Introduction

I wanted to do my final lab report on a topic that I am keen to discover more about, climate change, with the goal of promoting awareness and hopefully inspiring action – big or small. I approached this lab with an investigative mindset. In a prior lab report I had created visualizations on deforestation and its possible correlation with climate change. I wanted to expand my knowledge and analyze carbon emission contributions globally and by country together with temperature changes.  

Some of the questions I wanted to answer were:

Data & Methodology

I gathered various data sources for this report. I started off first by getting CO2 emissions data. I initially found a dataset that showed total emissions by country, but chose another dataset that had country emissions per capita. I decided to use the latter because it was scaled by population and normalizes the data when comparing across each country’s contribution.

Additionally, I had previously used a global temperature dataset that measures the land and ocean temperature anomalies, using a baseline average from 1901-2000. I also found a more specific dataset with temperature changes on a per country basis. The second temperature dataset measures the mean surface temperature using the years 1951-1980 as a baseline. 

Carbon Dioxide Emissions Data

Temperature data global

Temperature data by country

Continents list (used for vlookup)

I started by downloading the data into Google Sheets. Because I knew that I wanted to break down my analysis via continents, I knew I had to add a “continent” column to the existing worksheet. But before doing this, I had to first tidy the data and transpose the “year” columns into rows. I therefore exported the sheet into XLS and used Open Refine to tidy the data by transposing columns into rows. 

Afterwards, I loaded the data back in Google Sheets and was ready to bring over continent data into the same sheet. I did this by using another data list that contained countries and their continents. I appended continents to my sheet using a VLOOKUP formula. I had some issues when the country names from one data set were labeled differently compared to the other data set (i.e. “Saint Vincent and Grenadines” versus “St. Vincent and Grenadines”). I had to manually change some of the names because they were not matching (country names don’t seem to be standardized).

Once I ran into this issue, I used the ISO code in my VLOOKUP, since it is a standardized country code.

I ended up having to do a similar process with multiple data sets. 

Once I had the data I needed, I exported the Google sheet and loaded it into Tableau to begin creating visualizations and analyzing the data further.

Findings

The map below showcases the data that was analyzed based on a 2019 snapshot. We can visualize the emissions by circle size within each continent and have an understanding of the number of countries we are analyzing per continent.

I first wanted to scope the carbon dioxide emissions by classifying the continents in a time series. I found that the only continent that considerably reduced carbon dioxide emissions in recent years was Europe. The rest of the continents have an upward trend.

There was an interesting spike in North America’s line in 2009 which looked anomalous. I investigated it further and found that the spike comes from the country Saint Vincent and Grenadines, where there was a jump in emissions. I validated by Googling this specific country’s CO2 emissions in 2019 and found another source that had the same value that the dataset I was using (https://www.ceicdata.com/en/saint-vincent-and-the-grenadines/environment-pollution/vc-co2-emissions). I was not able to confirm why there was a spike in emissions that year, however.

Since Asia and Europe were the top overall contributors to emissions, I was curious to see what countries within those continents created most emissions. Interestingly, Qatar was Asia’s country that emitted the most carbon dioxide emissions by far, followed by Bahrain. For Europe, Luxembourg was the highest polluter followed by Russia.

In order to get a sense of the change in emissions, I decided to compare data of two years, the start of the dataset (1990) and the end (2019). It was interesting to learn that Africa had the biggest percent change, at 57% (rounded to a whole number from two decimal points).  On the other hand, Europe decreased their emissions by 35% (rounded from two decimal points). Not only is that an astounding value, but also they are the only continent that decreased their carbon footprint. This finding was not surprising, yet appreciated, since it is well known that Europe has many regulations and policies that support climate change initiatives. It’s great to see the actual outcome based on the EU’s proposal of becoming climate-neutral (Climate Change: What the EU Is Doing, 2023).

Despite Europe decreasing much of their collective emissions, we can see they are still one of the top contributors of carbon dioxide globally in 2019.

I made a similar comparison by continent with the temperature changes in the years 1990 and 2019. By taking the average temperature changes for each continent, I plotted both years against each other and calculated the percentage change over the last 20 years. The results were a little alarming, shown in the chart below. 

Each continent in 2019 has a temperature change of more than one degree celsius, and almost all see an increase of over 100% from 1990, despite their carbon emissions not seeing a percent change of that magnitude.

I wanted to then only focus on the overall temperature information, so I took the dataset with each country’s temperature change and got a yearly average to plot across the years. It’s clear that since the 1990’s temperatures have consistently increased.  

In order to find any relationship between the temperature changes and CO2 emissions, I again used the temperature changes dataset for all countries and measured it as a yearly average from 1990 to 2020. Similar to carbon dioxide emissions, the temperature steadily increased overall throughout the years.

Based on these charts, it appears that the temperature increase is growing at a rate that is steeper than the carbon emissions. However it’s important to remember that the top chart’s units are small ticks in Celsius degrees whereas the bottom chart’s units are metric tons divided by the country’s population.

We can also see that in recent years the trend lines have further spreads compared to he beginning of the dataset. This leads me to interpret that there is a compounding effect of temperature changes from carbon emissions. A looming question arises: will we be able to reverse the damage, and how long would it take?

Usability Test

I created a scenario with three tasks and tested it with two participants. My goals for the usability test were to:

I tested three different charts: the CO2 emissions time series by continent (Figure 1), a line chart with total CO2 emissions and temperature anomalies (Figure 8 below), and a bar chart comparison between years 1990 and 2019 for all continents (Figure 4). I asked general questions to make sure the participants were able to decipher the information and interpret it. I additionally asked them to validate certain information such as “what is the highest carbon emission in 2019?”

My findings

Task 1

Overall, there were no glaring issues since this was a simple chart. The participants were able to find the outliers and were surprised by some of the data, for example the spike in emissions in North America in 2009, or that Europe has such high emissions. One participant I noticed kept looking at the chart and the legend, so prior to interviewing the second participant, I added labels to the continents in line. However, Tableau refrained from showing Africa for some reason, which was confusing to the second participant. One participant mentioned whether the data is “good or bad” and wondered what emissions we should strive for as a society? I thought that was a great question, but one I can’t answer with the information I currently have. Finally, there was confusion about when the trend lines began and ended in the chart. I removed the trend lines since the story is clearly shown on its own through each line.

Task 2

The second task took longer for both participants to process since there were multiple elements in the line chart. One of the feedback concerned a lack of knowledge of what the temperature anomalies are referencing in terms of a baseline. This was an important point to note that not everything can be explained in a visual. 

A positive observation was that both participants understood that the color and line thickness attributed to the temperature changes after studying the chart. However, there was a discrepancy between the interpretation of the chart. One participant interpreted the increased temperature changes despite decreased emissions meant there was a lag between climate change and the pollution. The other participant did not think there was any relationship that could be determined from that chart. 

Task 3

The last task was straightforward and aroused curiosity in both the participants. One of the participants wanted to see more details on the countries that contributed to the continents’ emissions, and the other was interested in seeing a trend of percent change year over year. Finally, I received good feedback about the color choice, and decided to change the colors to mute the 1990 year, but make 2019 more prominent.  

Concluding Questions 

Finally, during my concluding questions, I was able to confirm that both participants learned something new via the visualizations, while sparking some more questions and interest mentioned above.

Summary of Changes Made 

To conclude, we can see that there are still too many increasing carbon emissions across the continents, except for Europe who is making strides to become climate-neutral by 2050 (Climate Change: What the EU Is Doing, 2023). Despite Europe’s efforts, temperature changes have not immediately decreased in sync to decreasing emissions. In addition, we can see that the overall temperature has increased despite a total decrease in emissions per capita. These artifacts show that temperature changes are not immediate to changes in emissions, the effects have a lag or are compounded from prior years.

Self Reflection & Future Considerations

After investigating the climate and CO2 emissions data, I was curious how a country’s economic status influences the emissions it produces. Moreover, how might the advances in energy efficiency affect the output of emissions? If the emissions improve for those countries, how can we ensure that other lower income countries have the resources to clean energy? I started exploring this a little bit, but realized I did not have the time to flesh it out. Overall, I found that the higher income countries are the ones that do produce more emissions. However, seeing that Africa has increased its emissions by 55% from 1990 to 2019, what does that mean? According to an article by the Foreign Policy Research Institute, the Chinese government is investing heavily in Africa. In 2010, one-third of Africa’s power and infrastructure was financed and built by companies owned by China (Jones et al., 2022). It will be interesting to see whether the continued investment in Africa from higher income countries has a negative outcome in CO2 emissions.

This lab was more time consuming compared to other labs, yet fulfilling. I had to find new datasets after my initial proposal, then familiarize and play around with those datasets. This took a big portion of time. I also got caught up in toying with the idea of going in a new direction by looking for economic information on countries, as mentioned above, to see if there were any interesting relationships with carbon output. However, I had to make a decision to refocus on my initial plan on analyzing only the climate changes and CO2 emissions. In the end, I am able to tell a story about the climate changes and carbon dioxide emissions across the world. In addition to my questions posed around how income may play a role in emissions, I might also consider exploring in the future visualizations between deforestation (which I did in a prior lab), carbon emissions and climate change. 

References

Climate Change Data. (n.d.). Climate Change Indicators Dashboard. Retrieved May 2, 2023, from https://climatedata.imf.org/pages/climatechange-data

Climate change: what the EU is doing. (2023, February 7). Consilium.europa.eu. Retrieved May 2, 2023, from https://www.consilium.europa.eu/en/policies/climate-change/

CO2 emissions (metric tons per capita) – North America | Data. (n.d.). World Bank Data. Retrieved May 2, 2023, from https://data.worldbank.org/indicator/EN.ATM.CO2E.PC?most_recent_value_desc=false&locations=XU

Developing Countries Are Responsible for 63 Percent of Current Carbon Emissions. (n.d.). Center for Global Development. Retrieved May 2, 2023, from https://www.cgdev.org/media/developing-countries-are-responsible-63-percent-current-carbon-emissions

FAOSTAT. (2023, March 27). Food and Agriculture Organization of the United Nations. Retrieved May 2, 2023, from https://www.fao.org/faostat/en/#data/ET

Home. (2023, April). YouTube. Retrieved May 2, 2023, from https://www.ncei.noaa.gov/access/monitoring/climate-at-a-glance/global/time-series/globe/land_ocean/ann/1/1990-2020

Jones, C. D., Ndofor, H., & Li, M. (2022, January 24). Chinese Economic Engagement in Africa: Implications for U.S. Policy. Foreign Policy Research Institute. Retrieved May 2, 2023, from https://www.fpri.org/article/2022/01/chinese-economic-engagement-in-africa/

WDI – The World by Income and Region. (n.d.). World Bank. Retrieved May 2, 2023, from https://datatopics.worldbank.org/world-development-indicators/the-world-by-income-and-region.html

“What is this graph telling you?” (n.d.). 3iap. Retrieved May 2, 2023, from https://3iap.com/key-questions-for-user-testing-data-visualizations-5vJ8JychRVGIGWq-TpFIIg/