Herbivores, saprovores and natural enemies respond differently to within-field plant characteristics of wheat fields

UPCommons Portal del coneixement obert de la UPC http://upcommons.upc.edu/e-prints Aquesta és una còpia de la versió author’s final draft d'un article publicat a la revista [Journal of Insect Conservation by Springer link] Disponible online: http://link.springer.com/article/10.1007/s10841-016-9879-5 URL d'aquest document a UPCommons E-prints: http:// hdl.handle.net/2117/90156 Article publicat1 / Published paper : Caballero-López, B., Blanco-Moreon, J.M., Pujade-Villar, J., Ventura, D., Sánchex-Espigares, J.A., Sans F.X. (2016) Herbivores, saprovores and natural enemies respond differently to within-field plant characteristics of wheat fields Doi: 10.1007/s10841-016-9879-5 -RXUQDORI,QVHFW&RQVHUYDWLRQ  +HUELYRUHVVDSURYRUHVDQGQDWXUDOHQHPLHVUHVSRQGGLIIHUHQWO\WRZLWKLQILHOGSODQW FKDUDFWHULVWLFVRIZKHDWILHOGV 0DQXVFULSW'UDIW  0DQXVFULSW1XPEHU -,&2'5 )XOO7LWOH +HUELYRUHVVDSURYRUHVDQGQDWXUDOHQHPLHVUHVSRQGGLIIHUHQWO\WRZLWKLQILHOGSODQW FKDUDFWHULVWLFVRIZKHDWILHOGV $UWLFOH7\SH 2ULJLQDO$UWLFOH .H\ZRUGV IXQFWLRQDODSSURDFKSODQWDUWKURSRGLQWHUDFWLRQELRORJLFDOFRQWUROOHJXPHV HFRV\VWHPVHUYLFHVLQVHFWIXQFWLRQDOWUDLWV &RUUHVSRQGLQJ$XWKRU %HUWD&DEDOOHUR/µSH]3K' 8QLYHUVLW\RI%DUFHORQD %DUFHORQD&DWDORQLD63$,1 &RUUHVSRQGLQJ$XWKRU6HFRQGDU\ ,QIRUPDWLRQ &RUUHVSRQGLQJ$XWKRUV,QVWLWXWLRQ 8QLYHUVLW\RI%DUFHORQD &RUUHVSRQGLQJ$XWKRUV6HFRQGDU\ ,QVWLWXWLRQ )LUVW$XWKRU %HUWD&DEDOOHUR/µSH]3K' )LUVW$XWKRU6HFRQGDU\,QIRUPDWLRQ 2UGHURI$XWKRUV %HUWD&DEDOOHUR/µSH]3K' -RV«0DQXHO%ODQFR0RUHQR3URIHVVRU -XOL3XMDGH9LOODU3URIHVVRU 'DQLHO9HQWXUD3K'VWXGHQW -RVHS$QWRQ6£QFKH](VSLJDUHV3URIHVVRU )UDQFHVF;DYLHU6DQV3URIHVVVRU 2UGHURI$XWKRUV6HFRQGDU\,QIRUPDWLRQ )XQGLQJ,QIRUPDWLRQ %HFDV3UHGRFWRUDOHVSDUD3HUVRQDO ,QYHVWLJDGRU ), 0UV%HUWD&DEDOOHUR/µSH] 0LQLVWHULRGH&LHQFLD\7HFQRORJ¯D &*/F%26 'U)UDQFHVF;DYLHU6DQV 0LQLVWHULRGH&LHQFLD\7HFQRORJ¯D &*/F 'U)UDQFHVF;DYLHU6DQV 0LQLVWHULRGH&LHQFLDH,QQRYDFLµQ &*/ 'U)UDQFHVF;DYLHU6DQV $EVWUDFW 8QGHUVWDQGLQJHFRV\VWHPIXQFWLRQLQJLQDIDUPODQGFRQWH[WE\FRQVLGHULQJWKHYDULHW\ RIHFRORJLFDOVWUDWHJLHVHPSOR\HGE\DUWKURSRGVLVDFRUHFKDOOHQJHLQHFRORJ\DQG FRQVHUYDWLRQVFLHQFH:HDGRSWHGDIXQFWLRQDODSSURDFKLQDQDVVHVVPHQWRIWKH UHODWLRQVKLSEHWZHHQWKHWKUHHIXQFWLRQDOSODQWJURXSVJUDVVHVEURDGOHDYHGDQG OHJXPHVDQGWKHDUWKURSRGFRPPXQLW\LQZLQWHUZKHDWILHOGVLQD0HGLWHUUDQHDQ GU\ODQGFRQWH[W:HVDPSOHGWKHDUWKURSRGFRPPXQLW\DVWKRURXJKO\DVSRVVLEOHZLWK DFRPELQDWLRQRIVXFWLRQFDWFKLQJDQGIOLJKWLQWHUFHSWLRQWUDSSLQJ$OOVSHFLPHQVZHUH LGHQWLILHGWRWKHDSSURSULDWHWD[RQRPLFOHYHOIDPLO\JHQXVRUVSHFLHVDQGFODVVLILHG DFFRUGLQJWRWKHLUIRUPRIIHHGLQJFKHZLQJKHUELYRUHVVXFNLQJKHUELYRUHVIORZHU FRQVXPHUVRPQLYRUHVVDSURYRUHVSDUDVLWRLGVRUSUHGDWRUV $ULFKHUSODQWFRPPXQLW\IDYRXUVDJUHDWHUGLYHUVLW\RIKHUELYRUHVDQGLQWXUQD ULFKQHVVRIKHUELYRUHVDQGVDSURYRUHVHQKDQFHVWKHFRPPXQLWLHVRIWKHLUQDWXUDO HQHPLHVZKLFKVXSSRUWVWKHFODVVLFDOWURSKLFVWUXFWXUHK\SRWKHVLV7KHSRVLWLYHHIIHFW RIJUDVVFRYHURQVXFNLQJKHUELYRUHVVDSURYRUHVDQGWKHLUQDWXUDOHQHPLHVLVGXHWR Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation 1 Herbivores, saprovores and natural enemies respond differently to 1 within-field plant characteristics of wheat fields 2 3 Berta Caballero-López 1,4*, José M. Blanco-Moreno 2, 4, Juli Pujade-Villar3, Daniel Ventura5,6, 4 Josep A. Sánchez-Espigares 7 & F. Xavier Sans 2, 4 5 1 Dept. of Arthropods, Natural Sciences Museum of Barcelona, Barcelona, Spain; 2 Dept. of Plant Biology, 6 Faculty of Biology, University of Barcelona, Barcelona, Spain; 3 Dept. of Animal Biology, Faculty of Biology, 7 University of Barcelona, Barcelona, Spain; 4 IRBio, University of Barcelona, Spain; 5 Dept. of Food 8 Industries and Environmental Sciences, Polytechnic School, University of Vic, Vic, Spain; 6 Functional Ecology 9 and Climate Change Group (GAMES - ECOFUN), Forest Sciences Center of Catalonia (CTFC), Solsona, Spain; 10 7 Dept. of Statistics and Operations Research (UPC), Barcelona, Spain. 11 12 13 *Corresponding author’s address: Dept. of Arthropods, Lab. of Nature, Museu de Ciències Naturals de 14 Barcelona, Picasso Av., E-08003 Barcelona (Catalonia/Spain). E-mail address: [email protected], Phone: 15 (+34) 93 256 22 11. 16 17 Acknowledgements 18 We are indebted to Lluís Tarés and Joan Ramon Salla for their willingness to participate in this project and 19 for generously allowing us to work in their fields. We are grateful to Amador Viñolas (Coleoptera), Miguel 20 Carles-Tolrà (Diptera) and Marcos Roca-Cusachs (Hemiptera) for the huge task of identifying specimens and for 21 offering information about arthropod feeding habits, which was of great help when deciding upon the most 22 appropriate feeding categories. We would also like to thank Albert Ferré and Arnau Mercadé (Cartography 23 group, Plant Biology Department, University of Barcelona) for their technical assistance with the GIS analyses 24 in the margin assessment. The authors would like to thank the two anonymous referees whose suggestions 25 significantly contributed to improve our manuscript. This research represents part of the PhD project by the 26 leading author and was funded by the FI Fellowship (Agència de Gestió d’Ajuts Universitaris i de Recerca, 27 Generalitat de Catalunya) and the Spanish Ministry of Economy and Competitiveness (CGL2006-c03-01/BOS; 28 CGL2009-13497-c02-01; CGL2012-39442). 29 0DQXVFULSW &OLFNKHUHWRGRZQORDG0DQXVFULSW&DEDOOHUR/RSH]HWDO PVB$SUB'()GRF[ &OLFNKHUHWRYLHZOLQNHG5HIHUHQFHV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 2 Abstract 30 Understanding ecosystem functioning in a farmland context by considering the variety of ecological 31 strategies employed by arthropods is a core challenge in ecology and conservation science. We adopted a 32 functional approach in an assessment of the relationship between the three functional plant groups (grasses, 33 broad-leaved and legumes) and the arthropod community in winter wheat fields in a Mediterranean dryland 34 context. We sampled the arthropod community as thoroughly as possible with a combination of suction catching 35 and flight-interception trapping. All specimens were identified to the appropriate taxonomic level (family, genus 36 or species) and classified according to their form of feeding: chewing-herbivores, sucking-herbivores, flower-37 consumers, omnivores, saprovores, parasitoids or predators. 38 A richer plant community favours a greater diversity of herbivores and, in turn, a richness of herbivores and 39 saprovores enhances the communities of their natural enemies, which supports the classical trophic structure 40 hypothesis. The positive effect of grass cover on sucking-herbivores, saprovores and their natural enemies is due 41 to grasses’ ability to provide – either directly or indirectly alternative resources or simply by offering better 42 conditions of environmental parameters. By the inclusion of legumes in agroecosystems we can improve the 43 conservation of beneficial arthropods like predators or parasitoids, and enhance the provision of ecosystem 44 services like the natural pest control. 45 46 47 Keywords: functional approach, plant-arthropod interaction, biological control, legumes, ecosystem services, 48 insect functional traits. 49 50 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 3 Introduction 52 Although traditionally considered as mere competitors of crop plants (Albajes et al. 2011), weeds do in fact 53 play a key role in the aboveground food chain in agro-ecosystems (Clough et al. 2007) by providing resources 54 for pollinators and herbivorous insects, and by supporting prey species for natural enemies (Norris and Kogan 55 2000; Hyvönen and Huusela-Veistola 2008). Nonetheless, how herbivores and natural enemies respond to the 56 within-field plant community is still a matter of debate and the information in the literature is rather 57 contradictory. Birkhofer et al. (2008) and Harwood et al. (2001) reported more predators in weedy fields – 58 probably as a response to increased prey availability – but other authors have found that the abundance of 59 predatory invertebrates seldom responds significantly to the weed community (Fuller et al. 2005). Some authors 60 state that weedy plots do not necessarily have higher predator densities as other authors have claimed (Altieri 61 and Nicholls 1999; Amaral et al. 2013) 62 These discrepancies arise because most predictions are limited to particular species groups that are unable 63 to provide accurate generalizations of observed patterns that are applicable to the entire arthropod community 64 (Perner and Voigt 2007). Indeed, arthropods account for over 80% of all known living animal species and play a 65 wide range of functional roles in ecosystems (Maleque et al. 2006). On the other hand, complete community-66 level assessments are rarely conducted given the huge amount of time, money and human resources (i.e. 67 taxonomists) that are required (Cardoso et al. 2004). Nevertheless, several authors have adopted a community 68 approach using higher taxonomic levels such as families as surrogates for inventories at species level (Balmford 69 et al. 1996a; Balmford et al. 1996b; Wickramasinghe et al. 2004; Biaggini et al. 2007), which is a way of 70 circumventing the enormous amount of resources required for close-to-complete inventories (Cardoso et al. 71 2004). The use of families as a taxonomic level not only allows parataxonomists to complete the required 72 classification tasks – which permits groups that had not previously been considered to be bioindicators (due to 73 taxonomic difficulties) to be included – but can also save time and money (Balmford et al. 1996a; Balmford et 74 al. 1996b). 75 Here we adopt a community approach and work at family level. We use a functional approach based on 76 species’ way-of-feeding strategies and, rather than relying on traditional taxonomic analyses, we amalgamate 77 different groups according to their trophic behaviour. This combination of a community approach at family level 78 and a functional approach is novel, and provides a link between taxonomic diversity and ecosystem functioning 79 (Grimm 1995; McCann 2000; Hawes et al. 2009). 80 Assessing how within-field plant communities affect whole arthropod assemblages is therefore essential for 81 understanding local processes related to agro-ecosystem functioning, and to accomplish this task it is crucial to 82 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 4 gain a broader picture of the different players on the scene. 83 Floristic richness as well as vegetation structure has been widely recognised as key factors influencing 84 insect assemblage (Schaffers et al. 2008). As plant species richness and vegetation complexity tend to cascade up 85 to higher trophic levels leading to high invertebrate diversity (Landis et al. 2000). Therefore we would expect 86 that with a richer assemblage in the within-plant community is likely to improve the conservation of multiple 87 arthropod groups. In this study we were interested in assessing the effect of richer within-field plant communities 88 as a component of habitat restoration strategies to improve and sustain biological control in an arable cropping 89 system. The research reported here aimed to examine how contrasting within-field plant communities in wheat 90 fields affect the whole community of insects associated to this crop. We hypothesised that plant-feeders and 91 saprovores would respond to the within-field plant assemblage according to the classical diversity-trophic 92 structure hypothesis, and that the abundance and richness of potential prey items would enhance the parasitoid 93 and predator assemblages. 94 95 Material and Methods 96 Study area 97 The study was carried out about 150 km south of Barcelona (41º29’0.9’’N, 1º7’16.4’’E; 627 m a.s.l.). The 98 arable fields – mainly cereal crops – represented only 40% of the agricultural landscape and formed a mosaic 99 with patches of natural vegetation. Field boundaries consisted of perennial grasslands dominated by 100 Brachypodium phoenicoides (L.) Roemer & Schultes, as well as a mix of Prunus spinosa L., and Rubus 101 ulmifolius L. thickets, and Rosmarinus officinalis L. scrub. 102 Four organically and four conventionally managed winter wheat fields (Triticum aestivum L.) were selected 103 in an area of 2×2 km. First, the organic fields were randomly selected from the 12 such fields in the area and, 104 then, the conventional fields were selected, none of which were further than 1 km from or adjoining the organic 105 fields. All selected fields were flat in order to avoid any differences due to slope or aspect. The selected organic 106 fields had been managed for over a decade along Catalan organic guidelines (Consell Català de la Producció 107 Agrària Ecològica 2013) and were certified by the Catalan Council for Organic Farming following the European 108 guidelines (EEC 2007). The management of organic fields relies on mechanical weed control and organic 109 fertilisation using green manure and occasionally chicken manure. Conventional fields were regularly sprayed 110 with herbicides – but not insecticides or fungicides – and fertilised with a combination of pig slurry and mineral 111 fertilisers. Although we tried to select fields of similar size and shape, we considered the homogeneity of the 112 boundary vegetation to be more important than the homogeneity of the fields’ dimensions, above all because the 113 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 5 fields were all relatively small. Even so, conventional fields were significantly larger (mean ± SE; 4.08 ± 0.8 ha) 114 than organic fields (2.19 ± 0.3 ha, ߯ଵௗ௙ ଶ= 5.78, P value = 0.016). By contrast, the perimeter-to-area ratio was 115 significantly greater in organic (mean ± SE; 0.09 ± 0.01) than in conventional fields (0.06 ± 0.01, ߯ଵௗ௙ ଶ= 4.85, P 116 value = 0.028). All selected fields were sown with winter wheat between the 27 October and 7 November 2003 117 (for further agronomic details, see Caballero-López et al. 2010). 118 The contrast in a common area between organic and conventional cereal fields whose boundaries share the 119 same vegetation but differ in terms of the non-crop plants they host appears to be a suitable model for exploring 120 the relationship between plant and arthropod communities. In addition, the comparison of fields under organic 121 and under conventional insecticide-free management in a Mediterranean context also avoids the confounding 122 indirect effects of insecticide application on the plant-arthropod interactions (Hole et al., 2005). 123 In each field we established an 80m transect diagonally across the centre of the field, starting at 55m from 124 the edge. Within each transect, five 1m×1m plots at 20m intervals were surveyed. Arthropod suction-sampling 125 and plant surveys were carried out successively in each plot. In addition, three flight-interception traps (FIT) 126 were positioned along each transect at 40m intervals. 127 128 129 Sampling 130 Arthropod communities were sampled using (i) flight interception traps (hereafter FIT) to assess aerial 131 communities and (ii) a petrol-driven Blow&Vac (McCullogh BVM250, Italy; sampling cylinder 60cm high and 132 12 cm in diameter) converted to suction sampler following Stewart and Wright (1995) to survey terrestrial 133 communities. 134 Each FIT consisted of an outer white plastic cup (150 mm in height, 200 mm internal diameter) mounted on 135 a 1-m-high wooden pole and an inner plastic cup (140×180 mm) with two 30×30 cm Plexiglas pieces fixed along 136 their midline in a cross-shape. The inner plastic cup contained approximately 1 litre of a NaCl-solution as a 137 preservative, with a drop of detergent added to decrease the surface tension. FIT are useful for catching many of 138 the small flying insects that tend to fly downwards when they hit a wall (Koricheva et al. 2000). 139 The petrol-driven suction sampler was operated on full power to produce an estimated constant airflow of 140 0.142 m3/s (according to manufacturer’s operating instructions). The pipe was held vertically and slowly passed 141 over the wheat plants in the 1-m2 quadrat and suction was performed for 60 seconds. After each plot sampling, 142 the bag was removed from the machine, placed in a labelled plastic bag and stored in a portable refrigerator to 143 prevent predatory activity in the bag. The sampling campaign lasted for two days and the eight fields were 144 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 6 sampled in a random order to avoid any systematic bias due to daytime sampling. All samples were taken by the 145 same two people to reduce sampling variability. This method has been shown to provide a good representation of 146 all trophic levels interacting with vegetation (Letourneau and Goldstein 2001), and is used extensively to study 147 arthropods in crops (Stewart and Wright 1995; Elliott et al. 2006). 148 Fit trapping took place on 20 May 2004 and 26 June 2004 to coincide with the wheat’s anthesis stage and 149 the mid-milk-ripe cereal development stage (Zadoks et al. 1974), respectively. In total, 24 traps were active 150 during two periods of eight days. Suction sampling was also performed twice to coincide with the two chosen 151 growth stages, the first campaign taking place on 25–27 May and the second on 24–26 June 2004, both at 10:00–152 19:00 and under sunny weather conditions (temperature > 20ºC). Thus, in all, 40 m2 of plots were assessed twice 153 during the study period. 154 Vegetation was surveyed twice and concomitant with the suction-sampling. The cover of crop species and 155 each weed species was recorded in each plot by means of a ground cover scale. Weed species were identified 156 according to Bolòs et al. (2005). Plant species were classified into three functional groups (grasses, forbs and 157 legumes) following Koricheva et al. (2000). Legumes have been separated from the other forbs due to the 158 generally higher nitrogen content of their tissues, which would make them a higher-quality resource for 159 herbivores, whereas grasses have tough tissues with low nitrogen content and structural characteristics that deter 160 plant-feeders (Koricheva et al. 2000). 161 162 Arthropod processing 163 Arthropods captured by suction sampling were frozen for subsequent sorting and identification, whilst FIT 164 trap catches were preserved in 70% alcohol. All samples were hand-sorted using a dissecting microscope to 165 separate animals from debris. Catches were quantified as the total numbers of individuals (adults and immature 166 stages) and with a few exceptions most arthropods were identified to family level; due to taxonomic difficulties, 167 some taxa were only identified to superfamily level (e.g. Apoidea, Curculionoidea and Staphylinoidea) or to 168 order level (e.g. Acari and Thysanoptera). Lepidoptera were only identified to order level because specimens 169 were too badly damaged by the sampling process to be properly identified. 170 The use of higher taxonomic levels is particularly useful when a functional-group perspective is required as 171 the majority of family members belong to the same feeding group (see the considerations below). Nevertheless, 172 the process of amalgamating taxa into functional groups requires the acceptance of assumptions regarding the 173 importance of certain common features (Hawes et al. 2009). 174 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 7 When taxa of the same family had different feeding preferences (e.g. Drosophilidae, Opomyzidae), 175 specimens were determined to genus or species level, and the predominant feeding habit of the most abundant 176 genus or species was used to classify the entire family and its feeding group. We initially considered splitting 177 families possessing several species into similar proportions and different feeding strategies; although in the end 178 no family fulfilled this condition. 179 All identified taxa were classified into one of the seven feeding groups: chewing-herbivores, flower-180 consumers, omnivores, parasitoids, predators, saprovores and sucking-herbivores. The definition of each feeding 181 group was based on field observations, a literature review and specialist advice (see Acknowledgements), and 182 contained different ways-of-feeding strategies. Granivores, plant-chewers and miners were included in the 183 chewing-herbivore category, while plant sapsuckers were added to the suction-herbivore category. Flower 184 consumers consisted of flower predators, pollen consumers and nectarivores. Saprovores included 185 mycetophages, plant saprovores, animal saprovores and scavengers. 186 Arthropods with different feeding preferences in larval and adult stages were counted in both feeding 187 groups in order to consider the impact of their whole life cycles. A small number of difficult-to-classify larvae 188 were taken into account only for total abundance but were excluded from the feeding group analyses. Other 189 groups were also excluded from the analyses due to their scarcity (families with less than three individuals were 190 excluded from the data) or a lack of available information about their biology. In addition, other groups such as 191 most parasitoids, which do not feed in the adult stage or whose effect is so small as to be insignificant, were 192 categorised as not having any trophic interaction (for further details, see Supplementary material). All the 193 specimens are now deposited in the Arthropod collection of the Natural Sciences Museum of Barcelona. 194 195 Data analysis 196 In order to simplify the statistical analyses and results section, the results are grouped into two categories: 197 primary and secondary consumers. Chewing-herbivores, sucking-herbivores and flower-consumers were 198 considered primary consumers and so are mainly herbivores, while parasitoids and predators were categorised as 199 secondary consumers given that they are entomophagous. Saprovores chew dead organic matter, bacteria and 200 fungi, and occasionally soil arthropods, and thus theoretically occupy an intermediate position between primary 201 and secondary consumers. However, they were included arbitrarily as primary consumers owing to the lack of 202 reliable information about their consumption rate of potential prey items. 203 The models for primary consumers and secondary consumers were analysed according to sampling method 204 (FIT vs. suction) and sampling period (first vs. second), with a common set of covariates (cover of broad-leaved 205 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 14 managed grass-clover fields: implications for conservation biological control. Ann Appl Biol 153:271–391 280. 392 Caballero-López B, Blanco-Moreno JM, Pérez N, Pujade-Villar J, Ventura D, Oliva F, Sans FX (2010) A 393 functional approach to assessing plant-arthropod interaction in winter wheat. Agric Ecosyst Environ 394 137:288–293. 395 Cardoso P, Silva I, de Oliveira NG, Serrano ARM (2004) Higher taxa surrogates of spider (Araneae) diversity 396 and their efficiency in conservation. Biol Conserv 117:453–459. 397 Casas J, Djemai I (2002) Canopy architecture and multitrophic interactions. In: Tscharntke T, Hawkins BA (eds) 398 Multitrophic level interactions. Cambridge University Press, Cambridge, pp 174–196 399 Clough Y, Holzschuh A, Gabriel D, Purtauf T, Kleijn D, Kruess A, Steffan-Dewenter I, Tscharntke T (2007) 400 Alpha and beta diversity of arthropods and plants in organically and conventionally managed wheat fields. 401 J Appl Ecol 44:804–812. doi: doi: 10.1111/j.1365-2664.2007.01294 402 Consell Català de la Producció Agrària Ecològica (2013) Guia per a la certificació ecològica i informació 403 pública. Generlaitat de Catalunya, Barcelona 404 Duffield SJ, Bryson RJ, Young JEB, Sylvester-Bradley R, Scott RK (1997) The influence of nitrogen fertiliser 405 on the population development of the cereal aphids Sitobion avenae (F) and Metopolophium dirhodum 406 (Wlk) on field grown winter wheat. Ann Appl Biol 130:13–26. doi: doi: 10.1111/j.1744-407 7348.1997.tb05779 408 EEC (2007) Council Regulation (EC) No 2007, of 28 June 2092/91, on organic production and labelling of 409 organic products and repealing Regulation (EEC). 410 Elliott NC, Tao FL, Fuentes-Granados R, Giles KL, Elliott DT, Greenstone MH, Shufran KA, Royer TA (2006) 411 D-Vac sampling for predatory arthropods in winter wheat. Biol Control 38:325–330. 412 Evans EW (2008) Multitrophic interactions among plants, aphids, alternate prey and shared natural enemiesa 413 review. Eur J Entomol 105:369–380. 414 Fuller RJ, Norton LR, Feber RE, Johnson PJ, Chamberlain DE, Joys AC, Mathews F, Stuart RC, Townsend MC, 415 Manley WJ, Wolfe MS, Macdonald DW, Firbank LG (2005) Benefits of organic farming to biodiversity 416 vary among taxa. Biol Lett 1:431–434. 417 Ghorbani R., Mousavi SK, Nazari S, Hatami B (2010) Study of the effect of irrigation and nitrogen fertilizer on 418 sugar beet growth and population dynamism of Empoasca decipiens (Hom.: Cicadellidae) and Thrips 419 tabaci (Thys.: Thripidae). Appl Entomol Phytopathol 78:61–80. 420 Grimm NB (1995) Why link species and ecosystems? A perspective from ecosystem ecology. In: Jones CG, 421 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 15 Lawton JH (eds) Linking Species and Ecosystems. Chapman & Hall, New York, pp 5–15 422 Haddad NM, Tilman D, Haarstad J, Ritchie M, Knops MH (2001) Contrasting effects of plant richness and 423 composition on insects communities: a field experiment. Am Nat 158:17–35. 424 Harwood JD, Sunderland KD, Symondson WOC (2001) Living where the food is: web location by linyphiid 425 Wheat, in relation to prey availability in winter. J Appl Ecol 38:88–99. 426 Hasken KH, Poehling HM (1995) Effects of different intensities of fertilisers and pesticides on aphids and 427 aphids predators in winter wheat. Agric Ecosyst Environ 52:45–50. 428 Hawes C, Haughton AJ, Bohanb DA, Squire GR (2009) Functional approaches for assessing plant and 429 invertebrate abundance patterns in arable systems. Basic Appl Ecol 10:34–42. doi: 430 doi:10.1016/j.baae.2007.11.007 431 Hodek I, Honek A (1996) Ecology of Coccinellidae. Kluwer Academic Publishers, Dordrecht, The Netherlands 432 Hyvönen T, Huusela-Veistola E (2008) Arable weeds as indicators of agricultural intensity. A case study from 433 Finland. Biol Conserv 141:2857–2864. 434 Isaacs R, Tuell J, Fiedler A, Gardiner M, Landis D (2009) Maximizing arthropod-mediated ecosystem services 435 in agricultural landscapes: the role of native plants. Front Ecol Environ 7:196–203. 436 Knops JMH, Tilman D, Haddad NM, Naeem S, Mitchell CE, Haarstad J, Ritchie ME, Howe KM, Reich PB, 437 Siemann E, Groth J (1999) Effects of plant species richness on invasion dynamics, disease outbreaks, 438 insect abundances and diversity. Ecol Lett 2:286–293. 439 Koricheva J, Mulder CPH, Schmid B, Joshi J, Huss-Danell K (2000) Numerical responses of different trophic 440 groups of invertebrates to manipulation of plant diversity in grasslands. Oecologia 125:271–282. doi: doi: 441 10.1007/s004420000450 442 Kutner MH, Nachtsheim CJ, Neter J (2004) Applied Linear Regression Models, 4th edn. McGraw-Hill, Irwin 443 Landis DA, Wratten SD, Gurr GM (2000) Habitat managment to conserve natural enemies of arthropods pests in 444 agriculture. 445 Letourneau DK, Goldstein B (2001) Pest damage and arthropod community structure in organic vs. conventional 446 tomato production in California. J Appl Ecol 38:557–570. 447 Lundgren JG, Seagraves MP (2011) Physiological benefits of nectar feeding by a predatory beetle. Biol J Linn 448 Soc 104:661–669. doi: 10.1111/j.1095-8312.2011.01729.x 449 Mäder P, Fließbach A, Dubois D, Gunst L, Fried P, Niggli U, Masters G (2002) Soil fertility and biodiversity in 450 organic farming. Science (80- ) 296:1694–97. 451 Maleque MA, Ishii HT, Maeto K (2006) The use of arthropods as indicators of ecosystem integrity in forest 452 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 16 management. J For 104:113–117. 453 McCann KS (2000) The diversity-stability debate. Nature 405:228–233. 454 Moreby SJ, Aebischer NJ, Southway SE, Sotherton NW (1994) A comparison of the flora and arthropod fauna 455 of organically and conventionally grown winter-wheat in southern England. Ann Appl Biol 125:13–27. 456 Müller C, Godfray H (1998) The response of aphid secondary parasitoids to different patch densities of their 457 host. BioControl 43:129–139. 458 Murdoch WW, Evans FC, Peterson CH (1972) Diversity and pattern in plants and insects. Ecology 53:819–829. 459 Nicholls CI, Altieri MA (2012) Plant biodiversity enhances bees and other insect pollinators in agroecosystems. 460 A review. Agron Sustain Dev 33:257–274. doi: 10.1007/s13593-012-0092-y 461 Norris RF, Kogan M (2000) Interactions between weeds, arthropod pests, and their natural enemies in managed 462 ecosystems. Weed Sci 48:94–158. doi: 10.1614/0043-1745(2000)048[0094:IBWAPA]2.0.CO;2 463 Perner J, Voigt W (2007) Measuring the complexity of interaction webs using vertical links between functional 464 groups. Agric Ecosyst Environ 120:192–200. 465 Pinheiro JB, Bates DM (2000) Mixed-Effects Models in S and S-Plus. Springer, New York, USA 466 R Development Core Team (2013) R: A Language and Environment for Statistical Computing. R Foundation for 467 Statistical Computing. R Dev. Core Team 468 Rostami M, Zamani AA, Goldasteh S, Shoushtari RV, Kheradmand K (2012) Influence of nitrogen fertilization 469 on biology of Aphis gossypii (Hemiptera: Aphididae) reared on Chrysanthemum indicum (Asteraceae). J 470 Plant Prot Res 52:118–121. doi: 10.2478/v10045-012-0019-2 471 Schaffers AP, Raemakers IP, Sýkora K V, Ter Braak CJF (2008) Arthropods assemblages are best predicted by 472 plant species composition. Ecology 89:782–794. 473 Shreeve TG., Dennis RLH., Roy DB., Mos D (2001) An ecological classification of British butterflies: 474 Ecological attributes and biotope occupancy. J Insect Conserv 5:145–161. 475 Siemann E (1998) Experimental tests of effects of plant productivity and diversity on grassland arthropod 476 diversity. Ecology 79:2057–2070. 477 Souza ALT De, Martins RP (2004) Distribution of plant-dwelling spiders: Inflorescences versus vegetative 478 branches. Austral Ecol 29:342–349. doi: 10.1111/j.1442-9993.2004.01371.x 479 Stewart AJA, Wright AF (1995) A new inexpensive suction apparatus for sampling arthropods in grassland. Ecol 480 Entomol 20:98–102. 481 Vucic-Pestic O, Birkhofer K, Rall BC, Scheu S, Brose U (2010) Habitat structure and prey aggregation 482 determine the functional response in a soil predator-prey interaction. Pedobiologia (Jena) 53:307–312. 483 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 17 Wardle DA, Nicholson KS, Bonner KI, Yeates GW (1999) Effects of agricultural intensification on soil-484 associated arthropod population dynamics, community structure, diversity and temporal variability over a 485 seven-year period. Soil Biol Biochem 31:1691–1706. doi: doi:10.1016/S0038-0717(99)00089-9 486 Wickramasinghe LP, Harris S, Jones G (2004) Abundance and species richness of nocturnal insects on organic 487 and conventional farms: Effects of agricultural intensification on bat foraging. Conserv Biol 18:1283–488 1292. doi: doi: 10.1111/j.1523-1739.2004.00152 489 Willmer P, Hughes J, Woodford J, Gordon S (1996) The effects of crop microclimate and associated 490 physiological constraints on the seasonal and diurnal distribution patterns of raspberry beetle ( Byturus 491 tomentosus ) on the host plant Rubus idaeus. Ecol Entomol 21:87–97. 492 Wilson RJ, Bennie J, Lawson CR, Pearson D, Ort??zar-Ugarte G, Guti??rrez D (2014) Population turnover, 493 habitat use and microclimate at the contracting range margin of a butterfly. J Insect Conserv 19:205–216. 494 doi: 10.1007/s10841-014-9710-0 495 Zadoks JC, Chang TT, Konzak CF (1974) A decimal code for the growth stages of cereals. Weed Res 14:415–496 421. 497 498 499 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 18 FIGURE CAPTIONS 500 Figure 1. Total number of individuals of primary consumers caught by interception traps (FIT) and suction 501 sampling (VAC) in May and June. CH = Chewing-herbivores, FC = Flower-consumers, SH = Sucking-502 herbivores, S = Saprovores and O= Omnivores. Symbols indicate mean values and bars indicate the standard 503 error. 504 Figure 2. Total number of individuals of secondary consumers caught by interception traps (FIT) and 505 suction sampling (VAC) in May and June. Pa = Parasitoids and Pr = Predators. Symbols indicate mean values 506 and bars indicate the standard error. 507 Figure 3. Total family richness of primary consumers caught by interception traps (FIT) and suction 508 sampling (VAC) in May and June. CH = Chewing-herbivores, FC = Flower-consumers, SH = Sucking-509 herbivores, S = Saprovores and O = Omnivores. Symbols indicate mean values and bars indicate the standard 510 error. 511 Figure 4. Total family richness of secondary consumers caught by interception traps (FIT) and suction 512 sampling (VAC) in May and June. Pa = Parasitoids, and Pr = Predators. Symbols indicate mean values and bars 513 indicate the standard error. 514 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Abundance (number of individuals) 0 50 100 150 May June CH May June FC May June O 0 50 100 150 S SH FIT VAC )LJXUH$EXQGDQFH3ULPDU\&RQVXPHUV &OLFNKHUHWRGRZQORDG)LJXUH $EXQGDQFHBSULPDU\B'()SGI Abundance (number of individuals) 5 10 15 20 25 30 May June Pa May June Pr FIT VAC )LJXUH$EXQGDQFH6HFRQGDU\&RQVXPHUV &OLFNKHUHWRGRZQORDG)LJXUH $EXQGDQFHBVHFRQGDU\B'()SGI Richness (number of species) 5 10 15 May June CH May June FC May June O 5 10 15 S SH FIT VAC )LJXUH5LFKQHVV3ULPDU\&RQVXPHUV &OLFNKHUHWRGRZQORDG)LJXUH 5LFKQHVVBSULPDU\B'()SGI Richness (number of species) 2 4 6 8 10 May June Pa May June Pr FIT VAC )LJXUH5LFKQHVV6HFRQGDU\&RQVXPHUV &OLFNKHUHWRGRZQORDG)LJXUH 5LFKQHVVBVHFRQGDU\B'()SGI Table 1 Effects of sampling method (SM), sampling period (SP), their interaction (SM*SP) and plant descriptors such as plant species richness (SR), legume cover (LC), broad-leaved herb cover (BC) and grass cover (GC) on the abundance of primary consumers. The level of significance for the different predictors included in the models was obtained using Markov Chain Monte Carlo methods. Chewing-herbivore Flower-consumer Suction-herbivores Saprovores abundance abundance abundance abundance X ± SE P X ± SE P X ± SE P X ± SE P Intercept 2.62 ± 0.52 0.000 4.09 ± 0.38 0.000 6.12 ± 1.78 0.001 2.43 ± 0.42 0.000 SM -2.09 ± 0.18 0.000 -3.27 ± 0.15 0.000 -1.33 ± 0.65 0.044 -2.61 ± 0.19 0.000 SP -0.66 ± 0.20 0.002 -0.12 ± 0.17 0.470 4.62 ± 0.73 0.000 -0.51 ± 0.22 0.020 SM*SP 1.02 ± 0.25 0.000 0.64 ± 0.21 0.002 -2.46 ± 0.92 0.010 1.53 ± 0.27 0.000 SR -0.02 ± 0.04 0.519 -0.02 ± 0.03 0.490 0.06 ± 0.15 0.827 0.06 ± 0.04 0.154 LC 0.02 ± 0.02 0.320 0.03 ± 0.02 0.052 0.05 ± 0.07 0.483 0.02 ± 0.02 0.285 BC 0.01 ± 0.01 0.491 0.00 ± 0.01 0.580 0.02 ± 0.03 0.530 0.01 ± 0.01 0.435 GC 0.01 ± 0.01 0.058 0.01 ± 0.00 0.003 0.02 ± 0.02 0.468 0.02 ± 0.00 0.000 Table 2 Effects of sampling method (SM), sampling period (SP), their interaction (SM*SP) and plant descriptors such as plant species richness (SR), legume cover (LC), broad-leaved herb cover (BC) and grass cover (GC) on the abundance of parasitoids and predators. The abundance of primary consumers was also included in this model. CH = Chewing-herbivores, FC = Flower-consumers, S = Saprovores, SH = Suckingherbivores (see text for further details). The level of significance for the different predictors included in the models was obtained using Markov Chain Monte Carlo methods. Parasitoids Predators abundance abundance X ± SE P X ± SE P Intercept 1.51 ± 0.28 0.000 1.53 ± 0.38 0.000 SM -1.13 ± 0.22 0.000 -1.07 ± 0.29 0.000 SP -0.03 ± 0.18 0.869 -0.37 ± 0.24 0.107 SM*SP 0.76 ± 0.22 0.001 1.33 ± 0.29 0.000 SR 0.00 ± 0.02 0.818 0.04 ± 0.03 0.240 LC 0.03 ± 0.01 0.027 0.03 ± 0.01 0.033 BC -0.00 ± 0.00 0.886 0.00 ± 0.00 0.798 GC 0.01 ± 0.00 0.001 0.01 ± 0.00 0.002 Ab.CH 0.01 ± 0.01 0.138 -0.00 ± 0.01 0.971 Ab.FC -0.00 ± 0.00 0.898 0.00 ± 0.00 0.561 Ab.SH 0.00 ± 0.00 0.000 0.00 ± 0.00 0.005 Ab.S 0.00 ± 0.00 0.204 0.00 ± 0.00 0.165 WDEOH &OLFNKHUHWRGRZQORDGWDEOH&DEDOOHUR/RSH]HW DO7DEOHVB$SUB'()GRF