Quasi-periodic Fast Propagating Magnetosonic Waves (QFPs)

• QFPs are coherent, arc‐shaped fast-mode magnetosonic waves in the corona, observed in EUV with speeds up to 2000 km/s and periods from tens of seconds to minutes.
• They are classified into narrow and broad categories, distinguished by angular extent, intensity variation, and propagation geometry, which aids in understanding their source and waveguide effects.
• Advanced imaging and Fourier analyses extract key kinematic and spectral diagnostics, enabling coronal seismology to estimate magnetic field strengths and probe flare-CME interactions.

Searching arXiv for recent and foundational QFP literature to ground the article. Quasi-periodic fast propagating magnetosonic wave trains, usually abbreviated QFP wave trains or QFPs, are spatially resolved trains of fast-mode magnetosonic disturbances in the solar corona. In EUV observations they appear as multiple coherent and concentric wavefronts emanating successively near the epicenter of accompanying flares and propagating outward either along or across coronal loops at fast-mode magnetosonic speeds from several hundred to more than $2000~\mathrm{km\,s^{-1}}$, with periods in the range of tens of seconds to several minutes (Shen et al., 2021). Their direct imaging became possible in the SDO/AIA era, and one foundational event on 2010 August 1 showed arc-shaped wave trains of $1$–$5\%$ intensity variations, a lifetime of $\sim 200$ s for individual fronts, propagation up to $\sim 400$ Mm along a funnel of coronal loops, and a Fourier $k$-$\omega$ ridge whose strongest signal at $5.5$ mHz temporally coincided with flare quasi-periodic pulsations (Liu et al., 2011).

1. Historical emergence and physical identification

QFPs were anticipated theoretically before SDO, but the modern observational field is anchored in AIA’s full-Sun field of view, 7 EUV channels, $12$ s cadence, and $1.2$ arcsec spatial resolution (Shen et al., 2021). A short synthesis published later characterized QFPs as a distinct coronal wave phenomenon discovered in EUV by SDO/AIA, typically propagating at speeds up to $1$0 within funnel-shaped waveguides in the wakes of CMEs (Liu et al., 2015). In that literature, the preferred physical interpretation is that QFPs are propagating, compressive, fast-mode magnetosonic waves rather than standing oscillations or generic brightness fluctuations.

The basic identification relies on a conjunction of morphology, kinematics, and spectral behavior. QFPs are observed as multiple arc-shaped fronts rather than a single pulse; they propagate much faster than typical slow-mode disturbances; and in Fourier space they generate a steep, narrow ridge in the $1$1-$1$2 diagram, interpreted as the wave dispersion relation (Liu et al., 2011). The review literature further notes that they undergo reflection, refraction, transmission, and trapping, which strongly supports their classification as genuine fast-mode disturbances (Shen et al., 2021).

Historically, the phenomenon quickly expanded beyond the original flare-funnel picture. Subsequent observations established loop-guided, open-funnel, broad-surface, bidirectional, and counter-propagating cases, together with associations with flares, CMEs, jets, mini-filament eruptions, flux-rope eruptions, and direct coronal-loop reconnection (Liu et al., 2015).

2. Observational phenomenology and the narrow–broad division

A major review conclusion is that QFP wave trains can be divided into two distinct categories, narrow and broad, on the basis of angular extent, thermal visibility, amplitude, and propagation geometry (Shen et al., 2021). Narrow QFPs usually propagate along coronal loops or funnels, whereas broad QFPs propagate on the solar surface or across the quiet Sun and are more closely connected to large-scale EUV-wave behavior.

Category	Typical properties	Preferred visibility
Narrow QFPs	Speeds $1$3–$1$4, periods $1$5–$1$6, wavelengths $1$7–$1$8, angular extent $1$9–$5\%$0, intensity amplitudes $5\%$1–$5\%$2	Best seen in AIA 171 Å
Broad QFPs	Speeds $5\%$3–$5\%$4, periods $5\%$5–$5\%$6, wavelengths $5\%$7–$5\%$8, angular extent $5\%$9–$\sim 200$0, intensity amplitudes $\sim 200$1–$\sim 200$2	Seen in all AIA EUV channels, best in 193 Å and 211 Å

The same review gives deceleration ranges of $\sim 200$3–$\sim 200$4 for narrow QFPs and $\sim 200$5–$\sim 200$6 for broad QFPs, typical lifetimes of $\sim 200$7–$\sim 200$8 min, and a longest reported duration of about $\sim 200$9 h (Shen et al., 2021). It also notes that QFPs can propagate a long distance over $\sim 400$0, usually first becoming visible at distances greater than $\sim 400$1 from the flare epicenter.

An especially instructive case of both classes occurring in one eruptive complex involved a GOES C2.9 flare in NOAA AR 11868 on 2013 October 20. During successive filament eruptions in a fan–spine magnetic system, a narrow QFP first propagated along the outer spine/open loop system and was followed by a broad QFP over the quiet Sun. The narrow train had width $\sim 400$2, corrected speed $\sim 400$3, dominant period $\sim 400$4, relative intensity perturbation $\sim 400$5, and lower-limit energy flux $\sim 400$6, whereas the broad train had width $\sim 400$7, speed $\sim 400$8, period $\sim 400$9, $k$0, and lower-limit energy flux $k$1 (Zhou et al., 2024). This suggests that the narrow–broad division is not merely descriptive but can encode genuinely different source and propagation physics.

3. Measurement strategies and spectral diagnostics

QFP analysis is dominated by imaging-based diagnostics. Running-difference, base-difference, and running-ratio images are used because many narrow QFPs have intensity amplitudes of only a few percent and may be difficult to detect in direct images (Shen et al., 2021). Time–distance diagrams are then constructed by extracting one-dimensional intensity profiles along a slit, curved path, or sector and stacking them in time; the propagating fronts appear as oblique ridges whose slopes yield projected phase speeds.

A standard Fourier treatment transforms a three-dimensional $k$2 AIA data cube into $k$3, converts to cylindrical wavenumber $k$4, and produces a $k$5-$k$6 diagram in which QFPs appear as a steep ridge (Shen et al., 2012). The basic diagnostics are

$k$7

so the ridge slope gives the phase and group speeds. In the 2011 May 30 event, the $k$8-$k$9 ridge was well fit by a straight line through the origin, giving $\omega$0, while the ridge nodes revealed frequencies spanning $\omega$1–$\omega$2 (Shen et al., 2012).

Wavelet analysis is used to extract periods from local intensity time series and from flare light curves. In QFP work this is commonly combined with Morlet wavelets and significance testing, as in the 2011 May 30 event, where five of seven main flare frequencies coincided with QFP frequencies, and in many later studies comparing QFPs with GOES, RHESSI, or radio quasi-periodic pulsations (Shen et al., 2012). A different methodological emphasis appeared in a 2022 narrow-QFP study, where Fourier $\omega$3 analysis, spatial sinusoidal fitting, wavelet analysis, and DEM inversion were combined to argue that the strongest train had wavelength $\omega$4, dominant period about $\omega$5–$\omega$6 s, and nearly equal phase and group speeds of $\omega$7, which the authors interpreted as nearly non-dispersive propagation (Zhou et al., 2022).

Thermal and plasma diagnostics also enter through DEM inversion and channel-dependent visibility. The review literature emphasizes that narrow QFPs are best seen in 171 Å, occasionally in 193 Å and 211 Å, whereas broad QFPs can be seen in all AIA EUV channels and are best seen in 193 Å and 211 Å (Shen et al., 2021). This temperature dependence is not merely instrumental; it constrains the thermodynamic state of the propagation medium.

4. Excitation mechanisms

Current interpretations of QFP generation fall into two main categories: pulsed energy excitation associated with magnetic reconnection, and dispersive evolution of impulsively generated broadband perturbations (Shen et al., 2021). A third, more limited possibility is that some periods are supplied by leakage of three- and five-minute oscillations from the lower atmosphere (Shen et al., 2021).

Evidence for flare-linked, periodically driven excitation is strongest in cases where QFP periods match flare pulsations. In the 2011 May 30 event, almost all the main frequencies of the flare were consistent with those of the QFP wave, suggesting that the flare and the QFPs were possibly excited by a common physical origin, likely periodic or oscillatory reconnection (Shen et al., 2012). A 2022 event sharpened this argument: three recurrent narrow QFP wave trains during a GOES C4.2 flare had onset times tightly correlated with bumps in the GOES derivative, and the dominant wave period $\omega$8 matched the flare periodicities $\omega$9 to $5.5$0; because $5.5$1, the authors argued against dispersive formation in that case (Zhou et al., 2022).

Dispersive formation is favored in events where the wave periods do not match flare pulsations. A clear example is the 2015 July 12 mini-filament event, where a nearby mini-filament eruption and only a small B4 GOES flare produced a QFP with average speed $5.5$2, acceleration $5.5$3, and reliable periods $5.5$4 s and $5.5$5 s, while the flare periodicities were qualitatively different (Shen et al., 2018). That study proposed dispersive evolution of an initially broadband disturbance generated by the filament eruption. A different non-flare-dominated route was identified in loop–loop reconnection on 2012 January 19, where quasi-periodic fast propagating disturbances with projected speeds $5.5$6–$5.5$7, mean speed $5.5$8, and peak period $5.5$9 min originated directly from the reconnection region during coronal condensation; the authors interpreted them as reconnection-driven QFPM waves rather than flow ejecta (Li et al., 2018).

More recent work has broadened the source inventory further. A 2023 numerical study proposed that large-scale QFPs observed on both sides of a CME can arise when a disturbance within an erupting magnetic flux rope leaks successively through the rope surface, the flux rope acting as an imperfect waveguide; synthesized images were found consistent with observations (Hu et al., 2023). A 2025 three-dimensional MHD simulation identified QFPs originating from both ends of a reconnecting current sheet during a solar eruption, with speeds of approximately $12$0 and period $12$1 s, and attributed them to a tuning-fork effect involving reconnection outflows, termination shocks, and oscillating fork-like current-sheet-end structures (Hu et al., 16 Oct 2025). Taken together, these results suggest that QFP excitation is not monolithic: periodic flare energy release, dispersive filtering, direct reconnection dynamics, eruptive filament or jet disturbances, flux-rope leakage, and current-sheet tuning-fork oscillations can all, in different regimes, generate QFP-like fast-mode wave trains.

5. Propagation, interactions, and coronal seismology

QFP propagation is strongly shaped by coronal magnetic structure. One well-observed flux-rope event on 2014 March 23 showed two branches with distinct kinematics and morphologies: the northern branch propagated at $12$2, lasted about $12$3 min, had intensity variation about $12$4, and exhibited obvious refraction when passing through a region of strong magnetic field; the southern branch propagated at $12$5, lasted about $12$6 min, and had intensity variation about $12$7 (Shen et al., 2017). The authors interpreted the refraction as evidence that different magnetic distributions along different paths control QFP speeds and morphological evolution.

Propagation can also be altered abruptly by waveguide interactions. In the 2012 April 23 event, a QFP initially observed in 171 Å at $12$8 along a divergent open loop system decelerated sharply to $12$9 after interaction with an underlying nearly perpendicular loop system, and only then became visible in 193 Å; this was interpreted through geometry, compression-induced density increase, and possible adiabatic heating of the guiding structure (Shen et al., 2013). A jet-driven event on 2011 February 14 showed a QFP with average speed $1.2$0, lifetime about $1.2$1 min, angular extent $1.2$2, period $1.2$3 s, and clear refraction through strong magnetic regions, while a coexisting broad EUV wave and a loop kink oscillation were interpreted as distinct disturbances launched by the same jet–loop interaction (Shen et al., 2018).

Counter-propagation and mode coupling further enrich the phenomenology. On 2013 May 22, neighboring flares generated the first direct observation of counter-propagating QFPs along large-scale trans-equatorial loops, with periods in the $1.2$4–$1.2$5 min range, intensity variations $1.2$6–$1.2$7, and propagation speeds of order $1.2$8; the same event also excited trapped kink and slow modes and yielded a seismological loop magnetic field of about $1.2$9 G (Ofman et al., 2018). A 2019 bidirectional event showed two QFPs guided by oppositely oriented funnels with essentially identical periods of about one minute, matching the periods of GOES soft X-ray derivative and 17 GHz radio oscillatory signals, which argued that the periodicity was imposed by the flaring core rather than independently by the two waveguides (Miao et al., 2021).

QFPs are therefore valuable seismological probes. The review literature uses the fast-mode relations

$1$00

with the perpendicular and parallel limits

$1$01

to estimate magnetic field strengths from measured QFP speeds and plasma densities (Shen et al., 2021). Reported seismological values include $1$02 G for a narrow flare-driven QFP waveguide (Zhou et al., 2022), about $1$03 G and $1$04 G for bidirectional funnel-guided QFPs (Miao et al., 2021), and, in a recent multi-path event, $1$05–$1$06 G for a narrow funnel and $1$07–$1$08 G for a broad low-coronal path (Miao, 9 Jul 2026). This suggests that multi-path QFPs can perform differential coronal magnetometry within the same flare system.

Energetically, QFPs can carry non-negligible flux. The review gives $1$09–$1$10 for narrow QFPs and $1$11–$1$12 for broad QFPs (Shen et al., 2021). A commonly used lower-limit estimate is

$1$13

derived from $1$14 and the perturbed kinetic-energy flux (Shen et al., 2021). In the 2010 August 1 seminal event, the estimated base energy flux $1$15–$1$16 was stated to be comparable to the steady-state heating requirement of active-region loops, though the transient and localized nature of such events limits any simple global-heating inference (Liu et al., 2011).

6. Outstanding problems and recent directions

Despite a decade of rapid progress, the review literature remains explicit that the excitation mechanism of QFPs is unresolved in general (Shen et al., 2021). The central ambiguity is whether an observed period is imposed at the source by pulsed reconnection or eruption dynamics, or instead emerges through waveguide dispersion from a broadband impulse. The relation to flare QPPs is correspondingly mixed: some events show nearly complete period matching, some only partial overlap, and some none at all (Shen et al., 2021).

Statistical understanding is also incomplete. A preliminary survey of 355 global EUV waves from June 2010 to December 2014 identified 155 preliminary QFP events and 112 definitive ones, implying that about one third of global EUV-wave events were accompanied by QFPs; the median flare class was M1.0, and no clear correlation was found between flare class and QFP significance (Liu et al., 2015). The same synthesis suggested that QFPs appear preferentially associated with eruptive flares rather than confined ones (Liu et al., 2015). The 2021 review argued that the one-third occurrence rate is probably an underestimate, because many QFPs occur without global EUV waves and because the low amplitudes of narrow QFPs make them difficult to detect, motivating automatic detection tools and larger homogeneous surveys (Shen et al., 2021).

Several recent developments indicate where the field is moving. Realistic active-region MHD modeling has begun to replace idealized waveguides: a 2025 study using a potential-field extrapolation of AR 11166 found that a localized source in the observed QFP region produced directional, localized propagation in a diffuse, leaky waveguide, with damping dominated mainly by geometrical spreading and aided by resistive dissipation (Ofman et al., 3 Oct 2025). The 2025 tuning-fork current-sheet model and the 2023 flux-rope-leakage model both extend the source physics beyond standard flare-kernel forcing (Hu et al., 16 Oct 2025, Hu et al., 2023). The 2026 multi-path study, in which a single M6.0 flare simultaneously launched oppositely oriented narrow and broad QFPs with similar $1$17–$1$18 s periods but very different speeds and magnetic environments, points toward using QFPs not only as wave diagnostics but also as probes of flare-core periodicity and coronal magnetic topology in the same event (Miao, 9 Jul 2026).

A plausible implication is that the term “QFP” now designates a phenomenological family rather than a single-generation mechanism. What unifies the family is the combination of fast-mode magnetosonic propagation, quasi-periodic or wave-train morphology, and strong sensitivity to magnetic structuring. What remains unsettled is how much of the observed periodicity is source-driven, how much is waveguide-driven, and under what conditions the two become observationally distinguishable.