! ! "#$%&''()*!+**'! ! ! ,-./! 0 ! Euglossine bee composition along an altitudinal gradient on the Caribbean and Pacific Slop es of the TilarÂ‡n Mountains, Costa Rica Sakoya Hart Department of Ecology and Evolutionary Biology, University of California, Santa Cruz EAP Tropical Biology & Conservation, Spring 2018 8 June , 2018 Abstract Bees not only contribute significantly to ecosystem biodiversity, but as pollinators they are responsible for facilitating the reproductive processes of many tropical forest plants . As tropical forests are fragmented and destroyed , euglossine bees are under increasing threat of habitat loss and population declines. Understanding their altitudinal and geographical distributions is a critical aspect needed for future conservation efforts. I sampled at four different altitudes along the Pacific slope and three different altitudes along the Caribbean slope . I used three different chemical scents : Methyl Salicylate, Eugeno l, and Cineole , as baits to attract male E uglossine bee s to sample their populations . The chemical Methyl Salicylate had the highest visitation rate by the highest number of species. Bee visitation was higher on warm, clear days, early in the morning. I captured and identified 128 Euglossine bees representing 11 species at various sampling sites across the TilarÂ‡n Mountains of Costa Rica . On both the Pacific and Caribbean slopes , I found a negative correlation between species richness and abundance as altitude increased. The Caribbean slope showed higher abundance and diversity in euglossine bees. ComposiciÂ—n de abejas euglosinas a lo largo de un gradiente altitudinal en las vertientes PacÂ’fico y Caribe de la Cordillera de TilarÂ‡n, Costa Rica Resumen Las abejas no solo contribuyen significativamente a la biodiversidad de los ecosistemas, sino que, como polinizadores, son responsables de facilitar los procesos reproductivos de muchas plantas tropicales. A medida que los bosques tropicales se fragmentan y destruyen, las abejas de las orquÂ’deas se ven cada vez mÂ‡s amenazadas por la pÂŽrdida de hÂ‡bitats y la disminuciÂ—n de la poblaciÂ—n. Comprender sus distribuciones altitudinales y geogrÂ‡ficas es un aspecto crÂ’tico necesario para los futuros esfuerzos de con servaciÂ—n. RealicÂŽ muestreos en tres altitudes diferentes a lo largo de la vertiente PacÂ’fico, dos altitudes diferentes a lo largo de la vertiente Caribe y en el punto mÂ‡s alto en la divisiÂ—n continental de Monteverde, Costa Rica. UtilicÂŽ tres aromas quÂ’mi cos diferentes Salicilato de metilo, Eugenol y Cineol como cebos para atraer a las abejas de las orquÂ’deas macho para obtener una muestra de sus poblaciones. El quÂ’mico Salicilato de metilo tuvo la mayor tasa de visitas por el mayor nÂœmero de especies. Las visitas a
! ! "#$%&''()*!+**'! ! ! ,-./! 1 ! las abejas fueron mayores en los dÂ’as cÂ‡lidos y despejados, temprano en la maÂ–ana. CapturÂŽ e identifiquÂŽ 128 abejas de 11 especies en varios sitios de muestreo en las montaÂ–as TilarÂ‡n de Costa Rica. Tanto en la vertiente del PacÂ’fico como en el Caribe se observÂ— una correlaciÂ—n negativa entre la riqueza de especies y la abundancia a medida que aumentaba la altitud. La pendiente del Caribe mostrÂ— una mayor abundancia y diversidad en las abejas euglosinas. Euglossine bees (tribe Euglossini, su bfamil y Bombinae, family Apidae) are the main bee component in many neotropical forests. Euglossines , otherwise known as orchid bees , play a quintessential ecological role in the complex ecology of tropical forest communities . While searching for resources they are responsible for pollinating thousands of plants, trees, and rare orchids found ranging from up in the forest canopy down to the forest floor at various al titudinal levels (Roubik and Hanson 2004). Male euglossines g ather v aporous chemica ls from orchids among other plants and store them in their hind tibial storage organ to later present these odor bouquets to females for mating purposes (Dressler 1982). Because of this , male euglossines are easily attracted to chemical scents that contain the same compounds as what they would collect in the wild. The floral odors that attract male euglossine bees require warm temperatures to be released into the air. It is probable that euglossine activity is limited at higher elevations because of the overall lower temperatures and biomass of plants they need. Because of this it is speculated that in the lowland habitats of the tropics there is the highest euglossine species richness and abundance (Roubik and Hanson 2004). Understanding a species local abundance and distribution over biogeographical boundaries is vital to habitat restorati on and conservation of wildlife ( Verberk 2011). Proper documentation of Euglossine bee population s , species richness and distribution is necessary for preservation of their ecological role as pollinators (Roubik and Hanson 2004). Understanding bees' adaptability to different habitats and environmental conditions is integral in predicting the future influence that global climate change c ould have . As knowledge on their altitudinal and geographical ranges across the TilarÂ‡n mountains is limited to only a few past UCEAP studies, gathering more data for comparison could provide key insight on these bees' habitat distribution to be utilized for future conservation efforts. The questions I set out to answer were: (1) does elevation affect euglossine abundance and diversity on the Pacific and Caribbean slopes ? and (2) does the species abundance and diversity differ between both slopes? Materials and Methods Study areas My research was conducted during a two week period at the start of the rainy season beginning May 9 to May 25. In order to research Euglossine abundance and diversity along an altitudinal gradient, I presented chemical baits in four different altitudinal ranges on the Pacific slope along with three on the Caribbean slope of th e TilarÂ‡n Mountain Range. In order of increasing elevation , the first study area was located in San Luis Arriba at the University of
! ! "#$%&''()*!+**'! ! ! ,-./! 2 ! Georgia campus along the Camino Real trail (1,064m). The next study area I sampled at was within the Children's Eternal Rainforest Bajo del Tigre Reserve along t he Mirador trail forest (1,253m). The subsequent sampling ar ea was located within the Crande ll Memorial Reserve behind th e Monteverde Institute (1,490m). The first three stud y sites were located within pre montane wet forest. The highest elevation used as a study site was in lower montane rain forest on the continental divide (1,790m). I considered this as an elevation for both t he Pacific and Caribbean slopes. Additionally, I sampled at two different altitudes along the Caribbe an slope, the lowest being in tropical wet forest at Maximo Ramirez's farm in the San Gerardo valley (794m), the second site was in pre montane wet forest along the Congo trail near the San Gerardo Biological Station (1,350m). Euglossine bee abundance was compared between altitudina l ranges on each slope of the TilarÂ‡n mountain range . Additionally, the Euglossine diversity was compared per altitudi nal range on each slope . Chemical preparation, bait presentation, and bee capture I sampled a total of 32 hours, averaging from 4 6 hours at each site over the spans of two to three days. Sampling took place during the estimated period of peak fragrance collection in the morning between the hours of 7:30 a.m. to 12:30 p.m. I used the chemicals Eugenol, Cineole , and Methyl Salicylate as baits to attract male euglossine bees. I hung three pieces of single ply toilet paper, one for each chemical attractant at approximately one meter apart on a line strung between two trees at chest height. I scented every piece of toilet paper with 0.5ml o f chemical fragrance using a syringe. I reapplied 0.5ml of each chemical attractant to the toilet pape r baits every hour. After male e uglossines began collecti ng from the chemical baits, the bee's preference of bait was recorded and the bees were captured using a butterfly net and my bare hands. I placed bees in plastic bags for the duration of my sampling period so as not to recapture bees. I made certain to keep bees out of direct sunlight and gave them fresh air every 30 minutes. After bees were captured and bagged I identified them to genus and species level in the field using the book Ã”The Orchid bees of Tropical America Biology and Field Guide' (Roubik and Hanson 2004). I released all bees that were captured at the end of each sampling period. I compared abundance dividing the number of bees per hour in order to standardize between sampling sites and the different amount of hours sampled. I also recorded the abundance and diversity of bees attracted to each of the three specific chemical scents. Results Over the duration of my study , a total of 172 Euglossine bees were sighted at the chemical baits, 128 were captured and identified. The average ( Â± standard deviation) number of bees per site a t each sampling area was 21.3 Â± 12.8 . The average number of species per site was 4.5 Â± 2.4. A total 13 taxa of Euglossines was identified (11 species and 2 morpho species). The most common species listed in order of decreasing abundance were Euglossa virridissima , Eulaema bombiformis, Euglossa imperialis , on both the Caribbean and Pacific slopes (Table 1). Sites differed in their number of individuals caught per hour (Fig.1) , as well as in their species composition . The lowest elevation on the Caribbean side held the highest species count at nine different species whi le also having the five unique individuals that were captured and identified only at that location. The lowest elevation on the Pacific side had the highest abundance of bees
! ! "#$%&''()*!+**'! ! ! ,-./! 3 ! overall with four different species and had one unique species captur ed only at that location (Table 1) . Species Composition Variation Between Study Sites Using the SÂ¿renson similarity index to compare the differences in species composition between sites, I found that the continental divide compared to the two lowest elevations on both the Pacific and Caribbean slopes provided to have the highest difference in species composition at 33% for San Luis Arriba and 22% for Maximo's. (Table 2.) The two mid altitudinal ranges at Bajo del Tigre and Congo trail had the same species composition at 100%. Altitudinal Variation in Species Abundance on the Pacific and Caribbean Slopes Both the Caribbean and Pacific s lopes had higher abundances of E uglossine bees as the altitude decreased. The Caribbean slope overall had a higher abundance of euglossine bees captured per study site . The Caribbean slope had a higher species count overall. Both the Caribbean and Pacific slopes showed a negative correlation between increasing altitude and total number of individual speci es with the lowest elevation al range sampled having the highest number of individual species. Species preference for specific chemical baits I found that bees overall had the highest preference for Methyl Salicylate, an intermediate preference for Eugenol and the least preference for Cineole across all of my sample sites. Table 1: The total number of individuals per euglossine bee species present at each location. Table 2: Comparison of euglossine bee species composition across all study sites.
! ! "#$%&''()*!+**'! ! ! ,-./! 4 ! Figure 1: The average abundance and species richness of euglossine bees at three elevations on the Pacific and Caribbean slopes. The error bars are illustrating the standard deviation.
! ! "#$%&''()*!+**'! ! ! ,-./! 5 ! Figure 2: The abundance per site of euglossine bees across four different altitudinal ranges on both the Pacific and Caribbean slopes.
! ! "#$%&''()*!+**'! ! ! ,-./! 6 ! Figure 3 : The species composition per site of euglossine bees across four different altitudinal ranges on both the Pacific and Caribbean slopes. Discussion My study showed a notable variation in Euglossine species composition between different altitudes on the Caribbean and Pacific slopes of the TilarÂ‡n Mountain Range in Costa Rica . Additionally , I found that there is both a hig her diversity and abundance of E uglossine bees on the Caribbean slope. Variation in Species Composition on the Pacific and Caribbean Slopes Eulaema bombiformis, Euglossa viridissima, and Euglossa imperialis accounted for the largest portion of the E uglossine bee's composition I found on both s lopes. Studies conducted during fall of 1999 and spring of 2008 by former UCEAP students on both slopes have continuously found these three species to be abundant (Asarian 1999, Christie 2008). During these two studies , the stu dents found Euglossa imperialis and Euglossa viridissima to be the two most common species, respectively. My data shows a shift in the order of e uglossine bee abundance in San Luis with Euglossa viridissima being the most abundant and Euglossa imperialis being the second most abundant. An explanation for this shift in abundance could be related to the abundance of scent and nectar sources within an area. Tongue length and body size in bees are positively correlated . A longer tongue permits a bee access to a wider range of nectar sources. Therefore, a bee's abundance can be dependent upon both body size and the tempora l patterns of flowers blooming thr oughout the wet and dry seasons (Ackerman 1985). As Euglossa imperialis has a long tongue extending past it' s sternum, this bee might have advantages over other species in that it has a wider access to nectar sources. 7! 0! 1! 2! 3! 4! 5! 6! 8! 9! 07! 977:0077! 0077:0277! 0277:0477! ;&)<)*)/-%! =(>(?*! @&/-%!A#B+*.!&C!DE*F(*'G'(/*! H%#?()-%!I-)$*'! ;-.(++*-)! J-F(KF!
! ! "#$%&''()*!+**'! ! ! ,-./! 8 ! The abundance of Eulaema bombiformis can be related to its body size. Larger euglossines have larger flight ranges. In addition to a wider distribution , this enables them to forage for nectar and scents on seasonally scarce resou rces making them more abundant at all times of the year. However , many species in the Eufriesea genus, which are also larger bees , are more seasonal and are active during dry season and into the start of the rainy season. I took this seasonality into cons ideration when counting Eufriese a rufocauda and Eufriesea chrysopyga in the species composition at Maximo' s as it was during the end of their most active period . Comparison of Species Composition Between Sites The SÂ¿renson Index revealed differences in the similarities of species richness between sites. S tudies have shown in some cases s econdary forests have been shown to have higher species richness than primary forest. The comparison between the mid elevation study sites of the Congo trail along the Caribbean slope and the Bajo del Tigre preserve both within the Children's Eternal Rain forest showed a 100% similarity in species composition. Probable explanation for this are the study sites location which were both within pre montane wet forest of similar elevation. The comparison between the continental divide and the two lowest elevations on both the Pacific and Caribbean slopes provided to have the greatest difference in species composition a t 33% for San Luis Arriba and 20 % for Maximo's. The site comparison between the continental divide an d the mid elevation Pacific slope site locations were Bajo del Tigre was 40% , followed by Crandall Memorial R eserve at 50%, and on the Caribbean slope 40% at the Congo Trail. These results illustrate the effects elevation has on species composition as both slopes showed that the percent difference lessened the closer in elevation the sit e comparison was. Altitudinal Variation in Species Abundance There was a decreasing trend in species abundance as altitude increased. The lowest altitudinal ranges sampled at on both the Pacific and Caribbean slopes showed the highest number of Euglossines. These findings are consistent with bee pollinator communities being heavily influenced by variation in food resource availability and climatic factors (Ramire z et al . 2015) . Plant families that provide nectar and scents for euglossines, especially in the family Orchidae have a low biomass at high altitudes in the cloud forest . In addition to having fewer plants to gather resources from , higher elevation cloud forests have lower temperatures and more variable and intense weather in regards to rainfall, horizontal precipitation, and wind. These are contributing factors in accounting for lower numbers of euglossine bees. My findings of eug lossine abundance having an inverse correlation with altitude corroborates past research on Euglossine bees conducted on the Pacific and Caribbean slopes of Costa Rica by former UCEAP students showing that bees' abundance and diversity is highe r at lower e levations (Christie 2008). The Caribbean slope showed a higher abundance of orchid bees across all study sites, especially taking into consideration the total time sampled at each location was less than the Pacific by two hours. Although the Caribbean slo pe was inhabited in the past, over the last several decades' anthropogenic disturbance has lessened considerably with the preservation of forest within the Children's Eternal Rainforest . Altitudinal Variation in Species Diversity
! ! "#$%&''()*!+**'! ! ! ,-./! 9 ! A decreasing trend in species richness as altitude increased was also found to be significant in both the Caribbean and Pacific slopes. The lowest altitudes sampled at yielded the highest number of species and were the only two locations that unique euglossine bee species were captured at. The Caribbean slope had a higher variation in euglossine species composition compared with the Pacific slope. Studies have shown that euglossine species richness decreas es in habitat that is disturbed (Roubik and Hanson 2004) . My research furt her support s these past findings as both euglossine bee abundance and diversity was significantly higher across all altitudinal ranges sampled at along the Caribbean slope . A potential explanation for this is all of the study sites were a part of the Child ren's Eternal Rainforest which is a minimally disturbed area in comparison to the study sites located on the pacific slope which are surrounded by unprotected, highly disturbed , and human developed areas. Confounding factors influencing euglossine bee capture and distribution I observed significantly less bees during sampling hours of high precipitation and lower temperatures. Most euglossines fly when conditions are favorable, usually when weather is both warm and humid (Roubik and Hanson 2004). As r ainfall and temperature play an important role in euglossine bees foraging behavior, and weather conditions were very variable throughout all of my days sampling, it is necessary to consider these factors effects on the accuracy of comparison between sites . It is important to consider the lesser known affects that changing environmental conditions such as habitat fragmentation has on euglossine bees. S tudying and documenting euglossine bees ' abundance and diversity alon g an altitudinal gradient on both the Caribbean and P acific slopes is vital knowledge that can be used for future conservation efforts of their species. As forest fragmentation continues to decrease bees ' habitat , having k nowledge about the ir current geographic distributions and the l evel of anthropogenic disturbance that is around them can be used in future reforestation and conservation methods . Other valuable information that can be used for their conservation is d etermining what factors other than deforestation have the biggest influence on euglo ssine bees, such as temperature and resource availability . As temperatures slowly increase, bee populations migrate upwards t o seek cooler habitats (Huebner 2017) . As orchid bees are endemic to the neo tropics and are critically important pollinators, their conservation is vital to the continued survival and dynamics of the plant families that they pollinate and all that is connected to them. Acknowledgements First off I would like to extend my utmost thanks to AndrÂŽs Camacho (Chamack) for being a patient and encouraging mentor . He not only put up with my never ending questions and chatter and for bringing my butterfly net back to me e very time I forgot it somewhere (even the continental divide). I would like to thank Frank Joyce for acc identally setting me up to study orchid bees by assigning them as my species to research in Isla Violin . I learned from him the importance in respecting and treating these bees as literal jewels of the tropic s. He offered, encouragement, transparency and s upport at a time when I needed it the most. Last but not least , I would like to thank FÂŽ lix Salazar for bee e ing an awesome sport and accompanying me out into the field to help me film these seÂ–or bees. On one occasion this involved a rain drenched hike,
! ! "#$%&''()*!+**'! ! ! ,-./! 07 ! and on all of them my endless dad jokes, overall foolishness, and bee puns. Without all of them this study would not have been possible. Thank your peer editor and the person who translated your abstra ct into Spanish. References Ackerman, J. D. 1983. Diversity and seasonality of male euglossine bees (Hymenoptera: Apid ae) in central panama'. Ecology 64: 274 283 . Asarian, J. E. (1999). Comparison of Male Euglossine Bee Abundance, Species Richness and Movements in Fragmented and Continuous Forest in San Luis, Costa Rica. UCEAP W inter 1999. Christie, J. M. (2008). Euglossine Species Composition at Different Elevatio ns and on Different Slopes of the TilarÂ‡n Mountains, Costa Rica. UCEAP Fall 2008. Dressler, R. L. 1982. Biology of the orchid bees (Euglossini). Annual Review of Ecology and Systematics 13: 373 394. Huebner, L. (2017). Higher abundance and species richne ss at lower elevations in Monteverde altitudinal survery on euglossine bee diversity. UCEAP Spring 2017. Janzen, D.H., P.G. Devries, M.L. Higgins & L.S. Kimsey. 1982. Seasonal and site variation in Costa Rican Euglossine bees at chemical baits in lowlan d deciduous and evergreen forests. Ecology 63: 66 74. Ramirez, S.R.et al. (2015). Seasonal cycles, phylogenetic assembly, and functional diversity of orchid bee communities. Ecology and Evolution 5 : 1896 Ã 1907. Roubik, D. W., & Hanson, P. E. (2004). Orchid Bees of Tropical America Biology and field guide. Santo Domingo, Heredia: Instituto Nacional de Biodiversidad. Schmale, D. G., Differential attraction of euglossine bees to varying strengths of chemical baits in two Costa Rican primary forests. Nature Education Knowledge 3 : 38 . Appendice s Appendix A . Photos A,B ,C are Eulaema c ingulata , pho tos D, E are Eufriesea chrysopyga, both found at Maximo's river . Photos F, G, H are Euglossa viridissima . Photos I, J, K are Eufriesea rufocauda at Maximo's River. Photos L,M,N,O are Euglossa imperialis at Bajo del Tigre. Phot os P,Q,R are Eulaema bombiformis.
! ! "#$%&''()*!+**'! ! ! ,-./! 00 !
! ! "#$%&''()*!+**'! ! ! ,-./! 01 !
! ! "#$%&''()*!+**'! ! ! ,-./! 02 !
! ! "#$%&''()*!+**'! ! ! ,-./! 03 !
! ! "#$%&''()*!+**'! ! ! ,-./! 04 !