Cumulative human impacts on global marine fauna highlight risk to biological and functional diversity

RESEARCH ARTICLE Cumulative human impacts on global marine fauna highlight risk to biological and functional diversity Casey C. O’HaraID 1 *, Melanie Frazier 1 , Mireia ValleID 1,2,3 , Nathalie Butt 4,5 , Kristin Kaschner 6 , Carissa Klein 5 , Benjamin S. Halpern 1,7 1National Center for Ecological Analysis and Synthesis, University of California, Santa Barbara, Santa Barbara, California, United States of America, 2AZTI, Marine Research, Basque Research and Technology Alliance (BRTA), Sukarrieta, Spain, 3Basque Centre for Climate Change (BC3), Scientific Campus of the University of the Basque Country (UPV-EHU), Leioa, Spain, 4The Nature Conservancy, South Brisbane, Queensland, Australia, 5Centre for Biodiversity and Conservation Science, School of the Environment, The University of Queensland, Brisbane, Queensland, Australia, 6Department of Biometry and Environmental System Analysis, Albert-Ludwigs-University of Freiburg, Freiburg, Germany, 7Bren School of Environmental Science & Management, University of California, Santa Barbara, Santa Barbara, California, United States of America *[email protected]b.edu Abstract Anthropogenic stressors to marine ecosystems from climate change and human activities increase extinction risk of species, disrupt ecosystem integrity, and threaten important ecosystem services. Addressing these stressors requires understanding where and to what extent they are impacting marine biological and functional diversity. We model cumulative risk of human impact upon 21,159 marine animal species by combining information on species-level vulnerability and spatial exposure to a range of anthropogenic stressors. We apply this species-level assessment of human impacts to examine patterns of speciesstressor interactions within taxonomic groups. We then spatially map impacts across the global ocean, identifying locations where climate-driven impacts overlap with fishing, shipping, and land-based stressors to help inform conservation needs and opportunities. Comparing species-level modeled impacts to those based on marine habitats that represent important marine ecosystems, we find that even relatively untouched habitats may still be home to species at elevated risk, and that many species-rich coastal regions may be at greater risk than indicated from habitat-based methods alone. Finally, we incorporate a traitbased metric of functional diversity to identify where impacts to functionally unique species might pose greater risk to community structure and ecosystem integrity. These complementary lenses of species, function, and habitat provide a richer understanding of threats to marine biodiversity to help inform efforts to meet conservation targets and ensure sustainability of nature’s contributions to people. PLOS ONE PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 1 / 20 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: O’Hara CC, Frazier M, Valle M, Butt N, Kaschner K, Klein C, et al. (2024) Cumulative human impacts on global marine fauna highlight risk to biological and functional diversity. PLoS ONE 19(9): e0309788. https://doi.org/10.1371/ journal.pone.0309788 Editor: Abdul Azeez Pokkathappada, Central Marine Fisheries Research Institute, INDIA Received: June 17, 2024 Accepted: August 19, 2024 Published: September 18, 2024 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0309788 Copyright: ©2024 O’Hara et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All data used as inputs for this assessment are freely available from original sources. All original code and data resulting from this analysis has been deposited at Introduction Anthropogenic stressors from human activities on land and sea coupled with those from anthropogenic climate change are driving degradation of coastal ecosystems, marine regime shifts, and increased extinction risk of threatened species [1–6], threatening the biological and functional diversity that underpin ecosystem services upon which we depend [7–11]. The Kunming-Montreal Global Biodiversity Framework [12] calls for urgent action to reduce extinction risk, improve ecosystem integrity, and ensure sustainability of nature’s contributions to people. To these ends, understanding where and to what degree anthropogenic stressors impose impacts on marine biological and functional diversity will be critical to designing, allocating, and monitoring effective conservation actions at scale [13]. To assess the risk of adverse impacts on marine ecosystems, cumulative human impact models have been developed to integrate information on vulnerability of elements of ecological concern (e.g., habitats, species, functional groups) and exposure of those elements to anthropogenic stressors [14]. Habitat-based estimates of ecosystem vulnerability to (e.g., [15]) and impact from (e.g., [7,14,16]) various stressors rely upon an understanding of fundamental structural and functional similarity between, say, a Caribbean coral reef and an Indonesian coral reef, or a Californian kelp forest and a Namibian kelp forest ecosystem, though there may be little overlap in the species that inhabit each system. Such a habitat-based approach provides valuable insights on potential impacts to the general trophic structure and functioning of an ecosystem as well as the types of ecosystem services that may be affected, but it may miss important differences in vulnerability stemming from heterogeneity of species composition between otherwise similar marine communities. A species-based approach to estimating impacts may be better able to capture heterogeneity of species’ vulnerability to stressors [3,17–19], providing insights on impacts to individual species and community structure. Studies assessing anthropogenic impacts on marine species (e.g., [18,20–23]) generally have been limited in scope to a few select taxa and/or stressors, largely due to lack of a systematic means of estimating species vulnerability across multiple taxa and stressors [19,24]. O’Hara et al [3] parsed threat information from IUCN Red List assessments to estimate sensitivity and impacts from multiple anthropogenic stressors across multiple marine taxa globally over an eleven-year period, but that study was necessarily restricted to those species categorized as threatened and near-threatened, excluding the vast majority of marine species. However, a recently developed trait-based framework to estimate species vulnerability broadly across taxa and stressors [19] greatly expands the set of marine species available for such cumulative impact assessment. Additionally, a species-based approach to estimating risk of impact, coupled with information on traits associated with ecological function, enables identification of functionally unique species and groups whose loss may pose greater risk to ecosystem functioning and resilience than similar impacts on functionally redundant species [25–28]. Here we provide a taxonomically-diverse spatial analysis of cumulative human impacts of 13 anthropogenic stressors on 21,159 marine animal species and subpopulations. For each species/stressor combination, we intersect the species’ range with the spatial distribution of the stressor; impact in each pixel is modeled as the product of stressor intensity and the species’ estimated vulnerability to that stressor (See S1 Fig in S1 File. for conceptual overview of methods). We spatially aggregate species impact distributions to estimate mean impact across species and taxa, providing a taxonomically detailed understanding of how anthropogenic pressures impose risk of impact to marine biodiversity. We then compare these results to cumulative impact estimates based on representative habitats to determine where and how speciesvs. ecosystem-level vulnerability drives potential impact. Finally, we use a set of traits PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 2 / 20 Figshare. DOI: 10.6084/m9.figshare.26454106. A living version of the code and data can be found at https://github.com/mapping-marine-spp-vuln/spp_ vuln_mapping. Funding: CK, BSH: National Philanthropic Trust (https://www.nptrust.org/), SB200099 CCO: National Science Foundation via Belmont Forum (https://belmontforum.org/about), 2019902 No funders played any role in study design, data collection, analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. to classify species into functional entities (sensu Mouillot et al. [26]) and use these groupings to calculate a weighted mean impact, emphasizing impacts to species who uniquely represent a position in functional trait space. The species, functional entity, and habitat approaches to estimating cumulative impact provide different but synergistic lenses through which to estimate our impact on marine ecosystems: areas of agreement between these methods reinforce urgency for conservation, while areas of significant difference may provide conservation insights by highlighting impacts on vulnerable and functionally important species in otherwise resilient marine ecosystems. Methods Analysis grid All spatial analyses were calculated on a gridded global map using a Mollweide equal-area projection coordinate reference system (CRS), gridded to 10 km x 10 km resolution. See SI Methods in S1 File for additional details on preparing the analysis grid. Species distributions The 21,159 species (including subpopulations) considered in this assessment are limited to those animal species with data on spatial distribution as well as sufficient trait data to estimate vulnerability and assign species to functional entities. These species represent only a small subset of the >240,000 marine species identified in the World Register of Marine Species (WoRMS, [29]); however, this subset includes most known marine mammals, marine reptiles, seabirds, and cartilaginous fishes, as well as about half of marine bony fishes and warm-water corals (S1 Table in S1 File by class, S2 Table in S1 File vertebrates by order). Together these species represent most top predators, many mid-trophic species, and ecologically critical habitat-forming species. Relatively fewer other invertebrates were included, as most lacked spatial data, trait data, or both. Species distribution data were taken from AquaMaps [30] (n = 18,480) and IUCN species distribution maps [31,32] (n = 2,679). For species appearing in both distribution map datasets, the AquaMaps distribution maps, based on transparent and repeatable algorithms using publicly available data, were preferred over IUCN range maps, which integrate data and expert knowledge but may include mapping decisions that are difficult to replicate. For species represented by the AquaMaps dataset, presence was calculated as any 0.5˚ cell with a probability of occurrence of 0.5 or greater; the resulting cells were then reprojected to the 10 km Mollweide analysis grid. For species represented by IUCN Red List rangemaps, the polygons were reprojected and rasterized to the resolution and CRS of the analysis grid. See SI Methods in S1 File for additional details on preparing species distributions. See S2 Fig in S1 File. for a map of species richness generated from these species distributions. Vulnerability estimates Vulnerability weights, i.e., the relative effect of a given stressor on the fitness/health of a given species, were determined based on methods of Butt et al. [19]. Briefly, that study estimated vulnerability of species to each of a suite of stressors based on presence of certain traits that are likely to increase the species’ physiological sensitivity (e.g., calcium external structures indicate higher sensitivity to ocean acidification), ability to adapt to or avoid that specific stressor (e.g., high mobility makes it easier to avoid localized stressors), and life history and population-level traits that affect the population’s ability to adapt to or recover from disturbances in general (e.g., high fecundity suggests easier recovery from a disturbance). A binary exposure multiplier PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 3 / 20 (zero or one) prevents nonsensical results for certain stressors where exposure is limited to certain depths or ocean zones, e.g., ship strikes will not affect mesopelagic species. Trait values for species were gathered through expert elicitation and provided as ordinal or nominal categorical values. Vulnerability weights range from 0 (a stressor does not affect a species) to 1 (a stressor imposes extreme adverse effects on a species). See SI Methods in S1 File and S3 Table in S1 File for details on traits and calculations. Stressor layers for species-focused analysis For the species-focused analysis, the intensity of exposure to a stressor depends on the spatial distribution of the stressor relative to the spatial distribution of the species. Spatial data for stressors is typically available as gridded data of some physical quantity related to anthropogenic activity, e.g., brightness of nighttime lights, tonnes of nutrient fertilizer runoff, population density within 25 km of coast, or value of aragonite saturation state. For each stressor, a reference value was determined from the data (typically 99.9th percentile of observed values), a historic baseline (e.g., mean/standard deviation of sea surface temperature from 1985–2015), or ecologically relevant value (e.g., aragonite saturation state of 1) (S4 Table in S1 File). We calculated stressor distributions as a value from 0 (stressor not present) to 1 (stressor at reference point, indicating maximum intensity). For most of the included stressors, a single map of relative stressor intensity was created from gridded data and applied to all species, although vulnerability to the stressor varied by species. These stressors include sea surface temperature (SST) extremes, ocean acidification, ultraviolet radiation, sea level rise, nutrient pollution (runoff), direct human disturbance, light pollution, shipping (ship strikes), and habitat destruction driven by demersal destructive fishing and the footprint of benthic structures. However, there were also several stressors for which intensity (again ranging from 0 to 1) depends on species-specific information. These stressors include bycatch (dependent on water column position, i.e., benthic, pelagic, or both), biomass removal (dependent on catch that is directly targeting that species), and increase in mean SST (dependent on species thermal tolerance). For these stressors, individual maps were generated for each species (biomass removal, SST rise) or for each water-column position category (bycatch). See SI Methods in S1 File and S4 Table in S1 File for details on the data source, transformation, and reference point used for these stressor layers. Stressor layers for habitat-focused analysis The habitat-focused analysis was similar to that for species, with the intensity of exposure to a stressor depending on the spatial distribution of the stressor relative to the distribution of the habitat. For this approach, fisheries stressors were calculated using the same source as the species-level stressors, i.e., Watson [33], but aggregated by fishing gear, depth, and scale according to their effects on various habitat types as described in Halpern et al. [7]: commercial pelagic and demersal low bycatch, commercial pelagic high bycatch, commercial demersal destructive, and artisanal/small scale fishing. For SST extremes, ocean acidification, ultraviolet radiation, sea level rise, nutrient pollution (runoff), direct human disturbance, light pollution, shipping, benthic structures, and demersal destructive fishing, we used the identical stressor layers prepared for the species-level analysis. The species-specific stress of increasing mean SST relative to their thermal tolerance was omitted, as it would not be feasible to determine an analogous habitat-level thermal tolerance. See SI Methods in S1 File and S4 Table in S1 File for details on the data source, transformation, and reference point used for these stressor layers. PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 4 / 20 Cumulative human impacts: Species method Estimating impact at species level per grid cell. We modeled the impact on species iof stressor j21:Jin a given location (i.e., grid cell) as the product of stressor intensity s j and vulnerability of that species to that stressor v ij : Ij i¼vijsj Cumulative impact on species iin a given location was determined by summing impacts across all stressors (or subset, e.g., climate vs. non-climate stressors) in that location: Icml i¼X J j¼1 vijsj Note that this additive model does not account for compound effects of multiple stressors acting in combination, i.e., synergistic or antagonistic effects. Meta-analyses examining two-stressor interactions [34,35] have observed some non-additive stressor interactions, but additive effects were more commonly reported. Additionally, an additive model requires fewer assumptions, is conceptually tractable, and likely results in more conservative results. Estimating species-level mean cumulative impact across species range. For each species i, we calculated a cumulative impact score Xaccounting for impacts across its entire range as an average of per-grid-cell impacts for all cells cin the species’ range R i . For a single stressor j: Xj i¼1 RiX c2Ri Ij ic Cumulative impact scores across multiple stressors (climate, non-climate, and total) were determined as the sum of single-stressor impact scores. Estimating impact across species per grid cell. The species-mean method for calculating the impact score for stressor jin a given cell was determined by taking an unweighted mean across all Nspecies present (or a taxonomic subset, e.g., all elasmobranchs): Ij spp ¼1 NX N i¼1 vijsj and the cumulative impact is the sum of impacts across all (or a subset of) stressors within that cell. Cumulative human impacts: Habitat method To compare the results of our species-based cumulative impact approach to those of a habitatbased approach (e.g., [14]; [7]), we recreated habitat maps at the same resolution and projection as the species-based analysis, aggregating habitat presence maps at ~1 km resolution to determine proportional habitat representation within each 10 km grid cell. Using these habitat maps, we applied habitat vulnerability weights from Halpern et al. [7] to determine impacts based on largely the same stressor maps data sources used for the species-based assessment. To identify vulnerability of each habitat to various stressors we used the matrix of habitat vulnerability from Halpern et al. [7]. Per-grid-cell habitat impact scores for each stressor jwere created as the product of habitat vulnerability for each habitat h21: Hand intensity of stressor j, averaged over the proportional PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 5 / 20 inclusion of that habitat p h in a given cell: Ij hab ¼X H h¼1 phvhj Cumulative impact per pixel is the sum of habitat-based impacts across all (or subset) of stressors. Cumulative human impacts: Functional entity method Functional entities. To estimate cumulative impact on functional diversity, we first assigned species to functional entities based on categorical values of four traits (maximum body length, adult mobility, position in water column, and adult trophic level) that roughly determine a species’ ecological niche with regard to regulation of food webs and nutrient cycling, following Mouillot et al. [26]. Due to limited trait data available across a broad range of taxa, we relied on a smaller set of traits (those four noted previously) for assignment of functional entity than the six traits used in Mouillot et al. [26], resulting in fewer but more populous functional entities and therefore a more conservative estimate of functional vulnerability. Trait values were gleaned from [19,36,37]; missing values were imputed using Multiple Imputation by Chained Equation (MICE) in the R package mice [38] using all other traits plus fecundity (where available), generation time (where available), order, and family. See SI Methods in S1 File for details on the trait values used to assign functional entities, along with analyses to test sensitivity of functional vulnerability and cumulative impact to potential error in imputation of traits. Estimating impact at functional entity level per grid cell. For each functional entity k21: Kconsisting of some subset of species in a particular location, the impact of stressor jon the functional entity is the mean impact across all species in that functional entity in that location: Ij k¼1 NFE X NFE i¼1 Ij i Cumulative impact of all stressors on this functional entity in this location is the sum of impacts across all stressors (or a subset). Estimating impact across functional entities per grid cell. The functional entity method for calculating the impact score for stressor jin a given location was determined by taking a weighted mean across all Kfunctional entities present. Weighting for each functional entity was based on the functional vulnerability, sensu Mouillot et al. [26] with a slight modification (see below). Ij FE ¼1 PK k¼1FVkX K k¼1 FVkIj k Mouillot et al. [26] scored vulnerability of a functional entity as 1 if that entity was represented by a single species and 0 otherwise. Here we calculated functional vulnerability based on an inverse exponential of the number of species that represent that functional entity in that location, where functional vulnerability of entity kwas calculated as FVk¼1 2 �Nk1, accounting for low-membership entities but rapidly asymptotically approaching zero as membership increases. As for the species-based approach, the cumulative impact is the sum of impacts across all (or a subset of) stressors within that cell. PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 6 / 20 Code and packages All analysis was performed in R statistical software, version 4.0.4 [39], relying primarily on packages tidyverse [40], terra [41], sf [42], taxize [43,44], rfishbase [45]. Results Mean impacts on the assessed species varied dramatically across and within taxa (Fig 1A–1C). Comparing across taxa, mean risk of impact was highest for corals, followed by other invertebrate groups, driven in large part by higher vulnerability to increasing sea surface temperature and ocean acidification. Of vertebrate taxa, elasmobranchs were on average at greatest risk, driven by rising temperatures and high fishing pressure. For species and taxa whose range extended beyond the continental shelf, coastal impacts from non-climate stressors were generally higher than when assessed across their full range (i.e., suggesting relatively lower impacts away from the continental shelf), though coastal and full-range climate stressors were of similar magnitude. Average impact scores for vertebrate taxa fell below those of invertebrates, though there is considerable variation within each taxon, resulting in outliers: the top 1% of Fig 1. Cumulative productivity-weighted mean impacts on species ranges by taxon. (A) Distribution of mean predicted impact across species ranges (full range and coastal portions of range) from anthropogenic stressors by taxonomic group. (B) Distribution of mean cumulative impact from five climate stressors. (C) Distribution of mean cumulative impact from eight non-climate stressors. Vertical black line indicates mean across all species in taxon; white point indicates median. Boxes represent interquartile range (IQR, quartile Q1 to Q3); whiskers indicate observations 1.5x IQR below (above) Q1 (Q3) of box; outliers omitted from plot for clarity. (D) Contribution of individual stressors to mean cumulative impact across species ranges by taxon. Climate stressors outlined in red. https://doi.org/10.1371/journal.pone.0309788.g001 PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 7 / 20 species by impact score (n = 211 of 21,159) contain a disproportionately high number of elasmobranchs (n = 66, 31.3% of the highest impact subset), molluscs (n = 70, 33.2%), and corals (n = 13, 6.1%) relative to their representation in the overall dataset (5.3%, 13.8%, and 4.7% respectively). The bottom 1% of species by impact score also contain a disproportionate number of elasmobranchs (n = 46, 21.8% of the lowest impact subset), as well as polychaetes (n = 22, 10.4%) and ray-finned fishes (n = 140, 66.4%). Sea surface temperature rise (long-term trends) and extremes (short-term events, e.g., marine heat waves) were substantial contributors to impacts across all taxa (Fig 1D). Ocean acidification was a dominant stressor on all invertebrate species, but imposes little to no direct effect on vertebrates. Ultraviolet radiation impacted most invertebrates primarily due to vulnerability of planktonic larval life stages. Targeted fishing imposed the greatest risk of impact of all non-climate stressors on average, even considering that many taxa are not targeted and therefore not directly impacted (and thus are scored as zero in the calculation of mean impact). Fisheries bycatch impacted species broadly across all taxa. Spatial distribution of impacts is calculated as the mean impact from all stressors, or some subset of stressors, across all species present in each analysis cell. Cumulative impacts tend to be highest along coastlines, particularly heavily populated coastal zones where human populations generate more localized stressors while overlapping with higher magnitude of diffuse climate stressors (e.g., Northern European waters, the Mediterranean Sea, and the Indo-Pacific region, Fig 2A). Lower cumulative impacts were observed in remote areas where low fishing and shipping activity coincided with relatively low climate stressors (e.g., Southeastern Australia, equatorial Eastern Pacific, Southern Atlantic, Northern Atlantic near Greenland). The very lowest cumulative impacts were observed in areas dominated by permanent or seasonal sea ice (Arctic Ocean, coastal Antarctica). Climate stressors broadly impact species across coastal and open ocean regions, with highest values in tropical Indo-Pacific waters, the Caribbean Sea, and northern Pacific and Atlantic (Fig 2B). Non-climate stressors are predominantly driven by fisheries stressors (targeted fishing and bycatch), particularly along coastlines and international waters just beyond the border of national Exclusive Economic Zones (Fig 2C). Examining modeled impacts by quartile, rather than magnitude, allows for comparison of where particularly high impacts from climate stressors (which cannot be mitigated in the short run, but may benefit from adaptation) overlap with high impacts from non-climate stressors (which can be effectively mitigated through actions such as marine protected areas or fisheries management). Reclassifying impacts across these two aggregated categories into quartiles, based on global values, reveals 10.3% of ocean area where climate impact hot spots (i.e., spatial cells in the top quartile of global aggregated impact within the category) overlap with hot spots of non-climate impacts (e.g., Southeast Asia, East China Sea, Gulf of Mexico, Caribbean Sea, international tropical Pacific waters, Barents Sea, Bering Sea, Fig 2D), and 8.8% of area where cool spots (i.e., bottom quartile) of climate and non-climate impacts overlap (e.g., Southern portions of Pacific, Atlantic, and Indian oceans; Weddell Sea; national waters for several southern Pacific and Atlantic small island states, Fig 2D). These hot spots and cool spots represent the two most common instances of impact quartile overlap; conversely, high climate/low nonclimate (2.9%) and low climate/high non-climate (3.1%) represent the rarest instances of overlap (S3A Fig in S1 File). Patterns of predicted impacts based on the species method vs. representative habitats method (Fig 3; see S3B and S4 Figs in S1 File. for habitat analog to Fig 2 and S3A Fig in S1 File. respectively) show some clear differences, driven by variations in vulnerability across species and ecosystem type, despite the underlying stressor distributions being largely the same. Comparing overlap between habitat and species methods of climate impacts (Fig 3A) reveals areas where the two methods disagree (Sørensen similarity index 61%), indicating the species PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 8 / 20 method predicts greater risk from climate change in equatorial Indian and Indo-Pacific waters (purple tones) relative to the habitat method, largely due to the inclusion of species-specific impacts as sea surface temperature rises relative to species thermal tolerance. Results based on Fig 2. Distribution of modeled risk of impact based on species-specific vulnerability and exposure to anthropogenic stressors. (A) Mean cumulative impact across all species, summing across all stressors. (B) Mean cumulative impact across all species, summing across all climate-related stressors. (C) Mean cumulative impact across all species, summing across all non-climate stressors. (D) Bivariate comparison of distributions of climate impacts (orange) vs. non-climate impacts (purple) by quartile within each stressor group. https://doi.org/10.1371/journal.pone.0309788.g002 PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 9 / 20 necessitating conservation action to mitigate non-climate stressors (e.g., reduction of landbased runoff) to allow for improved ecosystem resilience to climate change [55,56]. While well-enforced no-take marine protected areas are an effective conservation tool that can provide multiple co-benefits [55,71–73], in certain cases, sustainable-use marine protected areas with targeted exclusions may provide substantial ecological benefit at lower social cost [74]. Consideration of human impacts across lenses of species, function, and habitat provides a richer understanding of marine ecosystems, and highlights that impacts in species-rich coastal regions may pose greater risk to biodiversity than indicated from habitat-based methods alone. Our data can be used with socioeconomic information to help prioritize effective, economically efficient, and socially equitable conservation actions to best benefit nature and people. Supporting information S1 File. (PDF) Acknowledgments We thank the National Center for Ecological Analysis and Synthesis (NCEAS) for computational support. Author Contributions Conceptualization: Casey C. O’Hara, Melanie Frazier, Nathalie Butt, Carissa Klein, Benjamin S. Halpern. Data curation: Casey C. O’Hara. Formal analysis: Casey C. O’Hara. Funding acquisition: Carissa Klein, Benjamin S. Halpern. Methodology: Casey C. O’Hara, Melanie Frazier, Mireia Valle, Nathalie Butt, Carissa Klein, Benjamin S. Halpern. Software: Casey C. O’Hara. Supervision: Carissa Klein, Benjamin S. Halpern. Visualization: Casey C. O’Hara. Writing – original draft: Casey C. O’Hara. Writing – review & editing: Casey C. O’Hara, Melanie Frazier, Mireia Valle, Nathalie Butt, Kristin Kaschner, Carissa Klein, Benjamin S. Halpern. References 1. He Q, Silliman BR. Climate Change, Human Impacts, and Coastal Ecosystems in the Anthropocene. Current Biology. 2019; 29: R1021–R1035. https://doi.org/10.1016/j.cub.2019.08.042 PMID: 31593661 2. Andrello M, Darling ES, Wenger A, Sua ´rez-Castro AF, Gelfand S, Ahmadia GN. A global map of human pressures on tropical coral reefs. Conservation Letters. 2022; 15: e12858. https://doi.org/10.1111/conl. 12858 3. O’Hara CC, Frazier M, Halpern BS. At-risk marine biodiversity faces extensive, expanding, and intensifying human impacts. Science. 2021; 372: 84–87. https://doi.org/10.1126/science.abe6731 PMID: 33795456 PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 16 / 20 4. Rocha J, Yletyinen J, Biggs R, Blenckner T, Peterson G. Marine regime shifts: Drivers and impacts on ecosystems services. Philosophical Transactions of the Royal Society B: Biological Sciences. 2015; 370: 20130273. https://doi.org/10.1098/rstb.2013.0273 5. Boyce DG, Tittensor DP, Garilao C, Henson S, Kaschner K, Kesner-Reyes K, et al. A climate risk index for marine life. Nature Climate Change. 2022; 12: 854–862. https://doi.org/10.1038/s41558-022-01437y 6. O’Hara CC, Halpern BS. Anticipating the Future of the World’s Ocean. Annual Review of Environment and Resources. 2022; 47: annurev-environ-120120-053645. https://doi.org/10.1146/annurev-environ120120-053645 7. Halpern BS, Frazier M, Afflerbach J, Lowndes JS, Micheli F, O’Hara C, et al. Recent pace of change in human impact on the world’s ocean. Scientific Reports. 2019; 9: 1–8. https://doi.org/10.1038/s41598019-47201-9 PMID: 31406130 8. Georgian S, Hameed S, Morgan L, Amon DJ, Sumaila UR, Johns D, et al. Scientists’ warning of an imperiled ocean. Biological Conservation. 2022; 272: 109595. https://doi.org/10.1016/j.biocon.2022. 109595 9. Dı ´az S, Settele J, Brondı ´zio ES, Ngo HT, Agard J, Arneth A, et al. Pervasive human-driven decline of life on Earth points to the need for transformative change. Science. 2019;366. https://doi.org/10.1126/ science.aax3100 PMID: 31831642 10. Dı ´az S, Settele J, Brondı ´zio ES, Ngo HT, Guèze M, Agard J, et al. Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental SciencePolicy Platform on Biodiversity and Ecosystem Services. Bonn, Germany: IPBES secretariat; 2019 p. 60. 11. Eddy T, Lam V, Reygondeau G, Cisneros-Montemayor A, Greer K, Palomares M, et al. Global decline in capacity of coral reefs to provide ecosystem services. One Earth. 2021; 4. https://doi.org/10.1016/j. oneear.2021.08.016 12. UN CBD. Kunming-Montreal Global biodiversity framework Draft decision submitted by the President. Montreal, Canada: United Nations Environment Programme Convention on Biological Diversity; 2022 Dec. 13. Borja A, Elliott M, Teixeira H, Stelzenmu¨ller V, Katsanevakis S, Coll M, et al. Addressing the cumulative impacts of multiple human pressures in marine systems, for the sustainable use of the seas. Frontiers in Ocean Sustainability. 2024; 1: 1308125. https://doi.org/10.3389/focsu.2023.1308125 14. Halpern BS, Walbridge S, Selkoe KA, Kappel CV, Micheli F, D’Agrosa C, et al. A Global Map of Human Impact on Marine Ecosystems. Science. 2008; 319: 948–952. https://doi.org/10.1126/science.1149345 PMID: 18276889 15. Halpern BS, Selkoe KA, Micheli F, Kappel CV. Evaluating and Ranking the Vulnerability of Global Marine Ecosystems to Anthropogenic Threats. Conservation Biology. 2007; 21: 1301–1315. https://doi. org/10.1111/j.1523-1739.2007.00752.x PMID: 17883495 16. Hammar L, Molander S, Pålsson J, Schmidtbauer Crona J, Carneiro G, Johansson T, et al. Cumulative impact assessment for ecosystem-based marine spatial planning. Science of The Total Environment. 2020; 734: 139024. https://doi.org/10.1016/j.scitotenv.2020.139024 PMID: 32464374 17. Jones MC, Cheung WWL. Using fuzzy logic to determine the vulnerability of marine species to climate change. Global Change Biology. 2018; 24: e719–e731. https://doi.org/10.1111/gcb.13869 PMID: 28948655 18. Avila IC, Kaschner K, Dormann CF. Current global risks to marine mammals: Taking stock of the threats. Biological Conservation. 2018; 221: 44–58. https://doi.org/10.1016/j.biocon.2018.02.021 19. Butt N, Halpern BS, O’Hara CC, Allcock AL, Polidoro B, Sherman S, et al. A trait-based framework for assessing the vulnerability of marine species to human impacts. Ecosphere. 2022; 13: e3919. https:// doi.org/10.1002/ecs2.3919 20. Maxwell SM, Hazen EL, Bograd SJ, Halpern BS, Breed GA, Nickel B, et al. Cumulative human impacts on marine predators. Nature Communications. 2013; 4: 1–9. https://doi.org/10.1038/ncomms3688 PMID: 24162104 21. Queiroz N, Humphries NE, Couto A, Vedor M, da Costa I, Sequeira AMM, et al. Global spatial risk assessment of sharks under the footprint of fisheries. Nature. 2019; 572: 461–466. https://doi.org/10. 1038/s41586-019-1444-4 PMID: 31340216 22. Davidson AD, Boyer AG, Kim H, Pompa-Mansilla S, Hamilton MJ, Costa DP, et al. Drivers and hotspots of extinction risk in marine mammals. Proceedings of the National Academy of Sciences. 2012; 109: 3395–3400. https://doi.org/10.1073/pnas.1121469109 PMID: 22308490 23. Witt MJ, Augowet Bonguno E, Broderick AC, Coyne MS, Formia A, Gibudi A, et al. Tracking leatherback turtles from the world’s largest rookery: Assessing threats across the South Atlantic. Proceedings of the PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 17 / 20 Royal Society B: Biological Sciences. 2011; 278: 2338–2347. https://doi.org/10.1098/rspb.2010.2467 PMID: 21208949 24. Auber A, Waldock C, Maire A, Goberville E, Albouy C, Algar AC, et al. A functional vulnerability framework for biodiversity conservation. Nature Communications. 2022; 13: 4774. https://doi.org/10.1038/ s41467-022-32331-y PMID: 36050297 25. Naeem S. Species Redundancy and Ecosystem Reliability. Conservation Biology. 1998; 12: 39–45. https://doi.org/10.1111/j.1523-1739.1998.96379.x 26. Mouillot D, Ville ´ger S, Parravicini V, Kulbicki M, Arias-Gonza ´lez JE, Bender M, et al. Functional overredundancy and high functional vulnerability in global fish faunas on tropical reefs. Proceedings of the National Academy of Sciences. 2014; 111: 13757–13762. https://doi.org/10.1073/pnas.1317625111 PMID: 25225388 27. Micheli F, Halpern BS. Low functional redundancy in coastal marine assemblages. Ecology Letters. 2005; 8: 391–400. https://doi.org/10.1111/j.1461-0248.2005.00731.x 28. Griffin JN, Leprieur F, Silvestro D, Lefcheck JS, Albouy C, Rasher DB, et al. Functionally unique, specialised, and endangered (FUSE) species: Towards integrated metrics for the conservation prioritisation toolbox. Cold Spring Harbor Laboratory; 2020. p. 2020.05.09.084871. https://doi.org/10.1101/2020.05. 09.084871 29. Vandepitte L, Vanhoorne B, Decock W, Vranken S, Lanssens T, Dekeyzer S, et al. A decade of the World Register of Marine Species–General insights and experiences from the Data Management Team: Where are we, what have we learned and how can we continue? PLoS ONE. 2018; 13: e0194599. https://doi.org/10.1371/journal.pone.0194599 PMID: 29624577 30. Kaschner K, Rius-Barile J, Kesner-Reyes K, Garilao C, Kullander S, Rees T, et al. AquaMaps: Predicted range maps for aquatic species. www.aquamaps.org; 2019. 31. IUCN. The IUCN Red List of Threatened Species. Version 2021–3. http://www.iucnredlist.org; 2021. 32. BirdLife International and Handbook of the Birds of the World. Bird species distribution maps of the world. Version 7.0. BirdLife International, Cambridge, UK and NatureServe, Arlington, USA; 2018. 33. Watson RA. A database of global marine commercial, small-scale, illegal and unreported fisheries catch 1950–2014. Scientific Data. 2017; 4: 170039. https://doi.org/10.1038/sdata.2017.39 PMID: 28398351 34. Crain CM, Kroeker K, Halpern BS. Interactive and cumulative effects of multiple human stressors in marine systems. Ecology Letters. 2008; 11: 1304–1315. https://doi.org/10.1111/j.1461-0248.2008. 01253.x PMID: 19046359 35. Stockbridge J, Jones AR, Gillanders BM. A meta-analysis of multiple stressors on seagrasses in the context of marine spatial cumulative impacts assessment. Scientific Reports. 2020; 10: 11934. https:// doi.org/10.1038/s41598-020-68801-w PMID: 32686719 36. Froese R, Pauly D. FishBase. FishBase. www.fishbase.org, version (06/2022); 2022. 37. Palomares ML‘Deng’, Pauly D. SeaLifeBase. World Wide Web electronic publication. www.sealifebase. org, version (04/2022); 2022. 38. van Buuren S, Groothuis-Oudshoorn K. mice: Multivariate imputation by chained equations in R. Journal of Statistical Software. 2011; 45: 1–67. https://doi.org/10.18637/jss.v045.i03 39. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; R Foundation for Statistical Computing; 2022. 40. Wickham H, Averick M, Bryan J, Chang W, McGowan LD, Franc¸ois R, et al. Welcome to the tidyverse. Journal of Open Source Software. 2019; 4: 1686. https://doi.org/10.21105/joss.01686 41. Hijmans RJ. Terra: Spatial data analysis. 2022. 42. Pebesma E. Simple features for R: Standardized support for spatial vector data. The R Journal. 2018; 10: 439–446. https://doi.org/10.32614/RJ-2018-009 43. Chamberlain S, Szocs E. Taxize—taxonomic search and retrieval in R. F1000Research. 2013. https:// doi.org/10.12688/f1000research.2-191.v2 PMID: 24555091 44. Chamberlain S, Szoecs E, Foster Z, Arendsee Z, Boettiger C, Ram K, et al. Taxize: Taxonomic information from around the web. R package version 0.9.98. 2020. 45. Boettiger C, Temple Lang D, Wainwright P. Rfishbase: Exploring, manipulating and visualizing FishBase data from R. Journal of Fish Biology. 2012. https://doi.org/10.1111/j.1095-8649.2012.03464.x PMID: 23130696 46. Spalding MD, Fox HE, Allen GR, Davidson N, Ferdaña ZA, Finlayson MAX, et al. Marine ecoregions of the world: A bioregionalization of coastal and shelf areas. BioScience. 2007; 57: 573–583. PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 18 / 20 47. Williams R, Grand J, Hooker SK, Buckland ST, Reeves RR, Rojas-Bracho L, et al. Prioritizing global marine mammal habitats using density maps in place of range maps. Ecography. 2014; 37: 212–220. https://doi.org/10.1111/j.1600-0587.2013.00479.x 48. Selig ER, Turner WR, Troe ¨ng S, Wallace BP, Halpern BS, Kaschner K, et al. Global Priorities for Marine Biodiversity Conservation. Rotjan RD, editor. PLoS ONE. 2014; 9: e82898. https://doi.org/10.1371/ journal.pone.0082898 PMID: 24416151 49. O’Hara CC, Villaseñor-Derbez JC, Ralph GM, Halpern BS. Mapping status and conservation of global at-risk marine biodiversity. Conservation Letters. 2019; e12651. https://doi.org/10.1111/conl.12651 50. Klein CJ, Brown CJ, Halpern BS, Segan DB, McGowan J, Beger M, et al. Shortfalls in the global protected area network at representing marine biodiversity. Scientific Reports. 2015; 5: 17539. https://doi. org/10.1038/srep17539 PMID: 26631984 51. Harris NC, Ga ´mez S, Gadsden GI, Malhotra R. Textured species range maps enhance interdisciplinary science capacity across scales. Frontiers in Ecology and the Environment. 2022; 20: 319–326. https:// doi.org/10.1002/fee.2474 52. Henson SA, Beaulieu C, Ilyina T, John JG, Long M, Se ´fe ´rian R, et al. Rapid emergence of climate change in environmental drivers of marine ecosystems. Nature Communications. 2017; 8: 14682. https://doi.org/10.1038/ncomms14682 PMID: 28267144 53. Wåhlstro ¨m I, Hammar L, Hume D, Pålsson J, Almroth-Rosell E, Dieterich C, et al. Projected climate change impact on a coastal sea—As significant as all current pressures combined. Global Change Biology. 2022; 28: 5310–5319. https://doi.org/10.1111/gcb.16312 PMID: 35733273 54. IPCC. Summary for Policymakers. In: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Masson-Delmotte V, Zhai P, Pirani A, Connors SL, Pe ´an C, Berger S, et al., editors. Cambridge University Press; 2021. 55. Sala E, Giakoumi S. No-take marine reserves are the most effective protected areas in the ocean. ICES Journal of Marine Science. 2018; 75: 1166–1168. https://doi.org/10.1093/icesjms/fsx059 56. Roberts CM, O’Leary BC, McCauley DJ, Cury PM, Duarte CM, Lubchenco J, et al. Marine reserves can mitigate and promote adaptation to climate change. Proceedings of the National Academy of Sciences. 2017; 114: 6167–6175. https://doi.org/10.1073/pnas.1701262114 PMID: 28584096 57. Medoff S, Lynham J, Raynor J. Spillover benefits from the world’s largest fully protected MPA. Science. 2022; 378: 313–316. https://doi.org/10.1126/science.abn0098 PMID: 36264800 58. Saunders MI, Bode M, Atkinson S, Klein CJ, Metaxas A, Beher J, et al. Simple rules can guide whether landor ocean-based conservation will best benefit marine ecosystems. PLOS Biology. 2017; 15: e2001886. https://doi.org/10.1371/journal.pbio.2001886 PMID: 28877168 59. Klein CJ, Ban NC, Halpern BS, Beger M, Game ET, Grantham HS, et al. Prioritizing Land and Sea Conservation Investments to Protect Coral Reefs. PLOS ONE. 2010; 5: e12431. https://doi.org/10.1371/ journal.pone.0012431 PMID: 20814570 60. Violle C, Thuiller W, Mouquet N, Munoz F, Kraft NJB, Cadotte MW, et al. Functional Rarity: The Ecology of Outliers. Trends in Ecology & Evolution. 2017; 32: 356–367. https://doi.org/10.1016/j.tree.2017.02. 002 PMID: 28389103 61. Legras G, Loiseau N, Gaertner J-C. Functional richness: Overview of indices and underlying concepts. Acta Oecologica. 2018; 87: 34–44. https://doi.org/10.1016/j.actao.2018.02.007 62. Streit RP, Bellwood DR. To harness traits for ecology, let’s abandon “functionality.” Trends in Ecology & Evolution. 2023; 38: 402–411. https://doi.org/10.1016/j.tree.2022.11.009 PMID: 36522192 63. Halpern BS, Fujita R. Assumptions, challenges, and future directions in cumulative impact analysis. Ecosphere. 2013; 4: art131. https://doi.org/10.1890/ES13-00181.1 64. Hunsicker ME, Kappel CV, Selkoe KA, Halpern BS, Scarborough C, Mease L, et al. Characterizing driver–response relationships in marine pelagic ecosystems for improved ocean management. Ecological Applications. 2016; 26: 651–663. https://doi.org/10.1890/14-2200 PMID: 27411240 65. Kimmel K, Clark M, Tilman D. Impact of multiple small and persistent threats on extinction risk. Conservation Biology. 2022; 36: e13901. https://doi.org/10.1111/cobi.13901 PMID: 35212024 66. Wyborn C, Evans MC. Conservation needs to break free from global priority mapping. Nature Ecology & Evolution. 2021; 1–3. https://doi.org/10.1038/s41559-021-01540-x PMID: 34426678 67. Chaplin-Kramer R, Brauman KA, Cavender-Bares J, Dı ´az S, Duarte GT, Enquist BJ, et al. Conservation needs to integrate knowledge across scales. Nature Ecology & Evolution. 2022; 6: 118–119. https://doi. org/10.1038/s41559-021-01605-x PMID: 34824390 68. Pinsky ML, Selden RL, Kitchel ZJ. Climate-Driven Shifts in Marine Species Ranges: Scaling from Organisms to Communities. Annual Review of Marine Science. 2020; 12: 153–179. https://doi.org/10. 1146/annurev-marine-010419-010916 PMID: 31505130 PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 19 / 20 69. Garcı ´a Molinos J, Halpern BS, Schoeman DS, Brown CJ, Kiessling W, Moore PJ, et al. Climate velocity and the future global redistribution of marine biodiversity. Nature Climate Change. 2015; 6: 83–88. https://doi.org/10.1038/nclimate2769 70. Brown SC, Mellin C, Garcı ´a Molinos J, Lorenzen ED, Fordham DA. Faster ocean warming threatens richest areas of marine biodiversity. Global Change Biology. 2022; 1–10. https://doi.org/10.1111/gcb. 16328 PMID: 35795987 71. Lester SE, Halpern BS. Biological responses in marine no-take reserves versus partially protected areas. Marine Ecology Progress Series. 2008; 367: 49–56. https://doi.org/10.3354/meps07599 72. Edgar GJ, Stuart-Smith RD, Willis TJ, Kininmonth S, Baker SC, Banks S, et al. Global conservation outcomes depend on marine protected areas with five key features. Nature. 2014; 506: 216–220. https:// doi.org/10.1038/nature13022 PMID: 24499817 73. Sala E, Mayorga J, Bradley D, Cabral RB, Atwood TB, Auber A, et al. Protecting the global ocean for biodiversity, food and climate. Nature. 2021; 1–6. https://doi.org/10.1038/s41586-021-03371-z PMID: 33731930 74. Viana D, Gill D, Zvoleff A, Krueck N, Zamborain-Mason J, Free C, et al. Sustainable-use marine protected areas to improve human nutrition. In Review; 2022 Jul. https://doi.org/10.21203/rs.3.rs-1765829/ v1 PLOS ONE Cumulative human impacts on global marine biodiversity PLOS ONE | https://doi.org/10.1371/journal.pone.0309788 September 18, 2024 20 / 20