[论文解读] Calibrating redshift distributions at $z>2$ with Lyman-$α$ forest cross-correlations

Qianjun Hang, Laura Casas+4 more|arXiv (Cornell University)|Jan 23, 2026

Galaxies: Formation, Evolution, Phenomena被引用 0

一句话总结

论文证明 Lyα 森林互相关可以在 DESI 和 LSST 风格的 mocks 下校准高 redshift 尾部（2<z<3）的光度测量星等分布，在现实 continuum 拟合情形下实现对平均红移的高精度约束。

ABSTRACT

We explore the feasibility of using Lyman-$α$ (Ly$α$) forests to calibrate the ensemble redshift distribution of the high-redshift tail ($2

研究动机与目标

• Motivate the need to calibrate the redshift distribution tails of photometric samples at z>2 for Stage IV surveys.
• Develop a theoretical framework to model angular cross-correlations between Lyα forest and photometric galaxies when the usual n(z) estimator is inapplicable due to redshift-space distortions.
• Assess how continuum fitting, noise, and contaminants affect the SNR and precision of clustering redshifts in this regime.
• Forecast the performance of Lyα-based clustering redshifts with DESI Lyα forests and LSST-like galaxies under different binning and scale choices.

提出的方法

• Generate multi-tracer mock catalogs with CoLoRe for galaxies (z in [0,3]) and Lyα forests (z in [1.8,3]); include DESI-like spectra with and without contaminants.
• Simulate Lyα forest fluctuations with LyaCoLoRe, calibrating Lyα bias bF(z) and RSD parameter βF from the mocks.
• Apply two continuum fitting methods, Picca and LyCAN, to recover the fluctuating Lyα transmission field, and compare to true continuum cases.
• Bin Lyα fluctuations in redshift with Δz and compute the angular cross-correlation w_gI(θ) with photometric galaxies.
• Model the cross-correlation using a 3D ξ_gF with redshift-space distortions, incorporating galaxy bias evolution b_g(z) and Lyα bias/beta.
• Use two models for bn(z)=b_g(z)n_g(z): a shift in mean (δz) and a Gaussian Process-based approach (with a template bn^T).

Figure 1 : Redshift dependent quantities adopted in validation. The left axis shows redshift distributions of the galaxies, $n_{g}(z)$ , and of the Ly $\alpha$ forest using the raw mocks $n_{F,{\rm raw}}^{I}(z)$ , and the LyCAN mocks $n_{F,{\rm LyCAN}}^{I}(z)$ , with $\Delta z=0.1$ averaged over 10

实验结果

研究问题

• RQ1 Can Lyα forest cross-correlations calibrate the mean redshift of high-z photometric samples (2<z<3) to sub-percent levels?
• RQ2 How do continuum fitting, noise, and contaminants influence the signal-to-noise and accuracy of clustering redshifts in the Lyα–galaxy cross-correlation framework?
• RQ3 What is the impact of angular scale, bin width Δz, and survey area on the SNR and mean-redshift constraints?
• RQ4 How can the cross-correlation be modeled when the conventional n(z) estimator is invalid due to strong redshift-space distortions in Lyα?
• RQ5 What are practical bias-mitigation options for bn(z) in this high-z regime?

主要发现

• With LyCAN baseline continuum fitting, the cross-correlation signal reaches 24σ at θ~10 arcmin and Δz=0.1.
• If the shape of the redshift distribution and galaxy bias evolution are well known for z<2, the method can constrain the mean redshift to σ_z/(1+z̄)=0.006 at z̄=2.
• Continuum fitting strongly impacts the SNR of the measurements.
• The Lyα cross-correlation remains a reliable method to calibrate the high-redshift tail of Stage IV surveys under realistic DESI+LSST conditions.
• Modeling shows that the cross-correlation depends on bn(z)=b_g(z)n_g(z) and can be captured via a shift or a Gaussian Process template.
• The study demonstrates feasibility and provides a framework for using Lyα forest cross-correlations for high-z redshift calibration.

(a) Ly $\alpha$ forest flux transmission fluctuation field.