Introduction

A main challenge for the consolidation of environmental forensic entomology is to understand how the patterns of distribution of insects can be used to infer about environmental crimes, especially those related to deforestation, negligence in quarantine protocols, and inadequate management of conservation units. This is a recent area of forensic entomology that has produced sound evidence about the potential of blow flies and flesh flies species as indicator of the conservation status of forested areas in the Brazilian Amazon (Sousa et al 2014). This approach has not yet been tested on tropical islands, despite the relevance of insular environments for studies on ecological and evolutionary processes that can subsidize programs for biodiversity conservation (Whittaker & Fernández‐Palacios 2007).

Continental islands are those that somehow were once linked to the continent and may have appeared following a separation of continent portions due to erosion of a primitive peninsula or to variations in the sea level during glacial periods (Gillespie et al 2008). Oceanic islands are bodies of land distant from the coast, with limited portion above sea level; the endemism of the biota in this environment is a result of the oceanic dispersion of species, as these islands have never been connected to the continent (Cowie & Holland 2006). In either insular system, the presence of invasive species and the loss of habitat are the primary short-term threats to local biodiversity. The main impacts of invasive species on their native counterparts are related to the competitive exclusion, niche displacement, hybridization, predation, and extinction (Mooney & Cleland 2001).

Most quantification of impacts caused by invasive insect species was obtained from evidences in islands (Mooney & Cleland 2001), such as the mapping of harmful effects caused by ant species in Hawaii (Pimentel 2002). However, the role of insects from other guilds, such as necrophagous species, as indicators of the conservation status of an island is practically unknown. At least 27 Diptera families have species that explore carrion as food or as a site for copulation or oviposition/larviposition, of which Calliphoridae and Sarcophagidae seem to be the most important (Byrd & Castner 2010).

Calliphoridae has 1522 species in 109 genera worldwide (Thompson 2013). So far, 99 species from 29 genera have been described in the Neotropical region (Kosmann et al 2013). Sarcophagidae is a diverse and widely distributed family, with 3073 described species in 355 genera in the world; the Neotropical region harbors 750 species (Thompson 2013). Some species are important for forensic entomology, providing information about post-mortem interval (PMI) and as indicators of site of death (Byrd & Castner 2010).

Some species of Calliphoridae and Sarcophagidae are associated with environments exposed to anthropogenic action. Such association, called synanthropy, varies with the species and the geographical and climatic characteristics of human concentrations (Linhares 1981). Montoya et al (2009) highlighted the preference of blow fly species, particularly of the genus Chrysomya, for anthropogenic environments. On the other hand, species from the genus Mesembrinella may be used as monitors of environmental quality, as they are primarily found in preserved forest environments (Cabrini et al 2013). Few studies have been conducted on the association of Sarcophagidae with preserved or anthropogenic environments (but see Yepes-Gaurisas et al 2013). In a recent study performed in the Amazon forest, a strong association of Sarcophagidae species with clearing areas suggested that they could be efficient indicators of deforestation (Sousa et al 2014).

Because carrion is an important ephemeral resource, combined information about assemblage composition and geographical distribution of necrophagous species will help in understanding processes of competition and biological invasion. In this scenario, field data can help to explain questions of endemism and monitoring of invasive species, which may subsidize programs for monitoring and conservation of threatened environments. Islands, particularly those under increasing rate of human presence, are sensitive to anthropogenic action, so that the detection of invasive species may clarify aspects of adaptation and competition.

The main objectives of this study were to (i) describe assemblages of necrophagous Diptera (Calliphoridae and Sarcophagidae) in two insular environments of different origins and located at different distances from mainland, (ii) investigate the effect of anthropogenic impact on the assemblage composition of carrion flies, (iii) to quantify the establishment of invasive species in the two systems (islands), and (iv) to infer about the conservation status of the islands based on the ecological parameters in each environment.

Material and Methods

Study sites and sampling design

The archipelago of Fernando de Noronha (3°51′13″S, 32°25′25″W; total area, 18.4 km2; permanent population, 2630) is located 345 km offshore from the Brazilian coast, and it consists of 20 islets of volcanic origin (Brasil 2012) (Fig 1). The archipelago is legally protected, and economic activities (agriculture, tourism, commerce, industry) are limited or prohibited (Freitas & Vasconcelos 2008). The climate is tropical with average annual precipitation of 1418 mm/year, mean temperature of 27.0°C, and relative humidity of 78%. The flora is poor, with mainly seasonal deciduous vegetation, and most plant species are exotic. The island has a chronic problem of water scarcity, as it has no permanent reservoirs of fresh water (Freitas & Vasconcelos 2008).

Fig 1
figure 1

Location of Fernando de Noronha (a) and Itamaracá (b) islands, Brazil, with indication of the sampling sites according to the level of human presence.

Full size image

Itamaracá Island (07°44′52″S, 34°49′33″W; total area, 65.1 km2; permanent population, 23,923) is located 1.5 km offshore from the Brazilian coast (IBGE 2013). It has an urban zone, with houses, services, stores, and surrounding areas of subsistence agriculture. Vegetation is mostly comprised of restinga evergreen forests and sparsely distributed mangrove areas. It has tropical rainforest climate, average annual temperature of 25.3°C, and average annual precipitation of 1867 mm (IBGE 2013). It is considered part of the Metropolitan Region of Recife, the capital of the state of Pernambuco.

To analyze the impact of anthropic action on the presence and spatial distribution of species of necrophagous Diptera on the islands, different environments located in areas under variable degrees of human influence were selected and categorized (Fig 2). The categorization was based mainly on four factors: (i) permanent population and access to visitors, (ii) presence of housing within 1 km radius from the sampling site, (iii) production of solid waste, and (iv) adjacent economic activity. After a comparative categorization in each island, areas under high anthropogenic impact (HAI), moderate anthropogenic impact (MAI), or low anthropogenic impact (LAI) were selected (Fig 1). Collection was authorized by the license number 32657-1.

Fig 2
figure 2

Brief description of the collection areas to Fernando de Noronha (top) and Itamaracá (bottom), according to the degree of human presence.

Full size image

Diptera sampling and identification

The insects were collected by using suspended traps containing chicken liver or sardine baits, ca. 200 g, previously exposed to 25°C for 24 h to achieve a preliminary degree of decomposition (Oliveira & Vasconcelos 2010). In each sampling area, a grid of six traps separated from each other by 20 m was assembled, containing three traps with chicken liver baits and three with sardine baits. Each grid was separated by a minimal distance of 500 m. Traps were exposed for 48 h. Sampling was performed in the rainy season and dry season in each island. Three sampling expeditions (replicate) were conducted in each island. D escriptive summary of the collection is as follows: (a) grids per island = 6; (b) traps per grid = 3; (c) types of baits = 2; (d) replicates = 3; and (e) season = 2, which makes up 216 independent collections per island. The insects were kept in 70% ethanol and identified using the taxonomical keys of Carvalho & Mello-Patiu (2008), Carvalho & Ribeiro (2000), Kosmann et al (2013), and Vairo (2011). Voucher specimens were deposited at the UFPE Collection (Curator Dr. L. Iannuzzi).

Data analysis

Species of Calliphoridae and Sarcophagidae were classified as constant (when present in over 50% of samples), accessory (25% to 50%), or accidental (present in <25% of samples), according to the equation: C = (p × 100) / N, where C = constancy (%), p = number of collections with the species, and N = total number of collections (Silveira Neto et al 1976). Dominance was calculated using the Simpson’s index (D), associated with a dominance ranking developed using a logarithmic transformation of species’ abundance in each environment.

Species richness in both islands was compared using Chao1 and Jacknife1 estimators using traps as sampling units in each island. Furthermore, a chi-square test was used to compare species richness according the gradient of anthropogenic impact and to evaluate the attractiveness of baits. Diversity was evaluated using the Shannon’s index (H′), and evenness through Pielou’s index (J). An analysis of variance (ANOVA) was conducted to evaluate the abundance distribution in relation to the gradient of anthropogenic impact. A multidimensional scaling (MDS) followed by ANOSIM was used to evaluate the structure of insect assemblages according to the degree of anthropogenic impact. BioEstat® 5.3 and Primer® 6.0 programs with α = 5% significance level were used in the statistical and ecological analyses.

Results

Overall diversity

When all 432 samples were combined, a total of 99,862 insects from 21 species of Calliphoridae and Sarcophagidae were collected in both islands. Only three species of Calliphoridae and three of Sarcophagidae were recorded in Fernando de Noronha, while in Itamaracá, eight species of Calliphoridae and 12 of Sarcophagidae were registered. Overall abundance was higher in the oceanic island (88,752 individuals) when compared to the continental island (χ 2 = 60.366; df = 1; p < 0.0001) (Table 1).

Table 1 Abundance (A) and relative frequency (RF) of Calliphoridae and Sarcophagidae species sampled in Fernando de Noronha and Itamaracá islands, Brazil.
Full size table

In both Fernando de Noronha and Itamaracá, the effort was sufficient to sample the number of species from both families. No difference in the predicted richness was observed for Chao1 and Jacknife1 estimators. In the oceanic island, the expected richness according to Chao1 and Jacknife1 estimators was four species of Calliphoridae and three of Sarcophagidae. In Itamaracá, eight and ten species were predicted for Calliphoridae and Sarcophagidae, respectively.

The type of bait did not influence the richness of Calliphoridae or Sarcophagidae species sampled in both islands. The abundance of Calliphoridae did not differ among baits in the oceanic island (χ 2 = 0.31; df = 1; p = 0.581); however, sardine attracted a higher number of individuals in the continental island (χ 2 = 7.54; df = 1; p < 0.05). For Sarcophagidae, sardine baits attracted more adults in both islands—Fernando de Noronha (χ 2 = 401.99; df = 1; p < 0.0001) and Itamaracá (χ 2 = 204.37; df = 1; p < 0.0001).

Ecological aspects and gradient of anthropogenic impact

No difference was observed in total richness of both families according to the gradient of anthropogenic impact, in both islands (Tables 2 and 3). Multivariate analysis of MDS revealed that the species do not follow patterns of distribution according to the impact factor, with the global R statistics close to zero in both islands. This absence of specificity in the distribution was validated by ANOSIM (Fig 3).

Table 2 Abundance (A), relative frequency (RF), and constancy category (C) of Calliphoridae and Sarcophagidae species under different anthropogenic impact degrees in Fernando de Noronha island, Brazil.
Full size table
Table 3 Abundance (A), relative frequency (RF), and constancy category (C) of Calliphoridae and Sarcophagidae species under different anthropogenic impact degrees in Itamaracá island, Brazil.
Full size table
Fig 3
figure 3

Multidimensional scaling plot based on Bray-Curtis similarities of the sampling over different environments in Fernando de Noronha (a) and Itamaracá (b) with their respect values of global R and statistical p by ANOSIM.

Full size image

In Itamaracá, Chloroprocta idioidea (Robineau-Desvoidy), Hemilucilia segmentaria (Fabricius), Hemilucilia semidiaphana (Rondani), Mesembrinella bicolor (Fabricius) (Calliphoridae), Oxysarcodexia culmiforceps Dodge, Oxysarcodexia amorosa (Schiner), P eckia (Euboettcheria) collusor (Curran & Walley), and Ravinia belforti (Prado & Fonseca) (Sarcophagidae) were registered only in environments under low anthropogenic impact (Table 3). Nevertheless, no significant difference was observed according to the gradient (χ 2 = 1.55; df = 2; p = 0.4607). No significant difference was observed in insect abundance according to the degree of anthropogenic impact for both Calliphoridae (F 5; 18 = 0.311; p = 0.899) and Sarcophagidae species (F 5; 18 = 0.253; p = 0.931) in Fernando de Noronha and Itamaracá (Calliphoridae: F 5; 42 = 0.975; p = 0.554; Sarcophagidae: F 5; 54 = 1.067; p = 0.389).

In Fernando de Noronha, both Chrysomya megacephala (Fabricius) and Cochliomyia macellaria (Fabricius) were classified as constant in all sampling areas, regardless of the anthropogenic impact (Table 2). Among the Sarcophagidae species, only P. chrysostoma was classified as constant while the others were defined as accidental species. In the continental island, C. megacephala was constant in the three types of environment; Chrysomya albiceps (Wiedemann) was constant only in the environment under low anthropogenic impact, while the other species were accidental or accessory (Table 3). No Sarcophagidae species was constant in Itamaracá.

Chrysomya megacephala was classified as dominant in all environments in the oceanic island, irrespective of the anthropogenic impact: D LAI = 0.564; D MAI = 0.541; and D HAI = 0.526 (Fig 4a). Peckia chrysostoma was also dominant, regardless of the anthropogenic impact (D LAI = 0.861; D MAI = 0.888; D HAI = 0.972) (Fig 4b). In Itamaracá, C. megacephala and C. albiceps were dominant in all environments (D LAI = 0.741; D MAI = 0.773; D HAI = 0.799) (Fig 4c), while no Sarcophagidae species was dominant (D LAI = 0.129; D MAI = 0.205; D HAI = 0.205) (Fig 4d).

Fig 4
figure 4

Dominance ranking of Calliphoridae (a, c) and Sarcophagidae (b, d) species in Fernando de Noronha (top) and Itamaracá (bottom) islands, according to different anthropogenic impact degrees. O.amo: Oxysarcodexia amorosa; O.cul: Oxysarcodexia culmiforceps; O.int: Oxysarcodexia intona; O.tim: Oxysarcodexia timida; P.ang: Peckia (E) anguilla; P.chr: Peckia (P) chrysostoma; T.occ: Tricharaea (S) occidua.

Full size image

The diversity indices for Calliphoridae and Sarcophagidae in Fernando de Noronha were H′ = 0.679 and H′ = 0.226, respectively, and in Itamaracá, H′ = 0.477 (Calliphoridae) and H′ = 2.035 (Sarcophagidae). When analyzed individually by the degree of anthropogenic impact, diversity was also low in both islands (Table 4). The Pielou’s evenness index (J) for families Calliphoridae and Sarcophagidae in Fernando de Noronha were J = 0.618 and J = 0.206, respectively, and in Itamaracá, J = 0.229 (Calliphoridae) and J = 0.819 (Sarcophagidae). The evenness indices were high for Sarcophagidae in Itamaracá, regardless of the degree anthropogenic impact in the environment (Table 4).

Table 4 Ecological indicators: Shannon diversity (H′) and Pielou evenness (J) of Calliphoridae and Sarcophagidae under different anthropogenic impact degrees in Fernando de Noronha and Itamaracá islands.
Full size table

Discussion

Overall diversity and gradient of anthropogenic impact

The high abundance of Calliphoridae species in both islands is largely due to the reproductive characteristics of this family, typically r-strategist of high fecundity (Goodbrod & Goff 1990, Gabre et al 2005) and with fast larval development (Aguiar-Coelho & Milward-de-Azevedo 1995). On the other hand, Sarcophagidae species produce fewer descendants and have longer life cycles so that a pattern of high diversity associated with low abundance is found in studies conducted in distinct environments, like forests, urban areas, and agroecosystems (Denno & Cothran 1976, Vasconcelos & Araújo 2012).

The richness of Calliphoridae (three species) in the oceanic island was low when compared to other studies performed in islands worldwide (Early & Goff 1986, Tantawi & Sinclair 2013), probably because of its small portion of emerged land and considerable distance from the coast. The continental island Itamaracá, with a total richness of 20 species, has a greater variety of microhabitats and a high flow between the continent and the island. Additionally, effects like habitat fragmentation and landscape alteration substantially affect richness, abundance, and species composition (Debinski & Holt 2000, Heleno et al 2009). Furthermore, the effect of scale is an important factor on blow fly ecology because the distribution of abundance and species composition depend directly on the potential use of landscape and capacity of dispersion in 1 day by each species (Zabala et al 2014).

The effectiveness of decomposing animal tissues as attractants to necrophagous flies has been documented elsewhere (Vasconcelos et al 2015, Oliveira et al 2016). In this study, both sardine and chicken liver attracted a high richness of dipterans, with a slightly higher effectiveness of sardine, probably due to the stronger putrid odor released by decomposing fish.

A practical step towards the use of flesh flies and blow flies as tools for monitoring conservation in South America was provided by Sousa et al (2014), who suggest that sarcophagids are not suitable for the detection of minor variations in forest cover. In contrast, some blow fly species, especially of the genus Mesembrinella, prefer forested areas under limited human interference (Gadelha et al 2009, Cabrini et al 2013), so that their presence in Itamaracá suggests that parts of its territory are still under a certain degree of conservation. However, it should be noted that the species classified as constant in this environment, like C. macellaria, are considered by many authors as synanthropic (Baumgartner & Greenberg 1985, Montoya et al 2009).

In Fernando de Noronha, the distribution pattern of Calliphoridae revealed habitat sharing among the species regardless of the degree of anthropogenic impact. Several authors disagree about the synanthropic degree of C. macellaria, which was previously classified as constant not only in typically urban (Montoya et al 2009) but also in preserved environments (Vasconcelos & Salgado 2014). Chrysomya megacephala has the status of a synanthropic species (Linhares 1981), but it also thrives in preserved environments (Vasconcelos et al 2015). This environmental plasticity results in a niche overlap among several Neotropical species (D’Almeida & Almeida 1998).

The classification of synanthropy usually takes into consideration the presence and the relative frequency of a given species in an environment but leaves out the effect of scale. When the composition of species is correlated with the use of landscape incorporating different scales, the degree of association with human-modified environments can differ, as overlap patterns are more commonly detected under lower scale refinements of the area (Zabala et al 2014). This may explain the co-dominance of C. megacephala and C. macellaria in the limited area and isolated nature of Fernando de Noronha, despite its conservation status.

Among Sarcophagidae species, Peckia (P.) chrysostoma and Oxysarcodexia intona (Curran & Walley 1934) stood out in terms of abundance in both islands. These genera are the most abundant and diverse in the Neotropical region (Pape 1996). Flesh flies are typically larviparous (Carvalho et al 2012), and although this strategy may ensure a pioneering role in resource colonization, it yields low abundance when compared to Calliphoridae species. The investment of more resources and time in the development of immature individuals in this case means fewer descendants (Denno & Cothran 1976).

Yet, empirical data about the synanthropic degree of Sarcophagidae species are still scarce. Yepes-Gaurisas et al (2013) described the association of Peckia and Oxysarcodexia species with environments modified by human action in Colombia. Peckia chrysostoma is commonly found in urban environments and practically absent in preserved forests, a fact confirmed by its occurrence in urban environments in northeastern (Oliveira & Vasconcelos 2010, Vasconcelos & Salgado 2014) and southeastern (Linhares 1981, D’Almeida 1984) Brazil. However, the synanthropy status of necrophagous species seems to be based on an insufficient number of assemblage parameters, without taking into account factors like constancy, dominance, and niche overlap.

Aspects of conservation and bioinvasion

Chrysomya megacephala, C. albiceps, and C. putoria (Wiedemann) were introduced in the 1970s in southern Brazil (Guimarães et al 1978). We provide here further evidence of the establishment of Chrysomya species in insular environments, as reported originally by Carmo & Vasconcelos (2014), and register for the first time the presence of three species in a Southern Atlantic continental island (Itamaracá). The dominance of C. megacephala in both islands substantiates its ability to adapt to different environments irrespective of the degree of human presence.

Invasive species may cause significant harmful effects on local community dynamics (Kenis et al 2009). So far, it is impossible to prove that the establishment of C. megacephala caused extinction of other necrophagous species. The only reference to Diptera diversity in Fernando de Noronha was based on samples collected in the 1970s, when only two calliphorid species—C. macellaria and Lucilia eximia (Wiedemann)—were reported (Couri et al 2008).

The increased traffic between Fernando de Noronha and the Brazilian continent and other countries in the last decades has considerably increased the possibility of accidental introduction of exotic species in the archipelago. A similar situation takes place in Itamaracá, especially due to the easy access and short distance from the coast and deficient inspection of traded foods and living organisms. However, due to their small size and transportation as immature or adults, flies are hard to detect and refrain. In Fernando de Noronha, the strong dependence on human transport combined with a high degree of environmental plasticity is a requirement fulfilled by C. megacephala.

The displacement of C. macellaria populations, in the field or in experimental arenas, as a result of the introduction of C. albiceps and Chrysomya rufifacies (Macquart), was described by Wells & Greenberg (1992) and Faria & Godoy (2001). From that perspective, it is possible that the high abundance and constancy of C. albiceps, combined with its co-dominant status, may lead to displacement of species directly affected by C. albiceps in the Atlantic islands studied here.

This is, to our knowledge, the first study to investigate the pattern of assemblage of necrophagous insects in Southern Atlantic islands, incorporating the effect of human presence. This study shows that assemblages of necrophagous Diptera in remote oceanic islands can vary dramatically in a short period (less than 40 years). Surprisingly, the concept of habitat conservation using entomological evidence seems to have low priority in long-term ecological studies. Some groups of arthropods are often the first to become extinct as a result of loss of habitat or disturbance (Cardoso et al 2010). In this scenario, we propose the incorporation of data on necrophagous Diptera in the development of models to assess the conservation status of an environment.