My project proposal, just in case you're interested

Livin’ On the Edge:
Variance in Mammalian Species Richness and Abundance
Between Eucalyptus and Pasture Edges of the Atlantic Forest

Introduction

The Atlantic Forest is one of the most unique environments on the planet—providing hundreds of priceless resources for human beings around the world—but after decades of degradation, pollution, and exploitation, only 7-8% of the original Mata Atlântica ecosystem remains (Galindo-Leal and Câmara 2003). Unfortunately, this small percentage is still threatened by unsustainable human practices and a lack of adequate preservation efforts. One of the most imminent issues jeopardizing the survival of these existing forest remnants is fragmentation, which isolates populations, curtails biodiversity, and exacerbates extinction risks (Chiarello 1999). A vital step in understanding and eliminating the consequences of fragmentation is determining species diversity at the forest edges, and measuring how this diversity varies across different edge types, in order to improve conservation techniques, develop more efficient management strategies, and advance policies regarding forest preservation (Marsden et al. 2001).

In my project, “Variance in mammalian species richness and abundance between eucalyptus and pasture edges of the Atlantic Forest,” I will attempt to evaluate both species richness (number of species) and abundance (number of individuals) at eucalyptus and pasture edges of the Atlantic Forest using baited sand traps to collect mammalian tracks. The results of this project will also be applicable when assessing the Green Hug concept, an important measure being taken to diversify forest edges. Eucalyptus forests are planted at the edge of the Atlantic Forest, in order to create a buffer between the forest and surrounding pastures and human inhabited areas. Theoretically, the presence of this buffer should indicate elevated species presence at this edge because the eucalyptus forest “hugs” the Atlantic Forest and mitigates the division between the two environs (Ferinmore and Cullen 2002).

The neighboring habitats that surround the forest fragment edges—such as eucalyptus stands and pastures—are known as matrixes (Gascon et al. 2000). These buffers are critical for the survival of the forest fragments that they converge with; the harshness of a matrix (how closely the matrix resembles the ecosystem it encircles) can determine whether or not it provides crucial protection for the forest interior because it limits the harmful outside factors—such as pollution—that infiltrate the forest (Umetsu and Pardini 2007). In my project, the forest edges and their corresponding matrixes are the areas that I’m concerned with—through my sand trap experiment, I will attempt to determine the difference between species diversity of edges adjacent to eucalyptus and pasture matrixes.

Question and Hypotheses

In this project, the main question I will be asking is: How does the richness and abundance of mammalian species at Atlantic Forest edges vary according to the presence of eucalyptus and pasture matrixes? Based on research regarding this topic and my own prior knowledge, I have formulated a hypothesis in answer to this question: There will be a difference in species diversity between eucalyptus and pasture edges; there will be more mammalian species richness and abundance present at eucalyptus edges than at pasture edges (Ha).

H0: Peucalyptus edge = Ppasture edge
Ha: Peucalyptus edge ≠ Ppasture edge

Methods

In order to obtain a more accurate measure of species richness and abundance at forest edges in the Mata Atlântica ecosystem, I have chosen three different sites in the Nazaré Paulista region of Brazil where I will lay my sand traps. Each of these selected sites is a forest fragment that has both a eucalyptus edge and a pasture edge, in an attempt to increase the statistical relevance of my results. At each site, I will lay six 50x50cm sand traps (three replicates at each edge) at randomly selected points along a 50m line, for a total of 18 traps. These traps will be baited with bacon and bananas, in order to attract mammals. In a study of Atlantic Forest matrix quality, Umetsu and Pardini (2007) analyze small mammal populations because they are the most diverse genre of mammals present in Neo-tropical regions. If my results are similar to those of previous sand trap projects completed by SEE-U students at IPÊ, most of the wild animal tracks in the traps will belong to small mammal species (Peterson 2006 and Mintz 2007).

After setting and baiting the traps, I will return early the following morning (to avoid disturbances and loss of data) to record the tracks; I will photograph each track and include a scale in each photo, in order to gauge size and aid in identification. In both previous SEE-U sand trap projects, students checked their traps 6-7 times. Neither of these students was able to disprove their null hypothesis (Peterson 2006 and Mintz 2007). I will check my traps a total of ten times; I would like to collect more information and possibly determine if these students were unable to disprove their null hypotheses because they did not gather sufficient data.

When analyzing the results of my experiment, I will use a two-tailed t-test to determine the difference in species diversity between the two matrix types. By including multiple replicates, selecting random trap sites, and checking the traps additional times, I hope to collect statistically viable data that will prove my hypothesis. In previous SEE-U sand trap experiments, students faced two main obstacles when laying their traps: steep inclines and a lack of forest clearings (Peterson 2006 and Mintz 2007). In order to avoid the first issue, I have chosen sites with as little incline as possible, although some was unavoidable. Regarding the latter issue, I may have to adapt the random selection of my trap locations if I am unable to place them in relatively clear areas. Another complication I anticipate is inclement weather, which may disrupt my sand traps (so I will check the traps as early in the day as possible).

A notable limiting factor of my project is the proximity of my sites; I cannot measure the general species diversity of the Atlantic Forest because all of my sites are located in the Bairro do Moinho, but I have chosen three different sites instead of using pseudo-replicates, which will hopefully expand my data and increase their pertinence regarding this specific location.
 
References Cited

Chiarello, A. G. 1999. Effects of fragmentation of the Atlantic Forest on mammal communities in south-eastern Brazil. Biological Conservation 89 (1): 71-82.

Ferinmore, S. C. and L. Cullen Jr. 2002. Projecto Abraco verde: A practice-based approach to Brazilian Atlantic Forest conservation. (cases). Endangered Species Update 1-10.

Galindo-Leal, C. and I. G. Câmara. 2003. Atlantic Forest hotspot status: An overview. The Atlantic Forest of South America: Biodiversity Status, Threats, and Outlook 3-11.

Gascon, C., G. B. Williamson, and G. A. B. da Fonseca. 2000. Receding forest edges and vanishing reserves. Science 288 (5470): 1356-1358.

Marsden, S. J., Whiffin, M., and M. Galetti. 2001. Bird diversity and abundance in forest fragments and Eucalyptus plantations around an Atlantic Forest reserve, Brazil. Biodiversity and Conservation 10 (5): 737-751.

Mintz, E. 2007. Do small forest fragments lack mammalian diversity? SEE-U Brazil. Session 1. PowerPoint.

Peterson, J. 2006. Are eucalyptus forests devoid of mammals? SEE-U Brazil. Session 1. PowerPoint.

Umetsu, F. and Pardini, R. Small mammals in a mosaic of forest remnants and anthropogenic habitats—evaluating matrix quality in an Atlantic Forest landscape. Landscape Ecol 22: 517-530.