A comparison of ground dwelling insect communities in gaps and closed canopy forest in the elfin forest of Monteverde, Costa Rica Elizabeth Green Colleges of Natural Science and Liberal Arts, University of Texas, Austin, Texas, USA ABSTRACT This experiment compared communities of ground dwelling insects in gaps and closed canopy forest in the elfin forest of the Estacin Biolgica Monteverde in Puntarenas, Costa Rica. Insects were trapped with pitfall traps in six gaps and adjacent forest are as in October and November 2000. Shannon Weiner Diversity indices, Margalef species richness indices, evenness HÂ€ /In S, mean abundance per site, and Morisita Similarity indices were compared. Diversity differed only for Coleopterans, which were more d iverse in gaps HÂ€ = 1.201 than in forest HÂ€ = 0.8387; modified t test t= 1.979, p < 0.001. The Margalef index of species richness did not vary between paired gap and forest sites for total insects captured, but was significantly higher for Coleopterans in the gaps than in the forest sign test, a level 0.05. Coleopterans also had greater evenness in gaps than in forest sign test, a level 0.05. Mean abundances in gap 8.04 +/ 7.669 mean number of insects/trap and forest 5.09 +/ 3.191 insects/trap sites were not different. The Morisita Similarity index was 0.4846, with 47.05% of ground dwelling insect morphospecies shared between gaps and closed forest. Perhaps conditions in gaps and closed canopy areas in this forest are sufficiently different to allow specialization by ground dwelling insects to one or the other habitat type. RESUMEN Este experiment o comparo las comunidades de insectos que viven en la tierra en los espacios de arboles cados y en la sombra en el bosque de los duendes en la Estacin Biolgica Monteverde, Puntarenas, Costa Rica. Atrape los insectos con trampas en seis espacios y en el bosque adyacente en octubre y noviembre 2000. Compare los ndices de la diversidad Shannon Wei ner, los ndices de riqueza de las especies Margalef, la igualdad HÂ€/In S, la abundancia promedia y los Ândices de la similitud Morisita. La diversidad estuvo diferente solamente por el orden Coleoptera, que estuvo mas diverso en los espacios HÂ€ = 1.2 01 que en el bosque HÂ€ = 0.8387; prueba t modificado t= 1.979, p < 0.001. El ndice Margalef no vario entre los sitios de los espacios y el bosque pero estuvo de manera significativa mayor por la Coleoptera en los espacios que en el bosque prueba signo nivel a 0.05. Las abundancias promedias en los espacios y en el bosque no estuvieron diferentes. El ndice Moricita estuvo 0.4846 y 47.05% de las especies de insectos fueron compartidos entre los espacios y el bosque. Es posible que las condiciones de e stos lugares sean diferentes suficientemente para permitir la especializacin de los insectos en cada uno. INTRODUCTION
Many studies of p lants have found large differences in species composition between gaps in the Forest and closed Canopy areas Terborgh 1992. Fewer studies have been conducted to find the effects of gaps on animal diversity Alvarez and Willig 1993. Studies of birds Shelly 1998, snails Alvarez and Willig 1993, and ants Feener and Schupp 1998 have shown that some species of these g roups show a preference for either gap of forest conditions, while other species inhabit both with equal frequency. Arthropods are the most abundant animals in terrestrial ecosystems and among the least studied Alvarez and Willig 1993. Ground dwelling a rthropods play a vital ecological role through their activity as herbivores, seed predators, pollinators, predators, food for vertebrates and other invertebrates, and decomposers Erwin and Erwin 1976, Janzen 1983, Pearson 1986. An understanding of the ef fect of gaps on ground dwelling arthropods would allow the placement of a few more essential pieces in the growing picture emerging through studies of gap ecology. In studies of short horned grasshoppers Orthoptera: Acrididae, Braker 1991 found that th e species composition of acridids differed in gaps and closed canopy understory at La Selva Biological Reserve in Costa Rica Shelly 1988 observed that damselfly Odonata: Zygoptera and robber fly Diptera: Asilidae species show a preference for either deep shade or gap areas. In a study of day flying insects, Shelly 1988 also found that individuals of the Hymenopteran family Formicidae and of the orders Coleoptera and Psocoptera are more abundant in closed canopy forest than in gaps. He found no prefe rence among other orders of day flying insects Shelly 1988. The differing solar insolation, soil moisture, and temperature in gaps compared to surrounding forest is expected to influence species composition of ground dwelling insects in these habitats F eener and Schupp 1998. The high productivity associated with gaps is expected to be associated with a reduction in diversity, as found in studies of rodents and carnivores Owen 1988 and predicted by some models Abrams 1988. The uppermost slopes elev ation 1400 1800m of the Cordillera de Tilarn near Monteverde, Costa Rica, are covered with elfin forest Haber et al. 2000. This forest is characterized by stunted trees the canopy is 15 20m high that are exposed to north east trade winds and seasonal windstorms blowing over the ridge from the Atlantic Ocean as they grow Haber et al. 2000. This elfin cloud forest lies within the lower montane rain forest life zone Hartshorn 1983. The three meters of rain falling here each year, along with the daily mist and clouds, create a hospitable environment for an exuberance of life Haber et al. 2000. The heavy moisture and cool temperatures of this forest also keep the leaf litter damp or wet for nearly the entire year Haber et al. 2000, a factor known to be important to the survival of ground dwelling arthropods Darlington 1970, Erwin and Erwin 1976, Feener and Schupp 1998. The strong winds knock off branches and push down trees, creating an uneven canopy with frequent gaps that allow sunlight to penetr ate to the forest floor, nourishing a boisterous understory Haber et al. 2000. This experiment tested the hypothesis that gaps and continuous canopy understory in the elfin forest of the Estacin Biolgica Monteverde have different species compositions of ground dwelling insects, with a greater diversity of species in closed canopy forest than in gaps. This would follow the trends discovered by Shelly 1988 in
day flying insects and Owen 1988 in rodents and carnivores, as well as predicted by models r elating productivity to diversity Abrams 1988. MATERIALS AND METHODS This study compared the diversity of ground dwelling insects in six treefall gaps and adjacent forest areas along the trail through the elfin forest at the Estacin Biolgica Monteverde, Puntarenas, Costa Rica 10Â‚20Â€ N, 84Â‚48Â€ W during the late wet season. All gaps met BrokawÂ€s 1982 definition as Âƒa Â„holeÂ€ in the forest extending through all levels down to an average height of two m above groundÂ…. The gaps ranged in size fr om 90 to 140m calculated from radii measured with a Range Finder. Gap vegetation was less than a meter high, suggesting that the gaps were fairly young. The opening in the canopy created by each fallen tree permitted higher light conditions than in neig hboring areas, with a dense understory plant community visibly different from the more widely spaced and smaller understory plants in the closed canopy forest. Near the center of each gap four pitfall traps were placed in a square shaped arrangement approx imately three meters apart from one another. Four traps were placed in a similar arrangement in a forest area 10 15m from the edge of each gap. Pitfalls were made of 250 mL paper cups half filled with soapy water and placed in the ground. The traps were sh eltered from rain by Styrofoam plates supported five cm above the traps by bamboo skewers. Insects were trapped for varying lengths of time at each site to ensure a sufficient sample size despite the effects of variable weather conditions, with traps at t wo gap forest sites collecting for three days, one site for six days, and three sites for nine days in late October and early November 2000. A gap and its associated forest area were always surveyed together. Contents of traps were emptied into alcohol. No n insect arthropods and non ground dwelling insects were excluded, and the remaining insects were sorted to family and morphospecies. A reference collection was created and placed in the collection of the Estacin Biolgica Monteverde. The Shannon Weiner diversity index was used to compare diversity within orders and for all ground dwelling insects found in gaps and forest. A modified t test was used to test for significant differences. The Margalef indices of species richness and the evenness HÂ€/In S for paired gap and closed canopy forest sites were compared. Richness and evenness of paired sites were tested for significant differences with sign tests. Mean number of insects per trap and mean numbers of beetles per trap were compared between gaps and forest. The Morisita Similarity Index was used to compare morphospecies composition between gaps and closed canopy areas. Percent shared morphospecies between gaps and closed canopy forest was calculated as well. Traps from gaps and from forest sites were grouped to create a species area curve of Coleopteran morphospecies in gap and forest with increasing number of traps. RESULTS
A total of 265 individuals of 45 morphospecies were collected. One hundred sixty seven individuals of 35 morphospecies were fo und in the gaps, and 98 individuals of 28 morphospecies were trapped in the forest. These individuals represented six orders. Representatives from one family of each of the orders Orthoptera, Blattaria, Dermaptera, and Hemenoptera were trapped. Insects bel onging to two Hemipteran families and ten Coleopteran families were caught as well Figure 1. Coleopterans were the most abundant in numbers of individuals as well as the most speciose order found 127 individuals of 32 morphospecies Figure 2. Abundanc e of total insects captured did not vary in gaps 8.04 +/ 7.669 mean number of insects/trap and forest 5.09 +/ 3.191 insect/trap, nor did abundance of beetles 3.36 +/ 2.936 beetles/trap in gaps, 3.14 +/ 2.867 beetles/trap in forest Table 1. The Shannon Weiner Diversity Index showed a significant difference only among morphospecies of Coleopterans, which were more diverse in gaps HÂ€ = 1. 2011 than in closed canopy forest HÂ€ = 0.8387; modified t test t = 1.979, p < 0.001. There was no signific ant difference in overall diversity in gaps HÂ€ = 1.1868 and closed canopy forest HÂ€ = 1.124; modified t test t= 1.1969, 0.20 < p < 0.50. The Margalef index of species richness did not vary between paired gap and closed canopy sites for total insects ca ptured sign test, H0 not rejected at a level 0.05 Table 2, but was significantly higher for Coleopterans in the gaps than in the shade sign test, a level 0.05 Table 3. Evenness of all ground dwelling insects was not significantly different in the t wo habitats sign test, H0 not rejected at a level 0.05 Table 4, but Coleopterans had a higher evenness in the gaps than in closed canopy forest sign test, a level 0.05 Table 5. The Morisita Similarity Index for all ground dwelling insects captured in gap and closed canopy sites was 0.4846, with 47.05% shared morphospecies between gaps and closed canopy areas Table 1. The Morisita Similarity Index for Coleopterans in gaps and closed canopy forest was 0.4575, with 47.07% shared beetle morphospecies The number of morphospecies of Coleopterans in gaps increased more rapidly with increasing number of traps y = 3.84 x p v p R = 0.8997 than did the number of Coleopteran morphospecies caught in closed canopy forest y = 4.98 x p R = 0.9029 Figure 3. DISCUSSION Species diversity was not higher in closed canopy forest than in gaps. Coleopterans, the only group that did show a significant difference, had a greater diversity in gaps than in forest. This contrasts with ShellyÂ€s 1988 findings of greater diversity of day flying Formicids, Coleopterans, and Psocopterans in closed canopy forest than in gaps. This contrast merits further investigation to determine what factors might influence diversification in closed canopy or gap areas. One possibility is that the grea ter abundance of rotting logs and branches in gaps as compared with closed canopy forest is attractive to beetles, who frequently live in or eat dead and rotting wood. In particular, Carabid beetles, the most abundant family found in gap sites, are known t o live and hunt in rotting logs in their larval and adult stages Darlington 1970, Erwin and Erwin 1976. The greater abundance of young undefended vegetation in gaps may also attract an
increased diversity of beetles in comparison to the more strongly def ended older plants in closed canopy forest Coley and Barone 1996. While a larger sample size is required to determine whether there are significant trends in other taxa, this experiment supported ShellyÂ€s 1988 findings that variation in diversity betwe en gaps and shade is different with different taxa. The constancy of abundance and species richness and evenness between gap and forest sites indicates that gap and forest are able to host comparable communities of ground dwelling insects. While the habit ats are different enough to allow specialization to one or the other, neither habitat appears to be significantly more suitable for insects than the other. Species composition varies between gap and closed canopy areas in the elfin forest of Monteverde. F ewer than half of the morphospecies found were shared by adjacent gap and closed canopy areas of the forest. These results are consistent with the findings in previous studies of plants Terborgh 1992, birds Shelly 1988, snails Alvarez and Willig 1993 and other insects Shelly 1988, Braker 1991, Feener and Schupp 1998 that found some animals to occupy both gap and closed canopy forest while others specialized to one or the other habitat. This suggests that even in the elfin forest, with its great abu ndance and close proximity of gaps due to high wind, conditions are different enough in gaps and closed canopy forest to allow specialization to one habitat or the other. Sampling more thoroughly through the use of more pitfall traps per site and with the addition of other sampling methods would yield a more accurate picture of the ground dwelling insect communities in gap and forest habitats. It would also be interesting to survey insects on vegetation and insects flying in gap and forest areas to give a clearer understanding of the forest insect community. A comparison of insect communities in gaps of different sizes and different ages at different times of the year would be informative as well. The differing species composition and diversity of ground d welling insects in gaps and closed canopy forest may influence species composition of birds, reptiles, amphibians, mammals and other insects who rely on insects as prey Shelly 1988. Examination of the determining factors that make gaps or shade appealing to ground dwelling insects could give insight into factors influencing the distribution of predaceous insect and vertebrate communities in the forest as well. ACKNOWLEDGEMENTS Muchisimas gracias a Mauricio Garca por su ayuda paciente en todos los pasos de este experimento Thanks to Andrew Rodstrom for his frequent and cheerful help keying out tiny Brown beetles and for his revisions to this paper. Much appreciation goes to Tim Kuhman, Laurel Klein, and Seema She th for their thoughtful comments on earlier drafts of this paper. Thanks to my mama tica, Virginia Leiton for her prayers, support, and yummy lunches, all of which served me well on many days of crawling through gaps in the elfin forest. Thanks to the Esta cin Biolgica Monteverde for the use of its magical elfin forest. Thanks to Alan and Karen Masters and Mauricio Garcia for giving me the opportunity to perform this experiment.
LITERATURE CITED Abrams, P.A. 1988. Resource productivity consumer species diversity: simple models of competition in spatially heterogeneous environments. Ecology 69: 1418 1433. Alvarez, J. and M.R. Willig. 1993. Effects of treefall gaps on the density of land snails in the Luquillo Experimental Forest of Puerto Rico. Biotr opica 25:100 110. Braker, E. 1991. Natural history of a Neotropical gap inhabiting grasshopper. Biotropica 23: 41 50. Brokaw, N.V.L. 1982. The definition of treefall gap and its effect on measures of forest dynamics. Biotopica 14: 158 160. Coley, P.D and J.A. Barone. 1996. Herbivory and plant defenses in tropical forests. Ann. Rev. Ecol. Sust. 27:305 335. Darlington, P.J., Jr. 1970. Carabidae on tropical islands, especially the West indies. Biotropica 2: 7 15. Erwin, T.L. and L.V.J. M Erwin. 1 976. Relationships of predaceous beetles to tropical forest wood decay. Part II. The natural history of Neotropical Eurycoleus macularis Chevrolat Carabidae: Lebiini and its implications in the evolution of ectoparasitoidism. Biotropica 8: 215 223. F eener, D.H., Jr. and E.W> Schupp. 1998. Effect of treefall gaps on the patchiness and species richness of Neotropical ant assemblages. Oecologia 116: 191 201. Haber, W.A., W. Zuchowski and E. Bello 2000. An Introduction to Cloud Forest Trees: Montever de, Costa Rica. Monteverde de Puntarenas, Costa Rica, Mountain Gem Publications. Hartshorn, G.S. 1983. Plants. ed. D.H. Janzen. In Costa Rican Natural History. Chicago, University of Chicago Press. Janzen, D.H. 1983. Insects. ed. D.H. Janzen. In Costa Rican Natural History. Chicago, University of Chicago Press. Owen, J.G. 1988. On productivity as a predictor of rodent and carnivore diversity. Ecology 69: 1161 1165. Pearson, D.L. 1986. Seasonal patterns of lowland forest arthropod abundance in southeastern Peru. Biotropica 18: 244 256 Shelly, T.E. 1988. Relative abundance of day flying insects in treefall gaps and shaded understory in a Neotropical forest. Biotropica 20: 144 119. Terborgh, J. 1992. Diversity and the Tropical Rainforest. New York, HPHLP, Scientific American Library.
Table 1. Number of individuals trapped in gap and forest sites by morphospecies Order Family Morphospecies Number in gap Number in forest Orthoptera Tettigoniidae A 4 3 B 0 1 C 4 3 Blatteria Blatellidae A 11 10 Dermaptera Forficulidae A 1 0 Hemiptera Reduviidae A 3 1 Cydnidae A 0 1 B 2 5 Coleoptera Carabidae A 19 36 B 2 0 C 8 1 D 4 0 E 3 3 F 5 1 G 2 1 H 1 0 I 2 0 J 2 0 K 2 0 L 2 0 Scarabaeidae A 0 5 B 1 4 Eucnemidae A 1 0 Cleridae A 0 1 Chrysomelidae A 0 1 Curculionidae A 0 1 B 0 1 C 1 0 D 1 0 Staphylinidae A 1 3 B 3 2 C 3 0 D 2 0 E 1 2 F 0 2 G 4 4 H 1 0 Scolytidae A 1 0 Unidentified A 0 1 Unidentified B 1 0 Hymenoptera Formicidae A 16 15 B 1 1 C 51 1 D 0 2 E 1 0 Total insects 167 98 Total beetles 73 69
Table 2. Margalef indices of richness for all ground dwelling insects trapped in paired gap and forest sites. A sign test did not find a significant difference in these values at an Â† level of 0.05. Site D mg gap D mg forest Sign A 4.37 2.47 + B 4.80 3.00 + C 2.67 2.38 + D 2.34 1.34 + E 2.34 3.68 F 1.44 2.17 Table 3. Margalef indices of morphospecies richness of Coleopterans trapped in paired gap and forest sites. A sign test found gap and forest sites to be significantly different in morphospecies richness Â† level 0.05 Site D mg gap D mg forest Sign A 3.61 1.77 + B 3.40 2.34 + C 1.56 1.80 D 1.82 0.39 + E 2.40 2.06 + F 1.44 0.56 + Table 4. Evenness HÂ€/ln S of ground dwelling insects trapped in gap and forest sites. A sign test found no significant difference in evenness in gap and shade sites at an Â† level of 0.05 Site Evenness Gap Evenness Forest Sign A 0.403 0.391 + B 0.414 0.385 + C 0.380 0.383 D 0.350 0.324 + E 0.208 0.390 F 0.434 0.363 + Table 5. Evenness HÂ€/ln S of Coleopterans in gap and forest sites. A sign test found a significant difference in evenness in gap and forest sites Â† level 0.05 Site Evenness Gap Evenness Forest Sign A 0.424 0.367 + B 0.409 0.397 + C 0.352 0.374 D 0.395 0.170 + E 0.421 0.382 + F 0.434 0.282 +