midterm.tex

\documentclass{informalLabReport}
\urlstyle{same}

\usepackage{listings} 
\let\verbatim\undefined 
\let\verbatimend\undefined 
\lstnewenvironment{verbatim}{\lstset{breaklines,basicstyle=\ttfamily}}{}
\usepackage{booktabs}
\usepackage{graphicx}
\usepackage{multicol}
\usepackage{blindtext}
\usepackage{siunitx}
\usepackage{tabularx} 
\usepackage{float}
\usepackage{biblatex}
\usepackage{rotating}
\usepackage{longtable}   
\usepackage{wrapfig}
\addbibresource{midterm_csc2621.bib}
% \definecolor{lightblue}{rgb}{0.93,0.95,1.0}
\title{Lab 6: Statistical Exploratory Data Analysis}

\author{Leigh Goetsch}
\prof{Kedziora}
\className{Data Science}
\classCode{CSC 2621}
\semester{Spring Semester 2024}
\submissionDate{02/26/2024}

% %For Informal Reports
\labWeek{6}
\laboratoryDate{02/20/2024}

% \lfoot{\footnotesize{Lab 3}}

\begin{document}
\maketitle

\section*{Learning Outcomes}
\begin{itemize}
    \item Demonstrate knowledge of techniques to clean and explore data to perform an analysis of a provided data set
\end{itemize}

\newpage
\tableofcontents

\newpage
\begin{multicols*}{2}

% In a 2015 paper, Dr. Reid reported on an effort to determine if it is possible to distinguish between bobcat, coyote, and fox scats from morphological (e.g., length, volume, etc.), biogeochemical (e.g., chemical composition), and contextual traits (e.g., location). 
% Morphological traits are the least costly properties to evaluate and can be done in the field, while biogeochemical traits may require expensive equipment and need to be analyzed in a lab. 

% Your job is to apply your data science knowledge to analyze the data and answer the research questions posed by Dr. Reid. 
% You will also have to do some background research on the three species to interpret and explain the differences.

% Using this data set, you need to answer two research questions:
% 1. Which (if any) morphological and biogeochemical traits distinguish between originating species of the scat samples?
% 2. Why do you think those traits differ across species?

\section{Introduction and Background}
Understanding the population patterns of wildlife species is important for effective conservation and wildlife management. One set of  species of interest are the coyote (Canis latrans), the bobcat (Lynx rufus), and the grey fox (Urocyon cinereoargenteus). 

\addcontentsline{toc}{subsection}{Coyote}
The coyote is a medium-sized wild canid. It has a canine-like appearance and they live in a diverse range of habitats, being found between 10 and 70 degrees north latitude.\cite{Bekoff1977} While coyotes are often blamed for targeting livestock and deer, "there is little evidence that coyote predation is a primary limiting factor on populations of big game or domestic livestock." \parencite{Bekoff1977} Coyotes are opportunistic predators, so they have a wide variety of food, but it is primarily meat.\cite{Bekoff1977}

\addcontentsline{toc}{subsection}{Bobcat}
The bobcat is a medium-sized wild cat. It has a distinct appearance with tufted ears, a spotted coat, and bobbed tail. They live in most of the United States and Canada, but not Hawaii or Alaska, and is gone from much of the Ohio Valley, Mississippi Valley, and southern great Lakes region.\cite{Lariviere1997} The primary food source of bobcats is lagomorphs, which are wild rabbits and hares, and they are purely carnivorous.\cite{Lariviere1997}

\addcontentsline{toc}{subsection}{Grey Fox}
The grey fox is a small to medium canid. It has a grey coat with black tips and a black-tipped tail. They live in wooded, brushy, and rocky land, with their range going from southern Canada to Mexico and Central America.\cite{Fritzell1982} Grey foxes are omnivores, with their primary prey being us rabbits and rodents, and their diet changing depending on the season.\cite{Fritzell1982}

All three species are mammalian mesopredators that are distributed throughout North America. While none of these species are currently listed as endangered, their ranges overlap with human development. This creates threats of habitat loss, fragmentation, and human-wildlife conflict. 

While this is not a direct threat to human society, it does create conflict and without proper wildlife management, can cause issues. Understanding the population dynamics of these species provides insights into conservation efforts, allowing conclusions to be made about the efficacy of programs and wildlife management.

\addcontentsline{toc}{subsection}{Population Estimation Methods}
There are a couple different field techniques that can be used for tracking population of medium mammals. These techniques can fit into three categories: observational, capture, and marking.\cite{McComb2018}

Marking is attaching something identifying to the animal, such as pigment, a band or collar, or mutilating the animal in some way, such as nicking an ear. This is mainly for birds and small mammals, though. \\
Capture can be box traps. These take time to check and can potentially endanger the animal. \\
The last category is observational. This type has many methods such as tracking signs using foot track surveys or scat collection. There is also remote tracking using track plates, photo and video stations, or hair traps. While foot track surveys are low cost, they take time and depend on the seasons. Additionally, many remote tracking options cost money to implement and time to setup.

\addcontentsline{toc}{subsection}{Scat Sampling}
Scat samples are very common, easy to spot, and cause little disturbance to collect.\cite{Reid2015} This makes them a good choice for estimating population size. The set of speices that we are using are also good candidates for this analysis because they are of similar size, and their ranges overlap quite a bit, so their scat is likely going to be collected near each other, meaning that being able to identify which species the sample belongs to becomes an important observation.

% Using the given research questions, clearly state your hypotheses.
% 1. Which (if any) morphological and biogeochemical traits distinguish between originating species of the scat samples?
% 2. Why do you think those traits differ across species?
Using the dataset provided by Dr. Rachel Reid used in their article \textit{A morphometric modeling approach to distinguishing among
bobcat, coyote and gray fox scats}, I aim to analyze the relationships within the set to figure out which morphological and biogeochemical traits distinguish between originating species of the scat samples. I also am going to speculate why those traits differ across species.

The information contained within Dr. Reid's dataset are: Species, Month, Year, Site, Location, Age, Number, Length, Diameter, Taper, TI, Mass, d13C, d15N, CN, Ropey, Segmented, Flat, and Scrape.
My hypothesis is that the species' diets will affect the makeup of their sample, so the C:N variable will be a distinguishing variable. I also think that the bobcat will be more different than the coyote and grey fox because its diet is purely carnivorous when assessing the C:N ratio. I also hold the hypothesis that the mass variable will mke the grey fox distinct because the species is on average smaller than the other two, so their waste will be smaller.

\vfill
\section{Analysis}

\addcontentsline{toc}{subsection}{Load Dataset}
First I loaded the dataset. The set looked like before cleaning and conversions is here: Table \ref{tab:initial}. Here is the df.describe() results to show the sort of variables and data we have:

\begin{table}[H]
\centering
\small
\caption{Initial Dataset}
\begin{tabular}{lcc}
    \toprule
    \textbf{Column} & \textbf{Non-Null Count} & \textbf{Dtype} \\
    \midrule
    Species & 110 & object \\
    Month & 110 & object \\
    Year & 110 & int64 \\
    Site & 110 & object \\
    Location & 110 & object \\
    Age & 110 & int64 \\
    Number & 110 & int64 \\
    Length & 110 & float64 \\
    Diameter & 104 & float64 \\
    Taper & 93 & float64 \\
    TI & 93 & float64 \\
    Mass & 109 & float64 \\
    d13C & 108 & float64 \\
    d15N & 108 & float64 \\
    CN & 108 & float64 \\
    Ropey & 110 & int64 \\
    Segmented & 110 & int64 \\
    Flat & 110 & int64 \\
    Scrape & 110 & int64 \\
    \bottomrule
    \end{tabular}%
\end{table}

Then I assessed the data types of the variables. Here is a summary of the variables:\cite{Reid2015}

\begin{itemize}
    \item \textbf{\textit{Species:}} Categorical variable describing species. Target variable.\\
    \textit{Data Type:} Nominal.
    \item \textbf{\textit{Month:}} Month sample was collected.\\
    \textit{Data Type:} Ordinal.
    \item \textbf{\textit{Year:}} Year sample was collected(2011-13).\\
    \textit{Data Type:} Ordinal.
    \item \textbf{\textit{Site:}} Categorical variable for site the sample was collected from - Año Nuevo State Park and Reserve(ANNU) or Younger Lagoon Natural Reserve and Moore Creek Preserve(YOLA).\\
    \textit{Data Type:} Nominal.
    \item \textbf{\textit{Location:}} Categorical variable (3-point scale) describing scat location on the trail/road – middle, edge, or off edge.\\
    \textit{Data Type:} Ordinal.
    \item \textbf{\textit{Age:}} Integer between 1 and 5. I'm not sure what this measures the age of.\\
    \textit{Data Type:} Discrete.
    \item \textbf{\textit{Number:}} Integer number of separate scat pieces.\\
    \textit{Data Type:} Discrete.
    \item \textbf{\textit{Length:}} Length (cm) of longest piece to the nearest 0.5 cm.\\
    \textit{Data Type:} Ratio.
    \item \textbf{\textit{Diameter:}} Scat Diameter (mm) measurement at widest point to the nearest 10th of a millimeter.\\
    \textit{Data Type:} Ratio.    
    \item \textbf{\textit{Taper:}} Length (mm) of longest taper down the axis of the scat.\\
    \textit{Data Type:} Ratio.
    \item \textbf{\textit{TI / Taper\_deg:}} Degree of Taper. Unitless ratio of taper length to scat diameter.\\
    \textit{Data Type:} Ratio.
    \item \textbf{\textit{Mass:}} Total dry weight (grams) after freeze drying and baking.\\
    \textit{Data Type:} Ratio.
    \item \textbf{\textit{d13C / Carbon13:}} Carbon isotope analysis measurement.\\
    \textit{Data Type:} Ratio.
    \item \textbf{\textit{d15N / Nitrogen15:}} Nitrogen isotope analysis measurement.\\
    \textit{Data Type:} Ratio.
    \item \textbf{\textit{CN:}} C:N Ratio. Unitless ratio of carbon to nitrogen atoms in the scat, which is a proxy for the degree of carnivory of the animal.\\
    \textit{Data Type:} Ratio.    
    \item \textbf{\textit{Ropey:}} Does the scat appear ropey/twisted/woven? (1 - yes, 0 - no)\\
    \textit{Data Type:} Nominal.
    \item \textbf{\textit{Segmented:}} Does the scat show segmentation? (1 - yes, 0 - no)\\
    \textit{Data Type:} Nominal.    
    \item \textbf{\textit{Flat:}} Is the scat a flat puddle that lacks other morphological traits? (1 - yes, 0 - no)\\
    \textit{Data Type:} Nominal.
    \item \textbf{\textit{Scrape:}} Is there a scrape mark near the scat? (1 - yes, 0 - no)\\
    \textit{Data Type:} Nominal.
\end{itemize}

\addcontentsline{toc}{subsection}{Clean Dataset}
I then converted the variables into their prospective types and renamed some of the variables with more descriptive names. When looking at the set, there are missing variables in some of the ratio variables. There are 19 total entries missing values, which is around 17\% of the dataset. Reid\cite{Reid2015} explains the 8 missing values diameter and taper that those samples have irregular morphologies, and therefore the measurements were irrelevant. He also states that he marked those entries as positive for flat. Considering the context, fill in the missing morphological variables with zero and the missing biogeochemical with the mean value and save them as new columns. Here is a table representing the output for df.describe() after converting :

\begin{table}[H]
\centering
\small 
\caption{Dataset after Type Conversion}
\begin{tabular}{lcc}
    \toprule
    \textbf{Column} & \textbf{Non-Null Count} & \textbf{Dtype} \\
    \midrule
    Species & 110 & category \\
    Month & 110 & category \\
    Year & 110 & category \\
    Site & 110 & category \\
    Location & 110 & category \\
    Age & 110 & int64 \\
    Number & 110 & int64 \\
    Length & 110 & float64 \\
    Diameter & 104 & float64 \\
    Taper & 93 & float64 \\
    Taper\_deg & 93 & float64 \\
    Mass & 109 & float64 \\
    Carbon13 & 108 & float64 \\
    Nitrogen15 & 108 & float64 \\
    CN & 108 & float64 \\
    Ropey & 110 & bool \\
    Segmented & 110 & bool \\
    Flat & 110 & bool \\
    Scrape & 110 & bool \\
    \bottomrule
\end{tabular}%  
\end{table}

\addcontentsline{toc}{subsection}{Categorize Variables}
After cleaning the data, I categorized each variable in the data set as morphological, biogeochemical, contextual, or not a trait. Here are the results:
\begin{table}[H]
\centering
\caption{Statistical Analysis Predictor Variables}
    \begin{tabular}{|l|l|} \hline
    \textbf{Variable} & \textbf{Trait Type} \\ \midrule
    Species    & Not a trait   \\ 
    Month      & Contextual    \\ 
    Year       & Contextual    \\ 
    Site       & Contextual    \\ 
    Location   & Contextual    \\ 
    Age        & Contextual \\ 
    Number     & Morphological \\ 
    Length     & Morphological \\ 
    Diameter   & Morphological \\ 
    Taper      & Morphological \\ 
    TI         & Morphological \\ 
    Mass       & Morphological \\ 
    d13C       & Biogeochemical\\ 
    d15N       & Biogeochemical\\ 
    CN         & Biogeochemical\\ 
    Ropey      & Morphological \\ 
    Segmented  & Morphological \\ 
    Flat       & Morphological \\ 
    Scrape     & Morphological \\ \bottomrule
\end{tabular}
\end{table}

% 3. Use the visualization and statistical testing techniques you’ve learned to evaluate the relationship between the morphological and biogeochemical traits and species. 
% Determine which traits can be used to distinguish at least one species from the rest.
\addcontentsline{toc}{subsection}{Visualization}
I then used visualisation techniques to find stand-out relationships between the morphological and biogeochemical traits and species. Some of the traits that stood out were mass, diameter, and CN out of the numeric variables and scrape of the categorical variables. The following pages contain the visualizations of those factors.

\end{multicols*}
\begin{figure}
    \centering
    \includegraphics[width=1\linewidth]{mass1.png}
\end{figure}
\begin{figure}
    \centering
    \includegraphics[width=1\linewidth]{mass.png}
    \caption{The three mass distributions are unique, making it stand out as a good variable for identifying species}
    \label{fig:mass}
\end{figure}

\begin{figure}
    \centering
    \includegraphics[width=1\linewidth]{diameter.png}
    \label{fig:diameter}
\end{figure}
\begin{figure}
    \centering
    \includegraphics[width=1\linewidth]{diamFill.png}
        \caption{The distribution of diameter for the GrayFox is different from the other two, making it stand out as a good variable for identifying GrayFox}
    \label{fig:diameterFill}
\end{figure}

\begin{figure}
    \centering
    \includegraphics[width=1\linewidth]{CN.png}
    \caption{The C:N Ratio distributions of the three species are pretty unique, with bobcat having a more limited range to its distribution}
    \label{fig:cn}
\end{figure}

\begin{figure}
    \centering
    \includegraphics[width=1\linewidth]{scrape.png}
    \caption{Scrape is only observed in bobcats, so it can be an indicator that the sample is from a bobcat}
    \label{fig:scrape}
\end{figure}

\newpage
\addcontentsline{toc}{subsection}{Statistical Testing}
I also used a Kruskal-Wallis test for each numerical variable versus the Species variable and a $\chi^2$ test of independence between each categorical variable versus the Species variable. 

I used the Kruskal-Wallis test to tell if one or more of the distributions
-----------------------------
% For each of the categorical values, I used a Kruskal- Wallis test via the scipy.kruskal() function. I did this by first grouping the samples by their category in a list, then putting it through the function. 

\begin{table}[H]
\centering
\caption{Statistical Analysis Results}
\begin{tabular}{lrrl}
\toprule
{} &  test statistic & p-value &  stat significant \\
\midrule
Number           &  8.023204 &  1.810437e-02 &   \textbf{True} \\
Length           &  0.170969 &  9.180672e-01 &  False \\
Diameter         & 28.149458 &  7.716545e-07 &   \textbf{True} \\
Taper            &  7.369808 &  2.509958e-02 &   \textbf{True} \\
Taper\_deg       &  1.753889 &  4.160523e-01 &  False \\
Mass             & 32.813354 &  7.493260e-08 &   \textbf{True} \\ \midrule
Carbon13         & 18.324181 &  1.049433e-04 &   \textbf{True} \\
Carbon13\_fill   & 20.361705 &  3.788889e-05 &   \textbf{True} \\
Nitrogen15       & 26.753599 &  1.550707e-06 &   \textbf{True} \\
Nitrogen15\_fill & 25.757070 &  2.552250e-06 &   \textbf{True} \\
CN               & 29.758638 &  3.451388e-07 &   \textbf{True} \\
CN\_fill         & 32.449464 &  8.988524e-08 &   \textbf{True} \\ \midrule
Ropey            &  0.144431 &  9.303303e-01 &  False \\
Segmented        & 18.060201 &  1.197505e-04 &   \textbf{True} \\
Flat             & 13.737408 &  1.039824e-03 &   \textbf{True} \\
Scrape           &  4.870510 &  8.757543e-02 &  False \\
\bottomrule
\end{tabular}
\end{table}

\newpage
\begin{table}[H]
\centering
\caption{Descriptive Statistics for Traits Distinguishing Species}
\label{tab:species_traits}
\begin{tabular}{lcccccc}
\toprule
\textbf{Trait} & \textbf{Species} & \textbf{Mean} & \textbf{Median} & \textbf{Mode} & \textbf{Std. Dev.} & \textbf{Range} \\
\midrule
\textbf{Diameter (cm)} &
  Bobcat &
  \cellcolor[HTML]{ECF4FF}\textbf{19.01} &
  \cellcolor[HTML]{ECF4FF}\textbf{18.2} &
  15.7 &
  2.95 &
  13.0 - 25.8 \\
\textbf{} &
  Coyote &
  \cellcolor[HTML]{ECF4FF}\textbf{20.27} &
  \cellcolor[HTML]{ECF4FF}\textbf{20.7} &
  18.1 &
  4.48 &
  11.0 - 30.0 \\
\textbf{} &
  \cellcolor[HTML]{CBCEFB}\textbf{Gray Fox} &
  \cellcolor[HTML]{9698ED}\textbf{14.97} &
  \cellcolor[HTML]{9698ED}\textbf{15.8} &
  \cellcolor[HTML]{CBCEFB}15.8 &
  \cellcolor[HTML]{CBCEFB}3.21 &
  \cellcolor[HTML]{CBCEFB}7.8 - 19.1 \\ \midrule
\textbf{Taper} &
  Bobcat &
  \cellcolor[HTML]{ECF4FF}\textbf{26.26} &
  25.9 &
  2.3 &
  12.74 &
  \cellcolor[HTML]{ECF4FF}\textbf{2.3 - 53.4} \\
\textbf{} &
  \cellcolor[HTML]{CBCEFB}\textbf{Coyote} &
  \cellcolor[HTML]{9698ED}\textbf{33.23} &
  \cellcolor[HTML]{CBCEFB}29.95 &
  \cellcolor[HTML]{CBCEFB}20.1 &
  \cellcolor[HTML]{CBCEFB}17.92 &
  \cellcolor[HTML]{9698ED}\textbf{5.0 - 91.5} \\
\textbf{} &
  Gray Fox &
  \cellcolor[HTML]{ECF4FF}\textbf{23.04} &
  19.4 &
  20.0 &
  15.56 &
  \cellcolor[HTML]{ECF4FF}\textbf{2.5 - 67.7} \\ \midrule
\textbf{Mass (grams)} &
  Bobcat &
  \cellcolor[HTML]{ECF4FF}\textbf{12.48} &
  11.25 &
  13.0 &
  6.53 &
  1.5 - 26.89 \\
\textbf{} &
  Coyote &
  \cellcolor[HTML]{ECF4FF}\textbf{18.25} &
  16.75 &
  0.94 &
  11.78 &
  0.94 - 53.7 \\
\textbf{} &
  \cellcolor[HTML]{CBCEFB}\textbf{Gray Fox} &
  \cellcolor[HTML]{9698ED}\textbf{5.64} &
  \cellcolor[HTML]{CBCEFB}4.605 &
  \cellcolor[HTML]{CBCEFB}2.05 &
  \cellcolor[HTML]{CBCEFB}3.38 &
  \cellcolor[HTML]{CBCEFB}2.05 - 18.14 \\ \midrule
\textbf{Carbon13} &
  Bobcat &
  \cellcolor[HTML]{ECF4FF}\textbf{-27.70} &
  -27.80 &
  -27.92 &
  1.04 &
  \cellcolor[HTML]{ECF4FF}\textbf{-29.85 - -24.55} \\
\textbf{} &
  \cellcolor[HTML]{CBCEFB}\textbf{Coyote} &
  \cellcolor[HTML]{9698ED}\textbf{-24.82} &
  \cellcolor[HTML]{CBCEFB}-25.42 &
  \cellcolor[HTML]{CBCEFB}-29.62 &
  \cellcolor[HTML]{CBCEFB}3.08 &
  \cellcolor[HTML]{9698ED}\textbf{-29.62 - -19.67} \\
\textbf{} &
  Gray Fox &
  \cellcolor[HTML]{ECF4FF}\textbf{-27.28} &
  -27.30 &
  -26.99 &
  1.15 &
  \cellcolor[HTML]{ECF4FF}\textbf{-28.97 - -25.18} \\ \midrule
\textbf{Nitrogen15} &
  Bobcat &
  \cellcolor[HTML]{ECF4FF}\textbf{6.38} &
  6.39 &
  5.36 &
  1.81 &
  1.84 - 11.27 \\
\textbf{} &
  \cellcolor[HTML]{CBCEFB}\textbf{Coyote} &
  \cellcolor[HTML]{9698ED}\textbf{10.38} &
  \cellcolor[HTML]{CBCEFB}8.92 &
  \cellcolor[HTML]{CBCEFB}3.97 &
  \cellcolor[HTML]{CBCEFB}3.83 &
  \cellcolor[HTML]{CBCEFB}3.97 - 18.00 \\
\textbf{} &
  Gray Fox &
  \cellcolor[HTML]{ECF4FF}\textbf{6.46} &
  6.13 &
  3.90 &
  1.75 &
  3.90 - 10.10 \\ \midrule
\textbf{CN Ratio} &
  \cellcolor[HTML]{CBCEFB}\textbf{Bobcat} &
  \cellcolor[HTML]{9698ED}\textbf{6.71} &
  \cellcolor[HTML]{9698ED}\textbf{6.70} &
  \cellcolor[HTML]{CBCEFB}7.0 &
  \cellcolor[HTML]{CBCEFB}1.10 &
  \cellcolor[HTML]{CBCEFB}4.5 - 10.4 \\
\textbf{} &
  Coyote &
  \cellcolor[HTML]{ECF4FF}\textbf{8.59} &
  \cellcolor[HTML]{ECF4FF}\textbf{7.70} &
  7.7 &
  2.34 &
  4.9 - 15.4 \\
\textbf{} &
  \cellcolor[HTML]{CBCEFB}\textbf{Gray Fox} &
  \cellcolor[HTML]{9698ED}\textbf{12.12} &
  \cellcolor[HTML]{9698ED}\textbf{9.65} &
  \cellcolor[HTML]{CBCEFB}7.7 &
  \cellcolor[HTML]{CBCEFB}5.62 &
  \cellcolor[HTML]{CBCEFB}5.7 - 23.6 \\  \midrule
\textbf{Segmented} &
  Bobcat &
  \multicolumn{5}{c}{True (42), False (15)} \\
\textbf{} &
  Coyote &
  \multicolumn{5}{c}{False (14), True (14)} \\
\textbf{} &
  Gray Fox &
  \multicolumn{5}{c}{False (19), True (6)} \\ \midrule
\textbf{Flat} &
  \cellcolor[HTML]{CBCEFB}\textbf{Bobcat} &
  \multicolumn{5}{c}{\cellcolor[HTML]{CBCEFB}False (57)} \\
\textbf{} &
  Coyote &
  \multicolumn{5}{c}{False (27), True (1)} \\
\textbf{} &
  Gray Fox &
  \multicolumn{5}{c}{False (20), True (5)} \\ \midrule
\end{tabular}
\end{table}

\begin{multicols*}{2}
\section{Discussion and Interpretation}
From the analysis conducted, several traits were found to separate between the three species: length, mass, diameter, and the C:N ratio (CN). The summary of the descriptive statistics for these traits is provided in table \ref{tab:species_traits} above.


% 2. Using what you learned from your research, explain how the traits with differences relate to differences in the biology of the three species.


% 3. Explain why you think that predictive morphological and biogeochemical traits might be more useful than contextual traits to ecologists.


% \vfill 
\section{Conclusion}

% 1. Write a conclusion. The conclusion should summarize your key findings and describe what someone could do to continue or expand on this work



\end{multicols*}

\newpage

\printbibliography

\newpage
\section{Appendix}

\begin{sidewaystable}[tb]
\centering
\caption{Initial Dataset}
\label{tab:initial}
\resizebox{\linewidth}{!}{%
\begin{tabularx}{1.375\linewidth}{|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|x|}
\hline
Species & Month & Year & Site & Location & Age & Number & Length & Diameter & Taper & TI & Mass & d13C & d15N & CN & Ropey & Segmented & Flat & Scrape \\
\hline
Coyote & January & 2012 & YOLA & Edge & 5 & 2 & 9.5 & 25.7 & 41.9 & 1.63 & 15.89 & -26.85 & 6.94 & 8.5 & 0 & 0 & 0 & 0 \\
Coyote & January & 2012 & YOLA & Edge & 3 & 2 & 14.0 & 25.4 & 37.1 & 1.46 & 17.61 & -29.62 & 9.87 & 11.3 & 0 & 0 & 0 & 0 \\
Bobcat & January & 2012 & YOLA & Middle & 3 & 2 & 9.0 & 18.8 & 16.5 & 0.88 & 8.40 & -28.73 & 8.52 & 8.1 & 1 & 1 & 0 & 1 \\
Coyote & January & 2012 & YOLA & Middle & 5 & 2 & 8.5 & 18.1 & 24.7 & 1.36 & 7.40 & -20.07 & 5.79 & 11.5 & 1 & 0 & 0 & 0 \\
Coyote & January & 2012 & YOLA & Edge & 5 & 4 & 8.0 & 20.7 & 20.1 & 0.97 & 25.45 & -23.24 & 7.01 & 10.6 & 0 & 1 & 0 & 0 \\
Coyote & January & 2012 & YOLA & Edge & 5 & 3 & 9.0 & 21.2 & 28.5 & 1.34 & 14.14 & -29.00 & 8.28 & 9.0 & 1 & 0 & 0 & 0 \\
Bobcat & January & 2012 & ANNU & OffEdge & 1 & 5 & 6.0 & 15.7 & 8.2 & 0.52 & 14.82 & -28.06 & 4.20 & 5.4 & 1 & 1 & 0 & 1 \\
Bobcat & January & 2012 & ANNU & OffEdge & 3 & 7 & 5.5 & 21.9 & 19.3 & 0.88 & 26.41 & -27.60 & 3.89 & 5.6 & 0 & 1 & 0 & 0 \\
Bobcat & January & 2012 & ANNU & OffEdge & 5 & 2 & 11.0 & 17.5 & 29.1 & 1.66 & 16.24 & -28.64 & 7.34 & 5.8 & 0 & 1 & 0 & 0 \\
Bobcat & January & 2012 & ANNU & Middle & 5 & 1 & 20.5 & 18.0 & 21.4 & 1.19 & 11.22 & -27.35 & 6.06 & 7.7 & 1 & 1 & 0 & 0 \\
GrayFox & January & 2012 & ANNU & Middle & 3 & 1 & 8.0 & NaN & NaN & NaN & 2.51 & -25.79 & 7.83 & 20.5 & 0 & 0 & 1 & 0 \\
GrayFox & January & 2012 & ANNU & Middle & 1 & 1 & 8.0 & 12.9 & 14.7 & 1.14 & 8.55 & -25.71 & 8.47 & 18.1 & 1 & 0 & 0 & 0 \\
GrayFox & January & 2012 & ANNU & Middle & 3 & 1 & 12.0 & NaN & NaN & NaN & 18.14 & -25.18 & 10.10 & 15.5 & 0 & 0 & 1 & 0 \\
GrayFox & January & 2012 & ANNU & Middle & 3 & 1 & 11.5 & NaN & NaN & NaN & 8.17 & -25.73 & 9.72 & 18.9 & 0 & 0 & 1 & 0 \\
GrayFox & January & 2012 & ANNU & Middle & 1 & 1 & 8.5 & NaN & NaN & NaN & 3.43 & -26.17 & 8.07 & 19.9 & 0 & 0 & 1 & 0 \\
\bottomrule
\end{tabularx}%
}
\end{sidewaystable}

\end{document}