Hover fly (Diptera: Syrphidae) diversity and seasonality in North Georgia apple and peach orchards

Abstract

Crop pollination and natural biological control provided by beneficial insects have an economic worth of hundreds of billions of dollars annually. Apple and peach production in North Georgia are economically important industries that benefit from these ecological services. Hover flies are dual ecosystem service providers that have been relatively understudied in orchard ecosystems. We investigated the diversity and seasonal activity of hover flies in apple and peach orchards at 2 sites in North Georgia from March to October 2020 and 2021. Bowl traps were used to sample hover flies in orchard edge and interior habitats. The aphidophagous species Toxomerus geminatus (Say) (Diptera: Syrphidae) and Toxomerus marginatus (Say) (Diptera: Syrphidae) comprised 86.6% of the total hover flies collected. Apple orchards yielded the greatest hover fly presence, species richness, and Toxomerus spp. abundance. Hover fly richness and diversity were greatest during postbloom, but Toxomerus spp. abundance was greatest during the bloom period. No differences in presence, richness, diversity, or Toxomerus spp. abundance were found between edge and interior habitats. Toxomerus geminatus and T. marginatus were dominant from March through August, with T. geminatus being more abundant than T. marginatus in March, early April, and August. October sampling produced the greatest hover fly richness. Our results suggest that hover flies are abundant in North Georgia orchards and exhibit substantial spatial and temporal variation in richness and diversity. Expanded studies incorporating additional sampling efforts and methods are needed to further characterize the hover fly fauna and their impact on North Georgia apple and peach orchards.

Graphical Abstract

Open in new tabDownload slide

Introduction

Crop pollination by insects has an economic value of about $350 billion worldwide (Lautenbach et al. 2012), and natural enemies of insect pests of crops reduce expenditures on pest management by at least an order of magnitude (Losey and Vaughan 2006). Despite the enormous value of insect pollination and natural biological control, there is growing evidence that anthropogenic drivers such as pollution, land use changes, habitat fragmentation, and climate change are contributing to declining insect populations worldwide, including important crop pollinators such as wild bees. Mitigating these trends requires a multifaceted approach, including the conservation of alternative crop pollinators and natural enemies.

The loss of beneficial insects could have a devastating worldwide impact, which would also be felt locally. For example, apple and peach production in Georgia, USA, depend heavily on beneficial insects and had a combined value of $90.5 million in 2022 (University of Georgia Extension 2023a, 2023b). Apples are grown commercially primarily in the north Georgia counties of Fannin and Gilmer, and in this region, apples and peaches are often grown in adjacent orchards. Apple trees must cross-pollinate and thus depend on insect pollinators, whereas peach trees are primarily wind-pollinated and can self-pollinate. However, supplemental insect pollination can increase the fruit set of peaches, as well as the size and weight of the fruit (Partap et al. 2000, Tanda 2021). Both apple and peach trees are subject to attack by a variety of insects. Pests such as plum curculio, scale insects, fruit and bud moths, borers, and aphids attack various tree structures and can cause serious economic losses. Thus, these orchard crops depend greatly on insect-mediated pollination and pest management but also often receive intensive treatment with chemical pesticides to maintain yield and fruit quality (Blaauw et al. 2023). One such group of insects that can deliver these ecosystem services are the hover flies (Diptera: Syrphidae).

Hover flies, also known as flower flies or syrphid flies, are a diverse family that contains over 6,800 species worldwide within 284 genera (GBIF Secretariat 2023). Most adult hover flies are flower visitors, feeding on nectar and/or pollen, and as larvae, many hover fly species are predators. Two of the 4 subfamilies of hover flies (Syrphinae and Pipizinae) primarily contain species that prey on aphids, thrips, and other small insects that can be serious pests of crops (Skevington et al. 2019). These hover fly species have been described as dual ecosystem service providers because they play an important role in both crop pollination and pest management (Schneider 1969, Omkar and Mishra 2016; Rader et al. 2016, Klecka et al. 2018, Dunn et al. 2020). Studies in a variety of regions have identified hover flies, along with Apis mellifera and various wild bees, as key pollinators in apple orchards. In a review of the worldwide importance of insect pollination in apple orchards, Pardo and Borges (2020) found that, while bees are consistently important pollinators in apple orchards, hover flies were key pollinators in roughly 25% of studies. In organic apple orchards in southeastern France, hover flies were the most efficient natural enemies of rosy apple aphid (Dysaphis plantaginea Passerini) (Dib et al. 2010). In Mediterranean peach orchards, the hover fly Episyrphus balteatus (DeGeer) was one of the key natural enemies of the green peach aphid (Myzus persicae (Sulzer)) (Aparicio et al. 2021). However, little is known about the hover flies associated with orchard crops in the southeastern United States. Our objective was to document the spatial and temporal diversity and distribution of hover flies in North Georgia peach and apple orchards.

Materials and Methods

This study was conducted from March to October 2020 and 2021 in 4 orchards, 2 (1 peach and 1 apple) at each of 2 sites in north-central Georgia: “Site I,” in Fannin County (peach: 34.900536°, −84.360770°; apple: 34.900575°, −84.361577°; 21 ha total area), and “Site II,” in Gilmer County (peach: 34.660865°, −84.422715°; apple: 34.660295°, −84.422752°; 28 ha total area). At each site, peach and apple orchards are grown in adjacent stands. At site I (apple), the rootstock was M9, the scion was Gold Rush, spacing was 1.5 m between trees within rows and 6.1 m between rows, and tree age was 8 years. At site I (peach), the rootstock was Bailey, the scion was Gala, spacing was 3.0 m between trees within rows and 6.1 m between rows, and tree age was 6 years. Tree rows were oriented in an east-west direction at site I, and the orchards shared one border, with each orchard having 2 borders adjacent to other orchards of the same crop and one border ca. 10 m from a windbreak with a diversity of unmanaged vegetation, including key hardwood tree species, such as black cherry (Prunus serotina Ehrh.), sassafras (Sassafras albidum (Nutt.) Nees), pignut hickory (Carya glabra Miller), catalpa (Catalpa bignonioides Walter), sweetgum (Liquidambar styraciflua L.), oak (Quercus spp.), red maple (Acer rubrum L.), tulip poplar (Liriodendron tulipifera L.), dogwood (Cornus florida L.), sycamore (Platanus occidentalis L.), and elm (Ulmus americana L.). Shrubs and vines included blackberry (Rubus spp.), hawthorn (Crataegus spp.), pokeweed (Phytolacca americana L.), muscadine (Vitus rotundifolia Michx.), and elderberry (Sambucus nigra subsp. canadensis (L.) R. Bolli). At site II (apple), the rootstock was M9, the scion was Granny Smith, spacing was 5.5 m between trees within rows and 6.1 m between rows, and tree age was 8 years. At site II (peach), the rootstock was Lovell, the scion was Loring, spacing was 5.5 m between trees within rows and 6.1 m between rows, and tree age was 12 years. Tree rows were oriented southeast to northwest at site II, and the orchards shared one border, with the remaining 3 borders of each orchard adjacent to other orchards of the same crop (Fig. 1A). At both sites, vegetation between rows consisted predominantly of grasses (Poaceae) and clover (Trifolium L.), with primarily bare soil within rows. Pest management was conventional, with a pyrethroid-based program targeting primarily plum curculio (Conotrachelus nenuphar (Herbst)), oriental fruit moth (Grapholita molesta (Busck)), and several species of stink bugs (Pentatomidae). Neonicotinoids were used on an as-needed basis for woolly apple aphids, Eriosoma lanigerum (Hausmann), which were commonly present in apple orchards.

Twenty-four bowl traps were placed within each orchard at each site, arranged in 8 groups of 3, with one each of blue, white, and yellow to represent a diverse spectrum of floral coloration (Fig. 1B). Bowl traps were composed of ~96.1 ml clear plastic soufflé cups, 7.0 cm in diameter (Dart Container Corporation, Mason, MI, USA), which were then spray-painted (Rust-Oleum Corporation, Vernon Hills, IL, USA). Each group of bowls was placed on a wooden platform and placed at the base of the apple or peach tree (Fig. 1B) to maximize the likelihood of collecting hover flies associated with the orchards rather than vagrant species from outside of the orchards. A transect of 4 groups of bowls, spaced 20 m apart, was placed along the edge of the orchard, and another similar parallel transect was placed 50 m within the orchard (Fig. 1A). Bowls were filled with unscented soapy water (Seventh Generation Incorporated, Burlington, VT, USA) and deployed for one 48-h period every 5–15 days during bloom periods (mid-March to early April for peach, early to mid-April for apple), and one 48-h period monthly otherwise. A 48-h trapping period was implemented to decrease the likelihood of trap disruption due to weather or animal disturbance. Collected hover flies were stored in 70% EtOH and identified to species level by CSC and KWM using reference specimens and Skevington et al. (2019). Voucher specimens were deposited into the University of Georgia Collection of Arthropods, Georgia Museum of Natural History, Athens, Georgia.

Statistical Analyses

Five dominance classes were established, following the classification of Wojciechowicz-Żytko and Wilk (2023a): eudominants (>10% of total), dominants (5.1%–10%), subdominants (2.1%–5%), recedents (1.1%–2%), and subrecedents (≤1%). To analyze hover fly presence, species richness, diversity, and Toxomerus spp. abundance, data collected from yellow, blue, and white bowls from each trap set and each year were pooled. Toxomerus spp. abundance was chosen as a distinct parameter due to their dominance among sampled fauna. All statistical analyses were performed in R version 4.2.2. (R Core Team 2021), and data were plotted using packages tidyverse (Wickham et al. 2019) and ggplot2 (Wickham 2016).

Analysis of Hover Fly Presence

Due to zero inflation, overall hover fly abundance was transformed into a presence vs. absence dataset by assigning a 1 to traps that contained hover flies and a 0 to traps without hover flies. A generalized linear mixed model (GLMM) was used to assess the effects of crop on hover fly presence, specifying the distribution as Bernoulli (with a binomial link function), crop species (peach vs. apple), bloom period (bloom vs. postbloom), and trap location (interior of orchard vs. edge of orchard) as fixed effects, and trap within orchard site (1|orchard/trap) and year (1|year) as random effects.

Analysis of Hover Fly Species Richness

A GLMM was used to assess differences in hover fly species richness between crop types, specifying the distribution as Poisson (with a log link function), crop species (peach vs. apple), bloom period (bloom vs. postbloom), and trap location (interior of orchard vs. edge of the orchard) as fixed effects, and trap within orchard site (1|orchard/trap) and year (1|year) as random effects.

Analysis of Hover Fly Species Diversity

Hover fly species diversity was analyzed based on the Shannon diversity index, calculated using the diversity() function in the R package vegan (Oksanen et al. 2022). A linear mixed model was used to assess differences in hover fly species diversity between crop types, specifying crop species (peach vs. apple), bloom period (bloom vs. postbloom), and trap location (interior of orchard vs. edge of the orchard) as fixed effects, and trap within orchard site (1|orchard/trap) and year (1|year) as random effects.

Analysis of Toxomerus spp. Abundance

Due to zero-inflation, a hurdle model was used to subset data to only include traps in which hover flies were present prior to analysis of the effect of crop on the abundance of the dominant genus collected, Toxomerus spp. A GLMM was used to assess the effects of the crop on Toxomerus spp. abundance, specifying the distribution as Poisson (with a log link function), crop species (peach vs. apple), bloom period (bloom vs. postbloom), and trap location (interior of orchard vs. edge of the orchard) as fixed effects, and trap within orchard site (1|orchard/trap) and year (1|year) as random effects.

Analysis of Hover Fly Community Similarity

Permanova tests coupled with principal coordinate analyses (PCoA) were used to assess hover fly community similarity as a function of bloom period (bloom vs postbloom), orchard type (apple vs peach), and trap location (interior vs. edge). All analyses were conducted using the vegan package in R: Bray-Curtis dissimilarity matrices were built using the “vegdist” function, “adonis2” was used to conduct permanova analyses, and “wcmdscale” was used for PCoA. Due to test constraints and dataset limitations, trap samples were pooled based on site, orchard type, and trap location, yielding 4 replicates for each test variable.

Results

Overall, 491 hover flies representing 11 genera and 15 species were collected (Table 1). The 2 most abundant species were T. geminatus and T. marginatus, which comprised 86.6% of total hover flies collected, with Xylota ejuncida and Toxomerus boscii accounting for another 9.6%. Hover fly abundance was relatively evenly distributed between sites, with 249 collected at site I and 242 collected at site II. A total of 157 hover flies were collected in the apple orchard edge, 137 in the apple orchard interior, 102 in the peach orchard edge, and 95 in the peach orchard interior. The 2 most commonly represented trophic groups were aphidophagy (7 of 15 species) and detritivory (5 of 15 species). Four species were classified as eudominant, dominant, or subdominant overall, whereas the remaining 11 species were subrecedents. Yellow bowls collected 378 hover flies, white bowls collected 66, and blue bowls collected 47 (Table 2). Three of the 4 most abundant species were collected predominately (60%–83%) in yellow bowls, but X. ejuncida was well represented in both white (41%) and yellow (37%) bowls.

Significantly greater hover fly presence (ANOVA, χ² = 5.157, df = 1, P = 0.023; Fig. 2A) and species richness (ANOVA, χ² = 6.508, df = 1, P = 0.011; Fig. 2B) were detected in apple compared with peach orchards. However, Shannon-Weiner diversity (ANOVA, χ² = 2.439, df = 1, P = 0.118; Fig. 2C) was similar between the 2 orchard types. There was no significant difference in hover fly presence with respect to bloom period (ANOVA, χ² = 2.755, df = 1, P = 0.097; Fig. 3A), but hover fly richness (ANOVA, χ² = 8.743, df = 1, P = 0.003; Fig. 3B) and diversity (ANOVA, χ² = 13.327, df = 1, P < 0.001; Fig. 3C) were significantly higher postbloom when compared to during bloom. Trap location (edge vs. interior) did not have a significant effect on hover fly presence (ANOVA, χ² = 0.106, df = 1, P = 0.744, Fig. 4A), species richness (ANOVA, χ² = 0.029, df = 1, P = 0.865, Fig. 4B), or diversity (ANOVA, χ² = 0.009, df = 1, P = 0.926, Fig. 4C). Toxomerus spp. were significantly more abundant in apple than in peach orchards (ANOVA, χ² = 4.130, df = 1, P = 0.042, Fig. 5A) and during bloom when compared to postbloom (ANOVA, χ² = 6.407, df = 1, P = 0.011, Fig. 5B). There was no difference in Toxomerus spp. abundance between edge and interior habitats (ANOVA, χ² = 1.397, df = 1, P = 0.237, Fig. 5C). PCoA plots (Supplementary Fig. S1) and permanova analyses reveal that hover fly community structure was highly similar for orchard type (F = 1.28, P = 0.32) and trap location (F = 0.85, P = 0.49), but there were marginal differences for bloom period (F = 3.26, P = 0.05).

From March through August, hover fly collections were dominated by T. geminatus and T. marginatus, with these species comprising 96.9% of hover flies collected (Table 3). Toxomerus geminatus comprised 93.3% of hover flies collected in March and the 2 early April collections and 83.3% of hover flies collected in August. This species was particularly abundant in the 3 Apr site I collection, with 74 individuals collected (44 in the apple orchard, 30 in the peach orchard). In September and October, T. geminatus and T. marginatus comprised 66.7% (112 of 168) of hover flies collected, with X. ejuncida and T. boscii increasing in abundance. The greatest overall hover fly abundance and richness occurred in October, with 134 individuals and 9 species.

Discussion

Our results suggest that hover fly presence and species richness, particularly Toxomerus spp. abundance, are greater in apples than in peach orchards (Figs. 2 and 5A). Apple trees are self-incompatible and require cross-fertilization, whereas peach trees can self-pollinate and are wind-pollinated, so insect pollination would be expected to be more important in apple trees. Honey bees, Apis mellifera, are widespread pollinators of apples, but there is increasing evidence that non-Apis bees and hover flies provide important complementary pollination services in this crop (Pardo and Borges 2020, and references therein). Toxomerus geminatus was especially abundant during bloom periods (Table 3; Fig. 5), suggesting that this species could contribute to pollination. Apple orchards in North Georgia also have an abundance of aphid pests (Little et al. 2020, Walgenbach et al. 2023), which may attract ovipositing hover fly species with aphidophagous larvae, such as T. geminatus, T. marginatus, and T. boscii. Aphidophagous hover flies have potential for effective biological control of aphids (Chambers 1988). Our results are consistent with those of Khan and Riyaz (2017), who also found greater hover fly abundance and richness in apple than in peach orchards in Kashmir, India.

There is evidence in the literature that nearby windbreaks, hedgerows, and other sources of diverse vegetation can help maintain beneficial insects such as hover flies that can contribute to pollination and pest control in the United Kingdom (MacLeod 1999), Sweden (Lindgren et al. 2018), and in various other locations in North America, Europe, and New Zealand (Albrecht et al. 2021). Wojciechowicz-Żytko and Wilk (2023a,b), for instance, found that apple orchard boundaries with flowering vegetation had a positive influence on aphidophagous hover flies and increased the abundance and species richness of hover flies within the orchards in southeastern Poland. Miñarro and Prida (2013) likewise found hover flies to be the dominant natural enemies in hedgerows surrounding organic apple orchards in northwest Spain. Haenke et al. (2009) also found an association between hover fly richness and density and the proximity of sown flower strips adjacent to wheat fields in North Germany. They found that the flower strips were most effective in relatively simple landscapes where the strips would presumably have the greatest impact due to less competition from alternate floral resources. However, studies in southwest England suggest that increased natural enemy densities associated with flower strips may not result in decreased pest aphid densities (Campbell et al. 2017). Intriguingly, we found no significant difference in hover fly presence, richness, diversity, community structure, or Toxomerus spp. abundance between orchard edge and interior trap locations (Table 1; Fig. 4C), suggesting that nearby windbreaks had little effect on the orchard hover fly fauna. The surrounding landscape in the location where this study was done is relatively undisturbed and complex, suggesting, in light of the Haenke et al. (2009) study, that hover flies might be more diffuse across the landscape rather than concentrated in the windbreaks.

Our results show that T. geminatus and T. marginatus dominate the hover fly fauna of North Georgia orchards throughout most of the year. This is unsurprising, given that these 2 species are among the most common hover flies found in eastern North America and are highly generalist floral visitors that can be found in heavily disturbed habitats (Skevington et al. 2019). We found T. geminatus to be much more abundant than T. marginatus through early April, which corresponds closely to the bloom periods of these 2 orchard crops. The degree to which these species visit apple/peach flowers is unknown but should be investigated. More detailed studies are needed to evaluate the relative importance of these 2 species as possible pollinators and early season biological control agents in apple and peach orchards. We note that bowl trap effectiveness is biased in favor of smaller-bodied pollinators like Toxomerus spp. (Gill and O’Neal 2015), so employing additional sampling methodologies will likely reveal additional taxa. The relatively large number of hover flies that we collected in yellow bowl traps is consistent with the results of most other studies that have used bowl traps to sample hover flies and other flies, including Wheelock and O’Neal (2016), Harris et al. (2017), and Sowmiya et al. (2019). However, Gill and O’Neal (2015) found no difference in the abundance or richness of flies collected based on bowl trap color.

Hover fly abundance and diversity peaked late in the season in our study, which is consistent with previous knowledge that hover flies are typically most diverse in spring and fall (Skevington et al. 2019). Cooler temperatures during these months are likely optimal for hover fly foraging, but some species, such as Eupeodes americanus and Syrphus rectus, are likely to migrate into North Georgia from northern latitudes (Clem et al. 2022). Orchard management practices may be particularly impactful during autumn. Certain species, such as Xylota ejuncida, were only present during September and October (Table 3). Larvae of this species are detritivores, and while specific habitats are not well-known, other members of the genus inhabit rotting logs and crop debris (Skevington et al. 2019).

Hover flies are important dual ecosystem service providers in orchards, and knowledge of the fauna of these orchards is needed to help guide management practices that preserve these and other beneficial insects. Our results suggest that North Georgia apple and peach orchards harbor a diverse hover fly assemblage, with T. geminatus and T. marginatus being the most dominant. These 2 species are aphidophagous, and aphids, particularly E. lanigerum, are key pests of apples in north Georgia. Apple orchards yielded greater hover fly abundance and richness than did peach orchards, possibly reflecting the greater floral resources and abundance of aphids associated with apple trees. The 2 dominant hover fly species, particularly T. geminatus, were most abundant early in the season, whereas overall hover fly diversity was greatest late in the season. This suggests that orchard management practices at these times should be planned with the conservation of hover flies and other beneficial insects in mind. Further sampling and additional sampling methods would likely yield additional species, particularly hover flies of larger body size. As such, future studies incorporating greater numbers of orchards as well as the surrounding landscape, multiple years, and more sampling methods are needed to further clarify the diversity and importance of hover flies in these agroecosystems.

Supplementary Material

Supplementary material is available at Journal of Economic Entomology online.

Acknowledgments

We thank Reilly Farrell and Jamal Hunter for their assistance with field and lab work. We also thank our grower cooperators for their collaboration and support.

Funding

This work was supported by Hatch Funds (#GEO00853). Funding to C.S.C. provided by USDA NIFA EWD fellowship number 2021-67034-35132.

Author Contributions

Kenneth McCravy (Data curation [equal], Investigation [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), Scott Clem (Formal analysis [equal], Investigation [equal], Validation [equal], Writing—review & editing [equal]), Jordan Bailey (Conceptualization [equal], Investigation [equal], Methodology [equal], Validation [equal]), Sabrina Elgar (Formal analysis [lead], Validation [equal], Visualization [equal], Writing—review & editing [equal]), and Brett Blaauw (Conceptualization [equal], Formal analysis [equal], Funding acquisition [lead], Methodology [equal], Project administration [lead], Resources [lead], Supervision [lead], Validation [equal], Writing—review & editing [equal])

References