Helicity Reversal Patterns in Solar Plasmas

• Current helicity reversal pattern is defined as the sign change in B · (curl B) used to diagnose magnetic field dynamics in solar active regions.
• Methods include global 3D convective dynamo simulations, 2D mean-field models, and vector-magnetogram surveys that reveal scale, temporal, and field-strength dependencies.
• Insights from these reversals advance our understanding of solar eruptions, active region evolution, and their implications for space weather forecasting.

Searching arXiv for relevant papers on current helicity reversal patterns and related solar/heliity studies. Current helicity reversal pattern denotes a class of sign changes in helicity-related diagnostics of magnetized plasmas. In solar physics, the phrase most commonly refers to departures from the empirical hemispheric sign rule—negative current helicity in the northern hemisphere and positive in the southern—or to temporal sign-definite transitions during the evolution of an active region or an eruption. The pattern has been identified in global 3D convective dynamos, 2D mean-field dynamo models, photospheric vector-magnetogram surveys, helicity-flux studies of emerging regions, and flare/CME analyses. Across these settings, reversals are linked to cross-equatorial toroidal-band interaction, dynamo-wave migration, scale- and field-strength segregation, successive injection of opposite helicity, or redistribution of electric currents relative to the polarity inversion line (Miesch et al., 2016, Sokoloff et al., 2012, Sun et al., 15 Jul 2025).

1. Definitions and diagnostic conventions

In MHD treatments, magnetic helicity is defined as

$H_m \equiv \int_V A\cdot B\, dV,$

and, in the spherical-shell simulation with an outer boundary matched to a potential field, this coincides with the relative helicity and is gauge-invariant. Using the poloidal–toroidal decomposition

$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$

the local magnetic-helicity density is written as

$h_m(r,\theta,\phi)=A\,B_r,$

with

$H_m=2\int_V h_m\, dV.$

Current helicity density is defined by

$h_c \equiv B\cdot(\nabla\times B),$

with associated volume integral

$H_c=\frac{c}{4\pi}\int_V h_c\, dV$

(Miesch et al., 2016).

Photospheric observations usually replace the full volume quantity with a surface proxy. In Hinode/SOT-SP analyses this proxy is

$H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$

or, magnetogram by magnetogram,

$H_C(z)\equiv \langle B_zJ_z\rangle,$

where

$$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$

Equivalent notation appears in flare-oriented work as

$H_c=\int j_zB_z\, dS.$

A related twist parameter is

$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$0

with

$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$1

Hinode/SP studies also use the normalized average

$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$2

(Otsuji et al., 2014, Sun et al., 15 Jul 2025, Gosain et al., 2013, Hao et al., 2011).

These definitions are not interchangeable in a strict dynamical sense. The solar literature repeatedly treats $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$3, $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$4, and $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$5 as distinct but related observables or proxies, with the choice depending on whether the target is the photosphere, the large-scale dynamo field, or the fluctuating component (Sokoloff et al., 2012).

2. Reversals in cyclic dynamo models

A global 3D convective dynamo provides one of the clearest cycle-scale realizations of helicity reversal. In Case K3S, computed with the ASH code under the anelastic approximation in a full spherical shell $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$6, coherent toroidal “wreaths” form within the convection zone, with opposite polarities in the two hemispheres. Throughout most of the cycle, the zonally averaged magnetic helicity density $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$7 is negative in the northern hemisphere and positive in the southern, while during the declining phase this hemispheric rule reverses. Near $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$8, the rising phase gives $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$9–$h_m(r,\theta,\phi)=A\,B_r,$0 and $h_m(r,\theta,\phi)=A\,B_r,$1–$h_m(r,\theta,\phi)=A\,B_r,$2; during decline, a brief reversal appears, with the northern hemisphere becoming weakly positive at approximately $h_m(r,\theta,\phi)=A\,B_r,$3 for about $h_m(r,\theta,\phi)=A\,B_r,$4–$h_m(r,\theta,\phi)=A\,B_r,$5. The current-helicity proxy flips sign slightly earlier than the magnetic-helicity proxy, consistent with the resistive contribution $h_m(r,\theta,\phi)=A\,B_r,$6 (Miesch et al., 2016).

The same simulation ties the reversal to a specific topological sequence. Toroidal wreaths migrate equatorward, come into magnetic contact across the equator, reconnect, and generate an isolated, axisymmetric “magnetic bubble” of opposite poloidal polarity. This bubble carries reversed helicity, rises on an Alfvénic time scale, expands poleward, and reverses the sign of the polar fields. The causal chain is stated explicitly as equatorward drift $h_m(r,\theta,\phi)=A\,B_r,$7 cross-equatorial reconnection $h_m(r,\theta,\phi)=A\,B_r,$8 bubble of opposite helicity $h_m(r,\theta,\phi)=A\,B_r,$9 poleward expansion $H_m=2\int_V h_m\, dV.$0 flip in both toroidal and poloidal fields (Miesch et al., 2016).

Mean-field models reproduce related patterns, but with an important diagnostic distinction. In the advanced 2D model with helicity evolution, the small-scale current helicity

$H_m=2\int_V h_m\, dV.$1

forms a strong, nearly stationary mid-latitude belt with only weak oscillatory departures, whereas the large-scale magnetic helicity

$H_m=2\int_V h_m\, dV.$2

migrates in a familiar butterfly pattern. The proxy for active-region current helicity,

$H_m=2\int_V h_m\, dV.$3

sampled at the top of the overshoot layer, reproduces equatorward-drifting wings, a northern-hemisphere negative sign during cycle rise and maximum, and brief reversal at the tails of the wings. The simpler algebraic-quenching model gives a qualitatively similar proxy butterfly (Sokoloff et al., 2012).

A separate mean-field solar-type dynamo study attributes helicity sign-rule reversals to the propagation of the dynamo wave itself. In that model, small-scale magnetic helicity reversal bands follow the equatorward-migrating toroidal-field butterfly; their penetration to the surface depends strongly on the top boundary condition for helicity. With a random, non-equatorially-symmetric variation of up to $H_m=2\int_V h_m\, dV.$4 per cycle in the $H_m=2\int_V h_m\, dV.$5-effect amplitude, low-amplitude cycles produce especially strong and extended reversal bands, reversal latitude varies by about $H_m=2\int_V h_m\, dV.$6–$H_m=2\int_V h_m\, dV.$7, and the depth from which the reversal band emerges can vary by approximately $H_m=2\int_V h_m\, dV.$8 (Pipin et al., 2013).

3. Hemispheric rule, cycle phase, and observational reversals

The observational baseline for solar current helicity is the hemispheric sign rule: active regions are predominantly negative in the northern hemisphere and positive in the southern. The rule is statistical rather than exact. One survey notes about $H_m=2\int_V h_m\, dV.$9–$h_c \equiv B\cdot(\nabla\times B),$0-level consistency, while another reports that sign dominance exceeds about $h_c \equiv B\cdot(\nabla\times B),$1–$h_c \equiv B\cdot(\nabla\times B),$2 of regions in the weak-field regime (Miesch et al., 2016, Otsuji et al., 2014).

Cycle phase modulates the pattern. Reports summarized in the convective-dynamo study state that Tiwari et al. (2009) and Hao & Zhang (2011) found a weak reversal of the northern-hemisphere rule in the declining phase of Cycle 23, and Hao & Zhang (2011) found only $h_c \equiv B\cdot(\nabla\times B),$3 of regions obeying the usual rule during the decline (Miesch et al., 2016). A dedicated Hinode/SP study of $h_c \equiv B\cdot(\nabla\times B),$4 active regions, spanning the descending phase of Cycle 23 and the ascending phase of Cycle 24, found that the $h_c \equiv B\cdot(\nabla\times B),$5 active regions of Cycle 24 follow the hemispheric helicity rule, whereas the $h_c \equiv B\cdot(\nabla\times B),$6 active regions of Cycle 23 do not. In the descending phase of Cycle 23, only $h_c \equiv B\cdot(\nabla\times B),$7 active regions obey the usual sign rule when measured by $h_c \equiv B\cdot(\nabla\times B),$8, and $h_c \equiv B\cdot(\nabla\times B),$9 by $H_c=\frac{c}{4\pi}\int_V h_c\, dV$0; in the ascending phase of Cycle 24, $H_c=\frac{c}{4\pi}\int_V h_c\, dV$1 follow the rule in both $H_c=\frac{c}{4\pi}\int_V h_c\, dV$2 and $H_c=\frac{c}{4\pi}\int_V h_c\, dV$3. The combined sample returns the usual negative slope with latitude, but with large scatter and correlation coefficients only up to about $H_c=\frac{c}{4\pi}\int_V h_c\, dV$4 (Hao et al., 2011).

Synoptic vector maps provide a complementary large-sample view. Using the first Carrington maps of the photospheric vector field from SOLIS/VSM, covering $H_c=\frac{c}{4\pi}\int_V h_c\, dV$5 rotations from CR 2109 to CR 2131 and $H_c=\frac{c}{4\pi}\int_V h_c\, dV$6 NOAA active regions, longitudinal averages show the standard negative-north/positive-south pattern during the rising phase of solar cycle 24. In that interval, there is no evidence for a systematic reversal of the hemispheric helicity rule at cycle onset, despite predictions from some dynamo models. Patches of opposite sign appear, but the global rule remains in place through the analyzed rotations (Gosain et al., 2013).

These results establish that “reversal” in the observational literature does not denote a single phenomenon. It may refer to a decline-phase departure from the usual hemispheric rule, to short wrong-sign patches at the beginning and end of butterfly wings, or to the absence of an expected cycle-start reversal in synoptic averages. The common element is a sign change relative to the negative-north/positive-south baseline, but the time scale and spatial support differ across diagnostics (Sokoloff et al., 2012, Gosain et al., 2013).

4. Field-strength, inclination, and scale dependence

One persistent result is that current-helicity reversals depend on which magnetic structures are sampled. In Hinode/SP statistics covering $H_c=\frac{c}{4\pi}\int_V h_c\, dV$7 vector magnetograms of $H_c=\frac{c}{4\pi}\int_V h_c\, dV$8 active regions from 2006 to 2012, weak fields defined by $H_c=\frac{c}{4\pi}\int_V h_c\, dV$9 obey the standard hemispheric rule when no smoothing is applied, whereas medium fields with $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$0 violate it, and strong fields show a qualitatively similar violation. With a $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$1 arcsec Gaussian smoothing, weak fields still retain the rule, but the medium/strong-field violation becomes more mixed. The same study finds that weak, highly inclined fields carry the “correct-sign” helicity, while strong, more vertical fields systematically show “wrong-sign” helicity; the slope of the mean $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$2 crosses zero at about $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$3–$H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$4 and $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$5–$H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$6 (Otsuji et al., 2014).

SOLIS/VSM synoptic maps produce a related but not identical partition. In active-region latitudes, strong fields defined by $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$7 obey the usual hemispheric pattern, while weak fields with $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$8 exhibit the inverse hemispheric behavior, albeit with large statistical scatter. Two scenarios are proposed: a two-scale helicity separation in mean-field dynamo theory, and near-surface flow-driven twist, with cyclonic flows beneath active regions and anticyclonic flows in quiet-Sun regions contributing opposite signs (Gosain et al., 2013).

Spatial segregation also appears inside individual sunspots. Two Hinode/SP examples show that both $H_C(z)=\int_{z=0} B_zJ_z\, dx\,dy,$9 and $H_C(z)\equiv \langle B_zJ_z\rangle,$0 change sign from inner umbra to outer penumbra, with the zero crossing at about $H_C(z)\equiv \langle B_zJ_z\rangle,$1–$H_C(z)\equiv \langle B_zJ_z\rangle,$2. In NOAA 10940, the inner umbra has positive mean $H_C(z)\equiv \langle B_zJ_z\rangle,$3 and $H_C(z)\equiv \langle B_zJ_z\rangle,$4, while the penumbra is negative; in NOAA 11084 the polarity is reversed, but the same umbra–penumbra sign flip occurs. Whole-active-region averages follow the sign of the penumbra rather than the umbral core. By raising the $H_C(z)\equiv \langle B_zJ_z\rangle,$5 threshold from $H_C(z)\equiv \langle B_zJ_z\rangle,$6 to approximately $H_C(z)\equiv \langle B_zJ_z\rangle,$7, the fitted latitude slope reverses above about $H_C(z)\equiv \langle B_zJ_z\rangle,$8 in the flux-density case or about $H_C(z)\equiv \langle B_zJ_z\rangle,$9 in the field-strength case (Hao et al., 2011).

Spectral analysis adds a scale-space formulation of the same issue. Under isotropy, the current-helicity spectrum satisfies

$$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$0

For NOAA 11515, located at about $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$1 south, the usual southern-hemisphere rule would predict positive current helicity, yet the integrated $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$2 is strongly negative. The signed spectrum shows that for $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$3, corresponding to scales $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$4, $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$5, while for $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$6, corresponding to $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$7 up to about $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$8, $$J_z=\frac{\partial B_y}{\partial x}-\frac{\partial B_x}{\partial y}.$$9. Because the spectrum is steep at small $H_c=\int j_zB_z\, dS.$0, the negative large-scale helicity dominates the integral and reverses the net sign. The same study interprets this as a bihelical organization in which small- and large-scale fields carry opposite helicity, with reversals becoming more likely when large-scale power is enhanced (Zhang et al., 2015).

5. Emergence-, transport-, and eruption-driven reversals

Reversals are also observed as temporal transitions within a single active region. During the emergence of NOAA 11928, helicity-flux computations from SDO/HMI vector magnetograms and DAVE4VM/DAVE flow tracking show a change in the sign of the photospheric helicity flux. The net helicity flux rises from zero to a positive maximum of about $H_c=\int j_zB_z\, dS.$1 by 18:00 UT on 17 December 2013, then declines, crosses zero, and becomes negative. The zero crossing occurs at 22:00 UT on 18 December with DAVE on line-of-sight data and at 00:20 UT on 20 December with DAVE4VM. The tangential-motion term dominates the net injection by a factor of about $H_c=\int j_zB_z\, dS.$2 over the emergence term. Independent corroboration comes from the summed vertical currents in each polarity, which both reach about $H_c=\int j_zB_z\, dS.$3 and then reverse sign at 22:00 UT on 18 December, and from the mean twist proxy $H_c=\int j_zB_z\, dS.$4, which flips from positive to negative at the same time. EUV morphology evolves from an S-shaped sigmoid to a more potential-like arcade after the reversal (Vemareddy et al., 2016).

A distinct event-scale reversal pattern appears in the response to coronal mass ejections. In a 3D MHD model using the DARE–MHD code, with steady rotational driving at the bottom boundary, the photospheric current helicity

$H_c=\int j_zB_z\, dS.$5

shows a sign-definite temporal sequence relative to eruption onset: a pre-eruption decrease and a post-eruption increase. In the simulation, magnetic flux remains essentially constant, total unsigned current rises by about $H_c=\int j_zB_z\, dS.$6 over the driving interval, and $H_c=\int j_zB_z\, dS.$7 drops by about $H_c=\int j_zB_z\, dS.$8–$H_c=\int j_zB_z\, dS.$9 in the $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$00–$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$01 minutes before eruption before rising by about $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$02–$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$03 in the first $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$04–$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$05 minutes afterward. The mechanism is current redistribution: before eruption, shearing drives currents inward toward the polarity inversion line, reducing overlap between strong $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$06 and $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$07 in the regions that dominate the integral; after eruption, reconnection and flare-ribbon separation move the $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$08 enhancements outward again, increasing $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$09 in strong-field areas (Sun et al., 15 Jul 2025).

An observational survey of $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$10 eruptive flares with GOES class $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$11M5.0 recovers the same pattern in a statistical sense. Using SDO/HMI SHARP vector magnetograms at 12-minute cadence and AIA 1600 Å flare-ribbon masks, and counting only pixels with $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$12 inside the ribbon mask, $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$13 of events show a clear pre-eruption decrease and $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$14 show a post-eruption increase in current helicity. The detection criteria are stringent: a pre-eruption drop must be at least $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$15 lasting at least $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$16, and a post-eruption rise must be at least $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$17 lasting at least $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$18 minutes. Among the decrease cases, the mean drop is $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$19 over $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$20; among the increase cases, the mean rise is $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$21 over $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$22. Long-term evolution in two representative cases shows that $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$23 can begin a gradual decline $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$24–$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$25 before the main flare (Sun et al., 15 Jul 2025).

Within solar active-region physics, this event-driven use of “reversal pattern” differs from the hemispheric-rule literature. Here the central object is not north-versus-south sign, but a reproducible temporal dip-and-rise sequence in the same eruptive region. The data suggest that current helicity can act as a tracer of where currents are concentrated before eruption and how they relax after reconnection (Sun et al., 15 Jul 2025).

6. Conceptual scope and non-solar usages

Outside solar MHD, the word “helicity” is used for different magnetic attributes, and reversal denotes different operations. In a nanostructured frustrated Fe$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$26Sn$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$27 magnet, a skyrmionic bubble is described by

$$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$28

where $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$29 is the helicity. Two Bloch-type states exist with $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$30 and $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$31. Current pulses induce reversal between these two states, with threshold current density decreasing from about $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$32 at $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$33 to about $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$34 at $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$35. Simulations show that strong pinning favors helicity reversal, weak pinning favors depinning motion, and dipole-dipole interaction provides the energy barrier between the two helicity minima (Hou et al., 2019).

In ferromagnetic Pt/Co/Pt films, “helicity-dependent” refers instead to the circular polarization of optical pulses. The switching efficiency increases by a factor of $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$36 when the pump wavelength is increased from $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$37 to $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$38 and reaches $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$39 for wavelengths $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$40. The spectral dependence tracks magnetic circular dichroism rather than the inverse Faraday effect. This is a helicity-controlled magnetization reversal, but not a current-helicity reversal in the solar-MHD sense (Yamada et al., 2022).

These non-solar usages underscore a terminological point. In the solar literature, current helicity reversal concerns changes in the sign of $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$41 or its photospheric proxies; in skyrmion physics it concerns reversal of the in-plane swirling offset $$B=\nabla\times\nabla\times(C\,e_r)+\nabla\times(A\,e_r),$$42; in all-optical switching it concerns the helicity of the incident light. The shared term “helicity” therefore spans different observables, different control parameters, and different reversal mechanisms (Hou et al., 2019, Yamada et al., 2022).