Diversity, abundance, richness, and composition of avian communities and avian foraging behavior at Ficus pertusa Heather Whelpley Biological Aspects of Conservation, University of Wisconsin Madison ABSTRACT This study investigated differences in success between fruiting Ficus pertusa trees in a single pasture in Monteverde, Costa Rica. Success was qualitatively measured by the abundance, richness and diversity of avian dispersers and the total number of seeds dispersed as measured by the number of visits by birds that swallow fruits. Each of the four sites was observed for five days from 7:00am Â€ 11:00 am. Significant differences in diversity were found in five of seven comparisons between the sites, but nei ther diversity nor number of bird visits was related to dbh or crop size of the tree. Significant differences were found in the number of swallowing birds at each site, affecting the dispersal success of each tree. Proximity to forest patches and crop size may affect the overall success of a tree and differences in success are due to the interaction of several factors that change over very small spatial scales. RESUMEN Este estudio investigo las diferencias del xito entre cuatro Ficus pertusa con frutas en un potrero en Monteverde, Costa Rica. El xito fue medido por la abundancia, la riqueza de especies y la diversidad de pjaros que comen frutas y tambin el nmero de las semillas dispersaron medido por el nmero de visitas de pjaros que t ragan frutas. Cada de los cuatro sitios fue observado por cinco das, de las siete hasta las once de la maana. Diferencias significativas con la diversidad fueron encontradas en cinco de los siete comparaciones entre los sitios, pero ni la diversidad ni el numero de los visitas de pjaros estuvo correlacionado con el dimetro del arbol o el numero de las frutas en el arbol. Diferencias significativas fueron encontradas en el nmero de pjaros en cada sitio, afectando el xito de cada arbol. Es posible que la proximidad a fragmentos de bosque y el nmero de frutas del arbol afecten el xito del arbol. Diferencias en el xito son causadas por la interaccin de varios factores que cambian sobre espacios pequeos, afectando el xito de un arbol. INTRODUCTION The genus Ficus with over 900 species worldwide, is among the most widespread tropical genera Janzen 1979. It is a keystone resource for many birds and mammals because trees produce fruits asynchronously throughout the year, guaranteeing a food source during the dry season Janzen 1979, Terborgh 1986. Terborgh 1986 goes so far as to say that tropical ecosystems would collapse without the presence of fig trees. All Ficus species have extremely large seed shadows due to the wide diversity of their dis persers Janzen 1979, Jordano 1983. In Monteverde, Costa Rica, Ficus species receive a disproportionate number of frugivores when compared to other common genera
in Monteverde Wheelwright, et al. 1984. They are thought to attract diverse and abundant di spersers because the seeds have no secondary compounds and they are a constant food source Janzen 1979. Furthermore, because the fruits are soft, there is no limitation on the gape size of dispersers; animals can peck or mash the fruits, in addition to s wallowing them whole Janzen 1979, Wheelwright, 1985. Ficus pertusa the focal species in this study, is found in both forest and open areas and is common in pastures and fragments in Monteverde. It is either a hemiepiphyte or an independent tree and gro ws 10 15 meters tall. Populations fruit asynchronously and each tree produces a crop approximately once a year. Crop size varies from 100 200,000 fruits per tree and the fruiting period lasts between three weeks and three months Bronstein and Hoffmann 198 7. Fruits are dispersed by a variety of vertebrates, especially avian species, and are not dispersed by bats. Bronstein and Hoffmann 1987, Dinerstein 1983 in Bronstein and Hoffmann 1987. The foraging behavior of dispersers is one factor determining the overall success of F. persusa. Birds that peck at fruits are the least effective dispersers since they eat few, if any, seeds. Birds eating fruits by mashing disperse some seeds, however a large portion of each fruit is dropped below the parent tree, allo wing little hope for germination. Swallowing birds are likely the best dispersers since they donÂt drop any seeds below the parent tree and disperse seeds through defecation at other locations. Peck was defined as the action of a bird foraging and eating a small part of the fruit without removing it from the tree. Mashing occurred when a bird removed a fruit from a tree and chewed it while perching, allowing some pieces of the fruit to fall to the ground. Swallowing was defined as the removal and ingestion of a whole fruit. Spatial and temporal variations between trees are very important in determining the diversity, richness and composition of dispersers at F. pertusa Bronstein and Hoffmann 1987. Temporal variation can be defined as changes between diffe rent times, such as seasons, and spatial variation is changes over an area such as ecosystems. This study investigated the differences in the successes of an individual tree without the influence of these two factors. Success was qualitatively measured by the total abundance of bird visits, the diversity and richness of dispersers and the quantity of seeds dispersed as measured by the number of visits by swallowing birds. MATERIALS AND METHODS This study was conducted in the Monteverde region of Costa Rica 1450 meters, which is located on the northern Pacific slope of the Cordillera de Tilarn. One pasture was chosen for its abundance of Ficus pertusa trees planted for windbreaks and borders. Four tree sites with ripe fruit were selected in the pasture. Sites One, Two and Three were located beside each other in a line. Site One was continuous with a small forest patch. Site Four was located on the other side of the pasture, approximately 50 met es away and was growing in a windbreak of pine trees. Site One had two trees with their branches entwined, while sites two, three and four were single trees.
Data were collected from October 25, 2000 Â€ November 12, 2000. Each site was observed for five da ys, from 7:00am Â€ 11:00am and Sites One, Two and Three could be observed simultaneously due to their close proximity. During the observation periods, each bird was identified and the site and foraging behavior of the bird were noted. Foraging behavior was recorded as a peck, mash, swallow, glean, sally and/or perch. The crop size of each site was estimated by counting the number of fruits on an accessible branch and multiplying it by the number of similarly sized areas on the tree. Diameter at breast heigh t dbh was also measured at each site. Dbh was measured at Site One by adding the individual dbh of each tree and treating it as one measurement. To analyze diversity, the abundance of each species was counted and the Shannon Weiner Diversity Index HÂ Zar 1984 and Evenness were calculated for each site. A modified t test on diversity was used to determine if differences between the sites were significant. Furthermore, SorensonÂs Index was used to assess similarity of species composition between the si tes. For both the t test and SorensonÂs index, pairwise comparisons were calculated between all sites. Also, Sites One, Two and Three were tested against Site Four to determine if birds see a close grouping of fruiting trees as a single target of food. To analyze the relationship between bird species and foraging behavior the average weight of birds in the categories peck, mash and swallow was determined and compared using a one way ANOVA. Furthermore, Chi Square tests were used to compare the abundance of visits in all foraging categories and the abundance of the foraging categories peck, mash, and swallow between sites. RESULTS There was great variation in the number of visits, species richness and diversity between the four sites in the pasture Table 1. Neither the number of visits nor diversity was related to crop size or dbh Figures 1 4. However, Site One, with the largest tree and the largest crop size, had a far greater number of visits and species richness than the three smaller trees 57% of t he 754 total bird visits and 18 of 22 recorded species. Further, there were significant differences in the diversity of avian species between all sites except Sites One and Four and Sites Two and Four modified t test, p Â€ values < .05, Table 2. Also, wh en Sites One, Two and Three were tested as a single data point against site Four, there was significantly different diversity p value < .001, Table 2 Zar, 1984. When analyzing the similarity of species composition, pairwise comparisons between each si te were calculated as well as a comparison of Sites One, Two and Three to Site Four. The values of SorensonÂs index ranged from 0.50 to 0.71 and five of the seven comparisons gave values between 0.50 and 0.55, indicating that species composition changed li ttle between the sites Table 2. Similarity was not related to distance between the trees, i.e., trees located closer to one another did not show a greater similarity of species composition. The differences between weights of bird in a foraging categorie s peck, mash and swallow were analyzed using a one way ANOVA. The average weight of birds pecking,
mashing, and swallowing was 8.5 g, 20.8 g and 84.2 g, respectively. A significant difference F value = 959.39, df = 2, p < .001 between the average weight of birds in each category was found Table 3. Birds foraging for fruit comprised the majority of bird visits, with 745 98% of the 758 total bird visits. Frugivores also showed the greatest richness at Ficus pertusa with 13 of the 17 species of identif ied birds. Birds gleaning, sallying, and perching were only 13 2% of the total bird visits and six species. There was overlap between categories because HoffmanÂs Woodpecker was observed gleaning and swallowing and the Black Throated Green Warbler was s een pecking and gleaning. A Chi Square was calculated between total number of visits of birds eating fruit and birds sallying, gleaning and perching and a significant difference was found X = 1120.27, df = 1, p value < .01. There were large difference s in the behavior of frugivorous birds in F. pertusa Birds pecking at the fruit comprised 57.6% of total bird visits. However, the species richness in this category was very low s = 2 and Tennessee Warblers made up 99.8% of the total visits. Swallowing birds had the highest richness s= 8 and comprised 26.0% of total visits while mashing birds comprised 14.6% of visits with a richness of four species. Using a Chi Square test a significant difference in number of bird visits was found between the three c ategories of frugivore behavior X = 706.89, df = 5, p < .01. There were also significant differences between the sites for the number of birds in each foraging category peck: X = 248.8, df = 3, p < .001; mash: X = 104.3, df = 3, p < .001; swallow: = 112.1, df = 3, p< .001, Table 4. Site One and especially Site Three were dominated by Tennessee Warblers pecking birds, although Site One also had a high number of swallowers. Site Two was more evenly distributed between the three foraging categories an d Site Four was dominated by swallowing birds, such as Clay Colored Robins, Blue Crowned Motmots and Mountain Elaenias. DISCUSSION An individual treesÂ success can be measured by the abundance, richness and diversity of its dispersers, as well as the number of seeds dispersed. When combining these factors, Site One was the most successful tree. It had the highest abundance of visits n = 368, the highest richness s = 18 and relatively high diversity H Â = 0.704006. It also received the highest number of visits n= 109 and richness s = 8 of swallowing birds. There are many possible reasons for Site OneÂs success. First, it was continuous with a small patch of forest. Some birds, such as the Blue Crow ned Motmot, were reluctant to cross open spaces and were found mainly in Sites One and Four which was continuous with pine trees. Second, Site One was continuous with an orange tree and many birds were observed using both the F. pertusa and the orange tr ee. Site Four also had a high measure of success. It had the highest diversity and a very high proportion of swallowing birds visited the tree 49 of 74 total visits. It did have a low number of bird visits, however, which may be due to its isolation fro m other fruiting F. pertusa Birds were observed hopping between Sites One, Two and Three, but
this couldnÂt happen at Site Four. One possible explanation for its high diversity and number of swallowers is its proximity to pine trees, as explained above. Site two had a much lower measure of success than both Sites One and Four. Although it was fairly even E = .705, it had lower diversity. Twenty nine of its 81 bird visits were from swallowers, a decent proportion, but a fairly low total number. Its total abundance was likely constrained by its size; there was not room for more than five or six birds on the tree at once. In contrast, Sites One and Three often had ten or more birds in their branches at one time. The lowest success was found at Site Three. Despite its high number of total visits n = 230, its overall diversity was extremely low HÂ = 0.225745 because it was dominated by Tennessee Warblers. These birds peck at fruit, dispersing few seeds. This disproportionate number of Tennessee Warblers m ay be due to its proximity to the tree they used for perching. Tennessee Warblers almost always flew from an orange tree across the pasture into Site Three. They would often pass into Sites One or Two after foraging in Site Three, but they would rarely fly directly into these trees. There was also a very low abundance of swallowers at Site Three n = 10, therefore the tree likely had limited dispersal. Overall success of a given tree is due to many factors. This study has indicated that proximity to fores t patches and other fruiting trees may be a determining factor for a treesÂ success. Crop size did not directly relate to diversity or abundance, but there was a relationship between crop size and overall success with the more successful trees having a gre ater number of fruits. Although dbh did not relate to success, this is not an accurate measure of crown size and the relationship between tree size and success should be further studied. Both Jordano 1983 and Bronstein and Hoffmann 1987 found that spa tial and temporal variation were important factors in determining the abundance, richness and diversity of avian dispersers at Ficus trees. However, neither study assessed whether abundance and diversity differed between trees with little or no spatial and temporal variation. This study found that, in the absence of these two factors, diversity, abundance and richness of birds was different between otherwise similar sites. It has been demonstrated that there are multiple factors involved in the overall succ ess of a F. pertusa However, the reasons that birds prefer foraging in certain fruiting F. pertusa over other nearby trees is still unclear. Future studies could address the question further, possibly doing more careful analysis of proximity to forest pat ches and other fruiting trees. However, the main conclusion is that several factors interact to attract dispersers to F. pertusa and these factors are able to change over very small scales, thus affecting the overall success of a given tree. ACKNOWLEDGEME NTS I would like to thank my advisor, Karen Masters, for her ideas and help formulating in this study. Tim Kuhman generously offered his time to read and criticize the contents of this paper. I would also like to thank Tina and Ricardo Sander for the daily use of their land, without which this study would not have been possible.
LITERATURE CITED Bronstein, J.L., K. Hoffmann. 1987. Spatial and temporal variation in frugivory at a Neotropical fig, Ficus pertusa Oikos 49: 261 268. Dinerstein, E. 198 3. Reproductive ecology of fruit bats and seasonality of fruit production in a Costa Rica cloud forest. Ph. D. Diss., Univ. of Washington, Seattle, WA. Janzen, D.H. 1979. How to be a fig. Annual Review of Ecology and Systematics 10: 13 51. Jordano, P. 1983. Fig seed predation and dispersal by birds. Biotropica 15: 38 41. Stiles, G.F. and A.F. Skutch. 1989. A Guide to the Birds of Costa Rica. Cornell University Press: Ithaca, New York. Terborgh, J. 1986. Keystone plant resources in the tropical f orest, In: Soule, M. E. ed., Conservation biology: The science of scarcity and diversity. Sinauer, Sunderland, MA, pp. 330 344. Wheelwright, N.T., W.A. Haber, K.G. urray, C. Guinfon. 1984. Tropical fruit eating birds and their food plants: a survey of Costa Rican lower montane forest. Biotropica 16: 173 192 Wheelwright, N.T. 1985. Fruit size, gape width, and the diets of fruit eating birds. Ecology 66: 808 818. Zar, J.H. 1984. Biostatistical Analysis. Prentice Hall: Englewood Cliffs, New Jersey.
Table 1. Summary of all birds seen at each F. pertusa. Total abundances of birds at each site total abundance of each species, richness, Shannon Weiner Diversity Index and Evenness were included for comparison. Richness and total bird visits doesnÂt no t correlate directly to diversity. See Appendix for abbreviations. Species Name Site 1 Site 2 Site 3 Site 4 Total Abundance Tenn warbler 177 37 204 17 435 C c robin 79 27 8 27 141 Y t euphonia 50 10 11 1 72 B g tanager 20 5 3 4 32 B c motmot 12 0 1 9 22 Mtn elaenia 1 0 0 9 10 Br jay 5 1 0 3 9 H woodpecker 6 0 0 0 6 W tanager 3 0 1 2 6 B b flycatcher 2 1 1 0 4 T warbler 3 0 0 0 3 Y f grassquit 2 0 0 0 2 Soc flycatcher 2 0 0 0 2 Flycatcher, spp. a 2 0 0 0 2 B w warbler 1 0 0 0 1 R b grosbeak 1 0 0 0 1 Y warbler 0 0 1 0 1 B t g warbler 0 0 0 1 1 Bird A a 1 0 0 0 1 Pewee, spp. a 1 0 0 0 1 Warbler, spp. a 0 0 0 1 1 W t robin 0 0 0 1 1 Total Abundance 368 81 230 75 754 Richness 18 6 8 11 22 HÂ€ 0.704006 0.548422 0.22575 0.792628 Evenness .5840424 .70477057 .24997 .72511119 Crop Size 10,200 2,100 500 3,300 Dbh cm 43.29 28.33 23.87 15.27 a Species could not be determined
Table 2. Pairwise comparisons of sites. All sites were roughly similar, with SorensonÂs Index ranging from 0.50 to 0.71. Five of the 7 comparisons showed significant differences diversity between the sites, showing that trees with little or no special and temporal variation show different spe cies abundance and richness. SorensonÂs Index HÂ t Values Degrees of Freedom HÂ p values Site 1 vs. Site 2 0.50 3.38 1.82E+14 <.001 Site 1 vs. Site 3 0.53 6.64 280.46 <.001 Site 1 vs. Site 4 0.55 0.50 75.75 Ns Site 2 vs. Site 3 0.71 4.25 254.80 <.001 Site 2 vs. Site 4 0.53 1.35 81.04 Ns Site 3 vs. Site 4 0.63 3.00 82.01 <.005 Site 1, 2, 3 vs. Site 4 0.55 7.36 227.30 <.001 Table 3. Summary of foraging behavior of birds in F. pertusa. Average weight is based on the average weight per visit, more heavily weighing the more abundant birds in the average weight per category. See Appendix for abbreviations. Foraging Behavior Peck Mash Swallow Species Common Name Tenn Warbler B t g B g T anager Y t Euohonia R b Grosbeak W Tanager C c robin B c motmot H woodpecker Soc flycatcher B b flycatcher Mtn elaenia Br. jay Flycatcher, spp. a Abundance 437 111 199 Richness 2 4 8 Average Weight g 8.5 20.8 84.2 a Species could not be determined Table 4. Summary of foraging behavior at each site in the pasture. Site One has the highest abundance of swallowers, the most efficient dispersers, although Site Four has the highest proportion of swallowes. Both trees are likely being dispersed well. In contrast, Site Three has a high number of pecking birds, the worst dispersers, and low number of swallowing birds. Thus, it likely has limited dispersal. Site Two is being dispersed better than Site Three, due to its higher number of swallowers, but w orse than Sites One and Four. Foraging Behavior Site One Site Two Site Three Site Four Total Chi Square p value Peck 177 37 204 18 437 Ash 74 15 15 7 111 Swallow 109 29 10 49 197 Unknown 8 0 1 1 10 Total 368 81 229 75 754
Appendix Scientific and common names of birds identified at F. pertusa including abbreviations used in tables. Foraging behavior also included. Scientific Name Common Name Abbreviation Foraging Behavior Family Motmotidae Motmotus momota Blue Crowned Motmot B c motmot Swallow Family Picidae Melanerpes hoffmannii HoffmanÂs Woodpecker H woodpecker Swallow, Glean Family Tyrannidae Myiozetetes similis Social Flycatcher Soc flycatcher Swallow Megarhynchus pitangua Boat Billed Flycatcher b b flycatcher Swallow Elaenia franzii Mountain Elaenia Mtn elaenia Swallow Family Muscicapidae Turdus assimilis White Throated Robin W t robin Swallow Turdus grayi Clay Colored Robin C c robin Swallow Family Corvidae Psilorhinus morio Brown Jay Br jay Swallow Family Praulidae Vermivora peregrine Tennessee Warbler Tenn warbler Peck Mniotilta varia Black and White Warbler B w warbler Glean Dendroica petechia Yellow Warbler Y warbler Perch Dendroica townsendi TownsendÂs Warbler T warbler Perch Dendroica virens Black Throated Green Warbler B t g warbler Peck, Glean Family Thraupidae Euphonia hirundinacea Yellow Throated Euphonia Y t euphonia Mash Thraupis episcopus Blue Gray Tanager B g tanager Mash Piranga ludovician a Western Tanager W tanager Mash Family Emberizidae Pheuciticus ludovicianus Rose Breasted Grosbeak R b grosbeak Mash Tiaris olivacea Yellow Faced Grassquit Y f grassquit Perch Family Ramphastidae a Aulacorhybchus prasinus Emerald Toucanet Em toucanet Ramphastos sulfuratus Keel Billed Touca K b toucan Family Icteridae a Icterus g. galbula Northern Oriole N oriole Psarocolius montezuma Montezuma Oropendula M oropendula a Species seen foraging on F. pertusa outside of the observation period.