Davis 1 Presence and diversity of mammals across m icrohabitats in San Luis , Monteverde , Costa Rica Liesel Davis Department of Ecology and Evolutionary Biology University of California, Los Angeles EAP Tropical Biology and Conservation Program , Spring 2018 8 June 2018 ______________________________________________________________________________ ABSTRACT The transformat i on from natural to human dominated landscapes is occurring rapidly throughout much of the world. In S an Luis, Costa Rica, undisturbed forest land has been converted into farmland, introducing crops for agriculture and free ranging domestic animals. The aim of this study was to observe how relative local mammal abundance has been impacted by these anthropological factors. I sa mpled three locations in San Luis: a coffee plantation (Caf Ã© Bella Tica), a farm containi ng domestic animals (Marco MarÃn farm), and a forest microhabitat. To collect data, I used trail transects, miscellaneous observations, Sherman rodent traps, and cam era traps over the course of 14 days. I found a total of seven wild mammal species and 42 wild mammal sightings across the three microhabitats. Four mammal species were f ound among the coffee plantation (variegated squirrel Sciurus variegatoides , olingo Bassaricyon gabbii , spiny pocket mouse Heteromys desmarestianus , deer mouse Peromyscus mexicanus ), three were on the farm with domestic mammals (variegated squirrel, agouti Dasyprocta punctata , hispid cotton rat Sigmodon hispidus ), and three were in the fo rest (white faced capuchin Cebus imitator , agouti, deer mouse). The farm with domestic dogs and cats contained more arboreal wild mammals than terrestrial, which may suggest substantial predatory behavior from the domestic pets. With no significant species richness difference across microhabitats, these results indicate that either human presence is not negatively impacting mammal diversity in San Luis, or that the effect from nearby plantations actually diminished diversity in the studied forest habitat. _ _____ ______________________________________________________________________ Presencia y diversidad de mamÃferos entre microh Ã¡bitats en San Luis de Monteverde , Costa Rica RESUMEN La transformaciÃ³n de los paisajes natural es a dominados por actividades humanas estÃ¡ ocurriendo rÃ¡pidamente en gran parte del mundo. En San Luis Monteverde , Costa Rica, el bosque original se ha converti do en Ã¡reas agrÃcolas , con la introduc ciÃ³n de cultivos y animales domÃ©sticos. El propÃ³sito de este estudio fue observ a r cÃ³m o la abundancia relativa de mamÃ feros locales se ha visto afectada por estos factores antropolÃ³gicos. EstudiÃ© tres sitios en San Luis: una plantaciÃ³n de cafÃ© (CafÃ© Bella Tica), una finca pequeÃ±a con animales dom Ã©sticos ( finca de Marco Mar Ãn y Lorena L e i tÃ³n ), y un bosque remanente . U tilic Ã© transectos , trampas para roedores y trampas cÃ¡ma ras durante 14 dÃ as. EncontrÃ© siete especies de mamÃferos silvestres con 42
Mammal diversity in San Luis Davis 2 avistamientos en los tres microhÃ¡bitats. Cuatro especies de mamÃferos en la plantaciÃ³n de caf Ã© (ardilla Sciurus variegatoides , olingo Bassaricyon gabbii , ratÃ³n espinoso de bolsillo s Heteromys desmarestianus , y rat Ã³n ve nado Peromyscus mexicanus ), tres especies en la finca con animal e s domÃ©sticos ( ardilla , cherenga Dasyprocta punctata , rata de algodÃ³n hÃspida Sigmodon hispidus , y rat Ã³n venado), y tres especies en el bosque ( mono cara blanca Cebus imitator , cherenga , rat Ã³n venado). La finca c on perros y gatos domÃ©sticos presentÃ³ mÃ¡s mamÃferos silvestres arb Ã³reos que terrestres, lo que puede sugerir un comportamiento predatorio s ustancial de parte de las mascotas dom Ã©sticas. S in diferencia significativa en la riqueza de especies en tre microhÃ¡bitats . C onsidero que la presencia humana no estÃ¡ impactando negativamente la diversidad de mamÃferos en San Luis ; podrÃa ser que que el efecto de las plantaciones cercanas ya disminuyÃ³ la diversidad e n el hÃ¡bitat boscoso estudiado. ______________________________________________________________________________ Countryside biogeography, defined as distribution of biological variation over space and time in human dominated is quickly becomi ng more crucial to understa nd ing the increase in human exploitatio n of natural resources (Mendenhall 2013). While island biogeography has previously been used to explain habitat fragmentation, countryside biogeography focuses on anthropological impacts (Mendenhall 2014). In 2014, Mendenhall concluded that consequences, either positive or negative, of human dominated landscapes will ultimately be determined by the of these human created habitats . Vegetation modification , introduction of other food sources, and predatory domestic pets all have the potential to create a significant impact on local mammal populations . Coffee, one of Costa Rica most important agricultural exports, requires alterations to the natural landscape, inevitably impacting local fauna. While shade grown coffee is advocated to increase diversity of volant spec ies such as birds, bats, and insects, one study revealed that those farms are unable to support comparable numbers of non volant mammal species (Caudill 2013). Caudill (2013) explain s that even though coffee is one of Costa major exports, the continuous exploitation of forest land is causing detrimental effects to ecosystem health by reducing mammal richness . This will indirectly impact the quality and longevity of coffee farmi ng. However, other research has shown that many mammal species utilize coffee plantations, such as the Mexican deer mouse, Peromyscus mexicanus. Deer mice feed on coffee beans, hoarding them in caches to eat later (Reid 1997). interested in understanding if the resources introduced by shade grown coffee farms can counter more detrimental effects for wild mammals . Another major disruption humans have introduced to local environments are domestic animals, primarily dogs and cats. Unlike wild predators, domestic animals are not as strongly influenced by the bottom up influence of food scarcity and have the potential t o hunt species to extinction without fear of starvation (Coleman 1997). While wild mammals rely on successful reproduction to maintain their presence in a habitat, domestic pets, specifically cats and dogs , can be introduced to new areas frequently via hum ans. Loss (2013) stated that domestic cats kill between 6.3 22.3 billion mammals every year, earning them a spot on the invasive list. One s tudy revealed dogs to be interference competitors, directly causing the mortality of local foxes (Va nak 2010). Vanak (2010) found that dogs disrupt spatial distribution, as a result of intraguild competition. Considering all local wild mammal species, a previous study in 2012 by Hammoud found that wild mammal species richness was negatively
Mammal diversity in San Luis Davis 3 corre lated with the presence of dogs. However, this study concluded that it was unclear whether this was due to the dog presence or from neighboring human disturbances (Hammoud 2012). By comparing two human dominated habitats, one with a high prevalence of domestic animals and one without, I hope to clarify this distinction. While agriculture and domestic mammals can be destructive to surrounding areas, they also have the potential to introduce new resources for local fauna. Mammal diversity among three microhabitats in relatively close proximity can expand upon the data these previous studies gathered. The purpose of this study is to address the question of what mammals are present around a coffee plantation, a farm containing domestic animals, and in a forested area in San Luis, Costa Rica? How does habitat type affect the presence and diversity of wild mammals ? MATERIALS AND METHODS Study Sites The three sites in this study were within one kilometer of each other in San Luis, Costa Rica. The first site was a coffee plantation, Caf Ã© Bella Tica. The house is surrounded by shade grown coffee plants which are in turn surrounded by some forest fragments. The second site was across the street at Marco and Lorena MarÃn farm, an organic farm with two free ranging domestic cats and three free ranging dogs. The third site was along a forest edge trail, approximately 200 meters from the two farms. The forest this trail bordered is a large plot of undisturbed dense vegetation. I used four different methods to observe mammal presence: trail transects, miscella neous sightings, Sherman rodent traps, and camera traps. To identify the mammal species I found, I used Fiona Reid field guide (Reid 1997). Trail Transects and Miscellaneous Sightings I performed ten t rail transects in each microhabitat, from 14 May 2018 to 25 May 2018. Each transect was approximately 50 meters, surrounded by trees on either side. I walked each transect for 30 minutes, observing all wild mammals from the trail and recording their behavi or while identifying any repeat observations. I walked the transects either in the morning between 5:00AM and 7:00AM or in the evening from 4:00PM to 6:00PM. I completed seven morning transects and three evening transects. I also recorded any other mammal sightings along the sites that were not explicitly seen during trail transects, which were named miscellaneous observations. Sherman Rodent Traps Over five separate nights around 5:00PM, I placed ten Sherman rodent traps along each of the three trail tran sects. The traps were evenly dispersed throughout the 50 meter transect. Each trap was baited with approximately two tablespoons of a vanilla, oat, and dry rice mixture. I marked nearby trees using colored flagging labeled with their trap number to identif y the trap location. The next morning at 7:00AM, I identified any rodents inside the traps using gloves and fabric bags to handle them. I marked each individual to track re catches by re moving a small segment of fur from their back using scissors. The rode nts were then released where they were caught. Camera Traps
Mammal diversity in San Luis Davis 4 To record any other mammals not seen in person, I attached one Bushnell camera trap to trees in each location. The camera traps were placed approximately three feet off the ground and were record ing between 16 May to 25 May 2018. Each camera trap was padlocked to a tree to ensure that they would not be moved . The memory cards were checked periodically, and I recorded both domestic and wild mammal sightings . Data from each method was combined to understand mammal species richness in each microhabitat. I performed two chi squared tests to analyze the significance of domestic and wild mammal distribution across the three microhabitats. RESULTS A total of seven wild mammal species were seen acro ss the three microhabitats. I used trail transects, Sherman rodent traps, and camera traps to observe 42 wild ma mmal sightings overall . When comparing domestic mammals, the three microhabitats differed significantly in the number of sighted domestic mammals , with Marco farm being significantly higher (Fig. 1 2 = 10.73, df = 2, p < 0.001). At Marco MarÃn farm, I observed seven domestic animals using camera traps . One individual cat was observed on the camera trap 12 times (Appendix 1). Thi s was a much higher number of domestic animals compared to the one dog observed passing through the coffee plantation trail . No domestic mammals were sighted in the forest transect (Fig. 1). The camera traps recorded four new domestic dogs and cats on MarÃ farm, in addition to the ones that were known to reside there. Therefore, the with domestic indeed had a significantly higher number of dogs and cats than the other microhabitats. Figure 1: Species richness of wild and domestic mammals in three microhabitats in San Luis, Costa Rica : a coffee plantation, a farm with domestic animals, and a forest habitat . For wild mammals observed, there was no significant difference between species richness among the three microhabitats (Fig. 1). I observed three wild mammal species on both the forest and farm with domestic animals. I observed four on the coffee plantation. However, the type of species differed among the three sites (Fig. 2). The calculated beta diversity for wild mammals resul ted in a value of 2.1 02. Regarding specific mammal presence, I observed one olingo on the
Mammal diversity in San Luis Davis 5 coffee farm. With transects and camera traps, I saw one agouti on the farm with domestic animals and three individual agoutis in the forest. A troop of white faced ca puchin monkeys was observed twice, and once were accompanied by two agoutis eating fruit the monkeys dropped from the trees (Fig. 2). The first monkey troop observed on 21 May consisted of five individuals and the troop observed on 25 May consisted of eight individuals . Regarding smaller rodent presence, I observed variegated squirrels on both farms but not in the forest habitat . I caught spiny pocket mice an d deer mice on the coffee plantation , and of those, two male deer mice were recaptured across several days in the same traps . I observed a hispid cotton rat once on the farm with domestic animals. Referring to number of wild mammal sightings, I observed mo re arboreal than terrestrial mammals on the farm with domestic animals, largely due to the variegated squirrel presence. Figure 2: Wild mammal sightings by mammal type in three microhabitats in San Luis, Costa Rica. The three microhabitats used were a coffee plantation, a farm with domestic animals, and a forest trail. DISCUSSION This study found no significant difference in mammal species richness among three different microhabitats: a coffee plantation, a farm containing domestic animals, and a section of forest. Surprisingly, the forest habitat did not dominate in species richness, with only three species observed. This may either be bec ause the farm sites were habitable enough to sustain the same species richness as the forest, or because the f orest transect utilized was not truly representative of an undisturbed habitat. Even though the difference was not statistically significant, the coffee plantation contained one more observed species than the other sites the olingo. This differs from Cau dill findings (2013) of more mammals in undisturbed locations rather than coffee far ms. Frequent rodent sightings along the coffee farm I studied suggest that this disrupted environment is still suitable for survival. It provides additional food sources such as coffee beans and insects frequenting the coffee plants, both which deer mice and spiny pocket mice prey on. Daily (2003)
Mammal diversity in San Luis Davis 6 found similar results that plantations enhance the conservation value of small forest By enticing smaller ma mmals like mice, coffee plantations may provide desirable habitats for larger animals such as olingos, who are known to feed on deer mice (Wainwright 2002). At tracting small prey, the coffee plantation may be indirectly providing a suitable hunting location for olingos and other larger mammals . Higher levels of species richness throughout these human influenced habitats could reflect the Intermediate Disturbanc e Hypothesis originally suggested by Connell (1978), who argued that only non equilibrium can maintain high er diversity in tropical ecosystems. Even though human created microhabitats like agriculture disrupt natural interactions, they also introduce novel vegetation, food sources, and environmental textures micro niches that can then be filled by any species capable of enduring disruptions. The three habitats I studied in San Luis represent a microscopic view of this theory. If these anthropologically con structed habitats are still able to maintain functional communities , then they may prove to not be entirely harmful additions. Also pertinent is that both farms were within 200 meters of a larger forest habitat. While corridor effects and mammal home range distance were not explicitly measured in this study, they have the potential to impact habitat choice . Therefore, it is possible that some wild mammals may reside in more isolated locations while choosing to hunt or forage closer to humans. The farms were nonetheless able to attract small animals such deer mice, spiny pocket mice, hispid cotton rats, variegated squirrels, and larger mammals like agoutis and an olingo (Fig. 2). Another variable heavily prevalent throughout this study was the presence of d omestic mammals. Seven domestic mammals were spotted walking the transect of Marco farm . Wild mammals were also spotted throughout the site, including an agouti, a hispid cotton rat, and variegated squirrels (Fig. 2) . Compared to the nine deer mice sightings on the coffee farm and the one sighting in the forest, there were surprisingly zero deer mice caught on the farm with more domestic animals . The frequent appearance of dogs and cats and few mice observations among this transect suggests predatio n on local mammal and rodent populations. There were no large wild mammal predators spotted on this farm, as compared to the olingo and capuchin monkeys identi fied in the other two microhabitats. These wild predators were possibly replaced by domestic animals. This is consistent with study in 2012 that found a negative correlation between wild mammal species richness and domestic dogs after observing frequent predatorial chases . Another study in Zimbabwe discovered that domestic dogs were the most common predator within the observed sites (Butler 2004). This monopol ization by domestic animals may lower the numbers of smaller terrestrial prey, such as agoutis and mice , while allowing arboreal and less obtainable prey to thrive, such as variegated squirrel s (Fig. 2 ). The true impact of domestic animals, regarding the distance they are willing to traverse in order to hunt, requires further research to fully understand how they influence prey levels. Edge effects are another possible expla nation for the similar mammal species richness found among the three microhabitats. The forest may not have had the most wild mammal species because of negative effects from the nearby farm s. The forest transect I used followed a trail along the forest edg e. Similar studies on habitat gradients found decreased abundance in animals such as deer mice along forest edges than deeper into the forest (Menzel 1999). This may be due to a combinat ion of factors , including vegetation changes, air temperature, and lig ht intensity differing between the forest edge and interior. In turn, this may have negatively affected how many wild mammals preferred the forest edge. On 22 May, I noticed a pile of motmot feathers along the forest transect trail. While the origin of these feathers cannot be confirmed,
Mammal diversity in San Luis Davis 7 their presence suggests predatory behavior, possibly from a larger mammal present. Therefore, more mammals than th e ones observed could have been using this forest edge. However, t he lack of sightings throughout may indicate that they do not frequent ly utilize the e dge, possibly as a result of edge effects and negative impacts from neighboring farmland. These human do minated landscapes may be able to reach and indirectly impact even forest habitats. This study documented non flying wild and domestic mammal presence among three different microhabitats in San Luis. With no significant difference in wild ma mmal species richness among locations, these findings suggest that while microhabitat does have an effect on what species are present, the number of species is less variable. A possible implication suggests that higher habitat variation in small areas, inc luding forest, agriculture, and domestic mammal presence, increases overall species diversity . As long as these habitats maintain sufficient food and shelter, human and predator presence may increase species richness , allowing different species to take adv antage of new niches . Moderate disturbances prevent one species from dominating, allowing multiple species of different trophic levels to thrive. The longevity of this cohabitation requires future studies, while currently suggesting sustainability in habit at variability. The second approach is that the true effects of human dominated landscapes may st retch past their physical boundaries . Even though humans and their domestic animals may not be frequenting the forest habitat I studied, their impact still has th e potential to negatively impact forest dwelling mammals by encroaching upon forest edges . While the true implications of human dominated landscapes continue t o be discovered , our true ecological impact necessitates reevaluation. ACKNOWLEDGEMENTS I w ould like to thank everyone who supported this project, either by opening up their la nd to me or by providing scientific guidance. Special thanks to my teacher, mentor, and friend, Federico Chinchilla, who taught me so much every day . Thank you for patiently and happily accomp anied me on site visits, even when it included climbing a hill every morning at 7:00AM to catch rodents . Also, thank you to my wonderful secondary advisor Sofia Flores for introducing me to beautiful San Luis. Thank you to all o f the amazing students on this Costa Rica journey with me, including Madi f or assisting with the editing process of this paper. A sincere thanks to Marcos and Lorena MarÃn , and Oldemar and Ersi Salazar for graciously allowing me to use their farms as resea rch sites. Finally, thank you to Bobby, my loyal field assistant who stuck with me throughout my San Luis adventure. LITERATURE CITED Butler, J. R. A. and du Toit, J. T. & Bingham, J. 2004. Free ranging domestic dogs ( Canis familiaris ) as predators and prey in rural Zimbabwe: threats of competition and disease to large wild carnivores. Biological Conservation 115, 369 378. Caudill, S. A. 2013. Assessment of mammal diversity in coffee dominated landscapes of India and Costa Rica. Un iversity of Rhode Island Digital Commons. Coleman, J. S. and Temple, S. A. & Craven, S. R. 1997. Cats and wildlife a conservation dilemma. United States Fish and Wildlife Service.
Mammal diversity in San Luis Davis 8 Connell, J. H. 1978. Diversity in tropical rain forests and coral reefs. Science, 199(4335), 1302 1310. Daily, G. G. Ceballos . Pacheco, J. SuzÃ¡n G. and SÃ¡nchez Azofeifa, A. 2003. Countryside biogeography of neotropical mammals: conservation opportunities in agricultural landscapes of Costa Rica. Conservation Biology, 17(6). Hammoud, K. 2012. The effect of free roaming Canis lupus familiaris on wild mammal species richness and abundance in Monteverde, Costa Rica. EAP Tropical Biology and Conservation . Loss , S. R. and Will, T. & Marra, P. P. 2013. The impacts of free ranging domestic cats on wildlife of the United States. Nature Communications. Mendenhall, C. D. Kappel, C. V. and Ehrlich, P. R. 2013. Countryside biogeography. Encyclopedia of Biodiversity , 347 360. Mendenhall, C. D. Karp, D. S. Meyer, C. F. J. Hadly, E. A. and Daily, G. C. 2014. Predicting biodiversity change and averting collapse in agricultural landscapes. Nature. 213 217. Menzel, M. A. Ford, W. M. Laerm, J. and Krishon, Dian e. 1999. Forest to wildlife opening: habitat gradient analysis among small mammals in the southern Appalachians. Forest Ecology and Management, 114: 227 232. Reid, F. A. 1997. A field guide to the mammals of Central America and Southeast Mexico. Oxfo rd University Press. Vanak, A. T. and Gompper, M. E. 2010. Interference competition at the landscape level: the effect of free ranging dogs on a native mesocarnivore. Journal of Applied Ecology, 47(6), 1225 1232. Wainwright, M. 2002. The natural history of Costa Rican mammals. Zona Tropical.
Mammal diversity in San Luis Davis 9 Appendices Appendix 1: Camera Trap Data for Marco farm, the farm with domestic animals between 16 May to 25 May. Observations are separated by microhabitat type. Coffee Plantation Farm with Domestic Animals Forest Day 4 16 May 2018 All camera traps set 2:48AM unknown cat with cropped tail walking left 8:38PM unknown gray and white cat walking left Day 5 17 May 2018 9:15AM Dixie (domestic cat) walking left Day 6 18 May 2018 7:54AM farmer with domestic poodle 8:37AM German Shepherd dog walking left 8:40AM Dixie walking left 5:10PM black dog) walking left 6:54PM unknown cat with long tail walking left 9:50PM little black domestic cat walking right Day 7 19 May 2018 Checked camera traps for 1st time 1:23AM Dixie walking right 1:39AM unknown cat with long tail walking left 9:51AM Dixie walking left 10:14AM Dixie walking right 10:59AM Dixie walking right 11:15AM Dixie walking left 1:31PM Dixie walking left 6:51PM unknown animal, with long tail walking right No Camera Trap Data from previous days, technical difficulty, fixed camera 4:30PM agouti walking right (into forest trail) Day 8 20 May 2018 2:46AM unknown gray and white cat walking left 7:57AM Dixie walking right 8:23AM agouti walking left Day 9 21 May 2018 Checked camera traps for 2nd time 10:12AM farmer with domestic poodle Day 10 22 May 2018 8:46AM agouti walking out of forest onto path
Mammal diversity in San Luis Davis 10 Day 11 23 May 2018 11:52PM unknown animal with long tail walking right 8:03AM agouti 8:20AM agouti 8:58AM agouti 9:00AM agouti 9:21AM agouti 9:29AM agouti 1:41PM agouti Day 12 24 May 2018 10:18AM Dixie walking right 10:36AM agouti walking right 11:22AM Dixie walking left 7:35AM agouti 1:08PM agouti Day 13 25 May 2018 Checked and removed camera traps Appendix 2 : Camera trap data from Marco farm with domestic anim als. The data is separate d by number of domestic and wild mammal sightings between 16 May and 24 May.