Extracting intrinsic alignments in the Dark Energy Survey's year 1 data, using the self-calibration method and LSST-DESC tools

summary

The gist

We present an extension to the Vera C.

In short

Researchers extended a weak lensing pipeline to estimate intrinsic alignment (IA) contamination in galaxy two-point correlations using a self-calibration method from Zhang (2010). Testing on DES year one data showed non-zero IA signals in higher redshift bins, though separation between different IA components remained uncertain due to limitations in photometric redshift estimation. The work provides the first implementation of this self-calibration software for LSST-DESC tools.

Key concepts

Intrinsic Alignment (IA)
This signal occurs because galaxies are physically aligned with the large-scale structure of the universe. This physical alignment manifests as a systematic bias in weak lensing measurements, making it difficult to distinguish between true cosmic shear and this intrinsic alignment effect.
Self-Calibration Method
This technique is used to obtain model-independent estimates of IA by separating different contributions to the two-point correlation function. It works by analyzing the asymmetry of the lensing signal based on whether galaxies are ordered as sources or lenses, helping to isolate specific IA components.
Quality Parameter Q (Q)
This parameter measures how accurately researchers can estimate the redshift of individual galaxies within a specific cosmic bin. A high Q indicates good redshift estimation quality, which is crucial for reliably separating different types of intrinsic alignment signals in the analysis.

Terminology used across episodes

This episode discusses

The paper

Extracting intrinsic alignments in the Dark Energy Survey's year 1 data, using the self-calibration method and LSST-DESC tools · Read on arXiv

Eske M. Pedersen, Leonel Medina-Varela, Emily Phillips Longley, Mustapha Ishak, Joe Zuntz, Chihway Chang, C. Danielle Leonard

School of Mathematics, Statistics and Physics, Newcastle University · Department of Physics, The University of Texas at Dallas Department of Physics, Institute for Astronomy, University of Edinburgh Department of Physics, Duke University Department of Astronomy and Astrophysics, University of Chicago Department of Physics Harvard University

We present the implementation of a Self-Calibration of Intrinsic Alignments of galaxies as an extension to the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) Dark Energy Science Collaboration (DESC)'s weak lensing 3x2pt pipeline (TXPipe). As a demonstration, we have run this pipeline on the Dark Energy Survey (DES) year one data set. We find indications of a non-zero intrinsic alignment signal in the higher redshift bins, while in the lower bins our results look more uncertain. We believe this is caused by known issues with the individual galaxies photo-z estimation. This effect is particularly harmful for the self-calibration method, since it has high requirements for reliable estimation of the photo- z s, and the need for individual galaxy point estimates and tomographic binning to match. We show how different methods of recreating the redshift probability distribution can affect the detection of intrinsic alignment.

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: We're looking at the paper: "Extracting intrinsic alignments in the Dark Energy Survey's year 1 data, using the self-calibration method and LSST-DESC tools".

Jocelyn: Here's the gist of the paper. We present an extension to the Vera C.

Vera: Rubin Observatory’s Legacy Survey of Space and Time (LSST) Dark Energy Science Collaboration (DESC)’s weak lensing 3x2pt pipeline (TXPipe) that will attempt to utilize the self-calibration method suggested in Zhang (2010) to achieve a model independent estimate of the intrinsic alignment signal contaminating two-point correlations.

Jocelyn: As a demonstration, we have run this pipeline on the Dark Energy Survey (DES) year one data set and find indications of a non-zero intrinsic alignment signal in the higher redshift bins while in the lower bins our results look more uncertain.

Vera: The Problem with Weak Lensing and Intrinsic Alignment Weak lensing is plagued by systematic effects, including photometric redshift estimation and the intrinsic alignment of galaxies.

Title and authors: Vera: Moving on to the title of the paper, "Extracting intrinsic alignments in the Dark Energy Survey's year one data, using the self-calibration method and LSST-DESC tools" <ref:2601.10314#pg1,Extracting intrinsic alignments in the Dark Energy Survey's year 1 data>. It’s pretty descriptive about what they did.

Jocelyn: It clearly signals that they are taking existing DES data and applying a new technique involving self-calibration, which is then extended to tools for the future LSST Dark Energy Science Collaboration.

Subrahmanyan: The focus on intrinsic alignments in year one data is important because it’s testing this method against real observational constraints before we apply it to the full LSST dataset.

Vera: They are essentially showing how to handle that contamination—that intrinsic alignment—by using a self-calibration technique so they can get a model independent estimate for those two-point correlations.

Jocelyn: It’s not just measuring the effect; it’s building an extension to the TXPipe pipeline, which is where the actual analysis happens, taking catalogs of galaxy positions and shapes along with survey information.

Subrahmanyan: So they are tackling the systematic error budget head-on by trying to mathematically separate the clustering signal from that intrinsic alignment signal.

Vera: The implications are that if this self-calibration method works robustly, it gives us a way to get cleaner measurements of cosmic shear, which is how we study the growth of structure in the universe.

Jocelyn: It means we can reduce the dependence on assuming a specific model for intrinsic alignment when fitting cosmological parameters.

Subrahmanyan: That’s the goal; less dependence on assumptions and more reliance on what we measure directly from the data, which is what matters most for cosmology.

The paper's summary: Vera: So to summarize the paper, they present this implementation of a self-calibration of intrinsic alignments in DES year one data that uses LSST-DESC tools.

Jocelyn: They are extending the existing TXPipe pipeline to robustly estimate the three typical two-point correlations: position-position, position-shape, and shape-shape shear.

Subrahmanyan: Those three correlations are essential because they help break degeneracies that other probes might have in our measurements of cosmic shear.

Vera: They introduce the self-calibration method to focus on the term gamma I G i, gamma I j being zero and instead focus on the IG term, which is what they shorten to Ig.

Jocelyn: They write out how they get the two-point correlation of galaxy position and shape as a sum: delta g i, gamma obs j = delta g i, gamma G j + delta g i, gamma I j + noise, where delta g i is the galaxy number count over-density.

Subrahmanyan: The way they relate the galaxy overdensity to the matter overdensity usually by a galaxy bias parameter see Pandey et al. (two thousand twenty) <ref:2601.10314#pg1>. That connection is crucial for connecting the galaxy measurements back to cosmology.

Vera: They then discuss two main methods of alleviating intrinsic alignment: modeling and parameterizing, which includes the nonlinear-alignment model or NLA, like what Abbott et al. (two thousand eighteen) used <ref:2601.10314#pg1>.

Jocelyn: But they are specifically using the self-calibration method here to try and isolate the Ig signal from that IA-galaxy lensing correlation.

Subrahmanyan: The paper is testing if this separation actually works better than just applying a pre-defined model, which is a major methodological question.

The paper's improvements: Vera: The authors suggest some key improvements to the original work. They point out that their method has high requirements for reliable estimation of the photo-zs, and needs individual galaxy point estimates and tomographic binning to match.

Jocelyn: That means the method isn't plug-and-play; it’s sensitive to how well you manage those redshift estimates, which is a major hurdle they have to overcome.

Subrahmanyan: And that sensitivity directly relates to the quality parameter Q, which they define as the ratio between two correlation measurements used to assess how well we are able to estimate galaxies redshift within the tomographic bin.

Vera: They suggest using this quality parameter Q as a predictor for whether their self-calibration algorithm is successful, essentially using it as an indicator of success or failure.

Jocelyn: So they are proposing an AI module that can estimate this quality parameter Q and use it to predict the success of the self-calibration algorithm. That sounds like an interesting way to manage the inherent uncertainty in their PDF estimation.

Subrahmanyan: From a theoretical standpoint, having that predictor module could be very powerful because it moves us away from just hoping the method works and gives us a more quantitative measure of its reliability.

Conclusion: Vera: So to wrap up this paper, they found that while the isolated Gg signal aligns fairly well with theoretical expectations, there are more issues between the Ig signal and the NLA model.

Jocelyn: They estimate the potential cross-bin two-point correlation for intrinsic alignment is one point eight six times ten-five sigma using Q, and zero point four eight sigma using eta.

Subrahmanyan: That means while their data looks similar to the model, the uncertainties are too large to make any additional statements about that cross-bin signal being significant right now.

Vera: They are presenting this as the first implementation of LSST-DESC’s self-calibration of intrinsic alignment software and plan to apply this pipeline on other precursor surveys and simulated data sets as verification of its methodology.

Jocelyn: So the main message is that they have a new software extension ready, and they're moving toward testing it on more data to prove its methodology works in practice.

Subrahmanyan: For us, it means this method needs rigorous testing on those simulated sets to really build confidence before we can expect cleaner constraints from real survey data.

More episodes

← Home