Scientists, extolled by James Webb, were “a bit carried away” with his data

The James Webb Space Telescope may be revolutionary, but it is not without its flaws. Its instruments were not yet fully calibrated when the first data were made available, leading scientists to some small errors.

The general public discovered the very first images taken by the James Webb Space Telescope on July 12, 2022. The resounding announcement attracted worldwide media attention, including outside the trade press. Two days later, the first “real” data from James Webb was revealed and made available to researchers, who set about dissecting it.

It took less than a week for the first studies to be published on the ArXiv server. This server is dedicated to scientific articles that have not yet been peer-reviewed and therefore have not been published in journals. However, these papers will mostly need to undergo significant revisions before being validated. Because it turns out that data is wrong, as pointed out by the journal Nature on September 28.

First data from preliminary James Webb

We may have gotten a little carried awayacknowledges Véronique Buat, researcher at the Marseille Astrophysics Laboratory. There was so much excitement about this data that we published it very quickly, and now we have to revise most of our results. »

Could JWST have failed its mission? Absolutely not. This is a problem related to the calibration of the instruments. Before starting, the telescope’s various sensors are adjusted. But during the cycle, between the movements and then the temperature changes, everything moves and you have to start making the adjustments again in order to use all this technology properly.

This obstacle is well known by the scientific community. Every mission is entitled to its small imperfections in the beginning, and sometimes it is necessary to wait several months to have something that is really exploitable. But still three months later, the calibration of James Webb is not quite finished and the data is constantly being revised. ” We got the data uploaded very quicklysays Véronique Buat. It’s quite unusual to go that fast and we thought they were already strong enough, but they weren’t. »

Specifically, the scientist works on distant galaxies. These are the results that have been most affected, as it is extremely weak signals that require the greatest precision. Last July, Numerama told you about a study about the most distant galaxy ever discovered, and specified that it was only a pre-publication that should be taken with caution. If this study in particular remains quite cautious and has not been retracted, Véronique Buat had to review some of its results (which are no longer available online because they have become unpublished): ” We had a redshift (the redshift used to date galaxies) of 14, which is very far. But with the revision we have reached 12, which is quite a significant gap for physicists. »

Zoom in on GLASS-z13, the distant galaxy that James Webb would have discovered. // Source: NASA/CSA/ESA/STScI

At first, many of the signals corresponded to very distant galaxies, but after a re-calibration of the instruments, it turned out that these objects ended up being a little less spectacular than they appeared to be. Other studies involving closer objects had more robust results that did not need to be reworked since the data were clear enough. The question arises more for distant signals.

At the moment we have almost a new calibration every weekassures Véronique Buat, and it should take a little longer before we have reliable data. »

“We have to be careful and wait”

Usually, the first data arrives several months or even a year after the final installation of the mission. But for JWST, everything happened very quickly, surprising the community. ” Calibration can sometimes last throughout the mission, including aftersays Olivier Berné, astrophysicist at IRAP in Toulouse. We can always do better, and we know that in the beginning it is better not to overinterpret the data. »

The scientist was given the direction of one of the first programs to be studied with JWST. He has yet to publish any studies, but was tempted to see the flow of publications from the early days. ” There is a very frustrating side because we have been working on our project for years and some people are rushing to be the first to publish. But we must be careful and wait until we are sure of what we say. »

For the researcher, this precipitation is indicative of a competition problem within the scientific community. ” This is especially the case in the United States: if a researcher wants work, he must publish and be cited, which encourages him to be the first and therefore to hurry. »

JWST, in stowed position.  // Source: Northrop Grumman via Flickr (cropped photo)
JWST, in stowed position. // Source: Northrop Grumman via Flickr (cropped photo)

In France, the status of researchers is more protected. Still, that didn’t stop early releases. ” As the teams are international, there is a sense of urgencysays Véronique Buat. We have a joint Slack where there were messages 24 hours a day, from the US, from Japan… All this gives the impression that we have to move quickly, I had never experienced that in my entire career! »

What should be taken away from this case? For Véronique Buat, this confusion should not make us forget the prowess of JWST: ” An entire career can be made in a single photoassures the researcher. His results are incredible and when all his instruments are optimized we will discover a lot of things. »

For Olivier Berné, this ultimately shows that the scientific method and the peer review system are solid. ” The revisions were fast, the journals did not miss these articles, which remained in the state of pre-publication. This is proof that the system can withstand this overrun. The astrophysicist hopes that his first publication on the subject will be ready by the end of October, when the storm has subsided a bit.

Leave a Comment