Understanding the distribution and extent of suitable habitats is critical for the conservation of endangered and endemic taxa. Such knowledge is limited for many Central African species, including the rare and globally threatened Grey-necked Picathartes Picathartes oreas, one of only two species in the family Picathartidae endemic to the forests of Central Africa. Despite growing concerns about land-use change resulting in fragmentation and loss of forest cover in the region, neither the extent of suitable habitat nor the potential species’ distribution is well known. We combine 339 (new and historical) occurrence records of Grey-necked Picathartes with environmental variables to model the potential global distribution. We used a Maximum Entropy modelling approach that accounted for sampling bias. Our model suggests that Grey-necked Picathartes distribution is strongly associated with steeper slopes and high levels of forest cover, while bioclimatic, vegetation health, and habitat condition variables were all excluded from the final model. We predicted 17,327 km2 of suitable habitat for the species, of which only 2,490 km2 (14.4%) are within protected areas where conservation designations are strictly enforced. These findings show a smaller global distribution of predicted suitable habitat forthe Grey-necked Picathartes than previously thought. This work provides evidence to inform a revision of the International Union for Conservation of Nature (IUCN) Red List status, and may warrant upgrading the status of the species from “Near Threatened” to “Vulnerable”.

Since it was first described in 1901, the Príncipe thrush Turdus xanthorhynchus has been rare and restricted to the native forest in the south of Príncipe Island. The Obô giant land snail Archachatina bicarinata, however, was widespread across the island and at least locally abundant until the 1990s. Since then its population has collapsed, and now, like the thrush, it is also restricted to the native forest in the south of the island. Using species distribution modelling, we show that both species are currently strongly associated with rugged and remote areas of native forest at high altitudes. We argue that their current distribution might be negatively affected by anthropogenic pressures, as both are harvested, and also because invasive alien species are expected to have deleterious effects on these species, although further studies are needed to clarify interactions between these native and introduced species. The diachronic stories of these species highlight an overlooked value of native ecosystems: their role in the conservation of widespread species that might be unable to use anthropogenic landscapes in the future. They also reinforce the need for protected areas that strive to exclude most human activities in the context of particularly sensitive biodiversity, as is often the case on oceanic islands.

Across South Africa, Lesotho, and Eswatini, long-term citizen science atlas data have suggested concerning declines in the population of Black Stork Ciconia nigra. Unlike the Asian and European populations, the southern African Black Stork population is described as resident and is listed as “Vulnerable” in South Africa, Lesotho, and Eswatini. Here we report on surveys of historical nesting locations across northern South Africa, finding evidence for nest site abandonment and limited evidence of recent breeding. We undertook detailed species distribution modelling within a maximum entropy framework, using occurrence records from the BirdLasser mobile app. We cross-validated the models against information in the Southern African Bird Atlas Project (SABAP2) database, highlighting Lesotho as an important potential breeding area. Additionally, we used SABAP2 to assess population trends by investigating interannual patterns in reporting rate. Comparing current reporting rates with those from SABAP1 (1987–1992), we found that there has been a dramatic decrease. We noted that a large proportion of the population occurs outside the breeding range during the breeding season, suggesting a considerable non-breeding population, especially in the extensive wildlife refuge of the Kruger National Park. The slow declines observed might be indicative of a population which is not losing many adults but is failing to recruit significant numbers of juveniles due to limited breeding. Using densities derived from transect surveys, we used predictive models to derive estimates of breeding range carrying capacity and a population estimate, which suggested declines to numbers around 600 for this subregion. Minimising disturbance at breeding sites of this cliff-nesting species and improving water quality at key population strongholds are pathways to improving the status of the species in the subregion.

Species baseline information is required for effective biodiversity conservation. Here we provide sound ecological data to support effective conservation of the Principe Scops-owl, Otus bikegila Melo, Freitas, Verbelen, Costa, Pereira, Fuchs, Sangster, Correia, de Lima & Crottini 2022, discovered in 2016. We mapped the observed and potential distribution, inferred habitat associations, estimated population size, and assessed the International Union for Conservation of Nature (IUCN) Red List category for this species. Surveys were carried out across Príncipe Island during the long and short dry seasons, recording owl presence in sampling points along transects. These data were used to model distribution, revealing that the Principe Scops-owl is restricted to 34.4 km2 in the south of the island, inside the Príncipe Obô Natural Park. Most records were in lower altitude native forest. Remoteness was the most important variable to explain presence at island level, followed by land use, which showed that the species is restricted to forest, clearly depending on native forest. Distance sampling using different models suggested a population density ranging between 33.4 individuals/km2 (95% CI: 23.6–47.2) and 46.4 individuals/km2 (95% CI: 29.3–73.6), which extrapolates to an estimated population size ranging from 1,149 individuals (95% CI: 813–1,623) to 1,597 individuals (95% CI: 1,007–2,533). We propose that the species is classified as “Critically Endangered” due to the small extent of occurrence, coupled with occurrence in a single location and inferred continuing declines in the extent of occurrence, area of occupancy, number of mature individuals and area, extent, and quality of habitat. Effective conservation of the Principe Scops-owl requires regular monitoring and further studies focusing on reproduction and potential nest predation by introduced mammals. Widespread support for the conservation of the Natural Park is vital to ensure the protection of this species and the endemic-rich native forests of Príncipe on which it depends.

The prevention of tropical forest deforestation is essential for mitigating climate change. We tested the machine learning algorithm Maxent to predict deforestation across the Peruvian Amazon. We used official annual 2001–2019 deforestation data to develop a predictive model and to test the model’s accuracy using near-real-time forest loss data for 2020. Distance from agricultural land and distance from roads were the predictor variables that contributed most to the final model, indicating that a narrower set of variables contribute nearly 80% of the information necessary for prediction at scale. The permutation importance indicating variable information not present in the other variables was also highest for distance from agricultural land and distance from roads, at 40.5% and 14.3%, respectively. The predictive model registered 73.2% of the 2020 early alerts in a high or very high risk category; less than 1% of forest cover in national protected areas were registered as very high risk, but buffer zones were far more vulnerable, with 15% of forest cover being in this category. To our knowledge, this is the first study to use 19 years of annual data for deforestation risk. The open-source machine learning method could be applied to other forest regions, at scale, to improve strategies for reducing future deforestation.

Asian Houbara Chlamydotis macqueenii is a vulnerable flagship species specific to steppe, desert and semi-desert habitats of the Middle East and Central Asia. Iran provides a critical corridor in the middle of Asian Houbara’s migratory route and also hosts a relatively large proportion of wintering and breeding populations. The aim of this study was to assess the distribution and habitat suitability of both wintering and breeding populations of Asian Houbara in Iran and evaluate the effectiveness of the existing protected area network for long-term protection of the species. For this purpose, 644 occurrence points for wintering and 216 points for breeding birds were collected from 17 Iranian provinces during 2015 to 2017. We then used a consensus species distribution modelling (SDM) approach using 11 uncorrelated environmental variables to explore the distribution of Asian Houbara habitats. Results showed that climatic and topographic variations have the most significant influence on the regional-scale distribution of Asian Houbara. Of the suitable habitats recognised for the wintering and breeding populations, 40.6% and 29.6% respectively overlapped with the extent of the protected network. A high level of spatial niche similarity (78%) was observed between wintering and breeding populations. The central, eastern and south-central Iranian semi-arid regions and desert landscapes hosted the majority of both wintering and breeding houbara occurrences. Results of this study could be used for adopting direct management planning and raising the protection level of important no-hunting areas.

The Caatinga of north-eastern Brazil is the largest and most diverse seasonally dry tropical forest in the Americas and is home to numerous endemic species. However, only 1.2% of the area is under full protection, and given the ongoing decline of this biome there is an urgent need to expand the protected area network. The Endangered Caatinga howler monkey Alouatta ululata is almost endemic to the Caatinga, and is a potential umbrella species for the protection of its biodiversity. Using all available distribution data and our own surveys we applied Maxent and Zonation spatial modelling to identify the range of A. ululata, and priority conservation areas for the species, maximizing habitat quality and connectivity, and minimizing conservation constraints. The top 10% priority areas cover 34,400 km2 and mostly coincide with good remnants of Caatinga. Only priority areas in the northern part of the species’ range are protected, so it is essential to create new protected areas in the centre and south of the range. Maxent modelling indicates that the species depends on good tree cover, but even inside protected areas we observed recent deforestation, illustrating the urgency to improve management. Maxent also indicated that aridity limits the species' range, and therefore the ongoing aridification of the Caatinga is a threat to its future. The protection of A. ululata requires establishing new protected areas in priority locations and improving management of existing protected areas. Preservation of priority areas for the Caatinga howler monkey also represents an opportunity for the conservation of other important biodiversity in the region.

Forest ecosystems are rich in biodiversity and provide valuable ecosystem services, but are declining worldwide. Malus niedzwetzkyana, an Endangered wild relative of domesticated apples, is an important component of the walnut–fruit forests of Central Asia. Its iconic pink blossom and genetic properties give it special cultural and scientific significance, but livestock grazing and firewood collection threaten its survival. The conservation of the species and its native forest ecosystem is critical and urgent. This study provides information on the ecology and population of M. niedzwetzkyana and the threats affecting its habitat, improving our understanding of its distribution and proposing measures to reduce threats. We collected ecological data and assessed population structure and threats at four forest sites in southern Kyrgyzstan. We mapped 149 individuals, creating the largest known dataset for this species. We developed species distribution models for M. niedzwetzkyana to identify climatically suitable regions and potential areas for restoration. Sary-Chelek Biosphere Reserve contained the largest expanse of pristine forest and the most stable M. niedzwetzkyana population, followed by Kara-Alma Forestry Unit. Forests in the Gava Forestry Unit and Dashman Reserve were most extensively damaged by humans and livestock. The wild apple's favoured habitat was south-west facing slopes with a gradient < 30° and a relatively open canopy. Overall, the study population was vulnerable to extinction with limited regeneration potential. We recommend short-term population enhancement through planting projects and increased protection of individuals in pristine sites. Community-based conservation initiatives should be prioritized in extensively damaged sites, and larger-scale reforestation of these forests needs to be considered.

Organisms adapted to life at high elevations are particularly threatened by climate change, which can cause them to become isolated on mountain tops, yet their responses may vary according to their position in the food chain and their ecological flexibility. Predicting the future distributions of such organisms requires fine-tuned species-specific models. Building on a previous ecological niche model, we explored shifts in the suitability of habitats for the Endangered Andean cat Leopardus jacobita, and assessed how these will be represented within existing protected areas in the future. Using a robust set of presence records and corrected climate surfaces, we applied the Maxent algorithm to model habitat suitability for this carnivore and for its preferred prey, the mountain viscacha Lagidium viscacia. Our predictions indicate that the areas climatically suitable for Andean cats could contract by up to 30% by 2080 under the most pessimistic scenario, with an overall upwards shift of 225 m and a polewards displacement of 98–180 km. The predicted range contraction was more pronounced in the species’ core range, in the Bolivian and Peruvian Andes, whereas suitable conditions may increase in the southern range in Patagonia. Bolivia and Peru are predicted to suffer the most marked decline in habitat representativeness within protected areas. The southern range appears to be less vulnerable to climate change, offering opportunities for the conservation of this genetically distinct population. We discuss the value and limitations of using species distribution modelling to assess changes in the potential distribution and conservation status of this and other Andean species.

Monitoring species ranges and suitable and occupied habitat are core components of biogeography, ecology and conservation biology, but it is difficult to do for rare, cryptic, wide-ranging, migratory or nomadic species. We present a transparent and objective process to combine multiple types of locality data (peer-reviewed and grey literature, museum collections, camera-trap inventories, and citizen science reports). We illustrate the advantages of this pooled approach by assessing change in range and patch occupancy for a data-poor and threatened nomadic keystone species, the bearded pig Sus barbatus, in Borneo, Sumatra and Peninsular Malaysia. We used a collated set of all occurrence observations (n = 240) to create minimum convex polygons for forested habitats for two time periods. We evaluated confidence that a patch was truly occupied by the overlap among data types. We found that 62% of the forest habitat of the Sumatran bearded pig S. barbatus oi was lost during 1990–2010 and that its range contracted by 76%; the Bornean bearded pig S. barbatus barbatus lost 23 and 24% of its forest habitat and range, respectively, and in Peninsular Malaysia the 93% range collapse of this subspecies during 1985–2010 is more severe than the 33% habitat loss alone would suggest. We conclude that integrating data types can improve mapping of the ranges of many data-poor species.

Climate variables are commonly used to predict suitability for species occurrence, but local processes, such as landscape changes, may affect habitat suitability. We identified levels of exposure to deforestation of suitable climatic areas for eight bat species in the Brazilian Cerrado and explored how sensitivity to land-use changes could reduce their persistence. We created scenarios of sensitivity to land-use changes using theoretical species persistence thresholds to natural vegetation loss in landscapes (70%, 50% and 30% of loss). We also assessed sensitivity to land-use changes using empirical data. Species are under higher exposure to land-use changes in the southern Cerrado, a region more affected by humans due to its proximity to major urban areas. Changes in land use in the Cerrado mostly affect Myotis nigricans, Artibeus cinereus and Platyrrhinus lineatus. Empirically derived scenarios encountered significant thresholds at 50% of natural vegetation loss in landscapes for Artibeus lituratus and P. lineatus. Deforestation has already affected a half of the Cerrado area, but in terms of possibly vulnerable suitable areas, a larger proportion has been lost, amounting to up to 80% of the suitable area. We propose that information on species-specific sensitivity thresholds to habitat loss and on the exposure of suitable landscapes to land-use changes can be useful to assessing species vulnerability.

Species with specific roosting, foraging or breeding requirements are particularly vulnerable to habitat loss and degradation. For bats, the availability and environmental condition of caves can be a limiting factor. The cave specialist Natalus macrourus (formerly Natalus espiritosantensis) is categorized as Near Threatened on the IUCN Red List but as Vulnerable in Brazil, based on a projected population reduction and a decline in its area of occupancy, extent of occurrence and/or quality of habitat. There is a lack of knowledge about the species’ distribution, natural history and ecology, information that is required for conservation. Using new occurrence data and potential distribution modelling we evaluated the distribution of N. macrourus in Brazil, analysed pressures on and threats to the species, and assessed the species’ conservation needs. Natalus macrourus is positively associated with areas with higher probability of cave occurrence and negatively associated with areas of high variation in mean daily temperature and mean annual rainfall. Areas with high environmental suitability for N. macrourus correspond to only 3% of the potential distribution modelled. We estimate that the species has already lost 54% of its natural habitat and that there is < 35% of habitat remaining in areas with high environmental suitability. We calculated that approximately half of the caves in areas with high environmental suitability are < 5 km from mining operations and only 4% of the species’ potential distribution lies within protected areas. Given the strong association of N. macrourus with caves, it is important to protect these habitats, and we recommend that caves where the species is present should receive immediate protection.

IUCN Red List assessments are important for conservation and management initiatives. However, status assessments are often challenging because of poor sampling between biogeographical regions. Researchers sometimes assess poorly known species, which can have unforeseen ramifications, including the trade of rare and cryptic species under common species names. Here, we address this issue in relation to economically important reptile species in Indonesia. We reviewed examples of single species categorized as Least Concern for which the assessments probably encompassed multiple closely related species. We also examined Red List assessments that utilized species distribution modelling techniques, and identified biogeography as a major barrier to using such methods. To test how biogeography may affect status assessments we used our own model lizard system from Indonesia, taking an integrative phylogeographical approach to quantify status assessments under contrasting scenarios. We show that failure to account for biogeographical breaks leads to significant variation in Red List status. Our model system fluctuates from Least Concern to Endangered, depending upon whether biogeographical boundaries are considered in taxonomic evaluations. We identify Sauria (lizards) and Serpentes (snakes) as major lineages requiring taxonomic and conservation attention in Indonesia. We also make the following recommendations: (1) Indonesia's trade quotas should further subdivide management zones to account for gaps in taxonomic evaluations; (2) genetic sampling should be considered a high priority during wildlife exportation processes from poorly studied geographical areas; and (3) continuation of thorough biological inventory is critical for conservation initiatives across heterogeneous mountain and island landscapes.

Protected areas (PAs) have been selected using either subjective or objective criteria applied to an extremely limited subset of biodiversity. Improved availability of species distribution data, better statistical tools to predict species distributions and algorithms to optimize spatial conservation planning allow many impediments to be overcome, particularly on small islands. This study analyses whether 219 species are adequately protected by PAs on Pico Island (the Azores, Portugal), and if they are as efficient as possible, maximizing species protection while minimizing costs. We performed distribution modelling of species’ potential distributions, proposed individual conservation targets (considering the context of each species in the archipelago and their current conservation status) to determine the efficiency of current PAs in meeting such targets and identify alternative or complementary areas relevant for conservation. Results showed that current PAs do not cover all taxa, leaving out important areas for conservation. We demonstrate that by using optimization algorithms it is possible to include most species groups in spatial conservation planning in the Azores with the current resources. With increasing availability of data and methods, this approach could be readily extended to other islands and regions with high endemism levels.

A population of Eurasian lynx Lynx lynx was established by reintroductions in the Bohemian Forest Ecosystem in the 1970s and 1980s. The most recent information on the population status indicates that the distribution has stagnated since the late 1990s, for unknown reasons. We assessed the availability of suitable habitat along the Austrian–German–Czech border, and hypothesized that the Bohemian–Bavarian lynx population is not in equilibrium with habitat suitability. Based on global positioning system data from 10 radio-collared lynx, we used a maximum entropy approach to model suitable habitat. Variables reflecting anthropogenic influence contributed most to the model and were negatively associated with the occurrence of lynx. We evaluated the model prediction using independent records of lynx from monitoring in Bavaria, Germany. Using our habitat approach we estimated the area of potential habitat, based on a mean annual home range of 445 km2 for males and 122 km2 for females. Our results indicated there were 12,415 km2 of suitable habitat, distributed among 13 patches, for a potential population of c. 142 (93–160) resident lynx. We assessed connectivity via least-cost paths and found that all suitable patches could be reached by the lynx. A comparison with the current distribution of lynx, however, confirms that a significant proportion of suitable habitat is not occupied, which indicates that the distribution is limited by factors other than habitat, with illegal killing being the most likely cause. Our study provides crucial information for the development of a conservation strategy and regional planning for the Bohemian–Bavarian lynx population.

For over a century the foundation of biological conservation has been the development of open space networks either through outright public land acquisition or appropriate management of private lands. Because both approaches come with significant trade-offs, it is critical to understand which species are found across various land ownership types so that policy tools and management actions can efficiently be targeted to do the most good. In this paper, presence-only biological data were used to create species distribution maps for 18 imperilled forest bird species that breed within the deciduous forests of New Jersey (USA). These maps, combined with publicly available, spatially explicit information on land ownership, document who owns the habitat relied on by each of these 18 species. There were significant variations in both species- and guild-specific reliance on public versus private lands, with the latter preferentially supporting nearly twice as many species as the former. Subcategories of land ownership provided support for the role of both state-owned forests and privately-owned agricultural lands in forest bird conservation; however, each landownership type supports a distinct set of species. While explicitly recognizing the need to employ diverse conservation strategies, the approach provides a solid framework for structuring forest conservation planning and policy at regional scales.

Kelps, fucoids and other large brown seaweeds are common and important features of temperate coastal zones. The British Isles is a centre for seaweed diversity in the NE Atlantic, but, despite numerous surveys, an incomplete picture of the distribution remains. Survey data and herbarium specimens were used to examine the environmental preference of 15 species of large brown seaweeds, covering the orders Laminariales (kelps), Fucales (wracks) and one species of Tilopteridales. Habitat suitability models were developed to estimate broad-scale distribution and area of habitat created by these species around the British Isles. Topographic parameters were important factors limiting distributions. Generally, temperature did not appear to be a limiting factor, probably because the British Isles lies in the centre of the NE Atlantic distribution for most species, and not at climatic tolerance limits. However, for the recent migrant Laminaria ochroleuca, temperature was found to be important for the model, thus range expansion could continue northwards provided dispersal is possible. In contrast, the widespread Alaria esculenta showed a negative association with warmer summer temperatures. The total potential habitat around the British and Irish coastline is more than 19,000 km2 for kelps and 11,000 km2 for wracks, which represents a significant habitat area similar in scale to British broadleaf forest. We conclude that large brown algal species need to be managed and conserved in a manner that reflects their scale and importance.

Climate change and habitat fragmentation are considered key pressures on biodiversity, and mammalian carnivores with a limited geographical distribution are particularly vulnerable. The kodkod Leopardus guigna, a small felid endemic to the temperate forests of southern Chile and Argentina, has the smallest geographical range of any New World felid. Although the species occurs in protected areas in both countries, it is not known how well these areas protect the kodkod either currently or under climate change scenarios. We used species distribution models and spatial analyses to assess the distribution of the kodkod, examining the effects of changes in human land use and future climate change. We also assessed the species’ present representation in protected areas and in light of climate change scenarios. We found that the kodkod has already lost 5.5% of its range as a result of human land use, particularly in central areas of its distribution with intermediate habitat suitability. Climate change, together with human land use, will affect 40% of the kodkod's present potential distribution by the year 2050. Currently, 12.5% of the species’ potential distribution lies in protected areas and this will increase to 14% in the future. This increase does not, however, mean an increase in protected habitat but rather a reduction of the species' total potential range; a relatively larger percentage will be protected in Argentina than in Chile but the species is more susceptible to extinction in Argentina and the Chilean Matorral.

Niche modelling software can be used to assess the probability of detecting a population of a plant species at a certain location. In this study, we used the distribution of the wild relatives of lettuce (Lactuca spp.) to investigate the applicability of Maxent species distribution models for collecting missions. Geographic origin data of genebank and herbarium specimens and climatic data of the origin locations were used as input. For Lactuca saligna, we varied the input data by omitting the specimens from different parts of the known distribution area to assess the robustness of the predicted distributions. Furthermore, we examined the accuracy of the modelling by comparing the predicted probabilities of population presence against recent expedition data for the endemic Lactuca georgica and the cosmopolitan Lactuca serriola. We found Maxent to be quite robust in its predictions, although its usefulness was higher for endemic taxa than for more widespread species. The exclusion of occurrence data from the perceived range margins of the species can result in important information about local adaptation to distinct climatic conditions. We discuss the potential for enhanced use of Maxent in germplasm collecting planning.