Estuarine benthic sediments are important habitats that support ecosystem health and productivity, but the drivers of large‐scale patterns in benthic diversity are not well documented. Using 18S amplicon sequencing of 425 sediment DNA samples from estuaries spanning 20° of latitude along Australia's east coast, we assessed tropical–temperate patterns in benthic eukaryotic richness, community composition, and environmental structuring. Contrary to classical expectations, ASV richness declined towards the tropics, with higher richness in temperate estuaries. This inverse gradient was evident for total eukaryotes, microeukaryotes, macrofaunal metazoans, and multicellular algae, while microbial metazoans and unicellular algae showed weaker or no latitudinal trends. Temperature was one of the strongest predictors of richness and community composition, although its high correlation with latitude means that thermal effects cannot be separated from other latitudinally structured processes. Sediment properties were also important, particularly sediment heterogeneity, which was consistently positively associated with richness. Nutrients and water‐column physicochemical variables, including salinity, dissolved oxygen and pH, had more group‐specific effects on richness but contributed more clearly to shifts in community composition with latitude and taxon‐level abundances. Over the latitudinal gradient, differences in habitat conditions may have contributed to these observed patterns. Compared with higher‐latitude temperate estuaries, lower‐latitude tropical estuaries tended to have more heterogeneous sediments but lower nutrient levels, lower dissolved oxygen and lower pH, suggesting that warmer environments were associated with less favourable benthic habitat. These findings show that estuarine benthic eukaryotic diversity does not necessarily follow classical tropical‐richness expectations, but instead reflects interactions between large‐scale gradients, local sediment habitat conditions, and taxon‐specific environmental responses. More broadly, our results suggest that warming and changes in sediment dynamics could reduce benthic biodiversity and compromise the habitat quality and functioning of estuarine ecosystems.
The local effects of atmospheric nitrogen deposition on plants are well established, yet a global empirical analysis of nitrogen deposition effects on vegetation dynamics is missing. Here, we assess how nitrogen deposition, alongside other global change drivers such as fire, and changes in temperature, precipitation and human land use, is related to changes in vegetation greenness (enhanced vegetation index; EVI) from 2001 to 2024, and whether protected areas (PAs) mitigate these impacts. We found that increases in greenness were strongly and positively linked to nitrogen deposition and this pattern held across major ecosystem types. Nitrogen deposition also had a larger coefficient estimate and explained more variation in the spatial pattern of vegetation greening than the other drivers, suggesting the association of greening with nitrogen deposition may be even stronger than with climate and land‐use change. Although EVI increased slightly slower inside PAs, protection status did not alter the relationship between nitrogen deposition and greening. Our findings suggest atmospheric nitrogen deposition as a critical but often underestimated potential driver of global vegetation dynamics. Reducing nitrogen deposition, alongside greenhouse gas emissions and land‐use pressures, is therefore essential for sustaining ecosystem functioning and biodiversity.
Theory predicts that the strength and direction of species interactions can shift from being competitive in benign environments toward being facilitative in stressful environments. However, the environmental context dependency of species interactions has rarely been tested in animal communities. We capitalized on a 15‐year, landscape‐scale dataset, collected annually in a relatively stable old‐growth forest environment, to test the long‐held hypothesis that the strength and direction of species interactions might be mediated by climatic conditions. It is generally accepted that competitive and facilitative interactions drive the distributions of many species, but are rarely included in climate envelope models, and may partly explain poor predictive performance of species distribution models in the face of climate change. Using multi‐species dynamic occupancy models applied to long‐term data, we tested whether annual settlement by bird species could affect either the persistence or settlement by other phylogenetically related species, and whether these interactions are mediated by microclimate. We found that species interactions were influenced by microclimate for a high proportion of species pairs (80%). Related species pairs more often showed settlement dynamics that were indicative of attraction rather than repulsion (55 versus 30%, respectively). In some cases, competitive interactions in colder microclimates flipped to become facilitative in warmer ones (15%) and vice versa (5%). Furthermore, species pairs that were closely related were more likely to exhibit competitive relationships along at least part of the microclimatic gradient. Our results highlight the importance of using long‐term data to incorporate competitive and facilitative interactions into species distribution models, and support the notion that the strength and direction of species interactions can be dependent on microclimatic conditions.
Animal coloration exhibits substantial biogeographic variation, reflecting complex adaptations to both abiotic and biotic environments. Although the evolutionary mechanisms underlying avian plumage coloration have been widely studied, geographic variation in plumage patterning has received far less attention. To extend our understanding of avian plumage evolution, we quantified global plumage coloration and patterning across 10 290 bird species using an extensive red‐green‐blue (RGB) dataset derived from illustrations in the Handbook of the Birds of the World. To test ecogeographical predictions, we modeled variation in plumage lightness, blue–yellow chromaticity, and pattern granularity using phylogenetic mixed models and path analyses. Our analyses revealed pronounced geographic heterogeneity in both plumage coloration and patterning. Variation in plumage lightness followed broad climatic gradients, and darker coloration was associated with humid and colder environments, consistent with the combined predictions of Gloger's rule (darker in warm, humid areas) and Bogert's rule (darker in cold areas). These rules are not contradictory but coexist; their relative influence is determined by whether humidity or temperature is the dominant selective force in a given region. While resident species in cold environments largely supported Bogert's rule for thermoregulation, migratory species and proficient fliers (e.g. gulls) deviated from this pattern by maintaining lighter plumage to avoid overheating during flight. Furthermore, we identified a strong latitudinal gradient along the blue–yellow axis: tropical humidity promotes carotenoid‐rich yellow plumage through dietary shifts, whereas higher latitudes favor structural blue hues. Patterning also showed large‐scale structure, with larger pattern elements linked to humid environments and larger body sizes rather than to specific foraging habitats. By integrating classical ecological rules with the thermoregulation hypothesis, our study provides a unified framework explaining how climatic constraints and ecological processes interact to shape the global distribution of avian coloration patterns.
Ecosystem dynamics are shaped by interacting top‐down and bottom‐up processes, yet their relative importance may vary through time following disturbances such as disease outbreaks and changes in apex predator communities. Using a 52‐year time series in south‐central Sweden, we examined how disease, apex predators, and prey availability jointly impact a mesopredator, the red fox Vulpes vulpes , and how these dynamics propagate to secondary prey. Potential drivers of fox dynamics (number of litters) were vole Clethrionomys glareolus and Microtus agrestis abundance, sarcoptic mange outbreaks, and the re‐establishment of Eurasian lynx Lynx lynx and wolves Canis lupus , and we assessed potential subsequent effects on hare Lepus europaeus and L. timidus and woodland grouse Lyrurus tetrix and Tetrao urogallus densities. Fox litter numbers were positively associated with vole abundance throughout the study period, indicating persistent bottom‐up regulation. In contrast, top‐down effects were evident through reductions in the number of fox litters during sarcoptic mange outbreaks and following the simultaneous presence of lynx and wolves. Support for the mesopredator release hypothesis was context‐dependent: densities of secondary prey increased during the mange outbreak but declined after apex predator re‐establishment, despite reduced number of fox litters. Moreover, the characteristic synchrony between voles and secondary prey weakened following the return of apex predators, suggesting direct and indirect effects beyond mesopredator suppression. Our results demonstrate that fox reproduction is regulated by concurrent top‐down and bottom‐up processes and that apex predator recovery can alter trophic relationships and not only reverse mesopredator release.
The Pan‐Himalayan region harbors exceptional biodiversity, yet the origins and underlying mechanisms driving this remarkable diversity remain poorly understood, especially among species‐rich invertebrates. Pimoa spiders exhibit an intercontinental disjunct distribution across three major mountain regions: the Rockies, the Alps, and the Pan‐Himalaya. Notably, in the Pan‐Himalaya, Pimoa occurs at significantly higher elevations and exhibits far greater species diversity than in the other two regions, making it an ideal model for studying invertebrate diversification in this area. We integrated extensive genetic data (59 new transcriptomes and over 22 700 newly generated DNA sequences from 393 samples), along with distribution and climatic data, to explore the origins and drivers of the rich diversity of Asian Pimoa in the Pan‐Himalaya. Our findings indicate that Pan‐Himalayan Pimoa spiders originated from widely distributed ancestors in North America and Asia. During the Miocene, they dispersed southward from northeast Asia to south China and the Pan‐Himalaya, experiencing rapid diversification. Climatic niche modeling and ancestral reconstruction analyses revealed niche divergence between Pan‐Himalayan Pimoa and their ancestors, particularly in elevation and temperature. Gene selection analyses showed that high‐elevation adaptations primarily involve enhanced energy metabolism, hypoxia resistance, and DNA repair mechanisms. These results suggest that ecological niche differentiation, high‐elevation adaptation, and rapid diversification during the Miocene are crucial factors shaping the rich diversity of Asian Pimoa in the Pan‐Himalaya. Our study, which integrates genetic, distributional, and climatic data, provides a novel framework for understanding invertebrate diversification in the Pan‐Himalaya.
Land-use intensification threatens biodiversity, and restoring degraded ecosystems remains challenging due to the difficulty of identifying the rules governing community assembly and dynamics. Investigating the temporal dynamics of trait-abundance distributions (TADs) along long-term time series offers a promising approach to disentangle the influence of stochastic (e.g. climatic variability, ecological drift) and deterministic processes (e.g. biotic interactions, abiotic filtering) in shaping ecological communities and guiding biodiversity restoration. We used a long-term restoration experiment conducted in a species-poor, historically intensively managed grassland (Massif-central, France) to evaluate the relationships between land-use intensity, TADs dynamics, and the recovery of grassland plant species richness. TADs were quantified annually from 2004 to 2021 based on four traits related to plant architecture and leaf carbon economy. We analysed the temporal variability of TADs using the skewness-kurtosis relationship (SKR), a framework that explicitly accounts for stochasticity while revealing the imprint of deterministic processes on community assembly. TADs dynamics were not random, but were constrained according to trait- and land-use-specific SKR patterns. The cessation of fertilisation shifted TADs for all traits, from peaked distributions under fertilised management to more uniform distributions under unfertilised conditions. In the unfertilised treatment, TADs were also more stable than expected by chance and converged toward maximal trait evenness. Maximisation in trait evenness occurred during the early years of the experiment and preceded the long-term recovery of plant species richness. By explicitly accounting for stochastic community dynamics, our study reveals how plant communities reassemble during ecological restoration of grasslands historically degraded by intensive management and identifies the maximisation of trait evenness as a predictive benchmark linking early community reorganisation to long-term biodiversity recovery.
Mammals play important roles in redistributing elements across ecosystems, concentrating biogeochemical inputs across both space and time. However, research on zoogeochemical inputs is often constrained by logistical considerations, potentially limiting our knowledge of mammals' impacts on biogeochemical patterns and processes. Here, we present a bibliometric analysis that synthesizes both the spatiotemporal scope of research and range of methodological approaches used to study zoogeochemical inputs from mammals. Our assessment focuses on the major material pathways – fecal matter, urine, carcasses, and other body wastes – that are directly deposited by mammals. Our goal was to identify the ecological variables, ecosystem processes, and the spatial and temporal scales investigated by these studies, characterize geographic and taxonomic biases, and draw attention to opportunities for improved conceptual continuity. We found that while many studies effectively characterized the biogeochemical composition of mammalian inputs themselves, there is little methodological standardization across measurements that characterize the fates and functional impacts of these inputs within ecosystems. The diversity of approaches reflects the wide range of research questions in the field; however, paired with a lack of standardized measurement protocols and limited data sharing, this diversity prevents cross‐study empirical and conceptual synthesis. Notably, almost all studies were limited in duration (< 3 years) and did not follow ecosystem processes long enough to detect when (or if) the input's effects tapered off – highlighting a key opportunity for future research. Geographically, North American and European sites were relatively well represented, while deserts, boreal and tropical forests, and tropical systems were under‐represented relative to their global area. Addressing geographic biases, standardizing measurement protocols, and extending the duration of field studies to capture the full impacts of zoogeochemical inputs will enhance the ability to reconcile empirical and theoretical approaches and develop a more robust understanding of the spatiotemporal scale of mammalian control over ecosystem processes.
Over the past decades, Arctic and alpine biomes have warmed faster than the global average. Substantial vegetation changes have been reported as a response to this warming, including increased productivity and the spread of warm‐adapted (thermophilic) species that are less resistant to grazing pressure into higher latitudes and altitudes. At a local and regional level, however, reported vegetation changes are much more variable. Here, we investigate community changes of vascular plant species in the mountain tundra in Sweden during the period 2003 to 2020 by combining remote sensing and vegetation monitoring at 419 permanent plots. We first investigated changes in vegetation productivity by analysing the trend in the normalized difference vegetation index (NDVI). We then used plot averages of species‐specific environmental indicators weighted by species occurrences to test whether the vegetation changed towards a more thermophilic state with a decreased resistance to grazing by mammalian herbivores during that time. We found an increase in vegetation productivity in ca 23% of the sites, and no trend in ca 55% of the sites. There was no general trend towards an increased thermophilization of vascular plant communities. Furthermore, at the plot scale, the change in NDVI during the study period was not explained by thermophilization of the plant community. However, warm‐adapted forbs increased more than cold‐adapted species. We conclude that, although part of the vascular vegetation is getting more thermophilic, the observed increase in vegetation productivity during the last two decades is not driven by an increase in warm‐adapted species. We also suggest that reindeer grazing, which occurs in almost the entire study region, contributes to the preservation of cold‐adapted Arctic–alpine biodiversity by selectively feeding on warm‐adapted vascular species.
Abstract Ecological niche models (ENMs) are commonly used to predict species' potential distributions across space and time, allowing researchers and practitioners to assess invasion risk, estimate the impacts of future climate change, and reconstruct past distributions. These predictions, however, often require extrapolation into environmental conditions that fall outside the range of values used for model training (i.e. ‘non‐analog environments'). Two primary strategies exist for extrapolating into new conditions for some modeling algorithms: unconstrained extrapolation (allowing the modeled response curve to maintain its trajectory) and clamping (holding the response constant at the values observed at the limits of the training data). As these can lead to considerable differences in model predictions, it is important to inspect the behavior of modeled responses under different extrapolation strategies, which is rarely done in ENM studies. Such examination can help promote biologically realistic model transfers, for example by avoiding unconstrained extrapolation when the relevant tail of the modeled response has a sharply increasing trend. Here, we present a proposed workflow and new visualization tools to guide researchers in choosing appropriate extrapolation strategies, including plots for marginal response curves and non‐analog pixel densities. As an illustration, we used this workflow and tools to compare extrapolation strategies for two hindcast scenarios at the Last Glacial Maximum (~ 21 kya) for the black‐eared mouse Peromyscus melanotis in Mexico, which currently occurs in montane forests. Making extrapolation choices individually for each tail per variable (‘custom clamping') helped avoid unrealistically high suitability predictions for non‐analog environments. This study emphasizes the importance of both inspecting response curves and assessing the total area with non‐analog conditions. By making these choices by variable and tail – a first for ENMs – this study and associated code empower researchers to understand how individual extrapolation strategies affect results and to make decisions that should lead to more realistic predictions.
Understanding how deterministic and stochastic processes influence the shape of microorganism community assembly across different spatial scales is essential for disentangling biodiversity patterns. Mires are nutrient-poor and heterogeneous wetlands that form isolated habitats supporting highly diverse diatom assemblages, particularly in mountainous areas, allowing the evaluation of dispersal and species sorting forces. Diatoms are known as ubiquitous microalgae that disperse passively and whose community can endure changing environmental conditions. While their diversity is well documented, the mechanisms driving diatom community assembly across spatial scales remain unclear. Here, we quantified the influence of ecological drift, environmental selection, and dispersal following a null-modelling procedure based on morphology-based phylogenetic distance (beta NTI) and compositional turnover (RCBray) in diatom communities from 18 mountain mires across four Iberian national parks. We conducted the study using a hierarchical spatial design: local (within mire): < 1 km, landscape (within national park): 1.5-45 km, and regional (Iberian Peninsula): 200-800 km (between national parks). Our results reveal a spatial continuum in community organization, shifting from a predominance of homogenizing dispersal at local scales (explaining up to 55.6 +/- 28.5% of community variation) to ecological drift (27.8 +/- 24.7%) and heterogeneous selection (environmental filtering) (22.5 +/- 10.0%) at landscape and regional scales. These findings shed light on the scale-dependent interplay of community assembly processes in shaping diatom biodiversity in aquatic ecosystems, highlighting how the balance between stochastic and deterministic processes shifts across multiple spatial scales.
Understanding how global change reshapes mountain plant communities is essential for predicting biodiversity and ecosystem function in a warming world. Using resurvey data from over 1,400 alpine and subalpine vegetation plots across the European Alps, we show that community-weighted means of key functional traits – specific leaf area, leaf nitrogen, and seed mass – have increased significantly over recent decades, reflecting a widespread shift toward more resource-acquisitive strategies. Yet trait–environment relationships along the elevational gradient have remained remarkably stable, pointing to persistent abiotic filtering. Contrary to expectations, plant height declined slightly, and increases in seed mass were confined to lowland communities mainly, likely due to edaphic constraints and reduced uphill dispersal by large herbivores, respectively. These findings indicate that global change is reshaping the functional structure of montane plant communities by shifting baseline trait composition while leaving the underlying elevational filters largely intact.
Treelines are moving upslope, but the rates and drivers differ among different regions, globally. Many studies have examined the relationship between treeline movement and climate change, particularly rising temperature, while the role of topographical factors has received much less attention, despite the longstanding recognition of its importance. We examined treeline dynamics from 1980 to 2021 in two mountain ranges of Taiwan, the Yushan Mountain Range (YMR) and the Central Mountain Range (CMR), in relation to topography and climate using multi-temporal aerial imagery and U-Net deep-learning segmentation. Treeline elevation advanced approximately 60 m on the two mountains and was accompanied by forest area expansion of 197 ha in the YMR and 158 ha in the CMR, which was mainly due to upward forest movement, not tree densification. Elevation, increases in growing-season temperature and minimum temperature, and growing-season precipitation were important predictors of forest area expansion. In contrast, elevation, increases in growing-season precipitation and maximum temperature, and slope steepness were key predictors of treeline movement. Treeline upslope migration was approximately two times greater during the 1980-2001 period (2.1 in the YMR and 1.6 m year-1 in the CMR) compared to the 2001-2021 period (1.0 in the YMR and 0.75 m year-1 in the CMR), likely because forests already occupied most of the suitable habitat, and topographic factors constrained upward movement. Our study highlights the importance of accounting for both climatic and topographic factors in understanding treeline responses to climate change. In addition, our results indicate that treeline advance in mountain island forests, such as those in Taiwan, may be approaching an elevational limit, above which further upward migration is likely to exhibit a more muted response to further warming.
Accurately measuring biodiversity change remains a central challenge in ecology. Beyond the general idea of quantifying temporal species frequency changes, several sampling-related biases in data collection remain key methodological challenges to consider. Long-term standardized ecological data are rare, and most available datasets exhibit considerable spatial and temporal variation in sampling effort (i.e. unstructured data). Among the available methods, the local frequency scaling approach (Frescalo) has proven particularly effective at addressing these biases. By applying successive spatial and temporal corrections, Frescalo leverages emergent patterns in species assemblages to correct for variation in survey effort. Compared to other similar approaches, Frescalo is particularly well suited to long-term datasets and those with a high number of species. It is also a versatile method, allowing simultaneous estimation of temporal and spatial changes, or even providing diagnostics for survey design or bias assessment. The method's technical complexity, the level of ecological knowledge required, and the challenges of implementation raise a number of practical issues in its application. In this paper, we present a clear and accessible explanation of the Frescalo methodology, offer a step-by-step roadmap to guide users, and highlight the wide range of applications it supports. To further facilitate its adoption, we also introduce an R package designed to simplify implementation.
Species' niche positions and breadths within a region's environmental space, measured through ecological niche factor analysis (ENFA) as marginality and specialization, can reflect evolutionary constraints related to lineage age. The 'internal incumbency' hypothesis predicts that older species, due to competitive preemption, occupy more central niche positions than younger ones. In addition, regarding specialization, there are two contrasting predictions: the 'time-and-specialization' hypothesis posits that older species are more specialized given that they had more time to adapt to an environment, whereas the 'resource-use' hypothesis proposes that younger species are more specialized due to occupying narrower ecological niches, which may drive higher speciation and extinction rates. This study explores the relationship between species age and their climatic niche position and specialization among Neotropical terrestrial vertebrates. Using ENFA, we estimated the marginality (niche position) and specialization (niche breadth) of 1175 mammals, 3001 birds, 1826 reptiles and 1435 amphibians across the Neotropical biomes occupied by each species. Phylogenetic generalized least squares (PGLS) models were used to analyze the relationships between species age, marginality, and specialization across vertebrate classes. Although regression coefficients were predominantly negative, suggesting that older species tend to occupy more central niche positions and exhibit lower specialization, the explanatory power of the models was extremely low. Taken together, the consistently negligible variance explained across taxa and biomes indicates that species age is not a general driver of niche position and specialization, contrary to the expectations articulated in our hypotheses.
Organisms' energy requirements increase with body mass, leading to larger home range areas and lower population density. Previous research has highlighted the differential scaling of these variables in mammals, where species with large home ranges have higher density than expected due to increased home range overlap. Here we investigate this phenomenon in mammals, and hypothesize that home range overlap is influenced by group size, territoriality and habitat dimensionality. We address the question from both a correlative and a mechanistic perspective. First, we compiled data on home range, density, group size, territoriality and habitat dimensionality for 319 species. We then estimated average home range overlap, and explained its variation considering average home range, group size, territoriality and habitat dimensionality. We then complement the comparative analysis with a random-encounter mechanistic model (i.e. gas model) - treating individuals or social units as moving particles whose encounter rates depend on density, movement and an effective interaction distance - to provide independent support to the correlative analysis. Our results support previous studies showing that home range overlap with increasing home range sizes. However, this relationship is lower for species with larger group size due to the effect of spatial clumping on the relationship between body mass and group density, and body mass and individual home range area. Furthermore, such a relationship is higher and steeper in non-territorial species and those foraging in tridimensional habitats. The mechanistic model further supports these hypotheses. Taken all together, our results suggest that home range overlap ultimately depends on both species' area demands and their capacity to monopolize resources. Our study highlights the importance of considering both population density and home range as non-redundant measures of habitat requirements in conservation analyses.
Spatial sampling bias in occurrence data can generate spurious environmental associations in models of species distributions and ecological niches, and can also undermine inferences based on permutation tests made using these models. Geographic randomization tests are often used to generate distributions of expected behavior under the null hypothesis that species occurrences are unrelated to environmental predictors. Here we quantify how ignoring spatial sampling bias in these tests affects Type I error (i.e. false positives) for hypothesis tests and distorts estimates of methodological bias, and we evaluate a simple remedy that samples pseudoabsences in proportion to empirical bias patterns. We used a virtual species approach to simulate scenarios where no ecological processes were operating but where sampling was spatially biased to represent a worst-case scenario. We evaluated the effects of ignoring this bias on Type I error rates for multiple tests of model significance, tests of niche similarity, and tests for measuring methodological bias in model transfer scenarios. When randomization tests were conducted without accounting for sampling bias, Type I error was substantially inflated for hypothesis tests. Incorporating the correct spatial sampling bias model into randomization analyses reduced error rates for all randomization tests, and in most cases brought them down to approximately nominal levels. Analyses of empirical data for a group of South American ant species produced qualitatively similar but generally attenuated effects, but in some cases test results became significant only after bias was accounted for. We conclude that incorporating empirical spatial sampling bias estimates into geographic randomization tests substantially mitigates inflated Type I error and can materially alter estimates of methodological bias. We recommend including spatial sampling bias estimates in randomization tests where possible, and estimates of spatial autocorrelation where reliable bias estimates are not practical.
Plant-pollinator interactions are essential for maintaining biodiversity and supporting food production, yet inferences drawn from network syntheses may be shaped by where interaction data are generated and which datasets are most reused. Here, we quantify the global distribution of published plant-pollinator networks, assess how publication rates vary across continents after accounting for socioeconomic, geographic, and ecological predictors, and evaluate whether network reuse is driven primarily by time since publication and/or by continent-specific reuse trajectories. We systematically screened 3390 studies and assembled 721 plant-pollinator networks published between 1923 and 2024. Although networks were available from 58 countries, nearly half originated from just six - Brazil, the US, Spain, Germany, the UK, and Argentina - highlighting strong concentration in data generation. Across countries, publication rates increased with national research investment, country area, and bee species richness. After accounting for these predictors, publication rates in South America, Africa, and Oceania were comparable to those in North America, whereas Europe remained underrepresented relative to most other continents, and Asia showed persistently low representation. Network reuse increased with time since publication but varied among continents: South American networks exhibited higher reuse after roughly a decade since publication than networks from North America, Europe, Asia, and Oceania (and marginally more than Africa), while reuse accumulated more rapidly through time for networks from Europe, Africa, and Asia than for networks from North America. Together, these results show that plant-pollinator network knowledge is shaped by coupled biases in both data sampling and reuse, underscoring the need for targeted sampling, improved data mobilization and discoverability, and more equitable collaborations to broaden the geographic basis of global network synthesis.
Grassland ecosystems are facing rapid and ongoing change driven by intensified land-use and accelerated climate change, highlighting the urgent need to understand their potential adaptation and response to environmental change. We analyzed data from 52 980 vegetation plots spanning all major grassland habitats in Europe (including alpine, rocky, sandy, saline, dry, mesic and wet grasslands). We quantified competitive (C), stress-tolerant (S), and ruderal (R) strategies for 7858 plant species based on key functional traits and mapped the spatial patterns of C-, S- and R-strategies across European grasslands. Using random forest models, we evaluated the relative importance of environmental factors in shaping these patterns and explored potential changes in the distribution of C-, S- and R-strategies under future climate scenarios. We further investigated how these strategies and their environmental drivers vary across grassland habitats. Our results revealed a clear biogeographical gradient in the distribution of C-, S- and R-strategies from predominantly stress-tolerant strategies in Mediterranean grasslands to greater representation of competitive and ruderal strategies in temperate regions. Climate and soil factors emerged as major drivers shaping these patterns at the continental scale. Projected responses to future climate change varied among regions: grasslands in the Atlantic and Continental regions were projected to decrease in C-strategy and increase in both S- and R-strategy representation, whereas grasslands in Arctic and boreal regions exhibited contrasting trends. Mediterranean grasslands were projected to undergo a transition from mainly R-strategy toward S-strategy species predominance. Furthermore, the strategy patterns and their key drivers differed among grassland habitats, with patterns largely reflecting habitat-specific environmental constraints. This study demonstrates the utility of Grime's CSR framework for characterizing broad-scale patterns of plant adaptive strategies across diverse grassland habitats. It highlights region- and habitat-specific differences in potential responses to future climate change, with implications for targeted grassland management and restoration.
