Spectral shape of the nanohertz background: SMBH binaries with environments versus cosmological sources
Unchanged from the calibrated 33/6.5/38.5: the problem is the old one with the spectral discrimination made explicit, and more pulsars and longer baselines will largely decide it. Leverage stays below dark-matter identity and the CMB problems because a cosmological origin, though high-impact, is the less likely outcome.
The problem
Is the new low-pitched gravitational-wave hum from pairs of giant black holes, or from the early universe?
- 1Pulsars show a low-pitched gravitational hum
- 2Black-hole pairs predict one spectrum
- 3Early-universe sources predict others
- 4Compare shapes with Bayesian evidence
- 5Need more pulsars and years to decide
The signal. The background is described by its characteristic strain h_c(f), which follows a power law h_c ∝ f^α with α = (3 − γ)/2. For a population of circular binaries driven only by gravitational-wave emission, γ is 13/3. The 15-year NANOGrav data prefer γ near 3.2, with large uncertainties, and an amplitude larger than many binary-population predictions.
Built from
earlier problems → this oneThe problems this one is built from: progress or tools it needs (builds on), special cases that come first (generalizes), and weaker results on the way (implies). On the right, what builds on it.
- Start with
- Then
- Spectral shape of the nanohertz background: SMBH binaries with environments versus cosmological sourcesThis problem
Path to understanding
foundations → problemThe research entries this problem rests on, in the order to read them. Each links to its explainer.
- Start with
- Step 2
- Pulsar timing arraysAlgorithm · host
- Then
- Spectral shape of the nanohertz background: SMBH binaries with environments versus cosmological sourcesThe problem
State of the art
NANOGrav 15-year (arXiv:2306.16213), EPTA DR2, PPTA DR3 and CPTA all reported evidence for a correlated background in June 2023, with Hellings–Downs correlation significance of about 2–4.6σ across arrays and an amplitude A ≈ 2.4×10⁻¹⁵ at 1/yr for γ = 13/3. NANOGrav's new-physics analysis (arXiv:2306.16219) found Bayes factors of order 10–100 for several cosmological models over the baseline SMBHB model with fixed population assumptions, and the SMBHB paper (arXiv:2306.16220) found the amplitude requires a high normalisation of the black-hole mass function or overmassive black holes. Taylor et al. (arXiv:1606.09180) introduced phase-shift and sky-scramble backgrounds for assessing the significance of correlation searches. Both explanations remain viable and the cosmological-model comparison depends strongly on the astrophysical-prior choices. Later data have not crossed 5σ. The MeerKAT PTA (4.5 yr, 83 pulsars) finds 3–3.4σ Hellings–Downs evidence under some noise models and none under others, with a higher amplitude (Miles et al., arXiv:2412.01153). Re-modelling chromatic noise in NANOGrav 15-yr data raises the frequentist significance only from 3.16σ to 3.32σ (arXiv:2606.28554, June 2026), and a five-array combination with 121 pulsars stays below 5σ (arXiv:2512.08666). A February 2026 review of the decade describes PTAs as "on the threshold" of a confident detection (Liu et al., arXiv:2602.14014). IPTA DR3 and the NANOGrav 20-year data set had not been released as of October 2026.
Realistic applications
A confirmed cosmological component would be a window on physics at 10–100 MeV temperatures, constraining first-order phase transitions and cosmic strings. If the background is astrophysical, it measures the supermassive-black-hole binary population and merger rate. No practical application.
What counts as solved
Resolved when data (IPTA DR3, NANOGrav 20-year or later) with 5σ Hellings–Downs evidence give a spectral shape and anisotropy for which the Bayes factor between best astrophysical and best cosmological models exceeds 100, or when an individual binary or a non-Gaussianity signature is detected.
How it gets verified
ObservationSettled by new measurements or observations.
Needs new pulsar-timing data (IPTA DR3, NANOGrav 20-year or later) with Hellings-Downs evidence at 5 sigma that gives a spectral shape and anisotropy for which the Bayes factor between the best astrophysical and best cosmological models exceeds 100, or a detected individual binary or non-Gaussianity.
Claims under verification
+ add a claimNo claimed results on the Verification board for this problem yet.
Where the work is
Data analysis and population modelling dominate, with pulsar timing observations adding data.
Bottlenecks
ranked by votes- #1EmpiricalPulsar count and baseline
A 5σ detection and spectral discrimination needs more pulsars and longer baselines (IPTA DR3, 20-year data), and radio-telescope time.
Community grades
Sign in to add your own grade.
Connected problems
Progress
- Feb 15, 2026PTA decade reviewsource
Review of the decade: 2023 Hellings–Downs evidence of about 2–4.6σ across arrays; PTAs described as on the threshold of a confident detection.
- Dec 2, 2024MeerKAT PTAsource
Noise-model-dependent 3–3.4σ Hellings–Downs evidence from 83 pulsars over 4.5 yr.
- Jun 28, 2023PTA evidence for a GW backgroundsource
Four PTA collaborations report Hellings–Downs-correlated signals.