by From its orbit a million miles from Earth, the now-operational James Webb Space Telescope has finally returned its first images, including a deep-field view of thousands of galaxies, sparkling like gems billions of light-years away. But as impressive as those images are, they would be nothing more than a series of black pixels bypassing the Steven Muller Building, a modest khaki-colored brick structure hidden among the trees on the Johns Hopkins campus in Baltimore, Maryland. .
There are a few permanent features that alert the casual passerby that the building is the headquarters of the Space Telescope Science Institute (STScI), although a blue and gold banner hung above the main entrance proclaiming “Come on, Webb, come on!” provides an obvious clue. STScI began operating the Hubble Space Telescope on behalf of NASA and scientists in 1990, and the institution’s mission has now expanded to include Webb. STScI controllers helped guide the new space telescope through the deployment and commissioning process, and in early June, they began taking the first images with the big gold telescope.
And those images don’t magically appear in bright colors and balanced brightness. The raw data captured by Webb must be processed, artifact cleaned, and colorized by STScI specialists who work behind the scenes to process all Webb images published in the press over the years that the telescope does science. And it is, in many ways, an artistic as well as a technical process.
shedding light
On June 24, about two weeks before the first Webb images were released to the public, visual science developers Joseph DePasquale and Alyssa Pagan sat in their shared office surrounded by large computer screens, demonstrating how they processed the first images. from Webb transmitted to Earth. With a flick of the mouse, Mr. DePasquale took Webb’s first deep-field image, a series of glowing gems, actually thousands of impossibly distant galaxies, and returned the image to the form it came to him: a black screen. .
“The pixel values are mostly dark, because the sky is mostly dark, and only the brightest regions show up when you first see it,” he said. Mr. DePasquale and Mrs. Pagans’ task is to use a software package to increase the brightness of the image to allow people to see the darkest details, without washing out the bright regions. “All this information is hidden here, because it’s really very tenuous.”
With a few more clicks on the keyboard, DePasquale increases the brightness in a process known as “scaling” the data, revealing a grayscale version of Webb’s deep field. Adding color comes in a later step, but that has to wait until Mr. DePasquale deals with another issue presented by scaling the image to make it bright enough to view.
“Bright stars in Webb will tend to become saturated to the point where the detector is no longer providing valid information,” said DePasquale. “When that goes down the pipeline, you end up getting a black hole in the center of a bright star.”
This effect can be seen in Webb’s image released July 6 as a sneak peek, an orange star field captured by the space telescope’s guidance instrument. In the center of the bright, spiky stars are black circles that look like holes burned through a film negative.
“We were sweating this out as we got closer and closer to the [Webb image release] date,” DePasquale said, but he eventually found a computer script that would fill black holes with the values of neighboring pixels. It’s the kind of novel solution required with Webb data, he adds, because unlike the familiar workflow for developing images from Hubble data, with Webb “the process is changing right now because everything is new.”
Which Webb first?
Webb’s deep field image was the first of five images selected by STScI and NASA to show the tangible results of the more than 20 years and $10 billion it took to design, develop, build, test, launch, deploy, set up and launch the most sophisticated telescope ever built. US President Joe Biden previewed the deep-field image of the White House on July 11, while the remaining four images were revealed the following morning via NASA’s website. The complete set of images includes the deep field, spectrum, or pattern of light filtered through the atmosphere of exoplanet Wasp 96 b, and images of the Carina Nebula, the South Ring Nebula, and Stephan’s Quintet, a collection of five galaxies locked in a tight gravitational dance.
But as of June 24, which images the public would see first, and what exactly they would look like, remained a matter of discussion.
“The letter we have is to show the world that the observatory is ready to do science, to celebrate that it is ready to do science,” said Klaus Pontoppidan, associate astronomer at STScI. He was one of a dozen people in a small conference room on June 24 to discuss the images to be released.
“Hardly anyone else in this building or even at NASA has seen this,” added Dr. Pontoppidan. “It’s just this room.”
The small group had been meeting most mornings throughout the month to discuss the latest images processed by Mr. DePasquale and Mrs. Pagan and displayed on a huge monitor hanging on the wall. On June 24, the discussion turned to which version of the Carina nebula image would be made public, an image taken with Webb’s near-infrared instrument, NIRcam, or his mid-infrared instrument, MIRI.
While the NIRCam image highlighted orange and gold dust clouds, MIRI peered through the dust to reveal more stars, but the gas clouds were shown in shades of blue-gray against a red “sky,” a controversial aesthetic.
“To me, the grayish blue, the way it turned out in the MIRI image, that’s not attractive,” was one of many overlapping comments in the room.
But there was a third option presented by Mr. DePasquale and Ms. Pagan: a combination of NIRCam and MIRI images, a combination of perspectives that preserve the contrast of the MIRI image while overlaying the many details and impressive colors of the image. NIRCam.
“It’s like the best of both worlds,” said Ms. Pagan.
The group finally settled on Carina’s combined image, which is what the public saw on July 12.
color coordination
But creating Carina’s image highlights another way in which creating viewable images from Webb’s data is a creative process in its own right, especially when it comes to color processing.
Go back to the fact that most of Webb’s raw images are essentially blank to the human eye. The distant objects he captures are, in many cases, incredibly faint, too faint to register on the color-sensing cone cells in the human eye. That’s usually true even with less exotic astronomical observations.
“Look through a telescope at a planet like Jupiter or Saturn, and it looks almost black and white, because the light is so dim that it actually only activates the rods in your eyes and not the cones,” DePasquale said. “You don’t really get color information.”
In Webb’s case, add to that the fact that the telescope sees only in infrared, wavelengths of light too long for human eyes to see, no matter how bright. So, to make Webb’s images visible, Ms. Pagan and Mr. DePasquale must transpose frequencies of light invisible to human eyes into the visible part of the spectrum.
“Telescopes are designed with filters to separate the different colors and then we assign those colors chromatically,” he said. “The shorter wavelengths of light are assigned to the blue colors, then you go from blue to green to red as the wavelength increases.”
That’s a system that worked well for Hubble, which only saw in the near infrared, and so far it seems to work well for Webb’s NIRCam, according to Ms Pagan.
“But when we go into the mid-infrared with MIRI, what we get is very different, which is challenging,” he said. To avoid garish color combinations like Carina’s MIRI image, they had to get a little creative with the color mapping, “so it could be red, orange, and cyan” instead of red, green, and blue.
The process could be completely different for scientists using Webb to study a particular aspect of a distant object, Ms Pagan noted. Instead of trying to transpose non-visual wavelengths of light into the visible spectrum in a way that makes visual sense, a researcher could request color enhancement based on some phenomenon of interest, such as clouds of organic gas. Researchers can also use the services of your office when publishing the results of their research with Webb.
“There is a web page for scientists to submit their proposals for a press release,” said Mr. DePasquale. “You can go down that avenue, contact the news bureau here, and then we’ll determine if it’s really worth it for the press. If so, it’s up to us to process the data.”
Rendering can be a lot of work, especially with Webb (Carina Nebula image development took 16 hours) and Pagan and DePasquale worked through weekends in the days leading up to the release of the first Webb images. But the work is also so captivating that they would have processed the new images even without the urgency of the impending public release.
“The first data set came in on a Saturday morning and I had to drive to Philly for a family party,” DePasquale said. I’m at the party. And I’m like, ‘I just want to work on that image.'”
