Diversity Patterns of Pollinating Insect Communities in the Central Qinling Mountains and Their Responses to Environmental Change

Diversity Patterns of Pollinating Insect Communities in the Central Qinling Mountains and Their Responses to Environmental Change | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Diversity Patterns of Pollinating Insect Communities in the Central Qinling Mountains and Their Responses to Environmental Change Yaoyao Si, Qiqi Deng, Qingquan Xue, Yalin Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9444014/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study investigated pollinator community structure and its environmental drivers across different habitats in the central Qinling Mountains, a key biodiversity hotspot in China. Field surveys were conducted at approximately 40 sampling sites from March to September during 2024–2025, covering multiple seasons, with concurrent measurements of climatic, topographic, and vegetation variables. Alpha diversity patterns revealed strong habitat-dependent differences, with natural habitats supporting higher species richness and Shannon diversity, whereas agricultural habitats exhibited lower diversity but higher dominance. Semi-natural habitats showed intermediate characteristics. Taxonomic groups responded differently, indicating uneven sensitivity to habitat change. Beta diversity analyses further demonstrated significant community differentiation among habitats, with natural and agricultural habitats showing the greatest dissimilarity and semi-natural habitats acting as transitional systems. Environmental analyses indicated that pollinator community structure and diversity were jointly shaped by landscape composition, climatic conditions, and plant resource availability. Forest cover was positively associated with diversity maintenance, whereas increasing agricultural proportion and wind disturbance generally exerted negative effects. Precipitation and humidity showed overall positive influences, but responses varied among taxonomic groups. Importantly, interaction analyses revealed that environmental effects were strongly context-dependent. Landscape structure and floral resource availability significantly mediated climatic influences on community structure, diversity, and abundance. Forest cover tended to buffer climatic stress, while agricultural expansion amplified adverse environmental effects. These findings highlight that pollinator community responses are governed by the combined and interactive effects of climate, landscape, and resource gradients, emphasizing the importance of multi-factor mechanisms in shaping community dynamics. pollinators community structure environmental drivers interaction effects habitat gradients biodiversity Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Pollinators are essential components of terrestrial ecosystems because they support plant reproduction, biodiversity maintenance, and ecosystem stability (Liu et al., 2018 ). Although bees are often emphasized in pollination research, pollinator assemblages are taxonomically and functionally diverse, including Coleoptera, Diptera, Hymenoptera, Lepidoptera, and some Hemiptera (Luo & Lei, 2003 ; Yang, 2018 ; Wu & Zheng, 2019; Orford et al., 2015 ; Muinde & Katumo, 2024 ). These groups differ substantially in life-history traits, dispersal ability, and resource use, and therefore may respond differently to environmental change. Community-level approaches are thus necessary to understand the mechanisms shaping pollinator diversity and composition. Global declines in pollinator diversity and abundance have raised major concerns about ecosystem functioning and food security. Long-term monitoring has revealed marked reductions in insect biomass and diversity in parts of Europe (Müller et al., 2024 ), and similar patterns have been increasingly reported worldwide (Garratt et al., 2021 ; Dicks et al., 2021 ; Ratto et al., 2021 ). The Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) identified multiple interacting drivers of pollinator decline, including land-use change, agricultural intensification, pesticide use, invasive species, and climate change (Dicks et al., 2021 ). These drivers rarely act in isolation, and their combined effects may strongly influence pollinator community structure. Climate change is one of the most important drivers affecting pollinators. Temperature, precipitation, humidity, and wind can influence insect physiology, dispersal, survival, and phenology, while also altering floral resource availability and plant–pollinator synchrony (Scaven & Rafferty, 2013 ; Freimuth et al., 2022 ; Skendžić et al., 2021 ; Nealis, 2020 ). Warming has been associated with shifts in insect distributions toward higher elevations and latitudes (McCain & Garfinkel, 2021 ; Halsch et al., 2020 ), whereas climatic extremes may disrupt ecological interactions and alter community composition (Harvey et al., 2023 ; Miao et al., 2021 ). However, responses are often taxon-specific: some groups may benefit from warmer conditions, whereas others may decline under physiological stress or changing resource regimes (Fan et al., 2024 ; Adams et al., 2020 ; Ghisbain et al., 2021 ). In addition, climatic variability and extreme events may exert stronger effects than mean climatic conditions alone (Kellermann & van Heerwaarden, 2019 ; Sridhar et al., 2020 ). Landscape structure and land-use change are also key determinants of pollinator diversity. Agricultural intensification and habitat fragmentation reduce habitat availability, simplify landscapes, and limit nesting and foraging resources, thereby decreasing pollinator diversity and promoting community homogenization (Buchori et al., 2019 ; Kovács-Hostyánszki et al., 2017 ; Miao et al., 2021 ; Fahrig, 2003 ). Urbanization may further intensify these pressures by increasing habitat isolation and environmental stress (Fenoglio et al., 2021 ; Ryalls et al., 2022 ). In contrast, semi-natural habitats and heterogeneous landscapes often support higher pollinator diversity and more stable communities by providing complementary habitats and continuous resource supply (Crist & Peters, 2014 ; Zhou et al., 2020 ; Chowdhury et al., 2023 ). Landscape configuration, including patch size and connectivity, can further influence species dispersal and community assembly (Meloni et al., 2020 ; Riva et al., 2024 ), and historical land use may leave persistent legacy effects on present-day insect communities (Hahn & Orrock, 2015 ). Floral resources represent another key component of pollinator community dynamics. The abundance and diversity of flowering plants determine nectar and pollen availability, thereby shaping pollinator richness, abundance, and foraging behavior (Miñarro Prado et al., 2018 ; Galloway et al., 2021 ; Zhang et al., 2023 ). However, floral-resource effects are often context dependent. Under resource-limited conditions, increased flower diversity may enhance pollinator richness, whereas in resource-rich environments additional floral resources may lead to saturation effects or intensified competition (Outhwaite et al., 2022 ). Consequently, interactions among climate, landscape structure, and floral resources are likely to be crucial for understanding pollinator community responses. Mountain ecosystems provide ideal natural laboratories for investigating these interactions because they contain strong environmental gradients and high habitat heterogeneity. The Qinling Mountains form a major biogeographical boundary and biodiversity hotspot in China (Fang et al., 2017 ; Li et al., 2021 ). This region is characterized by complex topography, diverse habitat types, and pronounced climatic gradients, but it has also experienced increasing anthropogenic pressures, including land-use change, forest degradation, and habitat fragmentation (Liu & Qiu, 2024 ; Yang et al., 2024 ; Liu et al., 2024 ). Spatial heterogeneity in habitat quality remains substantial, providing an ideal setting for examining how multiple environmental gradients shape pollinator communities. Despite growing attention to pollinator ecology, studies that simultaneously integrate climatic factors, landscape structure, habitat differentiation, and floral resources at the community level remain limited, particularly in montane ecosystems such as the central Qinling Mountains. Moreover, most previous studies have focused on single drivers or specific taxa, whereas the combined and interactive effects of multiple environmental factors on pollinator diversity and community composition remain poorly understood. To address these gaps, we investigated pollinator communities across multiple habitat types in the central Qinling Mountains. Specifically, we aimed to: (1) compare pollinator alpha diversity and community composition among habitat types; (2) quantify the relative contributions of climatic, topographic, landscape, and floral-resource variables; and (3) evaluate how interactions among these factors mediate pollinator diversity, abundance, and community structure. By linking community patterns with multiple environmental drivers, this study provides a community-level perspective on pollinator responses to environmental change in a montane biodiversity hotspot. Materials and methods Study area and sampling design The study area was selected within the Qinling Mountains based on spatial analyses conducted in ArcGIS. Multi-year mean annual temperature (MAT) data (Zhang et al., 2025) and land-cover data (Yang & Huang, 2025) were used to characterize environmental gradients. The study region was classified into five temperature-gradient categories according to MAT values (Table 2-1). In addition, thresholds for land-cover types were defined based on regional characteristics of the Qinling Mountains. According to the proportions of forest cover (For_prop) and cropland cover (Cro_prop), the sampling sites were classified into three habitat types: agricultural habitat (Cro_prop > 50%), natural habitat (For_prop > 75%), and semi-natural habitat (Cro_prop > 25% and For_prop > 50%) (Table S1). Approximately 40 sampling sites were selected in Foping County, Yangxian County, Chenggu County, and Ningshan County, Ankang City (Fig. S1). Field surveys were conducted from March to September in 2024-2025 across multiple seasons to capture temporal variation in pollinator communities and to ensure data representativeness and reliability. Environmental factors Land-cover data Land-cover data were obtained from the annual 30-m resolution land-cover dataset for China (Yang & Huang, 2025). Using ArcGIS, a circular buffer with a radius of 150 m was established around each sampling transect. The proportion of each land-cover type within the buffer was calculated to quantify landscape composition, including forest cover (For_prop), cropland cover (Cro_prop), and impervious surface cover (Imp_prop). In addition, elevation (Ele) and flowering plant species richness (Flower_richness) were recorded during field surveys. Climatic data Climatic variables were obtained from the National Cryosphere Desert Data Center (Zhang et al., 2025). The extracted variables included minimum temperature at 2 m (MATmin), mean temperature at 2 m (MAT), maximum temperature at 2 m (MATmax), relative humidity (RH), mean annual precipitation (MAP), and wind speed at 10 m height (WS10) for each sampling site. All environmental variables were standardized prior to statistical analyses. Pollinator sampling and identification Pollinating insects were surveyed using a combination of sweep-net sampling and transect-based visual observations. Non-butterfly taxa were collected along approximately 2-km transects at each site using insect nets in the surrounding herbaceous and shrub vegetation. Ecological photographs were taken during sampling to assist subsequent identification. Butterflies were surveyed using the Pollard walk method at a walking speed of 1.5–2 km/h. All individuals observed within 2.5 m on each side of the transect, within 5 m above ground, and along a total transect length of 2 km were recorded. Individuals that could not be reliably identified in the field were collected for laboratory identification. Small insect specimens were preserved in 75% ethanol, whereas butterfly specimens were stored in triangular paper envelopes. In the laboratory, specimens were frozen at −80 °C for 1–2 days, then pinned, labeled, and preserved in specimen boxes. Specimens were identified morphologically using taxonomic literature and reference collections. DNA barcoding was used when necessary, and COI sequences were compared against the NCBI database using BLAST; sequence similarity ≥ 98% was considered supportive of species-level identification. Diversity and community analyses Data processing and diversity analyses Geographic and environmental data were extracted using ArcGIS, and insect and environmental datasets were organized in Excel. All statistical analyses were conducted in R 4.4.2. Species abundance and environmental matrices were matched by sampling site, and variables were converted to numeric format prior to analysis. Alpha diversity Sampling completeness and expected species richness were evaluated using rarefaction–extrapolation analyses implemented in vegan and iNEXT. Alpha diversity was quantified using species richness, Shannon diversity, Simpson dominance, and total abundance. Differences among habitat types were tested using one-way ANOVA followed by Tukey’s post hoc comparisons. Species abundance distribution was described using Preston’s octave analysis. Sampling completeness was evaluated using sample-size-based rarefaction and extrapolation analyses implemented in vegan and iNEXT, with 95% confidence intervals estimated from 500 bootstrap replicates. Detailed descriptions of diversity indices and statistical analyses are provided in the Supplementary Information (Section S1.1). Beta diversity and community structure Community composition was analyzed using Bray–Curtis dissimilarity and visualized by non-metric multidimensional scaling (NMDS). Jaccard similarity was used to quantify species sharing among habitats (see Supplementary Methods, Section S1.2). Dominant species were identified using importance values (IV), and indicator species were identified using the IndVal method. Significant indicator species were visualized using Z-score standardized heatmaps with hierarchical clustering. Environmental gradients and driver analyses Environmental variables and drivers Environmental variables included landscape, climatic, topographic, and floral-resource factors. Land-cover data (Yang and Huang, 2025) were used to calculate forest cover (For_prop), cropland cover (Cro_prop), and impervious surface cover (Imp_prop) within a 150 m buffer around each transect in ArcGIS. Elevation (Ele) and flowering plant species richness (Flower_richness) were recorded during field surveys. Climatic variables, including MATmin, MAT, MATmax, RH, MAP, and WS10, were obtained from the National Cryosphere Desert Data Center (Zhang et al., 2025). Principal component analysis (PCA) and Spearman correlation analysis were used to summarize environmental variation and assess collinearity among variables. Relationships between community composition and environmental gradients were examined using RDA or CCA, selected according to DCA results, with model significance assessed by permutation tests. Statistical modeling Generalized linear models (GLMs) and linear models (LMs) were used to evaluate environmental effects on diversity metrics. Species richness was analyzed using Poisson or negative binomial GLMs, whereas Shannon diversity and abundance were analyzed using Gaussian linear models. Three model types were fitted: (1) main-effect models to assess independent environmental effects; (2) habitat interaction models to test environment × habitat interactions; and (3) environmental interaction models to test pairwise interactions among environmental variables using binary grouping (low vs. high). Model significance was assessed using parameter estimates and associated p-values, and fitted response curves were used to visualize effect directions and strengths. Model structures and parameter definitions are provided in Supplementary Information (Section S1.3). Results Sampling completeness and alpha diversity patterns of pollinator communities Sample-size-based rarefaction and extrapolation analyses (iNEXT) indicated high sampling completeness across habitats and taxonomic groups, with sample coverage generally exceeding 96% and species accumulation curves approaching asymptotes, suggesting that sampling effort was sufficient to capture pollinator diversity (Fig. S2). Table 1. Taxonomic composition and diversity indices of insect communities across different habitats Habitat Family Genus Richness Abundance Shannon Diversity Index Simpson Dominance Index Agricultural 35 178 230 4215 3.7823 0.0533 Semi-natural 41 261 362 4586 4.3587 0.0421 Natural 40 272 376 3647 4.7181 0.0216 Total 47 389 601 12448 4.5458 0.0321 Data in the table 1. show the taxonomic composition and diversity indices of insect communities across different habitats. Family and genus refer to the numbers of families and genera, respectively; richness refers to species richness, and abundance refers to the total number of individuals. The Shannon diversity index describes community diversity, whereas the Simpson dominance index reflects the degree of dominance concentration. “Total” indicates the pooled value across all habitats. Table 2. Taxonomic composition and diversity indices of different insect community groups Data in the table 2. show the taxonomic composition and diversity indices of different insect community groups. Family and genus refer to the numbers of families and genera, respectively; richness refers to species richness, and abundance refers to the total number of individuals. The Shannon diversity index describes the diversity level of each group, whereas the Simpson dominance index reflects the degree of dominance concentration. “Total” indicates the pooled value across all groups. Icons represent different taxonomic groups. In total, 12,448individuals representing5orders, 47families, 389genera, and 601 species were recorded, with natural habitats harboring the highest species richness, followed by semi-natural and agricultural habitats (Table 1-2). Alpha diversity patterns showed that species richness was highest in Coleoptera and lowest in Diptera, whereas Shannon diversity peaked in Lepidoptera; across habitats, both richness and Shannon diversity followed the gradient natural > semi-natural > agricultural, while Simpson dominance showed the opposite trend. Agricultural habitats and Diptera groups were characterized by relatively low diversity but high dominance, whereas natural habitats and Coleoptera groups exhibited higher diversity and more even species distributions. Although abundance was higher in semi-natural and agricultural habitats, no significant differences in abundance were detected among habitats, indicating that habitat effects were mainly reflected in species composition and evenness rather than total individual numbers. Significant habitat differences in Shannon diversity were observed only for Diptera, while most other groups showed similar patterns without significant differences between natural and semi-natural habitats. Across all habitats, Lepidoptera consistently exhibited the highest diversity and represented the dominant pollinator group, whereas Hemiptera showed relatively high abundance in semi-natural habitats. Overall, these results suggest that habitat differences primarily influence community structure by altering species composition and dominance patterns, rather than total abundance, thereby contributing to increased homogenization under disturbed conditions. Beta diversity and community differentiation Pollinator community composition differed significantly among habitat types at both sampling-event and site scales. Non-metric multidimensional scaling (NMDS) based on individual sampling events revealed a clear separation among habitats (PERMANOVA, p = 0.001), although the explained variance was relatively low (R² = 0.06; Fig. 1a), indicating substantial within-group variability, potentially driven by seasonal or temporal dynamics. At the site scale, where repeated sampling events were aggregated for each site, community differentiation among habitats remained significant (p = 0.001), with a notably higher explanatory power (R² = 0.13; Fig. 1b). Overall, pollinator communities exhibited partial separation among habitat types. The greatest dissimilarity was observed between natural and agricultural habitats, whereas semi-natural habitats displayed an intermediate position. Despite this separation, considerable overlap among habitats was still evident, suggesting a degree of community similarity. Species composition and habitat specificity Principal component analysis (PCA) (Fig. S5) showed that agricultural intensity, temperature extremes, elevation, and flowering plant richness were the main sources of environmental differentiation among sampling sites. Habitat types were clearly separated in environmental space, with agricultural habitats associated with greater agricultural intensity, higher temperatures, lower humidity, and stronger disturbance, natural habitats associated with higher elevation, more humid conditions, and richer floral resources, and semi-natural habitats occupying an intermediate position. Indicator species analysis and importance value (IV) results were consistent with this pattern (Fig. S6, Fig. S7), revealing clear habitat-specific structuring of pollinator communities. Agricultural habitats were characterized by disturbance-tolerant and generalist taxa, such as Pieris rapae and Apis ceran a, whereas natural habitats supported distinct dominant and indicator species associated with less disturbed vegetation, including Neptis sappho and Celastrina argiolus . Semi-natural habitats exhibited intermediate composition, with both agricultural and natural-associated species (e.g., Vanessa cardui ) occurring across habitats. Together, these results support the interpretation that environmental filtering along gradients of land use, resource availability, and disturbance intensity underlies the observed community differentiation. Effects of environmental variables on diversity Detrended correspondence analysis (DCA) indicated that canonical correspondence analysis (CCA) was appropriate for ordination of community data. After removing highly collinear variables using variance inflation factor (VIF < 10), the retained environmental variables represented independent contributions to variation in community composition. The CCA ordination revealed clear separation among habitat types along environmental gradients. Site scores were jointly structured by species composition and their relationships with environmental variables. The length of environmental vectors indicated their relative influence on community structure, while ellipses represented 95% confidence intervals for each habitat type, illustrating the degree of separation among agricultural, semi-natural, and natural communities. Natural habitat sites (blue points) were primarily distributed in the upper-right quadrant and were associated with higher elevation (Ele) and stronger wind speed (WS10). Semi-natural habitat sites (orange points) were located in the central region of the ordination space, showing moderate associations with cropland proportion (Cro_prop) and impervious surface proportion (Imp_prop). In contrast, agricultural habitat sites (yellow points) were mainly clustered in the lower-left quadrant and were strongly associated with higher cropland and impervious surface proportions. The vectors of MATmax and MATmin also pointed toward the lower-left quadrant, indicating that temperature extremes influenced community composition in both agricultural and natural habitats (Fig. 2). To further evaluate how communities respond to environmental variation, we analyzed the effects of multiple environmental drivers on insect diversity. Environmental driver analyses were conducted for three diversity metrics—species richness (S), Shannon diversity (H′), and abundance (A)—across six pollinator groups. These analyses included main-effect models, habitat interaction models, and environmental interaction models, and only significant results (p ≤ 0.05) were retained. Among the significant results (n = 574), environmental interaction models accounted for the largest number of significant patterns (n = 412), followed by main-effect models (n = 104) and habitat interaction models (n = 58). Across response variables, species richness showed the highest sensitivity (n = 246), followed by Shannon diversity (n = 167), whereas abundance showed the fewest significant responses (n = 161). Main effects analysis Based on the main-effect models, multiple environmental variables—including temperature (MAT, MATmax, MATmin), habitat structure, elevation (Ele), wind speed (WS10), precipitation (MAP), relative humidity (RH), and flowering plant richness—had significant effects on pollinator communities. However, the direction of these effects varied among taxa and diversity metrics. where “+” and “−” indicate the number of positive and negative significant effects, respectively. Overall, forest cover (For_prop) showed predominantly positive effects on community metrics (+24/−1), whereas cropland cover (Cro_prop) was mainly associated with negative effects (+0/−24). Precipitation (MAP) and relative humidity (RH) generally exhibited positive relationships with diversity metrics (MAP: +17/−0; RH: +9/−0), while wind speed (WS10) and temperature (MAT) were mostly associated with negative responses (WS10: +0/−8; MAT: +0/−5). These patterns suggest that increased forest cover and higher moisture availability are generally linked to enhanced pollinator diversity, whereas higher cropland proportion and elevated temperature or wind disturbance are more consistently associated with declines in community diversity. Temperature variables showed contrasting effects across taxonomic groups and diversity metrics. With increasing maximum temperature (MATmax), Shannon diversity of the overall community (sampling-event scale) and both Shannon diversity and species richness of Lepidoptera declined (Fig. 3a), indicating negative effects of higher temperature extremes. In contrast, Shannon diversity of Hymenoptera increased with rising minimum temperature (MATmin) (Fig. 3b), suggesting a potential positive effect of milder minimum temperatures on this group. Increasing mean temperature (MAT) was associated with declines in species richness of the overall community, Coleoptera (site scale), and Lepidoptera (sampling-event scale), as well as reductions in Shannon diversity of the overall community and Lepidoptera (Fig. 3c; Fig. S8(f–g)), indicating an overall negative effect of warming. Landscape structure showed consistent but contrasting effects between forest and cropland proportions. Increasing forest cover was associated with higher species richness and Shannon diversity across all six groups (Fig. 3d), suggesting a generally positive effect on community diversity. However, abundance responses varied among taxa, with Hymenoptera abundance decreasing and Coleoptera abundance (sampling-event scale) increasing with forest cover (Fig. 3e). In contrast, increasing cropland proportion was associated with declines in species richness and Shannon diversity across all groups, along with reduced Coleoptera abundance (Fig. 3f; Fig. S8(c–e)), indicating an overall negative effect of agricultural expansion. Among topographic and climatic factors, species richness of the overall community (site scale) increased with elevation (Fig. 3g), suggesting greater diversity at higher altitudes. Wind speed generally showed negative effects, with Shannon diversity and/or species richness of multiple groups (including the overall community, Coleoptera, Lepidoptera, and Hymenoptera) decreasing as wind speed increased (Fig. 3h; Fig. S8(i)). In contrast, precipitation and relative humidity were generally associated with positive effects. Shannon diversity increased with precipitation across all six groups (Fig. 3i), and species richness of several groups (including the overall community, Coleoptera, Lepidoptera, and Hymenoptera) also increased, accompanied by higher Hymenoptera abundance (Fig. S8 (e–f)). Similarly, increasing humidity was associated with higher Shannon diversity in multiple groups (Fig. 3j) and increased species richness in the overall community, Lepidoptera, and Hymenoptera (Fig. S8 (c–e)), suggesting that wetter conditions generally promote community diversity. The effects of floral resources differed markedly among taxa. Shannon diversity of Diptera (site scale) decreased with increasing flowering plant richness (Fig. 3l), whereas Shannon diversity of Lepidoptera (Fig. 3k), as well as species richness of the overall community and Lepidoptera, increased. Hemiptera abundance decreased with increasing floral richness, whereas Lepidoptera abundance increased (Fig. S8(b–c)). These results indicate strong taxon-specific responses to floral resource availability. Overall, the main effects of environmental variables suggest that forest cover and higher moisture conditions generally promote pollinator diversity, whereas cropland expansion and strong wind disturbance have negative impacts. In contrast, the effects of temperature and floral resources are more complex and vary depending on taxonomic group and diversity metric. Habitat interaction effects analysis In the habitat interaction models, the direction of interaction effects was approximately balanced overall (+33/−34), but clear differences emerged among environmental variables. Interaction terms involving precipitation (MAP) were predominantly negative (+0/−10), whereas those involving temperature variables (MAT and MATmax) were mainly positive (MAT: +8/−0; MATmax: +6/−0). These results indicate that the effects of climatic variables are habitat-dependent. Responses of pollinator communities to climatic factors differed markedly among habitat types. Natural, agricultural, and semi-natural habitats frequently showed contrasting, and in some cases opposite, response patterns to temperature, wind speed, precipitation, humidity, and elevation. For temperature variables, natural habitats generally exhibited response patterns distinct from those of agricultural and semi-natural habitats. With increasing MATmax, Shannon diversity of Coleoptera at the sampling-event scale (Fig. 4a), species richness of Diptera at the site scale, and abundance of the overall assemblage and Hemiptera at the sampling-event scale increased in natural habitats, whereas opposite trends were observed in agricultural and semi-natural habitats. Hymenoptera showed a more complex pattern: abundance at the sampling-event scale increased with MATmax in natural and semi-natural habitats but decreased in agricultural habitats, while abundance at the site scale decreased in natural and agricultural habitats but showed the opposite trend in semi-natural habitats (Fig. S9(c)). Under MATmin, species richness of Hemiptera and Diptera at the site scale increased with temperature in natural habitats (Fig. 4b), but declined in agricultural and semi-natural habitats. In contrast, Hymenoptera abundance at the site scale decreased in natural and agricultural habitats, but increased in semi-natural habitats. Species richness of Lepidoptera at the site scale declined in natural habitats but increased in the other two habitats (Fig. 4c). Similar patterns were observed for MAT: species richness of Diptera, abundance of Hymenoptera at the site scale (Fig. 4d), Shannon diversity of Coleoptera at the sampling-event scale, and abundance of the overall assemblage and Hemiptera generally increased with temperature in natural habitats, but showed opposite responses in agricultural and semi-natural habitats (Fig. S9(c)). Along the elevational gradient, species richness of Diptera (Fig. 4e), as well as abundance of the overall assemblage and Hemiptera, decreased with elevation in natural habitats, whereas the opposite trends were generally observed in agricultural and semi-natural habitats. By contrast, species richness of Lepidoptera at the site scale increased with elevation in natural habitats (Fig. 4f) but declined in the other two habitats (Fig. S9(a)). Wind speed also produced habitat-specific effects. In natural habitats, Shannon diversity and species richness of Lepidoptera increased with wind speed (Fig. 4h), whereas species richness of Diptera at the site scale (Fig. 4g), species richness of Coleoptera at the sampling-event scale, and abundance of Diptera and Hemiptera at the sampling-event scale generally declined. Most of these relationships were reversed in agricultural and semi-natural habitats. Abundance of Coleoptera at the site scale increased with wind speed in natural and semi-natural habitats, but decreased in agricultural habitats (Fig. S9(d)). Precipitation effects also differed among habitats. Shannon diversity of Hemiptera at the site scale declined with increasing precipitation in natural habitats, but increased in agricultural and semi-natural habitats. Shannon diversity of the overall assemblage at the sampling-event scale and of Coleoptera at the site scale increased with precipitation in natural and semi-natural habitats, but decreased in agricultural habitats. Meanwhile, abundance of the overall assemblage, Lepidoptera (Fig. 4i), Coleoptera, and Hemiptera declined with precipitation in natural habitats, but showed the opposite trends in the other two habitats (Fig. S9(a–b)). For humidity, abundance (Fig. 4j) and species richness of Diptera at the site scale increased with relative humidity in natural habitats, whereas Shannon diversity and species richness of the overall assemblage, Coleoptera, and Hemiptera generally declined. Agricultural and semi-natural habitats often exhibited opposite response patterns (Fig. S9(c–d)). Overall, natural habitats differed markedly from the more disturbed agricultural and semi-natural habitats in their responses to climatic factors. In natural habitats, increasing temperature often promoted some community metrics, whereas increasing precipitation and humidity were frequently associated with declines in diversity or abundance. By contrast, agricultural and semi-natural habitats more often exhibited response patterns opposite to those observed in natural habitats. Environmental interaction effects analysis The results further demonstrated that these moderating effects followed clear directional patterns. Interaction terms involving cropland cover and temperature, including Cro_prop × MAT and Cro_prop × MATmax, were predominantly negative (+0/−19 and +0/−15, respectively). In contrast, interactions involving forest cover and temperature, including For_prop × MAT and For_prop × MATmax, were mainly positive (+19/−0 and +14/−0, respectively), whereas For_prop × Ele was predominantly negative (+0/−19). Similarly, MAP × MAT was mainly positive (+15/−0), while MAP × Ele was predominantly negative (+0/−14). Taken together, these results indicate that pollinator communities in the study region were not driven by single environmental factors in a simple linear manner. Instead, land-use variables, hydrothermal conditions, and topographic gradients jointly shaped community structure through significant interactions, producing differential response patterns in species richness, abundance, and Shannon diversity. From a landscape perspective, habitat structure not only modified the direction of environmental effects, but also altered the strength of climatic influences on pollinator communities. In areas with high forest cover and high cropland cover, the effects of temperature, wind speed, precipitation, and humidity were frequently opposite. Under high forest cover, increasing MATmax was associated with higher Shannon diversity of Diptera at the site scale, higher species richness of Diptera and the overall assemblage at the sampling-event scale, and greater abundance of the overall assemblage, Hemiptera, and Hymenoptera at the site scale, whereas opposite trends were observed in areas with low forest cover. In contrast, Shannon diversity of Hemiptera at the site scale declined with increasing MATmax under high forest cover (Fig. 5a), but increased in low-forest landscapes, indicating divergent responses among taxa and community metrics (Fig. S10(u-v)). This general pattern was maintained for MATmin and MAT. Under high forest cover, species richness of Diptera at the site scale, abundance of Hemiptera at the site scale, and abundance of the overall assemblage, Hemiptera, and Hymenoptera generally increased with temperature (Fig. 5c), whereas the opposite pattern was found under low forest cover. However, Shannon diversity of Hemiptera at the site scale and species richness of Lepidoptera at the site scale declined with increasing MATmin under high forest cover (Fig. 5b), suggesting that forested landscapes may simultaneously enhance and filter pollinator responses to warming (Fig. S10(t, v)). In contrast, under high cropland cover, increasing MATmax, MATmin, and MAT generally reduced abundance of the overall assemblage, Hymenoptera, and Hemiptera (Fig. S10(s–t)), while species richness of Diptera at the site scale also declined (Fig. 5h). Low-cropland landscapes usually showed opposite trends, suggesting that cropland-dominated areas are more vulnerable to warming. Wind effects were also strongly mediated by landscape context. Under high forest cover, abundance of Coleoptera at the site scale, Shannon diversity of Hemiptera at the site scale, and species richness of Lepidoptera increased with wind speed (Fig. 5d), whereas abundance of Diptera at the site scale declined; most of these trends were reversed under low forest cover (Fig. S10(ac)). Under high cropland cover, species richness of Coleoptera and Lepidoptera at the sampling-event scale, as well as abundance of Coleoptera at the site scale, declined with increasing wind speed, whereas abundance of Diptera at the site scale increased (Fig. 5j; Fig. S10(aa)), indicating that wind disturbance in cropland landscapes may suppress most groups while favoring some disturbance-tolerant taxa. Precipitation and humidity also showed contrasting effects under different landscape backgrounds. In high-forest landscapes, several community metrics showed weak or negligible responses to increasing precipitation, including Shannon diversity of the overall assemblage at the sampling-event scale, species richness of the overall assemblage (Fig. 5e), and species richness of Coleoptera at the site scale. By contrast, Shannon diversity of Hemiptera at the site scale, species richness of Lepidoptera at the sampling-event scale, and abundance of Lepidoptera at the site scale declined with increasing precipitation, whereas most of these indicators increased under low forest cover (Fig. S10(p)). In high-cropland landscapes, abundance of the overall assemblage and richness- or diversity-related metrics of Lepidoptera, Hemiptera, and Coleoptera generally increased with precipitation (Fig. 5k; Fig. S10(m-n)), whereas opposite trends were found in low-cropland landscapes. Humidity responses showed a similar landscape dependency. Under high forest cover, species richness and Shannon diversity of Coleoptera at the sampling-event scale declined with increasing humidity, whereas abundance of Diptera at the site scale and species richness of Hymenoptera at the site scale increased (Fig. 5f; Fig. S10(x)). Under high cropland cover, species richness and Shannon diversity of the overall assemblage, Coleoptera (Fig. 5l), and Hemiptera generally increased with humidity, whereas the opposite pattern was observed under low cropland cover. In addition, abundance of Diptera at the site scale declined with increasing precipitation in cropland-dominated landscapes, further indicating that landscape background modifies community adaptation to moist conditions (Fig. S10(v)). Overall, warming tended to promote multiple community metrics in forest-dominated landscapes, but more often led to community decline in cropland-dominated landscapes. Likewise, the effects of wind, precipitation, and humidity frequently reversed depending on landscape context, highlighting landscape structure as a key moderator of climatic effects on pollinator communities. Significant interactions were detected between climatic variables and landscape structure, with both forest and cropland proportions substantially modifying the direction and strength of climatic effects on pollinator communities. For temperature–forest interactions, MATmax, MATmin, and MAT showed consistent patterns. Under high-temperature conditions, increasing forest cover was generally associated with higher Shannon diversity, species richness, and abundance across the overall assemblage (Fig. 6a, e) and multiple taxa, including Coleoptera, Diptera, Hymenoptera, and Hemiptera (Fig. 6b). In contrast, under low-temperature conditions, these relationships were often reversed. This pattern indicates that forested landscapes buffer thermal stress and promote community performance under warming, whereas such effects weaken or reverse under cooler conditions (Fig. S10(i-l)). In contrast, temperature–cropland interactions showed predominantly negative effects. Under high-temperature conditions, increasing cropland cover was associated with declines in Shannon diversity, species richness, and abundance of the overall assemblage and major taxa, whereas opposite trends were generally observed under low-temperature conditions. This pattern was consistent across MATmax, MATmin, and MAT, and was particularly evident for the overall assemblage (Fig. 6b, f) and for Hymenoptera, Diptera, Coleoptera, and Hemiptera (Fig. 6d; Fig. S10(c-f)). These results suggest that cropland-dominated landscapes amplify thermal stress on pollinator communities. Wind-landscape interactions also exhibited contrasting effects between forest and cropland. Under high wind speed, increasing forest cover was associated with higher species richness of Lepidoptera (Fig. 6g), higher Shannon diversity of Hemiptera (sampling-event scale), and increased Lepidoptera abundance, whereas opposite trends were observed under low wind speed (Fig. S10(m)). In contrast, under high wind speed, increasing cropland cover reduced species richness of Lepidoptera (Fig. 6h) and Shannon diversity of Hemiptera, while the opposite trends occurred under low wind conditions (Fig. S10(f)). These results indicate that forest and cropland landscapes exert opposing moderating effects on wind disturbance. Precipitation–landscape interactions showed similar contrasts. Under high precipitation, increasing forest cover promoted Shannon diversity of Lepidoptera (sampling-event scale) (Fig. 6i) and Hemiptera (site scale), species richness of Lepidoptera, and abundance of Diptera at the site scale, but reduced Lepidoptera abundance. Under low precipitation, most of these relationships were reversed (Fig. S10(i)). Conversely, under high precipitation, increasing cropland cover reduced most diversity and richness metrics, while increasing Lepidoptera abundance (Fig. 6g; Fig. S10(b-c)), suggesting that cropland landscapes under wet conditions may suppress diversity but facilitate the dominance of certain taxa. Humidity–landscape interactions also depended strongly on landscape context. Under high humidity, increasing forest cover reduced Lepidoptera abundance (Fig. 6k), Diptera abundance at the site scale, and Shannon diversity of Hemiptera, whereas opposite trends were observed under low humidity (Fig. S10(l-m)). In contrast, increasing cropland cover reduced species richness of Lepidoptera (Fig. 6l) and Shannon diversity of Hemiptera under high humidity, while opposite patterns occurred under low humidity (Fig. S10(f)). Overall, forest and cropland proportions exhibited contrasting moderating roles. Forest-dominated landscapes generally enhanced or buffered pollinator diversity under stressful climatic conditions (e.g., high temperature, wind, or precipitation), whereas cropland-dominated landscapes tended to suppress community metrics. These findings highlight landscape structure as a critical context-dependent regulator of climatic effects on pollinator community composition and stability. Flowering plant richness interacted strongly with climatic variables, and both the direction and magnitude of responses to temperature, wind speed, precipitation, humidity, and elevation varied among taxa under contrasting resource conditions. For temperature variables, flowering plant richness clearly modified thermal effects. At the sampling-event scale, Shannon diversity of Diptera increased with MATmax and MAT under high flowering plant richness (Fig. 7a, h), indicating that abundant floral resources strengthened the positive response of Diptera to warmer conditions. In contrast, Hymenoptera at the site scale showed a different pattern: both species richness and Shannon diversity increased with MATmin under low flowering plant richness, whereas responses were weak or nearly absent under high flowering plant richness (Fig. 7i; Fig. S10(u-v)). This suggests that abundant floral resources may weaken the positive effect of higher minimum temperature on Hymenoptera communities. Wind-speed effects were also resource dependent. Species richness and Shannon diversity of the overall assemblage at the sampling-event scale (Fig. 7e) and of Hymenoptera at the site scale declined with increasing wind speed, but these declines were stronger under low flowering plant richness, indicating that abundant floral resources may buffer wind stress. In contrast, species richness of Lepidoptera at the site scale showed the opposite pattern (Fig. 7f), increasing with wind speed under high flowering plant richness but decreasing under low flowering plant richness, suggesting that Lepidoptera responses to wind are strongly conditioned by resource availability (Fig. S10(a-b)). Under increasing precipitation, different taxa and community metrics showed more complex resource-dependent patterns. Abundance of the overall assemblage (Fig. 7d) and Hymenoptera, as well as species richness and Shannon diversity of Hemiptera at the sampling-event scale, declined with precipitation when flowering plant richness was high, but showed opposite trends under low flowering plant richness. By contrast, species richness and Shannon diversity of the overall assemblage at the sampling-event scale (Fig. 7g), along with species richness of Coleoptera at the sampling-event scale, increased only slightly with precipitation under high flowering plant richness, but increased markedly under low flowering plant richness (Fig. S10(o-p)). These results suggest that precipitation has a stronger positive effect on some community metrics when floral resources are limited. Humidity effects were similarly altered by flowering plant richness. Species richness of the overall assemblage at the sampling-event scale and of Hymenoptera at the site scale (Fig. 7c) increased with relative humidity under high flowering plant richness, but the increase was stronger under low flowering plant richness. In contrast, abundance of the overall assemblage at the sampling-event scale (Fig. 7b), species richness of Lepidoptera at the site scale, abundance of Diptera, and species richness of Diptera at the sampling-event scale declined with increasing humidity under high flowering plant richness, whereas opposite trends were observed under low flowering plant richness (Fig. S10(w-x)). This indicates that humidity effects frequently reverse under contrasting floral-resource conditions. In addition, along the elevational gradient, Shannon diversity of Diptera at the sampling-event scale declined with elevation under high flowering plant richness, but showed the opposite trend under low flowering plant richness (Fig. S10(f)), further demonstrating that resource conditions strongly modify the effects of climatic and topographic factors on community structure. Overall, flowering plant richness not only influenced pollinator communities directly, but also substantially altered the direction and strength of climatic effects, generating pronounced resource-dependent responses in diversity, species richness, and abundance across taxonomic group Significant interactions were detected between climatic variables and flowering plant richness, with contrasting effects across pollinator taxa. For temperature-related variables, MATmax, MATmin, and MAT showed broadly consistent interaction patterns. Under warmer conditions, species richness (Fig. 8a, b) and Shannon diversity (Fig. 8c) of Hymenoptera at the site scale declined with increasing flowering plant richness. In contrast, Shannon diversity of Diptera at the sampling-event scale increased with flowering plant richness under high MATmax and MAT conditions (Fig. 8d, f). In addition, under high MATmin, abundance of Lepidoptera at the sampling-event scale also increased with flowering plant richness (Fig. 8e). These results indicate clear taxon-specific differences in responses to floral-resource availability under warm conditions. Under high precipitation, Shannon diversity of the overall assemblage at the sampling-event scale (Fig. 8g), species richness and Shannon diversity of Hemiptera, and abundance of Coleoptera at the site scale all declined with increasing flowering plant richness (Fig. S10(g)), suggesting that under wetter conditions, greater floral-resource availability may have suppressive effects on some community metrics. Along the elevational gradient, abundance of Lepidoptera at the sampling-event scale decreased with increasing flowering plant richness under high-elevation conditions (Fig. 8h), indicating that the positive effects of floral resources may be constrained at higher elevations. Under high wind speed, abundance of Coleoptera at the site scale increased with flowering plant richness (Fig. 8i), suggesting that this group may be particularly effective at exploiting floral resources under strong wind disturbance. Under high humidity, Shannon diversity of the overall assemblage at the sampling-event scale, as well as Shannon diversity and species richness of Hymenoptera at the site scale, and Shannon diversity of Lepidoptera at the sampling-event scale, all declined with increasing flowering plant richness (Fig. S10(g)). This pattern suggests that under highly humid conditions, greater floral-resource availability may negatively affect pollinator diversity. Overall, the effects of flowering plant richness on pollinator communities were strongly climate dependent. The direction of resource effects varied substantially among climatic conditions and differed markedly among taxonomic groups, indicating pronounced inconsistency in community responses. Discussion Habitat type shapes both α- and β-diversity of pollinator communities Our results indicate that habitat type is a major determinant of both α- and β-diversity in pollinator insect communities. Natural habitats consistently supported higher species richness and Shannon diversity, whereas agricultural habitats were characterized by lower diversity, stronger dominance, and increased community homogenization. Semi-natural habitats generally showed intermediate patterns, highlighting their transitional role along the land-use gradient. Notably, abundance was often higher in semi-natural and agricultural habitats than in natural habitats, suggesting that a greater number of individuals does not necessarily correspond to higher diversity. Instead, elevated abundance in disturbed habitats may reflect the proliferation of a limited number of dominant or disturbance-tolerant taxa, accompanied by reduced evenness and stronger homogenization. Taxon-specific patterns further support this interpretation. Coleoptera exhibited the highest species richness, Lepidoptera showed the highest Shannon diversity, and Diptera had the lowest diversity and appeared to be the most sensitive to habitat change. Together, these results suggest that natural and semi-natural habitats are more effective than agricultural habitats in maintaining diverse and relatively stable pollinator communities. Community turnover along the land-use gradient reflects strong environmental filtering These results demonstrate that pollinator community structure and diversity in the Qinling Mountains are jointly shaped by landscape structure, climatic conditions, and floral-resource availability, with both main effects and pairwise interactions determining the magnitude and direction of community responses. Environmental differentiation among sites was primarily associated with agricultural intensity, temperature extremes, elevation, and flowering plant richness. Agricultural habitats were characterized by higher cropland cover, warmer and drier conditions, and stronger disturbance, whereas natural habitats were associated with higher elevation, greater humidity, and richer floral resources; semi-natural habitats were intermediate. The clear separation of habitats in ordination space indicates that environmental filtering is a key mechanism underlying community differentiation. Considering main effects, forest cover was consistently associated with higher species richness and diversity, whereas cropland proportion showed broadly negative relationships across taxa. This pattern reflects the stabilizing role of forested landscapes in providing continuous resources and buffered microclimates, in contrast to the homogenizing effects of agricultural expansion (Ulyshen et al., 2023 ). Temperature emerged as one of the strongest drivers, although responses varied among taxa. In most cases, warming was associated with reduced diversity or abundance in the overall assemblage, Lepidoptera, and part of Coleoptera, whereas Hymenoptera showed positive responses under some thermal conditions. These differences are consistent with variation among taxa in thermal tolerance and activity constraints (Johnson et al., 2023 ; Karbassioon et al., 2023 ; Miyashita et al., 2023 ). Wind speed showed predominantly negative effects, indicating constraints on pollinator activity (Hennessy et al., 2020 ; Hennessy et al., 2021 ), whereas precipitation and humidity were generally associated with positive responses, likely reflecting enhanced vegetation and floral-resource supply. Flowering plant richness showed taxon-dependent effects, promoting Lepidoptera but showing neutral or negative relationships for some Diptera and Hemiptera metrics, consistent with differences in resource-use strategies (Hyjazie & Sargent, 2022 ; Davis et al., 2023 ). Environmental effects were strongly context dependent. Habitat type significantly modified climatic responses, with the same climatic factor often producing different or opposite patterns among habitats. In particular, temperature increases were sometimes associated with positive responses in natural habitats but negative responses in agricultural and semi-natural habitats, whereas precipitation and humidity showed the reverse tendency. These patterns indicate that habitat structure determines the extent to which communities buffer climatic variability (Ganuza et al., 2022 ). Climate also interacted strongly with landscape structure. Forest cover generally mitigated the negative effects of high temperature and wind, whereas cropland proportion amplified them, highlighting the dual role of landscape composition in directly shaping habitat conditions and indirectly modifying climatic stress (Ulyshen et al., 2023 ; Ganuza et al., 2022 ). Similar context dependence was observed for precipitation and humidity, with forest-dominated landscapes more often maintaining diversity, while cropland-dominated landscapes favored disturbance-tolerant assemblages. Floral resources further modulated climatic responses. Under high-temperature conditions, increased flowering plant richness enhanced positive responses in Diptera but weakened or reversed responses in Hymenoptera. Under high wind conditions, negative effects on diversity were stronger when floral resources were limited and weaker when resources were abundant, indicating a buffering role of floral availability. This pattern is consistent with evidence that wind reduces pollinator activity, whereas resource-rich environments can partially offset these constraints (Hennessy et al., 2020 ; Hennessy et al., 2021 ; Balfour & Ratnieks, 2025 ). Interactions involving precipitation, humidity, and floral resources showed strong context dependence, and similar patterns were observed along elevational gradients, consistent with shifts in pollinator assemblages along temperature–elevation axes (McCabe & Cobb, 2021 ). Overall, pollinator community responses were governed by multiple interacting drivers rather than single environmental factors. Temperature acted as a key regulator, but its effects depended on habitat context, landscape structure, and resource availability. Forest cover generally buffered environmental stress, whereas agricultural expansion amplified it, and floral resources further mediated taxon-specific responses. These findings highlight the need for integrated management strategies that combine forest conservation, limitation of agricultural expansion, and maintenance of floral and non-floral resources to sustain pollinator diversity (Ulyshen et al., 2023 ; Davis et al., 2023 ). Conclusions and recommendations This study demonstrates that pollinator insect communities in the Qinling Mountains vary markedly along the land-use gradient in terms of α-diversity, β-diversity, community composition, and environmental responses. Compared with agricultural habitats, natural and semi-natural habitats play a more important role in maintaining pollinator diversity and community stability. Natural habitats are particularly irreplaceable because they support higher diversity, contain more endemic and indicator species, and provide stronger buffering against environmental fluctuations. Semi-natural habitats, although generally intermediate in community metrics, also contribute substantially by preserving landscape heterogeneity and offering transitional habitat conditions that help sustain relatively diverse assemblages. By contrast, agricultural habitats may support high pollinator abundance, but this should not be interpreted as high conservation value, because such abundance is likely to be driven by a limited number of widespread, disturbance-tolerant taxa, together with reduced evenness and increased biotic homogenization. These findings have direct implications for pollinator conservation in the Qinling Mountains and similar mountain agroecosystems. Conservation efforts should not focus solely on maintaining pollinator abundance, but should place greater emphasis on preserving community composition, diversity, and ecological stability. In particular, retaining natural and semi-natural habitats, maintaining or increasing forest cover, and ensuring abundant and continuous floral resources should be prioritized as key strategies to buffer environmental stress and mitigate the negative effects of agricultural intensification. More broadly, limiting excessive cropland expansion, enhancing landscape heterogeneity, and conserving both floral and non-floral habitat resources are likely to be essential for sustaining pollinator biodiversity and pollination services under ongoing land-use and climate change. Several limitations should also be acknowledged. First, although this study identified broad patterns along the land-use gradient, the underlying mechanisms remain largely inferential because the analyses are based primarily on correlations. Second, taxonomic composition and conventional diversity indices alone cannot fully capture the functional consequences of pollinator community change. Third, additional attention should be given to multi-scale landscape processes, including habitat configuration, connectivity, and fragmentation, which may further influence community assembly. Accordingly, future research should combine long-term monitoring across seasons and years with stronger causal approaches, such as structural equation modelling, hierarchical models, or manipulative experiments, and should incorporate functional traits, flower-visitation networks, pollination efficiency, and plant reproductive success. Integrating community patterns, environmental drivers, and ecosystem functioning will provide a stronger scientific basis for pollinator conservation and habitat management in the Qinling region and other mountain agricultural landscapes. Declarations Funding This research was funded by the National Key R&D Program of China (2022YFE0115200), Horizon 2020 – the Framework Programme for Research and Innovation (Safeguarding European Wild Pollinators, SAFEGUARD, SC5-32-2020); the Biodiversity Survey and the Assessment Project of the Ministry of Ecology and Environment, China (2019HJ2096001006) and the National Animal Collection Resource Center, China. Competing interests The authors declare no competing interests. Author contributions YS conceived the study, conducted field sampling, curated the data, performed the analyses, prepared the figures, and wrote the first draft of the manuscript. QD contributed to field investigation, analytical methodology, and data curation. QX assisted with the work and revised the manuscript. YZ designed the project, supervised the study, acquired funding, and critically revised the manuscript. All authors read and approved the final manuscript. Supplementary material Supplementary material is available online as three separate files: Supplementary Methods, Supplementary Tables (Tables S1–S7), and Supplementary Figures (Figs. S1–S10). Data availability The datasets supporting the findings of this study are available from the corresponding author upon reasonable request. Code availability The custom R scripts used for data processing and analysis are available from the corresponding author upon reasonable request. Ethics approval Not applicable. This study involved field sampling of insects and did not include experiments on vertebrate animals or human participants. Field sampling was conducted in accordance with local regulations. Consent to participate Not applicable. This study did not involve human participants. Consent for publication All authors approved publication. References Adams BJ, Li E, Bahlai CA, Meineke EK, McGlynn TP, Brown BV (2020) Local- and landscape-scale variables shape insect diversity in an urban biodiversity hot spot. Ecological Applications 30:e02089. Balfour NJ, Ratnieks FLW (2025) Wind alters plant-pollinator community structure, bee foraging rate and movements between plants. Behavioral Ecology 36: araf067. Buchori D, Rizali A, Larasati A, Hidayat P, Ngo H, Gemmill-Herren B (2019) Natural habitat fragments obscured the distance effect on maintaining the diversity of insect pollinators and crop productivity in tropical agricultural landscapes. Heliyon 5: e01373. Chowdhury S, Jennions MD, Zalucki MP, Maron M, Watson JEM, Fuller RA (2023) Protected areas and the future of insect conservation. Trends in Ecology & Evolution 38:85–95. https://doi.org/10.1016/j.tree.2022.09.004 Crist TO, Peters VE (2014) Landscape and local controls of insect biodiversity in conservation grasslands: implications for the conservation of ecosystem service providers in agricultural environments. Land 3:693–718. https://doi.org/10.3390/land3030693 Davis AE, Bickel DJ, Saunders ME, Rader R (2023) Crop-pollinating Diptera have diverse diets and habitat needs in both larval and adult stages. Ecological Applications 33: e2859. Dicks LV, Breeze TD, Ngo HT, Senapathi D, An J, Aizen MA, et al. (2021) A global-scale expert assessment of drivers and risks associated with pollinator decline. Nature Ecology & Evolution 5:1453–1461. Fahrig L (2003) Effects of habitat fragmentation on biodiversity. Annual Review of Ecology, Evolution, and Systematics 34:487–515. Fan H, Liu T, Chen Y, Liao Z, Chen J, Hu Y, et al. (2024) Geographical patterns and determinants of insect biodiversity in China. Science China Life Sciences 67:1255–1265. Fang Z, Xu WH, Zhang JJ, Xiao Y, Zhang L (2017) Planning of the natural reserve system in the Qinling Mountains based on biodiversity and ecosystem services. Acta Ecologica Sinica 37:5334–5341 [in Chinese]. Fenoglio MS, Calviño A, González E, Salvo A, Videla M (2021) Urbanisation drivers and underlying mechanisms of terrestrial insect diversity loss in cities. Ecological Entomology 46:757–771. https://doi.org/10.1111/een.13041 Freimuth J, Bossdorf O, Scheepens JF, Willems FM (2022) Climate warming changes synchrony of plants and pollinators. Proceedings of the Royal Society B 289:20212142. Galloway AD, Seymour CL, Gaigher R, Pryke JS (2021) Organic farming promotes arthropod predators, but this depends on neighbouring patches of natural vegetation. Agriculture, Ecosystems & Environment 310:107295. https://doi.org/10.1016/j.agee.2020.107295 Ganuza C, Redlich S, Uhler J, et al. (2022) Interactive effects of climate and land use on pollinator diversity differ among taxa and scales. Science Advances 8: eabm9359. Garratt MPD, De Groot GA, Albrecht M, Bosch J, Breeze TD, Fountain MT, et al. (2021) Opportunities to reduce pollination deficits and address production shortfalls in an important insect-pollinated crop. Ecological Applications 31: e02445. Ghisbain G, Gérard M, Wood TJ, Hines HM, Michez D (2021) Expanding insect pollinators in the Anthropocene. Biological Reviews 96:2755–2770. Hahn PG, Orrock JL (2015) Land-use history alters contemporary insect herbivore community composition and decouples plant–herbivore relationships. Journal of Animal Ecology 84:745–754. Halsch CA, Shapiro AM, Fordyce JA, Nice CC, Thorne JH, Waetjen DP, et al. (2020) Insects and recent climate change. bioRxiv. https://doi.org/10.1101/2020.03.09.984328 Harvey JA, Heinen R, Gols R, Thakur MP (2020) Climate change-mediated temperature extremes and insects: from outbreaks to breakdowns. Global Change Biology 26:6685–6701. Harvey JA, Tougeron K, Gols R, et al. (2023) Scientists’ warning on climate change and insects. Ecological Monographs 93: e1553. Hennessy G, Harris C, Eaton C, Wright P, Jackson E, Goulson D, et al. (2020) Gone with the wind: effects of wind on honey bee visit rate and foraging behaviour. Animal Behaviour 161:23–31. Hennessy G, Harris C, Pirot LMP, Wright P, Jackson E, Goulson D, et al. (2021) Wind slows play: increasing wind speed reduces flower visiting rate in honey bees. Animal Behaviour 178:87–93. Hyjazie BF, Sargent RD (2022) Floral resources predict the local bee community: implications for conservation. Biological Conservation 273:109679. Jaccard P (1908) Nouvelles recherches sur la distribution florale. Bulletin de la Société Vaudoise des Sciences Naturelles 44:223–270 Johnson MG, Glass JR, Dillon ME, Harrison JF (2023) How will climatic warming affect insect pollinators? Advances in Insect Physiology 64:1–115. Karbassioon A, Yearsley J, Dirilgen T, Hayes B, de Ibarra NH, Nunes-Silva P (2023) Responses in honeybee and bumblebee activity to changes in weather conditions. Oecologia 201:689–701. Kellermann V, van Heerwaarden B (2019) Terrestrial insects and climate change: adaptive responses in key traits. Physiological Entomology 44:99–115. https://doi.org/10.1111/phen.12282 Kovács-Hostyánszki A, Espíndola A, Vanbergen AJ, Settele J, Kremen C, Dicks LV (2017) Ecological intensification to mitigate impacts of conventional intensive land use on pollinators and pollination. Ecology Letters 20:673–689. Li JY, Fu BJ, Sun JL, Hong ZL, Zhang BP, Wang XF, et al. (2021) Ecological civilization construction in the Qinling Mountains in the new era: existing problems and development pathways. Journal of Natural Resources 36:2449–2463 [in Chinese]. Liu J, Xie T, Lyu D, Cui L, Liu Q (2024) Analyzing the spatiotemporal dynamics and driving forces of ecological environment quality in the Qinling Mountains, China. Sustainability 16:3251. Liu XW, Wu CS, Zhou QS, Zhu CD (2018) Effects of environmental change on wild bee diversity in China. Biodiversity Science 26:760 [in Chinese]. Liu YL, Qiu XK (2024) Ecological environmental problems and countermeasures in the Qinling region. Modern Horticulture 47:173–175 [in Chinese]. Luo F, Lei CL (2003) Research progress on pollinating beetles. Chinese Bulletin of Entomology 40:313–317 [in Chinese]. McCabe LM, Cobb NS (2021) From bees to flies: global shift in pollinator communities along elevation gradients. Frontiers in Ecology and Evolution 8:626124. McCain CM, Garfinkel CF (2021) Climate change and elevational range shifts in insects. Current Opinion in Insect Science 47:111–118. https://doi.org/10.1016/j.cois.2021.06.003 Meloni F, Civieta BF, Zaragoza JA, Moraza ML, Bautista S (2020) Vegetation pattern modulates ground arthropod diversity in semi-arid Mediterranean steppes. Insects 11:59. https://doi.org/10.3390/insects11010059 Miao B, Peng Y, Yang D, Kubota Y, Economo EP, Liu C (2021) Climate and land-use interactively shape butterfly diversity in tropical rainforest and savanna ecosystems of southwestern China. Insect Science 28:1109–1120 Miñarro Prado M, García García D, Martínez Sastre R (2018) Impact of insect pollinators in agriculture: importance and management of their biodiversity. Ecosistemas 27:81–90. https://doi.org/10.7818/ecos.1394 Miyashita T, Hayashi S, Natsume K, Izumi Y, Sueyoshi M, Deto H, et al. (2023) Diverse flower-visiting responses among pollinators to multiple weather variables in buckwheat pollination. Scientific Reports 13:3099. Muinde J, Katumo DM (2024) Beyond bees and butterflies: the role of beetles in pollination systems. Journal for Nature Conservation 77:126523. Müller J, Hothorn T, Yuan Y, Seibold S, Mitesser O, Rothacher J, et al. (2024) Weather explains the decline and rise of insect biomass over 34 years. Nature 628:349–354. Nealis VG (2020) Weather and insects in a changing climate. Journal of the Entomological Society of British Columbia. Orford KA, Vaughan IP, Memmott J (2015) The forgotten flies: the importance of non-syrphid Diptera as pollinators. Proceedings of the Royal Society B 282:20142934. Outhwaite CL, McCann P, Newbold T (2022) Agriculture and climate change are reshaping insect biodiversity worldwide. Nature 605:97–102. Ratto F, Steward P, Sait SM, Pryke JS, Gaigher R, Samways MJ, et al. (2021) Proximity to natural habitat and flower plantings increases insect populations and pollination services in South African apple orchards. Journal of Applied Ecology 58:2540–2551. Riva F, Martin CJ, Galán-Acedo C, Bellon EN, Keil P, Morán-Ordóñez A, et al. (2024) Incorporating effects of habitat patches into species distribution models. Journal of Ecology 112:2162–2182. Ryalls JM, Langford B, Mullinger NJ, Bromfield LM, Nemitz E, Pfrang C, et al. (2022) Anthropogenic air pollutants reduce insect-mediated pollination services. Environmental Pollution 297:118847. Scaven VL, Rafferty NE (2013) Physiological effects of climate warming on flowering plants and insect pollinators and potential consequences for their interactions. Current Zoology 59:418–426. https://doi.org/10.1093/czoolo/59.3.418 Skendžić S, Zovko M, Živković IP, Lešić V, Lemić D (2021) The impact of climate change on agricultural insect pests. Insects 12:440. https://doi.org/10.3390/insects12050440 Sridhar J, Kumar KK, Murali-Baskaran RK, Senthil-Nathan S, Sharma S, Nagesh M, et al. (2020) Impact of climate change on communities, response and migration of insects, nematodes, vectors and natural enemies in diverse ecosystems. In: Global climate change: resilient and smart agriculture. (pp. 69-93). Singapore: Springer Singapore. https://doi.org/10.1007/978-981-32-9856-9_4 Ulyshen MD, Urban-Mead KR, Dorey JB, et al. (2023) Forests are critically important to global pollinator diversity and enhance pollination in adjacent crops. Biological Reviews 98:1118–1141. Yang D (2018) Research progress on pollinating insects. Yunnan Agricultural Science and Technology A01:3–7 [in Chinese]. Yang J, Huang X (2025) The 30 m annual land cover datasets and its dynamics in China from 1985 to 2024 [Data set]. Earth System Science Data 13:3907–3925. Zenodo. https://doi.org/10.5281/zenodo.15853565 Yang ZY, Zhang H, Hao XJ, Li N (2024) Spatial pattern changes of ecological environment in the Qinling Mountains over the past 20 years based on remote sensing. Geomatics & Spatial Information Technology 47:14–17 [in Chinese]. Zhang L, Hu YY, Zhao YB, Che T (2025) ChinaMet: a high-resolution multi-factor meteorological forcing dataset for China based on multi-source data fusion. National Cryosphere Desert Data Center. https://cstr.cn/CSTR:11738.11.NCDC.NIEER.DB6722.2025 [in Chinese] Zhang LL, Cheng JW, Li AG, Yang YJ (2023) Effects of planting nectar plants on the diversity of natural enemy insects in paddy fields. Acta Agriculturae Zhejiangensis 35:1360–1367 [in Chinese]. Zhou LY, Ding SY, Lu XL, Liu YM (2020) Effects of human disturbance on pollinator diversity and niche of dominant groups. Acta Ecologica Sinica 40:2111–2121 [in Chinese]. Table Table 2 is available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformation.pdf Table2.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9444014","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":626366211,"identity":"0ceb0b1e-d475-4a03-9880-0ef883b93877","order_by":0,"name":"Yaoyao Si","email":"","orcid":"","institution":"Northwest A\u0026F University","correspondingAuthor":false,"prefix":"","firstName":"Yaoyao","middleName":"","lastName":"Si","suffix":""},{"id":626366212,"identity":"0cd7334f-3a0d-4ef1-ac8c-7faa612dd8b6","order_by":1,"name":"Qiqi Deng","email":"","orcid":"","institution":"Northwest A\u0026F University","correspondingAuthor":false,"prefix":"","firstName":"Qiqi","middleName":"","lastName":"Deng","suffix":""},{"id":626366213,"identity":"fc645764-d6d9-44a6-980d-848747fc4aab","order_by":2,"name":"Qingquan Xue","email":"","orcid":"","institution":"Northwest A\u0026F University","correspondingAuthor":false,"prefix":"","firstName":"Qingquan","middleName":"","lastName":"Xue","suffix":""},{"id":626366215,"identity":"665b7ada-be75-4edd-a6ac-01c6c44591e3","order_by":3,"name":"Yalin Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIie2OPQvCMBCGI4VOwbkOfvyESMEu4m9pKdhFQTfHiLtzBX9EnMTt5IYuha5CBRWhk4PiKmji5pJ2dMgDd9xwD+9LiMHwl1gcCIGWOu9yarxcqX0VV11xZUUuCHhlhSX7OU5ex2jrhJcFJf2mAKs4a5U04BjTYryLh12pDF0Btse0iqyE1MGxyH3vSgkGAqjtaJXsIhWGEcujh0x5V1AOKsVHn+UjVQzKlcZBpQB2xek2na9Z6K7Q7mmVehbik76wzdJkw2+zQXOZLAqt0oHfnnIs3b+kzUseDAaDwUA+vdVWSAdKxOIAAAAASUVORK5CYII=","orcid":"","institution":"Northwest A\u0026F University","correspondingAuthor":true,"prefix":"","firstName":"Yalin","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2026-04-17 04:39:57","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9444014/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9444014/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107926083,"identity":"287817c6-720b-4ed7-a02b-1885d66c29e0","added_by":"auto","created_at":"2026-04-27 15:42:09","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4832096,"visible":true,"origin":"","legend":"\u003cp\u003eNMDS ordination of pollinator community composition across habitat types. (a) NMDS based on sampling-event data. (b) NMDS based on site-level aggregated data. Points represent samples and are colored by habitat type (agricultural, semi-natural, and natural). Polygons indicate the convex hull of each habitat group. Boxplots show the distribution of NMDS axis scores. Stress values are provided to indicate ordination fit. Differences in community composition among habitats were tested using PERMANOVA (p-values and R² shown in panels).\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/f050e2a57d1a9eb8b47a5b81.png"},{"id":107926059,"identity":"5e6b0dc4-54cb-43cc-ba5f-2354621d60c3","added_by":"auto","created_at":"2026-04-27 15:41:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":5079876,"visible":true,"origin":"","legend":"\u003cp\u003eCanonical correspondence analysis (CCA) of species composition across habitats in relation to environmental variables.Points represent sampling sites colored by habitat type (agricultural, semi-natural, and natural). Ellipses indicate 95% confidence intervals of each habitat group. Arrows represent environmental variables, with direction indicating gradients and length representing the strength of their influence on community composition.\u003c/p\u003e","description":"","filename":"image9.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/2709c7b0d491b01e0216f6cc.png"},{"id":107926151,"identity":"cd80ce85-1785-472c-abd3-1d2fc175f3a5","added_by":"auto","created_at":"2026-04-27 15:42:16","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6103072,"visible":true,"origin":"","legend":"\u003cp\u003eEffects of environmental factors on the diversity of the overall pollinating insect community. Panels show relationships between environmental variables (MATmax, MATmin, MAT, forest proportion, cropland proportion, elevation, wind speed, precipitation, relative humidity, and flower richness) and diversity metrics (species richness, Shannon diversity, and abundance). Points represent observations, lines indicate model predictions, and shaded areas represent 95% confidence intervals. Black insect icons indicate the focal pollinator taxonomic group represented in each panel.\u003c/p\u003e","description":"","filename":"image10.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/421d3a0468486cd62c31b8bf.png"},{"id":107926081,"identity":"07612ff1-e498-4b7c-b86d-80d48fb2ffbb","added_by":"auto","created_at":"2026-04-27 15:42:09","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":8784031,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential effects of climatic factors on pollinator communities across habitat types. Panels show relationships between environmental variables and diversity metrics (species richness, Shannon diversity, and abundance) under different habitat types (agricultural, semi-natural, and natural). Lines represent model predictions for each habitat, and shaded areas indicate 95% confidence intervals. Black insect icons indicate the focal pollinator taxonomic group represented in each panel.\u003c/p\u003e","description":"","filename":"image11.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/248f56feb6842f248091f225.png"},{"id":107926088,"identity":"f7189ac9-6058-4b68-9ef8-6b0abed8dbb9","added_by":"auto","created_at":"2026-04-27 15:42:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":8092509,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential effects of climatic factors under varying forest and cropland cover. Panels show relationships between environmental variables and diversity metrics (species richness, Shannon diversity, and abundance) under different habitat types (agricultural, semi-natural, and natural). Lines represent model predictions for each habitat, and shaded areas indicate 95% confidence intervals. Black insect icons indicate the focal pollinator taxonomic group represented in each panel.\u003c/p\u003e","description":"","filename":"image12.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/850dd73106286eaf4b85efc6.png"},{"id":107926150,"identity":"5e37cd01-dd0b-4bd5-aeab-78978fb25636","added_by":"auto","created_at":"2026-04-27 15:42:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1759103,"visible":true,"origin":"","legend":"\u003cp\u003eInteractive effects of climatic factors and forest/cropland cover. Panels show relationships between landscape variables (forest proportion, For_prop; cropland proportion, Cro_prop) and diversity metrics (species richness, Shannon diversity, and abundance) under contrasting climatic conditions, including maximum temperature (MATmax), minimum temperature (MATmin), mean temperature (MAT), wind speed (WS10), precipitation (MAP), and relative humidity (RH). Lines represent model predictions, shaded areas indicate 95% confidence intervals, and points represent observations. Colors indicate groups defined by threshold values of climatic variables. Black insect icons indicate the focal pollinator taxonomic group represented in each panel.\u003c/p\u003e","description":"","filename":"image13.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/bba507d5048a8413b3886ac8.png"},{"id":107926060,"identity":"d12093b6-4129-45a5-babf-9ce82f432bec","added_by":"auto","created_at":"2026-04-27 15:41:58","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1441690,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential effects of environmental factors under varying flowering plant species richness. Panels show relationships between climatic variables (MATmax, MAT, MATmin, WS10, MAP, and RH) and diversity metrics (species richness, Shannon diversity, and abundance) under contrasting flowering plant richness conditions. Lines represent model predictions, shaded areas indicate 95% confidence intervals, and points represent observations. Colors indicate groups defined by threshold values of flowering plant richness. Black insect icons indicate the focal pollinator taxonomic group represented in each panel.\u003c/p\u003e","description":"","filename":"image14.png","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/6d3fc784e86590ce8a4d5eb7.png"},{"id":107926319,"identity":"50a0063d-a191-461d-9845-1a878e422859","added_by":"auto","created_at":"2026-04-27 15:43:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":37095986,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/b3783de2-b206-4161-944d-dd7dae969b40.pdf"},{"id":107926032,"identity":"dc3a01b8-84bb-4711-9e45-6facbfe6c4e5","added_by":"auto","created_at":"2026-04-27 15:41:55","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":10140981,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/ecc23ccbbd93fdf00c68a89e.pdf"},{"id":107926152,"identity":"08135b90-9a92-4bb0-80ad-72f7194b4bd8","added_by":"auto","created_at":"2026-04-27 15:42:16","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":22763,"visible":true,"origin":"","legend":"","description":"","filename":"Table2.docx","url":"https://assets-eu.researchsquare.com/files/rs-9444014/v1/aabc5b3a9907ef90505a86bc.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Diversity Patterns of Pollinating Insect Communities in the Central Qinling Mountains and Their Responses to Environmental Change","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePollinators are essential components of terrestrial ecosystems because they support plant reproduction, biodiversity maintenance, and ecosystem stability (Liu et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Although bees are often emphasized in pollination research, pollinator assemblages are taxonomically and functionally diverse, including Coleoptera, Diptera, Hymenoptera, Lepidoptera, and some Hemiptera (Luo \u0026amp; Lei, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Yang, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wu \u0026amp; Zheng, 2019; Orford et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Muinde \u0026amp; Katumo, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These groups differ substantially in life-history traits, dispersal ability, and resource use, and therefore may respond differently to environmental change. Community-level approaches are thus necessary to understand the mechanisms shaping pollinator diversity and composition.\u003c/p\u003e \u003cp\u003eGlobal declines in pollinator diversity and abundance have raised major concerns about ecosystem functioning and food security. Long-term monitoring has revealed marked reductions in insect biomass and diversity in parts of Europe (M\u0026uuml;ller et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and similar patterns have been increasingly reported worldwide (Garratt et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Dicks et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Ratto et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) identified multiple interacting drivers of pollinator decline, including land-use change, agricultural intensification, pesticide use, invasive species, and climate change (Dicks et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These drivers rarely act in isolation, and their combined effects may strongly influence pollinator community structure.\u003c/p\u003e \u003cp\u003eClimate change is one of the most important drivers affecting pollinators. Temperature, precipitation, humidity, and wind can influence insect physiology, dispersal, survival, and phenology, while also altering floral resource availability and plant\u0026ndash;pollinator synchrony (Scaven \u0026amp; Rafferty, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Freimuth et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Skendžić et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Nealis, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Warming has been associated with shifts in insect distributions toward higher elevations and latitudes (McCain \u0026amp; Garfinkel, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Halsch et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), whereas climatic extremes may disrupt ecological interactions and alter community composition (Harvey et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Miao et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). However, responses are often taxon-specific: some groups may benefit from warmer conditions, whereas others may decline under physiological stress or changing resource regimes (Fan et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Adams et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Ghisbain et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In addition, climatic variability and extreme events may exert stronger effects than mean climatic conditions alone (Kellermann \u0026amp; van Heerwaarden, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Sridhar et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eLandscape structure and land-use change are also key determinants of pollinator diversity. Agricultural intensification and habitat fragmentation reduce habitat availability, simplify landscapes, and limit nesting and foraging resources, thereby decreasing pollinator diversity and promoting community homogenization (Buchori et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Kov\u0026aacute;cs-Hosty\u0026aacute;nszki et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Miao et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Fahrig, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Urbanization may further intensify these pressures by increasing habitat isolation and environmental stress (Fenoglio et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Ryalls et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In contrast, semi-natural habitats and heterogeneous landscapes often support higher pollinator diversity and more stable communities by providing complementary habitats and continuous resource supply (Crist \u0026amp; Peters, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Zhou et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Chowdhury et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Landscape configuration, including patch size and connectivity, can further influence species dispersal and community assembly (Meloni et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Riva et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and historical land use may leave persistent legacy effects on present-day insect communities (Hahn \u0026amp; Orrock, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFloral resources represent another key component of pollinator community dynamics. The abundance and diversity of flowering plants determine nectar and pollen availability, thereby shaping pollinator richness, abundance, and foraging behavior (Mi\u0026ntilde;arro Prado et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Galloway et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, floral-resource effects are often context dependent. Under resource-limited conditions, increased flower diversity may enhance pollinator richness, whereas in resource-rich environments additional floral resources may lead to saturation effects or intensified competition (Outhwaite et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Consequently, interactions among climate, landscape structure, and floral resources are likely to be crucial for understanding pollinator community responses.\u003c/p\u003e \u003cp\u003eMountain ecosystems provide ideal natural laboratories for investigating these interactions because they contain strong environmental gradients and high habitat heterogeneity. The Qinling Mountains form a major biogeographical boundary and biodiversity hotspot in China (Fang et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This region is characterized by complex topography, diverse habitat types, and pronounced climatic gradients, but it has also experienced increasing anthropogenic pressures, including land-use change, forest degradation, and habitat fragmentation (Liu \u0026amp; Qiu, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Spatial heterogeneity in habitat quality remains substantial, providing an ideal setting for examining how multiple environmental gradients shape pollinator communities.\u003c/p\u003e \u003cp\u003eDespite growing attention to pollinator ecology, studies that simultaneously integrate climatic factors, landscape structure, habitat differentiation, and floral resources at the community level remain limited, particularly in montane ecosystems such as the central Qinling Mountains. Moreover, most previous studies have focused on single drivers or specific taxa, whereas the combined and interactive effects of multiple environmental factors on pollinator diversity and community composition remain poorly understood.\u003c/p\u003e \u003cp\u003eTo address these gaps, we investigated pollinator communities across multiple habitat types in the central Qinling Mountains. Specifically, we aimed to: (1) compare pollinator alpha diversity and community composition among habitat types; (2) quantify the relative contributions of climatic, topographic, landscape, and floral-resource variables; and (3) evaluate how interactions among these factors mediate pollinator diversity, abundance, and community structure. By linking community patterns with multiple environmental drivers, this study provides a community-level perspective on pollinator responses to environmental change in a montane biodiversity hotspot.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003ch2\u003eStudy area and sampling design\u003c/h2\u003e\n\u003cp\u003eThe study area was selected within the Qinling Mountains based on spatial analyses conducted in ArcGIS. Multi-year mean annual temperature (MAT) data (Zhang et al., 2025) and land-cover data (Yang \u0026amp; Huang, 2025) were used to characterize environmental gradients. The study region was classified into five temperature-gradient categories according to MAT values (Table 2-1). In addition, thresholds for land-cover types were defined based on regional characteristics of the Qinling Mountains. According to the proportions of forest cover (For_prop) and cropland cover (Cro_prop), the sampling sites were classified into three habitat types: agricultural habitat (Cro_prop \u0026gt; 50%), natural habitat (For_prop \u0026gt; 75%), and semi-natural habitat (Cro_prop \u0026gt; 25% and For_prop \u0026gt; 50%) (Table S1).\u003c/p\u003e\n\u003cp\u003eApproximately 40 sampling sites were selected in Foping County, Yangxian County, Chenggu County, and Ningshan County, Ankang City (Fig. S1). Field surveys were conducted from March to September in 2024-2025 across multiple seasons to capture temporal variation in pollinator communities and to ensure data representativeness and reliability.\u003c/p\u003e\n\u003ch2\u003eEnvironmental factors\u003c/h2\u003e\n\u003ch2\u003eLand-cover data\u003c/h2\u003e\n\u003cp\u003eLand-cover data were obtained from the annual 30-m resolution land-cover dataset for China (Yang \u0026amp; Huang, 2025). Using ArcGIS, a circular buffer with a radius of 150 m was established around each sampling transect. The proportion of each land-cover type within the buffer was calculated to quantify landscape composition, including forest cover (For_prop), cropland cover (Cro_prop), and impervious surface cover (Imp_prop). In addition, elevation (Ele) and flowering plant species richness (Flower_richness) were recorded during field surveys.\u003c/p\u003e\n\u003ch2\u003eClimatic data\u003c/h2\u003e\n\u003cp\u003eClimatic variables were obtained from the National Cryosphere Desert Data Center (Zhang et al., 2025). The extracted variables included minimum temperature at 2 m (MATmin), mean temperature at 2 m (MAT), maximum temperature at 2 m (MATmax), relative humidity (RH), mean annual precipitation (MAP), and wind speed at 10 m height (WS10) for each sampling site.\u003c/p\u003e\n\u003cp\u003eAll environmental variables were standardized prior to statistical analyses.\u003c/p\u003e\n\u003ch2\u003ePollinator sampling and identification\u003c/h2\u003e\n\u003cp\u003ePollinating insects were surveyed using a combination of sweep-net sampling and transect-based visual observations. Non-butterfly taxa were collected along approximately 2-km transects at each site using insect nets in the surrounding herbaceous and shrub vegetation. Ecological photographs were taken during sampling to assist subsequent identification.\u003c/p\u003e\n\u003cp\u003eButterflies were surveyed using the Pollard walk method at a walking speed of 1.5\u0026ndash;2 km/h. All individuals observed within 2.5 m on each side of the transect, within 5 m above ground, and along a total transect length of 2 km were recorded. Individuals that could not be reliably identified in the field were collected for laboratory identification.\u003c/p\u003e\n\u003cp\u003eSmall insect specimens were preserved in 75% ethanol, whereas butterfly specimens were stored in triangular paper envelopes. In the laboratory, specimens were frozen at \u0026minus;80 \u0026deg;C for 1\u0026ndash;2 days, then pinned, labeled, and preserved in specimen boxes. Specimens were identified morphologically using taxonomic literature and reference collections. DNA barcoding was used when necessary, and COI sequences were compared against the NCBI database using BLAST; sequence similarity \u0026ge; 98% was considered supportive of species-level identification.\u003c/p\u003e\n\u003ch2\u003eDiversity and community analyses\u003c/h2\u003e\n\u003ch2\u003eData processing and diversity analyses\u003c/h2\u003e\n\u003cp\u003eGeographic and environmental data were extracted using ArcGIS, and insect and environmental datasets were organized in Excel. All statistical analyses were conducted in R 4.4.2. Species abundance and environmental matrices were matched by sampling site, and variables were converted to numeric format prior to analysis.\u003c/p\u003e\n\u003ch2\u003eAlpha diversity\u003c/h2\u003e\n\u003cp\u003eSampling completeness and expected species richness were evaluated using rarefaction\u0026ndash;extrapolation analyses implemented in vegan and iNEXT. Alpha diversity was quantified using species richness, Shannon diversity, Simpson dominance, and total abundance. Differences among habitat types were tested using one-way ANOVA followed by Tukey\u0026rsquo;s post hoc comparisons. Species abundance distribution was described using Preston\u0026rsquo;s octave analysis. Sampling completeness was evaluated using sample-size-based rarefaction and extrapolation analyses implemented in vegan and iNEXT, with 95% confidence intervals estimated from 500 bootstrap replicates. Detailed descriptions of diversity indices and statistical analyses are provided in the Supplementary Information (Section S1.1).\u003c/p\u003e\n\u003ch2\u003eBeta diversity and community structure\u003c/h2\u003e\n\u003cp\u003eCommunity composition was analyzed using Bray\u0026ndash;Curtis dissimilarity and visualized by non-metric multidimensional scaling (NMDS). Jaccard similarity was used to quantify species sharing among habitats (see Supplementary Methods, Section S1.2). Dominant species were identified using importance values (IV), and indicator species were identified using the IndVal method. Significant indicator species were visualized using Z-score standardized heatmaps with hierarchical clustering.\u003c/p\u003e\n\u003ch2\u003eEnvironmental gradients and driver analyses\u003c/h2\u003e\n\u003ch2\u003eEnvironmental variables and drivers\u003c/h2\u003e\n\u003cp\u003eEnvironmental variables included landscape, climatic, topographic, and floral-resource factors. Land-cover data (Yang and Huang, 2025) were used to calculate forest cover (For_prop), cropland cover (Cro_prop), and impervious surface cover (Imp_prop) within a 150 m buffer around each transect in ArcGIS. Elevation (Ele) and flowering plant species richness (Flower_richness) were recorded during field surveys. Climatic variables, including MATmin, MAT, MATmax, RH, MAP, and WS10, were obtained from the National Cryosphere Desert Data Center (Zhang et al., 2025).\u003c/p\u003e\n\u003cp\u003ePrincipal component analysis (PCA) and Spearman correlation analysis were used to summarize environmental variation and assess collinearity among variables. Relationships between community composition and environmental gradients were examined using RDA or CCA, selected according to DCA results, with model significance assessed by permutation tests.\u003c/p\u003e\n\u003ch2\u003eStatistical modeling\u003c/h2\u003e\n\u003cp\u003eGeneralized linear models (GLMs) and linear models (LMs) were used to evaluate environmental effects on diversity metrics. Species richness was analyzed using Poisson or negative binomial GLMs, whereas Shannon diversity and abundance were analyzed using Gaussian linear models.\u003c/p\u003e\n\u003cp\u003eThree model types were fitted: (1) main-effect models to assess independent environmental effects; (2) habitat interaction models to test environment \u0026times; habitat interactions; and (3) environmental interaction models to test pairwise interactions among environmental variables using binary grouping (low vs. high). Model significance was assessed using parameter estimates and associated p-values, and fitted response curves were used to visualize effect directions and strengths. Model structures and parameter definitions are provided in Supplementary Information (Section S1.3).\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eSampling completeness and alpha diversity patterns of pollinator communities\u003c/h2\u003e\n\u003cp\u003eSample-size-based rarefaction and extrapolation analyses (iNEXT) indicated high sampling completeness across habitats and taxonomic groups, with sample coverage generally exceeding 96% and species accumulation curves approaching asymptotes, suggesting that sampling effort was sufficient to capture pollinator diversity (Fig. S2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 1. Taxonomic composition and diversity indices of insect communities across different habitats\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003eHabitat\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eFamily\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eGenus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eRichness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003eAbundance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eShannon\u003c/p\u003e\n \u003cp\u003eDiversity Index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003eSimpson\u003c/p\u003e\n \u003cp\u003eDominance\u003c/p\u003e\n \u003cp\u003eIndex\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003eAgricultural\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e178\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e4215\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003e3.7823\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.0533\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003eSemi-natural\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e261\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e362\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e4586\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003e4.3587\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.0421\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003eNatural\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e272\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e376\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e3647\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003e4.7181\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.0216\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003eTotal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003e47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003e389\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003e601\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e12448\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003e4.5458\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 15px;\"\u003e\n \u003cp\u003e0.0321\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData in the table\u0026nbsp;1.\u0026nbsp;show the taxonomic composition and diversity indices of insect communities across different habitats. Family and genus refer to the numbers of families and genera, respectively; richness refers to species richness, and abundance refers to the total number of individuals. The Shannon diversity index describes community diversity, whereas the Simpson dominance index reflects the degree of dominance concentration. \u0026ldquo;Total\u0026rdquo; indicates the pooled value across all habitats.\u003c/p\u003e\n\u003cp\u003eTable 2. Taxonomic composition and diversity indices of different insect community groups\u003c/p\u003e\n\u003cp\u003eData in the table 2. show the taxonomic composition and diversity indices of different insect community groups. Family and genus refer to the numbers of families and genera, respectively; richness refers to species richness, and abundance refers to the total number of individuals. The Shannon diversity index describes the diversity level of each group, whereas the Simpson dominance index reflects the degree of dominance concentration. \u0026ldquo;Total\u0026rdquo; indicates the pooled value across all groups. Icons represent different taxonomic groups.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn total, 12,448individuals representing5orders, 47families, 389genera, and 601 species were recorded, with natural habitats harboring the highest species richness, followed by semi-natural and agricultural habitats (Table 1-2). Alpha diversity patterns showed that species richness was highest in Coleoptera and lowest in Diptera, whereas Shannon diversity peaked in Lepidoptera; across habitats, both richness and Shannon diversity followed the gradient natural \u0026gt; semi-natural \u0026gt; agricultural, while Simpson dominance showed the opposite trend. Agricultural habitats and Diptera groups were characterized by relatively low diversity but high dominance, whereas natural habitats and Coleoptera groups exhibited higher diversity and more even species distributions. Although abundance was higher in semi-natural and agricultural habitats, no significant differences in abundance were detected among habitats, indicating that habitat effects were mainly reflected in species composition and evenness rather than total individual numbers. Significant habitat differences in Shannon diversity were observed only for Diptera, while most other groups showed similar patterns without significant differences between natural and semi-natural habitats. Across all habitats, Lepidoptera consistently exhibited the highest diversity and represented the dominant pollinator group, whereas Hemiptera showed relatively high abundance in semi-natural habitats.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOverall, these results suggest that habitat differences primarily influence community structure by altering species composition and dominance patterns, rather than total abundance, thereby contributing to increased homogenization under disturbed conditions.\u003c/p\u003e\n\u003ch2\u003eBeta diversity and community differentiation\u003c/h2\u003e\n\u003cp\u003ePollinator community composition differed significantly among habitat types at both sampling-event and site scales. Non-metric multidimensional scaling (NMDS) based on individual sampling events revealed a clear separation among habitats (PERMANOVA, p = 0.001), although the explained variance was relatively low (R\u0026sup2; = 0.06; Fig. 1a), indicating substantial within-group variability, potentially driven by seasonal or temporal dynamics.\u003c/p\u003e\n\u003cp\u003eAt the site scale, where repeated sampling events were aggregated for each site, community differentiation among habitats remained significant (p = 0.001), with a notably higher explanatory power (R\u0026sup2; = 0.13; Fig. 1b).\u003c/p\u003e\n\u003cp\u003eOverall, pollinator communities exhibited partial separation among habitat types. The greatest dissimilarity was observed between natural and agricultural habitats, whereas semi-natural habitats displayed an intermediate position. Despite this separation, considerable overlap among habitats was still evident, suggesting a degree of community similarity.\u003c/p\u003e\n\u003ch2\u003eSpecies composition and habitat specificity\u003c/h2\u003e\n\u003cp\u003ePrincipal component analysis (PCA) (Fig. S5) showed that agricultural intensity, temperature extremes, elevation, and flowering plant richness were the main sources of environmental differentiation among sampling sites. Habitat types were clearly separated in environmental space, with agricultural habitats associated with greater agricultural intensity, higher temperatures, lower humidity, and stronger disturbance, natural habitats associated with higher elevation, more humid conditions, and richer floral resources, and semi-natural habitats occupying an intermediate position.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIndicator species analysis and importance value (IV) results were consistent with this pattern (Fig. S6, Fig. S7), revealing clear habitat-specific structuring of pollinator communities. Agricultural habitats were characterized by disturbance-tolerant and generalist taxa, such as \u003cem\u003ePieris rapae\u003c/em\u003e and \u003cem\u003eApis ceran\u003c/em\u003ea, whereas natural habitats supported distinct dominant and indicator species associated with less disturbed vegetation, including \u003cem\u003eNeptis sappho\u003c/em\u003e and \u003cem\u003eCelastrina argiolus\u003c/em\u003e. Semi-natural habitats exhibited intermediate composition, with both agricultural and natural-associated species (e.g., \u003cem\u003eVanessa cardui\u003c/em\u003e) occurring across habitats. Together, these results support the interpretation that environmental filtering along gradients of land use, resource availability, and disturbance intensity underlies the observed community differentiation.\u003c/p\u003e\n\u003ch2\u003eEffects of environmental variables on diversity\u003c/h2\u003e\n\u003cp\u003eDetrended correspondence analysis (DCA) indicated that canonical correspondence analysis (CCA) was appropriate for ordination of community data. After removing highly collinear variables using variance inflation factor (VIF \u0026lt; 10), the retained environmental variables represented independent contributions to variation in community composition.\u003c/p\u003e\n\u003cp\u003eThe CCA ordination revealed clear separation among habitat types along environmental gradients. Site scores were jointly structured by species composition and their relationships with environmental variables. The length of environmental vectors indicated their relative influence on community structure, while ellipses represented 95% confidence intervals for each habitat type, illustrating the degree of separation among agricultural, semi-natural, and natural communities.\u003c/p\u003e\n\u003cp\u003eNatural habitat sites (blue points) were primarily distributed in the upper-right quadrant and were associated with higher elevation (Ele) and stronger wind speed (WS10). Semi-natural habitat sites (orange points) were located in the central region of the ordination space, showing moderate associations with cropland proportion (Cro_prop) and impervious surface proportion (Imp_prop). In contrast, agricultural habitat sites (yellow points) were mainly clustered in the lower-left quadrant and were strongly associated with higher cropland and impervious surface proportions. The vectors of MATmax and MATmin also pointed toward the lower-left quadrant, indicating that temperature extremes influenced community composition in both agricultural and natural habitats (Fig. 2).\u003c/p\u003e\n\u003cp\u003eTo further evaluate how communities respond to environmental variation, we analyzed the effects of multiple environmental drivers on insect diversity.\u003c/p\u003e\n\u003cp\u003eEnvironmental driver analyses were conducted for three diversity metrics\u0026mdash;species richness (S), Shannon diversity (H\u0026prime;), and abundance (A)\u0026mdash;across six pollinator groups. These analyses included main-effect models, habitat interaction models, and environmental interaction models, and only significant results (p \u0026le; 0.05) were retained.\u003c/p\u003e\n\u003cp\u003eAmong the significant results (n = 574), environmental interaction models accounted for the largest number of significant patterns (n = 412), followed by main-effect models (n = 104) and habitat interaction models (n = 58). Across response variables, species richness showed the highest sensitivity (n = 246), followed by Shannon diversity (n = 167), whereas abundance showed the fewest significant responses (n = 161).\u003c/p\u003e\n\u003ch2\u003eMain effects analysis\u003c/h2\u003e\n\u003cp\u003eBased on the main-effect models, multiple environmental variables\u0026mdash;including temperature (MAT, MATmax, MATmin), habitat structure, elevation (Ele), wind speed (WS10), precipitation (MAP), relative humidity (RH), and flowering plant richness\u0026mdash;had significant effects on pollinator communities. However, the direction of these effects varied among taxa and diversity metrics.\u003c/p\u003e\n\u003cp\u003ewhere \u0026ldquo;+\u0026rdquo; and \u0026ldquo;\u0026minus;\u0026rdquo; indicate the number of positive and negative significant effects, respectively. Overall, forest cover (For_prop) showed predominantly positive effects on community metrics (+24/\u0026minus;1), whereas cropland cover (Cro_prop) was mainly associated with negative effects (+0/\u0026minus;24). Precipitation (MAP) and relative humidity (RH) generally exhibited positive relationships with diversity metrics (MAP: +17/\u0026minus;0; RH: +9/\u0026minus;0), while wind speed (WS10) and temperature (MAT) were mostly associated with negative responses (WS10: +0/\u0026minus;8; MAT: +0/\u0026minus;5).\u003c/p\u003e\n\u003cp\u003eThese patterns suggest that increased forest cover and higher moisture availability are generally linked to enhanced pollinator diversity, whereas higher cropland proportion and elevated temperature or wind disturbance are more consistently associated with declines in community diversity.\u003c/p\u003e\n\u003cp\u003eTemperature variables showed contrasting effects across taxonomic groups and diversity metrics. With increasing maximum temperature (MATmax), Shannon diversity of the overall community (sampling-event scale) and both Shannon diversity and species richness of Lepidoptera declined (Fig. 3a), indicating negative effects of higher temperature extremes. In contrast, Shannon diversity of Hymenoptera increased with rising minimum temperature (MATmin) (Fig. 3b), suggesting a potential positive effect of milder minimum temperatures on this group. Increasing mean temperature (MAT) was associated with declines in species richness of the overall community, Coleoptera (site scale), and Lepidoptera (sampling-event scale), as well as reductions in Shannon diversity of the overall community and Lepidoptera (Fig. 3c; Fig. S8(f\u0026ndash;g)), indicating an overall negative effect of warming.\u003c/p\u003e\n\u003cp\u003eLandscape structure showed consistent but contrasting effects between forest and cropland proportions. Increasing forest cover was associated with higher species richness and Shannon diversity across all six groups (Fig. 3d), suggesting a generally positive effect on community diversity. However, abundance responses varied among taxa, with Hymenoptera abundance decreasing and Coleoptera abundance (sampling-event scale) increasing with forest cover (Fig. 3e). In contrast, increasing cropland proportion was associated with declines in species richness and Shannon diversity across all groups, along with reduced Coleoptera abundance (Fig. 3f; Fig. S8(c\u0026ndash;e)), indicating an overall negative effect of agricultural expansion.\u003c/p\u003e\n\u003cp\u003eAmong topographic and climatic factors, species richness of the overall community (site scale) increased with elevation (Fig. 3g), suggesting greater diversity at higher altitudes. Wind speed generally showed negative effects, with Shannon diversity and/or species richness of multiple groups (including the overall community, Coleoptera, Lepidoptera, and Hymenoptera) decreasing as wind speed increased (Fig. 3h; Fig. S8(i)).\u003c/p\u003e\n\u003cp\u003eIn contrast, precipitation and relative humidity were generally associated with positive effects. Shannon diversity increased with precipitation across all six groups (Fig. 3i), and species richness of several groups (including the overall community, Coleoptera, Lepidoptera, and Hymenoptera) also increased, accompanied by higher Hymenoptera abundance (Fig. S8 (e\u0026ndash;f)). Similarly, increasing humidity was associated with higher Shannon diversity in multiple groups (Fig. 3j) and increased species richness in the overall community, Lepidoptera, and Hymenoptera (Fig. S8 (c\u0026ndash;e)), suggesting that wetter conditions generally promote community diversity.\u003c/p\u003e\n\u003cp\u003eThe effects of floral resources differed markedly among taxa. Shannon diversity of Diptera (site scale) decreased with increasing flowering plant richness (Fig. 3l), whereas Shannon diversity of Lepidoptera (Fig. 3k), as well as species richness of the overall community and Lepidoptera, increased. Hemiptera abundance decreased with increasing floral richness, whereas Lepidoptera abundance increased (Fig. S8(b\u0026ndash;c)). These results indicate strong taxon-specific responses to floral resource availability.\u003c/p\u003e\n\u003cp\u003eOverall, the main effects of environmental variables suggest that forest cover and higher moisture conditions generally promote pollinator diversity, whereas cropland expansion and strong wind disturbance have negative impacts. In contrast, the effects of temperature and floral resources are more complex and vary depending on taxonomic group and diversity metric.\u003c/p\u003e\n\u003ch2\u003eHabitat interaction effects analysis\u003c/h2\u003e\n\u003cp\u003eIn the habitat interaction models, the direction of interaction effects was approximately balanced overall (+33/\u0026minus;34), but clear differences emerged among environmental variables. Interaction terms involving precipitation (MAP) were predominantly negative (+0/\u0026minus;10), whereas those involving temperature variables (MAT and MATmax) were mainly positive (MAT: +8/\u0026minus;0; MATmax: +6/\u0026minus;0). These results indicate that the effects of climatic variables are habitat-dependent.\u003c/p\u003e\n\u003cp\u003eResponses of pollinator communities to climatic factors differed markedly among habitat types. Natural, agricultural, and semi-natural habitats frequently showed contrasting, and in some cases opposite, response patterns to temperature, wind speed, precipitation, humidity, and elevation.\u003c/p\u003e\n\u003cp\u003eFor temperature variables, natural habitats generally exhibited response patterns distinct from those of agricultural and semi-natural habitats. With increasing MATmax, Shannon diversity of Coleoptera at the sampling-event scale (Fig. 4a), species richness of Diptera at the site scale, and abundance of the overall assemblage and Hemiptera at the sampling-event scale increased in natural habitats, whereas opposite trends were observed in agricultural and semi-natural habitats. Hymenoptera showed a more complex pattern: abundance at the sampling-event scale increased with MATmax in natural and semi-natural habitats but decreased in agricultural habitats, while abundance at the site scale decreased in natural and agricultural habitats but showed the opposite trend in semi-natural habitats (Fig. S9(c)).\u003c/p\u003e\n\u003cp\u003eUnder MATmin, species richness of Hemiptera and Diptera at the site scale increased with temperature in natural habitats (Fig. 4b), but declined in agricultural and semi-natural habitats. In contrast, Hymenoptera abundance at the site scale decreased in natural and agricultural habitats, but increased in semi-natural habitats. Species richness of Lepidoptera at the site scale declined in natural habitats but increased in the other two habitats (Fig. 4c). Similar patterns were observed for MAT: species richness of Diptera, abundance of Hymenoptera at the site scale (Fig. 4d), Shannon diversity of Coleoptera at the sampling-event scale, and abundance of the overall assemblage and Hemiptera generally increased with temperature in natural habitats, but showed opposite responses in agricultural and semi-natural habitats (Fig. S9(c)).\u003c/p\u003e\n\u003cp\u003eAlong the elevational gradient, species richness of Diptera (Fig. 4e), as well as abundance of the overall assemblage and Hemiptera, decreased with elevation in natural habitats, whereas the opposite trends were generally observed in agricultural and semi-natural habitats. By contrast, species richness of Lepidoptera at the site scale increased with elevation in natural habitats (Fig. 4f) but declined in the other two habitats (Fig. S9(a)).\u003c/p\u003e\n\u003cp\u003eWind speed also produced habitat-specific effects. In natural habitats, Shannon diversity and species richness of Lepidoptera increased with wind speed (Fig. 4h), whereas species richness of Diptera at the site scale (Fig. 4g), species richness of Coleoptera at the sampling-event scale, and abundance of Diptera and Hemiptera at the sampling-event scale generally declined. Most of these relationships were reversed in agricultural and semi-natural habitats. Abundance of Coleoptera at the site scale increased with wind speed in natural and semi-natural habitats, but decreased in agricultural habitats (Fig. S9(d)).\u003c/p\u003e\n\u003cp\u003ePrecipitation effects also differed among habitats. Shannon diversity of Hemiptera at the site scale declined with increasing precipitation in natural habitats, but increased in agricultural and semi-natural habitats. Shannon diversity of the overall assemblage at the sampling-event scale and of Coleoptera at the site scale increased with precipitation in natural and semi-natural habitats, but decreased in agricultural habitats. Meanwhile, abundance of the overall assemblage, Lepidoptera (Fig. 4i), Coleoptera, and Hemiptera declined with precipitation in natural habitats, but showed the opposite trends in the other two habitats (Fig. S9(a\u0026ndash;b)).\u003c/p\u003e\n\u003cp\u003eFor humidity, abundance (Fig. 4j) and species richness of Diptera at the site scale increased with relative humidity in natural habitats, whereas Shannon diversity and species richness of the overall assemblage, Coleoptera, and Hemiptera generally declined. Agricultural and semi-natural habitats often exhibited opposite response patterns (Fig. S9(c\u0026ndash;d)).\u003c/p\u003e\n\u003cp\u003eOverall, natural habitats differed markedly from the more disturbed agricultural and semi-natural habitats in their responses to climatic factors. In natural habitats, increasing temperature often promoted some community metrics, whereas increasing precipitation and humidity were frequently associated with declines in diversity or abundance. By contrast, agricultural and semi-natural habitats more often exhibited response patterns opposite to those observed in natural habitats.\u003c/p\u003e\n\u003ch2\u003eEnvironmental interaction effects analysis\u003c/h2\u003e\n\u003cp\u003eThe results further demonstrated that these moderating effects followed clear directional patterns. Interaction terms involving cropland cover and temperature, including Cro_prop \u0026times; MAT and Cro_prop \u0026times; MATmax, were predominantly negative (+0/\u0026minus;19 and +0/\u0026minus;15, respectively). In contrast, interactions involving forest cover and temperature, including For_prop \u0026times; MAT and For_prop \u0026times; MATmax, were mainly positive (+19/\u0026minus;0 and +14/\u0026minus;0, respectively), whereas For_prop \u0026times; Ele was predominantly negative (+0/\u0026minus;19). Similarly, MAP \u0026times; MAT was mainly positive (+15/\u0026minus;0), while MAP \u0026times; Ele was predominantly negative (+0/\u0026minus;14).\u003c/p\u003e\n\u003cp\u003eTaken together, these results indicate that pollinator communities in the study region were not driven by single environmental factors in a simple linear manner. Instead, land-use variables, hydrothermal conditions, and topographic gradients jointly shaped community structure through significant interactions, producing differential response patterns in species richness, abundance, and Shannon diversity.\u003c/p\u003e\n\u003cp\u003eFrom a landscape perspective, habitat structure not only modified the direction of environmental effects, but also altered the strength of climatic influences on pollinator communities. In areas with high forest cover and high cropland cover, the effects of temperature, wind speed, precipitation, and humidity were frequently opposite.\u003c/p\u003e\n\u003cp\u003eUnder high forest cover, increasing MATmax was associated with higher Shannon diversity of Diptera at the site scale, higher species richness of Diptera and the overall assemblage at the sampling-event scale, and greater abundance of the overall assemblage, Hemiptera, and Hymenoptera at the site scale, whereas opposite trends were observed in areas with low forest cover. In contrast, Shannon diversity of Hemiptera at the site scale declined with increasing MATmax under high forest cover (Fig. 5a), but increased in low-forest landscapes, indicating divergent responses among taxa and community metrics (Fig. S10(u-v)).\u003c/p\u003e\n\u003cp\u003eThis general pattern was maintained for MATmin and MAT. Under high forest cover, species richness of Diptera at the site scale, abundance of Hemiptera at the site scale, and abundance of the overall assemblage, Hemiptera, and Hymenoptera generally increased with temperature (Fig. 5c), whereas the opposite pattern was found under low forest cover. However, Shannon diversity of Hemiptera at the site scale and species richness of Lepidoptera at the site scale declined with increasing MATmin under high forest cover (Fig. 5b), suggesting that forested landscapes may simultaneously enhance and filter pollinator responses to warming (Fig. S10(t, v)).\u003c/p\u003e\n\u003cp\u003eIn contrast, under high cropland cover, increasing MATmax, MATmin, and MAT generally reduced abundance of the overall assemblage, Hymenoptera, and Hemiptera (Fig. S10(s\u0026ndash;t)), while species richness of Diptera at the site scale also declined (Fig. 5h). Low-cropland landscapes usually showed opposite trends, suggesting that cropland-dominated areas are more vulnerable to warming.\u003c/p\u003e\n\u003cp\u003eWind effects were also strongly mediated by landscape context. Under high forest cover, abundance of Coleoptera at the site scale, Shannon diversity of Hemiptera at the site scale, and species richness of Lepidoptera increased with wind speed (Fig. 5d), whereas abundance of Diptera at the site scale declined; most of these trends were reversed under low forest cover (Fig. S10(ac)). Under high cropland cover, species richness of Coleoptera and Lepidoptera at the sampling-event scale, as well as abundance of Coleoptera at the site scale, declined with increasing wind speed, whereas abundance of Diptera at the site scale increased (Fig. 5j; Fig. S10(aa)), indicating that wind disturbance in cropland landscapes may suppress most groups while favoring some disturbance-tolerant taxa.\u003c/p\u003e\n\u003cp\u003ePrecipitation and humidity also showed contrasting effects under different landscape backgrounds. In high-forest landscapes, several community metrics showed weak or negligible responses to increasing precipitation, including Shannon diversity of the overall assemblage at the sampling-event scale, species richness of the overall assemblage (Fig. 5e), and species richness of Coleoptera at the site scale. By contrast, Shannon diversity of Hemiptera at the site scale, species richness of Lepidoptera at the sampling-event scale, and abundance of Lepidoptera at the site scale declined with increasing precipitation, whereas most of these indicators increased under low forest cover (Fig. S10(p)). In high-cropland landscapes, abundance of the overall assemblage and richness- or diversity-related metrics of Lepidoptera, Hemiptera, and Coleoptera generally increased with precipitation (Fig. 5k; Fig. S10(m-n)), whereas opposite trends were found in low-cropland landscapes.\u003c/p\u003e\n\u003cp\u003eHumidity responses showed a similar landscape dependency. Under high forest cover, species richness and Shannon diversity of Coleoptera at the sampling-event scale declined with increasing humidity, whereas abundance of Diptera at the site scale and species richness of Hymenoptera at the site scale increased (Fig. 5f; Fig. S10(x)). Under high cropland cover, species richness and Shannon diversity of the overall assemblage, Coleoptera (Fig. 5l), and Hemiptera generally increased with humidity, whereas the opposite pattern was observed under low cropland cover. In addition, abundance of Diptera at the site scale declined with increasing precipitation in cropland-dominated landscapes, further indicating that landscape background modifies community adaptation to moist conditions (Fig. S10(v)).\u003c/p\u003e\n\u003cp\u003eOverall, warming tended to promote multiple community metrics in forest-dominated landscapes, but more often led to community decline in cropland-dominated landscapes. Likewise, the effects of wind, precipitation, and humidity frequently reversed depending on landscape context, highlighting landscape structure as a key moderator of climatic effects on pollinator communities.\u003c/p\u003e\n\u003cp\u003eSignificant interactions were detected between climatic variables and landscape structure, with both forest and cropland proportions substantially modifying the direction and strength of climatic effects on pollinator communities.\u003c/p\u003e\n\u003cp\u003eFor temperature\u0026ndash;forest interactions, MATmax, MATmin, and MAT showed consistent patterns. Under high-temperature conditions, increasing forest cover was generally associated with higher Shannon diversity, species richness, and abundance across the overall assemblage (Fig. 6a, e) and multiple taxa, including Coleoptera, Diptera, Hymenoptera, and Hemiptera (Fig. 6b). In contrast, under low-temperature conditions, these relationships were often reversed. This pattern indicates that forested landscapes buffer thermal stress and promote community performance under warming, whereas such effects weaken or reverse under cooler conditions (Fig. S10(i-l)).\u003c/p\u003e\n\u003cp\u003eIn contrast, temperature\u0026ndash;cropland interactions showed predominantly negative effects. Under high-temperature conditions, increasing cropland cover was associated with declines in Shannon diversity, species richness, and abundance of the overall assemblage and major taxa, whereas opposite trends were generally observed under low-temperature conditions. This pattern was consistent across MATmax, MATmin, and MAT, and was particularly evident for the overall assemblage (Fig. 6b, f) and for Hymenoptera, Diptera, Coleoptera, and Hemiptera (Fig. 6d; Fig. S10(c-f)). These results suggest that cropland-dominated landscapes amplify thermal stress on pollinator communities.\u003c/p\u003e\n\u003cp\u003eWind-landscape interactions also exhibited contrasting effects between forest and cropland. Under high wind speed, increasing forest cover was associated with higher species richness of Lepidoptera (Fig. 6g), higher Shannon diversity of Hemiptera (sampling-event scale), and increased Lepidoptera abundance, whereas opposite trends were observed under low wind speed (Fig. S10(m)). In contrast, under high wind speed, increasing cropland cover reduced species richness of Lepidoptera (Fig. 6h) and Shannon diversity of Hemiptera, while the opposite trends occurred under low wind conditions (Fig. S10(f)). These results indicate that forest and cropland landscapes exert opposing moderating effects on wind disturbance.\u003c/p\u003e\n\u003cp\u003ePrecipitation\u0026ndash;landscape interactions showed similar contrasts. Under high precipitation, increasing forest cover promoted Shannon diversity of Lepidoptera (sampling-event scale) (Fig. 6i) and Hemiptera (site scale), species richness of Lepidoptera, and abundance of Diptera at the site scale, but reduced Lepidoptera abundance. Under low precipitation, most of these relationships were reversed (Fig. S10(i)). Conversely, under high precipitation, increasing cropland cover reduced most diversity and richness metrics, while increasing Lepidoptera abundance (Fig. 6g; Fig. S10(b-c)), suggesting that cropland landscapes under wet conditions may suppress diversity but facilitate the dominance of certain taxa.\u003c/p\u003e\n\u003cp\u003eHumidity\u0026ndash;landscape interactions also depended strongly on landscape context. Under high humidity, increasing forest cover reduced Lepidoptera abundance (Fig. 6k), Diptera abundance at the site scale, and Shannon diversity of Hemiptera, whereas opposite trends were observed under low humidity (Fig. S10(l-m)). In contrast, increasing cropland cover reduced species richness of Lepidoptera (Fig. 6l) and Shannon diversity of Hemiptera under high humidity, while opposite patterns occurred under low humidity (Fig. S10(f)).\u003c/p\u003e\n\u003cp\u003eOverall, forest and cropland proportions exhibited contrasting moderating roles. Forest-dominated landscapes generally enhanced or buffered pollinator diversity under stressful climatic conditions (e.g., high temperature, wind, or precipitation), whereas cropland-dominated landscapes tended to suppress community metrics. These findings highlight landscape structure as a critical context-dependent regulator of climatic effects on pollinator community composition and stability.\u003c/p\u003e\n\u003cp\u003eFlowering plant richness interacted strongly with climatic variables, and both the direction and magnitude of responses to temperature, wind speed, precipitation, humidity, and elevation varied among taxa under contrasting resource conditions.\u003c/p\u003e\n\u003cp\u003eFor temperature variables, flowering plant richness clearly modified thermal effects. At the sampling-event scale, Shannon diversity of Diptera increased with MATmax and MAT under high flowering plant richness (Fig. 7a, h), indicating that abundant floral resources strengthened the positive response of Diptera to warmer conditions. In contrast, Hymenoptera at the site scale showed a different pattern: both species richness and Shannon diversity increased with MATmin under low flowering plant richness, whereas responses were weak or nearly absent under high flowering plant richness (Fig. 7i; Fig. S10(u-v)). This suggests that abundant floral resources may weaken the positive effect of higher minimum temperature on Hymenoptera communities.\u003c/p\u003e\n\u003cp\u003eWind-speed effects were also resource dependent. Species richness and Shannon diversity of the overall assemblage at the sampling-event scale (Fig. 7e) and of Hymenoptera at the site scale declined with increasing wind speed, but these declines were stronger under low flowering plant richness, indicating that abundant floral resources may buffer wind stress. In contrast, species richness of Lepidoptera at the site scale showed the opposite pattern (Fig. 7f), increasing with wind speed under high flowering plant richness but decreasing under low flowering plant richness, suggesting that Lepidoptera responses to wind are strongly conditioned by resource availability (Fig. S10(a-b)).\u003c/p\u003e\n\u003cp\u003eUnder increasing precipitation, different taxa and community metrics showed more complex resource-dependent patterns. Abundance of the overall assemblage (Fig. 7d) and Hymenoptera, as well as species richness and Shannon diversity of Hemiptera at the sampling-event scale, declined with precipitation when flowering plant richness was high, but showed opposite trends under low flowering plant richness. By contrast, species richness and Shannon diversity of the overall assemblage at the sampling-event scale (Fig. 7g), along with species richness of Coleoptera at the sampling-event scale, increased only slightly with precipitation under high flowering plant richness, but increased markedly under low flowering plant richness (Fig. S10(o-p)). These results suggest that precipitation has a stronger positive effect on some community metrics when floral resources are limited.\u003c/p\u003e\n\u003cp\u003eHumidity effects were similarly altered by flowering plant richness. Species richness of the overall assemblage at the sampling-event scale and of Hymenoptera at the site scale (Fig. 7c) increased with relative humidity under high flowering plant richness, but the increase was stronger under low flowering plant richness. In contrast, abundance of the overall assemblage at the sampling-event scale (Fig. 7b), species richness of Lepidoptera at the site scale, abundance of Diptera, and species richness of Diptera at the sampling-event scale declined with increasing humidity under high flowering plant richness, whereas opposite trends were observed under low flowering plant richness (Fig. S10(w-x)). This indicates that humidity effects frequently reverse under contrasting floral-resource conditions.\u003c/p\u003e\n\u003cp\u003eIn addition, along the elevational gradient, Shannon diversity of Diptera at the sampling-event scale declined with elevation under high flowering plant richness, but showed the opposite trend under low flowering plant richness (Fig. S10(f)), further demonstrating that resource conditions strongly modify the effects of climatic and topographic factors on community structure.\u003c/p\u003e\n\u003cp\u003eOverall, flowering plant richness not only influenced pollinator communities directly, but also substantially altered the direction and strength of climatic effects, generating pronounced resource-dependent responses in diversity, species richness, and abundance across taxonomic group\u003c/p\u003e\n\u003cp\u003eSignificant interactions were detected between climatic variables and flowering plant richness, with contrasting effects across pollinator taxa.\u003c/p\u003e\n\u003cp\u003eFor temperature-related variables, MATmax, MATmin, and MAT showed broadly consistent interaction patterns. Under warmer conditions, species richness (Fig. 8a, b) and Shannon diversity (Fig. 8c) of Hymenoptera at the site scale declined with increasing flowering plant richness. In contrast, Shannon diversity of Diptera at the sampling-event scale increased with flowering plant richness under high MATmax and MAT conditions (Fig. 8d, f). In addition, under high MATmin, abundance of Lepidoptera at the sampling-event scale also increased with flowering plant richness (Fig. 8e). These results indicate clear taxon-specific differences in responses to floral-resource availability under warm conditions.\u003c/p\u003e\n\u003cp\u003eUnder high precipitation, Shannon diversity of the overall assemblage at the sampling-event scale (Fig. 8g), species richness and Shannon diversity of Hemiptera, and abundance of Coleoptera at the site scale all declined with increasing flowering plant richness (Fig. S10(g)), suggesting that under wetter conditions, greater floral-resource availability may have suppressive effects on some community metrics.\u003c/p\u003e\n\u003cp\u003eAlong the elevational gradient, abundance of Lepidoptera at the sampling-event scale decreased with increasing flowering plant richness under high-elevation conditions (Fig. 8h), indicating that the positive effects of floral resources may be constrained at higher elevations.\u003c/p\u003e\n\u003cp\u003eUnder high wind speed, abundance of Coleoptera at the site scale increased with flowering plant richness (Fig. 8i), suggesting that this group may be particularly effective at exploiting floral resources under strong wind disturbance.\u003c/p\u003e\n\u003cp\u003eUnder high humidity, Shannon diversity of the overall assemblage at the sampling-event scale, as well as Shannon diversity and species richness of Hymenoptera at the site scale, and Shannon diversity of Lepidoptera at the sampling-event scale, all declined with increasing flowering plant richness (Fig. S10(g)). This pattern suggests that under highly humid conditions, greater floral-resource availability may negatively affect pollinator diversity.\u003c/p\u003e\n\u003cp\u003eOverall, the effects of flowering plant richness on pollinator communities were strongly climate dependent. The direction of resource effects varied substantially among climatic conditions and differed markedly among taxonomic groups, indicating pronounced inconsistency in community responses.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eHabitat type shapes both α- and β-diversity of pollinator communities\u003c/h2\u003e \u003cp\u003eOur results indicate that habitat type is a major determinant of both α- and β-diversity in pollinator insect communities. Natural habitats consistently supported higher species richness and Shannon diversity, whereas agricultural habitats were characterized by lower diversity, stronger dominance, and increased community homogenization. Semi-natural habitats generally showed intermediate patterns, highlighting their transitional role along the land-use gradient. Notably, abundance was often higher in semi-natural and agricultural habitats than in natural habitats, suggesting that a greater number of individuals does not necessarily correspond to higher diversity. Instead, elevated abundance in disturbed habitats may reflect the proliferation of a limited number of dominant or disturbance-tolerant taxa, accompanied by reduced evenness and stronger homogenization.\u003c/p\u003e \u003cp\u003eTaxon-specific patterns further support this interpretation. Coleoptera exhibited the highest species richness, Lepidoptera showed the highest Shannon diversity, and Diptera had the lowest diversity and appeared to be the most sensitive to habitat change. Together, these results suggest that natural and semi-natural habitats are more effective than agricultural habitats in maintaining diverse and relatively stable pollinator communities.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eCommunity turnover along the land-use gradient reflects strong environmental filtering\u003c/h2\u003e \u003cp\u003eThese results demonstrate that pollinator community structure and diversity in the Qinling Mountains are jointly shaped by landscape structure, climatic conditions, and floral-resource availability, with both main effects and pairwise interactions determining the magnitude and direction of community responses. Environmental differentiation among sites was primarily associated with agricultural intensity, temperature extremes, elevation, and flowering plant richness. Agricultural habitats were characterized by higher cropland cover, warmer and drier conditions, and stronger disturbance, whereas natural habitats were associated with higher elevation, greater humidity, and richer floral resources; semi-natural habitats were intermediate. The clear separation of habitats in ordination space indicates that environmental filtering is a key mechanism underlying community differentiation.\u003c/p\u003e \u003cp\u003eConsidering main effects, forest cover was consistently associated with higher species richness and diversity, whereas cropland proportion showed broadly negative relationships across taxa. This pattern reflects the stabilizing role of forested landscapes in providing continuous resources and buffered microclimates, in contrast to the homogenizing effects of agricultural expansion (Ulyshen et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Temperature emerged as one of the strongest drivers, although responses varied among taxa. In most cases, warming was associated with reduced diversity or abundance in the overall assemblage, Lepidoptera, and part of Coleoptera, whereas Hymenoptera showed positive responses under some thermal conditions. These differences are consistent with variation among taxa in thermal tolerance and activity constraints (Johnson et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Karbassioon et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Miyashita et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Wind speed showed predominantly negative effects, indicating constraints on pollinator activity (Hennessy et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Hennessy et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), whereas precipitation and humidity were generally associated with positive responses, likely reflecting enhanced vegetation and floral-resource supply. Flowering plant richness showed taxon-dependent effects, promoting Lepidoptera but showing neutral or negative relationships for some Diptera and Hemiptera metrics, consistent with differences in resource-use strategies (Hyjazie \u0026amp; Sargent, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Davis et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eEnvironmental effects were strongly context dependent. Habitat type significantly modified climatic responses, with the same climatic factor often producing different or opposite patterns among habitats. In particular, temperature increases were sometimes associated with positive responses in natural habitats but negative responses in agricultural and semi-natural habitats, whereas precipitation and humidity showed the reverse tendency. These patterns indicate that habitat structure determines the extent to which communities buffer climatic variability (Ganuza et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eClimate also interacted strongly with landscape structure. Forest cover generally mitigated the negative effects of high temperature and wind, whereas cropland proportion amplified them, highlighting the dual role of landscape composition in directly shaping habitat conditions and indirectly modifying climatic stress (Ulyshen et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Ganuza et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Similar context dependence was observed for precipitation and humidity, with forest-dominated landscapes more often maintaining diversity, while cropland-dominated landscapes favored disturbance-tolerant assemblages.\u003c/p\u003e \u003cp\u003eFloral resources further modulated climatic responses. Under high-temperature conditions, increased flowering plant richness enhanced positive responses in Diptera but weakened or reversed responses in Hymenoptera. Under high wind conditions, negative effects on diversity were stronger when floral resources were limited and weaker when resources were abundant, indicating a buffering role of floral availability. This pattern is consistent with evidence that wind reduces pollinator activity, whereas resource-rich environments can partially offset these constraints (Hennessy et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Hennessy et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Balfour \u0026amp; Ratnieks, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Interactions involving precipitation, humidity, and floral resources showed strong context dependence, and similar patterns were observed along elevational gradients, consistent with shifts in pollinator assemblages along temperature\u0026ndash;elevation axes (McCabe \u0026amp; Cobb, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOverall, pollinator community responses were governed by multiple interacting drivers rather than single environmental factors. Temperature acted as a key regulator, but its effects depended on habitat context, landscape structure, and resource availability. Forest cover generally buffered environmental stress, whereas agricultural expansion amplified it, and floral resources further mediated taxon-specific responses. These findings highlight the need for integrated management strategies that combine forest conservation, limitation of agricultural expansion, and maintenance of floral and non-floral resources to sustain pollinator diversity (Ulyshen et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Davis et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eConclusions and recommendations\u003c/h2\u003e \u003cp\u003eThis study demonstrates that pollinator insect communities in the Qinling Mountains vary markedly along the land-use gradient in terms of α-diversity, β-diversity, community composition, and environmental responses. Compared with agricultural habitats, natural and semi-natural habitats play a more important role in maintaining pollinator diversity and community stability. Natural habitats are particularly irreplaceable because they support higher diversity, contain more endemic and indicator species, and provide stronger buffering against environmental fluctuations. Semi-natural habitats, although generally intermediate in community metrics, also contribute substantially by preserving landscape heterogeneity and offering transitional habitat conditions that help sustain relatively diverse assemblages. By contrast, agricultural habitats may support high pollinator abundance, but this should not be interpreted as high conservation value, because such abundance is likely to be driven by a limited number of widespread, disturbance-tolerant taxa, together with reduced evenness and increased biotic homogenization.\u003c/p\u003e \u003cp\u003eThese findings have direct implications for pollinator conservation in the Qinling Mountains and similar mountain agroecosystems. Conservation efforts should not focus solely on maintaining pollinator abundance, but should place greater emphasis on preserving community composition, diversity, and ecological stability. In particular, retaining natural and semi-natural habitats, maintaining or increasing forest cover, and ensuring abundant and continuous floral resources should be prioritized as key strategies to buffer environmental stress and mitigate the negative effects of agricultural intensification. More broadly, limiting excessive cropland expansion, enhancing landscape heterogeneity, and conserving both floral and non-floral habitat resources are likely to be essential for sustaining pollinator biodiversity and pollination services under ongoing land-use and climate change.\u003c/p\u003e \u003cp\u003eSeveral limitations should also be acknowledged. First, although this study identified broad patterns along the land-use gradient, the underlying mechanisms remain largely inferential because the analyses are based primarily on correlations. Second, taxonomic composition and conventional diversity indices alone cannot fully capture the functional consequences of pollinator community change. Third, additional attention should be given to multi-scale landscape processes, including habitat configuration, connectivity, and fragmentation, which may further influence community assembly. Accordingly, future research should combine long-term monitoring across seasons and years with stronger causal approaches, such as structural equation modelling, hierarchical models, or manipulative experiments, and should incorporate functional traits, flower-visitation networks, pollination efficiency, and plant reproductive success. Integrating community patterns, environmental drivers, and ecosystem functioning will provide a stronger scientific basis for pollinator conservation and habitat management in the Qinling region and other mountain agricultural landscapes.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis research was funded by the National Key R\u0026amp;D Program of China (2022YFE0115200), Horizon 2020 \u0026ndash; the Framework Programme for Research and Innovation (Safeguarding European Wild Pollinators, SAFEGUARD, SC5-32-2020); the Biodiversity Survey and the Assessment Project of the Ministry of Ecology and Environment, China (2019HJ2096001006) and the National Animal Collection Resource Center, China.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eAuthor contributions\u003c/h2\u003e\n\u003cp\u003eYS conceived the study, conducted field sampling, curated the data, performed the analyses, prepared the figures, and wrote the first draft of the manuscript. QD contributed to field investigation, analytical methodology, and data curation. QX assisted with the work and revised the manuscript. YZ designed the project, supervised the study, acquired funding, and critically revised the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eSupplementary material\u003c/h2\u003e\n\u003cp\u003eSupplementary material is available online as three separate files: Supplementary Methods, Supplementary Tables (Tables S1\u0026ndash;S7), and Supplementary Figures (Figs. S1\u0026ndash;S10).\u003c/p\u003e\n\u003ch2\u003eData availability\u003c/h2\u003e\n\u003cp\u003eThe datasets supporting the findings of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003ch2\u003eCode availability\u003c/h2\u003e\n\u003cp\u003eThe custom R scripts used for data processing and analysis are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003ch2\u003eEthics approval\u003c/h2\u003e\n\u003cp\u003eNot applicable. This study involved field sampling of insects and did not include experiments on vertebrate animals or human participants.\u0026nbsp;Field sampling was conducted in accordance with local regulations.\u003c/p\u003e\n\u003ch2\u003eConsent to participate\u003c/h2\u003e\n\u003cp\u003e\u0026nbsp;Not applicable. This study did not involve human participants.\u003c/p\u003e\n\u003ch2\u003eConsent for publication\u003c/h2\u003e\n\u003cp\u003eAll authors approved publication.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAdams BJ, Li E, Bahlai CA, Meineke EK, McGlynn TP, Brown BV (2020) Local- and landscape-scale variables shape insect diversity in an urban biodiversity hot spot. Ecological Applications 30:e02089.\u003c/li\u003e\n\u003cli\u003eBalfour NJ, Ratnieks FLW (2025) Wind alters plant-pollinator community structure, bee foraging rate and movements between plants. Behavioral Ecology 36: araf067.\u003c/li\u003e\n\u003cli\u003eBuchori D, Rizali A, Larasati A, Hidayat P, Ngo H, Gemmill-Herren B (2019) Natural habitat fragments obscured the distance effect on maintaining the diversity of insect pollinators and crop productivity in tropical agricultural landscapes. Heliyon 5: e01373.\u003c/li\u003e\n\u003cli\u003eChowdhury S, Jennions MD, Zalucki MP, Maron M, Watson JEM, Fuller RA (2023) Protected areas and the future of insect conservation. Trends in Ecology \u0026amp; Evolution 38:85\u0026ndash;95. https://doi.org/10.1016/j.tree.2022.09.004\u003c/li\u003e\n\u003cli\u003eCrist TO, Peters VE (2014) Landscape and local controls of insect biodiversity in conservation grasslands: implications for the conservation of ecosystem service providers in agricultural environments. Land 3:693\u0026ndash;718. https://doi.org/10.3390/land3030693\u003c/li\u003e\n\u003cli\u003eDavis AE, Bickel DJ, Saunders ME, Rader R (2023) Crop-pollinating Diptera have diverse diets and habitat needs in both larval and adult stages. Ecological Applications 33: e2859.\u003c/li\u003e\n\u003cli\u003eDicks LV, Breeze TD, Ngo HT, Senapathi D, An J, Aizen MA, et al. (2021) A global-scale expert assessment of drivers and risks associated with pollinator decline. Nature Ecology \u0026amp; Evolution 5:1453\u0026ndash;1461.\u003c/li\u003e\n\u003cli\u003eFahrig L (2003) Effects of habitat fragmentation on biodiversity. Annual Review of Ecology, Evolution, and Systematics 34:487\u0026ndash;515.\u003c/li\u003e\n\u003cli\u003eFan H, Liu T, Chen Y, Liao Z, Chen J, Hu Y, et al. (2024) Geographical patterns and determinants of insect biodiversity in China. Science China Life Sciences 67:1255\u0026ndash;1265.\u003c/li\u003e\n\u003cli\u003eFang Z, Xu WH, Zhang JJ, Xiao Y, Zhang L (2017) Planning of the natural reserve system in the Qinling Mountains based on biodiversity and ecosystem services. Acta Ecologica Sinica 37:5334\u0026ndash;5341 [in Chinese].\u003c/li\u003e\n\u003cli\u003eFenoglio MS, Calvi\u0026ntilde;o A, Gonz\u0026aacute;lez E, Salvo A, Videla M (2021) Urbanisation drivers and underlying mechanisms of terrestrial insect diversity loss in cities. Ecological Entomology 46:757\u0026ndash;771. https://doi.org/10.1111/een.13041\u003c/li\u003e\n\u003cli\u003eFreimuth J, Bossdorf O, Scheepens JF, Willems FM (2022) Climate warming changes synchrony of plants and pollinators. Proceedings of the Royal Society B 289:20212142.\u003c/li\u003e\n\u003cli\u003eGalloway AD, Seymour CL, Gaigher R, Pryke JS (2021) Organic farming promotes arthropod predators, but this depends on neighbouring patches of natural vegetation. Agriculture, Ecosystems \u0026amp; Environment 310:107295. https://doi.org/10.1016/j.agee.2020.107295\u003c/li\u003e\n\u003cli\u003eGanuza C, Redlich S, Uhler J, et al. (2022) Interactive effects of climate and land use on pollinator diversity differ among taxa and scales. Science Advances 8: eabm9359.\u003c/li\u003e\n\u003cli\u003eGarratt MPD, De Groot GA, Albrecht M, Bosch J, Breeze TD, Fountain MT, et al. (2021) Opportunities to reduce pollination deficits and address production shortfalls in an important insect-pollinated crop. Ecological Applications 31: e02445.\u003c/li\u003e\n\u003cli\u003eGhisbain G, G\u0026eacute;rard M, Wood TJ, Hines HM, Michez D (2021) Expanding insect pollinators in the Anthropocene. Biological Reviews 96:2755\u0026ndash;2770.\u003c/li\u003e\n\u003cli\u003eHahn PG, Orrock JL (2015) Land-use history alters contemporary insect herbivore community composition and decouples plant\u0026ndash;herbivore relationships. Journal of Animal Ecology 84:745\u0026ndash;754.\u003c/li\u003e\n\u003cli\u003eHalsch CA, Shapiro AM, Fordyce JA, Nice CC, Thorne JH, Waetjen DP, et al. (2020) Insects and recent climate change. bioRxiv. https://doi.org/10.1101/2020.03.09.984328\u003c/li\u003e\n\u003cli\u003eHarvey JA, Heinen R, Gols R, Thakur MP (2020) Climate change-mediated temperature extremes and insects: from outbreaks to breakdowns. Global Change Biology 26:6685\u0026ndash;6701.\u003c/li\u003e\n\u003cli\u003eHarvey JA, Tougeron K, Gols R, et al. (2023) Scientists\u0026rsquo; warning on climate change and insects. Ecological Monographs 93: e1553.\u003c/li\u003e\n\u003cli\u003eHennessy G, Harris C, Eaton C, Wright P, Jackson E, Goulson D, et al. (2020) Gone with the wind: effects of wind on honey bee visit rate and foraging behaviour. Animal Behaviour 161:23\u0026ndash;31.\u003c/li\u003e\n\u003cli\u003eHennessy G, Harris C, Pirot LMP, Wright P, Jackson E, Goulson D, et al. (2021) Wind slows play: increasing wind speed reduces flower visiting rate in honey bees. Animal Behaviour 178:87\u0026ndash;93.\u003c/li\u003e\n\u003cli\u003eHyjazie BF, Sargent RD (2022) Floral resources predict the local bee community: implications for conservation. Biological Conservation 273:109679.\u003c/li\u003e\n\u003cli\u003eJaccard P (1908) Nouvelles recherches sur la distribution florale. Bulletin de la Soci\u0026eacute;t\u0026eacute; Vaudoise des Sciences Naturelles 44:223\u0026ndash;270\u003c/li\u003e\n\u003cli\u003eJohnson MG, Glass JR, Dillon ME, Harrison JF (2023) How will climatic warming affect insect pollinators? Advances in Insect Physiology 64:1\u0026ndash;115.\u003c/li\u003e\n\u003cli\u003eKarbassioon A, Yearsley J, Dirilgen T, Hayes B, de Ibarra NH, Nunes-Silva P (2023) Responses in honeybee and bumblebee activity to changes in weather conditions. Oecologia 201:689\u0026ndash;701.\u003c/li\u003e\n\u003cli\u003eKellermann V, van Heerwaarden B (2019) Terrestrial insects and climate change: adaptive responses in key traits. Physiological Entomology 44:99\u0026ndash;115. https://doi.org/10.1111/phen.12282\u003c/li\u003e\n\u003cli\u003eKov\u0026aacute;cs-Hosty\u0026aacute;nszki A, Esp\u0026iacute;ndola A, Vanbergen AJ, Settele J, Kremen C, Dicks LV (2017) Ecological intensification to mitigate impacts of conventional intensive land use on pollinators and pollination. Ecology Letters 20:673\u0026ndash;689.\u003c/li\u003e\n\u003cli\u003eLi JY, Fu BJ, Sun JL, Hong ZL, Zhang BP, Wang XF, et al. (2021) Ecological civilization construction in the Qinling Mountains in the new era: existing problems and development pathways. Journal of Natural Resources 36:2449\u0026ndash;2463 [in Chinese].\u003c/li\u003e\n\u003cli\u003eLiu J, Xie T, Lyu D, Cui L, Liu Q (2024) Analyzing the spatiotemporal dynamics and driving forces of ecological environment quality in the Qinling Mountains, China. Sustainability 16:3251.\u003c/li\u003e\n\u003cli\u003eLiu XW, Wu CS, Zhou QS, Zhu CD (2018) Effects of environmental change on wild bee diversity in China. Biodiversity Science 26:760 [in Chinese].\u003c/li\u003e\n\u003cli\u003eLiu YL, Qiu XK (2024) Ecological environmental problems and countermeasures in the Qinling region. Modern Horticulture 47:173\u0026ndash;175 [in Chinese].\u003c/li\u003e\n\u003cli\u003eLuo F, Lei CL (2003) Research progress on pollinating beetles. Chinese Bulletin of Entomology 40:313\u0026ndash;317 [in Chinese].\u003c/li\u003e\n\u003cli\u003eMcCabe LM, Cobb NS (2021) From bees to flies: global shift in pollinator communities along elevation gradients. Frontiers in Ecology and Evolution 8:626124.\u003c/li\u003e\n\u003cli\u003eMcCain CM, Garfinkel CF (2021) Climate change and elevational range shifts in insects. Current Opinion in Insect Science 47:111\u0026ndash;118. https://doi.org/10.1016/j.cois.2021.06.003\u003c/li\u003e\n\u003cli\u003eMeloni F, Civieta BF, Zaragoza JA, Moraza ML, Bautista S (2020) Vegetation pattern modulates ground arthropod diversity in semi-arid Mediterranean steppes. Insects 11:59. https://doi.org/10.3390/insects11010059\u003c/li\u003e\n\u003cli\u003eMiao B, Peng Y, Yang D, Kubota Y, Economo EP, Liu C (2021) Climate and land-use interactively shape butterfly diversity in tropical rainforest and savanna ecosystems of southwestern China. Insect Science 28:1109\u0026ndash;1120\u003c/li\u003e\n\u003cli\u003eMi\u0026ntilde;arro Prado M, Garc\u0026iacute;a Garc\u0026iacute;a D, Mart\u0026iacute;nez Sastre R (2018) Impact of insect pollinators in agriculture: importance and management of their biodiversity. Ecosistemas 27:81\u0026ndash;90. https://doi.org/10.7818/ecos.1394\u003c/li\u003e\n\u003cli\u003eMiyashita T, Hayashi S, Natsume K, Izumi Y, Sueyoshi M, Deto H, et al. (2023) Diverse flower-visiting responses among pollinators to multiple weather variables in buckwheat pollination. Scientific Reports 13:3099.\u003c/li\u003e\n\u003cli\u003eMuinde J, Katumo DM (2024) Beyond bees and butterflies: the role of beetles in pollination systems. Journal for Nature Conservation 77:126523.\u003c/li\u003e\n\u003cli\u003eM\u0026uuml;ller J, Hothorn T, Yuan Y, Seibold S, Mitesser O, Rothacher J, et al. (2024) Weather explains the decline and rise of insect biomass over 34 years. Nature 628:349\u0026ndash;354.\u003c/li\u003e\n\u003cli\u003eNealis VG (2020) Weather and insects in a changing climate. Journal of the Entomological Society of British Columbia.\u003c/li\u003e\n\u003cli\u003eOrford KA, Vaughan IP, Memmott J (2015) The forgotten flies: the importance of non-syrphid Diptera as pollinators. Proceedings of the Royal Society B 282:20142934.\u003c/li\u003e\n\u003cli\u003eOuthwaite CL, McCann P, Newbold T (2022) Agriculture and climate change are reshaping insect biodiversity worldwide. Nature 605:97\u0026ndash;102.\u003c/li\u003e\n\u003cli\u003eRatto F, Steward P, Sait SM, Pryke JS, Gaigher R, Samways MJ, et al. (2021) Proximity to natural habitat and flower plantings increases insect populations and pollination services in South African apple orchards. Journal of Applied Ecology 58:2540\u0026ndash;2551.\u003c/li\u003e\n\u003cli\u003eRiva F, Martin CJ, Gal\u0026aacute;n-Acedo C, Bellon EN, Keil P, Mor\u0026aacute;n-Ord\u0026oacute;\u0026ntilde;ez A, et al. (2024) Incorporating effects of habitat patches into species distribution models. Journal of Ecology 112:2162\u0026ndash;2182.\u003c/li\u003e\n\u003cli\u003eRyalls JM, Langford B, Mullinger NJ, Bromfield LM, Nemitz E, Pfrang C, et al. (2022) Anthropogenic air pollutants reduce insect-mediated pollination services. Environmental Pollution 297:118847.\u003c/li\u003e\n\u003cli\u003eScaven VL, Rafferty NE (2013) Physiological effects of climate warming on flowering plants and insect pollinators and potential consequences for their interactions. Current Zoology 59:418\u0026ndash;426. https://doi.org/10.1093/czoolo/59.3.418\u003c/li\u003e\n\u003cli\u003eSkendžić S, Zovko M, Živković IP, Le\u0026scaron;ić V, Lemić D (2021) The impact of climate change on agricultural insect pests. Insects 12:440. https://doi.org/10.3390/insects12050440\u003c/li\u003e\n\u003cli\u003eSridhar J, Kumar KK, Murali-Baskaran RK, Senthil-Nathan S, Sharma S, Nagesh M, et al. (2020) Impact of climate change on communities, response and migration of insects, nematodes, vectors and natural enemies in diverse ecosystems. In: Global climate change: resilient and smart agriculture. (pp. 69-93). Singapore: Springer Singapore. https://doi.org/10.1007/978-981-32-9856-9_4\u003c/li\u003e\n\u003cli\u003eUlyshen MD, Urban-Mead KR, Dorey JB, et al. (2023) Forests are critically important to global pollinator diversity and enhance pollination in adjacent crops. Biological Reviews 98:1118\u0026ndash;1141.\u003c/li\u003e\n\u003cli\u003eYang D (2018) Research progress on pollinating insects. Yunnan Agricultural Science and Technology A01:3\u0026ndash;7 [in Chinese].\u003c/li\u003e\n\u003cli\u003eYang J, Huang X (2025) The 30 m annual land cover datasets and its dynamics in China from 1985 to 2024 [Data set]. Earth System Science Data 13:3907\u0026ndash;3925. Zenodo. https://doi.org/10.5281/zenodo.15853565\u003c/li\u003e\n\u003cli\u003eYang ZY, Zhang H, Hao XJ, Li N (2024) Spatial pattern changes of ecological environment in the Qinling Mountains over the past 20 years based on remote sensing. Geomatics \u0026amp; Spatial Information Technology 47:14\u0026ndash;17 [in Chinese].\u003c/li\u003e\n\u003cli\u003eZhang L, Hu YY, Zhao YB, Che T (2025) ChinaMet: a high-resolution multi-factor meteorological forcing dataset for China based on multi-source data fusion. National Cryosphere Desert Data Center. https://cstr.cn/CSTR:11738.11.NCDC.NIEER.DB6722.2025 [in Chinese]\u003c/li\u003e\n\u003cli\u003eZhang LL, Cheng JW, Li AG, Yang YJ (2023) Effects of planting nectar plants on the diversity of natural enemy insects in paddy fields. Acta Agriculturae Zhejiangensis 35:1360\u0026ndash;1367 [in Chinese].\u003c/li\u003e\n\u003cli\u003eZhou LY, Ding SY, Lu XL, Liu YM (2020) Effects of human disturbance on pollinator diversity and niche of dominant groups. Acta Ecologica Sinica 40:2111\u0026ndash;2121 [in Chinese].\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003eTable 2 is available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"pollinators, community structure, environmental drivers, interaction effects, habitat gradients, biodiversity","lastPublishedDoi":"10.21203/rs.3.rs-9444014/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9444014/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study investigated pollinator community structure and its environmental drivers across different habitats in the central Qinling Mountains, a key biodiversity hotspot in China. Field surveys were conducted at approximately 40 sampling sites from March to September during 2024\u0026ndash;2025, covering multiple seasons, with concurrent measurements of climatic, topographic, and vegetation variables.\u003c/p\u003e \u003cp\u003eAlpha diversity patterns revealed strong habitat-dependent differences, with natural habitats supporting higher species richness and Shannon diversity, whereas agricultural habitats exhibited lower diversity but higher dominance. Semi-natural habitats showed intermediate characteristics. Taxonomic groups responded differently, indicating uneven sensitivity to habitat change. Beta diversity analyses further demonstrated significant community differentiation among habitats, with natural and agricultural habitats showing the greatest dissimilarity and semi-natural habitats acting as transitional systems.\u003c/p\u003e \u003cp\u003eEnvironmental analyses indicated that pollinator community structure and diversity were jointly shaped by landscape composition, climatic conditions, and plant resource availability. Forest cover was positively associated with diversity maintenance, whereas increasing agricultural proportion and wind disturbance generally exerted negative effects. Precipitation and humidity showed overall positive influences, but responses varied among taxonomic groups.\u003c/p\u003e \u003cp\u003eImportantly, interaction analyses revealed that environmental effects were strongly context-dependent. Landscape structure and floral resource availability significantly mediated climatic influences on community structure, diversity, and abundance. Forest cover tended to buffer climatic stress, while agricultural expansion amplified adverse environmental effects. These findings highlight that pollinator community responses are governed by the combined and interactive effects of climate, landscape, and resource gradients, emphasizing the importance of multi-factor mechanisms in shaping community dynamics.\u003c/p\u003e","manuscriptTitle":"Diversity Patterns of Pollinating Insect Communities in the Central Qinling Mountains and Their Responses to Environmental Change","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-27 15:41:11","doi":"10.21203/rs.3.rs-9444014/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"230f469c-2157-4a63-a3df-750ba51db4e5","owner":[],"postedDate":"April 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T15:41:12+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-27 15:41:11","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9444014","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9444014","identity":"rs-9444014","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}