Solar Orbiter’s CoSEE-Cat: A large statistical study of the acceleration and transport of energetic electrons in the corona and inner heliosphere

(Solar Orbiter Nugget #72 by Alexander Warmuth¹ and the CoSEE-Cat Working Group²)

Introduction: Solar energetic electrons (SEEs)

Our Sun is the most energetic particle accelerator in the solar system, and therefore the acceleration of solar energetic particles (SEPs) and their propagation through interplanetary space are key topics in heliophysics [1]. SEP events are generally classified into two groups: gradual events are associated with large X-ray flares and coronal mass ejection (CME)-driven shocks, and can be measured over wide angular ranges, while impulsive events are electron-rich, associated with small X-ray flares, type III radio bursts, and are highly enriched in 3He and Fe/O [2, 3]. Note that while these terms originated from the time evolution of the associated flares, they are now commonly used to indicate the elemental composition of SEPs.
Compared to energetic ions, solar energetic electrons (SEEs) which are observed in-situ offer a unique advantage since electrons can be also observed remotely using X-ray and radio observations. Hard X-ray (HXR) nonthermal bremsstrahlung constrains electrons that have been accelerated in the corona and precipitate into the chromosphere. Conversely, electrons escaping into interplanetary space can be traced as type III radio bursts generated by plasma emission. Despite this observational advantage, acceleration and transport of SEEs has remained insufficiently understood. For instance, while a flare-related origin of impulsive SEEs is commonly assumed, there are discrepancies which question whether downward- and upward-moving electrons (detected remotely in HXRs and in-situ, respectively) are really accelerated by the same mechanism. SEEs generally show a solar release time (SRT; the time at which the particles are injected into interplanetary space) that is delayed with respect to the peak of the associated HXR flares and the onsets of type III bursts [4]. In addition, while the spectra of the flare electrons and the SEEs do show a correlation, the relation is inconsistent with either thick-target or thin-target X-ray emission that is usually adopted for inferring the accelerated electron spectrum from the observed photon spectrum [5, 6]. It is unclear whether these discrepancies imply that upward- and downward propagating electrons are completely different populations, or if this could be caused by modification by shock waves or various transport effects.

Observations: Solar Orbiter

ESA’s Solar Orbiter provides two unique advantages for the study of SEEs. Firstly, its comprehensive suite of in-situ and remote-sensing instruments covers all relevant types of observables. Having all these assets on a single platform is a great advantage and results in a high duty cycle. Secondly, Solar Orbiter samples SEEs over a wide distance range. This allows us for the first time to probe potential particle transport effects. To achieve this, a larger sample of events is required, and eight of Solar Orbiter’s ten instrument teams have joined forces and formed a joint working group dedicated to characterizing all observed SEE events as well as their potential solar source. 

Figure 1: Overview of an impulsive SEE event. From the top, the plot shows the electron fluxes as recorded by EPD, a dynamic electron spectrum (plotted as a function of v/c) from EPD showing a clear velocity dispersion, a dynamic radio spectrum from RPW showing a type III radioburst, and STIX X-ray count rates. Overplotted dashed vertical lines show the inferred solar release times of the SEEs using TSA and VDA (red and turquoise, respectively), HXR peak time (blue), type III onset time (black), as well as the electron onset time at Solar Orbiter (black dotted line).

Specifically, we use the Energetic Particle Detector (EPD) to detect SEE events and measure their key parameters, including timing, fluxes, anisotropies, and the composition of the associated energetic ions. EPD data is also used to derive the solar release time of the electrons, using both time-shift analysis (TSA) and velocity dispersion analysis (VDA) methods. Based on timing, we then associate solar events. The Spectrometer/Telescope for Imaging X-rays (STIX) provides information on the associated flares (timing, fluxes, source locations), which is supplemented by EUV data provided by the Extreme Ultraviolet Imager (EUI). Potentially associated CMEs are characterized by the coronagraph Metis and the Solar Orbiter Heliospheric Imager (SoloHI). Radio type III bursts observed with the Radio and Plasma Waves (RPW) trace electron beams from the Sun into interplanetary providing a link between the Sun and the spacecraft that helps us to identify the associated solar sources. Since particle propagation is strongly influenced by the characteristics of the interplanetary medium, we use local constraints on the solar wind plasma and magnetic fields given by the Solar Wind Analyzer (SWA) and the Solar Orbiter Magnetometer (MAG), respectively. These observations were supplemented by the Magnetic Connectivity Tool (http://connect-tool.irap.omp.eu/) which is based on data-driven simulations yielding magnetic field lines connecting Solar Orbiter and the Sun. 

Catalogue: CoSEE-Cat

From November 2020 until the end of 2022, EPD detected 303 SEE events suitable for further analysis. Using all assets described above, we determined 99 parameters for each event. We have compiled this information in the Comprehensive Solar Energetic Electron event Catalogue (CoSEE-Cat). An online version of this catalogue (https://coseecat.aip.de/ ) provides easy access to this data set. The catalogue is implemented as an SQL database, which means that advanced filtering options can be applied. The results as well as the full catalogue can be downloaded as CSV files. In addition, we provide quicklook plots and movies for all events.

First statistical results

About three quarters of the events show impulsive composition, which is consistent with previous results. Impulsive SEE events generally have shorter rise times and are more anisotropic than gradual ones. Nearly all SEE events are associated with X-ray and EUV flares or eruptive phenomena.

Figure 2: Left: Locations of STIX flares (rescaled to 1 au) associated to SEEs. Colors indicate the composition of the events, sizes of the circles show the equivalent GOES flare class. Note the clustering of impulsive events in the western hemisphere. Right: Histogram of the difference in heliolongitude between the STIX flare positions and the predicted locations of the connecting magnetic field lines. Note the small differences for impulsive events (blue) and the flat distribution for gradual events (red).

Concerning the X-ray flare source positions, we found that impulsive events were predominantly located on the western hemisphere, while gradual events did not show such clustering. This is also clearly shown when we compare the flare locations to the predicted footpoints of magnetic connectivity. While the difference in heliolongitude is strongly peaked around zero for impulsive events, the distribution for gradual events is flat. This implies that impulsive events have to originate from spatially small regions that are magnetically well connected to the spacecraft, such as flares or small-scale eruptions. Conversely, the source position seems to play no role for gradual SEE events, which means that the injection region has to be spatially extended. This suggests acceleration at a CME-driven shock or injection from a CME-related erupting flux rope. 

Figure 3: Left: Histogram of the difference between the solar release time (SRT) of SEEs (determined with velocity dispersion analysis) and the peak of the nonthermal HXR emission observed with STIX. The temporal agreement is better for impulsive events as compared to gradual ones. Right: SRT difference as a function of distance from the Sun for impulsive events that occur under quiet solar wind conditions. Note that the apparent SRT delays increase with distance, which is consistent with propagation effects.

With regards to timing, we found that the difference between the solar release time of the electrons and the peak of the hard X-ray flare or the 
onset of the type III radio burst is generally significantly smaller for impulsive events as opposed to gradual ones. For example, the mean delay 
between the VDA release time and the HXR peak was 3.7 min and 11 min for impulsive and gradual events, respectively.

With Solar Orbiter, we can now go one step further and study the solar release time delays as a function of distance from the Sun. While gradual events show no correlation with distance, a low correlation is found for impulsive events, i.e., events detected farther from the Sun show longer delays. This is qualitatively consistent with recent results obtained from Parker Solar Probe [7]. When just considering impulsive events that were observed in ambient solar wind conditions, the correlation was significantly enhanced. We conclude that transport effects that accumulate with distance are at least partially responsible for the observed SRT delays. In addition, we find that timing is strongly dependent on the conditions of the interplanetary magnetic field, which is of course what is to be expected.

Conclusion and Outlook

Using the unique data provided by Solar Orbiter, we have found strong evidence that impulsive SEE events are associated with acceleration at solar flares, while gradual events are more consistent with a CME-related origin. While this has already been suggested by previous studies, this distinction is very clearly seen in our sample, most probably due to the very homogeneous dataset provided by state-of-the-art instruments on a single platform, combined with the application of refined analysis methods and data-driven modelling of the magnetic connectivity. In addition, we have found evidence for transport effects influencing the travel time of SEEs in interplanetary space.
However, this has just been the first step. CoSEE-Cat is a living catalogue and will be updated as the mission progresses. An important next step will be to perform a spectral analysis of the in-situ electrons as measured by EPD and the electrons precipitating on the Sun constrained with hard X-rays (STIX) and to investigate whether their relation changes with distance from the Sun. Finally, CoSEE-Cat will provide an ideal starting point for new multi-spacecraft studies that fully exploit the fleet of missions currently active in the inner heliosphere.

This work has been published in Warmuth et al. 2025, A&A, 701, A20 https://doi.org/10.1051/0004-6361/202554830 
The link to the online catalogue is: https://coseecat.aip.de/ .
 

Affiliations/CoSEE-Cat Team

(1) Leibniz-Institute for Astrophysics Potsdam (AIP)

(2) Fredric Schuller, Raúl Gómez-Herrero, Ignacio Cernuda, Fernando Carcaboso, Glenn Mason, Nina Dresing, Daniel Pacheco, Laura Rodríguez-García, Manon Jarry, Matthieu Kretzschmar, Krzysztof Barczynski, Daria Shukhobodskaia, 
Luciano Rodriguez, Song Tan, David Paipa-Leon , Nicole Vilmer, Alexis  Rouillard, Clementina Sasso, Silvio Giordano, Giuliana Russano, Catio Grimani, Federico Landini, Cecilia Mac Cormack, Jake Mitchell,, Annamaria Fedeli, Laura Vuorinen, Aleksi Yli-Laurila, David Lario, Hamish Reid, Frederic Effenberger, Sophie Musset, Antonio Vecchio, Oleksiy Dudnik, Andrea 
Battaglia, Hannah Collier, Radoslav Bucik, George Ho, Vratislav Krupar,  Camille Lorfing, Xu Zigong, Arun Awasthi, Karl-Ludwig Klein, Säm Krucker, Milan Maksimovic, Javier Rodríguez-Pacheco, Marco Romoli, Robert Wimmer-Schweingruber
 

References

[1] Reames, D. V. 1999, Space Sci. Rev., 90, 413 https://doi.org/10.1023/A:1005105831781
[2] Desai, M., & Giacalone, J. 2016, Liv. Rev. Sol. Phys., 13, 3 https://doi.org/10.1007/s41116-016-0002-5
[3] Reames, D. V. 2018, Space Sci. Rev., 214, 61 https://doi.org/10.1007/s11214-018-0495-4
[4] Haggerty, D. K., & Roelof, E. C. 2002, ApJ, 579, 841 https://doi.org/10.1086/342870
[5] Krucker, S., Kontar, E. P., Christe, S., & Lin, R. P. 2007, ApJ, 663, L109 https://doi.org/10.1086/519373
[6] Dresing, N., Warmuth, A., Effenberger, F., et al. 2021, A&A, 654, A92 https://doi.org/10.1051/0004-6361/202141365
[7] http://connect-tool.irap.omp.eu/
[8] Mitchell, J. G., Christian, E. R., de Nolfo, G. A., et al. 2025, ApJ, 980, 96 https://doi.org/10.3847/1538-4357/adaa7c