Changes in Altitudinal Distribution of Hummingbird Diversity between Forest and Pasture Habitats Jessica Chute Department of Biology, Northeastern University ABSTRACT Climate change has affected the composition of bird communities elevationally in Mon teverde, Costa Rica (Pounds et al. 1999). In addition, habitat transformation from forest to pasture has favored weedy species there (Feinsinger 1988). This study sought to combine hummingbird distributions across both habitat type and elevation, as the interaction of habitat disturbance and climate change is not well understood. Past studies have shown changes in hummingbird communities for either climate or habitat changed (Oliver 1993; Donnelly 1998; Smith 2000; Lynn 2001; Winchell 2001; Spear 2004). In order to observe these changes, I hung hummingbird feeders in different elevational zones in both forest and pasture. I found that Purple throated Mountain Gems ( Lampornis calolaemus ) have shifted upward in elevation, and Violet Sabrewings ( Campylopte rus hemileucurus ) and Green crowned Brilliants ( Heliodoxa jacula ) show trends towards upward movement as well. Also, I observed that the communities of both fields and forest habitats have been altered. I saw no open area specialist species, very low num bers of any species in pasture sites, and I did not observe some common forest species. My data suggest possible changes in the composition of the hummingbird communities of Monteverde. RESUMEN El cambio climÂ‡tico ha afectado la composiciÂ—n de las com unidades en Monteverde, Costa Rica en tÂŽrminos de elevaciÂ—n. AdemÂ‡s la transformaciÂ—n del hÂ‡bitat de bosques a potreros a favorecido ciertas especies. Este estudio busca estudiar la interacciÂ—n entre la elevaciÂ—n y el tipo de hÂ‡bitat, debido a que esto no ha sido estudiado. Estudios en el pasado han demostrado los cambios de las comunidades en diferentes elevaciones y en diferentes hÂ‡bitats. Utilizando comederos artificiales en potreros y bosque a diferentes alturas estudie estos cambios. EncontrÂŽ que Lampo rnis calolaemos ha subido en elevaciÂ—n y existe una tendencia a que Campylopterus hemileucurus y Heliodoxa jacula muestran una tendencia a subir en su rango altitudinal. AdemÂ‡s observe que las comunidades de potrero y bosque tambiÂŽn han sido alteradas. No observe especialistas en Â‡reas abiertas, poca abundancia de especies en general y no observe algunas especies comunes de bosque. Estos datos sugieren cambios en la composiciÂ—n de las comunidades de colibrÂ’es en la comunidad de Monteverde.. INTRODUCTION Climate change has been a problem since the Industrial Revolution, and has skyrocketed since the 1970s (McCarty 2001). In addition, human impacts on the environment have become more and more apparent (IPCC 2007), with heavier repercussions on species th at are more habitat specific (Iverson and Prasad 1998). The tropics are among the most affected of all ecosystems because of the high biodiversity they contain (Wright 2005). Climate change has hit Monteverde of the TilarÂ‡n Mountains particularly hard, a s the area has a large number of habitat types and transitions in comparison to the area it encompasses (Pounds et al. 1999). In Monteverde, a 0.5 o C global change in temperature has a huge impact on biological communities where the
temperature difference between one altitudinal community and another may differ very little, and where temperature plays such an important role in the climates that define each community. Holdridge Life Zones based on elevation, mean annual temperature and rainfall, classify t hese habitats and provide scientists with a basis for identifying species ranges (Haber 2000). In Monteverde an overall change of 0.5 o C in temperature results in an expected 100 meter upward climb in elevational range (Pounds et al. 1999). Birds are goo d indicators of climate and human impacts as they respond quickly to changes because of their high mobility (Walther et al. 2002). Indeed, birds in Monteverde appear to be moving up altitudinally (Pounds et al. 1999). The hummingbird species that are typ ically found in Monteverde have known altitudinal distributions (Feinsinger 1977; Fogden 1993; Young et al. 1998). Initial comparisons of studies already show upward movement in some forest species such as Green crowned Brilliants ( Heliodoxa jacula ). (Fo gden 1993; Lynn 2001. Other forest species simply show lower than expected frequency of observance (Lynn 2001, Winchell 2001). Many species also have specific habitat preferences, with some favoring forest habitats and others forest edges or clearings (F einsinger 1977; Stiles and Skutch, 1987 the bird book; Fogden 1993). As forest was transformed to pasture in Monteverde, the abundances of these species have shifted. Comparisons of hummingbird habitat studies show changes in abundances of species as a r esult of habitat disturbance (Oliver 1993; Stiles and Skutch 1999; Smith 2000; Winchell 2001; Spear 2004). My goal was to track changes in species diversity and abundance along an altitudinal gradient as well as between habitat types. I predicted that s pecies in the forest would show greater upward movement than pasture species. Pasture species have a greater ability to disperse over area and altitude because of their more general requirements. This is because disturbances in habitats (clearings, field s, etc.) "loosen" species interactions and allow other species to move in that are more adapted for these less specific habitats. Hummingbird species in disturbed sites visit flowers in a more haphazardly and opportunistically than species in undisturbed sites, and so open habitats favor different species than undisturbed forest (Feinsinger et al. 1988). Hummingbirds circulate freely among bean fields, scrub, and other open areas, which also share many plant species and so, whether spatially adjacent or n ot, the plants in all three habitats are linked through their common set of pollinators (Feinsinger 1978). Therefore, pasture species were not expected to move as much elevationally, because their habitat is not subject to the interaction dynamics found i n natural forest habitats (Terborgh 1971). METHODS Eight sites were chosen at four elevations that each fell into different Holdridge Life Zones as defined by Fogden (1993) (Tables 1&2). At each elevation, four hummingbird feeders were hung from trees by yellow rope, two in a pasture or field clearing, and two in forest. The first set of sites at the lowest elevation was located at the Ecolodge in San Luis. The forest site was in the secondary forest patch ten meters off El Zapote Trail and had an el evation of 1185 meters. The field site was located in the middle of the horse pasture directly in front of the building complex and had an elevation of 1135 meters. The second set of sites was in the Bajo del Tigre area. The forest site was in the reser ve ten meters off the Sendero de los Monos at marker 20 with an elevation of 1340 meters. The field site was in the lowest horse pasture owned by Elias
Newswanger at 1385 meters. The third set of sites was at the Villa. The field site was in the front y ard at 1510 meters. The forest site was in the secondary forest patch behind the house at 1540 meters. The last sites were along the TV tower road at 1650 meters. The field site was in a clearing along the road created by the construction of telephone w ires. The forest site was approximately 30 meters into the forest from the field site. TABLE 1 Description of Fogden Zones used in finding study site elevations for hummingbird feeders. Zone Altitude Range Locations 1 700 m 1300m San Luis to cliff edge of SW Monteverde 2 1300m 1500m Santa Elena and Monteverde 3 1500m 1600m Upper Monteverde and lower levels of Monteverde Cloud Forest Preserve 4 1600m+ Pac. Upper levels of Monteverde Cloud Forest Preserve Table 2 Description of the Holdridge Life Zones of Monteverde on the Pacific slope. Life Zones were used in conjunction with Fogden Zones 1 through 4 to determine elevations for hummingbird feeders. Zones in grey were not used. Life Zone Eleva tion (m) Mean Annual Rainfall (mm) Mean Annual Temp (Ã»C) Dry Season Duration (months) Canopy (m) Premontane moist forest 700 1000 1200 2200 17 24 3.5 5 25 Premontane moist wet forest transition 1000 1200 Premontane wet forest 800 1450 2000 4000 17 2 4 0 5 30 40 Lower montane wet forest 1450 1600 1850 4000 12 17 0 3 25 35 Lower montane rain forest 1550 1850 3600 8000 12 17 0 3 20 30 The feeders had red bases with white flowers and perches, and were filled with a 20% sugar solution. The feeders wer e left for a week before commencing data collection to allow the hummingbird community composition to reach equilibrium. Each site was observed every other day over a period of ten days. Observation times were rotated to allow for variances in feeding be havior during different times of the day. During observation periods, each species that visited the feeders was recorded, as well as the number of individual visits of each species. Pseudo replication was limited as much as possible by not re recording i ndividuals that made obvious repeat visits. A one way ANOVA was used to test differences in the distribution of hummingbird species across elevations. Multinomial tests were used to find deviations of observed elevational distributions from both even d istributions and expected distributions from Fogden (1993) within each species. A ! 2 test for independence was used to test differences in habitat distribution between species. RESULTS Four of Monteverde's reported hummingbird species were observed: t he Violet Sabrewing ( Campylopterus hemileucurus ), Stripe tailed Hummingbird ( Eupherusa eximia ), Purple throated Mountain Gem ( Lampornis calolaemus ), and Green crowned Brilliant. Of the total number of
individuals observed, 25% were found in field habitats and 75% in forest habitats. Elevational distribution of both habitats was grouped into Fogden Zones 1, 2, 3 & 4. Individuals in Zone 1 accounted for 14.6% of total, Zone 2 had 16.7% of all individuals, in Zone 3 there were 47.9% of total individuals and 20.8% were in Zone 4. Of the observed species, only Violet Sabrewings were in both habitats in all three zones they were expected at. Stripe tailed hummingbirds were the only other species seen in a field site, and only in Zone 3, but were found in th e three other zones in forest habitats. Green crowned Brilliants and Purple throated Mountain Gems were only found in forest sites in Zones 3 and 4. FIGURE 1 Distribution in elevation (Fogden zones) and habitat (field or fores t) of observed hummingbird species in Monteverde, Costa Rica. Habitat There was a significant difference in the distribution of species across the two habitat types studied with a combined total of 12 individuals from two species found in field sites a nd 36 individuals from all four species found in forest sites ( ! 2 = 7.8195, d.f. = 3, p = 0.049892). Very few individuals were observed at most field sites, with no visits recorded at all for the field site at the lowest elevation (1135 meters). In addit ion none of the expected field specialists (Rufous tailed Hummingbird, Steely vented Hummingbird) were observed, and no visits were recorded for other expected species, like the Green Hermit and Green Violet ear.
Elevation There was a significant diff erence in elevational distribution between the species I observed (F = 10.5030, d.f. = 3, p < 0.0001). There were two results for the multinomial tests, one for even distribution across all four zones, while the other was weighted for expected distributio n according to Fogden's Annotated Checklist (1993). Significant differences from even distribution were found for Purple throated Mountain Gems, which were observed eight times in Zone 3, twice in Zone 4 and never in Zones 1 and 2 ( ! 2 = 17.2000, d.f. = 3, p = 0.0006); and Green crowned Brilliants, which were only observed in Zones 3, and 4 with four and two individuals respectively ( ! 2 = 8.6, d.f. = 3, p = 0.0351). No significant difference for observed data was found for even distribution of Violet Sabre wings for which three, five and six individuals were observed in Zones 2, 3 and 4 ( ! 2 = 6.0, d.f. = 3, p = 0.1116); and Stripe tailed Hummingbirds observed in all four Zones with seven, five, six and one individuals observed in Zones 1 through 4 ( ! 2 = 4.36 84, d.f. = 3, p = 0.2243). In the comparison to expected distribution, which was derived from ranking Fogden's expected observance rates of Common (ranked 4), Fairly Common (3), Uncommon (2) and Rare (1), two species show significant movement. Stripe t ailed Hummingbirds were expected to be uncommon in Zone 1, fairly common in Zone 2, and common in Zones 3 and 4, but my observations indicate they have moved down from their expected distribution as there were higher abundances in Zones 1 through 3 (seven, five and six individuals respectively) than in Zone 4 (one individual) ( ! 2 = 9.7939, d.f. = 3, p = 0.0204). Purple throated Mountain Gems show upward movement from Fogden's expected distribution of fairly common in Zone 2, and common in Zones 3 and 4, wi th eight individuals observed in Zone 3, and one individual in Zone 4 ( ! 2 = 8.7001, d.f. = 2, p = 0.0129). Two species show no significant difference between expected and observed distribution: Green crowned Brilliants were observed four times in Zone 3 a nd once in Zone 4, and were expected to be common in both Zones they were observed in ( ! 2 = 1.8000, d.f. = 1, p = 0.1797). Violet Sabrewing observations of three, five and six individuals in Zones 2 through 4 also did not show significant differences from the expected distribution of common throughout the Zones they were observed in ( ! 2 = 1.0, d.f. = 2, p = 0.6065), though from visual comparison of the graphs Violet Sabrewings show a slight trend of upward movement. While there is no uniform trend of upwa rd movement between all species, there are significant changes in their elevational distributions.
FIGURE 2 Frequency rank is assigned on a scale from 1 to 4 based on Michael Fogden's expected observance frequencies of Com mon (4), Fairly Common (3), Uncommon (2) and Rare (1) in each zone. FIGURE 3 Abundance of species across habitats was combined for each zone to show total number of individuals found in each zone per species. DISCUSSION The distribution of species across both elevations and habitats has changed. Chiefly, there were significant differences between observed elevational distributions and Fogden's expected
distributions, but they did not uniformly show upward or no movement . The one outlier from this prediction in these results was the Stripe tailed Hummingbird, which exhibited downward movement instead of up. However, according to the field guide The Birds of Costa Rica , Stripe tailed Hummingbirds are found at middle elev ations from 800 2000 meters with some seasonal altitudinal movements up to 500 meters higher or lower than their normal range (Garrigues & Dean, 2007). Seasonal movement occurs from March to June (Fogden 1993), a time period in which the complete duration of my study fell into. In addition, according to a compilation of bird species abundance data (Young et al. 1998), Stripe tailed Hummingbirds are abundant at much lower altitudes than reported by Fogden with 77.28% of the captured sample found between 10 00 and 1450 meters, 17.15% between 1400 and 1550 meters, and 5.56% 1500 Ã 1700 meters. These data imply that the abundance of Stripe tailed Hummingbirds along the altitudinal gradient of their range may vary from year to year, which would place my data wi thin the overall normal range and abundance of Stripe tailed Hummingbirds, and follow that they have not actually moved down. Beyond the behavior of Stripe tailed Hummingbirds, there are overall trends of upward or no movement for the hummingbird specie s observed in my data analysis. This is also shown to be true through comparisons between Fogden's (1993) expected distributions, Young et al.'s (1998) data, and my data; Green crowned Brilliants are the only species that do not show trends of upward move ment. Their natural range is for the most part located at higher elevations than human disturbances occur. As a result, Green crowned Brilliants may be more resilient than the other species observed in this study, as they are subjected to only one type o f pressure, climate change, to a great degree and not to a combination of two pressures. The absence of expected species in this study could be a result of climate change and habitat transformation affecting their individual habitats, causing those spec ies to decrease in abundance in my study area. Green Hermits are found in wet mountain forest edges, understory, and second growth from 1650 Ã 2600 meters (Stiles and Skutch 1989). As a result, they may be sensitive to climate change, since their preferr ed habitats of edges and second growth may not be as abundant at higher elevations in Monteverde due to a lower frequency of disturbance. This would result in a decrease in abundance of this species. Also, my study ended at 1650 meters, which is the repor ted beginning of their range. I may not have seen this species because their range has moved further up the mountain and out of my study area. Rufous tailed Hummingbirds were also not observed in this study. This species prefers non forest habitats suc h as open scrub, secondary growth, thickety edges, semi open areas, coffee, and gardens, and inhabit all elevations up to 1850 m (Stiles and Skutch 1989). However, many hummingbird studies conducted by other students also reported no Rufous tailed humming birds. These studies as well as mine had no observations of this species even at sites that would seem ideal, such as a pasture in Zone 1. The apparent disappearance of Rufous tailed hummingbirds may be because climate change has shifted their preferred elevations up the mountain, but since their preferred habitat, pasture, is quite scarce at higher elevations in Monteverde, their abundance may have drastically decreased instead. This showed that my prediction of greater resistance to climate change by field specialist species was false. My data does not show that "weedier" hummingbird species with broader habitat
requirements are more resistant to climate change. Instead these species may be suffering more than forest species, as their preferred habit ats are not present at the higher altitudes all species are being forced into, and may be undergoing local disappearances as a result. ACKNOWLEDGMENTS Thank you first and foremost to Alan Masters for all his support on this project from conception to c ulmination. Thank you also to Yi men Araya and Anjali Kumar for helping me with an exhaustive amount of statistics, and to Pablo Allen and Moncho Calderon for helping with whatever I asked of them. I would also like to thank the Monteverde Conservation L eague, the Tropical Science Center, the University of Georgia Ecolodge, and all the Monteverde residents who allowed me to conduct my research on their property. Lastly, thank you to my family and friends for supporting me throughout. LITERATURE CITE D Bornstein et al. 2007. Climate Change 2007: Synthesis Report. IPCC. Donnelly, E. 1998. Changes in Bird Species Composition in Four Habitat Zones in Monteverde, Costa Rica. CIEE Spring 1998. Feinsinger, P. 1977. "Notes on the Hummingbirds of Monteverde, Cordillera de TilarÂ‡n, Costa Rica." The Wilson Bulletin, Vol. 89, No. 1 (March 1977), pp 159 164. Feinsinger, P. 1978. "Ecological Interactions between Plants and Hummingbirds in a Successional Tropical Community": Ecological Monogra phs, Vol. 43, No. 3, pp. 268 287. Feinsinger, P. et al. 1988. "Mixed Support for Spatial Heterogeneity in Species Interactions: Hummingbirds in a Tropical Disturbance Mosaic": The American Naturalist, Vol. 131, No. 1, pp. 35 57. Fogden, M. 1993. An Annotated Checklist of the Birds of Monteverde and PeÂ–as Blancas . San Jose, Costa Rica. Garrigues, R. and R. Dean. 2007. The Birds of Costa Rica: A Field Guide . Cornell University Press, Ithaca, New York. Haber, W. A. 2000. "Plants and Vege tation." In Monteverde: Ecology and Conservation of a Tropical Cloud Forest , p. 43 eds. Nalini M. Nadkarni and Nathaniel T. Wheelwright. New York, NY, Oxford University Press. Iverson, L.R. and A.M. Prasad. 1998. "Predicting Abundance of 80 Tree Sp ecies Following Climate Change in the Eastern united States": Ecological Monographs, Vol. 68, No4, pp. 465 485. Lynn, A. 2001. "Altitudinal Effects on Cloudforest Hummingbird Communities". CIEE Spring 2001. Monteverde, Costa Rica. McCarty, J.P. 2001. "Ecological Consequences of Recent Climate Change": Conservation Biology, Vol. 15, No. 2, pp. 320 331. Blackwell Publishing for Society for Conservation Biology. Oliver, M. 1993. Frequency of Hummingbird Activity in Forest and Pasture of Mo nteverde, Costa Rica. CIEE Summer 1993.
Pounds, J.A. et al. 2006. "Widespread amphibian extinctions from epidemic disease driven by global warming." Nature, Vol. 439, pp. 161 167. Pounds, J.A., M.P.L. Fogden and J.H. Campbell. 1999. "Biologic al response to climate change on a tropical mountain": Nature, Vol. 398, pp. 611 615. Smith, S. 2000. Changes in Hummingbird Species Richness and Abundance in a Forest Fragment and Agricultural Ecotones. CIEE Fall 2000. Spear, E. 2004. Edge Ef fect and Hummingbird Specialization. CIEE Spring 2004. Stiles, F.G., and A.F. Skutch. 1989. A Guide to the Birds of Costa Rica . Cornell University Press, Ithaca, New York. Terborgh, J. 1971. "Distribution on environmental gradients: theory and preliminary interpretation of distributional patterns in the avifauna of Cordillera Vilcabamba, Peru": Ecology, Vol. 52, pp. 22 40. Walther, G.R. et al. 2002. "Ecological Responses to Recent Climate Change": Nature, Vol. 416, pp. 389 395. Winchell , M.M. 2001. "Responses of Canopy and Understory Hummingbird Communities to Habitat Disturbance". CIEE Spring 2001. Monteverde, Costa Rica. Wright, J. 2005. "Tropical Forests in a Changing Environment": TRENDS in Ecology and Evolution, Vol. 20, N o. 10, pp. 553 560. Young, B.E. et al. 1998. "Diversity and Conservation of Understory Birds in the TilarÂ‡n Mountains, Costa Rica. The Auk, Vol. 115, No. 4, pp. 998 1016. University of California Press on behalf of the American Ornithologists' Uni on.
FIGURE 4 Average elevations of species for individuals combined from both field and forest habitats. FIGURE 5 Abundance of Violet Sabrewing (VS), Stripe tailed Hummingbirds (STH), Purple thro ated Mountain Gem (PTMG), and Green crowned Brilliant (GCB) in field and forest habitats distributed across elevational zones.
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 M39-00100
Cambios en la distribucin altitudinal de la diversidad del colibr entre el bosque y los hbitats en los potreros
Changes in altitudinal distribution of hummingbird diversity between forest and pasture habitats
Climate change has affected the composition of bird communities elevationally in Monteverde, Costa Rica (Pounds et al. 1999). In addition, habitat transformation from forest to pasture has favored weedy species there (Feinsinger 1988). This study sought to combine hummingbird distributions across both habitat type and elevation, as the
interaction of habitat disturbance and climate change is not well understood. Past studies have shown changes in hummingbird communities for either climate or habitat changes (Oliver 1993; Donnelly 1998; Smith 2000; Lynn 2001; Winchell 2001; Spear 2004). In order to observe these changes, I hung hummingbird feeders in different elevational zones in both forest and pasture. I found that Purple-throated Mountain Gems (Lampornis calolaemus) have shifted upward in elevation, and Violet Sabrewings (Campylopterus hemileucurus) and Green-crowned Brilliants (Heliodoxa jacula) show trends towards upward movement as well. Also, I observed that the communities of both fields and forest habitats have been altered. I saw no open-area specialist species, very low numbers of any species in pasture sites, and I did not observe some common forest species. My data suggest possible changes in the composition of the hummingbird communities of Monteverde.
El cambio climtico ha afectado la composicin de las comunidades de aves en Monteverde, Costa Rica en trminos de elevacin (Pounds et al. 1999). Adems la transformacin del hbitat de bosques a potreros a favorecido ciertas especies de malas hierbas all (Feinsinger 1988). Este estudio trato de combinar la distribucion del colibr en funcin del tipo de hbitat y la elevacin, debido a que no se ha entendido muy bien la interaccin de la perturbacin del hbitat y el cambio climtico.
Text in English.
Hummingbirds--Costa Rica--Puntarenas--Monteverde Zone
Colibres--Costa Rica--Puntarenas--Zona de Monteverde
Colibres--Variacin--Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology 2009
Ecologa Tropical 2009
Hbitats del potrero
Hbitats del bosque
t Monteverde Institute : Tropical Ecology