Diversity of butterflies (Lepidoptera: Rhopalocera) in Bhubaneswar, Odisha, India

The loss and fragmentation of habitat caused by rapid urbanization can have devastating effects, both at regional and global level. In this study, butterfly species diversity has been assessed in Bhubaneswar, India, as a model geographical region for understanding the biology of the local population and its dynamics. In total 107 butterfly species have been documented, with the highest number of species being recorded from the family Nymphalidae (31.77%), followed by Lycaenidae (25.23%), Hesperiidae (23.36%), Pieridae (11.21%) and Papilionidae (8.41%). Out of these, 17 species are new reports for the city and nine species are legally protected in India under the Wildlife (Protection) Act, 1972. Sørensen’s diversity index and one-way ANOVA have been used to establish the relation between species diversity and habitat. The present investigation provides baseline data for future research and conservation of species in places like the model city, which face rapid urbanization.

Therefore, butterflies are regarded as good indicators of the quality of microhabitat and the extent of associated anthropogenic disturbances (Ruszczyk and Silva, 1997;Kehimkar, 2016). Approximately, 18,768 species of butterflies have been recorded worldwide (Van Nieukerken et al., 2011) and recent findings suggest that India hosts 1318 species in its subcontinent, out of which 89 species belong to Papilionidae, 277 species belong to Hesperiidae, 92 species belong to Pieridae, 19 species belong to Riodinidae, 380 species belong to Though literature related to the butterfly diversity and abundance is available for the state (Mishra et al., 2010;Mohapatra et al., 2013;Priyamvada and Mohapatra, 2016;Boruah et al., 2018;Payra et al., 2019;Singh et al., 2020), the species diversity and conservation in relation to their habitats in the rapidly urbanizing city of Bhubaneswar is yet to be explored. As part of long-term conservation strategies, local population dynamics as well as species diversity become extremely important since habitat fragmentation and loss of microhabitat pose threats to population persistence (Thomas et al., 1992;Roy et al., 2010;Fernandez-Chacon et al., 2014;MacDonald et al., 2018). In the past decade, topology and climate of the city have changed significantly due to increased urbanization, decreased vegetation, and a rise in temperature owing to the interplay of biotic and abiotic factors (Swain et al., 2016;Gogoi et al., 2019). The study aims to document the butterfly species diversity in relation to their habitats such as open scrub, garden habitat, urban habitat, urban park, urban forest, fragmented forest, and cropland in the city and its outskirts. The present study documented a list of 107 butterfly species out of which 17 species are new reports for this region and nine species are enlisted under the Wildlife (Protection) Act, 1972(WPA, 1972.

Study site
The study was conducted in Bhubaneswar and its outskirts in Khordha district of Odisha, India, covering approximately 440 km . The study range was divided into 11 different study sites and maps indicating the same were created using ArcGIS software (version 10.3) and Google Earth ( Figure 1). The detailed information about Global Positioning System (GPS) coordinates, elevation, and habitats for each study sites are provided (Table 1). The study range is predominated by seven types of habitats ( Figure 2).

Survey and monitoring methods
A survey and documentation of the study sites were conducted from July 2018 to August 2020. For data collection, opportunistic survey and random sightings were used along with modified Pollard walk method (Pollard and Yates, 1992;Royer et al., 1998;Wood and Gillman, 1998;Pellet et al., 2012). All observations were carried out between 6:30 hrs to 11:30 hrs and 14:30 hrs to 17:30 hrs in good weathered conditions. During the study tenure, each study site was visited at least two times in each quarter of a year. Information about the coordinates and elevation of study sites were obtained by using Google's location services. Butterflies were photographed using Nikon gears (D3500 and D5300 DSLR cameras mounted with 18-55 mm and 70-300 mm Nikkor lenses), identified using field guides, books and the butterflies of India website (Kunte, 2000;Kehimkar, 2016;Smetacek, 2018; Butterflies of India, 2021: https://www.ifoundbutterflies.org/). The WPA, Protected Areas (Scheduled Insect species, 2014: http://wiienvis.nic.in/). An entomological net was occasionally used and the butterflies were released unharmed to their natural habitat as soon as they were photographed. In this study none of the butterfly species were collected, euthanized or killed by any means.
The host plants were identified and recorded as per the literature available (Karmakar et al., 2018;Nitin et al., 2018).  (3) Khordha district indicating the study sites of Bhubaneswar and its outskirts; (4) Google map indicating the topology of the study range with marked study sites S1-S11 are the study sites

Statistical analysis
Sørensen's method was applied to find the similarity between number of species in different study sites and one-way Analysis of Variance (ANOVA) was used to analyse the dependence of relative distribution of species on habitat.

Sørensen's similarity index (β)
The Sørensen index is an indicator of similarity between two communities based on the number of common species shared by them. β can be defined as: β = 2c / (s1 + s2) where, s1 = total number of species recorded in the first community, s2 = number of species recorded in the second community, and c = number of species common to both communities.
The value of β lies between 0 and 1, wherein as the value approaches 0, the species overlap between the communities decrease and as the value moves towards 1, both the communities start having a greater number of shared species (Christopher, 2020). All the calculations were performed using Microsoft Excel (version 2019).
One-way ANOVA It is used to determine the differences in the means of three or more independent groups. One-way ANOVA has been used in this study to establish a substantial relation between the variation in habitat and the species diversity by considering a null hypothesis (H0) that the mean of family wise species diversity of each site is the same. The alternative hypothesis (H1) is that at least one of the means is different. A significance level (α) of 0.05 implies a 5% risk of concluding that a difference exists when there is no actual difference. The variance ratio, F statistic (F stat), probability value (P-value), and critical value of F distribution (F crit) in the ANOVA table serve as the base to conclude the analysis. If P < 0.05 and the value of F stat is more than F crit, then the null hypothesis is rejected whereas if P > 0.05 and F stat is less than F crit, then the null hypothesis is accepted (Kim, 2017). In this study, one-way ANOVA was calculated using GraphPad Prism Software (version 5.0) and Microsoft Excel (version 2019).

Results Results Results
A total of 107 species of butterflies, belonging to 76 genera and five families (Papilionidae, Hesperiidae, Pieridae, Lycaenidae, and Nymphalidae) were recorded in this study whereas no butterflies from the families Hedylidae and Riodinidae were encountered during this study (Table 2). Information about host plants of the recorded butterfly species were enlisted (Table 3). Butterfly hierarchy was represented showing the number of species observed from each family ( Figure 3). Photographs of each species categorised into their respective families were represented (Figure 4-8). Out of the total number of species recorded, nine species are legally protected in India under WPA, 1972, and 17 species are new reports for the city (Table 3). The highest number of species have been observed in Nymphalidae (34 species, 31.77%), followed by Lycaenidae (27 species, 25.23%), Hesperiidae (25 species, 23.36%), Pieridae (12 species, 11.21%), and Papilionidae (nine species, 8.41%) as illustrated ( Figure 9A). The species to genus ratio (S/G) determines distribution of species among genera, and is calculated to be 3.000, 1.316, 1.500, 1.125, and 1.545 for Papilionidae, Hesperiidae, Pieridae, Lycaenidae, and Nymphalidae respectively ( Figure 9B, Table 4). The family wise species distribution in different study sites showed that the highest number is observed in S8 (88 species) followed by S7 (77 species), S9 (74 species), S5 (63 species), S6 (48 species), S11 (43 species), S10 (42 species), S1 (36 species), S3 (34 species), S2 (31 species), and the lowest number is observed in S4 (27 species) as illustrated ( Figure 9C). The distribution range of species number is found to be more scattered for Hesperiidae, Lycaenidae, and Nymphalidae as compared to Papilionidae and Pieridae ( Figure 9D). According to the Sørensen's similarity index, a maximum value of 0.896 is observed between study sites S1 and S2 whereas a minimum value of 0.470 is observed between study sites S4 and S8 (Table 5). One-way ANOVA data shows that F stat value is 13.92, value of F crit is 2.55, and P < 0.05 for butterfly species in different families within the study range. Between the butterfly species among the study sites the result shows that F stat value is 2.34, value of F crit is 2.05, and P < 0.05. In both the cases P < 0.05 and the value of F stat is greater than the F crit value (Table 6).      Table 5. Table 5. Table 5. Table 5. Sørensen's similarity index for butterfly species recorded in the study sites Study Study Study Study site site site site S1 S1 S1 S1 S9 S9 S9 S10 S10 S10 S10 S11 S11 S11 S11 S1 Similarity index value marked (*) is highest and marked (^) is the lowest. S1-S11, study sites.

Discussion Discussion Discussion
Out of the total 107 species documented, three species (Papilio clytia, Pachliopta hector and Hypolimnas misippus) are listed in Schedule I, four species (Euchrysops cnejus, Lampides boeticus, Anthene lycaenina and Mahathala ameria) in Schedule II and two species (Hasora vitta and Hyarotis adrastus) in Schedule IV of WPA, 1972. The study shows highest S/G ratio for Papilionidae, probably due to their less genetic diversity as compared to other four families. Highest species diversity is observed in the study sites (S8, S7, and S9 respectively), which are mainly dominated by fragmented forest, open scrubs, and croplands having many host plants to sustain their lifecycle. Least number of butterfly species was recorded in the study sites S2 and S4 comprising urban habitat with continuous anthropogenic activities and less availability of host plants. 17 species are new additions to the existing butterfly data of Bhubaneswar. In this study, Painted Lady (Vanessa cardui) is reported only once in the month of April 2019, which could be a result of its well-known long-range migration (Stefanescu et al., 2016). The cluster is more scattered in case of Hesperiidae, Lycaenidae, and Nymphalidae as these families exhibit a greater number of species as compared to Papilionidae and Pieridae. Sørensen's similarity index showed the highest value between the study sites S1 and S2 indicating the highest species overlap, perhaps because they are in close proximity and share a similar type of habitat. Least Sørensen's similarity index has been recorded between the sites S4 and S8 which might be due to the presence of contrasting habitats. One-way ANOVA data suggest there is a significant variation between the butterfly species of different families within the study range. Moreover, a significant variation is observed between the butterfly species among the study sites. The ANOVA results show that the P < 0.05 and the value of F statistic is greater than the critical value of F distribution, thereby rejecting the null hypothesis. Butterflies are an integral part of our ecology and play an important role in maintaining the ecological balance. The present documentation provides a reference point and opens new ways of analytical research scopes. Researchers would find better sustainable approaches towards global conservation by understanding the biology of local populations and their dynamics pertaining to rapidly urbanizing geographical ranges. Studying species diversity has become more important in today's world as it serves as a checkpoint for awareness and understanding of interspecific and intraspecific interactions.

Conclusions Conclusions Conclusions Conclusions
The study range supports a rich diversity of butterflies with a wide variety of plants which provide them an ideal breeding habitat. An attempt is made in this study to show the importance of a local area as a model geographical region with diverse habitats, suggesting the importance of local population in long term biodiversity studies and conservation. Therefore, it is imperative to understand the relative dependence of the butterfly species on their habitat. Considering such correlations would help in putting a check on the decreasing number of butterflies due to rapid urbanization and habitat destruction. Further studies on local butterflies of this region, covering various other aspects would undoubtedly contribute a lot towards solving the global issue of conservation of nature and its depleting species.

Authors' Contributio Authors' Contributio Authors' Contributio Authors' Contributions ns ns ns
The manuscript is conceptualized, drafted and supervised by SKS. AS and NP helped equally in data curation and all the authors analysed the data. All the authors contributed equally for the field work while review and editing were done by SKS. All the authors read and approved the final manuscript.
Ethical approval Ethical approval Ethical approval Ethical approval (for researches involving animals or humans) In this study none of the butterfly species were collected, euthanized or killed by any means. Thus, the images of butterflies represented in the figures are the result of live photography. An entomological net was occasionally used and the butterflies were released unharmed to their natural habitat as soon as they were photographed.