Analysis of Worldwide Carbon Emissions

Emma Schou (Data Science at Reed College)https://reed-statistics.github.io/math241-spring2022/ , Aaron Berlau (Data Science at Reed College)https://reed-statistics.github.io/math241-spring2022/ , Tess Rutstein (Data Science at Reed College)https://reed-statistics.github.io/math241-spring2022/
May 5, 2022

Introduction

Climate change is an issue that grows in relevance everyday. This being said, the issue has not been given enough attention to what it deserves. Climate change, as it occurs through different chemical emissions being let into our atmosphere, has decades of science and data behind it. For our final project, we wanted to show just how emissions rates have grown or changed throughout the countries of the world. Through looking at the countries that output the most emissions, we can try and see what influences this output. A few factors we are looking at are the population of the country and the number of billionaires in the country. To do this, we are using three different data sets. The first data set is our Emissions.csv data set (Dennis Kafura, 2019), which has information on countries and levels of different emissions from 1970-2012. We will also be using the global_development.cvs dataset, which has data from 1980-2013 for each country in a variety of different categories. For the purpose of this project, this is where we are drawing our population data from. The final data set we are using is the billionaires.cvs (Billionaires CSV File, 2016) data set, which gives in depth information on the top billionaires from three different points within the last 40 years. Our interest in studying the climate crisis is driven by an awareness of how significant and catastrophic climate change is, and a hope that understanding the relationship between population size and emissions might shed light on some of the variables driving the climate crisis.

hide 
#load libraries
library(tidyverse)
library(ggraph)
library(sf)
library(ggthemes)
library(plotly)
library(gganimate)
library(gifski)
library(transformr)
library(png)
library(purrr)
library(maps)
library(tmap)
library(reshape2)
hide 
#load CSVs
emissions<-read_csv("emissions.csv")
billionaires <- read_csv("billionaires.csv")
glob_dev <- read.csv("global_development.csv")

Methods and Results

Our key objective in this study is to increase understanding of the factors that increase or reduce a country’s annual emissions. To complete this objective, we used the R package “tmap” and the Emissions csv to create visualizations of the countries that have the highest emissions in each category from year to year. Tmap has shapefiles and data for every country in the world, which we joined with the emissions data. We used tidyverse to get many of the data processing functions for the visualization and to do exploratory analysis and data wrangling. We then used “plotly”, “gganimate”, “gifski”, “transfomr”, and “png” to make the maps animated so that one could visualize the changes in emissions levels for each country per year.

Our maps demonstrate that emissions have risen dramatically since the emissions dataset was first started. Figure 1 shows the increase in CO2 emissions from 1980-2012, and, since we know the warming impact that CO2 has on the environment, this provides a grim projection for the future climate since the rise in CO2 emissions does not seem to be slowing. Other greenhouse gases (N2O and CH4 have also been increasing. Though they are generally released in lower quantities (when measured in tons), the standard metric for these gases is their warming potential when compared to CO2. For example, 1 unit of N2O released into the atmosphere is the equivalent of about 300 units of CO2 released. Figure 2 shows the amount of gases released in the 1980-2012 time period when the N2O and CH4 warming potentials are adjusted to CO2. We know that CO2 is rising rapidly (Fig 1) and Figure 2 demonstrates that N2O and CH4 also play a significant part in that warming. When looking at the breakdown of greenhouse gas emissions by country (Figure 3), we see that the United States and China, followed by Russia and India, appear to have the highest emissions, and are trailed by almost every other country in the world. Figure 4 also shows us that when we look at where most of the billionaires are located, in the most recent recorded year in the figure (2014), most of them claim China and the United States as their country of origin. We can’t claim a causal relationship between the presence of billionaires in a country and their level of greenhouse gas emissions due to having too few data points, however, this could be an area for further study since many of the economic systems that create billionaires are the same ones that encourage climate destruction.

hide 
##DATA WRANGLING
#importing world data
data("World")
st_crs(World) <- 4326

#Sorting Global Development dataset
globdev2 <- glob_dev%>%
  select(Country, Year, Data.Health.Total.Population)

#creating a more linear scale for population
World2<-World%>%
  mutate(logpop=log10(pop_est))

#combining Billionairs and World datasets
worldbillionaires <- billionaires %>%
  filter(!year > 2012) %>%
  left_join(World2, by = c("location.citizenship"="name"))

#joining emissions and world
worldemissions<-World2 %>%
 left_join(emissions, by=c("name"="Country"))

#joining emissions and global development
worldemissions2 <- worldemissions %>%
  left_join(globdev2, by = c("name"="Country", "Year"="Year")) %>%
  filter(!Year < 1980) %>%
  filter(!Year > 2013)
hide 
#CO2
worldemissionsco2<-aggregate(Emissions.Type.CO2 ~ Year, worldemissions2, sum)

#N2O
worldemissionsn2o<-aggregate(Emissions.Type.N2O ~ Year, worldemissions2, sum)

#CH4
worldemissionsch4<-aggregate(Emissions.Type.CH4 ~ Year, worldemissions2, sum)

#joining
worldemissionsall<-worldemissionsco2%>%
    full_join(worldemissionsn2o, by = c("Year"="Year"))

worldemissionsall<-worldemissionsall%>%
    full_join(worldemissionsch4, by = c("Year"="Year"))

worldemissionsall <- melt(worldemissionsall ,  id.vars = 'Year', variable.name = 'series')

Figures

hide 
#co2
worldemissionsco2%>%
  ggplot(aes(x=Year, y=Emissions.Type.CO2))+
  geom_point()+
  geom_line()+
  labs(title="World emissions of CO2 over time", y="Kilotons CO2")+
  theme_bw()
Kilotons of carbon dioxide released across the world between 1980 and 2012.

Figure 1: Kilotons of carbon dioxide released across the world between 1980 and 2012.

hide 
#all
worldemissionsall%>%
ggplot(aes(x=Year, y=value, color=series))+
  geom_point()+
  geom_line()+
  scale_color_colorblind(name = "Greenhouse Gas", labels = c("CO2", "N2O", "CH4"))+
  labs(title="World emissions over time by greenhouse gas type", y="Amount in kilotons CO2")+
  theme_bw()
Emissions of greenhouse gases (Carbon Dioxide, Nitrous Oxide, and Methane) between years 1980 and 2012. Amounts of N2O and CH4 are converted into kilotons of CO2 because this shows a consistent unit measurement for the atmostpheric impact of these emissions.

Figure 2: Emissions of greenhouse gases (Carbon Dioxide, Nitrous Oxide, and Methane) between years 1980 and 2012. Amounts of N2O and CH4 are converted into kilotons of CO2 because this shows a consistent unit measurement for the atmostpheric impact of these emissions.

hide 
#maps
#CO2 Emissions
mco2 <- tm_shape(worldemissions2) +
  tm_polygons("Emissions.Type.CO2") +
  tm_facets(along = "Year")

tmap_animation(mco2, delay=40, filename = "mco2.gif")
hide 
knitr::include_graphics("mco2.gif")
Maps of the greenhouse gas emissions around the world over time

Figure 3: Maps of the greenhouse gas emissions around the world over time

hide 
#N2O emissions
mn2o <- tm_shape(worldemissions2) +
  tm_polygons("Emissions.Type.N2O") +
  tm_facets(along = "Year")

tmap_animation(mn2o, delay=40, filename = "mn2o.gif")
hide 
knitr::include_graphics("mn2o.gif")
Maps of the greenhouse gas emissions around the world over time

Figure 4: Maps of the greenhouse gas emissions around the world over time

hide 
#CH4 emissions
mch4 <- tm_shape(worldemissions2) +
  tm_polygons("Emissions.Type.CH4") +
  tm_facets(along = "Year")

tmap_animation(mch4, delay=40, filename = "mch4.gif")
hide 
knitr::include_graphics("mch4.gif")
Maps of the greenhouse gas emissions around the world over time

Figure 5: Maps of the greenhouse gas emissions around the world over time

hide 
#making years in billionaires dataset categorical
bill<-billionaires%>%
  mutate(year=case_when(year==1996~"1996",
                        year==2001~"2001",
                        year==2014~"2014"))%>%
#removing outliers  
  filter(demographics.age>0,
         company.founded>=1000)

#graphs
  ggplot(data=bill, aes(y=fct_infreq(location.citizenship), fill=year))+
  geom_bar()+
    facet_wrap(~year, scales="free")+
    theme(axis.text = element_text(size = 5), legend.position = "bottom")+
    theme_bw()+
    labs(title="Number of billionaires by country of origin", x="Number of billionaires", y="Country of Origin", fill="Year")
Graph of the countries that have the most billionaires, measured in 1996, 2001, and 2014

Figure 6: Graph of the countries that have the most billionaires, measured in 1996, 2001, and 2014

Conclusions

Through our visualizations, we see that emissions have spiked dramatically since the beginning of the 21st century. Looking at specific emission data, you can see that CH4 poses the most danger to the environment. It is the emission with both the highest output level throughout our range of years as well as the main cause of the spike in emissions from the onset of the 21st century. Looking at the countries through our range of years, we can see that the US and China are the biggest contributors to the output of the various emissions. Other countries with large populations such as Brazil, Russia. and India start to increase the rate of their emissions as the graph goes on. These graphs show the direness of the climate crisis because, as more greenhouse gases are emitted and the climate continues to warms, there is more and more that people will have to do to recapture those emissions and avoid irreparable atmospheric damage. Though we cannot prove that the number of billionaires are necessarily related to the increased pollution of the atmosphere, we can say that some of the biggest polluters also have a large number of billionaires, and that many billionaires have made their money, at least partly, through climate endangerment. We believe that investigating the relationship between concentrated wealth and climate degradation would be a good direction to continue this research.

Class Peer Reviews

Billionaires CSV File. (2016). Billionaires CSV file. CORGIS Dataset Project. https://corgis-edu.github.io/corgis/csv/billionaires/
Dennis Kafura. (2019). Emissions CSV file. CORGIS Dataset Project. https://corgis-edu.github.io/corgis/csv/emissions/

References

Reuse

Text and figures are licensed under Creative Commons Attribution CC BY 4.0. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".