Probing The Dark Matter Halo of High-redshift Quasar from Wide-Field Clustering Analysis

High-redshift quasars have been an excellent tracer to study the astrophysics and cosmology at early Universe. Using 216,949 high-redshift quasar candidates ($5.0 \leq z < 6.3$) selected via machine learning from the Legacy Survey Data Release 9 and the Wide-field Infrared Survey Explorer, we perform wide-field clustering analysis to investigate the large-scale environment of those high-redshift quasars. We construct the projected auto correlation function of those high-redshift quasars that is weighted by its predicted probability of being a true high-redshift quasar, from which we derive the bias parameter and the typical dark matter halo mass of those quasars. The dark matter halo mass of quasars estimated from the projected auto correlation function is $\log(M_h/M_{\odot})=12.2 ^{+0.2}{-0.7}$ ($11.9^{+0.3}{-0.7}$), with the bias parameter $b$ of $12.34 ^{+4.26}{-4.37}$ ($11.52^{+4.02}{-4.14}$) for the redshift interval of $5.0 \leq z <5.7$ ($5.7 \leq z <6.3$). Our results, combined with other measurements of dark matter halo masses for quasars or active galactic nucleus which obtain a lower dark matter halo mass of $\sim 10^{11.5}$ M$\odot$ at similar redshift, suggest a more complex, and possibly non-monotonic evolution of quasar hosting dark matter halo. Moreover, we estimate the duty cycle of those quasars, which is $0.008^{+0.135}{-0.007}$ ($0.003+^{+0.047}_{-0.003}$) for the redshift interval of $5.0 \leq z <5.7$ ($5.7 \leq z <6.3$).

💡 Research Summary

This paper presents a wide‑field clustering analysis of high‑redshift (5.0 ≤ z < 6.3) quasar candidates to infer the typical dark‑matter halo (DMH) mass and duty cycle of early quasars. The authors first construct a large photometric sample by applying a machine‑learning (random‑forest) classifier to the Legacy Survey DR9 optical data (g, r, z) combined with WISE infrared bands (W1, W2). Two independent classifiers (“mag model” using magnitudes and “flux model” using fluxes) are cross‑matched, yielding 216,949 candidates that are identified by both models. To maximize purity, a probability threshold p_th = 0.8 is imposed, leaving 2,429 objects in the lower‑z bin (5.0 ≤ z < 5.7) and 1,923 objects in the higher‑z bin (5.7 ≤ z < 6.3).

A uniform random catalog is generated with a surface density of 1 arcmin⁻² over the ≈ 17,000 deg² survey footprint, and redshifts are assigned to match the photometric redshift distribution of the data. Pair counts (DD, DR, RR) are computed using the Corrfunc library, and each data–data pair is weighted by the product of the two objects’ predicted quasar probabilities, thereby reducing contamination from mis‑identified galaxies or stars. The projected two‑point correlation function ω_p(r_p) is obtained via the Landy–Szalay estimator, integrating the two‑dimensional correlation ξ(r_p, π) over line‑of‑sight separations to mitigate redshift‑space distortions.

The measured ω_p(r_p) is well described by a power‑law ξ(r) = (r/r₀)^‑γ with γ ≈ 1.8 and a correlation length r₀ ≈ 15 h⁻¹ Mpc in both redshift bins. By comparing the amplitude of ω_p(r_p) with the linear matter correlation function predicted by a ΛCDM cosmology (H₀ = 67.32 km s⁻¹ Mpc⁻¹, Ω_m = 0.3158), the authors infer linear bias parameters b = 12.34 ± 4.26 for 5.0 ≤ z < 5.7 and b = 11.52 ± 4.02 for 5.7 ≤ z < 6.3. Using the bias‑mass relation of Tinker et al. (2010), these biases correspond to typical halo masses log M_h/M_⊙ = 12.2 +0.2/‑0.7 and 11.9 +0.3/‑0.7, respectively. These values are higher than many previous estimates (∼10^11.5 M_⊙) for quasars at similar redshifts, suggesting that the most luminous high‑z quasars may reside in relatively massive halos, but the large uncertainties also allow for a non‑monotonic evolution of halo mass with redshift.

The duty cycle f_duty is calculated as the ratio of the observed quasar number density (weighted by probability) to the number density of halos above the inferred mass threshold, yielding f_duty = 0.008 +0.135/‑0.007 for the lower‑z bin and f_duty = 0.003 +0.047/‑0.003 for the higher‑z bin. These low values imply that the luminous quasar phase occupies only a few percent of a halo’s lifetime at z ≈ 5–6, consistent with short quasar lifetimes (∼10⁶–10⁷ yr) and supporting scenarios where most supermassive black‑hole growth occurs in obscured or radiatively inefficient phases.

Systematic tests are performed by varying the probability threshold (p_th = 0.41 and 0.6). Lower thresholds increase the sample size dramatically (up to ~98 000 objects) but reduce the inferred bias and halo mass by ~15–20 %, illustrating the trade‑off between purity and statistical power. The authors also compare their results with recent clustering studies that used much smaller samples (e.g., Arita et al. 2023, Lin et al. 2025) and find that their large area and sample size substantially reduce cosmic variance, leading to more robust estimates.

In conclusion, the paper demonstrates that a combination of modern machine‑learning selection, probability‑weighted clustering, and wide‑field photometric surveys can deliver precise constraints on the large‑scale environment of high‑redshift quasars. The inferred high bias and halo masses point to a scenario where early luminous quasars are associated with massive dark‑matter structures, while the very low duty cycles suggest that the observable quasar phase is fleeting. Future spectroscopic confirmation from DESI, JWST, Euclid, and the Rubin Observatory will enable refinement of these measurements, allowing a deeper understanding of the co‑evolution of supermassive black holes, their host galaxies, and the underlying dark‑matter scaffolding in the first billion years of cosmic history.