🔭 astrophysics

Line-of-sight shear in SLACS strong lenses II: validation tests with an extended sample

This paper extends the analysis of line-of-sight shear in SLACS strong lenses to a total of 45 systems, revealing a significant mean shear magnitude of $0.11\pm 0.024$ with many lenses exhibiting unexpectedly large shears that cannot be attributed to observational factors or lens mass model complexities.

Original authors: Natalie B. Hogg, Daniel P. Johnson, Anowar J. Shajib, Julien Larena

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Natalie B. Hogg, Daniel P. Johnson, Anowar J. Shajib, Julien Larena

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

Imagine the universe as a giant, slightly wobbly funhouse mirror. When light from a distant galaxy travels toward us, it doesn't always travel in a straight line; it gets bent by the gravity of massive objects in its path, like giant elliptical galaxies. Sometimes, if the alignment is perfect, this bending creates a cosmic "ring" or multiple images of that background galaxy. This is strong gravitational lensing, and it's a powerful tool for astronomers to weigh the invisible dark matter holding these galaxies together.

But here's the twist: the universe isn't empty space between us and those galaxies. It's filled with other galaxies, clusters, and clumps of matter along the same line of sight. These extra bits of mass act like subtle, invisible hands that stretch and squeeze the light, distorting the shape of the lensed images. This distortion is called shear.

In this new study, the authors (Hogg and team) decided to play detective with a specific collection of 50 cosmic lenses known as the SLACS sample. They had already solved the puzzle for 23 of them in a previous paper (Paper I). In this follow-up, they tackled the remaining 27. They successfully modeled 22 of these new lenses, bringing their total solved cases to 45.

The Big Surprise: The "Too-Strong" Stretch
When the team measured the shear in these 45 lenses, they found something strange. The average amount of stretching they measured was 0.11 ± 0.024 for the new batch, and 0.085 ± 0.019 when they combined it with the previous 23.

To put that in perspective, if you imagine the universe as a calm pond, these measurements suggest the water is rippling much more violently than the standard "calm pond" simulations predicted. The authors found that a significant chunk of these lenses had shear magnitudes greater than 0.1, which is surprisingly large. It's as if they expected the mirror to be slightly warped, but instead, they found it was being pulled by invisible giants.

What They Ruled Out (The "Not the Culprit" List)
The team didn't just measure the stretch; they tried to figure out why it was so strong. They tested several suspects, but the evidence cleared them all:

The "Octupole" Suspect (The Weird Shape): One leading theory was that the main lensing galaxies might have a weird, boxy, or disk-like shape (called an "octupole") that was confusing the math, making the shear look bigger than it was. The authors tested this by adding an octupole shape to their models.
- The Verdict: Nope. For the most part, adding this weird shape didn't fix the problem. In fact, it often made the math break down or led to impossible results. In only one single case did the tension with the simulations drop significantly, but even then, the shear was still too high. The paper explicitly states there is no systematic evidence that an unmodeled octupole is the reason for these large shear values.
The "Bad Data" Suspect (Filters and Cameras): Maybe the cameras were just playing tricks? The team checked if using a specific camera filter (F606W vs. F555W) or the specific blur of the telescope (the Point Spread Function, or PSF) caused the high numbers.
- The Verdict: Nope. While they did find a statistical quirk where lenses observed in the F606W filter were more likely to be in the "high shear" group, their simulations showed that the difference in how the telescope blurs the image (the PSF) does not statistically explain the huge shear values. The blur isn't the culprit.
The "Distance" Suspect (Redshift): Maybe the lenses with high shear are just further away, giving the light more time to get stretched by more stuff?
- The Verdict: Nope. They compared the distances (redshifts) of the high-shear lenses against the low-shear ones and found no difference. They are drawn from the same distribution.
The "Brightness" Suspect (Flux and Noise): Maybe the bright, noisy images were just harder to measure?
- The Verdict: Nope. While the high-shear lenses did happen to be brighter, the signal-to-noise ratio (how clear the image is compared to the background static) was actually the same for both groups. Brightness alone isn't the magic ingredient.

The Mystery Remains
So, what's going on? The authors are left with a puzzle. They have measured these large shears (|γLOS| > 0.1 in many cases), and they have ruled out the usual suspects: the shape of the galaxy, the camera filters, the distance, and the image noise.

They also checked if the direction of the stretch (shear) matched the direction of the galaxy's shape. They found that the shear and the galaxy's shape are randomly oriented, like two people spinning in a room without bumping into each other. This contradicts some other studies that suggested they should line up.

The Bottom Line
The paper concludes that there is no obvious feature in the photos or the basic setup that explains why these shear values are so unexpectedly large. The answer isn't a simple "bad camera" or "weird galaxy shape."

The authors suggest the real answer might lie in the complex, messy details of how the lens and the source galaxy are modeled—details that current computer models might still be missing. It's a reminder that even when we think we've reconstructed a cosmic image down to the very last pixel of noise, there might still be a hidden layer of complexity we haven't figured out yet. The hunt for the true cause of these "super-stretches" continues, waiting for better models and more data from future telescopes.

Technical Summary: Line-of-Sight Shear in SLACS Strong Lenses II: Validation Tests with an Extended Sample

Problem Statement
Strong gravitational lensing images are subject to shape distortions caused by inhomogeneities along the line of sight (LOS). While the leading-order distortion is shear, which offers a potential complementary cosmological probe to traditional cosmic shear, its measurement is complicated by degeneracies with the main lens mass model. A significant portion of measured shear in strong lenses often exhibits magnitudes larger than expected from N-body simulations. Previous work (Hogg et al. 2026, hereafter Paper I) modeled 23 SLACS lenses and found that including an octupolar distortion (representing "boxyness" or "diskyness" in the lens mass) did not systematically suppress these unexpectedly large shear magnitudes. This paper addresses whether the large shear values found in the initial sample are systematic artifacts of specific data properties (e.g., redshift, filter choice, PSF, flux) or if they persist across a larger, independent sample of lenses.

Methodology
The authors extended the analysis from Paper I to an additional 27 strong gravitational lenses from the SLACS catalogue, selected for their simple source morphologies and lack of satellite galaxies.

Modeling Framework: The study utilizes the dolphin package (powered by lenstronomy) for semi-automated forward modeling. Each deflector is modeled with an Elliptical Power Law (EPL) mass profile and a double elliptical S'ersic profile for the lens light. The source is modeled using an elliptical S'ersic profile and a basis set of Gaussian shapelets.
Shear Parameterization: The analysis employs the "minimal LOS shear model" (Fleury et al. 2021), which re-parametrizes the lens equation to isolate the LOS shear ( $\gamma_{LOS}$ ) from the main lens parameters, aiming to break degeneracies between the main lens mass and external perturbations.
Statistical Analysis: Parameters are optimized via particle swarm optimization, followed by joint posterior sampling using the affine-invariant ensemble Markov chain Monte Carlo (MCMC) sampler emcee.
Validation Tests: The authors investigated correlations between the measured $|\gamma_{LOS}|$ $∣ γ_{L O S} ∣$ and various data properties:
- Lens and source redshifts.
- Sky location (clustering).
- Observational filters (F555W vs. F606W) and Point Spread Function (PSF) characteristics.
- Flux and Signal-to-Noise Ratio (SNR) in the lensed arcs.
- The inclusion of an octupolar distortion in the lens mass model for a subset of lenses.

Key Results

Sample Expansion: The authors successfully modeled 22 out of the 27 new lenses, bringing the total sample of successfully modeled SLACS lenses with LOS shear constraints to 45. Five lenses were excluded due to complex morphologies, blending of lens/source light, or MCMC convergence failures.
Shear Magnitudes: The mean LOS shear magnitude for the 22 new lenses is $0.11 \pm 0.024$ . When combined with the 23 lenses from Paper I, the mean magnitude for the full sample of 45 is $0.085 \pm 0.019$ . A significant fraction of lenses (12 out of 22 in this work) exhibit shear magnitudes $|\gamma_{LOS}| > 0.1$ , which is unexpectedly large compared to N-body simulation expectations.
Octupole Test: Re-fitting the lenses with an octupolar distortion (EPL BOXYDISKY profile) resulted in pathological model parameters (e.g., unphysically shallow EPL slopes) or non-convergence for most systems. For the few successful cases, the inclusion of the octupole did not systematically reduce the shear magnitude or the tension with simulation expectations.
Orientation Analysis: Unlike previous findings in simulations and other datasets, this study found no preferential alignment or anti-alignment between the LOS shear position angle and the lens mass ellipticity or lens light ellipticity. The relative orientations are consistent with a random distribution.
Data Property Correlations:
- Redshift/Sky Location: No significant difference in redshift distributions or sky clustering was found between lenses with "large" shear (above median) and "small" shear.
- Filters: A statistically significant correlation ( $p=0.042$ ) was found between the F606W filter and large shear magnitudes (95.7% of large shear lenses were in F606W). However, simulations comparing F555W and F606W PSFs showed no statistically significant difference in shear recovery bias ( $p=0.58$ ), suggesting the correlation arises from image resolution and feature separability rather than PSF-induced bias.
- Flux/SNR: Lenses with large shear have significantly higher mean flux in their arcs ($43.8$ vs $22.2 $e$ ^-$/s), but this is counteracted by higher background RMS, resulting in no significant difference in SNR distributions between the two subsets.

Significance and Conclusions
The paper confirms that unexpectedly large LOS shear magnitudes are a persistent feature in strong lens modeling, even when expanding the sample size and testing for data-specific biases. The authors conclude that there is no obvious feature in the photometry (redshift, filter, SNR) that systematically leads to these large inferred shears.

The study explicitly rejects the hypothesis that unmodeled octupolar distortions in the lens mass are the primary driver of these large shear values, as their inclusion did not systematically suppress the shear. Furthermore, the lack of alignment between shear and lens mass ellipticity contradicts some previous simulation results.

The authors posit that the origin of these large shears likely lies within the complexities of the lens and source models themselves, which current techniques reconstruct down to the noise level. The paper concludes that determining the appropriate level of model complexity justified by the data remains a primary challenge. Future resolution of this "conundrum" is expected to come from advancements in modeling techniques, hydrodynamical simulations, and larger samples of high-quality multi-wavelength data (e.g., from the Euclid mission and COWLS catalogue), which will enable precision cosmology through cross-correlation with cosmic shear.

More like this