Community structure of Coleoptera in Bethuadahari Wildlife Sanctuary , West Bengal , India

We focused on the coleopteran species assemblage in a tropical deciduous forest in the Bethuadahari Wildlife Sanctuary, West Bengal, India. During a 2-year survey, we collected 56 species belonging to 13 families of Coleoptera, in varying relative abundance. Among the species, 15 belong to the family Chrysomelidae, nine to the Staphyllinidae, and four to the Coccinellidae. Our results substantiate the importance of the Bethuadahari Wildlife Sanctuary for the conservation of coleopteran insects.


INTRODUCTION
Insects are distributed worldwide in all habitats and constitute about half of the global diversity of animals (Gullan & Cranston 1994, Gillott 2005).A majority of insect species are beetles, grouped under the order Coleoptera.Of coleopterans, there are about 15,500 described species (nearly 4.5% of the world fauna) recorded from India to date, being with 3,100 endemic for the country (Ramkrishna & Alfred 2007).Ecosystems are dominated by beetle species with diverse morphologies that correspond to their functional roles within the ecosystem.In almost all habitat types, the richness and abundance of beetles are being explored worldwide, including beetle assemblages associated with agroecosystems (Bambaradeniya et al. 2004, Petrova et al. 2006, Kumar et al. 2009), as well as forests and protected areas (Halme & Niemelä 1993, Chung et al. 2000, Lucky et al. 2002, Apigian et al. 2006, Joshi et al. 2008, Grimbacher & Stork 2009, Sabu et al. 2011, Chandra & Gupta 2012, Thakare & Zade 2012, Campos & Hernández 2013).By assessment of beetle species assemblages, the functional role of beetles in their respective ecosystems can be better understood.
Tropical forests, with ample resources and hospitable habitats, provide an ideal ecosystem for coleopterans (Basset et al. 2003).In comparison to the similar ecosystems, the functions of the forest ecosystems are in part dependent on the foraging (Mattson & Addy 1975, Brown 1985, Schowalter 1995) and litter processing (Pfeiffer 1996, Nair 2007) by the beetles and other insects.The beetles represent the single major group of insects that explore both live and dead plant parts thereby accounting for the dynamic nature of the flow of energy and matter in the ecosystem concerned (Losey & Vaughan 2006, 2008).Appraisal of the beetle assemblages in forest ecosystems is being emphasized in several studies spanning different geographical areas and suggests their dominance over other insect orders (Schowalter et al. 1986, Novotny & Basset 2005).Our study is an account of the diversity of Coleoptera in the tropical moist deciduous forest of Bethuadahari Wildlife Sanctuary, West Bengal, India, and highlights the need for conservation planning and management in the reserve.Keeping in view the significance of the beetles in forest ecosystems, the information of the present study may be utilized for their conservation and maintaining the forest as a reserve area.

Study area
The 66.77 ha Bethuadahari Wildlife Sanctuary (BWS; 23°35ʹ N, 088°23ʹ E; 5 m above sea level) is located in the Nadia district of West Bengal, India (Fig. 1).The wildlife sanctuary now supports a tropical moist deciduous forest, but previously it was degraded and planted forest.The wildlife sanctuary was declared as such on 19 August 1998, under the Notification No. of 2772 of the Department of Forest, Government of West Bengal, India.

Data collection and analysis
To collect coleopterans, we employed the quadrat method (Brower et al. 1998, Krebs 1999), following the random selection of five survey plots in various parts of the forest.Each plot constituting an area of 625 m 2 (25 × 25 m 2 ) which were separated one from another by at least a distance of 200 m.The plots consisted of a heterogeneous mixture of grass and barren soil as well as shrubs and trees.Within each plot, three quadrats, each 5 × 5 m 2 were randomly selected, in compliance with the norms of interspersion and randomization such that the samples represent true replicates (Hurlbert 1984).
In each quadrat, we used sweep netting, pitfall trapping, fluorescent light trapping, and hand picking with a net to capture beetles.Different collection methods were employed to ensure that most species were captured.The sweep net used was 30 cm in diameter, 90 cm deep, and with 1.5 mm mesh.Each sweep represented a horizontal swing 0.5 to 2 m above the ground and with an arc between 135° to 180° (against the horizontal plane).At each quadrat, 100 random sweeps were made.One pitfall trap was placed at the center of each quadrat and left for 24 h.These were transparent plastic jars (10 × 10 cm), ⅓ filled with a 2:1 solution of propylene glycol in water (Halme & Niemelä 1993, Taboada et al. 2010, Earnst & Buddle 2013).A light trap was set up at the center of each plot for 10 h (18:00-4:00 h) (Lee et al. 2008).Therefore, beetles collected from each plot comprised of cumulative collections by 3 light trap, 3 pitfall traps, and 3 sweep net samples.Additional collections by hand picking was also employed to include insects directly observed within the quadrat.The hand picking method used a 15 cm net, with each search lasting15 min each.Consequently, we collected 45 samples from the 5 plots at each sampling operation involving all the methods.Twenty-four sampling operations were carried out, 1 in each month, from January 2010 to December 2011.
Collected insects were placed in plastic bags or specimen containers and brought to the laboratory in Zoological Survey of India (ZSI), Kolkata.Specimens were separated into families using Crowson (1956) and Choate (1999), and mainly the numerous volumes on Coleoptera in the Fauna of British India, including Ceylon and Burma (Gahan 1906, Jacoby 1908, Fowler 1912, Marshall 1916, Maulik 1919, 1926, 1936, Andrewes 1929, 1935, Cameron 1930, 1932, 1934, 1939a, 1939b).Identifications of genera or species were made with the help of the Scientists of the Coleoptera section in the Zoological Survey of India (ZSI), Kolkata.Specimens were deposited in the ZSI (Kolkata).
The data on the relative abundance of each species was used for diversity analysis using BDPro software (McAleece et al. 1997).In order to estimate the diversity indices, raw data of each species collected from the plots were pooled together into 24 monthly samples.Using relative abundance as a measure, a discriminant function analysis (Manly 1994, Legendre & Legendre 1998) was performed to show differences between families of Coleoptera.
On the basis of the species richness and abundance of coleopteran in the 24 monthly samples, the Shannon-Weaver diversity index (Hʹ) was between 1.17 and 3.44, with a mean value of 3.11.The corresponding evenness values (H even ) was 0.65 and 0.85 with a mean value of 0.81.However, the monthly  beetle individuals in the samples and the species richness, the saturation was reached at the sample number 19 (Fig. 2), which is also reflected through the non-parametric measures of the species diversity such as Jackknife 1 (species saturation in 20th sample) (Fig. 2).The differences in the species richness among the samples were also reflected through the Chao 1 estimator, estimator of the unique and doubleton species in the samples.In comparison to the Chao 1, the Jackknife 1 estimate provided a better descriptor owing to similarity with the observed satura-tion of the species richness in the samples.
The discriminant function analysis indicates considerable differences in the relative representations of families of Coleoptera as shown in the biplot of Table .3. The axes of the biplot explained about 87% of the variability of the data on the relative abundance of the coleopteran families.On comparison, the two families, Elateridae and Tenebrionidae, were oriented in different coordinates to rest of the families.The Fisher's distance among the pairs of the families remained significant for  most of the cases reflecting the differences in the representation of the different species under the families.

DISCUSSION
Locally, terrestrial insect diversity depends on the resource availability and habitat, while climate determines diversity of insects and terrestrial arthropods, which is evident from biogeographical studies worldwide (Barberena-Arias & Aide 2002, Deans et al. 2005, Hirao et al. 2007, Richards & Windsor 2007).For example, about 1470 beetle species were recorded over a 4-year period in lowland tropical rainforests of Australia, but with a patchy distribution of species over time (Grimbacher & Stork 2009).In comparison, 50 species of beetles belonging to 11 families recorded from mesic undisturbed tundra in Nunavut, Canada (Ernst & Buddle 2013).
In tropical forests of Brazil, dung beetle assemblages vary between locations (21 species in Bahia, Vieira & Silva 2012; 33 species in Santa Catarina, Campos & Hernández 2013) and are considerably greater than the present observation of just 3 species.In pine-dominated forests, Vanderwel et al. (2006), found that xylophagous families Cerambycidae and Scolytidae were more abundant; in the present study, these families were represented by 4 species.The number of carabid species recorded (4 species) in the present study also is low compared to a study by Abdel-Dayem (2012) in Egypt (39 species in 28 genera).In Bandhavgarh National Park, Madhya Pradesh, India, a moist deciduous forest bordered with marsh and interspersed hillocks, is rich in beetles, with at least 44 scarabaeid species (Chandra & Ahirwar 2005).However, in terms of the number of species, the present observation (57 species) was greater than the 26 species recorded from deciduous forests of Achanakmar-Amarkantak Biosphere Reserve, central India (Chandra & Gupta 2012) and the 12 species recorded from Melghat Tiger Reserve, Maharashtra, India (Thakare & Zade 2012).In comparison to the tropical moist deciduous forests of Bethuadahari Wildlife Sanctuary, the steppe grasslands of Czech Republic appears to have a more diverse beetle assemblage with 103 being weevil (Curculionoidea), while rest were representatives of Anthribidae (4 species), Rhynchitidae (5 species), Apionidae (14 species) and Nanophyidae (1 species) (Stejskal 2004).From these comparisons, it is apparent that plant assemblages, landscapes, and climatic factors strongly influence beetle species richness.Apart from human-induced disturbance, the seasonal factors and resource availability in Bethuadahari Wildlife Sanctuary may cause variations in beetle abundance in the survey area.In Buxa National Park, West Bengal, high quality habitats in a mosaic of forest types account for a much greater richness of beetles (Sarkar et al. 2014(Sarkar et al. , 2015a(Sarkar et al. , 2015b)).
The dynamics of plant succession and the resultant variations in the habitat complexity influences the insect species assemblages (Brown 1985).Such changes (Buskirk & Buskirk 1976, Arun & Vijayan 2004) may limit the spatial distribution and encounter in course of collection using the pitfall traps or sweep nets.The cryptic species residing in the litters and within the degrading woods may have been missed by the sampling methods used.
In moist deciduous forests, such as found in the Bethuadahari Wildlife Sanctuary, the depth of leaf litter, as well as the litter-dwelling invertebrate fauna, including coleopterans, varies with the seasons.The forest of the Bethuadahari Wildlife Sanctuary is dominated by Tectona grandis, which may account for less variability in the detritus on the forest floor.In a stable forest ecosystem, the availability of the detritus, including decaying wood and faecal matter of larger herbivores, are high in relation to live leaves.Although abundance of detritivore species of coleopterans was higher than the strictly phytophagous species, in general, the 2 groups varied in accordance with seasonal variations of the resources available.Among the coleopterans, the Scarabaeidae are relatively specialized detritus-associated taxa and involved in the decomposition process (Chittenden 1915, Weiss 1922, Weinreich 1968).The presence of scarabaeids accelerates litter decomposition and helps maintain soil quality (Halffter & Edmonds 1982, Hanski & Cambefort 1991, Slade et al. 2007, Nichols et al. 2008, Simmons & Ridsdill-Smith 2011).The representatives of Staphylinidae, Tenebrionidae, and Elateridae, contribute to the multiple functional roles (Mattson & Addy 1975, Brown 1985, Schowalter 1995, Pfeiffer 1996, Dixon 2000, Omkar & Pervez 2000, Basset et al. 2003, Nair 2007) in sustaining the forest ecosystem.The ecological role of coleopteran and other insect groups in deciduous forests such as in Bethuadahari Wildlife Sanctuary may be better assessed by evaluating the trophic guild structure and the resources available over the entire season.Further studies would be beneficial to decipher interactions between the insect species and forest resources and sustaining the forest ecosystem.

Figure 1 .
Figure 1.Map of Bethuadahari Wildlife Sanctuary, West Bengal, India, with the sampling plots denoted by red dots.

Figure 2 .
Figure 2. The SHE analysis [S (species richness), H (information) and E (evenness) in the samples] (a), species richness S, and abundance N (b), and various measures of species richness (Unique species, Chao1, and Jacknife1) (c), for Coleoptera, calculated on the data of 24 samples of different months from January, 2010 to December, 2011, in Bethuadahari Wildlife Sanctuary, West Bengal, India.The filled in triangles mark the sample where the species saturation was reached in the estimates in b and c.

Table 1 .
Representative families of Coleoptera with ascertained species numbers observed in course of sampling of the Bethuadahari Wildlife Sanctuary, West Bengal, India, during 2010-2011.

Table 2 .
Species of Coleoptera recorded from Bethuadahari Wildlife Sanctuary, West Bengal, India (n i represents relative proportion of the species collected in the total sample; description of 56 species given).