step_6_final_report.Rmd

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

```{r}
knitr::opts_chunk$set(echo = TRUE)
```


```{r include = FALSE}
library(tidyverse)
library(sf)

# Parameters
  # File where generated answers are saved, by default in the home directory
file_answers <- "/Users/graciegoheen/Downloads/dcl/own/file_answers.rds"
  # Maximum temperature .grid file for December 29 2019
max_temp_file <- "/Users/graciegoheen/Downloads/dcl/own/2019122920191229.grid"
  # Geometry data for countries
countries_data <- "/Users/graciegoheen/Downloads/dcl/ne_110m_admin_0_countries/ne_110m_admin_0_countries.shp"
  # Geometry data for cities
places_data <- "/Users/graciegoheen/Downloads/dcl/own/ne_110m_populated_places/ne_110m_populated_places.shp"
  # Map colors
colors <- 
  c(
    "18" = "#f6d387",
    "21" = "#f3bc67",
    "24" = "#eda455",
    "27" = "#ea955a",
    "30" = "#df8561",
    "33" = "#da8368",
    "36" = "#d16768",
    "39" = "#bf191f",
    "42" = "#97040b",
    "45" = "#6f0206"
  )
  # Australian climate variability & change - Time series graphs - mean temperature
temp_anon_file <- "/Users/graciegoheen/Downloads/dcl/own/latestsort_temp.txt"
  # Temperature average from 1961 - 1990) in °C according to the Australian Government Bureau of Meteorology
temp_avg <- 21.8
  # Australian climate variability & change - Time series graphs - rainfall anonmaly
rain_anon_file <- "/Users/graciegoheen/Downloads/dcl/own/latestsort_rainfall_anon.txt"
 # Australian climate variability & change - Time series graphs - rainfall total
rain_total_file <- "/Users/graciegoheen/Downloads/dcl/own/latestsort_rainfall.txt"
  # Rainfall average from 1961 - 1990 in mm according to the Australian Government Bureau of Meteorology
rain_avg <- 465.2
  # Color for 2019 column
color_2019 <- "#c7432e"
  # Colors for year range graph
range_colors <- 
  c(
    "1910-1980s" = "#dcd0c5", 
    "1990s" = "#b8dae4",
    "2000s" = "#6284a4",
    "2010s" = "#a35066"
  )
  # NASA "Active Fires Dataset" via FIRMS (1 km estimates)
nasa_fire_link <- "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-01-07/MODIS_C6_Australia_and_New_Zealand_7d.csv"

#===============================================================================

temp_anon <-
  read_table(
    temp_anon_file, 
    col_names = c("year", "temp_anomaly_c")
  ) %>% 
  mutate(year = year %>% str_extract("^\\d{4}") %>% as.double())

rain_anon <-
  read_table(
    rain_anon_file, 
    col_names = c("year", "rainfall_anomaly_mm")
  ) %>% 
  mutate(year = year %>% str_extract("^\\d{4}") %>% as.double())

rain_total <-
  read_table(
    rain_total_file, 
    col_names = c("year", "rainfall_mm")
  ) %>% 
  mutate(year = year %>% str_extract("^\\d{4}") %>% as.double())

nasa_fire <- read_csv(nasa_fire_link)

countries <- read_sf(dsn = countries_data)

australia <- 
  countries %>% 
  filter(NAME == "Australia")

places <- read_sf(dsn = places_data)

aus_places <- 
  places %>% 
  filter(SOV0NAME == "Australia")

max_temp_matrix <-
  max_temp_file %>% 
  read_delim(
    delim = " ", 
    col_names = FALSE, 
    na = "99999.90", 
    trim_ws = TRUE, 
    skip = 6, 
    n_max = 691) %>% 
  select(-X887) %>% 
  as.matrix()

max_temp <- 
  crossing(i = 1:691, j = 1:886) %>% 
  mutate(
    max_temp = map2_dbl(i, j, ~ max_temp_matrix[.x,.y]),
    lat = -44.500 + 0.050 * 691 - 0.050 * i,
    long = 112.000 + 0.050 * j
  ) %>%
  select(long, lat, max_temp)
```

Austrailia's most recent fire season was one of the most extreme in its history, with around 11m hectares burnt and over than 120 fires still raging as of January 21st. This total area is more than five times the area destroyed by the 2018 California wildfires. More than 20% of Austrailia's forest has been destroyed and many highly-populated areas have faced severe burning. According to [The Guardian](https://www.theguardian.com/australia-news/datablog/ng-interactive/2019/dec/07/how-big-are-the-fires-burning-on-the-east-coast-of-australia-interactive-map), "this season's bushfires in south-eastern Australia have killed at least 32 people and destroyed almost 2,000 homes in New South Wales, Victoria and South Australia." 

So, why was this year so bad for bushfires in Australia? 

Note: This project uses data from a [tidytuesday challenge](https://github.com/rfordatascience/tidytuesday/blob/master/data/2020/2020-01-07/readme.md).

Source articles: 

*[Here’s Where Australia’s Destructive Wildfires Are Burning, New York Times](https://www.nytimes.com/interactive/2020/01/02/climate/australia-fires-map.html)

*[Australia fires: A visual guide to the bushfire crisis, BBC News](https://www.bbc.com/news/world-australia-50951043)

*[Australia’s deadly wildfires in numbers, Financial Times](https://www.ft.com/content/c068339e-3c55-11ea-b232-000f4477fbca)

```{r}
temp_2019 <-
  temp_anon %>% 
  filter(year == 2019)

temp_anon %>% 
  ggplot(aes(year, temp_anomaly_c)) +
  geom_col(fill = "grey") +
  geom_col(data = temp_2019, aes(year, temp_anomaly_c, fill = color_2019)) +
  scale_y_continuous(
    breaks = seq(-1, 1.5, 0.5),
    labels = c("-1°", "-0.5°", "0°", "+0.5°", "+1°", "+1.5°")
  ) +
  annotate(
    "text",
    x = c(2013, 1998, 1917, 1915.8, 1917),
    y = c(1.5, 1.05, -1.35, 1.5, 0.5),
    label = 
      c(
        "2019 →", 
        "1998", 
        "1917", 
        "warmer than average", 
        "Annual temperature above or below \nthe 1961–1990 average"
        ),
    color = c("black", "gray", "gray", "black", "cornflowerblue"),
    hjust = c(0.5, 0.5, 0.5, 0.5, 0),
    size = 3.5
  ) +
  theme(
    axis.title.x = element_blank(),
    axis.text.x = element_blank(),
    axis.text.y = element_text(color = "black"),
    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 = "gray97")
  ) +
  labs(
    title = "2019 was Australia’s hottest year",
    x = NULL,
    y = NULL,
    caption = "Source: Australian Government Bureau of Meteorology"
  )
```


```{r}
rain_2019 <-
  rain_anon %>% 
  filter(year == 2019)

rain_anon %>% 
  ggplot(aes(year, rainfall_anomaly_mm)) +
  geom_col(fill = "grey") +
  geom_col(data = rain_2019, aes(year, rainfall_anomaly_mm, fill = color_2019)) +
  scale_y_continuous(
    breaks = seq(-200, 200, 100),
    labels = c("-200", "-100", "0", "100", "200")
  ) +
  annotate(
    "text",
    x = c(2013, 1974, 1902, 1900, 1924),
    y = c(-180, 310, -165, 200, 100), 
    label = 
      c(
        "2019 →",
        "1974",
        "1902",
        "millimeters",
        "Rainfall difference \nfrom average"
      ),
    color = c("black", "gray", "gray", "black", "cornflowerblue"),
    hjust = c(0.5, 0.5, 0.5, 0.5, 0),
    size = 3.5
  ) +
  theme(
    axis.title.x = element_blank(),
    axis.text.x = element_blank(),
    axis.text.y = element_text(color = "black"),
    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 = "gray97")
  ) +
  labs(
    title = "2019 was Australia’s driest year",
    x = NULL,
    y = NULL,
    caption = "Source: Australian Government Bureau of Meteorology"
  )
```


The ideal conditions for the spread of wildfires are a hot and dry climate matched with strong winds. 2019 was both the hotest and driest year on record for Austrailia. 

```{r}
temp_total <-
  temp_anon %>% 
  mutate(temp_c = temp_anomaly_c + temp_avg) %>% 
  left_join(
    y = rain_total,
    by = "year"
  ) %>% 
  mutate(
    year_range = 
      case_when(
        year < 1990 ~ "1910-1980s",
        year < 2000 ~ "1990s",
        year < 2010 ~ "2000s",
        year < 2020 ~ "2010s"
      )
  )

top_temp_recent <-
  temp_total %>% 
  filter(year > 2010) %>% 
  top_n(temp_c, n = 4)

temp_total %>% 
  ggplot(aes(temp_c, rainfall_mm, fill = factor(year_range))) +
  geom_hline(yintercept = rain_avg, color = "gray40", linetype = "dashed") +
  geom_vline(xintercept = temp_avg, color = "gray40", linetype = "dashed") +
  geom_point(shape = 21, color = "white", size = 3, alpha = 0.8) +
  geom_point(shape = 21, color = "black", size = 3, data = top_temp_recent) +
  ggrepel::geom_text_repel(
    aes(label = year),
    data = top_temp_recent,
    size = 3,
    nudge_x = c(-0.25, -0.25, 0.05, -0.25),
    hjust = 0
  ) +
  annotate(
    geom = "text",
    x = c(-Inf, Inf, Inf, -Inf),
    y = c(Inf, Inf, -Inf, -Inf),
    label = c("Cold and wet", "Hot and wet", "Hot and dry", "Cold and dry"),
    hjust = c(-0.1, 1.1, 1.1, -0.1),
    vjust = c(2, 2, -1, -1),
    color = "gray40"
  ) +
  scale_x_continuous(expand = expand_scale(add = c(0.6, 0.8))) +
  scale_fill_manual(
    values = range_colors
  ) +
  coord_fixed(ratio = 0.004) +
  guides(
    fill = 
      guide_legend(
        title.position = "top", 
        title.hjust = 0.5, 
        label.position = "bottom", 
        nrow = 1)
  ) +
  theme(
    legend.position = "top",
    legend.spacing.x = unit(1.0, "cm"),
    legend.key = element_blank(),
    legend.text = element_text(margin = margin(t = -5)),
    panel.background = element_rect(fill = "transparent"),
    panel.grid = element_line(color = "transparent"),
    panel.grid.major.y = element_line(color = "gray97"),
    panel.grid.minor = element_blank()
  ) +
  labs(
    title = "Record heat and drought have fuelled Austrailia's bushfires",
    subtitle = "Average annual temperature and rainfall 1910-2019",
    x = "Temperature (°C)",
    y = "Rainfall (mm)",
    fill = NULL,
    caption = "Source: Australian Government Bureau of Meteorology"
  ) 
```

Drought and heatwaves are becoming more and more prevelant. Looking towards Austrailia's next fire season, there is a growing risk that fires of this magnitude could happen again.

```{r}
nasa_fire_aus_only <-
  st_intersection(
    x = nasa_fire %>% st_as_sf(coords = c("longitude", "latitude"), crs = 4326) ,
    y = australia
  )

nasa_fire_aus_only %>% 
  transmute(
    longitude = st_coordinates(geometry) %>% as_tibble() %>% pull(X),
    latitude = st_coordinates(geometry) %>% as_tibble() %>% pull(Y)
  ) %>% 
  ggplot() +
  geom_sf(data = australia, size = 0.1, fill = "seashell") +
  geom_point(
    aes(x = longitude, y = latitude), 
    color = "firebrick3", 
    size = 1, 
    alpha = 0.2
  ) +
  geom_sf(data = aus_places, color = "cornflowerblue", alpha = 0.8) +
  geom_sf_text(
    aes(label = NAME),
    size = 3,
    data = aus_places,
    nudge_x = 0.5,
    hjust = 0
  ) +
  theme_void() +
  labs(
    title = "Location of Active Fires (1 km estimations)",
    subtitle = "Points represent any detection of fire, not actual area burned",
    caption = "Source: NASA MODIS"
  )
```

Many of these active fires have been concentrated in New South Wales - one of the most populated parts of Austrailia. 

```{r}
max_temp_aus_only <-
  st_intersection(
    x = max_temp %>% st_as_sf(coords = c("long", "lat"), crs = 4326) ,
    y = australia
  )
```

```{r}
max_temp_aus_only %>% 
  transmute(
    max_temp = pmin(45, 3 * round(max_temp / 3)),
    long = st_coordinates(geometry) %>% as_tibble() %>% pull(X),
    lat = st_coordinates(geometry) %>% as_tibble() %>% pull(Y)
  ) %>% 
  ggplot() +
  geom_raster(aes(x = long, y = lat, fill = factor(max_temp))) +
  geom_sf(data = australia, size = 0.1, fill = NA) +
  geom_sf(data = aus_places, color = "white") +
  geom_sf_label(
    aes(label = NAME),
    size = 2,
    data = aus_places
  ) +
  scale_fill_manual(
    values = colors,
    breaks = seq(18, 45, 3)
  ) +
  guides(
    fill = 
      guide_legend(
        title.position = "top", 
        title.hjust = 0.5, 
        label.position = "bottom", 
        nrow = 1)
  ) +
  theme_void() +
  theme(
    legend.position = "bottom",
    legend.direction = "horizontal"
  ) +
  labs(
    title = "Mean maximum temperature",
    subtitle = "Dec 29 2019",
    fill = "Mean maximum temperature (°C)", 
    caption = "Source: BOM Austrailia, Natural Earth"
  )
```

Australia surpassed its all-time temperature record twice in December: An average maximum of 40.9°C on December 17 and 41.9°C on December 18. As the temperature continues to rise and rainfall becomes less frequent, the elevated fire risk in Australia will continue.