1 Effects of fragmentation and human disturbance on mammal communities in San Luis and Monteverde, Costa Rica Lindsay Crawford Environmental Analysis Department, Pomona College ABSTRACT Mammals occupy important roles in tropical ecosystems and can often be keystone species, controlling the composition of their communities. This study compared mammal communities in Monteverde and San Luis, Costa Rica, two regions that differ in their current and historical patterns of fragmentation. Three habitats per re gion, representing a disturbance gradient, were surveyed using scent stations. Prints as well as visual observations were recorded. No significant differences were found in species richness across zones or across habitats. However, overall diversity was consistently higher in Monteverde and locally highest in edge habitat in both regions. Local richness patterns (across habitats) follow the Intermediate Disturbance Hypothesis and show that edge habitat acts as an area of overlap between disturbed and int act fragmentation and increased pressure exerted by the largely agricultural community. This study points to the conservation benefits of large, continuous reserves and reports on the current and future threats of fragmentation and disturbance to mammal communities. RESUMEN Los mamferos tienen posiciones importantes en ecosistemas tropicales y a menudo pueden ser especies clave, controlando la composicin de sus comunidades. Este estudio compar las comunidades de mamferos de Monteverde y San Luis, Costa Rica; estas son dos regiones que difieren en sus patrones actuales e histricos de fragmentacin. Tres hbitats por regin, representando un gradiente de perturbaciones, fueron examinados usando estaciones de olor. Huellas y observaciones visuales fueron registradas. Ninguna diferencia significativa fue encontrada ni entre las regiones ni entre los hbitats. Sin embargo, la riqueza general fue consistentemente ms alta en Monteverde y ms alta localmente al borde del bosque en las dos regiones. Los patrones locales de riqueza (entre hbitats) se explican por la hiptesis de perturbacin intermedia y mostraron que el hbitat de borde acta c omo un rea de solapacin entre el hbitat perturbado y el intacto. La diversidad total de San Luis se atribuye a una composicin y distribucin de especies que han sido distorcionadas por la fragmentacin y la mayor presin proveniente de la comunidad ag rcola. Este estudio seala los beneficios de conservacin de reservas grandes y continuas e indica las amenazas actuales y futuras de la perturbacin y la fragmentacin para las comunidades de mamferos. INTRODUCTION Mammals play a pivotal role in trop ical ecosystems. Simply due to the sheer amount of resources they consume, their collective effect on the community is significant (Eisenberg 1989). Seed and seedling predators, including agoutis, pacas, and other herbivorous mammals, heavily impact the floral community composition. Mammals like They tend to feed opportunistically on whichever food source is currently abundant.
2 These two mammal groups in turn are prey fo r the top predators, such as pumas and jaguars. The top carnivores are often keystone species, whose presence maintains a healthy balance of prey populations and thus dictates the composition of the community through top down control. Ecosystem wide co nsequences can result from the absence of a keystone predator, as was demonstrated on Barro Colorado Island (BCI) in Panama. The land mass was isolated from the mainland in the early 20 th century during the construction of the Panama Canal, and jaguars an d pumas subsequently disappeared, along with aerial predator species, due to insufficient area to support viable populations (Terborgh 1992). Consequently, populations of agoutis, pacas, and coatis exploded, reaching abundances an order of magnitude highe r than natural mainland populations. By 1970, 45 bird species had disappeared from the island due to nest predation by mesopredators. Additionally, the abnormal abundance of seed predators has resulted in reduced floral diversity and a marked change in f loral community composition (Terborgh 1992). In this way, changes in top predator populations can reverberate through the trophic pyramid to 2004). The case of BCI is only one well documented example of the extensive impact that humans can have on the natural environment. Fragmentation may not always form a habitat, as suggested by MacAr Species area relationships suggest that a smaller fragment of habitat will not be able to support as many species as a large, continuous habitat. The isolation of habitat remnants reduces habitat quality and limits genetic exchange with surrounding populations, increasing risk of extinction (Bentley 2000). Fragmentation also increases the habitat area that is exposed to disturbance. Edge effects change microclimatic conditions by bringing increased temp eratures, wind, and light penetration, and reduced humidity (Laurance and Bierregaard 1997). Habitat quality is further degraded by introduction of exotic or invasive species (e.g. domestic animals), noise, air, and light pollution, and human presence (Go osem 1997). Worse, large mammals are often directly targeted by humans as food, game, or agricultural pests. The aim of this study was to determine the extent of human impact on mammal communities in the Monteverde area. To examine the factor of f ragmentation, two regions differing in level of deforestation and land use history were compared. To investigate the effects of disturbance, three habitats representing a gradient in the intensity of disturbance were surveyed. It was hypothesized that th e diversity and composition of mammal populations would reflect differences in current and historical anthropogenic impacts. Specifically, diversity was expected to have an inverse relationship with fragmentation and disturbance.
3 MATERIALS AND METHOD S Study Sites This study was conducted in November 2004 for two adjacent life zones in the region of Monteverde, Costa Rica (Appendix I, Fig. 1). The Monteverde site was in and around the Biological Station property at an average elevation of 1550 m, pl acing it into the Lower Montane Wet Forest Life Zone (Hayes and Laval 1989). This forest is part of the 29,000 ha Monteverde Reserve Complex, most of which was established in the 1970s (Burlingame 2000). The San Luis site was in and around the EcoLodge. At an altitude of 1100 m, this region falls into the Premontane Moist Forest Life Zone (Hayes and Laval 1989). The EcoLodge property is 66 ha total. Forty hectares are covered by secondary forest, which had coffee planted in the understory about 15 year s ago (EcoLodge staff personal communication). As a result, the forest structure is characterized by large canopy trees and regenerating growth in the understory. It is important to note that the San Luis Valley is still largely agricultural and therefor e deforestation is more extensive than in the neighboring region of Monteverde, where conservation has been a priority and ecotourism is a main industry. Although it has habitat corridors to the Monteverde Reserve and Bosque Eterno de los Nios (pers. comm .), the EcoLodge forest is significantly more fragmented and isolated than the Monteverde site. Three replicate sites were set up for each habitat type: disturbed, edge, and forest. Disturbed habitat was completely open (no tree cover), having been cleare d by humans. Being in the vicinity of dwellings and roads, it was constantly exposed to human activity and light and noise pollution. Edge sites were situated on the border of secondary forest and disturbed areas. This habitat was subjected to edge effe cts but was also sheltered somewhat by the forest. The final habitat type was in secondary forest, under complete tree cover and at least 100 m from the forest edge. All forest stations were alongside established trails, which, despite disturbance caused by human use, are also well used by large mammals. The nine sites in San Luis were scattered along the road to the EcoLodge and on the Camino Real trail (Appendix I, Fig. 2). The Monteverde plots were located along a loop including the road to the station the Lower Loop trail, and the TV tower road (Appendix I, Fig. 3). The two loops were of comparable distance, each taking approximately an hour and a half walking. Data Collection The primary sampling method was scent stations, a technique pioneered in North America to census foxes that has been commonly used to survey terrestrial carnivorous or omnivorous mammal populations (Beltrn et al. 1991). Animals are attracted by scent, either of a food source or the urine of another animal, which is especia lly effective for felids (Reid 1997). Due to equipment limitations, canned tuna was used as bait in this study. It was complemented by either bananas or bread to attract non carnivorous species, as well.
4 Each station was one square meter, in which the t opsoil was loosened and homogenized with a shovel and by hand. This was then covered with a one to two centimeter layer of sand. Bait was placed in the center of the plot on a plastic plate between four and six in the evening. Tracks were noted the fol lowing morning between seven and ten. Tracks found along the trail were also recorded, as well as direct visual observation of mammals during data collection periods. Animal tracks were carefully examined, taking note of width and length of each print in millimeters, stride length if possible, and compression of soil to infer weight. Tracks were often drawn and/or digitally photographed for further reference. The main aids in field identification were plates drawn by Reid (1997) and Wainwright (2001). Assistance and verification was subsequently provided by Dr. Federico Chinchilla. RESULTS Twenty species were found in Monteverde, while there were only 11 in San Luis (Appendix II, Fig. 1). However, this difference can be attributed to disparate samp le sizes. Maintenance of the substrate was the main source of difficulty during data collection. Heavy rain blurred or erased prints, compacted the soil, and washed away the sand, while extended dry periods desiccated the sand and reduced its ability to retain animal tracks. Days on which bait had been eaten but prints were absent or impossible to read were discounted, resulting in unequal sample sizes. Additionally, data were collected on seven nights in San Luis and eight nights in Monteverde due to v arying weather conditions. In order to resolve the variance in sample size for comparison of richness, chi squared expected values were scaled to match the relative numbers of successful collection incidents between sites. Once sample size was standardize d, the difference in species richness between Monteverde and San Luis was not shown to be significant ( 2 = 1.83, df = 1, p > 0.05). Both Monteverde and San Luis followed a pattern of highest richness in edge habitat (MV = 18, SL = 8), intermediate richne ss in forest habitat (MV = 12, SL = 7), and lowest richness in disturbed areas (MV = 11, SL = 5). Again, after standardizing the sample size, the across habitat differences among zones were not found to be significant (MV: 2 = 1.47, SL: 2 = 0.32, df = 2 p > 0.05). To determine patterns in diversity, two indices were used. Shannon Weiner values indicated that diversity was consistently higher in Monteverde for each habitat type (Figure 1). In addition, the local diversity was highest in edge habitat and lowest in disturbed areas in both Monteverde and San Luis.
5 Figure 1. Shannon Weiner index values for mammal diversity organized by habitat type in Monteverde and San Luis, Costa Rica. Diversity is consistently higher in Monteverde and highest in edge habitat. Since Shannon showed higher diversity in Monteverde acros s all habitats, a peak in diversity in edge habitat, and lowest diversity in disturbed areas. Figure 3. Monteverde and San Luis, Costa Rica. Results agree w ith Shannon Weiner index values, indicating higher mammal diversity in Monteverde and highest local diversity in edge habitat.
6 composition between regions and between habitats. Wit h a value of one indicating full similarity (same composition), Monteverde and San Luis showed low overlap (C S = 0.45). Comparisons of habitat types within the regions yielded lower values in San Luis than in Monteverde (Table 1). A relevant observation i s that only one species, an opossum, was found in all three habitats in San Luis. There were seven such species in Monteverde, including non generalists such as pumas and agoutis. Table 1. habitats within Monteverde and San Luis, Costa Rica. Highest possible value is 1, representing the same species composition. Data show less habitat overlap in San Luis. Disturbed vs. Edge Edge vs. Forest Disturbed vs. Forest MONTEVERDE 0.62 0.73 0.7 SAN LUIS 0.62 0.4 0.33 Community composition was compared by organizing mammals into functional groups (Figure 4). Data show that San Luis is dominated by generalist omnivores, while the more specialized feeders (such as herbivores and insectivores) have a reduced presence or are absent. By contrast, Monteverde has representatives from four groups, and omnivores make up a much smaller proportion. MONTEVERDE SAN LUIS Figure 4. Mammal community composition by proportion of species in each feeding guild. Monteverde has a more even distribution between guilds, while San Luis is dominated by omnivores. Composition was also compar ed by relative abundance of each species (Figure 5). Monteverde showed a more even distribution among species, while San Luis was dominated by a small number of common species.
7 Figure 5. Mammal community composition by relative proportion of individuals of each species. Each color represents a different species. Distribution is more even in Monteverde than in San Luis. DISCUSSION Since species richness was not shown to vary significantly between Monteverde and San Luis, th e difference in diversity must be attributed to relative abundances. In Monteverde, the community was much more evenly distributed among species, while San Luis was dominated by only a few common species. The best explanation for this trend is the elevat ed degree of fragmentation in San Luis. Bentley (2000) found that fragments support lowered abundances of forest species, due to decreased connectivity, and tend to favor a subset of habitat generalists. Fragmentation works to produce a mammal community of distorted abundances and ultimately leads to a reduction in diversity. The effects of fragmentation are also evident in the differing compositions of the two communities. Fox (2000) found that smaller fragment size confers reduced habitat variety. Th e result is lower diversity of resources and fewer available niches. Mammals with specialized needs usually suffer, while generalists benefit. The low overlap between habitats in San Luis can possibly be attributed to an increased intensity of disturban ce due to the prevalence of agriculture, which has produced a greater disparity between disturbed and intact areas. Goosem (1997) calls this which they have evolved The surrounding community in San Luis reinforces these barriers through practices of hunting large mammals or killing them as threats to livestock. In Monteverde, the lines between habitats may be more blurred. Since agriculture is not common around th e reserve, there is less external pressure confining the animals to the forest. Both San Luis and Monteverde showed the same local pattern of diversity between habitats highest at the forest edge, intermediate within the forest, and lowest in disturbed
8 which postulates that diversity is highest at an intermediate intensity or frequency of disturbance (Mackey et al. 2001). The theory explains that environments with low leve ls of disturbance tend to be dominated by better competitors, while frequent or extreme disturbance causes low diversity by making it very difficult for species to establish and/or maintain viable populations. At intermediate levels of disturbance, the en vironment is in flux. Better competitors are kept from dominance, while species that are better adapted for disturbance are favored. The result is an equilibrium at a higher level of diversity. Also, edge habitat is the zone of overlap between highly disturbed and relatively intact areas. Since it has elements of both habitats, animals that have adapted to one of these environments can often also be found at forest edge, resulting in higher species richness there. Clearly, the mammal community in S an Luis shows signs of human impact in its lowered diversity, distorted composition, and reduced habitat overlap. However, with a small sample size, it is difficult to quantify the contribution of causal factors, because fragmentation and disturbance cann ot easily be isolated. Fox (2000) found that they have a tendency to amplify each other. For example, prolonged hunting pressure combined with forest clearing to distort the mammal community in the Amazon, but the two factors could not be separated due t o their correlation (Lopes and Ferrari 2000). This case shows marked resemblance to the situation in San Luis. Further study with a larger sample size and a longer duration are necessary to assess the relative contributions of each factor. Lopes and Fer rari (2000) also found that the area of intact forest that remained was the largest determinant of mammal communities in fragments. Preserving undisturbed areas not only benefits its inhabitants, but it can also enhance nearby disturbed areas, providing a n additional incentive for conservation. In this way, the Monteverde Reserve Complex may have helped to maintain mammal populations in the fragmented area of San Luis. Nevertheless, human impact is evident in the mammal community in San Luis, which is no t yet out of danger, as indirect effects have not fully run their course. Increasing the area of the fragment and its connectivity with the MRC would help to counter these effects and would offer great benefits to the mammal community and the ecosystem as a whole. ACKNOWLEDGMENTS I would like to thank Alan Masters for his patient guidance, Javier Mndez for his willing and prompt assistance, Ollie Hyman for his generous donation of time and physical labor, and Federico Chinchilla for his much needed ad vice and wealth of knowledge on the subject. Thanks to the EcoLodge staff for the gracious and timely approval of the study. Last and most, thank you to my compaera Holle Harjer for allowing and encouraging me to partake in the project and carrying it t hrough with enough humor and enthusiasm for both of us. LITERATURE CITED Beltrn, Juan Francisco, Miguel Delibes, and Jaime R. Rau. 1991. Methods of censusing red fox ( Vulpes vulpes ) populations. Hystrix 3: 199 214. Bentley, J.M., C.P. Catterall, and G.C. Smith. 2000. Effects of fragmentation of Araucarian vine forest on small mammal communities. Conservation Biology 14(4) 1075 1087.
9 Burlingame, Leslie J. 2000. Conservation in the Monteverde zone: contributions of conservation organizations. In Nadk arni, N.M. and N.T. Wheelwright. Monteverde: Ecology and Conservation of a Tropical Cloud Forest Oxford University Press, New York. Pp. 351 375. Eisenberg, John F. 1989. Mammals of the Neotropics. Vol. 2. The University of Chicago Press, Chicago. Fox, Barry J. and Marilyn D. Fox. 2000. Factors determining mammal species richness on habitat islands and isolates: habitat diversity, disturbance, species interactions and guild assembly rules. Global Ecology and Biogeography 9(1): 19 37. Goosem, Miriam. 19 97. Internal fragmentation: the effects of roads, highways, and powerline clearings on movements and mortality of rainforest vertebrates. In Laurance, W.F. and R.O. Bierregaard, Jr., eds. Tropical Forest Remnants: Ecology, Management, and Conservation of Fragmented Communities University of Chicago Press, Chicago. Pp. 241 255. Hayes, Marc and Richard Laval. 1989. The Mammals of Monteverde: An annotated checklist to the mammals of Monteverde. Tropical Science Center, San Jose, Costa Rica. Laidlaw, Ruth K. 2000. Effects of habitat disturbance and protected areas on mammals of peninsular Malaysia. Conservation Biology 14(6): 1639 1648. Laurance, W.F. and R.O. Bierregaard, Jr., eds. Tropical Forest Remnants: Ecology, Management, and Conservation of Fragme nted Communities. University of Chicago Press, Chicago. Lopes, M.A. and S.F. Ferrari. 2000. Effects of human colonization on the abundance and diversity of mammals in Eastern Brazilian Amazonia. Conservation Biology 14(6): 1658 1665. MacArthur, R.H. and E.O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, NJ. Mackey, Robin L. and David J. Currie. 2001. The diversity disturbance relationship: is it generally strong and peaked? Ecology 82(12): 3479 3492. Masters, Al an. 2004. Tropical Community Ecology Reader. CIEE, Monteverde, Costa Rica. Reid, Fiona A. 1997. A Field Guide to the Mammals of Central America and Southeast Mexico. Oxford University Press, New York. Terborgh, John. 1992. Diversity and the Tropical Rain forest. Scientific American Library, New York. Wainwright, Mark, Illustrator. 2001. Costa Rica Field Guide: Animal Tracks. Rainforest Publications, Costa Rica.
10 APPENDIX I Figure 1. Map of Monteverde region showing location of two study sites (marked with stars). Dark shading indicates protected areas, while lighter shading is forested. Unshaded areas are deforested. Source: Nadkarni, N.M. and N.T. Wheelwright. 2000. Monteverde: Ecology and Conservation of a Tropical Cloud Forest. Oxford University Press, New York.
12 Figure 3. Map (not to scale) showing location of 9 study sites in Monteverde, Costa Rica. Sites labeled with D for disturbed, E for edge, or F for forest habitat, accompanied by number of replicate.
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 M39-00376
Efectos de la fragmentacin y perturbacin humana en las comunidades de mamferos en San Luis y Monteverde, Costa Rica
Effects of fragmentation and human disturbance on mammal communities in San Luis and Monteverde, Costa Rica
Mammals occupy important roles in tropical ecosystems and can often be keystone species, controlling the composition of their communities. This study compared mammal communities in Monteverde and San Luis, Costa Rica, two regions that differ in their current and historical patterns of fragmentation. Three habitats per region, representing a disturbance gradient, were surveyed using scent stations. Prints as well as visual observations were recorded. No significant differences were found in species richness across zones or across habitats. However, overall diversity was consistently higher in Monteverde and locally
highest in edge habitats in both regions. Local richness patterns (across habitats) follow the Intermediate Disturbance Hypothesis and show that edge habitat acts as an area of overlap between disturbed and intact habitats. San Luiss reduced diversity is attributed to a skewed composition and distribution of species due to fragmentation and increased pressure exerted by the largely agricultural community. This study points to the conservation benefits of large, continuous reserves and reports on the current and future threats of fragmentation and disturbance to mammal communities.
Los mamferos tienen posiciones importantes en ecosistemas tropicales y a menudo pueden ser especies clave, controlando la composicin de sus comunidades. Este estudio compar las comunidades de mamferos de Monteverde y San Luis, Costa Rica; estas son dos regiones que difieren en sus patrones
actuales e histricos de fragmentacin. Tres hbitats por regin, representando un gradiente de perturbaciones, fueron examinados usando estaciones de olor. Huellas y observaciones visuales fueron
registradas. Ninguna diferencia significativa fue encontrada ni entre las regiones ni entre los hbitats. Sin embargo, la riqueza general fue consistentemente ms alta en Monteverde y ms alta localmente al borde del bosque en las dos regiones. Los patrones locales de riqueza (entre hbitats) se explican por la hiptesis de perturbacin intermedia y mostraron que el hbitat de borde acta como un rea de solapacin entre el hbitat perturbado y el intacto. La diversidad total de San Luis se atribuye a una composicin y distribucin de especies que han sido distorcionadas por la fragmentacin y la mayor presin proveniente
de la comunidad agrcola. Este estudio seala los beneficios de conservacin de reservas grandes y continuas e indica las amenazas actuales y futuras de la perturbacin y la fragmentacin para las
comunidades de mamferos.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone
Fragmentacin del hbitat
Diversidad de especies
Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology Fall 2004
Ecologa Tropical Otoo 2004
t Monteverde Institute : Tropical Ecology