CO_Turtles_Exploratory.Rmd

---
title: "CO Turtles Exploratory Analyses"
explanation: "Exploratory analysis of spatial-capture-recapture data recorded summer 2018"
author: "Natalie Haydt"
output:
  pdf_document: default
  html_notebook: default
---



```{r, checking working directory}
# getwd()
library(lme4)
library(ggplot2)
```

```{r, read in scr data}
scr.data <- read.csv(file = "Data/canalscr_9_1_18.csv", stringsAsFactors = FALSE)
head(scr.data)
```

```{r, data summary}
summary(scr.data)
```

```{r, histograms}
# need to separate by site?

par(mfrow = c(1,2))
sizes <- hist(scr.data$carapace, main = "Carapace Size Distribution", xlab = "Carapace Size Per Turtle (mm)")
leeches <- hist(scr.data$leeches, main = "Leech Count Distribution", xlab = "Number of Leeches Per Turtle" )
par(mfrow = c(1,1))
```

```{r}
scr.mf <- scr.data[which(scr.data$sex != "U" & !is.na(scr.data$sex)), ]
ggplot(data = scr.mf, aes(carapace)) + geom_histogram(aes(fill = species)) + facet_wrap(~sex) +
  ggtitle("Frequency of Turtles Caught by Size, Sex, and Species") +
  theme(plot.title = element_text(hjust = 0.5, size = 16),
        axis.title.x = element_text(size=14),
        axis.title.y = element_text(size=14))

 # scale_color_manual(values=c("lightgodlenrod4", "#darkorange3", "rosybrown3", "wheat2", "bisque2", "grey50"))+
 # scale_fill_manual(values=c("lightgodlenrod4", "#darkorange3", "rosybrown3", "wheat2", "bisque2", "grey50"))


ggplot(data = scr.mf, aes(carapace)) + geom_histogram(aes(fill = sex)) + facet_wrap(~species)
```

```{r, recap rate histogram}

recaps <- read.csv(file = "Data/recaprates_9_1_18.csv")
hist(recaps$prop_recap, breaks = 10, xlab = "Proportion of Recaps", main = "Proportion of Recaptures Across C&O Sites", col = "darkorange3")



p <- ggplot(recaps, aes(x=prop_recap)) + 
  geom_histogram(binwidth = 0.05, color="darkorange3", fill = "darkorange3", position = position_dodge())
print(p)

p + labs(x ="Proportion of Recaps", y = "Frequency")+
   theme_classic() +
   scale_fill_manual(values=c('#999999','#E69F00')) +
  ggtitle("Proportion of Recaptures Across C&O Sites") +
  theme(plot.title = element_text(hjust = 0.5, size = 16),
        axis.title.x = element_text(size=14),
        axis.title.y = element_text(size=14)
)


```

```{r, subsetting by species}
scr_cpic <- subset(scr.data, species == "CPIC")
scr_sodo <- subset(scr.data, species == "SODO")
scr_cser <- subset(scr.data, species == "CSER")
scr_prub <- subset(scr.data, species == "PRUB")
scr_gins <- subset(scr.data, species == "GINS")
scr_tscr <- subset(scr.data, species == "TSCR")
```


```{r, species summaries}
summary(scr_cpic)
summary(scr_sodo)
summary(scr_cser)
summary(scr_prub)
summary(scr_gins)
## only 1 tscr caught, so did not create summary
```
```{r, histograms per species}
hist(scr_cpic$carapace)
hist(scr_sodo$carapace)
hist(scr_cser$carapace)
hist(scr_prub$carapace)

hist(scr_cpic$leeches)
hist(scr_sodo$leeches)
hist(scr_cser$leeches)
hist(scr_prub$leeches)
```


```{r, size by site per species, ## add in overall}

cpic_size_site <- aov(data = scr_cpic, carapace ~ site)
cpic_size_summary <- summary(cpic_size_site)
cpic_size_summary
cpic_posthoc <- TukeyHSD(cpic_size_site, 'site', conf.level = 0.95)
cpic_posthoc

sodo_size_site <- aov(data = scr_sodo, carapace ~ site)
sodo_size_summary <- summary(sodo_size_site)
sodo_size_summary
sodo_posthoc <- TukeyHSD(sodo_size_site, 'site', conf.level = 0.95)
sodo_posthoc

cser_size_site <- aov(data = scr_cser, carapace ~ site)
cser_size_summary <- summary(cser_size_site)
cser_size_summary
cser_posthoc <- TukeyHSD(cser_size_site, 'site', conf.level = 0.95)
cser_posthoc

prub_size_site <- aov(data = scr_prub, carapace ~ site)
prub_size_summary <- summary(prub_size_site)
prub_size_summary
prub_posthoc <- TukeyHSD(prub_size_site, 'site', conf.level = 0.95)
prub_posthoc

all_size_site <- aov(data = scr.data, carapace ~ site)
all_size_summary <- summary(all_size_site)
all_size_summary
all_size_posthoc <- TukeyHSD(all_size_site, 'site', conf.level = 0.95)
all_size_posthoc

```

## Exploration of leech data

```{r, leeches and site; per species and ##overall}

cpic_leeches_site <- aov(data = scr_cpic, leeches ~ site)
cpic_leeches_summary <- summary(cpic_leeches_site)
cpic_leeches_summary
cpic_leeches_posthoc <- TukeyHSD(cpic_leeches_site, 'site', conf.level = 0.95)
cpic_leeches_posthoc

sodo_leeches_site <- aov(data = scr_sodo, leeches ~ site)
sodo_leeches_summary <- summary(sodo_leeches_site)
sodo_leeches_summary
sodo_leeches_posthoc <- TukeyHSD(sodo_leeches_site, 'site', conf.level = 0.95)
sodo_leeches_posthoc

cser_leeches_site <- aov(data = scr_cser, leeches ~ site)
cser_leeches_summary <- summary(cser_leeches_site)
cser_leeches_summary
cser_leeches_posthoc <- TukeyHSD(cser_leeches_site, 'site', conf.level = 0.95)
cser_leeches_posthoc

prub_leeches_site <- aov(data = scr_prub, leeches ~ site)
prub_leeches_summary <- summary(prub_leeches_site)
prub_leeches_summary
prub_leeches_posthoc <- TukeyHSD(prub_leeches_site, 'site', conf.level = 0.95)
prub_leeches_posthoc

all_leeches_site <- aov(data = scr.data, leeches ~ site)
all_leeches_summary <- summary(all_leeches_site)
all_leeches_summary
all_leeches_posthoc <- TukeyHSD(all_leeches_site, 'site', conf.level = 0.95)
all_leeches_posthoc

```

### Regression of number of leeches as a function of the species

Q: do different species have different numbers of leeches

H0: There no diff in the number of leeches among species

HA: Musk turtles (SODO) will have more leeches than basking turtles

*Will eventually want to consider leeches per unit area since the turtle species are different sizes or adding size as a covariate interacting with species*

```{r, leeches and species}
 #add in site as random effect

leeches_species <- aov(data = scr.data, leeches ~ species)
leeches_species_summary <- summary(leeches_species)
leeches_species_summary
##leeches_species_posthoc <- TukeyHSD(leeches_species, 'leeches', conf.level = 0.95)
##NaNs produced (need to use NA.omit?)
##leeches_species_posthoc

glmer1 <- glmer(leeches ~ species + (1|site), data = scr.data, family = poisson)
summary(glmer1)

hist(rnorm(10000, exp(0.31932), sd = 0.4694))

library(dplyr)
scr.scaled <- scr.data %>%
  filter(species != "GINS",
         species != "TSCR") %>%
  mutate(mass.s = (mass - mean(mass)) / sd(mass))
summary(scr.scaled)

glmer2 <- glmer(leeches ~ species + mass.s + (1|site), data = scr.scaled, family = poisson)
summary(glmer2)
```

```{r, leeches and size - regression}
#add in site as random effect

leeches_size <- aov(data = scr.data, leeches ~ carapace)
leeches_size_summary <- summary(leeches_size)
leeches_size_summary

cpic_leeches_size <- aov(data = scr_cpic, leeches ~ carapace)
cpic_leeches_size_summary <- summary(cpic_leeches_size)
cpic_leeches_size_summary
##cpic_leeches_size_posthoc <- TukeyHSD(cpic_leeches_size, 'carapace', conf.level = 0.95)
##cpic_leeches_size_posthoc

##sodo_leeches_size <- aov(data = scr_sodo, leeches ~ carapace)
##sodo_leeches_size_summary <- summary(sodo_leeches_size)
##sodo_leeches_size_summary
##sodo_leeches_size_posthoc <- TukeyHSD(sodo_leeches_size, 'carapace', conf.level = 0.95)
##sodo_leeches_size_posthoc

##cser_leeches_size <- aov(data = scr_cser, leeches ~ carapace)
##cser_leeches_size_summary <- summary(cser_leeches_size)
##cser_leeches_size_summary
##cser_leeches_size_posthoc <- TukeyHSD(cser_leeches_size, 'carapace', conf.level = 0.95)
##cser_leeches_size_posthoc

##prub_leeches_size <- aov(data = scr_prub, leeches ~ carapace)
##prub_leeches_size_summary <- summary(prub_leeches_size)
##prub_leeches_size_summary
##prub_leeches_size_posthoc <- TukeyHSD(prub_leeches_size, 'carapace', conf.level = 0.95)
##prub_leeches_size_posthoc

```

```{r, total caugt vs filled status}

```


Analyses Based off of Raabe et al. 2014:


Analyses Based off of Kornilev et al. 2010: