Mapping Controversies: Wildlife conservation

Michael M.
11 min readFeb 16, 2019

--

Mehroze and Michael

Introduction to controversy

Introduction to our controversy

We have chosen to deal with the conservation of the wildlife around the world. As the global climate changes due to global warming, the nature gets affected and this also includes animals. Cities are also expanding and thus moving closer to the wildlife itself, which causes clashes between humans and animals.

Some wild animals are getting closer to a global extinction while some wild animals start to appear in places where these animals have not appeared before. However, Wild animals that have been extinct from a certain area or country have started to reappear hundreds of years later, which concerns people living around any the affected areas, which means that people start to take precautions and try to stop wild animals from settling down. There is debate about whether living wild animals should be reintroduced to former areas of living or not.

In April 2018 a 66-year old farmer shot and killed a wolf who was wandering around on his fields in Denmark, the murder of the wolf started a huge debate in Denmark, where people were split on the fact that whether it was okay or not. The 66-year old farmer and his attorney received several life threats from the pro-wolf “camp”. The farmer was later in 2018 convicted of shooting and killing a protected animal.

With the use of the news media Infomedia, we been granted access to articles concerning wildlife in relation to search keywords that we have chosen to work with. These keywords that we have chosen, are based on articles that we have read in regards of the wildlife topic. By combining these keywords with the search topics, we have gained an overview of the numbers of articles that could be found through Infomedia, which shows us how debated a certain topic was in the Danish news press.

Protocol for finding our Wikipedia pages

We started our data collection by choosing an overall category on Wikipedia. In our case, the category was: Wildlife Conservation.

Then we used a custom Python script to scrape this category in order to get a set of pages according to depth level. We chose only to go down one level, such that the network would be more manageable. By including all pages found on the category itself and all the pages that is part of each subcategory we ended up with a total of 344 Wikipedia pages represented as .csv/.json files.

Then we wanted to explore how all these pages link to each other and how they are connected.
We opted to use a python script that looked for all internal wiki-links within the 344 pages, including links in templates to get as many clusters as possible.

We also ran a script to extract all text from all of the 344 pages, for use in our semantic analysis.

Network of Wildlife Conservation with keyword overlay

Network of the category Wildlife Conservation with internal wiki-links. The green circles represent to which degree the word “controver*” occured in the text.

On Wikipedia wildlife conservation do not include obvious controversies in itself. However when you look around on the various pages, there are some hints of controversies. For example on the page for WWF (World Wide Fund for Nature) which include a chapter on “controversies and disputes”.

In order to find out if more Wikipedia pages linked to wildlife conservation included the mention of controversies, a keyword search was conducted.
By using a custom python script and inputting the keyword of “controver” we were able to extract information on how many of the pages the word was mentioned.
The word was mentioned 27 times, which indicate that conclusions made on the background of the vizualisation should properly be taken with a grain of salt. However, this method of searching for controversies across pages, could indicate potential areas of further investigations.
In this case it shows that the most cases of mentions of controversies is centered around rewilding. This aligns quite well with our initial interest in the subject, which was about animal reintroduction, which has sparked a lot of media debate in Denmark. This vizualisation shows that rewilding in general is a potential controversial topic.
It is also interesting to note that several different acts are also shrouded in potential controversies. However, without digging deeper into each case, it is impossible to conclude whether or not controversies actually are taking place.

A network of co-occurring noun phrases.

Network of the top 300 terms co-occuring between wikipedia pages. The raw data used to make the network is from an extraction of all the text within the pages of the Wildlife conservation category.

Initially we had an assumption that Wildlife conservation would be very broad and include many different areas. While this is certainly true, this network illustrates that many of the co-occurring terms are used in relation to bird species management. In regards to the overall debate about wildlife conservation, this is somewhat surprising for us, given that birds are not the first thing you hear about in the news in relation to conservation of animal species. Animals such as polar bears, elephants, rhinos and so on are perhaps more present in the media, but this network at least shows that many of the terms used on the Wiki-pages stems from bird related pages.

The network also shows that the co-occurring nouns are divided into several distinct clusters, where some clusters are more connected than others.
There is also a large center of nodes, that are more loosely connected to each other and thus does not have as many co-occurring terms.

Revision timeline of the Wikipedia page: World Wide Fund for Nature.

One of the larger pages within the Wildlife Conservation category is the World Wide Fund for Nature page (WWF).

Quickly skimming through the edit history also reveals that the page includes a large number of edits. Therefore we found it interesting to investigate that specific page for revisions.

First we inputted the Wikipage into a script that extracted all the revisons made and produced a table that contained a list of number of revisions made per month. This table was parsed into the RAWGraphs website to get an overview.
The RAWGraphs takes the table input and produce a simply graph as output:

This graph shows that there seems to have been most edits around 2006–2009. To examine this further we can use another Python script that can narrow the search to a specific time period:

This graph shows number of revisions running from 2005 to 2009. The graph shows some spikes throughout the time period. We can estimate whether those edit spikes was caused by unique users or many of the same users:

Comparing this graph with the previous one, we can see that the number of edits corresponds somewhat with the number of unique users making edits. Some new knowledge regarding the WWF might be the cause of the spikes, where outsiders who have not had any previous affiliation to the Wiki-page come in and add some specific details.
Looking deeper into the edits themselves through a .csv file containing all the edits, it is possible to better see what areas the edits has revolved around:

Screenshot of a portion of the table containing all revisions made to the page. This shows an example of what some of the edits have concerned. In the third column it is possible to determine whether the edit was made from an anonymous user (IP-address) or a registered user (username).

The former graph shows that many of the revisions happened in November 2006. Looking at the above table, one can see that some of those edits has been reverted due to accusation of vandalism and that is the reason behind some of the revisions.

Another example is looking at August of the same year, which is another spike according to the graph. The revision history .csv file shows that many of those edits are concerning a legal dispute between World Wide Fund for Nature and the World Wrestling Federation. Reading those edit comments, a battle between the communities representing each organization seems to have taken place.

The next part will focus on the medium from which we have collected data.

Data protocol for havesting Infomedia

For hand-in 2 we chose the comparative route. This decision was mainly based on the fact that our initial interest was the public debate about the wolf situation in Jutland and we wanted to return to that issue.

Therefore we decided to focus on the news media as our chosen medium. Because of our focus on the wolf debate in particular, we were only interested in Danish news articles. We quickly realized however, that a broader approach to the topic might be even more interesting. In the recent years, rewilding of certain areas around Denmark has been a hot topic, which includes a variety of animals being reintroduced (though wolves in Jutland is not technically rewilding). Thus our main focus in this second part of the exam is about rewilding as a conservation method.

Our data has been harvested exclusively from the Infomedia platform, which is a search engine for Danish news outlets.

In the beginning we read numerous articles in relation to rewilding and wolves in Denmark, in order to get an overview of actors and issues.

Afterwards we made a cross-search for each term, going back the last 10 years. The number of hits for each combination was parsed into an excel sheet.

This large table of cells containing hits was then translated into a Gephi network for further analysis.

Screenshot of our Infomedia search settings. This combination shows 4110 unique articles.

Timeline over wolves-related news articles

This visualization gives us insight into the development of the number of articles on infomedia, from the 1st quarter of 2012 through the 1st quarter of 2019.

The reason we have chosen to start from Q1 2012 is that during the first quarter of 2012, for the first time in recent times, wolves were observed in Denmark.

How did we gain insight to the number of articles?

We chose to use the following keywords on Infomedia; Ulv* and Debat*

By choosing these keywords, we were able to display the articles on infomedia, where “Ulv” and “Debat” appear together in an article. This means that we get articles that mainly deal with wolves in Denmark and the debate that they cause.

As can be seen from the chart, the number of articles increases between April 2012 and August 2013, as during this period information have been received about the wolf reappearing in Denmark after many years of absent. The information about the wolves in Denmark are; Observations, wolves that have been caught by camera traps and episodes of dead animals near places where the wolves were believed to be observed.

Murder of a wolf in Denmark in April 2018

16. April 2018, a 66-year-old man shot a wolf who was walking around on his fields. The killing of the wolf created a massive public debate. The episode of the wolf-murder and the massive public debate it caused, can be seen on the chart. The number of articles explodes right after the wolf-murder and the following months.

Timeline of wolves-related news articles.

This visualization shows us that, if there has been an incident the numbers of articles rises, and when by saying that the numbers of articles rises, it can be said that the debate rises. This chart gives a visualization on number of articles found on infomedia that contains our keywords Ulv* og Debat* from the period of 1st quarter of 2012 until 1st quarter 2019. The Y-axis is number of articles, while the X-axis is date. This chart gives an insight of the wolf debate in Denmark during our time span.

Bar chart of a specific search combination

By choosing rewild as our main search keyword, we were able to gain insight on which actors appears in articles regarding rewild and the number of articles where both rewild and the actor are mentioned.

Bar chart with rewild* as the main search term, and each entry as a second search term. This helps to give an overview of some of the actors in the rewilding debate and how much they show up in the public media.

This chart shows us number of articles where the main search keyword Rewild* has been combined with different sub keywordd/actors. The data is based from articles from found on Infomedia. This chart shows number of articles where the main keyword and actors are appearing in the same article. The Y-axis shows number of articles, while the X-axis shows the sub keyword/actors used in the search combined with the main keyword; Rewild*.

Rewild* and Jens-Christian Svenning:

As can be seen on the chart an actor that appear in many articles containing rewild is Jens-Christian Svenning, who is a professor in biology from Aarhus University.

According to Jens-Christian Svenning: There is a globally increasing attention on ecology and biodiversity, this attention is not only focused on plants, but also wild animals. A global agenda among some experts is to allow animal wildlife back to areas where they previously were living, this trend is called “rewildning”. By “allowing” wild animals to rewild, to reappear in areas where they previously lived, it is believed that nature will regain its natural growth and increase the biodiversity naturally.

However, as seen with the wolf murder in Denmark which happened in April the controversy starts when people living in the affected areas starts to fight back and make the wild animals go away once they start to reappear. So, there are actors that are pro-rewild and then there are some which are anti-rewild.

Rewild* and Naturstyrelsen

Another actor that appears in numerous articles with rewild are Naturstyrelsen and the reason for that can be that Naturstyrelsen is the Danish Nature Agency, who deals with the topics such as rewild. Danish Nature Agency manages over 200.000 hectares of forest and natural areas. The Danish Nature Agency carries out practical tasks within hunting and game management and specific outdoor and nature projects — often in collaboration with other authorities, organizations and volunteers.

Actors that are not appearing with rewild*

As can be seen on the chart some actors do not appear in articles with containing rewild, and the reason for that some articles uses different word than rewild, or the actors doesn’t simply not deal with rewild. Actors such as Landbrug og Fødevare and Miljø- og Fødevareministeret doesn’t not appear in many articles where rewild is mentioned.

Network of all the search terms

Network of all the different search terms used in Infomedia. The thicker the edges, the more hits the search combination (the two nodes) lead to.

This visulization was made with an applied edge weight filter of 30, which means that all the combinations that had less than 30 hits was excluded. This has kept the focus on the main actors in a more clear way.

In the middle is a cluster of all the major Danish nature organizations. Almost all nodes points to that cluster with relatively many hits.

The actors on the top of the center cluster, has more or less only been active in the wolf debate, and not in the general rewilding debate about other animals.

At the same time Jens-Christian Svenning has been active in all the debates regarding rewilding in Denmark.

Another interesting aspect of the network, is the fact that the other animals concerned, are not as highly linked compared to wolf. It seems that the Danish news media has really had a big focus on wolves in particular. This could of course be down to the fact that the wolves, were the only species that wandered into the country themselves, compared to Bison oxes and beavers (which were reintroduced).

It is also interesting to see that the two politicians included, are mostly active in the wolves debate and not nearly as much in the general rewilding debate.

--

--