1 Hymenoptera diversity and abundance in fragmented and continuous forest in San Luis, Costa Rica Katie MacDiarmid Department of Environmental Science, University of Oregon ABSTRACT Fragmentation of forest habitats can have a negative impact on the dive rsity of many species. When fragmentation impacts pollination, it can result in major changes to the composition of forest plant communities as well as a reduction in fruit production for nearby agriculture. I studied Hymenoptera diversity in a four hect are fragment and a continuous forest using Euglossine traps as well as a mixture of honey and water to attract Hymenoptera species. Hymenoptera diversity did not differ significantly between the two habitats but abundance of Hymenoptera, Euglossine, and A pidae were all significantly greater in the continuous forest site. This indicates that continuous forest supports a larger, more stable population of Hymenoptera species, thereby supplying steadier pollination services to the forest and nearby farms . RESUMEN La fragmentaciÃ³n de los bosques puede tener un impacto negativo en la diversidad de muchas especies. Cuando la fragmentaci Ã³n impacta procesos como la pol inizaciÃ³n, puede causar cambios grandes de la composiciÃ³n de especies de plantas en las comu nidades del bosque. AdemÃ¡s podrÃa ocasionar una reducciÃ³n en la producciÃ³n de frutos en los cultivos aledaÃ±os . EstudiÃ© la diversidad de HimenÃ³ptera en un fragmento de bosque y un bosque continuo usando trampas de Euglossinos, ademÃ¡s de una mezcla de miel y agua. La diversidad de HimenÃ³ptera no fue diferente en los dos hÃ¡bitats . Sin embargo, la abundancia de HimenÃ³ptera , Euglossinos, y la famili a Apidae fue mÃ¡s alta en el sit io del bosque continuo. Esto indica que el bosque continuo mantiene poblaciones mÃ¡s gra ndes y mÃ¡s estables de especies de HimenÃ³ptera . Por lo que el abastecimiento de polinizadores a lo s cultivos es mÃ¡s estable en sitios prÃ³ximos al bosque continuo. INTRODUCTION Forests throughout the tropics are experiencing fragmentation as patche s are cleared for agriculture and development Aizen and Feinsinger 1994a. A growing body of evidence is showing that this practice significantly affects forest communities Aizen and Feinsinger 1994a, Aizen and Feinsinger 1994b, Powell and Powell 1987, Sekercioglu et al 2001. Fragmented habitats differ from continuous forests in several crucial ways. Fragments are discontinuous with reduced area and a greater edge to interior ratio. Wind
2 and insolation are stronger in fragments, putting organisms at a greater risk for desiccation Aizen and Feinsinger 1994a. Consequently, susceptible species will suffer in fragments, potentially reducing diversity and abundance in these harsher habitats. Studies of different species have shown that many species suffe r when their forest habitat is fragmented Powell and Powell 1987, Aizen and Feinsinger 1994a. For example, a study by Sekercioglu et al. 2001 found significantly fewer insectivorous birds in the understory of fragments compared to continuous forests. They reported that the best determinant of success by a bird species in fragments was the ability to disperse through deforested habitats Sekercioglu et al. 2001. By changing the conditions of the habitat, fragments can add new selective pressures for species, favoring those that can adapt to the new conditions. Changes in species composition have a stronger impact on the forest when they disrupt pollination and seed dispersal mutualisms Aizen and Feinsinger 1994a. When pollinator populations change , this can impact both native plant species and nearby crops Aizen and Feinsinger 1994b because diversity of pollinators may impact pollination and consequently seed and fruit set Ricketts 2003. A study by Ricketts et al. 2004 showed that coffee pla nts near large forest patches greater than 46 ha produced more fruit than those located far from forest. Pollination studies showed that the increase in production was due to the pollination services received from the nearby forest Ricketts et al. 2004 . However, it is still unclear if small fragments will yield the same benefit. This study addresses the effects of fragmentation on the abundance and diversity of Hymenoptera. Hymenopterans have a large impact on the forest community because they provi de pollination services through bees Apidae and biological control of agricultural pests through wasps Kean and Barlow 2001. My purpose was to determine how fragmentation impacts Hymenoptera populations. My hypothesis was that Hymenoptera diversity an d abundance would differ between the four hectare fragment and the continuous forest. I predicted that diversity and abundance would be greater in the continuous forest. The results of my study will provide insight into what land use strategies are necess ary to preserve forest Hymenoptera species for conservation and agricultural benefits. MATERIALS AND METHODS My two study sites were located in San Luis, a farming community in Puntarenas, Costa Rica. The farms in San Luis are bordered by either con tinuous forest or forest fragments. The continuous forest site was located near the e d g e of a forest that is contiguous with the Monteverde Cloud Forest Preserve. The fragment site, located less than one km away from the continuous forest, was a four hec tare patch situated behind a series of small coffee farms collectively known as Finca la Bella. In each site, I placed my traps 100 m in from the edge of the forest. To catch Euglossine bees, I hung two commercial, yellow euglossine bee traps in each sit e. The traps were hung from branches at eye level 1.5 m and were located roughly two meters apart. I placed blotter paper covered with eugenol scent inside each trap. I visited each site in the morning for 11 days between October 26 and November 15. Each day I took three samples of the number of bees present inside or outside the trap. I recorded the bees present when I arrived, ten minutes later, and again
3 after another ten minutes. I preserved a sample of each species in a vial of alcohol and late r identified the species to the genus level. To study overall Hymenoptera diversity and abundance, I visited each site six times in the morning from November 8 to November 15. In each site, I selected two palm branches at eye level 1.5 m roughly three meters apart. I sprayed the underside and topside of the leaves with a mixture of one part honey to four parts water. Ten minutes later, I observed each branch for one minute and recorded the number of each Hymenoptera species present. I repeated this two more times for a total of three observations per site each day. I used the Shannon Weiner diversity index to calculate the Shannon Weiner diversity indices for the continuous forest and the fragment. I then used a t test to determine if the indices were significantly different. I used three separate chi squared tests to compare the abundances of Hymenoptera, bees, and Euglossine bees in the continuous and fragment sites. RESULTS I observed nine species in the continuous forest site and found al l but two of the same species in the fragment site Fig 1. Apis mellifera and the clear wing morphospecies were not found in the fragment site Fig 1. No species were found only in the fragment and all except two species were observed more often in the continuous site Fig 1. Eulaema sp. was observed only once in each habitat Fig 1. The yellow morphospecies of Ichneumonidae was the only species that was more abundant in the fragment site with ten sightings in the fragment and seven in the continuou s site Fig 1. Hymenoptera diversity was not significantly different between continuous forest H = 0.659 and fragmented forest habitats H = 0.631 t test, t = 0.343, df = 94.31, p>.05, Fig 1. Hymenoptera abundance was significantly greater in con tinuous forest compared to the fragment Ã°c 2 = 100.77, df =1, p<.05, Fig 2 as was total bee abundance Ã°c 2 = 19.932, df =1, p<.05, Fig 3. Euglossine bee abundance was significantly greater in the continuous forest site Ã°c 2 =7.136 df =1, p<.05, Fig 4. DISC USSION I expected the diversity and abundance of Hymenoptera to be greater in continuous forest than in fragmented forest but I found no significant difference. I may have found no significant difference in diversity because my sites were located within 100 meters of the edge of the forest. I chose to sample from the edge to accurately sample the Hymenoptera populations available for pollinating nearby crops. However, the edge habitat of a forest contains some of the same conditions, such as increased w ind and insolation that are found in fragments Aizen and Feinsinger 1994a. This could have altered the community on the edge of the continuous forest. However, it is also possible that not enough time has passed since fragmentation to show all its ef fects on the forest. The San Luis area was first cleared for agriculture 50 60 years ago. Often, forest communities take longer than this to fully adapt to a new pressure; it is possible that diversity may decrease in this fragment in the future. This possibility is supported by the higher abundance of Hymenoptera, Apidae, and
4 Euglossine that I found in the continuous site. These results indicate that fragmentation may lead to smaller populations of Hymenoptera species. Due to genetic drift, a reduced population size is likely to lead to reduced genetic variability, which in turn leads to less stable populations with a greater likelihood of extinction Aizen and Feinsinger 1994a. Some of the observed Hymenoptera species may later experience local ext inctions in this fragment due to their decreased abundance. This would reduce the local diversity of Hymenopterans in the fragment. In the future, the continuous forest may contain greater Hymenoptera diversity, especially if further fragmentation of the four hectare patch occurs. It remains to be seen whether fragmentation will ultimately lead to a decrease in Hymenoptera diversity. However, we do know that as conditions change with fragmentation, species that do well in disturbed areas, such as me mbers of the Ichneumonidae family and Apis mellifera Gauld 1995 and Hanson et al.1995, will have an advantage over species that suffer in fragments, such as Euglossians Powell and Powell 1987. This is not only true for Hymenoptera species; as an order with many plant animal interactions, a change in either Hymenoptera abundance, diversity, or composition is likely to mean changes for plant communities as well Aizen and Feinsinger 1994b. Plants that are dependent on declining pollinator species will suffer declines themselves. Euglossine bees are significant pollinators for over 30 tropical plant families including Orchidaceae, Solanaceae, Bignoniaceae, and Luguminaceae. If euglossine bee populations decline in fragmented forest, as it seems they ha ve done in San Luis, these plant families are likely to follow Powell and Powell 1987. It is likely that this change in pollinator composition will ultimately result in a less diverse community of plants. Aizen and Feisinger s study on forest fragment ation revealed that 13 out of the 16 plant species studied showed declines in pollination levels from continuous forest to fragments. Seed production also decreased for many of these species in the fragmented forest patches as a result of reduced pollinat ion. This suggests that a smaller group of species will contribute more seeds to the seed set. These few species will eventually come to dominate the plant community in fragments Aizen and Feinsinger 1994a. In a farming community like San Luis, preserv ation of forest diversity is not just for conservation s sake alone. The pollination services provided by forest bee species are highly valuable to nearby agricultural production Ricketts et al 2004. More diverse pollinator populations stabilize pollina tion services and help buffer against declines in any one species Ricketts 2004. Taylor Ricketts has shown that coffee farms near large patches of forest 46 and 111 ha produce more and higher quality coffee than farms near a narrow riparian strip 2.5 km long by 30 70 m wideRickets 2004. In addition, many species of wasps can also provide ecosystem services as biological control agents for agricultural pests Kean and Barlow 2001. The results of my study demonstrate that small forest fragments, ev en within one km of continuous forest, cannot support the same size of Hymenoptera populations that continuous forest can. This indicates that communities like San Luis cannot continue to fragment their surrounding forests to produce more land for agricul ture. As a result, farmers may lose money in the end due to decreased crop yield following decreased pollinator abundance. Further directions for research include repeating this study with a
5 larger sample size in sites further into the forest to more accu rately sample Hymenoptera populations in the forest interior. ACKNOWLEDGEMENTS Thanks to Milton Brennis for letting me use his forest patch and walking me around. Thanks to the Leit Ã³ n familiy for use of their forest as well. Also thanks to Tanya for he r help identifying and catching bees. Thanks to Cam and Tom for answering my questions and always being available for help. Thanks to Alan and Karen for their advising help. LITERATURE CITED Aizen, M.A. and P. Feinsinger. 1994a. Forest fragmentation, pollination, and plant reproduction in a Chaco dry forest, Argentina. Ecology 75 : 330 351 Aizen, M.A. and P. Feinsinger. 1994b. Habitat fragmentation, native insect pollinators, and feral honey bees in Argentine Chaco Serrano . Ecological Applicatio ns 4 : 378 392 Gauld, I.D. and S.R. Shaw. 1995. The Icheneumonidae families. In: The Hymenoptera of Costa Rica, P.E. Handson and I.D. Gauld, eds. Oxford University Press, New York, NY, 890 pp Griswold, T., F.D. Parker, and P.E. Hanson. The bees Api dae In: The Hymenoptera of Costa Rica , P.E. Handson and I.D. Gauld, eds. Oxford University Press, New York, NY, 890 pp Kean, J.M. and N.D. Barlow. 2001. A spatial model for the successful biological control of Sitona discoideus by Microctonus aethiop oides . Ecology 38 : 162 169 Powell, H.A. and G.V.N. Powell. 1987. Population dynamics of male euglossian bees in Amazonian forest fragments. Biotropica 19 : 176 179 Ricketts, T.H. 2003.Tropical forest fragments enhance pollinator activity in nearby crops. Conservation Biology 18 : 1262 1271 Ricketts, T. H., G.C. Daily, P.R. Ehrlich, and C.D. Michener. 2004. Economic value of tropical forest to coffee production. Conservation Biology 15 : 378 388 Sekercioglu, C.H., P.R.vEhrlich, G. C.Daily, D.Aygen, D. Go ehring, and R.F. Sandi. 2002. Disappearance of insectivorous birds from tropical forest fragments. Proceedings of the National Academy of Sciences of the United States : 99 : 263 267
6 0 20 40 60 80 100 120 140 Apidae Euglossia sp. Apidae Eulaema sp. Apidae Apis mellifera Apidae Parthemona sp. Ichneumonidae yellow morph Ichneumonidae clear wing morph Ichneumonidae green morph Scolidae Vespidae Hymenoptera families observed Total number of individuals observed # Continous # fragment Fig 1. Hymenoptera species observed in cont inuous and fragment sites. Hymenoptera diversity was not significantly different between sites t = 0.343, df = 94.31, p>.05.
7 0 50 100 150 200 250 300 350 Continuous Fragment Site Total number of Hymenoptera individuals observed Fig 2. Total number Hymenoptera observed in continuous and fragment sites. Hymeno ptera abundance was significantly greater in the continuous forest site Ã°c 2 = 100.77, df =1, p<.05.
8 0 20 40 60 80 100 120 continuous fragment Site Total number of Apidae individuals observed Fig 3. Total number of Apidae individuals observed in continuous and fragment sites. Apidae abundance was signific antly greater in the continuous forest site Ã°c 2 = 19.932, df = 1, p<.05.
9 0 10 20 30 40 50 60 70 80 continuous fragment Site Total number of Euglossinae individuals observed Fig 4. Total number of Euglossinae individuals observed in continuous and fragment sites. Euglossinae abundance was significantly greater in the cont inuous site Ã°c 2 = 7.136, df = 1, p<.05.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00174
La diversidad y la abundancia de Hymenoptera en los bosques fragmentados y continuos en San Luis, Costa Rica
Hymenoptera diversity and abundance in fragmented and continuous forest in San Luis, Costa Rica
Fragmentation of forest habitats can have a negative impact on the diversity of many species. When fragmentation impacts pollination, it can result in major changes to the composition of forest plant communities as well as a reduction in fruit production for nearby agriculture. I studied Hymenoptera diversity in a four hectare fragment and a continuous forest using Euglossine traps as well as a mixture of honey and water to attract Hymenoptera species. Hymenoptera diversity did not differ significantly between the two habitats but abundance of Hymenoptera, Euglossine, and Apidae were all significantly greater in the continuous forest site. This indicates that continuous forest supports a larger, more stable population of Hymenoptera species, thereby supplying steadier pollination services to the forest and nearby farms.
La fragmentacin de los hbitats forestales puede tener un impacto negativo en la diversidad de muchas especies. Cuando la fragmentacin impacta los procesos como la polinizacin, puede causar cambios grandes de la composicin de especies de plantas en las comunidades del bosque. Adems podra ocasionar una reduccin en la produccin de frutos en los cultivos aledaos. Estudie la diversidad de Hymenoptera en un fragmento de cuatro hectreas de bosque continuo usando trampas de Euglossines, adems de una mezcla de miel y agua para atraer a las especies de Hymenoptera.
Text in English.
Hymenoptera--Costa Rica--Puntareans--Monteverde Zone
Hymenoptera--Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology 2006
Ecologa Tropical 2006
t Monteverde Institute : Tropical Ecology