About

Patrick N. Halpin is a Professor of Marine Geospatial Ecology in the Marine Science and Conservation Division of the Nicholas School of the Environment at Duke University. He leads the Marine Geospatial Ecology Lab, which has facilities at both the main campus of Duke University and the Duke University Marine Lab. Prof. Halpin received his Ph.D. in Environmental Sciences from the University of Virginia in 1995. His research focuses on marine geospatial analysis, ecological applications of geographic information systems and remote sensing, as well as marine conservation and ecosystem-based management. He serves on several international scientific and conservation program steering committees, including the Ocean Biogeographic Information System (OBIS), the Global Oceans Biodiversity Initiative (GOBI), the Marine Working Group for the Group on Earth Observations - Biodiversity Observing Networks (GEO-BON), and the Google Oceans Advisory Council.

Research topics

• Oceanography
• Environmental science
• Geography
• Political Science
• Geology
• Ecology
• Environmental resource management
• Archaeology
• Law
• History
• Business
• Environmental planning
• Fishery

Selected publications

• Models Incorporating Non‐Stationarity Improve Detection of Climate‐Driven Range Shifts in Odontocetes

  Diversity and Distributions · 2026-02-01 · 1 citations

  articleOpen access

  ABSTRACT Aim Climate change is causing distributional shifts in many species globally. identifying and anticipating these shifts is critical to understanding ecosystem impacts and implementing successful management strategies. species distribution models ( SDMs ) are useful tools often employed to describe current and changing habitat use, particularly for marine predators. However, most SDMs assume the statistical relationships between species and their environment are temporally static, which may not be true. We examined how incorporating temporal variability improved SDM performance and estimated range shifts for six Odontocete species. We used a high performing model to quantify changes in Odontocete distribution over a 24‐year period. Location Waters of the United States, east coast, from Florida to Nova Scotia. Methods We utilised nearly 1.4 million kilometres of line transect survey data collected from 1997 to 2020 along the East Coast of the United States to evaluate changes in the distribution of six Odontocete species. We assessed six model specifications of generalise additive models that varied in the extent of temporal and spatial variability incorporated. Results We found that the best performing model specifications included temporally dynamic species–environment relationships and temporally dynamic spatial terms. These model specifications identified significant poleward range shifts in all species for which we had sufficient data across their range. In contrast, model specifications which only included static terms performed poorly and identified limited or no spatial shifts. Main Conclusions These results advance our predictive capabilities from static species–environment relationships for marine predators and demonstrate the importance of carefully considering assumptions and model specifications when modelling changes to distributions. The odontocete range shifts we identified are likely to have substantial ecosystem impacts, and the framework we present offers a diagnostic approach for modelling and identifying range shifts in other wide‐ranging species.

  PublisherDOI

• Ecological Insights and Management Implications of the Global Migratory Connectivity of Green Turtles

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-09-16 · 1 citations

  preprintOpen access

  ABSTRACT Aim Green turtles are a widely distributed and highly migratory species; despite extensive data on their movement, there is no species‐specific global synthesis on the subject. Based on three decades of published literature and building on previous global analyses, we developed a global network model of migratory connectivity for green turtles to better understand their spatial biology. Location Global. Time Period 1990–2022. Major Taxa Studied Green Sea Turtle ( Chelonia mydas ). Methods We conducted a structured literature review extracting georeferenced information on the movement of green turtles from 1990 to 2022, aggregating this information into a single connectivity model, defining nodes of connectivity. We evaluated connectivity routes from nesting areas to foraging sites for each RMU, identifying trajectories moving outside and across the boundaries of these areas. Results We found a total of 113 sources informing migratory connectivity globally. We identified 474 sites, representing locations where green turtles were observed (124 being nesting sites). Migratory connections, derived from both long‐term (≥ 1 year) and short‐term (< 12 months) tracking data, ranged from resident turtles that never left their nesting sites to rookeries connected to as many as 20 different locations, some over 5000 km apart. This long‐distance connectivity exposes populations to threats across disparate locations. Most connections traversed national jurisdictions, including crossing different Regional Management Units. Main Conclusions We compiled the largest available dataset describing movement of green turtles worldwide and present the most comprehensive global model of their migratory connectivity. This model provides ecological insights into regional differences in life histories, identifies geographic and demographic gaps in sampling, and provides baseline information on connectivity to support transboundary management of green turtle populations. The study reiterates the need for larger collaborative efforts to aggregate knowledge beyond local jurisdictions, to inform and align effective management measures to protect this historically threatened species.

  PublisherOA PDFDOI

• Marine megavertebrate migrations connect the global ocean

  Nature Communications · 2025-05-08 · 6 citations

  articleOpen access

  Animal migrations are extensive, ubiquitous, and in decline. To effectively protect migratory species, it is often crucial to identify the interconnected sets of sites they rely upon. Gaps between primary ecological research and synthesised information that is useful to policymakers has limited effective conservation of long-distance migrants, particularly in the marine realm. By synthesising 1304 references to identify 1787 sites and develop model migratory networks for 109 species, we show the minimum extent of marine megafauna connectivity across the global oceans. Our analyses underscore the importance of transboundary cooperation for migratory species conservation at scales larger than current regional structures afford and provide a free online system that will enable policymakers to efficiently summarise how marine migrants use and connect their jurisdictions. Animal migrations are extensive and crucial for ecosystem health but are in decline. This study identifies 1,787 sites and links among them for 109 marine species, highlighting the need for international cooperation and providing policymakers with essential knowledge for effective conservation.

  PublisherOA PDFDOI

• Leveraging built marine structures to benefit and minimize impacts on natural habitats

  BioScience · 2025-02-01 · 15 citations

  articleOpen access

  Abstract Many natural marine habitats are decreasing in extent despite global conservation and restoration efforts. In contrast, built marine structures, such as hardened shorelines, offshore energy and aquaculture infrastructure, and artificial reefs, are increasing in extent—and, in some locations, represent over 80% of nearshore, structured habitat. When introduced into the seascape, built marine structures inevitably interact with natural habitats, but these structures are not typically designed to support natural systems. This approach often results in overall harm to natural systems, further impeding marine conservation goals. However, there is growing recognition within the ocean management and engineering community that built marine structures can be strategically designed to minimize their negative impacts and potentially support ecosystems and associated biota. We synthesize the best available science and provide bright spot examples of how leveraging built marine structures to mimic or facilitate natural habitats can help recover biodiversity, augment ecosystem services, and rehabilitate degraded habitats, providing positive outcomes for people and nature in a changing climate. Despite these bright spots, we caution that built structures typically have overall negative environmental consequences for natural habitats and should not be used in lieu of conventional habitat restoration or conservation or to justify the destruction of natural habitats.

  PublisherOA PDFDOI

• Expected occurrence of wildlife in US Atlantic offshore wind areas

  Frontiers in Marine Science · 2025-06-18 · 2 citations

  articleOpen accessSenior author

  Introduction Offshore wind energy has entered a pivotal phase of development for the U.S. Atlantic Outer Continental Shelf (OCS), a region that supports critical habitats, migratory corridors and flyways for many marine species. Assessing where and when marine wildlife occurs is a crucial first step in developing a risk assessment framework to evaluate potential risks and impacts of offshore wind development. Methods In this study, we perform this initial assessment by evaluating the expected occurrence of marine mammal, seabird and sea turtle taxa in areas of interest to identify patterns and potential areas of concern. Specifically, this work depicts the expected monthly density of 84 marine species and taxa within each of the 29 active wind energy lease areas plus a 10 km buffer to account for nearby activity. We then compare these densities to subregional thresholds, evaluated as the 90th percentile of the subregion’s monthly density, to provide comparisons across the shelf region. Results This analysis synthesizes the most recent spatial distribution models of 31 marine mammal taxa (26 species and 5 guilds), 49 seabird species and 4 sea turtle species to provide a unified evaluation of the major marine wildlife in the region. Out of the 84 species and taxa analyzed, 56 exhibit levels of expected density in wind energy areas that exceed the corresponding 90th percentile subregional threshold at some point throughout the year. Discussion These results represent an initial assessment in the broader Occurrence, Exposure, Response, and Consequence (OERC) framework, originally developed by the U.S. Navy for marine species risk assessments. These results offer valuable guidance to marine spatial planners, management agencies and offshore wind developers on the expected locations and timing of interaction risk to wildlife species in or near wind energy areas across the region.

  PublisherOA PDFDOI

• Correction: Expected occurrence of wildlife in US Atlantic offshore wind areas

  Frontiers in Marine Science · 2025-08-21

  articleOpen accessSenior author

  Amidst advancements in offshore wind (OSW) on the Atlantic Outer Continental Shelf (OCS) of the U.S., driven by the growing demand for renewable energy infrastructure, it remains crucial to evaluate the potential impacts on the wildlife sharing these habitats (Sinclair et al., 2018;Copping et al., 2020;Galparsoro et al., 2022;Methratta et al., 2023;Congressional Research Service, 2024). With an ambitious goal of reaching 30 gigawatts of OSW energy capacity by 2030, more than 50 projects have been proposed across US Atlantic, Pacific and Gulf of America waters, and over 30 leases issued in the Atlantic OCS alone (Croll et al., 2022;Methratta et al., 2023;Congressional Given the complexity of the potential impacts of OSW, a more nuanced, systematic approach to risk assessments is crucial to help integrate the spatiotemporal variations across species and habitats. One approach for compound evaluations is the "Occurrence, Exposure, Response, Consequence" (OERC) framework, originally adapted by the U.S. Navy for their Integrated Comprehensive Monitoring Program for marine species. This framework provides a systematic progression, starting with the foundational understanding of species occurrence and then moving to the more complex aspects of exposure, response, and consequence, ultimately leading to an assessment of potential risk (Bell, 2013). Each element of the OERC framework, described in detail in figure 1, can be evaluated quantitatively, as retrospective assessments, models, or simulations. By organizing information in a ladder structure, the OERC framework accommodates data from diverse sources and can integrate cumulative impacts. The focus of our study on species occurrence serves as the critical first step in the OERC framework, providing essential baseline knowledge of temporal and spatial habitat use patterns and lays the groundwork for subsequent assessments.Here, we employ widely used Spatial Distribution Models (SDMs) for 84 marine life taxa to evaluate their occurrence within Wind Energy Lease Areas (WELAs) and five subregions along the U.S. Atlantic coast to identify spatiotemporal overlaps of marine life and offshore wind activities. This analysis aims to better understand the species, time periods, and areas that may be most vulnerable to potential OSW impacts, helping inform prioritization of mitigation and monitoring efforts. For each of the 29 active Atlantic WELAs, we calculated the average monthly predicted densities for 13 marine mammal taxa (12 species and one guild), 49 seabird species and four sea turtle species. We also calculated average annual density for 18 additional marine mammal taxa (14 species and four guilds) for which only year-round average predictions were available. To provide a more focused and representative subset for this assessment and enable clearer interpretation of risk across multiple taxa, we concentrated more closely on eleven focal species, chosen to capture a range of spatiotemporal patterns, behaviors, conservation statuses and ecologies. The focal species include four marine mammal, three avian, and four sea turtle species.The marine mammal focal species are fin whales (Balaenoptera physalus), humpback whales (Megaptera novaengliae), minke whales (Balaenoptera acutorostrata), and North Atlantic right whales (Eubalaena glacialis). North Atlantic right whales, listed as "Critically Endangered" on the IUCN Red List (Cook, 2020), have been a primary focus of monitoring and mitigation in the region. While minke and humpback whales (Northwest Atlantic population) are currently listed as "Least Concern" (Cook, 2018a(Cook, , 2018c)), both species, as well as North Atlantic right whales, are currently experiencing an unusual mortality event in the region, as defined by the Marine Mammal Protection Act (MMPA 16 U.S.C. § 1421h(9)), highlighting the importance of continued study (Silber et al., 2023;Stepanuk et al., 2023;Congressional Research Service, 2024). Fin whales are listed as "Vulnerable" and have limited migratory movement making them more sensitive to displacement risk (Cook, 2018b;Davis et al., 2020).The avian species selected -Red-throated Loons (Gavia stellata), Northern Gannets (Morus bassanus), and Great Black-backed Gulls (Larus marinus) -were chosen due to their high sensitivity to collision and displacement, and habitat use patterns (Willmott, Forcey and Kent, 2013;Heinänen et al., 2020;Peschko et al., 2021;Fauchald et al., 2024). While all three bird species are listed as "Least Concern" on the IUCN Red List (BirdLife International, 2018a, 2018c, 2018b), Red-throated Loons and Northern Gannets are both designated as "Priority" under the Bird Conservation Region 30 (BCR30) from the North American Bird Conservation Initiative and as "High Concern" by the Atlantic Marine Bird Conservation Cooperative (AMBCC) (Steinkamp, 2008;Marine Bird Species Priority List -July 2014, 2014;Curtice et al., 2019). Additionally, Great Black-backed Gulls are year-long residents in the northeast making them more vulnerable to potential habitat degradation (Welcker and Nehls, 2016;Goodale, Milman and Griffin, 2019).Although sea turtles are not often the focus of OSW studies, they may be particularly vulnerable due to their strong fidelity to migratory routes and critical habitats, as well as their attraction to artificial light and structures, which could increase overlap with WELAs (Secor, O'Brien and Bailey, 2025). All four of the modeled sea turtle species -green (Chelonia mydas), Kemp's ridley (Lepidochelys kempii), leatherback (Dermochelys coriacea), and loggerhead (Caretta caretta) -were considered focal species for this analysis and are all listed on the IUCN Red List. Greens, who are considered "Endangered", and Kemp's ridleys, who are listed as "Critically Endangered", are especially tied to specific nesting sites, raising concerns about displacement risk (Wibbels and Bevan, 2019;Seminoff, 2023;Secor, O'Brien and Bailey, 2025), while leatherbacks and loggerheads, both classified as "Vulnerable", may be sensitive to the noise frequencies produced by OSW activities (Wallace, Tiwari and Girondot, 2013;Bailey, Brookes and Thompson, 2014;Casale and Tucker, 2017).To provide further context for this analysis, we use subregional divisions -Gulf of Maine (GOM), Southern New England (SNE), New York/New Jersey Bight (NYNJB), Central U.S. Atlantic (CA), and Southeastern U.S. Atlantic (SEA) -established by the Regional Wildlife Science Collaborative (figure 2). To account for regional variations in species distribution and to create a standardized context for comparability across the region, we calculated the 90 th percentile of monthly density, hereafter referred to as "thresholds", for each species across these divisions. This approach highlights locations and time periods of particularly high species occurrence, identifying potential areas of elevated risk and facilitating the prioritization of monitoring and mitigation.As OSW development expands across the U.S. OCS, understanding the cumulative and interconnected nature of its ecological effects -across regions, seasons, and life stages -will be critical. However, major knowledge gaps persist in predicting the long-term, population-level outcomes of these overlapping effects, particularly due to the complexity of ecological connectivity and impacts of concurrent stressors (Methratta et al., 2023;Silber et al., 2023). By establishing a robust baseline of species occurrence, this work provides a critical foundation for integrating wildlife distribution patterns into broader, multi-faceted risk frameworks. These insights are essential not only for informing near-term mitigation and monitoring efforts, but also for supporting more proactive spatial planning that aligns OSW expansion with long-term conservation and sustainability goals.Materials and methodsThe species distribution products used in this study are Density Surface Models (DSMs) that cover 84 marine taxa, including 31 marine mammal taxa (table 1; Roberts et al., 2016;Roberts, Yack and Halpin, 2023), 49 seabird species (table 2; Winship et al., 2023) and 4 sea turtle species (table 3; DiMatteo et al., 2024). These publicly available DSMs were generated by relating qualitychecked observation data from aerial or ship-based line-transect surveys to environmental variables using models to predict species density across spatial grid cells -marine mammals at a 5 km X 5 km resolution, and a 10 km X 10 km resolution for seabirds and sea turtles. Each dataset included detailed metadata, such as platform type and characteristics (e.g. observation height, flight speed), and environmental conditions. This auxiliary information was used to develop detection probability functions that were incorporated into the models. All DSMs reflect composite predictions across their respective period of data collection (table 4), averaged per calendar month, except 18 marine mammal taxa that were at a year-round temporal resolution. For more details about each of the datasets and access information see table 4.Density calculations were used instead of abundance to account for the varying sizes of WELAs and subregions, facilitating comparability between these areas. To maintain data integrity, each dataset was used at its maximum temporal and spatial resolution and retained in its original Coordinate Reference System. It is important to note that while the marine mammal and sea turtle models applied distance sampling techniques to estimate absolute density (individuals per unit area), the seabird models estimate relative density (an index of presence rather than of absolute abundance). This is because distance sampling is generally not feasible for seabird surveys, due to missed detections and movement bias. To allow for comparison across avian species we convert our calculations to proportional density (see section 2.2 Analysis).The study area covers over 886,000 square kilometers, comprising the U.S. East Coast Exclusive Economic Zone (EEZ), extending from shore out approximately 200 nautical miles, from Maine to Florida. To subdivide this large area and provide ecological context for the calculations, we used delineations provided by the RWSC that were based on ecological, political, and geographic characteristics, yielding 5 subregions: Gulf of Maine, Southern New England, New York/New Jersey Bight, Central U.S. Atlantic, and Southeastern U.S. Atlantic seen in figure 2.To delineate the WELAs, we use publicly available polygon shapefiles distributed by the of Energy sites, and the Research were from the because of their and yielding a of 29 WELAs that were 10 km was applied to each to account for the movement of marine species and potential of noise and OSW effects This aligns with of for the of whales and sea turtles and Halpin, et al., 2020), the detection range for whales et al., and seabird displacement et al., 2023). assessments that the density with and the were the 10 km that potential were for in a with a estimate of species each and all 84 taxa, we species occurrence within WELAs by predicted density for all grid cells that within the of each We then the spatial average of these to a monthly density per species per These were used to identify patterns and with subregional evaluate WELAs with high species density, we a for comparison between the WELAs and their respective subregions: Gulf of Maine, Southern New England, New York/New Jersey Bight, U.S. Central Atlantic, and U.S. We selected the 90 th percentile of the monthly densities as the from marine life risk and et al., sea et al., sea et al., and with These were for each calendar and used as to identify each densities in and regional ecological These were as and to identify patterns and potential areas of that could inform OSW the Gulf of Maine was included in these it active WELAs, its are not further in this account for the relative density from the seabird DSMs and species we standardized the predicted relative density using a the average monthly densities for each and each 90 th percentile This adapted from Winship the predicted relative density by each grid of the by the density across all grid cells for each species and month, to monthly and subregional This in a one areas of density, while one areas of in this highlights spatial patterns and areas of particularly high or relative density, However, it is important to note that these are not to absolute density and be as because the DSMs of the taxa in this study are as absolute this approach not the calculations were for all 84 marine taxa, we selected eleven species to as focal species. These were chosen to a of and ecological and were also by regional and The focal species include four marine mammals whales, humpback whales, minke whales, and North Atlantic right three avian species Northern and Great and all four modeled sea turtle species Kemp's and study calculated the predicted densities of 84 marine taxa across 29 WELAs in the US Atlantic region. For all taxa, average monthly for the 18 year-round marine mammal density for each 90 th percentile for all subregions, between the densities and and of species with the densities and densities the 90 th percentile in WELAs can be in the 13 monthly marine mammal taxa were predicted to have of presence across the the and presence for species minke and in Additionally, three taxa North Atlantic right whales, predicted presence in of the WELAs, and selected the 18 year-round marine mammal taxa, were not in of the WELAs Northern and and and presence in the densities were in and in in in in in in in and Atlantic in in in The species with the between density and the subregional 90 th percentile for each were in in in in in in and Atlantic in and in in in in the four focal species we can identify patterns of and spatial density (figure Fin whales density from with in and particularly in However, these densities the subregional the WELAs presence from to but only one the in a and in the the in whales the densities with their density in in and leases and the in with it in and as The over the for in in with additional the in and whales a in all subregions with the densities all 10 leases the these with it for out of the not elevated densities from to were in leases in and The was in in Atlantic right whales a in from to of the leases and densities for the with only the in all The in was the with and for and five we temporal patterns of the 18 taxa with only year-round we can identify spatial patterns (figure these 18 taxa, whales the densities in all areas the sites, and the and the leases the Outer in North the most taxa with average annual densities than whales, and of the densities the regional for of the year-round 49 species were predicted to be in all 29 areas the The species with the relative densities and relative densities the 90 th percentile in WELAs for each are in The species with the relative densities were in and in in in in in in and Red in in in species with the between density and the 90 th percentile for each was in in in in and in in in in and were in and in in in in in in and Red in in in the focal species, Northern Gannets and Red-throated Loons patterns, while Great Black-backed Gulls maintain a presence year-round across all regions, except for (figure Northern density across the coast over the of the relative densities are from to especially in the between density and subregional is in This area also the in all except relative densities from to and to are from to and to while in the densities are from with densities the are more while in and these in For in densities the for from relative densities in but a from and WELAs densities the for or more with the in all except the of the densities were more from to in to in and to in spatial was in the WELAs further in and further in in these areas densities the for or more including three in and which the also the in The leases in also over the the presence of Great Black-backed Gulls was more were densities in than in with the in in also with than were leases over the in and in across the turtles were predicted to in all WELAs the temporal gaps were in the densities of Kemp's ridley and sea turtles (figure These species predicted density for at three in of the 29 the and year-round presence of all four sea turtle species. The densities for each and densities the 90 th percentile for each are in All species a density from the regional and of this by species.The species with the densities were leatherback for in in loggerhead in and in in in in and in in in The species with the between density and the 90 th percentile for each was leatherback for and in in in in and loggerhead in in in species the subregional in of the sea turtles from in the and from in of the the densities the for in with the for the species in densities were the for five leases in as particularly in and turtles the and most with elevated densities from across the leases in and three or more all between The in density for leatherbacks in in in ridley a their presence was in and densities more 10 of the from the four leases a including which the in and from in and While densities in were and not the with densities the of and from to all but one at one over the with and the Additionally, leases in and in The in density from the in in in study the first of Density Surface Models (DSMs) to the potential of OSW development to marine mammal, seabird and sea turtle species along the U.S. Atlantic The the spatiotemporal in species by the complex and patterns of these taxa and Halpin, et al., et al., O'Brien and Bailey, 2025). By the OERC framework, we can these by risk to provide the foundation for and mitigation the focus of this assessment is on species occurrence, the of the OERC risk framework, the also insights into of species such as potential information and for in OSW planning and mitigation.As the species distribution patterns tied to and habitat Brookes and Thompson, and Halpin, et al., et al., O'Brien and Bailey, 2025). For avian species, including of the focal species, Northern and Red-throated relative densities the in the more while sea turtles the and the four focal species, the three migratory species and North Atlantic right more of the 90 th percentile density than fin whales, who have been to have limited migratory and maintain a distribution year-round (figure et al., et al., et al., 2023;Secor, O'Brien and Bailey, 2025). the presence of Kemp's and in the and the in Great Black-backed that may be for species that or within WELAs the Great Black-backed Gulls are not they are of growing conservation due to their large and activities in the and regions, making them more vulnerable to OSW collision and displacement (Willmott, Forcey and Kent, et al., Milman and Griffin, species density on a monthly critical in species that have on OSW mitigation For North Atlantic right whales were in the while and humpback whales high densities in that at this time in by the in figure patterns have such as activities to the and to risk to the North Atlantic right (Silber et al., 2023;Stepanuk et al., 2023). However, these temporal in migratory patterns have been further in by the effects of and highlighting the for and et al., et al., 2023;Silber et al., 2023;Secor, O'Brien and Bailey, critical spatial to is the distance to shore of the of the avian species densities in the within 50 of shore (figure such as which the regional density for all three focal avian species in at (see figure This is also for Red-throated Loons in and all with elevated densities and were within 50 of the of the year-round marine mammals densities in the leases further such as whales, whales, and whales (figure The offshore of these more and marine mammal species an important as wind more and OSW activities further offshore habitats. This of patterns the for species data to inform mitigation to highlighting species of high these are also for identifying species at risk from OSW development that have not been as in that sea turtles were concentrated in the three the the for more focused on their habitat use in that area to OSW marine and the densities and density a risk of (figure For species such as the and were to have densities than subregional the subregions (figure these species are considered species under with high sensitivity to collision and displacement, making them at risk from OSW activities (Steinkamp, 2008;Marine Bird Species Priority List -July 2014, 2014;Curtice et al., Milman and Griffin, 2019). While is on the potential risk of OSW to of these species -green sea turtles (Secor, O'Brien and Bailey, 2025), et al., et al., and Halpin, and Halpin, Milman and Griffin, and et al., and Halpin, et al., further the importance of a range of species for OSW planning and mitigation this study was based on predictions from DSMs using environmental it is essential to the with a and marine The DSMs on which are to from and et al., and Yack and Halpin, et al., 2023). For densities predicted in the for species, may reflect limited rather than a Yack and Halpin, 2023). in patterns, by further and the for monitoring and Yack and Halpin, 2023;Silber et al., et al., 2023;Secor, O'Brien and Bailey, 2025). These are for species, of which IUCN or ecological making it to these is essential to not only to and but to more proactive and mitigation continued baseline studies, the of additional data and in advancements et al., et al., et al., foundational assessment provides the first assessment of predicted occurrence across all WELAs the U.S. Atlantic Outer Continental We identify both and species are most to overlap with OSW activities to a crucial ecological baseline to inform OSW potential impacts are not but by behaviors, and conservation be the for nuanced, mitigation to ecological The patterns in our such as high densities in specific areas or of subregional for OSW development with conservation that each of the selected focal species of occurrence in WELAs that the 90 th percentile subregional density at the of the This that all focal species are to to OSW in the U.S. Atlantic in one or more each the in the OERC risk framework, these the groundwork for species and a more understanding of risk across taxa and its to OSW the OERC risk framework and approach provide for spatial environmental assessments, the of mitigation and for development across multiple and OSW development for renewable energy it with assessments such as and be by monitoring all of development to models, inform and impacts to marine species and these are only essential for renewable energy development with the conservation of marine habitats. By the knowledge we can better that the to renewable energy with of marine and This was as an account of work by an of the the of their or or or for the or of or or that its use not Reference to specific or by or not or its or by the or The and of not or reflect of the or to the of and that were to the of this We Roberts for providing the marine mammal density models, Winship for the avian density models, and DiMatteo and for the sea turtle density models. We also the and from these three data which were crucial in the with and from the of work on risk assessment as of the of this were also We also our to the for their foundational which as the for this we our for and the Reference are as a as as the is and be applied in the of or and for For and include in the For more of and to the of in the of or the in and in and the for be in of in the by in the for and

  PublisherOA PDFDOI

• Recommendations for built marine infrastructure that supports natural habitats

  Frontiers in Ecology and the Environment · 2025-03-11 · 3 citations

  reviewOpen access

  The extent of built marine infrastructure—from energy infrastructure and ports to artificial reefs and aquaculture—is increasing globally. The rise in built structure coverage is concurrent with losses and degradation of many natural habitats. Although historically associated with net negative impacts on natural systems, built infrastructure—with proper design and innovation—could offer a largely unrealized opportunity to reduce those impacts and support natural habitats. We present nine recommendations that could catalyze momentum toward using built structures to both serve their original function and benefit natural habitats (relative to the status quo, for example). These recommendations integrate functional, economic, and social considerations with marine spatial planning and holistic ecosystem management. As the footprint of the Anthropocene expands into ocean spaces, adopting these nine recommendations at global scales can help to ensure that ecological harm is minimized and that, where feasible, ecological benefits from marine built structures are accrued.

  PublisherOA PDFDOI

• Protecting for or from?: Unique high-seas oceanographic features demonstrate contrasting protection priorities for the high seas 

  2025-03-25

  preprintOpen accessSenior author

  Nearly half the surface of the Earth consists of the high seas, the ocean 200 nautical miles beyond a nation’s coastline, with a recent 2023 UN international agreement (BBNJ Treaty) requiring renewed political efforts to protect this vast area. Unique to this agreement, when ratified, will be its ability to create marine protected areas on the high seas. With various organizations and institutions beginning the race to champion their chosen biodiversity hotspot, two unique high seas oceanographic features have emerged: the Costa Rica Thermal Dome and the Sargasso Sea. In these high seas case studies, we investigate the fishing intensity, the corporate actors benefitting from the fishing effort, and shipping traffic to understand the current anthropogenic stress in both regions. We find anthropogenic activity is deeply intertwined with the biogeochemistry in these contrasting oceanographic high seas features. The Sargasso Sea is a sub-tropical gyre and is thus highly oligotrophic, while the Costa Rica thermal dome is a seasonal upwelling that brings nutrient-rich water to the euphotic zone, attracting highly migratory species and the corporate actors that aim to extract them. Understanding that the greatest current impact from anthropogenic activity on the high seas is industrialized fishing, the high seas case studies analyzed here pose the question, should we protect marine life and high seas habitat from current activity (such as in the Costa Rica thermal dome), or areas with very little anthropogenic activity that have endemic species and intrinsic natural value (the Sargasso Sea)? With the implementation of the BBNJ treaty on the horizon and limited tools to keep pace with industrial corporations who work to overexploit the high seas, it will be essential to identify the values behind the models we create to identify and champion high seas areas’ and if they are protecting marine life from harm or not.

  PublisherDOI

• An expanded evaluation of global fisheries management organizations on the high seas

  Environmental Research Letters · 2025-11-04

  articleOpen accessSenior author

  Abstract The high seas comprise nearly half the Earth’s surface and the 2023 United Nations Biodiversity Beyond National Jurisdiction agreement renewed efforts to better manage this vast area. Regional fisheries management organizations (RFMOs) are tasked with a UN dual-mandate to ensure the (1) long-term conservation of fish stocks and their associated ecosystems while managing the (2) sustainable use of high seas fisheries. Building on a previous analysis, we developed 10 questions each for our 10 categories (e.g. catch targets, transparency, bycatch). These 100 questions were both descriptive (e.g. are there quantitative results of bycatch improvement?) while others indicated intent (e.g. are there plans to rebuild all declining stocks?) and used publicly available information to evaluate 16 RFMOs (with a score of 0, 0.5, or 1). The average score for all RFMOs out of 100 was 46 (max 61.5, min 29.5). There were six questions for which all RFMOs received a 0, 90 questions for which at least one RFMO scored a one, and a ‘best case’ score of 76.5 that aggregated the highest RFMO score across the 10 categories. This indicates a competence gap; no RFMO is achieving what is empirically documented as possible in one RFMO or another. We also analyzed satellite-derived fishing activity and target stock status. Five RFMO convention areas had the highest overall and density of fishing effort. On average, 56% of stocks targeted in RFMO convention areas were considered overexploited or collapsed. We did not find that RFMO scores were correlated with better management outcomes (i.e. target stock status and fishing effort). However, we found correlations between better target stock status and questions assessing protected area adoption, banning transshipment at-sea, and reducing allowable catches in response to overfishing—policies that few RFMOs have fully adopted. These findings indicate ways for RFMOs to improve high seas fisheries management that are immediately available and effective.

  PublisherDOI

• What is an ecologically or biologically significant area?

  npj Ocean Sustainability · 2025-05-31

  articleOpen accessSenior author

  The first iteration of the ecologically or biologically significant areas (EBSA) process, which aims to ascribe ecological value to marine and coastal regions, has drawn to a close. This Convention on Biological Diversity process has collated vast amounts of information to describe 338 EBSAs that span from estuaries to ocean trenches. To increase the utility and accessibility of the ocean of knowledge generated by the EBSA process, and to support appropriate application of the dataset, clarity is required around the types of areas described, the biodiversity they hold, and the rationale for their selection. In this study, we provide a holistic answer to the question: What is an EBSA? We identify geographic and taxonomic gaps in EBSA descriptions, trends in the levels of protection observed, and ways forward to improve the uptake and appropriate use of the outputs of this singular intergovernmental process.

  PublisherOA PDFDOI

Recent grants

• Expansion of Metadata Management, Visualization and Data Processing Functionality of OBIS-SEAMAP

  NSF · $314k · 2011–2014

• Collaborative Research: U.S. SO GLOBEC Synthesis and Modeling - Habitat Utilization and Predator-Prey interactions In Western Antarctic Peninsula

  NSF · $201k · 2006–2009

• Developing the Next Generation Marine Mammal Information Center for Integrated Ocean Observing: OBIS-SEAMAP 2.0

  NSF · $1.7M · 2007–2011

Frequent coauthors

• Daniel C. Dunn

  83 shared

• Andrew J. Read

  Duke University

  46 shared

• Jason J. Roberts

  Duke University

  43 shared

• André M. Boustany

  California State University, Monterey Bay

  36 shared

• Jesse Cleary

  Duke University

  29 shared

• Benjamin D. Best

  28 shared

• Connie Y. Kot

  Duke University

  23 shared

• Kristina M. Gjerde

  Middlebury Institute of International Studies at Monterey

  23 shared

Education

• Ph.D., Environmmental Sciences

  University of Virgina

  1995