Cosmic Himalayas in CROCODILE : Probing the Extreme Quasar Overdensities by Count-in-Cells analysis and Nearest Neighbor Distribution

TL;DR

The paper addresses the tension between an extreme quasar overdensity (Cosmic Himalayas) and LCDM expectations. It uses the CROCODILE cosmological hydrodynamic simulation to study quasar clustering with two complementary statistics: count-in-cells (CIC) and nearest-neighbor distribution (NND). The authors show that the CIC distribution is heavy-tailed and non-Gaussian, and that fitting with an asymmetric generalized normal distribution (AGND) substantially reduces the inferred rarity of CH-like structures, reconciling them with LCDM; NND analysis confirms that CH-like clustering can arise from sample selection biases and cosmic variance, without requiring new physics. These results highlight the importance of non-Gaussian statistics in quantifying extreme overdensities and support CH as a natural outcome of structure formation in $\Lambda$CDM.

Abstract

The recently reported Cosmic Himalayas (CH) -- an extreme quasar overdensity at z~2 -- poses an apparent challenge to the Lambda CDM framework, with a reported significance of 16.9-sigma under Gaussian assumptions. Such an event appears improbably rare, with a formal probability of P ~ 10^-68. In this work, we investigate whether CH-like structures can naturally arise in cosmological hydrodynamic simulations. Using the CROCODILE simulation, which self-consistently models galaxy-black hole coevolution, we examine quasar clustering through two complementary approaches: the count-in-cells (CIC) statistic, which probes large-scale overdensities, and the nearest-neighbor distribution (NND), sensitive to small-scale environments. CIC analysis reveals that the underlying distribution is heavy-tailed and non-Gaussian, and that conventional Gaussian-based evaluation substantially overestimates the significance of extreme events. When modeled with an asymmetric generalized normal distribution (AGND), the inferred rarity of the CH is substantially reduced and reconciled with standard Lambda CDM; for instance, regions appearing as 12-sigma outliers under Gaussian assumptions (P ~ 10^-33) are found to occur in the AGND regime with a probability of P ~ 10^-4. NND analysis further demonstrates that extreme overdense regions within the simulation can naturally sustain two-point correlation function values similar to those observed in the CH (r0 ~ 30 Mpc/h), suggesting that the strong clustering stems from sample selection biases and local environmental variations. These two analyses conclusively highlight the importance of adopting non-Gaussian statistics when quantifying extreme overdensities of quasars and establish that the CH is not an anomaly, but a natural outcome of structure formation in the Lambda CDM universe.

Figures (14)

• Figure 1: Eddington ratio ($\lambda_{\rm Edd}$) distribution function for BHs in our simulation BF50NEL (solid orange line). Colored lines and shaded regions indicate observational constraints: sky blue and green shades correspond to georgakakis_observational_2017 ($z=2.25$) and laloux_accretion_2024 ($1.5 < z<2.0$), respectively, while yellow, blue, and thick orange dotted lines with shades represent the constraints in stellar mass bins of aird_x-rays_2018 ($2.0 < z < 2.5$).
• Figure 2: X-ray luminosity function (XLF) from simulations and observations. The orange solid line shows the result from our BF50NEL simulation. Colored dotted curves indicate results from other simulation projects compiled by habouzit_supermassive_2022: EAGLE (sky blue; schaye_eagle_2015crain_eagle_2015), SIMBA (green; dave_simba_2019), TNG100 (yellow) and TNG300 (blue; nelson_first_2018), and Illustris (red; vogelsberger_properties_2014). Observational estimates are shown from hopkins_observational_2007 (pink), aird_evolution_2015 (orange dotted line), and buchner_obscuration-dependent_2015 (gray shade).
• Figure 3: Black hole mass function (BHMF) from simulations. The orange solid line shows the result from our BF50NEL simulation. Other colored curves represent results from major cosmological hydrodynamic simulations compiled by habouzit_supermassive_2021: EAGLE (sky blue; schaye_eagle_2015crain_eagle_2015), SIMBA (green; dave_simba_2019), TNG100 (yellow) and TNG300 (blue; nelson_first_2018), Illustris (red; vogelsberger_properties_2014), and Horizon-AGN (pink; dubois_dancing_2014).
• Figure 4: CIC analysis for Case A. $N_{\rm QSO}^{\rm max} = 6\; ( \delta_{\rm QSO}^{\rm max}\sim23)$.
• Figure 5: CIC analysis for Case B. $N_{\rm QSO}^{\rm max} = 17 \; (\delta_{\rm QSO}^{\rm max} \sim 10)$.