Global CO₂ Emissions Analysis

A comprehensive R-based analysis that fetches and processes global CO₂ emissions data (per-capita and total) from Our World in Data (OWID), generates a suite of 12 exploratory and analytical plots, and performs a regression of per-capita emissions on GDP per capita.

Table of Contents

1. Project Overview
2. Features & Charts
3. Prerequisites
4. Installation
5. Usage
6. Script Breakdown
7. Interpreting the Outputs
8. Extending & Customizing
9. Data Source & Citations
10. License

---

Project Overview

This repository contains:

• CO2_Analysis.R: A self-contained R script that

  1. Downloads the OWID CO₂ dataset (yearly, since 1750)
  2. Cleans & prepares: CO₂ per capita, total CO₂, GDP per capita, date conversion
  3. Produces 12 visualizations exploring temporal trends, economic relationships, top emitters, quartile analyses, cumulative sums, and year-over-year changes
  4. Fits a linear regression model: CO₂ per capita ~ GDP per capita, with summary statistics and an annotated plot

• report.Rmd: An R Markdown report that weaves narrative, code, and plots into a shareable HTML document.

• scripts/, Makefile, Dockerfile, docker-compose.yml, and .env.example to automate data fetching, analysis, report building, and containerized execution.

---

Features & Charts

1. Global Trend
   • Line plot of global average CO₂ per capita over time
2. GDP vs. CO₂
   • Scatter of GDP per capita vs. CO₂ per capita (latest) with LOESS smoothing
3. Regression Analysis
   • Linear model summary printed to console
   • Scatter + regression line with R² and p-value annotation
4. Top 10 Emitters (per capita)
   • Horizontal bar chart of the 10 highest per-capita emitters
5. Top 5 Time Series
   • Line plots of CO₂ per capita over time for the top 5 countries
6. Quartile Boxplot
   • CO₂ per capita by GDP per capita quartile (boxplot)
7. Quartile Violin
   • Distribution of CO₂ per capita by GDP quartile (violin plot)
8. Quartile Heatmap
   • Heatmap of average CO₂ per capita by year & GDP quartile
9. Cumulative Sum
   • Cumulative sum of per-capita CO₂ over time
10. Year-over-Year Change
    • Line plot of YOY % change in global average CO₂ per capita
11. Total vs. Per Capita
    • Scatter of total CO₂ emissions vs. per-capita emissions (log scale)
12. Global Time Series
    • (Already #1) — the core global trend

---

Prerequisites

• R ≥ 4.0
• Internet access to fetch OWID data

R Packages

• ggplot2, dplyr, lubridate, tidyr, forcats, scales, viridis, zoo, corrplot

The main script auto-installs any missing packages.

---

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/co2-analysis.git
   cd co2-analysis

2. Optional: copy .env.example → .env to set DATA_DIR if you prefer a custom data folder.

---

Usage

From R or RStudio

# Set working directory to the project root
setwd("path/to/co2-analysis")

# Source the analysis script
source("CO2_Analysis.R")

All charts will render in sequence and the regression summary will print to the console.

Via Make

make report

This will:

1. Fetch the CO₂ CSV into data/
2. Run CO2_Analysis.R
3. Render CO2_Analysis.Rmd → CO2_report.html

Via Docker

docker-compose up --build

This runs the entire pipeline in a reproducible container.

---

Script Breakdown

1. Setup

   • Defines package list, auto-installs, and loads libraries.

2. Data Fetch & Prep

   • Downloads OWID CSV, filters to years ≥1960, computes co2_pc, total_co2, gdp_pc.
   • Derives date from year.

3. Visualization Sections

   • Global trend, GDP relationship, top emitter charts, quartile analyses, cumulative sums, YOY change, total vs. per-capita.

4. Regression

   • Fits lm(co2_pc ~ gdp_pc), prints summary(), and generates an annotated plot.

---

Interpreting the Outputs

• Global Average: captures long-term trends in per-capita emissions.
• GDP vs CO₂: shows the positive correlation between wealth and emissions.
• Top Emitters: highlights the countries with the highest per-capita footprint.
• Quartile Plots: reveal distributional differences across income strata.
• Cumulative & YOY: expose acceleration or deceleration in emission growth.
• Regression: quantifies how much GDP per capita explains variation in emissions.

---

Extending & Customizing

• Adjust filters (e.g., include earlier years).
• Swap color palettes or themes.
• Add new predictors (population density, energy mix).
• Save plots by adding ggsave() calls.
• Integrate regional maps using sf and rnaturalearth.

---

Data Source & Citations

• Our World in Data CO₂ dataset: https://github.com/owid/co2-data
• R & Packages: R Core Team (2023). R: A language and environment for statistical computing.
• Visualization: Wickham H. et al. “ggplot2: Elegant Graphics for Data Analysis.”

---

License

This project is released under the MIT License. See LICENSE for details.