temperature.Rmd

---
title: "temperature"
author: "Gracie Goheen"
date: "`r Sys.Date()`"
output: 
  github_document:
    toc: true
---

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

```{r message=FALSE, warning=FALSE}
# Libraries
library(tidyverse)
library(lubridate)

# Parameters
temperature_data <- here::here("c01-own/data/temperature.rds")
temperature <- read_rds(temperature_data)
#===============================================================================

```

## Temperature by City 

```{r}
summary(temperature)
```

```{r}
sum(is.na(temperature))
```

```{r}
temperature %>% 
  count(city_name)
```

```{r}
temperature %>% 
  mutate(date = ymd(date)) %>% 
  filter(year(date) == 2019)
```

```{r}
temperature %>% 
  top_n(n = 1, wt = temperature)
```

```{r}
temperature %>% 
  mutate(
    date = ymd(date),
    year = year(date)
  ) %>% 
  count(year) %>% 
  arrange(desc(year))
```


```{r}
temperature %>% 
  filter(city_name == "PERTH") %>% 
  ggplot(aes(date, temperature, color = temp_type)) +
  geom_line()
```

```{r}
temperature %>% 
  mutate(date = ymd(date)) %>% 
  filter(
    year(date) == 2019, 
  ) %>% 
  ggplot(aes(date, temperature, color = temp_type)) +
  geom_line() +
  facet_wrap(vars(city_name))

```


```{r}
temperature %>% 
  group_by(city_name, temp_type) %>% 
  summarise(avg_temp = median(temperature, na.rm = TRUE)) %>% 
  ggplot(aes(fct_reorder(city_name, avg_temp), avg_temp, color = temp_type)) +
  geom_point()
```

```{r}
temperature %>% 
  mutate(
    city_name = as_factor(city_name)
  ) %>% 
  ggplot(aes(city_name, temperature, color = temp_type)) +
  geom_boxplot(position = "dodge")
```


```{r}
temperature %>% 
  mutate(
    year = year(date)
  ) %>% 
  group_by(year, temp_type) %>% 
  summarise(avg_temp = median(temperature, na.rm = TRUE)) %>%
  ggplot(aes(year, avg_temp, color = temp_type)) +
  geom_line()
```

```{r}
temperature %>% 
  mutate(
    year = year(date)
  ) %>% 
  filter(year %in% c(2016:2019)) %>% 
  group_by(year, temp_type) %>% 
  summarise(avg_temp = median(temperature, na.rm = TRUE)) %>%
  ggplot(aes(year, avg_temp, color = temp_type)) +
  geom_line()
```


```{r}
# Attempt to recreate the New York Times "2019 was Australia’s hottest year" graph.

temperature %>% 
  drop_na() %>% 
  mutate(year = year(date)) %>% 
  group_by(year) %>% 
  summarise(temp_year = mean(temperature)) %>% 
  mutate(
    avg_temp = mean(temp_year),
    temp_dif = temp_year - avg_temp
  ) %>% 
  ggplot(aes(year, temp_dif)) +
  geom_col()
```

```{r}
# Attempt to recreate the New York Times "2019 was Australia’s hottest year" graph.

temperature %>% 
  drop_na() %>% 
  filter(temp_type == "max") %>%
  mutate(year = year(date)) %>% 
  group_by(year) %>% 
  # median or mean? max or min?
  summarise(temp_year = mean(temperature)) %>% 
  mutate(
    avg_temp = mean(temp_year),
    temp_dif = temp_year - avg_temp
  ) %>% 
  ggplot(aes(year, temp_dif)) +
  geom_col()
```

```{r}
# Attempt to recreate the New York Times "2019 was Australia’s hottest year" graph.

# The 1961–1990 average temperature
avg_temp_range <- 
  temperature %>% 
  drop_na() %>% 
  mutate(year = year(date)) %>% 
  filter(year %in% c(1961:1990)) %>% 
  summarize(mean(temperature))

temp_2019 <-
  temperature %>% 
  drop_na() %>% 
  mutate(year = year(date)) %>% 
  group_by(year) %>% 
  summarise(temp_year = mean(temperature)) %>% 
  mutate(
    temp_dif =  temp_year - as.double(avg_temp_range)
  ) %>% 
  filter(year == 2019)

temperature %>% 
  drop_na() %>% 
  mutate(year = year(date)) %>% 
  group_by(year) %>% 
  summarise(temp_year = mean(temperature)) %>% 
  mutate(
    temp_dif =  temp_year - as.double(avg_temp_range)
  ) %>% 
  ggplot(aes(year, temp_dif)) +
  geom_col(fill = "grey") +
  geom_col(data = temp_2019, aes(year, temp_dif, fill = "red")) +
  theme(
    axis.title.x = element_blank(),
    axis.ticks = element_blank(),
    legend.position = "none",
    panel.background = element_rect(fill = "transparent"),
    panel.grid = element_line(color = "transparent"),
    panel.grid.major.y = element_line(color = "White"),
    panel.ontop = TRUE
  ) +
  labs(
    title = "2019 was Australia’s hottest year.",
    y = "Annual temperature above or below the 1961–1990 average"
  )
```


Something seems to be wrong with my data, because my graph looks very different. I also need to figure out how to do the y scale and caption / arrows.