Abstract Environmental heterogeneity generates taxonomic and functional beta diversity, but a key question regards whether heterogeneity can maintain functional diversity when global change causes range shifts, potentially causing biotic homogenization by favouring species sharing certain traits over others. We tested the capacity of environmental filtering to conserve taxonomic and functional beta diversity through time, using recent changes to butterfly communities across elevation gradients as a focal system. We sampled species in historic (1985–2005) and recent periods (2017–2022) in 163 discrete sites in four mountain ranges of the central Iberian Peninsula, a reservoir of genetic and taxonomic diversity during glacial periods. We used 35 binary traits to calculate functional diversity and partitioned pairwise Sørensen dissimilarity into turnover and nestedness for taxonomic beta diversity and functional replacement. In historical and recent periods, pairwise taxonomic and functional beta diversity were related to geographic distance and elevation. Distance had a stronger effect on taxonomic beta diversity than on functional replacement, suggesting different species played similar functional roles at equivalent elevations in different mountain ranges. However, the effects of distance on taxonomic and functional beta diversity decreased over time, whereas strong effects of elevation persisted in the recent period. Turnover was the main component of taxonomic and functional beta diversity, but the relative contribution of nestedness increased in the recent period, implying non‐random losses of rare species or gains of widespread species. Effects of mountain range and microclimate on taxonomic and functional differences at equivalent elevations also decreased over time. Variation over the elevation gradient in species richness per functional group generally persisted over time, although at some higher elevations, average richness per functional group increased. Despite some evidence of biotic and functional homogenization across four mountain ranges, effects of elevation on taxonomic and functional beta diversity persisted over time. These results emphasize the importance of elevational filtering to maintain taxonomic and functional differences related to ecosystem resilience, and support the role of environmental heterogeneity to sustain functional diversity and provide refugia from global change. Read the free Plain Language Summary for this article on the Journal blog.

Researching the properties of mutualistic networks over environmental gradients is a promising but underexplored means to test how global change can affect ecosystem assembly and functioning. We examined how elevation and microclimate influenced butterfly-flower interaction networks at the hottest time of the year in a Mediterranean mountain range. Throughout July 2023, we recorded weekly butterfly-flower interaction networks from 36 transects in nine sites, across an 800-m elevation gradient in the Sierra de Guadarrama (Central Spain). We quantified metrics relating to network topology (modularity, nestedness), structural complexity (connectance, linkage density) and resilience to species loss (robustness, generality) and related these descriptors using generalised additive mixed models to elevation, microclimate temperature (modelled using Microclima) and time of day. The networks were dominated at all sites by one or two abundant butterfly and flower species, but these species varied with elevation. Butterfly networks in July were more robust at higher elevations to plant species loss. However, there were no clear effects of summer microclimatic temperatures on butterfly-plant networks, for which nestedness and modularity were low. Network properties also varied through the day, with connectance decreasing from morning to afternoon. In the Mediterranean mountains studied, summer butterfly-flower interaction networks appeared to show greater resilience to disturbance at high elevations. Nectar availability could become an important limiting factor for insects in a warming climate, and understanding the mechanisms influencing the properties of flower visitor networks is therefore likely to become increasingly important for adapting the conservation of insects to climate change.

Anthropogenic climate change is projected to become a major driver of biodiversity loss, destabilizing the ecosystems on which human society depends. As the planet rapidly warms, the disruption of ecological interactions among populations, species and their environment, will likely drive positive feedback loops, accelerating the pace and magnitude of biodiversity losses. We propose that, even without invoking such amplifying feedback, biodiversity loss should increase nonlinearly with warming because of the non-uniform distribution of biodiversity. Whether these non-uniformities are the uneven distribution of populations across a species’ thermal niche, or the uneven distribution of thermal niche limits among species within an ecological community, we show that in both cases, the resulting clustering in population warming tolerances drives nonlinear increases in the risk to biodiversity. We discuss how fundamental constraints on species’ physiologies and geographical distributions give rise to clustered warming tolerances, and how population responses to changing climates could variously temper, delay or intensify nonlinear dynamics. We argue that nonlinear increases in risks to biodiversity should be the null expectation under warming, and highlight the empirical research needed to understand the causes, commonness and consequences of clustered warming tolerances to better predict where, when and why nonlinear biodiversity losses will occur. This article is part of the discussion meeting issue ‘Bending the curve towards nature recovery: building on Georgina Mace’s legacy for a biodiverse future’.

ABSTRACTAimThe capacity of cool refugia to protect cold‐adapted species against climate change may depend on both their initial climatic conditions and how quickly these change. We test how local climatic conditions influence mountain butterfly communities via their effects on colonisation and local extinction.LocationFour mountain ranges in Central Spain.MethodsWe used community temperature index (CTI), based on the climatic niches of constituent species (species temperature index, STI), to estimate thermal affinities for butterfly communities sampled in 1984–2005 to 2017–2022. We related CTI to local temperature, estimated using the model Microclima, and tested for changes to local temperature and CTI over time. We used standard deviation in CTI (CTISD) and species richness to detect effects of colonisation and local extinction on community change. Finally, we tested for differences in thermal affinity and thermal niche breadth (STISD) between species undergoing local extinction or colonisation at each site.ResultsCTI was positively related to local temperature in both periods. However, there were regional differences in rates of change in CTI and local temperature. CTI increased overall, even though temperatures decreased at many sites; and CTI increases were greatest in historically cool sites. Neither CTISD nor species richness changed overall, suggesting that communities experienced equivalent numbers of colonisations and extinctions. Colonising species had warmer thermal affinities than those undergoing local extinction, and species with broader thermal niches increased their occupancy most over time.Main ConclusionsLocal climatic conditions influenced changes to community composition based on species thermal tolerances, resulting in the loss of communities where cool‐affinity species predominated, and a narrower range of community thermal affinities overall. Our results suggest that a regional perspective to identifying climate change refugia is needed to provide a wide range of local climate conditions and rates of change to help adapt conservation to climate change.

Climate change poses extreme risks to biodiversity, threatening the ecosystems upon which humanity depends. Understanding the traits that mediate how organisms respond to climatic gradients in space and time will enable better predictions of the taxa and ecological communities most at risk from warming. In insects, colouration influences thermoregulation, but warming responses must be traded off against other selective pressures. We hypothesise that butterfly wing reflectance in visible and near-infrared (NIR) spectra respond differently to temperature and ultraviolet radiation over space and time. We combined existing reflectance data of 97 butterfly species in both NIR and visible wavelengths, with long-term abundance monitoring data of 120 butterfly communities sampled across 1650 m altitude gradients from May to August in two time periods (2004/5 and 2017). The visible and NIR reflectances of butterfly communities showed a non-linear relationship with altitude, with the lowest reflectance (darkest butterflies) at the coolest, highest sites. In contrast, community visible reflectance decreased through the year as temperatures warmed over spring-summer, whereas community NIR reflectance remained constant, revealing divergent responses of reflectance types to seasonal changes. Temperature had opposing effects on visible and NIR reflectance of butterfly communities, where increasing temperature reduced community visible reflectance strongly while increasing community NIR reflectance slightly. Considering the effects on reflectance of shared evolutionary history in a Bayesian hierarchical model for individual species, lighter-coloured (more reflective) species were associated with warmer temperatures - flying later in spring-summer or at lower altitudes. However, instead of decreasing in reflectance through the year and across temperature gradients, species instead became lighter, we expect as a result of a Simpson’s paradox. These results emphasise how visible and NIR reflectance wavelength bands mediate butterflies’ responses to environmental gradients in distinct ways, despite being highly correlated across species. We also show that incorporating phylogeny into trait-environment models is essential; relying on traits alone would lead to incorrect inferences and predictions of taxa most at risk from warming climates.

Gradients in community diversity and composition rarely track rates of warming, prompting efforts to understand the factors causing non-equilibrium responses to climatic change and their implications for conservation. Here, we test the roles of fine-resolution habitat heterogeneity and microclimate in delaying butterfly community responses to warming over a mountain elevation gradient. We assess species diversity and Community Temperature Index (CTI) in 2004–2005 and 2017 across 120 transect sites in the Sierra de Guadarrama (Spain), modelling temperatures based on topography, and vegetation structure based on LiDAR. A humped elevation gradient in species diversity was maintained over time. However, diversity in the later period was more positively related to vegetation heterogeneity, and sites with reduced rates of warming and greater forest cover maintained or increased their diversity. Site CTI declines with increasing elevation, showing little evidence of systematic change over the gradient between periods, although CTI increased most in locations with the greatest rates of spring microclimatic warming. Our results show that delays in community tracking of climatic conditions could result partly from positive effects of habitat and topographic heterogeneity providing microclimatic buffering against warming to existing communities; although barriers to colonization could also delay diversity increases and community thermophilization at high elevations. We suggest that protecting and managing complex habitats with high community diversity, and identifying localized microclimates that buffer populations against negative effects of warming, are more immediate conservation priorities over elevation gradients than efforts to ensure that communities track prevailing rates of warming.

Phenological responses to climate change vary across trophic levels. However, how trophic phenological synchrony determines species’ distributions through its effects on population dynamics has rarely been addressed. Here, we show that phenological variation underlies population and geographical range dynamics in a range-shifting herbivore, and demonstrate its interplay with changing trophic interactions. Using a novel modelling approach, we identify drivers of variation in phenology and population growth (productivity) for populations of the brown argus butterfly ( Aricia agestis ) feeding on ancestral and novel host plants in the UK. We demonstrate host plant-specific links between phenology and productivity, highlighting their role in the consumer’s range expansion. Critically, later butterfly phenology is associated with higher productivity in the annual second brood, especially on novel annual hosts where later activity improves synchrony with germinating plants. In turn, later phenology and higher second brood productivity are associated with more rapid range expansion, particularly in regions where only the novel hosts occur. Therefore, phenological asynchrony imposes limits on local population growth, influencing consumer resource selection, evolutionary responses and emergent range dynamics. How existing and future trophic phenological synchrony determine population dynamics will be critical for the ecological and evolutionary outcomes of climate change.

Abstract Climate change refugia are landscapes, topographic features or ecosystems that buffer organisms against exposure to climate change by providing conditions that are stable or spatially heterogeneous (macrorefugia) or distinct from their surroundings (microrefugia). Refugia allow taxa to persist through unfavorable climatic conditions and act as foci for subsequent recovery or range expansion, and could therefore underpin measures for adapting conservation to climate change. This chapter outlines physical and microclimatic features of refugia, and the evidence for their influence on insect responses to historical and recent climate change. It considers how vulnerability to climate change depends on how climate varies over space and time in refugia, but also on biotic traits (endemism, narrow climatic tolerance or habitat specialism) that may increase climate change sensitivity for species occupying these locations. The chapter concludes by reviewing proposals and remaining challenges for the practical application of climate change refugia to insect conservation.

Shifts over time (phenology) and space (latitude and elevation range) represent common ecological responses to climate change. However, the factors determining how changes in phenology and distribution interact, and the consequences for conservation, remain uncertain. Here, we assess how phenology responded to temperature over four decades of warming across the elevation ranges of 18 univoltine butterfly species in four mountain regions of Spain. Using count data from intermittent surveys in 166 sites between 1985 and 2022, we tested for (1) effects of monthly temperature and elevation on mean annual flight date; (2) changes to flight dates between 1985–2005 and 2017–2022; and (3) whether shifts in flight date were related to shifts in the average elevation occupied. Mean flight dates were later in years with cooler springs, and at higher sites, with a mean delay of nearly twenty days per km elevation increase. As conditions warmed over time, average flight date advanced for two thirds of species, especially those whose average elevation was stable over time. Species with stable flight dates showed greater indication of upward range shifts, although only one species showed a significant shift in average elevation. Implications for insect conservation: We show that spring temperatures influence mountain butterfly phenology, and that shifts in phenology and elevation range could compensate for each other in determining population exposure and responses to climate change. Monitoring these changes over time, including by employing evidence from historical surveys and scientific collections, can help to understand constraints on species adaptive capacity to climate change.

Landscape‐scale approaches are increasingly advocated for species conservation but ensuring landscape level persistence by enlarging the size of patches or increasing their physical connectivity is often impractical. Here, we test how such barriers can be overcome by management of habitat at the local (site‐based) level, using a rare butterfly as an exemplar. We used four surveys of the entire UK distribution of the Lulworth skipper Thymelicus acteon over 40 years to test how local habitat influences population density and colonization/extinction dynamics, and parameterized, validated and applied a metapopulation model to simulate effects of varying local habitat quality on regional persistence. We found the total number of populations in four distribution snapshots between 1978 and 2017 varied between 59 and 84, and from 1997 to 2017 34% of local populations showed turnover (colonization or extinction). Population density was closely linked to vegetation characteristics indicative of management, namely height and food plant frequency, both of which changed through time. Simulating effects of habitat quality on metapopulation dynamics 40 years into the future suggests coordinated changes to two key components of quality (vegetation height and food plant frequency) would increase patch occupancy above the range observed in the past 40 years (50–80%). In contrast, deterioration of either component below threshold levels leads to metapopulation retraction to core sub‐networks of patches, or eventual extirpation. Our results indicate that changes to habitat quality can overcome constraints imposed by habitat patch area and spatial location on relative rates of colonization and local extinction, demonstrating the sensitivity of regional dynamics to targeted in situ management. Local habitat management therefore plays a key role in landscape‐scale conservation. Monitoring of population density, and the monitoring and management of local (site‐level) habitat quality, therefore represent effective and important components of conservation strategies in fragmented landscapes.

Taxa restricted to mountains may be vulnerable to global warming, unless local-scale topographic variation and conservation actions can protect them against expected changes to the climate.We tested how climate change will affect the 19 mountain-restricted Erebia species of the Iberian Peninsula, of which 7 are endemic.To examine the scope for local topographic variation to protect against warming, we applied species distribution models (HadGEM2 and MPI) at two spatial scales (10 x 10 and 1 x 1 km) for two representative concentration pathways (RCP4.5 and RCP8.5) in 2050 and 2070. We also superimposed current and future ranges on the protected area (PA) network to identify priority areas for adapting Erebia conservation to climate change.In 10 x 10 km HadGEM2 models, climatically suitable areas for all species decreased in 2050 and 2070 (average -95.7%). Modelled decreases at 1 x 1 km were marginally less drastic (-95.3%), and 14 out of 19 species were still expected to lose their entire climatically favourable range by 2070.The PA network is well located to conserve the species that are expected to retain some climatically suitable areas in 2070. However, we identify 25 separate 10 x 10 km squares where new PAs would help to adapt the network to expected range shifts or contractions by Erebia.Based on our results, adapting the conservation of range-restricted mountain taxa to projected climate change will require the implementation of complementary in situ and ex situ measures alongside urgent climate change mitigation.

Context Efforts to adapt conservation to climate change often focus on facilitating range shifts to higher latitudes, by enhancing landscape capacity for poleward expansion. The need to protect populations at trailing edges of species distributions, and how and where to do so, has received less attention. Objectives We assess how population declines caused by variation over space and time in exposure to climate change can necessitate conservation adaptation to climate change throughout species’ geographic ranges. We propose approaches for conservation in landscapes where species are vulnerable. Methods We synthesize primary literature relating to recent landscape-scale changes to species distributions to identify evidence for patchy patterns of climate-driven decline. We use this evidence to propose a framework to adapt terrestrial species conservation. Results Patchy retractions occur throughout species ranges as environmental heterogeneity results in spatial variation in climate and rates of climate change, whereas equatorward range margins are often not the first place to exceed climatic limits. Furthermore, climate effects on fitness, survival and reproduction interact with habitat quality, creating both localized extinction hotspots and climatically resilient microrefugial landscapes across species ranges. Conservation can benefit from the identification of vulnerable versus microrefugial landscapes, and implementation of targeted interventions. Conclusions A focus on expansions and retractions at broad latitudinal range margins risks overlooking declines throughout species’ distributions. Understanding fine-resolution ecological responses to the climate can help to identify resilient microrefugial landscapes, and targeted management to promote cooler or more stable conditions can complement facilitation of broader-scale range shifts.

The hazel dormouse is predominantly an arboreal species that moves down to the ground to hibernate in the autumn in temperate parts of its distributional ranges at locations not yet well understood. The main objective of this study is to test whether environmental characteristics surrounding hazel dormouse hibernacula can be identified using high-resolution remote sensing and data collected in situ. To achieve this, remotely sensed variables, including canopy height and cover, topographic slope, sky view, solar radiation and cold air drainage, were modelled around 83 dormouse hibernacula in England ( n = 62) and the Netherlands ( n = 21), and environmental characteristics that may be favoured by pre-hibernating dormice were identified. Data on leaf litter depth, temperature, canopy cover and distance to the nearest tree were collected in situ and analysed at hibernaculum locations in England. The findings indicated that remotely sensed data were effective in identifying attributes surrounding the locations of dormouse hibernacula and when compared to in situ information, provided more conclusive results. This study suggests that remotely sensed topographic slope, canopy height and sky view have an influence on hazel dormice choosing suitable locations to hibernate; whilst in situ data suggested that average daily mean temperature at the hibernaculum may also have an effect. Remote sensing proved capable of identifying localised environmental characteristics in the wider landscape that may be important for hibernating dormice. This study proposes that this method can provide a novel progression from habitat modelling to conservation management for the hazel dormouse, as well as other species using habitats where topography and vegetation structure influence fine-resolution favourability.

Climate-driven geographic range shifts have been associated with transitions between dietary specialism and generalism at range margins. The mechanisms underpinning these often transient niche breadth modifications are poorly known, but utilization of novel resources likely depends on phenological synchrony between the consumer and resource. We use a climate-driven range and host shift by the butterfly Aricia agestis to test how climate-driven changes in host phenology and condition affect phenological synchrony, and consider implications for host use. Our data suggest that the perennial plant that was the primary host before range expansion is a more reliable resource than the annual Geraniaceae upon which the butterfly has become specialized in newly colonized parts of its range. In particular, climate-driven phenological variation in the novel host Geranium dissectum generates a narrow and variable 'window of opportunity' for larval productivity in summer. Therefore, although climatic change may allow species to shift hosts and colonise novel environments, specialization on phenologically limited hosts may not persist at ecological margins as climate change continues. We highlight the potential role for phenological (a)synchrony in determining lability of consumer-resource associations at range margins and the importance of considering causes of synchrony in biotic interactions when predicting range shifts. This article is part of the theme issue 'Species' ranges in the face of changing environments (Part II)'.

Trade-offs between specialist use of particular resources and opportunistic use of widespread materials may present different strategies for survival. Hazel dormice Muscardinus avellanarius are arboreal mammals that in Great Britain hibernate from late autumn to mid spring in nests that are specially built at ground level. Hibernation nests are rarely encountered, and little is known about the types of construction or materials used. Specifically, it is not known whether nest materials are selected opportunistically, based on their availability, or are specialised to suit local environmental conditions. We therefore conducted a study to characterise the main materials used to construct these nests, explore the distances travelled to collect materials and investigate whether regional climate and/or local microclimate have an impact on the types of nests built. Thirty-three hibernation nests were located through radio-tracking, systematic searches and incidental finds. Structurally, hibernation nests were built similarly to summer nests and were most commonly constructed with an outer layer of leaves and distinct core section made of woven material. We found no correlation between nest type and the temperature recorded at nest sites. Nests were built using a mean of two materials per nest, which were both in every case available within 3m of the nest. The most frequently used materials were bracken, hazel and beech leaves, and grasses. Dormice were flexible in their use of nest materials, using various materials harvested very locally. However, dormice travelled further to collect grasses, ferns, bracken and honeysuckle, and these materials made up most of the nests in which they were found. There were also positive correlations between material abundance and usage, and suitable materials for hibernation nest construction were therefore readily available within their home range.

How the species that form ecological communities respond to climate change will affect the future resilience of ecosystems, and their capacity to support humankind. The responses of animals and plants to four decades of warming demonstrate the sensitivity of high-latitude ecosystems to increasing temperatures.

The Hazel Dormouse is predominantly an arboreal species that moves down to the ground to hibernate in the autumn in temperate parts of its distributional ranges at locations not yet well understood. In this study, we tested whether environmental characteristics surrounding Hazel Dormouse hibernacula can be identified using high-resolution remote sensing and data collected in situ. We modelled remotely sensed variables, including canopy height and cover, topographic slope, sky view, solar radiation and cold air drainage around 83 dormouse hibernacula in England (n=62) and the Netherlands (n=21), and identified environmental characteristics that may be favoured by pre-hibernating dormice. We also collected and analysed data on leaf litter depth, temperature, canopy cover and distance to the nearest tree collected in situ at hibernaculum locations in England. We found that remotely sensed data were effective in identifying attributes surrounding the locations of dormouse hibernacula and, when compared to in situ information, provided more conclusive results. Our study suggests that remotely sensed topographic slope, canopy height and sky view have an influence on animals choosing suitable locations to hibernate; whilst in situ data suggested that average daily mean temperature at the hibernaculum may also have an effect. Remote sensing proved capable of identifying localised environmental characteristics in the wider landscape that may be important for hibernating dormice. We also propose that this method can provide a novel progression from habitat modelling to conservation management for the Hazel Dormouse, as well as other species using habitats where topography and vegetation structure influence fine-resolution favourability.