Updates to the WFC3/IR Photometric Stability Stellar Cluster Study

Imagine the Wide Field Camera 3 (WFC3) on the Hubble Space Telescope as a super-precise digital camera floating in space. Like any camera, its sensor—the part that catches the light from distant stars—doesn't stay perfect forever. Over the years, scientists have noticed that the infrared (IR) channel of this camera is slowly getting a little "dimmer," meaning it loses a tiny bit of sensitivity to light every year.

This report, written by K. Huynh and V. Bajaj in May 2026, is like a detective story where the investigators go back to the crime scene to see if the "dimming" is speeding up, slowing down, or staying the same. They didn't just look at random stars; they focused on two famous cosmic "beehives" of stars called globular clusters: 47 Tucanae and Messier 4 (M4). By taking pictures of the same un-crowded patches of these clusters over and over again between 2023 and 2026, they could measure exactly how much the camera's sensitivity has changed.

The "Ghost Light" Problem
Here is the tricky part: sometimes, when the camera takes a picture of a very bright star, it leaves a "ghost" of that image behind on the sensor. This is called "persistence." If the camera takes another picture of a dim star right after, that ghost light can make the dim star look brighter than it really is.

In the past, some of the data used to measure the camera's dimming might have been tainted by these ghosts. The authors suggest that earlier measurements, which didn't account for this persistence as well, might have made the camera look like it was losing sensitivity faster than it actually is. It's like trying to measure how fast a candle burns while someone is occasionally shining a flashlight on it; you might think the candle is flickering wildly when it's actually quite steady.

What They Found
To get a clearer picture, the team looked at their new data in two ways:

1. The "Full Visit" Method: They used every single photo taken during an observation session. This is like looking at a whole album, but the later photos in the album might have those pesky "ghost" lights from earlier photos in the same session.
2. The "First Shot" Method: They only looked at the very first photo taken in each session. This is like only looking at the first page of the album, where no ghosts from previous photos in that session have had time to appear yet.

Here is what their measurements showed:

• For 47 Tucanae (using the F160W filter):

  • When looking at all photos, the camera is losing sensitivity at a rate of -0.07% ± 0.01% per year.
  • When looking only at the first photos (to avoid the ghost light), the rate is -0.06% ± 0.02% per year.
  • The Takeaway: The camera is dimming, but the rate is slower than what was measured in a previous study from 2022 (which found a loss of about -0.12% to -0.10%). The authors suggest this is because the newer observations were designed to avoid those "ghost" effects, giving a more accurate, slower rate of loss.

• For Messier 4 (M4) using the F110W filter:

  • Using all photos, the loss rate is -0.13% ± 0.02% per year.
  • Using only the first photos, the rate drops to -0.07% ± 0.03% per year.
  • The Takeaway: Again, the "first shot" method shows a much slower loss of sensitivity. The authors note that the earlier data for this filter was heavily affected by persistence, so the newer, cleaner data suggests the camera isn't fading as fast as we once thought.

• For Messier 4 (M4) using the F160W filter:

  • Using all photos, the rate is -0.10% ± 0.01% per year.
  • Using only the first photos, the rate is a very flat -0.01% ± 0.03% per year.
  • The Takeaway: The authors are cautious here. They point out that the early photos of M4 in this filter were "heavily affected by persistence" because the telescope moved in a way that kept the ghost lights in the same spot. Because of this, the results from the "all photos" method are considered "less trustworthy." The "first shot" method suggests the sensitivity loss is almost non-existent for this specific filter, but the authors don't claim this is the final answer yet.

The Big Picture
The main finding is that the rate at which the WFC3/IR camera is losing sensitivity appears to have decreased since the last major study. The authors suggest this isn't necessarily because the camera has suddenly gotten better, but because we are finally looking at data that isn't messed up by "ghost lights" from bright stars. The older, faster rates of loss were likely overestimates caused by these persistence effects.

However, the paper is careful not to say the problem is solved. The authors note that for the bluer filter (F110W), there is still a discrepancy between their new "clean" data and the correction values currently used by the telescope's software. They suggest this might just be because we don't have enough early data points for that specific filter yet, and the curve might flatten out as more time passes.

What Happens Next?
The team isn't stopping. They are continuing to monitor these star clusters every year. In fact, a new program is already scheduled to take another picture of 47 Tucanae in mid-2026. For now, the official correction file used to fix the telescope's data (called IMPHTTAB) remains unchanged, sticking to the values established in 2024, because the new data supports the idea that the older, faster loss rates were likely too high, but the new, slower rates need a bit more time to be fully confirmed across all filters.

In short: The camera is still getting a tiny bit dimmer, but it's probably not fading as fast as we feared, once we wipe away the "ghosts" from the lens.