Percolation of the Cosmic Web: Threshold, Topology, and the Signature of Halo Bias

We present a multi-scale topological analysis of the dark matter cosmic web at z = 0 using the Millennium Simulation suite. At the halo scale, friend-of-friends percolation of N = 37,028 halos (n_p ≥ 20) in a 62.5 Mpc/h periodic volume yields a threshold R_c = 1.23 ± 0.17 Mpc/h with critical mean degree ⟨k⟩_c = 7.96 — 2.76× the Poisson expectation (2.88), demonstrating that filamentary structure percolates at a smaller linking length yet with 2.76× more links per halo than a random network of the same density. Density dilution experiments spanning 2.3 decades in halo number reveal R_c ∝ n^{-0.379} for the cosmic web versus n^{-0.329} for Poisson (−1/3 theoretical), with the exponent ratio 1.15 quantifying the enhancement from spatial clustering; this advantage persists down to ~0.4% of the full density, where the web and Poisson thresholds converge at R_c ≈ 10 Mpc/h. Mass-resolved analysis shows that halo bias systematically enhances connectivity: at fixed number density, mass-selected subsamples percolate earlier by up to 22%, and the critical mean degree ⟨k⟩_c decreases from 7.96 to ~4.5 along the resolution sequence — demonstrating that percolation universality does not hold for clustered distributions. Persistent homology analysis finds that the cosmic web contains only 6.2% of the voids and 18% of the loops of matched random networks, with void characteristic scales 1.67× larger. At the galaxy scale, a complementary network analysis of 12,050 galaxies in a (500 Mpc/h)^3 volume identifies percolation at R_c = 19.74 ± 0.08 Mpc/h with ⟨k⟩_c = 3.89 (1.35× above the Poisson reference 2.88). The tracer dependence — both halos (2.76× above Poisson) and galaxies (1.35× above) exceed the Poisson reference, with the elevation factor set by tracer-dependent filamentary clustering rather than by a universal constant. Small-world analysis against degree-preserving random ensembles reveals extreme local clustering (γ ≈ 460–870) but characteristic path lengths 4–8× longer than random (λ ≫ 1), establishing the cosmic web as a modular large-world topology fundamentally distinct from small-world networks.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-31
DOI
https://doi.org/10.5281/zenodo.23832895
Primary Topic
Galaxies: Formation, Evolution, Phenomena
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

Percolation of the Cosmic Web: Threshold, Topology, and the Signature of Halo Bias

FatJack
Zenodo (CERN European Organization for Nuclear Research)
Galaxies: Formation, Evolution, Phenomena
preprint

Percolation of the Cosmic Web: Threshold, Topology, and the Signature of Halo Bias

FatJack
preprint en

Abstract

We present a multi-scale topological analysis of the dark matter cosmic web at z = 0 using the Millennium Simulation suite. At the halo scale, friend-of-friends percolation of N = 37,028 halos (n_p ≥ 20) in a 62.5 Mpc/h periodic volume yields a threshold R_c = 1.23 ± 0.17 Mpc/h with critical mean degree ⟨k⟩_c = 7.96 — 2.76× the Poisson expectation (2.88), demonstrating that filamentary structure percolates at a smaller linking length yet with 2.76× more links per halo than a random network of the same density. Density dilution experiments spanning 2.3 decades in halo number reveal R_c ∝ n^{-0.379} for the cosmic web versus n^{-0.329} for Poisson (−1/3 theoretical), with the exponent ratio 1.15 quantifying the enhancement from spatial clustering; this advantage persists down to ~0.4% of the full density, where the web and Poisson thresholds converge at R_c ≈ 10 Mpc/h. Mass-resolved analysis shows that halo bias systematically enhances connectivity: at fixed number density, mass-selected subsamples percolate earlier by up to 22%, and the critical mean degree ⟨k⟩_c decreases from 7.96 to ~4.5 along the resolution sequence — demonstrating that percolation universality does not hold for clustered distributions. Persistent homology analysis finds that the cosmic web contains only 6.2% of the voids and 18% of the loops of matched random networks, with void characteristic scales 1.67× larger. At the galaxy scale, a complementary network analysis of 12,050 galaxies in a (500 Mpc/h)^3 volume identifies percolation at R_c = 19.74 ± 0.08 Mpc/h with ⟨k⟩_c = 3.89 (1.35× above the Poisson reference 2.88). The tracer dependence — both halos (2.76× above Poisson) and galaxies (1.35× above) exceed the Poisson reference, with the elevation factor set by tracer-dependent filamentary clustering rather than by a universal constant. Small-world analysis against degree-preserving random ensembles reveals extreme local clustering (γ ≈ 460–870) but characteristic path lengths 4–8× longer than random (λ ≫ 1), establishing the cosmic web as a modular large-world topology fundamentally distinct from small-world networks.

Zenodo (CERN European Organization for Nuclear Research)
Galaxies: Formation, Evolution, Phenomena
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
Percolation of the Cosmic Web: Threshold, Topology, and the Signature of Halo Bias — FatJack · Zenodo (CERN European Organization for Nuclear Research) (2026) | TGRS Research Map | TGRS