Skip to main content
NCBI home page
Wiley - PMC COVID-19 Collection logoLink to Wiley - PMC COVID-19 Collection
. 2017 Jul 10;11(1):72–87. doi: 10.1111/icad.12245

Conservation of hoverflies (Diptera, Syrphidae) requires complementary resources at the landscape and local scales

Laura Moquet 1,✉, Estelle Laurent 1,2, Rossana Bacchetta 1,3, Anne‐Laure Jacquemart 1
Editors: Raphael Didham, Francis Gilbert
PMCID: PMC7165621  PMID: 32336985

Abstract

  1. Accumulating evidence shows that landscape fragmentation drives the observed worldwide decline in populations of pollinators, particularly in species of Lepidoptera and Hymenoptera. However, Little is known about the effects of landscape fragmentation on hoverfly (Diptera, Syrphidae) communities. Hoverflies provide varied ecosystem services: larvae contribute to waste decomposition (saprophagous species) and pest control (aphidophagous species), and adults pollinate a wide range of flowers.

  2. To determine how the diversity and quantity of resources for larvae and adults affect hoverfly abundance and species richness at three spatial scales, we recorded insect visitors of five target plant species in Belgian heathlands, habitats that have decreased considerably due to human activities.

  3. Hoverflies represented the most abundant visitors on two plant species, and the second most abundant visitors (after bumblebees) on the other target plant species. A large proportion of hoverflies observed were aphidophagous species associated with coniferous and deciduous forests. Resources for the larvae and floral resources for the adults influenced interactions among hoverflies and plants, but acted at different scales: larval habitat availability (distance to larval habitat) was relevant at the landscape scale, whereas adult resource availability (floral density) was relevant at the plot scale.

  4. Hoverfly abundance and species richness decreased with distance to larval habitat but increased with floral density. Moreover, landscape structure and composition had different effects according to hoverfly ecological traits. Landscape composition influenced aphidophagous but not saprophagous hoverflies, in that their abundance and species richness decreased with distance to forests. Maintenance of the interactions between plants and their hoverfly visitors requires complementary resources at both landscape and local scales.

Keywords: Aphidophagous, ericaceous species, forest, heathlands, pollinators

Introduction

As an ecosystem service, pollination was estimated to be worth €153 billion in 2005 (Gallai et al., 2009). Bees (Hymenoptera), hoverflies and other anthophilous flies (Diptera), butterflies (Lepidoptera) and beetles (Coleoptera) are the main pollinators in temperate regions (Kevan & Baker, 1983; Vanbergen & The Insect Pollinators Initiative, 2013; Rader et al., 2016). Many studies have shown that the abundance and species richness of pollinators has decreased over the past 50 years, and that the ecological homogenisation of bee and butterfly communities has increased (Biesmeijer et al., 2006; Carvalheiro et al., 2013). These studies have highlighted land‐use changes that destroy, fragment and degrade habitats as one of the main causes of this decline in pollinators (Goulson et al., 2005; Stokstad, 2006; Potts et al., 2010). Although many studies have examined the effects of fragmentation on bees and butterflies (Cane, 2001; Maes & Van Dyck, 2001; Van Dyck et al., 2009; Winfree et al., 2009; Hadley & Betts, 2011), few studies have examined the effects on hoverfly communities, and these studies have produced limited, often contradictory, results (Biesmeijer et al., 2006; Keil et al., 2011; Carvalheiro et al., 2013).

Adult hoverflies (Diptera, Syrphidae) depend on floral resources, that is, pollen and nectar. Pollen is consumed mainly by females to provide nutrients (proteins, sterols) for egg production, and nectar is mainly consumed by males and females to provide energy (sugars) for flight while searching for partners (Haslett, 1989; Woodcock et al., 2014). Hoverflies are pollinators or co‐pollinators of many plant species (Larson et al., 2001; Inouye et al., 2015). Although there appears to be little specialisation at the species level, individuals can be highly flower constant over short time scales (Branquart & Hemptinne, 2000). Their degree of specialisation is linked to distinct morphological traits. For example, large‐body species with a short proboscis are limited to flowers that produce unconcealed nectar, such as several species of the Rosaceae or Apiaceae family. By contrast, ubiquitous hoverfly species generally have a small body with a long proboscis (Gilbert, 1981, 1986; Branquart & Hemptinne, 2000).

Hoverfly larvae have extremely diverse diets. Some species are saprophagous, feeding on decaying plant or animal matter and thereby helping to decompose waste. Other species are insectivorous (aphids, thrips, etc.) and provide biological pest control, especially in agricultural landscapes. A number of these species are migrants that exploit local fluctuations in aphid abundance (Stubbs & Falk, 1983). Some species are phytophagous or mycophagous (Ball & Morris, 2015; Speight et al., 2015). Their diets, therefore, restrict each species to specific microhabitats (identifiable structural features used by hoverfly larvae such as plant debris, foliage or cow dung) during larval development. Their restricted microhabitats allow hoverflies to be used as biological indicators to evaluate the state of conservation of a specific ecosystem (Sommaggio, 1999; Speight et al., 2000; Burgio & Sommaggio, 2007).

Ecological and morphological traits of hoverflies could influence their responses to landscape composition and resource availability. For example, several studies have shown that fragmentation affects species differently depending on their level of specialisation and ability to disperse (Schweiger et al., 2007). Nevertheless, community analyses supported by ecological and morphological data are infrequent in animal ecology, mainly due to a lack of information at the community level. The Syrph the Net database (Speight et al., 2015), which contains data on the ecological traits of hoverfly species, has facilitated community‐level analyses for hoverflies. To analyse the responses of hoverfly communities to environmental conditions, studies group species with similar ecological traits (Branquart & Hemptinne, 2000; Ouin et al., 2006; Schweiger et al., 2007). Most studies separate hoverfly groups by ecological traits such as larval diet, larval microhabitat or adult body size (Dziock, 2006; Schweiger et al., 2007; Keil et al., 2008), or a combination of larval and adult traits (Ouin et al., 2006).

Because hoverflies require different resources at the larval and adult stages, the persistence of hoverfly communities requires a set of varied resources. For example, larvae of Scaeva selenitica grow under closed‐canopy conditions, whereas adults need open biotopes rich in floral resources (Gittings et al., 2006). Thus, the composition of the landscape matrix matters at the landscape scale (Kleijn & Van Langevelde, 2006; Meyer et al., 2009; Öckinger et al., 2012). Furthermore, dispersal capability influences the spatial scale at which individuals perceive their environment. Species of hoverflies vary in their dispersal capability from a few metres to 2 km per day (Schönrogge et al., 2006; Schweiger et al., 2007). In general, hoverflies are affected by the landscape matrix in a radius of 500–1000 m (Kleijn & Van Langevelde, 2006; Meyer et al., 2009).

In addition to landscape scale resources, the resources available at smaller scales also affect the spatial distribution of adult hoverflies. At the patch scale (habitat fragments), the probability of finding suitable microhabitats and floral resources increases with patch size. Thus, the abundance and species richness of hoverflies are predicted to be higher in large patches than in small patches (Sarthou et al., 2005; Ouin et al., 2006; Herrault et al., 2016). At the local scale, optimal foraging theory suggests that hoverflies should preferentially visit plots with high floral density because such plots provide a high quantity of resources at a low foraging cost (Meyer et al., 2009).

The literature provides increasing information about the effects of landscape changes on hoverfly populations and hoverfly–plant interactions in forests (Gittings et al., 2006; Ouin et al., 2006; Meyer et al., 2009) and agricultural landscapes, but these studies have mainly focused on aphidophagous species (Sutherland et al., 2001; Haenke et al., 2014; Raymond et al., 2015). Semi‐natural open habitats such as heathlands remain poorly studied, although they constitute habitats for many hoverfly species (77 species according to Speight et al., 2015; and see also Stubbs & Falk, 1983; Jacquemart, 1993; Mahy et al., 1998; Meyer et al., 2009; Descamps et al., 2015). Heathlands were extensively destroyed and fragmented in Europe during the last two centuries (Gimingham, 1972; Webb, 1998) and have been largely converted into agricultural or afforested areas (e.g. spruce plantations in Belgium, Clicheroux, 1957). Compared to forests and wetlands, heathlands do not offer many microhabitats for larvae (Speight & Castella, 2015), but they do offer floral resources for adult hoverflies during the entire flight period (Moquet et al., 2017a). The main floral resources in heathlands are ericaceous species, which provide pollen and nectar. Ericaceous species have poricidal anthers (which require vibrations to release pollen) and are therefore considered bee pollinated because their pollen is not easily accessible; nevertheless, these species are visited by hoverflies. Moreover, the presence of Salix spp. in wet heathlands in early spring provides floral resources for early emerging species (Moquet et al., 2015).

Here, we analysed the effects of the landscape at different spatial scales on hoverfly–plant interactions in Belgian heathlands. We examined the following aspects of these interactions:

  1. We hypothesised that the dominant hoverfly species observed on ericaceous flowers are species associated with heathlands, according to the Syrph the Net database (Speight & Castella, 2015).

  2. Larval and adult requirements differ substantially among hoverfly species. We supposed that resources for adults and larvae vary across different spatial scales and influence communities of hoverfly visitors. For adults, crucial resources include floral density at the local plot scale, floral cover at the habitat patch scale and availability of floral‐rich habitats at the landscape scale. For larvae, crucial resources include availability of larval microhabitats at both patch and landscape scales.

  3. Hoverflies have diverse ecological and morphological traits. We hypothesised that each ecological hoverfly group is influenced differently by landscape composition and resource availability.

Methods

Studied heathland patches

The study was carried out in 10 heathland patches in the Plateau des Tailles, Upper Ardenne, Belgium (50°10′N; 5°43′E). Eight patches were an assemblage of wet and dry heathlands and the two others, Fange aux Mochettes and Grand Passage, were ombrotrophic mires (raised bogs that receive all water from direct precipitation). Heathland patch elevation ranged from 550 to 630 m and heathland patch size ranged from 0.3 to 40.7 ha (Table 1).

Table 1.

Characteristics of the studied heathland patches located in the Upper Ardenne, Belgium

Heathland patches Coordinates Localities Heathland area (ha) Patch perimeter (m) Altitude (m) Spruce plantation cover Deciduous forest cover Meadow cover Wetland cover Number of macrohabitats (100 m) Distance to spruce plantations (m) Distance to deciduous forests (m) Distance to meadows (m)
Grande fange 50°14′40″N 5°46′45″E Vielsalm 0.3 308 560 8.2 35.6 5.9 50.3 15 195 ± 130 15 ± 20 400 ± 60
Wé des pourceaux 50°14′42″N 5°44′57 Vielsalm 1.1 447 630 57.9 2.5 24.5 1.5 4 35 ± 5 240 ± 20 250 ± 10
Pisserotte 50°13′13″N 5°47′54″E Gouvy 1.6 698 594 42.3 36.5 7.3 10.7 7 85 ± 60 20 ± 15 235 ± 55
Crépale 50°16′40″N 5°44′07″E Lierneux 7.6 1800 557 79.6 0.0 0.3 20.1 7 30 ± 15 585 ± 0 390 ± 20
Fange aux mochettes 50°13′21″N 5°40′55″E La Roche‐en‐Ardenne 9.6 1225 602 9.8 35.6 8.0 45.8 15 195 ± 85 15 ± 30 150 ± 30
Grand Passage 50°13′44″N 5°45′46″E Houffalize 10.9 1800 611 3.9 5.1 7.6 79.6 13 445 ± 140 230 ± 150 300 ± 150
Pouhon 50°14′32″N 5°41′51″E Manhay 18.6 1937 580 23.2 9.0 0.0 64.1 8 180 ± 20 215 ± 20 590 ± 10
Robiefa 50°15′27″N 5°42′10″E Manhay 25.4 2216 565 32.9 25.4 6.2 31.8 13 180 ± 35 55 ± 20 180 ± 45
Nazieufà 50°15′ 05″N 5°43′09″E Manhay 35.9 2605 625 20.4 0.0 0.0 53.6 8 5 ± 0 480 ± 0 1000 ± 0
Sacrawé 50°14′33″N 5°45′46″E Vielsalm 40.7 3678 600 35.6 15.9 0.0 44.9 9 70 ± 50 m 55 ± 50 360 ± 50
Open in a new tab

Heathland patch area is the surface dominated by ericaceous species. Percentage of each type of land cover was analysed in 2013 within a radius of 500 m from the centre of heathland patches. Number of macrohabitats is the habitat diversity according to Speight and Castella (2015). The minimal distances between the studied plot of 10 m2 of heathland and other types of land cover are indicated (mean ± SD). Heathland patches are arranged by increasing heathland patch size.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

The landscape matrix consisted mainly of forests: spruce plantations of Picea abies (48%) and deciduous forests (8%) of beech (Fagus sylvatica) or oaks (Quercus robur and Q. petraea). These closed habitats were interspersed with open habitats such as pastures and hay meadows (31%) and wetlands and heathlands (8%). Urbanisation and population density were very low (<50 inhabitants per km2, Grimmeau et al., 2013).

Target plant species

Over three successive years (2013–2015), we observed the visitors to the four most abundant ericaceous species, Vaccinium myrtillus L., V. vitis‐idaea L., Erica tetralix L. and Calluna vulgaris (L.) Hull. These ericaceous species are dwarf evergreen shrubs (except V. myrtillus, which is deciduous). Floral morphology differs among species: E. tetralix and V. myrtillus have a deep and urceolate corolla (5–7 mm), V. vitis‐idaea has a campanulate corolla and C. vulgaris has an open corolla (Ritchie, 1955, 1956; Gimingham, 1960; Bannister, 1966). Flowers are tetra‐ or pentameric and secrete nectar at the base of the style. Except for C. vulgaris, pollen grains are contained in poricidal anthers only accessible to species able to generate vibrations (some bee species, Buchmann, 1983). In addition, in 2014, we observed visitors to Salix x multinervis Döll, the main floral resource in early spring in the studied heathlands (Moquet et al., 2015). We observed Salix x multinervis in April, V. myrtillus between April and June, V. vitis‐idaea between May and July, E. tetralix between July and August and C. vulgaris in August.

Hoverfly observations

Insect visitors were surveyed between 9:00 am and 6:20 pm during one to four full days per heathland patch (depending on the weather conditions), per target plant species, and per year. For each target plant species, we performed observations in at least six of the 10 heathland patches using a standardised method (Mayer et al., 2012; Descamps et al., 2015; Moquet et al., 2015). Observations took place when the weather conditions were appropriate (i.e. sunny days and wind speed <5 on the Beaufort scale), by one observer, in a plot of 10 m2 of ericaceous target plant species during 20 min. We conducted a total of 1190 sessions (396 h) over the 3 years. For S. x multinervis, observations were performed on a section of one grove with approximately 500 catkins of male individuals. When plant populations were sufficiently large, the position of the plot was changed each day of observation. Insects were considered visitors when they collected floral resources (pollen or nectar). The observer collected all insects visiting target plant species with an insect net. Hoverflies were identified in the field, to the species level when no confusion was possible (Episyrphus balteatus De Geer, Myathropa florea L., Rhingia campestris Meigen, Scaeva pyrastri L., S. selenitica Meigen, Sericomyia silentis Harris, S. lappona L., Syritta pipiens L., Volucella bombylans L.) or to morphotype, grouping morphologically similar species for the following genera: Chelosia, Chrysotoxum, Eristalis, Eupeodes, Helophilus, Platycheirus, Sphaerophoria, Syrphus. Insects were released on the plot immediately after the 20‐min session, except several individuals per morphotype (a total of 215 individuals) that were killed to determine the precise species diversity within each morphotype. Then, precise identification in the laboratory was conducted, using identification keys (Verlinden, 1994; Van Veen, 2004) and the reference collection of the Royal Belgian Institute of Natural Sciences. Abundance was calculated as the number of individuals and species richness was estimated as the number of morphotypes observed during 1 day (10 sessions of 20 min, Table 2).

Table 2.

Abundance and species richness by observation day (10 × 20 min, mean ± SD) according to heathland patches and plant species

Abundance Species richness
S. x multinervis V. myrtillus V. vitis‐idaea E. tetralix C. vulgaris S. x multinervis V. myrtillus V. vitis‐idaea E. tetralix C. vulgaris
Grande fange 45.0 ± 11.3 (2) 4.0 ± 0.0 (1) 6.0 ± 5.4 (4) 56.4 ± 32.3 (5) 2.0 ± 0.0 (2) 2.0 ± 0.0 (1) 3.0 ± 1.1 (4) 8.0 ± 2.1 (5)
Wé des pourceaux 12.7 ± 9.2 (4) 5.0 ± 3.6 (4)
Pisserotte 17.0 ± 2.8 (2) 9.5 ± 0.7 (2) 2.5 ± 2.5 (6) 68.8 ± 35.5 (5) 3.5 ± 3.5 (2) 5.0 ± 1.4 (2) 1.8 ± 1.3 (6) 8.2 ± 3.4 (5)
Crépale 15.5 ± 9.7 (4) 4.2 ± 1.5 (4)
Fange aux mochettes 72.0 ± 60.8 (2) 2.0 ± 2.5 (5) 10.2 ± 8.5 (7) 54.0 ± 31.4 (4) 1.5 ± 0.7 (2) 1.2 ± 1.3 (5) 4.1 ± 3.5 (7) 8.2 ± 2.3 (4)
Grand passage 32.5 ± 34.6 (2) 0.6 ± 1.1 (3) 1.7 ± 1.7 (4) 6.0 ± 8.1 (7) 39.6 ± 35.3 (3) 2.0 ± 0.0 (2) 0.3 ± 0.5 (3) 1.2 ± 0.9 (4) 2.2 ± 2.2 (7) 5.0 ± 3.4 (3)
Pouhon 14.1 ± 8.3 (6) 55.3 ± 27.0 (3) 3.8 ± 1.1 (6) 8.0 ± 3.6 (3)
Robiefa 4.5 ± 2.3 (4) 1.2 ± 1.7 (5) 8.2 ± 10.5 (9) 82.6 ± 53.0 (3) 1.0 ± 0.0 (4) 0.8 ± 1.3 (5) 2.0 ± 1.6 (9) 10.3 ± 2.0 (3)
Nazieufà 9.5 ± 8.5 (4) 3.2 ± 2.2 (4)
Sacrawé 44.0 ± 32.6 (3) 1.3 ± 2.3 (6) 1.5 ± 1.5 (8) 2.1 ± 3.2 (7) 64.1 ± 37.9 (7) 3.3 ± 3.2 (3) 1.0 ± 1.5 (6) 0.7 ± 0.7 (8) 1.4 ± 2.1 (7) 6.4 ± 2.9 (7)
Open in a new tab

Heathland patches are arranged by increasing heathland patch size. N values are indicated between brackets.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Hoverfly ecological traits

We used the Syrph the Net database (a list of 926 European species including ecological traits, geographical repartition, and preferred environment, Speight et al., 2015) to establish a list of expected hoverfly species. We applied three selection filters: (i) Belgian species (321 species); (ii) species visiting shrub flowers for pollen or nectar (148 species); and (iii) species living in recorded macrohabitats (94 species). Macrohabitat categories were equivalent to categories used in the CORINE habitat classification system, and supplementary macrohabitats were small habitat features that occurred in association with macrohabitats (e.g. a temporary pool in open ground, Appendix S1). We divided the total species observed and precisely identified into three categories: predicted and observed, predicted but missing and non‐predicted but observed.

Each ecological group (clustering species with the same ecological traits) has a species richness and abundance determined by different factors (Schweiger et al., 2007; Keil et al., 2008). This suggests that an ecological group approach provides a better base for conservation research than a global community approach. Hoverfly species were ranked according to six ecological traits: (i) larval microhabitat (aquatic or terrestrial); (ii) larval diet (aphidophagous, bumblebee larval predator, mycophagous, phytophagous or saprophagous); (iii) migratory status (not migratory, occasionally migratory or strongly migratory); (iv) preferred environment (coniferous and deciduous forests, grasslands and heathlands, urbanised areas, wet forests such as Betula swamps or fen carr or wetlands); (v) body size of the adults; (vi) length of proboscis (Gilbert, 1985; Ssymank, 1991; Speight et al., 2015). Continuous variables were categorised: species with mean proboscis length <3 mm were considered as small proboscis species, from 3 to 5 mm as medium proboscis species and more than 5 mm as long proboscis species. In the same way, three body size categories were created: small species when mean body size was <6 mm, medium species between 7 and 10 mm and large species when mean body size was more than 10 mm.

Spatial characteristics

Immediately before performing hoverfly observations, the floral density of the target plant species was estimated on the studied plots. This was done by counting the number of open flowers on four randomly selected quadrats of 1 m2. For C. vulgaris, due to its high flower density, flower numbers were assessed on five quadrats of 0.25 m2.

Plot, heathland patch and landscape spatial characteristics were recorded (Table 1) with GPS (GPSMAPS 62; Garmin Olathe, KS, USA) and georeferenced images (cartopro1.wallonie.be) and were mapped with QGIS (version 2.2.0). The following characteristics related to resource diversity for the larval stage were calculated: (i) at the heathland patch scale: the number of macrohabitats and supplementary macrohabitats in the studied heathland patches and 50 m around it was recorded according to Speight et al., 2015 and (ii) at the landscape scale: minimal distances between studied plots of 10 m2 of heathland and other types of land cover.

Moreover, parameters related to floral availability for adult hoverflies at different scales were measured: (i) at the plot scale: floral density; (ii) at the heathland patch scale: heathland patch size defined by the surface dominated by ericaceous species; (iii) at the landscape scale: we described the proportion of each type of land cover within 500 m from the centre of the heathland patches. Proportions were adjusted each year according to restoration projects (LIFE10 NAT/BE/000706) and to spruce cuts.

Statistical analyses

Statistical analyses were performed with R (R Development Core Team, 2013). Unless indicated otherwise, data are presented as a mean ± standard deviation.

We tested for correlations between the different spatial characteristics (Table 1) with Spearman's rank tests. The proportion of spruce plantations was negatively correlated with the proportion of wetlands around the studied heathland patches (P = 0.05, r s = −0.85) and the number of macrohabitat (P = 0.04, r s = −0.85). We therefore chose to analyse three parameters that directly and indirectly affect larvae: number of macrohabitats (heathland patch scale), distance to deciduous forests and distance to spruce plantations (landscape scale), as well as three parameters related to floral availability: floral density (plot scale), heathland patch size (heathland patch scale) and dominant type of land cover (landscape scale).

We used multiple correspondence analysis (MCA command, R‐package FactoMineR) to describe the relationships between total hoverfly species and the six selected ecological traits. We used hierarchical clustering on principal components (HCPC command, R‐package FactoMineR) to identify hoverfly main ecological groups. Canonical correspondence analysis (CCA command, R‐package vegan) was performed to link hoverfly abundance in each ecological group to the environmental spatial characteristics. All species identified in each morphotype were included in MCA, HCPC and CCA.

We used generalised linear mixed models (GLMM, glmer command, R‐package lme4, Bates et al., 2014) and information‐theoretic approaches to assess to what extent the explanatory variables were related to the abundance and species richness (to morphotype) of hoverfly communities. GLMMs are an excellent tool for analysing non‐normal data that involve random effects. GLMMs make it possible to analyse hierarchically structured and unbalanced data sets such as ours and effectively eliminate the statistical problem of pseudoreplication due to statistically non‐independent data points (i.e. repeated observations, Bolker et al., 2009; Zuur et al., 2009). We pooled our observations by day (10 × 20 min sessions) resulting in 137 data points. The global model consisted of the six parameters as fixed effects and heathland patches, target plant species and year as random effects. These three factors were included as random effects because their levels were sampled from a larger sample (patches and years) and because we are interested in variation among levels rather than the specific effects of each level (target plant species, Bolker et al., 2009). We used a Poisson distribution, and ranked all possible models according to their Akaike's information criterion corrected for small sample size (AICc) and selected the model with the smallest AICc (dredge command, R‐package MuMIn). When necessary, factors were rescaled with the standardised or rescale command of the R‐package arm. The significance of each of the fixed effects specified in the model was tested with Type III Tests of Fixed Effects.

Results

Abundance and species richness

During the 395 h of observation, we observed a total of 6887 visitors, including 2476 individuals belonging to 49 hoverfly species. Other visitors belonged to the Apidae (Apis 3% and Bombus 45%, see Moquet et al., 2017a,b), other Diptera families (Bibionidae, Calliphoridae, Conopidae, Empididae, Muscidae, Sarcophagidae, Sepsidae, Tachinidae, 7%), solitary bees (Andrenidae, Halictidae, 4%) and butterflies (Hesperiidae, Noctuidae, Nymphalidae, 2%). Hoverflies represented the most abundant visitors on S. x multinervis (67%) and C. vulgaris (43%) and the second most abundant visitors on our other target plant species (between 12% and 20%). Visitor species richness varied among target plant species from 11 species on V. myrtillus to 32 species on C. vulgaris (Appendix S2), and for all studied plant species, hoverflies represented the most diverse family.

Most hoverfly species observed in the heathland patches (56%) had forests (wet, coniferous or deciduous forests) listed as their preferred habitat. Other species preferred wetlands and open habitats like grasslands and heathlands (Table 3). Based on the Syrph the Net database, 88 hoverfly species were expected. Missing species were mainly non‐migrant forest and wetland species. More species were observed on the target plant species flowering late in the season (C. vulgaris) than in early spring plant species (S. x multinervis).

Table 3.

Numbers of species predicted according to the database Syrph the Net, observed in heathlands on studied target plant species, predicted and observed, predicted but missing, and observed but non‐predicted according to the preferred habitats

Preferred habitat Predicted Observed Predicted and observed Missing Not predicted
Urbanised areas 3 5 3 0 2
Coniferous and deciduous forests 42 18 12 30 7
Wet forests 14 10 6 8 4
Wetlands 15 11 7 8 4
Grasslands and heathlands 15 5 5 10 0
Total 88 49 33 55 17
Open in a new tab

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

The five target plant species were visited by hoverflies with different ecological traits (Fig. 1). The main visitors to S. x multinervis belonged to large and saprophagous species with a long proboscis (Eristalis pertinax). The number of wetland and saprophagous individuals decreased from approximately 80% on V. myrtillus and V. vitis‐idaea in the spring to approximately 40% on E. tetralix and C. vulgaris in the summer. Furthermore, there were more individuals with a long proboscis on V. myrtillus (60%) than on other ericaceous species (between 25% and 30%). The number of migrant hoverflies varied between 2% (S. x multinervis) and 80% (V. vitis‐idaea).

Figure 1.

Figure 1

Open in a new tab

Ecological traits of hoverfly individuals visiting the five target plant species, ordered according to their flowering period: larval diet (a), larval habitat (b), adult preferred habitat (c), adult size (d), proboscis length (e) and migratory status (f). Small species <6 mm; 7 mm < medium species < 10 mm; large species > 10 mm. Short proboscis, length < 3 mm; medium proboscis, between 3 and 5 mm; long proboscis, >5 mm.

The abundance of hoverfly individuals per day varied from 0 to 125 depending on the target plant species. Per day of observations (200 min) on the 10 m2 plot of the target plant species, the number of hoverflies was higher on C. vulgaris (60.8 ± 34.2 individuals) than on S. x multinervis (32.2 ± 32.3), V. myrtillus (2.2 ± 3.1), V. vitis‐idaea (4.9 ± 6.2), and E. tetralix (8.8 ± 9.0).

Hoverfly groups

Hoverfly species were separated into two groups according to larval diet (P < 0.001), larval habitat (P < 0.001), proboscis length (P < 0.001) and preferred habitats (P = 0.007; Fig. 2; Appendix S3). The first group (group 1) was mainly composed of aphidophagous species (93% of individuals, P < 0.001), with terrestrial larvae (100%, P < 0.001), adults having a short or medium proboscis (100%, P < 0.001), and a preference for forests (coniferous and deciduous, 50%, P = 0.030) or grasslands and heathlands (17%). This group was mainly represented by the Syrphini tribe (Chrysotoxum, Didea, Epistrophe, Episyrphus, Eupeodes, Meliscaeva, Leucozona, Parasyrphus, Scaeva, Sphaerophoria, Syrphus). The second group (group 2) was composed of species with saprophagous (95%, P < 0.001), aquatic larvae (90%, P < 0.001), having a long proboscis (71%, P < 0.001) and a preference for wetlands (29%) or wet forests (38%, P = 0.030). This group was mainly represented by the Eristalini tribe (Eristalis, Helophilus, Myathropa).

Figure 2.

Figure 2

Open in a new tab

Multiple correspondence analysis (MCA) plot showing hoverfly species according to the six life‐history traits. (a) 95% confidence ellipses around groups of hoverfly species according to hierarchical clustering on principal components. (b–e) 95% confidence ellipses around significant ecological traits: (b) larval diet, (c) larval habitat, (d) preferred habitat, (e) proboscis length. Adult body size and migratory status were not significant factors in MCA (not shown). Aphido = Aphidophagous, Sapro = Saprophagous, Myco = Mycophagous, Phyto = Phytophagous in (b). Open habitats included heathlands and grasslands in (d). Short proboscis, length < 3 mm; medium proboscis, between 3 and 5 mm; long proboscis, >5 mm.

Parameters influencing hoverfly abundance

Model averaging across our set of candidates revealed that distances from the observation plots to spruce plantations and deciduous forests were the two most important factors associated with hoverfly abundance. Abundance significantly decreased with the distance to forests (GLMM, z‐value = 3.79, P < 0.001).

The abundance of group 1 hoverflies (aphidophagous and terrestrial) was significantly positively influenced by floral density (plot scale, GLMM, z‐value = −4.46, P = <0.005; Table 4; Fig. 4a). Moreover, abundance of hoverflies in this group significantly decreased with distance to spruce plantations (landscape scale, GLMM, z‐value = −2.59, P = 0.010; Fig. 4c) and to deciduous forests (landscape scale, GLMM, z‐value = 2.06, P = 0.040). The CCA plot (Fig. 3) shows the relationships between the abundance of group 1 hoverflies and the spatial characteristics of the landscape. The eigenvalues of axis 1 (CCA1) and axis 2 (CCA2) were 0.79 and 0.32 respectively. We observed a distinct distribution of species according to spatial characteristics. Hoverflies were ordered in relation to the proportions of forest (spruce plantations and deciduous forest) and distances in the landscape. For example, Meliscaeva cinctella and Platycheirus albimanus were found in landscapes dominated by forests, at close proximity to the forest edges, but Parasyrphus lineolus, Epistrophe grossulariae and Episyrphus balteatus were mainly observed in landscapes dominated by wetlands, distant from forest edges. No tested factor had a significant effect on the abundance of group 2 hoverflies (saprophagous and aquatic hoverflies, Fig. 4b, see Appendix S4 for CCA illustrating the relationships between the abundance of group 2 hoverflies and spatial characteristics).

Table 4.

Model selection and results for candidate models explaining hoverfly abundance and species richness for the two functional groups

Response variables Models Intercept Dominant cover land Floral density Distance to deciduous forest Distance to spruce plantation Number of macrohabitat Heathland area AICc d.f. ΔAICc ωm
Abundance group 1 Model 1 0.09242 −0.3987 *** 0.7202 * −0.8454 ** 704.7 7 0 0.2
Model 2 0.1841 −0.4048 −0.6307 706.5 6 1.8 0.081
Model 3 0.09065 −0.3966 0.6396 −0.7663 −0.2325 706.7 8 2.02 0.073
Model 4 0.09291 −0.2236 −0.3988 0.746 −0.89 706.7 8 2.04 0.072
Model 5 0.07525 −0.3991 0.7238 −0.8547 −0.1351 706.9 8 2.18 0.067
Abundance group 2 Model 1 2.048 −0.131 −0.2169 0.1402 1444 6 0 0.087
Model 2 2.044 −0.1377 −0.2598 0.2005 −0.07182 1445 7 1.14 0.049
Model 3 2.054 −0.09135 −0.1283 −0.1477 −0.08886 1445 7 1.39 0.043
Model 4 2.048 −0.04873 −0.128 −0.2157 0.1236 1445 7 1.53 0.04
Model 5 2.057 −0.1216 −0.1344 1445 5 1.61 0.039
Species richness group 1 Model 1 −0.1318 −0.3491 * −0.3466. 398.9 6 0 0.061
Model 2 −0.1339 0.2374 −0.402 −0.3353 399 7 0.12 0.057
Model 3 −0.1223 −0.2976 −0.3462 399.2 6 0.34 0.051
Model 4 −0.07546 399.6 4 0.67 0.044
Model 5 −0.1092 −0.2885 399.8 5 0.88 0.039
Species richness group 2 Model 1 0.4247 0.3804 ** −0.3007 −0.4239 * 423.6 7 0 0.111
Model 2 0.4398 0.3955 −0.3623 424.2 6 0.62 0.081
Model 3 0.4297 0.3988 −0.2144 −0.3898 424.9 7 1.29 0.058
Model 4 0.4767 0.3717 424.9 5 1.3 0.058
Model 5 0.4223 0.3852 −0.2502 −0.1144 −0.4284 425.5 8 1.87 0.044
Open in a new tab

Group 1 mainly consisted of aphidophagous, terrestrial hoverflies and group 2 mainly consisted of saprophagous, aquatic hoverflies. Candidate models are ranked in order of increasing corrected Akaike's information criterion (Δ AICc). Explanatory variables were standardised. In bold, the selected model with the degree of factor significance (***P < 0.001, **P < 0.01 and *P < 0.05).

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Figure 4.

Figure 4

Open in a new tab

Abundance of group 1 (a, c, e) and group 2 hoverfly individuals (b, d, f) per day of observation (10 sessions × 20 min on 10 m2) according to floral density (a, b), distances to spruce plantations (c, d) and distances to deciduous forests (e, f). Observations were carried out on S. x multinervis (▵), V. vitis‐idaea (⨉), V. myrtillus (+), E. tetralix (○) and C. vulgaris (□). Group 1 mainly consisted of aphidophagous, terrestrial larvae species and group 2 contained principally saprophagous, aquatic larvae species (see Fig. 2). Lines are linear regression lines for each target plant species (only shown when the factor was significant).

Figure 3.

Figure 3

Open in a new tab

Canonical correspondence analysis (CCA) ordination plots of hoverfly abundance (+), site (•) and environmental data (→) for group 1 hoverfly species. The length and angle of the arrows shows the contribution of a particular environmental variable to the CCA axes. The eigenvalues of axis 1 (CCA1) and axis 2 (CCA2) are 0.79 and 0.32 respectively. P_land cover is the percentage of each type of land cover (spruce plantations, deciduous forests, meadows, and wetlands). D_land cover is the minimal distance between the studied plot of 10 m2 of heathland and other types of land cover. Number of macrohabitats is the habitat diversity according to Speight and Castella (2015). Closure is the proportion of the heathland patch edge that consisted of woody elements. Patch size is heathland patch area. Floral density is mean number of open target flowers per m2 (see Appendix S2 for hoverfly species name). [Colour figure can be viewed at http://wileyonlinelibrary.com]

Parameters influencing hoverfly species richness

The best model to explain differences in hoverfly species richness included the distances between observation plots and spruce plantations and heathland land cover. Species richness significantly decreased with distance to spruce plantations (landscape scale, GLMM, z‐value = −2.11, P = 0.035) and with heathland patch area (heathland patch scale, GLMM, z‐value = −1,88 P = 0.051). Similarly, the species richness of group 1 decreased with the distance to spruce plantations (landscape scale, GLMM, z‐value = −1.76, P = 0.011; Fig. 5c) and heathland area (heathland patch scale, GLMM, z‐value = −1.85, P = 0.015; Fig. 5e.). In contrast, for group 2, species richness was significantly negatively influenced by heathland patch area (heathland patch scale, GLMM, z‐value = −2.18, P = 0.029; Fig. 5b) and positively influenced by floral density (plot scale, GLMM, z‐value = 3.07, P < 0.002; Table 4; Fig. 5f).

Figure 5.

Figure 5

Open in a new tab

Species richness of group 1 (a, c, e) and group 2 hoverflies (b, d, f) per day of observation (10 sessions × 20 min on 10 m2) according to floral density (a, b), distances to spruce plantations (c, d) and heathland patch area (e, f). Observations were carried out on S. x multinervis (▵), V. vitis‐idaea (⨉), V. myrtillus (+), E. tetralix (○) and C. vulgaris (□). Group 1 mainly consisted of aphidophagous, terrestrial hoverfly larvae species. Group 2 mainly consisted of saprophagous and aquatic hoverfly larvae species (see Fig. 2). Lines are linear regression lines for each target plant species (only shown when the factor was significant).

Discussion

Hoverfly species associated with forests were major visitors to ericaceous plants

Hoverflies were abundant visitors to flowers in the heathlands and represented more than one third of all recorded visitors. For the 49 hoverfly species recorded in this study, species richness was comparable to that observed in other studies in heathlands (Dupont & Nielsen, 2006; Descamps et al., 2015). For C. vulgaris alone, Descamps et al. (2015) observed 25 species at Mont Lozère (Languedoc‐Roussillon, France) compared with our 39 species, with 19 species in common. The presence of rare or threatened species such as Blera fallax and Epistrophe flava on Vaccinium species further emphasises the value of heathlands for hoverfly conservation.

Most of the observed hoverfly species were associated with coniferous forests, wet forests and wetlands; only five of the observed species were associated with heathlands. These results are consistent with previous studies indicating that most of the hoverfly species in northwest Europe have larval stages in flower‐poor forests and the adults disperse to open habitats, such as heathlands, to acquire floral resources, that is, pollen and nectar (Humphrey et al., 1999; Branquart & Hemptinne, 2000; Speight et al., 2015). Within forest hoverfly species, the predicted but missing species were non‐migratory, suggesting that they have a limited capacity for large‐scale dispersal and probably a greater susceptibility to ongoing heathland degradation.

Our results showed that the relevant hoverfly traits for the separation of ecological groups were larval diet, larval habitat and proboscis size. Other studies using similar approaches have also highlighted the importance of these two larval traits (Dziock, 2006; Schweiger et al., 2007; Keil et al., 2008), which Thompson and Rotheray (1998) considered to be the most significant characteristics of the Syrphidae. Moreover, because we focused on flower visitors mainly on ericaceous species, it is not surprising that proboscis size explains a large amount of the variation in our study. In consequence, hoverfly species on heathlands can be separated into two groups: one composed of species with aphidophagous and terrestrial larvae, adults with a short or medium proboscis and having a preference for forests (coniferous and deciduous) or grasslands and heathlands; and the other composed of species with saprophagous and aquatic larvae, a long proboscis and a preference for wetlands and wet forests.

Species richness and abundance driven by the availability of resources for both larvae and adults

Factors associated with the availability of resources for larvae (distance to larval habitat, macrohabitat diversity) as well as floral resources for adults (floral density) influenced hoverfly abundance and species richness. Studies in agricultural landscapes have also suggested that both adult and larval habitat requirements have to be considered when analysing hoverfly communities (Kleijn & Van Langevelde, 2006; Meyer et al., 2009). Hoverfly abundance and species richness depend on factors related to floral resources for adults, such as the quantity and quality of pollen and nectar, as well as factors associated with larval development, such as microhabitat diversity (Meyer et al., 2009). Different scales were relevant depending on whether factors were related to larval or to adult resources. For example, larval habitat availability influenced hoverfly community composition at the landscape scale (distance to larval habitat), whereas floral resource availability for adults influenced community composition at the plot scale (floral density).

Our study showed that the distance to larval habitat, especially forests, was the main factor related to larval habitats that influenced hoverfly visitor composition. These results are consistent with other studies (Dupont & Nielsen, 2006; Meyer et al., 2009; Bortolotto et al., 2016) in which adult hoverflies were highly influenced by the types of larval macrohabitats, especially forest fragments, in the surrounding landscape. For an insect species to persist in a heathland fragment, larval and adult resources have to be available within the flight distance of the specific insect species. When most flower visitors could fly only a short distance, communities were strongly influenced by the surrounding vegetation (Dupont & Nielsen, 2006).

For factors associated with adult resources, floral density (plot scale) positively influenced the abundance of hoverfly individuals at the plot level, whereas heathland patch area (patch scale) and landscape composition (landscape scale) had no effect. A previous study found the opposite results for bumblebees in the same heathland patches and on the same target plant species, showing that bumblebee abundance was influenced by floral availability at the landscape scale (Moquet et al., 2017a). Differences in the spatial scale affecting Diptera and Hymenoptera imply that their perception of the landscape, foraging ranges and resource requirements differ (Kleijn & Van Langevelde, 2006; Jauker et al., 2009; Öckinger et al., 2012). Hoverflies disperse over very short distances (except during migration events), compared with bumblebees (200 m vs. up to 3000 m, Lövei et al., 1998; Wratten et al., 2003; Westphal et al., 2006). Therefore, only floral resources at a small scale influenced hoverfly abundance. Hoverfly abundance increased with floral density, which is consistent with other studies of hoverflies in agricultural landscapes (Meyer et al., 2009) and even other pollinators such as bumblebees and solitary bees (Sih & Baltus, 1987; Ohashi & Yahara, 2002; Somme et al., 2014). All these results are consistent with optimal foraging theory, which predicts higher abundance of pollinators in dense patches of flowers (Pyke et al., 1977). Indeed, high floral density provides a greater quantity of resources at lower foraging costs for all anthophilous insects (Hegland & Boeke, 2006).

Contrasting responses according to ecological group

Factors that affected the abundance and species richness of hoverfly visitors of our target plant species differed according to ecological group. Species richness and abundance of saprophagous hoverflies were not affected, in the context of our study, by parameters related to larval wetland macrohabitats, but the abundance and species richness of aphidophagous hoverflies decreased with distance to forests. One other study has shown that landscape context affects differently according to hoverfly ecological traits (Schweiger et al., 2007); in this study, the main traits influencing response to landscape structure were degree of specialisation in habitat or in feeding. Aphidophagous species were probably more affected by landscape structure than saprophagous wetland species, because they need different resource patches for larval and adult stages. Such species require different landscape elements at different life stages (Dunning et al., 1992) and thus are particularly dependent on landscape heterogeneity (Fahrig & Nuttle, 2005). This sensitivity will depend on the interaction between heterogeneity and dispersal capability. Because we observed no aphidophagous species farther than 500 m from spruce plantations, we suppose that distance limited adult dispersal to heathlands. Some studies have found a low dispersal capability for aphidophagous forest species (Wratten et al., 2003; Herrault et al., 2016). Our results showed that, in this range, adult dispersal varied according to species, as some forest species were mainly found in forest‐dominated landscapes at close proximity to forest edges (M. cinctella and P. albimanus), while others were abundant in wetland‐dominated landscapes at larger distances from forest edges (Parasyrphus lineolus, Epistrophe grossulariae and Episyrphus balteatus).

Conclusion

We showed that Belgian heathlands are habitats where diverse hoverfly species, especially species from forests, can feed on flower resources. These resources are rare under coniferous canopies, but abundant in heathlands. Forest hoverfly species were mainly aphidophagous species, more affected by landscape structure than saprophagous wetland species, probably because they need two different resource patches for larval and adult stages, yet have limited dispersal capability. Landscape complementation (patches for larval development and patches for adult requirements) is therefore needed to preserve the species pool of hoverflies and the ecosystem services they provide.

Supporting information

Appendix S1. Macrohabitats observed on each studied sites in the Upper Ardenne (Belgium).

Appendix S2. Abundance of hoverfly species observed on the five target plant species during 3 years in 10 heathland patches located in the Upper Ardenne, Belgium.

Appendix S3. Life‐history traits of hoverfly species observed on the five target plant species in the Upper Ardenne (Belgium), according to Speight et al. (2015), Gilbert (1985) and Ssymank (1991).

Appendix S4. Canonical correspondence analysis (CCA) ordination plots of hoverfly abundance (cross), site (black points) and environmental data (arrows) for group 2 hoverfly species.

Click here for additional data file. (255KB, docx)

Acknowledgements

We would like to thank the Département de la Nature et des Forêts for the permission to study in nature reserves and for the derogation concerning the sampling of plant and insect individuals. We thank Pol Limbourg and Wouter Dekoninck (Royal Belgian Institute of Natural Sciences) for access to the insect collections and Marc Dufrêne and Alain Guillet for statistical recommendations. We further thank Kristin Sherrard and Jennifer Mach for language improvement and the two anonymous reviewers for their valuable comments on the first version of the manuscript. The study constituted master thesis of R.B and E.L., the PhD thesis of L.M. and was conducted in accordance with current Belgian laws. Funding for M.L. was provided by an FSR grant (Fonds spéciaux de recherche, Université catholique de Louvain) and FNRS (Fonds National de la Recherche Scientifique, Web Impact project, FRFC 2.4613.12).

References

  1. Ball, S. & Morris, R. (2015) Britain's Hoverflies: A Field Guide, 2nd edn Princeton University Press, Hampshire, UK. [Google Scholar]
  2. Bannister, P. (1966) Biological flora of the British Isles: Erica tetralix L. Journal of Ecology, 54, 795–813. [Google Scholar]
  3. Bates, D. , Mächler, M. , Bolker, B. & Walker, S. (2014) Fitting linear mixed‐effects models using lme4. Journal of Statistical Software, 67, 1–48. [Google Scholar]
  4. Biesmeijer, J.C. , Roberts, S.P.M. , Reemer, M. , Ohlemüller, R. , Edwards, M. , Peeters, T. , Schaffers, A.P. , Potts, S.G. , Kleukers, R. & Thomas, C.D. (2006) Parallel declines in pollinators and insect‐pollinated plants in Britain and the Netherlands. Science, 313, 351–354. [DOI] [PubMed] [Google Scholar]
  5. Bolker, B.M. , Brooks, M.E. , Clark, C.J. , Geange, S.W. , Poulsen, J.R. , Stevens, M.H.H. & White, J.‐S.S. (2009) Generalized linear mixed models: a practical guide for ecology and evolution. Trends in Ecology & Evolution, 24, 127–135. [DOI] [PubMed] [Google Scholar]
  6. Bortolotto, O.C. , Júnior, M. , de Oliveira, A. , Hoshino, A.T. , Campos, T.A. , Bortolotto, O.C. , Júnior, M. , de Oliveira, A. , Hoshino, A.T. & Campos, T.A. (2016) Distance from the edge of forest fragments influence the abundance of aphidophagous hoverflies (Diptera: Syrphidae) in wheat fields. Acta Scientiarum. Agronomy, 38, 157–164. [Google Scholar]
  7. Branquart, E. & Hemptinne, J.‐L. (2000) Selectivity in the exploitation of floral resources by hoverflies (Diptera: Syrphinae). Ecography, 23, 732–742. [Google Scholar]
  8. Buchmann, S.L. (1983) Buzz pollination in angiosperms Handbook of Experimental Pollination Biology (ed. by Jones C.E. and Little R.J.), pp. 73–113. Van Nostrand Reinhold, New York City, New York. [Google Scholar]
  9. Burgio, G. & Sommaggio, D. (2007) Syrphids as landscape bioindicators in Italian agroecosystems. Agriculture, Ecosystems & Environment, 120, 416–422. [Google Scholar]
  10. Cane, J. (2001) Habitat fragmentation and native bees: a premature verdict? Conservation Ecology, 5, 1–9. [Google Scholar]
  11. Carvalheiro, L.G. , Kunin, W.E. , Keil, P. , Aguirre‐Gutiérrez, J. , Ellis, W.N. , Fox, R. , Groom, Q. , Hennekens, S. , Van Landuyt, W. , Maes, D. , Van de Meutter, F. , Michez, D. , Rasmont, P. , Ode, B. , Potts, S.G. , Reemer, M. , Roberts, S.P.M. , Schaminée, J. , WallisDeVries, M.F. & Biesmeijer, J.C. (2013) Species richness declines and biotic homogenisation have slowed down for NW‐European pollinators and plants. Ecology Letters, 16, 870–878. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Clicheroux, E. (1957) L'évolution des terrains incultes en Belgique. Bulletin de L'Institut de Recherches Économiques et Sociales, 23, 497–524. [Google Scholar]
  13. Descamps, C. , Moquet, L. , Migon, M. & Jacquemart, A.‐L. (2015) Diversity of the insect visitors on Calluna vulgaris (Ericaceae) in Southern France heathlands. Journal of Insect Science, 15, 130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dunning, J.B. , Danielson, B.J. & Pulliam, H.R. (1992) Ecological processes that affect populations in complex landscapes. Oikos, 65, 169–175. [Google Scholar]
  15. Dupont, Y.L. & Nielsen, B.O. (2006) Species composition, feeding specificity and larval trophic level of flower‐visiting insects in fragmented versus continuous heathlands in Denmark. Biological Conservation, 131, 475–485. [Google Scholar]
  16. Dziock, F. (2006) Life‐history data in bioindication procedures, using the example of hoverflies (Diptera, Syrphidae) in the elbe floodplain. International Review of Hydrobiology, 91, 341–363. [Google Scholar]
  17. Fahrig, L. & Nuttle, W.K. (2005) Population ecology in spatially heterogeneous environments Ecosystem Function in Heterogeneous Landscapes (ed. by Lovett G.M., Jones C.E., Turner M.G. and Weathers K.C.), pp. 95–118. Springer, New York City, New York. [Google Scholar]
  18. Gallai, N. , Salles, J.‐M. , Settele, J. & Vaissiére, B.E. (2009) Economic valuation of the vulnerability of world agriculture confronted with pollinator decline. Ecological Economics, 68, 810–821. [Google Scholar]
  19. Gilbert, F.S. (1981) Foraging ecology of hoverflies: morphology of the mouthparts in relation to feeding on nectar and pollen in some common urban species. Ecological Entomology, 6, 245–262. [Google Scholar]
  20. Gilbert, F.S. (1985) Morphometric patterns in hoverflies (Diptera, Syrphidae). Proceedings of the Royal Society of London. Series B, Biological Sciences, 224, 79–90. [Google Scholar]
  21. Gilbert, F.S. (1986) Ecomorphology of Syrphidae Ecology of Aphidophaga (ed. by Hodek I.), pp. 23–28. Dr. W. Junk, Dordrecht, The Netherlands. [Google Scholar]
  22. Gimingham, C.H. (1960) Biological flora of the British Isles: Calluna Salisb. Journal of Ecology, 48, 455–483. [Google Scholar]
  23. Gimingham, C.H. (1972) Ecology of Heathlands. Chapman & Hall, London, UK. [Google Scholar]
  24. Gittings, T. , O'Halloran, J. , Kelly, T. & Giller, P.S. (2006) The contribution of open spaces to the maintenance of hoverfly (Diptera, Syrphidae) biodiversity in Irish plantation forests. Forest Ecology and Management, 237, 290–300. [Google Scholar]
  25. Goulson, D. , Hanley, M.E. , Darvill, B. , Ellis, J.S. & Knight, M.E. (2005) Causes of rarity in bumblebees. Biological Conservation, 122, 1–8. [Google Scholar]
  26. Grimmeau, J.‐P. , Decroly, J. & Wertz, I. (2013) La démographie des communes belges de 1980 à 2010. Courrier Hebdomadaire du CRISP, 37, 1–90. [Google Scholar]
  27. Hadley, A.S. & Betts, M.G. (2011) The effects of landscape fragmentation on pollination dynamics: absence of evidence not evidence of absence. Biological Reviews, 87, 526–544. [DOI] [PubMed] [Google Scholar]
  28. Haenke, S. , Kovács‐Hostyánszki, A. , Fründ, J. , Batáry, P. , Jauker, B. , Tscharntke, T. & Holzschuh, A. (2014) Landscape configuration of crops and hedgerows drives local syrphid fly abundance. Journal of Applied Ecology, 51, 505–513. [Google Scholar]
  29. Haslett, J.R. (1989) Adult feeding by holometabolous insects: pollen and nectar as complementary nutrient sources for Rhingia campestris (Diptera: Syrphidae). Oecologia, 81, 361–363. [DOI] [PubMed] [Google Scholar]
  30. Hegland, S.J. & Boeke, L. (2006) Relationships between the density and diversity of floral resources and flower visitor activity in a temperate grassland community. Ecological Entomology, 31, 532–538. [Google Scholar]
  31. Herrault, P.‐A. , Larrieu, L. , Cordier, S. , Gimmi, U. , Lachat, T. , Ouin, A. , Sarthou, J.‐P. & Sheeren, D. (2016) Combined effects of area, connectivity, history and structural heterogeneity of woodlands on the species richness of hoverflies (Diptera: Syrphidae). Landscape Ecology, 31, 877–893. [Google Scholar]
  32. Humphrey, J.W. , Hawes, C. , Peace, A.J. , Ferris‐Kaan, R. & Jukes, M.R. (1999) Relationships between insect diversity and habitat characteristics in plantation forests. Forest Ecology and Management, 113, 11–21. [Google Scholar]
  33. Inouye, D.W. , Larson, B.M.H. , Ssymank, A. & Kevan, P.G. (2015) Flies and flowers III: ecology of foraging and pollination. Journal of Pollination Ecology, 16, 115–133. [Google Scholar]
  34. Jacquemart, A.‐L. (1993) Floral visitors of Vaccinium species in the high Ardenne, Belgium. Flora, 188, 263–273. [Google Scholar]
  35. Jauker, F. , Diekötter, T. , Schwarzbach, F. & Wolters, V. (2009) Pollinator dispersal in an agricultural matrix: opposing responses of wild bees and hoverflies to landscape structure and distance from main habitat. Landscape Ecology, 24, 547–555. [Google Scholar]
  36. Keil, P. , Biesmeijer, J.C. , Barendregt, A. , Reemer, M. & Kunin, W.E. (2011) Biodiversity change is scale‐dependent: an example from Dutch and UK hoverflies (Diptera, Syrphidae). Ecography, 34, 392–401. [Google Scholar]
  37. Keil, P. , Dziock, F. & Storch, D. (2008) Geographical patterns of hoverfly (Diptera, Syrphidae) functional groups in Europe: inconsistency in environmental correlates and latitudinal trends. Ecological Entomology, 33, 748–757. [Google Scholar]
  38. Kevan, P.G. & Baker, H.G. (1983) Insects as flower visitors and pollinators. Annual Review of Entomology, 28, 407–453. [Google Scholar]
  39. Kleijn, D. & Van Langevelde, F. (2006) Interacting effects of landscape context and habitat quality on flower visiting insects in agricultural landscapes. Basic and Applied Ecology, 7, 201–214. [Google Scholar]
  40. Larson, B.M.H. , Kevan, P.G. & Inouye, D.W. (2001) Flies and flowers: taxonomic diversity of anthophiles and pollinators. The Canadian Entomologist, 133, 439–465. [Google Scholar]
  41. Lövei, G.L. , Macleod, A. & Hickman, J.M. (1998) Dispersal and effects of barriers on the movement of the New Zealand hover fly Melanostoma fasciatum (Dipt., Syrphidae) on cultivated land. Journal of Applied Entomology, 122, 115–120. [Google Scholar]
  42. Maes, D. & Van Dyck, H. (2001) Butterfly diversity loss in Flanders (north Belgium): Europe's worst case scenario? Biological Conservation, 99, 263–276. [Google Scholar]
  43. Mahy, G. , De Sloover, J.R. & Jacquemart, A.‐L. (1998) The generalist pollination system and reproductive success of Calluna vulgaris in the Upper Ardenne. Canadian Journal of Botany, 76, 1843–1851. [Google Scholar]
  44. Mayer, C. , Michez, D. , Chyzy, A. , Brédat, E. & Jacquemart, A.‐L. (2012) The abundance and pollen foraging behaviour of bumble bees in relation to population size of whortleberry (Vaccinium uliginosum). PLoS ONE, 7, e50353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Meyer, B. , Jauker, F. & Steffan‐Dewenter, I. (2009) Contrasting resource‐dependent responses of hoverfly richness and density to landscape structure. Basic and Applied Ecology, 10, 178–186. [Google Scholar]
  46. Moquet, L. , Bacchetta, R. , Laurent, E. & Jacquemart, A.‐L. (2017a) Spatial and temporal variations in floral resource availability affect bumblebee communities in heathlands. Biodiversity and Conservation, 26, 687–702. [Google Scholar]
  47. Moquet, L. , Mayer, C. , Michez, D. , Wathelet, B. & Jacquemart, A.‐L. (2015) Early spring floral foraging resources for pollinators in wet heathlands in Belgium. Journal of Insect Conservation, 19, 837–848. [Google Scholar]
  48. Moquet, L. , Vanderplanck, M. , Moerman, R. , Quinet, M. , Roger, N. , Michez, D. & Jacquemart, A.‐L. (2017b) Bumblebees depend on ericaceous species to survive in temperate heathlands. Insect Conservation and Diversity, 10, 78–93. [Google Scholar]
  49. Öckinger, E. , Lindborg, R. , Sjödin, N.E. & Bommarco, R. (2012) Landscape matrix modifies richness of plants and insects in grassland fragments. Ecography, 35, 259–267. [Google Scholar]
  50. Ohashi, K. & Yahara, T. (2002) Visit larger displays but probe proportionally fewer flowers: counterintuitive behaviour of nectar‐collecting bumble bees achieves an ideal free distribution. Functional Ecology, 16, 492–503. [Google Scholar]
  51. Ouin, A. , Sarthou, J.‐P. , Bouyjou, B. , Deconchat, M. , Lacombe, J.‐P. & Monteil, C. (2006) The species‐area relationship in the hoverfly (Diptera, Syrphidae) communities of forest fragments in southern France. Ecography, 29, 183–190. [Google Scholar]
  52. Potts, S.G. , Biesmeijer, J.C. , Kremen, C. , Neumann, P. , Schweiger, O. & Kunin, W.E. (2010) Global pollinator declines: trends, impacts and drivers. Trends in Ecology & Evolution, 25, 345–353. [DOI] [PubMed] [Google Scholar]
  53. Pyke, G.H. , Pulliam, H.R. & Charnov, E.L. (1977) Optimal foraging: a selective review of theory and tests. The Quarterly Review of Biology, 52, 137–154. [Google Scholar]
  54. R Development Core Team (2013) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. [Google Scholar]
  55. Rader, R. , Bartomeus, I. , Garibaldi, L.A. , Garratt, M.P.D. , Howlett, B.G. , Winfree, R. , Cunningham, S.A. , Mayfield, M.M. , Arthur, A.D. , Andersson, G.K.S. , Bommarco, R. , Brittain, C. , Carvalheiro, L.G. , Chacoff, N.P. , Entling, M.H. , Foully, B. , Freitast, B.M. , Gemmill‐Herren, B. , Ghazoul, J. , Griffin, S.R. , Gross, C.L. , Herbertsson, L. , Herzog, F. , Hipólito, J. , Jaggar, S. , Jauker, F. , Klein, A.‐M. , Kleijn, D. , Krishnan, S. , Lemos, C.Q. , Lindström, S.A.M. , Mandelik, Y. , Monteiro, V.M. , Nelson, W. , Nilsson, L. , Pattemore, D.E. , de Pereira, N.O. , Pisanty, G. , Potts, S.G. , Reemer, M. , Rundlöf, M. , Sheffield, C.S. , Scheper, J. , Schüepp, C. , Smith, H.G. , Stanley, D.A. , Stout, J.C. , Szentgyörgyi, H. , Taki, H. , Vergara, C.H. , Viana, B.F. & Woyciechowski, M. (2016) Non‐bee insects are important contributors to global crop pollination. Proceedings of the National Academy of Sciences of the United States of America, 113, 146–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Raymond, L. , Ortiz‐Martínez, S.A. & Lavandero, B. (2015) Temporal variability of aphid biological control in contrasting landscape contexts. Biological Control, 90, 148–156. [Google Scholar]
  57. Ritchie, J.C. (1955) Biological flora of the British Isles: Vaccinium vitis‐idaea L. Journal of Ecology, 43, 701–708. [Google Scholar]
  58. Ritchie, J.C. (1956) Biological flora of the British Isles: Vaccinium myrtillus L. Journal of Ecology, 44, 291–299. [Google Scholar]
  59. Sarthou, J.P. , Ouin, A. , Arrignon, F. , Barreau, G. & Bouyjou, B. (2005) Landscape parameters explain the distribution and abundance of Episyrphus balteatus (Diptera: Syrphidae). European Journal of Entomology (Czech Republic), 102, 539–545. [Google Scholar]
  60. Schönrogge, K. , Gardner, M.G. , Elmes, G.W. , Napper, E.K.V. , Simcox, D.J. , Wardlaw, J.C. , Breen, J. , Barr, B. , Knapp, J.J. , Pickett, J.A. & Thomas, J.A. (2006) Host propagation permits extreme local adaptation in a social parasite of ants. Ecology Letters, 9, 1032–1040. [DOI] [PubMed] [Google Scholar]
  61. Schweiger, O. , Musche, M. , Bailey, D. , Billeter, R. , Diekötter, T. , Hendrickx, F. , Herzog, F. , Liira, J. , Maelfait, J.‐P. , Speelmans, M. & Dziock, F. (2007) Functional richness of local hoverfly communities (Diptera, Syrphidae) in response to land use across temperate Europe. Oikos, 116, 461–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Sih, A. & Baltus, M.‐S. (1987) Patch size, pollinator behavior, and pollinator limitation in catnip. Ecology, 68, 1679–1690. [DOI] [PubMed] [Google Scholar]
  63. Sommaggio, D. (1999) Syrphidae: can they be used as environmental bioindicators? Agriculture, Ecosystems & Environment, 74, 343–356. [Google Scholar]
  64. Somme, L. , Mayer, C. & Jacquemart, A.‐L. (2014) Multilevel spatial structure impacts on the pollination services of Comarum palustre (Rosaceae). PLoS ONE, 9, e99295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Speight, M.C.D. & Castella, E. (2015) StN Database: Content and Glossary of Terms, 2015. Syrph the Net, the Database of European Syrphidae. Syrph the Net publications, Dublin, Ireland. [Google Scholar]
  66. Speight, M.C.D. , Castella, E. & Obrdlik, P. (2000) Syrph the Net: A Database of Biological Information About European Syrphidae (Diptera) and Its Use in Relation to the Conservation of Biodiversity. Syrph the Net publications, Dublin, Ireland. [Google Scholar]
  67. Speight, M.C.D. , Castella, E. & Sarthou, J.P. (2015) StN 2015 Syrphe the Net on CD, Issue10. (ed. by Speight M.C.D., Castella E., Sarthou J.P. and Vanappelghem C.), Dublin, Ireland. [Google Scholar]
  68. Ssymank, A. (1991) Rüssel‐und Körperlänge von Schwebfliegen (Diptera: Syrphidae) unter Berücksichtigung der Verwendung von Alkoholmaterial. Mitteilungen der Schweizerischen Entomologischen Gesellschaft, 64, 67–80. [Google Scholar]
  69. Stokstad, E. (2006) Pollinator diversity declining in Europe. Science, 313, 286. [DOI] [PubMed] [Google Scholar]
  70. Stubbs, A.E. & Falk, S.J. (1983) British Hoverflies. An Illustrated Identification Guide. British Entomological and Natural History Society, London, UK. [Google Scholar]
  71. Sutherland, J.P. , Sullivan, M.S. & Poppy, G.M. (2001) Distribution and abundance of aphidophagous hoverflies (Diptera: Syrphidae) in wildflower patches and field margin habitats. Agricultural and Forest Entomology, 3, 57–64. [Google Scholar]
  72. Thompson, F.C. & Rotheray, G. (1998) Family Syrphidae Contributions to a Manual of Palaearctic Diptera (with Special Reference to Flies of Economic Importance) (ed. by Papp L. and Darvas B.), pp. 81–139. Science Herald, Budapest, Hungary. [Google Scholar]
  73. Van Dyck, H. , Van Strien, A.J. , Maes, D. & Van Swaay, C.A.M. (2009) Declines in common, widespread butterflies in a landscape under intense human use. Conservation Biology, 23, 957–965. [DOI] [PubMed] [Google Scholar]
  74. Van Veen, M.P. (2004) Hoverflies of Northwest Europe. KNNV Publishing, Utrecht, The Netherlands. [Google Scholar]
  75. Vanbergen, A.J. ; The Insect Pollinators Initiative (2013) Threats to an ecosystem service: pressures on pollinators. Frontiers in Ecology and the Environment, 11, 251–259. [Google Scholar]
  76. Verlinden, L. (1994) Faune de Belgique, Syrphides (Syrphidae). Edition de l'Institut Royal des Sciences Naturelles de Belgique, Bruxelles. [Google Scholar]
  77. Webb, N.R. (1998) The traditional management of European heathlands. Journal of Applied Ecology, 35, 987–990. [Google Scholar]
  78. Westphal, C. , Steffan‐Dewenter, I. & Tscharntke, T. (2006) Bumblebees experience landscapes at different spatial scales: possible implications for coexistence. Oecologia, 149, 289–300. [DOI] [PubMed] [Google Scholar]
  79. Winfree, R. , Aguilar, R. , Vázquez, D.P. , LeBuhn, G. & Aizen, M.A. (2009) A meta‐analysis of bees' responses to anthropogenic disturbance. Ecology, 90, 2068–2076. [DOI] [PubMed] [Google Scholar]
  80. Woodcock, T.S. , Larson, B.M. , Kevan, P.G. , Inouye, D.W. & Lunau, K. (2014) Flies and flowers II: floral attractants and rewards. Journal of Pollination Ecology, 12, 63–94. [Google Scholar]
  81. Wratten, S.D. , Bowie, M.H. , Hickman, J.M. , Evans, A.M. , Sedcole, J.R. & Tylianakis, J.M. (2003) Field boundaries as barriers to movement of hover flies (Diptera: Syrphidae) in cultivated land. Oecologia, 134, 605–611. [DOI] [PubMed] [Google Scholar]
  82. Zuur, A.F. , Ieno, E.N. , Walker, N.J. , Saveliev, A.A. & Smith, G.M. (2009) Mixed Effects Models and Extensions in Ecology. Springer, New York City, New York. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix S1. Macrohabitats observed on each studied sites in the Upper Ardenne (Belgium).

Appendix S2. Abundance of hoverfly species observed on the five target plant species during 3 years in 10 heathland patches located in the Upper Ardenne, Belgium.

Appendix S3. Life‐history traits of hoverfly species observed on the five target plant species in the Upper Ardenne (Belgium), according to Speight et al. (2015), Gilbert (1985) and Ssymank (1991).

Appendix S4. Canonical correspondence analysis (CCA) ordination plots of hoverfly abundance (cross), site (black points) and environmental data (arrows) for group 2 hoverfly species.

Click here for additional data file. (255KB, docx)

Articles from Insect Conservation and Diversity are provided here courtesy of Wiley

RESOURCES