Increasing deforestation worldwide has expanded the interfaces between fragmented forests and non-forest habitats. Human-made edges are very different from the original forest cover, with different microclimatic conditions. Conversely, the natural transitions (i.e., ecotones) are distinct from human-made forest edges. The human-made forest edges are usually sharp associated with disturbances, with abrupt changes in temperature, humidity, luminosity and wind incidence towards the forest interior. However, the natural forest-lake ecotones, even when abrupt, are composed of a complex vegetal physiognomy, with canopy structures close to the ground level and a composition of herbaceous and arboreal species well adapted to this transition range. In the present study, fruit-feeding butterflies were used as models to investigate whether faunal assemblages in natural ecotones are more similar to the forest interior than to the anthropic edges. Butterflies were sampled monthly over one year in the Rio Doce State Park, Southeastern Brazil, following a standardized design using a total of 90 bait traps, in three different forest habitats (forest interior, forest ecotone and anthropic edges), in both canopy and understory. A total of 11,594 individuals from 98 butterfly species were collected (3,151 individuals from 79 species in the forest interior, 4,321 individuals from 87 species in the ecotone and 4,122 individuals from 83 species in the edge). The results indicated that the butterfly richness and diversity were higher in transition areas (ecotones and edges). The ecotone included a combination of butterfly species from the forest interior and from anthropic edges. However, species composition and dominance in the ecotone were similar to the forest interior in both vertical strata. These results suggest that human made forest edges are quite distinct from ecotones. Moreover, ecotones represent unique habitats accommodating species adapted to distinct ecological conditions, while anthropic edges accommodate only opportunistic species from open areas or upper canopies.

Butterflies are one of the best‐known insect groups, and they have been the subject of numerous studies in ecology and evolution, especially in the tropics. Much attention has been given to the fruit‐feeding butterfly guild in biodiversity conservation studies, due to the relative ease with which taxa may be identified and specimens sampled using bait traps. However, there remain many uncertainties about the macroecological and biogeographical patterns of butterflies in tropical ecosystems. In the present study, we gathered information about fruit‐feeding butterfly species in local communities from the Atlantic Forests of South America. The ATLANTIC BUTTERFLIES data set, which is part of ATLANTIC SERIES data papers, results from a compilation of 145 unpublished inventories and 64 other references, including articles, theses, and book chapters published from 1949 to 2018. In total, the data set contains 7,062 records (presence) of 279 species of fruit‐feeding butterflies identified with taxonomic certainty, from 122 study locations. The Satyrini is the tribe with highest number of species (45%) and records (30%), followed by Brassolini, with 13% of species and 12.5% of records. The 10 most common species correspond to 14.2% of all records. This data set represents a major effort to compile inventories of fruit‐feeding butterfly communities, filling a knowledge gap about the diversity and distribution of these butterflies in the Atlantic Forest. We hope that the present data set can provide guidelines for future studies and planning of new inventories of fruit‐feeding butterflies in this biome. The information presented here also has potential use in studies across a great variety of spatial scales, from local and landscape levels to macroecological research and biogeographical research. We expect that such studies be very important for the better implementation of conservation initiatives, and for understanding the multiple ecological processes that involve fruit‐feeding butterflies as biological indicators. No copyright restrictions apply to the use of this data set. Please cite this Data paper when using the current data in publications or teaching events.

No abstract

No abstract