Effects of agricultural land use on plant-pollinator interactions in peri- urban grasslands over 3 years

Effects of agricultural land use on plant-pollinator interactions in peri- urban grasslands over 3 years | 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 Article Effects of agricultural land use on plant-pollinator interactions in peri- urban grasslands over 3 years G Austrheim, R Straume, A Dhukuchhu, H Heggøy, S. H Eldøy, F Ødegaard This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8086473/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Land use is a major driver for plant-pollinator interactions in grasslands. This study examined abundance and taxa richness of flowering herbs and interacting insects in semi-natural (SN) and successional (SS) grasslands in a peri-urban landscape over 3 years. All insects found on flowers were recorded. Strict pollinating insects included bumblebees (Bombus), hoverflies (Syrphidae), wild bees and honeybees ( Apis mellifera ). We found higher abundances and richness of flowering herbs in SN vs SS grasslands all years. Pollinators had higher abundances at SN sites in 2 out of 3 years, mainly driven by hoverflies. In contrast to our expectations, we found no land use effect on bumblebees. However, bumblebee and strict pollinator abundance correlated with flower abundance, and a linear relationship between bumblebee and herb species diversity suggests that flower resources and diversity are good predictors of pollinator abundance and diversity. Analyses of plant-bumblebee network structure showed higher asymmetry in SS grasslands. SN grasslands scored higher on Shannon diversity and linkage density, which suggests that SS networks are more vulnerable to both plant and bumblebee species loss. Overall, the highest abundance and diversity of flowers and strict pollinating insects were found in SN grasslands, but many plant-pollinator parameters varied more within than among grassland type. Our study identified herb species favored by pollinators, which should be facilitated to increase SN grasslands with high quality flower resources to prevent pollinator declines. Biological sciences/Ecology Earth and environmental sciences/Ecology Biological sciences/Zoology Semi-natural grasslands successional grasslands herbs bumblebees hoverflies Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction A large proportion of grasslands are semi-natural; grass and herb dominated ecosystems induced and maintained by agricultural land use, such as animal husbandry and fodder production. The ecological value of these man-made habitats is strongly associated with low intensive land use, preventing secondary succession to the benefit of field layer plant diversity 1 , 2 , 3 . For pollinating insects, these herb rich semi-natural grasslands provide floral resources but also host plants and nesting sites for many species 4 , 5,6 . The biodiversity value for semi-natural grassland is thus found to be high for both plant and pollinator communities 7 , and most likely also for predators feeding on insects 8 , 9 , resulting in biodiversity hotspots across trophic levels. In parallel with a gradual increase in knowledge on the biodiversity and ecosystem services that semi-natural grasslands support, these low intensively used ecosystems are in decline mainly due to a change in land use. Two processes could be identified. First, agricultural intensification, such as ploughing frequency, fertilization and pesticide use to increase plant productivity, triggers loss of plants and pollinators and their network interactions 10 – 12 . Secondly, abandonment of marginal agricultural sites implies an ecosystem shift from grassland to forest but agricultural legacies such as fertilization often prevent the development of species rich ecosystems for any group of species independent of tree cover 2 , 13 . Although these processes and the resulting negative effects on plants and pollinating insects are known 14 , 15 , we have less knowledge on how different land use in grasslands affects the interactions between plants and pollinators 16 . In general, pollinator loss is driven by flower plant loss, and less so the other way around 17 . Hence, a high abundance of flowers is often found to promote high pollinator abundances. There is also some evidence that flowering plant diversity could positively impact pollinator diversity 18 , 19 suggesting that more diverse plant communities can attract specialized pollinators dependent on specialized floral characters. Historical data also show a parallel decline in plant and bee diversity in both the UK and the Netherlands 20 . Plant identity could thus be more important than plant diversity per se, especially species of high connectance, interacting with several other species. However, the overall size of the network (pool of both plants and pollinators) is expected to impact ecosystem properties such as stability. For example, diverse networks are found to be more resilient to disturbances and thus less vulnerable to plant or pollinator loss 21 . In many European countries, such as Norway, the abandonment of former infields in marginal areas for agricultural production has been increasing 22 . A significant proportion of these abandoned infields are found in peri-urban areas where former agricultural land is fragmented by housing and other infrastructure, leaving mainly grasslands as patches for secondary succession 23 , 24 . Nitrophilous species dominate the vegetation, reflecting the high levels of nutrients historically added to facilitate fodder production. Although these areas still resemble green lungs in a built environment, their values for biodiversity and ecosystem services are considered low. This includes plant diversity, pollinators, and other invertebrates typically associated with grassland ecosystems. However, our understanding of the biodiversity structure and function of these former intensively used agricultural grasslands is limited in comparison to more species rich semi-natural grasslands valued for nature conservation. This study examined the abundance and diversity of flowering herb plants and pollinating insects in peri-urban semi-natural (SN) and successional (SS) grasslands. The invertebrate community associated with grasslands is highly complex and includes a broad range of taxa from strict pollinators to more poor pollinating insects. We quantified all insects found on flowers. This included strict pollinators such as bumblebees (Bombus), wild bees, honeybees ( Apis mellifera ), hoverflies (Syrphidae) and butterflies (Lepidoptera), but also insects considered to be of less importance for pollination such as beetles (Coleoptera), other flies (Brachycera), wasps (Vespidae), and true bugs (Heteroptera). Secondly, we addressed the land use impact on interactions between these two trophic levels in terms of network size and structure. The study was conducted over three years to assess temporal treatment effects. Our two main expectations were, firstly, that plant resources in terms of overall flower abundance and their species richness are expected to follow habitat and species-specific phenological patterns throughout the growing season 17 . Peak pollinator abundance and taxonomic richness often coincide with the peak in flower resources. However, key drivers such as land use and weather influence the phenology of both plants and pollinators and the strengths of their interactions. We anticipate lower flower resources, pollinator abundance, and species richness in successional grasslands (abandoned agricultural infields) compared to semi-natural grasslands (nutrient-poor, less intensively managed habitats), regardless of season and year. Secondly, based on the assumption that an increase in the availability of plant resources will positively affect pollinator abundance and richness independent of land use and year, we expect more interactions and a higher stability (less variation) in plant resources, pollinator abundances and pollinator taxa richness across years in semi-natural (SN) grassland than in successional (SS) grassland. Moreover, since the plant-pollinator network can range from simple structures with few species of plants and pollinators to complex structures involving many species at both levels, regardless of the network size (number of interactions), we expect semi-natural grasslands to have more complex network structures in terms of diversity and connectance (higher proportion of possible interactions). 2. Results 3.1 Abundance and species richness (SR) of insects and flowering plants We found 38 taxa of insects of which 18 were identified to species level, and 20 to higher taxonomic levels (Table S3). Sixty-six flowering plant taxa included taxa at genus (9) and species (57) level (Table S4) across years, seasons and sites. Bumblebee species richness did not differ between grassland types (mean = 2 in SN and mean = 1 in SS across years, pe=-0.39, se 0.31, Z=-1.26, p = 0.20). In total, ten bumblebee species were found, and two were exclusive to the SN grasslands. For the flowering plants, 27 and 5 species were unique for the SN and SS grassland, respectively. Overall, the overlap in plant community composition between grasslands was low, while the insect community showed a large overlap (Fig. 2 a and b). Both abundance (pe=-0.43, se = 0.13, Z=-3.25, p = 0.001) and species richness (pe=-0.51, se = 0.17, Z=-3.02, P = 0.002) of flowering plants were higher in SN than SS grasslands (Fig. 3 ). Flower abundance peaked in both grassland types in 2020 (Mean 140 and 89 in SN and SS respectively) and were at the lowest in 2022 (mean 103 in SN and 23 in SS). Also, SR of flowering species peaked in 2020 (mean = 9 in SN and mean = 5 in SS) while 7 (SN) and 3 (SS) were the mean values in 2022 (Table S1 ). NMDS and environmental fitting revealed a strong effect of season on the community composition of flowering herbs (Fig. 2 a) (r2 = 0.465, p < 0.001), followed by grassland type (r2 = 0.116 p < 0.001), while year showed no effects (r2 = 0.023, p = 0.212). For the insect community (Fig. 2 b), we found variation among years (r2 = 0.252, p < 0.001) and season (r2 = 0.166, p < 0.001), while grassland type had no effect (r2 = 0.007, p = 0.428). Flowering peaked in early summer, and the abundances of both flowers (Fig. S3a) and insects (Fig. S3b) were at their lowest in spring. 3.2 Plant-pollinator interactions A total of 7369 interactions were identified across years, seasons and sites. Flies dominated (3446) before bumblebees (1552), honeybees (924), beetles (438), hoverflies (346), wasps (209), Heteroptera (133), while the remaining had < 100 interactions. For Bumblebees, we also recorded flying individuals (226 Bombus, most of them in SN sites) which were not further included in the analyses (for comparable values between all insect taxa for all years and grassland types but across sites, see Table. S3). Flower abundance had a positive effect on the abundance of strict pollinating species (Fig. 4 a, estimate 0.11, se = 0.05, DF = 23 t-value = 2.24, p = 0.035, marginal R 2 = 0.068, conditional R 2 = 0.227) and a marginal positive effect on bumblebee abundance (Fig. 4 b, estimate 0.07, se = 0.03, DF = 23, t-value = 2.06, p = 0.050, marginal R 2 = 0.087, conditional R 2 = 0.256). Flower diversity had a strong positive effect on bumblebee diversity (Fig. 4 c, estimate 0.72, se = 0.13, DF = 23, t-value 5.403, p < 0.001, marginal R 2 = 0.351, conditional R 2 = 0.407). The most important plant species for pollinators and bumblebees, especially across all years in SN grasslands, were Knautia arvensis, Hieracium sp., Hypericum maculatum, Trifolium pratense, and Campanula rotundifolia (Fig. S7a . In SS grasslands Epilobium angustifolium and Cirsium arvense were the most frequently visited species (Fig. S7b). 3.3 Network size and structure Number of interactions was examined across and within taxa of strict pollinating insects (bumblebees, hoverflies, wild bees and honeybees, Fig. 5 , Table 1 ) and insects of low importance for pollination (beetles, flies and wasps, Fig. S2, Table Sn). For “all pollinators”, the effect of land use varied over the years, with lower pollinator interactions in successional grasslands in both 2021 and 2022. Also for honeybees, the number of interactions in SS was lower in 2021 and in 2022, independent of land use. In hoverflies, interactions were higher in SN grasslands, independent of years. Only land use was retained in the best model for bumblebees and wild bees, but with no significant effects for either pollinator group. Table 1 Mixed effect models on the effects of land use, year and their interactions and additive effects on different groups of pollinating insects. The best model is presented with the AIC scores, and the other models as the change in AIC (Δ) as compared to the best model. Main effects from the best model is given as t-statistics. Abbreviations are Y = year, LSS = land use successional. Response variables # observations Trans. Landuse Year Interaction Additive All pollinators 144 Poisson Δ6.180 Δ68.597 3813.198 Y2021 7.525*** LSS:Y2021 -8.161*** LSS:Y2022 -8.229*** Δ 69.606 Hoverflies 144 Poisson 968.372 LSS -3.295 *** Δ288.679 Δ 276.416 Δ 282.373 Bumblebees 144 Poisson 2736.168 Δ 52.060 Δ11.990 Δ 53.523 Honeybees 144 Poisson Δ104.17115 Δ93.157 2162.251 Y2022 -4.643*** LSS:Y2021 -8.920*** Δ 95.135 Wild bees 144 Poisson 439.652 Δ22.631 Δ24.304 Δ 22.354 “Poor pollinators” had a higher number of interactions in both 2021 and 2022 compared to 2020, and interactions varied with land use in 2021 (higher in SS). These differences were probably driven by the high number of interacting flies both in 2021 and 2022 as compared to 2020 (Fig. S2c). Beetle and wasp abundance varied over the years, independent of land use, with especially high values in 2022 (Fig. S2a, S2b). All empirical networks differed significantly from the simulated null model networks. H2 was higher than null model networks, while Connectivity, Web asymmetry, Links per species, Nestedness, Linkage density, and Shannon diversity were all lower than null models. We examined land use effects on network size and structure for two subgroups of insects: (1) bumblebees (Fig. 6 , Table 2 ) and (2) strict pollinating insects (Fig. S5, Table S5). Bumblebees had a higher web asymmetry in SS grasslands. Shannon diversity of interactions (i.e. network entries) was lower in SS vs SN grasslands networks (pe = -0.59, se = 0.26, p = 0.026), while linkage density showed a tendency of higher values in SN vs SS grasslands (pe=-0.38, se = 0.23, z=-1.68, p = 0.091). For all strict pollinating insects which also included, wild bees, hoverflies, and honeybees, the network patterns very much corresponded with network indices for bumblebees alone for Shannon diversity (marginally not significantly higher in SN pe = 0.44, se = 0.23, Z=-1.89, p = 0.057) and linkage density (marginal higher number of interactions in SN vs SS grasslands, pe=-0.35, se = 0.21, Z=-1.65, p = 0.097). Web asymmetry for all strict pollinators was more driven by differences among years as compared to bumblebees alone, but the proportion of pollinator taxa vs plant taxa was still higher in SS vs SN grasslands. Table 2 Mixed effect models on the effects of land use, year and their interactions and additive effects on different plant-bumblebee network indices. The best model is presented with the AIC scores, and the other models as the change in AIC (Δ) as compared to the best model. Main effects from the best model is given as t-statistics. Abbreviations are Y = year, LSS = land use successional. The network index H2 had a model convergence problem and is not shown. Note that linkage density had marginally lower values in SN grasslands (Z= -1.688, p = 0.0914). Response variables # observations Trans. Landuse Year Interaction Additive Shannon's diversity 36 Poisson LSS 100.315 -2.217* Δ 3.932 Δ 4.366 Δ 0.797 connectance 36 Poisson 59.033 Δ 1.349 Δ 5.205 Δ 1.233 web_asymmetry 36 Gaussian -13.131 LSS 3.733*** Δ -5.999 -13.612 -14.250 linkage_density 36 Poisson 109.840 Δ 4.230 Δ 4.230 Δ 1.139 links_per_species 36 Poisson 78.656 Δ 2.669 Δ7.530 Δ1.740 weighted_NODF 36 Poisson Δ 80.006 Δ76.941 460.110 -4.643*** LSS:Y2022 -7.632*** Δ76.941 3. Discussion Semi-natural grasslands are expected to be key habitats for pollinating insects due to high abundances and richness of flowering herb species, but intensive land uses are often found to decrease flower food resources with negative effects for pollinators 14 . This study compared plant-pollinator interactions in semi-natural (SN) versus former intensively used agricultural land which could be classified as successional (SS) grassland over three years. We found higher abundance and richness of flowering herb species in SN than in the SS grasslands. In contrast to our expectations, the impact of grassland type on insects was less clear. Strict pollinating insects were more abundant in SN in 2 out of three years, while hoverflies were more abundant in SN independent of years. We did not find higher abundances or species richness of bumblebees in SN grasslands. However, the higher abundance of flower resources correlated with the abundances of bumblebees and pollinators as expected. Also, bumblebee species richness was strongly correlated with flowering plant diversity, suggesting that the abundance and species richness of flowers are better predictors for pollinator abundance and diversity than grassland type. We found clear differences in network size and structure for bumblebees with higher diversity of interactions (i.e. number of network entries) and a tendency of higher linkage density for both bumblebees and strict pollinators in SN vs SS grasslands. In contrast, SS grasslands had a higher proportion of bumblebees and pollinators in relation to plants (web asymmetry) than the SN grasslands. Although the evidences for loss of insect species richness and abundance are strong at a global level 14 as documented by an increasing number of local reports also from Norway 25 , the understanding of how different drivers effects these changes, and the relative importance of specific factors such as land use, climate, resources extraction are still limited 14 . Our study shows that semi-natural grassland is a heterogenous ecosystem with large variations in plant and insect community structure despite a common low intensive land use. Indeed, land use is a complex driver with varying impacts such as grazing, fodder harvest, fuel-wood cutting, use of fertilizers, and the temporal scale and legacies of these activities are difficult to quantify and disentangle 10 . Nevertheless, such differences in land use are probably an important reason for the large variation in both flower resources and insects among our SN sites. The clear difference in plant community composition between grassland types is characterised by the dominance of tall nitrophilous plants in former agricultural SS grasslands, while SN had a higher proportion of lower, less competitive herbs. Some of the tall herbs like Epilobium angustifolium and Cirsium arvense offer a high abundance of pollinator food and show a high abundance of interactions, but probably for a shorter period compared to a plant community with a higher diversity and more even distribution across the growing season. Plant identity could also be more important than plant diversity per se, which underlines the need to identify attractive plant species that could be used to facilitate networks in grassland restoration initiatives. We show that several pollinator favored plants are low abundant species e.g. Knautia arvensis , Trifolium pratense , Hypericum maculatum , and Campanula rotundifolia . None of these are rare in terms of red-listed but are restricted to areas with low-productivity agricultural practices of the past and absent in the dominating successional grasslands. The less clear differences in insect community composition between grassland types indicate a difference in overall community structure between the two trophic levels and more among year variation for insects. However, the lack of species level data for insects as compared to plants may have masked community differences in our study. The low abundance of butterflies across all years and seasons was surprising as SN grasslands are expected to be the main butterfly habitat in these peri-urban landscapes. Their rarity is confirmed by regional monitoring over the last decade, but lack of long-term monitoring makes it difficult to conclude on long-term trends. However, decreases in butterflies are shown in comparable habitats elsewhere in Europe 26 . Annual differences in external environmental conditions as well as internal life histories are expected to strongly impact plant properties such as flowering, as well as insect abundances and richness, with consequences for interaction networks 27 . This three-year study showed clear annual differences for several taxa of poor pollination importance such as flies and beetles. Also hoverflies varied among years with a low number of interactions the first year 28 , 29 possibly due to a dry summer which is known to limit recruitment of insects in general 30 , 31 . Little is known about the longer-term changes in plant-pollinator interactions. A regional study on bumblebee and butterfly showed no negative trends in abundances at the county level (Trøndelag) from 2012 to 2020 32,33 . In contrast, a national survey showed decreased insect biomass at the regional level (Trøndelag) over 5 years (2020–2024) 34 . A challenge for both studies is the lack of good reference values, but abundance values for both groups were considered to be lower than expected, possibly due to legacies of low recruitment in the past 35 . Although we found limited effects of grassland land use on plant pollinator network size, the network structure differed in three ways. First, the asymmetry values revealed a lower proportion of insects relative to flowering plants in the SN grasslands, which suggests that SS grasslands with lower flower abundance is still able to support an insect abundance not significantly lower than in the SN grasslands. A possible explanation could be a larger area of many of the SS grasslands. Indeed, SN grasslands had limited areas which could restrict food availability and insect population densities. Secondly, the higher number of network entries indicates that each plant and insect taxa interacted with more species in SN vs SS grasslands. Thirdly, we found a tendency for higher linkage density in SN grasslands, indicating that these grasslands rich in plant species supported larger networks with higher linkage densities 36 i.e. losing a partner in the network could make a species less vulnerable if it’s linked to many partners and the grassland more resilient to environmental fluctuations 37 . 4. Conclusions We found higher abundances and species richness of flowering herbs in SN as compared to SS grasslands. However, the land use impact on interacting insects was limited to higher abundances of hoverflies in SN and for pollinating insects in general for 2021 and 2022. A positive linear correlation between flower resources and pollinator abundance still supports the assumption that low food resources are a limiting factor for interacting pollinating insects. The high correlation between plant species richness and bumblebee diversity also suggests that SN grasslands have the potential to support a larger species pool of bumblebees if more low intensive grassland habitat is made available. Several studies have shown positive effects of restoring former intensively used agricultural habitats into low-productive SN grasslands 10 , 38 . An example is biomass and nutrient removal to increase flower resources, insect abundances and diverse networks, as shown for SN grasslands 10 . Our work will inform management of the plants and insect communities associated with key peri-urban grassland habitats on which plant species are the most important for strict pollinating insects such as bumblebees, hoverflies, wild bees, honeybees, and butterflies. Most of the key plant resources for pollinators identified in this study are common species expected to establish if SS grasslands currently out of agricultural use, could be restored to more low productive biodiversity friendly habitats. 5. Material and methods 5.1 Study design All study sites (n = 12) are found in Trondheim municipality, Trøndelag, Norway (63°26′24′′N 10°24′0′′E) (Fig. 1 , Table S1 ). A “wall-to-wall” survey of land cover and land use was done at a large scale prior to the selection of the grassland study sites 39 within the peri-urban landscape of the city of Trondheim using the classification from AR5, a national product which classifies areas based on land-use and primary environmental conditions (NIBIO 2025). Two different grassland habitats were selected. Semi-natural grasslands (SN, n = 6) had a low intensive land use with limited or no addition of fertilizer. Successional grasslands (SS, n = 6) were former infields with a legacy of winter-fodder production and use of fertilizer. SN grasslands were characterized by low productivity and moderate plant height (mean height 50.51 ± 2.2), while SS grasslands had a high productivity with tall biomass (mean height 119.19 ± 2.05) as measured in July 2020 28 . This pilot study 39 also documented the rarity of SN grasslands (0.4 km 2 , 0.7%) as well as a higher proportion of former infield (0.6 km 2 , 1.1%) with no further agricultural use and thus left for secondary succession. We visited each study site four times (in random order) during the summer season over three years (2020, 2021, 2022) to account for seasonal differences: (1) spring (late May, early June), (2) early summer (late June), (3) mid summer (late July) and (4) late summer (mid August). Plant flower abundance, species richness of flowering plants and insect abundance and taxa richness were recorded within a 250 m 2 area at the site level. For plants, we quantified flowering herbs within a 0.5 x 0.5 quadrat divided into 16 subquadrats. Presence-absence of flowering herbs within the subquadrats gave a maximum value of 16 per species at a quadrat level if all subplots had a flower of the same species independent of the number of flowers within each subplot. Species richness of flowering species was also recorded at the quadrat level as the total number of flowering species found within the plot. A total of 25 quadrats for examining flowering plants were placed systematically within the 250 m 2 area at each site. When quantifying insects at the site level, the same study area was examined by walking in the middle of 5 m broad transects (5 x 50 m = 250 m 2 ) 32 . All insects sitting on herb flowers were recorded 2.5 m on each side of the transect while walking at a slow pace. A sweep net was used to collect insects for identification. Only bumblebees (Bombus) and butterflies (Lepidoptera) were identified to the species level. For bumblebees, only males and workers were taken to the lab for species identification if needed. The naming of bumblebees follows Ødegaard 40 . All other insects were identified at the genus or family level. A complete list of insect and plant taxa included is given in Tables S3 and S4, respectively. Insect sampling was only done on sunny days, avoiding cold mornings and with an ambient temperature > 15°C. Days with strong winds were also avoided (< 15 km/h). Coordinates of both starting and ending points were recorded with the help of a GPS (Global Positioning System) device to have a reference for later sampling. Six different people were involved in plant and insect sampling across the years. To minimize personal errors and calibrate sampling procedures, 2 recorders were present at the start of the sampling in all years. All invertebrates along the transect were quantified and grouped at different taxonomic levels. For analysis on plant-pollinator networks, we included bumblebees, which were identified at a species level except for Bombus s. str . Subgroups of pollinating insects, including bumblebees (Bombus), butterflies (Lepidoptera parts, species level), solitary bees (grouped), hoverflies (Syrphidae, grouped) and the honeybee ( Apis mellifera) were also identified for network-level analyses. Invertebrates considered to be of low importance for pollination, such as other flies (Brachycera), beetles (Coleoptera), ladybugs (Coccinellidae), ants (Formicidae), moths (Lepidoptera parts), true bugs (Heteroptera) and stinging wasps (Vespidae) were grouped as poor pollinators. 5.2 Statistical analyses Data processing and analyses were done using RStudio (version 4.3.1, (Team 2024) R Core Team 2024). We used generalised linear mixed models (GLMM) (library glmmTMB) to test the effect of four candidate models: treatment (semi-natural vs successional grassland), year and their potential interactions (multiplicative, treatment*year) and additive effects (treatment + year). The responses included (1) herb flower abundance and species richness, (2) bumblebee abundance and species richness (3) abundance for all taxa of insects found at flowers, and (4) network level properties on size and structure for a) Plant-Bumblebee, b) Plant-strict pollinating insects’ networks. Site (six sites per treatment) was included as a random variable. In order to manage zero-inflation and problems with non-normal variation in the model residuals, plant and pollinator data were aggregated across the four seasons except for descriptive models on seasonal variations. To account for the non-normal distribution for count variables, we used Poisson distribution for all modelling except for negative responses (web asymmetry) for which we used a Gaussian distribution. AIC optimization was used to select the best model. All diagnostics of the selected model were assessed by checking the distribution of the residuals using the diagnostic tools. Non-metric multidimensional scaling (NMDS) followed by environmental fitting by the function envfit with 999 permutations was used for the ordination of plant and insect community data using Vegan 41 . All plots were made by the library ggplot 42 , except for the figures on plant-pollinator networks made by using Bipartite 43 . The Bipartite package 43 was also used to calculate plant-pollinator network level properties on size and structure for 1) strict pollinating species, 2) bumblebee species. The following indices were included: (1) Mean number of links per species (qualitative): sum of links divided by number of species. (2) Linkage density: Marginal totals-weighted diversity of interactions per species (quantitative). This is computed as the average of vulnerability and generality 44 . (3) Shannon's diversity of interactions (i.e. network entries). (4) Web asymmetry: Balance between numbers in the two levels: positive values indicate higher-trophic level species, negative lower-trophic level species; implemented as (ncol(web)-nrow(web))/sum(dim(web)); web asymmetry is a null model for what one might expect in dependence asymmetry 45 . (5) Connectance: Realised proportion of possible links 46 : sum of links divided by number of cells in the matrix (= number of higher times number of lower trophic level species). This is the standardised number of species combinations often used in co-occurrence analyses 47 (6) Specialisation (H2). (7) Nestedness (NODF). Networks and associated properties were constructed for each site every year (12 sites x 3 years = 36 networks). To check for possible spurious relationships in our empirical networks between plants and pollinators, we carried out Z-score calculations 43 based on 1000 random null model interactions for each of the seven network properties. A t-test was used to assess the difference between the empirical model and the null model. When modelling treatment and year effects on network properties, network size was included as a random variable to correct for differences in size. However, the inclusion of size did not improve any model as assessed by AIC. Declarations Funding This work was supported by the Research Council of Norway, Project No. 319892. Author Contribution GA: Conceptualization, Formal analysis, Methodology, Funding acquisition, Supervision, Writing. RS: Investigation, Data curation, Writing. AD: Investigation, Data curation, Writing. HH: Investigation, Data curation, Writing. SHE: Formal analysis, Writing. FØ: Conceptualization, Methodology, Supervision, Writing. Acknowledgement AcknowledgementsThis work was supported by the Research Council of Norway, Project No. 319892. We thank Aksel J. Fosse for help with the fieldwork. Data Availability Summary tables on plant and pollinator data are available in the supplementary (Table S2 and S3). Complete datasheets are available through figshare.com: [https://doi.org/10.6084/m9.figshare.30589385.v1](https:/doi.org/10.6084/m9.figshare.30589385.v1) References Le Féon, V. et al. Intensification of agriculture, landscape composition and wild bee communities: A large scale study in four European countries. Agriculture, Ecosystems & Environment 137 , 143–150 (2010). https://doi.org/https://doi.org/10.1016/j.agee.2010.01.015 Winberg, J., Ekroos, J. & Smith, H. G. Abandonment or biomass production? Phytodiversity responses to land-use changes of semi-natural grasslands in northern Europe. Biological Conservation 294 , 110632 (2024). https://doi.org/https://doi.org/10.1016/j.biocon.2024.110632 Hirayama, G. S., Tomita, M. & Ushimaru, A. Impacts of agricultural intensification and abandonment on plant and pollinator diversity, their mutualistic networks, and pollination services. Agriculture, Ecosystems & Environment 381 , 109437 (2025). https://doi.org/https://doi.org/10.1016/j.agee.2024.109437 ÖCKINGER, E. & SMITH, H. G. Semi-natural grasslands as population sources for pollinating insects in agricultural landscapes. Journal of Applied Ecology 44 , 50–59 (2007). https://doi.org/https://doi.org/10.1111/j.1365-2664.2006.01250.x Goulson, D., Lye, G. C. & Darvill, B. Decline and Conservation of Bumble Bees. Annual Review of Entomology 53 , 191–208 (2008). https://doi.org/https://doi.org/10.1146/annurev.ento.53.103106.093454 Larkin, M. & Stanley, D. A. Impacts of management at a local and landscape scale on pollinators in semi-natural grasslands. Journal of Applied Ecology 58 , 2505–2514 (2021). https://doi.org/https://doi.org/10.1111/1365-2664.13990 Bardgett, R. D. et al. Combatting global grassland degradation. Nature Reviews Earth & Environment 2 , 720–735 (2021). https://doi.org/10.1038/s43017-021-00207-2 Bowler, D. E., Heldbjerg, H., Fox, A. D., de Jong, M. & Böhning-Gaese, K. Long-term declines of European insectivorous bird populations and potential causes. Conservation Biology 33 , 1120–1130 (2019). https://doi.org/https://doi.org/10.1111/cobi.13307 Seress, G., Sándor, K., Evans, K. L. & Liker, A. Food availability limits avian reproduction in the city: An experimental study on great tits Parus major. Journal of Animal Ecology 89 , 1570–1580 (2020). https://doi.org/https://doi.org/10.1111/1365-2656.13211 Carvalheiro, L. G. et al. Species richness declines and biotic homogenisation have slowed down for NW-European pollinators and plants. Ecology Letters 16 , 870–878 (2013). https://doi.org/https://doi.org/10.1111/ele.12121 Vanbergen, A. J. & Initiative, t. I. P. Threats to an ecosystem service: pressures on pollinators. Frontiers in Ecology and the Environment 11 , 251–259 (2013). https://doi.org/https://doi.org/10.1890/120126 Deguines, N. et al. Large-scale trade-off between agricultural intensification and crop pollination services. Frontiers in Ecology and the Environment 12 , 212–217 (2014). https://doi.org/https://doi.org/10.1890/130054 Habas, J. F. et al. Long-term effects of nutrient addition on yield, diversity, and nutritional quality in a Mediterranean grassland under variable rainfall. Agriculture, Ecosystems & Environment 393 , 109811 (2025). https://doi.org/https://doi.org/10.1016/j.agee.2025.109811 IPBES. Report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services on Pollinators, Pollination and Food Production. . (Secretariat of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, Bonn, Germany., 2016). Raven, P. H. & Wagner, D. L. Agricultural intensification and climate change are rapidly decreasing insect biodiversity. Proceedings of the National Academy of Sciences 118 , e2002548117 (2021). https://doi.org/doi:10.1073/pnas.2002548117 Vanbergen, A. J. Landscape alteration and habitat modification: impacts on plant–pollinator systems. Current Opinion in Insect Science 5 , 44–49 (2014). https://doi.org/https://doi.org/10.1016/j.cois.2014.09.004 Willmer, P. Pollination and Floral Ecology. , (Princeton University Press, 2011). Ebeling, A., Klein, A.-M., Schumacher, J., Weisser, W. W. & Tscharntke, T. How does plant richness affect pollinator richness and temporal stability of flower visits? Oikos 117 , 1808–1815 (2008). https://doi.org/https://doi.org/10.1111/j.1600-0706.2008.16819.x Baude, M. et al. Historical nectar assessment reveals the fall and rise of floral resources in Britain. Nature 530 , 85–88 (2016). https://doi.org/10.1038/nature16532 Biesmeijer, J. C. et al. Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313 , 351–354 (2006). https://doi.org/10.1126/science.1127863 Redhead, J. W. et al. Potential landscape-scale pollinator networks across Great Britain: structure, stability and influence of agricultural land cover. Ecology Letters 21 , 1821–1832 (2018). https://doi.org/https://doi.org/10.1111/ele.13157 Terres, J.-M. et al. Farmland abandonment in Europe: Identification of drivers and indicators, and development of a composite indicator of risk. Land Use Policy 49 , 20–34 (2015). https://doi.org/https://doi.org/10.1016/j.landusepol.2015.06.009 Geslin, B., Gauzens, B., Thébault, E. & Dajoz, I. Plant Pollinator Networks along a Gradient of Urbanisation. PLOS ONE 8 , e63421 (2013). https://doi.org/10.1371/journal.pone.0063421 Petersen, T. K., Speed, J. D. M., Grøtan, V. & Austrheim, G. Urban aliens and threatened near-naturals: Land-cover affects the species richness of alien- and threatened species in an urban-rural setting. Scientific Reports 10 , 8513 (2020). https://doi.org/10.1038/s41598-020-65459-2 Burner, R. C., Selås, V., Kobro, S., Jacobsen, R. M. & Sverdrup-Thygeson, A. Moth species richness and diversity decline in a 30-year time series in Norway, irrespective of species’ latitudinal range extent and habitat. Journal of Insect Conservation 25 , 887–896 (2021). https://doi.org/10.1007/s10841-021-00353-4 Van Swaay, C. A. M., 2, Dennis, E.B.3, Schmucki, R.4, Sevilleja, C.G.1,2, Åström, S.6, Balalaikins, M.7, Barea-Azcón, J.M. 5, Bonelli, S.8, Botham, M.4, Cancela, J.P.51, Collins, S.1, De Flores, M.18, Dapporto, L.44, Dopagne, C.12, Dziekanska, I.13, Escobés, R.14, Faltynek Fric, Z.30, Fernández-García, J.M.16, Fontaine, B.17, Glogovčan, P.46, Gracianteparaluceta, A.16, Harpke, A.15, 50, Harrower, C.4, Heliölä, J.19, Houard, X., 18, Judge, M.25, Kolev, Z.20, Komac, B.21, Kühn, E.15, Kuussaari, M.19, Lang, A.23, Lysaght, L.25, Maes, D.26, McGowan,D.11, Mestdagh, X.27, Middlebrook, I.3, Monasterio, Y.14, Monteiro, E.28, Munguira, M.L.9,1, Musche, M.15, Olivares, F.J.24, Õunap, E.29, Ozden, O.30, Pavlíčko, A.32, Pendl, M.22, Pettersson, L.B.33, Rákosy, L.36, Roth, T.37, Rüdisser, J.38, Šašić, M.39, Scalercio, S.47, Settele, J.15, 48, 49, Sielezniew, M.40, Sobczyk-Moran, G.18, Stefanescu, C.31, Švitra, G.42, Szabadfalvi, A.43, Tiitsaar, A.29, Titeux, N.27, Tzirkalli, E.45, Ubach, A.31, Verovnik, R.10, Vray, S.27, Warren, M.S.3, Wynhoff, I.1,2 & Roy, D.B. European Grassland Butterfly Indicator 1990-2020 (2022). Fisogni, A. et al. Seasonal trajectories of plant-pollinator interaction networks differ following phenological mismatches along an urbanization gradient. Landscape and Urban Planning 226 , 104512 (2022). https://doi.org/https://doi.org/10.1016/j.landurbplan.2022.104512 Dhukuchhu, A. Effects of land use and season on flowering plants and plant-pollinator networks in urban grasslands Master’s thesis in Biology thesis, Norwegian University of Science and Technology, (2021). Heggøy, H. Effects of land-use on bumblebees and other pollinators in urban grasslands Master thesis, Norwegian University of Science and Technology, (2021). Huberty, A. F. & Denno, R. F. PLANT WATER STRESS AND ITS CONSEQUENCES FOR HERBIVOROUS INSECTS: A NEW SYNTHESIS. Ecology 85 , 1383–1398 (2004). https://doi.org/https://doi.org/10.1890/03-0352 Gely, C., Laurance, S. G. W. & Stork, N. E. How do herbivorous insects respond to drought stress in trees? Biological Reviews 95 , 434–448 (2020). https://doi.org/https://doi.org/10.1111/brv.12571 Öberg, S., Gjershaug, J.O., Certain, G. & Ødegaard, F. Development of methodology for area representative monitoring of chosen invertebrate groups. . 50 (NINA, 2010). Straume, R. Spatiotemporal variation and flowering plant preferences of insect pollinators in grassland ecosystems , Norwegian University of Science and Technology, (2022). Åström, J., Gohli, J., Birkemoe, T., Brealey, J., Dahle, S., Davey, M., Ekrem, T., Fossøy, F., Laugsand, A., Opsahl, N., A., Solvang Klokke, V., Staverløkk, A. & Sverdrup‐Thygeson, A. . Norsk insektovervåking. Rapport fra feltsesongen 2024., (Norwegian Insitute for Nature Research, Trondheim, 2025). Dupont, Y. L., Damgaard, C. & Simonsen, V. Quantitative Historical Change in Bumblebee (Bombus spp.) Assemblages of Red Clover Fields. PLOS ONE 6 , e25172 (2011). https://doi.org/10.1371/journal.pone.0025172 Larkin, M. & Stanley, D. A. Impacts of local and landscape grassland management on the structure of plant-pollinator networks. Basic and Applied Ecology 70 , 50–59 (2023). https://doi.org/https://doi.org/10.1016/j.baae.2023.04.003 Proesmans, W. et al. Urbanisation and agricultural intensification modulate plant–pollinator network structure and robustness. Functional Ecology 38 , 628–641 (2024). https://doi.org/https://doi.org/10.1111/1365-2435.14503 Sexton, A. N. & Emery, S. M. Grassland restorations improve pollinator communities: a meta-analysis. Journal of Insect Conservation 24 , 719–726 (2020). https://doi.org/10.1007/s10841-020-00247-x Kleppe, K. B. Grasslands in Trondheim; Land-use, plant species richness and soil characteristics Master’s thesis in biology thesis, Norwegian University of Science and Technology, (2019). Ødegaard, F., Staverløkk, A., Gjershaug, J.O., Bengtson, R. & Mjelde, A. Bumblebees in Norway: Characteristics, distrubution and life history. , (NINA, 2015). vegan: Community Ecology Package v. R package version 2.8-0 (2025). Wickham, H. Ggplot2: Elegant Graphics for Data Analysis. 2nd Edition edn, (Springer 2009). Bipartite vignette (github, 2019). Bersier, L.-F., Banašek-Richter, C. & Cattin, M.-F. Quantitative Descriptors of Food-Web Matrices. Ecology 83 , 2394–2407 (2002). https://doi.org/10.2307/3071801 Blüthgen, N., Menzel, F., Hovestadt, T., Fiala, B. & Blüthgen, N. Specialization, Constraints, and Conflicting Interests in Mutualistic Networks. Current Biology 17 , 341–346 (2007). https://doi.org/https://doi.org/10.1016/j.cub.2006.12.039 Dunne, J. A., Williams, R. J. & Martinez, N. D. Food-web structure and network theory: The role of connectance and size. Proceedings of the National Academy of Sciences 99 , 12917–12922 (2002). https://doi.org/doi:10.1073/pnas.192407699 Gotelli, N. J. a. G. R. G. Null Models in Ecology. , (Smithsonian Institution Press, , 1996). Additional Declarations No competing interests reported. Supplementary Files SupplementarySRAustrheim.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 19 May, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviews received at journal 20 Mar, 2026 Reviewers agreed at journal 23 Feb, 2026 Reviewers invited by journal 01 Dec, 2025 Editor assigned by journal 25 Nov, 2025 Editor invited by journal 20 Nov, 2025 Submission checks completed at journal 14 Nov, 2025 First submitted to journal 14 Nov, 2025 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-8086473","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":554160916,"identity":"befd3135-69b9-4446-82a5-5fd9528d4416","order_by":0,"name":"G Austrheim","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYLADZgYGCQZ+kjSAtUg2MJOmhYHB4AABLfIRucce/NzDIG9w/vyxzwUVFtHGt/uPPeZhuCOHS4vhjbx0w55nDIYbbiQzz55xRiJ3253D7MY8DM+McWqZkWMmwXOAgXHDDWZmZt42oJYbyWzSPAyHExvwaJH8c4DBfsP5w0At/yRyN88goEVeIsdMGmhL4oYDyUAtDRK5GyQIaDHgeWMmLXNAInnmjWRjZp5jErkzbiSbSc4xOIzTL/LtQIe9OWBj23f+4GNmnpq63P4Zic8k3lQcxhliBgfAlASGOC4NQFtwuXgUjIJRMApGARwAAACoT69dyKjKAAAAAElFTkSuQmCC","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"G","middleName":"","lastName":"Austrheim","suffix":""},{"id":554160917,"identity":"ef214874-e3bb-4d7f-9688-10a35cdd4437","order_by":1,"name":"R Straume","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"R","middleName":"","lastName":"Straume","suffix":""},{"id":554160918,"identity":"c6d47e98-f0f8-4ab9-adc7-b179d83b2259","order_by":2,"name":"A Dhukuchhu","email":"","orcid":"","institution":"North Dakota State University","correspondingAuthor":false,"prefix":"","firstName":"A","middleName":"","lastName":"Dhukuchhu","suffix":""},{"id":554160919,"identity":"6489b114-cca8-4baf-9fff-8d3de34a8586","order_by":3,"name":"H Heggøy","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"H","middleName":"","lastName":"Heggøy","suffix":""},{"id":554160920,"identity":"e9437761-1d7b-4acd-8c41-7712aa5d371c","order_by":4,"name":"S. H Eldøy","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"S.","middleName":"H","lastName":"Eldøy","suffix":""},{"id":554160921,"identity":"6497fb61-6050-4eb1-9fc0-a3c8a2dec4ea","order_by":5,"name":"F Ødegaard","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"F","middleName":"","lastName":"Ødegaard","suffix":""}],"badges":[],"createdAt":"2025-11-11 11:53:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8086473/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8086473/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97358755,"identity":"baf1c5ea-67f7-428e-8ee4-f79e416be5ac","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":761031,"visible":true,"origin":"","legend":"","description":"","filename":"MSSRAustrheim1111.docx","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/771ddc825f3801c3af8de35d.docx"},{"id":97370598,"identity":"b7055678-aec2-4020-9868-56d0ae5b367a","added_by":"auto","created_at":"2025-12-03 16:27:39","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7726,"visible":true,"origin":"","legend":"","description":"","filename":"c2288217dec44a31bf914e84683185c4.json","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/c1f7d86bb0e55073b9fcd3ff.json"},{"id":97358761,"identity":"caac9289-1722-4059-9535-f3787c9f92d5","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":768579,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarySRAustrheim.docx","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/d3c27813d87b85a5147350a2.docx"},{"id":97358765,"identity":"65cef791-cda8-42ff-b418-a93b5f3e3bcf","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":121188,"visible":true,"origin":"","legend":"","description":"","filename":"c2288217dec44a31bf914e84683185c41enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/2c0dd4f865ffe756a4eff5c9.xml"},{"id":97370959,"identity":"848be6d6-ea64-4970-a015-8bddc22d7b45","added_by":"auto","created_at":"2025-12-03 16:28:11","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":168613,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/f525d04c430c608f81322b37.jpeg"},{"id":97358762,"identity":"fab047dd-8762-4fa5-9da6-fd1dd4ccba3a","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":461655,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/19f185e7bddec784137d0058.jpeg"},{"id":97358768,"identity":"3ed59706-edd8-42ad-9634-c341ba9281af","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146150,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/5a339af64c17dd6916d8d451.jpeg"},{"id":97371069,"identity":"059d654c-2918-475a-80f9-6c290feed49f","added_by":"auto","created_at":"2025-12-03 16:28:23","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":163117,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/5ed8f8c7cab47159f2a5f480.jpeg"},{"id":97371144,"identity":"609aacc7-b840-4c85-9f85-5849a6a36f9b","added_by":"auto","created_at":"2025-12-03 16:28:26","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118538,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/3dd57a2aaf9292ad58abade4.png"},{"id":97370917,"identity":"5ce79f28-8bee-4d1d-a64d-3d072b0561f3","added_by":"auto","created_at":"2025-12-03 16:28:09","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56436,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/f73cb914d51028b1bda426e2.png"},{"id":97358772,"identity":"76b9a933-cc89-4dc9-819a-cb9b4dd12b6e","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50626,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/0b03a270cca60ec648cb5d1b.png"},{"id":97358773,"identity":"8a1f244b-f37a-4680-9f9c-f5eb1486d806","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96989,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/54b152a25bcc1837eafab608.png"},{"id":97358780,"identity":"30b8b603-6678-4089-80e5-ed9809e72e36","added_by":"auto","created_at":"2025-12-03 14:10:17","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25484,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/ca5da8d2aee99242a398c29a.png"},{"id":97371499,"identity":"775e0827-5c7e-4544-a4ba-5da1f6fbfb8b","added_by":"auto","created_at":"2025-12-03 16:29:03","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35306,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/891af54c4e143343daa61988.png"},{"id":97370763,"identity":"e36184b4-a01a-4f56-bbdc-8727b3d22c8e","added_by":"auto","created_at":"2025-12-03 16:27:54","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23960,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/795d730a1180f65a13799956.png"},{"id":97358777,"identity":"9074a093-1f06-4bad-b331-a14460bf436d","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17638,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/4f491662d3d2acda0d2e911a.png"},{"id":97358774,"identity":"39c2533a-b77b-41f6-b085-170a13ec5251","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118963,"visible":true,"origin":"","legend":"","description":"","filename":"c2288217dec44a31bf914e84683185c41structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/e74f26fd60e64172bdaef581.xml"},{"id":97358779,"identity":"67b4d3e9-7485-49c8-afdd-5aad87d4e713","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":130868,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/0d4d71b2aa67f9eb9166da07.html"},{"id":97358754,"identity":"1c1d3315-77de-487c-a0d7-9a41b42daaf1","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":168613,"visible":true,"origin":"","legend":"\u003cp\u003eMap of the study sites with the delimitation showing the border for selected urban areas in Trondheim municipality. Semi-natural (SN) sites are (1) Bjørndalen, (2) Flatåsen, (3) Grønlia, (4) Lade, (5) Lian Upper, (6) Lian Lower. Successional sites (SS) are (7) Buengveien N, (8) Buengveien S, (9) Forsøkslia, (10) Okstad, (11) Selsbakk N, (12) Selsbakk S.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/aa228b247b8a97175911b866.jpeg"},{"id":97358756,"identity":"a96f3d6d-19fa-4584-87a4-60cd23b614ca","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":461655,"visible":true,"origin":"","legend":"\u003cp\u003eNMDS-plot on (a) plant community and (b) insect community composition in SN and SS grasslands. Only plant species names of the ten most common plant and insect taxa in SN and SS habitats are shown. Year (2020, 2021, 2022) and season (1) spring, (2) early summer, (3) midsummer, (4) late summer) are shown as vectors.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/e809c75cdf64b5d328ecccb9.jpeg"},{"id":97358757,"identity":"5ea1ddba-2854-4f68-9e4f-b140b4783a4c","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":146150,"visible":true,"origin":"","legend":"\u003cp\u003eFlowering plant (a) Abundance and (b) species richness in semi-natural and successional grasslands across years. The boxplots show the min, max, upper and lower quartile, and median of the data.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/37f943dbbc6212ae478cc75b.jpeg"},{"id":97371450,"identity":"b76d922f-106d-4490-8ac8-5692a07dc945","added_by":"auto","created_at":"2025-12-03 16:28:58","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":152389,"visible":true,"origin":"","legend":"\u003cp\u003eFlower abundance affecting (a) pollinator abundance (b) bumblebee abundance and (c) flower species diversity affecting bumblebee diversity. Green dots represent the SN and orange dots the SS grassland sites.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/43151d0b1e9cf635c8d612a1.jpeg"},{"id":97358771,"identity":"c7075654-b323-4763-8042-a4de50504550","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":155506,"visible":true,"origin":"","legend":"\u003cp\u003eNumber of plant-pollinator interactions at successional (SS) and semi-natural (SN) grasslands in years 2020-2022 for (a), Bumble bees, (b) Hoverflies, (c) Wild bees, (d) Honeybees. The boxplots show the min, max, upper and lower quartile, and median of the data.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/27a77c59376adf8792e7cbc8.png"},{"id":97358764,"identity":"56bb3879-f395-41b6-81b0-5f80536bfe84","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":65277,"visible":true,"origin":"","legend":"\u003cp\u003eEffects of grassland type on bumblebee network level properties. (a) web asymmetry, (b) Shannon's diversity, (c) linkage density. The boxplots show the min, max, upper and lower quartile, and median of the data.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/39c838ea0f865a68583b4c37.png"},{"id":97373243,"identity":"e9cbf47b-75c6-4120-bcac-4f9d900c03eb","added_by":"auto","created_at":"2025-12-03 16:35:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1885094,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/afa6aca2-9303-4b29-b921-a1f3ad79914c.pdf"},{"id":97358758,"identity":"eff77f92-1cc8-43d0-be55-d0ad3eb56acd","added_by":"auto","created_at":"2025-12-03 14:10:16","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":768579,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarySRAustrheim.docx","url":"https://assets-eu.researchsquare.com/files/rs-8086473/v1/a30e0331dde363734ff824f2.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Effects of agricultural land use on plant-pollinator interactions in peri- urban grasslands over 3 years","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eA large proportion of grasslands are semi-natural; grass and herb dominated ecosystems induced and maintained by agricultural land use, such as animal husbandry and fodder production. The ecological value of these man-made habitats is strongly associated with low intensive land use, preventing secondary succession to the benefit of field layer plant diversity \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, 3\u003c/sup\u003e. For pollinating insects, these herb rich semi-natural grasslands provide floral resources but also host plants and nesting sites for many species \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, 5,6\u003c/sup\u003e. The biodiversity value for semi-natural grassland is thus found to be high for both plant and pollinator communities\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, and most likely also for predators feeding on insects\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e, resulting in biodiversity hotspots across trophic levels.\u003c/p\u003e\u003cp\u003eIn parallel with a gradual increase in knowledge on the biodiversity and ecosystem services that semi-natural grasslands support, these low intensively used ecosystems are in decline mainly due to a change in land use. Two processes could be identified. First, agricultural intensification, such as ploughing frequency, fertilization and pesticide use to increase plant productivity, triggers loss of plants and pollinators and their network interactions\u003csup\u003e\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Secondly, abandonment of marginal agricultural sites implies an ecosystem shift from grassland to forest but agricultural legacies such as fertilization often prevent the development of species rich ecosystems for any group of species independent of tree cover\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Although these processes and the resulting negative effects on plants and pollinating insects are known\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, we have less knowledge on how different land use in grasslands affects the interactions between plants and pollinators\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn general, pollinator loss is driven by flower plant loss, and less so the other way around\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Hence, a high abundance of flowers is often found to promote high pollinator abundances. There is also some evidence that flowering plant diversity could positively impact pollinator diversity \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e suggesting that more diverse plant communities can attract specialized pollinators dependent on specialized floral characters. Historical data also show a parallel decline in plant and bee diversity in both the UK and the Netherlands \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Plant identity could thus be more important than plant diversity per se, especially species of high connectance, interacting with several other species. However, the overall size of the network (pool of both plants and pollinators) is expected to impact ecosystem properties such as stability. For example, diverse networks are found to be more resilient to disturbances and thus less vulnerable to plant or pollinator loss \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn many European countries, such as Norway, the abandonment of former infields in marginal areas for agricultural production has been increasing \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. A significant proportion of these abandoned infields are found in peri-urban areas where former agricultural land is fragmented by housing and other infrastructure, leaving mainly grasslands as patches for secondary succession \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Nitrophilous species dominate the vegetation, reflecting the high levels of nutrients historically added to facilitate fodder production. Although these areas still resemble green lungs in a built environment, their values for biodiversity and ecosystem services are considered low. This includes plant diversity, pollinators, and other invertebrates typically associated with grassland ecosystems. However, our understanding of the biodiversity structure and function of these former intensively used agricultural grasslands is limited in comparison to more species rich semi-natural grasslands valued for nature conservation.\u003c/p\u003e\u003cp\u003eThis study examined the abundance and diversity of flowering herb plants and pollinating insects in peri-urban semi-natural (SN) and successional (SS) grasslands. The invertebrate community associated with grasslands is highly complex and includes a broad range of taxa from strict pollinators to more poor pollinating insects. We quantified all insects found on flowers. This included strict pollinators such as bumblebees (Bombus), wild bees, honeybees (\u003cem\u003eApis mellifera\u003c/em\u003e), hoverflies (Syrphidae) and butterflies (Lepidoptera), but also insects considered to be of less importance for pollination such as beetles (Coleoptera), other flies (Brachycera), wasps (Vespidae), and true bugs (Heteroptera). Secondly, we addressed the land use impact on interactions between these two trophic levels in terms of network size and structure. The study was conducted over three years to assess temporal treatment effects.\u003c/p\u003e\u003cp\u003eOur two main expectations were, firstly, that plant resources in terms of overall flower abundance and their species richness are expected to follow habitat and species-specific phenological patterns throughout the growing season\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Peak pollinator abundance and taxonomic richness often coincide with the peak in flower resources. However, key drivers such as land use and weather influence the phenology of both plants and pollinators and the strengths of their interactions. We anticipate lower flower resources, pollinator abundance, and species richness in successional grasslands (abandoned agricultural infields) compared to semi-natural grasslands (nutrient-poor, less intensively managed habitats), regardless of season and year. Secondly, based on the assumption that an increase in the availability of plant resources will positively affect pollinator abundance and richness independent of land use and year, we expect more interactions and a higher stability (less variation) in plant resources, pollinator abundances and pollinator taxa richness across years in semi-natural (SN) grassland than in successional (SS) grassland. Moreover, since the plant-pollinator network can range from simple structures with few species of plants and pollinators to complex structures involving many species at both levels, regardless of the network size (number of interactions), we expect semi-natural grasslands to have more complex network structures in terms of diversity and connectance (higher proportion of possible interactions).\u003c/p\u003e"},{"header":"2. Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Abundance and species richness (SR) of insects and flowering plants\u003c/h2\u003e\n \u003cp\u003eWe found 38 taxa of insects of which 18 were identified to species level, and 20 to higher taxonomic levels (Table S3). Sixty-six flowering plant taxa included taxa at genus (9) and species (57) level (Table S4) across years, seasons and sites.\u003c/p\u003e\n \u003cp\u003eBumblebee species richness did not differ between grassland types (mean\u0026thinsp;=\u0026thinsp;2 in SN and mean\u0026thinsp;=\u0026thinsp;1 in SS across years, pe=-0.39, se 0.31, Z=-1.26, p\u0026thinsp;=\u0026thinsp;0.20). In total, ten bumblebee species were found, and two were exclusive to the SN grasslands. For the flowering plants, 27 and 5 species were unique for the SN and SS grassland, respectively. Overall, the overlap in plant community composition between grasslands was low, while the insect community showed a large overlap (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ea and b).\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eBoth abundance (pe=-0.43, se\u0026thinsp;=\u0026thinsp;0.13, Z=-3.25, p\u0026thinsp;=\u0026thinsp;0.001) and species richness (pe=-0.51, se\u0026thinsp;=\u0026thinsp;0.17, Z=-3.02, P\u0026thinsp;=\u0026thinsp;0.002) of flowering plants were higher in SN than SS grasslands (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Flower abundance peaked in both grassland types in 2020 (Mean 140 and 89 in SN and SS respectively) and were at the lowest in 2022 (mean 103 in SN and 23 in SS). Also, SR of flowering species peaked in 2020 (mean\u0026thinsp;=\u0026thinsp;9 in SN and mean\u0026thinsp;=\u0026thinsp;5 in SS) while 7 (SN) and 3 (SS) were the mean values in 2022 (Table \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eNMDS and environmental fitting revealed a strong effect of season on the community composition of flowering herbs (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003ea) (r2\u0026thinsp;=\u0026thinsp;0.465, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), followed by grassland type (r2\u0026thinsp;=\u0026thinsp;0.116 p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while year showed no effects (r2\u0026thinsp;=\u0026thinsp;0.023, p\u0026thinsp;=\u0026thinsp;0.212). For the insect community (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eb), we found variation among years (r2\u0026thinsp;=\u0026thinsp;0.252, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and season (r2\u0026thinsp;=\u0026thinsp;0.166, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while grassland type had no effect (r2\u0026thinsp;=\u0026thinsp;0.007, p\u0026thinsp;=\u0026thinsp;0.428). Flowering peaked in early summer, and the abundances of both flowers (Fig. S3a) and insects (Fig. S3b) were at their lowest in spring.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Plant-pollinator interactions\u003c/h2\u003e\n \u003cp\u003eA total of 7369 interactions were identified across years, seasons and sites. Flies dominated (3446) before bumblebees (1552), honeybees (924), beetles (438), hoverflies (346), wasps (209), Heteroptera (133), while the remaining had\u0026thinsp;\u0026lt;\u0026thinsp;100 interactions. For Bumblebees, we also recorded flying individuals (226 Bombus, most of them in SN sites) which were not further included in the analyses (for comparable values between all insect taxa for all years and grassland types but across sites, see Table. S3).\u003c/p\u003e\n \u003cp\u003eFlower abundance had a positive effect on the abundance of strict pollinating species (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ea, estimate 0.11, se\u0026thinsp;=\u0026thinsp;0.05, DF\u0026thinsp;=\u0026thinsp;23 t-value\u0026thinsp;=\u0026thinsp;2.24, p\u0026thinsp;=\u0026thinsp;0.035, marginal R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.068, conditional R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.227) and a marginal positive effect on bumblebee abundance (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eb, estimate 0.07, se\u0026thinsp;=\u0026thinsp;0.03, DF\u0026thinsp;=\u0026thinsp;23, t-value\u0026thinsp;=\u0026thinsp;2.06, p\u0026thinsp;=\u0026thinsp;0.050, marginal R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.087, conditional R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.256). Flower diversity had a strong positive effect on bumblebee diversity (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003ec, estimate 0.72, se\u0026thinsp;=\u0026thinsp;0.13, DF\u0026thinsp;=\u0026thinsp;23, t-value 5.403, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, marginal R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.351, conditional R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.407).\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eThe most important plant species for pollinators and bumblebees, especially across all years in SN grasslands, were \u003cem\u003eKnautia arvensis, Hieracium sp., Hypericum maculatum, Trifolium pratense, and Campanula rotundifolia (Fig. S7a\u003c/em\u003e. In SS grasslands \u003cem\u003eEpilobium angustifolium\u003c/em\u003e and \u003cem\u003eCirsium arvense\u003c/em\u003e were the most frequently visited species (Fig. S7b).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Network size and structure\u003c/h2\u003e\n \u003cp\u003eNumber of interactions was examined across and within taxa of strict pollinating insects (bumblebees, hoverflies, wild bees and honeybees, Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e, Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e) and insects of low importance for pollination (beetles, flies and wasps, Fig. S2, Table Sn). For \u0026ldquo;all pollinators\u0026rdquo;, the effect of land use varied over the years, with lower pollinator interactions in successional grasslands in both 2021 and 2022. Also for honeybees, the number of interactions in SS was lower in 2021 and in 2022, independent of land use. In hoverflies, interactions were higher in SN grasslands, independent of years. Only land use was retained in the best model for bumblebees and wild bees, but with no significant effects for either pollinator group.\u003c/p\u003e\n \u003cp\u003e\u003c/p\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMixed effect models on the effects of land use, year and their interactions and additive effects on different groups of pollinating insects. The best model is presented with the AIC scores, and the other models as the change in AIC (\u0026Delta;) as compared to the best model. Main effects from the best model is given as t-statistics. Abbreviations are Y\u0026thinsp;=\u0026thinsp;year, LSS\u0026thinsp;=\u0026thinsp;land use successional.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eResponse variables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e# observations\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTrans.\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLanduse\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eYear\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eInteraction\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAdditive\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAll pollinators\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e144\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026Delta;6.180\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;68.597\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3813.198\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eY2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e7.525***\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLSS:Y2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-8.161***\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLSS:Y2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-8.229***\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 69.606\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHoverflies\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e144\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e968.372\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLSS\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-3.295 ***\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;288.679\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026Delta; 276.416\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 282.373\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBumblebees\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e144\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2736.168\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 52.060\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026Delta;11.990\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 53.523\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHoneybees\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e144\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026Delta;104.17115\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;93.157\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2162.251\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eY2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-4.643***\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLSS:Y2021\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-8.920***\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 95.135\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWild bees\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e144\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e439.652\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;22.631\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026Delta;24.304\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 22.354\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n \u003cp\u003e\u0026ldquo;Poor pollinators\u0026rdquo; had a higher number of interactions in both 2021 and 2022 compared to 2020, and interactions varied with land use in 2021 (higher in SS). These differences were probably driven by the high number of interacting flies both in 2021 and 2022 as compared to 2020 (Fig. S2c). Beetle and wasp abundance varied over the years, independent of land use, with especially high values in 2022 (Fig. S2a, S2b).\u003c/p\u003e\n \u003cp\u003eAll empirical networks differed significantly from the simulated null model networks. H2 was higher than null model networks, while Connectivity, Web asymmetry, Links per species, Nestedness, Linkage density, and Shannon diversity were all lower than null models.\u003c/p\u003e\n \u003cp\u003eWe examined land use effects on network size and structure for two subgroups of insects: (1) bumblebees (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e, Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e) and (2) strict pollinating insects (Fig. S5, Table S5). Bumblebees had a higher web asymmetry in SS grasslands. Shannon diversity of interactions (i.e. network entries) was lower in SS vs SN grasslands networks (pe = -0.59, se\u0026thinsp;=\u0026thinsp;0.26, p\u0026thinsp;=\u0026thinsp;0.026), while linkage density showed a tendency of higher values in SN vs SS grasslands (pe=-0.38, se\u0026thinsp;=\u0026thinsp;0.23, z=-1.68, p\u0026thinsp;=\u0026thinsp;0.091). For all strict pollinating insects which also included, wild bees, hoverflies, and honeybees, the network patterns very much corresponded with network indices for bumblebees alone for Shannon diversity (marginally not significantly higher in SN pe\u0026thinsp;=\u0026thinsp;0.44, se\u0026thinsp;=\u0026thinsp;0.23, Z=-1.89, p\u0026thinsp;=\u0026thinsp;0.057) and linkage density (marginal higher number of interactions in SN vs SS grasslands, pe=-0.35, se\u0026thinsp;=\u0026thinsp;0.21, Z=-1.65, p\u0026thinsp;=\u0026thinsp;0.097). Web asymmetry for all strict pollinators was more driven by differences among years as compared to bumblebees alone, but the proportion of pollinator taxa vs plant taxa was still higher in SS vs SN grasslands.\u003c/p\u003e\n \u003cp\u003e\u003c/p\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMixed effect models on the effects of land use, year and their interactions and additive effects on different plant-bumblebee network indices. The best model is presented with the AIC scores, and the other models as the change in AIC (\u0026Delta;) as compared to the best model. Main effects from the best model is given as t-statistics. Abbreviations are Y\u0026thinsp;=\u0026thinsp;year, LSS\u0026thinsp;=\u0026thinsp;land use successional. The network index H2 had a model convergence problem and is not shown. Note that linkage density had marginally lower values in SN grasslands (Z= -1.688, p\u0026thinsp;=\u0026thinsp;0.0914).\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eResponse variables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e# observations\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTrans.\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLanduse\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eYear\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eInteraction\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAdditive\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eShannon\u0026apos;s diversity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eLSS 100.315\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-2.217*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 3.932\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 4.366\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 0.797\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003econnectance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e59.033\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 1.349\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 5.205\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 1.233\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eweb_asymmetry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGaussian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e-13.131\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLSS 3.733***\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; -5.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-13.612\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-14.250\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003elinkage_density\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e109.840\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 4.230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 4.230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 1.139\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003elinks_per_species\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e78.656\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta; 2.669\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;7.530\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;1.740\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eweighted_NODF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoisson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026Delta; 80.006\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;76.941\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e460.110\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-4.643***\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLSS:Y2022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e-7.632***\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026Delta;76.941\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n\u003c/div\u003e"},{"header":"3. Discussion","content":"\u003cp\u003eSemi-natural grasslands are expected to be key habitats for pollinating insects due to high abundances and richness of flowering herb species, but intensive land uses are often found to decrease flower food resources with negative effects for pollinators\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThis study compared plant-pollinator interactions in semi-natural (SN) versus former intensively used agricultural land which could be classified as successional (SS) grassland over three years. We found higher abundance and richness of flowering herb species in SN than in the SS grasslands. In contrast to our expectations, the impact of grassland type on insects was less clear. Strict pollinating insects were more abundant in SN in 2 out of three years, while hoverflies were more abundant in SN independent of years. We did not find higher abundances or species richness of bumblebees in SN grasslands. However, the higher abundance of flower resources correlated with the abundances of bumblebees and pollinators as expected. Also, bumblebee species richness was strongly correlated with flowering plant diversity, suggesting that the abundance and species richness of flowers are better predictors for pollinator abundance and diversity than grassland type.\u003c/p\u003e\u003cp\u003eWe found clear differences in network size and structure for bumblebees with higher diversity of interactions (i.e. number of network entries) and a tendency of higher linkage density for both bumblebees and strict pollinators in SN vs SS grasslands. In contrast, SS grasslands had a higher proportion of bumblebees and pollinators in relation to plants (web asymmetry) than the SN grasslands.\u003c/p\u003e\u003cp\u003eAlthough the evidences for loss of insect species richness and abundance are strong at a global level\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e as documented by an increasing number of local reports also from Norway \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e, the understanding of how different drivers effects these changes, and the relative importance of specific factors such as land use, climate, resources extraction are still limited\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Our study shows that semi-natural grassland is a heterogenous ecosystem with large variations in plant and insect community structure despite a common low intensive land use. Indeed, land use is a complex driver with varying impacts such as grazing, fodder harvest, fuel-wood cutting, use of fertilizers, and the temporal scale and legacies of these activities are difficult to quantify and disentangle\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Nevertheless, such differences in land use are probably an important reason for the large variation in both flower resources and insects among our SN sites.\u003c/p\u003e\u003cp\u003eThe clear difference in plant community composition between grassland types is characterised by the dominance of tall nitrophilous plants in former agricultural SS grasslands, while SN had a higher proportion of lower, less competitive herbs. Some of the tall herbs like \u003cem\u003eEpilobium angustifolium\u003c/em\u003e and \u003cem\u003eCirsium arvense\u003c/em\u003e offer a high abundance of pollinator food and show a high abundance of interactions, but probably for a shorter period compared to a plant community with a higher diversity and more even distribution across the growing season. Plant identity could also be more important than plant diversity per se, which underlines the need to identify attractive plant species that could be used to facilitate networks in grassland restoration initiatives. We show that several pollinator favored plants are low abundant species e.g. \u003cem\u003eKnautia arvensis\u003c/em\u003e, \u003cem\u003eTrifolium pratense\u003c/em\u003e, \u003cem\u003eHypericum maculatum\u003c/em\u003e, and \u003cem\u003eCampanula rotundifolia\u003c/em\u003e. None of these are rare in terms of red-listed but are restricted to areas with low-productivity agricultural practices of the past and absent in the dominating successional grasslands.\u003c/p\u003e\u003cp\u003eThe less clear differences in insect community composition between grassland types indicate a difference in overall community structure between the two trophic levels and more among year variation for insects. However, the lack of species level data for insects as compared to plants may have masked community differences in our study. The low abundance of butterflies across all years and seasons was surprising as SN grasslands are expected to be the main butterfly habitat in these peri-urban landscapes. Their rarity is confirmed by regional monitoring over the last decade, but lack of long-term monitoring makes it difficult to conclude on long-term trends. However, decreases in butterflies are shown in comparable habitats elsewhere in Europe\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAnnual differences in external environmental conditions as well as internal life histories are expected to strongly impact plant properties such as flowering, as well as insect abundances and richness, with consequences for interaction networks\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. This three-year study showed clear annual differences for several taxa of poor pollination importance such as flies and beetles. Also hoverflies varied among years with a low number of interactions the first year\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e possibly due to a dry summer which is known to limit recruitment of insects in general \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Little is known about the longer-term changes in plant-pollinator interactions. A regional study on bumblebee and butterfly showed no negative trends in abundances at the county level (Tr\u0026oslash;ndelag) from 2012 to 2020\u003csup\u003e32,33\u003c/sup\u003e. In contrast, a national survey showed decreased insect biomass at the regional level (Tr\u0026oslash;ndelag) over 5 years (2020\u0026ndash;2024)\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. A challenge for both studies is the lack of good reference values, but abundance values for both groups were considered to be lower than expected, possibly due to legacies of low recruitment in the past\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAlthough we found limited effects of grassland land use on plant pollinator network size, the network structure differed in three ways. First, the asymmetry values revealed a lower proportion of insects relative to flowering plants in the SN grasslands, which suggests that SS grasslands with lower flower abundance is still able to support an insect abundance not significantly lower than in the SN grasslands. A possible explanation could be a larger area of many of the SS grasslands. Indeed, SN grasslands had limited areas which could restrict food availability and insect population densities. Secondly, the higher number of network entries indicates that each plant and insect taxa interacted with more species in SN vs SS grasslands. Thirdly, we found a tendency for higher linkage density in SN grasslands, indicating that these grasslands rich in plant species supported larger networks with higher linkage densities \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e i.e. losing a partner in the network could make a species less vulnerable if it\u0026rsquo;s linked to many partners and the grassland more resilient to environmental fluctuations\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"4. Conclusions","content":"\u003cp\u003eWe found higher abundances and species richness of flowering herbs in SN as compared to SS grasslands. However, the land use impact on interacting insects was limited to higher abundances of hoverflies in SN and for pollinating insects in general for 2021 and 2022. A positive linear correlation between flower resources and pollinator abundance still supports the assumption that low food resources are a limiting factor for interacting pollinating insects. The high correlation between plant species richness and bumblebee diversity also suggests that SN grasslands have the potential to support a larger species pool of bumblebees if more low intensive grassland habitat is made available.\u003c/p\u003e\u003cp\u003eSeveral studies have shown positive effects of restoring former intensively used agricultural habitats into low-productive SN grasslands\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. An example is biomass and nutrient removal to increase flower resources, insect abundances and diverse networks, as shown for SN grasslands\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Our work will inform management of the plants and insect communities associated with key peri-urban grassland habitats on which plant species are the most important for strict pollinating insects such as bumblebees, hoverflies, wild bees, honeybees, and butterflies. Most of the key plant resources for pollinators identified in this study are common species expected to establish if SS grasslands currently out of agricultural use, could be restored to more low productive biodiversity friendly habitats.\u003c/p\u003e"},{"header":"5. Material and methods","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e5.1 Study design\u003c/h2\u003e\u003cp\u003eAll study sites (n\u0026thinsp;=\u0026thinsp;12) are found in Trondheim municipality, Tr\u0026oslash;ndelag, Norway (63\u0026deg;26\u0026prime;24\u0026prime;\u0026prime;N 10\u0026deg;24\u0026prime;0\u0026prime;\u0026prime;E) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). A \u0026ldquo;wall-to-wall\u0026rdquo; survey of land cover and land use was done at a large scale prior to the selection of the grassland study sites\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e within the peri-urban landscape of the city of Trondheim using the classification from AR5, a national product which classifies areas based on land-use and primary environmental conditions (NIBIO 2025). Two different grassland habitats were selected. Semi-natural grasslands (SN, n\u0026thinsp;=\u0026thinsp;6) had a low intensive land use with limited or no addition of fertilizer. Successional grasslands (SS, n\u0026thinsp;=\u0026thinsp;6) were former infields with a legacy of winter-fodder production and use of fertilizer. SN grasslands were characterized by low productivity and moderate plant height (mean height 50.51\u0026thinsp;\u0026plusmn;\u0026thinsp;2.2), while SS grasslands had a high productivity with tall biomass (mean height 119.19\u0026thinsp;\u0026plusmn;\u0026thinsp;2.05) as measured in July 2020\u003csup\u003e28\u003c/sup\u003e. This pilot study\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e also documented the rarity of SN grasslands (0.4 km\u003csup\u003e2\u003c/sup\u003e, 0.7%) as well as a higher proportion of former infield (0.6 km\u003csup\u003e2\u003c/sup\u003e, 1.1%) with no further agricultural use and thus left for secondary succession.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe visited each study site four times (in random order) during the summer season over three years (2020, 2021, 2022) to account for seasonal differences: (1) spring (late May, early June), (2) early summer (late June), (3) mid summer (late July) and (4) late summer (mid August). Plant flower abundance, species richness of flowering plants and insect abundance and taxa richness were recorded within a 250 m\u003csup\u003e2\u003c/sup\u003e area at the site level. For plants, we quantified flowering herbs within a 0.5 x 0.5 quadrat divided into 16 subquadrats. Presence-absence of flowering herbs within the subquadrats gave a maximum value of 16 per species at a quadrat level if all subplots had a flower of the same species independent of the number of flowers within each subplot. Species richness of flowering species was also recorded at the quadrat level as the total number of flowering species found within the plot. A total of 25 quadrats for examining flowering plants were placed systematically within the 250 m\u003csup\u003e2\u003c/sup\u003e area at each site.\u003c/p\u003e\u003cp\u003eWhen quantifying insects at the site level, the same study area was examined by walking in the middle of 5 m broad transects (5 x 50 m\u0026thinsp;=\u0026thinsp;250 m\u003csup\u003e2\u003c/sup\u003e)\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. All insects sitting on herb flowers were recorded 2.5 m on each side of the transect while walking at a slow pace. A sweep net was used to collect insects for identification. Only bumblebees (Bombus) and butterflies (Lepidoptera) were identified to the species level. For bumblebees, only males and workers were taken to the lab for species identification if needed. The naming of bumblebees follows \u0026Oslash;degaard\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. All other insects were identified at the genus or family level. A complete list of insect and plant taxa included is given in Tables S3 and S4, respectively.\u003c/p\u003e\u003cp\u003eInsect sampling was only done on sunny days, avoiding cold mornings and with an ambient temperature\u0026thinsp;\u0026gt;\u0026thinsp;15\u0026deg;C. Days with strong winds were also avoided (\u0026lt;\u0026thinsp;15 km/h). Coordinates of both starting and ending points were recorded with the help of a GPS (Global Positioning System) device to have a reference for later sampling. Six different people were involved in plant and insect sampling across the years. To minimize personal errors and calibrate sampling procedures, 2 recorders were present at the start of the sampling in all years.\u003c/p\u003e\u003cp\u003eAll invertebrates along the transect were quantified and grouped at different taxonomic levels. For analysis on plant-pollinator networks, we included bumblebees, which were identified at a species level except for \u003cem\u003eBombus s. str\u003c/em\u003e. Subgroups of pollinating insects, including bumblebees (Bombus), butterflies (Lepidoptera parts, species level), solitary bees (grouped), hoverflies (Syrphidae, grouped) and the honeybee (\u003cem\u003eApis mellifera)\u003c/em\u003e were also identified for network-level analyses. Invertebrates considered to be of low importance for pollination, such as other flies (Brachycera), beetles (Coleoptera), ladybugs (Coccinellidae), ants (Formicidae), moths (Lepidoptera parts), true bugs (Heteroptera) and stinging wasps (Vespidae) were grouped as poor pollinators.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e5.2 Statistical analyses\u003c/h2\u003e\u003cp\u003eData processing and analyses were done using RStudio (version 4.3.1, (Team 2024) R Core Team 2024). We used generalised linear mixed models (GLMM) (library glmmTMB) to test the effect of four candidate models: treatment (semi-natural vs successional grassland), year and their potential interactions (multiplicative, treatment*year) and additive effects (treatment\u0026thinsp;+\u0026thinsp;year). The responses included (1) herb flower abundance and species richness, (2) bumblebee abundance and species richness (3) abundance for all taxa of insects found at flowers, and (4) network level properties on size and structure for a) Plant-Bumblebee, b) Plant-strict pollinating insects\u0026rsquo; networks. Site (six sites per treatment) was included as a random variable. In order to manage zero-inflation and problems with non-normal variation in the model residuals, plant and pollinator data were aggregated across the four seasons except for descriptive models on seasonal variations.\u003c/p\u003e\u003cp\u003eTo account for the non-normal distribution for count variables, we used Poisson distribution for all modelling except for negative responses (web asymmetry) for which we used a Gaussian distribution. AIC optimization was used to select the best model. All diagnostics of the selected model were assessed by checking the distribution of the residuals using the diagnostic tools. Non-metric multidimensional scaling (NMDS) followed by environmental fitting by the function envfit with 999 permutations was used for the ordination of plant and insect community data using Vegan\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. All plots were made by the library ggplot \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, except for the figures on plant-pollinator networks made by using Bipartite\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe Bipartite package\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e was also used to calculate plant-pollinator network level properties on size and structure for 1) strict pollinating species, 2) bumblebee species. The following indices were included: (1) Mean number of links per species (qualitative): sum of links divided by number of species. (2) Linkage density: Marginal totals-weighted diversity of interactions per species (quantitative). This is computed as the average of vulnerability and generality\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. (3) Shannon's diversity of interactions (i.e. network entries). (4) Web asymmetry: Balance between numbers in the two levels: positive values indicate higher-trophic level species, negative lower-trophic level species; implemented as (ncol(web)-nrow(web))/sum(dim(web)); web asymmetry is a null model for what one might expect in dependence asymmetry\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. (5) Connectance: Realised proportion of possible links\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e: sum of links divided by number of cells in the matrix (=\u0026thinsp;number of higher times number of lower trophic level species). This is the \u003cem\u003estandardised number of species combinations\u003c/em\u003e often used in co-occurrence analyses \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e (6) Specialisation (H2). (7) Nestedness (NODF).\u003c/p\u003e\u003cp\u003eNetworks and associated properties were constructed for each site every year (12 sites x 3 years\u0026thinsp;=\u0026thinsp;36 networks). To check for possible spurious relationships in our empirical networks between plants and pollinators, we carried out Z-score calculations\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e based on 1000 random null model interactions for each of the seven network properties. A t-test was used to assess the difference between the empirical model and the null model. When modelling treatment and year effects on network properties, network size was included as a random variable to correct for differences in size. However, the inclusion of size did not improve any model as assessed by AIC.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis work was supported by the Research Council of Norway, Project No. 319892.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eGA: Conceptualization, Formal analysis, Methodology, Funding acquisition, Supervision, Writing. RS: Investigation, Data curation, Writing. AD: Investigation, Data curation, Writing. HH: Investigation, Data curation, Writing. SHE: Formal analysis, Writing. F\u0026Oslash;: Conceptualization, Methodology, Supervision, Writing.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eAcknowledgementsThis work was supported by the Research Council of Norway, Project No. 319892. We thank Aksel J. Fosse for help with the fieldwork.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eSummary tables on plant and pollinator data are available in the supplementary (Table S2 and S3). Complete datasheets are available through figshare.com: [https://doi.org/10.6084/m9.figshare.30589385.v1](https:/doi.org/10.6084/m9.figshare.30589385.v1)\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLe F\u0026eacute;on, V.\u003cem\u003e et al.\u003c/em\u003e Intensification of agriculture, landscape composition and wild bee communities: A large scale study in four European countries. \u003cem\u003eAgriculture, Ecosystems \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e137\u003c/strong\u003e, 143\u0026ndash;150 (2010). https://doi.org/https://doi.org/10.1016/j.agee.2010.01.015\u003c/li\u003e\n\u003cli\u003eWinberg, J., Ekroos, J. \u0026amp; Smith, H. G. Abandonment or biomass production? Phytodiversity responses to land-use changes of semi-natural grasslands in northern Europe. \u003cem\u003eBiological Conservation\u003c/em\u003e \u003cstrong\u003e294\u003c/strong\u003e, 110632 (2024). https://doi.org/https://doi.org/10.1016/j.biocon.2024.110632\u003c/li\u003e\n\u003cli\u003eHirayama, G. S., Tomita, M. \u0026amp; Ushimaru, A. Impacts of agricultural intensification and abandonment on plant and pollinator diversity, their mutualistic networks, and pollination services. \u003cem\u003eAgriculture, Ecosystems \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e381\u003c/strong\u003e, 109437 (2025). https://doi.org/https://doi.org/10.1016/j.agee.2024.109437\u003c/li\u003e\n\u003cli\u003e\u0026Ouml;CKINGER, E. \u0026amp; SMITH, H. G. Semi-natural grasslands as population sources for pollinating insects in agricultural landscapes. \u003cem\u003eJournal of Applied Ecology\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 50\u0026ndash;59 (2007). https://doi.org/https://doi.org/10.1111/j.1365-2664.2006.01250.x\u003c/li\u003e\n\u003cli\u003eGoulson, D., Lye, G. C. \u0026amp; Darvill, B. Decline and Conservation of Bumble Bees. \u003cem\u003eAnnual Review of Entomology\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 191\u0026ndash;208 (2008). https://doi.org/https://doi.org/10.1146/annurev.ento.53.103106.093454\u003c/li\u003e\n\u003cli\u003eLarkin, M. \u0026amp; Stanley, D. A. Impacts of management at a local and landscape scale on pollinators in semi-natural grasslands. \u003cem\u003eJournal of Applied Ecology\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e, 2505\u0026ndash;2514 (2021). https://doi.org/https://doi.org/10.1111/1365-2664.13990\u003c/li\u003e\n\u003cli\u003eBardgett, R. D.\u003cem\u003e et al.\u003c/em\u003e Combatting global grassland degradation. \u003cem\u003eNature Reviews Earth \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 720\u0026ndash;735 (2021). https://doi.org/10.1038/s43017-021-00207-2\u003c/li\u003e\n\u003cli\u003eBowler, D. E., Heldbjerg, H., Fox, A. D., de Jong, M. \u0026amp; B\u0026ouml;hning-Gaese, K. Long-term declines of European insectivorous bird populations and potential causes. \u003cem\u003eConservation Biology\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 1120\u0026ndash;1130 (2019). https://doi.org/https://doi.org/10.1111/cobi.13307\u003c/li\u003e\n\u003cli\u003eSeress, G., S\u0026aacute;ndor, K., Evans, K. L. \u0026amp; Liker, A. Food availability limits avian reproduction in the city: An experimental study on great tits Parus major. \u003cem\u003eJournal of Animal Ecology\u003c/em\u003e \u003cstrong\u003e89\u003c/strong\u003e, 1570\u0026ndash;1580 (2020). https://doi.org/https://doi.org/10.1111/1365-2656.13211\u003c/li\u003e\n\u003cli\u003eCarvalheiro, L. G.\u003cem\u003e et al.\u003c/em\u003e Species richness declines and biotic homogenisation have slowed down for NW-European pollinators and plants. \u003cem\u003eEcology Letters\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 870\u0026ndash;878 (2013). https://doi.org/https://doi.org/10.1111/ele.12121\u003c/li\u003e\n\u003cli\u003eVanbergen, A. J. \u0026amp; Initiative, t. I. P. Threats to an ecosystem service: pressures on pollinators. \u003cem\u003eFrontiers in Ecology and the Environment\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 251\u0026ndash;259 (2013). https://doi.org/https://doi.org/10.1890/120126\u003c/li\u003e\n\u003cli\u003eDeguines, N.\u003cem\u003e et al.\u003c/em\u003e Large-scale trade-off between agricultural intensification and crop pollination services. \u003cem\u003eFrontiers in Ecology and the Environment\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 212\u0026ndash;217 (2014). https://doi.org/https://doi.org/10.1890/130054\u003c/li\u003e\n\u003cli\u003eHabas, J. F.\u003cem\u003e et al.\u003c/em\u003e Long-term effects of nutrient addition on yield, diversity, and nutritional quality in a Mediterranean grassland under variable rainfall. \u003cem\u003eAgriculture, Ecosystems \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e393\u003c/strong\u003e, 109811 (2025). https://doi.org/https://doi.org/10.1016/j.agee.2025.109811\u003c/li\u003e\n\u003cli\u003eIPBES. Report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services on Pollinators, Pollination and Food Production. . (Secretariat of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, Bonn, Germany., 2016).\u003c/li\u003e\n\u003cli\u003eRaven, P. H. \u0026amp; Wagner, D. L. Agricultural intensification and climate change are rapidly decreasing insect biodiversity. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2002548117 (2021). https://doi.org/doi:10.1073/pnas.2002548117\u003c/li\u003e\n\u003cli\u003eVanbergen, A. J. Landscape alteration and habitat modification: impacts on plant\u0026ndash;pollinator systems. \u003cem\u003eCurrent Opinion in Insect Science\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 44\u0026ndash;49 (2014). https://doi.org/https://doi.org/10.1016/j.cois.2014.09.004\u003c/li\u003e\n\u003cli\u003eWillmer, P. \u003cem\u003ePollination and Floral Ecology.\u003c/em\u003e, (Princeton University Press, 2011).\u003c/li\u003e\n\u003cli\u003eEbeling, A., Klein, A.-M., Schumacher, J., Weisser, W. W. \u0026amp; Tscharntke, T. How does plant richness affect pollinator richness and temporal stability of flower visits? \u003cem\u003eOikos\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 1808\u0026ndash;1815 (2008). https://doi.org/https://doi.org/10.1111/j.1600-0706.2008.16819.x\u003c/li\u003e\n\u003cli\u003eBaude, M.\u003cem\u003e et al.\u003c/em\u003e Historical nectar assessment reveals the fall and rise of floral resources in Britain. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e530\u003c/strong\u003e, 85\u0026ndash;88 (2016). https://doi.org/10.1038/nature16532\u003c/li\u003e\n\u003cli\u003eBiesmeijer, J. C.\u003cem\u003e et al.\u003c/em\u003e Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e313\u003c/strong\u003e, 351\u0026ndash;354 (2006). https://doi.org/10.1126/science.1127863\u003c/li\u003e\n\u003cli\u003eRedhead, J. W.\u003cem\u003e et al.\u003c/em\u003e Potential landscape-scale pollinator networks across Great Britain: structure, stability and influence of agricultural land cover. \u003cem\u003eEcology Letters\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 1821\u0026ndash;1832 (2018). https://doi.org/https://doi.org/10.1111/ele.13157\u003c/li\u003e\n\u003cli\u003eTerres, J.-M.\u003cem\u003e et al.\u003c/em\u003e Farmland abandonment in Europe: Identification of drivers and indicators, and development of a composite indicator of risk. \u003cem\u003eLand Use Policy\u003c/em\u003e \u003cstrong\u003e49\u003c/strong\u003e, 20\u0026ndash;34 (2015). https://doi.org/https://doi.org/10.1016/j.landusepol.2015.06.009\u003c/li\u003e\n\u003cli\u003eGeslin, B., Gauzens, B., Th\u0026eacute;bault, E. \u0026amp; Dajoz, I. Plant Pollinator Networks along a Gradient of Urbanisation. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, e63421 (2013). https://doi.org/10.1371/journal.pone.0063421\u003c/li\u003e\n\u003cli\u003ePetersen, T. K., Speed, J. D. M., Gr\u0026oslash;tan, V. \u0026amp; Austrheim, G. Urban aliens and threatened near-naturals: Land-cover affects the species richness of alien- and threatened species in an urban-rural setting. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 8513 (2020). https://doi.org/10.1038/s41598-020-65459-2\u003c/li\u003e\n\u003cli\u003eBurner, R. C., Sel\u0026aring;s, V., Kobro, S., Jacobsen, R. M. \u0026amp; Sverdrup-Thygeson, A. Moth species richness and diversity decline in a 30-year time series in Norway, irrespective of species\u0026rsquo; latitudinal range extent and habitat. \u003cem\u003eJournal of Insect Conservation\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 887\u0026ndash;896 (2021). https://doi.org/10.1007/s10841-021-00353-4\u003c/li\u003e\n\u003cli\u003eVan Swaay, C. A. M., 2, Dennis, E.B.3, Schmucki, R.4, Sevilleja, C.G.1,2, \u0026Aring;str\u0026ouml;m, S.6, Balalaikins, M.7, Barea-Azc\u0026oacute;n, J.M. 5, Bonelli, S.8, Botham, M.4, Cancela, J.P.51, Collins, S.1, De Flores, M.18, Dapporto, L.44, Dopagne, C.12, Dziekanska, I.13, Escob\u0026eacute;s, R.14, Faltynek Fric, Z.30, Fern\u0026aacute;ndez-Garc\u0026iacute;a, J.M.16, Fontaine, B.17, Glogovčan, P.46, Gracianteparaluceta, A.16, Harpke, A.15, 50, Harrower, C.4, Heli\u0026ouml;l\u0026auml;, J.19, Houard, X., 18, Judge, M.25, Kolev, Z.20, Komac, B.21, K\u0026uuml;hn, E.15, Kuussaari, M.19, Lang, A.23, Lysaght, L.25, Maes, D.26, McGowan,D.11, Mestdagh, X.27, Middlebrook, I.3, Monasterio, Y.14, Monteiro, E.28, Munguira, M.L.9,1, Musche, M.15, Olivares, F.J.24, \u0026Otilde;unap, E.29, Ozden, O.30, Pavl\u0026iacute;čko, A.32, Pendl, M.22, Pettersson, L.B.33, R\u0026aacute;kosy, L.36, Roth, T.37, R\u0026uuml;disser, J.38, \u0026Scaron;a\u0026scaron;ić, M.39, Scalercio, S.47, Settele, J.15, 48, 49, Sielezniew, M.40, Sobczyk-Moran, G.18, Stefanescu, C.31, \u0026Scaron;vitra, G.42, Szabadfalvi, A.43, Tiitsaar, A.29, Titeux, N.27, Tzirkalli, E.45, Ubach, A.31, Verovnik, R.10, Vray, S.27, Warren, M.S.3, Wynhoff, I.1,2 \u0026amp; Roy, D.B. European Grassland Butterfly Indicator 1990-2020 (2022).\u003c/li\u003e\n\u003cli\u003eFisogni, A.\u003cem\u003e et al.\u003c/em\u003e Seasonal trajectories of plant-pollinator interaction networks differ following phenological mismatches along an urbanization gradient. \u003cem\u003eLandscape and Urban Planning\u003c/em\u003e \u003cstrong\u003e226\u003c/strong\u003e, 104512 (2022). https://doi.org/https://doi.org/10.1016/j.landurbplan.2022.104512\u003c/li\u003e\n\u003cli\u003eDhukuchhu, A. \u003cem\u003eEffects of land use and season on flowering plants and plant-pollinator networks in urban grasslands\u003c/em\u003e Master\u0026rsquo;s thesis in Biology thesis, Norwegian University of Science and Technology, (2021).\u003c/li\u003e\n\u003cli\u003eHegg\u0026oslash;y, H. \u003cem\u003eEffects of land-use on bumblebees and other pollinators in urban grasslands\u003c/em\u003e Master thesis, Norwegian University of Science and Technology, (2021).\u003c/li\u003e\n\u003cli\u003eHuberty, A. F. \u0026amp; Denno, R. F. PLANT WATER STRESS AND ITS CONSEQUENCES FOR HERBIVOROUS INSECTS: A NEW SYNTHESIS. \u003cem\u003eEcology\u003c/em\u003e \u003cstrong\u003e85\u003c/strong\u003e, 1383\u0026ndash;1398 (2004). https://doi.org/https://doi.org/10.1890/03-0352\u003c/li\u003e\n\u003cli\u003eGely, C., Laurance, S. G. W. \u0026amp; Stork, N. E. How do herbivorous insects respond to drought stress in trees? \u003cem\u003eBiological Reviews\u003c/em\u003e \u003cstrong\u003e95\u003c/strong\u003e, 434\u0026ndash;448 (2020). https://doi.org/https://doi.org/10.1111/brv.12571\u003c/li\u003e\n\u003cli\u003e\u0026Ouml;berg, S., Gjershaug, J.O., Certain, G. \u0026amp; \u0026Oslash;degaard, F. Development of methodology for area representative monitoring of chosen invertebrate groups. . 50 (NINA, 2010).\u003c/li\u003e\n\u003cli\u003eStraume, R. \u003cem\u003eSpatiotemporal variation and flowering plant preferences of insect pollinators in grassland ecosystems\u003c/em\u003e, Norwegian University of Science and Technology, (2022).\u003c/li\u003e\n\u003cli\u003e\u0026Aring;str\u0026ouml;m, J., Gohli, J., Birkemoe, T., Brealey, J., Dahle, S., Davey, M., Ekrem, T., Foss\u0026oslash;y, F., Laugsand, A., Opsahl, N., A., Solvang Klokke, V., Staverl\u0026oslash;kk, A. \u0026amp; Sverdrup‐Thygeson, A. . Norsk insektoverv\u0026aring;king. Rapport fra feltsesongen 2024., (Norwegian Insitute for Nature Research, Trondheim, 2025).\u003c/li\u003e\n\u003cli\u003eDupont, Y. L., Damgaard, C. \u0026amp; Simonsen, V. Quantitative Historical Change in Bumblebee (Bombus spp.) Assemblages of Red Clover Fields. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e25172 (2011). https://doi.org/10.1371/journal.pone.0025172\u003c/li\u003e\n\u003cli\u003eLarkin, M. \u0026amp; Stanley, D. A. Impacts of local and landscape grassland management on the structure of plant-pollinator networks. \u003cem\u003eBasic and Applied Ecology\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 50\u0026ndash;59 (2023). https://doi.org/https://doi.org/10.1016/j.baae.2023.04.003\u003c/li\u003e\n\u003cli\u003eProesmans, W.\u003cem\u003e et al.\u003c/em\u003e Urbanisation and agricultural intensification modulate plant\u0026ndash;pollinator network structure and robustness. \u003cem\u003eFunctional Ecology\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 628\u0026ndash;641 (2024). https://doi.org/https://doi.org/10.1111/1365-2435.14503\u003c/li\u003e\n\u003cli\u003eSexton, A. N. \u0026amp; Emery, S. M. Grassland restorations improve pollinator communities: a meta-analysis. \u003cem\u003eJournal of Insect Conservation\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 719\u0026ndash;726 (2020). https://doi.org/10.1007/s10841-020-00247-x\u003c/li\u003e\n\u003cli\u003eKleppe, K. B. \u003cem\u003eGrasslands in Trondheim; Land-use, plant species richness and soil characteristics\u003c/em\u003e Master\u0026rsquo;s thesis in biology thesis, Norwegian University of Science and Technology, (2019).\u003c/li\u003e\n\u003cli\u003e\u0026Oslash;degaard, F., Staverl\u0026oslash;kk, A., Gjershaug, J.O., Bengtson, R. \u0026amp; Mjelde, A. \u003cem\u003eBumblebees in Norway: Characteristics, distrubution and life history.\u003c/em\u003e, (NINA, 2015).\u003c/li\u003e\n\u003cli\u003evegan: Community Ecology Package v. R package version 2.8-0 (2025).\u003c/li\u003e\n\u003cli\u003eWickham, H. \u003cem\u003eGgplot2: Elegant Graphics for Data Analysis.\u003c/em\u003e 2nd Edition edn, (Springer 2009).\u003c/li\u003e\n\u003cli\u003eBipartite vignette (github, 2019).\u003c/li\u003e\n\u003cli\u003eBersier, L.-F., Bana\u0026scaron;ek-Richter, C. \u0026amp; Cattin, M.-F. Quantitative Descriptors of Food-Web Matrices. \u003cem\u003eEcology\u003c/em\u003e \u003cstrong\u003e83\u003c/strong\u003e, 2394\u0026ndash;2407 (2002). https://doi.org/10.2307/3071801\u003c/li\u003e\n\u003cli\u003eBl\u0026uuml;thgen, N., Menzel, F., Hovestadt, T., Fiala, B. \u0026amp; Bl\u0026uuml;thgen, N. Specialization, Constraints, and Conflicting Interests in Mutualistic Networks. \u003cem\u003eCurrent Biology\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 341\u0026ndash;346 (2007). https://doi.org/https://doi.org/10.1016/j.cub.2006.12.039\u003c/li\u003e\n\u003cli\u003eDunne, J. A., Williams, R. J. \u0026amp; Martinez, N. D. Food-web structure and network theory: The role of connectance and size. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e99\u003c/strong\u003e, 12917\u0026ndash;12922 (2002). https://doi.org/doi:10.1073/pnas.192407699\u003c/li\u003e\n\u003cli\u003eGotelli, N. J. a. G. R. G. \u003cem\u003eNull Models in Ecology.\u003c/em\u003e, (Smithsonian Institution Press, , 1996).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Semi-natural grasslands, successional grasslands, herbs, bumblebees, hoverflies","lastPublishedDoi":"10.21203/rs.3.rs-8086473/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8086473/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLand use is a major driver for plant-pollinator interactions in grasslands. This study examined abundance and taxa richness of flowering herbs and interacting insects in semi-natural (SN) and successional (SS) grasslands in a peri-urban landscape over 3 years. All insects found on flowers were recorded. Strict pollinating insects included bumblebees (Bombus), hoverflies (Syrphidae), wild bees and honeybees (\u003cem\u003eApis mellifera\u003c/em\u003e). We found higher abundances and richness of flowering herbs in SN vs SS grasslands all years. Pollinators had higher abundances at SN sites in 2 out of 3 years, mainly driven by hoverflies. In contrast to our expectations, we found no land use effect on bumblebees. However, bumblebee and strict pollinator abundance correlated with flower abundance, and a linear relationship between bumblebee and herb species diversity suggests that flower resources and diversity are good predictors of pollinator abundance and diversity. Analyses of plant-bumblebee network structure showed higher asymmetry in SS grasslands. SN grasslands scored higher on Shannon diversity and linkage density, which suggests that SS networks are more vulnerable to both plant and bumblebee species loss. Overall, the highest abundance and diversity of flowers and strict pollinating insects were found in SN grasslands, but many plant-pollinator parameters varied more within than among grassland type. Our study identified herb species favored by pollinators, which should be facilitated to increase SN grasslands with high quality flower resources to prevent pollinator declines.\u003c/p\u003e","manuscriptTitle":"Effects of agricultural land use on plant-pollinator interactions in peri- urban grasslands over 3 years","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-03 14:10:11","doi":"10.21203/rs.3.rs-8086473/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-19T17:22:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"288787641928737253401497256040919194218","date":"2026-04-29T13:29:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"2485389088149409699859618196461797914","date":"2026-04-29T10:08:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"64653560443513030394258381915829427039","date":"2026-04-29T08:35:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"233240474769085025013210050466510792181","date":"2026-04-28T22:31:34+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-20T12:42:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"87329704510183984962667282725510327474","date":"2026-02-23T08:14:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-02T00:30:03+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-25T16:40:29+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-11-20T09:21:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-14T08:24:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-11-14T08:21:23+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"370e8a28-1b29-4296-9bff-f9da147e402a","owner":[],"postedDate":"December 3rd, 2025","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-19T17:22:50+00:00","index":102,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":58971965,"name":"Biological sciences/Ecology"},{"id":58971966,"name":"Earth and environmental sciences/Ecology"},{"id":58971967,"name":"Biological sciences/Zoology"}],"tags":[],"updatedAt":"2025-12-03T14:10:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-03 14:10:11","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8086473","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8086473","identity":"rs-8086473","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}