The homepage of Penguin Watch.
Penguin Watch is a citizen science project on Zooniverse, an online citizen science website collector, where participants identify and mark penguins (adult and child) in photographs captured in remote areas. In addition to marking penguins, participants mark eggs and other animals in the area. The purpose of the project is to help researchers scour through the overload of image data collected from various sites so that they can then start to piece together a picture of any changes in the ecosystem. Penguins are of importance because of their stance at the top of the food chain. The photos come from over 50 stations monitored by the project team that take several photos a day and year round. Thus, it is necessary to get help in gathering data about the changes in penguin population habits and growth over time.
The interface of Penguin Watch, like many other Zooniverse created websites, is designed with a user-friendly, aesthetically-pleasing design. Once signed in (which takes two seconds to create an account), participants get a quick tutorial (around 3 minutes or less) about tagging penguins before jumping right into the project. The object of the project is simple, click and identify adult penguins, baby penguins, eggs, and other animals. They show you a picture and you just have to click on any penguins you see. There is no time limit or amount you have to do, and you are welcome to return to the activity at any time.
The Variables and Issues
At first, I felt it strange to go ahead without understanding more about identifying adult penguins from adolescent penguins, or different types of penguins for that matter. Before the project, the only distinctions I really knew about were that baby penguins have poofy, brown fur covering their bodies and as adults, they gain the sleek black and white swim tuxedoes we typically see in photos (thanks Happy Feet). Penguin Watch designers thought of this and have implemented certain measures to counteract the general public’s lack of penguin identification knowledge.
Firstly, at the bottom of the page is a key to a) types of penguins and b) the differences between different species. Unfortunately, there is no scale to judge the different penguins, so you don’t know if the photo is small or in relative height. They also only provide one average shot of the penguins without listing particular characteristics to look out for or having other photos to compare to when determining which type of penguin you’re judging. How do you know from a penguin turned around if it has the black line of the chinstrap adult penguin versus the baby? If the photo is black and white, what am I looking for if not color? It’s also a step not blatantly listed in the directions.
Figure 1. Characterizing Penguins
Secondly, they check to make sure participants aren’t blindly looking at a photograph they don’t understand or doesn’t have any penguins in it at all (I have yet to come across a photo with no penguins).
Figure 2. An example of black and white penguin photos and the first question that asks if one can see penguins in a photo.
Thirdly, they take into account the number of penguins in the photo based on if the participant was able to mark all the penguins or not. This way, they can gauge how well the photo has been labeled. If there’s few penguins, it’s easier to get an accurate account versus a photo with a lot of penguins. They also darken portions of photos that are far away and contain a large mass of penguins.
Figure 3. Are all the penguins marked?
Fourthly, they put a warning to participants about having to mark all penguins in a photo once the counter hits 30 per single category. This step also makes it aware that you are not the only participant marking this photo, which alludes to their check to use many participants to get a more accurate count. This also makes it apparent why participants would need to mark the penguins’ center mass to hopefully overlap with numerous other markers by other participants. Having to only mark 30 (or more if you are so inclined) is good on an interactive stand point because once an activity becomes monotonous, people get bored and can get sloppy. Keeping the activity changing is important, especially so participants keep going. It’s also hard to see individual penguins when there are so many markers. The issue may be that some penguins in more concentrated areas may be miscounted in the photo.
Figure 5. Notification if 30 penguins are marked in a given photo.
Lastly, there’s an option to discuss or leave comments for researchers and other participants.
Figure 6. Discussion option.
Social media is utilized by the project to spread word about their work and to excite and thank participants for donating their time and energy. They currently have a Facebook page, Twitter handle, and Google Plus account. Recently, they issued a competition to get as many friends liking their page as possible in order to spread awareness and interest in their project. They also make funny posts like the following:
Figure 7. Twitter post.
Overall Penguin Watch is a relatively simple citizen science project that is great for anyone who enjoys looking at cute penguins and doing a little eye-spy detective work. The directions could use a bit of improvement, but overall, it seems to be a beneficial project that is liked my many participants and is easy to do. It only takes a few minutes to complete a photo or two so people can do it on a break or in the morning (as one participant said on a Facebook post).
