USF Libraries

Rodent diversity in relation to Atlantic and Pacific slopes in a neotropical cloud forest

MISSING IMAGE

Material Information

Title:
Rodent diversity in relation to Atlantic and Pacific slopes in a neotropical cloud forest
Translated Title:
La diversidad de roedores en relación a las pendientes del Atlántico y el Pacífico en un bosque nublado neotropical ( )
Physical Description:
Book
Language:
English
Creator:
Smith, Justine
Publication Date:

Subjects

Subjects / Keywords:
Rodents   ( lcsh )
Species diversity   ( lcsh )
Costa Rica--Puntarenas--Monteverde Zone   ( lcsh )
Roedores
Diversidad de especies
Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology Fall 2009
Ecología Tropical Otoño 2009
Genre:
Reports   ( lcsh )
Reports

Notes

Abstract:
Mountainous regions in Costa Rica are characterized by windward/leeward effects caused by easterly trade winds and orographic cloud formation. The Pacific slope is generally drier and more seasonal compared to the Atlantic slope. These factors tend to favor higher biodiversity for a given taxon on the Atlantic slope. Live traps were used to test this for a tropical Cloud Forest rodent community on either side of the Continental Divide. Six species were captured overall. All six species were present on the Atlantic slope and three were present on the Pacific slope. As predicted, rodent diversity, including richness and evenness, was significantly higher on the Atlantic slope than the Pacific slope. The proportion of P. mexicanus in comparison to other species was higher on the Pacific slope. The results indicate that climatic factors are principle drivers of rodent diversity. It appears that generalist and dry-adapted species are outcompeting and potentially causing range shifts in other native species. Alterations in abundance and range may be due to regional climate fluctuations and the flexibility of species to adapt to these changing conditions.
Abstract:
Las regiones montañosas en Costa Rica están caracterizadas por barloventos y sotaventos causados por los vientos alisios del este y la formación orográfica de las nubes. La costa pacífica es generalmente seca y más estacionaria comparada con la costa atlántica. Estos factores tienden a favorecer una mayor diversidad de diversos taxones en la costa atlántica. Se utilizaron trampas vivas para probar esta hipótesis para la comunidad de roedores en el bosque nuboso a ambos lados de la división continental. Seis especies fueron capturadas en total. Las seis especies se encontraron en la costa atlántica y tres en la costa pacífica. Como predije, la diversidad de roedores, incluyendo la riqueza y la equidad, fue significativamente mayor en la costa atlántica que en la pacífica. La proporción de P. mexicanus en comparación con otras especies fue mayor en la costa pacífica. Estos resultados indican que los factores climáticos son los principales causantes de la diversidad de roedores. Parece que generalistas y especies adaptadas a ambientes secos están compitiendo y causando cambios en el rango de otras especies nativas. Las alteraciones en la abundancia y el rango se puede deber a fluctuaciones climatológicas a nivel regional y la flexibilidad de las especies para adaptarse a estas condiciones cambiantes.
Language:
Text in English.
General Note:
Born Digital

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - M39-00332
usfldc handle - m39.332
System ID:
SFS0001501:00001


This item is only available as the following downloads:


Full Text
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 M39-00332
040
FHM
0 041
eng
049
FHmm
1 100
Smith, Justine
242
La diversidad de roedores en relacin a las pendientes del Atlntico y el Pacfico en un bosque nublado neotropical
245
Rodent diversity in relation to Atlantic and Pacific slopes in a neotropical cloud forest
260
c 2009-11
500
Born Digital
3 520
Mountainous regions in Costa Rica are characterized by windward/leeward effects caused by easterly trade winds and orographic cloud formation. The Pacific slope is generally drier and more seasonal compared to
the Atlantic slope. These factors tend to favor higher biodiversity for a given taxon on the Atlantic slope. Live traps were used to test this for a tropical Cloud Forest rodent community on either side of the Continental Divide. Six species were captured overall. All six species were present on the Atlantic slope and three were present on the Pacific slope. As predicted, rodent diversity, including richness and
evenness, was significantly higher on the Atlantic slope than the Pacific slope. The proportion of P. mexicanus in comparison to other species was higher on the Pacific slope. The results indicate that climatic
factors are principle drivers of rodent diversity. It appears that generalist and dry-adapted species are outcompeting and potentially causing range shifts in other native species. Alterations in abundance and
range may be due to regional climate fluctuations and the flexibility of species to adapt to these changing conditions.
Las regiones montaosas en Costa Rica estn caracterizadas por barloventos y sotaventos causados por los vientos alisios del este y la formacin orogrfica de las nubes. La costa pacfica es generalmente seca y ms estacionaria comparada con la costa atlntica. Estos factores tienden a favorecer una mayor diversidad de diversos taxones en la costa atlntica. Se utilizaron trampas vivas para probar esta hiptesis para la comunidad de roedores en el bosque nuboso a ambos lados de la divisin continental. Seis especies fueron capturadas en total. Las seis especies se encontraron en la costa atlntica y tres en la costa pacfica. Como predije, la diversidad de roedores, incluyendo la riqueza y la equidad, fue significativamente mayor en la costa atlntica que en la pacfica. La proporcin de P. mexicanus en comparacin con otras especies fue mayor en la costa pacfica. Estos resultados indican que los factores climticos son los principales causantes de la diversidad de roedores. Parece que generalistas y especies adaptadas a ambientes secos estn compitiendo y causando cambios en el rango de otras especies nativas. Las alteraciones en la abundancia y el rango se puede deber a fluctuaciones climatolgicas a nivel regional y la flexibilidad de las especies para adaptarse a estas condiciones cambiantes.
546
Text in English.
650
Rodents
Species diversity
Costa Rica--Puntarenas--Monteverde Zone
4
Roedores
Diversidad de especies
Costa Rica--Puntarenas--Zona de Monteverde
653
Tropical Ecology Fall 2009
Ecologa Tropical Otoo 2009
655
Reports
720
CIEE
773
t Monteverde Institute : Tropical Ecology
856
u http://digital.lib.usf.edu/?m39.332



PAGE 1

Rodent D iversity in Relation to Atlantic and Pacific Slopes in a Neotropical Cloud Forest Justine Smith Department of Ecology and Evolutionary Biology, University of Colorado ABSTRACT Mountainous regions in Costa Rica are characterized by windward/lee ward effects caused by easterly trade winds and orographic cloud formation. The Pacific slope is generally drier and more seasonal compared to the Atlantic slope. These factors tend to favor higher biodiversity for a given taxon on the Atlantic slope. L ive traps were used to test this for a tropical Cloud Forest rodent community on either side of the Continental Divide. Six species were captured overall. All six species were present on the Atlantic slope and three were present on the Pacific slope. As predicted, rodent diversity, including richness and evenness, was significantly higher on the Atlantic slope than the Pacific slope. The proportion of P. mexicanus in comparison to other species was higher on the Pacific slope. The results indicate that climatic factors are principle drivers of rodent diversity. It appears that generalist and dry adapted species are outcompeting and potentially causing range shifts in other native species. Alterations in abundance and range may be due to regional clima te fluctuations and the flexibility of species to adapt to these changing conditions. RESUMEN Las regiones montaosas en Costa Rica estn caracterizadas por barloventos y sotaventos causados por vientos alisios del este y la formacin orogrfica de nubes La costa pacfica es generalmente seca y ms estacionaria comparada con la costa atlntica. Estos factores tienden a favorecer un mayor diversidad de diversos taxones en la costa atlntica. Se utilizaron trampas vivas para probar esta hiptesis para l a comunidad de roedores en el bosque nuboso a ambos lados de la divisin continental. Seis especies fueron capturadas en total. Las seis especies se encontraron en la costa atlntica y tres en la costa pacfica. Como predije, la diversidad de roedores, incluyendo riqueza y equidad, fue significativamente mayor en la costa atlntica que en la pacfica. La proporcin de P. mexicanus en comparacin con otras especies fue mayor en la costa pacfica. Estos resultados indican que factores climticos son los principales causantes de la diversidad de roedores. Parece que generalistas y especies adaptadas a ambientes secos estn compitiendo y causando cambios en el rango de otras especies nativas. Alteraciones en la abundancia y rango se puede deber a fluctuac iones climatolgicas a nivel regional y la flexibilidad de las especies para adaptarse a estas condiciones cambiantes. INTRODUCTION abundance and fecundity. Rodent populati on fluctuations correlate with seasonal climatic variation in the Australian Tropics (Madsen and Shine 1999). Studies in Brazilian tropical forests have found that small mammals had diminished population m and Ernest 1995, Bergallo and Magnusson 1999). The effect of climate on different species of rodents in a community needs to be explored and understood to piece together how climate affects the community as a whole. Nevertheles s, the influence of climate remains

PAGE 2

important to note because of its vital role in rodent fecundity. The direct impact of climatic cues on rodent reproduction, and consequently relative abundance, may be especially influential in areas subject to climatic variation, such as the Monteverde Cloud Forest. Small mammals are an integral part of tropical communities, often acting as important seed dispersers and occasionally as pollinators (Lumer 1980, Howe and Smallwood 1982). They are common prey for many ca rnivores including snakes, birds of prey, and coatis, while also being key consumers of insects. Studies on rodents in the Tropics are vital because these animals are integral to ecosystem function. Studies on diversity of small mammals in Monteverde a re limited in that they have been confined to comparisons along elevational gradients on one slope. A recent study conducted in the Ro Peas Blancas watershed on the Atlantic slope of the Monteverde Cloud Forest from 750 1800 m surveyed small mammal dive rsity during three temporal seasons along an elevational gradient (McCain 2004). One of 16 species ( Peromyscus mexicanus ) dominated every elevation range, amounting to 54% of 1601 total captures. Studies completed on the Pacific slope as low as 1000 m sh ow a similar pattern but to a greater degree. Anderson (1982) found that at three elevations up to 1540 m on the Pacific slope, 76% of 1707 total captures were P. mexicanus of eight species recorded. A series of less extensive studies conducted on the Pa cific slope of the Monteverde cloud forest also found P. mexicanus to be the dominant species, representing 100% (Orlando 1994), 94% (Lee 1994), 92% (Rothman 1999), and 74% (Burdick 2008) of captures. There appears to have been a general increase in the a bundance of P. mexicanus on the the increased abundance of P. mexicanus have not yet been studied. P. mexicanus was des ago, while eight other small Tropical forests are generally understood to have high biodiversity and relatively stable communities on the regional scale. High productivity in trop ical forests results from more direct and consistent solar radiation, higher rainfall, and less seasonality, yielding more biomass than comparable temperate communities (Rosenzweig 1968). Variations in this pattern caused by differences in abiotic factors can be observed at the local scale (Holdridge 1967). Hence, local differences in diversity and productivity may be attributed to topographical or climatic factors that influence seasonality, total precipitation, and temperature. Temperature, precipitati on and seasonality in Costa Rica are largely driven by topography. Mountain ranges that run north to south separate the country into Atlantic and Pacific slopes (Haber 2000). Predominant winds come from the East, creating windward/leeward effects. Easte rly trade winds cause orographic cloud formation on the Atlantic slope of Costa Rica, resulting in high and consistent rainfall (Scholl et al. 2008). This produces a rainshadow on the Pacific slope that contributes to more pronounced seasonality and overa ll drier conditions.

PAGE 3

Species respond to these differences in climatic variation between the Pacific and Atlantic slopes. The upper Atlantic slope life zone in Monteverde (Lower Montane Rain Forest), as defined by Holdridge (1967), generally has higher b iodiversity than its drier elevational equivalent on the Pacific slope in Monteverde (Lower Montane Wet Forest). Increased precipitation often results in high productivity that increases biomass, allowing for the development of more specialized niches and therefore promoting species diversity (Pianka 1966). This characteristic of the landscape of Costa Rica makes it a particularly interesting place to study differences in biodiversity related to climatic features. The Monteverde Cloud Forest in the Cordi llera de Tilarn mountain range lies directly on the Continental Divide in Costa Rica, making it an ideal place to study these leeward/windward effects on biodiversity. This study seeks to determine if windward/leeward effects help to determine rodent d iversity on the Pacific and Atlantic slopes of the Monteverde cloud forest. If there is a difference in diversity of rodents between the two slopes, it is expected that the Atlantic slope will exhibit higher rodent diversity than the Pacific slope due to higher productivity, higher rainfall, and less seasonality. This study will also evaluate the relative abundance of P. mexicanus on each slope. It is predicted that P. mexicanus will be more dominant on the Pacific slope due to its ability to thrive in d ry and warm conditions. This study was designed to provide insight into the diversity of two rodent communities in relation to abiotic factors affecting each slope and perhaps expand our understanding of the importance of climate on the composition of tro pical rodent communities. MATERIALS AND METHODS Study Site This study took place in the Monteverde cloud forest on Cerro de los Amigos located in the Cordillera de Tilarn mountain range, Costa Rica. Two study sites were surveyed, the first on the Atla ntic slope of Cerro de los Amigos and the second on the Pacific slope. Both study sites were in a single patch of continuous forest that is part of the 22,000 hectare Monteverde Cloud Forest Reserve system. Atlantic Slope: Lower Montane Rain Forest The Atlantic slope of the Monteverde cloud forest receives high rainfall, has high humidity, and experiences little seasonality with respect to the Pacific slope (Haber 2000). The site sampled was in the Holdridge life zone Lower Montane Rain Forest (Hodlridg e 1967). The Lower Montane Rain Forest has a dense understory and a broken canopy that reaches 30 m (Hartshorn 1983). Epiphytes are common and often take the form of plants in thick mats of moss (Haber 2000).

PAGE 4

Pacific Slope: Lower Montane Wet Forest The Pacific slope site is characterized by a higher degree of seasonality and less rainfall than the Atlantic slope (Haber 2000). The site sampled was in the Holdridge life zone Tropical Lower Montane Wet Forest (Holdridge 1967). The Lower Montane Wet Forest is an evergreen forest with a distinct dry season that is abated by high mist levels (Hartshorn 1983, Haber 2000). The understory of this forest is less dense and the canopy more complete than the Lower Montane Rain Forest, harboring trees up to 50 m tall (Hartshorn 1983). Epiphytes are common but are less dense than in the Lower Montane Rain Forest and are comprised largely of orchids, bromeliads, and ferns. Trapping Twenty Sherman traps (26 x 10 x 12 cm) were placed at intervals of 5 m elevati on over a 100 m range on each slope. On Cerro de los Amigos, traps were set at 1620 1720 m on the Atlantic slope and at 1550 1650 m on the Pacific slope. Trapping took place at a lower altitude on the Pacific slope to avoid windward effects that carry ov er the Continental Divide. Traps were set at both sites for seven nights each, totaling 140 trap nights per site and 280 trap nights in total. Both sites were sampled each night of trapping to avoid weather biases. The traps were baited with an oat and peanut butter mixture. Upon capture, rodent occupants were identified to species, sexed, and recorded as new or recapture. To control for recaptures, hair was snipped from the back of each individual caught. RESULTS Two hundred and eighty total trap n ights yielded 115 captures, 69 of which were novel captures and 46 of which were recaptures. Thirty two novel individuals were captured on the Atlantic slope and 37 novel individuals were captured on the Pacific slope. Six species representing five gener a were captured overall: Heteromys desmarestianus (Forest spiny pocket mouse), Oryzomys albigularis (Montane rice rat), Oryzomys alfaroi Peromyscus mexicanus (Mexican deer mouse), Reithrodontomys gracilis (Slender harvest mouse), and S cotinomys teguina 1). All six species were present on the Atlantic slope, but only H. desmarestianus P. mexicanus and S. teguina were present on the Pacific slope. Data from the Pacific slope indicate the presence of a si ngle dominant rodent species. P. mexicanus was the predominant species, amounting to 70.3% of novel captures. The second most prominent species on the Pacific slope was S. teguina representing 24.3% of novel captures. The species with the lowest relati ve abundance was H. desmarestianus with only 5.4% of captures, representing two individuals. A very different pattern was observed on the Atlantic slope, as the three dominant species appeared equally distributed. S. teguina occupied 31.3% of traps, whil e both P. mexicanus and H. desmarestianus occupied 28.1%. The remaining three species were very rare, with abundance frequencies of 6.3% for O. alfaroi and 3.1% for O. albigularis and R. gracilis

PAGE 5

The relative abundance of rodent species captured per slop e was significantly different 2 =16.48, p<0.01, df=5). Species richness was higher on the Atlantic slope (S A =6, S P =3). Evenness was also higher on the Atlantic slope (E A =0.872, E P =0.648). To compare diversity estimates on each slope, Shannon Weiner indices of diversity were comput ed and compared. The Shannon A =1.560) and Pacific slope P =0.712) were significantly different (Zar modified t test; t=2.99, p<0.01, df=52.84; Fig. 1). Figure 1. Abundances of rodent species per slope. A s ample based rarefaction curve was produced to estimate richness on each slope using the statistical program EstimateS (Fig. 2). Based on the sampling data, expected richness for the Atlantic slope was ACE A =9. Expected richness for the Pacific slope was A CE P =3. This suggests that all species present on the Pacific slope were likely to have been detected while three species on the Atlantic slope were not represented in the data. The locations of occurrences of two species captured are inconsistent with th e published mammal list for Monteverde (Hayes and Laval 1989). Two species were found out of their expected ranges. R. gracilis is reported to live only on the Pacific slope and is not known from the Lower Montane Rain Forest (Hayes and LaVal 1989, Reid 1997); however, the one specimen captured was found on the Atlantic slope in the Lower Montane Rain Forest (at 1700 m). O. alfaroi is also not reported to live in the Lower Montane Rain Forest, yet both specimens captured were found there (at 1625 and 170 5 m). Both species were found above their local range upper limit of 1500 1550 m (McCain 2002).

PAGE 6

Figure 2. Rarefaction curve of predicted species observed based on sampling effort. This model indicates that sampling effort was sufficient to detect al l species present in the environment at the Pacific site but not the Atlantic site. The proportions of P. mexicanus on both slopes were significantly different ( 2 =12.19, p<0.01, df=1), the higher proportion of captures occurring on the Pacific slope. P mexicanus is reported to prefer secondary or semi deciduous forests, yet it was found in abundance in primary, evergreen forest (Reid 1997). Although its range extends to 3000 m elevation, no specimens were caught above 1685 m. P. mexicanus is most abu ndant to 1500 m (Reid 1997). A shift in abundances of individual species in the region overall was observed. R. gracilis and O. albigularis were both recorded as common in The Mammals of Monteverde but were each only captured on one occasion (Hayes and L aVal 1989). There appeared to be elevational partitioning in the Atlantic site surveyed, but not in the Pacific site. All five species captured on the Atlantic slope apart from P. mexicanus were detected above its upper range, while only three of them sp atially coexisted with P. mexicanus The two species not found coexisting with P. mexicanus were R. gracilis and O. albigularis There does not appear to be any trend on the Pacific slope in relation to competition leading to species exclusion within ele vational groups. A Chi square analysis of abundances of rodent species in four elevational groups per slope indicate that differences in abundances from expected abundances were marginally significant on the 2 =23.94, p<0.1, df=15; Fig. 3A) and not significant on the Pacific slope 2 =6.12, p>0.1, df=10; Fig. 3B).

PAGE 7

Figure 3. Abundances per elevation on the Pacific slop (3A) and the Atlantic slope (3B). Abundances of the three species captured on the Pacific slope were not signifi cantly different than expected values (3A). P. mexicanus and S. teguina were present at all elevations, while H. desmarestianus was present only in the middle of the range. Abundances of the six species captured on the Atlantic slope were marginally signif icantly different (3B). P. mexicanus and O. alfaroi were not present in the highest quadrant, S. teguina was not present in the lowest quadrant, and both R. gracilis and O. albigularis were only found in the highest quadrant. H. desmarestianus was present throughout the range. 3B 3A

PAGE 8

Table 1. Species Accounts. Heteromys desmarestianus Heteromys desmarestianus is a medium to large pocket mouse (Reid 1997). Its distribution is from Mexico to Columbia in lowlands to 2400 m. It is common in evergreen and semideciduous forests and often occurs in areas of secondary growth it is a burrowing mouse and stores seeds in caches. Breeding occurs year round, and individuals of this species are relatively long lived. It eats mostly seeds but may also feed on fruit and insects. Oryzomys albigularis Oryzomys albigularis is a large rice rat (Reid 1997). It is found from the mountains of Costa Rica to Northern Peru from 1000 3000 m. It is locally common in highland, evergreen, primary forest, often by streams and n ear mountaintops. It is known to be a good swimmer. It primarily feeds on fruits and seeds, but may eat insects as well. Oryzomys alfaroi Oryzomys alfaroi is a small rice rat (Reid 1997). Its distribution is from Mexico to Ecuador in lowlands to 3250 m. It is found in evergreen or semideciduous forests often near water, but is uncommon. It is known to make nests of shredded bark. It feeds mostly on fruits and seeds, but may supplement its diet with insects. Peromyscus mexicanus Peromyscus mexicanu s is a medium to large deer mouse (Reid 1997). It is found from Mexico to Panama in lowlands to 3000 m, but usually occurs between 600 1500 m. It is widespread and very common, living primarily in semideciduous or secondary forests but occurring in a var iety of habitats. Breeding occurs year round and adults reside in burrows. It is primarily insectivorous but also eats seeds and green plant material. Reithrodontomys gracilis Reithrodontomys gracilis is a small harvest mouse (Reid 1997). It occurs from Mexico to mountainous regions in Costa Rica in lowlands to 1800 m. It is locally common in a variety of habitats. Reproduction is largely constricted to the wet season and nests in hollows. It eats insects and seeds. Scotinomys teguina Scotinomys teguina is a tiny singing mouse (Reid 1997). Its distribution is in the highlands of Mexico to Panama from 900 2900 m. It is often abundant in highland forests, clearings, and edges. Breeding occurs year round and elaborate nests are made by both sexes It is primarily insectivorous, yet it occasionally eats fruits and seeds. DISCUSSION The data indicate that species abundances differ significantly from the expected values on the Atlantic and Pacific slopes. The Shannon index was significantly hi gher on the Atlantic slope than on the Pacific slope, indicating higher diversity on the Atlantic slope. Richness and evenness were both higher on the Atlantic slope. Sampling based richness projections predicted nine species on the Atlantic and three on the Pacific. The data show that sampling was likely incomplete on the Atlantic slope but complete on the Pacific, indicating a larger difference in richness than was observed. Abiotic factors contributing to the patterns observed are likely due to windw ard/leeward effects of differential cloud formation and precipitation on each slope. The expected relationship of higher biodiversity of rodents on the Atlantic slope was observed, likely

PAGE 9

due to higher productivity, increased rainfall, and less seasonalit y. These conditions have a greater ability to support high species richness. However, some data could not be explained by climatic factors attributed to slope. Range discrepancies observed in R. gracilis and O. alfaroi demand attention. In addition, su btle patterns in elevational partitioning were detected on the Atlantic slope. It has been suggested that changes in biotic populations in Monteverde may be due to greater levels of disturbance and fragmentation on the Pacific slope. This explanation doe s not transfer to rodent populations. There is no evidence for a correlation between degree of disturbance and rodent species richness (Van der Bergh and Kappelle 1998). Instead, the above atypical observations are likely elements of the reaction of the biotic community to climate change in Monteverde. Warming of sea surface temperatures is causing higher orographic cloud formation and decreased mist in the Monteverde cloud forest (Pounds et al. 1996). Costa Rican cloud forests are enduring microclimat e changes in precipitation and temperature, particularly in the dry season (Still et al. 1999). Leeward slopes in Costa Rica are experiencing a precipitation decrease, while windward slopes are receiving a precipitation increase (Vargas and Trejos 1994). In short, windward/leeward effects are magnifying, drastically affecting mountainous regions of Costa Rica. Climatic changes lead to a shift in the range of life zones by directly altering the biotic community. Life zones at high elevations have been fo und to be highly sensitive to changes in temperature (Enquist 2002). The Lower Montane Rain Forest has particularly low resistance to climate variation, possibly due to its traditionally stable conditions. These factors all make Monteverde susceptible to magnified biotic alteration as the climatic conditions continue to change. Montane environments are unique because they are often associated with high biodiversity despite limited range size (McCain 2006). Range size decreases with altitude due to spati al constraints as area decreases near the tops of mountains. However, elevational clines provide a wide range of habitats that can be specialized upon by competing species. They may also represent the presence of transition zones between species rich are as. A combination of environmental variation and range size constraints has created a spike in rodent diversity in Monteverde at mid elevations (McCain 2004). This spike coincides with high precipitation at mid elevations. This study was completed just above mid elevations, but many of the factors influencing diversity remain the same. P. mexicanus is a species primarily found in secondary or semi deciduous forests (Reid 1997). S. teguina and H. desmarestianus are present in secondary forests and edges but are more common in evergreen forests. All three species are reported to breed year round. These three species, P. mexicanus in particular, should be more resistant to changes in climate because of their ability to thrive and reproduce year round an d in disturbed habitats. The data exhibited that P. mexicanus was in fact more dominant on the Pacific slope. Only 20 years ago, all six species captured were reported to live in the Lower Montane Wet Forest of the Pacific slope (Hayes and LaVal 1989). Of the three species only present on the Atlantic slope, only O. albigularis was previously reported to live in the Lower Montane Rain Forest. The two species of Oryzomys captured usually live in wet highland forests. R. gracilis is known to reproduce th roughout the wet season

PAGE 10

and only sporadically during the dry season (Ried 1997). The biological constriction of habitat type and reproductive phenology in these three species accompanied with decreased precipitation may have led to their disappearance fro m the Pacific slope and appearance on the Atlantic slope. The capability for P. mexicanus S. teguina and H. desmarestianus to reproduce year round regardless of season favors them in variable environments. This can explain their exclusive dominance on the Pacific slope of Cerro de los Amigos, regardless of climate change. These three species were also dominant on the Atlantic slope, with the other three species representing a very small subset of specimens captured. Recent temperature and precipitatio n changes on the Atlantic slope are likely causes of this shift, decreasing populations of rodents more reliant on specific climatic conditions and opening up space for less specialized species like P. mexicanus S. teguina and H. desmarestianus This ma y explain the presence of O. albigularis and R. gracilis detected only above the observed range of P. mexicanus at the Atlantic site. Resource availability is an unlikely explanation for range shifts in the six observed rodent populations. P. mexicanus S. teguina and R. gracilis are all primarily insectivorous, while H. desmarestianus O. albigularis and O. alfaroi feed mostly on seeds and fruit (Reid 1997). All six species are omnivorous and supplement their diet with many different food types. Flex ibility of diet in all of these species indicates that resource availability is probably not the limiting factor on the ranges of rodent species in Monteverde. This study suggests that rodent diversity in a middle to high elevation Tropical Cloud Forest is likely limited by climatic factors and windward/leeward effects. Past studies indicate that resource availability and habitat disturbance are not major contributors to this pattern, although future studies could test these relationships in the Montever de Cloud Forest. While rodent diversity was higher on the Atlantic slope, it is unclear of how the magnitude of each climatic factor and their current shifts produces this pattern in the light of current climate change. If niche partitioning is limited d ue to less stable abiotic conditions, it may result in decreased rodent diversity. Monitoring of precipitation and their effects on fecundity is necessary to understand fluctuations in diversity. Further studies controlling for precipitation, temperature and resource availability should be conducted to further understand this phenomenon. ACKNOWLEDGEMENTS I would like to thank Alan Masters and Pablo Allen for their guidance in formulating and carrying out this study. I would like to thank Anjali Kuma r for her assistance in statistical analysis. I would like to thank Jose Caldarn for his expertise in trapping and handling small rodents and for all the materials supplied. I would like to thank my familia tica for their endless hospitality. I would li ke to thank the Estacin Biologa for the extensive use of their forest. Finally, I would like to thank my classmates for being the most fascinating personalities, the most curious minds, and the most phenomenal friends that I could have ever imagined.

PAGE 11

LITERATURE CITED A NDERSON S TEPHEN D 1982. Comparative population ecology of Peromyscus mexicanus in a Costa Rican wet forest. Unpublished data. B ERGALLO H ELENA G ODOY AND W ILLIAM E RNEST M AGNUSSON 1999. Effects of climate and food availability o n four rodent species in southeastern Brazil. Journal of Mammalogy 80.2: 472 486. B URDICK C AROLINE 2008. Determining population densitites of Peromyscus mexicanus (Muridae) in Monteverde, Costa Rica: A follow up of Rothman (1999). CIEE Tropical Ecolo gy and Conservation Spring: 273 283. E NQUIST C. A. F 2002. Predicted regional impacts of climate change on the geographical distribution and diversity of tropical forests in Costa Rica. Journal of Biogeography 29: 519 534. H ABER W ILLIAM A. 2000. Pla nts and vegetation. Monteverde Eds. Nalini M. Nadkarni and Nathaniel T. Wheelwright. New York: Oxford University Press. 39 90. H ARTSHORN G. S. 1983. Plants. Costa Rican Natural History Ed. Daniel Janzen. Chicago: University of Chicago Press. 1 18 157. H AYES M ARC AND R ICHARD L AVAL 1989. The Mammals of Monteverde San Jose, Costa Rica: Tropical Science Center. H OLDRIDGE L. R 1967. Life Zone Ecology San Jose: Tropical Science Center. H OWE H. F AND J. S MALLWOOD 1982. Ecology of se ed dispersal. Annual Review of Ecology and Systematics 13: 201 228. L EE C OLIN 1994. Windbreak effects on rodent communities. EAP Tropical Biology Fall: 61 69. L UMER C. 1980. Rodent pollination of Blakea (Melastomataceae) in a Costa Rican cloud for est. Brittonia 32.4: 512 517. M ADSEN T HOMAS AND R ICHARD S HINE 1999. Rainfall and rats: Climatically driven dynamics of a tropical rodent population. Australian Journal of Ecology 24: 80 89. M ARES M ICHAEL A. AND K RISTINA A. E RNEST 1995. Populatio n and community ecology of small mammals in a gallery forest of central Brazil. Journal of Mammalogy 76.3: 750 768. M C C AIN C HRISTY M 2004. The mid domain effect applied to elevational gradients: Species richness of small mammals in Costa Rica. Journa l of Biogeography 31: 19 31. M C C AIN C HRISTY M. 2006. Do elevational range size, abundance, and body size patterns mirror those documented for geographic ranges? A case study using Costa Rican rodents. Evolutionary Ecology Research 8: 435 454. O RLANDO A NNE 1994. Population structures and species diversity or small rodent communities in varied habitats. CIEE Tropical Ecology and Conservation Summer: 215 228. P IANKA E RIC R. 1966. Latitudinal gradients in species diversity: a review of concepts. Th e American Naturalist 100.910: 33 46. P OUNDS A LAN J ET AL 2006. Widespread amphibian extinctions from epidemic disease driven by global warming. Nature 439: 161 167. R EID F IONA A 1997. A Field Guide to the Mammals of Central America and Southeast Mexico. New York: Oxford University Press. R OSENZWEIG M ICHAEL L 1968. Net Primary Productivity of Terrestrial Communities: Prediction from Climatological Data. The American Naturalist 102.923: 67 74. R OTHMAN A NDREW 1999. Effects of climate chan ge and elevation on small rodent communities in Monteverde, Costa Rica. CIEE Tropical Ecology and Conservation Spring: 160 170. S CHOLL M. A. ET AL 2008. Cloud water in windward and leeward mountain forests: The stable isotope signature of orographic c loud water. Water Resources Research 43.W12411: 1 13. S TILL C HRISTOPHER J., P RUDENCE N. F OSTER AND S TEPHEN H. S CHNEIDER 1999. Simulating the effects of climate change on tropical montane cloud forests. Nature 398: 608 610. V AN DEN B ERGH M. B. AND M. K APPELLE 1998. Diversity and distribution of small terrestrial rodents along a disturbance gradient in montane Costa Rica. Revista de Biologa Tropical 46.2: 331 338. V ARGAS A LVARO B RENES AND V ICTOR F. S ABORIO T REJOS 1994. Changes in the general circulation and its influence on precipitation trends in Central America: Costa Rica. Ambio 23.1: 87 90.