Range Size and Extinction Risk in Forest Birds

Abstract:  Small geographical range size is the single best predictor of threat of extinction in terrestrial species. Knowing how small a species' range has to be before authorities consider it threatened with extinction would allow prediction of a species' risk from continued deforestation and warming climates and provide a baseline for conservation and management strategies aspiring to mitigate these threats. To determine the threshold at which forest-dependent bird species become threatened with extinction, we compared the range sizes of threatened and nonthreatened species. In doing so, we present a simple, repeatable, and practical protocol to quantify range size. We started with species' ranges published in field guides or comparable sources. We then trimmed these ranges, that is, we included only those parts of the ranges that met the species' requirements of elevation and types of forest preferred. Finally, we further trimmed the ranges to the amount of forest cover that remains. This protocol generated an estimate of the remaining suitable range for each species. We compared these range estimates with those from the World Conservation Union Red List. We used the smaller of the two estimates to determine the threshold, 11,000 km², below which birds should be considered threatened. Species considered threatened that have larger ranges than this qualified under other (nonspatial) red list criteria. We identified a suite of species (18) that have not yet qualified as threatened but that have perilously small ranges—about 11% of the nonthreatened birds we analyzed. These birds are likely at risk of extinction and reevaluation of their status is urgently needed.     

The Threat of Disease Increases as Species Move Toward Extinction

At local scales, infectious disease is a common driver of population declines, but globally it is an infrequent contributor to species extinction and endangerment. For species at risk of extinction from disease important questions remain unanswered, including when does disease become a threat to species and does it co‐occur, predictably, with other threats? Using newly compiled data from the International Union for Conservation of Nature (IUCN) Red List, we examined the relative role and co‐occurrence of threats associated with amphibians, birds, and mammals at 6 levels of extinction risk (i.e., Red List status categories: least concern, near threatened, vulnerable, endangered, critically endangered, and extinct in the wild/extinct). We tested the null hypothesis that the proportion of species threatened by disease is the same in all 6 Red List status categories. Our approach revealed a new method for determining when disease most frequently threatens species at risk of extinction. The proportion of species threatened by disease varied significantly between IUCN status categories and linearly increased for amphibians, birds, and all species combined as these taxa move from move from least concern to critically endangered. Disease was infrequently the single contributing threat. However, when a species was negatively affected by a major threat other than disease (e.g., invasive species, land‐use change) that species was more likely to be simultaneously threatened by disease than species that had no other threats. Potential drivers of these trends include ecological factors, clustering of phylogenetically related species in Red List status categories, discovery bias among species at greater risk of extinction, and availability of data. We echo earlier calls for baseline data on the presence of parasites and pathogens in species when they show the first signs of extinction risk and arguably before. La Amenaza de Enfermedades Incrementa a Medida que las Especies se Aproximan a la Extinción     

Trait‐based prediction of extinction risk of small‐bodied freshwater fishes

Small body size is generally correlated with r‐selected life‐history traits, including early maturation, short‐generation times, and rapid growth rates, that result in high population turnover and a reduced risk of extinction. Unlike other classes of vertebrates, however, small freshwater fishes appear to have an equal or greater risk of extinction than large fishes. We explored whether particular traits explain the International Union for Conservation of Nature (IUCN) Red List conservation status of small‐bodied freshwater fishes from 4 temperate river basins: Murray‐Darling, Australia; Danube, Europe; Mississippi‐Missouri, North America; and the Rio Grande, North America. Twenty‐three ecological and life‐history traits were collated for all 171 freshwater fishes of ≤120 mm total length. We used generalized linear mixed‐effects models to assess which combination of the 23 traits best explained whether a species was threatened or not threatened. We used the best models to predict the probability of 29 unclassified species being listed as threatened. With and without controlling for phylogeny at the family level, small body size—among small‐bodied species—was the most influential trait correlated with threatened species listings. The k‐folds cross‐validation demonstrated that body size and a random effect structure that included family predicted the threat status with an accuracy of 78% (SE 0.5). We identified 10 species likely to be threatened that are not listed as such on the IUCN Red List. Small body size is not a trait that provides universal resistance to extinction, particularly for vertebrates inhabiting environments affected by extreme habitat loss and fragmentation. We hypothesize that this is because small‐bodied species have smaller home ranges, lower dispersal capabilities, and heightened ecological specialization relative to larger vertebrates. Trait data and further model development are needed to predict the IUCN conservation status of the over 11,000 unclassified freshwater fishes, especially those under threat from proposed dam construction in the world's most biodiverse river basins.     

Use of Basic Biological Information for Rapid Prediction of the Response of Species to Habitat Loss

Abstract:  Much research has focused on identifying traits that can act as useful indicators of how habitat loss affects the extinction risk of species, and the results are mixed. We developed 2 simple, rapid-assessment models of the susceptibility of species to habitat loss. We based both on an index of range size, but one also incorporated an index of body mass and the other an index combining habitat and dietary specialization. We applied the models to samples of birds (Accipitridae and Bucerotidae) and to the lemurs of Madagascar and compared the models' classifications of risk with the IUCN's global threat status of each species. The model derived from ecological attributes was much more robust than the one derived from body mass. Ecological attributes identified threatened birds and lemurs with an average of 80% accuracy and endangered and critically endangered species with 100% accuracy and identified some species not currently listed as threatened that almost certainly warrant conservation consideration. Appropriate analysis of even fairly crude biological information can help raise early-warning flags to the relative susceptibilities of species to habitat loss and thus provide a useful and rapid technique for highlighting potential species-level conservation issues. Advantages of this approach to classifying risk include flexibility in the specialization parameters used as well as its applicability at a range of spatial scales.     

Using cumulative human-impact models to reveal global threat patterns for seahorses

Understanding threats acting on marine organisms and their conservation status is vital but challenging given a paucity of data. We studied the cumulative human impact (CHI) on and conservation status of seahorses (Hippocampus spp.)—a genus of rare and data-poor marine fishes. With expert knowledge and relevant spatial data sets, we built linear-additive models to assess and map the CHI of 12 anthropogenic stressors on 42 seahorse species. We examined the utility of indices of estimated impact (impact of each stressor and CHI) in predicting conservation status for species with random forest (RF) models. The CHI values for threatened species were significantly higher than those for nonthreatened species (category based on International Union for Conservation of Nature Red List). We derived high-accuracy RF models (87% and 96%) that predicted that 5 of the 17 data-deficient species were threatened. Demersal fishing practices with high bycatch and pollution were the best predictors of threat category. Major threat epicenters were in China, Southeast Asia, and Europe. Our results and maps of CHI may help guide global seahorse conservation and indicate that modeling and mapping human impacts can reveal threat patterns and conservation status for data-poor species. We found that for exploring threat patterns of focal species, species-level CHI models are better than existing ecosystem-level CHI models.     

Automated conservation assessment of the orchid family with deep learning

International Union for Conservation of Nature (IUCN) Red List assessments are essential for prioritizing conservation needs but are resource intensive and therefore available only for a fraction of global species richness. Automated conservation assessments based on digitally available geographic occurrence records can be a rapid alternative, but it is unclear how reliable these assessments are. We conducted automated conservation assessments for 13,910 species (47.3% of the known species in the family) of the diverse and globally distributed orchid family (Orchidaceae), for which most species (13,049) were previously unassessed by IUCN. We used a novel method based on a deep neural network (IUC-NN). We identified 4,342 orchid species (31.2% of the evaluated species) as possibly threatened with extinction (equivalent to IUCN categories critically endangered [CR], endangered [EN], or vulnerable [VU]) and Madagascar, East Africa, Southeast Asia, and several oceanic islands as priority areas for orchid conservation. Orchidaceae provided a model with which to test the sensitivity of automated assessment methods to problems with data availability, data quality, and geographic sampling bias. The IUC-NN identified possibly threatened species with an accuracy of 84.3%, with significantly lower geographic evaluation bias relative to the IUCN Red List and was robust even when data availability was low and there were geographic errors in the input data. Overall, our results demonstrate that automated assessments have an important role to play in identifying species at the greatest risk of extinction.     

Extinction risk of the endemic vascular flora of Kauai,Hawaii, based on IUCN assessments

The International Union for Conservation of Nature's Red List of Threatened Species (IUCN Red List) is the world's most comprehensive information source on the global conservation status of species. Governmental agencies and conservation organizations increasingly rely on IUCN Red List assessments to develop conservation policies and priorities. Funding agencies use the assessments as evaluation criteria, and researchers use meta-analysis of red-list data to address fundamental and applied conservation science questions. However, the circa 143,000 IUCN assessments represent a fraction of the world's biodiversity and are biased in regional and organismal coverage. These biases may affect conservation priorities, funding, and uses of these data to understand global patterns. Isolated oceanic islands are characterized by high endemicity, but the unique biodiversity of many islands is experiencing high extinction rates. The archipelago of Hawaii has one of the highest levels of endemism of any floristic region; 90% of its 1367 native vascular plant taxa are classified as endemic. We used the IUCN's assessment of the complete single-island endemic (SIE) vascular plant flora of Kauai, Hawaii, to assess the proportion and drivers of decline of threatened plants in an oceanic island setting. We compared the IUCN assessments with federal, state, and other local assessments of Kauai species or taxa of conservation concern. Finally, we conducted a preliminary assessment for all 1044 native vascular plants of Hawaii based on IUCN criterion B by estimating area of occupancy, extent of occurrence, and number of locations to determine whether the pattern found for the SIE vascular flora of Kauai is comparable to the native vascular flora of the Hawaiian Islands. We compared our results with patterns observed for assessments of other floras. According to IUCN, 256 SIE vascular plant taxa are threatened with extinction and 5% are already extinct. This is the highest extinction risk reported for any flora to date. The preliminary assessment of the native vascular flora of Hawaii showed that 72% (753 taxa) is threatened. The flora of Hawaii may be one of the world's most threatened; thus, increased and novel conservation measures in the state and on other remote oceanic islands are urgently needed.     

Exploring the conservation status of an endangered coastal plant over the last nine years: a case study in Andalusia,Spain

Human pressure has been exponentially growing during recent decades in coastal areas, which have led to drastic losses of biodiversity in coastal ecosystems. The current conservation status of many coastal plant species is directly related to a lack of environmental criteria in the urban planning of coastal areas over recent decades. This study aimed to evaluate the evolution, over the last 9 years, of the conservation status of various populations of the endangered plant Glaucium flavum, exploring the extent to which human pressure and different management strategies practiced in the coastal areas where the populations are established have affected the conservation status of the species. The populations analysed have evolved in a different manner over the last 9 years, as have their threat factors, and a relationship was evident between their conservation status and the evolution of these different threat factors. Our results indicate that an appropriate planning of local management actions, such as the installation of walkways or the successful eradication of invasive species, may be determinant factors for successful conservation of the coastal vegetation. The presence of species that are sensitive to slight changes in the ecosystem, and the main factors that govern the plant performance of these species, must be given full consideration in decision-making processes of coastal planning and management.     

Status of Southeast Asia's marine sharks and rays

In Southeast Asia, elasmobranchs are particularly threatened. We synthesized knowledge from the peer-reviewed and gray literature on elasmobranchs in the region, including their fisheries, status, trade, biology, and management. We found that 59% of assessed species are threatened with extinction and 72.5% are in decline; rays were more threatened than sharks. Research and conservation is complicated by the socioeconomic contexts of the countries, geopolitical issues in the South China Sea, and the overcapacity and multispecies nature of fisheries that incidentally capture elasmobranchs. The general paucity of data, funds, personnel, and enforcement hinders management. Reduced capacity in the general fishery sector and marine protected areas of sufficient size (for elasmobranchs and local enforcement capabilities) are among recommendations to strengthen conservation.     

Regional IUCN Red Listing: the Process as Applied to Birds in the United Kingdom

Abstract:  The World Conservation Union (IUCN) has developed guidelines that enable the assessment of extinction risk at a regional scale. We used these guidelines to assess the extinction risk of birds in the United Kingdom for comparison with an existing assessment of conservation status. Sixty-four species were categorized as regionally threatened, of which 12 were critically endangered. The categorizations of the 223 species assessed agreed broadly with those from the existing U.K. system, which considers more than extinction risk, thus giving a more complete assessment of conservation status. There was, however, a tendency for the IUCN process to give higher risk status to edge-of-range species (some of which are relatively recent colonists considered of comparatively low conservation concern) and low status to those that have declined substantially but remain common (such as many farmland birds, the focus of considerable conservation effort in the United Kingdom). The final red list depended heavily on the subjective decisions made during the assessment process. An alternative interpretation of the guidelines could have resulted in as many as 19 or as few as 6 species being listed as critically endangered. We recommend the revision of the IUCN guidelines to reduce this subjectivity, in particular with respect to the effect of extralimital populations on the likelihood of regional extinction, and hence the potential for variation in the manner of application between regional red-list assessors. Preventing extinction does not have to be the principal driving force behind conservation action at a regional scale if the continuance of a species is safeguarded in other regions.     

Quantification of Extinction Risk: IUCN's System for Classifying Threatened Species

Abstract: The International Union for Conservation of Nature (IUCN) Red List of Threatened Species was increasingly used during the 1980s to assess the conservation status of species for policy and planning purposes. This use stimulated the development of a new set of quantitative criteria for listing species in the categories of threat: critically endangered, endangered, and vulnerable. These criteria, which were intended to be applicable to all species except microorganisms, were part of a broader system for classifying threatened species and were fully implemented by IUCN in 2000. The system and the criteria have been widely used by conservation practitioners and scientists and now underpin one indicator being used to assess the Convention on Biological Diversity 2010 biodiversity target. We describe the process and the technical background to the IUCN Red List system. The criteria refer to fundamental biological processes underlying population decline and extinction. But given major differences between species, the threatening processes affecting them, and the paucity of knowledge relating to most species, the IUCN system had to be both broad and flexible to be applicable to the majority of described species. The system was designed to measure the symptoms of extinction risk, and uses 5 independent criteria relating to aspects of population loss and decline of range size. A species is assigned to a threat category if it meets the quantitative threshold for at least one criterion. The criteria and the accompanying rules and guidelines used by IUCN are intended to increase the consistency, transparency, and validity of its categorization system, but it necessitates some compromises that affect the applicability of the system and the species lists that result. In particular, choices were made over the assessment of uncertainty, poorly known species, depleted species, population decline, restricted ranges, and rarity; all of these affect the way red lists should be viewed and used. Processes related to priority setting and the development of national red lists need to take account of some assumptions in the formulation of the criteria.     

A Retrospective Evaluation of the Global Decline of Carnivores and Ungulates

Assessing temporal changes in species extinction risk is necessary for measuring conservation success or failure and for directing conservation resources toward species or regions that would benefit most. Yet, there is no long‐term picture of genuine change that allows one to associate species extinction risk trends with drivers of change or conservation actions. Through a review of 40 years of IUCN‐related literature sources on species conservation status (e.g., action plans, red‐data books), we assigned retrospective red‐list categories to the world's carnivores and ungulates (2 groups with relatively long generation times) to examine how their extinction risk has changed since the 1970s. We then aggregated species’ categories to calculate a global trend in their extinction risk over time. A decline in the conservation status of carnivores and ungulates was underway 40 years ago and has since accelerated. One quarter of all species (n = 498) moved one or more categories closer to extinction globally, while almost half of the species moved closer to extinction in Southeast Asia. The conservation status of some species improved (toward less threatened categories), but for each species that improved in status 8 deteriorated. The status of large‐bodied species, particularly those above 100 kg (including many iconic taxa), deteriorated significantly more than small‐bodied species (below 10 kg). The trends we found are likely related to geopolitical events (such as the collapse of Soviet Union), international regulations (such as CITES), shifting cultural values, and natural resource exploitation (e.g., in Southeast Asia). Retrospective assessments of global species extinction risk reduce the risk of a shifting baseline syndrome, which can affect decisions on the desirable conservation status of species. Such assessments can help conservationists identify which conservation policies and strategies are or are not helping safeguard biodiversity and thus can improve future strategies. Una Evaluación Retrospectiva de la Declinación Global de Carnívoros y Ungulados     

Phylogenetic Comparative Methods Strengthen Evidence for Reduced Genetic Diversity among Endangered Tetrapods

Abstract: The fitness of species with little genetic diversity is expected to be affected by inbreeding and an inability to respond to environmental change. Conservation theory suggests that endangered species will generally demonstrate lower genetic diversity than taxa that are not threatened. This hypothesis has been challenged because the time frame of anthropogenic extinction may be too fast to expect genetic factors to significantly contribute. I conducted a meta‐analysis to examine how genetic diversity in 894 tetrapods correlates with extinction threat level. Because species are not evolutionarily independent, I used a phylogenetic regression framework to address this issue. Mean genetic diversity of tetrapods, as assessed by protein heterozygosity, was 29.7–31.5% lower on average in threatened species than in their nonthreatened relatives, a highly significant reduction. Within amphibians as diversity decreased extinction risk increased in phylogenetic models, but not in nonphylogenetic regressions. The effects of threatened status on diversity also remained significant after accounting for body size in mammals. These results support the hypothesis that genetic effects on population fitness are important in the extinction process.     

Predicting the conservation status of data‐deficient species

There is little appreciation of the level of extinction risk faced by one‐sixth of the over 65,000 species assessed by the International Union for Conservation of Nature. Determining the status of these data‐deficient (DD) species is essential to developing an accurate picture of global biodiversity and identifying potentially threatened DD species. To address this knowledge gap, we used predictive models incorporating species’ life history, geography, and threat information to predict the conservation status of DD terrestrial mammals. We constructed the models with 7 machine learning (ML) tools trained on species of known status. The resultant models showed very high species classification accuracy (up to 92%) and ability to correctly identify centers of threatened species richness. Applying the best model to DD species, we predicted 313 of 493 DD species (64%) to be at risk of extinction, which increases the estimated proportion of threatened terrestrial mammals from 22% to 27%. Regions predicted to contain large numbers of threatened DD species are already conservation priorities, but species in these areas show considerably higher levels of risk than previously recognized. We conclude that unless directly targeted for monitoring, species classified as DD are likely to go extinct without notice. Taking into account information on DD species may therefore help alleviate data gaps in biodiversity indicators and conserve poorly known biodiversity. Predección del Estado de Conservación de Especies con Deficiencia de Datos     

Infectious Diseases and Extinction Risk in Wild Mammals

Abstract:  Parasite-driven declines in wildlife have become increasingly common and can pose significant risks to natural populations. We used the IUCN Red List of Threatened and Endangered Species and compiled data on hosts threatened by infectious disease and their parasites to better understand the role of infectious disease in contemporary host extinctions. The majority of mammal species considered threatened by parasites were either carnivores or artiodactyls, two clades that include the majority of domesticated animals. Parasites affecting host threat status were predominantly viruses and bacteria that infect a wide range of host species, including domesticated animals. Counter to our predictions, parasites transmitted by close contact were more likely to cause extinction risk than those transmitted by other routes. Mammal species threatened by parasites were not better studied for infectious diseases than other threatened mammals and did not have more parasites or differ in four key traits demonstrated to affect parasite species richness in other comparative studies. Our findings underscore the need for better information concerning the distribution and impacts of infectious diseases in populations of endangered mammals. In addition, our results suggest that evolutionary similarity to domesticated animals may be a key factor associated with parasite-mediated declines; thus, efforts to limit contact between domesticated hosts and wildlife could reduce extinction risk.     

Indigenous Uses, Population Density, and Conservation of Threatened Medicinal Plants in Protected Areas of the Indian Himalayas

Abstract:  For 10 years I monitored the population density of threatened medicinal plant species in seven protected areas in the Indian Himalayas. I also documented the indigenous uses of threatened medicinal plants through interviews with 138 herbal healers (83 Tibetan healers and 55 Ayurvedic healers) residing in the buffer zone villages of these protected areas. To assess the population status of threatened medicinal plant species, I sampled the 10 major habitat types in the protected areas. In all, I found 60 threatened medicinal plant species during the study period, of which 54 species occurred in the sampling plots. Twenty-two percent of threatened medicinal plant species were critically endangered, 16% were endangered, and 27% were vulnerable. Thirty-two threatened medicinal plant species were endemic to the Himalayan region. The density of threatened medicinal plant species varied with protected areas. The Valley of Flowers protected area had the highest number of threatened medicinal plant species. The "moist" habitat type was richest in these species among all 10 habitat types sampled. Arnebia euchroma (Royle ex Benth.) Johnston and Ephedra gerardiana Wall. ex Stapf. were the most common threatened medicinal plant species. The indigenous groups of healers used these threatened species in curing about 45 different ailments. Based on my findings, I believe that to ensure the long-term sustainability of threatened medicinal plants, medicinal-plant conservation areas should be established.     

Protected Areas and Prospects for Endangered Species Conservation in Canada

Abstract:  Reserve networks figure prominently in conservation strategies that aim to reduce extinction rates. We tested the effectiveness of the current reserve network at protecting species at risk in Canada, where relatively extensive wilderness areas remain. We compared numbers of terrestrial species at risk included in existing reserves to randomly generated networks with the same total area and number of reserves. Existing reserve networks rarely performed better than randomly selected areas and several included fewer endangered species than expected by chance, particularly in the most biologically imperiled regions. The extent of protected area and density of species at risk were unrelated at either broad (countrywide) or finer spatial scales (50 × 50 km grids), although there was a tendency for the most threatened regions of the country to have few or no protected areas (1.5% of areas with >30 endangered species were in reserves). Although reserves will play a useful role in conserving endangered species that occur within them, reducing extinction rates in a region with much of the world's remaining wilderness will require integrating conservation strategies with agricultural and urban land-use plans outside formally protected areas.     

Species Review of Amphibian Extinction Risks in Madagascar: Conclusions from the Global Amphibian Assessment

Abstract:  We assessed the extinction risks of Malagasy amphibians by evaluating their distribution, occurrence in protected areas, population trends, habitat quality, and prevalence in commercial trade. We estimated and mapped the distribution of each of the 220 described Malagasy species and applied, for the first time, the IUCN Red List categories and criteria to all species described at the time of the assessment. Nine species were categorized as critically endangered, 21 as endangered, and 25 as vulnerable. The most threatened species occur on the High Plateau and/or have been subjected to overcollection for the pet trade, but restricted extent of occurrence and ongoing habitat destruction were identified as the most important factors influencing extinction threats. The two areas with the majority of threatened species were the northern Tsaratanana-Marojejy-Masoala highlands and the southeastern Anosy Mountains. The current system of protected areas includes 82% of the threatened amphibian species. Of the critically endangered species, 6 did not occur in any protected area. For conservation of these species we recommend the creation of a reserve for the species of the Mantella aurantiaca group, the inclusion of two Scaphiophryne species in the Convention on the International Trade in Endangered Species Appendix II, and the suspension of commercial collecting for Mantella cowani . Field surveys during the last 15 years reveal no pervasive extinction of Malagasy amphibians resulting from disease or other agents, as has been reported in some other areas of the world.     

Variability in Population Abundance and the Classification of Extinction Risk

Abstract: Classifying species according to their risk of extinction is a common practice and underpins much conservation activity. The reliability of such classifications rests on the accuracy of threat categorizations, but very little is known about the magnitude and types of errors that might be expected. The process of risk classification involves combining information from many sources, and understanding the quality of each source is critical to evaluating the overall status of the species. One common criterion used to classify extinction risk is a decline in abundance. Because abundance is a direct measure of conservation status, counts of individuals are generally the preferred method of evaluating whether populations are declining. Using the thresholds from criterion A of the International Union for Conservation of Nature (IUCN) Red List (critically endangered, decline in abundance of >80% over 10 years or 3 generations; endangered, decline in abundance of 50–80%; vulnerable, decline in abundance of 30–50%; least concern or near threatened, decline in abundance of 0–30%), we assessed 3 methods used to detect declines solely from estimates of abundance: use of just 2 estimates of abundance; use of linear regression on a time series of abundance; and use of state‐space models on a time series of abundance. We generated simulation data from empirical estimates of the typical variability in abundance and assessed the 3 methods for classification errors. The estimates of the proportion of falsely detected declines for linear regression and the state‐space models were low (maximum 3–14%), but 33–75% of small declines (30–50% over 15 years) were not detected. Ignoring uncertainty in estimates of abundance (with just 2 estimates of abundance) allowed more power to detect small declines (95%), but there was a high percentage (50%) of false detections. For all 3 methods, the proportion of declines estimated to be >80% was higher than the true proportion. Use of abundance data to detect species at risk of extinction may either fail to detect initial declines in abundance or have a high error rate.     

Extrapolating cetacean densities to quantitatively assess human impacts on populations in the high seas

As human activities expand beyond national jurisdictions to the high seas, there is an increasing need to consider anthropogenic impacts to species inhabiting these waters. The current scarcity of scientific observations of cetaceans in the high seas impedes the assessment of population‐level impacts of these activities. We developed plausible density estimates to facilitate a quantitative assessment of anthropogenic impacts on cetacean populations in these waters. Our study region extended from a well‐surveyed region within the U.S. Exclusive Economic Zone into a large region of the western North Atlantic sparsely surveyed for cetaceans. We modeled densities of 15 cetacean taxa with available line transect survey data and habitat covariates and extrapolated predictions to sparsely surveyed regions. We formulated models to reduce the extent of extrapolation beyond covariate ranges, and constrained them to model simple and generalizable relationships. To evaluate confidence in the predictions, we mapped where predictions were made outside sampled covariate ranges, examined alternate models, and compared predicted densities with maps of sightings from sources that could not be integrated into our models. Confidence levels in model results depended on the taxon and geographic area and highlighted the need for additional surveying in environmentally distinct areas. With application of necessary caution, our density estimates can inform management needs in the high seas, such as the quantification of potential cetacean interactions with military training exercises, shipping, fisheries, and deep‐sea mining and be used to delineate areas of special biological significance in international waters. Our approach is generally applicable to other marine taxa and geographic regions for which management will be implemented but data are sparse.