Dinosaurs — Back for Another Bite
A visualization 65 million years in the making.
Dinosaurs are a topic that has captivated audiences for years. It seems that everyone likes them. From the critical success of the Jurassic Park franchise, to the popularity of dinosaur exhibits in museums, there is something about these beasts that never fails to captivate our attention.
The main goal of this project is to be able to see the correlation between dinosaur diet type and country with relation to various Mesozoic time periods. For example, this visualization might uncover that a particular region in the globe primarily was home to herbivores throughout the entire Mesozoic era.
Data visualization is not simply a way for us to visualize a dataset. Data visualizations, especially interactive ones, are a way to tell a story — to give the viewer a connection with the data. As Cole Cussbaumer Knaflic, founder & CEO of storytelling with data, argues: “Success in data visualization does not start with data visualization. Rather, … attention and time should be paid to understanding the context for the need to communicate.”
For my project, I wanted to use this approach to data visualization to tap into that sense of wonderment and curiosity that so many people seem to have towards dinosaurs. I had two main goals with this visualization. The first was to serve as a casual, yet informative, source about dinosaurs and the Mesozoic era. The second was to encourage exploration and foster a sense of curiosity from the viewer’s standpoint.
The main dataset that I used had the following fields:
- Species.
- Time period the dinosaur lived in (eg. Triassic, Jurassic, or Cretaceous periods).
- Diet type (eg. herbivore, omnivore, carnivore).
- Country that the fossils were found in.
Using this data, I began to think of ways that the data could tell a story, and how to best engage viewers with the data. I knew that at the very least, I would be using a world map in some way, but there was a lot of work I needed to do to figure out how to display the map, and what accompanying visualizations would be appropriate.
Before I could begin to work on the visualization, I needed to rework the data to better suit my goals. To accomplish this, I used pandas, a Python library designed for data analysis and manipulation.
Once my data was in a workable state, I began to create sketches of what I envisioned the visualization to look like.
The biggest challenge I faced early on was how best to make my visualization accessible. When I say “Everyone likes dinosaurs!”, I mean everyone does. This is no exception for people who have disabilities or information processing impairments.
Because my visualizations were shaping to be heavily reliant on text and color, I wanted to accommodate two groups of people who have issues directly related to processing text and color– dyslexia and color-blindness.
I spent some time trying to design a visualization that could work for these populations as well as keep the features I had originally envisioned for my visualization. However, I eventually came to realize that the best decision would be to make one main visualization, and then create alternate versions that are friendly to those with dyslexia or color-blindness.
Looping back to the two main goals I had for this project, I decided that an interactive dashboard would be the best way for viewers to interact with the data. The dashboard I designed has three main features.
The first is the world map that the viewer is greeted with. Once selecting a Mesozoic era to explore, countries with available data are filled with a color corresponding to what the majority diet type in that country is.
If the viewer selects a country, they then have two more components to interact with.
The first is an interactive bar chart that breaks down the dinosaurs by diet type for that country. If the user hovers over one of the bars, they will notice two things. The first is that all other bars will show the difference between their value and the selected value. For example, in China, there were 95 herbivores, 39 carnivores, 20 omnivores, and 4 dinosaurs whose diet was unknown. If you were to hover over the herbivores bar, the carnivores bar would display -56, the omnivores -75, and the other -91.
The second is that a dotted horizontal line will appear over the top of the bar, intersecting with the other bars of the chart. Both of these features allow for a more precise visual comparison between values.
Aside from the bar chart, the user also has a full list of dinosaurs from the selected country that they can scroll through. However, I recognize that a long list of dinosaur names can be overwhelming and a bit dull. In order to help engage the user with the data even more, if the user clicks on a dinosaur’s name, an image of the dinosaur is displayed. This picture is pulled from the Wikipedia article of the dinosaur, and by including a picture, I was hoping to add an extra layer of engagement and connection with the data from the user’s point of view.
From this visualization, there are a few interesting trends that I think are worth pointing out.
The first is that we see lots of groups of diet type during the Jurassic Period. In northwest Africa/western Europe, southern Africa, and Asia, the majority diet type was herbivorous. In central Europe, Russia, America, and Australia, the majority diet type was carnivorous. I thought it was interesting to see distinct groupings pop up together.
We can observe similar groupings in the Cretaceous and Triassic periods as well:
Another interesting thing to note is that throughout the entire Mesozoic era, countries in southern Africa were primarily home to herbivorous dinosaurs.
As I mentioned earlier, this visualization relies heavily on text and color to convey its data, and I recognize that this is not accessible to those who have visual impairments or information processing disorders. In order to address this lack of inclusion, I created alternate visualizations that can be enjoyed by people who have dyslexia and those who are color-blind.
For the dyslexia-friendly visualization, I swapped the font to be one that is specifically designed for dyslexia.
This font is considered to be dyslexic-friendly due to its heavy weight on the bottom of the characters. When testing this font with a person who has dyslexia, they responded very positively to it, saying that this font makes each character stand out, addressing the issue of their dyslexia “muddling together” individual letters, as they put it.
For the color-blind version of the visualization, I wanted to check how my existing color palette would look to various color impairments. My original color palette held up pretty well against various types of color-blindness, with the exception of achromatopsia.
The top row of colors is my original palette, and the bottom row is how someone with achromatopsia would see the palette. It is obvious that there is a poor contrast between colors from the perspective of someone with achromatopsia, with the third and fifth colors being nearly indistinguishable. To fix this, I had to find a color palette that can still be distinguishable between the different types of colorblindness.
As we can see, this new color palette I am proposing has a much better contrast when viewed as someone with achromatopsia.
Additionally, this is what the color palette looks like in various other forms of color-blindness.
The contrast between colors is high enough for each form that they are clearly distinguishable from each other.
As someone who has always been fascinated by dinosaurs, I was very excited at the idea of making this my project. The sense of curiosity I feel when learning about them was something that I wanted to share in my visualization by fostering a sense of engagement and exploration for everyone to enjoy.
