Papers. Oh they take some time. I am pleased to announce a new paper (well newish. Published September 2020 Tropical Conservation Science). This paper started in 2018, but it’s always been a side project so I’ve struggled to make the time to make it a focus.

I am happy to finally have this off my to do list, but I’m more happy to share the work with you here. This is from work I’ve been doing with one of my favourite collaborators, Dr. David Roberts at The University of Kent.

Illegal trade of wildlife is very difficult to monitor. Fortunately this is Dave’s expertise! One approach is to find patterns in legal online trade, so we suggest an analysis that gets the most out of this data.

Maximising what we know

Our dataset is from a study of online wildlife trade which investigated the trade of Convention on International Trade in Endangered Species animals (including live animals, parts and derivatives), over 280 open online marketplaces, across 16 countries, during a six week period in 2014 (Hastie & McCrea-Steele, 2014).

Our study was considerably more extensive than usual data on legal wildlife trade, which is currently not collected with a global mindset. Instead there is a focus on the extensively traded items from countries which trade large amounts, for example, China trades ivory products and Germany trades turtles and tortoises. However, by applying some statistical analysis, we can shine a light on the less dominating behaviours.

Looking at 31 different types of wildlife products, traded by 16 different countries, we cluster the products according to their trade patterns (using hierarchical clustering in R), so products which are traded similarly are grouped together. The 31 products were grouped into eight categories, five of which grouped items which were predominately traded by one country. For example, cluster 1 contained ivory, rhino and pangolins, which are all largely traded by China. During the time period of the dataset, 1662 ivory items were traded by China, but only 164 rhino items, and a measly 3 pangolins. With this distribution, it is understanding that reports tend to focus on the ivory items. However, by grouping them together, one can make assumptions about the illegal trade of all of these items for the price of one!

Other products do not have a strong association with a particular country (from our list of 16 countries). These products were exotic birds, primates, otters, antelopes, red panda, conches, cats, crocodiles and alligators, foxes, bears and whales, see the figure below. Therefore monitoring the illegal trade of these items would require surveillance in many countries.

Voice and Accountability

Returning to the five clusters which contain wildlife items that were traded by a dominating country. These countries were China, Germany, the Netherlands, the UK, and Poland. (The trade of items from the other three clusters were not dominated by a particular country.) Ordinating these clusters we looked for a relationship between the items traded and Worldwide Governance Indicators, which are six measures for each country that include factors such as control of corruption and government effectiveness.

These indicators come with controversy, of course! I personally find them nice measures for understanding the world. They are defined here (Kaufmann et al., 2010). We found a correlation between the items traded and the ‘Voice and Accountability’ score for each country. The  ‘Voice and Accountability’ captures perceptions of the extent to which a country’s citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media. So using this correlation (not causation!) we can make trade predictions for countries not included in our data. For example, comparing the Voice and Accountability score for the United States (a country not included in our dataset), with the average Voice and Accountability score for the clusters, we infer that the United States traded elephant items (not ivory) and owl items during 2014 (items from cluster 6).

What pets do you have?

Sharing the results of this paper with my international colleagues was fun, as my German and Dutch colleagues learnt that Germany and the Netherlands hosts Europe’s largest reptile trade shows, Hamm Terraristik (as well as its associated online trading platform) and Terraria-Houten. As my Dutch colleague said, “I don’t know anyone with a pet turtle…”. I guess she’s not asking the right questions at parties.