Landscape Ecology

, Volume 24, Issue 4, pp 547–555 | Cite as

Pollinator dispersal in an agricultural matrix: opposing responses of wild bees and hoverflies to landscape structure and distance from main habitat

  • Frank Jauker
  • Tim Diekötter
  • Franziska Schwarzbach
  • Volkmar Wolters
Research Article

Abstract

Semi-natural habitats provide essential resources for pollinators within agricultural landscapes and may help maintain pollination services in agroecosystems. Yet, whether or not pollinators disperse from semi-natural habitat elements into the adjacent agricultural matrix may to a large extent depend on the quality of this matrix and the corresponding pollinator-specific life history traits. To investigate the effects of matrix quality on the distance decay of wild bees and hoverflies, six transects along vegetated field tracks originating at a large semi-natural main habitat and leading into the adjacent agricultural matrix were established in the Wetterau Region, central Hesse, Germany. Species richness of wild bees did not change with distance from the main habitat in landscapes with sufficient grassland cover in the surrounding landscape, but significantly declined when semi-natural grasslands where scarce and isolated in the adjacent agricultural matrix. Abundance of wild bees declined with distance regardless of matrix quality. Species richness of hoverflies did not decline with increasing distance in any landscape. Abundance even increased with distance to the main habitat independently of matrix quality. Thus, our data show that taxa of the pollinator guild may perceive landscapes quite differently. Because of their differing dispersal modes and resource requirements as compared to wild bees, hoverflies may play an important role in maintaining pollination services in agricultural landscapes unsuitable for bee species. Our results highlight the need for considering these taxon-specific differences when predicting the effect of landscape structure on pollinators.

Keywords

Apidae Syrphidae Landscape context Distribution patterns 

Introduction

Traditionally, bees are considered the most important group of pollinators in agricultural landscapes. Consequently, a great deal of the current pollinator crisis debate centers on this taxon, and so does the existing knowledge about the effects of environmental change on pollination services (e.g., Ghazoul 2005; Steffan-Dewenter et al. 2005). Much less is known about the effects of land-use change on hoverflies, a taxon that only recently has been shown to provide significant pollination services to wild flowers and commercial crops (Fontaine et al. 2006; Jauker and Wolters 2008). Because even closely related bee species vary in their responses to landscape characteristics (Steffan-Dewenter et al. 2002; Cane et al. 2006), effects of land-use intensity and landscape structure may be expected to differ substantially between different insect families such as hoverflies and bees. Here, we aim at a better understanding of such taxon-specific responses of wild bees and hoverflies to landscape structure, which will be crucial in counteracting the ongoing decline in species and functional diversity of pollinators in agricultural landscapes.

Habitat loss and isolation due to agricultural intensification represent a major threat to farmland biodiversity (Pimm and Raven 2000; Benton et al. 2003; Watling and Donnelly 2006) and the associated interactions among species (Kruess and Tscharntke 1994; Tscharntke and Brandl 2004; Tylianakis et al. 2006). Many of the ecologically valuable habitats remaining in agricultural landscapes are confined to a relatively small proportion of semi-natural habitats or protected areas (Bengtsson et al. 2003). Yet, despite this limited extent, semi-natural habitats are often of great importance to the overall species richness (Hendrickx et al. 2007; Billeter et al. 2008) and thereby to the maintenance of ecosystem services in agricultural landscapes (Costanza et al. 1997; Kremen et al. 2004; Klein et al. 2007). Recently, some of the larger of these remaining habitat elements have been shown to serve as important population sources from which pollinators disperse and thereby contribute to higher densities and species richness in the adjacent agricultural area (Duelli and Obrist 2003; Ockinger and Smith 2007). The dispersal success of individuals between spatially separated habitat elements is strongly related to the isolation of these elements. The degree of isolation, however, not only depends on the geographic distance between habitat fragments but also on the intervening type of land-use—the so called matrix.

Dispersal distances for a given pollinator species could be shown to differ between different matrix types (Roland et al. 2000) and for a given matrix type the permeability could be shown to vary even among closely related pollinator species (Ricketts 2001). In accordance, pollinator community structure in focal habitats can often be explained to a certain extent by the diversity of adjacent land-use forms (Steffan-Dewenter et al. 2002; Morandin et al. 2007). Thus, whether or not habitat elements are effectively isolated, strongly depends on the surrounding matrix as well as on a species’ ability to utilize specific landscape features for movement (Tischendorf and Fahrig 2000).

Landscape features that have generally been reported to facilitate successful pollinator dispersal from source habitats are corridors (Debinski and Holt 2000; Haddad et al. 2003; Townsend and Levey 2005). These corridors (e.g., hedges, ditches, field track margins) may provide resources required for a species to successfully complete its life cycle (Cane 2001; Dennis et al. 2003). However, the effectiveness of corridors is likely conditional on the quality of the surrounding matrix as it is seldom entirely hostile but may instead provide additional resources for multi-habitat users (e.g., foraging or nesting habitats).

A particular difficulty associated with the evaluation of landscape suitability for pollinators relates to the enormous differences in the resource requirements between many pollinator groups such as bees and hoverflies. While all pollinators rely on floral resources as adults, larval requirements differ substantially within, but even more prominently between taxa. Wild bees, for instance, build brood cells and collect pollen for their offspring. For larval feeding, they often have to commute between nesting and nearby foraging habitats and are therefore—to various degrees—spatially constricted during their reproductive period. Hoverflies, in contrast, select suitable larval microhabitats for oviposition without any need to return to these sites later on. Their larval feeding modes are extremely diverse (i.e., carnivorous, phytophagous, saprophagous, mycophagous), but no pollen or nectar feeding larvae are known (Speight 2006). Therefore, generalizations of how landscape structure and resource availability will affect the diversity and distribution of pollinators in general may be inaccurate (Cane et al. 2006; Ewers and Didham 2006) and result in misleading recommendations for nature conservation. Instead, we argue that the dispersal and distribution of pollinators in the agricultural landscape strongly depend on the interplay of taxon-specific resource requirements and matrix quality.

Here, we surveyed dispersal and distribution patterns of wild bees and hoverflies in an intensively managed agricultural region in central Germany. Because of the described differences in their resource requirements, the two pollinator taxa were expected to show differences in distance decay along vegetated field track margins from semi-natural habitats into the agricultural matrix of varying quality. Specifically, we hypothesised that (1) both pollinator taxa show a distance decay along the corridors from large semi-natural main habitats into the agricultural matrix, (2) this distance decay is more pronounced in central place foragers like bees than in hoverflies that are less spatially restricted, and (3) the distance decay is less pronounced along corridors in landscapes of high matrix quality as opposed to landscapes of low matrix quality.

Methods

Study site

The study was conducted in the Wetterau, an intensively managed agricultural region of about 1,000 km2 in central Hesse, Germany. More than 50% of the area is farmland (ca. ¾ crops and ¼ grassland), about 30% is woodland. In the central region, semi-natural habitats mainly aggregate along the river Wetter, forming a continuous, linear, and structurally rich ‘green belt’. Within the surrounding landscape semi-natural habitats are only sparsely scattered throughout the arable land and are almost exclusively represented by grasslands like meadows, orchards, and fallows. As numerous conservation areas harbouring diverse wild bee communities are located within the ‘green belt’ (Frommer 2001), we treated this whole area as a large main habitat for pollinators in this region, with field margins providing the only continuous linear structures that potentially serve pollinators as corridors to the agricultural area.

Pollinator surveys

Originating in the main habitat, we established six 2,000 m transects for the sampling of wild bees and hoverflies. Each transect followed field margins along unpaved farm tracks that lead perpendicularly from the main habitat into the area under crop. Three transects were directed to the west and three to the east. They were separated from each other by at least 500 m (Fig. 1). Since grasslands scattered in the matrix most likely have a strong impact on floral resource availability, matrix quality within a 250 m buffer around each transect was classified according to grassland cover as high (grassland cover of more than 20%), medium (grassland cover of 10–20%), and low (grassland cover of less than 10%). Each category of matrix quality was represented by two transects. Landscapes were categorized by using digitized maps of land-use for which information was recorded in the field prior to the survey. Landscape analyses were done in ArcView GIS 3.3, ESRI, Redlands, CA.
Fig. 1

A schematic map of the study region and the six established transects. The river Wetter (grey band), three villages (from top: Gambach, Rockenberg, Steinfurth; dark grey), large, well connected and small scattered semi-natural habitats (grey) are embedded in the intensively managed agricultural region (light grey). Black dots indicate each of the 20 sampling points per transect, spacing between two sampling points of each transect was 100 m. The transects followed field margins in three different matrix quality classes: (1) high matrix quality (>20% grassland cover in a 250 m radius; top left and bottom right transects), (2) medium matrix quality (10–20% grassland cover; top right and bottom left transects), and (3) low matrix quality (<10% grassland cover; right and left middle transects)

We conducted five surveys in June and July 2006. Wild bees and hoverflies were recorded at point stops every 100 m along the transects, resulting in 20 recordings per transect and 120 recordings for six transects per survey. For each survey, the sequence in which transects were sampled was randomly determined. At point stops, wild bees and hoverflies were sampled for 20 min using insect nets and were almost exclusively caught during flower visitation. Records were taken along field margins following the field tracks for 10 m in each direction from the point stop. Pollinator sampling took place between 10 a.m. and 5 p.m. on sunny days with little wind. At each point stop, we also determined all flowering plant species to the genus level and estimated the percent flower cover within the sampled area.

Pollinators were identified to the species level in the lab. Some individuals could only be determined either to the genus level (hoverflies: Heringia sp., Paragus sp., Pipizella sp.) or to the level of species groups (hoverflies: Cheilosia vernalis—group, Eumerus strigatus/sogdianus; wild bees: Andrena minutula—group, Halictus simplex agg.). The most abundant taxa were not caught, but identified on the wing to the highest level of taxonomic resolution possible (species: Episyrphus balteatus, Eristalis tenax, Syritta pipiens, Bombus pascuorum, B. Sylvarum, genus: Melanostoma sp., Sphaerophoria sp., species groups: Bombus lapidarius/soroeensis, B. terrestris agg.).

Statistical analyses

For our analyses species richness and abundances recorded at each of 120 point stops were pooled over the five surveys. We calculated general linear mixed effect models in Statistica 7.1 (StatSoft, Tulsa, Okla.; Type III sums of squares) for the dependent variables wild bee species number, hoverfly species number, wild bee abundance, and hoverfly abundance. We included matrix quality of the surrounding landscape as a categorical variable with three categories (low, medium, high) and distance from the main semi-natural habitats as a continuous variable (100, 200,… 2,000 m), as well as the interaction of both. The predefined categories of matrix quality adequately represented the flowering plant richness and cover along the corridors (Supplement 1). Accordingly, matrix quality comprises the amount of potential foraging sites around and resource availability along the corridors.

As each of the 20 recording points of one transect belongs to the same landscape, we nested the random factor “transect” into the landscape factor to avoid pseudo replication (Underwood 2002). All dependent variables were log-transformed and met the assumptions of normal distribution and homogeneity of variances.

Model residuals for bee species richness, bee abundance, hoverfly species richness, and hoverfly abundance were assessed for spatial autocorrelation using Moran’s I correlograms with a maximum distance lag of 1,000 m (Fortin and Dale 2005; Wagner and Fortin 2005). Monte Carlo procedures with 5,000 permutations at each distance lag (100, 200, .., 1,000 m) were used to detect significant departures of the observed data from the reference distribution (Sawada 1999). Despite some significant values at rather large distance lags, there was no systematic pattern of spatial autocorrelation (Supplement 2).

Results

We recorded 2,137 bee individuals representing 76 species or species groups in 16 genera. The most abundant and widespread bee species were Bombus terrestris agg. (18.1%), B. lapidarius/soroeensis (17.8%), and B. pascuorum (12.4%). A total 1,484 hoverfly individuals were recorded representing 46 species or species groups in 28 genera. The most abundant and widespread species were Sphaerophoria sp. (33.2%), Episyrphus balteatus (25.8%), and Eupeodes corollae (9.4%).

Bee species richness was significantly affected by distance to the main habitat (Table 1). The significant interaction with matrix quality indicates, however, that bee species richness was only negatively affected by distance to the main habitat in landscapes of low matrix quality (single regression: P < 0.001, R² = 0.29), whereas no distance effect could be established for landscapes of medium and high matrix quality (Fig. 2). No effects of distance and matrix quality were detected for hoverfly species richness (Table 1).
Table 1

Results of the general linear mixed effect models (Type III sums of squares) for the dependent variables wild bee species number, hoverfly species number, wild bee abundance, and hoverfly abundance (Statistica 7.1 StatSoft, Tulsa, Okla.)

 

Bee species richness

Hoverfly species richness

Bee abundance

Hoverfly abundance

 

df

F/P

F/P

F/P

F/P

Matrix quality

2, 3

0.67/NS

0.62/NS

0.10/NS

0.11/NS

Distance

1, 3

20.59/0.020

0.93/NS

15.57/0.029

13.30/0.036

Matrix quality × distance

2, 3

9.78/0.048

0.71/NS

6.39/NS

1.28/NS

Transect

3, 108

1.46/NS

1.77/NS

2.24/NS

1.48/NS

Transect × distance

3, 108

0.82/NS

2.53/NS

1.02/NS

1.41/NS

Matrix quality refers to grassland cover at three levels: low (<10% grassland cover), medium (10–20% grassland cover), high (>20% grassland cover) in a buffer of 250 m around the established transects; distance refers to distance from the large, aggregated semi-natural habitats along the river Wetter within the study region (continuously 100, 200, … 2,000 m); matrix quality × distance indicates the interaction of the factors matrix quality and distance; transect (nested in landscape) assigns the 20 sampling points of one transect to each transect and two transects to each level of matrix quality, respectively; the interaction of transect and distance (transect × distance) is included for the correct calculation of the F-values (Underwood 2002). Given are the denominator degrees of freedom (df) and F- and P-values for each term (NS non significant)

Fig. 2

Relation between bee species richness (log-transformed) and distance from main habitat (F = 9.73, P = 0.048) along six transects following field track margins in an agricultural landscape. Matrix quality is defined by cover of semi-natural grasslands in a buffer of 250 m around each transect (low: <10% grassland cover; medium: between 10 and 20% grassland cover; high: >20% grassland cover)

Distance from the main semi-natural habitat had a significant effect on both bee and hoverfly abundance (Table 1). Yet, while bee abundance declined with increasing distance from the main habitat (Fig. 3a), hoverfly abundance increased (Fig. 3b). No effect of “matrix quality” or “matrix quality x distance” on the abundance of bees and hoverflies could be established.
Fig. 3

Relationship of bee abundance (a) and hoverfly abundance (b) with respect to distance from main habitat along six transects following field track margins in an agricultural landscape (bees: F = 15.57, P = 0.029; hoverflies: F = 13.30, P = 0.036)

The fact that the factors “transect” and “transect × distance” did neither affect species richness nor abundance of bees and hoverflies proves that the location of transects did not bias our findings.

Discussion

Field margins are landscape features effectively enhancing species immigration to and emigration from habitats and may thus be considered as classical corridors (Collinge 2000; Townsend and Levey 2005). Here, we showed that along field margins the abundance of wild bees declined while that of hover flies increased with the distance to the source habitat. Considering that abundance most probably is closely related to pollination efficiency (Hayter and Cresswell 2006), this shift in the bee-to-hoverfly ratio may dramatically impact plant-pollinator interactions in the surrounding landscape. This effect could be particularly strong in landscapes of low matrix quality, when the decline of bee abundance is accompanied by a decline of species richness.

In agricultural landscapes of low matrix quality the few grasslands and the scarce flower resources along the field margins fail in supporting diverse bee communities or are too isolated to allow species’ dispersal (Steffan-Dewenter 2003). In contrast, a heterogeneous mosaic of land-use types including a sufficient cover of grassland habitats of well over 10% within a 250 m radius, approximately the foraging distance of small bees (Gathmann and Tscharntke 2002), supports species’ dispersal and cannot be considered a hostile matrix.

Although declining bee species richness with distance to the main habitat was restricted to landscapes with poor matrix quality, the community structure might be affected in a more general way (Brosi et al. 2007). Accordingly, abundance of pollinators responded quite differently to distance than species numbers. Again, more bees were detected close to the main habitat, but the subsequent decline in the farmed areas with increasing distance was independent of matrix quality. Thus, linear habitats like field margins cannot sustain bee populations equal in size to those in large habitats (Steffan-Dewenter 2003). Instead, by providing ample rewards in food resources but not necessarily in nesting sites, they may either function as foraging corridors or true sinks and pseudosinks (cf. Thomas and Kunin 1999). Accordingly, the most dominant species found were bumble bees (ca 55% of the individuals), which are capable of long distance foraging and occasionally build nests along the unpaved field tracks (personal observation).

Compared to bees, hoverflies seem to cope better with managed agricultural areas, as species richness was generally not affected by distance from the main habitat even in landscapes of poor matrix quality. While diverse agricultural landscapes have previously been shown to facilitate hoverfly movement and support diverse hoverfly communities (Burgio and Sommaggio 2007; Schweiger et al. 2007; Meyer et al. 2008), our results suggest that even small grassland patches and scarce flower resources along field margins may be utilized by a variety of species. The main reason for the opposing pattern of distance decay in hoverfly and bee species can likely be seen in the differing breeding strategy. Hoverflies, unlike bees, do not collect pollen and nectar for their offspring. Thus, females can disperse into landscapes in a linear manner, alternating between feeding and ovipositioning, whereas bees need to return to their brood cells repeatedly after foraging (Kleijn and van Langevelde 2006). Most syrphids, however, show an extremely specialized selection of microhabitat sites for larval development that are usually not found in intensively managed agroecosystems (Speight 2006). Accordingly, most species detected in the study area are associated with agricultural habitats, as larvae feed on aphids in crops (e.g., Sphaerophoria sp., Episyrphus balteatus, Eupeodes corollae, Melanostoma sp.), on rotting organic matter in dung heaps and the like (e.g., Syritta pipiens), or live aquatically in small, extremely eutrophic water bodies like drainages or liquid manure assemblages (e.g., Eristalis tenax).

The dependency of hoverflies on managed fields becomes even more obvious when looking at the observed abundances. The number of individuals increased with distance from the main semi-natural habitat. The overall abundance, however, was mainly made up by individuals that feed on aphids as larvae (>80% in total). Gravid females are likely to disperse from flower-rich semi-natural habitats, i.e., adult feeding habitat, and oviposit in the aphid-rich fields, i.e., spatially separated larval feeding habitat. As a result, for each female a many times fold number of offspring starts dispersal from the larval habitat. Because the offspring does not necessarily disperse into the very same adult feeding habitat of their ancestors, no accumulation of individuals might be expected. Larger areas of larval feeding habitat will therefore locally increase population sizes. A simultaneous increase in species richness due to larger heterogeneity in larger areas, however, may not be expected; increased resource heterogeneity in large fields is likely to be ineffective in the aphidophagous guild, because the larvae tend to be rather generalistic in prey species selection (Sadeghi and Gilbert 2000).

The contrasting patterns of the distance decay between wild bees and hoverflies observed in our study is in opposition to the recent ambitions to generalize relations of distance to natural or semi-natural habitats and both diversity and functioning of pollinators (Ricketts et al. 2008). Instead, our results substantiate Ricketts et al. (2008) assumption that in temperate regions distance decay in species numbers may be less consistent within the pollinator guild than in the tropics. The absence of any distance effect in hoverfly richness and the interaction of distance and matrix quality in bee species richness indicate that distance decay in temperate regions is strongly dependent upon dispersal-related life history traits of the examined pollinator taxon.

The reason why we could not establish a direct link to the landscape for wild bees might be explained by the spatial proximity of the origins of the transects in the main habitat, thus blurring any landscape effects at this spatial scale (Steffan-Dewenter et al. 2002). In the comparatively monotonous agricultural landscape of our study, these large continuous semi-natural habitats may represent rather one large source habitat for all landscape sections instead of several small ones for each section. This emphasizes the value of large conservation areas in order to conserve stable pollinator metapopulations at large scales (Pauw 2007). For hoverflies, the reason for the lack of a direct landscape effect might be a different one. In the dominant aphidophagous guild, reproducing in arable fields with no need for the adults to actively provide food for the offspring, the limitation of resources depressing abundances must be associated with larval requirements to a much larger extent than with adult requirements. This is clearly shown by the reversed pattern of the expected source–sink relationship. Thus, the agricultural landscape cannot generally be considered a matrix (Haila 2002) but is part of the habitat, alas possibly prone to shifts in species’ composition (Tscharntke et al. 2005). In terms of diversity conservation issues, however, protected areas and semi-natural grasslands might still be crucial for persistence of non-aphidophagous hoverfly species not detected in our study sites.

Conclusions

Large semi-natural habitats and protected areas are important elements for enhancing pollinator diversity in agricultural landscapes. They harbour diverse wild bee communities and can be considered as classical source habitats. Dispersal from these main habitats into the agricultural area, however, is strongly dependent on the surrounding landscape and dispersal-related life history traits of species. Both linear and patchy semi-natural habitat elements in the agricultural matrix could be shown to lower the distance decay of pollinators from source habitats into the agricultural matrix. The rapid decline of bee species richness and abundance with increasing distance from main habitats in landscapes of high land-use intensity, however, indicates that pollination provided by wild bees might be limited by too long distances or insufficient floral resources in the matrix around the main habitats. Because of different habitat requirements, hoverflies may play a very important role in maintaining pollination services in agricultural landscapes unsuitable for specialized or less mobile bee species. Thus, our results show that generalizations regarding the effects of landscape structure on pollinators across the various taxa in this guild are not justified and must lead to erroneous recommendations to landscape planning and nature conservation. Yet, with regard to global change it seems wise to create landscape conditions that maintain a maximal diversity of pollinators in order to ensure this essential ecosystem service also in the long-run.

Notes

Acknowledgments

We would like to thank Yann Clough, Birgit Meyer, and an anonymous referee for valuable comments on the manuscript. Dean Anderson provided valuable help with the spatial autocorrelation procedures. Ulrich Frommer (Apidae) and Paul-Walter Lohr (Syrphidae) helped to identify dubious species. This work was supported by a doctoral scholarship from the German Environmental Foundation (DBU) to F. Jauker.

Supplementary material

10980_2009_9331_MOESM1_ESM.doc (2.9 mb)
(DOC 3014 kb)

References

  1. Bengtsson J, Angelstam JP, Elmqvist T et al (2003) Reserves, resilience, and dynamic landscapes. Ambio 32:389–396. doi:10.1639/0044-7447(2003)032[0389:RRADL]2.0.CO;2 PubMedCrossRefGoogle Scholar
  2. Benton TG, Vickery JA, Wilson JD (2003) Farmland biodiversity: is habitat heterogeneity the key? Trends Ecol Evol 18:182–188. doi:10.1016/S0169-5347(03)00011-9 CrossRefGoogle Scholar
  3. Billeter R, Liira J, Bailey D et al (2008) Indicators for biodiversity in agricultural landscapes: a pan-European study. J Appl Ecol 45:141–150CrossRefGoogle Scholar
  4. Brosi BJ, Daily GC, Ehrlich PR (2007) Bee community shifts with landscape context in a tropical countryside. Ecol Appl 17:418–430. doi:10.1890/06-0029 PubMedCrossRefGoogle Scholar
  5. Burgio G, Sommaggio D (2007) Syrphids as landscape indicators in Italian agroecosystems. Agric Ecosyst Environ 120:416–422. doi:10.1016/j.agee.2006.10.021 CrossRefGoogle Scholar
  6. Cane JH (2001) Habitat fragmentation and native bees: a premature verdict? Conserv Ecol 5: art. no. 3 [online]. http://www.consecol.org/vol5/iss1/art3/
  7. Cane J, Minckley RL, Kervin LJ et al (2006) Complex responses within a desert bee guild (Hymenoptera: Apiformes) to urban habitat fragmentation. Ecol Appl 16:632–644. doi:10.1890/1051-0761(2006)016[0632:CRWADB]2.0.CO;2 PubMedCrossRefGoogle Scholar
  8. Collinge SK (2000) Effects of grassland fragmentation on insect species loss, colonization, and movement patterns. Ecology 81:2211–2226Google Scholar
  9. Costanza R, d’Arge R, deGroot R et al (1997) The value of the world’s ecosystem services, and natural capital. Nature 387:253–260. doi:10.1038/387253a0 CrossRefGoogle Scholar
  10. Debinski DM, Holt RD (2000) A survey and overview of habitat fragmentation experiments. Conserv Biol 14:342–355. doi:10.1046/j.1523-1739.2000.98081.x CrossRefGoogle Scholar
  11. Dennis RHL, Shreeve TG, van Dyck H (2003) Towards a functional resource-based concept for habitat: a butterfly biology viewpoint. Oikos 102:417–426. doi:10.1034/j.1600-0706.2003.12387.x CrossRefGoogle Scholar
  12. Duelli P, Obrist MK (2003) Regional biodiversity in an agricultural landscape: the contribution of seminatural habitat islands. Basic Appl Ecol 4:129–138. doi:10.1078/1439-1791-00140 CrossRefGoogle Scholar
  13. Ewers RM, Didham RK (2006) Confounding factors in the detection of species responses to habitat fragmentation. Biol Rev Camb Philos Soc 81:117–142. doi:10.1017/S1464793105006949 PubMedGoogle Scholar
  14. Fontaine C, Dajoz I, Meriguet J et al (2006) Functional diversity of plant-pollinator interaction webs enhances the persistence of plant communities. PLoS Biol 4:129–135. doi:10.1371/journal.pbio.0040001 CrossRefGoogle Scholar
  15. Fortin M-J, Dale M (2005) Spatial Analysis. A guide for ecologists, Cambridge University Press, CambridgeGoogle Scholar
  16. Frommer U (2001) Bestandsaufnahme der Bienenfauna im mittleren Hessen (Hymenoptera, Apidae). Naturwiss Ver Darmstadt Ber 24:129–191Google Scholar
  17. Gathmann A, Tscharntke T (2002) Foraging ranges of solitary bees. J Anim Ecol 71:757–764. doi:10.1046/j.1365-2656.2002.00641.x CrossRefGoogle Scholar
  18. Ghazoul J (2005) Buzziness as usual? Questioning the global pollination crisis. Trends Ecol Evol 20:367–373. doi:10.1016/j.tree.2005.04.026 PubMedCrossRefGoogle Scholar
  19. Haddad NM, Bowne DR, Cunningham A et al (2003) Corridor use by diverse taxa. Ecology 84:609–615. doi:10.1890/0012-9658(2003)084[0609:CUBDT]2.0.CO;2 CrossRefGoogle Scholar
  20. Haila Y (2002) A conceptual genealogy of fragmentation research: from island biogeography to landscape ecology. Ecol Appl 12:321–334Google Scholar
  21. Hayter KE, Cresswell JE (2006) The influence of pollinator abundance on the dynamics and efficiency of pollination in agricultural Brassica napus: implications for landscape-scale gene dispersal. J Appl Ecol 43:1196–1202. doi:10.1111/j.1365-2664.2006.01219.x CrossRefGoogle Scholar
  22. Hendrickx F, Maelfait JP, van Wingerden W et al (2007) How landscape structure, land-use intensity and habitat diversity affect components of total arthropod diversity in agricultural landscapes. J Appl Ecol 44:340–351. doi:10.1111/j.1365-2664.2006.01270.x CrossRefGoogle Scholar
  23. Jauker F, Wolters V (2008) Hover flies are efficient pollinators of oilseed rape. Oecologia 156:819–823. doi:10.1007/s00442-008-1034-x PubMedCrossRefGoogle Scholar
  24. Kleijn D, van Langevelde F (2006) Interacting effects of landscape context and habitat quality on flower visiting insects in agricultural landscapes. Basic Appl Ecol 7:214–301. doi:10.1016/j.baae.2005.07.011 CrossRefGoogle Scholar
  25. Klein AM, Vaissiere BE, Cane JH et al (2007) Importance of pollinators in changing landscapes for world crops. P R Soc Lond B Biol 274:303–313. doi:10.1098/rspb.2006.3721 CrossRefGoogle Scholar
  26. Kremen C, Williams NM, Bugg RL et al (2004) The area requirements of an ecosystem service: crop pollination by native bee communities in California. Ecol Lett 7:1109–1119. doi:10.1111/j.1461-0248.2004.00662.x CrossRefGoogle Scholar
  27. Kruess A, Tscharntke T (1994) Habitat fragmentation, species loss, and biological control. Science 264:1581–1584. doi:10.1126/science.264.5165.1581 PubMedCrossRefGoogle Scholar
  28. Meyer B, Jauker F, Steffan-Dewenter I (2008) Contrasting resource-dependent responses of hoverfly richness and density to landscape structure. Basic Appl Ecol. doi:10.1016/j.baae.2008.01.001 Google Scholar
  29. Morandin LA, Winston ML, Abbott VA et al (2007) Can pastureland increase wild bee abundance in agriculturally intense areas? Basic Appl Ecol 8:117–124. doi:10.1016/j.baae.2006.06.003 CrossRefGoogle Scholar
  30. Ockinger E, Smith HG (2007) Semi-natural grasslands as population sources for pollinating insects in agricultural landscapes. J Appl Ecol 44:50–59. doi:10.1111/j.1365-2664.2006.01250.x CrossRefGoogle Scholar
  31. Pauw A (2007) Collapse of a pollination web in small conservation areas. Ecology 88:1759–1769. doi:10.1890/06-1383.1 PubMedCrossRefGoogle Scholar
  32. Pimm SL, Raven P (2000) Biodiversity—Extinction by numbers. Nature 403:843–845. doi:10.1038/35002708 PubMedCrossRefGoogle Scholar
  33. Ricketts TH (2001) The matrix matters: effective isolation in fragmented landscapes. Am Nat 158:87–99. doi:10.1086/320863 PubMedCrossRefGoogle Scholar
  34. Ricketts TH, Regetz J, Steffan-Dewenter I et al (2008) Landscape effects on crop pollination services: aere there general patterns? Ecol Lett 11:499–515. doi:10.1111/j.1461-0248.2008.01157.x PubMedCrossRefGoogle Scholar
  35. Roland J, Keyghobadi N, Fownes S (2000) Alpine Parnassius butterfly dispersal: effects of landscape and population size. Ecology 81:1642–1653Google Scholar
  36. Sadeghi H, Gilbert F (2000) Aphid suitability and its relationship to oviposition preference in predatory hoverflies. J Anim Ecol 69:771–784. doi:10.1046/j.1365-2656.2000.00433.x CrossRefGoogle Scholar
  37. Sawada M (1999) ROOKCASE: an Excel 97/2000 visual basic (VB) add-in for exploring global and local spatial autocorrelation. Bull Ecol Soc Am 80:231–234. doi:10.1890/0012-9623(1999)080[0231:TT]2.0.CO;2 CrossRefGoogle Scholar
  38. Schweiger O, Musche M, Bailey D et al (2007) Functional richness of local hoverfly communities (Diptera, Syrphidae) in response to land use across temperate Europe. Oikos 116:461–472. doi:10.1111/j.2007.0030-1299.15372.x CrossRefGoogle Scholar
  39. Speight MCD (2006) Species accounts of European Syrphidae (Diptera), Ferrara 2006. In: Speight MCD, Castella E, Sarthou J-P, Monteil C (eds) Syrph the Net on CD, Issue 5. The database of European Syrphidae. Syrph the Net Publications, DublinGoogle Scholar
  40. Steffan-Dewenter I (2003) Importance of habitat area and landscape context for species richness of bees and wasps in fragmented orchard meadows. Conserv Biol 17:1036–1044. doi:10.1046/j.1523-1739.2003.01575.x CrossRefGoogle Scholar
  41. Steffan-Dewenter I, Münzenberg U, Bürger C et al (2002) Scale-dependent effects of landscape context on three pollinator guilds. Ecology 83:1421–1432CrossRefGoogle Scholar
  42. Steffan-Dewenter I, Potts SG, Packer L (2005) Pollinator diversity and crop pollination services are at risk. Trends Ecol Evol 20:651–652. doi:10.1016/j.tree.2005.09.004 PubMedCrossRefGoogle Scholar
  43. Thomas CD, Kunin E (1999) The spatial structure of populations. J Anim Ecol 68:647–657. doi:10.1046/j.1365-2656.1999.00330.x CrossRefGoogle Scholar
  44. Tischendorf L, Fahrig L (2000) How should we measure landscape connectivity? Landscape Ecol 15:633–641. doi:10.1023/A:1008177324187 CrossRefGoogle Scholar
  45. Townsend PA, Levey DJ (2005) An experimental test whether habitat corridors affect pollen transfer. Ecology 86:466–475. doi:10.1890/03-0607 CrossRefGoogle Scholar
  46. Tscharntke T, Brandl R (2004) Plant-insect interactions in fragmented landscapes. Annu Rev Entomol 49:405–430. doi:10.1146/annurev.ento.49.061802.123339 PubMedCrossRefGoogle Scholar
  47. Tscharntke T, Klein AM, Kruess A et al (2005) Landscape perspectives on agricultural intensification and biodiversity—ecosystem service management. Ecol Lett 8:857–874. doi:10.1111/j.1461-0248.2005.00782.x CrossRefGoogle Scholar
  48. Tylianakis JM, Tscharntke T, Klein AM (2006) Diversity, ecosystem function, and stability of parasitoid host interactions across a tropical habitat gradient. Ecology 87:3047–3057. doi:10.1890/0012-9658(2006)87[3047:DEFASO]2.0.CO;2 PubMedCrossRefGoogle Scholar
  49. Underwood AJ (2002) Experiments in ecology. Cambridge University Press, CambridgeGoogle Scholar
  50. Wagner HH, Fortin M-J (2005) Spatial analysis of landscapes: concepts and statistics. Ecology 86:1975–1987. doi:10.1890/04-0914 CrossRefGoogle Scholar
  51. Watling JI, Donnelly MA (2006) Fragments as islands: a synthesis of faunal responses to habitat patches. Conserv Biol 20:1016–1025. doi:10.1111/j.1523-1739.2006.00482.x PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  • Frank Jauker
    • 1
  • Tim Diekötter
    • 1
  • Franziska Schwarzbach
    • 1
  • Volkmar Wolters
    • 1
  1. 1.Department of Animal EcologyJustus Liebig UniversityGiessenGermany

Personalised recommendations