Introduction
Annually, international wildlife trade is estimated to be worth billions of dollars and to include hundreds of millions of plant and animal. The trade is diverse, ranging from live animals and plants to a vast array of wildlife products derived from them, including food products, exotic leather goods, wooden musical instruments, timber, tourist curios, and medicines.
Because the trade in wild animals and plants crosses borders between countries, the effort to regulate it requires international cooperation to safeguard certain species from over-exploitation. Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) is such an organization tasked with monitoring, reporting, and providing recommendations on the international species trade. For many years, CITES has been among the conservation agreements with the largest membership, with now 183 Parties. CITES affords different levels of protection for species depending on how threatened they are. In cases where trade is allowed, countries are supposed to grant export permits only if scientific evidence shows that trade won’t undermine the animals’ survival in the wild. [1] [2]
Dataset
I found this interesting dataset on CITES website. It has a database of wildlife trade from 1975 to 2021, and allows users to download the dataset. The dataset contains records on every international import or export conducted with legal species from the CITES lists. It contains columns identifying the species, the import and export countries, and the amount and characteristics of the goods being traded (which range from live animals to skins and cadavers). I think this will be an interesting dataset to explore.
Process
First of all, I listed a couple of interesting problems that I wanted to explore. For example, the top 3 things of US import and export, the percentage of the live trade, as opposed to animal products, etc. So many interesting questions came to my mind, but I could only pick some of them for my final project. Therefore, based on the dimensions of significance, difficulty, and storytelling, I chose the following four questions for this final project.
1. Tusks/Ivory Products Trade (2011-2020)
The large tusks on either side of elephants ‘face—used for foraging for food and water—have long been desired by people. The ivory trade is driven by consumer demand for products made from tusks and supplied by a sophisticated international network of traffickers. In order to remove the tusks, poachers necessarily kill the elephant in the process. These tusks are then sold and made into anything from jewelry and crafts to musical instruments to religious objects. [4]
In 2015, UN urged its member states to take decisive steps to prevent, combat and eradicate the illegal trade in wildlife, on both the supply and demand sides. [5] Since then, more and more countries announced that they started taking actions on stopping illegal ivory trade. In this context, I’m curious about the effect of what they have been doing, so I would like to make a chart to visualize the trend of tusks/ivory trade in the past decade.
- Data cleaning & Visualization
Talking about trend, a line chart is often used to visualize the data over intervals of time. It took me some time to prepare the dataset for this chart. Firstly, I downloaded the dataset for each year from CITES trade database, filtering with the 5 terms about tusks and ivory. [Figure 1,2]
Each piece of record in the dataset represents an international trade. Then I found that there are some other species included in this dataset because they also have tusks such as Physeter macrocephalus (Sperm Whale). Therefore, I deleted these data and only kept the data about elephants.[Figure 3] After doing this, I counted the number of records for each year, and input these numbers in a table. [Figure 4] Then I got a standard data structure for the line chart. I used Tableau to visualize it.
- Result
From the line chart[Figure 5], we know the trade of tusks/ivory products increased a lot before 2014, but it is obvious that the tusks/ivory products trade dropped significantly since 2015. Although the trades in this dataset are legal for some reason, we can still see the efforts of so many countries. In 2020, the number of trade records was 159, which is much lower than 10 years ago.
2. International Wildlife Trade (2020)
Since the dataset is about international trade, I thought I could make a Sankey diagram to show the flow of these wildlife among different countries. However, I found it is complicated to make a Sankey diagram by Tableau. There are some tutorials of Sankey diagram made with Power BI which is a data visualization and analysis software by Microsoft. Power BI is similar to Tableau. Both of them are friendly to beginners. I decided to use Power BI to do the Sankey diagram of international wildlife trade. However, since there are nearly 200 countries/regions in my dataset, the diagram looks so messy. [Figure 6] My idea of Sankey diagram failed.
As I was thinking of giving up this new attempt, the chord diagram suddenly came in my mind. Because each trade is between two countries/regions, a circle of all countries/regions is suitable for visualizing the flow among them. It is even better than Sankey diagram! However, I met the problem again. Although Power BI is also able to make the chord diagram, it cannot process all data and looks strange. [Figure 7]
I did not want to waste this idea of chord diagram, so I searched which software could be used to make it. Some people wrote the tutorial of using R to make chord diagrams. I refused to open it in the beginning, because R is too hard to learn quickly. However, after trying several online software, and finding that they had the same problem with Power BI which cannot process so much data, I finally turned to R. I have seen its capability to process thousands of data in my previous project.
- Data cleaning & Visualization
The chord diagram needs a data structure that is different from the original one. There are more than 30,000 pieces of data, and I really did not know how to transform the data structure to what I need. Luckily, I finally got the right data structure with the help of Prof James using R. [Figure 8]
There is no doubt that using R to make visualization is very challenging to me. I looked at so many tutorials and tried over and over again during the whole process. I was thrilled when I saw the final chord diagram. It was awesome! [Figure 9] The outside circle shows the code of each country/region. However, they are too small to be seen clearly. In order to improve the user experience, I added the flag and name of some main countries/regions beside the code using Sketch. [Figure 10]
library(circlize)
df_group <- read.csv("Desktop/group 2.csv")
chordDiagram(df_group, directional = 1, direction.type = c("diffHeight", "arrows"), link.arr.type = "big.arrow", annotationTrackHeight = c(0.05, 0.05),diffHeight = -0.04)
- Result
From the diagram, it surprised me that Netherlands (NL) is the largest exporter of wildlife all over the world in 2020, and Germany (DE) is the largest importer of wildlife.
3. Top 10 Wildlife Export In Netherlands
From the chord diagram, we got to know that Netherlands (NL) is the largest exporter of wildlife in 2020. It made me want to explore what kind of wildlife that Netherlands exports.
- Data cleaning & Visualization
I used OpenRefine to sort all the exporters and only kept Netherlands (NL). [Figure 11] Then I used Tableau to make a bar chart. It ranks all the wildlife exported from Netherlands by the numbers of records. [Figure 12]
- Result
From the bar chart, I found the number of records of Phalaenopsis hybrid is more than twice that of other species. Then I selected the Top 10 to figure out what are they exactly. After some searches, I found all of the Top 10 wildlife exported by Netherlands are plants. Since Taxon is the formal name of the wildlife, I do not know what they are. I decided to use Sketch to add the images of that plants beside their names. [Figure 13]
4. Top 10 Wildlife Import In Germany
From the chord diagram, we also got to know that Germany (DE) is the largest importer of wildlife. So what kind of wildlife does Germany import?
- Data cleaning & Visualization
The same with last data cleaning method, I used OpenRefine to sort all the importers and only kept Germany (DE). Then I used Tableau to make a bar chart. It ranks all the wildlife Imported to Germany by the numbers of records. I also selected the Top 10 wildlife imported to Germany and added their images beside their names in Sketch.
In the last bar chart of Netherlands, the terms of all the plants are “live”, so it is not necessary to show the terms. However, there are many kinds of terms in the chart of Germany including skins, leather products, trophies, etc. In addition, I would like to know the purpose of these imports. I used Tableau to make a stacked bar chart to show what I want to know above.
- Result
The result in Germany is totally different from Netherlands. All of the imports are related to animals. From the stacked bar chart [Figure 14], most of the terms we found are skins and leather products. The colors of bars represent the purpose of imports. Macaca fascicularis is mostly imported for medical reasons. Apart from the purpose of importing Macaca fascicularis, the purposes of other imports mostly are “Commercial” and “Hunting trophy”. I guess most of them are used to make clothes, shoes, handbags, etc.
UX research
Methods
I plan to use moderated usability testing to evaluate my visualization work. Moderated usability testing is a qualitative study where the researcher’s goal was to discover the problems users face with looking at the visualizations. The core elements of this method are a facilitator, a participant and a set of tasks which are completed by the participants while they are observed by the facilitator. All participants were asked to employ the “think-aloud technique” during the test. This meant that through the help of the facilitator, the participant was constantly verbalizing their thoughts and feelings while looking at the visualization work. This is an effective method of discovering problems without biasing participants.
Since this topic is able to be understood by everyone, I think my target viewers should not be limited by their age, education level, etc. I asked two participants in my network to do the usability testing. Neither of them has learned information visualization before.
They were asked to look at the visualizations one by one. For each image, they were encouraged to speak out their thoughts at first glance. Then they were asked to complete the following tasks for each visualization.[Figure 15]
Tasks
1. Tusks/Ivory Products Trade (2011-2020)
- Tell me what this visualization is talking about.
- What did you see from the pattern of line chart?
2. International Wildlife Trade (2020)
- Tell me what this visualization is talking about.
- Find out which country/region is the largest importer.
- Find out which country/region is the largest exporter.
3. Top 10 Wildlife Export In Netherlands (2020)
- Tell me what this visualization is talking about.
4. Top 10 Wildlife Import In Germany (2020)
- Tell me what this visualization is talking about.
- “Term” refers to the form or the parts of these wildlife. Find out what the term of crocodylus porosus is.
- Find out the purpose of importing alligator mississippiensis.
Findings & Improvements
Overall, both of the participants found the visualizations easy to understand. In particular, although neither of them heard of chord diagram before, they can understand it easily. Having said that, my usability testing showed that there are some problems that could be improved. The findings and the corresponding improvements are listed below.
1. Tusks/Ivory Products Trade (2011-2020)
Findings: One of the participants did not notice the meaning of the row. She thought the number on each data point means the quantity of tusks/ivory trade rather than the number of records.
Improvements: Since it is difficult to explain the meaning of the visualization, I added the explanation of the dataset in the report.
2. International Wildlife Trade (2020)
Findings: Both of the participants feel confused about how to differentiate importer and exporter.
Improvements: I changed the end of the chord from flat to the arrow so that the viewers can see the direction of the chord. [Figure 16]
3. Top 10 Wildlife Export In Netherlands (2020)
Findings: Both of the participants thought this chart is easy to understand.
Improvements: Since this chart and the chart about Germany below follow the same design style, I swapped the row and the column to make them look consistent. I also got some good feedback from my viewer. He suggests adding the color palette to differentiate the different species being traded. [Figure 17]
4. Top 10 Wildlife Import In Germany (2020)
Findings: 1. Some words in the color blocks are not displayed. 2. If some terms (skin, leather products…) are concentrated in one color block, they did not know the number of each term.
Improvements: Firstly, I swapped the row and the column of this stacked bar chart so that leaving more space for the long words. Then I added the border to divide the blocks more clearly. [Figure 18]
Final Work
Tusks/Ivory Products Trade (2011-2020)
This chart looks very simple, but I think it is important to show the efforts of the whole world.
International Wildlife Trade in 2020
Since the last 3 visualizations tell one story on International Wildlife Trade, I combined them together.
Reflection
For the future direction, I have an idea that making a world map showing the paths between the importer and exporter. I tried several times, but I was stuck at the data structure. I would find a way to solve it if I had more time. What’s more, this idea could even be made on a 3D earth, which would be really cool!
Overall, although there were a couple of setbacks in the process, I really enjoy the time that solving problems over and over again, which gives me a sense of achievement!
References
2. https://en.wikipedia.org/wiki/Wildlife_trade
3. https://www.nationalgeographic.com/animals/article/luxury-fashion-wildlife-imports-seized
4. https://www.animallaw.info/article/detailed-discussion-elephants-and-ivory-trade