science Monday, September 28, 2009. Post by David Bradley
The opportunity to have hundreds of thousands of people analyze a scientific dataset allows research teams to consider research projects that would otherwise be impossible – only imagination is the limit!
That’s the ethos behind Galaxy Zoo, one of the world’s largest citizen science projects.
The Galaxy Zoo project was created to solve a specific problem – the classification of galaxies, something humans can do so much better than any machine. In early 2007, Kevin Schawinski (then a graduatestudent at Oxford University, England) was manually inspecting 50,000 galaxy images from the Sloan Digital Sky Survey (SDSS). He realised that he would need to inspect the other 950,000 images to achieve his science goals and so he enlisted the help of Chris Lintott and others to create the Galaxy Zoo website and asked members of the general public to help.
This is an entirely different approach to the likes of SETI@home, Folding@home, and other distributed computing projects which simply exploit our computers’ idle processors to crunch vast quantities of data, whether that’s extraterrestrial radio signals or protein structures.
Since the launch of Galaxy Zoo, almost a quarter of a million users have contributed over 100 million classifications of galaxies from the SDSS. version 2 was launched recently and has already collected over 40 million classification clicks in just 7 months.
The team has over fifteen papers submitted or published in scientific journals based upon the results of the original project and has had follow-up time on some of the world’s largest telescopes, including the Hubble Space Telescope, Gemini, XMM-Newton and the William Herschel Telescope. With Galaxy Zoo 2 asking the public for a more detailed classification of 250,000 SDSS galaxies, it is expected that this second incarnation will result in even more scientific output.
Some of the most exciting discoveries to date from Galaxy Zoo have been serendipitous in nature and have come as a direct consequence of human inspections of the Sloan targets. The most famous of these is ‘Hanny’s Voorwerp’, a strange green blob discovered by the Dutch school teacher Hanny Van Arkel while classifying galaxies.
This astronomical anomaly was missed entirely by the automated pipeline software of the SDSS. The Voorwerp (Dutch for ‘object’) is thought to be an extremely rare example of a quasar light echo. Further observations are scheduled on Hubble Space Telescope and the Voorwerp is being studied by numerous research teams.
In addition to working on scientific goals, the Galaxy Zoo team conducted a large-scale survey to investigate what motivates their community. The results were a pleasant surprise: the most important motivation by far was a desire to contribute to research. With this in mind the team are developing a suite of new projects for their army of citizen scientists to contribute to. Early attempts at a galaxy merger simulation have shown great promise and a recent test of a supernova-hunting website discovered 20 new supernovae in three days. As such, Galaxy Zoo has demonstrated that the citizen scientist has an important role to play in modern science. The Galaxy Zoo forum, launched after early media attention increased user interest substantially, now allows the community to answer common questions about the site.
The Galaxy Zoo team believe that the success of a citizen science project should not be measured by the number of visitors or media attention received but instead by the contribution to research.
Early papers by the team focused on assessing the quality of the results; it turns out that because each galaxy is seen numerous times by different people, not only are the Galaxy Zoo classifications as good as those of a professional astronomer, for the first time the team also have a statistical measure of the certainty of the classification.
SciScoop Science is owned and operated by David Bradley Science Writer.