Community Composition of Frugivorous Butterflies in a Neotropical Land Mosaic Emily Heim Department of Biological Aspects of Conservation, University of Madison Wisconsin ABST R ACT It is important to consider land transformation and its effects on speci es abundance, richness, and diversity. Frugivorous Nymphalid butterflies, baited using fruit traps, were used to assess trends in diversity for a mosaic of three habitat types forest, edge, and pasture in a premontane wet Costa Rica n landscape . Results showed more species in the pasture habitat, but a higher richness, evenness, and diversity in the forest habitat. Integrati on of conservation areas in the expanding array of human impacted landscapes is necessary to bes t maintain regional biodiversity in the long term. RESUMEN Es importante considerar cÃ³mo la transformaciÃ³n de la tierra afecta a la abundancia, a la riqueza y a la diversidad de la s especies. Las mariposas frugÃvoras Nymphalidae, atrapadas por las trampas de fruta, fueron usadas para evaluar las te ndencias de diversidad en un mosaico de tres tipos de hÃ¡bitat el bosque, el borde y el potrero en una zona premontan a y hÃºmeda de Costa Rica . Los resultados mostraron que habÃa mÃ¡s especies en el hÃ¡bitat del potrero pero que habÃa una riq ueza, una uniformidad y una di versidad mayores en el hÃ¡bitat del bosque. La integraciÃ³n de las zonas de conservaciÃ³n dentro de los paisajes impactados por los seres humanos es necesaria para mantener con Ã©xito la biodiversidad regional a largo plazo. INT RODUCTION Human dominion over the EarthÂ€s ecosystems has a negative impact on biodiversity worldwide Vitrousek et al. 1997. The population of tropical countries increased from 1.8 billion in 1950 to 4.9 billions in 2000 and is projected to grow by a fur ther 2 billion before 2030 Wright 2005. By clearing tropical forests, practicing subsistence agriculture, intensifying farmland production, or expanding urban centers, human actions are changing the worldÂ€s landscapes in pervasive ways Foley 2005. Be twe en one third and one half of usable land surface has been altered by human action Vitrousek et al. 1997. According to the 2005 FAO Global Forest Resource Assessment, 13 million hectares of forest are lost worldwide each year, primarily for conversion into agricultural landscapes, and Central America has the highest rate of tropical forest transformation at a rate of 1.3% 285,000 ha per year. Accounting for almost half of the deforestation worldwide, tropical rainforests are disappearing at a rate o f approximately six million hectares per year Willis et al. 2004. Tropical forests support approximately 50% of described species and an even larger number of undescribed species Wright 2005. Land transforma tion has been the primary driving force in the loss of biological diversity worldwide Vitrousek et al. 1997. As a result, how much tropical biodiversity remains will affect how organisms respond to altered landscapes Warner 2005. It is important to consi der land transformation and its effect s on species abundance, richness, and diversity to predict future regional composition . The ability of species to respond to such altered environments will indicate which species will persist and which are lost.
The fates of organisms that once made thei r homes in large, unbroken tracts of native habitats form a continuum: at one end is population decline to local, and eventually global, extinction; at the other extreme is the expansion into human dominated landscapes Daily et al. 2001. Understandi ng trade offs between alternative agricultural systems and biodiversity conservation is vital to predict changes in biodiversity and maximize conservation efforts Daily et al. 2001. Land transformation such as fragmentation and deforestation largely imp act the use of ecosystems, the size of intact natural habitats, and community composition of the habitat. M any ecosystem services e.g., flood control, soil fertility, pollination, pest control, timber production depend on the capacity of countryside spe cies to generate these services on a local scale Daily 1997 . As species go extinct, ecosystems are unable to function properly. D ue to intensifying human impact, a small number of large, undisturbed areas of natur al habitat are likely to remain Daily et al. 1998. The reduction of habitat size creates islands of forest in a human dominated system, therefore creating a mosaic landscape. The fragmented landscape creates a preferred habitat for biota that may be described as Â weedyÂ‚ Daily et al. 2001. These habitat generalists and open habitat species often replace forest habitat species that are often endemic or restricted to pristine or remn ant forests Warner 2001. Replacement of forest endemics by weedy species leads to a decline in regional div ersity because land transformation is a leading cause of tropical extinctions . Therefore, p astures , edge, and forest are crucial habitats to consider when studying effects of land use on community composition changes in a ccompanying land transformation . I n this study, the frugivorous butterflies in a mosaic landscape were used to predict future changes in community composition due to land transformations . Frugivorous butterflies are found throughout these three habitat types and their community compositi on will serve as an environmental marker of habitat degradation Nadkarni and Wheelwright 2000. T his paper reports on the frugivorous butterfly community in the premontane wet forest life zone in Monteverde , Costa Rica Holdridge, 1967. The main goa l of this study is to compare the species richness and composition of the butterfly community between forests , edge, and pasture habitats. METHODS S tudy Site Six sites of similar size were chosen that contained forest, edge, and pasture habitats. These s ites six pastures , six edges , six forests were located on or ne ar La Finca Santamaria in Monteverde , Costa Rica. Butterfly data were collected eight days at the end of the dry season from April 29 Âƒ May 7, 2006 between 10AM and 2PM. A butterfly trap was h ung from a tree in the center of each habitat type at the six different sites. Fishing line attached to a tree branch suspended each trap 1 to 3 meters above the ground. A total of eighteen butterfly traps were used in this study. The butterflies were c aptured in Van Someren rydon traps that consist of a frame of two wire loops 0.25 m dia meter surrounded by nylon mesh Daily and Ehrlich 1995. The nylon creates a cylinder open at the bottom and closed at the top. Each trap has a sheet of plywood susp ended below the netting with a disposable plastic dish located in the center to hold the bait. Butterfly bait was placed in the plastic dish. The butterfly bait was made in a plastic bucket with a secure lid. The bucket was filled halfway with beer, mas hed bananas, and a generous
dose of honey until it forms a thick consistency. The concoction was left to ferment for three days and was then placed in the traps. Butterfly traps were checked eight times and each time fermented bait was added to the trap after butterflies were identified and released. Data Collection Butterflies were identified with Butterflies of Costa Rica DeVries 198 7. A small net was used to get the butterfly out of the trap through a hole cut into the nylon mesh. A permanent paint marker was used to mark the forewing of each butterfly with a unique number. Each individual was marked to avoid recount ing it in the same trap and to see if any individuals travel between habitats. The chronological numbe r, habitat type, trap number, and scientific name were recorded for each individual captured. RESULTS Butterflies Collected In total, 288 individual butterflies of six subfamilies of Nymphalidae were identified. Four species and nine individuals were ca ptured in the Charaxinae subfamily, two species and 132 individuals in the Nymphalinae subfamily, two species and three individuals in the Ithomiinae subfamily, one species and 22 individuals in the Morphinae subfamily, four species and 116 individuals in the Satyrinae subfamily, and two species and six individuals in the Br assolinae subfamily Table 1. TABLE 1: Nymphalidae subfamily, number of species, and number of individuals captured across a habitat mosaic of premontane moist areas in Monteverde , Cos ta Rica from April 29 May 7, 2006 Nymphalidae subfamily Number of species Number of individuals Charaxinae 4 9 Nymphalinae 2 132 Ithomiinae 2 3 Morphinae 1 22 Satyrinae 4 116 Site 1 Pasture Edge Forest Site 2 Site 3 Site 4 Site 5 Site 6 Example of Butterfly Traps in Study Sites
Brassolinae 2 6 Abundance Four species and 178 individuals were capture d in the pasture, six species and 46 individuals in the edge, and 14 species a nd 61 individuals in the forest Table 2. Abundance values differ significantly between the three habitat types. The pasture had significantly more butterfly individuals X 2 = 72.5, df = 2, P < 0.05 than edge and forest X 2 = 25.27, df = 2, P < 0.05; X 2 = 12.17, df = 2, P < 0.05 , respectively . Large numbers of individuals of two butterfly species, Cissia gigas 46% and Smyrna blomfildia 52%, were found in the pasture habi tat. Table 2: Habitat type, number of individuals, and number of species of Nymphalid butterflies captured across a mosaic landscape in Monteverde, Costa Rica from April 29 Âƒ May 7, 2006 Habitat Number of individuals Number of species Pasture 178 4 Edg e 46 6 Forest 61 14 Species richness Species richness declined from forest through edge to pasture . The butterflies ca ptured in the forest had higher species richness than pasture and edge habitats Figure 1 . The forest habitat had highest species ri chness X 2 = 4.5, df = 2, P < 0.05 with 14 out of the 15 species present . The pasture habitat had a lower species richness than the forest habitat X 2 = 2, df = 2, P < 0.05 with six species and t he edge habitat had the lowest species richness X 2 = 0.5, df = 2, P < 0.05 with four species . Evenness E venness varied significantly between the three habitat types. Evenness values are a representation of how evenly species in a community are distributed by abundance, with 1 being perfectly even and 0 comp letely uneven. Evenness was highest in the forest habitat E = 0.51 followed by the edge habitat E = 0.35 and pasture habitat E = 0.15. Diversity The Shannon Weiner Index was calculated to find the range in species diversity. Butterfly diversity di ffered significantly between the three habitat types using a modified t test for HÂ€ Magurran 1988. These calculations showed that the butterfly community captured in the forest habitat was more diverse than edge habitat and pasture hab itat Figure 1. The forest had a significantly higher diversity than both the pasture and the edge P F: t = 11.09, df = 74.13, P < 0.05; P E: t = 4.80, df = 104.64, P < 0.05. T he edge had significantly higher diversity of butterfly species than the pasture E P: t = 4.74, df = 56.09, P < 0.05.
Forest 1 3 3 1 5 1 15 1 1 15 2 2 1 10 Archaeoprepona Caligo illioneus Cissia gigas C. polyphemus Consul electra Eretis suzannae H. acheronta M. maculata Memphis beatrix Memphis pithyusa Morpho peleides Napeogenes Opsiphanes Pseudoscada S. blomfildia Edge 1 7 6 7 1 24 A B
Pasture 82 2 1 93 FIGURE 1: Number of Nymphalid butterfly individuals that were found in forest, edge, and forest habitats in study sites in Monteverde , Costa Rica from April 29 May 7, 2006. These graphs show the difference in species diversity, demonstrating that the forest habitat has the highest diversity and the pasture habitat is the least diverse habitat. Similarity Similarity of species between h abitat s was calculated using SorensonÂ€s Quantative Index, which determines similarity based upon relative abundance of co occurring species Southwood 1966. A value is assigned from 0 1, with 0 indicating no similarity and 1 indicating complete homogeny betwee n sites. The highest similarity of butterfly species was found between edge and forest with seven species in common Cn = 0.523. The pasture and forest have least similar butterfly communities with only thre e of the same butterfly species out of 15 tota l species Cn = 0.117. Butterfly natural history and distribution patterns The butterflies identified in this study can be classified into four different habitat preferences DeVries 198 7. These include forest habitat, forest/edge habitat, all habita t, and open habitat butterfly species . In all, e ight species and 142 individuals of forest habitat butterflies were captured along with one species and five individuals of forest edge habitat, four species and 46 individuals of all habitat, and two specie s and 95 individuals of open habitat butterflies Table 2. Out of the eight forest species, only four were found in only forest habitat: E retis suzannae , Memphis pithyusa . Napeogenes tolosa amara , and Greta nero . Two of the forest species were located i n both forest and edge: Archaeopreona meander amphimachus and Opsiphanes quiteria quirinus . Oddly, one C
forest species, Historis acheronta , was located only in pasture. Another unusual finding is that 73% of S. blomfildia was found in pa sture habitat, whe n DeVries 198 7 describes this species as occurring in forest. The only edge species, Consul electra , was only identified in forest. Two out of the four all habitat species were only found in forest: Cissia polyphemus and Memphis beatrix . Morpho peleid es , an all habitat species, was discovered in both pasture and forest habitat, but not in the edge. Manataria maculata , an all habitat species, was identifi ed in all three habitats. One of the open habitat species, Caligo memnon , was found in only forest habitat. The other open habitat species, Cissia hermes , was found in all three habitats. TABLE 3 : Species, number of individuals, and distribution and habitat type on Nymphalid butterflies caught across three habitat types at six study sites in Montev erde , Costa Rica. Chart adapted from DeVries 1987. Scientific name Number of Individuals Butterfly distribution and habitat type Archaeoprepona meander amphimachus 2 Occurs from 0 1,800 m on both slopes in all forest habitats. Consul electra 5 Occ urs from 500 1,400 m on both slopes in all forest habitats. Found near edges or canopy. Moderately common, but local. Memphis beatrix 1 Occurs from 700 1,600 m. Widespread and common. Memphis pithyusa 1 Occurs from 0 1,500 m in both slopes in all for ests habitats. Widespread and common Historis acheronta 2 Occurs from 0 1,200 m on both slopes in all forest habitats. Less common and it may be rare in many areas. Smyrna blomfildia datis 130 Occurs from 0 1,200 m on both slopes in all forest habitat s. Napeogenes tolosa amara 2 Occurs from 0 1,500 m in all forest habitats. Common. Greta nero 1 Occurs from 600 2,000 m on both slopes in all forest habitats. Morpho peleides limpida 22 Occurs from 0 1,800 m on both slopes in all habitats. Commonest species seen along edges, rivers, and coffee plantations. Cissia polyphemus 1 Occurs from 800 2,500 m in all habitats. Opsiphanes quiteria quirinus 3 Occurs from 500 1,800 m on both slopes in association with cloud forest habitats. Rare. Caligo memno n 3 Occurs from 0 1,400 m on both slopes. Common on Pacific slope. Can live in severely disturbed agricultural habitats. Most common in rainy season. Mantaria maculata 22 Occurs from 0 2,500 m in all habitats. Migratory .
Cissia hermes 92 Occur s from 0 1,500 m in all habitats. Widespread and common in pastures and open areas. Eretis suzannae 1 Occurs from 900 1,400 m associated with premontane rain forest. Additional Findings One open habitat species, C. hermes , was mainly captured in only o ne of six pasture traps. 88% of the total C. hermes butterflies identified were captured there. DISCUSSION The highest number of butterfly individuals were collected in pasture traps as a result of two ÂweedyÂ‚ butterfly species: C. hermes and S. blomf ildia . These two species were found in all three habitat types and increased in abundance from the forest to the pasture habitats. C. hermes is a hab itat generalist and may be replacing rarer forest species. One main finding that differs from previous s tudies is the abundance of S. blomfildia in the pasture habitat. In The Butterflies of Costa Rica , S. blomfildia is described primarily as a forest species DeVries 1987. If S. blomfildia is mostly a forest species then why were 73% of the total species captured in traps in the pasture habitat? It is possible that this butterfly species is increasing in abundance because it is able to benefit from the mosaic landscape. This species, along with C. hermes , is likely to persist and grow in numbers in the future. In contrast to the abundance of individuals in the pasture habitat, this study found that butterfly richness, evenness, and diversity are the highest in forest habitats. This could indicate that large number s of species are negatively affected when primary forest habitats are destroyed. Previous studies suggest that small fragments of primary forest may be incapable of sustaining population of Archaeoprepona , Memphis , Morpho , and Caligo species Daily and E hrlich 1995. T hese same four genera were all captured in the forest habitat traps and none of them were found in the pasture habitat traps. It is possible that these species are likely to decline in the future as human transformations negatively impact forest habitats. Composition of th is butterfly community is most similar between forest and edge habitats. Perhaps more butterfly species are able to survive in less modified landscape in comparison to a pasture habitat. For example, temperate studies have suggested that fewer forest spe cies remain in pastureland than in agricultural cropland after land transformation has occurred Tucker 1997. In another study on tropical avian communities, it was concluded that forested coffee farms may serve to better maintain native biodiversity tha n pastureland after habitat transformation Warner 2001. These studies suggest that the use of alternative agricultural practices may be the most effective way to preserve biodiversity in the long term. Education programs for farmers in local communitie s could increase the amount of forest like habitat therefore increasing the probability the more species will survive into the future. Management strategies that incorporate forest like habitats are crucial for future preservation of natural biodiversity. Ecologists have assumed that few tropical forest animals and plants survive in agricultural landscapes. Yet deforested countryside may retain substantial ÂforestÂ‚ biodiversity, especially in landscapes with a variety of land uses Hughes et al. 2002. Countryside habitats include active agricultural plots, plantation or managed forest,
fallow land, gardens, and small remnants of native vegetation embedded in landscapes devoted primarily to human activities Daily et al. 2001. These management strategi es that integrate different land use practices are required to maintain diversity long term. It is important to consider the preservation of a mosaic of landscapes with continuously varying degrees of connectedness. As a species fails to keep pace with t he changes in environment, it is ultimately doomed to extinction. The goal of conservation is to halt the accelerating rate of species loss associated with human dominance of the biosphe re Ehrlich and Ehrlich 1981. Conversation efforts that incorporate alternative agricultural management strategies are important to maintain biodiversity in the long term. ACKNOWLEDGMENTS I am extremely grateful for the help of Alan Masters, who encouraged and supported me throughout this project, from the proposal to the final analysis. Thanks to the TAÂ€s, Ollie Hyman and Maria Jost for answering endless amounts of questions and always making me laugh. A special thank you to Katie Stalland for helping me set up traps and sharing her Costa Rican home with me. A huge thank you to Adam Trigg, who flew here and helped fix my project when everything fell apart. Also, I am very grateful to Norman Santamaria, without him I would still be looking for study sites. To Marisol, Roxanne, and Eleina who reminded me to look at each butterfly with curiosity and excitement . Thanks to my sister, Diana, for the help with Spanish translations. Finally, to Mom and Dad, for supporting me and giving me the opportunity to explore the world. LITERATURE CITED Daily, G.C., P.R. Ehrlich, and G.A. Sanchez Azofiefa. 2001. Countryside biogeography: use of human dominated habitats bye the avifauna of southern Costa Rica . Ecological Applications. 11: 1 13. Daily, G.C. and P. Ehrlich. 1995. Preservation of biodiversity in small rainfore st patches: rapid evaluations using butterfly trapping. Biodiversi ty and Conservation. 4: 35 55. Daily, G.D., editor. 1997. NatureÂ€s services: societal dependence on natural ecosystems . Island Press, Washington, D.C., USA. Daily, G.D., et al. 1998 . Food production, population growth, and the environment . Science 281:1291 1292. DeVries, P.J 1987. The butterflies of Costa Rica and their natural history. Princeton University Press, Princeton, New Jersey. Ehrlich, P. R., and A. H. Ehrlich. 1981 . Extinction: The Causes and Consequences of the Disappearance of Species. Random House, New York. As seen in Warner, M. 2005. The role of primary forest on tropical avian communities in a mixed agricultural landscape. Tropical Ecology and Conse rvation. CIEE. Foley, J.A., R. DeFries, G.P. Asner. Global Consequences of Land Use. Science. Vol 309: 570 574. Holdridge, L.R. 1967. San Jose, Tropical Science Center. Hughes, J.B., G.C. Daily, and P.R. Ehrlich. 2002. Conservation of tropical fore st birds in countryside habitats. Ecology letters. 5:121 129. NA. 2005. Global Forest Resources Assessment 2005. FAO. Available online: http://www.fao.org/forestry/site/fra/en Nadkarni, N.M. and N.T. Wheelwright. 2000. Monteverde Ecology and Cons ervation of a Tropical Cloud Forest. Oxford University Press, New York Perkins, B. 2001. Diversity and population dynamics of frugivorous butterflies in a fragmented landscape. Tropical Ecology and Conservation. CIEE. Southwood, T.R.E. 1966. Ecol ogical Methods. Chapman and Hall, New York , New York. USA. Tucker, G.M. 1997. Priorities for bird conservation in Europe: the importance of the farmed landscape. Page 79 116 in D.J. Pain and M. W. P ienkowski, editors. Farming and birds in Europe: the common agricultur al policy and its implications for bird conservation. Academic Press, San
Diego, California, USA . As shown in Warner, M. 2005. The role of primary forest on tropical avian communities in a mixed agricultural landscape. Tropical Ecolo gy and Conservation. CIEE. Vitousek P.M, H. Mooney, J. Lubchenco, J. Melillo. 1997. Human Domination of EarthÂ€s Ecosystems. Science Vol. 277 pp. 494 499. Warner, M. 2005. The role of primary forest on tropical avian communities in a mixed agricultur al landscape. Tropical Ecology and Conservation. CIEE. Willis, K.J., L. Gillson, T.M Brncic. How ÂvirginÂ‚ is virgin rainforest? Science. Vol. 304 pp. 402 403. Wright, J.S. 2005. Tropical forests in a changing environment. Trends in Ecology and Ev olution. 2010: 553 560