Frugivorous butterflies (Lepidoptera: Nymphalidae) as a habitat quality indicator in Cerrado urban fragment

Environmental monitoring programs use recurrently insects to assess the quality of habitats, mainly frugivorous butterflies. These have ample availability of information, easy and lowcost method, in addition to responding easily to environmental changes, making an efficient tool in environmental diagnosis. The objective of this work was inventory the frugivorous butterflies in an urban Cerrado fragment belonging to the Universidade Federal de Rondonópolis, state of Mato Grosso, and verify the efficient as biological indicator of environmental quality. We hypothesize differences in species richness, abundance and composition between edge and center of fragment. The sampling was carried out between February and May 2019, using 10 Van Someren-Rydon traps, with six has distributed at the edges and four in the center of the fragment. The collections have taken once a week lasting three days at each event. A total of 105 individuals belonging to nine species of Nymphalidae were captured, respectively, as subfamilies: Biblidinae (71), Satyrinae (31) and Charaxinae (03). There was no significant difference between species richness, abundance and composition in the edge area and in the center. The predominance of taxa such as Hamadryas feronia Hübner, Hamadryas februa Linnaeus and Paryththimades phronius Godart evidences that fragment is in regeneration process, once the species are characteristics of regenerating environments. Therefore, the present work collaborates and demonstrates the efficiency of the use of frugivorous butterflies as indicators of habitat quality and the contribution for implantation and monitorng of protected areas, serving as parameters for future analyzes of the quality of the studied area.

The Brazilian Cerrado contributes about 5% of the diversity of fauna and flora worldwide and about 1/3 of the Brazilian biota (alho & martinS 1995). For this reason, it is considered an important hotsposts due to its function of richness at the endemic level and degree of threat for the species. Therefore, when conducting the study on the presence of frugivorous butterflies, it is possible to obtain information on the quality of the habitat to consequently contribute to the elaboration of mitigation measures. Thus, the objective of this work was inventory frugivorous butterflies using them as environmental indicator in an urban fragment of the Brazilian Cerrado. In addition, the hypotheses for comparing species richness, abundance of individuals and community composition between the edge area and the central region of the fragment were tested, we expecting that there will be a difference between the two regions of the study area of the metrics performed by checking the fragment quality for conservation implications.

MATERIAL AND METHODS
Study area. The study area is located at the Universidade Federal de Rondonópolis, Rondonópolis, MT (16º 27' 23.5' 'S;54º 36' 35.8' 'W). The municipality has an urban area with approximately 232,500 inhabitants (iBge 2018). The study site is a fragment with influence of anthropic actions, since in the south of the state of Mato Grosso due to the favorable relief to agricultural practice and an accelerated agriculture and livestock expansion. Consequently, the remaining vegetation is fragmented and isolated, being currently, representative of the regional native biodiversity characterized by several typical Cerrado formations (FelFili & Silva Júnior 2001).
The region's predominant climate corresponds to the hot and sub-humid tropical type and its vegetation belongs to the Cerrado domain (Sette 1996), with two climatic patterns: the dry season (May to September) and the rainy season (October to April). The average rainfall index in Rondonópolis is 1,500 mm/year with the concentration of rainfall in the summer (Sette & tariFa 2001). The campus covers approximately 60 ha, part of which is urbanized and paved by university buildings. The sampled points are located in an area of native vegetation of the Cerrado with 17.35 ha. The vegetation is classified as Cerrado sensu stricto located on a red-yellow latosol. It is an area in a process of natural regeneration since 1980, the result of a cattle farm donation for university implantation which was surrounded for the installation of the campus, and which is currently decreasing due to new buildings in this area (De CampoS et al. 2018).
Sampling. Sampling was carried out from February to May 2019 (SISBIO Authorization 61938-1) during the wet season, as it is known that the richness and abundance of frugivorous butterflies tends to increase in the wet season due to the greater availability of food resources (DevrieS et al. 1997). The traps were arranged in two distinct areas in the fragment: central region and edge area (Figure 1). Distances from the internal points of approximately 120 m between each were considered. The distances between the internal points and the edge region were around 150 m. Thus, we believe that the distance established between the central region and the edge region is sufficient to ensure that if there is a difference in the composition of the community in relation to the preference for the use of the specific habitat, individuals would not move between regions. We consider the approximate values found by BoSSart & opuni-Frimpong (2009)  We use Van Someren-Rydon attractive traps for frugivorous butterflies, which consisted of 90 cm high thin-screen cylinders, closed at the top end, in which at the top there is a metal frame, and at the bottom a 3 cm opening for the entrance of the butterflies (DevrieS 1987). In this opening, a plastic dish with fermented fruit baits (bananas and sugar) which were prepared 48h before the start of sampling. This fermentation process is responsible for attracting frugivorous lepidopteran into the trap through the odor (FreitaS et al. 2014). The trap remained suspended from the ground, hanging from the low branches of the trees with the help of a rope. Along the study area, 10 traps were used, of which 6 were located on the edge of the fragment, and the other four were located inside it ( Figure 1).
Thus, sampling was carried out weekly, with an interval of three days (72 h specimens over 20 weeks, totaling 14,400 h/trap. The captured specimens were placed in entomological envelopes enumerated according to the collected trap and location, taken to the entomology laboratory at UFR for the material classification process, assembly on an entomological pin and subsequent identification. The identification was carried out through the butterfly monitoring and biodiversity catalog for the Cerrado (SantoS et al. 2014).
Data analysis. The data were analyzed through the description of the community, using the species richness and abundance. To compare these parameters between the center and the fragment's edge, the Mann-Whitney test was used; to compare the composition of the community the Nonparametric Multivariate Analysis test (NPMANOVA) was used; and Non-parametric Multidimensional Scaling (NMDS) for data visualization. The similarity index (SIMPER) was used to define the species responsible for the similarity percentage of the community of the captured frugivorous butterflies, in order to observe which species were more important in the different study areas. For all tests, p 0.05 was assigned, and performed using the Past® program.

RESULTS AND DISCUSSION
A total of 105 individuals were registered, distributed in nine species belonging to the three subfamilies of Nymphalidae: Biblidinae with 71 individuals, Satyrinae with 31 and Charaxinae with 3 individuals (Table 1) Low richness was to be expected since, the closer to dense urban area, the smaller the number of species found in forest fragments, due to the disappearance of species sensitive to urbanization (SantoS 2016). However, when compared to works carried out on urbanized fragments such as those recorded by Silva et al. (2012)  Comparing the regions of the edge and central area, we found that there is no significant difference between them in relation to species richness (U = 40, p = 0.47) and abundance of individuals (U = 41.5, p = 0.54 ) (Figure 2A and 2B), the same occurs with the composition, which is similar (NPMANOVA F = 1.71; p = 0.1; NMDS, Stress = 0.61) ( Figure  3), with the same species inhabit the the edge are also found at interior of the fragment. The SIMPER test showed and reinforced the similarity of 77% of the sampled composition between the different points of the center and the edge of the fragment represented by similar taxa Hamadryas feronia, Linnaeus, Hamadryas februa, Huebner and Paryphthimoides phronius, Godartthat are responsible for more than 57% similarity (Table 2). Comparing the communities at the edge and center of the fragment separately, it is clear that there is a great similarity between them with regard to composition, richness and abundance of species. These results suggest a homogeneous fauna, which may be the result of an extreme degree of environmental disturbance (mCkinney & loCkwooD 1999). The absence of difference between the edge and the center can be justified by the high level of alteration and the very small size of the fragment, since its size directly influences the ecological processes, due to the changes induced by the formation of the edge, that is, small fragments have a higher proportion of altered environment, due to the extension of the edge to the interior (ShahaBuDDin & terBorgh 1996;ShahaBuDDin & ponte 2005;FeiDen et al. 2008;SChmiDt et al. 2012).
It is notable that the presence and abundance of certain species such as H. feronia, H. februa e P. phronius, known as indicators of impacted environments (Silva et al. 2007Brown 1992, it shows that the anthropized urban fragment, resulting mainly from agricultural activities, is still a regenerating environment. This habitat is not yet complex, that is, a large edge, since these species preferentially inhabit the edge of forests, open environments and with the presence of clearings, where the food resource, probably in its larval stage, is more abundant in vines (Euphorbiaceae) in case of H. feronia and H. februa and monocotyledons in the case of butterflies of the subfamily Satyrinae, typical of open-field habitats, resulting in the high abundance of individuals of these species in this environment (gomeS Filho 2003). This demonstrates that the anthropic disturbance in the fragment favored organisms more tolerant to the sun and desiccation, typical of these environments. In addition to the majority of species, including the most abundant of Biblidinae as Hamadryas spp., have caterpillars that feed on plants typical of early stages of succession, for using during their development host plants present in early stages of regeneration (Silva et al. 2007). It is known that some individuals of the Nymphalidae family are common in edges of forests and secondary forests, because they prefer to fly in the sun (paim 1995; ramoS 2000).
Studies have demonstrated that fragmentation can affect the richness, diversity and composition of the frugivorous butterfly community (e.g. horner- Devine et al. 2003;ShahaBuDDin & ponte 2005) and the abundance and distribution of populations in relation to their viable size for long-term stay (hanSki et al. 1996). Highlighting that certain components present in the fragment, such as the degree of isolation and size directly affect the assembly of fruit-feeding butterflies and different biological groups may respond in different ways with respect to fragmentation (Caitano et al 2020). Furthermore, it should be noted that the characteristics of the vegetation and the permeability of the matrix are factors which are correlated with the distribution of frugivorous butterflies (e.g. ShahaBuDDin & terBorgh 1999;ramoS 2000;uehara-praDo et al. 2005). In general, urbanization has negative effects on species richness and abundance of frugivorous butterfly individuals, often being able to cause local extinctions of certain species more sensitive to drastic changes caused by the conversion of native areas (ramirez-reStepo & maCgregor-ForS 2016).
Many groups of insects can assist in the definition of small areas and fragmented habitats with a long history of anthropic influence (Cullen et al. 2003). In this way, the preservation of urban green areas can contribute to the conservation of butterfly fauna. These environments can sustain a high diversity of species, by providing different resources and a less disturbed location in cities, promoting the improvement of the urban ecosystem. (Fortunato & ruSzCzyk 1997). Distances greater than 150-200 m show distinct community structures of frugivorous butterflies from the edge in fragments of dry forests in the Afrotropical region (BoSSat & opuni-Frimpong 2009). Considering that the average distance from the edges to the central area of the studied fragment is in the range of 150-200 m we can see that for the locality, the fragment can still be considered as a large edge region, as it is an area in natural regeneration coming from regeneration in the last 30 years since its implantation, being surrounded by urban and pasture areas. In this way, we see that because it is a small Cerrado area, fauna regeneration processes can be more sensitive and time consuming because it does not guarantee the minimum functionality of the ecosystem to ensure the presence of species considered typical of preserved/ conserved areas.
Therefore, it is concluded that the urbanization process directly influences the number and presence of butterfly species, highlighting the potential that they have as biological  indicators in the study area. It is worth mentioning that due to the fragment, it is still in a state of regeneration and has been impacted by anthropic activities. We must continue with efficient tools in the process of diagnosis and environmental monitoring the area, thus contributing to the adoption of mitigating actions and measures responsible for ensuring the maintenance of habitat and local biodiversity. Siderone galanthis Hubner 4,24 100 Table 2. Percentage of similarity with individual and cumulative importance of each species between edge and center, indicating which species were more important in the similarity between areas.