The gases that trap heat in the atmosphere are known as greenhouse gases. While these gases play a pivotal role in keeping the planet warm, human activities in the past 150 years have led to an overall increase in greenhouse gases. A leading source of greenhouse gas emissions in the United States is the combustion of fossil fuels to produce heat, electricity, and transportation.
In order to reduce greenhouse gas emissions, the world needs to rapidly switch to low-carbon sources of energy like nuclear and renewable technologies. Renewable energy usually tops the discussion related to climate change as sources like solar and wind energy do not emit carbon dioxide and other greenhouse gases that contribute to global warming.
As energy production is a leading source of GHG emission, the goal of my study was to understand the correlation between renewable energy production and GHG emission for leading countries. I chose to create a dashboard that helps a user get some insight from the data. I made use of Tableau to create the dashboard while made use of Our World in Data to get the datasets.
Since the industrial revolution, the energy source for most countries has been dominated by fossil fuels. This has not just an adverse effect on the environment but also on human health. Even today, around 80% of our energy comes from crude oil, coal, and natural gas. This has been responsible for global warming, air pollution as well as water pollution.
The burning of fossil fuels produces large amounts of carbon dioxide which traps heat in the atmosphere that contributes to climate change. In the United States, the burning of fossil fuels accounts for 75% of carbon emissions. In addition to carbon dioxide, the coal-fired power plants produce 42% of mercury emissions in the United States, as well as 67% of U.S. sulfur dioxide emissions that lead to acid rains.
In order to stave off the detrimental effects of climate change, renewable sources of energy can be extremely helpful. Renewable sources of energy like wind and solar power do not emit carbon dioxide and other greenhouse gases that lead to global warming. In addition to being “green”, it also plays a role in creating jobs, making electric grids more resilient, and expanding energy access in developing countries. By the year 2035, renewable sources of energy are projected to account for half of the electricity production globally. As of now, Hydropower is the most widely used renewable power source.
Choosing a dataset
In order to understand the correlation between renewable energy production and greenhouse gas emission, I retrieved the dataset for annual change in renewable energy production and greenhouse gas emission for each country. Between the years 2006 and 2016, I analyzed the countries that had the maximum change in renewable energy production by absolute value and the greenhouse gas emission for those countries in the same time period.
For the research, I conducted interviews in two phases. Following the user interviews, I made a note of the pain points and keywords mentioned in the interviews. Subsequently, I framed the goals for my study
I conducted semi-structured user interviews with two participants. One of these interviews was conducted early-on while the dataset had not yet been finalised. The second phase of interviews were conducted after the dataset was finalized in order to start working in the desired direction.
Before starting to look for a dataset, I interviewed a student pursuing his master’s in Renewable Energy from KTH Royal Institute of Technology, Sweden. From the initial interviews, I understood that there was a lot of data available that could be useful. Not just were there enough datasets available, but also enough dashboards and visualizations. However, it would be more useful if this data could provide some insights. This could be done by correlating two variables. He also mentioned the Paris Agreement, an agreement within the United Nations Framework Convention on Climate Change, that deals with greenhouse-gas-emissions alleviation, adaptation, and finance, signed in the year 2016.
A student at KTH Royal Institute of Technology, Sweden pursuing a double Master’s degree in Energy Innovation and Renewable Energy. Previously, he has completed his bachelor’s degree in Mechanical Engineering.
- One dataset is not sufficient
- There is plenty of data available
- Processing data to get meaningful insights is a challenge
- Showing a correlation can be helpful
A Product Operations Analyst with more than a year’s experience working in the internet industry, skilled in SQL, Hive, Tableau, and Microsoft Excel. As someone who works with dashboards on a daily basis, she made for an ideal user.
- Prefers dashboards to make first level judgments
- Uses MS Excel when a deeper dive in the raw data needs to be seen and thus prefers downloading data on excel and cleaning it
- Prefers dashboards with simple interfaces that are easy to use even if the use case cannot be extended much.
Once I completed taking the interviews, I synthesised the results of the interview to create a user persona shown below.
Based on the user persona and the insights received in the interviews, I defined the goals that I wished to accomplish by means of my dashboard. I collated all the insights into a Miro board as shown below. The main goal was that the data must provide some useful insight but at the same time be easy to use.
The term dashboard, which is a metaphor for a car dashboard can be effective in providing crucial information at a glance. In order to design a better dashboard, I conducted some secondary research to understand the best practices in the design of dashboards.
The two types of dashboards I came across were Operational dashboards and Analytical dashboards. While operational dashboards provides critical information to users engaged in time-sensitive tasks, analytical dashboards provide users with information at a glance. Based on the user interviews, I decided to move forward with an analytical dashboard.
In order to create the dashboard, I made use of Tableau. Previously, I had made use of Tableau for a similar project on Hydroelectricity Production. While the first time I made use of trend-lines, this time I implemented a few different visualizations and also incorporated the feedback received for the first report.
Our World in Data
Our World in Data is an online platform that seeks to make knowledge of bigger concerns accessible and lucid. In addition to drawing inspiration from the existing visuals on the website, I also retrieved my dataset from it.
Miro is an online visual collaboration software that aids your UX Research. I used Miro to gather insights from the user interviews as well as define the goals for my study.
Once I had the dataset, I refined the same using MS Excel. I also combined two different datasets for the first chart.
The biggest challenge was to choose which visuals that must go into the dashboard. Towards the end, I added a dual-axis chart on the top, five charts to show the absolute change in renewable energy as well as a trend-line to show the greenhouse gas emission across the globe.
The different visualizations considered included Maps, Separate bar charts, and combo charts. Towards the end, I decided to go ahead with a dual-axis chart and use the relative values for both the quantities. This allows the user to look at the values of both variables for each country, making it easier to compare, consequently making it easier to derive inferences.
The dual axis chart was placed on the top of the dashboard to provide information at a glance.
Trend Lines for specific countries
Below the dual-axis chart, trend-lines that showed the absolute change in the renewable energy production for the same five countries between the same time period to get an idea about how the values have changed over the span of ten years.
Trend Lines for Global GHG Emission
In order to serve to a larger audience, a chart was placed at the bottom of the dashboard that showed the global greenhouse gas emission for the users to get a deeper perspective of how the global GHG Emission had increased for all the countries as well as for the world. By default, it shows the top five GHG producing countries and allows the user to further choose the country of their choice.
For choosing the color palette, I took inspiration from existing visualizations on the Our World in Data Website.
Based on these, I decided to use two color gradients for my Dual-Axis Chart
- Red-Green: When the Greenhouse Gas Emissions increase, the color tends towards red, otherwise towards green.
- Cyan-Blue: As there is an increase in renewable energy production, the color tends towards Blue.
For the singular trend lines, I chose green that resonates with the topic. For the Global GHG Emission chart, the colors were chosen in such a way that they are easy to distinguish. Similar shades of colors tend to be confusing in case of trend lines.
The final dashboard can be seen here.
As it can be noted, out of 5 countries listed in the Dual-Axis chart, 4 of them have either shown a decrease in GHG Emissions, or the same has been constant. However, the same does not hold true for China. This can be either be an aberration or can be attributed to several other factors. Considering that energy production is a major source of GHG Emission, switching to renewable sources has definitely had an impact on the GHG Emissions.
It must also be noted that between the years 2006 and 2016, there was a 13% increase in the Global Greenhouse Gas emission.
After completing the visualization, I tested the same with the users to get feedback. The users were asked to think-aloud while surfing through the dashboard. The following things were noted:
- The users really liked how the dashboard looked, as the graphs were neat and easy to read.
- The visuals are deemed to be good as long as you can derive meaning out of them.
- The legends in the Dual-Axis chart were confusing and it was suggested that the same should be made clearer.
- In some places, the units were not clear.
- The users really liked the Global GHG Emission chart that allowed them to select countries of their choice which would make it easy for them to compare.
While working on the project, I had the opportunity to expand my skillset in Tableau and learn about a few more visualizations. One of the most important things I learnt from working on this project was understanding a user’s perspective and iterate upon the design based on the feedback I received. I realized that just mapping datapoints into a graph may not suffice and may not defeat the purpose, as visuals help in recognizing patterns and put information to good use.
As for further steps, I wish to collaborate with my users and incorporate their feedback that I received after testing the interface. I also wish to retrieve datasets related to transportation that can further bolster my research and help me uncover more insights that may or may not support my hypothesis.