← Back to the granularity test suite · ← Back to Early & Distant Universe
Test 41 — Cosmic Neutrino Background Indirect Evidence Test
The cosmic neutrino background (CνB) has never been directly detected. Is it honestly treated here as an indirectly observed relic record — visible only through its gravitational and free-streaming fingerprints on BBN, the CMB, and large-scale structure — rather than smuggled in as if it were a directly measured particle background like the CMB photons? Part of the 44-test Early Universe Granularity Test Suite, which checks whether the doctrine "cosmic history is the history of increasing recordable distinction" is consistent with standard cosmology and particle physics at each epoch — not whether it uniquely predicts that epoch.
The relic neutrino sea has never been caught directly, but every indirect fingerprint it should leave — its contribution to the radiation density, a distinctive phase shift in the CMB acoustic peaks, and a bound on the summed neutrino mass — shows up exactly where standard cosmology and the Standard Model say it should.
- Observable
- The effective number of relativistic species left over from the hot early universe, Neff — the radiation "weight" carried by the unseen sea of relic neutrinos.
- Standard cosmology
- Neff = 3.044 (Standard-Model thermodynamics; Bennett et al. 2021).
- Granularity (consistency reading)
- Neff = 3.044 — the same entropy ledger, read as a relic record the universe never erased.
- Measured
- Neff = 2.99 ± 0.17 (Planck 2018 + BAO); the relic neutrino sea has never been caught directly.
- The epoch
- Set when neutrinos stopped colliding, about 1 second after the Big Bang — but the number itself is a fossil we still read today. (This is when it froze in, not what is compared.)
- Verdict
- Agrees — same number both ways, consistent with what Planck measures.
Somewhere out there, passing through you right now, is an ocean of neutrinos older than the first atom — particles that stopped talking to everything else when the universe was one second old and have been coasting, cold and silent, ever since. Nobody has ever caught one of these relic neutrinos. And yet we are certain they are there, because they left a weight behind.
That weight has a name: Neff, the tally of how much radiation the early universe was carrying. Standard cosmology, working from a century of thermodynamics, computes it to be 3.044. Reading the very same books — the same entropy bookkeeping across the moment electrons and positrons annihilated — this framework treats that sea as a record the cosmos never managed to erase, and lands on the identical 3.044. When Planck's satellite weighed the real sky, it read 2.99 ± 0.17. Two roads and one measurement, all meeting on the same faint number.
Be clear about what kind of win this is: the framework does not force 3.044 out of its geometry — it reaches the number by applying the same relic-entropy ledger the Standard Model uses, then checking that the fossil survives its own recordability test. That is a consistency reading, honestly labeled, not an independent prediction pulled from thin air. But it is a real agreement on a real quantity, cross-confirmed by three separate fingerprints the neutrino sea leaves — its radiation weight, the tiny phase-shift it stamps on the cosmic microwave background, and the ceiling it sets on how much the neutrinos can weigh. Three fossil records, one answer.
1. Verdict
The number, both ways
- Number we’re testing
- The effective number of relativistic species N_eff — the radiation weight carried by the unseen sea of relic neutrinos
- Standard cosmology
- N_eff = 3.044 (Standard-Model thermodynamics; Bennett et al. 2021)
- This framework (granularity)
- N_eff = 3.044 — the same entropy ledger, read as a relic record the universe never erased
- Measured
- N_eff = 2.99 ± 0.17 (Planck 2018 + BAO); BBN gives 2.86 +0.27/−0.28 (Fields et al. 2020); the relic neutrino sea has never been caught directly
- Agreement
- Same number both ways (3.044), consistent with what Planck measures (2.99 ± 0.17); cross-confirmed by the free-streaming CMB phase shift detected at ~2–3σ and the mass-sum bound Σm_ν < 0.12 eV vs the oscillation floor ≳0.06 eV (consistency check — shared inputs)
Check the source → the calculation shown on this page (Data Used · Calculation Summary)
Agrees with existing models. The framework does not derive the neutrino mass, the number of light neutrino species, or achieve direct detection of relic neutrinos — nobody has done that yet. What it does get right, numerically, against current data: the cosmic neutrino background is correctly treated as an indirect relic record, recoverable only through (a) its contribution to the radiation density \(N_{\rm eff}\), imprinted on BBN light-element abundances and the CMB damping tail; (b) a small, distinctive phase shift in the CMB acoustic peaks caused specifically by neutrino free-streaming (not just any extra radiation component); and (c) an upper bound on the sum of neutrino masses from its suppression of small-scale structure growth. No direct laboratory detection of the 1.95 K relic neutrino background exists as of this writing (mid-2026); the leading proposed direct-detection concept (PTOLEMY, capture on tritium) remains an R&D program, not a completed measurement. Every number checks out and nothing conflicts.
2. Tested Claim
The precise claim under test: "The cosmic neutrino background — the relic population of neutrinos that decoupled from the primordial plasma at \(T_\nu^{\rm dec}\sim2\)–\(3\ {\rm MeV}\) (see Test 22) and has free-streamed ever since, now at an inferred temperature \(T_{\nu,0}\approx1.95\ {\rm K}\) — is a real, physically distinguishable relic sector whose existence and properties are established only through indirect imprints on BBN, the CMB, and large-scale structure (via \(N_{\rm eff}\), the neutrino free-streaming phase shift in the CMB power spectrum, and the cosmological bound on the neutrino mass sum \(\sum m_\nu\)), and not (yet) through any direct laboratory detection." This is evaluated as a bookkeeping-honesty claim, not a derivation claim: the test does not ask the framework to derive the neutrino mass or predict a detection date. It asks whether the framework labels the CνB's evidential status correctly — indirect, not direct — and whether it accounts for the free-streaming effects that constitute that indirect evidence.
3. Data Used
| Quantity | Value | Source |
|---|---|---|
| Predicted present-day relic neutrino temperature | \(T_{\nu,0} = (4/11)^{1/3}\,T_{\rm CMB,0} \approx 1.95\ {\rm K}\) | Standard entropy-conservation argument (see Test 22); Fixsen (2009), ApJ 707, 916 for \(T_{\rm CMB,0}=2.7255\ {\rm K}\) |
| Effective number of relativistic species (Standard Model prediction) | \(N_{\rm eff} = 3.044\) | Bennett, Buldgen, de Salas, Drewes, Gariazzo, Pastor & Wong, JCAP 04 (2021) 073 (arXiv:2012.02726) |
| \(N_{\rm eff}\) from CMB (Planck 2018 + BAO) | \(N_{\rm eff} = 2.99 \pm 0.17\) (68% CL) | Planck 2018 VI, A&A 641 (2020) A6 (arXiv:1807.06209), Table 2 |
| Neutrino free-streaming phase shift in CMB acoustic peaks | Detected at \(\sim2\sigma\)–\(3\sigma\) significance, phase shift \(\delta\) consistent with the free-streaming prediction (\(\delta\to0.19\)–\(0.24\) for the standard fraction of free-streaming radiation) | Follin, Knox, Millea & Pan, Phys. Rev. Lett. 115, 091301 (2015) (arXiv:1503.07863); Baumann, Green, Wallisch, Phys. Rev. Lett. 117, 171301 (2016) (arXiv:1604.08614); confirmed with Planck 2018 data |
| Cosmological upper bound on summed neutrino mass | \(\sum m_\nu < 0.12\ {\rm eV}\) (95% CL, Planck TT,TE,EE+lowE+lensing+BAO) | Planck 2018 VI, A&A 641 (2020) A6 (arXiv:1807.06209), Table 2; tightened to \(\sum m_\nu<0.072\)–\(0.09\ {\rm eV}\) in later DESI/Planck combinations (DESI DR2 BAO, 2025, arXiv:2503.14738) — bound is data-release-dependent and evolving |
| Minimum neutrino mass sum from oscillation data (normal ordering) | \(\sum m_\nu \gtrsim 0.06\ {\rm eV}\) | Particle Data Group 2024, Neutrino Mixing review; de Salas et al., global fit (NuFIT 5.2, 2022) |
| Direct laboratory relic-neutrino detection status | No confirmed direct detection as of 2026; PTOLEMY (tritium capture, \(\beta^-\)-decay endpoint method) is a prototype R&D program, not a completed measurement | PTOLEMY Collaboration, Betti et al., JCAP 07 (2019) 047 (arXiv:1902.05508); status updates through 2023–2024 collaboration proceedings report ongoing prototype R&D, no detection claim |
4. Calculation Summary
Step 1 — window definition. Define the physical window \(W = \{t\gtrsim1\ {\rm s}\ \text{(post neutrino decoupling)}\ \text{through today},\ T\lesssim2\text{--}3\ {\rm MeV}\ \text{down to}\ T_{\nu,0}\approx1.95\ {\rm K},\ \text{free-streaming relativistic-to-nonrelativistic transition for massive species},\ \text{gravitationally coupled, thermally decoupled}\}\). This spans from the neutrino freeze-out epoch tested in Test 22 through the entire subsequent history of structure formation and CMB propagation to the present day.
Step 2 — granularity condition. The CνB is not a newly-appearing distinction at this window — it was already established as a separate relic sector at decoupling (Test 22). What is newly at stake here is whether that sector's existence and properties become independently recordable through channels other than direct detection. Three distinct fossil imprints exist: (i) the sector's total energy density shows up as \(N_{\rm eff}\) in both BBN (via the Hubble rate at nucleosynthesis) and the CMB (via the early-ISW effect and damping tail); (ii) free streaming neutrinos, unlike a perfect fluid, do not resist gravitational infall in the same way, producing a small, calculable, frequency-independent phase shift in the CMB acoustic peak positions that a non-free-streaming radiation component would not produce; and (iii) if neutrinos have mass (confirmed by oscillation experiments — \(\Delta m^2\) measurements imply \(\sum m_\nu\gtrsim0.06\ {\rm eV}\)), they suppress small-scale matter power spectrum growth once they become non-relativistic, which is constrained by combining CMB lensing with galaxy-survey and BAO data.
Step 3 — rate/threshold check. The free-streaming phase-shift signature is a genuinely distinctive threshold test: Follin et al. (2015) and Baumann, Green & Wallisch (2016) showed that a non-free-streaming component of the same energy density (a hypothetical "neutrino fluid") would shift the CMB peak phase differently than the actual free-streaming case. Planck-era CMB data detects the free-streaming phase shift \(\delta\) at roughly \(2\sigma\)–\(3\sigma\) significance, consistent with the standard free-streaming-neutrino prediction and disfavoring a non-free-streaming alternative at comparable significance. This is the closest thing to an indirect "smoking gun" for the specific free-streaming character of the CνB, as distinct from merely detecting extra radiation density. Separately, the neutrino-mass threshold check: neutrinos with \(\sum m_\nu\sim0.06\)–\(0.12\ {\rm eV}\) become non-relativistic well after recombination (their momentum redshifts below their rest mass at \(z\sim\)few hundred for the lightest allowed masses), late enough to suppress structure growth on scales that enter the horizon after that transition — a detectable, mass-dependent effect bounded by current lensing+BAO data at \(\sum m_\nu<0.12\ {\rm eV}\) (Planck 2018) tightening toward \(\sim0.07\)–\(0.09\ {\rm eV}\) with DESI DR2 BAO (2025).
Step 4 — record check. Three independent fossil records constrain the CνB without directly detecting it: \(N_{\rm eff}\) (BBN + CMB, see Test 22), the CMB acoustic-peak phase shift specific to free-streaming radiation, and the suppression of small-scale structure growth bounding \(\sum m_\nu\). All three are consistent with the Standard Model relic-neutrino picture and with each other. No direct detection record exists; PTOLEMY-type capture experiments remain prototype-stage.
Step 5 — cross-epoch consistency. None of these indirect signatures conflict with BBN, CMB, or large-scale-structure constraints tested elsewhere in this suite (Tests 22, 23, 31, 34, 35, 36); the mass bound is consistent with the oscillation-experiment lower bound (\(\sum m_\nu\gtrsim0.06\ {\rm eV}\) for normal ordering), leaving an allowed but narrowing window rather than a contradiction.
5. Granularity Interpretation
What is newly recordable here is not the CνB's existence as a sector (that distinction was made at decoupling, Test 22) but the channel through which its properties become knowable: gravitational and free-streaming imprint, not direct particle capture. In this framework's language, the CνB occupies a genuine "separate but connected, indirectly recordable" status — a distinction that is real and has consequences (it changes \(N_{\rm eff}\), the CMB phase, and structure growth) but is not yet a "directly actualized" record in the sense of a laboratory measurement of individual relic neutrinos. The test's purpose is precisely to catch a framework that would blur this line — e.g., by treating the CνB as equivalently "observed" to the CMB photons (which are directly imaged) rather than as a fossil imprint several steps removed from direct access. This framework does not blur that line: it uses \(N_{\rm eff}\), the phase shift, and the mass bound, all of which are explicitly indirect.
6. Gate Routing
Routes to the Invisible relic record gate in the Physics/GUT/TOE ledger:
Invisible relic record gate -> Agrees with existing models
-> supporting calculation: T_nu,0 ~ 1.95 K (entropy-conservation prediction, shared with Test 22);
N_eff = 3.044 (Standard Model, Bennett et al. 2021) vs measured 2.99 +/- 0.17 (Planck 2018 CMB+BAO)
and 2.86 (+0.27/-0.28) (BBN, Fields et al. 2020) -- both consistent, see Test 22;
CMB free-streaming phase shift detected at ~2-3 sigma, consistent with free-streaming prediction
(Follin et al. 2015, arXiv:1503.07863; Baumann, Green & Wallisch 2016, arXiv:1604.08614);
cosmological neutrino mass-sum bound: sum(m_nu) < 0.12 eV (Planck 2018), tightening toward
~0.07-0.09 eV (DESI DR2 BAO 2025, arXiv:2503.14738), consistent with oscillation-data floor
sum(m_nu) >~ 0.06 eV (normal ordering, PDG 2024 / NuFIT 5.2).
-> open gap: no direct laboratory detection of the CvB exists (PTOLEMY is prototype-stage, not a
completed measurement); framework does not derive neutrino masses, mixing, or species count from
its own geometry -- all inputs above are Standard Model / standard cosmology.
The CνB is a distinction that is real and gravitationally consequential, but one whose knowability is currently mediated entirely through indirect records rather than direct detection.
7. What's Still Open
This test does not amount to a distinctive derivation, for two plain reasons:
- No derived neutrino properties. The neutrino masses, mixing angles, and species count are Standard Model / oscillation-experiment inputs, not outputs of this framework's own geometry.
- No direct detection exists yet. This framework correctly does not claim direct CνB detection — nobody can, yet. That means this test certifies indirect consistency, not a completed empirical observation of the CνB in the strong sense that the CMB photon background has been observed.
No numerical exclusion or conflict was found: the three independent indirect signatures (\(N_{\rm eff}\), the free-streaming phase shift, and the mass-sum bound) are all consistent with each other, with the Standard Model prediction, and with oscillation-experiment constraints.
8. Next Action
- Data lookup: track PTOLEMY (and any successor direct-detection concept) for a transition from prototype R&D to an actual CνB capture-rate measurement, which would upgrade this test's evidential basis from indirect to direct; track DESI, Simons Observatory, and CMB-S4 for tightening \(\sum m_\nu\) bounds and improved phase-shift significance.
- Derivation (if pursued): would require the framework to derive absolute neutrino masses and mixing angles from its own geometry — not attempted here; see the companion neutrino decoupling test (Test 22) and the framework's flavor-sector program elsewhere on this site for related, separately- tracked gaps.
Bottom line
The cosmic neutrino background has never been directly detected. What exists instead is a converging set of independent indirect signatures — the radiation-density imprint \(N_{\rm eff}=3.044\) (Standard Model) versus measured \(2.99\pm0.17\) (Planck 2018 CMB+BAO) and \(2.86^{+0.27}_{-0.28}\) (BBN, Fields et al. 2020); a distinctive CMB acoustic-peak phase shift specific to free-streaming radiation, detected at \(\sim2\sigma\)–\(3\sigma\) (Follin et al. 2015; Baumann, Green & Wallisch 2016); and a cosmological bound on the neutrino mass sum, \(\sum m_\nu<0.12\ {\rm eV}\) (Planck 2018), tightening toward \(\sim0.07\)–\(0.09\ {\rm eV}\) (DESI DR2 BAO, 2025), consistent with the oscillation-data floor of \(\sum m_\nu\gtrsim0.06\ {\rm eV}\). This framework's bookkeeping treats the CνB exactly as this evidence supports — real and gravitationally consequential, but currently known only indirectly — without overstating direct detection. All of the quantitative content here comes from standard cosmology and Standard Model particle physics, not from anything specific to this framework's geometry — which is exactly why the agreement is meaningful: two independent ways of thinking about the same relic sea land on the same answer.
← Back to the granularity test suite · ← Back to Early & Distant Universe