Tidy Tuesday: Energy (2023 - Week 23)

Introduction

In this document I’ll analyze and visualize some energy data that were the focus of Tidy Tuesday 2023 week 23. The data comes from Our World In Data and the full data set is available here. Data Source Citation: Ritchie, Roser, and Rosado (2022).

Analysis and Visualization

I’ll start by loading the tidyverse (Wickham et al. (2019)) library and the data set. The result is a dataframe with a row for each country and year, from 1900-2002.

Code

suppressPackageStartupMessages(library(tidyverse))
suppressPackageStartupMessages(library(plotly))
ggplot2::theme_set(theme_grey(base_size = 15))

owid_energy <- readr::read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2023/2023-06-06/owid-energy.csv", show_col_types = FALSE)

head(owid_energy)

# A tibble: 6 × 129
  country      year iso_code population   gdp biofuel_cons_change_pct
  <chr>       <dbl> <chr>         <dbl> <dbl>                   <dbl>
1 Afghanistan  1900 AFG         4832414    NA                      NA
2 Afghanistan  1901 AFG         4879685    NA                      NA
3 Afghanistan  1902 AFG         4935122    NA                      NA
4 Afghanistan  1903 AFG         4998861    NA                      NA
5 Afghanistan  1904 AFG         5063419    NA                      NA
6 Afghanistan  1905 AFG         5128808    NA                      NA
# ℹ 123 more variables: biofuel_cons_change_twh <dbl>,
#   biofuel_cons_per_capita <dbl>, biofuel_consumption <dbl>,
#   biofuel_elec_per_capita <dbl>, biofuel_electricity <dbl>,
#   biofuel_share_elec <dbl>, biofuel_share_energy <dbl>,
#   carbon_intensity_elec <dbl>, coal_cons_change_pct <dbl>,
#   coal_cons_change_twh <dbl>, coal_cons_per_capita <dbl>,
#   coal_consumption <dbl>, coal_elec_per_capita <dbl>, …

How many countries are in the dataset?

Code

length(unique(owid_energy$country))

[1] 306

That’s a lot! I’ll focus on just the United States for now.

I also noticed that the data set goes back to 1900 but a a lot of the data for earlier years are missing/NA so I’ll filter those out as well.
It looks like we have data for the USA from 2000-2021.

Make a new dataframe for just the USA data and remove years without data:

Code

usa <- owid_energy %>%
  filter(country == "United States") %>%
  filter(!is.na(electricity_demand))

usa |>
  DT::datatable(rownames = FALSE, options = list(pageLength = 5))

Share of electricity generation by fuel type

For this analysis, I’ve chosen to investigate how the mix of fuel types used to generate electricity has changed over time. We need to reduce carbon emissions in order to prevent or mitigate the effects of climate change, and electricity generation is a large component of these emissions. I’m interested to see what progress has been made in transitioning to more renewable/low-carbon fuels for electricity generation.

Conveniently, the data already contain fields for the share of total electricity generation for each fuel type! I’ll make a new data frame with just these fields. I can select these columns (all ending in share_elec), using the ends_with function from the dplyr (Wickham et al. (2023)) package.

Code

usa %>%
  select(year, dplyr::ends_with("share_elec")) %>%
  DT::datatable(rownames = FALSE, options = list(pageLength = 5))

Table 1: Percent of electricity generation by fueltype

Now I have a dataframe (Table 1) with just the variables I want to plot, but the format isn’t ideal; I would have to specify/plot each variable separately. What I want to do is give a single plot command and group/color the lines by the variable (fuel type). To achieve this, I am going to pivot the data frame from wide to long format, using the pivot_longer function from the tidyr (Wickham, Vaughan, and Girlich (2023)) package.

Note

The pivot_ functions in tidyr (Wickham, Vaughan, and Girlich (2023)) package were previously referred to as gather or melt.

Code

usa_share <- usa %>%
  select(year, ends_with("share_elec")) %>%
  tidyr::pivot_longer(
    cols = ends_with("share_elec"),
    names_to = "FuelType",
    values_to = "Percentage") |>
  mutate(FuelType = str_remove(FuelType,'_share_elec')) 

usa_share |>
    DT::datatable(rownames = FALSE, options = list(pageLength = 5))

Table 2: Percent of electricity generation by fueltype

Now my dataframe (Table 2) has a row for each year, fuel type, and value, and I can simply group or color by the fuel type when I plot.

Code

g <- usa_share %>%
  ggplot(aes(year, Percentage)) +
  geom_line(aes(color = FuelType), linewidth = 1.5) +
  ggtitle("Percent of US electricity Generation By Fuel type") +
  xlab("Year") +
  ylab("Percent")

plotly::ggplotly(g)

Here we finally have a plot (Figure 1) of the share of electricity generation by fuel type. We can see that the share of fossil fuels and coal has decreased, and renewable have increased. But there’s a lot on this plot and it’s hard to read, so I’ll focus on some more specific subsets of the data.

Total Fossil, renewables, and nuclear shares

First we can look at the total shares of fossil (oil,coal, gas), renewable (wind, solar, hydro), and nuclear generation. Grouping into these categories de-clutters the plot and makes it easier to interpret.

Code

g <- usa %>%
  select(year, fossil_share_elec, renewables_share_elec, nuclear_share_elec) %>%
  tidyr::pivot_longer(
    cols = dplyr::ends_with("share_elec"),
    names_to = "FuelType",
    values_to = "Percentage"
  ) %>%
  mutate(FuelType = str_remove(FuelType,'_share_elec')) |>
  ggplot(aes(year, Percentage)) +
  geom_line(aes(color = FuelType), linewidth = 1.5) +
  ggtitle("Percent of US electricity Generation") +
  xlab("Year") +
  ylab("Percent")

plotly::ggplotly(g)

Observations from this plot (Figure 2):

Fossil fuel share has been decreasing steadily since about 2007
Renewable share has been increasing steadily since about 2007
Nuclear has remained relatively constant at around 20%.
Fossil share remains the majority of generation, but is decreasing. Renewables became approximately equal to nuclear around 2020 and are continuing to increase.

Breakdown of fossil fuel shares

In this dataset, fossil fuels include coal, gas, and oil.

Code

g <- usa %>%
  select(year, fossil_share_elec, oil_share_elec, gas_share_elec, coal_share_elec) %>%
  tidyr::pivot_longer(
    cols = dplyr::ends_with("share_elec"),
    names_to = "FuelType",
    values_to = "Percentage"
  ) %>%
  mutate(FuelType = str_remove(FuelType,'_share_elec')) |>
  ggplot(aes(year, Percentage)) +
  geom_line(aes(color = FuelType), linewidth = 1.5) +
  ggtitle("Percent of US electricity Generation: Fossil Fuels") +
  xlab("Year") +
  ylab("Percent")

plotly::ggplotly(g)

Observations from this plot (Figure 3):

We can see that the fossil fuel share of electricity generation has been decreasing, starting around 2008.
Coal and gas make up the majority of the fossil fuel generation.
Coal share has been decreasing while the gas share has increased. Coal was much higher than gas previously, but their shares became equal around 2015 and gas now makes up a larger share of the fossil fuel generation.

Renewables breakdown

In this dataset, renewables include wind, solar, and hydro.

Code

g <- usa %>%
  select(
    year, renewables_share_elec, solar_share_elec,
    hydro_share_elec, wind_share_elec
  ) %>%
  tidyr::pivot_longer(
    cols = dplyr::ends_with("share_elec"),
    names_to = "FuelType",
    values_to = "Percentage"
  ) %>%
  mutate(FuelType = str_remove(FuelType,'_share_elec')) |>
  ggplot(aes(year, Percentage)) +
  geom_line(aes(color = FuelType), linewidth = 1.5) +
  ggtitle("Percent of US electricity Generation: Renewables") +
  xlab("Year") +
  ylab("Percent")


plotly::ggplotly(g)

Observations from this plot (Figure 4):

The share of renewable electricity production has increased sharply, approximately doubling from 2008 to 2020.
The share of hydro generation has remained relatively constant.
Solar and wind shares have increased significantly.
- Wind started to increase earlier, around 2005.
- Solar started increasing around 2012

SessionInfo

Code

sessionInfo()

R version 4.3.1 (2023-06-16)
Platform: x86_64-apple-darwin20 (64-bit)
Running under: macOS Sonoma 14.2

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/4.3-x86_64/Resources/lib/libRblas.0.dylib 
LAPACK: /Library/Frameworks/R.framework/Versions/4.3-x86_64/Resources/lib/libRlapack.dylib;  LAPACK version 3.11.0

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

time zone: America/Denver
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices datasets  utils     methods   base     

other attached packages:
 [1] plotly_4.10.3   lubridate_1.9.3 forcats_1.0.0   stringr_1.5.0  
 [5] dplyr_1.1.3     purrr_1.0.2     readr_2.1.4     tidyr_1.3.0    
 [9] tibble_3.2.1    ggplot2_3.4.4   tidyverse_2.0.0

loaded via a namespace (and not attached):
 [1] sass_0.4.7        utf8_1.2.4        generics_0.1.3    renv_1.0.3       
 [5] stringi_1.7.12    hms_1.1.3         digest_0.6.33     magrittr_2.0.3   
 [9] evaluate_0.22     grid_4.3.1        timechange_0.2.0  fastmap_1.1.1    
[13] jsonlite_1.8.7    httr_1.4.7        fansi_1.0.5       crosstalk_1.2.0  
[17] viridisLite_0.4.2 scales_1.2.1      jquerylib_0.1.4   lazyeval_0.2.2   
[21] cli_3.6.1         rlang_1.1.1       crayon_1.5.2      ellipsis_0.3.2   
[25] bit64_4.0.5       munsell_0.5.0     cachem_1.0.8      withr_2.5.1      
[29] yaml_2.3.7        parallel_4.3.1    tools_4.3.1       tzdb_0.4.0       
[33] colorspace_2.1-0  DT_0.30           curl_5.1.0        vctrs_0.6.4      
[37] R6_2.5.1          lifecycle_1.0.3   htmlwidgets_1.6.2 bit_4.0.5        
[41] vroom_1.6.4       pkgconfig_2.0.3   bslib_0.5.1       pillar_1.9.0     
[45] gtable_0.3.4      glue_1.6.2        data.table_1.14.8 xfun_0.40        
[49] tidyselect_1.2.0  rstudioapi_0.15.0 knitr_1.44        farver_2.1.1     
[53] htmltools_0.5.6.1 labeling_0.4.3    rmarkdown_2.25    compiler_4.3.1

References

Ritchie, Hannah, Max Roser, and Pablo Rosado. 2022. “Energy.” Our World in Data.

Wickham, Hadley, Mara Averick, Jennifer Bryan, Winston Chang, Lucy D’Agostino McGowan, Romain François, Garrett Grolemund, et al. 2019. “Welcome to the Tidyverse” 4: 1686. https://doi.org/10.21105/joss.01686.

Wickham, Hadley, Romain François, Lionel Henry, Kirill Müller, and Davis Vaughan. 2023. “Dplyr: A Grammar of Data Manipulation.” https://CRAN.R-project.org/package=dplyr.

Wickham, Hadley, Davis Vaughan, and Maximilian Girlich. 2023. “Tidyr: Tidy Messy Data.” https://CRAN.R-project.org/package=tidyr.