3  Results

Code 
#install.packages('naniar');
library(ggplot2);
library(tidyverse);
── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.5
✔ forcats   1.0.0     ✔ stringr   1.5.1
✔ lubridate 1.9.3     ✔ tibble    3.2.1
✔ purrr     1.0.2     ✔ tidyr     1.3.1
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
Code 
library(readxl);
library(naniar);
library(dplyr);
library(forcats);

3.1 Load dataset

Code 
df <- read_excel("CoralBleaching.xlsx")
df <- df |> drop_na(REGION)

3.2 Regional Distributions

Code 
ggplot(data = df, aes(x = fct_infreq(REGION))) +
  geom_bar(fill = "lightblue", color = 'black') +
  labs(y = "Number of Observations", x = "Region", title = "Number of Observations by Region") +
  theme_minimal() +
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.text = element_text(size = 6)
  )

This bar chart demonstrates the total number of coral bleaching observations, all of various bleaching degrees, across six regions. Australia and the Americas have the highest number of observations, both over 2000, due to the presence of extensive coral reef systems (e.g., the Great Barrier Reef) and stronger monitoring efforts. The Pacific and Asia follow with moderate observation counts, reflecting their significant reef areas but potentially less monitoring capacity compared to Australia and the Americas. Africa and the Middle East have the fewest observations, possibly due to smaller reef systems and limited monitoring resources.

Code 
country_counts <- df |>
  filter(BLEACHING_SEVERITY != "Severity Unknown") |>
  group_by(COUNTRY) |>
  summarize(num_observations = n()) |>
  arrange(desc(num_observations))

# Cleveland dot plot
ggplot(country_counts, aes(x = num_observations, y = reorder(COUNTRY, num_observations))) +
  geom_point(color = "lightblue", size = 1.5) +
  geom_segment(aes(xend = 0, yend = COUNTRY), color = "darkgray", size = 0.5) + 
  labs(
    title = "Number of Observations by Country",
    x = "Number of Observations",
    y = "Country"
  ) +
  theme_minimal() +
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.text = element_text(size = 6)
  )
Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
ℹ Please use `linewidth` instead.

This Cleveland dot plot displays the total number of observations for each country. Australia, particularly the Torres Strait & Great Barrier Reef region, dominates the data with the highest number of observations, reflecting its extensive reef system and well-funded research initiatives. Other countries, such as Mexico, Malaysia, and the United States (specifically Florida), also show high observation counts, likely due to their substantial reef coverage and strong conservation efforts. In contrast, many countries have significantly fewer observations, possibly indicating limited monitoring resources or smaller reef areas. The steep drop-off in observation counts after the top few countries highlights a variable global distribution of monitoring efforts.

Code 
df$BLEACHING_SEVERITY <- factor(df$BLEACHING_SEVERITY)

# Violin plot
ggplot(df, aes(x = REGION, y = as.numeric(BLEACHING_SEVERITY), fill = REGION)) + 
  geom_violin(trim = TRUE, color = "black") +
  labs(x = "Region", y= "Bleaching Severity", title = "Violin Plot of Bleaching Severity by Region") + 
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.text = element_text(size = 6)
  )

This violin plot demonstrates the distribution of bleaching severity across different regions, with the width of the violins representing the density of observations for each severity level. Bleaching severity ranges from lower values to higher values, providing insights into the most commonly observed severity levels within each region:

  1. Americas & Australia: These regions exhibit a wide range of bleaching severity and diversity across every level. The Americas has more emphasis on lower levels (level 2) while Australia has higher severity levels (level 4).
  2. Africa & Middle East: Africa shows a concentration of observations at lower severity levels, suggesting less frequent high-severity bleaching. Contrastly, the Middle East has a similar shape, but reversed—emphasizing higher severity levels.
  3. Asia & Pacific: These regions appear to have more uniform severity levels; however, the Pacific differs in that it demonstrates a significant increase in density at severity level 5, the highest of any region.

This visualization highlights the variation in bleaching severity across regions, and suggests that regions like Australia, the Americas, and the Pacific experience a broader spectrum of bleaching severity. This variability potentially reflects the presence of larger and more closely monitored reef systems. In contrast, regions like Africa and the Middle East may require additional monitoring to better understand coral health dynamics.

3.5 Summary

Our analysis reveals several key insights about coral bleaching trends and their global impact. First, regions like Australia and the Americas dominate the number of observations, reflecting strong monitoring capacity, while Africa and the Middle East demonstrate significantly smaller numbers of observations, likely due to limited resources. Spatial distribution maps highlight the Great Barrier Reef as a hotspot for high-severity bleaching events, emphasizing the urgent need for conservation in the region. Temporal trends show a concerning increase in bleaching severity since the 1990s, closely tied to global warming and rising ocean temperatures. Seasonal patterns suggest a connection between bleaching events and warmer months, but variations across regions reflect differences in local climates.