Outdoor recreation is expanding across public lands, yet managers lack evidence for how long wildlife responses to recreation persist and whether response duration depends on disturbance type and frequency. Using a field-based acoustic playback experiment in the Bridger–Teton National Forest, Wyoming, USA, we isolated recreation noise from other components of human presence and quantified how long mammal detections took to return to baseline. Across 12 sites over two field seasons, we broadcast motorized and non-motorized recreation sounds at rates mimicking high- and low-use trails, paired with nature-sound controls. We quantified treatment effects as deviations from expected weekly detections at our control sites, defining time-to-return as the first week when deviations overlapped zero. Response duration differed between disturbance types. High-rate motorized playbacks produced the longest suppression: all species except mule deer did not return to baseline within our 15-week season, whereas mule deer recovered by six weeks. For high-rate non-motorized playbacks, all species except for mule deer had suppressed detections until seven weeks while mule deer detections were not different from baseline. For low-rate non-motorized playbacks, mule deer detections returned to baseline at nine weeks while all other species returned to baseline at seven weeks. Species differences indicate that apparent overall recovery can mask prolonged avoidance by less tolerant taxa. Recreation effects persisted for weeks to months and varied with disturbance type and exposure frequency, highlighting how temporal patterns of use influence wildlife responses and where recreational planning of spatial refugia or seasonal restrictions could reduce impacts on sensitive species.
Insular bat populations are particularly vulnerable to human disturbance and environmental change, yet long-term demographic data remain scarce. The Egyptian fruit bat (Rousettus aegyptiacus) is the only native fruit bat in Europe, with Cyprus harbouring one of its most isolated populations. We analysed long-term monitoring data from 21 roosts surveyed between 2005 and 2022 to quantify population trends and evaluate the roles of roost accessibility and food availability. Population trends were estimated using log-linear TRIM models, while generalized linear mixed models tested whether trends differed between roosts with contrasting accessibility. Commercial fruit production was analysed as a proxy for food availability. The population declined from approximately 7200 individuals in 2005 to about 1050 in 2022, corresponding to an overall decline of 85.4% and an average annual decrease of 8.8%. Trend analyses identified three phases: a rapid decline (2005–2011), a slower decline (2012–2017), and a partial recovery after 2018. Colonies in easily accessible roosts declined significantly faster than those in less accessible roosts, indicating that human disturbance likely contributed substantially to the population collapse. Commercial fruit production also declined markedly during this period, coinciding with exceptional drought on Cyprus. Our results suggest that roost disturbance was a major driver of the decline, whereas reduced food availability, likely associated with increasing aridity and climate-driven changes in agricultural production, further contributed to population losses. Our findings highlight the need for strict protection of roosts and suggest that securing food and water resources may represent an important complementary conservation measure.
The Western Ghats of peninsular India, a globally recognized biodiversity hotspot, support exceptional levels of endemism, with protected areas serving as refuges for threatened and endemic species. In the Kerala segment of this mountain system, these reserves occur close to human dominated landscapes and are increasingly exposed to artificial light at night, an emerging ecological stressor affecting nocturnal environments. Using 2024 VIIRS Day/Night Band radiance data together with mapped infrastructure variables, including roads, buildings, places of worship, and points of interest derived from OpenStreetMap (OSM), we quantified light exposure within protected areas and across surrounding distance zones (0–2, 2–5, 5–10, 10–20, and 20–30 km). Radiance varied markedly among sites, with Periyar Tiger Reserve West showing the highest mean radiance (0.369) and maximum (24.0) radiance. Internal illumination was most strongly associated with places of worship density (ρ = 0.496, p = 0.022), indicating localized cultural lighting hotspots. Across the surrounding zones, radiance increased significantly with distance (χ2 = 46.895, p < 0.001), and modelled skyglow contributions were greatest in the broader 10–30 km surrounding landscape rather than in the nearest buffer zones. Road networks and clustered activity nodes emerged as the dominant predictors, demonstrating that light exposure is primarily structured by surrounding landscape context. These findings provide a practical framework for prioritizing monitoring and targeted mitigation to minimize artificial light intrusion into protected areas, while also laying a foundation for future research on its ecological impacts and management.
Dieback is an emerging threat to forests globally, but its long-term impact on biodiversity remains sparsely investigated. In this study, we integrate multiple datasets to assess whether canopy dieback in stringybark eucalypt (Eucalyptus obliqua, L'Hérit. and Eucalyptus baxteri (Benth.) Maiden & Blakely ex J.M. Black) woodlands of the Mount Lofty Ranges in South Australia contributes to declines in bird populations. Using 20 years of monitoring data on bird populations and tree canopy condition, we investigate which woodland species are impacted by canopy dieback. Overall, results suggest that many woodland birds have been gradually declining over time and that canopy dieback may be accentuating declines. While only the Striated Thornbill (Acanthiza lineata) significantly declined over time with the amount of dieback in generalised mixed-effect models, estimated marginal means suggested aggravated abundance declines for species foraging in the canopy or on invertebrates when severe dieback occurs. These findings support previous work suggesting that canopy dieback can negatively affect habitat specialists and canopy-dependent species. Future increases in canopy dieback, driven by increases in drought and heat stress severity under climate change, are therefore likely to add pressure on already declining bird populations that are vulnerable to canopy loss. Consequently, management of biodiversity dependent on tree canopies will need to account for future habitat loss from dieback.
Defaunation causes the loss of top predators, increases the abundance of herbivores, and has impacts on plant populations. Forest herbivores commonly consume small plant individuals, reducing regeneration. Yet, some herbivores target large, reproductive individuals, with potentially more detrimental demographic consequences. In Atlantic Forest fragments in Brazil, abundant capuchin monkeys (Sapajus nigritus) consume the apical meristem of mature Juçara palms (Euterpe edulis), inflicting adult mortality of this vulnerable palm species. Utilizing four years of demographic monitoring of Juçara palms in four populations differing in monkey predation, we constructed Integral Projection Models to assess the effects of herbivory. Populations experiencing more intense predation exhibited consistent declines, with population growth rate (λ) ranging from 0.6 (95% CI: 0.5740–0.6340) to 0.9 (95% CI: 0.8921–0.9159), the lowest λ (λ = 0.6042 and λ = 0.6396) values were observed in the most intensively predated populations. The size structures of the most impacted populations become dominated by smaller-sized individuals, as the large ones are decimated. Projections for the future of these populations indicate local extinction, where the population size reaches zero, within 15 years. Such low population growth rates are unprecedented in the studies on demographic effects of herbivory and predation, showcasing the possible dramatic impacts of multi-trophic interactions. Although capuchin monkeys are a native species that can thrive in the defaunated and deforested context of the Atlantic Forest, their foraging behavior threatens an already vulnerable palm species. These impacts extend further than the palm populations because Juçara fruits are a key resource for frugivorous species in certain communities in the Atlantic Forest.
Conservation of the Black-Footed Ferret (Mustela nigripes) exemplifies the ethical complexities inherent in species recovery efforts. Through a Delphi study, we examined how compassionate conservation practitioners evaluate intrusive management practices—including captive breeding, reintroduction, disease management, and cloning—when confronted with a species facing extinction. Participants assessed six critical stages in the Ferret's conservation history, revealing how ethical frameworks shape decision-making in constrained circumstances. Thematic analysis of responses identified interrelated dimensions: tensions between species survival and individual wellbeing; ethical concerns with intrusive management; importance of habitat protection; emotional responses as ethical guides; and the influence of responsibility, culpability, and structural constraints on decision-making. Rather than representing inaction, participants advocated for robust conservation responses while demanding strict ethical and scientific justification for harmful interventions. Emotional responses—particularly moral distress, frustration, and hope—provided ethical clarity rather than confusion, signaling moral boundaries and guiding practitioners toward less harmful alternatives. Notably, participants demonstrated a sense of responsibility that compels engagement even when options were constrained by past failures to address root causes. This manifested as reluctant concession of some interventions as “lesser evils” while simultaneously advocating for systemic changes. The findings challenge characterizations of compassionate conservation as absolutist or impractical, instead revealing principled engagement that navigates ethical commitments with recognition of real-world constraints. The study demonstrates how compassionate conservation represents not merely an ethical overlay on traditional practice, but a fundamentally different approach integrating wellbeing considerations throughout decision-making processes.
Global biodiversity loss threatens not only species survival but also the loss of unique evolutionary history, highlighting the need to understand how macroevolutionary processes shape the evolutionary diversity and its vulnerability to extinction. Radiations of arid-adapted plants provide a useful system to examine how diversification and dispersal generate present-day biodiversity patterns.Here, we reconstruct the evolutionary history of Agave across the Americas using a time-calibrated phylogenomic framework (207 taxa, 286 nuclear genes) and integrate these results with biogeographic inference and Evolutionarily Distinct and Globally Endangered (EDGE) metrics to identify conservation priorities.We find that Agave originated in Aridoamerica during the late Miocene (∼7 Ma), followed by dispersal into Mesoamerica, Central America, and the Caribbean. This history has generated a spatially uneven distribution of evolutionary distinctiveness, with arid ecosystems harboring a disproportionate share of evolutionarily unique and threatened lineages. We identify 70 EDGE taxa, with exceptional conservation priority (the most endangered species were: Agave montium-sancticaroli, A. aurea var. promontorii, and A. albopilosa in EDGE and EDGE2). We identify key regions in Mexico, USA, and Guatemala—such as the Tehuacán–Cuicatlán Valley, Oaxaca, the Chihuahuan Desert and Arizona—that combine high phylogenetic diversity with elevated concentrations of threatened evolutionary history.These patterns reflect the interaction between macroevolutionary dynamics, and ecological specialization, and are further intensified by anthropogenic pressures, as evidenced by the high proportion of Agave species listed under the IUCN Red List. Our results demonstrate that integrating evolutionary history with extinction risk can improve spatial conservation prioritization.
Persistent taxonomic bias in conservation biology affects research funding and public perceptions of biodiversity. Public communication of conservation policy plays a key role in this context, and postage stamps represent an effective medium for official public outreach and a potential tool to help mitigate this bias. In this study, we use wildlife stamps issued in mainland China to examine whether the mainland Chinese government's public communication about wildlife conservation exhibits taxonomic bias and whether the broader conservation policies proposed in recent years have exerted any influence. We collected 87 sets of wildlife-related stamps, from which 245 wild species were identified, corresponding to 343 issuance events. Taxonomically, we found birds and mammals accounted for the majority of animal species featured, while angiosperms dominated plant species on stamps. Overrepresented taxa on stamps were also overrepresented in IUCN assessments. This bias has progressively intensified over the last decades, with most groups exhibiting consistent directional deviations. Although stamp issuances often coincided with major conservation policies or social events, these milestones did not correspond to reduced taxonomic bias. This indicates that, without explicit scientific guidance, the aesthetic and communicative functions of stamps inherently limit the inclusion of less charismatic taxa. These biases likely extend beyond stamps to other government-led public communication initiatives facing the same challenge. Bridging the gap between decision-makers and researchers, and embedding conservation priorities into strategic outreach planning, will be essential not only to raise awareness of such biases but to begin addressing them.
Protected areas face chronic financing gaps, and governments are increasingly experimenting with innovative mechanisms to mobilize additional resources for conservation. However, finance reform can also redirect management towards revenue generation, visitor growth, infrastructure delivery, or investor-facing projects unless biodiversity outcomes, legal limits, and public accountability remain decisive. This paper analyzes Indonesia's emerging initiative to establish a national task force on innovative financing and national-park management as a live policy case. We conduct a criterion-based documentary policy analysis using a purposive corpus of 29 official, legal, planning, technical, international, and peer-reviewed sources. The corpus was coded into 112 relevant text segments and 174 criterion-code applications against five literature-grounded criteria: ecological primacy, legal and zoning integrity, institutional coherence, outcome accountability, and justice and legitimacy. Landscape connectivity was treated as a cross-cutting biodiversity condition. We show that the Indonesian case is distinctive because national-park finance reform is being advanced through a high-level task-force model within an existing environmental-fund architecture, a national-park system containing a World Heritage property on the List of World Heritage in Danger, and strategic indicators that already connect conservation management with environmental-service revenues and carbon-value readiness. The analysis identifies six governance risks with plausible pathways to biodiversity harm: conservation drift, legal and zoning drift, institutional duplication, metric failure, weak social legitimacy, and protected-area insularity. We compare these risks with other conservation-finance mechanisms, including conservation trust funds, tourism user fees, biodiversity credits, wildlife conservation bonds, Indonesia's Environmental Fund Management Agency and Indonesia Biodiversity Fund, and regional public-service financing. We translate the risk analysis into an operational safeguard device linking minimum approval rules, evidence requirements, accountable institutions, and measurable indicators. The core argument is that innovative finance is defensible only when designed as a biodiversity-governance instrument rather than as a revenue strategy attached to protected areas.
Shade coffee agroforests are potential refuges for biodiversity in human-modified tropical landscapes. However, biodiversity is influenced by coffee cultivation methods and climates, both of which vary widely and are increasingly dynamic. One significant but understudied dynamic is the shift in cropping system from arabica (Coffea arabica) to robusta (C. canephora), which alongside deforestation and climate change, is reshaping global coffee landscapes. We compared land use types (secondary rainforest, arabica agroforest, robusta agroforest) and climate zones (coffee in wet vs. dry zones) for amphibian abundance, community composition, multidimensional diversity (taxonomic, functional, phylogenetic), and ecologically-sensitive and conservation-priority species in India's Western Ghats mountains. We sampled amphibians using line transects (total 12.9 km), collected primary and secondary data on 12 functional traits, geographic distributions, threat status, phylogeny, and estimated multidimensional diversity (Hill-Chao numbers q = 0–2) and species occurrence probabilities across land uses and climate zones. While overall amphibian abundance and richness (q = 0) were similar across land uses and climates, the secondary rainforest had distinct community composition, higher multidimensional diversity (q = 1, 2), and higher occurrence probabilities of stream breeding and threatened-endemic species than coffee. Amphibian multidimensional diversity (q = 1, 2) was generally higher in arabica than robusta agroforest in both climate zones, and alongside stream breeding and threatened-endemic species was lower in dry than wet zones. Our findings suggest that conversion from arabica to robusta systems and climate drying can diminish amphibian multidimensional diversity, and highlight the value of safeguarding and restoring forests and streams for ecologically-sensitive and conservation-priority species in changing coffee landscapes.
Wave disturbance from tides, waves, and storm surge is widely recognized as a threat to sea turtle reproduction, yet its frequency, spatial distribution, and biological consequences have rarely been quantified at broad scales. Using >230,000 clutch records from 2016 to 2024 for loggerhead turtles (Caretta caretta), green turtles (Chelonia mydas), and leatherback turtles (Dermochelys coriacea), we provide the first large-scale, multi-species assessment of recorded wave disturbance, defined here as wash-over, partial wash-out, and complete wash-out, across Florida, an important global nesting region and low-elevation, storm-exposed coastline. We quantified emergence success (ES) and incubation duration (ID) associated with wave disturbance, and evaluated regional, repeated-exposure, and clutch placement patterns. Recorded wave disturbance was associated with lower ES across species, with statewide estimated reductions of about 12% in loggerheads and green turtles and 4% in leatherbacks; the strongest regional effects occurred on broad, low-relief Gulf-coast beaches. Among clutches exposed only to wash-over, repeated wash-over events were associated with further ES reductions, especially in loggerheads, and wash-over exposure later in incubation was associated with lower ES. Washed-over clutches had longer IDs by approximately 0.8–1.5 days, suggesting altered incubation conditions with potential implications for primary sex ratios. Clutch placement was associated with exposure risk, yet disturbed and undisturbed clutches overlapped broadly across the beach profile, and complete wash-outs occurred even in upper-beach zones. These results illustrate the stochastic nature of wave-driven flooding and highlight the need to consider survival and potential sex-ratio effects when evaluating clutch-management strategies under sea-level rise, with implications for rookeries worldwide.
Effective conservation of at-risk, wide-ranging mammals in human-disturbed landscapes requires management decisions grounded in evidence of how animals use landscapes, particularly areas critical for persistence and recovery. An ecotype of woodland caribou, the southern mountain caribou (Rangifer tarandus caribou), is being extirpated in Canada despite federal and provincial legislative and policy protections. The ultimate cause of population declines is habitat destruction, yet protection is stymied by the absence of current habitat quality maps across southern mountain caribou herd extents. We identify drivers of habitat selection and predict key habitats for five southern mountain caribou herds in the Columbia Mountains, British Columbia, one of which became extirpated during this analysis. We modeled caribou resource selection in early winter, late winter, spring calving, and summer under a mixed-effects regression framework using accessible public data of natural features and disturbance threats. We then produced third-order spatial predictions of habitat quality. Seasonal caribou migrations show differences in elevation, canopy closure, and biogeoclimatic zones. Caribou select old growth forest (>140 years) during calving and forest stands greater than 75 years old in all other seasons. We identified a threshold forestry road density value where relative probability of use declines to half: 0.78 km/km2 (late winter), 1.05 km/km2 (calving), 1.08 km/km2 (summer). Effective conservation of southern mountain caribou requires consideration of forestry-related disturbance and seasonal variation in caribou habitat use to accurately identify and protect key habitats. Protecting mature and old-growth forests and restoring roads are urgently needed measures to maintain caribou in this landscape.
The growing uncertainty about the impacts of climate change and increasing disturbances on terrestrial ecosystems highlights the need to evaluate their resilience. Old-growth forests are particularly vulnerable to these changes. This study explores the dynamics and ecological resilience over the last 2700 years in an old-growth beech forest in the Eastern Carpathians, employing a multiproxy approach integrating pollen analysis, macro charcoal influx, fungal spores, and biomarkers. The research focuses on how vegetation reacted to disturbances (fire, fungal outbreaks, and human-related activities) and its response to these changes. The sedimentary record from Lake Morské oko reveals a stable forest community dominated by beech under low disturbance regimes, characterised by fires and fungal outbreaks. However, an increase in human impact around 1150 cal yr BP, shown by macro-charcoal influx and coprophilous fungi spores, introduced grazing and fire practices that shifted forest composition towards stress-tolerant and light-demanding species like birch and hazel. Despite the forest's resilience and partial recovery following reduced disturbances, modern pressures, such as climate change, fires, and prolonged anthropogenic impact, have further shifted the species composition of this community. The findings highlight the high ecological resilience of old-growth forests under low to moderate disturbances while also emphasising their susceptibility to intensified human activity, fires and climate change. This research highlights the importance of conservation strategies that mitigate human pressure and prioritise the regeneration of late-successional species such as beech. Findings from this long-term ecological analysis provide direction for the management and protection of these ecosystems against global environmental change.
Land and sea are ecologically connected, and effective conservation of coastal ecosystems requires integrated planning that addresses threats spanning both realms. Deforestation in tropical catchments delivers sediment, nutrients, and other pollutants to downstream coral reefs, where these can impair coral physiology and alter reef community structure and resilience. Failing to address these cross-realm threats impacts ecosystem services vital to forest and coastal communities. A growing body of work has developed ridge-to-reef modelling approaches to link land-use change to downstream marine impacts. Yet, integrating these dynamics into spatial conservation planning remains challenging, particularly at large spatial scales and while accounting for socio-economic costs. Here, we compare alternative ridge-to-reef spatial planning strategies that integrate conservation actions on land and sea, showing that explicit representations of land–sea connectivity can substantially reduce future sediment exposure on reefs. Across the hyper-diverse Wallacea bioregion in Indonesia, we estimated sediment yields from terrestrial catchments for present and future deforestation scenarios. We then quantified and mapped how runoff affected downstream turbidity on reefs. Projected future sediment runoff to reefs varied widely, with the greatest impacts concentrated near large river catchments and deltas, especially in central-southeast Sulawesi. Joint land–sea conservation scenarios reduced reef sedimentation by 25% more than strategies that treated land and sea separately, with approaches that account for where sediment reaches reefs outperforming simpler proxies based on adjacency or distance. In the face of widespread deforestation and reef decline across the tropics, applying data-driven ridge-to-reef conservation approaches could provide major benefits for people and biodiversity.
North American grassland birds have experienced steeper population declines than any other avian guild, yet conservation efforts remain largely reactive and fragmented. We used nearly four decades of North American Breeding Bird Survey data to identify biome-scale spatial patterns (clustering) of grassland bird abundance for the Great Plains. Our results reveal an ecological core in the north-central Plains where community-level abundance is either increasing by >100% or remains high and stable, providing a strategic roadmap for a “Defend the Core” conservation approach. This approach flips the script from reactive triage centered on isolated population fragments to a proactive strategy of maintaining large-scale ecosystem integrity. Conversely, we found that population losses are more spatially clustered than wins, reflecting the relentless, one-way movement of woody encroachment and agricultural conversion. This asymmetry supports prioritizing intact landscapes, as current restoration rates are often outpaced by the scale of habitat loss. Notably, we found that community-level spatial clustering is a more robust indicator of biome condition than trends of individual flagship species, suggesting that managing for ecosystem integrity provides a more effective multi-species umbrella. Given our results, there is an opportunity for operationalizing a Great Plains Conservation Design that is ecosystem-centric and rooted in the sustainability of the private-land cattle production that maintains these open spaces. By leveraging avian abundance as a biological sensor, managers and producers can deploy a shared vision that matches the spatial scale of the threats, moving from reactive triage to proactive defense of core working grasslands in North America.
Remote tracking of animals complements traditional marking and recapture methods across ecological studies, improving estimates of demographic rates, migration connectivity, habitat selection, and social interactions. Its insights into how individual behaviour scales up to population dynamics are especially critical for effective transnational conservation, particularly for threatened migratory species. Here, we synthesise over two decades of GPS tracking data from 626 Eurasian curlews (Numenius arquata) to assess how movement ecology informs conservation and policy for this species of high conservation concern across Europe. We show how tracking has refined our understanding of migration connectivity, demography, and habitat selection, and how it can be translated into actionable insights for conservation. Tracking data provided unprecedented information confirming strong site fidelity among adult curlews and evidence of local recruitment. It highlights the importance of young birds' survival until they become reproductively mature and recruit for the population dynamics, and therefore the need for appropriate conservation measures. At the same time, this review shows how movement data contributes to or can contribute to characterize most of the threats to curlews, such as human disturbance or the development of human infrastructure. These findings translate into policy recommendations, including habitat protection, network-based conservation of migratory nodes, mitigation of anthropogenic threats, and the strategic use of population reinforcement. Overall, this synthesis highlights the critical role of individual-based tracking in bridging ecological knowledge and conservation action and provides guidance for the future use of tracking data to improve the conservation status of the Eurasian Curlew along its flyway.
Human-elephant conflict (HEC) poses a significant challenge to wildlife conservation and rural community well-being in agricultural landscapes. Understanding the environmental and anthropogenic drivers of conflict and their spatiotemporal dynamics is essential for effective mitigation. Using a Bayesian spatiotemporal occupancy modeling framework, we analyzed over 20,000 verified, farmer-reported HEC incidents (2015–2023) surrounding Bandipur and Nagarahole National Parks in Karnataka, India. Situated within the Western Ghats biodiversity hotspot, these parks support some of India's largest elephant populations. Our results identified cropland area and seasonal nighttime light radiance, an indicator of human economic and agricultural activity, as the strongest predictors of HEC occurrence. Additional landscape attributes, such as proximity to refuge sites, slope, and seasonal vegetation, further explained variation in conflict occurrence. Detection probability increased with cropland extent and survey effort but declined with higher seasonal temperatures and greater distances from protected areas and refuge sites. Across the modeled landscape, mean HEC occupancy, or the probability that HEC occurs at least once per season, was estimated at 16%. Mean detection probability, that an HEC incident was reported after having occurred, was 31.7%. Occupancy was highest during the post-monsoon season, coinciding with peak harvest periods, while detection was consistently lower during the pre-monsoon dry season. Our findings highlight pronounced spatial and seasonal heterogeneity in both HEC occurrence and detection. They underscore the need for adaptive, spatiotemporally targeted mitigation and monitoring strategies to effectively reduce the impact of HEC incidents on local communities and promote coexistence in this biodiversity-rich and conflict-prone landscape.
Mapping large-scale landscape characteristics is essential for biodiversity monitoring and conservation. Remotely sensed image texture quantifies landscape heterogeneity and vegetation structure, key determinants of biodiversity patterns. However, its ecological application has been limited by the large number of metrics, difficulties in interpreting landscape features they capture, and sensitivity to methodological choices. Here, we present the first cross-biome synthesis of image texture analysis to evaluate how key parameters influence metric behavior and ecological interpretation. We focus on how methodological decisions influence texture metrics across contrasting landscapes rather than directly evaluating relationships with biodiversity observations. Our objectives were to assess the utility of image texture across multiple biomes, evaluate how methodological decisions affect texture metrics and their interpretation, and provide practical guidance for ecological and conservation applications. We analyzed twelve landscapes representing diverse global biomes and representative landscapes to demonstrate how parameter selection influences texture outputs and interpretation. Homogeneity and entropy provided complementary, low redundancy information and were robust, whereas correlation was more sensitive to moving window size, spatial resolution, and image acquisition timing. Texture metrics were robust to image quantization, but spatial resolution and acquisition timing strongly influenced texture outputs. We recommend using vegetation indices rather than individual spectral bands, selecting one to three uncorrelated texture metrics, keeping image acquisition and processing parameters consistent, and matching moving window size and spatial resolution to the ecological question. Image acquisition timing should also reflect the study objective. These recommendations provide a practical workflow for designing robust, consistent, and ecologically meaningful image texture analyses.
Populations of amphibians and reptiles have undergone substantial declines in recent decades worldwide. The Iberian Peninsula, a European diversity hotspot for this group, is suffering severe droughts, rising temperatures, and abandonment of traditional agricultural practices. Despite the importance and sensitivity of Iberian herpetofauna, a data-driven, national, long-term, and multi-species analysis of the distribution and demographic situation of amphibians and reptiles has been lacking. We used dynamic occupancy models to study how the distributions of 25 amphibian and 37 reptile species have changed over the last three decades in Spain. We analysed a detection dataset compiled from multiple sources including citizen science, monitoring programmes and opportunistic observations. Our results showed widespread declines in amphibian and reptile occupancy, which could also suggest generalised population declines. Moreover, we observed a mismatch between the IUCN Red List categories and the estimated occupancy trend, with some species appearing stable at the IUCN global level but presenting substantial range contractions at the regional level. The causes for these declines are unclear. According to our models, amphibian persistence was negatively correlated with rainfall and colonization negatively correlated with altitude. Reptile persistence was negatively correlated with temperature. Our analysis highlights the value of national-scale assessments to identify occupancy trends at fine resolution and investigate regional threats driving these changes. We encourage further research and early conservation action before regional trends and threats become global issues. We further highlight the need for long-term, standardized monitoring programmes to contribute high-quality data to regional and global assessments.
