Galactic Magnetic Fields

Halo toroids, X-fields, and magnetic arms — the canonical loop’s ledger, read in Faraday rotation

Summary

A galaxy is threaded by an organized magnetic field a few microgauss strong and tens of kiloparsecs across. Four decades of Faraday-rotation measurements — J.L. Han’s program above all — have resolved its architecture: a spiral field in the thin disk that runs along the arms but is strongest between them, with reversals inside the arms; a pair of enormous magnetic toroids in the halo, one above the plane and one below, carrying opposite field directions; a poloidal dipole standing vertically at the Galactic center; and, in external edge-on galaxies, a near-universal X-shaped field centered on the nucleus. These are precision observations of a structure that standard theory struggles to produce: the mean-field dynamo generically prefers the even-parity (quadrupole) mode in a thin disk, while the halo field is unmistakably odd (dipole) — and the literature’s escape is that odd modes “might take of order a Hubble time to develop.”

The substrate framework reads the entire architecture at once. The magnetism chapter established what a magnetic field is in this framework: the velocity field of organized co-rotating dc1 flow — the substrate’s bookkeeping of rotational organization, at whatever scale that organization exists. A galaxy is the canonical loop — co-rotating disk, polar outflow, counter-rotating return sheath — at galactic scale. Each observed field component is the ledger of one loop component, and the halo’s opposite-sign toroid pair is the loop’s counter-rotating sheath: the anti-phase breath made directly visible in Faraday rotation. Odd parity is not a marginal dynamo mode that needs a Hubble time to grow. It is forced by the topology of the flow, from the moment the flow exists.

The claims in this chapter are topological, not quantitative — the framework maps which field structure appears where and why it is stable, but does not derive the microgauss amplitudes (see Honest Accounting). What it buys in exchange is a single mechanism for observations that currently require separate stories, and a set of clean observational forks (Predictions).

How the Field Is Read

The line-of-sight magnetic field is measured through Faraday rotation: a polarized radio signal traversing magnetized plasma has its polarization angle rotated by \lambda^2 \times RM, where the rotation measure is

\text{RM} = 0.81 \int n_e\, \mathbf{B} \cdot d\mathbf{l}

integrated from source to observer. Extragalactic radio sources give an all-sky grid of total path integrals; pulsars, distributed through the Galaxy with distances estimated from their dispersion measures, slice that integral radially — the combination is a crude but genuine tomography of \mathbf{B}_\parallel. Han’s group has pushed this program since the early 1990s, from ~200 usable pulsar RMs to today’s combined sample of ~59,000 extragalactic sources and 634 high-latitude pulsars (many measured by FAST).¹

What the Ledger Shows

The disk: a spiral field that ignores the starlight. The thin-disk field is a bisymmetric spiral with pitch angle -8.2° \pm 0.5° and amplitude \sim 1.8\;\muG, running along the Carina-Sagittarius arm — but strong in the interarm regions and reversing inside the arms [R161]. The same inversion is seen face-on in NGC 6946, whose polarized “magnetic arms” sit squarely between the optical spiral arms.² If the field were simply frozen into the gas, it would pile up where the matter piles up. It does not: the field keeps its own spiral, phase-shifted from the material one. Han & Qiao also found that the regular field extends far beyond the optical disk, and that the local field has a weak vertical component, 0.2–0.3\;\muG, pointing from the south Galactic pole to the north — a piece of poloidal field threading our own neighborhood.

The halo: an antisymmetric sky. In 1997, Han, Manchester, Berkhuijsen & Beck showed that the RM sky toward the inner Galaxy is antisymmetric about both the Galactic plane and the meridian through the Galactic center: positive RMs in the first quadrant above the plane, negative below, and the mirror image in the fourth quadrant.³ The unique field structure that produces this pattern is a pair of toroids — azimuthal field rings — above and below the plane with opposite circulation, plus a poloidal dipole perpendicular to the plane at the Galactic center, where vertical field filaments are directly observed. In dynamo language this is odd (dipole) vertical symmetry: the A0 mode.

The toroids are huge. The lingering objection was that the antisymmetry might be local perturbation. Xu & Han settled it by using pulsars — mostly within 5 kpc — to measure and subtract the local RM contribution, leaving the RM of the halo beyond the pulsars.⁴ The antisymmetry survives — and extends past the solar circle to the Galactic anticenter. Their model fit requires the toroids to run from a Galactocentric radius of less than 2 kpc out to at least 15 kpc with no field-direction reversals anywhere in that range: one single, coherent, counter-circulating pair of magnetic rings the size of the entire galaxy.

External galaxies: the X. The CHANG-ES survey of 35 edge-on spirals finds, by stacking, a characteristic X-shaped polarization pattern centered on each galactic nucleus — large-scale field limbs rising from the disk into the halo at both ends.⁵ In NGC 4631, RM synthesis resolves something further: regular kpc-scale sign reversals of the halo field with distance from the minor axis. Modeling these with scale-invariant dynamo solutions, Woodfinden et al. found that rotation-only velocity fields fit poorly, outflow reasonably — and accretion onto the disk fits best.⁶

Note“A0” is not a_0

An unfortunate collision of notation: the A0 dynamo mode (A-zero: Axisymmetric, odd vertical parity) of Han et al. [R162] is a classification label from mean-field dynamo theory and has nothing to do with the MOND acceleration scale a_0 derived in Galactic Dynamics. The two appear in adjacent chapters of this framework by coincidence of the literature’s naming, not by any claimed identity. The connection this chapter does draw between galactic magnetism and galactic gravity runs through the shared boundary topology, never through this label.

Two Problems the Standard Picture Carries

The parity problem. Mean-field dynamo theory in a thin disk generically favors the even (quadrupole, S0) mode — toroidal field with the same sign above and below the plane. The observed halo field is odd. The standard escape is that odd modes are favored in thick disks and halos but “might take of order a Hubble time to develop” [R162]; Xu & Han’s own discussion concedes the halo field “may be still in the state of development” [R163]. The theory’s preferred mode is not the one observed, and the observed mode is the one the theory grows slowest.

The origin problem. A dynamo amplifies a seed field over gigayears of differential rotation and turbulent helicity. But Han & Qiao already argued in 1994 that the bisymmetric spiral structure of the disk field is difficult for a dynamo to generate at all, and concluded that the field is “of primordial origin,” with the dynamo acting only to maintain the preserved field against diffusion rather than to create it [R161]. Meanwhile Henriksen showed that the halo’s X-field and its RM quadrant pattern require no dynamo growth whatsoever: pure advection — a galactic wind increasing with radius, plus rotation with a halo lag, acting on a disk field whose radial component reverses across the plane — produces both, with the sign of every feature set by the actual flow.⁷ The competing “battery” proposal reads the same X pattern as evidence for a radiation-driven mechanism that always drives current outward along the axis, in every galaxy.⁸ The fork between them — flow-determined sign versus universal sign — is open and, as Henriksen notes, unusually decidable.

The Loop’s Ledger, Component by Component

The framework needs no new machinery here. Two established pieces do all the work:

1. What a field is: the magnetic field is the velocity field of leaked, organized, co-rotating dc1 flow — the substrate’s bookkeeping of whatever rotational organization the local matter sustains (Magnetism).
2. What a galaxy is: the canonical loop — a co-rotating equatorial disk, polar outflow along the spin axis, and a counter-rotating sheath wrapping the disk that returns material inward at the equator — the same three-part topology documented from the substrate lattice through accretion disks, the geodynamo, and the Sun.

Put them together and the galaxy’s magnetic architecture is the loop’s flow structure, written down:

Observation	Loop component	Substrate reading
Disk spiral field, in-plane, along arms	Co-rotating disk	Ledger of the disk’s own rotation; spiral pitch follows the flow, not the starlight
Halo toroid pair, opposite signs above/below plane [R162, R163]	Counter-rotating return sheath	The anti-phase pair’s ledger — odd parity forced by the pairing
Central poloidal dipole + vertical filaments; local vertical field 0.2–0.3\;\muG [R161, R162]	Polar jet axis	Ledger of the axial outflow channel
X-field in every edge-on halo [R166]	Jet + return streamlines, seen in projection	The loop’s poloidal circulation, edge-on
No reversals from <2 to \geq 15 kpc [R163]	One boundary, whole galaxy	The sheath is a single coherent structure, not a patchwork of dynamo cells
Regular field extends beyond the optical disk [R161]	Loop closes at the flow boundary	The organization tracks the coherent substrate response, not the baryon light
NGC 4631 halo reversals best fit by accretion [R165]	The return flow	The sheath’s defining job — conveying material back inward — detected kinematically

The centerpiece is the parity argument, and it is worth stating plainly. The counter-rotating sheath above the plane and its mirror partner below are an anti-phase pair — the same paired motion the framework identifies in Cooper pairs, antiferromagnets, and the lattice’s own breath. The magnetic ledger of an anti-phase pair is necessarily odd: azimuthal field of one sign above the plane, the opposite sign below. What dynamo theory treats as its slowest, most marginal mode is, in the substrate, the only structure the halo flow can write. Conversely the co-rotating disk, which has no sign change across the plane, writes an even in-plane spiral — and the observed decomposition of the Galactic field into a spiral thin-disk component plus an odd halo-toroid component (implemented literally as two separate models in Xu & Han’s combined field code [R163]) is the loop’s two flow components, ledgered separately.

This completes a series the framework had already started. The feedback topology chapter identified the dynamo seat with the counter-rotating boundary at every scale where one is observed: Earth’s field is generated at the inner-core and core-mantle boundaries; the Sun’s at the tachocline (Solar & Stellar Dynamics). The galaxy was the missing rung. Its “dynamo seat” is the disk-halo interface where the co-rotating disk meets the counter-rotating sheath — and the huge toroids are that boundary’s field, exactly where the loop demands it and with exactly the parity the pairing demands.

Why the Timescale Problem Dissolves

In the substrate reading, the field’s topology is not grown — it is imposed. The magnetic field is bookkeeping: it exists with the structure of the flow from the moment the flow exists, and the MHD dynamo’s real role is the one Han & Qiao assigned it in 1994 — maintaining the organized field against turbulent diffusion, not manufacturing its architecture from a seed [R161]. This is also precisely what Henriksen’s calculation demonstrates from within standard MHD: given the flow, the X-field and its RM signature follow by frozen-flux advection alone, with no growth time [R164]. The framework simply supplies what that calculation must assume — why the flow has the disk-wind-rotation structure it does (the canonical loop), and why the initial field is organized rather than random (it is the substrate’s ledger of that same loop).

The Hubble-time embarrassment of the A0 mode therefore inverts into a discriminating prediction: if odd-parity halo fields must be slow-grown, young rotation-supported disks should lack them; if they are flow-imposed, any galaxy with a settled rotating disk and halo circulation should show organized fields at essentially full structure, at any epoch (see Predictions).

The Field Lives in the Medium, Not the Gas

The interarm placement of the magnetic arms — in the Milky Way’s field-strength maxima [R161] and spectacularly in NGC 6946 [R167] — is the most audience-legible clue in this chapter. A field passively frozen into the interstellar gas should be compressed and amplified where the gas is: in the arms. Instead the ordered field is strongest where the matter is not, holding its own spiral at its own phase. In the substrate reading this is expected: the ordered component of the field is the ledger of the substrate’s rotational organization, which the spiral density wave of stars and gas propagates through. The matter arm is a pressure wave in the baryons; the magnetic arm is the standing organization of the medium those baryons swim in. Where the gas piles up, turbulence scrambles the local ledger (field reversals in the arms); between the arms, the quiet flow lets the large-scale organization show through.

The v_L Cross-Check

Galactic Dynamics sorts systems by the substrate’s coherence threshold v_L \approx 750 km/s: galaxies (v_\text{disp} \ll v_L) sit in the coherent, superfluid-response phase; clusters (v_\text{disp} \gtrsim v_L) in the incoherent, normal phase. The magnetic ledger honors the same split. Galaxies below v_L carry galaxy-spanning ordered fields — the toroids, the X, the magnetic arms. Galaxy clusters carry comparable-strength (\sim\muG) fields that are turbulent and tangled, with coherence lengths of order 10 kpc and no cluster-scale ordered toroids, as probed by radio halos and RM scatter through the intracluster medium. Where the substrate’s response is coherent, its bookkeeping is coherent; where the lattice has shredded into vortex tangle, the ledger is scribble. The framework did not tune anything to get this — the phase table was built for gravity, and the magnetism falls on the same line.

Honest Accounting

Field amplitudes are not derived. The framework maps topology — which structure, what parity, where — and stability. It does not compute why the disk field is 1.8\;\muG and the halo toroids 0.7\;\muG. The microgauss scale lives in the coupling between the substrate’s organization and the plasma that carries the actual currents; deriving it would require the HVBK boundary-transmission calculation that is also the open item behind the CPR coefficient I_2 (Open Problems).

The arm/interarm reversal pattern has no derived mechanism. Reading it as turbulent scrambling of the ledger inside density waves is consistent but qualitative. The kpc spacing of the disk reversals has no computed connection to the \xi-spaced boundary chain of the gravity sector, and none is claimed.

Standard MHD is not being replaced. As with the solar chapter, the induction equation, frozen flux, and turbulent diffusion remain the correct description of what the plasma does. The framework’s contribution is structural selection: which flow topology the medium organizes into, and therefore which field topology the plasma must write.

The local sky is messy. Loop I, the Gum Nebula, Region A, and other nearby magnetized bubbles perturb the RM sky at tens-of-degrees scale; every claim above concerns the large-scale organized component that survives Han’s selection and Xu & Han’s local-discounting procedures.

Predictions

1. No universal axial current. The sign pattern of a galaxy’s halo RM quadrants should be set by its rotation sense and disk-field orientation — Henriksen’s flow rule [R164] — so stacked samples should show no preferred axial current direction. This is a three-way fork: robust confirmation of the battery signature (universal outward current [R168]) in improved CHANG-ES-class stacking would falsify the flow-ledger reading — unless the excess tracks the substrate’s global chirality preference, which would be a distinct, larger claim requiring its own accounting. The framework registers the null (flow-determined, no universal sense) as its prediction.

2. Odd-parity halo fields are generic and early. Every galaxy with a settled rotating disk and halo circulation should carry the opposite-sign toroid pair, including young systems: organized, near-full-strength halo fields should be present in rotation-supported disks at high redshift, with no gigayear growth delay. A confirmed population of dynamically settled disks with absent or even-parity halo fields would count directly against the flow-imposed reading; conversely, slow-grown dynamo fields predict a visible build-up epoch.

3. Halo RM reversals mark the return flow. Where kpc-scale halo RM sign reversals are found (NGC 4631 and, per CHANG-ES, likely others), kinematic modeling should continue to prefer accretion/inflow velocity fields over rotation-only [R165] — the reversal zones are the sheath’s inward conveyor, and should correlate spatially with independent accretion tracers.

4. No coherent toroids above v_L. Galaxy clusters should never show cluster-scale ordered toroidal fields, only tangled fields, regardless of age or dynamical relaxation — the ordered ledger requires the coherent substrate phase that clusters’ velocity dispersions preclude.

5. The magnetic boundary tracks the coherent-rotation extent, not the light. The ordered field’s radial reach (already \geq 15 kpc with no reversals in the Milky Way [R163], and beyond the optical disk in Han & Qiao’s ERS fits [R161]) should track the region of coherent rotation — the flat rotation curve’s domain — rather than the stellar disk, tying the magnetic and gravitational ledgers to the same boundary structure.

Context

This chapter closes a loop the framework left open. Feedback Topology claimed the canonical loop’s boundary sheath “always becomes the dynamo seat” and cashed that claim for Earth and Sun; the galaxy now pays out the same way, with the added twist that at this scale the boundary’s pairing — the anti-phase breath that Galactic Dynamics reads as the origin of MOND — is directly photographed in the Faraday sky as two counter-circulating magnetic toroids. Gravity and magnetism at galactic scale are two ledgers of one boundary: the parity-even response of the paired breath is the flat rotation curve, and the parity-odd record of the paired flow is the halo field. One structure, two bookkeeping columns.

Outward, the same logic scales: the cosmic web’s filaments should carry the ledger of their own flows, and the modons-in-space chapter’s coupled stellar systems show the same entrainment fields at stellar scale. Inward, the solar and planetary dynamos are the same seat at smaller rungs — one substrate, one loop, one ledger, kept at every scale the rotation reaches.

Footnotes

1. Han, J.L. & Qiao, G.J., “The magnetic field in the disk of our Galaxy,” Astronomy & Astrophysics 288, 759, 1994. The careful-selection methodology — discarding pulsars behind local magnetized bubbles and beyond 3.5 kpc — is what first stabilized the disk-field fit. [R161]

2. Beck, R. & Hoernes, P., “Magnetic spiral arms in the galaxy NGC 6946,” Nature 379, 47, 1996. [R167]

3. Han, J.L., Manchester, R.N., Berkhuijsen, E.M. & Beck, R., “Antisymmetric rotation measures in our Galaxy: evidence for an A0 dynamo,” Astronomy & Astrophysics 322, 98, 1997. Pulsar RMs with |b| > 8° show the same pattern, and the growth of |RM| with pulsar distance out to 3–4 kpc rules out a local-bubble origin. [R162]

4. Xu, J. & Han, J.L., “The huge magnetic toroids in the Milky Way halo,” Astrophysical Journal 966, 240, 2024. Local-discounted RMs from 543 pulsar sightlines; best-fit toroid model B_0 = 0.73\;\muG, scale height z_0 = 3.0 kpc, Gaussian radial profile peaking at R_0 \approx 8 kpc. [R163]

5. Krause, M., Irwin, J., Schmidt, P., et al., “CHANG-ES XXII: Coherent magnetic fields in the halos of spiral galaxies,” Astronomy & Astrophysics 639, A112, 2020. [R166]

6. Woodfinden, A., Henriksen, R.N., Irwin, J. & Mora-Partiarroyo, S.C., “Evolving galactic dynamos and fits to the reversing rotation measures in the halo of NGC 4631,” MNRAS 487, 1498, 2019. [R165]

7. Henriksen, R.N., “Galactic magnetic X fields,” Astronomy & Astrophysics 658, A101, 2022. The Cauchy (frozen-flux Lagrangian) evolution of an initial disk field under wind plus lagged rotation yields the X polarization and the XRM quadrant signature; reversing either the rotation sense or the radial-field sign flips the pattern. [R164]

8. Myserlis, I. & Contopoulos, I., “A universal X-shaped rotation measure pattern,” Astronomy & Astrophysics 649, A94, 2021. [R168]