Geography of Indian Butterflies: Patterns Revealed by Checklists of Federal States

Abstract

Butterflies are widely used to analyze biogeographical patterns, both at the global and regional scales. Thus far, most of the latter originated from well-surveyed northern regions, while the species-rich tropical areas lag due to a lack of appropriate data. We used checklists of 1379 butterfly species recorded in 36 federal states of the Republic of India (1) to explore the basic macroecological rules, and (2) to relate species richness and the distribution of endemics and geographic elements to geography, climate, land covers and socioeconomic conditions of the states. The area, land covers diversity and latitude did not affect species richness, whereas topographic diversity and the precipitation/temperature ratio (energy availability) were positive predictors. This is due the geographic and climatic idiosyncrasies of the Indian subcontinent, with its highest species richness in the small, densely forested mountainous northeast that receives summer monsoons. The peninsular effect that decreases the richness towards the tip of subcontinent is counterbalanced by the mountainous forested Western Ghats. Afrotropical elements are associated with savannahs, while Palearctic elements are associated with treeless habitats. The bulk of Indian butterfly richness, and the highest conservation priorities, overlap with global biodiversity hotspots, but the mountainous states of the Western Himalayas and the savannah states of peninsular India host distinctive faunas.

Keywords: Oriental realm; biogeographic elements; climate; faunal turnover; latitudinal gradient; peninsular effect.

Figure 1

(a) Map of Indian federal states, with state names, abbreviations and tone of blue indicating numbers of butterfly species. (b) Ordination scatterplot of Indian states, obtained from a principal component analysis organizing the states according to values of predictors used in the regression models (eigenvalues 1–4: 0.336, 0.197, 0.135, 0.076, explained variation 35.0%); the darts indicating per-state butterfly numbers were entered as supplementary variables, not influencing the positions of the states. (c) PCA scatterplot of the predictors used in the regression models. Legend to Figure 1: Indian states with respective butterfly species richness: AN—Andaman and Nicobar (242), AP—Andhra Pradesh (229), AR—Arunachal Pradesh (745), AS—Assam (686), BR—Bihar (174), CG—Chhattisgarh (203), CH—Chandigarh (100), DD&DN—Daman and Diu and Dadra and Nagar Haveli (37), DL—Delhi (106), GA—Goa (275), GJ—Gujarat (200), HP—Himachal Pradesh (386), HR—Haryana (118), JH—Jharkhand (162), JK—Jammu and Kashmir (315), KA—Karnataka (328), KL—Kerala (330), LA—Ladakh (180), LD—Lakshadweep (15), MH—Maharashtra (284), ML—Meghalaya (712), MN—Manipur (730), MP—Madhya Pradesh (184), MZ—Mizoram (276), NL—Nagaland (643), OD—Odisha (246), PB—Punjab (129), PY—Puducherry (62), RJ—Rajasthan (124), SK—Sikkim (729), TN—Tamil Nadu (329), TR—Tripura (349), TS—Telangana (151), UK—Uttarakhand (518), UP—Uttar Pradesh (153), WB—West Bengal (731). Variables in panel (c): AltDiff—altitude difference, Clim1–4—composite climatic variables obtained by PCA analysis of 19 bioclimatic variables, GDPpercap—GDP per capita, HumPopDn—human population density, LanCovDv—land cover diversity, Lat—latitude, LitRate—literacy rate, Long—longitude, MeanAlt—average altitude, MDF—moderate dense forest, NonFor—non forest, OF—open forest, PdT—precipitation/temperature ratio, RuPop—rural population, TreeOut—scattered tree patch, UrPop—urban population, VDF—very dense forest.

Figure 2

Maps showing the richness of all, endemic, Oriental, Afrotropical and Palearctic elements in 36 federal states and territories of the Republic of India (raw data: left column), and predictions of the regression models (four right columns), explaining the left column patterns by sets of geographic, climatic, land cover and socioeconomic predictors. See Table 3 and Table 4 for the models’ terms and related statistics. The color scales run from highest values (light yellow) to lowest values (dark blue).

Figure 3

Maps showing the turnover of butterfly species compositions in 36 federal states and territories of the Republic of India (raw data: left), and predictions of the regression models, explaining the turnover pattern by sets of geography, climate, land covers and socioeconomic predictors. See Table 3 for the models’ terms and related statistics.