Received: 6 November 2020 | Accepted: 16 August 2021 DOI: 10.1111/1365-2664.14022 RESEARCH ARTICLE Management-­dependent effects of pollinator functional diversity on apple pollination services: A response–­effect trait approach Laura Roquer-­Beni1,2 | Georgina Alins3 | Xavier Arnan1,4 | Virginie Boreux5 | Daniel García6 | Peter A. Hambäck7 | Anne-­Kathrin Happe8 | Alexandra-­Maria Klein5 | Marcos Miñarro9 | Karsten Mody8,10 | Mario Porcel11,12 | Anselm Rodrigo1 | 7,13,14 11,15 1 Ulrika Samnegård | Marco Tasin | Jordi Bosch 1 CREAF, Universitat Autònoma de Barcelona, Bellaterra, Spain Abstract 2 BETA, University of Vic–­Central University 1. Functional traits mediate the response of communities to disturbances (response of Catalonia, Vic, Spain traits) and their contribution to ecosystem functions (effect traits). To predict 3 IRTA Fruitcentre, PCiTAL, Lleida, Spain 4 how anthropogenic disturbances influence ecosystem services requires a dual Department of Biological Sciences, University of Pernambuco, Garanhuns, Brazil approach including both trait concepts. Here, we used a response–­effect trait 5 Chair of Nature Conservation and conceptual framework to understand how local and landscape features affect pol- Landscape Ecology, University of Freiburg, linator functional diversity and pollination services in apple orchards. Freiburg, Germany 6 Instituto Mixto de Investigación en 2. We worked in 110 apple orchards across four European regions. Orchards dif- Biodiversidad (CSIC-­Uo-­PA), Oviedo, Spain fered in management practices. Low-­intensity (LI) orchards were certified organic 7 Department of Ecology, Environment or followed close-­to-­organic practices. High-­intensity (HI) orchards followed in- and Plant Sciences, Stockholm University, Stockholm, Sweden tegrated pest management practices. Within each management type, orchards 8 Department of Biology, Technical University encompassed a range of local (flower diversity, agri-­environmental structures) and of Darmstadt, Darmstadt, Germany 9 landscape features (orchard and pollinator-­friendly habitat cover). We measured Servicio Regional de Investigación y Desarrollo Agroalimentario, Villaviciosa, pollinator visitation rates and calculated trait composition metrics based on 10 Spain pollinator traits. We used initial fruit set as a measure of pollination service. 10 Department of Applied Ecology, 3. Some pollinator traits (body size and hairiness) were negatively related to orchard Hochschule Geisenheim University, Geisenheim, Germany cover and positively affected by pollinator-­friendly habitat cover. Bee functional 11 Department of Plant Protection Biology, diversity was lower in HI orchards and decreased with increased landscape or- Integrated Plant Protection Unit, Swedish University of Agricultural Sciences, Alnarp, chard cover. Pollination service was not associated with any particular trait but Sweden increased with pollinator trait diversity in LI orchards. As a result, LI orchards with 12 Corporación Colombiana de Investigación high pollinator trait diversity reached levels of pollination service similar to those Agropecuaria, Meta, Colombia 13 Department of Biology, Lund University, of HI orchards. Lund, Sweden 4. Synthesis and applications. Pollinator functional diversity enables pollinator com- 14 School of Environmental & Rural Sciences, munities to respond to agricultural intensification and to increase pollination func- University of New England, Armidale, Australia tion. Our results show that efforts to promote biodiversity provide greater returns 15 Department of Chemistry, University of in low-­intensity than in high-­intensity orchards. The fact that low-­intensity or- Padova, Padova, Italy chards with high pollinator functional diversity reach levels of pollination services This is an open access article under the terms of the Creat​ive Commo​ns Attri​bution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. Journal of Applied Ecology published by John Wiley & Sons Ltd on behalf of British Ecological Society. J Appl Ecol. 2021;58:2843–2853.  wileyonlinelibrary.com/journal/jpe | 2843 2844 | Journal of Applied Ecology ROQUER-­BENI et al. Correspondence Laura Roquer-­Beni similar to those of high-­intensity orchards provides a compelling argument for the Email: laura.roquer@uvic.cat conversion of high-­intensity into low-­intensity farms. Funding information KEYWORDS BiodivERsA/FACCE-­JPI, Grant/Award Number: 2014-­74; Agència de Gestió agricultural intensification, agri-­environmental structures, integrated pest management, d'Ajuts Universitaris i de Recerca, organic management, response–­effect trait framework, trait diversity, trait identity Grant/Award Number: FI; Stiftelsen Lantbruksforskning, Grant/Award Number: H1256150; Bundesministerium für Bildung und Forschung, Grant/Award Number: 01LC1403 and PT-­DLR/BMBF; Ministerio de Economía y Competitividad, Grant/ Award Number: CGL2015-­68963-­C2-­2-­R , PCIN-­2014-­145-­C02 and RYC-­2015-­18448; Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Grant/ Award Number: RTA2013-­00039-­C03-­0 0; Svenska Forskningsrådet Formas, Grant/ Award Number: 2013-­934 and 2014-­1784 Handling Editor: Guadalupe Peralta 1 | I NTRO D U C TI O N pollinator richness and abundance have been widely studied (Marja et al., 2019; Scheper et al., 2013), but much less is known about their Environmental disturbances such as agricultural intensification alter consequences on pollinator functional composition. ecosystem function through changes in functional composition of Pollinator responses to agricultural intensification depend on plant and animal communities (Larsen et al., 2005; McGill et al., 2006). traits related to mobility, feeding and nesting requirements and Species have traits that affect their ability to cope with environmental physiological tolerance (De Palma et al., 2015; Forrest et al., 2015; changes (response traits) and traits that contribute to specific func- Rader et al., 2014). For example, large mobile bee species may be tions such as pollination (effect traits; Violle et al., 2007). Thus, the better suited to find floral resources in disturbed habitats com- degree to which disturbances will affect ecosystem functioning will pared to small species (Jauker et al., 2013; Klein et al., 2008), but depend on the overlap between response and effect traits (Schleuning they may also have higher levels of exposure to pesticides (Brittain et al., 2015; Suding et al., 2008). For this reason, a response–­effect & Potts, 2011). Although no general patterns have been found be- trait framework is necessary to fully understand how specific drivers tween single trait identity and responses to environmental changes affect ecosystem function and services (Lavorel & Garnier, 2002). At (Bartomeus et al., 2018; Bommarco et al., 2010), there is evidence the community level, the response to environmental changes and the that landscape intensification acts as a filter of specific traits caus- maintenance of ecosystem functions may be affected by the identity ing decreases in pollinator trait diversity (Forrest et al., 2015; Geslin and abundance of specific traits (functional identity) but also by the et al., 2016). Trait diversity is crucial, as it allows for a variety of re- diversity of traits (functional diversity; Leps et al., 2006). sponses to disturbances over space and time (Mori et al., 2013). Pollinator diversity and abundance have declined over the last As for effect traits, functional composition appears to be a better century, and agricultural intensification is considered one of the predictor of pollination function than taxonomic composition (Gagic main drivers of these declines (IPBES, 2016). Agricultural intensifi- et al., 2015). Pollinator traits such as body size, flower-­handling be- cation affects pollinators through landscape simplification, includ- haviour and hairiness have been associated with pollination success ing isolation and loss of natural and semi-­natural habitats, leading to (Phillips et al., 2018; Roquer-­Beni et al., 2020; Stavert et al., 2016). In ad- decreased availability of feeding resources and nesting substrates dition, based on the complementarity hypothesis (Díaz & Cabido, 2001; (Roulston & Goodell, 2011; Shuler et al., 2005). In addition, the in- Tilman, 2001), communities with high trait diversity should be better creased use of pesticides associated with intensive agriculture has suited to provide pollination services under a variety of environmental direct negative effects on pollinator fitness and survival (Woodcock scenarios (Blüthgen & Klein, 2011; Woodcock et al., 2019). et al., 2017). To reverse these effects and enhance on-­farm biodiver- Some studies have addressed the response of pollinator functional sity, agri-­environmental measures have been promoted at both local composition to agricultural intensification (De Palma et al., 2015; Forrest and landscape scales (Primdahl et al., 2003). These measures include et al., 2015; Geslin et al., 2016; Rader et al., 2014; Williams et al., 2010) reducing pesticide use, preserving historical land uses and implement- whereas others have addressed the effects of functional composition ing agri-­environmental structures (hereafter AES) such as hedgerows on pollination service (Gagic et al., 2015; Hoehn et al., 2008; Woodcock and buffer strips to increase connectivity with semi-­natural habitats et al., 2019). However, studies analysing pollinator responses and effects (Ekroos et al., 2016). The effects of agri-­environmental measures on simultaneously remain scarce (Bartomeus et al., 2018; Klein et al., 2008). ROQUER-­BENI et al. Journal of Applied Ecology | 2845 To address this knowledge gap, we measured functional trait identity surrounding the orchard. To estimate flower diversity (Shannon's and diversity of apple pollinator assemblages in 110 orchards differing in index), we used the mean cover of each flower species. Quadrat size management and in local and landscape features in four European apple-­ differences among regions were not expected to affect comparative growing regions. Apples Malus domestica are one of the most important trends in flower diversity because region variability was integrated in fruit crops worldwide and are highly dependent on insect pollination the statistical design (see below). (Garratt et al., 2016; Martins et al., 2014). We used a response–­effect trait framework to determine how local and landscape features affect pollination service through changes in pollinator functional composition. 2.3 | Landscape features Our objectives were: (a) to identify functional metrics responding to en- vironmental features and affecting pollination services; b) to establish We used ArcView 10.3.1, MiraMon v8.2e, R 3.2.3 (R Core Team, 2015) whether traits that determine pollinator responses to changes in local and digital maps (see Table S1) to measure the area covered by dif- and landscape factors overlap with traits that affect pollination; (c) to de- ferent habitat types within 1-­km buffers from the surveyed trees and termine whether pollination services are dependent on certain pollinator calculated two landscape variables. First, the per cent area occupied functional traits and/or on trait diversity; (d) to establish whether the by pollinator-­friendly habitats (henceforth PFH), defined as habitats relationship between pollinator trait composition and pollination service free of pesticides and hosting abundant floral resources and potential is dependent on orchard management. bee nesting substrates. PFH included shrublands, orchard meadows, semi-­natural grasslands, abandoned orchards and hedgerows. Second, the area occupied by orchards, as a proxy of agricultural landscape 2 | M ATE R I A L S A N D M E TH O DS homogenization (Table S1). 2.1 | Study sites 2.4 | Pollinator surveys The study was conducted in 2015 in 110 commercial apple orchards from four European apple-­growing regions: Skåne (Sweden), Baden-­ Pollinator surveys were conducted during apple bloom (April–­May). Württemberg (Germany), Asturias (Spain) and Catalonia (Spain) (see In Sweden, Germany and Catalonia, observers walked along the two Figure S1, Table S1; Happe et al., 2019; Miñarro & García, 2018; 20-­m transects (M ± SE: 35 ± 1.3 trees/transect) and recorded all apple Samnegård, Alins, et al., 2019 for further details on the study orchards). visitors contacting the reproductive parts of apple flowers. Transect In each region, orchards were selected to encompass a range of walks lasted 5 min and were repeated three times throughout the day, local and landscape features (Table S1). In Sweden, Germany and amounting to 30 min of pollinator survey per orchard. In Asturias, ob- Catalonia, half of the orchards followed IOBC guidelines for integrated servers surveyed a 1-­m diameter canopy area for 5 min in five trees per pest management (Cross, 2002) and used chemical insecticides, fun- orchard three times throughout the day, for a total of 75 survey min- gicides, fertilizers and herbicides (see Happe et al., 2019 for details). utes per orchard. Pollinators were mostly visually identified in the field, These orchards were considered high-­intensity (henceforth HI) or- but some specimens were captured and identified in the laboratory. chards. The rest of the orchards in Sweden, Germany and Catalonia From these surveys, we calculated abundance of each pollinator and all orchards in Asturias were either certified organic or followed species (number of individuals observed visiting flowers) and pollina- close-­to-­organic guidelines, with very low levels of synthetic inputs tor visitation rate (number of visits/100 flowers/5 min). and mechanical weed control. These orchards were grouped into a low-­intensity orchard management category (henceforth LI). 2.5 | Pollination service 2.2 | Local features At the onset of bloom, we marked two to three branches per tree on six to seven trees per orchard and counted the number of flower buds In each orchard, we established two consecutive 20-­m transects on (1,200–­1,300 per orchard). About 3 weeks after petal fall, we assessed which we conducted pollinator counts (see below). To assess local fea- initial fruit set as the percentage of flowers that developed into a fruit- tures, we used aerial photographs combined with on-­site inspection let. Initial fruit set is a better measure of pollination service than fruit to measure the area occupied by AES within a 20-­m buffer from the set at harvest because it is less influenced by post-­pollination factors first transect trees. AES included hedgerows, forests fallow lands, or- such as natural and/or artificially induced fruit abscission. chard meadows and semi-­natural grasslands (Table S1). During apple bloom, we estimated flower cover and diversity of entomophilous plant species within and around each orchard. These measures were 2.6 | Pollinator traits taken using 12–­14 gridded quadrats (1 m2 in Sweden, Germany and Catalonia; 0.25 m2 in Asturias). Half of the quadrats were laid between We selected pollinator traits that, based on our knowledge and/or the two orchard transects, and the other half within the 20-­m buffer previous studies (Table S2), could either influence the response of 2846 | Journal of Applied Ecology ROQUER-­BENI et al. pollinators to environmental conditions (response traits, hereafter R), 2.8.1 | Response traits pollination service (effect traits, hereafter E) or both (hereafter R-­E). All pollinator species recorded (99 species, Table S3) were characterized To assess the response of pollinator functional composition to local with three traits: body length (R-­E), hairiness (R-­E) and larval diet (R-­E). and landscape features, we performed separate linear mixed-­effect Because bees are the most frequent and effective apple pollinators models (LMMs) for each functional composition metric (CWM of (Garratt et al., 2016; Vicens & Bosch, 2000a), we conducted separate each trait and RaoQ of all traits combined). These analyses were analyses including only bees. We measured eight traits in this group done separately for all pollinator species and for bee species only. (45 species): intertegular span (a proxy of body size; R-­E), hairiness Full models included three local variables (orchard management, (R-­E), proboscis length (E), forewing aspect ratio (maximum length/ AES cover, flower diversity) and two landscape variables (orchard maximum width; R), sociality (R), voltinism (R), nesting substrate (R) cover, PFH cover) as fixed effects, and region as a random effect. and pollen transportation structure (E) (see Table S2 for details). Numerical explanatory variables were not highly correlated (r < 0.7, Quantitative traits were measured on pinned specimens from the Table S5) and VIF was <5 for all full models. survey orchards. Categorical traits were based on literature records and our expert knowledge (see Table S2 for methods and sample sizes). We worked with a single mean value per trait and species. We 2.8.2 | Effect traits explored correlation between each pair of numerical traits (Table S4). In bees, proboscis length was highly correlated to body size (r > 0.7). To analyse the effect of functional traits on pollination service, Thus, we only used body size in the analyses. we conducted LMMs with initial fruit set as the response variable. Because all pollinators potentially contribute to fruit set, these analy- ses included honeybees. To discriminate between trait identity and 2.7 | Functional composition metrics trait diversity effects, we run one model with trait CWMs and an- other with RaoQ as predictor variables. We checked for correlation For each orchard, we calculated the community-­weighted mean between pairs of predictors and excluded variables as required until (CWM) of each trait, a measure of functional identity (Garnier VIF was <5 (Table S6; Zuur et al., 2010). Whenever two predictors et al., 2004). CWM of numerical traits was calculated as the mean were strongly correlated, we kept the one that we considered more value across species weighted by the relative abundance of each spe- likely to be associated with pollination services. As a result, CWM cies (Ricotta & Moretti, 2011). CWM of categorical traits was calcu- of hairiness and pollinivorous larvae and RaoQ were the only func- lated as the proportion of individuals belonging to each trait category. tional predictor variables included. These models also included two To measure functional diversity, we used the Rao quadratic diversity additional variables that could affect initial fruit set: overall pollinator index (RaoQ), which measures the dissimilarity between two randomly visitation rate and orchard management. To establish whether the selected individuals and calculates the sum of weighted abundance relationship between functional composition and pollination ser- dissimilarity between each pair of species (Rao, 1982). The RaoQ index vices was management dependent, these models also included the has some advantages over other functional diversity metrics: (a) It is interaction between management and the selected functional com- easy to interpret (it ranges from 0 to the maximum of Simpson's diver- position metrics. Finally, apple variety and region were included as sity index); (b) in general, it is relatively unaffected by species richness independent random effects. Fruit set data were not available for (Botta-­Dukát, 2005) and (c) it can be used to measure both single-­and German orchards, so n = 81 orchards for these analyses. multi-­trait diversity (Ricotta & Moretti, 2011). Following model selection procedures (MuMin package; Barton & Honeybees were excluded from calculations of response trait met- Barton, 2019), we tested all possible explanatory variable combinations rics because their presence was mostly (or solely) attributable to man- (see above) through a multimodel inference approach (Anderson & aged colonies in all four regions. Conversely, honeybees were included Burnham, 2004). We then used a model averaging approach (with av- in calculations of effect trait metrics because they affect apple polli- eraged variable coefficients) based on AICc to assign a relative impor- nation (Vicens & Bosch, 2000b). Measures of CWM and RaoQ were tance to each variable. Models with ΔAICc < 2 were considered equally conducted using the dbFD and functomp functions from FD library suitable (see Appendix S1 for details on model selection procedures). (Laliberté & Legendre, 2010) with R (see Appendix S1 for details on Normality and homoscedasticity assumptions were graphically functional composition metric calculations). In addition to the RaoQ evaluated by plotting the distribution of residuals of each model. index, we also measured trait diversity with the Functional Divergence Response variables were square-­root-­ or log-­transformed as Index (FDiv; Villéger et al., 2008). The results were similar to those ob- needed. Numerical explanatory variables were standardized to fa- tained with the RaoQ and are provided in the Supporting Information. cilitate comparison across variables. To detect model outliers, we calculated Cook's distance and excluded sites with distances >4/N (Cook, 1977). The exclusion of outliers provided better model ad- 2.8 | Statistical analysis justments, but trends remained similar (Tables S7 and S8). To rule out spatial autocorrelation, we applied a Moran's I test with the re- All statistical analyses were conducted with R v3.2.3. siduals of all our models (Table S9). ROQUER-­BENI et al. TA B L E 1 Statistical outputs of model averaging relating wild pollinator and wild bee functional composition response variables to local and landscape features. Estimated coefficients, their 95% intervals (in parentheses) and relative importance (in brackets) are provided. Significant terms are in bold. ‘—­’ denotes variables not appearing in the model average. R2m and R2c are the range of marginal and conditional R 2 of the best-­fitted models respectively. R 2 of the best model is indicated in parentheses. ‘Sites’ indicates the number of orchards included in the model after outlier exclusion (Table S7) % Pollinator friendly Response variable Management* Flower diversity AES cover % Orchard cover habitat cover R2m R2c Sites All pollinators CWM body length —­ 0.145 [0.31] —­ −0.683 [1](−1.078, 0.986 [1](0.423, 1.550) 0.28–­0.29 0.28–­0.29 98 (−0.201, 0.491) −0.288) (0.28) (0.28) CWM Hairinessa —­ —­ −0.397 [0.44] −1.046 [1](−1.627, 0.676 [0.83](0.054, 0.16–­0.23 0.28–­0.35 99 (−0.889, 0.095) −0.465) 1.299) (0.23) (0.35) CWM Pollinivorous larvae 0.026 [0.18] —­ —­ −0.031 [0.33] 0.154 [1](0.073, 0.235) 0.11–­0.13 0.56–­0.57 94 (−0.076, 0.129) (−0.083, 0.021) (0.11) (0.56) CWM Insectivorous larvaea −0.026 [0.14] 0.033 [0.84] 0.075 [1](0.040, 0.110) 0.023 [0.20] −0.017 [0.17] 0.14–­0.17 0.36–­0.39 99 (−0.099, 0.047) (−0.000, 0.066) (−0.018, 0.063) (−0.050, 0.017) (0.16) (0.38) Bees RaoQ 4.946 [1](0.655, 9.236) —­ —­ −3.478 [1](−5.704, 1.657 [0.36] 0.18–­0.20 0.24–­0.26 105 −1.252) (−1.488, 4.802) (0.18) (0.24) Note: aSquare-­root Log(X + 1). *Low intensity: reference level of management. Journal of Applied Ecology | 2847 2848 | Journal of Applied Ecology ROQUER-­BENI et al. 3 | R E S U LT S orchards but not in HI orchards (Table 2). When trait diversity was low, initial fruit set was lower in LI orchards, but when trait diver- We recorded 8,253 pollinator individuals visiting apple flowers. sity was high, initial fruit set was similar in the two orchard types. In Most pollinators were honeybees (77.6%), followed by wild bees other words, differences between LI and HI orchards in initial fruit (9.1%), hoverflies (5.9%), other flies (6.1%), beetles (1%) and others set disappeared as pollinator trait diversity increased (Figure 2). (0.4%) (Table S3). Honeybee visitation rates were similar in LI and Models computed with FDiv showed similar results (Tables S12 HI orchards (LMM, estimate β = 0.08, p = 0.24), but overall pollina- and S13), except that FDiv was not affected by management and was tor visitation rates were higher in LI orchards (β = 0.15, p < 0.05; enhanced by PFH cover. Table S10). 4 | D I S CU S S I O N 3.1 | Response traits Our results show that bee trait diversity was enhanced by LI orchard At the local scale, AES cover favoured pollinators with insectivorous management and negatively affected by landscape orchard cover. larvae (Table 1). There were no effects of local features, landscape At the same time, pollinator trait diversity enhanced pollination ser- features or their interactions on pollinator trait diversity (RaoQ). vice in LI, but not in HI orchards. Our results provide evidence that Landscape orchard cover was negatively associated with body length pollinator functional diversity is an important mechanism linking re- and hairiness CWMs (Table 1). Pollinator-­friendly habitat cover was sponses to agricultural intensification and contribution to pollination positively associated with body length, hairiness and proportion of services in agricultural systems. pollinators with pollinivorous larvae (bees). Local factors were important determinants of pollinator func- When considering bees only, the best-­f itted models for all tional composition. AES enhanced the abundance of pollinators with CWM variables were the null models. Thus, we could not attribute insectivorous larvae, probably by providing prey for aphidophagous any effect of local or landscape features to any specific bee trait hoverflies and predatory wasps (Rodríguez-­Gasol et al., 2019). Some (Table 1). However, local and landscape features had important studies have reported AES to also benefit pollinators with pollini- effects on functional diversity, which was negatively affected by vorous larvae (bees; Blaauw & Isaacs, 2014). However, we found landscape orchard cover (Figure 1a) and enhanced by LI manage- no such effect, suggesting that floral resources may not be limiting ment (Figure 1b). during apple bloom. LI management enhanced bee trait diversity. Various studies show positive effects of organic farming on bee abundance and tax- 3.2 | Effect traits onomic diversity, particularly in homogeneous landscapes, and these results are attributed to reduced use of chemical inputs (Forrest Models including CWM metrics showed that initial fruit set was en- et al., 2015; Rundlöf et al., 2008). As far as we know, effects of agri- hanced by HI management but was not affected by any particular cultural management on bee trait diversity have not been previously trait (Table 2). Models including RaoQ revealed an interaction effect reported. Our results show that the increase in species richness between trait diversity and management on initial fruit set (Figure 2). caused by LI management also results in an increase in functional There was a positive effect of trait diversity on initial fruit set in LI diversity. F I G U R E 1 Effects of landscape orchard cover (a) and orchard management (high intensity vs. low intensity) (b) on bee multi-­trait functional diversity (RaoQ). Grey bands indicate 95% confidence intervals. Vertical bars indicate standard deviations ROQUER-­BENI et al. TA B L E 2 Statistical outputs of model averaging relating initial fruit set to management (low intensity vs. high intensity), functional composition metrics, the interaction between management and functional composition metrics and pollinator visitation rate. The first model includes single-­trait metrics (CWM of hairiness and pollinivorous larvae). The second model includes functional diversity (multi-­trait RaoQ). Estimated coefficients, their 95% intervals (in parentheses) and relative importance (in brackets) are provided. Significant terms are in bold. ‘—­’ denotes variables not appearing in the model average. R2m and R2c are the marginal and conditional R 2 range of the best-­fitted model respectively. R 2 of the best model is indicated in parentheses. ‘Sites’ indicates the number of orchards included in the model after outlier exclusion (Table S8) CWM CWM pollinivorous hairiness x x Response variable Management* CWM hairiness CWM pollinivorous management management Visitation rate R2m R2c Sites All pollinators Initial fruit set a −1.617 [1](−2.304, 0.113 [0.20] —­ —­ —­ 0.154 [0.25] 0.24–­ 0.24–­ 76 −0.930) (−0.251, 0.477) (−0.193, 0.500) 0.25 0.25 (0.24) (0.24) Response variable Management* RaoQ RaoQ × management Visitation rate R2m R2c Sites a Initial fruit set −1.580 [1](−2.193, −0.966) −0.301 0.749 [0.64](0.089, 1.409) —­ 0.27–­ 0.27–­ 76 [0.64] 0.32 0.32 (−0.821, (0.32) (0.32) 0.219) Note: Data transformations: aSquare-­root. *Low intensity: reference level of management. Journal of Applied Ecology | 2849 2850 | Journal of Applied Ecology ROQUER-­BENI et al. hindering consistent spatio-­temporal availability of flower resources not only during but especially before and after apple bloom (Marini et al., 2012). In sum, our results support the need of landscape level initiatives to promote bee communities in agroecosystems (Cole et al., 2020). HI orchards had higher initial fruit set than LI orchards. Because pollinator visitation rates were higher in LI orchards, we cannot attribute this result to lower pollination levels in LI orchards. We find two possible explanations for the higher initial fruit set in HI orchards. First, the use of chemical fertilizers and lower levels of pests and diseases may enhance fruitlet retention in HI orchards (Peck et al., 2006; Samnegård et al., 2019). Second, to enhance the production of large apples, fruit load is usually chemically thinned. In HI orchards, synthetic thinners are applied during fruitlet growth (Fallahi & Greene, 2010) whereas in LI orchards, organic thinners are applied during bloom (Lordan et al., 2018), effectively lowering the numbers of flowers available for fruitlet development. No specific pollinator traits emerged as important determi- nants of pollination service. We found that trait diversity enhanced F I G U R E 2 Effects of pollinator multi-­trait diversity (RaoQ, initial fruit set, but only in LI orchards. The positive effect of trait including wild and managed pollinators) on initial fruit set (square-­ diversity on pollination services has been previously recognized root transformed) in low-­intensity (estimate β = 0.192; p = 0.05) (Hoehn et al., 2008; Martins et al., 2014; Woodcock et al., 2019), and high-­intensity (β = −0.108; p = 0.213) orchards (Table S11). but ours is, as far as we know, the first study showing interactive Grey bands indicate 95% confidence intervals effects between trait diversity and agricultural management on pol- lination services. Functionally diverse pollinator communities may Landscape features also affected pollinator functional compo- enhance pollination function through various spatial and temporal sition. Large and hairy pollinators were negatively affected by or- complementarity mechanisms (Blüthgen & Klein, 2011). Our results chard cover and enhanced by pollinator-­friendly habitat cover. The show that, through the enhancement of pollinator trait diversity, response of body size and other traits to landscape intensification LI orchards may reach levels of initial fruit set similar to those of is controversial and appears to be context dependent. According HI orchards. Importantly, the positive effect of bee trait diversity to some studies, small bee species are more sensitive to landscape on initial fruit set in LI orchards was detected despite a very strong intensity and isolation from natural habitats (De Palma et al., 2015; background of honeybee visitation. Klein et al., 2008). However, other studies show opposite trends Some pollinator traits responded to local and/or landscape fac- (Larsen et al., 2005; Rader et al., 2014) or no clear patterns tors but, as mentioned, no specific traits influenced pollination ser- (Bartomeus et al., 2018; Forrest et al., 2015; Williams et al., 2010). vices. In other words, pollination service could not be explained by We cannot think of any reason why hairier pollinators should be sampling effects of dominant traits (Mokany et al., 2008). A recent more abundant in semi-­natural habitats than in agricultural areas. study shows that pollen deposition in apple flowers is influenced We also found that landscapes with high levels of pollinator-­ by various morphological and behavioural pollinator traits, possi- friendly habitat promoted the relative abundance of pollinators with bly diluting the effect of any single trait by itself (L. Roquer-­Beni, pollinivorous larvae (bees). Semi-­natural habitats provide spatio-­ unpublished data). Bee trait diversity responded negatively to two temporal stability to flower visitors (Klein, 2009) through wide- important features associated with agricultural intensification: HI spread and continued accessibility to flower resources, especially management and increased orchard cover, and pollinator trait diver- before and after the crop's flowering period. Pollinator-­friendly sity enhanced initial fruit set in LI orchards. These results suggest habitats also provide nesting resources and pesticide-­free refuge that the response–­effect framework is more relevant for integrative areas (Holzschuh et al., 2010), thus promoting the colonization and measures (multi-­trait diversity) than single traits (Peña et al., 2020). establishment of bee communities in agricultural areas (Kremen Our results have important implications in the face of the new et al., 2004). Bee trait diversity was not affected by pollinator-­ European Common Agricultural Policy, one of whose strategic ob- friendly habitat in our study, but it decreased with landscape or- jectives is the preservation of landscapes and biodiversity (European chard cover. Bee trait diversity is known to be lower in farmland Commission, 2019). First, efforts to promote functional biodiver- compared to natural habitats (Forrest et al., 2015; Hass et al., 2018; sity provide greater returns in low-­intensity than in high-­intensity Woodcock et al., 2014). In addition to higher levels of pesticide ex- farms. Second, the fact that high levels of ecosystem services can posure, orchard-­dominated landscapes are characterized by land- be reached in low-­intensity farms (as long as functional diversity is scape homogenization and isolation from semi-­natural habitats, thus preserved) provides a compelling argument for the conversion of ROQUER-­BENI et al. Journal of Applied Ecology | 2851 high-­intensity into low-­intensity farms. Measures promoting pollina- Bartomeus, I., Cariveau, D. P., Harrison, T., & Winfree, R. (2018). On the inconsistency of pollinator species traits for predicting either re- tor functional diversity (e.g. reduction of chemical inputs, implemen- sponse to land-­use change or functional contribution. Oikos, 127(2), tation of AES enhancing diversified nesting and flower resources, 306–­315. https://doi.org/10.1111/oik.04507 increased landscape heterogeneity and connectivity to semi-­natural Barton, K., & Barton, M. K. (2019). Package ‘MuMIn’. Multi-­model infer- habitats) could provide environmental benefits and, at the same ence. Version, 1(6). Blaauw, B. R., & Isaacs, R. (2014). Flower plantings increase wild bee time, reduce farmer dependence on costly external inputs while abundance and the pollination services provided to a pollination-­ maintaining competitive production. dependent crop. Journal of Applied Ecology, 51(4), 890–­898. https:// doi.org/10.1111/1365-­2664.12257 AC K N OW L E D G E M E N T S Blüthgen, N., & Klein, A. M. (2011). Functional complementarity and spe- We thank all producers for permission to work in their orchards and cialisation: The role of biodiversity in plant-­pollinator interactions. Basic and Applied Ecology, 12(4), 282–­291. https://doi.org/10.1016/j. several fruit associations (ADV-­Ponent, ADV-­Fluvià, ACTEL, KOB baae.2010.11.001 Bavendorf, FÖKO e.V. Äppelriket in Kivik) for their advice and as- Bommarco, R., Biesmeijer, J. C., Meyer, B., Potts, S. G., Pöyry, J., Roberts, sistance. We are very grateful to I. Fraile, S. Muntada and several S. P. M., Steffan-­Dewenter, I., & Ockinger, E. (2010). Dispersal students for their invaluable help. This research (EcoFruit project) capacity and diet breadth modify the response of wild bees to habitat loss. Proceedings of the Royal Society B: Biological Sciences, was funded through the 2013–­2014 BiodivERsA/FACCE-­JPI joint 277(1690), 2075–­2082. https://doi.org/10.1098/rspb.2009.2221 call (2014-­74), Spanish MinECo (PCIN-­2014-­145-­C02), German Botta-­Dukát, Z. (2005). Rao's quadratic entropy as a measure of functional BMBF (PT-­DLR/BMBF, 01LC1403) and Swedish Research Council diversity based on multiple traits. Journal of Vegetation Science, 16(5), Formas (2014-­1784) by Formas (2013-­934 to M.T.), Stiftelsen 533–­540. https://doi.org/10.1111/j.1654-­1103.2005.tb023​93.x Brittain, C., & Potts, S. G. (2011). The potential impacts of insecticides Lantbruksforskning (H1256150 to M.P.), INIA (RTA2013-­0 0039-­ on the life-­history traits of bees and the consequences for polli- C03-­0 0 to G.A. and M.M.), MinECo/FEDER (CGL2015-­68963-­C2-­ nation. Basic and Applied Ecology, 12(4), 321–­331. https://doi. 2-­R to D.G.), FI-­AGAUR (to L.R.-­B.) and MinECo (RYC-­2015-­18448 org/10.1016/J.BAAE.2010.12.004 to X.A.). The use of IACS (Sweden) was developed within projects Cole, L. J., Kleijn, D., Dicks, L. V., Stout, J. C., Potts, S. G., Albrecht, M., Balzan, M. V., Bartomeus, I., Bebeli, P. J., Bevk, D., Biesmeijer, J. SAPES and MULTAGRI and adapted by M. Stjernman and P. Olsson. C., Chlebo, R., Dautartė, A., Emmanouil, N., Hartfield, C., Holland, J. M., Holzschuh, A., Knoben, N. T. J., Kovács-­Hostyánszki, A., C O N FL I C T O F I N T E R E S T … Scheper, J. (2020). A critical analysis of the potential for EU The authors declare that there is no conflict of interest. Common Agricultural Policy measures to support wild pollinators on farmland. Journal of Applied Ecology, 57(4), 681–­694. https://doi. org/10.1111/1365-­2664.13572 AU T H O R S ' C O N T R I B U T I O N S Cook, R. D. (1977). Detection of influential observation in linear regres- A.-­M.K., A.-­K .H., A.R., D.G., J.B., K.M., L.R.-­B., M.M., P.A.H., U.S., sion. Technometrics, 19(1), 15–­18. V.B. and X.A. conceived and designed the study; A.-­K .H., D.G., G.A., Cross, J. V. (2002). Guidelines for integrated production of pome fruits in J.B., L.R.-­B., M.M., M.P., M.T., U.S. and V.B. collected the data; L.R.-­B. Europe. Bulletin OILB SROP, 25, 8. De Palma, A., Kuhlmann, M., Roberts, S. P. M., Potts, S. G., Börger, led data analysis and manuscript writing with assistance from A.R., L., Hudson, L. N., Lysenko, I., Newbold, T., & Purvis, A. (2015). X.A. and J.B. All authors contributed to draft development and gave Ecological traits affect the sensitivity of bees to land-­use pressures final approval for publication. in European agricultural landscapes. Journal of Applied Ecology, 52(6), 1567–­1577. https://doi.org/10.1111/1365-­2664.12524 Díaz, S., & Cabido, M. (2001). Vive la différence: Plant functional diversity DATA AVA I L A B I L I T Y S TAT E M E N T matters to ecosystem processes. Trends in Ecology & Evolution, 16(11), Data available via the Dryad Digital Repository https://doi. 646–­655. https://doi.org/10.1016/S0169​-­5347(01)02283​-­2 org/10.5061/dryad.63xsj​3v39 (Roquer-­Beni et al., 2021). Ekroos, J., Ödman, A. M., Andersson, G. K. S., Birkhofer, K., Herbertsson, L., Klatt, B. K., Olsson, O., Olsson, P. A., Persson, A. S., Prentice, H. C., Rundlöf, M., & Smith, H. G. (2016). Sparing land for biodiversity ORCID at multiple spatial scales. Frontiers in Ecology and Evolution, 3, 145. Laura Roquer-­Beni https://orcid.org/0000-0001-8454-6745 https://doi.org/10.3389/fevo.2015.00145 Xavier Arnan https://orcid.org/0000-0002-9904-274X European Commission. (2019). The post-­2020 common agricultural policy: Daniel García https://orcid.org/0000-0002-7334-7836 Environmental benefits and simplification. European Commission. Fallahi, E., & Greene, D. W. (2010). The impact of blossom and postbloom Peter A. Hambäck https://orcid.org/0000-0001-6362-6199 thinners on fruit set and fruit quality in apples and stone fruits. Alexandra-­Maria Klein https://orcid.org/0000-0003-2139-8575 Acta Horticulturae, 884, 179–­187. https://doi.org/10.17660/​ActaH​ Marcos Miñarro https://orcid.org/0000-0002-5851-6873 ortic.2010.884.20 Anselm Rodrigo https://orcid.org/0000-0001-6341-0363 Forrest, J. R. K., Thorp, R. W., Kremen, C., & Williams, N. M. (2015). Contrasting patterns in species and functional-­trait diversity of Marco Tasin https://orcid.org/0000-0001-7379-4954 bees in an agricultural landscape. Journal of Applied Ecology, 52(3), Jordi Bosch https://orcid.org/0000-0002-8088-9457 706–­715. https://doi.org/10.1111/1365-­2664.12433 Gagic, V., Bartomeus, I., Jonsson, T., Taylor, A., Winqvist, C., Fischer, C., REFERENCES Slade, E. M., Steffan-­Dewenter, I., Emmerson, M., Potts, S. G., Anderson, D., & Burnham, K. (2004). Model selection and multi-­model in- Tscharntke, T., Weisser, W., & Bommarco, R. (2015). Functional ference (Vol. 63, 2nd ed.). Springer-­Verlag. identity and diversity of animals predict ecosystem functioning 2852 | Journal of Applied Ecology ROQUER-­BENI et al. better than species-­based indices. Proceedings of the Royal Society B: Leps, J., de Bello, F., Lavorel, S., & Berman, S. (2006). Quantifying and Biological Sciences, 282(1801), 20142620. https://doi.org/10.1098/ interpreting functional diversity of natural communities: Practical rspb.2014.2620 considerations matter. Preslia, 78, 481–­501. Garnier, E., Cortez, J., Billès, G., Navas, M. L., Roumet, C., Debussche, Lordan, J., Alins, G., Àvila, G., Torres, E., Carbó, J., Bonany, J., & Alegre, M., Laurent, G., Blanchard, A., Aubry, D., Bellmann, A., Neill, C., & S. (2018). Screening of eco-­friendly thinning agents and adjusting Toussaint, J. P. (2004). Plant functional markers capture ecosystem mechanical thinning on ‘Gala’, ‘Golden Delicious’ and ‘Fuji’ apple properties during secondary succession. Ecology, 85(9), 2630–­2637. trees. Scientia Horticulturae, 239(January), 141–­155. https://doi. https://doi.org/10.1890/03-­0799 org/10.1016/j.scien​t a.2018.05.027 Garratt, M. P. D., Breeze, T. D., Boreux, V., Fountain, M. T., McKerchar, M., Marini, L., Quaranta, M., Fontana, P., Biesmeijer, J. C., & Bommarco, R. Webber, S. M., Coston, D. J., Jenner, N., Dean, R., Westbury, D. B., (2012). Landscape context and elevation affect pollinator commu- Biesmeijer, J. C., & Potts, S. G. (2016). Apple pollination: Demand de- nities in intensive apple orchards. Basic and Applied Ecology, 13(8), pends on variety and supply depends on pollinator identity. PLoS ONE, 681–­689. https://doi.org/10.1016/j.baae.2012.09.003 11(5), e0153889. https://doi.org/10.1371/journ​al.pone.0153889 Marja, R., Kleijn, D., Tscharntke, T., Klein, A., Frank, T., & Batáry, P. (2019). Geslin, B., Oddie, M., Folschweiller, M., Legras, G., Seymour, C. L., van Veen, F. Effectiveness of agri-­environmental management on pollinators is J. F., & Thébault, E. (2016). Spatiotemporal changes in flying insect abun- moderated more by ecological contrast than by landscape structure dance and their functional diversity as a function of distance to natural or land-­use intensity. Ecology Letters, 22(9), 1493–­1500. https://doi. habitats in a mass flowering crop. Agriculture, Ecosystems & Environment, org/10.1111/ele.13339 229, 21–­29. https://doi.org/10.1016/J.AGEE.2016.05.010 Martins, K. T., Gonzalez, A., & Lechowicz, M. J. (2014). Pollination ser- Happe, A. K., Alins, G., Blüthgen, N., Boreux, V., Bosch, J., García, D., vices are mediated by bee functional diversity and landscape con- Hambäck, P. A., Klein, A. M., Martínez-­Sastre, R., Miñarro, M., Müller, A. text. Agriculture, Ecosystems and Environment, 200(November), 12–­ K., Porcel, M., Rodrigo, A., Roquer-­Beni, L., Samnegård, U., Tasin, M., & 20. https://doi.org/10.1016/j.agee.2014.10.018 Mody, K. (2019). Predatory arthropods in apple orchards across Europe: McGill, B. J., Enquist, B. J., Weiher, E., & Westoby, M. (2006). Rebuilding Responses to agricultural management, adjacent habitat, landscape community ecology from functional traits. Trends in Ecology & Evolution, composition and country. Agriculture, Ecosystems and Environment, 21(4), 178–­185. https://doi.org/10.1016/j.tree.2006.02.002 273, 141–­150. https://doi.org/10.1016/j.agee.2018.12.012 Miñarro, M., & García, D. (2018). Complementarity and redundancy in Hass, A. L., Liese, B., Heong, K. L., Settele, J., Tscharntke, T., & Westphal, the functional niche of cider apple pollinators. Apidologie, 49(6), C. (2018). Plant-­pollinator interactions and bee functional diversity 789–­8 02. https://doi.org/10.1007/s1359​2-­018-­0600-­4 are driven by agroforests in rice-­dominated landscapes. Agriculture, Mokany, K., Ash, J., & Roxburgh, S. (2008). Functional identity is more Ecosystems & Environment, 253, 140–­147. https://doi.org/10.1016/J. important than diversity in influencing ecosystem processes in a AGEE.2017.10.019 temperate native grassland. Journal of Ecology, 96(5), 884–­893. Hoehn, P., Tscharntke, T., Tylianakis, J. M., & Steffan-­Dewenter, I. (2008). https://doi.org/10.1111/J.1365-­2745.2008.01395.x Functional group diversity of bee pollinators increases crop yield. Mori, A. S., Furukawa, T., & Sasaki, T. (2013). Response diversity determines Proceedings of the Royal Society B: Biological Sciences, 275(1648), the resilience of ecosystems to environmental change. Biological 2283–­2291. https://doi.org/10.1098/rspb.2008.0405 Reviews, 88(2), 349–­364. https://doi.org/10.1111/brv.12004 Holzschuh, A., Steffan-­Dewenter, I., & Tscharntke, T. (2010). How do Peck, G. M., Andrews, P. K., Reganold, J. P., & Fellman, J. K. (2006). landscape composition and configuration, organic farming and Apple orchard productivity and fruit quality under organic. fallow strips affect the diversity of bees, wasps and their par- Conventional, and Integrated Management, 41(1), 99–­107. https://doi. asitoids? Journal of Animal Ecology, 79(2), 491–­500. https://doi. org/10.21273/​HORTS​CI.41.1.99 org/10.1111/j.1365-­2656.2009.01642.x Peña, R., Schleuning, M., Donoso, I., Rodríguez-­Pérez, J., Dalerum, IPBES. (2016). The assessment report of the Intergovernmental Science-­ F., & García, D. (2020). Biodiversity components mediate the re- Policy Platform on Biodiversity and Ecosystem Services on pollinators, sponse to forest loss and the effect on ecological processes of pollination and food production. IPBES. plant–­frugivore assemblages. Functional Ecology, 34(6), 1257–­1267. Jauker, B., Krauss, J., Jauker, F., & Steffan-­Dewenter, I. (2013). Linking life history https://doi.org/10.1111/1365-­2435.13566 traits to pollinator loss in fragmented calcareous grasslands. Landscape Phillips, B. B., Williams, A., Osborne, J. L., & Shaw, R. F. (2018). Shared Ecology, 28(1), 107–­120. https://doi.org/10.1007/s1098​0-­012-­9820-­6 traits make flies and bees effective pollinators of oilseed rape Klein, A. M. (2009). Nearby rainforest promotes coffee pollination (Brassica napus L.). Basic and Applied Ecology, 32, 66–­76. https://doi. org/10.1016/j.baae.2018.06.004 by increasing spatio-­temporal stability in bee species richness. Primdahl, J., Peco, B., Schramek, J., Andersen, E., & Oñate, J. J. (2003). Forest Ecology and Management, 258(9), 1838–­1845. https://doi. Environmental effects of agri-­environmental schemes in Western org/10.1016/j.foreco.2009.05.005 Europe. Journal of Environmental Management, 67(2), 129–­138. Klein, A. M., Cunningham, S. A., Bos, M., & Steffan-­Dewenter, I. (2008). https://doi.org/10.1016/S0301​- ­4797(02)00192​-­5 Advances in pollination ecology from tropical plantation crops. R Core Team. (2015). R: A language and environment for statistical comput- Ecology, 89(4), 935–­943. https://doi.org/10.1890/07-­0 088.1 ing. R Foundation for Statistical Computing. Kremen, C., Williams, N. M., Bugg, R. L., Fay, J. P., & Thorp, R. W. (2004). Rader, R., Bartomeus, I., Tylianakis, J. M., & Laliberté, E. (2014). The win- The area requirements of an ecosystem service: Crop pollination by ners and losers of land use intensification: Pollinator community dis- native bee communities in California. Ecology Letters, 7(11), 1109–­ assembly is non-­random and alters functional diversity. Diversity and 1119. https://doi.org/10.1111/j.1461-­0248.2004.00662.x Distributions, 20(8), 908–­917. https://doi.org/10.1111/ddi.12221 Laliberté, E., & Legendre, P. (2010). A distance-­based framework for Rao, R. (1982). Diversity and dissimilarity. Theoretical Population Biology, measuring functional diversity from multiple traits. Ecology, 91, 21(1), 24–­43. https://doi.org/10.13140/​RG.2.1.3901.9924 299–­3 05. https://doi.org/10.1890/08-­2244.1 Ricotta, C., & Moretti, M. (2011). CWM and Rao's quadratic diversity: A Larsen, T. H., Williams, N. M., & Kremen, C. (2005). Extinction unified framework for functional ecology. Oecologia, 167(1), 181–­ order and altered community structure rapidly disrupt eco- 188. https://doi.org/10.1007/s0044​2-­011-­1965-­5 system functioning. Ecology Letters, 8(5), 538–­547. https://doi. Rodríguez-­Gasol, N., Avilla, J., Aparicio, Y., Arnó, J., Gabarra, R., org/10.1111/j.1461-­0248.2005.00749.x Riudavets, J., Alegre, S., Lordan, J., & Alins, G. (2019). The contri- Lavorel, S., & Garnier, E. (2002). Predicting changes in community compo- bution of surrounding margins in the promotion of natural enemies sition and ecosystem functioning from plant traits. Functional Ecology, in mediterranean apple orchards. Insects, 10(5), 148. https://doi. 16(5), 545–­556. https://doi.org/10.1046/j.1365-­2435.​2002.00664.x org/10.3390/insec​t s100​50148 ROQUER-­BENI et al. Journal of Applied Ecology | 2853 Roquer-­Beni, L., Alins, G., Arnan, X., Boreux, V., García, D., Hambäck, P. Delicious’ Apple. Environmental Entomology, 29(2), 235–­240. A., Happe, A.-­K ., Klein, A.-­M., Miñarro, M., Mody, K., Porcel, M., https://doi.org/10.1093/ee/29.2.235 Rodrigo, A., Samnegård, U., Tasin, M., & Bosch, J. (2021). Data from: Villéger, S., Mason, N. W. H., & Mouillot, D. (2008). New multidimen- Management-­dependent effects of pollinator functional diversity sional functional diversity indices for a multifaceted framework on apple pollination services: A response-­effect trait approach. in functional ecology. Ecology, 89(8), 2290–­2301. https://doi. Dryad Digital Repository, https://doi.org/10.5061/dryad.63xsj​3v39 org/10.1890/07-­1206.1 Roquer-­Beni, L., Rodrigo, A., Arnan, X., Klein, A., Fornoff, F., Boreux, V., & Violle, C., Navas, M. L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., & Bosch, J. (2020). A novel method to measure hairiness in bees and Garnier, E. (2007). Let the concept of trait be functional! Oikos, 116(5), other insect pollinators. Ecology and Evolution, 10(6), 2979–­2990. 882–­892. https://doi.org/10.1111/j.2007.0030-­1299.15559.x https://doi.org/10.1002/ece3.6112 Williams, N. M., Crone, E. E., Roulston, T. H., Minckley, R. L., Packer, Roulston, T. H., & Goodell, K. (2011). The role of resources and risks in L., & Potts, S. G. (2010). Ecological and life-­history traits predict regulating wild bee populations. Annual Review of Entomology, 56(1), bee species responses to environmental disturbances. Biological 293–­312. https://doi.org/10.1146/annur​ev-­ento-­12070​9-­144802 Conservation, 143(10), 2280–­2291. https://doi.org/10.1016/j. Rundlöf, M., Nilsson, H., & Smith, H. G. (2008). Interacting effects biocon.2010.03.024 of farming practice and landscape context on bumble bees. Woodcock, B. A., Bullock, J. M., Shore, R. F., Heard, M. S., Pereira, M. G., Biological Conservation, 141(2), 417–­426. https://doi.org/10.1016/j. Redhead, J., Ridding, L., Dean, H., Sleep, D., Henrys, P., Peyton, J., Hulmes, biocon.2007.10.011 S., Hulmes, L., Sárospataki, M., Saure, C., Edwards, M., Genersch, E., Samnegård, U., Alins, G., Boreux, V., Bosch, J., García, D., Happe, A. K., Knäbe, S., & Pywell, R. F. (2017). Country-­specific effects of neonicoti- Klein, A. M., Miñarro, M., Mody, K., Porcel, M., Rodrigo, A., Roquer-­ noid pesticides on honey bees and wild bees. Science, 356(6345), 1393–­ Beni, L., Tasin, M., & Hambäck, P. A. (2019). Management trade-­offs 1395. https://doi.org/10.1126/scien​ce.aaa1190 on ecosystem services in apple orchards across Europe: Direct and Woodcock, B. A., Garratt, M. P. D., Powney, G. D., Shaw, R. F., Osborne, J. L., indirect effects of organic production. Journal of Applied Ecology, Soroka, J., Lindström, S. A. M., Stanley, D., Ouvrard, P., Edwards, M. E., 56, 802–­811. https://doi.org/10.1111/1365-­2664.13292 Jauker, F., McCracken, M. E., Zou, Y., Potts, S. G., Rundlöf, M., Noriega, Samnegård, U., Hambäck, P. A., & Smith, H. G. (2019). Pollination treat- J. A., Greenop, A., Smith, H. G., Bommarco, R., … Pywell, R. F. (2019). ment affects fruit set and modifies marketable and storable fruit Meta-­analysis reveals that pollinator functional diversity and abun- quality of commercial apples. Royal Society Open Science, 6(12), dance enhance crop pollination and yield. Nature Communications, 190326. https://doi.org/10.1098/rsos.190326 10(1), 1481. https://doi.org/10.1038/s4146​7-­019-­09393​-6 ­ Scheper, J., Holzschuh, A., Kuussaari, M., Potts, S. G., Rundlöf, M., Smith, Woodcock, B. A., Harrower, C., Redhead, J., Edwards, M., Vanbergen, A. H. G., & Kleijn, D. (2013). Environmental factors driving the effec- J., Heard, M. S., Roy, D. B., & Pywell, R. F. (2014). National patterns tiveness of European agri-­environmental measures in mitigating of functional diversity and redundancy in predatory ground beetles pollinator loss –­ A meta-­analysis. Ecology Letters, 16(7), 912–­920. and bees associated with key UK arable crops. Journal of Applied https://doi.org/10.1111/ele.12128 Ecology, 51(1), 142–­151. https://doi.org/10.1111/1365-­2664.12171 Schleuning, M., Fründ, J., & García, D. (2015). Predicting ecosystem func- Zuur, A. F., Ieno, E. N., & Elphick, C. S. (2010). A protocol for tions from biodiversity and mutualistic networks: An extension of data exploration to avoid common statistical problems. trait-­based concepts to plant-­animal interactions. Ecography, 38(4), Methods in Ecology and Evolution, 1(1), 3–­14. https://doi. 380–­392. https://doi.org/10.1111/ecog.00983 org/10.1111/j.2041-­210X.2009.00001.x Shuler, R. E., Roulston, T. H., & Farris, G. E. (2005). Farming practices influence wild pollinator populations on squash and pumpkin. Journal of Economic Entomology, 98(3), 790–­795. https://doi.org/10.1603/0022-­0493-­98.3.790 Stavert, J. R., Liñán-­Cembrano, G., Beggs, J. R., Howlett, B. G., Pattemore, D. E., S U P P O R T I N G I N FO R M AT I O N & Bartomeus, I. (2016). Hairiness: The missing link between pollinators Additional Supporting Information may be found in the online ver- and pollination. PeerJ, 4, e2779. https://doi.org/10.7717/peerj.2779 sion of the article at the publisher’s website. Suding, K. N., Lavorel, S., Chapin, F. S., Cornelissen, J. H. C., Díaz, S., Garnier, E., Goldberg, D., Hooper, D. U., Jackson, S. T., & Navas, M. L. (2008). Scaling environmental change through the community-­level: A trait-­based response-­and-­effect framework How to cite this article: Roquer-­Beni, L., Alins, G., Arnan, X., for plants. Global Change Biology, 14(5), 1125–­1140. https://doi. Boreux, V., García, D., Hambäck, P. A., Happe, A.-­K., Klein, org/10.1111/j.1365-­2486.2008.01557.x A.-­M., Miñarro, M., Mody, K., Porcel, M., Rodrigo, A., Tilman, D. (2001). Functional diversity. Encyclopedia of Biodiversity, 3(3), 109–­120. https://doi.org/10.1016/B0-­12-­22686​5-­2/00132​-­2 Samnegård, U., Tasin, M., & Bosch, J. (2021). Management-­ Vicens, N., & Bosch, J. (2000a). Weather-­dependent pollinator ac- dependent effects of pollinator functional diversity on apple tivity in an apple orchard, with special reference to Osmia pollination services: A response–­effect trait approach. Journal cornuta and Apis mellifera (Hymenoptera: Megachilidae and of Applied Ecology, 58, 2843–­2853. https://doi. Apidae). Environmental Entomology, 29(3), 413–­420. https://doi. org/10.1111/1365-­2664.14022 org/10.1603/0046-­225X-­29.3.413 Vicens, N., & Bosch, J. (2000b). Pollinating efficacy of Osmia cornuta and Apis mellifera (Hymenoptera: Megachilidae, Apidae) on ‘Red