Tracking Greenhouse Gas Emissions in the U.S.

Maps, Visualization


It’s no secret that greenhouse gases contribute largely to the ongoing climate crisis in not only the United States, but the entire world. These emissions manifest in a number of different ways that cause damage to the Earth’s atmosphere. My intention for this project was to discover which industries and processes are the worst contributors to greenhouse gas emissions, and to break down the emissions data by state to compare regulations for each. In doing so, my hope is to provide visual context for which areas need to be highlighted to help curb emissions and preserve the integrity of the Earth’s ozone layer.


I began my research with transportation data in mind, hoping to find geospatial data for public transportation which I could map against emissions produced by cars. To my surprise, I came across a dataset from the World Resources Institute that included not only CO2 emissions data but all other types of greenhouse gas emissions as well. This led me to consider opening up my research to compare other types of emissions by state as well, and to find more data sources that pertain to certain emissions. From there, I conducted background research to understand how each greenhouse gas is produced as well as the severity level that each contributes to climate change. I decided to focus on three specific greenhouse gases for my visualizations: Carbon Dioxide (CO2), Methane (CH4), and Nitrous Oxide (N2O).

Tools and Data

For this unit, I used QGIS to map geospatial data layers and join other forms of data for plotting. QGIS is an open source geographic information system that allows users create maps using shape files, which contain the code needed to render layers that display location data. I first downloaded a base layer map from the U.S. Census Bureau’s open data portal in order to build any visualizations on top of it. Once I added the map layer to QGIS, I searched for state-level data on greenhouse gas emissions to add to my visualization. To my surprise, it was fairly difficult to find shape files for greenhouse gas emissions, so I settled on a CSV dataset which I joined with the data from the base layer map. This allowed me to create a graduated color ramp for my base map, which displayed darker colors for states with higher emission rates. I realized that the visualization this produced reflected that of a population map, so I uploaded the emissions data to OpenRefine to normalize the numbers based on each state’s population, resulting in a more informative map.

Total Greenhouse Gas Emissions by State (2020)

Once I was a bit more acclimated to using QGIS, I began searching for other types of data that directly contribute to each type of greenhouse gas emissions. After downloading public transportation records, agricultural emissions data, and location data for oil and natural gas wells, I began arranging my map layers to create more robust visualizations.

Results and Findings

The first greenhouse gas I began to analyze was carbon dioxide, which had the highest output per individuals by far. One of the main contributors to CO2 emissions doesn’t come from natural human processes, but from burning fossil fuels to power machines, most commonly automobiles. Because of this fact, I figured there may be a correlation between states with better access to public transportation and a reduced CO2 output. The results showed that this correlation was true, and that states with more public transportation tend to rely less on cars, which significantly reduces CO2 output.

Public Transportation Stops and Total CO2 Transportation Emissions

The next greenhouse gas I focused on was methane, or CH4. Although CH4 only makes up around 10% of the U.S.’s total greenhouse gas emissions, it is considered one of the more harmful gases for our atmosphere. The two areas that contribute to CH4 emissions come from livestock production and oil fracking. Because I wasn’t able to find any geographic livestock data, I chose to plot the locations of oil and natural gas wells against the total CH4 emissions produced in each state. The results also showed a correlation between drilling and CH4 output, and that states with high volumes of natural gas wells tend to produce up to 100x more methane than those without any wells.

Total CH4 Production and Natural Gas/Oil Wells

Lastly, I wanted to focus on the volume of nitrous oxide output due to the nuances of its origin. Nitrous oxide, or N2O, is generated from a variety of sources, largely associated with agriculture. Because animal waste and fertilizer are so high in N2O, emissions are most commonly attributed to those factors. To my luck, I was able to find a geospatial file that contains total emissions from crops that I simply layered on top of my base map. Again, the correlation between the crop emissions and N2O emissions specifically was present, and could help states create regulations to better control N2O emissions.

Total Crop-Specific Emissions and N2O emissions by state


Overall, I found my experience working with QGIS to be a bit tedious, but I can understand how this is a useful tool given the right data. Because I had such a difficult time finding map layer data, I feel as though my insights were slightly stunted. Although state-level data is still helpful context, it would be even more interesting to be able to visualize emissions data at a lower level, potentially by county or zip code. I also ran into some barriers to entry regarding mapping conventions simply because I wasn’t sure which projections to use. This also led to some anomalies and I was unable to include data from non-mainland states in the U.S., which was unfortunate. However, given more time to become familiar with QGIS and sources for data types supported by QGIS, I am more than confident that it would be incredibly useful for future maps I would like to create. For future work on my topic, I think it would be interesting to compare these emissions to past years in a more interactive way, to see if specific states in the U.S. are trending toward either a better or worse direction for greenhouse gas production.