A Study of Bat Diversity in Campanario S Primary Forest

A Study of Bat Diversity in Campanario’s Primary Forest

Cameron Incognito

Kyle Hovey

Julianne Pekny

AnnieKrichten

TramondBaisden

Introduction

Costa Rica is home to over eleven percent of bat species in the world, with 108 species observed thus far. Furthermore, over half of the mammalian species in Costa Rica are bats. Hollow trees, their bark, buttress roots, and some leaves are common roosting places for bats in Costa Rica due to the relative scarcity of buildings and caves. While roosting, multiple species of bats may be observed, but when foraging it is most common for single-species flocks.

Bats consume a wide variety of food sources, but individual species are classified into different guilds including frugivores, nectarivores, insectivores, carnivores, and omnivores.Frugivores, whose diets consist of mainly fruits, may supplement such protein-poor diets with insects when necessary. Captured examples in the present study include Carolliaperspicillata and Carolliacastanea. They are frequently observed in both primary and secondary forests since many fruiting plants (e.g. Piper and Cecropia) grow there. Nectarivoresconsume nectar from flowers, thereby promoting the pollination of a number of flower species, and obtain protein from pollen. In the wet season when nectar becomes scarce, nectarivores consume insects and fruit. Glossophagasoricina, a nectar-eating bat, was captured this year. Insectivores, like thePteronotusgymnonotus captured this year, feed on insects. Pteronotus is observed to forage in open areas, for example over roads and near beaches, for insects. They are commonly known to roost in caves, but still benefit from visiting fruit trees where many insects may be found (LaVal K, Rodriguez-H B, 2002). Omnivores have a diverse diet, including fruits, insects, nectar, and small animals. Phyllostomus discolor was found this year and is essential as a pollinator as shown in a Guanacaste study in which eighty-two percent of captured bats carried pollen (LaVal K, Rodriguez-H B, 2002). Carnivores consume small animals, and the first carnivore in this study, Lophostomasilvicola, was caught this year. Lophostomasilvicola’s diet consists mostly of frogs and lizards, though it may consume small amounts of fruit if necessary. It forages at levels of low vegetationexclusively in primary forests and roosts in hollow trees and termite nests. This species is an indicator of a high-quality, healthy environment because it relies on primary forests (Mora JM, 2013).

This study aimed to sample bat diversity in the primary forest located at Campanario Biological Station in an effort to determine relationships between the presence of bats and the density of plants and fruits. Comparisons between 2013 primary forest data and previous years’ data in both the primary and secondary forests were used to determine any correlations between bat populations in different habitats. Lastly, the bat richness, abundance, and spread were compared to the data obtained from a similar 2013 primary forest bird study in an effort to determine any correlations between bird and bat feeding and living conditions.

It was hypothesized that a higher abundance and richness of bats would be found in close proximity to their respective food sources. In an area of highplant and fruit density that may attract or serve as bat food sources, a higher number of bat captures would be expected. It was also expected that there would be similar trends between bird and bat captures; in areas of high bird captures, it was expected that there would be a high amount of bat captures. Finally, due to the fact that bats’ peak feeding periods are at dusk and dawn, it was expected that bat captures would be highest just after dusk (LaVal RK, Rodriguez-H B, 2002.).

Methods

Nine mist nets were assembled along the Bauhinia trail in the primary forest of Campanario Biological Station to capture both birds and bats. The first net was positioned closest to the field station at 3:20 pm, with net nine positioned at the furthest end of the trail at 5:00 pm. The mist nets, consisting of thin mesh netting divided into four panels with a pocket in between each layer, ranged from ten to fifteen meters long and were hung on 3 meter poles of various material. The poles were anchored to nearby trees in order to provide stability and to keep the net tense enough to capture bats and birds, but prevent bat ricochets, potentially freeing them. Once assembled, the nets were patrolled periodically(approximately every 30 minutes) for captures. In the event of a capture, the bat was untangled and classified according to time, net number, species, and gender. The length of the tibia and forearm were recorded, as well as age (adult or infant) and the presence of parasites for species identification purposes. Fecal samples were also collected from captured bats if possible to identify any seeds and plants that they previously fed upon. The captured bats were then released back into the forest. Bat captures took place between 6:00 pm and 9:00 pm.

Plant density and fruit density surrounding each net were also measured. At each net, plant density was measured at six locations by finding the number of leaves, branches, and stems that touched a 2 meter long bamboo pole at a distance of two meters on both sides of the net. This was carried out at both ends of the net and also in the center. Later, plants were classified by family. Fruit density was measured by counting, collecting, and classifying all berries within three meters of each side of each net. The total berry count was divided by the area surveyed to yield the fruit density for each net. All statistical analyses were carried out using Microsoft Excel and the JMP program.

Results

Table 1: Detailed measurements and calculations from bat capture at Campanario primary forest. 2013

Bat Number

Genus and Species

Gender

Adult (y/n)

Tibia Length (mm)

Forearm length (mm)

1 GlossophagaSoricina

Female

Y

n/a

Carolliaperspicillata

n/a

Unknown

n/a

Glossophagasoricina

n/a

Unknown

n/a

Carolliaperspicillata

Female

Pteronotusgymnonotus

Female

n/a

Phyllostmus discolor

Female

n/a

Carolliaperspicillata

Female

Glossophagasoricina

Female

n/a

Carolliacasteanea

Female

Carolliacastenea

Female

Lophostomasilvicola

n/a

Unknown

n/a

Figure 1: The number of individuals of each bat species captured at Campanario primary forest (2013) and the secondary forest (2005).

Table 2: The number of bat individuals captured at Campanario primary forest, their respective dietary guilds, and their capture location. 2013

Net number

Total number of Individuals

Number of Insectivores

Number of Frugivores

Number of Nectarivores

Number of Omnivores

Number of Carnivores

Figure 2: Comparison between the number of captures per net between bats and birds in Campanario primary forest. 2013

Figure 3: The number of bats captured in select time intervals in Campanario primary forest. 2013

Figure 4: The number of bat individuals captured as function of plant density in Campanario primary forest. 2013

Figure 5: The number of bat individuals captured as a function of fruit density in Campanario primary forest. 2013

Figure 6: The richness of bats as a function of plant richness in Campanario primary forest. 2013

Figure 7: A comparison of the number of bat individuals captured in both the Campanario primary and secondary forests. 2013 and 2005

Discussion

As seen in Figure 1, there was very little overlap between species found in Campanario’s primary and secondary forests. There was also a higher bat richness (total bat species captured) found in primary forests compared to secondary forests, a likely result of primary forests’ capability of providing more cover for the bats from predators and their higher abundanceof resources due to maturity. Similar capture success rates were also observed in the primary forests (0.055) and in secondary forests (0.07). While more nets were used in the primary forest than the secondary forest, the capture success rates were similar, suggesting that perhaps extra nets and the extra time are not necessary to have the same capture success.

Figure 2 shows that the captured bat individuals were located in the earlier nets (1-4) while the birds captured by peers ranged across all nets except 4.To determine why this was the case, the plant and fruit density surrounding each net was compared to the number of bats and birds captured to look for possible correlations and relationships.

While a direct relationship was found between bird richness and plant richness by peers, this relationship was not observed with bats. As seen in Figure 6, there is no direct correlation between these two variables as exemplified by the insignificant P-value (0.3950) and low R2 value (0.105).Plant diversity and richness surrounding the nets included five different genera of Melastomata and four different genera of Psychotria. NeitherPipernor Cecropia, both of which are pioneer species known to be consumed by some species of bats caught in this study, were observed in close proximity to the nets.

Alternatively, the number of birds captured was inversely proportional to the plant density surrounding the nets suggesting that birds preferred the more open areas. As seen in Figure 4, there is no correlation between the number of bat individuals and the surrounding plant density. The large P-value (0.8907) indicates that the relationship is not significant and the small R2 value (0.00289) indicates that there is no linear relationship present. Therefore no conclusions can be made about the habitat or feeding area preferences of bats in the primary forest.

Thirdly, the number of birds captured was positively related to the fruit density surrounding the nets suggesting that the areas of more fruit are more populated by birds. This is to be expected as the more fruit available, the more birds will be able to be sustained. As exemplified in Figure 5, no correlations were able to be determined between the number of bats captured and the surrounding fruit density. The plot possessed a large P-value (0.4206) and a very small R2 value (0.0919) showing that the points were insignificant and that a linear regression was not appropriate.

In summary, a higher abundance and richness of bats was not observed in proximity to the food sources near nets. Additionally, locations with higher fruit and plant densities did not indicate higher bat abundance, contradicting several of the original hypotheses.

This data also disproves the hypothesis that birds and bats would possess similar trends regarding captures and factors such as net location (Figure 2), abundance, richness (Figure 6), plant density (Figure 4),and fruit density (Figure 5). Several species of bats, using echolocation, are capable of avoiding mist nets easier than birds, which may have been a factor in the lack of trends.

As exemplified by Figure 7, the highest bat capture rate was after dusk for both primary and secondary forests.This is most likely observed because bats are beginning to leave their roosting sites to forage. Foraging is also important to consider when thinking about why some of our hypotheses were disproved, why there were so many differences between the bird and bat capture data, and why there were so few correlations between bat capture and the factors analyzed. While bats are known to roost together, it hasbeen observed that they do not forage together (Wainwright, M. 2002). This is not the case for birds, which are known to forage in flocks of multiple species ultimately providing protection during feeding. This may explain why there was a higher richness of birds captured in comparison to bats.

Bats are known to travel several kilometers per night in order to find their respective food sources. This may indicate that the bats captured in this study may have been captured en route to their foraging sites. Therefore the bats captured may not have been near the nets for the resources around, unlike the birds. This rationale is supported in this study as some of the collected bat feces were found to contain Piper seeds, a common food source forCorollia.Piper was not found near our net sites indicating that the particular bat had been traveling while foraging for food. In addition, the mist nets were originally placed in areas containing plants known for bird foraging, not bat foraging, such as Melastomata and Psychotria.

Admittedly, there are some potential sources of error present in our study. First, fruit may have been located near the nets but not within our sampling distance of three meters. Thus, abundance of fruit may have been underrepresented for some nets. Additionally, bats may have been caught near their roosts or at another point of a potentially long journey towards food sources other than those surrounding the net, such as Piper or Cecropia. It is possible for captured animals to escape from mist nets and, while incredibly unlikely, individuals can be captured more than once.Lastly, due to collaboration with a group studying birds, mist nets were set up in spots likely for birds to flock (e.g. those near Psychotria andMelastomatafruit sources) rather than bats, which may account for the relatively low number of individual bats captured.

Additional experiments, which would likely involve a significant amount of addition labor and cost, would be able to remove many of the sources of error and biases found in this experiment. Bat detectors, which have been show to record over 30% more captures than alternative means such as mist nets, can be utilized to provide a more accurate sample for measuring bat diversity and abundance. The time intervals for capture can also be increased to possibly provide a larger sample size and a more accurate comparison to the bird study conducted by peers. The larger sample size obtained would make it possible to examine the relationship between frugivoreand nectarivorebats and the fruit density surrounding the nets where they were captured. This was carried out for this study; however there were only four data points available to produce a plot of the number of frugivore bats captured versus fruit density. The plot contained an insignificant P-value (>o.o5) and a relatively low R2 value (0.66262). If there was a larger bat sampling size, which may become available after several years worth of data, a trend may become evident. Lastly, it would be interesting to study richness and abundance of bats in different seasonal and geographical conditions as they may differ greatly.

Conclusion

In this study a higher abundance and richness of bats was not observed in proximity to the food sources near nets. A higher plant density and higher fruit density also did not correspond to a higher bat capture number as predicted in the original hypothesis. No parallel trends were observed between birds and bats at the sampling locations as predicted. However, the frequency of bat captures was found to be highest just after dark as originally predicted. Lastly, the primary and secondary forests were found to contain different species of bats, with the primary forest being more diverse.

References

Carlo, T. Pers. Comm. 2013.

LaVal RK, Rodríguez-H B. 2002.Murciélagos de Costa Rica: BATS. 1 ed. Santo Domingo de Heredia, Costa Rica: InstitutoNacional de Biodiversidad, INBio, 2002.

Mora, JM. Pers. Comm. 2013.

Wainwright, M. 2002. The Natural History of Costa Rican Mammals. Ed: David Featherstone. DistribuidoresZona Tropical, S.A. Miami, FL. 2002.