Likely Common Coronal Source of Solar Wind and 3He-enriched Energetic Particles: Uncoupled Transport from the Low Corona to 0.2 au

Parker Solar Probe (PSP) observations of a small dispersive event on 2022 February 27 and 28 indicate scatter-free propagation as the dominant transport mechanism between the low corona and greater than 35 solar radii. The event occurred during unique orbital conditions that prevailed along specific flux tubes that PSP encountered repeatedly between 25 and 35 Rs during outbound orbit 11. This segment of the PSP orbit exhibits almost stationary angular motion relative to the rotating solar surface, such that in the rotating frame, PSP’s motion is essentially radial. The time dispersion often observed in impulsive solar energetic particle (SEP) events continues in this case down to velocities including the core solar-wind ion velocities. Especially at the onset of this event, the 3He content is much larger than the usual SEP abundances seen in the energy range from ∼100 keV to several MeV for helium. Later in the event, iron is enhanced. The compositional signatures suggest this to be an example of an acceleration mechanism for generating the seed energetic particles required by shock (or compression) acceleration models in SEP events to account for the enrichment of various species above solar abundances in such events. A preliminary search of similar orbital conditions over the PSP mission has not revealed additional such events, although favorable conditions (isolated impulsive acceleration and well-ordered magnetic field connection with minimal magnetic field fluctuation) that would be required are infrequently realized, given the small fraction of the PSP trajectory that meets these observation conditions.


Introduction
Solar energetic particle (SEP) events observed in interplanetary space have been associated both with flares and with coronal mass ejections (CMEs).Broadly speaking, the acceleration of SEPs is often associated with one or more of four processes: reconnection, compressions, shocks, and waveparticle heating/acceleration, with reconnection generally describing the acceleration of particles in solar flares (e.g., Zharkova et al. 2011) and diffusive shock acceleration for the case of CME-driven shocks (Reames 1999;Melrose 2009).Low in the corona, compressions likely serve as significant accelerators even prior to the development of shocks.Characteristic broken power-law spectra provide observational signatures of particle acceleration from the compressive structures in the low corona (Schwadron et al. 2015).
It has been widely known that energetic particle seed populations are often rich with nearly scatter-free electrons and species such as 3 He, known to be flare associated (Mason et al. 1986;Reames 1999;Mason et al. 2002;Desai et al. 2003).The enhancements in energetic particle seed populations of 2019 April 18-24 (Schwadron et al. 2020) demonstrate how the early evolution of CMEs enhance the fluxes of energetic particle seed populations, which precondition the particle-acceleration process at distances farther from the Sun where compressions can steepen into shocks.The ISeIS observations from Schwadron et al. (2020) below 1 MeV show a very hard energy spectrum, indicating that it is likely a superposition of particles from multiple flares.The spectrum is close to the E −1.5 limit of possible stationary-state plasma distributions out of equilibrium (Livadiotis & McComas 2009, 2010).
The compositional features observed for some of the events associated with SEP shock and compression acceleration processes rely on the prior production of energetic particle seed populations.The subsequent acceleration of seed populations by compressions and shocks occurs as structures propagate throughout the interplanetary medium.A key question remains as to how exactly energetic particle seed populations are formed to begin with.ISeIS data from Parker Solar Probe (PSP) close to the Sun provides essential new data Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
by disentangling isolated events from close to the Sun that feed into the seed populations of energetic particles.Raouafi et al. (2023a) demonstrated that ubiquitous magnetic reconnection at small scales generates tiny jets of hot plasma, known as jetlets (Raouafi & Stenborg 2014).They argued that these jetlets are the source of both regimes of the solar wind.Bale et al. (2023) proposed a model for the acceleration of solar wind in the low corona through reconnection between open field and closed loops.This model also results in a power-law tail in the energetic particles (protons and helium in the test cases), which may account for the transient energetic particle events seen, especially close to the Sun.Another mechanism has been invoked in an attempt to account for events with enhanced 3 He/ 4 He ratios, involving resonant acceleration by waves generated by outward-streaming beams of energetic electrons (Temerin & Roth 1992;Roth & Temerin 1997).This mechanism, though referred to in more recent literature (e.g., Reames 2015), has not gained broad acceptance because of perceived deficiencies in explaining the associated heavy-ion enrichment in the 3 He-rich events.However, Mitchell et al. (2020) argue that those deficiencies can be overcome, and suggest that current instabilities combined with parallel electric fields can account for the generation of seed populations with enrichment of particular ranges of charge-to-mass ratios depending on the wave-power frequency distribution.As the event observed by PSP on 2022 February 27 stands out both for its dispersion and for its high 3 He content, it may be that both features are a consequence of reconnection combined with wave-particle heating in the source.Only a small subset of jets/jetlets suggested by Raouafi et al. (2023a) as the source of the solar wind is thought likely to produce transient SEPs as seen in this event.

SEP Event of 2022 February 27
The angular velocity of the orbital motion of the PSP (Raouafi et al. 2023b) spacecraft exceeds the angular rotation velocity of the surface of the Sun at perihelion, while farther out in PSP's orbit the Sun's surface angular velocity exceeds that of PSP.For about 2 days during both the inbound and outbound legs of the orbit, PSP's angular velocity roughly matches that of the Sun's surface.During these limited segments of the orbit, PSP effectively corotates over a fixed longitude in the rotating Carrington longitude system.Therefore, during these unique orbital segments, the source for the solar wind measured at PSP remains approximately fixed at the longitude directly beneath the spacecraft.This means that variations in the solar wind (to this approximation) represent time variations in source-region behavior, rather than spatial variations, which often dominate as PSP quickly changes source locations during other segments of its orbit.
On 2022 February 27, PSP was outbound at a distance of about 0.11 au from the Sun, and was just entering a 2 day interval of matching the angular rotation rate of the solar surface (remaining fixed in Carrington longitude to within ∼±2.5°over the ensuing interval).This geometry is illustrated in Figures 1(a) and (b). Figure 1(a) shows the PSP orbit in inertial coordinates, with an inset in Carrington coordinates.That Carrington system is expanded in Figure 1(b), showing that the PSP orbital motion in (rotating) Carrington coordinates is primarily radial.
In Figure 2, we present an overview of the event.At about 0830UT on February 27, energetic particles were first detected in the ISeIS (McComas et al. 2016) EPI-Hi sensor.Although the EPI-Hi protons remained at background, a distinct onset was seen in 4 He at energies as high at 20 MeV.Subsequently, the ISeIS EPI-Lo sensor measured 4 He as high as 2 MeV total energy, with lower energies following in a typical timedispersed sequence characteristic of sudden-onset SEP events.EPI-Lo also measured the dispersive event in protons (weakly) and 3 He (see Figures 4 and 5; the 3 He fraction relative to 4 He was the highest of the PSP mission for this otherwise rather minor event).The ISeIS data types that most clearly show the event time history are the EPI-Hi 4 He (inset, plotted in energy/ nucleon) and the EPI-Lo ion energy/nucleon derived from measuring their velocities (both in panel (c)).The remaining panels include PSP solar-wind energy flux in panel (d) (SWEAP; Kasper et al. 2016); magnetic field and RTN components in panel (a), |B|, elevation, and azimuth in panel (e) (FIELDS; Bale et al. 2016), and radio emissions in panel (b) (Pulupa et al. 2017).An unusual aspect of this event was the observation of dispersed ions all the way down to just above solar-wind energies, as discussed in Alnussirat et al. (2023).The magenta curve overlaid on the particle data traces the loci for the expected time of arrival at PSP as a function of energy/ nucleon for an impulsive injection in the corona occurring at the time of the first in the series of Type III events seen in the radio wave spectrogram.The time dispersion in the energetic particles accelerated in the impulsive event stands out clearly.Also clearly seen are dropouts in the energetic particle intensities, the most prominent being from 1030UT to 1150UT.
In order to observe dispersion of this sort, the spacecraft must be magnetically well connected with the source location of the energetic particles, and the transport of the particles must be relatively scatter free.We now discuss the geometry of the magnetic field, and its relationship to the particle intensities.
In Figure 3, we have reproduced the plots in Figure 2, and have added notation to call out changes in the magnetic geometry.We have reduced the large-scale fluctuations in the field angles to three basic categories: radial, oblique, and transverse.These orientations are encountered repeatedly over the course of the event, and changes in the energetic particle intensity appear to align with changes in the field orientation.We interpret this to mean that PSP repeatedly enters and leaves flux tubes with three different field orientations, presumably connected to three different coronal foot points.This sort of transition among three flux tubes multiple times is probably unique to this segment of the PSP trajectory, when the spacecraft is matching the rotation rate of the Sun's surface and so remaining connected with approximately the same longitude over an interval of 2 days.Clearly, in the cases of the transverse and oblique orientations, this geometry cannot persist all the way back to the coronal.These flux tubes must "bend" during their transport outward, since it is precisely the nonradial orientations that are populated with particles injected in this event.The solar-wind strahl electrons (not shown) exhibit unidirectional anti-sunward streaming throughout this interval, indicative of a topologically open field rooted in the low corona, consistent with the continuous connection with the acceleration site required to explain the dispersive transport signature.
In addition to this apparent confinement of the event to specific flux tubes, the composition of the accelerated ions changes over the course of the event, indicating that it should not be considered simply an instantaneous energization of coronal plasma, but rather a process that begins with an impulsive onset, and evolves over a few hours such that the ongoing acceleration mechanism favors helium isotopes early in the energetic particle process and heavy ions (especially iron) later in the acceleration process.

Composition
In Figure 4, we present time-energy spectrograms for several major ion species.Panel (a) shows particles for which only time of flight (TOF) is measured; no solid-state detector (SSD) measurement of energy (E) is included.For these data, we  know only the velocity of the ion, not its energy.So we calculate energy/nucleon, with the added assumption that the energy losses in the TOF system foils correspond to calibrated losses for protons.Generally, this is a good approximation, although in some cases helium can dominate this product.The advantage of this product is that the EPI-Lo efficiency is higher than for the TOF × E products, resulting in better statistics, and the low-energy threshold for this measurement is well below any of the TOF × E species products, since the ion need not be measured in the SSD.These data show clearly the onset for the EPI-Lo energy range, as well as the near-dropout in intensity for the radial field interval from about 1020UT to 1150UT (the intensity in this interval does not fall to zero, however).
Following the dropout, the intensity increases again.The dispersion can no longer be followed in EPI-Lo because the energy of the leading edge is below the EPI-Lo energy range.It does continue in the SWEAP SPAN-I instrument (Livi et al. 2022), as noted earlier.The ion composition, however, is quite different at this time.Whereas 3 He was relatively abundant during the event onset (at higher energies/nucleon), it is no longer being energized sufficiently to be measured by EPI-Lo during this phase.Note that the ions observed at this time, being much farther in time from the leading edge of the event than any of the intensities were before the gap, represent a later phase in the event profile at the acceleration site.This indicates that, whereas4 He and 3 He were efficiently accelerated at the peak of the energy/nucleon values, the later phase of the energization in the corona also enriched heavier species, and especially iron.This could reflect a wave-particle mechanism for the energization (Temerin & Roth 1992;Roth & Temerin 1997;Mitchell et al. 2020) with more wave power at frequencies that favor the helium isotopes at the peak of the energization, and either a shift or a broadening to frequencies that favor iron energization later.
Figure 5 provides a more quantitative look at the evolution of the composition over the course of the event.Panel (a) is a color scatter plot of the events within the apertures that received most of the counts during the event, and also have the highest mass resolution among the EPI-Lo apertures (i.e., those apertures with longer TOF path lengths).Two intervals are called out using colored bars along the timeline, and accumulated histograms for those intervals appear in the panels below, (b) for the early portion and peak energization portion of the event, and (c) for the reintensification after the dropout.The latter interval, as discussed above at the end of Section 2, is relatively iron-rich, whereas the earlier interval, from onset through the peak of the particle energization, is relatively iron-poor.This variation in composition over the course of the event favors a wave-particle mechanism over a shock-acceleration mechanism, with higher wave power in frequencies that include resonance with lighter species early in the event, shifting later in the event to wave power in frequencies that include resonance with heavier species, especially iron.Shock acceleration would be expected to accelerate all species proportionate to their abundances in the source region, inconsistent with the observed time evolution in composition.Furthermore, no shock or abrupt increase in solar-wind speed was observed at any point throughout this event.Rather, the solar-wind speed remained between 270 and 235 km s −1 throughout the interval over which the solar wind associated with the dispersion was observed.

Dispersion
In Figure 6, we return to the topic of dispersion that was discussed in Alnussirat et al. (2023).That paper showed conclusively that the dispersion observed at higher energies extended all the way down to about 2 keV in solar-wind protons, just above the core solar-wind plasma.Here, we have extended the dispersion curve, focusing for now only on the traditional, white curve delineating the first-arriving energetic particles associated with what is assumed to be a sudden, impulsive injection in the low corona.Following that curve forward in time, it naturally extends down into the energy band typically occupied by the solar-wind core plasma.
In interpreting the data, it is useful to keep in mind that (in the approximation of radial, noncollisional propagation) this dispersion line should delineate the first-arrival time for ions even in the energy band encompassing the solar-wind core plasma.In that spirit, and following the structure in the core plasma energy flux as a function of time, the energy flux just above the dispersion curve is markedly lower (by over an order of magnitude) than the energy flux of the solar-wind plasma at and below that line.These lower-energy ions, again under the assumption of noncollisional, radial propagation, had to have been emitted from the low corona prior to the onset of the event.This structure (low-intensity flux just above the line, high-intensity flux at the line, and even higher, by nearly an order of magnitude, below the line) persists at least until about 1030UT on day 59, and affects the distribution down to about 200 eV at that point.This would be consistent with the particularly low density associated with the solar wind emitted at the injection site during, and even lower density for several hours following, the impulsive event.So while the eventassociated intensities (those on and just above the white dispersion line) at energies above about 1 keV are elevated, at lower energies they are depressed relative to the previously emitted solar wind (below the line).Whether this reflects a process that has moved the energy density from lower energies into higher energies in that locale, or the event simply takes place in a low-density region of the corona, is beyond the scope of this paper.Beyond that time (1030UT), the magnetic field orientation changes, suggesting that PSP has moved on to a different flux tube, which may no longer connect with the source region for this event.This would explain the abrupt ending of the low-intensity structure lying just above the dispersion curve at 1030UT.
Exploring this event a bit further, we have included two dispersion curves on the plot, the white curve based on the high-energy ion onset, and the other (purple) displaced assuming release from a low coronal source 3 hr 20 minutes after that white curve.These curves are meant to guide the eye to a couple of features suggesting that, whereas the initial acceleration of the ions to high energies must have had a very abrupt onset, the source region may have continued to energize ions for more than 3 hr following that onset.
The "late" (purple) dispersion curve, displaced later in time to bound the drop in intensity of the energetic ions greater than 100 keV, includes between it and the onset (white) dispersion curve the later enhancement in energetic hydrogen, and especially in energetic iron (see Figure 4).Those particles, extending for ∼3 hr 20 minutes after the first-arriving energetic particles, indicate that the region not only continued to produce (or at least release) energetic ions for hours after the onset, but also changed with regard to the composition of the energetic particles accelerated, favoring iron far more than at onset.Whether this is a different process, or simply the evolution of the same process into a parameter range that favors iron acceleration (we favor the latter), the data indicate that the region remained an active source for at least 3+ hr past the energetic ion onset.
We have also undertaken a survey of the PSP data, restricted to the ∼2 days, inbound and outbound segments of the PSP trajectory during which the Carrington longitude of the subspacecraft point on the Sun remains nearly constant, to look for additional examples of dispersive events for which we can have some degree of confidence that PSP remains connected with the event source location in the corona throughout the dispersion.This search yielded no additional events.There are, to be sure, additional events for which dispersion can be observed in the energy range just above the solar-wind core energy band, but these all occur during intervals when the PSP Carrington longitude changes by typically tens of degrees over the course of the dispersion, which raises questions about whether the spacecraft remains connected with relatively similar coronal source conditions throughout the event.The trajectory restrictions we require are an attempt to minimize such ambiguity.

Discussion and Summary
At ISeIS energies (∼50 keV to 10 MeV) the initial injection was helium-rich, with distinct 3 He enrichment.Protons appeared to be considerably less abundant than helium at the high energies.Later in the event, iron also was seen to be enriched relative to its usual abundances.The progression from helium-rich to iron-rich over the evolution of the source-region acceleration suggests a likely wave-particle mechanism for the energization of the energetic ions.The event profile began with an impulsive phase, with sudden generation of suprathermal electrons (not directly observed at PSP, but inferred as they are necessary for generating the Type III event) as well as energetic helium 3 and helium 4 (but very little hydrogen, and only weak heavier ions).After about another hour, conditions further evolved such that the ion energization favored heavy ions, especially iron.This time development is not a propagation effect, but rather evolution in the source-region behavior.Because of the selectivity for particular ion masses at different stages of the source-region development, we favor a waveparticle mechanism for the ion energization, with helium being efficiently accelerated/heated during the initial impulsive phase, and iron favored later, presumably due to a shift or expansion in the peak wave power to include frequencies resonant with iron.The other clue is the Type III event, which indicates the presence of energetic electron-driven instabilities in the corona.Field-aligned streaming electrons can also drive broadband electrostatic waves, which in turn can effectively heat the plasma ions, especially for charge-to-mass ratios for which the wave power matches the gyrofrequencies (e.g., Mitchell et al. 2020 and references therein).
The dispersed ions were observed all the way down to solarwind energies (down to ∼250 eV protons), an observation never seen at 1 au, observable on PSP presumably owing to its small radial distance.
The continued dispersion over more than a day is also consistent with solar-wind plasma spectral features, suggestive of a flux tube containing (transient) relatively low density at solar-wind plasma energies.Taken together, these observations support a picture whereby, under favorable circumstances, particles may be transported effectively scatter free out to distances as large as ∼35 R s , even down to solar-wind energies.
This would require little coupling of the plasma particles with each other during transport, and raises the question as to what the meaning of concepts like velocity, temperature, and a solar-wind frame mean under such conditions.
The observed composition enhancements and time histories make this event a good candidate example of a class of events proposed (e.g., Schwadron et al. 2020) to generate seed populations that feed diffusive shock (or compression) accelerated SEP events.In this instance, there is no evidence for a shock, or even a compression, but the acceleration/heating that did lead to these compositional features may be a common precursor in larger SEP events associated with CMEs and shocks.Particle data are in the spacecraft frame of reference, as are the dispersion curves.Overlaid on the solar-wind plasma data is a green curve, the energy of a proton traveling at the measured solar-wind velocity, in the spacecraft frame.That characteristic energy remains below the dispersion curves until after 1030UT on day 59, consistent with the notion that the bulk of the solar-wind plasma observed over the interval of interest was emitted prior to the impulsive onset on day 58, and that from about 0200UT until 1030UT on day 59, the solar-wind plasma in the energy range above the dispersion curves (emitted after the impulsive event) is depleted relative to the pre-event solar wind.

Figure 1 .
Figure 1.Parker Solar Probe orbit 11 perihelion.The left panel presents the trajectory in inertial coordinates (white), with a yellow arrow that covers a 2 day segment of the outbound orbit.The inset in that panel, expanded in the right panel, shows the perihelion segment of the trajectory in Carrington coordinates, where the longitude rotates according to Carrington longitude.This system tracks the subsolar Carrington longitude of PSP.The 2 day yellow segment is nearly radial, indicating the subsolar Carrington longitude remains nearly constant (to within ∼±2.5°) over that interval.

Figure 2 .
Figure 2. (a) PSP FIELDS magnetic field, RTN coordinates; (b) PSP FIELDS plasma wave data, highlighting Type III events; (c) PSP ISeIS EPI-Lo energy/nucleon spectrogram (with EPI-Hi, small box inset) showing 4 He data in E/n; (d) PSP SWEAP/SPAN solar-wind proton energy density spectrogram; (e) PSP FIELDS magnetic field azimuth and elevation.The magenta curve drawn through the particle data traces the particle first-arrival time as a function of energy assuming an impulsive injection in the low corona at the time of the first Type III event.

Figure 3 .
Figure 3. Same as Figure2, with shading to highlight three specific ranges of magnetic field orientation (interpreted as repeated encounters with three flux tubes, each connecting with unique locations in the corona).These flux tubes are named for the orientation of the magnetic field relative to the radial direction.

Figure 4 .
Figure 4. EPI-Lo species spectrograms.Top panel is a TOF-only measurement, converted to energy assuming all ions are protons.For other species, data can be considered energy/nucleon.The other panels are labeled according to their species determined by TOF × E. These show clearly a relatively robust population of 3 He early in the event, while heavy ions such as oxygen and especially iron are enriched later in the event.The gap between about 1015UT and 1145UT is attributed to magnetic connection geometry (see Figure 3 and discussion).

Figure 5 .
Figure 5. (a) Mass scatter plot as a function of time; (b) and (c) summed histograms by epoch.The differences ( 3 He-rich at onset, heavy-ion-rich later) may reflect the time evolution of the acceleration process(es).

Figure 6 .
Figure6.Dispersion curves (first-arriving energetic particles, white; late, purple).This is similar to panels (c), (d), and (e) of Figure2, but for a full 2 day time interval.Particle data are in the spacecraft frame of reference, as are the dispersion curves.Overlaid on the solar-wind plasma data is a green curve, the energy of a proton traveling at the measured solar-wind velocity, in the spacecraft frame.That characteristic energy remains below the dispersion curves until after 1030UT on day 59, consistent with the notion that the bulk of the solar-wind plasma observed over the interval of interest was emitted prior to the impulsive onset on day 58, and that from about 0200UT until 1030UT on day 59, the solar-wind plasma in the energy range above the dispersion curves (emitted after the impulsive event) is depleted relative to the pre-event solar wind.