map, whereas physical constraints would likely limit the range of accessible cells. Incorporating these physical processes would also require identifying species’ spawning grounds, as these are unlikely to be uniformly distributed across the adults’ habitat. Such refinements could provide a more realistic assessment of connectivity between reserves, as well as their contribution to fisheries through spill-over effects, thereby improving the evaluation of reserve network efficiency 97 . However, gathering this information for 95 species would be a daunting task. Third, our model considers the distribution of each species as determined by its climate niche but does not account for ontogenetic habitat shifts. To better reflect the importance of protecting certain areas over others, it would be beneficial to incorporate spatial distribution maps by stage or size class to identify major spawning and nursery grounds. These aspects make us confident in affirming that our results are most likely conservative. As data on the spatial distribution of fishing effort differentiated by fleet were not available at the Mediterranean scale, we did not account for fleet heterogeneity, which may introduce some bias in the estimated impact of MPAs on fisheries 99 . Distinguishing between large-scale and small-scale fisheries in future applications would enable the inclusion of more nuanced management options, such as multi-zone MPAs. In addition, fisheries impacts were assessed using catch as a primary indicator, but extending the framework to incorporate complementary socio-economic indicators, such as fishing income and costs, would provide a more comprehensive assessment of management outcomes. Despite these simplifying assumptions, our modeling framework is, to our knowledge, the first ecosystem modeling approach that combines such ecological detail with full basin-scale coverage for this region. Our findings consistently demonstrate the higher positive impacts of science-informed FPA networks over both random and existing expansions in the Mediterranean Sea. While our results do not directly demonstrate this outcome, the spatial analysis suggests that a well-designed FPA network could potentially support the recovery and rebuilding of Mediterranean exploited species, without major tradeoffs for fisheries. This could be feasible before 2030, in line with the CBD’s post-2020 strategic plan, if concertation and political support follow. Although our analysis was conducted for the Mediterranean Sea, some of our findings are generic enough to apply to many other regions, and science-informed networks are needed where exploitation levels are high, stocks are overexploited and overall ecosystem health is degraded 12 , 28 . Yet, spatial protection alone may not be sufficient in light of the current climate and biodiversity crises fish and fisheries are facing 100 . Therefore, alternative management measures, such as annual catch limits and shares, effort regulation or gear modifications, are also necessary 101 , 102 . These measures have the potential to significantly increase fisheries catches and profits 103 . This is especially evident in the Mediterranean Sea 104 , which suffers from excessively high fishing effort and inappropriate selectivity patterns, resulting in average fishing mortality rates for all species that are 2.5 times higher than the level at which maximum sustainable yield can be achieved 20 . Additionally, most commercial species have a low size at first capture 105 . Reducing fishing effort and increasing length at first capture by at least 30% could enhance total biomass and catch levels of high trophic level species, particularly demersal, large pelagic and benthic species 104 . Future large-scale studies should focus on assessing the combined effects of multiple management strategies. The primary challenge is to combine the results of basin-scale assessments such as here with more detailed and local impact studies carried out in close collaboration with multiple stakeholders. This will help promote effective and integrated strategies for conservation and fisheries management. Methods The OSMOSE end-to-end model of the Mediterranean Sea The ecological and fishery effects of the six FPA networks were simulated using a previously developed OSMOSE model (Object-oriented Simulator of Marine ecOSystEms) at whole-basin Mediterranean scale for the period 2006–2013 106 , here slightly modified to improve the spatial representation of fishing activity. OSMOSE is a spatially explicit age and size-structured individual-based ecosystem model representing the dynamics and main life cycle processes of key high trophic level species, including growth, reproduction, movement and different mortality sources (predation, starvation, fishing and natural mortalities). Each ‘super-individual’ or school of fish is independently modeled. The model makes the assumption that predation processes are opportunistic, driven by spatial overlap and size-based constraints between ‘super-individuals’, which allow the emergence of complex food-web structures. The detailed OSMOSE model documentation can be found at https://osmose-model.org/ and has been synthesized in the Supplementary Note 1 . The Mediterranean configuration of the OSMOSE model was extended through one-way coupling with the regional circulation model CNRM-RCSM4 107 and the biogeochemical model Eco3M-S 108 , simulating the dynamics of seven major planktonic functional groups in the Mediterranean for the same period 106 . The resulting end-to-end model, OSMOSE-MED, simulates the life cycles and trophic interactions of 95 marine species (82 fish, 5 cephalopods and 8 crustaceans) of ecological and commercial importance in the Mediterranean, accounting for 85% of total declared catches according to the Sea Around Us Project reconstructed catch data for the period 2006–2013, at relatively high spatiotemporal resolution for a basin-scale model (regular grid of 6229 cells of 20 x 20 km; 15-day time step). The specificities of the Mediterranean configuration OSMOSE-MED, its parameterization and calibration are fully described in 106 and in the Supplementary Note 1 . Model code and configuration are publicly available (see Code availability section). Spatial representation of fishing activity To improve the spatial representation of fishing activity, we estimated fishing effort at basin-scale using Synthetic Aperture Radar (SAR) satellite imagery data. SAR satellites provide high-resolution images that can detect vessels greater than 10 m in length, providing a solution for real-time vessel location 109 . The use of SAR satellite imagery data is not critically affected by weather conditions (e.g., cloud cover) and day-night cycles 110 . Although it was necessary to make several assumptions to distinguish fishing from other activities 111 , this technology allowed us to obtain a more accurate representation of fishing effort in the Mediterranean compared to using Automatic Identification System (AIS) data. Indeed, as in the Mediterranean, AIS is only mandatory in EU waters for vessels over 15 m in length and for ships of over 300 gross tonnage engaged on international voyages (International Maritime Organization), which are unlikely to be fishing vessels, significant gaps in AIS coverage are particularly noticeable in the southern and eastern parts of the basin. We obtained 14,278 Ground Range Detected High-Resolution images (i.e., images with a range detection of 10m per pixel) from Sentinel-1A and Sentinel-1B, the SAR satellite constellation of the European Union’s Copernicus program for Earth Observation, operated by the European Space Agency (ESA) from the Alaska Satellite Facility platform 112 . These images cover the entire Mediterranean basin from January 1 to December 31, 2019. The Search for Unidentified Maritime Objects (SUMO) algorithm, developed by the European Joint Research Center 113 was used to detect vessels. SUMO is a pixel-based Constant False Alarm Rate detector that uses several detection thresholds to differentiate vessels from sea clutter 114 . Data was post-processed to minimize the number of false positives and provide a more realistic estimation of fishing effort (see Supplementary Note 2 for details). MPA and fishing redistribution scenarios We evaluated the ecological and fisheries impacts of alternative FPA scenarios by systematically increasing protected area coverage from 1 to 30% in 1% increments. Each network was initialized with 62 marine protected area seed locations, corresponding to a single OSMOSE-MED grid cell (~ 1% of the Mediterranean simulation grid, comprising 6,229 cells). Seed placement differed by scenario: in S1 (“Current”), seeds were positioned at the barycenter of current MPAs; in S2 and S3, they were distributed randomly across the basin or within EEZs, respectively; and in science-informed scenarios S4-S6 (“Micheli”, “Mazor GFCM” and “Mazor SAUP”), seeds were located in literature-identified priority areas. All scenarios, except S4, which followed a prioritization scheme, were then expanded using a randomized rook-neighborhood approach, in which new cells were added to one of the four orthogonally adjacent grid cells. Expansion was either fully random (in the case of S2 and S3) or constrained by the existing or proposed conservation plans (S1, S5, S6). Further details on the design of the six FPA scenarios and the three fishing effort reallocation strategies are provided in the Supplementary Notes 4 and 3 , respectively. Each simulation ran for 200 years, with FPAs being implemented in year 110 to ensure sufficient spin-up time for the system to reach equilibrium. For all the analyses, we averaged the last 60 years (from year 140 to 200) to leave sufficient time after MPA establishment for the model to reach a new steady state. To account for the stochasticity of the model, each simulation was replicated 30 times, which offered a good compromise between simulation time and model stochasticity (see Supplementary Fig. 5. 1). A total of 2,160 sets of 30 replicated simulations for the 24 FPA network configurations, representing the six FPA scenarios, were ran. Each FPA scenario was modeled considering three fishing redistribution strategies for each coverage level ranging from 1 to 30% in 1% increments. We also modeled a reference state simulation without MPAs. To assess the impact of FPAs and compare the six network scenarios, we calculated the relative or absolute change before and after MPAs for the different indicators computed (i.e., total biomass and catch and by groups of species, the large fish indicator and taxonomic α-diversity). We only evaluated the magnitude of difference between scenarios as recommended by 115 . Spatial and aggregation metrics of MPA network scenarios The different spatial and aggregation metrics used for the redundancy analysis (RDA) were calculated using R version 4.4.0 and the landscapemetrics R package version 2.1.4 116 , except for mean MPA network distance to the coast. In total, fifty-six metrics were computed (see the landscapemetrics package documentation for details). To assess the relationship between the response variables (i.e., total biomass and catch and large fish indicator and taxonomic α-diversity inside and outside MPAs) and a suite of explanatory spatial and aggregation metrics, we first fitted a full RDA model using all fifty-six standardized explanatory variables. All metrics and response variables were centered and scaled using the ‘standardize’ method of the decostand() function from the vegan R package version 2.6-8. To reduce dimensionality and improve model parsimony, we applied backward stepwise selection using the ordistep() function from the vegan R package with 1000 permutations. To assess the stability of selection, we repeated the procedure 100 times with different seeds and recorded variable selection frequencies. To address multicollinearity, we calculated pairwise Pearson correlation coefficients among selected variables (see Supplementary Fig. 8.1 and Supplementary Note 8 ). Variables were iteratively removed based on redundancy, ecological relevance according to the scientific literature 117 , 118 , and ease of interpretation. The final RDA model (R 2 Adj = 0.81) was constructed with four explanatory variables (see Supplementary Fig. 8.2): number of MPAs “np” (inversely related to mean MPA size for a given coverage), level of aggregation of the network “pladj” (Percentage of Like Adjacencies, partially redundant with the number of MPAs but giving additional information on the distance between MPAs), mean MPA compactness of the network (inverse of mean shape index “shape_mn”) and mean MPA distance to the coast averaged over the entire network (“distance_to_coast”). Full description of selected metrics is found in Supplementary Table 8.1 . Declarations Author Contribution CA, BE, FM and YS contributed to the conception and design of the study. CA and FM created the model configuration used for the analysis. NB helped with code maintenance and methodology. CA and IP-V processed SAR images to spatialize the fishing effort. CA ran the simulations, performed the analysis and wrote the manuscript. YS, BE and FM guided and contributed to the results analysis, manuscript read and corrections. All authors approved the submitted version. Acknowledgement The authors would like to thank the Pôle de Calcul et de Données Marines (PCDM) for providing access to their DATARMOR computing resources (http://www.ifremer.fr/pcdm), and the two anonymous reviewers for their constructive feedback.This research has been funded by the European Union’s Horizon 2020 research and innovation program under grant agreements No 869300 (FutureMARES), the Horizon Europe research and innovation program under grant agreement No 101060072 (ActNow), the European Union’s Horizon 2020 research and innovation program under Grant Agreement No 101059823 (B-USEFUL), France Filière Pêche under grant agreement PH/2022/10 (ADAPT), the French Ministry of Ecological Transition and the Pew fellows program in Marine Conservation at the Pew Charitable Trusts. Data availability The datasets generated and/or analyzed during the current study are available at https://doi.org/10.5281/zenodo.14039492 in the data directory. Due to size constraints (> 30 TB), raw model outputs are not publicly available but can be provided upon request. Code availability Model code and configuration are available at https://doi.org/10.5281/zenodo.14039492 . R scripts for the analyses are available at https://github.com/cleaab/MPA_scenarios_Mediterranean and https://doi.org/10.5281/zenodo.16734084 . References Caselle, J. E., Rassweiler, A., Hamilton, S. L. & Warner, R. R. Recovery trajectories of kelp forest animals are rapid yet spatially variable across a network of temperate marine protected areas. Sci. Rep. 5, 14102 (2015). Grorud-Colvert, K. et al. Marine Protected Area Networks: Assessing Whether the Whole Is Greater than the Sum of Its Parts. PLOS ONE 9, e102298 (2014). McCook, L. J. et al. Adaptive management of the Great Barrier Reef: A globally significant demonstration of the benefits of networks of marine reserves. Proc. Natl. Acad. Sci. 107, 18278–18285 (2010). Smith, J. G. et al. Conservation benefits of a large marine protected area network that spans multiple ecosystems. Conserv. Biol. e14435 (2025) doi: 10.1111/cobi.14435 . Jantke, K. et al. Poor ecological representation by an expensive reserve system: Evaluating 35 years of marine protected area expansion. Conserv. Lett. 11, e12584 (2018). Gill, D. A. et al. Capacity shortfalls hinder the performance of marine protected areas globally. Nature 543, 665–669 (2017). Iacarella, J. C., Clyde, G., Bergseth, B. J. & Ban, N. C. A synthesis of the prevalence and drivers of non-compliance in marine protected areas. Biol. Conserv. 255, 108992 (2021). Di Franco, A. et al. Five key attributes can increase marine protected areas performance for small-scale fisheries management. Sci. Rep. 6, 38135 (2016). Giakoumi, S. et al. Ecological effects of full and partial protection in the crowded Mediterranean Sea: a regional meta-analysis. Sci. Rep. 7, 8940 (2017). Edgar, G. J. et al. Global conservation outcomes depend on marine protected areas with five key features. Nature 506, 216–220 (2014). Grorud-Colvert, K. et al. The MPA Guide: A framework to achieve global goals for the ocean. Science 373, eabf0861 (2021). Hilborn, R. et al. When can marine reserves improve fisheries management? (2004). Claudet, J., Loiseau, C., Sostres, M. & Zupan, M. Underprotected Marine Protected Areas in a Global Biodiversity Hotspot. One Earth 2, 380–384 (2020). Guilhaumon, F. et al. Representing taxonomic, phylogenetic and functional diversity: new challenges for Mediterranean marine-protected areas. Divers. Distrib. 21, 175–187 (2015). Abecasis, D., Fragkopoulou, E., Claro, B. & Assis, J. Biophysical modelling and graph theory identify key connectivity hubs in the Mediterranean marine reserve network. Front. Mar. Sci. 9, (2023). Andrello, M. et al. Low Connectivity between Mediterranean Marine Protected Areas: A Biophysical Modeling Approach for the Dusky Grouper Epinephelus marginatus. PLoS ONE 8, e68564 (2013). Assis, J. et al. Weak biodiversity connectivity in the European network of no-take marine protected areas. Sci. Total Environ. 773, 145664 (2021). Ospina-Alvarez, A. et al. MPA network design based on graph theory and emergent properties of larval dispersal. Mar. Ecol. Prog. Ser. 650, 309–326 (2020). Coll, M. et al. The Biodiversity of the Mediterranean Sea: Estimates, Patterns, and Threats. PLOS ONE 5, e11842 (2010). FAO. The State of Mediterranean and Black Sea Fisheries 2023: Special Edition . (FAO, Rome, Italy, 2023). doi: 10.4060/cc8888en . Cardinale, M., Osio, G. C. & Scarcella, G. Mediterranean Sea: A Failure of the European Fisheries Management System. Front. Mar. Sci. 4, (2017). Dimarchopoulou, D., Stergiou, K. I. & Tsikliras, A. C. Gap analysis on the biology of Mediterranean marine fishes. PLOS ONE 12, e0175949 (2017). Hilborn, R. et al. Effective fisheries management instrumental in improving fish stock status. Proc. Natl. Acad. Sci. 117, 2218–2224 (2020). Gill, D. A. et al. A diverse portfolio of marine protected areas can better advance global conservation and equity. Proc. Natl. Acad. Sci. 121, e2313205121 (2024). Melnychuk, M. C., Peterson, E., Elliott, M. & Hilborn, R. Fisheries management impacts on target species status. Proc. Natl. Acad. Sci. 114, 178–183 (2017). Ban, N. C. et al. Effects of management objectives and rules on marine conservation outcomes. Conserv. Biol. 37, e14156 (2023). Melnychuk, M. et al. Identifying management actions that promote sustainable fisheries. Nat. Sustain. 4, (2021). Hilborn, R. Are MPAs effective? ICES J. Mar. Sci. 75, 1160–1162 (2018). Dinerstein, E. et al. A Global Deal For Nature: Guiding principles, milestones, and targets. Sci. Adv. 5, eaaw2869 (2019). O’Leary, B. C. et al. Effective Coverage Targets for Ocean Protection. Conserv. Lett. 9, 398–404 (2016). Relano, V. & Pauly, D. The ‘Paper Park Index’: Evaluating Marine Protected Area effectiveness through a global study of stakeholder perceptions. Mar. Policy 151, 105571 (2023). Popova, E. et al. Ecological connectivity between the areas beyond national jurisdiction and coastal waters: Safeguarding interests of coastal communities in developing countries. Mar. Policy 104, 90–102 (2019). van Gennip, S. J. et al. Going with the flow: the role of ocean circulation in global marine ecosystems under a changing climate. Glob. Change Biol. 23, 2602–2617 (2017). Chamberlain, D. A., Possingham, H. P. & Phinn, S. R. Decision-making with ecological process for coastal and marine planning: current literature and future directions. Aquat. Ecol. 56, 1–19 (2022). Podda, C. & Porporato, E. M. D. Marine spatial planning for connectivity and conservation through ecological corridors between marine protected areas and other effective area-based conservation measures. Front. Mar. Sci. 10, (2023). D’Aloia, C. C. et al. A multiple-species framework for integrating movement processes across life stages into the design of marine protected areas. Biol. Conserv. 216, 93–100 (2017). Jenkins, T. & Stevens, J. Assessing connectivity between MPAs: Selecting taxa and translating genetic data to inform policy. Mar. Policy 94, 165–173 (2018). Di Franco, A. et al. Dispersal of larval and juvenile seabream: Implications for Mediterranean marine protected areas. Biol. Conserv. 192, 361–368 (2015). Huserbråten, M. B. O. et al. Conservation, Spillover and Gene Flow within a Network of Northern European Marine Protected Areas. PLOS ONE 8, e73388 (2013). van der Meer, M. H., Berumen, M. L., Hobbs, J.-P. A. & van Herwerden, L. Population connectivity and the effectiveness of marine protected areas to protect vulnerable, exploited and endemic coral reef fishes at an endemic hotspot. Coral Reefs 34, 393–402 (2015). Robinson, N. M., Nelson, W. A., Costello, M. J., Sutherland, J. E. & Lundquist, C. J. A Systematic Review of Marine-Based Species Distribution Models (SDMs) with Recommendations for Best Practice. Front. Mar. Sci. 4, (2017). Ball, I. R., Possingham, H. P. & Watts, M. Marxan and relatives: software for spatial conservation prioritisation. Spat. Conserv. Prioritisation Quant. Methods Comput. Tools 14, 185–196 (2009). Moilanen, A. et al. Novel methods for spatial prioritization with applications in conservation, land use planning and ecological impact avoidance. Methods Ecol. Evol. 13, 1062–1072 (2022). Giakoumi, S. et al. Advances in systematic conservation planning to meet global biodiversity goals. Trends Ecol. Evol. 0, (2025). Cardoso, I., Moura, T., Mendes, H., Silva, C. & Azevedo, M. An ecosystem approach to mixed fisheries: technical and biological interactions in the Portuguese multi-gear fleet. ICES J. Mar. Sci. 72, 2618–2626 (2015). Fulton, E. A. et al. Modelling marine protected areas: Insights and hurdles. Philos. Trans. R. Soc. B Biol. Sci. 370, (2015). Rassweiler, A., Costello, C., Hilborn, R. & Siegel, D. A. Integrating scientific guidance into marine spatial planning. Proc. R. Soc. B Biol. Sci. 281, 20132252 (2014). Couve, P., Bahamon, N., Canales, C. M. & Company, J. B. Systematic Review of Multi-Species Models in Fisheries: Key Features and Current Trends. Fishes 9, 372 (2024). Kaplan, I. C. & Marshall, K. N. A guinea pig’s tale: learning to review end-to-end marine ecosystem models for management applications. ICES J. Mar. Sci. 73, 1715–1724 (2016). Micheli, F. et al. Setting Priorities for Regional Conservation Planning in the Mediterranean Sea. PLoS ONE 8, e59038 (2013). Mazor, T., Giakoumi, S., Kark, S. & Possingham, H. P. Large-scale conservation planning in a multinational marine environment: cost matters. Ecol. Appl. 24, 1115–1130 (2014). Di Lorenzo, M., Claudet, J. & Guidetti, P. Spillover from marine protected areas to adjacent fisheries has an ecological and a fishery component. J. Nat. Conserv. 32, 62–66 (2016). Ohayon, S., Granot, I. & Belmaker, J. A meta-analysis reveals edge effects within marine protected areas. Nat. Ecol. Evol. 5, 1301–1308 (2021). Halpern, B. S. et al. Recent pace of change in human impact on the world’s ocean. Sci. Rep. 9, 11609 (2019). Albouy, C., Mouillot, D., Rocklin, D., Culioli, J. M. & Loc’h, F. L. Simulation of the combined effects of artisanal and recreational fisheries on a Mediterranean MPA ecosystem using a trophic model. Mar. Ecol. Prog. Ser. 412, 207–221 (2010). Malakhoff, K. D. & Miller, R. J. After 15 years, no evidence for trophic cascades in marine protected areas. Proc. R. Soc. B Biol. Sci. 288, 20203061 (2021). Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Torres, F. Fishing Down Marine Food Webs. Science 279, 860–863 (1998). Piroddi, C. et al. Historical changes of the Mediterranean Sea ecosystem: modelling the role and impact of primary productivity and fisheries changes over time. Sci. Rep. 7, 44491 (2017). Sandin, S. A. et al. Baselines and Degradation of Coral Reefs in the Northern Line Islands. PLOS ONE 3, e1548 (2008). Edelist, D., Golani, D. & Spanier, E. First implementation of the Large Fish Index (LFI) in the eastern Mediterranean. Sci. Mar. 78, 185–192 (2014). Chao, A. et al. An attribute-diversity approach to functional diversity, functional beta diversity, and related (dis)similarity measures. Ecol. Monogr. 89, e01343 (2019). Hollitzer, H. A. L., May, F. & Blowes, S. A. A meta-analysis examining how fish biodiversity varies with marine protected area size and age. Ecosphere 14, e4733 (2023). Davies, B. F. R., Holmes, L., Bicknell, A., Attrill, M. J. & Sheehan, E. V. A decade implementing ecosystem approach to fisheries management improves diversity of taxa and traits within a marine protected area in the UK. Divers. Distrib. 28, 173–188 (2022). Lester, S. E. et al. Biological effects within no-take marine reserves: a global synthesis. Mar. Ecol. Prog. Ser. 384, 33–46 (2009). Boulanger, E. et al. Environmental DNA metabarcoding reveals and unpacks a biodiversity conservation paradox in Mediterranean marine reserves. Proc. R. Soc. B Biol. Sci. 288, 20210112 (2021). Loiseau, N. et al. Maximizing regional biodiversity requires a mosaic of protection levels. PLOS Biol. 19, e3001195 (2021). Brander, L. M. et al. The global costs and benefits of expanding Marine Protected Areas. Mar. Policy 116, 1–12 (2020). Tallis, H. et al. Prioritizing actions: spatial action maps for conservation. Ann. N. Y. Acad. Sci. 1505, 118–141 (2021). Cinner, J. E. et al. Transitions toward co-management: The process of marine resource management devolution in three east African countries. Glob. Environ. Change 22, 651–658 (2012). Bergseth, B. J. et al. Closing the compliance gap in marine protected areas with human behavioural sciences. Fish Fish. 24, 695–704 (2023). Gammage, T., Gurney, G. G., Diedrich, A. & Dudayev, R. Equity Through Co-Management in Small-Scale Fisheries—A Review. Fish Fish. 26, 425–443 (2025). Browman, H. & Stergiou, K. Perspectives on ecosystem-based approaches to the management of marine resources. Mar. Ecol.-Prog. Ser. - MAR ECOL-PROGR SER 274, 269–303 (2004). Claudet, J. et al. Marine reserves: size and age do matter. Ecol. Lett. 11, 481–489 (2008). Halpern, B. S. The Impact of Marine Reserves: Do Reserves Work and Does Reserve Size Matter? Ecol. Appl. 13, 117–137 (2003). Sciberras, M. et al. Evaluating the relative conservation value of fully and partially protected marine areas. Fish Fish. 16, 58–77 (2015). Coll, M. et al. The Mediterranean Sea under siege: spatial overlap between marine biodiversity, cumulative threats and marine reserves. Glob. Ecol. Biogeogr. 21, 465–480 (2012). Kriegl, M., Elías Ilosvay, X. E., von Dorrien, C. & Oesterwind, D. Marine Protected Areas: At the Crossroads of Nature Conservation and Fisheries Management. Front. Mar. Sci. 8, (2021). Costello, M. J. Evidence of economic benefits from marine protected areas. Sci. Mar. 88, e080–e080 (2024). Hasan, M. M., Moitree, S. M., Wahi, C. A. M. A., Talukder, M. A. M. H. & Islam, M. S. Eastern Mediterranean Sea Dispute: Exploring Sustainable Resolutions. Open J. Mar. Sci. 15, 74–99 (2025). Mazor, T., Possingham, H. P. & Kark, S. Collaboration among countries in marine conservation can achieve substantial efficiencies. Divers. Distrib. 19, 1380–1393 (2013). Katsanevakis, S. et al. Marine conservation challenges in an era of economic crisis and geopolitical instability: The case of the Mediterranean Sea. Mar. Policy 51, 31–39 (2015). Mouillot, D. et al. The socioeconomic and environmental niche of protected areas reveals global conservation gaps and opportunities. Nat. Commun. 15, 9007 (2024). Hilborn, R. & Kaiser, M. J. A path forward for analysing the impacts of marine protected areas. Nature 607, E1–E2 (2022). Abbott, J. K. & Haynie, A. C. What are we protecting? Fisher behavior and the unintended consequences of spatial closures as a fishery management tool. Ecol. Appl. 22, 762–777 (2012). Abesamis, R. A., Alcala, A. C. & Russ, G. R. How much does the fishery at Apo Island benefit from spillover of adult fish from the adjacent marine reserve? Fish. Bull. 104, 360–375 (2006). Lédée, E. J. I., Sutton, S. G., Tobin, R. C. & De Freitas, D. M. Responses and adaptation strategies of commercial and charter fishers to zoning changes in the Great Barrier Reef Marine Park. Mar. Policy 36, 226–234 (2012). Goñi, R. et al. Spillover from six western Mediterranean marine protected areas: evidence from artisanal fisheries. Mar. Ecol. Prog. Ser. 366, 159–174 (2008). Murawski, S. A., Wigley, S. E., Fogarty, M. J., Rago, P. J. & Mountain, D. G. Effort distribution and catch patterns adjacent to temperate MPAs. ICES J. Mar. Sci. 62, 1150–1167 (2005). Cabral, R. B., Gaines, S. D., Johnson, B. A., Bell, T. W. & White, C. Drivers of redistribution of fishing and non-fishing effort after the implementation of a marine protected area network. Ecol. Appl. 27, 416–428 (2017). Forcada, A., Valle, C., Sánchez-Lizaso, J. L., Bayle-Sempere, J. T. & Corsi, F. Structure and spatio-temporal dynamics of artisanal fisheries around a Mediterranean marine protected area. ICES J. Mar. Sci. 67, 191–203 (2010). Goñi, R., Quetglas, A. & Reñones, O. Spillover of spiny lobsters Palinurus elephas from a marine reserve to an adjoining fishery. Mar. Ecol. Prog. Ser. 308, 207–219 (2006). Palacios-Abrantes, J. et al. Incorporating protected areas into global fish biomass projections under climate change. FACETS 8, 1–16 (2023). Sala, E. et al. Protecting the global ocean for biodiversity, food and climate. Nature 592, 397–402 (2021). Costa, B. H. e et al. Fishers’ Behaviour in Response to the Implementation of a Marine Protected Area. PLOS ONE 8, e65057 (2013). Di Franco, A. et al. Linking home ranges to protected area size: The case study of the Mediterranean Sea. Biol. Conserv. 221, 175–181 (2018). Grüss, A., Kaplan, D. M. & Hart, D. R. Relative Impacts of Adult Movement, Larval Dispersal and Harvester Movement on the Effectiveness of Reserve Networks. PLOS ONE 6, e19960 (2011). Moffitt, E. A., Botsford, L. W., Kaplan, D. M. & O’Farrell, M. R. Marine reserve networks for species that move within a home range. Ecol. Appl. Publ. Ecol. Soc. Am. 19, 1835–1847 (2009). Sève, C. et al. Fisheries outcomes of marine protected area networks. Conserv. Lett. n/a, e12983 (2023). Dépalle, M., Thébaud, O. & Sanchirico, J. N. Accounting for Fleet Heterogeneity in Estimating the Impacts of Large-Scale Fishery Closures. Mar. Resour. Econ. 35, 361–378 (2020). Boerder, K., Schiller, L. & Worm, B. Not all who wander are lost: Improving spatial protection for large pelagic fishes. Mar. Policy 105, 80–90 (2019). Erm, P., Balmford, A. & Holden, M. H. The biodiversity benefits of marine protected areas in well-regulated fisheries. Biol. Conserv. 284, 110049 (2023). Froese, R., Stern-Pirlot, A., Winker, H. & Gascuel, D. Size matters: How single-species management can contribute to ecosystem-based fisheries management. Fish. Res. 92, 231–241 (2008). Costello, C. et al. Global fishery prospects under contrasting management regimes. Proc. Natl. Acad. Sci. 113, 5125–5129 (2016). Moullec, F. et al. Rebuilding Mediterranean marine resources under climate change. Mar. Ecol. Prog. Ser. 708, (2023). Colloca, F. et al. Rebuilding Mediterranean fisheries: a new paradigm for ecological sustainability. Fish Fish. 14, 89–109 (2013). Moullec, F. et al. Capturing the big picture of Mediterranean marine biodiversity with an end-to-end model of climate and fishing impacts. Prog. Oceanogr. 178, 102179 (2019). Sevault, F. et al. A fully coupled Mediterranean regional climate system model: design and evaluation of the ocean component for the 1980–2012 period. Tellus Dyn. Meteorol. Oceanogr. 66, 23967 (2014). Auger, P. A. et al. Functioning of the planktonic ecosystem on the Gulf of Lions shelf (NW Mediterranean) during spring and its impact on the carbon deposition: a field data and 3-D modelling combined approach. Biogeosciences 8, 3231–3261 (2011). Santamaria, C. et al. Mass Processing of Sentinel-1 Images for Maritime Surveillance. Remote Sens. 9, 678 (2017). Fernandez Arguedas, V., Velotto, D., Tings, B., Greidanus, H. & Bentes da Silva, C. A. Ship classification in high and very high resolution satellite SAR imagery. in Security Research Conference, 11th Future Security, Berlin, September 13–14 , 2016, Proceedings (eds. Ambacher, O., Quai, R. & Wagner, J.) 347–354 (Fraunhofer Verlag, Berlin, Germany, 2016). Pita, I. et al. SAR Satellite Imagery Reveals the Impact of the Covid-19 Crisis on Ship Frequentation in the French Mediterranean Waters. Front. Mar. Sci. 9, (2022). Copernicus. Copernicus Sentinel Data 2019. Available at: https://search.asf.alaska.edu/#/ . (2022). JRC. The SUMO Ship Detection Software for Satellite Radar Images h [Er]:Short Installation and User Guide . (Publications Office, LU, 2017). Greidanus, H. et al. The SUMO Ship Detector Algorithm for Satellite Radar Images. Remote Sens. 9, 246 (2017). White, J. W., Rassweiler, A., Samhouri, J. F., Stier, A. C. & White, C. Ecologists should not use statistical significance tests to interpret simulation model results. Oikos 123, 385–388 (2014). Hesselbarth, M. H. K., Sciaini, M., With, K. A., Wiegand, K. & Nowosad, J. landscapemetrics : an open-source R tool to calculate landscape metrics. Ecography 42, 1648–1657 (2019). Rodríguez-Rodríguez, D., Rodríguez, J., Blanco, J. M. & Abdul Malak, D. Marine protected area design patterns in the Mediterranean Sea: Implications for conservation. Mar. Pollut. Bull. 110, 335–342 (2016). Abdou, K. et al. Exploring the potential effects of marine protected areas on the ecosystem structure of the Gulf of Gabes using the Ecospace model. Aquat. Living Resour. 29, 202 (2016). Additional Declarations No competing interests reported. Supplementary Files AppendicesAbelloetalNPJOceansustainabilityresubmission.pdf Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 08 May, 2026 Reviews received at journal 06 May, 2026 Reviews received at journal 30 Apr, 2026 Reviews received at journal 22 Apr, 2026 Reviewers agreed at journal 06 Apr, 2026 Reviewers agreed at journal 02 Apr, 2026 Reviewers agreed at journal 31 Mar, 2026 Reviewers invited by journal 29 Mar, 2026 Editor assigned by journal 22 Jan, 2026 Submission checks completed at journal 22 Jan, 2026 First submitted to journal 20 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8653042","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":615514541,"identity":"58217e0c-01c1-4dca-80bc-ad32e36728de","order_by":0,"name":"Clea Abello","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5UlEQVRIie3PvQrCMBDA8QsH1yXYNcWvV6gUxElfRSk4uTmJgkOhbj5OZiVQX0HQwQ9wVkTRRUyLg1Oqm2D+Q3JDfpADsNl+MSc7Z1BwgG30xAu5BF+EENBPCX1DSKRjLnEjnB9htK4QYjK49JolAtzulgYiFIUCkkNASN1VWYb6YxQEPQPxkfsCSHVi5PWVJ1ETTkUzcc83eKhxSvqeHH9COAkWqzZpwk5S5RO9S73RmR5qMVJYZHLB9VLmXdxJtF8er+uq60Tz010OW3rY7k0kq/26kWdn3vP32O2b1zabzfY3PQH5wjs0Acc9QwAAAABJRU5ErkJggg==","orcid":"","institution":"MARBEC, Univ. Montpellier, CNRS, Ifremer, IRD","correspondingAuthor":true,"prefix":"","firstName":"Clea","middleName":"","lastName":"Abello","suffix":""},{"id":615514542,"identity":"2b87ab93-576b-4c99-b392-0e257a76afda","order_by":1,"name":"Bruno Ernande","email":"","orcid":"","institution":"MARBEC, Univ. Montpellier, CNRS, Ifremer, IRD","correspondingAuthor":false,"prefix":"","firstName":"Bruno","middleName":"","lastName":"Ernande","suffix":""},{"id":615514543,"identity":"51268c1f-e82b-4035-80cc-ce74027b57b6","order_by":2,"name":"Fabien Moullec","email":"","orcid":"","institution":"MARBEC, Univ. Montpellier, CNRS, Ifremer, IRD","correspondingAuthor":false,"prefix":"","firstName":"Fabien","middleName":"","lastName":"Moullec","suffix":""},{"id":615514544,"identity":"66956779-497d-4e0a-83d9-7385a5f5188c","order_by":3,"name":"Nicolas Barrier","email":"","orcid":"","institution":"MARBEC, Univ. Montpellier, CNRS, Ifremer, IRD","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Barrier","suffix":""},{"id":615514545,"identity":"2a198af9-4e7b-478c-a245-1d0accd6b986","order_by":4,"name":"Ignacio Pita-Vaca","email":"","orcid":"","institution":"MARBEC, Univ. Montpellier, CNRS, Ifremer, IRD","correspondingAuthor":false,"prefix":"","firstName":"Ignacio","middleName":"","lastName":"Pita-Vaca","suffix":""},{"id":615514546,"identity":"7aaeb879-a711-49a6-9b8c-e73c3832a252","order_by":5,"name":"Yunne-Jai Shin","email":"","orcid":"","institution":"MARBEC, Univ. Montpellier, CNRS, Ifremer, IRD","correspondingAuthor":false,"prefix":"","firstName":"Yunne-Jai","middleName":"","lastName":"Shin","suffix":""}],"badges":[],"createdAt":"2026-01-20 21:20:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8653042/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8653042/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106402130,"identity":"cc2389b2-73e0-4a91-b3eb-28ba9b373ace","added_by":"auto","created_at":"2026-04-08 09:11:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":10885445,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSimulated changes in biomass and fisheries catch with the expansion of fully protected area (FPAs) networks relative to a scenario without protected areas.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eChanges in biomass (A) and fisheries catch (B) are simulated according to three different fishing redistribution strategies. For the random scenarios (S2 – S3), the average of the 10 FPA networks is represented in line, and the maximum and minimum in shaded. Variability due to the model's inherent stochasticity is not shown, as it is assumed to be stable when averaging 30 replicates (refer to Supplementary Figures 5. 1 and 5. 2)\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/24da8e5afc7db256288f7b7d.png"},{"id":106004360,"identity":"2aedca56-a5d3-4178-ab37-bc1557b53f6d","added_by":"auto","created_at":"2026-04-02 10:30:31","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":116129,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChanges in biomass and catch relative to before protected area establishment.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eChanges are presented for the most effective fully protected area (FPA) network scenario (“S5-Mazor GFCM”) at 10% coverage and for each fishing redistribution strategy according to the distance to the FPA (x-axis). The inside-edge and outside-edge areas are within 20 km from the edge of the FPA.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/d7d652d2303d926d35b8b893.png"},{"id":106094144,"identity":"c71bdb47-addf-4025-a5ee-523fb5c1b6fb","added_by":"auto","created_at":"2026-04-03 11:41:18","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":11914607,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSimulated changes in biomass and catch by groups of species due to fully protected areas.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Simulated changes in biomass and catch by groups of species for the scenario S5-Mazor GFCM, relative to a scenario without MPAs. (B) Simulated changes in small pelagic biomass inside fully protected areas for the six simulated scenarios. For scenarios S2 and S3, the shaded envelope represents the range of variability between the maximum and minimum values among the 10 random FPA scenarios. Variability due to the model's inherent stochasticity is not shown, as it is assumed to be stable when averaging 30 replicates (refer to Supplementary Figures 5. 1 and 5. 2).\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/e4e5d64258d93db1cf34a72a.png"},{"id":106004362,"identity":"025644a1-75b9-4e22-8f60-371464f77733","added_by":"auto","created_at":"2026-04-02 10:30:31","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":13416184,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLarge Fish Indicator and taxonomic α-diversity inside and outside fully protected areas for the six scenarios.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Large Fish Indicator (i.e., proportion of the biomass of fish ≥ 20 cm in length) inside and outside fully protected areas (FPAs). (B) Taxonomic α-diversity inside and outside FPAs calculated with the Hill-Shannon diversity index exponentiated (Hill number of order q = 1). Species are weighted in proportion to their abundances. The indicator is calculated by OSMOSE-MED grid cell and averaged between all cells inside and outside FPAs, respectively. Variability due to the model's inherent stochasticity is not shown, as it is assumed to be stable when averaging 30 replicates (refer to Supplementary Figures 5. 1).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/5ab61cf7a95ea9781b1acc2b.png"},{"id":106004363,"identity":"ef593cd0-72f9-4cce-bbe3-048368a62faa","added_by":"auto","created_at":"2026-04-02 10:30:31","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":394778,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRedundancy analysis on the effect of protected area network configuration on biomass, catch, the large fish indicator and taxonomic α-diversity.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe effect of fully protected area (FPA) network configuration on biomass, catch, the large fish indicator inside and outside FPAs and taxonomic α-diversity inside and outside FPAs is shown with the blue arrows on the redundancy analysis (RDA) triplot using scaling 2. Network configuration was characterized by four aggregation and shape metrics (red arrows): the total mean distance of the network from the coast calculated by MPA patch from its centroid; the percentage of like adjacencies (PLADJ) that describes the level of aggregation of the network (PLADJ equals 0 if the network is maximally disaggregated, i.e., every cell is a different patch, and 100 for a single aggregated patch); the mean shape index, calculated over the entire network, describing the ratio between the actual perimeter of a given patch and its hypothetical minimum perimeter if the patch would be maximally compact; and the number of patches (inversely related to mean patch size for a given coverage). All indicators were calculated for the 24 FPA network scenarios at 10% coverage and for the fishing redistribution strategy by GSA proportionally to the spatial distribution of fishing effort prior to FPA establishment. Scenarios S1 to S6 are projected on the triplot as dots with the same colors as in previous figures. Under scaling 2, the angle between response variables (blue arrows), between explanatory variables (red arrows) or between response and explanatory variables reflect their correlations. The projection of a scenario (dots S1 to S6) at right angle on a response or an explanatory variable approximates the value of this scenario along that variable. In contrast, distances between scenario projections (dots) do not approximate their Euclidian distances (see triplot using scaling 1 for this, Supplementary Figure 8. 3)\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/333b62781f7ab5baf8bf9c43.png"},{"id":106414869,"identity":"82b43f65-92c4-414a-9e01-19f9d7ce76c2","added_by":"auto","created_at":"2026-04-08 10:29:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":37904026,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/ca462eae-5a7f-4b2c-a940-f52895d0e54f.pdf"},{"id":106093700,"identity":"56084cf6-2b9c-47b1-bd2e-215b03d1aa6e","added_by":"auto","created_at":"2026-04-03 11:38:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":6052026,"visible":true,"origin":"","legend":"","description":"","filename":"AppendicesAbelloetalNPJOceansustainabilityresubmission.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8653042/v1/94274eb678f68cdcc54b9256.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Science-informed marine protected area networks outperform random and opportunistic designs","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAlthough the effectiveness of large-scale networks of marine reserves has been demonstrated in several regions of the world \u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, the current global MPA estate remains largely ineffective and costly \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. However, with sufficient dedicated financial and human resources for managing, monitoring and enforcing MPAs, the conservation benefits can be high \u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. These benefits are further enhanced by the implementation of fully or highly protected MPAs \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, defined by the \u003cem\u003eMPA Guide\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e as areas with no or minimal impact from destructive or extractive activities, particularly in overexploited areas where ecosystem structure and function are degraded \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe Mediterranean Sea is a particularly stark example of poor MPA implementation, with 95% of MPAs showing no regulatory difference from unprotected areas \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e and the majority not encompassing more biodiversity than expected by chance \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Moreover, apart from the fact that highly protected areas are almost non-existent (0.23% of the Mediterranean Sea area), the current network is highly fragmented, with the majority of protected areas concentrated in the western basin, resulting in poor network connectivity \u003csup\u003e\u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. The rich biodiversity of the Mediterranean Sea \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e is threatened by multiple anthropogenic pressures, including the overexploitation of marine resources \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Although the percentage of overexploited stocks has fallen under 60% for the first time since decades, fishing pressure is still twice above the level considered sustainable \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. This alarming situation is partly explained by the socio-political complexity of the region (surrounded by 22 countries), the lack of adherence to scientific advice, and the inadequacy of existing national management plans, which do not take sufficient account of the mixed nature of Mediterranean fisheries and the ecological interactions between fished stocks and with the environment \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Furthermore, the majority of Mediterranean fished stocks remain unassessed \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e and therefore cannot be managed conventionally, e.g. by setting quotas. However, studies have shown that the existence of scientific assessments and recommendations combined with well-developed management tools can positively correlate with improvements in stock status and harvest rates \u003csup\u003e\u003cspan additionalcitationids=\"CR24 CR25\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe implementation of management actions at the basin scale is greatly complicated by the lack of effective rebuilding plans to date \u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. The western Mediterranean multiannual plan (2019\u0026ndash;2025) is the first Europe-wide restoration plan for demersal stocks, but it focuses on a small part of the Mediterranean Sea, and involves only three countries (France, Italy, Spain). In this context, the development of a network of large-scale connected marine reserves stands out as an option to explore for biodiversity conservation and ecosystem-based fisheries management in the Mediterranean Sea \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. With conservation plans on the table to increase the extent of MPAs worldwide, such as the international thirty by thirty initiative (30 x 30), aiming at protecting 30% of the ocean by 2030 \u003csup\u003e29,30\u003c/sup\u003e, it is crucial that we seize this opportunity to implement effective MPAs, rather than more \u0026lsquo;paper parks\u0026rsquo; \u003csup\u003e31\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCurrent approaches in conservation planning increasingly emphasize the importance of maximizing ecological connectivity between reserves \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. In the MPA modeling field, ecological connectivity is mostly addressed through larval dispersal or gene flow (see \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e for a review), using biophysical models \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, graph theory \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, and, less frequently, genetic markers \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e, with the majority of studies focusing on a single emblematic species (e.g., 11, 13, 26\u0026ndash;28). Another widely used approach in conservation planning involves identifying biodiversity hotspots through species distribution models (see 29 for a review). Systematic conservation planning tools, such as Marxan \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e or Zonation \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e, are also commonly used to optimize the spatial design of reserves, aiming to achieve biodiversity targets while minimizing cost for human activities or maximizing biodiversity benefits under certain budget or area constraints \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWhile all these methods can be applied at fine spatial resolutions and provide science-based planning solutions, they commonly overlook demographic and ecosystem responses to changes in anthropogenic pressures such as fishing, especially those caused by the implementation of MPA networks. This is even more essential in a context of mixed fisheries, where biological (species predator-prey relationships) and technical (gears targeting multiple stocks and in competition for the same stock) interactions can result in non-linear and unexpected responses of species abundance and fisheries productivity (e.g., \u003csup\u003e45\u003c/sup\u003e). End-to-end ecosystem models, which explicitly represent the dynamics of ecosystem physical and biogeochemical properties, trophic networks, species habitats and human activities, can overcome these limitations and be used to explore the strategic implementation of MPAs \u003csup\u003e\u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. Yet, they are still rarely used for management purposes due to their complexity and inherent uncertainties that make them unfamiliar to decision makers \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHere, the conservation and fisheries outcomes of expanding the network of fully protected areas (FPAs) was investigated at the scale of the whole Mediterranean Sea. We hypothesize that carefully designed and strategically placed (i.e., science-informed) MPA networks will perform better in protecting biodiversity and rebuilding marine resources than the expansion of the current MPA network and/or randomly placed MPAs \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. To address this, we propose a novel framework using the spatially-explicit multi-species individual-based ecosystem model OSMOSE (see \u003cem\u003eSupplementary Note 1\u003c/em\u003e) coupled to a high-resolution regional climate model and a regional biogeochemical model. We fine-scale modeled the full life cycle of nearly 100 fish and macro-invertebrate species, representing ca. 85% of the total reported catches in the Mediterranean Sea \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e (see \u003cem\u003eSupplementary Table\u0026nbsp;6. 1\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eWe seek to inform policy recommendations for large-scale conservation planning in the Mediterranean Sea by comparing six types of FPA networks: one resulting from the expansion of current MPAs with existing fishing regulations (i.e., National MPAs, Marine Natura 2000 areas and Fisheries Restricted Areas) (Scenario \u0026ldquo;S1-Current\u0026rdquo;), two in which MPAs are randomly distributed, either over the entire Mediterranean Sea (Scenario \u0026ldquo;S2-Random basin\u0026rdquo;) or in existing or theoretical Economic Exclusive Zones (EEZ) in proportion to the size of each country\u0026rsquo;s maritime area (Scenario \u0026ldquo;S3-Random EEZ\u0026rdquo;), and three networks proposed in the scientific literature based on a combination of ecological, economic, social, political and cultural criteria, we hereafter name \u0026lsquo;science-informed\u0026rsquo; networks \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e (Scenarios S4 to S6) (see \u003cem\u003eSupplementary Note 4\u003c/em\u003e). Among the three science-informed FPA network scenarios, the first (Scenario \u0026ldquo;S4-Micheli\u0026rdquo;) results from the identification of areas with the strongest consensus on conservation priorities identified across more than five proposed whole-basin conservation initiatives. These initiatives are mainly biodiversity-driven, considering various criteria such as habitat types and species distributions, but also incorporate cultural and geological factors, as well as human threat intensity \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. In contrast, the second and third scenarios \u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e are designed to protect 10% of the distribution area of 77 threatened marine species while minimizing the impact on fisheries using the systematic planning tool Marxan \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. The two latter differ in the origin of the data used to estimate commercial fishing opportunity cost (i.e., lost revenues because of protection): either from the General Fisheries Commission for the Mediterranean (Scenario \u0026ldquo;S5-Mazor GFCM\u0026rdquo;) or from the Sea Around Us project (Scenario \u0026ldquo;S6-Mazor SAUP\u0026rdquo;). For each FPA scenario, we compared the impacts of different levels of coverage of the Mediterranean Sea on fish and fisheries by increasing all networks from 1 to 30% coverage in steps of 1%. Additionally, we considered three possible strategies of fishing effort redistribution after FPA establishment: (1) fishing-the-line redistribution, i.e., reallocation of the displaced effort from each FPA at its boundary, (2) uniform redistribution by GFCM Geographical Sub-Area (GSA), and (3) redistribution by GSA proportionally to the spatial distribution of fishing effort prior to FPA establishment (see \u003cem\u003eSupplementary Note 3\u003c/em\u003e for further details).\u003c/p\u003e"},{"header":"Results and Discussion","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eScience-informed MPA networks better balance biomass gains and catch losses\u003c/h2\u003e \u003cp\u003eAs expected, increasing FPA coverage resulted in an increase in total biomass of fish and macro-invertebrate species, regardless of network configuration (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). However, we found that for a given level of coverage, the simulation of science-informed MPA networks (S4-S6) led to higher biomass gains than random networks (S2-S3) and even than a network resulting from the expansion of current MPAs, Natura 2000 sites and Fisheries Restricted Areas in the Mediterranean assuming they become no-take areas (S1). For instance, at 10% coverage, the biomass increase ranged from 1 to 2.9% for the science-informed FPA networks (S4-S6), from 0.6 to 1.8% for the random networks (S2-S3), and from 0.09 to 2.1% for the current network expansion (S1), depending on the fishing effort redistribution strategy.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAll scenarios resulted in a reduction in catch as FPA coverage increased, but the decrease in catch was proportionally smaller than the increase in MPA coverage. Besides the expansion of the current MPA network (S1) which has the most negative impacts on catch, catch losses ranged between ca. -2% and − 6% at 10% coverage and between ca. -8% and − 19% at 30% coverage for the other MPA scenarios (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). Furthermore, the percent catch loss always exceeded the percent biomass gains in all MPA scenarios. Expanding current MPAs (S1) appeared to be the least optimal solution, with biomass gains similar to random networks (S2-S3), but with two to three times significant greater negative impact on fisheries catch (between − 10.4 and − 8.7% decrease for S1 vs. -1.8 to -3.5% for S2-S3 random networks at 10% coverage; Wilcoxon test p-value \u0026lt; 0.05). Among the science-informed networks (S4-S6), those that use fisheries data and minimize fishing opportunity cost when selecting areas to protect (S5-S6) achieved the best balance between biomass gains and catch losses: for example, 10% MPA coverage resulted in an increase in biomass ranging from 1 to 2.6% for S5-S6 vs. 1.5 and 2.9% for S4 and a decrease in catch between − 3.3 and − 1.4% for S5-S6 vs. -6 to -4.5% for S4. The changes in biomass and catch in the “S3-Random EEZ” scenario were very similar to the random placement of reserves throughout the Mediterranean Sea (“S2-Random basin”), suggesting that an even distribution of reserve coverage across Mediterranean countries may not be sufficient to achieve an effective MPA network and that concerted action among all Mediterranean countries is necessary.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBenefits of MPA networks extend beyond their boundaries\u003c/h3\u003e\n\u003cp\u003eThe conservation benefits of marine reserve networks are not confined within their boundaries, as biomass also increased outside MPAs (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). For the most effective FPA network tested (“S5-Mazor GFCM”) at 10% coverage, the increase of biomass outside MPAs (distance \u0026gt; 20 km from the edge of the MPA) ranged between 0.8% and 1.9% according to the fishing redistribution strategy considered. This is linked to a spillover effect \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e (i.e., the export of biomass from the protected area to neighboring areas through a density-dependent process), as evidenced by the gradual decrease in biomass from inside to outside MPAs (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, left panel). There is a notable opposite response of fish biomass for the “fishing-the-line” effort redistribution strategy at reserve boundaries (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, left panel) where catches increased dramatically (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, right panel) due to high fishing effort concentration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe also found evidence of an edge effect \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e (i.e., the undermining of the effective size of protected areas, caused by human-related stressors in their surroundings) as biomass gains at the inner edge (i.e., \u0026lt; 20 km from the edge of FPAs) were less than half of those in the core, especially in the case of the “fishing-the-line” strategy. Such decline in the “effective size” of protected areas was highlighted in a recent meta-analysis combining data from 27 no-take areas around the world. The study found significant reduction in population size within 1 km inside MPAs \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e and highlighted fishing as a major driver, reinforcing the importance of establishing buffer zones around no-take reserves, where fishing activities are sustainably managed \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. However, other human-induced environmental stressors, such as pollution and runoff, can also hinder MPA success \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cb\u003eIndirect effects of MPAs through top-down trophic cascades result in winners and losers between functional groups\u003c/b\u003e \u003c/p\u003e \u003cp\u003eExamining the changes in biomass and catch by major species group revealed more nuanced results (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA, see \u003cem\u003eSupplementary Figs.\u0026nbsp;6\u003c/em\u003e. 2–6. 5 for other scenarios). Focusing on the MPA scenario with the greatest potential to increase conservation gains while minimizing catch losses (“S5-Mazor GFCM”), we observed that the decrease in total catch was mostly driven by a decrease in the biomass and catch of small pelagic species, followed by crustaceans and cephalopods. At 10% coverage, with fishing effort redistributed proportionally to pre-MPA levels (by GSA), these groups experienced biomass declines of -3.6%, -3.3% and − 14.7%, respectively, with corresponding catch losses of -11.7%, -15.4% and − 19.6%. In contrast, the biomass and catch of high trophic level species, i.e. large and medium-sized demersal and pelagic species, increased substantially with MPA coverage, with biomass gains of + 12.8% and + 9.4%, and catch increases of + 12.1% and + 8%, respectively, at 10% coverage. This was observed for nearly all of the MPA scenarios regardless of the fishing redistribution strategy considered, except for the current (S1) and random-basin (S2) scenarios under a uniform fishing redistribution strategy for MPA coverage below 10% (see \u003cem\u003eSupplementary Figs.\u0026nbsp;6. 2–6. 5\u003c/em\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAmong small pelagic species, the European anchovy (\u003cem\u003eEngraulis encrasicolus\u003c/em\u003e) and the European pilchard (\u003cem\u003eSardina pilchardus\u003c/em\u003e) experienced the steepest declines under the “S5-Mazor GFCM scenario”. At 10% coverage, their biomass dropped by -6.6% and − 3.7%, respectively, and at 30% coverage, these losses intensified to -22.1% and − 10.9% (see \u003cem\u003eSupplementary Fig.\u0026nbsp;6. 1\u003c/em\u003e). The loss of biomass of low trophic level species, while the biomass of high trophic level species increased, is due to a higher predation pressure exerted by the latter on the former due to increased MPA coverage. Specifically, predation pressure on main low-trophic level groups (small pelagic fish, cephalopods and crustaceans) increased by 3.45%, 3.32% and 1.88%, respectively, under a 10% FPA network (see \u003cem\u003eSupplementary Fig.\u0026nbsp;6. 2\u003c/em\u003e). This rise in predation was primarily driven by increased ingestion by large demersal predators, which became more abundant with MPAs.\u003c/p\u003e \u003cp\u003eSuch top-down trophic cascade effects due to increased predation pressure within marine reserves are common \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, but not always the case \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Here, the decline in the biomass of small pelagic species within reserves started at different coverage levels depending on the MPA scenario (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). The decline started at a much lower coverage level for two of the three science-informed networks (MPA coverage between 2 and 3% for S5 and S6) than for the current network and the other science-informed network (MPA coverage between 10% and 12% for S1 and S4) or random networks (MPA coverage between 5 and 20% for S2-S3). This is consistent with the fact that scenarios S5 and S6 were primarily designed to protect threatened and vulnerable species, which are mostly high trophic level species that are targeted by fisheries and exert predation pressure on low trophic levels.\u003c/p\u003e \u003cp\u003eAlthough the implementation of reserve networks would most likely have a negative impact on low trophic level species, this is not necessarily alarming from both a conservation and fisheries economic perspective. From a conservation perspective, the Mediterranean Sea has experienced a \"fishing-down the food web\" effect \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e over the last 70 years with a decrease in the abundance and catch of overexploited high trophic level fish species and an increase in macroinvertebrates (ca. 23%) \u003csup\u003e58\u003c/sup\u003e. Thus, our results suggest that an increase in FPA coverage could partially reverse this trend, restoring the biomass pyramid to more closely resemble that of pristine areas \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. From a fisheries economic perspective, the negative impact on fisheries due to the loss of fishing grounds could be economically offset despite the reduction in yields, as high trophic level species, such as European hake or bluefin tuna, are of high commercial value.\u003c/p\u003e\n\u003ch3\u003eScience-informed MPA networks are effective in restoring large fish even if biodiversity benefits remain modest\u003c/h3\u003e\n\u003cp\u003eThe recovery of large-bodied fish with increasing MPA coverage was evidenced by the Large Fish Indicator (LFI), i.e., the proportion of the biomass of large fish (≥ 20 cm in length) \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA). We estimated that if a 10% FPA network were implemented, the LFI within reserves would shift from 29% to between 30 and 38%, depending on the network and fishing redistribution strategy (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA, left panel). The benefits of FPAs for the recovery of large fish were projected to extend beyond their boundaries, as the proportion of large fish in unprotected areas also increased for a coverage ≥ 5% (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA, right panel). Science-informed FPA networks (S4-S6) were also more effective than random networks (S2-S3) for this indicator, but equivalent to expanding the current MPA network (S1).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo assess the representativeness of biodiversity within the six FPA networks, we calculated the Hill-Shannon alpha-taxonomic diversity index \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e for the 95 species modelled (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB). We found that although across all scenarios, community structure within MPAs was extremely uneven and dominated by a few very abundant species, the current MPA network (S1) and consensual priority conservation areas \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e (S4) encompassed more biodiversity than areas protected randomly (S2-S3) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB, left panel). However, this was not the case for the other two science-informed networks (S5-S6) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB, left panel).\u003c/p\u003e \u003cp\u003eFurthermore, in scenarios S1 and S4, biodiversity within FPAs was, on average, 1.5 times higher compared to unprotected areas, consistent with the findings of a recent meta-analysis \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. In contrast, the difference between inside and outside FPAs was less pronounced in the other scenarios, with some random scenarios (S2-S3) even showing higher biodiversity outside FPAs (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB). These differences can be attributed to differences in network configuration, as diversity outside FPAs was positively correlated with the average distance of FPAs from the coastline (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e). Consequently, more coastal networks, such as scenarios S1 and S4, captured more biodiversity than more offshore networks (see section below).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhile our modelling framework enables comparison of diversity across MPA networks under a common baseline —specifically, the distributions of 95 fish and macroinvertebrate species— it cannot assess the effectiveness of MPAs aimed at protecting biodiversity components absent from our model. Therefore, we advise caution when interpreting this index, as it does not encompass all facets of the rich biodiversity of the Mediterranean Sea. Even if scenarios S5 and S6 appeared to capture less fish diversity than S1 and S4 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB, left), more than half of the species they were designed to protect (primarily, marine mammals, turtles, seabirds and elasmobranchs) were either absent or underrepresented in our model.\u003c/p\u003e \u003cp\u003eWe further explored the potential role of FPAs in ecosystem recovery by calculating the Hill-Shannon diversity index before and after MPA implementation. Our findings showed that nearly all FPA networks experienced increases in alpha diversity following the ban on fishing activities (see \u003cem\u003eSupplementary Fig.\u0026nbsp;7. 1\u003c/em\u003e), in line with previous studies \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e–\u003cspan class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. However, the positive effect on biodiversity remained modest and was particularly noticeable when MPA coverage exceeded 10%. Among the different networks, FPAs in S1 and S4 proved most effective in enhancing local taxonomic diversity, even though we observed that beyond a certain threshold of MPA coverage, conservation gains plateaued, suggesting diminishing returns. These patterns align with previous research indicating that biodiversity gains within marine reserves may be limited by ecological feedbacks, such as increased predation pressure within reserves, which can reduce the diversity of lower trophic level species \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e65\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAlthough a global study assessing the ecological and economic costs and benefits of expanding MPAs put forward that protecting pristine areas overall yielded greater net benefits than targeting low biodiversity areas or heavily impacted ecosystems \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, the high level of human activities in the Mediterranean Sea calls for a careful evaluation of socio-ecological trade-offs at a more local scale. Given the Mediterranean’s complex socio-political landscape, protecting high biodiversity areas might not always be feasible nor the first and best solution. Strategical prioritization that engages local stakeholders and decision makers \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e, and a stronger focus on actions rather than areas, as emphasized by \u003csup\u003e44,68\u003c/sup\u003e, is needed to inform decisions on most adapted conservation actions for each area in order to maximize conservation outcomes. Inclusive governance is particularly critical in areas where communities heavily depend on the sea and its resources \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. Empirical evidence \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e has demonstrated that high levels of local participation, capacity building and transparent decision-making are significantly more likely to achieve ecological success and reduce illegal fishing activities. Moreover, perceived legitimacy and equity are essential predictors of compliance in marine resource management \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eMPA network design is crucial to achieve effective outcomes\u003c/h3\u003e\n\u003cp\u003eThe size, shape, spacing and location of MPAs have a direct influence on the ecological effectiveness of MPA networks \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e72\u003c/span\u003e–\u003cspan class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e. To better understand the observed differences between our six FPA network scenarios, we computed several shape and aggregation metrics and used redundancy analysis (RDA) to explore their relationship with absolute changes in biomass and catch, the Large Fish Indicator (LFI) inside and outside protected areas and the Hill-Shannon diversity indicator inside and outside protected areas following MPA establishment using 10% coverage as an example (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe found that catch declined less when reserves were further from shore (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, Distance to coast red arrow), which was the case for nearly all of the random networks (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, S2-S3, gray and black dots). This is consistent with the spatial distribution of fishing effort in the Mediterranean Sea (see \u003cem\u003eSupplementary Fig.\u0026nbsp;2. 4\u003c/em\u003e), which is mainly concentrated close to the coast. Catch also decreased less (even if less strongly) as the average compactness of MPAs in the network decreased (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, Mean shape index red arrow). Compactness is defined as the ratio between an MPA’s actual perimeter and the minimum possible perimeter it would have if it were perfectly compact. Biomass and the proportion of large fish both inside and outside FPAs increased more when the networks were more aggregated (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, PLADJ red arrow) and MPA patches larger (i.e., fewer patches; Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, Number of patches red arrow). As the current MPA network consists of smaller reserves closer to shore (see \u003cem\u003eSupplementary Fig.\u0026nbsp;4. 1\u003c/em\u003e), it is not surprising that this network (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, S1, yellow dot) appeared suboptimal compared to the others in achieving the best balance between biomass gains and catch losses, despite encompassing areas with above-average biodiversity. The latter was related to the fact that taxonomic α-diversity inside protected areas was higher when reserves were more coastal (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, Diversity inside MPA blue arrow versus Distance to coast red arrow and \u003cem\u003eSupplementary Fig.\u0026nbsp;7. 2\u003c/em\u003e). This aligns with the findings of \u003csup\u003e76\u003c/sup\u003e, who reported fish biodiversity to be primarily concentrated in coastal regions, including Spain, France and Italy, the north-western coast of Africa, the Ionian and Aegean Sea, and the Adriatic Sea.\u003c/p\u003e \u003cp\u003eThe RDA also highlighted the structural differences between the three science-informed networks, the random networks and the current one (see \u003cem\u003eSupplementary Fig.\u0026nbsp;8. 3\u003c/em\u003e for a RDA triplot using scaling 1 under which the distances between scenario projections approximate their Euclidian distances). Our analysis indicates that compact networks of large reserves should be preferred over patchy ones to maximize conservation outcomes. To minimize the impact on fisheries, reserves should be placed further offshore. One of the downsides of MPA establishment in areas where fishing takes place is the emerging trade-off between fisheries management goals and conservation objectives that must be resolved \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. In this study, for example, all tested FPA networks, although yielding benefits for conservation, projected a decrease in catch. However, our spatial analysis suggests that a win-win strategy, balancing both conservation and fisheries benefits is possible. Notably, conservation indicators such as biomass and the Large Fish Indicator showed relative independence from fishing metrics, as evidenced by the 90 degree angle between biomass/LFI and catch indicators (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, blue arrows). This is supported by real-world examples where MPAs have delivered tangible benefits to fisheries \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e. Moreover, the conservation and fisheries indicators responded to spatial metrics that were relatively independent as reflected by the nearly 90 degree angle between the number of patches or PLADJ and the distance to the coast (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e). Hence our analysis shows that a more optimal network configuration in terms of both conservation and fisheries benefits could be achieved by establishing a few large offshore reserves that are relatively aggregated at the network level (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, bottom right quadrant). This would be done without sacrificing the protection of biodiversity hotspots, as although diversity would decrease inside MPAs, it would increase outside (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, Diversity blue arrows). Indeed, some offshore areas exhibit high α-diversity (see \u003cem\u003eSupplementary Fig.\u0026nbsp;7. 2\u003c/em\u003e for the distribution of α-diversity across the Mediterranean Sea).\u003c/p\u003e \u003cp\u003eSuch recommendations are conditional on multinational cooperation, which is not an easy task in the Mediterranean Sea, given the region’s geopolitical complexity. In certain areas, like in the Eastern Mediterranean Sea, ongoing disputes over maritime boundaries \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e exacerbated by resource competition, are hampering any conservation action. Yet, existing ties between countries, supported by collaborative legislation frameworks, such as the European Union or the Arab Maghreb Union, have facilitated coordinated conservation initiatives, such as the Natura 2000 marine network or the Pelagos Sanctuary. Research has shown that coordinated actions could be more cost-efficient for most Mediterranean countries, reducing costs by about two-thirds compared to fully independent national efforts \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e. Encouragingly, many Mediterranean countries are actively declaring or negotiating their EEZs, offering hope that clearer jurisdictional delineations will further catalyze coordinated conservation actions in the region in the near future\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e81\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAmong the 24 different networks we tested (incl. the expansion of the existing network (S1), 10 configurations for each random scenario (S2-S3) and 3 science-informed networks (S4-S6)), the three science-informed networks (S4-S6, Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e) came closest to this optimal configuration. Nevertheless, our results suggest that there is still significant room for improvement in MPA placement and overall network design, as most of the ensemble of optimal configurations remain unexplored (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, bottom right quadrant). In addition, national and local considerations, as well as refined socio-economic criteria have not been prioritized in the scenarios found in the literature \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e, nor in their simulated expansion, despite the fact that socio-economic factors appear to be critical to identify suitable areas for MPA establishment \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e. This leads to some unrealistic MPA placement and size in all scenarios.\u003c/p\u003e\n\u003ch3\u003eFishing redistribution strategies can alter the effectiveness of MPAs\u003c/h3\u003e\n\u003cp\u003eIn addition to the spatial design of the network, our study highlights the importance of the fishing effort redistribution strategy modeled. We found that the conservation and fishery benefits were sensitive to the strategy deployed. We modeled three simple fishing redistribution strategies and found that the higher effectiveness of one network over the other in terms of biomass and catch could be reduced or reversed depending on the fishing strategy considered. As an example, total biomass gains were more important in the current MPA network (S1) than in random networks (S2-S3) or in the “Mazor SAUP” network (S6) up to a 20% coverage with a fishing-the-line redistribution strategy, whereas this was not the case with a uniform redistribution strategy (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Redistributing fishing effort proportionally to pre-closure effort density resulted in higher conservation gains in terms of total biomass than either the fishing-the-line or uniform redistribution strategy (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). In contrast, the uniform redistribution strategy had the least impact on fishing activities in terms of catch loss, significantly lower than both the fishing-the-line and the proportional redistribution strategy (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e and see \u003cem\u003eSupplementary Fig.\u0026nbsp;5. 3\u003c/em\u003e for map of catch change outside protected areas at 10% coverage).\u003c/p\u003e \u003cp\u003eAlthough our archetypal strategies of fishing redistribution do not capture the complexity of fishing behavior following a fishery closure, the proportional redistribution strategy is commonly observed and could be considered as the default option \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e. Several fisheries around the world exhibit such site-fidelity behavior, resulting in increased fishing effort in areas of already high fishing pressure following spatial closures \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e84\u003c/span\u003e–\u003cspan class=\"CitationRef\"\u003e86\u003c/span\u003e\u003c/sup\u003e. Concentration of fishing effort around reserves, commonly referred to as “fishing-the-line”, is another frequently observed strategy \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e87\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e88\u003c/span\u003e\u003c/sup\u003e. This strategy adopted by fishers assumes that the net export of biomass from the reserve should increase catch rates in adjacent unprotected waters. However, it has been shown to be highly fleet and resource-dependent \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e89\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/sup\u003e. Furthermore, boats tend to concentrate in the first few kilometers beyond the reserve boundary \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e88\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e91\u003c/span\u003e\u003c/sup\u003e. As our model’s spatial resolution (20 x 20 km) does not allow us to account for the exponential decrease in effort from the reserve boundary, we may underestimate fishing competition at the edge of the reserve when simulating this strategy (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cem\u003eSupplementary Fig.\u0026nbsp;5. 3\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eOur results show that, although simplifying assumptions are sometimes needed when using already complex modeling tools, there are still opportunities for cost-effective improvements. For example, it may be more realistic to model a proportional fishing redistribution strategy rather than a uniform redistribution, or even the collapse of part of the fleets, as has been done in recent modeling studies \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e92\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e93\u003c/span\u003e\u003c/sup\u003e. A maximum redistribution distance within a sub-area of the system should also be considered in models to avoid unrealistic scenarios where fishers can travel implausible distances in a single day. Indeed, distance travelled has been identified as a key factor influencing the selection of new fishing grounds by artisanal fisheries \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e94\u003c/span\u003e\u003c/sup\u003e. This criterion is especially relevant in regions such as the Mediterranean Sea, where artisanal fisheries make up over 80% of the fishing fleet \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWithin this context, and in the interest of transparency, we highlight below key methodological considerations that both frame the interpretation of our results and point toward promising avenues for future model development.\u003c/p\u003e \u003cp\u003eFirst, as mentioned above, the spatial resolution of our model (20 x 20 km) prevents us from representing scenarios with smaller MPAs, as we are limited to an MPA minimum size of 400 km², while the mean size of current Mediterranean MPAs is 143 km². While MPAs with enhanced conservation benefits are often larger than 100 km² \u003csup\u003e9\u003c/sup\u003e, smaller MPAs have also been shown to yield positive outcomes \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, in particular when protection aligns with species home ranges \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e94\u003c/span\u003e\u003c/sup\u003e. However, since the objective is to simulate an increase in MPA coverage in the Mediterranean, the lack of fine spatial resolution may not prevent the model from capturing the general patterns across the different expansion scenarios, Adopting a finer resolution (e.g., 10 x 10 km), as recommended by the European Union’s Directive 2007/2/EC for spatial planning, would enhance direct tactical support to management. Here, our contribution is rather strategic in nature: the different scenarios help illuminate key ecological and fishing processes involved and inform the optimal design–both in terms of location and configuration–of MPAs to support conservation and fisheries objectives at the basin scale.\u003c/p\u003e \u003cp\u003eSecond, the model assumes random movement of fish —represented as super-individuals or “schools” of biologically identical individuals of the same species, born at the same time— across each species' distribution area. While this approach allows to account for climate niche of the species, it does not account for fine behavioral processes such as territoriality or ontogenetic migrations, nor for functional differences between species linked to mobility (whether active for adults or passive for larval pelagic stages) or habitat preferences. Previous studies have demonstrated that adult movement is a critical determinant of the effectiveness of reserve networks \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e96\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e97\u003c/span\u003e\u003c/sup\u003e. In our model, schools have an equal probability of remaining in their current cell or moving to one of the eight adjacent cells at every 15-day time step. However, 24 out of the 95 species modeled are primarily sedentary, and 60 have relatively low mobility (see \u003cem\u003eSupplementary Table\u0026nbsp;6. 1\u003c/em\u003e for list of species). Consequently, population persistence within reserve networks, particularly for sedentary species, is most likely underestimated in our study. Constraining movement within home ranges is complex and data-intensive; a simpler approach could involve differentiating movement capacity by species or groups of species. Another limitation of our modeling approach lies in the simplified representation of larval dispersal, which does not incorporate physical oceanographic processes. Instead, newborn larval schools are randomly distributed across the species’ distribution map, potentially leading to an overestimation of connectivity both among MPAs and between MPAs and unprotected areas within the distribution map \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e98\u003c/span\u003e\u003c/sup\u003e. This random distribution indeed assumes that larvae have an equal probability of reaching any cell within the map, whereas physical constraints would likely limit the range of accessible cells. Incorporating these physical processes would also require identifying species’ spawning grounds, as these are unlikely to be uniformly distributed across the adults’ habitat. Such refinements could provide a more realistic assessment of connectivity between reserves, as well as their contribution to fisheries through spill-over effects, thereby improving the evaluation of reserve network efficiency \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e97\u003c/span\u003e\u003c/sup\u003e. However, gathering this information for 95 species would be a daunting task.\u003c/p\u003e \u003cp\u003eThird, our model considers the distribution of each species as determined by its climate niche but does not account for ontogenetic habitat shifts. To better reflect the importance of protecting certain areas over others, it would be beneficial to incorporate spatial distribution maps by stage or size class to identify major spawning and nursery grounds. These aspects make us confident in affirming that our results are most likely conservative.\u003c/p\u003e \u003cp\u003eAs data on the spatial distribution of fishing effort differentiated by fleet were not available at the Mediterranean scale, we did not account for fleet heterogeneity, which may introduce some bias in the estimated impact of MPAs on fisheries \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e99\u003c/span\u003e\u003c/sup\u003e. Distinguishing between large-scale and small-scale fisheries in future applications would enable the inclusion of more nuanced management options, such as multi-zone MPAs. In addition, fisheries impacts were assessed using catch as a primary indicator, but extending the framework to incorporate complementary socio-economic indicators, such as fishing income and costs, would provide a more comprehensive assessment of management outcomes.\u003c/p\u003e \u003cp\u003eDespite these simplifying assumptions, our modeling framework is, to our knowledge, the first ecosystem modeling approach that combines such ecological detail with full basin-scale coverage for this region. Our findings consistently demonstrate the higher positive impacts of science-informed FPA networks over both random and existing expansions in the Mediterranean Sea. While our results do not directly demonstrate this outcome, the spatial analysis suggests that a well-designed FPA network could potentially support the recovery and rebuilding of Mediterranean exploited species, without major tradeoffs for fisheries. This could be feasible before 2030, in line with the CBD’s post-2020 strategic plan, if concertation and political support follow. Although our analysis was conducted for the Mediterranean Sea, some of our findings are generic enough to apply to many other regions, and science-informed networks are needed where exploitation levels are high, stocks are overexploited and overall ecosystem health is degraded \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eYet, spatial protection alone may not be sufficient in light of the current climate and biodiversity crises fish and fisheries are facing \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e100\u003c/span\u003e\u003c/sup\u003e. Therefore, alternative management measures, such as annual catch limits and shares, effort regulation or gear modifications, are also necessary \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e101\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e102\u003c/span\u003e\u003c/sup\u003e. These measures have the potential to significantly increase fisheries catches and profits \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e103\u003c/span\u003e\u003c/sup\u003e. This is especially evident in the Mediterranean Sea \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e104\u003c/span\u003e\u003c/sup\u003e, which suffers from excessively high fishing effort and inappropriate selectivity patterns, resulting in average fishing mortality rates for all species that are 2.5 times higher than the level at which maximum sustainable yield can be achieved \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Additionally, most commercial species have a low size at first capture \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e105\u003c/span\u003e\u003c/sup\u003e. Reducing fishing effort and increasing length at first capture by at least 30% could enhance total biomass and catch levels of high trophic level species, particularly demersal, large pelagic and benthic species \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e104\u003c/span\u003e\u003c/sup\u003e. Future large-scale studies should focus on assessing the combined effects of multiple management strategies. The primary challenge is to combine the results of basin-scale assessments such as here with more detailed and local impact studies carried out in close collaboration with multiple stakeholders. This will help promote effective and integrated strategies for conservation and fisheries management.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e\n\n "},{"header":"Methods","content":"\u003ch2\u003eThe OSMOSE end-to-end model of the Mediterranean Sea\u003c/h2\u003e\u003cp\u003eThe ecological and fishery effects of the six FPA networks were simulated using a previously developed OSMOSE model (Object-oriented Simulator of Marine ecOSystEms) at whole-basin Mediterranean scale for the period 2006–2013 \u003csup\u003e106\u003c/sup\u003e, here slightly modified to improve the spatial representation of fishing activity. OSMOSE is a spatially explicit age and size-structured individual-based ecosystem model representing the dynamics and main life cycle processes of key high trophic level species, including growth, reproduction, movement and different mortality sources (predation, starvation, fishing and natural mortalities). Each ‘super-individual’ or school of fish is independently modeled. The model makes the assumption that predation processes are opportunistic, driven by spatial overlap and size-based constraints between ‘super-individuals’, which allow the emergence of complex food-web structures. The detailed OSMOSE model documentation can be found at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osmose-model.org/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e and has been synthesized in the \u003cem\u003eSupplementary Note 1\u003c/em\u003e.\u003c/p\u003e\u003cp\u003eThe Mediterranean configuration of the OSMOSE model was extended through one-way coupling with the regional circulation model CNRM-RCSM4 \u003csup\u003e107\u003c/sup\u003e and the biogeochemical model Eco3M-S \u003csup\u003e108\u003c/sup\u003e, simulating the dynamics of seven major planktonic functional groups in the Mediterranean for the same period \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e106\u003c/span\u003e\u003c/sup\u003e. The resulting end-to-end model, OSMOSE-MED, simulates the life cycles and trophic interactions of 95 marine species (82 fish, 5 cephalopods and 8 crustaceans) of ecological and commercial importance in the Mediterranean, accounting for 85% of total declared catches according to the Sea Around Us Project reconstructed catch data for the period 2006–2013, at relatively high spatiotemporal resolution for a basin-scale model (regular grid of 6229 cells of 20 x 20 km; 15-day time step). The specificities of the Mediterranean configuration OSMOSE-MED, its parameterization and calibration are fully described in \u003csup\u003e106\u003c/sup\u003e and in the \u003cem\u003eSupplementary Note 1\u003c/em\u003e. Model code and configuration are publicly available (see Code availability section).\u003c/p\u003e\u003ch3\u003eSpatial representation of fishing activity\u003c/h3\u003e\u003cp\u003eTo improve the spatial representation of fishing activity, we estimated fishing effort at basin-scale using Synthetic Aperture Radar (SAR) satellite imagery data. SAR satellites provide high-resolution images that can detect vessels greater than 10 m in length, providing a solution for real-time vessel location \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e109\u003c/span\u003e\u003c/sup\u003e. The use of SAR satellite imagery data is not critically affected by weather conditions (e.g., cloud cover) and day-night cycles \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e110\u003c/span\u003e\u003c/sup\u003e. Although it was necessary to make several assumptions to distinguish fishing from other activities \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e111\u003c/span\u003e\u003c/sup\u003e, this technology allowed us to obtain a more accurate representation of fishing effort in the Mediterranean compared to using Automatic Identification System (AIS) data. Indeed, as in the Mediterranean, AIS is only mandatory in EU waters for vessels over 15 m in length and for ships of over 300 gross tonnage engaged on international voyages (International Maritime Organization), which are unlikely to be fishing vessels, significant gaps in AIS coverage are particularly noticeable in the southern and eastern parts of the basin. We obtained 14,278 Ground Range Detected High-Resolution images (i.e., images with a range detection of 10m per pixel) from Sentinel-1A and Sentinel-1B, the SAR satellite constellation of the European Union’s Copernicus program for Earth Observation, operated by the European Space Agency (ESA) from the Alaska Satellite Facility platform \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e112\u003c/span\u003e\u003c/sup\u003e. These images cover the entire Mediterranean basin from January 1 to December 31, 2019. The Search for Unidentified Maritime Objects (SUMO) algorithm, developed by the European Joint Research Center \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e113\u003c/span\u003e\u003c/sup\u003e was used to detect vessels. SUMO is a pixel-based Constant False Alarm Rate detector that uses several detection thresholds to differentiate vessels from sea clutter \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e114\u003c/span\u003e\u003c/sup\u003e. Data was post-processed to minimize the number of false positives and provide a more realistic estimation of fishing effort (see \u003cem\u003eSupplementary Note 2\u003c/em\u003e for details).\u003c/p\u003e\u003ch2\u003eMPA and fishing redistribution scenarios\u003c/h2\u003e\u003cp\u003eWe evaluated the ecological and fisheries impacts of alternative FPA scenarios by systematically increasing protected area coverage from 1 to 30% in 1% increments. Each network was initialized with 62 marine protected area seed locations, corresponding to a single OSMOSE-MED grid cell (~ 1% of the Mediterranean simulation grid, comprising 6,229 cells). Seed placement differed by scenario: in S1 (“Current”), seeds were positioned at the barycenter of current MPAs; in S2 and S3, they were distributed randomly across the basin or within EEZs, respectively; and in science-informed scenarios S4-S6 (“Micheli”, “Mazor GFCM” and “Mazor SAUP”), seeds were located in literature-identified priority areas. All scenarios, except S4, which followed a prioritization scheme, were then expanded using a randomized rook-neighborhood approach, in which new cells were added to one of the four orthogonally adjacent grid cells. Expansion was either fully random (in the case of S2 and S3) or constrained by the existing or proposed conservation plans (S1, S5, S6). Further details on the design of the six FPA scenarios and the three fishing effort reallocation strategies are provided in the \u003cem\u003eSupplementary Notes 4 and 3\u003c/em\u003e, respectively. Each simulation ran for 200 years, with FPAs being implemented in year 110 to ensure sufficient spin-up time for the system to reach equilibrium. For all the analyses, we averaged the last 60 years (from year 140 to 200) to leave sufficient time after MPA establishment for the model to reach a new steady state. To account for the stochasticity of the model, each simulation was replicated 30 times, which offered a good compromise between simulation time and model stochasticity (see Supplementary Fig.\u0026nbsp;5. 1). A total of 2,160 sets of 30 replicated simulations for the 24 FPA network configurations, representing the six FPA scenarios, were ran. Each FPA scenario was modeled considering three fishing redistribution strategies for each coverage level ranging from 1 to 30% in 1% increments. We also modeled a reference state simulation without MPAs. To assess the impact of FPAs and compare the six network scenarios, we calculated the relative or absolute change before and after MPAs for the different indicators computed (i.e., total biomass and catch and by groups of species, the large fish indicator and taxonomic α-diversity). We only evaluated the magnitude of difference between scenarios as recommended by \u003csup\u003e115\u003c/sup\u003e.\u003c/p\u003e\u003ch2\u003eSpatial and aggregation metrics of MPA network scenarios\u003c/h2\u003e\u003cp\u003eThe different spatial and aggregation metrics used for the redundancy analysis (RDA) were calculated using R version 4.4.0 and the \u003cem\u003elandscapemetrics\u003c/em\u003e R package version 2.1.4 \u003csup\u003e116\u003c/sup\u003e, except for mean MPA network distance to the coast. In total, fifty-six metrics were computed (see the \u003cem\u003elandscapemetrics\u003c/em\u003e package documentation for details). To assess the relationship between the response variables (i.e., total biomass and catch and large fish indicator and taxonomic α-diversity inside and outside MPAs) and a suite of explanatory spatial and aggregation metrics, we first fitted a full RDA model using all fifty-six standardized explanatory variables. All metrics and response variables were centered and scaled using the ‘standardize’ method of the decostand() function from the \u003cem\u003evegan\u003c/em\u003e R package version 2.6-8. To reduce dimensionality and improve model parsimony, we applied backward stepwise selection using the ordistep() function from the \u003cem\u003evegan\u003c/em\u003e R package with 1000 permutations. To assess the stability of selection, we repeated the procedure 100 times with different seeds and recorded variable selection frequencies. To address multicollinearity, we calculated pairwise Pearson correlation coefficients among selected variables (see Supplementary Fig.\u0026nbsp;8.1 and \u003cem\u003eSupplementary Note 8\u003c/em\u003e). Variables were iteratively removed based on redundancy, ecological relevance according to the scientific literature \u003csup\u003e\u003cspan class=\"CitationRef\"\u003e117\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e118\u003c/span\u003e\u003c/sup\u003e, and ease of interpretation. The final RDA model (R\u003csup\u003e2\u003c/sup\u003eAdj = 0.81) was constructed with four explanatory variables (see Supplementary Fig.\u0026nbsp;8.2): number of MPAs “np” (inversely related to mean MPA size for a given coverage), level of aggregation of the network “pladj” (Percentage of Like Adjacencies, partially redundant with the number of MPAs but giving additional information on the distance between MPAs), mean MPA compactness of the network (inverse of mean shape index “shape_mn”) and mean MPA distance to the coast averaged over the entire network (“distance_to_coast”). Full description of selected metrics is found in \u003cem\u003eSupplementary Table\u0026nbsp;8.1\u003c/em\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eCA, BE, FM and YS contributed to the conception and design of the study. CA and FM created the model configuration used for the analysis. NB helped with code maintenance and methodology. CA and IP-V processed SAR images to spatialize the fishing effort. CA ran the simulations, performed the analysis and wrote the manuscript. YS, BE and FM guided and contributed to the results analysis, manuscript read and corrections. All authors approved the submitted version.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors would like to thank the P\u0026ocirc;le de Calcul et de Donn\u0026eacute;es Marines (PCDM) for providing access to their DATARMOR computing resources (http://www.ifremer.fr/pcdm), and the two anonymous reviewers for their constructive feedback.This research has been funded by the European Union\u0026rsquo;s Horizon 2020 research and innovation program under grant agreements No 869300 (FutureMARES), the Horizon Europe research and innovation program under grant agreement No 101060072 (ActNow), the European Union\u0026rsquo;s Horizon 2020 research and innovation program under Grant Agreement No 101059823 (B-USEFUL), France Fili\u0026egrave;re P\u0026ecirc;che under grant agreement PH/2022/10 (ADAPT), the French Ministry of Ecological Transition and the Pew fellows program in Marine Conservation at the Pew Charitable Trusts.\u003c/p\u003e\n\u003ch3\u003eData availability\u003c/h3\u003e\n\u003cp\u003eThe datasets generated and/or analyzed during the current study are available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.14039492\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e in the data directory. Due to size constraints (\u0026gt; 30 TB), raw model outputs are not publicly available but can be provided upon request.\u003c/p\u003e\u003ch2\u003eCode availability\u003c/h2\u003e\u003cp\u003eModel code and configuration are available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.14039492\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. R scripts for the analyses are available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/cleaab/MPA_scenarios_Mediterranean\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.16734084\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCaselle, J. E., Rassweiler, A., Hamilton, S. L. \u0026amp; Warner, R. R. Recovery trajectories of kelp forest animals are rapid yet spatially variable across a network of temperate marine protected areas. \u003cem\u003eSci. Rep.\u003c/em\u003e 5, 14102 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrorud-Colvert, K. \u003cem\u003eet al.\u003c/em\u003e Marine Protected Area Networks: Assessing Whether the Whole Is Greater than the Sum of Its Parts. \u003cem\u003ePLOS ONE\u003c/em\u003e 9, e102298 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCook, L. J. \u003cem\u003eet al.\u003c/em\u003e Adaptive management of the Great Barrier Reef: A globally significant demonstration of the benefits of networks of marine reserves. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 107, 18278\u0026ndash;18285 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith, J. G. \u003cem\u003eet al.\u003c/em\u003e Conservation benefits of a large marine protected area network that spans multiple ecosystems. \u003cem\u003eConserv. Biol.\u003c/em\u003e e14435 (2025) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/cobi.14435\u003c/span\u003e\u003cspan address=\"10.1111/cobi.14435\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJantke, K. \u003cem\u003eet al.\u003c/em\u003e Poor ecological representation by an expensive reserve system: Evaluating 35 years of marine protected area expansion. \u003cem\u003eConserv. Lett.\u003c/em\u003e 11, e12584 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGill, D. A. \u003cem\u003eet al.\u003c/em\u003e Capacity shortfalls hinder the performance of marine protected areas globally. \u003cem\u003eNature\u003c/em\u003e 543, 665\u0026ndash;669 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIacarella, J. C., Clyde, G., Bergseth, B. J. \u0026amp; Ban, N. C. A synthesis of the prevalence and drivers of non-compliance in marine protected areas. \u003cem\u003eBiol. Conserv.\u003c/em\u003e 255, 108992 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Franco, A. \u003cem\u003eet al.\u003c/em\u003e Five key attributes can increase marine protected areas performance for small-scale fisheries management. \u003cem\u003eSci. Rep.\u003c/em\u003e 6, 38135 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiakoumi, S. \u003cem\u003eet al.\u003c/em\u003e Ecological effects of full and partial protection in the crowded Mediterranean Sea: a regional meta-analysis. \u003cem\u003eSci. Rep.\u003c/em\u003e 7, 8940 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdgar, G. J. \u003cem\u003eet al.\u003c/em\u003e Global conservation outcomes depend on marine protected areas with five key features. \u003cem\u003eNature\u003c/em\u003e 506, 216\u0026ndash;220 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrorud-Colvert, K. \u003cem\u003eet al.\u003c/em\u003e The MPA Guide: A framework to achieve global goals for the ocean. \u003cem\u003eScience\u003c/em\u003e 373, eabf0861 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilborn, R. \u003cem\u003eet al.\u003c/em\u003e When can marine reserves improve fisheries management? (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClaudet, J., Loiseau, C., Sostres, M. \u0026amp; Zupan, M. Underprotected Marine Protected Areas in a Global Biodiversity Hotspot. \u003cem\u003eOne Earth\u003c/em\u003e 2, 380\u0026ndash;384 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuilhaumon, F. \u003cem\u003eet al.\u003c/em\u003e Representing taxonomic, phylogenetic and functional diversity: new challenges for Mediterranean marine-protected areas. \u003cem\u003eDivers. Distrib.\u003c/em\u003e 21, 175\u0026ndash;187 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbecasis, D., Fragkopoulou, E., Claro, B. \u0026amp; Assis, J. Biophysical modelling and graph theory identify key connectivity hubs in the Mediterranean marine reserve network. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e 9, (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrello, M. \u003cem\u003eet al.\u003c/em\u003e Low Connectivity between Mediterranean Marine Protected Areas: A Biophysical Modeling Approach for the Dusky Grouper Epinephelus marginatus. \u003cem\u003ePLoS ONE\u003c/em\u003e 8, e68564 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAssis, J. \u003cem\u003eet al.\u003c/em\u003e Weak biodiversity connectivity in the European network of no-take marine protected areas. \u003cem\u003eSci. Total Environ.\u003c/em\u003e 773, 145664 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOspina-Alvarez, A. \u003cem\u003eet al.\u003c/em\u003e MPA network design based on graph theory and emergent properties of larval dispersal. \u003cem\u003eMar. Ecol. Prog. Ser.\u003c/em\u003e 650, 309\u0026ndash;326 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColl, M. \u003cem\u003eet al.\u003c/em\u003e The Biodiversity of the Mediterranean Sea: Estimates, Patterns, and Threats. \u003cem\u003ePLOS ONE\u003c/em\u003e 5, e11842 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAO. \u003cem\u003eThe State of Mediterranean and Black Sea Fisheries 2023: Special Edition\u003c/em\u003e. (FAO, Rome, Italy, 2023). doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4060/cc8888en\u003c/span\u003e\u003cspan address=\"10.4060/cc8888en\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCardinale, M., Osio, G. C. \u0026amp; Scarcella, G. Mediterranean Sea: A Failure of the European Fisheries Management System. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e 4, (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDimarchopoulou, D., Stergiou, K. I. \u0026amp; Tsikliras, A. C. Gap analysis on the biology of Mediterranean marine fishes. \u003cem\u003ePLOS ONE\u003c/em\u003e 12, e0175949 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilborn, R. \u003cem\u003eet al.\u003c/em\u003e Effective fisheries management instrumental in improving fish stock status. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 117, 2218\u0026ndash;2224 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGill, D. A. \u003cem\u003eet al.\u003c/em\u003e A diverse portfolio of marine protected areas can better advance global conservation and equity. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 121, e2313205121 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMelnychuk, M. C., Peterson, E., Elliott, M. \u0026amp; Hilborn, R. Fisheries management impacts on target species status. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 114, 178\u0026ndash;183 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBan, N. C. \u003cem\u003eet al.\u003c/em\u003e Effects of management objectives and rules on marine conservation outcomes. \u003cem\u003eConserv. Biol.\u003c/em\u003e 37, e14156 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMelnychuk, M. \u003cem\u003eet al.\u003c/em\u003e Identifying management actions that promote sustainable fisheries. \u003cem\u003eNat. Sustain.\u003c/em\u003e 4, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilborn, R. Are MPAs effective? \u003cem\u003eICES J. Mar. Sci.\u003c/em\u003e 75, 1160\u0026ndash;1162 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDinerstein, E. \u003cem\u003eet al.\u003c/em\u003e A Global Deal For Nature: Guiding principles, milestones, and targets. \u003cem\u003eSci. Adv.\u003c/em\u003e 5, eaaw2869 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Leary, B. C. \u003cem\u003eet al.\u003c/em\u003e Effective Coverage Targets for Ocean Protection. \u003cem\u003eConserv. Lett.\u003c/em\u003e 9, 398\u0026ndash;404 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRelano, V. \u0026amp; Pauly, D. The \u0026lsquo;Paper Park Index\u0026rsquo;: Evaluating Marine Protected Area effectiveness through a global study of stakeholder perceptions. \u003cem\u003eMar. Policy\u003c/em\u003e 151, 105571 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePopova, E. \u003cem\u003eet al.\u003c/em\u003e Ecological connectivity between the areas beyond national jurisdiction and coastal waters: Safeguarding interests of coastal communities in developing countries. \u003cem\u003eMar. Policy\u003c/em\u003e 104, 90\u0026ndash;102 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Gennip, S. J. \u003cem\u003eet al.\u003c/em\u003e Going with the flow: the role of ocean circulation in global marine ecosystems under a changing climate. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e 23, 2602\u0026ndash;2617 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChamberlain, D. A., Possingham, H. P. \u0026amp; Phinn, S. R. Decision-making with ecological process for coastal and marine planning: current literature and future directions. \u003cem\u003eAquat. Ecol.\u003c/em\u003e 56, 1\u0026ndash;19 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePodda, C. \u0026amp; Porporato, E. M. D. Marine spatial planning for connectivity and conservation through ecological corridors between marine protected areas and other effective area-based conservation measures. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e 10, (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD\u0026rsquo;Aloia, C. C. \u003cem\u003eet al.\u003c/em\u003e A multiple-species framework for integrating movement processes across life stages into the design of marine protected areas. \u003cem\u003eBiol. Conserv.\u003c/em\u003e 216, 93\u0026ndash;100 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJenkins, T. \u0026amp; Stevens, J. Assessing connectivity between MPAs: Selecting taxa and translating genetic data to inform policy. \u003cem\u003eMar. Policy\u003c/em\u003e 94, 165\u0026ndash;173 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Franco, A. \u003cem\u003eet al.\u003c/em\u003e Dispersal of larval and juvenile seabream: Implications for Mediterranean marine protected areas. \u003cem\u003eBiol. Conserv.\u003c/em\u003e 192, 361\u0026ndash;368 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuserbr\u0026aring;ten, M. B. O. \u003cem\u003eet al.\u003c/em\u003e Conservation, Spillover and Gene Flow within a Network of Northern European Marine Protected Areas. \u003cem\u003ePLOS ONE\u003c/em\u003e 8, e73388 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Meer, M. H., Berumen, M. L., Hobbs, J.-P. A. \u0026amp; van Herwerden, L. Population connectivity and the effectiveness of marine protected areas to protect vulnerable, exploited and endemic coral reef fishes at an endemic hotspot. \u003cem\u003eCoral Reefs\u003c/em\u003e 34, 393\u0026ndash;402 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobinson, N. M., Nelson, W. A., Costello, M. J., Sutherland, J. E. \u0026amp; Lundquist, C. J. A Systematic Review of Marine-Based Species Distribution Models (SDMs) with Recommendations for Best Practice. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e 4, (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBall, I. R., Possingham, H. P. \u0026amp; Watts, M. Marxan and relatives: software for spatial conservation prioritisation. \u003cem\u003eSpat. Conserv. Prioritisation Quant. Methods Comput. Tools\u003c/em\u003e 14, 185\u0026ndash;196 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoilanen, A. \u003cem\u003eet al.\u003c/em\u003e Novel methods for spatial prioritization with applications in conservation, land use planning and ecological impact avoidance. \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e 13, 1062\u0026ndash;1072 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiakoumi, S. \u003cem\u003eet al.\u003c/em\u003e Advances in systematic conservation planning to meet global biodiversity goals. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e 0, (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCardoso, I., Moura, T., Mendes, H., Silva, C. \u0026amp; Azevedo, M. An ecosystem approach to mixed fisheries: technical and biological interactions in the Portuguese multi-gear fleet. \u003cem\u003eICES J. Mar. Sci.\u003c/em\u003e 72, 2618\u0026ndash;2626 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFulton, E. A. \u003cem\u003eet al.\u003c/em\u003e Modelling marine protected areas: Insights and hurdles. \u003cem\u003ePhilos. Trans. R. Soc. B Biol. Sci.\u003c/em\u003e 370, (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRassweiler, A., Costello, C., Hilborn, R. \u0026amp; Siegel, D. A. Integrating scientific guidance into marine spatial planning. \u003cem\u003eProc. R. Soc. B Biol. Sci.\u003c/em\u003e 281, 20132252 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCouve, P., Bahamon, N., Canales, C. M. \u0026amp; Company, J. B. Systematic Review of Multi-Species Models in Fisheries: Key Features and Current Trends. \u003cem\u003eFishes\u003c/em\u003e 9, 372 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaplan, I. C. \u0026amp; Marshall, K. N. A guinea pig\u0026rsquo;s tale: learning to review end-to-end marine ecosystem models for management applications. \u003cem\u003eICES J. Mar. Sci.\u003c/em\u003e 73, 1715\u0026ndash;1724 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMicheli, F. \u003cem\u003eet al.\u003c/em\u003e Setting Priorities for Regional Conservation Planning in the Mediterranean Sea. \u003cem\u003ePLoS ONE\u003c/em\u003e 8, e59038 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMazor, T., Giakoumi, S., Kark, S. \u0026amp; Possingham, H. P. Large-scale conservation planning in a multinational marine environment: cost matters. \u003cem\u003eEcol. Appl.\u003c/em\u003e 24, 1115\u0026ndash;1130 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Lorenzo, M., Claudet, J. \u0026amp; Guidetti, P. Spillover from marine protected areas to adjacent fisheries has an ecological and a fishery component. \u003cem\u003eJ. Nat. Conserv.\u003c/em\u003e 32, 62\u0026ndash;66 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhayon, S., Granot, I. \u0026amp; Belmaker, J. A meta-analysis reveals edge effects within marine protected areas. \u003cem\u003eNat. Ecol. Evol.\u003c/em\u003e 5, 1301\u0026ndash;1308 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalpern, B. S. \u003cem\u003eet al.\u003c/em\u003e Recent pace of change in human impact on the world\u0026rsquo;s ocean. \u003cem\u003eSci. Rep.\u003c/em\u003e 9, 11609 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlbouy, C., Mouillot, D., Rocklin, D., Culioli, J. M. \u0026amp; Loc\u0026rsquo;h, F. L. Simulation of the combined effects of artisanal and recreational fisheries on a Mediterranean MPA ecosystem using a trophic model. \u003cem\u003eMar. Ecol. Prog. Ser.\u003c/em\u003e 412, 207\u0026ndash;221 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalakhoff, K. D. \u0026amp; Miller, R. J. After 15 years, no evidence for trophic cascades in marine protected areas. \u003cem\u003eProc. R. Soc. B Biol. Sci.\u003c/em\u003e 288, 20203061 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePauly, D., Christensen, V., Dalsgaard, J., Froese, R. \u0026amp; Torres, F. Fishing Down Marine Food Webs. \u003cem\u003eScience\u003c/em\u003e 279, 860\u0026ndash;863 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiroddi, C. \u003cem\u003eet al.\u003c/em\u003e Historical changes of the Mediterranean Sea ecosystem: modelling the role and impact of primary productivity and fisheries changes over time. \u003cem\u003eSci. Rep.\u003c/em\u003e 7, 44491 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandin, S. A. \u003cem\u003eet al.\u003c/em\u003e Baselines and Degradation of Coral Reefs in the Northern Line Islands. \u003cem\u003ePLOS ONE\u003c/em\u003e 3, e1548 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdelist, D., Golani, D. \u0026amp; Spanier, E. First implementation of the Large Fish Index (LFI) in the eastern Mediterranean. \u003cem\u003eSci. Mar.\u003c/em\u003e 78, 185\u0026ndash;192 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChao, A. \u003cem\u003eet al.\u003c/em\u003e An attribute-diversity approach to functional diversity, functional beta diversity, and related (dis)similarity measures. \u003cem\u003eEcol. Monogr.\u003c/em\u003e 89, e01343 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHollitzer, H. A. L., May, F. \u0026amp; Blowes, S. A. A meta-analysis examining how fish biodiversity varies with marine protected area size and age. \u003cem\u003eEcosphere\u003c/em\u003e 14, e4733 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavies, B. F. R., Holmes, L., Bicknell, A., Attrill, M. J. \u0026amp; Sheehan, E. V. A decade implementing ecosystem approach to fisheries management improves diversity of taxa and traits within a marine protected area in the UK. \u003cem\u003eDivers. Distrib.\u003c/em\u003e 28, 173\u0026ndash;188 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLester, S. E. \u003cem\u003eet al.\u003c/em\u003e Biological effects within no-take marine reserves: a global synthesis. \u003cem\u003eMar. Ecol. Prog. Ser.\u003c/em\u003e 384, 33\u0026ndash;46 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoulanger, E. \u003cem\u003eet al.\u003c/em\u003e Environmental DNA metabarcoding reveals and unpacks a biodiversity conservation paradox in Mediterranean marine reserves. \u003cem\u003eProc. R. Soc. B Biol. Sci.\u003c/em\u003e 288, 20210112 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoiseau, N. \u003cem\u003eet al.\u003c/em\u003e Maximizing regional biodiversity requires a mosaic of protection levels. \u003cem\u003ePLOS Biol.\u003c/em\u003e 19, e3001195 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrander, L. M. \u003cem\u003eet al.\u003c/em\u003e The global costs and benefits of expanding Marine Protected Areas. \u003cem\u003eMar. Policy\u003c/em\u003e 116, 1\u0026ndash;12 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTallis, H. \u003cem\u003eet al.\u003c/em\u003e Prioritizing actions: spatial action maps for conservation. \u003cem\u003eAnn. N. Y. Acad. Sci.\u003c/em\u003e 1505, 118\u0026ndash;141 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCinner, J. E. \u003cem\u003eet al.\u003c/em\u003e Transitions toward co-management: The process of marine resource management devolution in three east African countries. \u003cem\u003eGlob. Environ. Change\u003c/em\u003e 22, 651\u0026ndash;658 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBergseth, B. J. \u003cem\u003eet al.\u003c/em\u003e Closing the compliance gap in marine protected areas with human behavioural sciences. \u003cem\u003eFish Fish.\u003c/em\u003e 24, 695\u0026ndash;704 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGammage, T., Gurney, G. G., Diedrich, A. \u0026amp; Dudayev, R. Equity Through Co-Management in Small-Scale Fisheries\u0026mdash;A Review. \u003cem\u003eFish Fish.\u003c/em\u003e 26, 425\u0026ndash;443 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrowman, H. \u0026amp; Stergiou, K. Perspectives on ecosystem-based approaches to the management of marine resources. \u003cem\u003eMar. Ecol.-Prog. Ser. - MAR ECOL-PROGR SER\u003c/em\u003e 274, 269\u0026ndash;303 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClaudet, J. \u003cem\u003eet al.\u003c/em\u003e Marine reserves: size and age do matter. \u003cem\u003eEcol. Lett.\u003c/em\u003e 11, 481\u0026ndash;489 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalpern, B. S. The Impact of Marine Reserves: Do Reserves Work and Does Reserve Size Matter? \u003cem\u003eEcol. Appl.\u003c/em\u003e 13, 117\u0026ndash;137 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSciberras, M. \u003cem\u003eet al.\u003c/em\u003e Evaluating the relative conservation value of fully and partially protected marine areas. \u003cem\u003eFish Fish.\u003c/em\u003e 16, 58\u0026ndash;77 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColl, M. \u003cem\u003eet al.\u003c/em\u003e The Mediterranean Sea under siege: spatial overlap between marine biodiversity, cumulative threats and marine reserves. \u003cem\u003eGlob. Ecol. Biogeogr.\u003c/em\u003e 21, 465\u0026ndash;480 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKriegl, M., El\u0026iacute;as Ilosvay, X. E., von Dorrien, C. \u0026amp; Oesterwind, D. Marine Protected Areas: At the Crossroads of Nature Conservation and Fisheries Management. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e 8, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCostello, M. J. Evidence of economic benefits from marine protected areas. \u003cem\u003eSci. Mar.\u003c/em\u003e 88, e080\u0026ndash;e080 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHasan, M. M., Moitree, S. M., Wahi, C. A. M. A., Talukder, M. A. M. H. \u0026amp; Islam, M. S. Eastern Mediterranean Sea Dispute: Exploring Sustainable Resolutions. \u003cem\u003eOpen J. Mar. Sci.\u003c/em\u003e 15, 74\u0026ndash;99 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMazor, T., Possingham, H. P. \u0026amp; Kark, S. Collaboration among countries in marine conservation can achieve substantial efficiencies. \u003cem\u003eDivers. Distrib.\u003c/em\u003e 19, 1380\u0026ndash;1393 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKatsanevakis, S. \u003cem\u003eet al.\u003c/em\u003e Marine conservation challenges in an era of economic crisis and geopolitical instability: The case of the Mediterranean Sea. \u003cem\u003eMar. Policy\u003c/em\u003e 51, 31\u0026ndash;39 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMouillot, D. \u003cem\u003eet al.\u003c/em\u003e The socioeconomic and environmental niche of protected areas reveals global conservation gaps and opportunities. \u003cem\u003eNat. Commun.\u003c/em\u003e 15, 9007 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilborn, R. \u0026amp; Kaiser, M. J. A path forward for analysing the impacts of marine protected areas. \u003cem\u003eNature\u003c/em\u003e 607, E1\u0026ndash;E2 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbbott, J. K. \u0026amp; Haynie, A. C. What are we protecting? Fisher behavior and the unintended consequences of spatial closures as a fishery management tool. \u003cem\u003eEcol. Appl.\u003c/em\u003e 22, 762\u0026ndash;777 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbesamis, R. A., Alcala, A. C. \u0026amp; Russ, G. R. How much does the fishery at Apo Island benefit from spillover of adult fish from the adjacent marine reserve? \u003cem\u003eFish. Bull.\u003c/em\u003e 104, 360\u0026ndash;375 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eL\u0026eacute;d\u0026eacute;e, E. J. I., Sutton, S. G., Tobin, R. C. \u0026amp; De Freitas, D. M. Responses and adaptation strategies of commercial and charter fishers to zoning changes in the Great Barrier Reef Marine Park. \u003cem\u003eMar. Policy\u003c/em\u003e 36, 226\u0026ndash;234 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGo\u0026ntilde;i, R. \u003cem\u003eet al.\u003c/em\u003e Spillover from six western Mediterranean marine protected areas: evidence from artisanal fisheries. \u003cem\u003eMar. Ecol. Prog. Ser.\u003c/em\u003e 366, 159\u0026ndash;174 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMurawski, S. A., Wigley, S. E., Fogarty, M. J., Rago, P. J. \u0026amp; Mountain, D. G. Effort distribution and catch patterns adjacent to temperate MPAs. \u003cem\u003eICES J. Mar. Sci.\u003c/em\u003e 62, 1150\u0026ndash;1167 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCabral, R. B., Gaines, S. D., Johnson, B. A., Bell, T. W. \u0026amp; White, C. Drivers of redistribution of fishing and non-fishing effort after the implementation of a marine protected area network. \u003cem\u003eEcol. Appl.\u003c/em\u003e 27, 416\u0026ndash;428 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForcada, A., Valle, C., S\u0026aacute;nchez-Lizaso, J. L., Bayle-Sempere, J. T. \u0026amp; Corsi, F. Structure and spatio-temporal dynamics of artisanal fisheries around a Mediterranean marine protected area. \u003cem\u003eICES J. Mar. Sci.\u003c/em\u003e 67, 191\u0026ndash;203 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGo\u0026ntilde;i, R., Quetglas, A. \u0026amp; Re\u0026ntilde;ones, O. Spillover of spiny lobsters Palinurus elephas from a marine reserve to an adjoining fishery. \u003cem\u003eMar. Ecol. Prog. Ser.\u003c/em\u003e 308, 207\u0026ndash;219 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePalacios-Abrantes, J. \u003cem\u003eet al.\u003c/em\u003e Incorporating protected areas into global fish biomass projections under climate change. \u003cem\u003eFACETS\u003c/em\u003e 8, 1\u0026ndash;16 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSala, E. \u003cem\u003eet al.\u003c/em\u003e Protecting the global ocean for biodiversity, food and climate. \u003cem\u003eNature\u003c/em\u003e 592, 397\u0026ndash;402 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCosta, B. H. e \u003cem\u003eet al.\u003c/em\u003e Fishers\u0026rsquo; Behaviour in Response to the Implementation of a Marine Protected Area. \u003cem\u003ePLOS ONE\u003c/em\u003e 8, e65057 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Franco, A. \u003cem\u003eet al.\u003c/em\u003e Linking home ranges to protected area size: The case study of the Mediterranean Sea. \u003cem\u003eBiol. Conserv.\u003c/em\u003e 221, 175\u0026ndash;181 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGr\u0026uuml;ss, A., Kaplan, D. M. \u0026amp; Hart, D. R. Relative Impacts of Adult Movement, Larval Dispersal and Harvester Movement on the Effectiveness of Reserve Networks. \u003cem\u003ePLOS ONE\u003c/em\u003e 6, e19960 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoffitt, E. A., Botsford, L. W., Kaplan, D. M. \u0026amp; O\u0026rsquo;Farrell, M. R. Marine reserve networks for species that move within a home range. \u003cem\u003eEcol. Appl. Publ. Ecol. Soc. Am.\u003c/em\u003e 19, 1835\u0026ndash;1847 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS\u0026egrave;ve, C. \u003cem\u003eet al.\u003c/em\u003e Fisheries outcomes of marine protected area networks. \u003cem\u003eConserv. Lett.\u003c/em\u003e n/a, e12983 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD\u0026eacute;palle, M., Th\u0026eacute;baud, O. \u0026amp; Sanchirico, J. N. Accounting for Fleet Heterogeneity in Estimating the Impacts of Large-Scale Fishery Closures. \u003cem\u003eMar. Resour. Econ.\u003c/em\u003e 35, 361\u0026ndash;378 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoerder, K., Schiller, L. \u0026amp; Worm, B. Not all who wander are lost: Improving spatial protection for large pelagic fishes. \u003cem\u003eMar. Policy\u003c/em\u003e 105, 80\u0026ndash;90 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErm, P., Balmford, A. \u0026amp; Holden, M. H. The biodiversity benefits of marine protected areas in well-regulated fisheries. \u003cem\u003eBiol. Conserv.\u003c/em\u003e 284, 110049 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFroese, R., Stern-Pirlot, A., Winker, H. \u0026amp; Gascuel, D. Size matters: How single-species management can contribute to ecosystem-based fisheries management. \u003cem\u003eFish. Res.\u003c/em\u003e 92, 231\u0026ndash;241 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCostello, C. \u003cem\u003eet al.\u003c/em\u003e Global fishery prospects under contrasting management regimes. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 113, 5125\u0026ndash;5129 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoullec, F. \u003cem\u003eet al.\u003c/em\u003e Rebuilding Mediterranean marine resources under climate change. \u003cem\u003eMar. Ecol. Prog. Ser.\u003c/em\u003e 708, (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColloca, F. \u003cem\u003eet al.\u003c/em\u003e Rebuilding Mediterranean fisheries: a new paradigm for ecological sustainability. \u003cem\u003eFish Fish.\u003c/em\u003e 14, 89\u0026ndash;109 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoullec, F. \u003cem\u003eet al.\u003c/em\u003e Capturing the big picture of Mediterranean marine biodiversity with an end-to-end model of climate and fishing impacts. \u003cem\u003eProg. Oceanogr.\u003c/em\u003e 178, 102179 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSevault, F. \u003cem\u003eet al.\u003c/em\u003e A fully coupled Mediterranean regional climate system model: design and evaluation of the ocean component for the 1980\u0026ndash;2012 period. \u003cem\u003eTellus Dyn. Meteorol. Oceanogr.\u003c/em\u003e 66, 23967 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuger, P. A. \u003cem\u003eet al.\u003c/em\u003e Functioning of the planktonic ecosystem on the Gulf of Lions shelf (NW Mediterranean) during spring and its impact on the carbon deposition: a field data and 3-D modelling combined approach. \u003cem\u003eBiogeosciences\u003c/em\u003e 8, 3231\u0026ndash;3261 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantamaria, C. \u003cem\u003eet al.\u003c/em\u003e Mass Processing of Sentinel-1 Images for Maritime Surveillance. \u003cem\u003eRemote Sens.\u003c/em\u003e 9, 678 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFernandez Arguedas, V., Velotto, D., Tings, B., Greidanus, H. \u0026amp; Bentes da Silva, C. A. Ship classification in high and very high resolution satellite SAR imagery. in \u003cem\u003eSecurity Research Conference, 11th Future Security, Berlin, September 13\u0026ndash;14\u003c/em\u003e, 2016, \u003cem\u003eProceedings\u003c/em\u003e (eds. Ambacher, O., Quai, R. \u0026amp; Wagner, J.) 347\u0026ndash;354 (Fraunhofer Verlag, Berlin, Germany, 2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePita, I. \u003cem\u003eet al.\u003c/em\u003e SAR Satellite Imagery Reveals the Impact of the Covid-19 Crisis on Ship Frequentation in the French Mediterranean Waters. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e 9, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCopernicus. Copernicus Sentinel Data 2019. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://search.asf.alaska.edu/#/\u003c/span\u003e\u003cspan address=\"https://search.asf.alaska.edu/#/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJRC. \u003cem\u003eThe SUMO Ship Detection Software for Satellite Radar Images h [Er]:Short Installation and User Guide\u003c/em\u003e. (Publications Office, LU, 2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreidanus, H. \u003cem\u003eet al.\u003c/em\u003e The SUMO Ship Detector Algorithm for Satellite Radar Images. \u003cem\u003eRemote Sens.\u003c/em\u003e 9, 246 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhite, J. W., Rassweiler, A., Samhouri, J. F., Stier, A. C. \u0026amp; White, C. Ecologists should not use statistical significance tests to interpret simulation model results. \u003cem\u003eOikos\u003c/em\u003e 123, 385\u0026ndash;388 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHesselbarth, M. H. K., Sciaini, M., With, K. A., Wiegand, K. \u0026amp; Nowosad, J. \u003cem\u003elandscapemetrics\u003c/em\u003e: an open-source \u003cem\u003eR\u003c/em\u003e tool to calculate landscape metrics. \u003cem\u003eEcography\u003c/em\u003e 42, 1648\u0026ndash;1657 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodr\u0026iacute;guez-Rodr\u0026iacute;guez, D., Rodr\u0026iacute;guez, J., Blanco, J. M. \u0026amp; Abdul Malak, D. Marine protected area design patterns in the Mediterranean Sea: Implications for conservation. \u003cem\u003eMar. Pollut. Bull.\u003c/em\u003e 110, 335\u0026ndash;342 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbdou, K. \u003cem\u003eet al.\u003c/em\u003e Exploring the potential effects of marine protected areas on the ecosystem structure of the Gulf of Gabes using the Ecospace model. \u003cem\u003eAquat. Living Resour.\u003c/em\u003e 29, 202 (2016).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-ocean-sustainability","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjoceansustain","sideBox":"Learn more about [npj Ocean Sustainability](https://www.nature.com/npjoceansustain/)","snPcode":"44183","submissionUrl":"https://mts-npjoceansustain.nature.com/cgi-bin/main.plex","title":"npj Ocean Sustainability","twitterHandle":"@NaturePortfolio","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"npj","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8653042/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8653042/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe 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 \u0026ldquo;fishing down the food web\u0026rdquo; 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.\u003c/p\u003e","manuscriptTitle":"Science-informed marine protected area networks outperform random and opportunistic designs","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-02 10:30:20","doi":"10.21203/rs.3.rs-8653042/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-08T08:43:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T13:49:03+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-30T15:47:20+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-22T11:19:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"57348318338053507958567757969520239848","date":"2026-04-06T21:01:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"308583682349224423155539980653619109834","date":"2026-04-02T12:56:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"116321335169176235283930552671774376280","date":"2026-03-31T14:10:59+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-29T14:05:04+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-22T22:05:23+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-22T09:14:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Ocean Sustainability","date":"2026-01-20T20:43:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-ocean-sustainability","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjoceansustain","sideBox":"Learn more about [npj Ocean Sustainability](https://www.nature.com/npjoceansustain/)","snPcode":"44183","submissionUrl":"https://mts-npjoceansustain.nature.com/cgi-bin/main.plex","title":"npj Ocean Sustainability","twitterHandle":"@NaturePortfolio","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"npj","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"01d37d19-a357-4b51-a10a-9e0af245945e","owner":[],"postedDate":"April 2nd, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-08T08:43:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T13:49:03+00:00","index":27,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-30T15:47:20+00:00","index":26,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[{"id":65495290,"name":"Biological sciences/Ecology"},{"id":65495291,"name":"Earth and environmental sciences/Ecology"},{"id":65495292,"name":"Earth and environmental sciences/Environmental sciences"},{"id":65495293,"name":"Earth and environmental sciences/Ocean sciences"}],"tags":[],"updatedAt":"2026-05-08T08:57:16+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-02 10:30:20","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8653042","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8653042","identity":"rs-8653042","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}