“A black cat crossing your path signifies that the animal is going somewhere” – Groucho Marx
Animal-related open data
After last blogging about the emergence and benefits of open data, I decided to see what data was out there that related specifically to animals.
The first one that I found was the Movebank data repository1, hosted by the Max Planck Institute for Ornithology. This is an open data repository that focuses on animal tracking. Their aim is to “…help animal tracking researchers to manage, share, protect, analyze, and archive their data”.
At the time of writing, the site contained 248 data files on studies ranging from ‘Tracking the migration of a nocturnal aerial insectivore in the Americas’ to ‘Antarctic petrel 3D flights’ to ‘Migration timing in barnacle geese’.
Datasets can be located on a map, and data downloaded directly from the site. It also gives you the date of the last dataset update, as-well-as contact details for the data owner.
After browsing for a while, I stumbled across an intriguing dataset called ‘Movement coordination in trawling bats’2,3. This was a study conducted in 2015 which tracked the positions of bats over short time-frames, with the aim to infer what rules the bats used to coordinate their movement. The authors state that “When observing pairs of Daubenton’s bats foraging low over water for stranded insects, we found they have intriguing ‘traffic rules’ – they chase each other, perform tandem turns and even slow down to avoid collision”. The paper for this study is open access4, and was even the subject of a national geographic article5.
I downloaded this data and had a play. I took two of the bats (identified as being in a ‘paired, coordinated flight’ from the dataset’s reference data), and created an animation in R (an open-source statistical programming language). I coloured the bats red and blue, respectively, and drew a line between them that fades as the distance between them increases.
In a very similar vein, the Zoatrack website6 also has animal tracking data, mainly focussed on studies in the Australasian region. It also hosts a number of analysis and visualisation tools, to help make use of the data.
More specialised is Seaturtle7, which focusses on the tracking of (unsurprisingly) sea turtles. For this site, the use of the data appears to be not quite ‘open’, in that written consent of the data owner must first be obtained.
Moving away from animal tracking, the EMPRES-i website8 (hosted by the Food and Agriculture Organisation of the United Nations) contains open data on global animal disease. This large dataset provides information on outbreaks and occurrences of diseases such as swine fever, foot-and-mouth disease, and bird flu. All disease incidents are accompanied by such information as location (longitude and latitude), date, the number of deaths, etc. I downloaded the data for global bird-flu outbreaks and created a quick and simple animation.
A different type of animal-related data can be found on the FAOSTAT website9. Also hosted by the UN, this site focusses on trade and production by country, across many industries. One dataset shows all live animal transport, data heavily used by the ‘stop live transport’ charity. I grabbed some of this data (specifically that relating to all live pig exports across Europe in 2013), and created an interactive map with it.
The interactive version can be found here.
Different again is the animal-related data on the UK government website10. Searching here revealed (amongst others) all exotic animal import licences for 2015 (https://data.gov.uk/dataset/exotic-animal-import-licences-2015/resource/c2117ca8-1dad-4b6f-a14e-b1b8d14f34a3). If you ever wanted to know how many Giant otters were recently allowed into the UK, this is the dataset for you.
Looking at the US government’s open data portal11, at the time of writing there were 16,131 datasets matching the word ‘animals’. These include data on animal bites in Louisville, a plant and animal species atlas for Vermont, and Seattle pet licences. Searching specifically for ‘dogs’ resulted in data on declared dangerous dogs in Austin, control incidents in Baton Rouge, and registered dog names in Anchorage.
Putting data to use
The datasets mentioned above only scratch the surface of what’s available, and new datasets are being uploaded to such sites on a daily basis.
At one level, such data are useful purely from a ‘look and see’ point-of-view. If you find data on a subject that’s of relevance, simply downloading and exploring the data (typically through simple plots) can reveal things of interest.
Beyond that, open data’s real power stems from being able to merge datasets together, either using more than one open dataset, or even better, to enrich your own private data. For example, at Avacta Animal Health, we have years’ worth of allergy testing data. We’ve recently started to look at such data in conjunction with open environmental datasets, such as those available from the MET office, to gain new insights.
In short, open data repositories are awash with animal data, many sections of which may be of interest to veterinary professionals. So whatever data you’ve got, or whatever interests you have, take a look at what’s out there, and perhaps even share your own data, too.
Written by Rob Harrand – Technology & Data Science Lead
DID YOU FIND THIS USEFUL?
To register your email to receive further updates from Avacta specific to your areas of interest then click here.