Introduction

A holistic view of community structure, encompassing primary producers to consumers, is required to better characterize the impacts of global climate change and anthropogenic effects on ecosystems to inform mitigation measures [1]. However, traditional vertebrate diversity surveillance techniques, such as transect sampling and trapping, can be labor-intensive and require considerable expertise [2–4]. Active sampling of individuals misses other taxa that may be important in an organism’s food web, as well as its ecosystem function. More popular minimal-effort surveillance techniques, like camera trapping, also come with their own disadvantages and suffer from sampling biases that overlook important dimensions of the community. Camera trapping specifically may miss nearby taxa outside of the camera’s field of view and is often biased towards capturing images of larger animals [5–7]. Additionally, such methods limit the number of taxa or guilds that can be evaluated simultaneously, requiring implementation of several methods (increasing sampling effort and cost) for total community assessment.

The shortcomings of traditional vertebrate surveillance methods can be improved with the use of necrophagous insects. Blow flies (Diptera: Calliphoridae) develop on animal carcasses as larvae and then continue to “sample” (i.e., taste) vertebrate resources (e.g., carcasses, feces) during their adult lives. These natural behaviors can then be used by scientists to indirectly monitor vertebrate communities [8, 9]. Previous work has indicated that DNA from such vertebrate resources can be extracted from the blow fly guts and sequenced for species identification [10–12]. Blow fly-derived vertebrate richness assessments have been determined to out-perform traditional surveillance methods [11, 13], and as an added bonus, they require minimal field-sampling time with increased detection potential (e.g., in 240 min, 26–43% of common mammal species in an environment can be detected using blow flies [12]). Furthermore, a vertebrate-specific fecal test (targeting vertebrate urobilinoids) can determine if the resource fed on by the adult fly was feces vs. carrion [14], giving more information to the researcher about whether the vertebrates assessed were alive or dead at the time of sampling.

Here we present a complementary method to DNA analyses by using nitrogen (δ¹⁵N) and carbon (δ¹³C) stable isotopes of adult blow flies to understand ecosystem vertebrate trophic structure. Molecular DNA analysis of fly guts can only give information about recent resources the fly has visited, not the type of carcass used for larval development. In contrast, stable isotope signatures provide a record of resources utilized over a longer time period with the knowledge that the resource was indeed deceased. Knowing the prior development substrate of a fly is particularly important for gaining a more comprehensive view of the vertebrate community in an area, as DNA profiles from flies may have originated from resources other than carcasses (e.g., feces). The adult fly’s nitrogen isotopic composition, when calibrated for fractionation effects during feeding, can indicate the trophic level of the individual carcass that the fly used for its larval development [15, 16] (Fig 1). We propose that by employing this method in conjunction with gut content sequencing (which approximates community diversity at the time of sampling), scientists will be able to track faunal distributions of the recent past (i.e., two to three weeks before sampling, when the life cycle of the fly began).

Fig 1

Schematic illustrating the concept for determining trophic position of the carcass.

1) Blow fly larvae develop on a carcass (e.g., a deer) by consuming the animal’s tissue. 2) Larvae leave the carcass in order to pupate in the soil. 3) Adult blow flies emerge from puparia approximately 2 weeks after initial pupation and begin the free-roaming stage of their life cycle. 4) Analysis of a random adult blow fly can reveal the trophic position of the animal on which it developed as a larva. Different colors represent different carcass trophic positions and thus the variable isotopic signatures present within wild populations of adult blow flies. In the example illustrated here, the adult blow fly retains the δ¹⁵N value of the carcass (“herbivore”) from its larval stage.

Stable isotope analysis (SIA) is an important tool in trophic food web ecology. Animals incorporate isotopes from food into their own tissues, which provides a record of the animal’s diet over time [17]. Nitrogen isotopes are associated with the trophic position of a consumer due to the preferential incorporation of ¹⁵N into tissues, leading to an approximate 3‰ increase in δ¹⁵N compared to the consumer’s diet [18, 19]. However, this generalized enrichment value has been challenged, as isotopic variation among animal taxa can be significant, and additionally, the most relevant variation actually occurs in the trophic fractionation [20, 21]. Similarly, dietary carbon isotopes are associated with the source of carbon in an animal’s diet. Carbon at the base of the food web originates from primary producers (e.g., plants, phytoplankton, algae), which exhibit different ranges of carbon isotope fractionation based on the photosynthetic pathways (e.g., C₃ or C₄) used to fix CO₂ [18, 22]. Typically, the δ¹³C value of the carbon source becomes incorporated in the tissues of consumers with little fractionation (within 1‰), though the extent of fractionation can be variable for different taxa.

The usefulness of stable isotope ecology in entomological studies has increased over the last decade [23]. Though few and far between, studies involving necrophagous blow flies have revealed important ecological information on how scavengers contribute to an ecosystem. Previous ecological studies have shown that blow flies can be used to estimate the fate of nutrients and energy entering an ecosystem, as >80% of adult blow flies (Calliphora spp.) reared from salmon runs (estimated at 4–7 million larvae per watershed) exhibited salmon-based δ¹³C and δ¹⁵N values [16]. Furthermore, blow flies and other necrophagous insects show a reliable trend in isotopic signatures that tracks with seasonal and spatial changes in salmon breeding and die-offs [24]. More recently, researchers have realized the forensic significance of blow fly SIA. Bernhardt et al. (2017) analyzed isotopic fractionation in the green bottle blow fly (Lucilia sericata) to determine if SIA could discriminate between humans and other animals as larval substrates and found that the human trophic position cannot be deconvoluted from animals occupying the same position [15]. Additionally, Matos et al. (2018) found that although essential amino acids do not vary throughout the life stages of blow flies, non-essential amino acids become more depleted in ¹³C in the adult fly compared to its larval substrate, which could be utilized by forensic entomologists when making time since death estimations with entomological evidence [25].

At the intersection of decomposition ecology and conservation biology is the necrophagous blow fly, which, we propose, can be used as an unbiased sampler of its environment. The purpose of this study was to test the utility of using blow fly SIA to infer the trophic level structure of dead vertebrates in an area the size of a national park, which currently can only be done with traditional surveying methods and molecular analysis. However, in order to draw any meaningful conclusions from “wild” isotopic data, it was imperative to develop baseline, species-specific datasets for comparative purposes [19]. One of our objectives was to determine if the predictable fractionations for δ¹⁵N and δ¹³C expected from previous food web studies (3‰ for δ¹⁵N, ± 1‰ for δ¹³C) holds true for the black blow fly, Phormia regina (Meigen) (Diptera: Calliphoridae). Phormia regina is one of the most common and ubiquitous blow flies in North America, and it has been previously investigated as an environmental DNA sampler [4]. Furthermore, we investigated the systematic offset in δ¹⁵N and δ¹³C between the adult fly and its larval diet, and whether this fractionation (Δδ = δ_fly− δ_{larval resource}) can be affected by other variables, such as the preservation method of the fly, adult feeding events, and age and sex of the fly.

Results

As the overall goal of this study was to assess carcass trophic levels using SIA of adult blow flies, it was necessary to perform a series of experiments informing our analysis of wild-caught flies. First, it was important to determine if our field preservation method would significantly affect isotopic signatures in wild-caught flies (first experiment). In order to extract trophic structure from the wild fly isotope compositions, the fractionation effects were determined in controlled feeding scenarios (second experiment) as a basis for calculating the isotopic composition of the dietary sources. Finally, there was a need to understand how additional adult feeding (third experiment) and aging of the fly (fourth experiment) impacted isotopic fractionation values. All experiments were vital for determining how to use wild fly isotope values for carcass trophic level assessment in natural landscapes.

Effect of preservation

It was important to address the practicality of this method, therefore, we determined if isotopic composition effects could be introduced during sample preservation as field methods differ from laboratory-based methods. We assessed this potential impact by comparing samples stored in ethanol (a typical field-based preservation method) and the effect of freezing specimens at -20°C (a typical lab-based preservation method). There was no significant difference in δ¹⁵N or δ¹³C between the two preservation methods (P > 0.05 for both). Furthermore, there were no significant differences between male or female fly isotope compositions observed in either preservation method (Table 1).

Table 1

Summary of isotopic data from the preservation experiment.

Preservation Methods	Sex	δ15N	δ13C
Ethanol	Male	8.78 ± 0.63	-20.00 ± 0.64
	Female	8.39 ± 0.62	-20.97 ± 0.47
Frozen	Male	7.70 ±0.62	-20.36 ± 0.45
	Female	6.50 ± 0.76	-20.46 ± 1.25

Mean (± standard deviation) δ¹⁵N and δ¹³C isotopic values from male (N = 10 per treatment) and female (N = 10 per treatment) flies killed via ethanol of freezing.

Persistence of larval diet isotopic signature

The main premise of diet analysis using stable isotopes is that an animal’s isotopic signatures reflect that of its diet, with different diets producing different isotopic values in the consumer. However, the overall deviation (= fractionation) from diet to consumer should remain relatively stable for a given organism. Therefore, we wanted to test the hypothesis that different larval diets produce similar fractionation patterns among adult blow flies, regardless of differences in their raw isotopic values. This is important to investigate in order to provide a reliable method of determining the larval diet from wild-caught adult blow flies. Adult flies exhibited systematic trophic level offsets in δ¹⁵N and δ¹³C relative to their respective larval resources (beef liver, chicken liver, or salmon filet), with an overall 3.9‰ increase in δ¹⁵N and a 2.0‰ decrease in δ¹³C across all treatments (Table 2, Figs 2 and 3). The MANOVA revealed a significant effect of treatment (P < 0.001) and sex (P < 0.001) on δ¹⁵N and δ¹³C fractionation together. A post-hoc ANOVA showed a significant impact of treatment (P < 0.001) on both δ¹⁵N and δ¹³C fractionations and sex for δ¹³C (P < 0.001), yet there were no significant interaction effects. Specifically, flies fed salmon exhibited significantly lower nitrogen isotope fractionations compared to flies fed other animal tissues, and females tended to have lower δ¹³C values compared to males. In all treatments, female P. regina exhibited nearly identical δ¹⁵N compared to males but were always lower in δ¹³C by approximately 1.0‰.

Fig 2

Boxplots summarizing the fractionation values for all experiments.

Mean nitrogen fractionation (Δδ¹⁵N; panel A) and mean carbon fractionation (Δδ¹³C; panel B) are illustrated for each treatment across experiments. Open circles represent females and closed circles represent males. The red line indicates the mean value for both Δδ¹⁵N and Δδ¹³C across all experiments.

Fig 3

Scatterplot of δ¹⁵N and δ¹³C values for the persistence of larval diet experiment.

Isotope relationships observed in this experiment demonstrate the possible range of fractionations during larval uptake. Circled clusters of data represent adult flies that were reared on their respective larval diets: chicken liver (blue), beef liver (grey), or salmon filet (green).

Table 2

Summary of isotopic data from feeding experiments.

					Fractionation (‰)
Experiment	Larval Diet	Adult Diet	Other	Sex	Δδ15N	Δδ13C
Persistence of Larval Diet Isotopic Signatures	Beef	None	None	Male	4.2 ± 0.2	-2.4 ± 0.3
				Female	3.9 ± 0.3	-3.4 ± 0.5
				Overall	4.1 ± 0.3	-2.9 ± 0.7
	Chicken	None	None	Male	5.0 ± 0.4	-1.2 ± 0.7
				Female	4.9 ± 0.3	-2.4 ± 0.4
				Overall	4.9 ± 0.4	-1.8 ± 0.8
	Salmon	None	None	Male	2.8 ± 0.3	-1.0 ± 0.3
				Female	2.9 ± 0.2	-1.9 ± 0.4
				Overall	2.8 ± 0.2	-1.4 ± 0.6
Effects of Single Adult Feeding Event	Beef	None	None	Male	3.8 ± 0.5	1.2 ± 1.5
				Female	3.6 ± 0.6	0.4 ± 1.5
		Beef	None	Male	3.4 ± 0.7	1.3 ± 1.2
				Female	3.1 ± 0.8	0.4 ± 1.1
		Salmon	None	Male	3.8 ± 0.7	1.4 ± 1.5
				Female	4.0 ± 0.9	0.4 ± 1.3
		Feces	None	Male	3.8 ± 0.6	1.3 ± 1.6
				Female	3.9 ± 0.7	0.3 ± 1.4
				Overall	3.7 ± 0.7	1.4 ± 0.8
Effects of Age	Salmon		4h	Both	1.0 ± 0.4	2.0 ± 1.4
			3d	Both	1.1 ± 0.9	2.4 ± 0.9
		None	9d	Both	1.4 ± 0.6	2.6 ± 1.0
			15d	Both	1.3 ± 0.3	2.8 ± 1.2
				Overall	1.2 ± 0.5	2.4 ± 1.2
		Chicken	4h	Both	0.5 ± 0.4	1.3 ± 0.6
			3d	Both	-0.2 ± 0.6	2.3 ± 1.8
			9d	Both	-0.2 ± 1.1	3.0 ± 1.1
			15d	Both	0.3 ± 0.4	3.5 ± 1.7
				Overall	0.1 ± 0.6	2.5 ± 1.5)

Mean (± standard deviation) of isotopic fractionation factors (Δδ = δ_fly− δ_{larval resource}) of flies from various adult feeding treatments relative to their larval resource.

Urban and national park samples

Wild fly δ¹⁵N values ranged from 5.7 to 14.8‰ and δ¹³C ranged from -29.2 to -19.2‰. Raw data distributions grouped roughly according to region (urban, Smokies, Yellowstone). Flies collected in the urban environment exhibited significantly higher δ¹⁵N values compared to the other two environments (P < 0.001), with more similar δ¹⁵N values seen between the two national parks. In terms of δ¹³C, the values of Yellowstone flies were significantly lower than flies from the urban environment and from the Smokies (P < 0.001). In order to determine the carcass trophic positions derived from wild fly isotopic signatures, all fractionation data for female flies from all experiments (since only female flies were analyzed from the wild sampling collections) was averaged for both δ¹⁵N and δ¹³C, giving a mean of 2.1‰ enrichment in δ¹⁵N and 1.4‰ depletion in δ¹³C. As these fractionation values aligned with similar studies of blow flies [15], the average fractionation value of 2.1 was subtracted from wild fly δ¹⁵N values and the average value of 1.4 was added to δ¹³C values, resulting in carcass δ¹⁵N values ranging from 3.6 to 12.8‰ and carcass δ¹³C ranging from -30.6 to -20.6‰.

Based on the isotopic composition of animal tissues gathered from the literature, three conservative trophic categories were statistically differentiated based on δ¹⁵N values: herbivore, carnivore, and fish (Table 3). As an overlap did exist between herbivores and carnivores, an “herbivore/carnivore” category was also included to account for when the two trophic levels cannot be differentiated. Though fish do not technically represent a true trophic position, their high δ¹⁵N values are different enough from all other vertebrates that they were designated as their own category for the purposes of this study. Omnivores were excluded as a definitive category as their isotopic values (both δ¹⁵N and δ¹³C) overlapped with several other categories and could therefore not be statistically differentiated.

Table 3

Ranges of δ¹⁵N values for each vertebrate trophic position obtained from the literature review.

Trophic Level	δ15N
Herbivore	<5.7
Herbivore/Carnivore	5.8–5.9
Carnivore	6.0–10.9
Fish	>11.0

Overall, carcass trophic structure was significantly different between the urban environment and both national parks (P < 0.001; Fig 4). The national parks had greater similarity in trophic structure, with approximately three quarters of flies exhibiting carnivore signatures, approximately one quarter exhibiting herbivore signatures, and few exhibiting signatures indicative of fish. Conversely, 90% of flies from the urban environment had detectable carnivore carcass signatures, with the remainder of these flies mostly exhibiting signatures of fish.

Fig 4

Putative carcass trophic values transformed from wild fly data by mean experimental fractionation factors for females.

Panel A: Proportion (out of 100%) of wild flies from urban, Smokies, and Yellowstone regions with isotopic signatures indicative of four carcass trophic positions: herbivore (green), herbivore/carnivore (grey), carnivore (orange), fish (blue). Panel B: Summary of results from wild flies are illustrated via scatterplot with δ¹⁵N given on the y-axis and δ¹³C values on the x-axis. Black circles = flies from the Great Smoky Mountains National Park; grey triangles = flies from the urban environment (Indianapolis, IN, USA); black X’s = flies from Yellowstone National Park.

Discussion

This study has demonstrated that blow flies can be used as an informative surveillance tool for monitoring terrestrial vertebrate communities. Using our experimentally derived isotopic fractionation data, we were able to infer the broad vertebrate trophic level structure present in three different environments of varying degrees of human influence: Yellowstone National Park, a pristine environment; the Great Smoky Mountains National Park, a protected area that is also the most visited national park in the United States; and the greater Indianapolis area, which is an urban environment boasting a population of almost 2 million people. Specifically, we were able to determine what types of carcasses our adult flies used for their prior larval development, which cannot be done with DNA analysis of the adults as the carcass DNA is broken down during metamorphosis. By comparing data from the current SIA study to our previous molecular DNA analysis of the same flies, we have demonstrated that blow flies utilize primarily carnivore carcasses for larval development and primarily herbivore resources (either carcasses or feces) for adult consumption in the three environments sampled. We provide evidence that SIA is a powerful method for gleaning information about what these insects were doing before their metamorphosis into adults. With careful experimentation, we were able to demonstrate a predictable δ¹⁵N fractionation in adult flies compared to their larval diet that does not appear to vary significantly by adult diet or by the aging process of the adult fly.

The overall isotopic fractionation of female blow flies from our laboratory experiments aligns with similar values compiled from the extensive body of animal isotopic literature [27], lending more confidence to deriving carcass trophic diversity metrics from these flies. The trophic positions of carcasses given here align with much of the molecular and chemical data from the same flies previously reported [12]. Vertebrate DNA data from flies in the urban environment showed a greater proportion of medium sized carnivores (e.g., dogs and raccoons) visited by adult flies compared to flies collected in the national parks, and herbivores (e.g., white-tailed deer, groundhogs, eastern cottontails) made up a much smaller proportion of animals detected from molecular analysis of urban flies. The molecular data is supported by the isotopically-derived data presented here with overall higher δ¹⁵N values observed in flies from the urban environment compared to the other regions. This may be attributed to a higher number of carnivore carcasses available due to road mortality events caused by humans. Indianapolis is the 13^th largest city in the United States [28] with 20,000 km of roadways [29]. Subsampling of roadkill deaths from 9 locales in Indiana from 2005 to 2006 showed 457 mammal mortality events, with 39% and 32% of those deaths made up of Virginia opossums and raccoons, respectively [30]. This aligns with both molecular and isotopic data from the current study, as both opossums and raccoons made up a large portion of vertebrate DNA derived from flies in the urban environment, and they may have potentially served as a portion of the ~90% of carnivores utilized by flies for larval development [12]. Conversely, the DNA-derived vertebrate composition of the national parks contrasted with isotopic data, suggesting that carnivore carcasses are utilized more often for larval development and herbivore resources (e.g., feces) are used more often for adult survival [14, 31]. Given the sheer abundance of herbivore populations in the national parks, we initially expected a proportional number of flies collected in these environments to exhibit herbivore carcass signatures. However, we did not account for the presence of numerous vertebrate predators and scavengers in the national parks: coyotes, black bears, eagles, and vultures in the Smokies; black bears, grizzly bears, wolves, ravens, and eagles in Yellowstone. The presence of these scavengers likely introduces strong intra-guild competition with blow flies vying for opportunities to exploit carcasses. In Yellowstone specifically, wolves provision other vertebrate scavengers with carcasses from their hunts, most of which are herbivores (e.g., elk) [32]. Some vertebrate scavengers, like ravens and coyotes, track wolf packs in order to exploit wolf-kills [33, 34]. In order for blow flies to “win” this inter-class competition, they must first locate the carcass and successfully colonize it before other scavengers arrive, with their ensuing eggs, larvae, and pupae constantly facing the threat of predation and death by larger animals. Compared to highly sought-after herbivores, the relatively low abundance of predators and scavengers in these environments may give flies an advantage for exploiting carnivore carcasses simply due to decreased competition. In fact, some studies suggest that vertebrate scavengers actively avoid consuming carnivore carcasses, or at least conspecifics, as scavenging these resources may be associated with a greater likelihood of exposure to infectious diseases [35]. This avoidance strategy by vertebrate scavengers has also been shown to facilitate greater opportunities for invertebrate scavengers to exploit carnivore carcasses [36], which likely explains the carnivore bias in our blow flies collected in the national parks.

Higher δ¹³C values were observed more in flies from the urban and Smokies environments compared to flies from Yellowstone. This could likely be an influence of corn-based foods (generated by humans) being more available in this environment. Domestic or agricultural animals being fed or supplemented with corn-based diets make up a portion of the animals detected by DNA in these areas (e.g., cows, pigs, chickens), therefore their muscle tissues may be more enriched in ¹³C assimilated from their C₄ (e.g., corn-based) diet. Additionally, opportunistic vertebrate scavengers (e.g., raccoons, opossums, black bears) may also have access to human junk food, fast food (often deep-fried in corn oil, [37]), and corn-based food waste that could have the same isotopic effect on their tissues. These animals are the likely vectors for introducing corn-based diets into the fly tissues. If flies develop on the tissues of any of these animals, their δ¹³C isotopic signatures would more likely resemble those of corn-fed animals. Conversely, flies collected in Yellowstone had lower δ¹³C values compared to the other regions, which indicates that the herbivores they fed upon were potentially consuming a mixture of C₃ and C₄ plants.

Unlike nitrogen, there was a consistent sex-effect on carbon fractionation with female flies almost always exhibiting lower δ¹³C values compared to males. This could be explained by quicker turnover rates that might be occurring in female flies throughout their lifetimes as compared to males. Isotopic turnover can vary among different tissues within an individual, e.g., lipids that tend to have low δ¹³C (1–2 ‰) relative to body tissue [38]. This would make sense given that female blow flies have a higher fat content on average compared to males [39]. Larvae of other fly species, like Chironomus acerbiphilus (Diptera: Chironomidae), can exhibit increased δ¹³C during periods of starvation due to the preferential breakdown of body tissue (e.g., muscle) and respiration of ¹²CO₂ when dietary carbon is unavailable [40]. Thus, sex differences in δ¹³C seen in this study could be due to the breakdown of previously assimilated ¹²C from the larval diet in the females, perhaps due in part to reproductive needs of the maturing female flies. Conversely, males may assimilate ¹³C quicker than females, leading to the differences observed (males with higher δ¹³C). Additionally, males and females may metabolize components that have different δ¹³C values altogether. Regardless of the exact mechanism, it will be important for future studies to take sex into consideration when refining interpretations from carbon isotopic data derived from flies.

We also provide evidence that, for the most part, the adult diet does not alter the larval diet isotopic signatures in adult flies. This means that the carbon and nitrogen isotopic signatures obtained from flies can be reliably attributed to the fly’s larval resource and not environmental or physiological conditions experienced by the adult fly itself. We also showed that fly age does not affect the δ¹⁵N signatures in blow fly tissues for up to 15 days (an average lifespan of a wild fly is three to four weeks, [41]); however, δ¹³C increases as the fly gets older, and this is exacerbated if the fly has taken a protein meal during adult life. Other female holometabolous insects have shown similar decreases in δ¹³C throughout their adult lifetimes caused by rapid isotopic turnover in reproductive organs and fatty tissue [42]. The δ¹³C of invertebrate tissues can evolve with time as a result of metabolic activity combined with changes in dietary sources that have distinct isotope compositions [37]. Therefore, the significant enrichment in our chicken-fed flies over time may be due to the switch in diet from chicken to sugar and water for the remainder of the fly’s lifetime, as well as isotopic turnover in the tissues. Though this does represent an effect of adult diet on the isotopic signature of the fly, it is likely not seen in nature where flies have access to multiple protein sources (including carrion, refuse, and feces) throughout their lifetime.

Though the experiments performed in this study specifically targeted the effects of feeding and age on isotopic signatures, other intrinsic variables, such as fly size, were not explicitly investigated but could prove especially informative for this method. The size of the adult fly is influenced by the amount of food consumed as a larva, i.e., larger larvae = larger adults. Therefore, the density of larvae on a resource can significantly affect their size, and crowded resources (e.g., carcasses) tend to produce smaller adults due to larval competition [43]. Additionally, it is common that female blow flies are larger on average than males [44]. Therefore, there may be a sex, larval size, and carcass size effect on δ¹⁵N signatures in adult flies that should be accounted for if using flies to ascertain carcass trophic positions of a terrestrial environment. For example, the average carbon isotope fractionation for the diet experiments exhibited changes in magnitude and direction that ranged from -2.9‰ to 2.5‰. For each experiment, food sources were well-controlled and isotopically characterized, although the amount of sugar uptake by individual flies is unknown. The size of the rearing containers limited the number of larvae to ~150–200 and therefore mitigated resource limitations or density effects. Although it is unclear whether the variance observed in carbon isotope fractionation is environmental or intrinsic to the fly metabolism, the range of ~ ± 3‰ is small relative to the distinct and non-overlapping δ¹³C values of C₃ or C₄ plant-based carbon sources [18, 22] that would be available to temperate-zone biomes.

When used in conjunction with other analytical methods, SIA can add another rich dimension to community assessment survey methods using carrion insects. Today’s changing climate requires special tools to assess community diversity that can be quickly implemented and be as minimally invasive as possible to maintain the integrity of fragile ecosystems. The use of SIA of necrophagous insects, like blow flies, allows researchers to gain a snapshot of the overall vertebrate trophic level structure of an ecosystem with little disturbance. Simultaneous use of isotopic, molecular, and chemical data derived from flies in these areas can give a quick, comprehensive survey of vertebrate communities that rivals any traditional survey method currently being used. Additionally, SIA of blow flies allows researchers to assess the dead vertebrate community in a way that is difficult to accomplish using other techniques, including molecular methods. As most blow fly larvae are limited to developing on dead animals, isotopes reveal carcasses that are available in an environment, with no additional resources (e.g., feces) confounding interpretations.

Materials and methods

National park and urban samples

In order to determine the utility of using isotopic analysis of flies to characterize the carcass trophic level structure in the wild, we sampled wild blow flies in three regions: 1) Indianapolis, Indiana (“urban”), 2) the Great Smoky Mountains National Park (“Smokies”), and 3) Yellowstone National Park (“Yellowstone”), as previously reported [12]. The permits to sample in the national parks were obtained via application through the National Park Service Research Permit and Reporting System: Great Smoky Mountains: Permit # GRSM-2018-SCI-2036, study # GRSM-02039; Yellowstone National Park: Permit #YELL-2018-SCI-8046, study # YELL-08046). All locations in the parks were authorized and included areas close to roads/parking areas and away from developed areas. Sampling occurred at N = 4 separate timepoints from June to August at each of N = 4 different geographic sites per region (Table 4). Sampling in the urban environment occurred in 2016 while sampling in the national parks occurred in 2018. Flies were morphologically identified and sorted so that only females of P. regina were used. We used a random number generator to select a maximum of N = 10 random females from each individual spatiotemporal sample. Selected flies that underwent DNA [12] and fecal [14] analysis were also subjected to SIA (urban, N = 100; Smokies, N = 96; Yellowstone, N = 98).

Table 4

Summary of blow fly sampling sites per region.

Region	Site	City, State	Coordinates	Flies Analyzed (proportion)
Urban	Military Park	Indianapolis, IN	39.771, -86.169	20 (30.3%)
	Northwest Park	Greenwood, IN	39.629, -86.144	30 (30.9%)
	Skiles Test Park	Indianapolis, IN	39.868, -86.049	20 (12.3%)
	University Park	Greenwood, IN	39.611, -86.051	30 (28.3%)
Smokies	Site 1	near Gatlinburg, TN	35.735, -83.413	20 (13.6%)
	Site 2	near Gatlinburg, TN	35.704, -83.365	20 (13.0%)
	Site 3	near Gatlinburg, TN	35.663, -83.526	26 (17.1%)
	Site 4	near Gatlinburg, TN	35.671, -83.680	30 (5.2%)
Yellowstone	Site 1	near Gardiner, MT	44.614, -110.414	26 (39.4%)
	Site 2	near Gardiner, MT	44.958, -110.542	30 (22.6%)
	Site 3	near Gardiner, MT	44.958, -110.312	22 (32.4%)
	Site 4	near Gardiner, MT	44.886, -110.144	20 (37.7%)

Urban = Indianapolis, IN, USA, Smokies = Great Smoky Mountains National Park, Yellowstone = Yellowstone National Park. Only female P. regina were selected for analyses. The last column refers to the number of flies analyzed per site, as well as the proportion of flies analyzed out of the entire sample (in parentheses).