---
title: "Data Nugget 3"
author: "Elli Bosch"
date: "4/4/2019"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```


#Loading the Data into R 

```{r}
library(readxl)
Data <- read_excel("Data - Nutrients in the James River.xlsx")
View(Data)
summary(Data)
```

#Installing necessary packages
```{r}
install.packages("scales")
library("scales")

install.packages("tidyverse")
library(tidyverse)

options(scipen=5)
```

#Scatterplot of Nutrients v. Discharge 
```{r}
nutrientsvdischarge <- ggplot(data=Data, aes(x=Discharge2, y=Kgperyear, col=Nutrient)) + geom_point() + labs(x = "Discharge (m3/y)", y = "Nutrient Concentration (kg/y)") + ggtitle("Nutrient Concentration (kg/y) vs. Discharge (m3/y)") + scale_y_continuous(labels=comma) + scale_x_continuous(labels=comma) 

nutrientsvdischarge
```




#Looking at Nitrogen
#Testing for Normality of Nitrogen
```{r}
shapiro.test(Data$TN)
```
*Teacher note: this p-value means that we should accept the null hypothesis that our data for nitrogen is normally distributed*


##Nitrogen Linear Regression Plot and Summary (with diagnostic plots)
```{r}
plot( TN  ~ Discharge , Data)
mod <- lm(TN ~ Discharge, data=Data)
abline(mod) 
```


```{r}
summary(mod)
plot(mod)
```



```{r}
mod2 <- ggplot(Data, aes(x=Data$Discharge, y=Data$TN)) + xlab("Discharge (m3/y)") + ylab("Total Nitrogen (kg/y)") + 
  labs(title = "Linear model of total nitrogen concentration as a function of discharge") + geom_point(shape=16, colour="#CC6666") + 
  geom_smooth(method=lm, colour="#CC6666") + scale_fill_grey(start=0.8, end=0.2) + 
  theme_classic() + scale_y_continuous(limits = c(4000000, 12000000), labels=comma) + scale_x_continuous(labels=comma)
```

```{r}
plot(mod2)
```


#Looking at Phosphorus
#Testing for Normality of Phosphorus

```{r}
shapiro.test(Data$TP)
```
*Teacher note: this p-value means that we should accept the null hypothesis that our data for phosphorus is normally distributed*

##Phosphorus Linear Regression Plot and Summary (with diagnostic plots)
```{r}
plot( TP  ~ Discharge , Data)
mod3 <- lm(TP ~ Discharge, data=Data)
abline(mod3)
```


```{r}
summary(mod3)
plot(mod3)
```




##Linear model graph made nice

```{r}

mod4 <- ggplot(Data, aes(x=Data$Discharge, y=Data$TP)) + xlab("Discharge (m3/y)") + ylab("Total Phosphorus (kg/y)") + 
  labs(title = "Linear model of total phosphorus concentration as a function of discharge") + geom_point(shape=16, colour="#56B4E9") + 
  geom_smooth(method=lm, colour="#56B4E9") + scale_fill_grey(start=0.8, end=0.2) + 
  theme_classic() + scale_y_continuous(limits = c(200000, 1600000), labels=comma) + scale_x_continuous(labels=comma)
```

```{r}
plot(mod4)
```