App developers are used to debugging software, which is a tough enough task at the best of times. But when the bugs are the size of your hand, and bite, you know you’re in new territory. In Naturally Speaking reports this week, we find out about a project to develop a mobile app that translates research data into a citizen science tool for tourists.

Tracking bugs and beasts in the Serengeti National Park

 

In the summer of 2015, software engineering student Vlad Schnakovszki found himself in the Serengeti, Tanzania, helping to debug a mobile app. Fortunately, finding himself (and a million wildebeest) was precisely the function of the app he’d been developing—the Serengeti Animal Tracker.

The app provides real-time locations of wildebeest and zebras as they migrate across the savannahs of the Serengeti, allowing tourists on safari to interact with one of nature’s greatest spectacles. The app is the brainchild of Dr Grant Hopcraft, a Research Fellow in the Institute, and Dr Inah Omoronyia, a Lecturer in the School of computing Science. Vlad joined the project in 2013, and during his 3rd and 4th year he helped build out the app, helping to turn data into directions for tourists.

The data for the app come from a project led by Hopcraft, a landscape ecologist interested in how migratory animals such as wildebeest make decisions about where and when to move. “The research helps us to understand the natural and human-mediated factors that influence these movements,” says Hopcraft. “These may include natural fires or vegetation cover, but also large scale shifts in the climate that may bring wildebeest into closer contact with human livestock systems.”

“A single line up of all the Serengeti wildebeest would stretch from Glasgow to Casablanca”

A substantial component of Hopcraft’s research focuses on the knock-on effects that 1.3 million wildebeest and 250,000 zebra have on ecosystem processes. With numbers this big, they literally shape the land through which they move, harvesting vegetation in one region and fertilising with droppings in another—facilitating biodiversity and providing ecosystem services.

“Grant is among the few people in the world who can claim to get a txt message every morning from wildebeest in the Serengeti.”

The irony is that at peak migration it’s sometimes hard to find wildebeest in an area that is more than 20,000 km². To follow them, the research teams attach GPS collars around the neck of 10 lucky female wildebeest. While 10 individuals out of 1.3 million is a drop in the ocean, the technology gives the researchers a rare glimpse into the daily lives of a few animals as they navigate around the Serengeti.

With the installation of a mobile phone network covering roughly half of the park, the GSM-enabled collars collect GPS data four times a day and send it over the network. Grant is among the few people in the world who can claim to get a txt message every morning from wildebeest in the Serengeti.

While the data are important for the research, they’re also fascinating for non-specialists, if provided in a form they can use. Incorporated into a phone app, the data become a guide to where animals are, and if you’ve just spent £2,000 on a safari, this is where you want to be. The park receives roughly 180,000 tourists each year, most of whom will get a phone signal, yet the majority of them visit just a tiny fraction of the park. Hopcraft points out that the app can actually help ease congestion in the Serengeti.

“Safari tours tend to go where the known viewpoints of the great migration take place or where animals have recently been spotted,” says Hopcraft. “This results in 20 or 30 vehicles descending onto a few locations, which is not what tourists are looking for in an exclusive trip to this great wilderness. There are in fact plenty of places for breath-taking views of the zebra and wildebeest migration, and the entourage of predators.”

Hopcraft suggests that by allowing people to track the migration in real time, and to share their sightings of other animals, there is an opportunity for people to explore less frequented parts of the park, reducing the damage to honey-pot areas.

 

The app does more than just track wildebeest and zebra. Users can upload their own sightings, allowing them to contribute to a citizen science biodiversity project. As a user, you can tag the location of animals you see, along with a photo. If you can’t identify the animal, there’s a handy photographic animal directory with information about each of the major species. This also lets you personalise your journey, keeping a record of what you saw, where you saw it and when.

Your observations are also shared with other users of the app (and by social media), except tags of species at risk from poaching, such as elephants and rhinos—though you can log them. You can even report an animal in distress—these logs are forwarded automatically to one of the local ranger stations, along with a photo and a map, so they can follow up on critical reports. The more eyes watching and protecting these animals the better.

A feature-rich app isn’t without its technical issues. A necessary step in readying the app is to test it and check for bugs in the field. Software bugs that is. The aim is to make the app publicly available to the Serengeti National Park visitors by 2016. Clearly, the best field-test is to take your software developer out of their cozy office in Glasgow and plonk them in the middle of the Serengeti—far from any creature comfort and surrounded by bugs. The biting kind. This is where we left Vlad at the beginning of this post.

We caught up with Vlad, who has since graduated from the University, to ask about his experience.

“I was really excited about going to the Serengeti, but as a person who hasn’t left the continent before, I was a little nervous about things like malaria or wild animal attacks, especially since a journalist was killed by a lion in South Africa just days before. However, it soon turned out that the animals I needed to pay most attention to were baboons—this became clear after a baboon broke into our car and stole half of our bananas.”

“During my stay I joined the field team in the morning, testing the application and making small adjustments. The afternoons were then spent fixing the problems, while the evenings were generally used for leisure”.

These aren’t the only problems you have to fix in the field. The others include vehicles, which with little encouragement are only too happy to fail, leaving you stranded.

Vlad continues,

“It became clear as soon as I entered the park that the application had one major flaw—it assumed 3G internet would be available most of the time. However, in most places there is only a slow 2G service available from any of the three most popular Tanzanian networks, so I spent a lot of time making the application handle this correctly.”

With a final set of bugs under wraps, and feedback from the Serengeti researchers, Vlad returned to Glasgow to complete his version of the app. While the current app is for Android applications, the team plan to make the app available across platforms before launch next year.

You can keep up to date with the Serengeti Animal Tracker project at www.serengeti-tracker.org

App screenshot image credits: Vlad Schnakovszki (2015) Serengeti Tracker–A mobile application for safari tracking. Honours project. School of Computing Science, University of Glasgow.

Words and photo editing: Jim Caryl, NaturallySpeakingreports