Projected climate change threatens Himalayan brown bear habitat more than human land use

Animal Conservation, Journal Year: 2021, Volume and Issue: 24(4), P. 659 - 676

Published: Feb. 1, 2021

Abstract Climate change and land use jointly are the largest drivers of population declines, range contraction extinction for many species across globe. Wide‐ranging large‐bodied especially vulnerable to habitat loss fragmentation due their typically low densities, reflecting need extensive connected habitats. We used multi‐scale Random Forests machine learning algorithm identify factors driving selection future changes in Himalayan brown bear, an iconic wide‐ranging high conservation interest, a spatial scales. Habitat bears was scale‐dependent, with most variables selected at broad Climatic such as maximum temperature coldest month, minimum warmest month potential evapotranspiration wettest quarter strongly influenced bears. Future projections indicate strong difference between emission scenarios. Alarmingly, our model suggests that scenarios, or without change, may result decline bear >90% by end century. In contrast, scenarios projected reduce <23%, much shifting higher elevations. This study provides integrative understanding scale‐dependent selection, providing critical information prioritizing areas management conservation. Most importantly, imply traditional efforts, situ conservation, will not be sufficient protect climate mitigation. The incorporation mitigation adaptation strategies one pressing priorities biodiversity this region.

Language: Английский

---

A multi-scale Maxent approach to model habitat suitability for the giant pandas in the Qionglai mountain, China

Global Ecology and Conservation, Journal Year: 2021, Volume and Issue: 30, P. e01766 - e01766

Published: Aug. 20, 2021

Wild animals usually respond to different landscape features at spatial scales. The adoption of multi-scale modeling frameworks in habitat suitability studies have been shown improve model performance and provide greater insights into relationships between species components. Although the advantage modeling, implementation this framework lagged considerably. In present study, we used a approach assess for globally endangered giant panda (Ailuropoda melanoleuca) Qionglai mountain range, Sichuan, China an effort improved species-environment with aim informing conservation efforts. occurrence data collected from Fourth National Giant Panda Survey presence-only, Maxent were pandas. Our results showed that optimal scale identified each environmental variable varied, most variables strongly related relatively fine-scale (≤ 2000 m). Multi-scale models outperformed their analogous single-scale counterparts respect discrimination predictive ability. Additionally, there significant differences predictions model. This study reveals response pandas confirms modeling. Therefore, it is necessary beneficial take dependence consideration future panda.

Language: Английский

---

Investigating Carnivore Guild Structure: Spatial and Temporal Relationships amongst Threatened Felids in Myanmar

ISPRS International Journal of Geo-Information, Journal Year: 2021, Volume and Issue: 10(12), P. 808 - 808

Published: Nov. 30, 2021

The co-occurrence of felid species in Southeast Asia provides an unusual opportunity to investigate guild structure and the factors controlling it. Using camera-trap data, we quantified space use, temporal activity, multi-dimensional niche overlap tiger, clouded leopard, Asiatic golden cat, marbled leopard cat Htamanthi Wildlife Sanctuary, Myanmar. We hypothesised that spatio-temporal behaviour smaller cats would reflect avoidance larger cats, similar-sized members partition their niches or time reduce resource competition. Our approach involved modelling single-species occupancy, pairwise spatial using Bayesian inference, activity with kernel density estimation, multivariate analyses. assembly appeared be partitioned mainly on a rather than dimension, no significant evidence mesopredator release was observed. Nonetheless, association between three mesopredators inversely related similarity body sizes. largest differences use occurred smallest species. This study offers new insight into carnivore adds substantially knowledge five least known felids conservation concern.

Language: Английский

---

Combining multiscale niche modeling, landscape connectivity, and gap analysis to prioritize habitats for conservation of striped hyaena (Hyaena hyaena)

PLoS ONE, Journal Year: 2022, Volume and Issue: 17(2), P. e0260807 - e0260807

Published: Feb. 10, 2022

Identifying spatial gaps in conservation networks requires information on species-environment relationships, and prioritization of habitats corridors. We combined multi-extent niche modeling, landscape connectivity, gap analysis to investigate scale-dependent environmental identify core corridors for a little-known carnivore Iran, the striped hyaena (Hyaena hyaena). This species is threatened Iran by road vehicle collisions direct killing. Therefore, understanding factors that affect its habitat suitability, pattern distribution, connectivity among them are prerequisite steps delineate strategies aiming at human-striped co-existence. The results showed highest predictive power extent was obtained sizes 4 2 km, respectively. Also, revealed number changed with increasing dispersal distance, approximately 21% found support 15-17% overlapped areas. Given body size species, mobility, lack significant specialization we conclude this would be more strongly influenced changes amount rather than configuration. Our approach scale variables ability must accounted efforts prioritize corridors, designing could facilitate through identification habitats, establishment areas, mitigating conflicts

Language: Английский

---

Predicting biodiversity richness in rapidly changing landscapes: climate, low human pressure or protection as salvation?

Biodiversity and Conservation, Journal Year: 2020, Volume and Issue: 29(14), P. 4035 - 4057

Published: Oct. 9, 2020

Abstract Rates of biodiversity loss in Southeast Asia are among the highest world, and Indo-Burma South-Central China Biodiversity Hotspots rank world’s most threatened. Developing robust multi-species conservation models is critical for stemming both here globally. We used a large geographically extensive remote-camera survey multi-scale, multivariate optimization species distribution modelling to investigate factors driving across these two adjoining hotspots. Four major findings emerged from work. (i) identified clear spatial patterns richness, with main biodiverse centres Thai-Malay Peninsula mountainous region Southwest China. (ii) Carnivores particular, ungulates lesser degree, were strongest indicators richness. (iii) Climate had largest effect on biodiversity, followed by protected status human footprint. (iv) Gap analysis between model current system areas revealed that majority supporting predicted not protected. Our results highlighted several key locations should be prioritized expanding area network maximize effectiveness. demonstrated importance switching single-species approaches highlight high priority conservation. In addition, since mostly occur over multiple countries, we also advocate paradigmatic focus transboundary planning.

Language: Английский

---

Inferring predator–prey interactions from camera traps: A Bayesian co‐abundance modeling approach

Ecology and Evolution, Journal Year: 2022, Volume and Issue: 12(12)

Published: Dec. 1, 2022

Abstract Predator–prey dynamics are a fundamental part of ecology, but directly studying interactions has proven difficult. The proliferation camera trapping enabled the collection large datasets on wildlife, researchers face hurdles inferring from observational data. Recent advances in hierarchical co‐abundance models infer species while accounting for two species' detection probabilities, shared responses to environmental covariates, and propagate uncertainty throughout entire modeling process. However, current approaches remain unsuitable interacting whose natural densities differ by an order magnitude have contrasting such as predator–prey interactions, which introduce zero inflation overdispersion count histories. Here, we developed Bayesian N‐mixture model that is suitable interactions. We accounted excessive zeros histories using informed zero‐inflated Poisson distribution abundance formula including random effect per sampling unit occasion probability formula. demonstrate with these modifications outperform alternative approaches, improve goodness‐of‐fit, overcome parameter convergence failures. highlight its utility 20 10 tropical forest landscapes Southeast Asia estimate four relationships between tigers, clouded leopards, muntjac sambar deer. Tigers had negative abundance, providing support top‐down regulation, leopards positive deer, likely driven unmodelled covariates like hunting. This approach quantify widely applicable across species, ecosystems, may be useful forecasting cascading impacts following widespread predator declines. Taken together, this facilitates nuanced mechanistic understanding food‐web ecology.

Language: Английский

---

Variable importance and scale of influence across individual scottish wildcat hybrid habitat models

Ecological Modelling, Journal Year: 2024, Volume and Issue: 491, P. 110698 - 110698

Published: March 29, 2024

Understanding the scale dependence of species-habitat relationships is an important area research in species distribution modeling. There has been little focused on how habitat selection may depend individual variation among organisms, geographical location and ecological context that location. Furthermore, known about extent drivers heterogeneity individuals a inhabiting different contexts, few studies have compared variable importance spatially replicated framework. Two most factors for interpreting models include: (1) relative variables model (2) spatial at which each largest influence. Based existing evidence we hypothesize priori landcover will generally be predictors, followed by topography, then soil type (which influence both vegetation prey), Normalized Difference Vegetation Index (NDVI) as indicator total density perhaps proxy prey density, cover rabbit abundance. We also expected there would consistent patterns across wildcat hybrid related to groups. topographical features selected broad scales, they broad-scale climatic conditions. land classes relatively scales given past showing influences scales. NDVI finer their resources limiting conditions within landscapes. Finally, abundance linear affect occurrence finest these are vary over short distances strongly behavior use. Our results were with hypothesis consistency regarding or groups predictors Scotland. previous our identify clear trend increasing frequency inclusion increasingly This retained increased. suggests monotonic pattern more frequent retention

Language: Английский

---

Comparison of patch-based and synoptic connectivity algorithms with graph theory metrics. A case study in Reggio Calabria metropolitan area (South Italy)

Ecological Informatics, Journal Year: 2024, Volume and Issue: 82, P. 102678 - 102678

Published: June 13, 2024

Predicting and mapping connectivity between habitats populations is critical to addressing habitat loss biodiversity issues. Several strategies in the literature exist understand, restore, preserve ecological connectivity. The main issue of current research identify which modeling are most reliable for planning purposes. Our goals this paper were compare predictions using a wide variety commonly used approaches improve understanding similarities differences these methods. Specifically, we investigated related algorithm, number distribution source points, threshold distance at allowed locations. First, separately applied different methods model same study area. Then, going through series hypotheses, compared models confirm or disprove initial hypotheses. We proposed 4 hypotheses 14 combinations them, hypothesizing that what influences results dispersal thresholds; algorithms, especially kernel, path, graph theory-based approaches; produced by two software tools, UNICOR Graphab; use points derived from synoptic patch-based perspective. found dominant pattern analyses was method analysis, with clear graph-theory relatively little effect due density define capability. This work provides one first comparisons spatial methods, frameworks, parameterizations models. support environmental clarifying movement patterns how predicted networks differ analytical frameworks.

Language: Английский

---

Identifying gaps in the conservation of small wild cats of Southeast Asia

Biodiversity and Conservation, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 20, 2025

Abstract Southeast Asia hosts more felid species than any other region and, although smaller (< 30 kg) felids have important ecological roles, regional conservation has mainly focused on a few charismatic big cats. Information the ecology and status of small is often lacking or geographically limited. We used empirically derived scale-optimized models for seven in three regions (mainland, Borneo Sumatra) to evaluate effectiveness existing protected areas network preserving suitable habitats, map protection. Finally, we assessed whether are good proxies broader terrestrial biodiversity. On mainland, largest most habitats occurred Northern Forest Complex Myanmar between Eastern Myanmar, Laos Vietnam. In these also highlighted areas. Borneo, central highlands Sabah. Sumatra, strongholds habitat suitability were Barisan Mountains, western extent island, highly concentrated within found that aggregated was correlated strongly vertebrate biodiversity single individually, suggesting multiple an association with high overall Overall, our assessment distribution highlights fundamental importance conservation, given associated large extents forest. Our results clarion call expand extent, improve management, remaining core Asia, work enhance protect connectivity them ensure long-term demographic genetic exchange among region’s wildlife populations.

Language: Английский

---

Habitat suitability of common leopard in northern Pakistan

Ecosphere, Journal Year: 2025, Volume and Issue: 16(3)

Published: March 1, 2025

Abstract Habitat fragmentation and loss are considered primary threats to common leopards ( Panthera pardus ) across their geographical range. We investigated anthropogenic environmental factors influencing the habitat suitability of in northern Pakistan using an ensemble model direct indirect leopard signs during 2014–2022. Using location data from 206 sightings model's performance was good (true skill statistic, TSS = 0.52). highest forest cover negatively related density settlements roads. peaked at intermediate elevations (about 1000–2000 m). Based on model, we estimated 4543 km 2 Pakistan, which 3144 (69%) occurred six contiguous patches least 58 (range 65–951 ), minimum size support one female leopard. There patch within a protected area, overall, 36% total areas were as suitable. Our findings suggest that current network does not adequately represent suitable for leopards; increasing expanding area could improve suitability.

Language: Английский

---