BP Statistical Review of World Energy visualisation in R

energy_cons_blog.R

library(tidyverse)
library(scales)

# Read in data and clean (column) names
# Source: https://www.bp.com/en/global/corporate/energy-economics/statistical-review-of-world-energy/downloads.html
# Below assumes you save the CSV in a folder called DATA
bp_wide <- read_csv("DATA/bp-stats-review-2022-consolidated-dataset-panel-format.csv",
                    name_repair = ~ janitor::make_clean_names(., case = "snake")) 

bp_wide %>% glimpse()

bp_wide <- bp_wide %>% mutate(country = recode(country, "US" = "United States")) 

# Add consumption variables in TWhs by multiplying be constant.
# This step is not necessary (data can be viewed in EJ) just a personal preference.
bp_wide <- bp_wide %>%
  mutate(
    hydrocons_twh    = hydro_ej * 277.78,
    nuclearcons_twh  = nuclear_ej * 277.78,
    solarcons_twh    = solar_ej * 277.78,
    windcons_twh     = wind_ej * 277.78,
    gascons_twh      = gascons_ej * 277.78,
    coalcons_twh     = coalcons_ej * 277.78,
    oilcons_twh      = oilcons_ej * 277.78,
    biogeocons_twh   = biogeo_ej * 277.78)

# Setup custom colours - create your own or use a predefined palette.
my_colours <- c("Hydro"   = "#56B4E9",
                "Nuclear" = "#009E73",
                "Solar"   = "gold2",
                "Wind"    = "#CC79A7",
                "Gas"     = "sandybrown", 
                "Coal"    = "#8B5A2B",
                "Oil"     = "#36648B",
                "Geo, Bio, Oth" = "olivedrab4")

# Check colours look reasonable
scales::show_col(my_colours)

# Set theme
theme_set(theme_minimal())

# Review - easier with glimpse() when there are many columns
bp_wide %>% glimpse()

cons_lng <- bp_wide %>%
  # Remove existing aggregated values to avoid double-counting
  filter(!grepl(country, pattern = "Total")) %>% 
  
  select(country, year, 
         Hydro           = hydrocons_twh,
         Nuclear         = nuclearcons_twh,
         Solar           = solarcons_twh,
         Wind            = windcons_twh,
         Gas             = gascons_twh,
         Coal            = coalcons_twh,
         Oil             = oilcons_twh,
         "Geo, Bio, Oth" =  biogeocons_twh) %>%
  pivot_longer(
    cols = !contains(c("country", "year")), 
    names_to = "technology",
    values_to = "value",
    # The below will drop rows that contain only NAs in the value_to column
    # which is what we want.
    values_drop_na = TRUE)

# Create two aggregated tibbles with consumption & percentage variables
# Create global tibble
cons_lng_global <- cons_lng %>%
  filter(year >= 1965) %>%
  
  group_by(year, technology) %>%
  summarise(annual_cons = sum(value)) %>%
  mutate(percentage = annual_cons / sum(annual_cons),
         tech_ord = factor(technology, levels=c("Coal", "Oil", "Gas",
                                                "Geo, Bio, Oth", "Hydro",
                                                "Nuclear", "Solar", "Wind"))) %>% 
  ungroup()

# Create country level tibble
cons_lng_country <- cons_lng %>% 
  filter(year >= 1965) %>% 
  group_by(year, country, technology) %>%
  summarise(annual_cons = sum(value)) %>%
  mutate(percentage = annual_cons / sum(annual_cons),
         tech_ord = factor(technology, levels=c("Coal", "Oil", "Gas",
                                                "Geo, Bio, Oth", "Hydro",
                                                "Nuclear", "Solar", "Wind"))) %>% 
  ungroup()

# Review
glimpse(cons_lng_global)
glimpse(cons_lng_country)

# Area global plot global
cons_lng_global %>%
  ggplot(aes(x = year, y = annual_cons, fill = tech_ord)) +
  geom_area(alpha = 0.6 , size = 0.3, colour = "white") +
  scale_fill_manual(values = my_colours) +
  scale_x_continuous(breaks = seq(1965, 2021, 2), expand = c(0,0)) +
  scale_y_continuous(labels = scales::comma, expand = c(0,0)) +
  labs(title = "Primary Energy Consumption by Technology",
       y = "Primary Energy Consumption (TWh)", x = "Year", fill = "Technology",
       caption = "Source: BP Statistical Review of World Energy June 2022") +
  theme(plot.title = element_text(size = 12, face = "bold"),
        axis.title = element_text(size = 11), legend.position = "right",
        axis.text.x = element_text(angle = 45, size = 8, hjust = 1),
        plot.caption = element_text(size = 8, color = "gray50", face = "italic"),
        plot.background = element_rect(fill = 'antiquewhite', colour = 'antiquewhite'),
        panel.background = element_rect(fill = 'snow'),
        legend.background = element_rect(fill = 'antiquewhite', colour = 'antiquewhite'),
        legend.title = element_text(size = 9))

ggsave("energy_consumption_by_tech_global.png", path = "PLOTS", width = 10, height = 7)

# Area plot global percentage
cons_lng_global %>% 
  ggplot(aes(x = year, y = percentage, fill = tech_ord)) +
  geom_area(alpha = 0.6 , size = 0.3, colour = "white") +
 cale_fill_manual(values = my_colours) +
  scale_x_continuous(breaks = seq(1965, 2021, 2), expand = c(0,0)) +
 scale_y_continuous(labels = scales::percent, breaks = seq(0, 1, 0.1), expand = c(0,0)) +
  labs(title = "Primary Energy Consumption by Technology",
       y = "Energy Consumption (%)", x = "Year", fill = "Technology",
       caption = "Source: BP Statistical Review of World Energy June 2022") +
  theme(plot.title = element_text(size = 12, face = "bold"),
        axis.title = element_text(size = 11), legend.position = "right",
        axis.text.x = element_text(angle = 45, size = 8, hjust = 1),
        plot.caption = element_text(size = 8, color = "gray50", face = "italic"),
        plot.background = element_rect(fill = 'antiquewhite', colour = 'antiquewhite'),
        panel.background = element_rect(fill = 'snow'),
        legend.background = element_rect(fill = 'antiquewhite', colour = 'antiquewhite'),
        legend.title = element_text(size = 9))

ggsave("energy_consumption_perc_by_tech_global.png", path = "PLOTS", width = 9, height = 6)


# Facet 4 countries of interest
# Area plot using geom_area
cons_lng_country %>%
  filter(country %in% c("Italy", "Spain", "Belgium", "Germany")) %>% 
  ggplot(aes(x = year, y = annual_cons, fill = tech_ord)) +
  geom_area(alpha = 0.6 , size = 0.3, colour = "white") +
  scale_fill_manual(values = my_colours) +
  scale_x_continuous(breaks = seq(1965, 2021, 2), expand = c(0,0)) +
  scale_y_continuous(labels = scales::comma, expand = c(0,0)) +
  facet_wrap(~ country, scales = "free_y", ncol = 2) +
  labs(title = "Primary Energy Consumption by Technology",
       # subtitle = "United Kingdom",
       y = "Energy Consumption (TWh)", x = "Year", fill = "Technology",
       caption = "Source: BP Statistical Review of World Energy June 2022") +
  theme(plot.title = element_text(size = 12, face = "bold"),
        axis.title = element_text(size = 10), legend.position = "right",
        axis.text.x = element_text(angle = 45, size = 8, hjust = 1),
        strip.text.x = element_text(size = 10, colour = "darkgreen", face = "bold"),
        plot.caption = element_text(size = 8, color = "gray50", face = "italic"),
        plot.background = element_rect(fill = 'antiquewhite', colour = 'antiquewhite'),
        panel.background = element_rect(fill = 'snow'),
        legend.background = element_rect(fill = 'antiquewhite', colour = 'antiquewhite'),
        legend.title = element_text(size = 9), legend.text = element_text(size = 9))

ggsave("energy_consumption_by_tech_facet.png", path = "PLOTS", width = 14, height = 8)