Beyond Usual Charts: the Making of My Data Story “Sinking”
My design process, from inspiration to research, data and visual choices.
Not all data story projects start with data. It was the case for my data visualization “Sinking”, with which I won third place in the “Data Art and Data Storytelling” category of the “Make Your Data Speak” Award in February 2023.
In the visualization, I show the level the sea could reach around 13 Pacific Island Nations by 2090, according to research by the United Nations’ Intergovernmental Panel on Climate Change (IPCC).
Inspiration
Inspiration for a dataviz can come from different places: a topic of interest, a dataset, but also nature, art, design, or the desire to explore a particular technique or tool in visualization.
Inspiration has something unpredictable. That’s why, when I come across something that I feel could be a potential source of inspiration for a project, I make a conscious effort to fix it in my mind in case it bloomed into a more structured idea later. Sometimes I write it down.
Most of the time I don’t rush to translate straightaway new ideas into concrete projects, and that is for two reasons. The first one is I rarely — or let’s be honest, never! — have enough spare time on my schedule to take the luxury of diving into a new project whenever a new idea pops into my head.
The second is that I believe holding new ideas in the waiting room of my creativity is a very effective way of selecting only the most interesting ones. It’s a sort of natural selection.
The best sources of inspiration are those which stick for a while, and keep coming back and evolving.
The design process for my project “Sinking” started with one of those ideas: a topographic map. Topographic maps are those maps that represent relief features with contour lines connecting all points of equal elevation.
I find them elegant and mesmerizing. With their stubborn repetition of lines over lines, I thought they would be a good fit to express feelings of increasing intensity and urgency in a visualization.
I kept the idea in a drawer in my head until I found the right topic.
Topic
Climate change and ecological transition are topics I feel very concerned about, and data visualization can be a very appropriate way to raise awareness about these issues.
Also, feelings of increasing intensity and urgency — which I visually associated with topographic maps - are inherently associated with climate change.
The link came to me almost naturally: I would use the shape of a contour lines map not to show the elevation of land, but the rise of sea level due to global warming.
I remembered that in November 2021, the foreign minister of the Island State Tuvalu, in the Pacific region, addressed the climate summit COP26 while standing knee-deep in the ocean to raise awareness of the imminent threat sea level rise represents for Pacific islands.
I decided to focus on those islands. I wanted to create something beautiful and sad at the same time, conveying the idea of each State being surrounded by the mounting water.
Research
As interested as I was in the issue of sea level rise, I felt I didn’t know enough to make a dataviz about it. First I had to go through a learning phase.
How do we measure sea level? What causes waters to rise? How pressing is the issue? Is it happening everywhere at the same pace?
I learnt that sea level can be measured either with a satellite or with a tide gauge. According to NASA, both methods produce accurate results — which for me was very good news because it meant I didn’t have to worry about which one of the two methods had been used to collect the data.
I thought the only cause of sea level change was the ice melting from glaciers due to global warming. But I discovered global warming also has another consequence: water expands as it gets warmer, taking up more space in our oceans and making sea levels higher.
Global mean sea level has risen about 8–9 inches (21–24 centimeters) since 1880. But sea level rise at specific locations may be more or less than the global average due to local factors.
In much of the Pacific Islands Region, sea level is reported to have risen at a rate 3 to 4 times greater than the global mean. Changing sea levels can have devastating consequences on the Pacific islands' population: coastal flooding, shoreline erosion, infrastructure deterioration… and even the possibility of a fair portion of their land being under water by the end of the century.
Data
While searching for information to understand sea level change, I was looking at the same time for data to visualize it. I found the report “Sea Level Change in the Pacific Islands Region” published by the Asian Development Bank (ADB) in July 2022, which uses data from the United Nations’ Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report.
I chose to use this source of data for several reasons. First, because the ADB and the IPCC are sources of authority. Second, because it provided complete data and explanations to understand it. Third, because the report dated from July 2022, which was recent enough to be still relevant. And fourth, because it included projected changes in sea level rise until 2090. I wanted to focus on what could happen in the near future, so this was exactly the kind of data I needed.
Four possible scenarios are drawn by the IPCC, depending on how much action will be taken against climate change in the coming years. I decided to visualize the worst scenario, in which sea level could rise at least 60 cm around each island country.
But I felt that talking only about centimeters was not enough, I had to show the human side of the story.
The more vulnerable to sea rising are those living in low-lying coastal lands. I found data on population living in those areas on the website of the organization Pacific Community — SPC. I also downloaded total population data from the World Bank and Wikipedia, in order to highlight the fact that in many Pacific countries, the majority of the population lives in low elevation coastal zones and is therefore particularly exposed to the threat of sea level rise.
Design choices
Once done learning and collecting data, I was all set to start visualizing! Taking inspiration from topographic maps, I started by showing the islands on a map. I performed this step in Tableau software, then exported the map in pdf and reworked it in Illustrator.
I circled the islands of each Pacific Nation and repeated that shape several times to show the centimeters sea level could rise around each country in the next decades. As the sea could rise by at least 60 cm for each country alike, I decided to show lines only for centimeters above 60 to avoid visual clutter.
Pacific island countries are made up of small islands, sometimes scattered on a great surface of ocean — such as Kiribati — and sometimes counting only one tiny island — as Nauru or Niue do. In order to make them all visible without bigger countries giving a false impression of higher sea rise threat, I decided to resize and reposition each island State in an arrangement that, though not matching geographic reality, enables readability. This was a judgement call, and not easy to make, because it meant twisting geography. But the focus of my data story was not geography — readability, aesthetics, raising awareness and evoking emotions were much more important in this case.
For population data, I showed for each country two simple circles with proportional sizes, side by side: one for total population and another for people living in low-lying areas up to 10 meters above sea level. I decided not to write down the absolute values nor the percentages to avoid visual overload — I added a size legend instead. I was aware that it would reduce precision, but it was a trade-off for readability and aesthetics I was willing to make.
Finally, I added colors. I googled “Pacific Islands” and found a picture of a wonderful Samoan beach. I took inspiration from it for my color palette: two shades of water blue and two others of a soft light brown reminding the color of the sand.
Conclusion
Unlike other projects of mine, the creation process for this one was pleasantly smooth and linear, with no bumps or u-turns. I just followed my first idea based on contour line maps and let it guide me through the design process.
The final result is a custom type of visualization, not relying on any pre-existing chart we are used to see. I knew this could make it difficult for some to read, so I was particularly pleased by the positive feedback I received from several people — and hey, I also managed to win third place at a dataviz award!
At the end of the day, I guess something I learnt is that when designing a data story one should not be afraid of taking the liberty to choose unusual solutions if they are consistent with the goals of the project, as long as we can justify them.
Thanks for reading!
My name is Silvia Romanelli, I’m an Italian data visualization designer based in France. If you’d like to know more about me, you can check out my portfolio on Behance or my personal website.
Follow the Make Your Data Speak community on LinkedIn and Twitter!
