🔭 astrophysics

The Impact of Degraded Charge Transfer Efficiency on Extended Sources in ACS/WFC

This study analyzes seventeen years of HST/ACS/WFC data to demonstrate that degraded Charge Transfer Efficiency significantly impacts photometric accuracy and spatial symmetry of extended sources, particularly in low-background or faint observations, while establishing specific exposure thresholds and recommending the use of CTE-corrected images to mitigate these effects.

Original authors: David V. Stark, M. Chiaberge, N. A. Grogin

Published 2026-02-24

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: David V. Stark, M. Chiaberge, N. A. Grogin

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Fading Fingerprint: How Space Dust is Blurring Hubble's Photos

Imagine the Hubble Space Telescope as a world-class photographer who has been taking pictures of the universe for over two decades. One of their most trusted cameras, the ACS/WFC, is like a high-end digital sensor. But, just like a camera left in a dusty, radioactive desert, it has taken a beating. Tiny, high-energy space particles have damaged the camera's internal wiring over the years.

This damage has created a specific problem called Charge Transfer Inefficiency (CTI). To understand what this means, let's use an analogy.

The Analogy: The Bucket Brigade

Imagine a group of people (pixels) standing in a line, passing buckets of water (electrons/charge) from one person to the next to get them to the finish line (the computer).

In a perfect camera: Every person passes the full bucket. The water arrives exactly where it started.
In the damaged Hubble camera: Some people have holes in their buckets or are too tired to hold them tight. As the buckets move down the line, some water leaks out. Later, the people further down the line might accidentally pick up that leaked water and pass it along.

The result? The bucket arriving at the finish line has less water than it started with (the object looks dimmer), and there is a trail of spilled water behind it (an artificial "smear" in the image).

This report, written by astronomers in 2025, asks a crucial question: How bad is this "leaking bucket" problem for big, fuzzy objects like galaxies, and how can we fix our photos?

The Experiment: A 17-Year Time Capsule

The scientists didn't just guess; they looked at a specific galaxy cluster called CL0024+16.

The "Truth" Photo: They took a picture in 2004. At that time, the camera was relatively new, and the damage was minimal. They treat this as the "perfect" reference.
The "Damaged" Photos: They took pictures of the exact same spot in 2013 and 2021. By then, the camera had been in space for 11 and 19 years, respectively, and the "leaking" was much worse.

They compared the 2021 photos to the 2004 "truth" to see exactly how much the galaxies had faded or distorted.

The Findings: It Depends on Three Things

The study found that the damage isn't the same for every photo. It depends on three main factors:

1. The "Background Noise" (The Rainy Day Effect)

Imagine trying to hear a whisper in a quiet room versus a noisy party.

Quiet Room (Low Background): If the sky is very dark (low background light), the "leaking" is obvious. The signal gets lost in the noise, and the galaxy looks significantly dimmer.
Noisy Party (High Background): If you add a little bit of "noise" (like a post-flash LED light that brightens the whole sensor), the signal becomes stronger relative to the leak.
The Rule: To get a good photo, the background needs to be at least 20 electrons per pixel. If it's darker than that, the "leak" steals too much of the galaxy's light.

2. The Distance from the "Exit" (The Serial Register)

In our bucket brigade, the people at the very end of the line have to pass the bucket the fewest times. The people at the start have to pass it many times.

Close to the Exit: If a galaxy is near the "serial register" (the exit door of the camera), the buckets don't travel far. Very little water leaks out. The photo is accurate.
Far from the Exit: If a galaxy is on the other side of the detector, the buckets travel a long way. A lot of water leaks out. The galaxy looks dimmer, and a "tail" of spilled water appears behind it.
The Rule: For detailed studies of a galaxy's shape, it's best if the galaxy is within 512 pixels of the exit. If it's further away, the distortion gets messy.

3. The Brightness of the Galaxy (The Size of the Bucket)

Bright Galaxies: They have huge buckets of water. Even if a little leaks, the bucket is still mostly full. They are "self-shielded."
Faint Galaxies: They have tiny buckets. If even a drop leaks, the bucket is empty.
The Rule: If a galaxy is very faint (less than 300 electrons in a single exposure), the camera simply can't measure it accurately anymore, no matter what you do.

The "Magic Fix" (And Why It's Not Perfect)

The Hubble team has a software tool called DRC (CTE-corrected images). Think of this as a digital "spell" that tries to guess where the leaked water went and put it back in the bucket.

Does it work? Yes, but not perfectly.
The Catch: The software is great at fixing the total amount of light for big galaxies. However, it struggles to put the water back in the exact right spot for the center of the galaxy. This means that while the galaxy might look the right brightness overall, its center might still look dimmer than it should, and its shape might look slightly warped.

The "Smear" Effect

The report also found that this damage creates artificial asymmetries.

Imagine a perfectly round galaxy. Because of the leaking buckets, the side of the galaxy facing the "exit" looks dimmer, and the side facing the "start" looks brighter (because it's catching the spilled water from behind).
This makes the galaxy look like it's leaning or has a tail, even if it's perfectly round. This is dangerous for scientists studying galaxy shapes!

The Takeaway: How to Take Good Photos in 2025

Based on this study, here is the advice for anyone using the Hubble camera today:

Use the "Magic Fix" Software: Always use the DRC (corrected) images, not the raw ones. It's not perfect, but it's much better.
Don't Let the Sky Get Too Dark: Try to keep the background light above 30 electrons per pixel. If the sky is too dark, use a "post-flash" (a quick burst of LED light) to brighten the sensor slightly. This stops the "leak" from stealing your signal.
Aim for Bright Targets: Make sure your galaxy is bright enough (at least 300 electrons per exposure). If it's too faint, the camera can't save it.
Get Close to the Exit: If you need to study the detailed shape or center of a galaxy, try to point the telescope so the galaxy is close to the serial register (within 512 pixels). If it's far away, the distortion will be too strong to trust.

Summary

The Hubble camera is aging, and its "buckets" are leaking. But by understanding where the leak happens, how much light is lost, and how to add a little extra light to the scene, astronomers can still take stunning, reliable photos of the universe. They just need to be careful not to trust the photos of faint, distant galaxies too much without applying the right corrections!

1. Problem Statement

The Advanced Camera for Surveys (ACS) Wide Field Channel (WFC) on the Hubble Space Telescope (HST) has suffered from Charge Transfer Inefficiency (CTI) due to radiation damage accumulated over two decades in space. While the impact of CTI on point sources is well-characterized and corrected via empirical formulas and pixel-based algorithms, the impact on extended sources (e.g., galaxies) remains poorly understood.

Extended sources were hypothesized to be "self-shielded" against CTI because their internal charge density raises the local background, potentially reducing charge trapping. However, it was unclear if:

Global brightness measurements (large apertures) remain reliable.
Spatially resolved measurements (central brightness, 2D structure) suffer from significant biases.
Current pixel-based CTE correction algorithms (used to generate FLC/DRC images) effectively restore charge to its original location within extended structures.

2. Methodology

The authors conducted a comparative analysis using a 17-year baseline of imaging data for the galaxy cluster CL0024+16.

Data Selection:
- Reference ("Truth"): 2004 data (Program GO 10325). Taken early in the mission with deep exposures, minimizing CTI effects.
- Test Epochs: 2013 (11 years post-installation) and 2021 (19 years post-installation). These programs (CAL/ACS 13603 and 16870) were specifically designed with shorter exposures (4 × 500s) and variable LED post-flash durations (0s, 0.7s, 1.4s) to create a wide range of background levels (15–40 $e^-$ /pixel) to stress-test CTI effects.
Image Processing:
- Generated both non-corrected (DRZ) and CTE-corrected (DRC/FLC) images using Astrodrizzle.
- Aligned all epochs to the 2004 reference using phase cross-correlation on masked star regions to avoid proper motion bias.
Source Characterization:
- Detected sources and measured shape parameters (Kron radius, ellipticity) on the 2004 "truth" image.
- Applied identical elliptical apertures to all epochs to measure:
  - Total Magnitudes: $k=2.0$ Kron radius (global brightness).
  - Central Magnitudes: $k=0.5$ Kron radius (local/core brightness).
- Calculated magnitude differences ( $\Delta m = m_{measured} - m_{truth}$ $Δ m = m_{m e a s u r e d} - m_{t r u t h}$ ) as a function of:
  - $\Delta y$ : Distance from the serial register (number of pixel transfers).
  - Background: Local background level per exposure.
  - Fluence ( $f_{500}$ ): Total source counts in a 500s exposure.
2D Structure Analysis:
- Analyzed surface brightness asymmetries by splitting galaxies along their major axis into "read-out" (closer to serial register) and "anti-read-out" (further) halves.
- Stacked radial profiles to quantify artificial gradients induced by CTI.

3. Key Contributions

First Systematic Study of Extended Sources: This is the first comprehensive analysis quantifying how degraded CTE affects extended source photometry and morphology, distinguishing it from point-source behavior.
Validation of "Self-Shielding": Confirmed that extended sources are indeed more resilient to CTI than point sources due to self-shielding, but demonstrated that this protection is not absolute, particularly for faint regions and low backgrounds.
Evaluation of Correction Algorithms: Provided empirical evidence that while pixel-based CTE correction (DRC images) significantly improves data quality, it does not perfectly restore charge to its original location within extended sources, leaving residual biases in central brightness and 2D structure.
Quantification of Thresholds: Established specific quantitative thresholds for background, source brightness, and detector position where measurements remain reliable versus where they degrade.

4. Key Results

Global Photometry (Total Magnitudes)

Reliability: Global brightness measurements using large apertures are generally reliable within $\sim 0.05$ magnitudes if:
- Backgrounds are $> 20 e^-$ /pixel.
- Source fluence is $> 300 e^-$ per exposure.
Degradation: Below 20 $e^-$ /pixel, systematic offsets increase to $\sim 0.06$ –$0.1$ mag. For sources with fluence $< 300 e^-$ , brightness is significantly underestimated regardless of background or CTE correction.
CTE Correction: DRC images perform better than DRZ, but the correction is less effective for very faint sources ( $< 300 e^-$ ) where the algorithm avoids noise amplification.

Central Photometry (Spatially Resolved)

Significant Bias: Central brightness measurements suffer much larger deficits than global measurements.
Thresholds: To keep systematic errors $< 0.05$ mag:
- Backgrounds must be $> 20 e^-$ /pixel.
- Targets must be within $\sim 512$ pixels of the serial register.
Distance Dependence: For sources far from the serial register ( $\Delta y > 1536$ ) with low backgrounds, central brightness deficits can exceed 0.2–0.3 magnitudes in uncorrected data, and remain significant even in DRC data.

2D Structure and Asymmetry

Artificial Gradients: Degraded CTE creates artificial surface brightness gradients aligned with the read-out direction (charge "smearing" away from the register). This results in a deficit on the "front" (closer to register) and an excess on the "back."
Mitigation: These asymmetries are largely mitigated if:
- DRC images are used.
- Backgrounds are $> 20 e^-$ /pixel.
- Sources are within $\sim 1536$ pixels of the serial register.
Residuals: Even with DRC correction, significant asymmetries persist for faint sources far from the serial register in low-background data.

The Role of Background and Post-Flash

Increasing background (e.g., via LED post-flash) improves charge recovery.
For faint sources ( $f_{500} < 300 e^-$ ) in low background, increasing the background to $> 30 e^-$ /pixel recovers source counts and reduces $\Delta m$ to near zero, despite the added sky noise. This is because the recovered signal outweighs the Poisson noise increase.

5. Significance and Recommendations

This study provides critical guidelines for HST/ACS observers and data analysts to ensure scientific accuracy, particularly for studies involving galaxy morphology, stellar mass estimates, and AGN host galaxy properties.

Recommendations for Observers:

Use DRC Images: Always use CTE-corrected (FLC/DRC) images for analysis, acknowledging they are not perfect but superior to raw data.
Maintain High Backgrounds: Ensure total background per exposure is $> 30 e^-$ /pixel. If sky/dark current is insufficient, use LED post-flash.
Source Brightness: Ensure targets have $> 300 e^-$ per exposure.
Detector Placement:
- For global photometry: Placement is flexible if background/brightness thresholds are met.
- For spatially resolved/central photometry: Place targets as close to the serial register as possible (ideally $\Delta y < 512$ pixels).
Caution Zones: Be extremely cautious when analyzing 2D properties of faint sources in low-background data far from the serial register, as artificial asymmetries may mimic physical galaxy features.

Scientific Impact:
Failure to account for these effects can lead to underestimation of stellar masses, incorrect star formation histories, and biased morphological metrics (e.g., Sersic indices, concentration). The study highlights that while "self-shielding" offers some protection, it is insufficient for high-precision science in the later epochs of HST's life without strict observational constraints.