|USFLDC Home | Tropical Ecology Collection [Monteverde Institute]||| RSS|
This item is only available as the following downloads:
1 Sugar concentrations, hummingbird aggressiveness, and community composition in Monteverde, Costa Rica. Paul Brooks Natural Resources and Environmental Science, Purdue University Jennifer M. Gillen Department of Fisheries and Wildlife, University of Minnesota August 13, 2006 ABSTRACT We studied hummingbird communities in Caitas, Mont everde, Puntarenas, Costa Rica to observe territori al tendencies when given high and low percentage sugar concentration of food resources at artificial feed ers along the edge of a forest patch. We wanted to determine if community composition was disproportionately affect ed by more aggressive, and therefore territorial species. We observed interactions of hummingbirds at artificial feeders of 20 and 33% sugar concentration. We found a disproport ionately high number of visits by Rufous-tailed Hum mingbirds ( Amazilia tzacatl ) at both sets of feeders (52.5% at 20% feeders and 89.4% at 33% feeders). We found a higher frequency of visits by all species at low reward fe eder than at the high reward feeder (1016 visits at 20% feeders and 716 at 33% feeders) and a higher proportion of aggr essive interactions at the lower reward feeder. Ru fous-tailed hummingbirds were the most territorial species at o ur study site and their behavior influenced the rel ative abundance of subordinate species. Contrary to what we expect ed, we found that frequency of close chases and far chases at different sugar concentration feeders were not diff erent than random. RESUMEN Estudiamos comunidades de colibres en Caitas, Mon teverde, Puntarenas, Costa Rica para observar tende ncies territoriales cuando se les dio concentraciones de recursos alimenticios en comederos artificiales a l o largo del borde de un parche de bosque. Queriamos determinar si la composicin de la comunidad estaba desproporcionad amente afectada por especies ms agresivas, y por lo tanto territoriales. Observamos interacciones de colibr es en comederos artificiales con 20% y 30% de concentraci n de azcar. Encontramos un nmero de visitas desproporcionadamente alto del colibr de cola caf ( Amazilia tzacatl) en ambos grupos de comederos (52.5% en los comederos con concentracin con 20% y 89.4% en los comederos con 33%). Encontramos una frecuencia may or de visitas de todas las especies en el comedero con po ca recompensa en comparacin con el comedero con al ta recompensa. (1016 visitas en los comederos con 20% y 716 en los comederos con 33% ).y una proporcin m s alta de interacciones agresivas en el comedero con poca recompensa. Los colibres con cola caf fueron la especie ms territorial en nuestro sitio de estudio y su compor tamiento influenci la abundancia relativa de espec ies subordinadas.. Contrario a lo esperado, encontramo s que no la frequencia de persecusiones cercas y pe rsecusiones lejos en comedores de concentraciones de azucar dif erentes no fueran diferente que impensado. INTRODUCTION
2 There are 17 species of hummingbirds common on the Pacific slope of the Monteverde area of Costa Rica (Fogden and Fogden 2005). This hummingbi rd population is diverse despite the fact that it has a limited supply of nectar producing fl owers (Tiebout 2000). This species richness has come about because different hummingbird species sp ecialize on different types of flowers or use different foraging behaviors and competitive be haviors. Interand intraspecific interactions between hummingbirds can change their community com position (Fogden and Fogden 2005). Both interand intraspecific competition can be i nfluenced by resource value and availability, species-specific foraging behavior, d ifferences in body size, and densities of individuals. Intraspecific competition can also be influenced by mating pressures. Interspecific interactions in hummingbirds are often influenced b y the most dominant species in the area. They may limit subordinate species access to food sources, while dominant individuals may limit their own populations through intraspecific i nterference at rich food resources (Tiebout 1993). These dominant hummingbird species are typically cl assified as territorial. Territorial species are usually medium sized hummingbirds that are very aggressive and often brightly colored (Fogden and Fogden 2005). Their territori es usually consist of several flowers that are close enough together to defend. When different sp ecies of territorial hummingbirds are found at the same location, the smaller species tend to shif t behavior from territorial to subordinate (Lawlor and Maynard Smith 1976). Difference in resource value, such as the caloric v alue of nectar, can change hummingbird behaviors. When different resource val ues are available, birds will usually feed where there are higher quality resources (Pimm et a l. 1985). However, optimal foraging theory states that population densities of competing speci es will influence foraging strategies (Lawlor and Maynard Smith 1976). As the density of humming birds increases, interand intraspecific competition increases, causing a shift in how birds use the available resources. This would cause subordinate hummingbirds to rely on lower quality r esources because more dominant individuals defend the higher quality resources. Dominant hummingbirds should be more aggressive at resources of greater value and thus affect community composition. We hypothesized that there would be a difference in aggressive behavior between feeders of different su gar concentrations. Therefore, we predicted that higher sugar concentrations will lead to more chases, limiting weaker competitors access to food sources. Competitive interactions should be l ess common and intense when territoriality is not as pronounced. METHODS AND MATERIALS Study Site The site chosen for our study was on the Santamara farm in Caitas, Puntarenas, Costa Rica. Caitas is a rural community located approximately 2.2 km northwest of Santa Elena. The study site was located along the edge of a forest patch t hat bordered a coffee farm. Natural History of Local Hummingbirds The four common hummingbirds at our feeders in Cai tas were the Rufous-tailed Hummingbird ( Amazilia tzacatl), the Striped-tailed Hummingbird ( Eupherusa eximia ), the Steely-vented Hummingbird ( Amazilia saucerrottei ), and the Violet Sabrewing ( Campylopterus hemileucurus ). In Costa Rica, the Rufous-tailed Hummingbird is the most widespread
3 hummingbird (Fogden and Fogden 2005). Throughout i ts range, it prefers non-forest habitat such as second growth, coffee plantations, and alon g forest breaks and gaps (Stiles and Skutch 1989). Both sexes of Rufous-tailed Hummingbirds ar e aggressive and dominant to hummingbirds of similar size (Fogden and Fogden 200 5). Males commonly defend territories. Another common resident to the Monteverde area is t he Striped-tailed Hummingbird (Stiles and Skutch 1989). Male Striped-tails occur mainly in t he forest canopy, while females prefer the forest understory (Fogden and Fogden 2005). Steely-Vented Hummingbirds are common in the Monte verde area. There is no sexual dimorphism within the species and both defend terri tories, though not as effectively as larger hummingbirds. They frequent scrubby woodland, coff ee plantations, and gardens, foraging at many flowering trees (Fogden and Fogden). The Viol et Sabrewing is among one of the largest and most common hummingbirds in the Monteverde area It is a high-reward trapliner that prefers understory, edges of mountain forests, or p atches of woods in disturbed areas (Stiles and Skutch 1989, Fogden and Fogden 2005). Though rarel y territorial at flowers, Violet Sabrewings can easily dominate all other Monteverde hummingbir ds at feeders (Fogden and Fogden 2005). Procedure For this study, four hummingbird feeders were used. Two feeders were filled with a simulated nectar mixture of 20% sugar concentration by volume (approximately 240 g/946.2 ml) and two other feeders were filled with a simulated nectar mixture of 33% sugar concentration by volume (362 g/ 946.2 ml). The four feeders were ref illed with an identical sugar concentration every three days to prevent the sugar from degradin g. The feeders were alternately placed at a distance of 47 meters 1 meter away in order to as sure that each feeder could be defended separately from all others. Feeders were set up th ree days prior to data collection to allow hummingbirds to discover the food sources. Field g uides for hummingbirds were used to check accuracy in identification. Binoculars were used i n order to study at a distance that would not affect the interactions at the feeder. Data were collected for eight days at varying times to control for temporal variation. We spent one hour at each of the four feeders, with on e day spent together collecting data at all feeders. Sixty total hours were spent in the field Each feeder was stationed in a tree along the forest edge that was visible from the forest. We r ecorded the type of hummingbird species, the amount of time spent hovering and feeding, and the type of interaction. Visits were counted if the hummingbird approached the feeder within 15 cm. All interactions were classified as either chase or no chase, (after Dearborn 1998). An i nteraction was recorded as a chase if the hummingbird at the feeder either chased or was chas ed by another individual, while no chase implied no interaction between the hummingbird visi ting and another individual. Additionally, we noted which two species interacted. Far chases were counted if a chase interaction occurred farther than 15 cm away from the feeder. Statistical Analyses We used Chi-square analyses to determine if there w as a preference by any or all species for one sugar concentration feeder over another, if there was a difference in the frequency of chase and no-chase interactions at 20 and 33% s ugar concentration feeders, and if there was a difference in close and far chase interactions at the two sugar concentration feeders. A parametric t-test was used to compare the mean numb er of hummingbird visits per hour interval.
4 RESULTS At our study site, four of the 17 species common on the Monteverde Pacific slope were observed: Rufous-tailed Hummingbird ( Amazilia tzacatl ), Stripe-tailed Hummingbird ( Eupherusa eximia ), Steely-vented Hummingbird ( Amazilia saucerrottei ), and Violet Sabrewing ( Campylopterus hemileucurus ). We observed 1732 interactions at the feeders in 60 hours of observations and 1008 chases away from feeders in 5 1 hours. A total of 1016 visits were observed at the feeders with 20% sugar concentratio n, while 716 visits at feeders with 33% sugar concentration were observed (Figure 1). The freque ncy of visits by the four hummingbird species at different sugar concentrations were not different than random (Chi-square test, c 2 = 273. 67, p < 0.001, n = 1732). Rufous-tailed Hummi ngbirds were more common than expected at the high concentration feeders, while the other species were all less common than expected. The frequency of visitation to the 20% sugar concen tration feeder was significantly higher than expected by chance, and vice versa for the 33% suga r concentration feeder. We compared the number of no chase versus chase events at the two concentrations of feeders (Figure 2). At different sugar concentr ations, there was a significant difference in no chase and chase frequencies. (Chi-square test: c2 = 24.10, p < 0.001, n = 1735). The relative abundance of each species at different sugar concentrations was noted. Both feeders show a disproportionately high frequency of visits by the Rufous-tailed Hummingbird, although the 20% feeders show more species diversit y (Figure 3). The Rufous-tailed Hummingbird was the most common visitor at each fee der. 0 200 400 600 800Rufous-tailed Hummingbird Steely-vented Hummingbird Striped-tailed Hummingbird Violet Sabrewing Hummingbird Species 20% Sugar Concentration 33% Sugar Concentration Figure 1. The number of visits by four hummingbird species to feeders of 20 and 33% sugar concentration. The frequency of visits to feeders of the two concentrations was not random. (Chi-square test: c2 = 273.67, p < 0.001, n = 1732). 0 200 400 600 No ChaseChaseNumber of events 20% Sugar Concentration 33% Sugar Concentration Figure 2. Comparison of hummingbird chases versus no chases at feeders of 20 and 33% sugar concentration. Frequency of chase and no-chase events at feeders with different sugar concentrations. (Chi-square test: c2 = 24.10, p < 0.001, n = 1735). The average number of visits observed per hour inte rval was 34.0 and 23.8 for 20 and 33% sugar concentration feeders, respectively (Figu re 4). There was a significant difference in the average number of visits per hour to each feede r (T-test: t-value = -3.55, p < 0.001, n = 60, d.f. = 58).
5 n nr r r !! Figure 3. Proportion of visits by different hummin gbird species at feeders of 20 and 33% sugar concentration. Rufous-tailed Hummingbird were most common at both feeder types, while there was a larger proportion of Violet Sabrewings, Strip ed-tailed Hummingbirds, and Steely-vented Hummingbirds at the 20% feeder. " !" rnnn !! 0 200 400 600 800 Close ChaseFar ChaseNumber of events 20% Sugar Concentration 33% Sugar Concentration
6 Figure 4. The mean number of hummingbird visits per time interval (one hour) to feeders of 20 and 33 % sugar concentration ( SE). There was a significant difference in the mean number of visits to feeders of different sugar concentrations. (T-test, t = -3.55, p < 0.001, n = 60, d.f. = 58). Figure 5. A comparison of number of close and far chases at feeders of 20 and 33 % sugar concentrations. There was not a significant difference in close versus far chases at feeders of different sugar concentrations. (Chi-square test: c2 = 0.93, p-value = 0.33, n = 1725). We observed 463 and 254 close chases for 20 and 3 3% sugar concentrations, respectively. There were 628 and 380 far chases observed at 20 and 33% sugar concentrations recorded (Figure 5). The frequency of close and fa r chases at feeders of low and high sugar concentrations was not different than random (chi-s quare goodness of fit test, c2 = 0.93, p-value = 0.33, n = 1725). DISCUSSION Our first hypothesis was that the feeders with high er sugar concentrations, and therefore higher caloric rewards, would have more visits than the lo wer reward feeder. We found that the opposite was true, in that there were more visits t o the lower reward feeder. There were more visits by Rufous-tailed Hummingbirds than other spe cies at both types of feeders, but they were proportionately more abundant at the 33% feeders. We think optimal foraging theory supports our results. Optimal foraging theory suggests that pressures from high densities of a dominant species may cause a subordinate species to avoid hi gher reward resources and only feed in low reward areas (Pimm et al. 1985, Tiebout 1993). Sub ordinate species may perceive the lower reward as more beneficial and avoid increased aggre ssive interactions with the dominant species at the higher reward feeders. This could be an exp lanation for the lack of diversity at the feeders with higher sugar concentration. We predicted that, given the option of different su gar levels, there would be a difference in the number of aggressive interactions at low and high sugar concentration feeders. Our study found, contrary to what we hypothesized, that there was a higher frequency of chase interactions at the feeders with 20% sugar concentration than at the higher reward feeders. This could be explained by the possibility that Rufous-tailed Hum mingbirds at the 33% sugar concentration feeders are more aggressive and do not allow other individuals access to the resource and hence chase interactions could not be observed. Also, th e territorial hummingbirds were often perched nearby, allowing them to readily defend the feeder. Many interactions at the feeders with 20% sugar concentrations involved chases by species oth er than the Rufous-tailed Hummingbird. This could signify that Rufous-tails at lower rewar d feeders were not as efficient at defending their territory or not as aggressive when the stake s are lower at defending their territory, allowing individuals of other species to use or defend the r esource. These results are also predicted by the optimal foraging theory (Lawlor and Maynard Smith 1 976). We predicted that there would be a higher frequency of far chases than close chases at the higher reward feeder, because it would be beneficia l to defend the resource before intruders had a chance to steal the nectar. Our analysis showed that, although there was higher frequency of far chases at the 33% concentration feeder, they we re not different from random. Perhaps this could be due to dominant individuals desire to sav e energy and remain close to the resource, thereby avoiding theft of the resource during a far chase.
7 A previous study states that the Steely-vented Hum mingbird is territorially aggressive (Young and McDonald 2000), but in our study it took a subordinate role. This subordinate role is probably due to its smaller size relative to oth er species at our study site. Many times, a territorial birds decision to defend can be based on the size of the intruder (Dearbor 1998). In our study, the Steely-vented Hummingbird was the sm allest species of visitor, thereby severely limiting its ability to defend against other specie s and giving it a subordinate role in the community. Previous studies in the Monteverde area focused on territoriality and aggressiveness of hummingbirds. Matheson (2004) found Rufous-tailed Hummingbirds to be the most dominant species in the area, supporting our results that th e Rufous-tail was the most aggressive species. The Rufous-tail was the largest territorial species at our study site, giving it an advantage over the other species. Another study reported that the Violet Sabrewing was one of the most aggressive species observed (French 1992). Althoug h this study contradicts our findings, French did not compare aggressiveness between sexes, and o nly females were observed at our study site. Our results did agree with published accounts of th e Violet Sabrewing; that Violet Sabrewings are typically high reward trapliners (Stiles and Sk utch 1989, Fogden and Fogden 2005). They were also the least frequent visitor at both 20 and 33% feeders. One suggestion for future studies could be to measu re sugar concentrations by weight instead of volume intially. This would make it eas ier to compare results with many published articles. Another suggestion is to count far away chases from the beginning of data collection. During our data collection, the identities of speci es involved in far chases were not recorded. Having this information would make it possible to t est at what frequency subordinate species attempt to approach the resource and would be benef icial to future studies. A more pronounced difference in sugar concentrations can be used to d etermine tendencies in territoriality of resident hummingbird species. There are several follow-up projects appropriate fo r the Caitas area. The topography in Caitas, as well as throughout the Monteverde regio n, has strong elevational gradients. At a nearby location, which was more open than our study area, several other hummingbird species were observed, including Plain-capped Starthroats ( Heliomaster constantii ), Coppery-headed Emeralds ( Elvira cupreiceps ), and male Violet Sabrewings as well as all specie s observed at our study site. This location was less than 300 m away from the study site and only slightly lower in elevation, yet had a different hummingbird communit y composition. A future study could attempt to determine differences in habitat quality and preferences of local hummingbird species at these locations. Acknowledgements We would like to thank Pilo Jimnez Santamara and Roxana Santamara for providing the setting for our project (and for letting Paul borrow his boots). Also grac ias to Olman and Kailor Mndez for giving us the to ur of the farm and helping us find good spots for feeders. Lastly we want to thank Karen Masters for her guidance a nd materials required for the project, and staff. Literature Cited Dearborn, D.C. 1998. Interspecific territoriality by a Rufous-tailed Hummingbird ( Amazilia tzacatl): effects of intruder size and resource va lue. Biotropica 30(2): 306-313. Fogden, M. and P. Fogden. 2005. Hummingbirds of C osta Rica. D. Featherstone and A. Sheehan, eds. Zona Tropical Publication. Miami, FL, pp. 41-57.
8 French, A.R. 1992. Aggression and feeding patterns of seven tropical hummingbird species. EAP Tropical Biology. Fall 1992. Lawlor, L.R. and J.Maynard Smith. 1976. The coevo lution and stability of competing species. American Naturalist 110:79-99. Matheson, E. 2004. Aggressive behavior of humming birds and their response to nectar concentrations. CIEE Summer Program. Pimm, S.L., R.L. Rosenzweig, and W. Mitchell. 1985 Competition and food selection: Field tests of a theory. Ecology 66(3): 798-807. Stiles, F.G. and A.F. Skutch 1989. A Guide to the Birds of Costa Rica Cornell University Press. Ithaca, New York. Tiebout III, H.M. 1993. Mechanisms of competition in tropical hummingbirds: Metabolic costs for losers and winners. Ecology 72(2): 405-418. Tiebout, H.M. 2000. Do subordinate species have an advantage? Testing the pointer hypothesis with tropical hummingbirds. In: Monteve rde: Ecology and Conservation of a Tropical Cloud Forest N.M. Nadkarni and N.T Wheelwright, eds. Oxford University Press. New York, NY, pp. 216-218. Young, B.E. and D.B. McDonald. 2000. Chapter six: Birds. In: Monteverde: Ecology and Conservation of a Tropical Cloud Forest N.M. Nadkarni and N.T. Wheelwright, eds. Oxford University Press. New York, NY, pp.199.
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 M39-00063
Las concentraciones de azcar, la agresividad en el colibr, y la composicin de la comunidad en Monteverde, Costa Rica.
Sugar concentrations, hummingbird aggressiveness, and community composition in Monteverde, Costa Rica.
We studied hummingbird communities in Caitas, Monteverde, Puntarenas, Costa Rica to observe territorial tendencies when given high and low percentage sugar concentration of food resources at artificial feeders along the edge of a forest patch. We wanted to determine if community composition was disproportionately affected by more aggressive, and therefore territorial species. We observed interactions of hummingbirds at artificial feeders of 20 and 33% sugar concentration. We found a disproportionately high number of visits by Rufous-tailed Hummingbirds (Amazilia tzacatl) at both sets of feeders (52.5% at 20% feeders and 89.4% at 33% feeders). We found a higher frequency of visits by all species at low reward feeder than at the high reward feeder (1016 visits at 20% feeders and 716 at 33% feeders) and a higher proportion of aggressive interactions at the lower reward feeder. Rufous-tailed hummingbirds were the most territorial species at our study site and their behavior influenced the relative abundance of subordinate species. Contrary to what we expected, we found that frequency of close chases and far chases at different sugar concentration feeders were not different than random.
Estudiamos las comunidades de colibres en Caitas, Monteverde, Puntarenas, Costa Rica para observar las tendencias territoriales cuando se les dio concentraciones de recursos alimenticios en comederos artificiales a lo largo del borde de un parche de bosque. Queriamos determinar si la composicin de la comunidad estaba desproporcionadamente afectada por especies mas agresivas, y por lo tanto territoriales.
Text in English.
Hummingbirds--Feeds and feeding--Costa Rica--Puntarenas--Caitas
Species diversity--Costa Rica
Colibres--Alimentos y alimentacin--Costa Rica--Puntarenas--Caitas
Diversidad de especies--Costa Rica
Tropical Ecology 2006
Ecologa Tropical 2006
Gillen, Jennifer M.
t Monteverde Institute : Tropical Ecology