The MoGO app helps people report wildlife affected by the Gulf oil spill (Image: UMass Amherst)
Many people around the Gulf of Mexico have been asking themselves if there’s anything they can do to help the situation, as oil continues to drift on shore, and spew from the seafloor. A new iPhone app offers a possible answer, one that doesn’t involve the same kind of comittment as joining a crew cleaning beaches or birds.
Called MoGO - for Mobile Gulf Observatory - it asks users to take photos of any oiled wildlife and tar balls they see on the shore and prompts them to label what they see. For example, snapping a photo and labeling it as a bird brings up a screen asking if it is alive, injured or dead. The apps then sends the photo and that information, along with its location, to a database maintained at the University of Massachusetts, Amherst.
All that information is made available to NOAA’s Wildlife Hotline, to help them keep track of the extent of harm to wildlife and coordinate rescue efforts. The MoGO app will also connect users, at the press of a button, to the Hotline if they think an animal needs particularly urgent help.
UMass Amherst computer scientist professor Deepak Ganesan and colleagues built the app in just a week using a tool called mCrowd that his group created to enable rapid development of crowdsourcing mobile apps. This platform can also connect to web-based services like Amazon Mechanical Turk, so that people at home can help by solving simple tasks.
“There’s a lot of redundancy with this kind of app,” said Yan, “you get photos with misidentified species, for example.” It would be possible to use Mechanical Turk to have people screen submitted photos for obvious errors, for example if a misplaced screen tap tagged a bird as a turtle, he says. A similar approach could let wildlife experts make more specific judgements, for example about specific species in a photo.
Being able to connect users sat in front of home computers with an evolving event like the Gulf oil spill, via cellphone users, makes this kind of citizen science much more scaleable. “You can’t have a single user look at all of them, so you need the crowd,” Yan said.