Solar Origins of a Strong Stealth CME Detected by Solar Orbiter Jennifer O’Kane1, Lucie M

Astronomy & Astrophysics manuscript no. output ©ESO 2021 April 10, 2021

Letter to the Editor

Solar origins of a strong stealth CME detected by Solar Orbiter Jennifer O’Kane1, Lucie M. Green1, Emma E. Davies2, Christian Möstl3, Jürgen Hinterreiter3, 4, Johan L. Freiherr von Forstner5, Andreas J. Weiss3, 4, David M. Long1, and Tanja Amerstorfer3

1 Mullard Space Science Laboratory, UCL, Holmbury St Mary, Dorking, Surrey, RH5 6NT, UK e-mail: [email protected] 2 Department of Physics, Imperial College London, London, UK 3 Space Research Institute, Austrian Academy of Sciences, Graz, Austria 4 Institute of Physics, University of Graz, Graz, Austria 5 Institut für Experimentelle und Angewandte Physik, Christian-Albrechts-Universität zu Kiel, Kiel, Germany

April 10, 2021

ABSTRACT

Aims. We aim to locate the origin of a stealth coronal mass ejection (CME) detected in situ by the MAG instrument on board Solar Orbiter and make connections between the CME observed at the Sun and the interplanetary CME (ICME) measured in situ. Methods. Remote sensing data were analysed using advanced image processing techniques to identify the source region of the stealth CME, and the global magnetic field at the time of the eruption was examined using potential field source surface (PFSS) models. The observations of the stealth CME at the Sun were compared with the magnetic field measured by the Solar Orbiter spacecraft, and plasma properties were measured by the Wind spacecraft. Results. The source of the CME is found to be a quiet Sun cavity in the northern hemisphere. We find that the stealth CME has a strong magnetic field in situ, despite originating from a quiet Sun region with an extremely weak magnetic field. Conclusions. The interaction of the ICME with its surrounding environment is the likely cause of a higher magnetic field strength measured in situ. Stealth CMEs require multi-wavelength and multi-viewpoint observations in order to confidently locate the source region; however, their elusive signatures still pose many problems for space weather forecasting. The findings have implications for Solar Orbiter observing sequences with instruments such as EUI that are designed to capture stealth CMEs. Key words. Sun: coronal mass ejections (CMEs) – Sun: corona – Sun: magnetic fields – Sun: heliosphere

1. Introduction image processing techniques can be used to identify observational signatures, although these can be very faint and hard to de- Coronal mass ejections (CMEs) are large eruptions of the Sun’s tect (Alzate and Morgan 2017; O’Kane et al. 2019). Studies car- plasma and magnetic field into the heliosphere. If they are Earth- ried out so far indicate that stealth CMEs may originate at rela- directed, these events have the potential to drive space weather tively high heights ≈ 1.3 R in the corona (Robbrecht et al. 2009; effects. A subset of interplanetary CMEs (ICMEs), known as O’Kane et al. 2021), implying that the eruptive structure has a magnetic clouds, exhibit enhanced magnetic field strength, a ro- relatively low magnetic field strength and low plasma density, tation in at least one of the magnetic field components, and low leading to a correspondingly low energy release, lack of strong plasma density, temperature, and plasma β when detected in situ radiative emissions, and lower propagation speed (<500 km/s, (Burlaga et al. 1981b). The helical field configuration of a mag- D’Huys et al. 2014). Despite these characteristics, stealth CMEs netic cloud is known as a flux rope. The geoeffectiveness of a still have the potential to be geoeffective and therefore present magnetic cloud (or indeed any CME) is highly dependent on the a challenge for space weather forecasting (Kilpua et al. 2014; strength and orientation of the magnetic field. In particular, a Nitta and Mulligan 2017). Carrying out in situ and remote sens- strong southward Bz for a period of time longer than a few hours ing connection studies can help one to understand the origins, typically produces a strong geomagnetic storm. In order to im- evolution, and impact of CMEs in the heliosphere. prove space weather prediction capabilities, it is then necessary Here, we study the solar origins of the first CME detected to study the CME source region and use this knowledge to fore- in situ by Solar Orbiter (Müller et al. 2020). The encounter took cast the configuration and properties of the magnetic field in the place on 19 April 2020 at a heliocentric distance of 0.809 AU. interplanetary medium. The source region may or may not have a The details of the CME’s structure and propagation in the inner flux rope configuration itself (see Green et al. 2018; Patsourakos heliosphere are reported in Davies et al. (2021) and information et al. 2020, for reviews). However, if a flux rope is present, it on the corresponding Forbush decrease can be found in Frei- can be identified using a range of remote sensing data and the herr von Forstner et al. (2021). The interplanetary CME had a knowledge used to forecast its geoeffectiveness (see Palmerio clear flux rope structure and its arrival at Earth produced a Dst et al. 2017; Palmerio et al. 2018, and references therein). minimum of -60 nT on 20 April 2020 and a category G1 geo- Stealth CMEs lack the classic low coronal signatures typi- magnetic storm. At the Sun, the event was determined to be a cally associated with CMEs, such as solar flares, extreme ultra- streamer blowout with no clear signatures in the lower corona. violet (EUV) dimmings, and filament eruptions (see overview This places the CME in the stealth category and, here, we present by Howard and Harrison 2013). However, it has been shown that the identification and analysis of the CME source region that was

Article number, page 1 of 7

Article published by EDP Sciences, to be cited as https://doi.org/10.1051/0004-6361/202140622 A&A proofs: manuscript no. output

20 made possible using advanced image processing techniques. The Solar Orbiter Btot BR paper is organised as follows: Sect. 2 describes the in situ and re- 10 mote sensing data as well as the methods used within this study; BT B Sect. 3 describes the analysis of the stealth CME connecting the 0 N in situ and the solar source perspectives; and ﬁnally Sect. 4 dis- B [nT, RTN] cusses the results and conclusions. 10 20 2020-04-18 2020-04-19 2020-04-20 2020-04-21 2020-04-22 2020-04-23

2. Data and methods 20 Wind Btot BR 2.1. Data 10 BT B For the in situ analysis, magnetic field data were obtained from 0 N the magnetometer (MAG; Horbury et al. 2020) on board Solar Orbiter (Müller et al. 2020), which provides in situ observations B [nT, RTN] 10 of the CME at approximately 0.8 AU. The magnetic field inves- 20 tigation (MFI; Lepping et al. 1995) and solar wind experiment 2020-04-18 2020-04-19 2020-04-20 2020-04-21 2020-04-22 2020-04-23 (SWE; Ogilvie et al. 1995) of the Wind spacecraft provide mag- 600 120 Wind Vp netic field and solar wind plasma observations at 1 AU. 500 100 ] For the remote sensing observations, the white light CME ] 3

1 400 80 s

was analysed using data from the Large Angle and Spectrometric m c m 300 60 [ k

[ p

COronagraph (LASCO; Brueckner et al. 1995) on board the SO- V lar and Heliospheric Observatory (SOHO; Domingo et al. 1995) 200 40 N and the COR1 and 2 coronagraphs (part of the Sun Earth Con- 100 Np 20 nection Coronal and Heliospheric Investigation instrument suite, 0 0 SECCHI, Howard et al. 2008) on board the Solar TErrestrial 2020-04-18 2020-04-19 2020-04-20 2020-04-21 2020-04-22 2020-04-23 Relations Observatory Ahead (STEREO-A, Kaiser et al. 2008) Fig. 1. In situ observations of the CME. In each panel, the shock front spacecraft. EUV data were taken from the Atmospheric Imag- is delineated by the solid vertical line, and the magnetic flux rope ing Assembly (AIA, Lemen et al. 2011) on board the Solar Dy- boundaries are shown by the vertical dashed lines. Top panel: Solar namics Observatory (SDO, Pesnell et al. 2011) spacecraft, and Orbiter magnetic field data in radial-tangential-normal (RTN) coordi- the Extreme Ultraviolet Imager (EUVI; Howard et al. 2008) on nates, where the total magnetic field magnitude is shown in black and board STEREO-A. Line of sight magnetic field data were taken the magnetic field components, RTN, are in red, green, and blue, respectively. Middle panel: Wind magnetic field data in RTN coordinates, from the Helioseismic and Magnetic Imager (HMI) data (Scher- displayed similarly to that of Solar Orbiter. Bottom panel: The proton rer et al. 2012) on board SDO. speed (black) and density (red) measured by Wind.

2.2. Methods by Wind (0.996 AU) at 01:34 UT on 20 April 2020, arriving with Coronagraph data were processed using the Normalising-Radial- −1 Graded Filter (NRGF, Morgan et al. 2006). EUV data were a mean proton speed of 346 km s . processed using both the Multi-scale Gaussian Normalisation The in situ data reveal a clear magnetic flux rope structure, (MGN) technique (Morgan and Druckmüller 2014) and the constrained by the vertical dashed lines in Fig. 1, where the lead- NRGF technique. Stack plots were created using COR1 and ing edge times of the magnetic flux rope are 08:59 UT on 19 COR2 data by taking a radial slice from Sun centre. The slice April 2020 at Solar Orbiter and 07:56 UT on 20 April 2020 at taken at 90◦ is presented in this paper, as this most clearly shows Wind (defined by Davies et al. 2021). The CME can be classi- the CME curve. The plane-of-sky kinematics were obtained us- fied as a magnetic cloud, meeting criteria detailed by Burlaga ing the Savitzky-Golay bootstrapping technique (Byrne et al. et al. (1981a). The configuration of the magnetic flux rope can 2013), and the 3D shape and orientation of the CME in the COR2 be classified as south-east-north (Bothmer and Schwenn 1998; and LASCO/C2 fields of view were modelled using the gradu- Mulligan et al. 1998) and, therefore, has left-handed chirality ated cylindrical shell (GCS) model (Thernisien et al. 2006; Frei- and the flux rope axis is of low inclination to the ecliptic plane. herr von Forstner 2021). Longitude and latitude are defined in The maximum magnetic field strength reached within the Carrington coordinates. magnetic flux rope was 20.1 nT at Solar Orbiter and 16.3 nT at Wind. This is a relatively high field strength for an ICME at solar minimum and/or from quiet Sun regions. The proton den- 3. The stealth CME sity of the magnetic flux rope measured at Wind was relatively −3 3.1. In situ observations low throughout, with a mean density of 11.6 cm ; however, this includes the rise in density towards the trailing edge (discounting In situ magnetic field signatures observed by both Solar Or- this rise, the mean density in the flux rope is 8.3 cm−3). The rise biter and Wind as well as the solar wind plasma data observed in density towards the trailing edge is likely due to a compression by Wind whilst the spacecraft were close to radial alignment, of the magnetic flux rope caused by the high speed stream that but separated in longitude by less than 5◦, are shown in Fig. 1. follows, which we observe at Wind to have a mean proton speed The shock front (delineated by the solid vertical line in Fig. 1) of ∼450 km s−1. The high speed stream may also be the rea- driven by the CME was first observed in situ by Solar Orbiter at son we observe a stronger magnetic field strength than expected 05:06 UT on 19 April 2020 whilst the spacecraft was located at a within the CME, where the mean magnetic field strength of the heliocentric distance of 0.809 AU. The shock was then observed flux rope (18.4 nT at Solar Orbiter and 13.6 nT at Wind) is much

Article number, page 2 of 7 Jennifer O’Kane et al.: Solar origins of a strong stealth CME detected by Solar Orbiter

EUVI−A 171Å 13−April−2020 12:09 UT COR1-A Stack Plot COR1−A 14−April−2020 03:15 UT 2000 3.5

1000 3.0 ) • O

0 2.5 Height (R Y (arcsec)

−1000 2.0

−2000 1.5 0 1000 2000 3000 00:00 12:00 00:00 12:00 00:00 12:00 X (arcsec) Start Time (12-Apr-20 17:47:34) Fig. 2. Left: Combined EUVI-A 171Å and COR1-A, illustrating the connectivity between the high altitude structures observed in EUVI-A and the CME observed in COR1-A. The EUVI-A image was taken at 22:09 UT 13 April 2020, when the cavity was observed, whilst the COR1-A image was taken at 03:15 UT 14 April 2020, when the CME eruption was underway. The blue line represents the 90◦ slice taken for the stack plot. Right: A stack plot inverse image of COR1-A, taken at an angle of 90◦, showing the CME curve. The blue crosses indicate the points selected to track the CME curve. stronger than those calculated for ICMEs at similar heliocentric das and Romashets 2017), with the parameters derived from the distances (e.g. Janvier et al. 2019). 3DCORE fit, The magnetic field profiles observed at both spacecraft show a slight decline in magnetic field strength as the flux rope passes πB1AU Φax = ln(1 + τ1AU D1AU /2), (1) over the spacecraft. Such asymmetric profiles are consistent with τ1AU the model of an expanding flux rope (Nieves-Chinchilla et al. +0.16 21 2018; Janvier et al. 2019). However, the proton speed profile of gives Φax = 0.33−0.13 × 10 Mx, which is consistent with the the CME observed at Wind shows only a small decline through- numerical result; although, this should in general be seen only out the magnetic flux rope. Asymmetric magnetic field profiles as a rough approximation because the flux rope cross section with a relatively small expansion may correspond to ICMEs in 3DCORE is not cylindrical, but elliptical. The axial flux we with distorted structures (Nieves-Chinchilla et al. 2018), as ob- determined for this event is thus at the lower end of reported served in images taken by the STEREO-A Heliospheric Imager axial fluxes in ICME flux ropes (cf. Qiu et al. 2007). (HI) of this event (Davies et al. 2021). The mean propagation speed of the leading edge between Solar Orbiter and Wind was 3.2. Stealth CME source region 341 km s−1, assuming a constant transit velocity would suggest a CME onset time ∼5 days prior to its arrival at Wind (i.e. on 15 As seen in Fig.2, the stealth CME is observed in COR1-A as April 2020). a faint enhancement in density, and it becomes clear in COR2- We have applied a semi-empirical 3D flux rope model to the A by 03:54 UT 15 April 2020, following the brightening and Solar Orbiter data (3DCORE, Möstl et al. 2018; Weiss et al. swelling of a streamer. The CME exhibits a bright underside and 2021) in order to derive global parameters of the flux rope within a clear concave-up structure, which is suggestive of a flux rope this ICME. Detailed results are given in Davies et al. (2021) and configuration (Vourlidas et al. 2013) . In COR1-A, the underside Freiherr von Forstner et al. (2021), and work on a paper on si- of the CME can be clearly tracked from 10:16 UT 14 April 2020 multaneous multi-point fitting of the Solar Orbiter, Wind, and (Fig.2), using the 90◦ slice, and it has an average velocity of −1 BepiColombo flux rope magnetic field data is underway (Weiss 18 km s . Using this speed and a height of 2.5 R at 22:30 UT et al. 2021b, in prep.). Here we present additional information 14 April 2020, we estimate a more accurate CME onset time of on the magnetic flux derived from the 3DCORE fit on the Solar ∼19:30 UT 13 April 2020. In COR2-A, the CME has an average Orbiter MAG data only. This is of particular interest because for speed of 53 km s−1. The increase in CME speed between that stealth CMEs, large-scale magnetic reconnection during flares measured from COR2-A and that measured in situ at Wind is can be ruled out as a formation mechanism of the flux rope (e.g. likely due to the high speed stream following the CME and the Qiu et al. 2007; Möstl et al. 2009), and this yields parameters of background solar wind. a flux rope that was pre-existing in the solar corona. Primary observations of the Sun during this time period show Davies et al. (2021) report the fit results with the 3DCORE no active regions, and the photospheric magnetic field exhibited technique applied to the Solar Orbiter data, propagated to 1 AU: only weak small-scale field fragments (see Fig. 3). EUV and H- an axial field strength of B0 = 14.3 ± 0.9 nT, a flux rope diame- alpha data showed that no filaments or filament channels were ter of D1AU = 0.114 ± 0.022 AU, and a twist of τ = −3.7 ± 0.6 present either, making the source region of the CME very chal- field line turns over the full torus (τ < 0 implies left-handed lenging to identify. Using the images from COR2-A and C2 at chirality), or τ1AU = −0.6 ± 0.1 turns per AU. We have numer- 05:39 UT and 05:36 UT 15 April 2020, respectively, the CME ically integrated the flux rope cross section to find an axial flux was fitted with the GCS model (see Fig.4). The GCS fit found 21 ◦ ◦ of Φax = 0.30 ± 0.06 × 10 Mx. The application of the formula an approximate source region at 143±7 longitude and 3±3 lat- for a uniform twist cylindrical flux rope (e.g. equation 16 in Van- itude (see Table 2 in Freiherr von Forstner et al. 2021), and it is

Article number, page 3 of 7 A&A proofs: manuscript no. output

HMI 13−April−2020 20:00 UT AIA 193Å 13−April−2020 20:00 UT 800 800

600 600

400 400

Y (arcsec) 200 Y (arcsec) 200

0 0

−200 −200 −600 −400 −200 −0 200 400 600 −600 −400 −200 −0 200 400 600 X (arcsec) X (arcsec)

Fig. 3. Left: HMI image at 20:00 UT 13 April 2020, showing the magnetic ﬁeld just before the cavity began to rise. The blue ’X’ marks the location of the GCS approximated source region, whilst the red box marks the approximate location of the cavity that was observed at the solar limb. Right: Corresponding AIA 193Å image, with the GCS and cavity regions being highlighted as before. In both ﬁgures, the limb of EUVI-A is over-plotted as the white line.

COR2-A LASCO-C2 therefore deduce that this is the cavity responsible for the CME 2020-04-15 05:39 UT 2020-04-15 05:36 UT that is later detected in situ. The deflection of the cavity was measured by applying the GCS model to the structure in 171 Å EUVI-A. The cavity is observed to deflect southwards by ∼19◦ within the EUVI FOV, and a further ∼12◦ into the FOV of COR1. It is noted that due to applying the GCS model to one spacecraft instead of two, errors are larger. However, this is smaller for latitude such that the deflection can be measured with reasonable confidence. The cavity is an observational signature of a flux rope present in the low corona as the eruption takes place. We tracked the region back to the eastern limb of SDO to look for evidence of the flux rope present prior to the eruption. Three cavities are observed on the limb in AIA 171 Å between 6 and 10 April 2020. The first cavity at 10◦ latitude erupted off limb on 8 April 2020, and therefore was ruled out as being the source. The second cavity at 40◦ latitude, which is highlighted by the red-dashed circle in Fig. 6, closely matches the erupting cavity observed in EUVI- A (at 35.7◦ latitude) and with its approximate longitude. The ◦ Fig. 4. Remote sensing observations of the CME. STEREO-A COR2 third cavity at 30 latitude does not match in longitudinal ex- (direct images) and SOHO LASCO C2 (running difference images). In tent with the cavity observed at the limb of EUIV-A. Thus we the bottom panels, the GCS fitting according to the parameters obtained conclude the second cavity observed is the structure that later by Freiherr von Forstner et al. (2021) is indicated by the blue mesh. erupts as the stealth CME. Using these two observations, we ob- tain an approximate source region from the SDO perspective as 139◦±5◦ longitude and 38◦±2◦ latitude, indicated by the red box indicated by the blue cross in Fig. 3. This puts the approximate in Fig. 3. The difference between the source region and the es- source region on disk from SDO, and on limb from STEREO-A. timated region obtained from the GCS is due to the southwards deflection of the cavity as it erupts. Although both the MGN and difference imaging techniques were applied to each AIA passband, they did not reveal any clear We further investigated the magnetic topology of the source eruption signatures on disk. However, the NRGF and MGN pro- region and its surrounding environment with a potential field cessed 195 Å and 171 Å EUVI-A data revealed multiple high- source surface (PFSS) model, using the PFSS solver in Python altitude dynamic structures off the solar limb, including a bright (pfsspy, Yeates 2018; Stansby 2019). Fig. 7 shows the PFSS looped region, a large bright structure, and a cavity structure model as viewed from both the AIA (left) and STEREO-A (Fig. 5). The large bright structure is observed from 08:50 UT (right) perspectives, with the modelled magnetic field lines 13 April 2020, and then it slowly rises into COR1-A, coinciding showing the positive (red) and negative (blue) polarities. The with the brightening and swelling of the streamer. The concave- CME originates in the northern hemisphere, near the central up cavity structure is observed from 22:09 UT 13 April 2020 at meridian (west limb) from the AIA (STEREO-A) perspective. 35.7◦ latitude, and it is deflected southwards as it erupts from Open field lines are deflected southwards, towards the equator, the Sun, coinciding with initiation of the CME in COR1-A. We similar to the trajectory of the cavity observed in EUVI-A. Given

Article number, page 4 of 7 Jennifer O’Kane et al.: Solar origins of a strong stealth CME detected by Solar Orbiter

EUVI−A 171Å 12−April−2020 02:09 UT EUVI−A 171Å 13−April−2020 14:09 UT EUVI−A 171Å 13−April−2020 22:09 UT 800 800 800

600 600 600

400 400 400

200 200 200 Y (arcsec) Y (arcsec) Y (arcsec)

0 0 0

−200 −200 −200

400 600 800 1000 1200 1400 400 600 800 1000 1200 1400 400 600 800 1000 1200 1400 X (arcsec) X (arcsec) X (arcsec)

Fig. 5. MGN processed EUVI-A 171Å images, illustrating the structures contributing to the stealth CME. Left panel: The red arrows point out a bright loop structure that is above ﬁlament material. Middle panel: The large bright high altitude structure, illustrated by the red arrows. Right panel: The cavity structure, highlighted by the red-dashed circle.

cantly higher than that found in a comparison study of stealth CMEs at solar minimum by Kilpua et al. (2014), who show an AIA 171Å 7−April−2020 19:00 UT average Bmax of 9.7 nT across a sample of ten stealth CMEs, 1000 and the in situ stealth CME study of Nieves-Chinchilla et al. (2013). This maximum field strength and the south-east-north 800 configuration of the flux rope may have been the deciding fac- tors in this event producing a category G1 geomagnetic storm. 600 The slow speed of stealth CMEs means that we do not typically expect them to drive shocks. In this event, a forward shock is detected in situ, similar to the case of the stealth CME stud- 400 ied in Nieves-Chinchilla et al. (2013). The shock is likely due

Y (arcsec) to the slower solar wind conditions at 1 AU ahead of the ICME 200 (∼300 km s−1). Also, the slight expansion in the unperturbed section of the ICME means that a combination of a relatively slow 0 solar wind speed ahead of the CME and the CME expansion may drive the shock as proposed by Lugaz et al. (2017). This study reinforces the need for multi-wavelength, multi- −200 −1200 −1000 −800 −600 −400 −200 −0 view point observations of the solar corona, combined with im- X (arcsec) age processing techniques, to further our ability to detect and analyse stealth CME source regions. These observations indicate that for Solar Orbiter EUI to successfully observe stealth Fig. 6. MGN-processed AIA 171Å image at 19:00 UT 7 April 2020, CME sources, observing sequences should include regular syn- showing the cavity structure observed on the limb (red-dashed circle) optic images from the 174 Å waveband of the Full Sun Imager ∼6 days before the eruption. (FSI) telescope, which will have a field-of-view of 4 R at peri- helion. the absence of active regions and the solar minimum phase at this Acknowledgements. JO thanks the STFC for support via funding given in her time, the global magnetic field is close to that of a dipole field. PHD studentship. LMG is grateful to the Royal Society, which supported this research through the Royal Society University Research Fellowship scheme. DML is grateful to the Science Technology and Facilities Council for the award of an Ernest Rutherford Fellowship (ST/R003246/1). JvF thanks the German Space 4. Discussion and conclusions Agency (Deutsches Zentrum für Luft- und Raumfahrt e.V., DLR) for their support of his work on the Solar Orbiter EPD team under grant 50OT2002. CM, Here, we present a study of the first stealth CME observed in situ AJW, JH and TA thank the Austrian Science Fund (FWF): P31521-N27, P31659- by Solar Orbiter, with accompanying Wind in situ observations N27, P31265-N27. ED is supported by funding from the Science and Technol- and STEREO-A and SDO remote sensing data. The combined ogy Facilities Council (STFC) studentship ST/N504336/1. SDO is a mission of use of STEREO-A and SDO data revealed that the origin of the NASA’s Living With a Star Program. STEREO is the third mission in NASA’s Solar Terrestrial Probes program. SOHO is a mission of international coopera- CME was a quiet Sun cavity within small-scale magnetic field tion between ESA and NASA. The authors thank the SDO, STEREO, and SOHO and that there was no clear polarity inversion line. The cavity teams for making their data publicly accessible.The Solar Orbiter magnetometer identified in EUVI-A and AIA 171 Å passbands extended up was funded by the UK Space Agency (grant ST/T001062/1). Its data are avail- able in the Solar Orbiter Archive at http://soar.esac.esa.int/soar/. to 1.3 R (from Sun centre) in the corona. Such dark elliptical cavities in the quiet Sun indicate the presence of a magnetic flux rope and have been found to have a median height of 1.2 R (Gibson 2015). References Despite the quiet Sun origin of the stealth CME, the mag- Alzate, N. and Morgan, H. (2017), ‘Identification of low coronal sources of netic cloud still had a very high field strength in the interplan- “stealth” coronal mass ejections using new image processing techniques’, The etary medium (13.6 nT at Wind). The ∼1 AU value is signifi- Astrophysical Journal 840(2), 103.

Article number, page 5 of 7 A&A proofs: manuscript no. output

Fig. 7. PFSS model of Carrington rotation 2229, as observed from the AIA perspective (left) and STEREO-A perspective (right). Positive (negative) polarity is coloured in red (blue), on 13 April 2020. The approximate source region of the stealth CME is indicated by the white arrows.

Bothmer, V. and Schwenn, R. (1998), ‘The structure and origin of magnetic Goldstein, M. L., Hoeksema, T., Lockwood, M., Magnes, W., Maksimovic, clouds in the solar wind’, Ann. Geophys. 16, 1–24. M., Marsch, E., Matthaeus, W. H., Murphy, N., Nakariakov, V. M., Owen, Brueckner, G., Howard, R., Koomen, M., Korendyke, C., Michels, D., Moses, C. J., Owens, M., Rodriguez-Pacheco, J., Richter, I., Riley, P., Russell, C. T., J., Socker, D., Dere, K., Lamy, P., Llebaria, A. et al. (1995), The large an- Schwartz, S., Vainio, R., Velli, M., Vennerstrom, S., Walsh, R., Wimmer- gle spectroscopic coronagraph (lasco), in ‘The SOHO Mission’, Springer, Schweingruber, R. F., Zank, G., Müller, D., Zouganelis, I. and Walsh, A. P. pp. 357–402. (2020), ‘The Solar Orbiter magnetometer’, A&A 642, A9. Burlaga, L., Sittler, E., Mariani, F. and Schwenn, a. R. (1981a), ‘Magnetic loop Howard, R. A., Moses, J. D., Vourlidas, A., Newmark, J., Socker, D., Plunkett, behind an interplanetary shock: Voyager, helios, and imp 8 observations’, S., Korendyke, C., Cook, J., Hurley, A., Davila, J. et al. (2008), ‘Sun earth Journal of Geophysical Research: Space Physics 86(A8), 6673–6684. connection coronal and heliospheric investigation (secchi)’, Space Science Burlaga, L., Sittler, E., Mariani, F. and Schwenn, R. (1981b), ‘Magnetic loop Reviews 136(1-4), 67. behind an interplanetary shock: Voyager, Helios, and IMP 8 observations’, Howard, T. A. and Harrison, R. A. (2013), ‘Stealth coronal mass ejections: A J. Geophys. Res. 86(A8), 6673–6684. perspective’, Solar Physics 285(1-2), 269–280. Byrne, J., Long, D., Gallagher, P., Bloomfield, D., Maloney, S., McAteer, R., Janvier, M., Winslow, R. M., Good, S., Bonhomme, E., Démoulin, P., Dasso, S., Morgan, H. and Habbal, S. R. (2013), ‘Improved methods for determining Möstl, C., Lugaz, N., Amerstorfer, T., Soubrié, E. and Boakes, P. D. (2019), the kinematics of coronal mass ejections and coronal waves’, Astronomy & ‘Generic Magnetic Field Intensity Profiles of Interplanetary Coronal Mass Astrophysics 557, A96. Ejections at Mercury, Venus, and Earth From Superposed Epoch Analyses’, Davies, E. E., Möstl, C., Owens, M. J., Weiss, A., Amerstorfer, T., Hinterreiter, J., Journal of Geophysical Research (Space Physics) 124(2), 812–836. Bauer, M., Bailey, R. et al. (2021), ‘In situ multi-spacecraft and remote imag- Kaiser, M. L., Kucera, T., Davila, J., Cyr, O. S., Guhathakurta, M. and Chris- ing observations of the first cme detected by solar orbiter and bepicolombo’, tian, E. (2008), ‘The stereo mission: An introduction’, Space Science Reviews A&A . 136(1-4), 5–16. URL: https://doi.org/10.1051/0004-6361/202040113 Kilpua, E., Mierla, M., Zhukov, A., Rodriguez, L., Vourlidas, A. and Wood, B. D’Huys, E., Seaton, D., Poedts, S. and Berghmans, D. (2014), ‘Observational (2014), ‘Solar sources of interplanetary coronal mass ejections during the so- characteristics of coronal mass ejections without low-coronal signatures’, The lar cycle 23/24 minimum’, Solar Physics 289(10), 3773–3797. Astrophysical Journal 795(1), 49. Lemen, J. R., Akin, D. J., Boerner, P. F., Chou, C., Drake, J. F., Duncan, D. W., Domingo, V., Fleck, B. and Poland, A. I. (1995), ‘The soho mission: an Edwards, C. G., Friedlaender, F. M., Heyman, G. F., Hurlburt, N. E. et al. overview’, Solar Physics 162(1-2), 1–37. (2011), The atmospheric imaging assembly (aia) on the solar dynamics ob- Freiherr von Forstner, J. L. (2021), ‘johan12345/gcs_python: Version 0.2.0’. servatory (sdo), in ‘The Solar Dynamics Observatory’, Springer, pp. 17–40. URL: https://doi.org/10.5281/zenodo.4443203 Lepping, R. P., Acuna,˜ M. H., Burlaga, L. F., Farrell, W. M., Slavin, J. A., Schat- Freiherr von Forstner, J. L., Dumbovic,´ M., Möstl, C., Guo, J., Papaioannou, ten, K. H., Mariani, F., Ness, N. F., Neubauer, F. M., Whang, Y. C., Byrnes, A., Elftmann, R., Xu, Z., Terasa, J. C., Kollhoff, A., Wimmer-Schweingruber, J. B., Kennon, R. S., Panetta, P. V., Scheifele, J. and Worley, E. M. (1995), R. F., Rodríguez-Pacheco, J., Weiss, A. J., Hinterreiter, J., Amerstorfer, T., ‘The Wind Magnetic Field Investigation’, Space Sci. Rev. 71, 207–229. Bauer, M., Belov, A. V. et al. (2021), ‘Radial evolution of the april 2020 Lugaz, N., Farrugia, C. J., Winslow, R. M., Small, C. R., Manion, T. and Savani, stealth coronal mass ejection between 0.8 and 1 au. comparison of forbush N. P. (2017), ‘Importance of CME Radial Expansion on the Ability of Slow decreases at solar orbiter and near the earth’, A&A . CMEs to Drive Shocks’, ApJ 848(2), 75. URL: https://doi.org/10.1051/0004-6361/202039848 Morgan, H. and Druckmüller, M. (2014), ‘Multi-scale gaussian normalization Gibson, S. (2015), Coronal Cavities: Observations and Implications for the for solar image processing’, Solar physics 289(8), 2945–2955. Magnetic Environment of Prominences, Vol. 415, p. 323. Morgan, H., Habbal, S. R. and Woo, R. (2006), ‘The depiction of coronal struc- Green, L., Torok, T., Vršnak, B., Manchester, I., Veronig, A. et al. (2018), ‘The ture in white-light images’, Solar Physics 236(2), 263–272. origin, early evolution and predictability of solar eruptions’, arXiv preprint Möstl, C., Amerstorfer, T., Palmerio, E., Isavnin, A., Farrugia, C. J., Lowder, C., arXiv:1801.04608 . Winslow, R. M., Donnerer, J. M., Kilpua, E. K. J. and Boakes, P. D. (2018), Horbury, T. S., O’Brien, H., Carrasco Blazquez, I., Bendyk, M., Brown, P., Hud- ‘Forward Modeling of Coronal Mass Ejection Flux Ropes in the Inner Helio- son, R., Evans, V., Oddy, T. M., Carr, C. M., Beek, T. J., Cupido, E., Bhat- sphere with 3DCORE’, Space Weather 16(3), 216–229. tacharya, S., Dominguez, J. A., Matthews, L., Myklebust, V. R., Whiteside, Möstl, C., Farrugia, C. J., Miklenic, C., Temmer, M., Galvin, A. B., Luhmann, B., Bale, S. D., Baumjohann, W., Burgess, D., Carbone, V., Cargill, P., East- J. G., Kilpua, E. K. J., Leitner, M., Nieves-Chinchilla, T., Veronig, A. and wood, J., Erdös, G., Fletcher, L., Forsyth, R., Giacalone, J., Glassmeier, K. H., Biernat, H. K. (2009), ‘Multispacecraft recovery of a magnetic cloud and its

Article number, page 6 of 7 Jennifer O’Kane et al.: Solar origins of a strong stealth CME detected by Solar Orbiter

origin from magnetic reconnection on the Sun’, Journal of Geophysical Re- search (Space Physics) 114(A4), A04102. Müller, D., Cyr, O. S., Zouganelis, I., Gilbert, H. R., Marsden, R., Nieves- Chinchilla, T., Antonucci, E., Auchère, F., Berghmans, D., Horbury, T. et al. (2020), ‘The solar orbiter mission-science overview’, Astronomy & Astro- physics 642, A1. Mulligan, T., Russell, C. T. and Luhmann, J. G. (1998), ‘Solar cycle evolution of the structure of magnetic clouds in the inner heliosphere’, Geophys. Res. Lett. 25, 2959–2962. Nieves-Chinchilla, T., Vourlidas, A., Raymond, J. C., Linton, M. G., Al-haddad, N., Savani, N. P., Szabo, A. and Hidalgo, M. A. (2018), ‘Understanding the Internal Magnetic Field Configurations of ICMEs Using More than 20 Years of Wind Observations’, Sol. Phys. 293(2), 25. Nieves-Chinchilla, T., Vourlidas, A., Stenborg, G., Savani, N. P., Koval, A., Sz- abo, A. and Jian, L. K. (2013), ‘Inner Heliospheric Evolution of a “Stealth” CME Derived from Multi-view Imaging and Multipoint in Situ observations. I. Propagation to 1 AU’, ApJ 779(1), 55. Nitta, N. V. and Mulligan, T. (2017), ‘Earth-affecting coronal mass ejections without obvious low coronal signatures’, Solar Physics 292(9), 125. Ogilvie, K. W., Chornay, D. J., Fritzenreiter, R. J., Hunsaker, F., Keller, J., Lo- bell, J., Miller, G., Scudder, J. D., Sittler, Jr., E. C., Torbert, R. B., Bodet, D., Needell, G., Lazarus, A. J., Steinberg, J. T., Tappan, J. H., Mavretic, A. and Gergin, E. (1995), ‘SWE, A Comprehensive Plasma Instrument for the Wind Spacecraft’, Space Sci. Rev. 71, 55–77. O’Kane, J., Cormack, C. M., Mandrini, C. H., Démoulin, P., Green, L. M., Long, D. M. and Valori, G. (2021), ‘The magnetic environment of a stealth coronal mass ejection’, The Astrophysical Journal 908(1), 89. URL: https://doi.org/10.3847/1538-4357/abd2bf O’Kane, J., Green, L., Long, D. M. and Reid, H. (2019), ‘Stealth coronal mass ejections from active regions’, The Astrophysical Journal 882(2), 85. Palmerio, E., Kilpua, E. K. J., Möstl, C., Bothmer, V., James, A. W., Green, L. M., Isavnin, A., Davies, J. A. and Harrison, R. A. (2018), ‘Coronal Mag- netic Structure of Earthbound CMEs and In Situ Comparison’, Space Weather 16(5), 442–460. Palmerio, E., Kilpua, E. K., James, A. W., Green, L. M., Pomoell, J., Isavnin, A. and Valori, G. (2017), ‘Determining the intrinsic cme flux rope type using remote-sensing solar disk observations’, Solar Physics 292(2), 39. Patsourakos, S., Vourlidas, A., Török, T., Kliem, B., Antiochos, S. K., Archon- tis, V., Aulanier, G., Cheng, X., Chintzoglou, G., Georgoulis, M. K., Green, L. M., Leake, J. E., Moore, R., Nindos, A., Syntelis, P., Yardley, S. L., Yurchyshyn, V. and Zhang, J. (2020), ‘Decoding the Pre-Eruptive Magnetic Field Configurations of Coronal Mass Ejections’, Space Sci. Rev. 216(8), 131. Pesnell, W. D., Thompson, B. J. and Chamberlin, P. (2011), The solar dynamics observatory (sdo), in ‘The Solar Dynamics Observatory’, Springer, pp. 3–15. Qiu, J., Hu, Q., Howard, T. A. and Yurchyshyn, V. B. (2007), ‘On the Mag- netic Flux Budget in Low-Corona Magnetic Reconnection and Interplanetary Coronal Mass Ejections’, ApJ 659(1), 758–772. Robbrecht, E., Patsourakos, S. and Vourlidas, A. (2009), ‘No trace left behind: Stereo observation of a coronal mass ejection without low coronal signatures’, The Astrophysical Journal 701(1), 283. Scherrer, P. H., Schou, J., Bush, R., Kosovichev, A., Bogart, R., Hoeksema, J., Liu, Y., Duvall, T., Zhao, J., Schrijver, C. et al. (2012), ‘The helioseismic and magnetic imager (hmi) investigation for the solar dynamics observatory (sdo)’, Solar Physics 275(1-2), 207–227. Stansby, D. (2019), ‘dstansby/pfsspy: pfsspy 0.1.2’. URL: https://doi.org/10.5281/zenodo.2566462 Thernisien, A., Howard, R. and Vourlidas, A. (2006), ‘Modeling of flux rope coronal mass ejections’, The Astrophysical Journal 652(1), 763. Vandas, M. and Romashets, E. (2017), ‘Magnetic cloud fit by uniform-twist toroidal flux ropes’, A&A 608, A118. Vourlidas, A., Lynch, B. J., Howard, R. A. and Li, Y. (2013), ‘How Many CMEs Have Flux Ropes? Deciphering the Signatures of Shocks, Flux Ropes, and Prominences in Coronagraph Observations of CMEs’, Sol. Phys. 284, 179– 201. Weiss, A. J., Möstl, C., Amerstorfer, T., Bailey, R. L., Reiss, M. A., Hinterre- iter, J., Amerstorfer, U. A. and Bauer, M. (2021), ‘Analysis of Coronal Mass Ejection Flux Rope Signatures Using 3DCORE and Approximate Bayesian Computation’, ApJS 252(1), 9. Yeates, A. (2018), ‘First release of pfss code.’. URL: https://doi.org/10.5281/zenodo.1472183

Article number, page 7 of 7