A multi-scale, multi-species approach for assessing effectiveness of habitat and connectivity conservation for endangered felids

Biological Conservation, Journal Year: 2020, Volume and Issue: 245, P. 108523 - 108523

Published: April 4, 2020

Language: Английский

---

Multi-scale habitat modelling and predicting change in the distribution of tiger and leopard using random forest algorithm

Scientific Reports, Journal Year: 2020, Volume and Issue: 10(1)

Published: July 10, 2020

Tigers and leopards have experienced considerable declines in their population due to habitat loss fragmentation across historical ranges. Multi-scale suitability models (HSM) can inform forest managers aim conservation efforts at increasing the suitable for tigers by providing information regarding scale-dependent habitat-species relationships. However current gap of knowledge about ecological relationships driving species distribution reduces applicability traditional classical statistical approaches such as generalized linear (GLMs), or occupancy surveys produce accurate predictive maps. This study investigates multi-scale impacts future climate change on using a machine-learning algorithm random (RF). The recent advancements algorithms provide powerful tool building even when little is available species. We collected occurrence data camera traps indirect evidence animal presences (scats) field over 2 years rigorous sampling used (RF) predict maps tiger leopard under climatic scenarios. developed niche overlap based recently assess patterns similarity between leopards. Tiger utilized resources broadest spatial scales (28,000 m). Our model predicted 23% RCP 8.5 Scenario (2050). modeling provides valuable disturbed human-dominated landscapes concerning two large felid importance. These areas may act refugee habitats carnivores thus should be focus also methodological framework similar multi-species monitoring programs robust more machine learning forest.

Language: Английский

---

Species and space: a combined gap analysis to guide management planning of conservation areas

Landscape Ecology, Journal Year: 2020, Volume and Issue: 35(7), P. 1505 - 1517

Published: May 22, 2020

Language: Английский

---

Not seeing the forest for the trees: Generalised linear model out-performs random forest in species distribution modelling for Southeast Asian felids

Ecological Informatics, Journal Year: 2023, Volume and Issue: 75, P. 102026 - 102026

Published: Feb. 18, 2023

Species Distribution Models (SDMs) are a powerful tool to derive habitat suitability predictions relating species occurrence data with features. Two of the most frequently applied algorithms model species-habitat relationships Generalised Linear (GLM) and Random Forest (RF). The former is parametric regression providing functional models direct interpretability. latter machine learning non-parametric algorithm, more tolerant than other approaches in its assumptions, which has often been shown outperform algorithms. Other have developed produce robust SDMs, like training bootstrapping spatial scale optimisation. Using felid presence-absence from three study regions Southeast Asia (mainland, Borneo Sumatra), we tested performances SDMs by implementing four modelling frameworks: GLM RF bootstrapped non-bootstrapped data. With Mantel ANOVA tests explored how combinations influenced their predictive performances. Additionally, scale-optimisation responded species' size, taxonomic associations (species genus), area algorithm. We found that choice algorithm had strong effect determining differences between SDMs' predictions, while no effect. followed species, were main factors driving scales identified. trained showed higher performance, however, revealed significant only explaining variance observed sensitivity specificity and, when interacting bootstrapping, Percent Correctly Classified (PCC). Bootstrapping significantly explained specificity, PCC True Skills Statistics (TSS). Our results suggest there systematic identified produced vs. RF, but neither approach was consistently better other. divergent inconsistent abilities analysts should not assume inherently superior test multiple methods. implications for SDM development, revealing inconsistencies introduced on optimisation, selecting broader RF.

Language: Английский

---

Meta‐replication, sampling bias, and multi‐scale model selection: A case study on snow leopard (Panthera uncia) in western China

Ecology and Evolution, Journal Year: 2020, Volume and Issue: 10(14), P. 7686 - 7712

Published: July 1, 2020

Abstract Replicated multiple scale species distribution models (SDMs) have become increasingly important to identify the correct variables determining and their influences on ecological responses. This study explores multi‐scale habitat relationships of snow leopard ( Panthera uncia ) in two areas Qinghai–Tibetan Plateau western China. Our primary objectives were evaluate degree which relationships, expressed by predictors, scales response, magnitude effects, consistent across or locally landcape‐specific. We coupled univariate optimization maximum entropy algorithm produce multivariate SDMs, inferring relative suitability for ensembling top performing models. optimized SDMs based average omission rate ensembles’ overlap with a simulated reference model. Comparison highlighted landscape‐specific responses limiting factors. These dependent effects hydrological network, anthropogenic features, topographic complexity, heterogeneity landcover patch mosaic. Overall, even accounting specific local differences, we found general landscape attributes associated requirements, consisting positive association uplands ridges, aggregated low‐contrast landscapes, large extents grassy herbaceous vegetation. As means performance bias correction methods, explored three datasets showing range intensities. The corrections depends intensity; however, density kernels offered reliable strategy under all circumstances. reveals response leopards environmental confirms role meta‐replicated designs identification spatially varying Furthermore, this makes contributions ongoing discussion about best approaches sampling correction.

Language: Английский

---

Habitat suitability and connectivity implications for the conservation of the Persian leopard along the Iran–Iraq border

Ecology and Evolution, Journal Year: 2021, Volume and Issue: 11(19), P. 13464 - 13474

Published: Aug. 30, 2021

Abstract Habitat fragmentation has major negative impacts on wildlife populations, and the connectivity could reduce these impacts. This study was conducted to assess habitat suitability structural of Persian leopard along Iran–Iraq border (i.e., Zagros Mountains) compare situation identified core habitats with existing conservation areas (CAs). An ensemble modeling approach resulting from five models used predict suitability. To identify corridors border, factorial least‐cost path analyses were applied. The results revealed that topographic roughness, distance CAs, annual precipitation, vegetation/cropland density, rivers most influential variables for predicting occurrence in area. By an estimated dispersal 82 km (suggested by previous studies), three (two cores Iran one Iraq). largest located south center area, which had highest priorities. maintained within Iraqi side. Only about one‐fifth detected relative protected CAs Detected this be appropriate road map accomplish network regarding conservation. Establishing transboundary particularly is strongly recommended conserve large carnivores, including leopard.

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

---

Prey availability modulates predicted range contraction of two large felids in response to changing climate

Biological Conservation, Journal Year: 2021, Volume and Issue: 255, P. 109018 - 109018

Published: Feb. 24, 2021

Language: Английский

---

Identifying core habitats and corridors of a near threatened carnivore, striped hyaena (Hyaena hyaena) in southwestern Iran

Scientific Reports, Journal Year: 2022, Volume and Issue: 12(1)

Published: March 2, 2022

Abstract Conservation of large carnivores requires preservation extensive core habitats and linkages among them. The goal this study was to identify corridors by predicting habitat suitability (an ensemble approach), calculating resistant kernel factorial least-cost path modeling for a relatively unknown carnivore, the striped hyaena in Khuzestan area southwestern Iran. We used procedure spatial randomization test evaluate coincidence road crossing with predicted corridors. results revealed that elevation, distance conservation areas, categorical climate grasslands density were most influential variables occurrence area. In estimated dispersal 70 km, four identified. largest located northeast highest connectivity contribution. Only about 12% 1.5% protected respectively. Predicted corridors, crossed roads represented high risk hyaenas. Adaptive management plan throughout landscape (conservation reducing species mortality on roads) must be considered wildlife managers

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

---