Helping Experts Find Planets

Now you can read us on your iPhone and iPad! Check out the BTRtoday app.

It’s not easy to measure wind on Mars. That may come as no surprise, but if humanity hopes to land on the red planet, understanding wind and climate patterns on its surface will be of vital importance.

As of now, scientists use images of Mars’ surface to identify fans and blotches, which indicate wind direction and speed. By studying the wind patterns and how certain blotches form, scientists can gain a greater understanding of Mars’ climate.

The task seems simple enough—look at a bunch of images, identify the different kind of blotches on the Martian surface, and move on. The problem is that there are so many images to inspect, it would take a small group of researchers years just to get through them. Computers are an obvious solution, but can’t actually recognize patterns as efficiently as humans.

Where there is a problem in the way of manpower, the Internet provides solid respite. With more than 1.5 million registered users, Zooniverse is a citizen science platform that allows average individuals with an internet browser to contribute to large, encompassing projects like Planet Four–which asks volunteers to scan images to gain a better understanding of Mars’ climate.

Meg Schwamb, an astronomer at the Institute of Astronomy & Astrophysics at Academia Sinica in Taiwan, is a member of Planet Four’s science team and explains how our innate ability of pattern recognition is ideal for these kinds of projects.

“The reality is that machines are great and computers can do a lot of things,” Schwamb tells BTRtoday, “but there are many things we’ve developed from hunting and gathering that make the human brain the best at pattern recognition and noticing what doesn’t belong, because we had to do that to survive.”

Between some of its lofty concepts and the borderline illegible jargon of its dense studies, many avenues of science seem unattainable to those of us with no formal training. When we think of scientists, we envision people that went to school for years, performing experiments and studying loads of information about complex ideas and their effects on our species, our planet, and the universe that encapsulates us.

Given these preconceptions about the intelligence and experience levels required to participate in scientific inquiry, leaving gobs of data to randomized, untrained individuals seems counterintuitive. Zooniverse aims to break down the complexity into digestible bits that maximize common understanding, as well as the number of able participants. Some projects present tasks that are so simple—such as Sunspotter, which asks users to identify which image is more complex—that even elementary schoolers can complete them with ease.

“The way we design our projects is to boil the task down to basic pattern recognition that everyone is capable of, no matter their scientific education,” Zooniverse communications lead Grant Miller tells BTRtoday. “We take consensus answers from multiple users to assure that any individual mistakes are ironed out.”

This participation helps researchers twofold, because not only are human beings better at identifying patterns than machines, but the collective contribution of volunteers can actually outweigh the results of an expert–both in volume and accuracy.

“If I do this task, it might depend on whether I’ve had a cup of coffee, whether I’ve already classified a thousand images, whether I’m feeling under the weather,” she explains “That might bias how I complete the task, but when you combine the results of many people, you get rid of this effect. You keep a consistency, and you typically outperform machine algorithm proficiencies.”

One of the most well-known Zooniverse projects is Planet Hunters which—you guessed it—aims to locate exoplanets orbiting around distant stars located by NASA’s Kepler spacecraft. Like many other Zooniverse projects, the premise is based around the human brain’s capability to recognize patterns, and in this case to identify the drop in light characteristic of a transiting planet.

According to Miller, there have been 10 published discoveries and many more candidates uncovered by volunteers that are waiting to be confirmed.

“Some of these discoveries have been very important and unique,” Miller tells BTRtoday, “including the first ever planet found in a four-star system.”

That planet is PH1b, a Neptune-sized giant about 5,000 light years away from Earth. It was the first confirmed discovery of the project. Planet Hunters has also uncovered a planet in a system alongside six others, marking the largest planetary system known to man outside of our own.

Most of the Zooniverse projects have accompanying blogs which allows researchers on the science teams to share progress with volunteers, and discussion tools on the respective project sites also make direct contact possible.

“As a science team, we’re on there talking to volunteers, and they’re talking to us,” Schwamb says. “We are really trying to bring everyone that contributes along with us and keep updates of what we’re doing, and how the project’s going.”

Schwamb, who is on the science team for two other Zooniverse projects, believes it brings people in at the ground floor of science. Participants get to witness the full process, from forming a hypothesis to gathering data, testing that hypothesis, and figuring out where things go right and wrong. She also believes there’s a practical nature for the public as well, both from an understanding and economic point of view.

“People pay tax money for these NASA missions and these telescopes, so I think it’s a nice way to give people access to this data in a way that’s understandable,” Schwamb says. “There are lots of public archives where you can download NASA images and spacecraft data, but you have to know how to process it.”

Given the amount of time we waste toiling around on the Internet—U.S. consumers spend more than 40 minutes on Facebook per day—Zooniverse offers a productive alternative to spend our time online, while also exercising some of the abilities of our brain more efficient than machines.

“It’s amazing to think how complex the human brain is, and we don’t even think about it,” Schwamb says.