Solar System
Torsional Oscillations: The Sun's Migrating Rotation Bands
Every 11 years the Sun's surface reorganizes into faint zonal jets — belts of gas rotating just 5–10 m s⁻¹ (roughly 0.5% of the ~2 km s⁻¹ equatorial rotation speed) faster or slower than their surroundings — that drift steadily from mid-latitudes toward the equator, tracing out the coming sunspot cycle years in advance. These are the solar torsional oscillations: a small, cyclic modulation of the Sun's differential rotation, discovered in 1980, in which alternating fast and slow bands migrate across the solar disk in lockstep with the magnetic activity cycle.
Far from a surface curiosity, helioseismology has shown these bands are not skin-deep but penetrate a large fraction of the convection zone, making them one of the clearest observational windows into the solar dynamo that generates the Sun's magnetic field.
- RegimeSolar convection zone, cyclic zonal flows
- Key numberAmplitude ~5–10 m s⁻¹ (<1% of rotation)
- Driven byMagnetic feedback on differential rotation (solar dynamo)
- First describedHoward & LaBonte, 1980
- Observed withDoppler surface measurements + helioseismology (SOHO/MDI, SDO/HMI, GONG)
- Matters forSolar dynamo, cycle prediction, differential rotation
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
What the torsional oscillation is and why it matters
The Sun does not rotate as a solid body: its equator laps its poles, completing one turn in about 25 days versus ~34 days near the poles. Superimposed on this steady differential rotation is a tiny, time-varying pattern — the torsional oscillation. It consists of alternating bands where the plasma rotates slightly faster or slower than the long-term average, with a perturbation amplitude of only ~5–10 m s⁻¹, less than 1% of the mean rotation velocity.
What makes these bands remarkable is their behaviour: they are not fixed. As the ~11-year sunspot cycle proceeds, the bands drift systematically in latitude. A branch at mid-latitudes migrates toward the equator over the cycle, and the fast/slow boundary tracks — and slightly leads — the belt where sunspots emerge. Because the pattern appears before the corresponding sunspots, torsional oscillations are a diagnostic of the subsurface dynamo, revealing the magnetic cycle organizing itself beneath the visible surface.
The physics: magnetic feedback on the flow
The torsional oscillation is understood as the mechanical back-reaction of the Sun's cyclic magnetic field on its rotation — a two-way coupling within the solar dynamo. The dynamo converts differential rotation into toroidal (east–west) magnetic field via the Ω-effect; that field, in turn, exerts forces back on the flow.
Two mechanisms are proposed. In the Lorentz-force picture, the large-scale magnetic field feeds back through the Maxwell stress term (∝ B_r·B_φ / 4π), locally braking or accelerating the plasma as the dynamo wave sweeps through — the classic Malkus–Proctor magnetic back-reaction. In the thermal/geostrophic picture, magnetic activity modulates convective heat transport and turbulent angular-momentum transport (Reynolds stresses), driving weak zonal flows to maintain thermal-wind balance. The equatorward-migrating branch closely follows the dynamo's toroidal-field belt, while a poleward branch at high latitudes may reflect enhanced polar activity. Reproducing both branches self-consistently remains a benchmark test for global convective-dynamo simulations.
Characteristic numbers, scales, and the governing balance
The numbers are small but well-measured. The zonal-velocity perturbation is δv ≈ 5–10 m s⁻¹, against a mean equatorial surface speed of ~2 km s⁻¹ (2000 m s⁻¹) — a fractional modulation δΩ/Ω of only a few ×10⁻³. The full pattern completes one cycle in ~11 years (the magnetic period; ~22 years for the full Hale magnetic cycle), and the equatorward branch takes several years to migrate from ~30° latitude to the equator.
The relevant balance is angular-momentum conservation in a rotating, magnetized fluid. Schematically, ∂(ρ⟨v_φ⟩)/∂t is set by the divergence of Reynolds stresses, Maxwell stresses, meridional-flow advection, and viscous transport. Because the driving Maxwell/Reynolds stresses are weak, the resulting flow perturbation is tiny — the ~10 m s⁻¹ amplitude directly reflects how little momentum the cyclic field extracts. Helioseismic inversions show the perturbation extends through much of the ~200,000 km-deep convection zone, not just the surface.
How it is detected: Doppler measurements and helioseismology
The torsional oscillation was first found in surface Doppler measurements. Robert Howard and Barry LaBonte, using the Mount Wilson 150-foot tower's daily full-disk Doppler magnetograms, reported the pattern in 1980 by removing the mean differential rotation and revealing residual ±few-m s⁻¹ bands migrating with the cycle.
The decisive advance came from helioseismology, which uses the Sun's p-mode acoustic oscillations to probe interior rotation. Rotational splitting of the modes' frequencies encodes Ω as a function of radius and latitude. Time-dependent inversions from the SOHO/MDI instrument, NSF's ground-based GONG network, and later SDO/HMI (and local-helioseismology techniques like ring-diagram and time–distance analysis) tracked the bands into the convection zone. These data confirmed the flows are deep-seated and revealed that the equatorward branch descends over the cycle. The signal is extracted by differencing rotation-rate inversions across many months to isolate the small time-varying residual from the steady profile.
Where it operates and how it differs from related flows
Torsional oscillations are a property of the Sun's convection zone — the outer ~30% by radius where turbulent convection and the dynamo operate. They should not be confused with the steady differential rotation on which they ride: differential rotation is the fixed background profile (fast equator, slow poles), while the torsional oscillation is its small, cyclic, migrating perturbation.
They also differ from meridional circulation, the slow north–south flow (~10–20 m s⁻¹ poleward at the surface) that advects magnetic flux in flux-transport dynamo models. Torsional oscillations are purely zonal (east–west). By analogy, the Sun's alternating fast/slow bands resemble the jet-stream-like zonal flows seen on the giant planets, though the driving physics differs. Because they are dynamo-driven, similar cyclic zonal flows are expected on other cool, magnetically active stars, and asteroseismology of Sun-like stars is beginning to constrain their interior rotation — a route to testing whether stellar torsional oscillations are universal.
Open questions and significance
Key questions remain unresolved. Which force dominates — direct Lorentz-force back-reaction or thermally/turbulently mediated stresses — is still debated, and the two may operate at different depths and latitudes. The origin of the poleward-migrating high-latitude branch, and whether it is truly independent of the equatorward branch, is unsettled. It is also unclear how deep the pattern really extends near the base of the convection zone, where the tachocline anchors the dynamo.
The stakes are high because the torsional oscillation is one of the few observables that constrains the momentum budget of the working dynamo. Because the mid-latitude bands appear a year or two before the sunspots that follow them, they are studied as a space-weather forecasting tool for predicting the onset and strength of upcoming solar cycles. Matching the observed ~10 m s⁻¹ amplitude, dual branches, and depth profile is now a demanding validation test for the global MHD simulations that aim to explain the solar cycle from first principles.
| Property | Torsional oscillation | Steady differential rotation | Meridional circulation |
|---|---|---|---|
| Direction of flow | Zonal (E–W), latitudinal bands | Zonal (E–W), latitude-dependent | Meridional (N–S), pole-equator loop |
| Typical speed | ~5–10 m s⁻¹ (perturbation) | ~2 km s⁻¹ equator vs ~1.6 km s⁻¹ poles | ~10–20 m s⁻¹ surface poleward |
| Time behaviour | Cyclic, ~11-yr period, migrating | Quasi-steady background | Quasi-steady, weak cyclic modulation |
| Latitude drift | Poleward + equatorward branches migrate | None (fixed profile) | Poleward at surface, equatorward at base |
| Depth extent | Much of convection zone (helioseismic) | Full convection zone | Full convection zone (single/multi-cell) |
| Link to activity | Equatorward branch tracks sunspot belt | Sets sunspot rotation rate | Advects flux (flux-transport dynamo) |
Frequently asked questions
How fast are the torsional oscillation bands compared to the Sun's normal rotation?
The bands are extremely subtle: the fast/slow perturbation is only about 5–10 m s⁻¹, whereas the Sun's equator rotates at roughly 2 km s⁻¹ (2000 m s⁻¹) at the surface. That is a fractional modulation of just a few tenths of a percent of the mean rotation rate, which is why they went undetected until careful Doppler subtraction in 1980.
Why do the bands migrate toward the equator?
The equatorward drift mirrors the migration of the solar dynamo's toroidal-field belt, which is where sunspots emerge. As the dynamo wave propagates from mid-latitudes toward the equator over the ~11-year cycle, the magnetic feedback on the flow — and hence the torsional-oscillation bands — moves with it. The bands slightly lead the sunspots, appearing a year or two earlier.
Who discovered solar torsional oscillations?
Robert Howard and Barry LaBonte reported them in 1980, using daily full-disk Doppler measurements from the 150-foot solar tower at Mount Wilson Observatory. After removing the mean differential rotation, they found residual bands of slightly faster and slower rotation drifting in latitude in step with the sunspot cycle.
How do we know the pattern extends into the Sun's interior?
Helioseismology reveals it. The Sun's acoustic p-mode oscillations are split in frequency by rotation, and inverting those splittings gives the rotation rate as a function of depth and latitude. Time-dependent inversions from SOHO/MDI, the GONG network, and SDO/HMI show the torsional bands persist through much of the ~200,000 km-deep convection zone, not just at the surface.
What actually drives the torsional oscillation?
It is the mechanical back-reaction of the cyclic magnetic field on the Sun's rotation. Two mechanisms compete: a direct Lorentz force from Maxwell stresses (magnetic tension/pressure braking or accelerating the plasma), and thermal/turbulent effects where magnetic activity alters convective heat and angular-momentum transport, driving weak zonal flows. Which dominates, and where, is still debated.
How are torsional oscillations different from meridional circulation?
Torsional oscillations are zonal (east–west) flows — bands rotating slightly faster or slower than average. Meridional circulation is a north–south flow, roughly 10–20 m s⁻¹ poleward at the surface, that carries magnetic flux in flux-transport dynamo models. Both are subtle solar-cycle flows, but they act in perpendicular directions and play different roles in the dynamo.