WASHINGTON -- Whoops! Call the painter back. The color of the universe is not the turquoise that astronomers said. Try a rather ordinary beige.
Two Johns Hopkins University astronomers announced in January that they had averaged all colors from the light of 200,000 galaxies and concluded that if the human could see this combined hue, it would be a sprightly pale green. That, they said, was the color of the universe.
But Karl Glazebrook and Ivan Baldry said Thursday that their January conclusion was tripped up by flawed software that was uncovered by color engineers who checked their data.
"It is embarrassing," Glazebrook said. "But this is science. We're not like politicians. If we make a mistakes, we admit them. That's how science works."
The effect of the software error was that the computer picked a nonstandard white and mixed it with the other colors to come up with the turquoise. When the error was corrected and replaced with a standard white index, beige was the result, Glazebrook said.
"It looks like beige," he said. "I don't know what else to call it. I would welcome suggestions."
In January, Baldry called the turquoise "cosmic spectrum green." But the pair offered no fancy name for the new beige hue.
To find this average color, Glazebrook and Baldry gathered light from galaxies out to several billion light years. They processed the light to break it into the various colors - similar to the way a prism turns sunlight into a rainbow. They averaged the color values for all the light and converted it to the primary color scale seen by the human eye.
Glazebrook said the underlying data was correct; the problem came when the scientific data was converted into a hue compatible with the perception of the human eye.
The astronomer said that expressing the color for popular view was not even part of the original scientific experiment. They did it "as a lark."
"We were doing this as an amusing footnote to our paper," said Glazebrook. "Then there was huge media thing. We were completely overwhelmed. We didn't expect it to get so big."
The publicity attracted the attention of color engineers who contacted the astronomers and asked for copies of their program software. When the engineers ran the software, they concluded there was a mistake and notified Glazebrook and Baldry.
"It's our fault for not taking the color science seriously enough," said Glazebrook. "I don't like being wrong, but once I found I was, I knew I had to get the word out."
The problem was so complex, he said, that only a small number of color engineers had the expertise to determine that there was a flaw.
On the Net:
New universe color: http://www.pha.jhu.edu/~kgb/cosspec/