Visibility and Origin of Compact Interplanetary Radio Type IV Bursts
SSolar PhysicsDOI: 10.1007/ ••••• - ••• - ••• - •••• - • Visibility and Origin of Compact InterplanetaryRadio Type IV Bursts
Nasrin Talebpour Sheshvan · Silja Pohjolainen c (cid:13) Springer ••••
Abstract
We have analysed radio type IV bursts in the interplanetary (IP)space at decameter–hectometer (DH) wavelengths, to find out their source originand a reason for the observed directivity. We used radio dynamic spectra from theinstruments on three different spacecraft, STEREO-A,
Wind , and STEREO-B,that were located approximately 90 degrees apart from each other in 2011-2012,and thus gave a 360 degree view to the Sun. The radio data was compared towhite-light and extreme ultraviolet (EUV) observations of flares, EUV waves,and coronal mass ejections (CMEs) in five solar events. We find that the reasonfor observing compact and intense DH type IV burst emission from only onespacecraft at a time is due to the absorption of emission to one direction andthat the emission is blocked by the solar disk and dense corona to the otherdirection. The geometry also makes it possible to observe metric type IV burstsin the low corona from a direction where the higher-located DH type IV emissionis not detectable. In the absorbed direction we found streamers present, andthese were estimated to be the locations of type II bursts, caused by shocks atthe CME flanks. The high-density plasma was therefore most probably formedby shock–streamer interaction. In some cases the type II-emitting region was alsocapable of stopping later-accelerated electron beams, visible as type III burststhat ended near the type II burst lanes.
Keywords:
Coronal Mass Ejections, Initiation and Propagation, Radio Bursts,Meter-Wavelengths and Longer (m, dkm, hm, km), type II, type IV (cid:66)
N. Talebpour Sheshvan natash@utu.fi
S. Pohjolainen silpoh@utu.fi Department of Physics and Astronomy, University of Turku, Turku, Finland Tuorla Observatory, Department of Physics and Astronomy, University of Turku,Turku, Finland
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 1 a r X i v : . [ a s t r o - ph . S R ] O c t alebpour Sheshvan and Pohjolainen
1. Introduction
Coronal mass ejections (CMEs) are a large scale phenomena of plasma andmagnetic field explosion from the Sun into the interplanetary (IP) medium.Solar events such as flares and CMEs accelerate particles, with different mecha-nisms, and cause them to propagate from the solar corona into the IP space. Forreviews see, e.g. , Pick and Vilmer (2008) and Nindos et al. (2008). The particlepaths, their directions and locations, can be studied using radio bursts at a widewavelength range, from metric to kilometric wavelengths.Solar radio bursts in the plasma regime (below ≈ f p and/or its harmonics. The particles in type III bursts are typicallyflare-accelerated electrons travelling along coronal magnetic field lines (Krupar et al. , 2018) while type II bursts have been associated with CME-induced shockwaves (Cane, 1984; Leblanc et al. , 2001) .The more rare radio type IV bursts have been closely associated with ex-panding and/or rising magnetized plasma structures. The emission mechanismcan therefore be both synchrotron emission by trapped high-energy electronsgyrating in the magnetic field and plasma emission as the magnetic cloud lifts offfrom the Sun. Typically metric type IV bursts are divided into two categories,stationary and moving. The characteristics of stationary type IV bursts havebeen defined by Kundu (1965), to be located near a flare, to be strongly circu-larly polarized, smooth and broad-band, and located low in the corona near thecorresponding plasma level. The stationary type IV bursts show no systematicmovement and the bursts may also occur without any associated type II emission.It has been suggested that the source of emission is particles accelerated in aflare, trapped in post-flare loops and arcades. Moving type IV bursts on the otherhand show definite frequency drifts toward the lower frequencies, and they aremostly thought to represent particles trapped in rising CME structures. Kundu(1965) also stated that type IV bursts show directivity in their emission. Sometype IV bursts extend in the dynamic spectra to decameter–hectometer (DH)wavelengths and they can be recorded by space-borne instruments at frequenciesbelow ≈
20 MHz (Earth ionosphere limit). The characteristics of IP moving typeIV bursts have recently been listed and studied by Hillaris, Bouratzis, and Nindos(2016), and their catalogue contains bursts recorded in 1998–2012 by the radioexperiment on-board
Wind satellite. Directivity in IP type IV burst emissionwas later noticed by Gopalswamy et al. (2016).Closely associated with high-speed CMEs are EUV-waves, observed in the ex-treme ultra-violet wavelength range in the low corona. These bright features andthe dimming behind them typically start from the flare location and propagateglobally over the solar disk. However, no strong correlation was found betweenthe global EUV disturbance speeds and flare intensities, or CME magnitudes
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 2 isibility of Compact IP type IV bursts (Nitta et al. , 2013). Recently, Kwon and Vourlidas (2017) found that halo-CME widths are in agreement with the widths of EUV waves in the low corona,suggesting a common origin for these structures. In some cases the heights ofEUV waves have been found to be compatible with the heights of radio typeII burst sources (Cunha-Silva, Fernandes, and Selhorst, 2015). Since radio typeIV bursts can be observed over a wide frequency range, the result of the studiesmade over the past decades has been to define type IV bursts as any broad-band, long-duration, continuum-like emission occurring after a flare. Moreover,imaging has only been available for decimetric–metric bursts that occur in thelow corona, but not for DH type IV bursts in the IP space.In 2007 two more spacecrafts with radio spectrograph were launched to orbitthe Sun, and with
Wind , STEREO A and STEREO B instruments a 3D-viewof the Sun was made possible, if not real imaging. In this study we use the radiodata from these three spacecraft and the questions we try to answer are, whereare IP type IV bursts located, is there directivity in their emission, and are theya continuation of metric type IV burst emission.
2. Data Analysis
The time period to find DH type IV bursts was set to 2011–2012 in order tohave a full 3D-view of the Sun, as then the spacecraft were orbiting the Sunwith an ≈ ◦ angular separation, and the ’backside’ of the Sun could be imagedin EUV and white-light with the instruments on-board STEREO A and B, seeFigure 1. The Earth-side imaging were provided by Solar Dynamic Observatory (SDO) and
Solar and Heliospheric Observatory (SOHO).
Figure 1.
Locations of STEREO A (red circle) and B (blue circle) spacecraft relative to theEarth (green circle) on 4 June 2011 (left) and 5 March 2012 (right). These are the first andlast event dates in our sample, and the plots show how the orbital distance was increasingin respect to Earth and the
Wind satellite located near L1. During this time period the Suncould be viewed in 3D. ( https://stereo-ssc.nascom.nasa.gov ) The DH type IV bursts were selected from the catalogue of solar type IIand type IV bursts detected by the WAVES instruments on
Wind (Bougeret et al. , 1995) and STEREO A and B (Bougeret et al. , 2008). The cataloguehas been prepared by Michael L. Kaiser. Altogether eleven DH type IV events
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 3 alebpour Sheshvan and Pohjolainen were found from this time period but only five of them were found to show acompact, intense, and long-duration DH type IV burst, with radio and imagingobservations from all the three viewing angles. These five events and their sourceorigins are listed in Tables 1 and 2. Figure 1 shows the spacecraft positionsrelative to the Sun, on the first and last event date.The highest observing frequencies for
Wind /WAVES and STEREO/WAVESare 14 MHz and 16 MHz, respectively. In the case of plasma emission thesefrequencies correspond to densities of 2.4 × cm − and 3.2 × cm − (plasmaemission frequency f p ≈ √ n e , where f p is in Hz and n e in cm − ).Three of the DH type IV bursts were preceded by metric type IV bursts thatwere observed with ground-based instruments from Earth. The other two DHtype IVs had their source origin on the backside of the Sun and therefore theradio emission at lower atmospheric heights could not be observed (Table 2).Two of the five flares originated from the backside of the Sun, with no X-rayobservations, and therefore their GOES X-ray flare class is not known. The otherthree were GOES X-class flares. Table 1.
DH type IV bursts observed from three different viewing angles.Date Type IV Type IVyyyy-mm-dd appearance STEREO-B Wind STEREO-AUT (source loc.) (source loc.) (source loc.)2011-06-04 07:12–09:00 – – strong (W50)2011-06-04 22:15–23:30 – – strong (W70)2011-09-22 11:10–12:40 strong (W10) – –2012-01-27 18:30–20:10 – (N-limb) – strong (E50)2012-03-05 04:15–06:00 faint (W80) strong (E50) – Time of IP type IV start and end at 16–14 MHz (instrument limit). First part, a second more narrow-band enhancement follows at 23:45–01:20 UT.
Table 2.
Flares and metric type IV bursts prior to the DH type IV bursts.Date Flare Flare GOES Flare Metricyyyy-mm-dd start maximum class location type IVUT UT (Earth view)2011-06-04 06:20 N20W140 not available2011-06-04 21:45 N20W160 not available2011-09-22 10:29 11:01 X1.4 N13E78 10:40–13:402012-01-27 17:37 18:37 X1.7 N27W71 18:15–18:352012-03-05 02:30 04:09 X1.1 N17E52 04:20–05:20
All the type IV events were associated with EUV waves and they are listed inTable 3. All the observed strong, long-duration type IV bursts were associatedwith global EUV waves that were observed to cross the solar disk in this viewing
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 4 isibility of Compact IP type IV bursts angle. The EUV waves that were visible only near the limb, or which covered onlya small portion of the solar disk, had either no corresponding type IV emissionor the type IV was faint in intensity and ended earlier.The associated CMEs and type II shocks are listed in Table 4. All the CMEshad very high speed, in the range of 1600–2900 km s − near the time of typeIV burst appearance (2nd order fit to the observed plane-of-sky heights, fromthe LASCO CME Catalogue). The CME on 27 January 2012 was the only oneaccelerating while all the other CMEs were decelerating. Table 3.
EUV waves associated with the DH type IV eventsDate EUV EUV wave (radio)yyyy-mm-dd wave STEREO-B/EUVI SDO/AIA STEREO-A/EUVIkm s − (WAVES) ( Wind /WAVES) (WAVES)2011-06-04 – (no IV) – (no IV) global (strong IV)2011-06-04 at limb (no IV) – (no IV) global (strong IV)2011-09-22 595 global (strong IV) at limb (no IV) – (no IV)2012-01-27 635 N-quarter (no IV) W-quarter (no IV) global (strong IV)2012-03-05 915 half-disk (faint IV) global (strong IV) – (no IV) Large-scale Coronal Propagating Fronts observed by SDO/AIA (Nitta et al. , 2013), note thatthe velocity value comes from Earth-view observations only.
Table 4.
CMEs and type II bursts associated with the DH type IV events.Date CME CME Type IV Type IV Type II Type II CMEyyyy-mm-dd height speed lowest lowest freq. height height R (cid:12) km s − UT MHz MHz R (cid:12) R (cid:12) CME height at the time of the type IV appearance, interpolated or extrapolated when notobserved at that time. The type IV appearance at 16–14 MHz (instrument limit) correspondsto a height of 2–3 R (cid:12) , depending on the atmospheric density model. CME speed from a 2nd order fit to the observed plan-of-sky heights near the time of type IVburst appearance. Type II burst frequency at the time of lowest type IV frequency. Type II burst height from the density model of Vr˘snak, Magdaleni˘c, and Zlobec (2004). CME height at the time of the lowest type IV frequency, interpolated or extrapolated whennot observed at that time.
The lowest type IV frequencies mark the time when the particle supply endsand/or the trapping loops start to shrink. The lowest observed type IV fre-quencies were 9–6 MHz, which correspond to heliocentric heights of 3.4–4.1 R (cid:12) when calculated using the atmospheric density model of Vr˘snak, Magdaleni˘c,and Zlobec (2004). The type II bursts were at that time already at much lower
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 5 alebpour Sheshvan and Pohjolainen frequencies, at 1.6–0.7 MHz, which correspond to heights of 8.0–12.6 R (cid:12) . Theheight separation between the front of the type IV emission source and thepropagating type II shock was therefore in the range of 4–9 R (cid:12) .When comparing the type II heights with the simultaneous CME leading frontheights, it is evident that the type II bursts were not due to CME bow shocks,but were shocks at the CME flanks (height separation of about 4–8 R (cid:12) ).The magnetic field structure near the eruption location on each day is shownin Figure 2. Figures 3 – 7 present the radio dynamic spectra and the correspond-ing coronagraph and EUV difference images for the five events, with the threedifferent views from the three spacecraft.
Figure 2.
The Potential Field Source Surface (PFSS) maps made from the SOHO/MDImagnetograms (Earth view from L1) provide an approximation of the magnetic field linestructures in each event, on a) 4 June 2011, b) 22 September 2011, c) 27 January 2012, andd) 5 March 2012. In the above maps, the purple and green lines are open field lines indicatingnegative and positive polarity, respectively. The white lines are closed field lines. Open fieldlines show locations where accelerated particles can escape from the active region, for exampleelectron beams observed as type III bursts.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 6 isibility of Compact IP type IV bursts
Type IVType II
StreamersPossible type II source locations
Type II
Figure 3.
Solar event on 4 June 2011: Flare location on the backside of the Sun, nearN20W140 (Earth view).
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 7 alebpour Sheshvan and Pohjolainen
Type IVType II
StreamersPossible type II source location
Figure 4.
Solar event on 4 June 2011: Flare location on the backside of the Sun, nearN20W160 (Earth view).
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 8 isibility of Compact IP type IV bursts
Possible type II source locationStreamer
Type II
Type IVType II
Figure 5.
Solar event on 22 September 2011: Flare location at N13E78 in AR1302.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 9 alebpour Sheshvan and Pohjolainen
Type IVType II
Possible type II source locationStreamer
Type IIType II
Figure 6.
Solar event on 27 January 2012: Flare location at N27W71 in AR1402. The red linein the STEREO-B EUVI difference image indicates the maximum extent of the EUV wave onthe visible disk.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 10 isibility of Compact IP type IV bursts
StreamerPossible type II source location
Type IVType II
Type IIType IV (cut)
Figure 7.
Solar event on 5 March 2012: Flare location at N17E52 in AR1429. The red linein the STEREO-B EUVI difference image indicates the maximum extent of the EUV wave onthe visible disk.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 11 alebpour Sheshvan and Pohjolainen
3. Results
The flare started approximately at 06:20 UT from the backside of the Sun, nearN20W140. The halo CME was first observed by SOHO/LASCO at 06:48 UT atheight 2.58 R (cid:12) , propagating to north-west with a linear fit speed of 1407 km s − (LASCO CME Catalog). An EUV wave occurred on the backside, observed bySTEREO-A/EUVI, and it originated from the active region located at ∼ W50in this field of view. The first brightening phase of the EUV wave started at06:30 UT, after which a dimming phase was observed at 06:50 UT. The EUVwave crossed most part of the disk within the STEREO-A view, but the wavewas not observed on the disk in the other field-of-views.The DH type IV event became visible in the dynamic spectrum at 07:12 UTat 16 MHz, and it was observed by STEREO-A but not by
Wind or STEREO-B.At 08:00 UT it reached the lowest frequency of about 6 MHz. As the flare sitewas at ∼ W50 in STEREO-A view we cannot know if a metric type IV burstexisted, as radio emission originating from low heights would have been blockedby the solar disk toward the Earth.The type IV burst was atypical in a sense that it showed type III-like burststructures superposed, and starting, from the wide-band emission (Figure 3,top right dynamic spectrum). This indicates that several open field lines existedalong which particles could stream and/or escape from the trap, see the potentialmagnetic field lines plotted in Figure 2a.A type II burst was observed by STEREO-A and
Wind , but not by STEREO-B. As the most probable location for the type II burst source is in the westernflank of the CME (Earth view, Figure 3 middle left difference image), it wouldhave been behind the Sun in STEREO-B view and hence undetectable.
The next type IV event occurred on the same day, 15 hours later, on 4 June2011. The flare originated from the same active region on the backside of theSun, starting at 21:45 UT and located at N20W160 approximately ( ∼ W70 inSTEREO-A view). The magnetic field structures could be as in the earlier eventon the same day (Figure 2a), but as the active region had now rotated wellbehind the limb, we cannot be sure of the actual configuration. The halo CMEwas first observed by SOHO/LASCO at 22:05 UT, at height 4.35 R (cid:12) , with alinear fit speed of 2425 km s − .An EUV wave that was followed by a dimming originated from the flarelocation and it was well-observed by STEREO-A. The wave was visible near thelimb in the STEREO-B view, but no wave signatures were observed from theEarth view (Figure 4, difference images on the left).The DH type IV burst was observed to start at 22:15 UT, observed bySTEREO-A only, and it reached the lowest frequency of 6 MHz at 23:05 UT. Avery fast-drifting type II burst was observed by STEREO-A, but this emissionwas not detected by the other instruments (Figure 4, top right spectrum). The SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 12 isibility of Compact IP type IV bursts type II burst source is again estimated to be located in the western flank of theCME (Earth view), but this time in a more narrow region that is blocked inview from the other radio instruments.
Type IIIType IV
Type IIIs M H z NDA
Metric type IV
Figure 8.
Radio dynamic spectrum observed by STEREO-B/WAVES on 22 September 2011(bottom) and
Nancay Decameter Array (NDA) spectrum at 70–10 MHz (top). The flaringregion was located at E80 in NDA view and at W10 in STEREO-B view. The metric type IVemission observed by NDA was not detectable in the
Wind /WAVES data (same field-of-view,frequencies below 14 MHz). The DH type IV burst observed in the STEREO-B field-of-viewdoes not have a one-to-one match in duration with the metric type IV burst. The type IIIbursts stop at the type II burst emission lane (indicated by a white dashed line), suggestingthat the electron beams cannot pass the type II shock front.
A GOES X1.4 class flare was observed to start at 10:29 UT on 22 September2011 in active region NOAA 11302, located at N13E78. It reached maximum fluxat 11:01 UT. A halo CME was first observed by SOHO/LASCO at 10:48 UT,at height 2.98 R (cid:12) , and it had a linear fit speed of 1905 km s − . An EUV wave SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 13 alebpour Sheshvan and Pohjolainen related to this event was mainly observed by STEREO-B/EUVI on the disk(Figure 5, bottom left difference image), but it was also observed to propagatetoward the west in Earth view, with a speed of 595 km/s (Nitta et al. , 2013).The DH type IV burst was observed at 11:10 UT by STEREO-B/WAVES at16 MHz and it reached the lowest frequency of 8 MHz at 12:15 UT. The DH typeIV burst was preceded by a decimetric-metric type IV burst observed by NDA(Earth view). These bursts do not seem to be directly associated, as the metrictype IV emission continued after the DH emission had ended, see Figure 8. Alsothe drift rates of the emission envelopes do not look to match. The metric typeIV burst shows groups of type III bursts with varying frequency drifts (bothpositive and negative), similar to those reported by Melnik et al. (2018).Near 12:00 UT all the three space instruments recorded a type III burst intheir dynamic spectra. The type III burst (formed by a propagating electronbeam) is observed to cross the DH type IV burst in STEREO-B spectrum.This, and also the later type III bursts, stop near the type II emission lane,suggesting that the electron beams cannot pass the type II shock front (Figure8). The location of the type II burst source is most probably in the south-easternflank of the CME (Earth view, Figure 5 middle left difference image), makingit observable to STEREO-B and
Wind . Figure 2b shows some of the open fieldlines along which electrons could stream out from the active region and they aremostly directed toward STEREO-B.
The active region NOAA 11402 produced an X1.7 GOES class flare at locationN27W71 that started at 17:37 UT and peaked at 18:37 UT on 27 January2012. The flare was associated with a halo CME that was first observed bySOHO/LASCO at 18:27 UT at height 3.76 R (cid:12) , and had a linear fit speed of2508 km s − . The event was associated with an EUV wave that was observedglobally on the disk by STEREO-A/EUVI, and partly on the disk by STEREO-B/EUVI and SDO/AIA. The EUV wave speed from Earth view was determinedas 635 km s − (Nitta et al. , 2013).The DH type IV burst was observed at 16 MHz at 18:30 UT by STEREO-A, but not by the other instruments. The burst reached the lowest frequencyof 9 MHz at 19:30 UT. The DH type IV burst was preceded by a metric typeIV burst at 18:15–18:35 UT, observed by the Green Bank Solar Radio BurstSpectrometer (GBSRBS) down to 300 MHz and by the
Radio Solar TelescopeNetwork (RSTN) at 180–25 MHz. Due to the short duration of the metric typeIV burst it is not possible to determine if these two bursts were related.A type II burst was recorded by all the three space spectrographs, althoughthe type II emission lane was more patchy and narrow in the STEREO-B spec-trum (Figure 6). The most probable location for the type II burst source is inthe south-western flank of the CME (Earth view, Figure 6 middle left differenceimage), thus making it visible in all three field-of-views. No type III bursts wereobserved during the DH type IV burst emission, although some open field linesexisted, directed to the north-west (Figure 2c).
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 14 isibility of Compact IP type IV bursts H i R A S M H z Figure 9.
Radio dynamic spectrum observed by Wind/WAVES on 5 March 2012 (bottom)and by HiRAS at 2000–20 MHz (top).
An X1.1 GOES class flare started at 02:30 UT from the NOAA active region11429 located at N17E52. At 03:12 UT a filament eruption formed into a CME,first visible in the SOHO/LASCO images at height 3.0 R (cid:12) . The linear fit speedof this CME was 594 km s − . A fast halo CME appeared at 04:00 UT at height4.2 R (cid:12) , with a linear fit speed of 1531 km s − , overshadowing the earlier CME.The launch of the second CME was most probably related to the fast rise in X-ray flux, observed to start at 03:25 UT. An EUV wave was observed globally onthe disk from Earth view, and it was estimated to move toward the south-westdirection at a speed of 915 km s − (Nitta et al. , 2013). The wave was partlyobserved also by STEREO-B/EUVI (Figure 7, bottom left difference image).The DH type IV burst was observed as a compact event by Wind /WAVES,starting at 04:15 UT and reaching the lowest frequency of 7 MHz at 05:15 UT.It was also observed as shorter-duration (’cut’) emission in the STEREO-B view(Figure 7, bottom right dynamic spectrum). This sudden end of emission at
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 15 alebpour Sheshvan and Pohjolainen all the type IV burst frequencies suggests that the emission was blocked frombeing observed, rather than ending as an emission process. No significant typeIII burst activity was observed during the type IV burst emission. Some openfield lines are however visible in the PFSS maps, directed toward the north-east,see Figure 2d.At decimeter–meter waves drifting continuum emission was observed at 03:30–05:15 UT (Figure 9, on top, spectrum from NICT/HiRAS). This could be afaint type IV burst or a noise storm. HiRAS spectrum shows that the emissionenvelope has a drift toward the lower frequencies, with a drift and timing thatquite well matches those of the DH type IV burst.The
Wind and STEREO-B spectra show also intense type II burst emission.This event has been analysed by Magdaleni´c et al. (2014), who used radio trian-gulation technique to locate the type II burst source positions. These were foundto be close to the south-eastern flank of the CME (Earth view, Figure 7, middleleft difference image), where streamers were also located. The appearance of thetype II emission was hence suggested to be due to shock wave and streamerinteraction.
4. Discussion
The first ideas to explain why type IV bursts show directivity in their emissionwere reviewed by Kundu (1965), see especially Fig. 11-12 in the book andreferences therein. A schematic drawing in Figure 10 presents the suggestedconfiguration, based on observations of metric type IV bursts as observationsfrom space at low frequencies did not yet exist. A flare explosion ejects a columnof gas and the outward moving gas carries a magnetic field with it. A smallfraction of particles accelerated to very high energies can be trapped in thefrozen-in magnetic fields and will emit continuum (type IV) emission in a widefrequency range. The directivity of emission could then be associated with thelarge extent of the type IV source in the direction of motion. Later on, a segmentof the type IV burst source can separate and move outward, but it will no longeremit synchrotron emission due to the insufficient magnetic fields.
Faint or no type IV emissionfrom this angle when observed top−onStrong type IV emission
Figure 10.
Schematic drawing along the lines presented in Kundu (1965) on where the typeIV burst source may be located. Directivity could be related to the extent of the source, whichis larger when viewed top-on, in the direction of propagation.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 16 isibility of Compact IP type IV bursts
Gopalswamy et al. (2016) recently studied a DH type IV burst on 7 Novem-ber 2013 that was observed to be intense and complete only in STEREO-Bobservations, with STEREO-A seeing a partial burst and
Wind no burst atall. However, Melnik et al. (2018) reported later that the type IV burst wasobserved also at decameter waves by URAN-2, from Earth. Gopalswamy et al. (2016) concluded that the type IV emission was directed along a narrow cone,less than ≈
60 degrees in width from above the flare site. They suggested thatthe source of energy for the burst was the flare: electrons accelerated due to flarereconnection and then trapped in the post-eruption structures, producing radioemission at the local plasma frequency. As Melnik et al. (2018) were able toobserve the start of the type IV burst at decameter waves, they confirmed thatthe emission was indeed plasma emission since it was highly polarized. Melnik etal. (2018) suggested that the source region for the type IV burst was the CMEcore but they did not specify why URAN-2 could observe it while
Wind couldnot.We note that it is possible that the observed directivity may not be caused bythe emission mechanism and/or the extent of the type IV source. Alternatively,the produced radiation could be stopped in certain directions, by for examplehigher density plasma if it existed in between the source and the observer. Inthe solar corona such local density enhancements can occur in coronal streamersand shock-related compression fronts (Kwon and Vourlidas, 2018). Figure 11shows how emission from the n2-region cannot pass through the higher densityplasma in the n3-region, and hence it cannot be observed along this direction.The dense solar atmosphere can also block part of the emission, especially ifthe source region is located on the backside of the Sun and the direction ofpropagation is away from the observer.
No emission n2Sunobserved n3 Emission observedby higher density plasman1n1 Density:n1 < n2 < n3Emission from n2 absorbed
Figure 11.
Schematic drawing for not detecting radio emission from a coronal source (n2)which is located behind a region of higher density plasma (n3). Due to the plasma frequency,longer wavelengths produced by plasma emission and/or sychrotron emission in the n2-regionwill be stopped and absorbed in the n3-region. To directions not blocked by the n3-region orthe Sun itself, the emission can be observed in full.
The five events in our study presented configurations similar as in Figure 11:a CME source region was located in n2-region and a streamer was located in n3-region. Moreover, in all events a type II burst was estimated to be located nearthe streamer, as a CME flank shock. Therefore, a higher density region existed
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 17 alebpour Sheshvan and Pohjolainen toward a direction where the type IV burst could not be observed. In the two4 June 2011 events the type II – streamer region was in the south-western CMEflank, blocking type IV emission toward the Earth but being fully observable withSTEREO-A. In the 22 September event the type II – streamer region was in thesouth-eastern CME flank, and the type IV burst was observed by STEREO-B butnot by
Wind from the Earth direction. In the 27 January 2012 event the type II –streamer region was in the south-western CME flank, with a similar configurationas in the 4 June 2011 events. In the 5 March 2012 event the type II – streamerregion was located in the south-eastern CME flank (projected Earth view), butthe CME propagation direction indicated that the type IV burst source wasdirected more toward the Earth and the streamer was located behind it. Theeffect was that the type IV burst was observed by
Wind near Earth, but it wasonly partly visible for STEREO-B.In the 22 September 2011 event we observed several type III bursts thatended near the type II burst emission frequency. This type of ’cut-off’ has beenreported earlier by Al-Hamadani et al. (2017). One possible explanation forthe disappearance of type III radio emission could be a reduced level of beam-driven Langmuir waves, like in the case when two electron beams cross and theradio emission gets depleted (Briand, Henri, and Hoang, 2014). However, in thisscenario the level of radio waves is recovered after the beam crossing, unlike inour event on 22 September 2011. This suggests that also the type III burstscould not pass the type II – streamer region, but were stopped.
5. Summary and Conclusions
In this paper, we have studied five solar events that showed intense and compactDH type IV bursts, with the duration of around one hour for most of them, butvisible only from one viewing angle. The associated flares had high intensities(three GOES X-class, two unclassified as they appeared on the backside of theSun). All the CMEs had very high speed and were halo-type. Our analysis showedthat EUV waves were observed in all of the events, and intense and compact typeIV bursts were observed only when the EUV wave propagated globally acrossthe whole visible disk. This may simply indicate the propagation direction of thedepleted CME material, if not the directivity of the type IV radio emission.All the CMEs were associated with type II radio emission at DH wavelengths,which indicates propagating shock fronts driven by CMEs. The calculated es-timates for the type II heights showed that the shocks were not bow shocksat the leading fronts of the CMEs, but were located near the CME flanks. Inone event the type II location at the CME flank had also earlier been verifiedwith the radio triangulation technique. In all five events there were high-densitystreamers present at the deduced locations of the type II shocks. Shock–streamerinteraction has been found to be the source of type II burst emission in severalearlier studies ( e.g. , Al-Hamadani et al. , 2017, and references therein). Our anal-ysis showed that in all five events the type IV emission was not observed from adirection where the type II – streamer region was located in between the CMEsource region and the observer.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 18 isibility of Compact IP type IV bursts
We therefore conclude that a high density region existed in all events towardthe direction where the DH type IV burst could not be observed. As the typeII burst sources were found to be located higher in the corona than the type IVsources, the type IV emissions could have been stopped in these shock regions.Dense type II emitting regions could simply be optically thick for the type IVemission. In one of the events the type II region was also capable of stoppinglater-accelerated electron beams, visible as type III bursts that ended near thetype II burst lane.
Type II Metric type IVType IIMetric type IV +DH type IV
Figure 12.
Schematic cartoon that explains why metric type IV emission sources can beobserved when DH type IV sources can not, from the same viewing angle. The green regionsindicate possible type II burst locations and the filled grey region shows streamer position. Inthis scenario DH type IV emission is observed only from one viewing angle, because the plasmanear the type II emitting region is optically thick for the type IV emission to pass through.The higher density region coincides here with a type II shock – streamer region and it blocksthe type IV emission only at longer DH wavelenghts, so that the metric type IV emission canstill be observed.
To summarize our results and also to tie up our suggested scenario with anearlier analysed event – where a metric type IV burst could be observed froma direction where the DH type IV burst could not – we present a schematiccartoon in Figure 12. Basically, a type IV burst can be observed from a direc-tion where nothing blocks or absorbs the radio waves. Based on our analysis,possible blocking regions could be located at CME flanks, where shock-streamerinteractions may form type II radio bursts and where electron densities can bemuch higher than in the surrounding space. These regions may form only higherin the corona, making it possible for the metric type IV bursts to be visibleat lower coronal heights. We also note that in most cases the type II burststhemselves are visible even if the type IV bursts are not.
Acknowledgments
We thank the anonymous referee for comments and suggestions thathelped to improve this article. We are grateful to all the individuals who have contributedin creating and updating the various solar event catalogues. The CME catalog is generatedand maintained at the CDAW Data Center by NASA and the Catholic University of Americain cooperation with the Naval Research Laboratory. The
Wind
WAVES radio type II burstcatalog has been prepared by Michael L. Kaiser and is maintained at the Goddard SpaceFlight Center. SOHO is a project of international cooperation between ESA and NASA. N.Talebpour Sheshvan wishes to thank CIMO (The Centre for International Mobility, Finland)for financial support, contract TM-16-10298.
SOLA: SOLAtype4final.tex; 23 October 2018; 1:05; p. 19 alebpour Sheshvan and Pohjolainen
Disclosure of Potential Conflicts of Interest
The authors declare thatthey have no conflicts of interest.
References
Al-Hamadani, F., Pohjolainen, S., Valtonen, E.: 2017,
Solar Phys. , 127. DOI10.1007/s11207-017-1148-6Bougeret, J.-L., Goetz, K., Kaiser, M.L., Bale, S.D., Kellogg, P.J., Maksimovic, M., et al. :2008,
Space Sci. Rev. , 487. DOI 10.1007/s11214-007-9298-8Bougeret, J.-L., Kaiser, M.L., Kellogg, P.J., Manning, R., Goetz, K., Monson, S.J., Monge, N.,Friel, L., Meetre, C. A., Perche, C., Sitruk, L., Hoang, S.: 1995,
Space Sci. Rev. , 231.DOI 10.1007/BF00751331Briand, C., Henri, P., Hoang, S.: 2014, J. Geophys. Res. , 2365. DOI 10.1002/2013JA019688Cane, H.V.: 1984,
Astron. Astrophys. , 205.Cunha-Silva, R.D., Fernandes, F.C.R., Selhorst, C.L.: 2015,
Astron. Astrophys. , id.A38.DOI 10.1051/0004-6361/201425388Gopalswamy, N., Akiyama, S., M¨akel¨a, P., Yashiro, S., Cairns, I.H.: 2016, URSI Asia-PacificRadio Science Conference in Seoul, August 21-25, 2016. 2016arXiv160502223G.Hillaris, A., Bouratzis, C., Nindos, A.: 2016,
Solar Phys. , 2049. DOI 10.1007/s11207-016-0946-6Krupar, V., Maksimovic, M., Kontar, E.P., Zaslavsky, A., Santolik, O., Soucek, J., Kru-parova, O., Eastwood, J.P., Szabo, A.: 2018,
Astrophys. J. , id. 82 DOI10.3847/1538-4357/aab60fKundu, M.R.; 1965, Solar Radio Astronomy, Interscience Publishers.Kwon, R-Y., Vourlidas, A.: 2017,
Astrophys. J. , 246. DOI 10.3847/1538-4357/aa5b92Kwon, R-Y., Vourlidas, A.: 2018,
J. Space Weather Space Clim. , id. A08 DOI10.1051/swsc/2017045Leblanc, Y., Dulk, G. A., Vourlidas, A., Bougeret, J.-L.: 2001, J. Geophys. Res. , 25301.DOI 10.1029/2000JA000260Melnik, V. N., Brazhenko, A. I., Konovalenko, A. A., Dorovskyy, V. V., Rucker, H. O.,Panchenko, M., Frantsuzenko, A. V., Shevchuk, M. V.: 2018,
Solar Phys. , 53. DOI10.1007/s11207-018-1271-zMagdaleni´c, J., Marqu´e, C., Krupar, V., Mierla, M., Zhukov, A.N., Rodriguez, L., Maksimovi´c,M., Cecconi, B.: 2014,
Astrophys. J. , 115. DOI 10.1088/0004-637X/791/2/115Nindos, A., Aurass, H., Klein, K.-L., Trottet, G.: 2008,
Solar Phys. , 3. DOI10.1007/s11207-008-9258-9Nitta, N.V., Schrijver, C.J., Title, A.M., Liu, W:
Astrophys. J. , 58. DOI 10.1088/0004-637X/776/1/58Olmedo, O., Vourlidas, A., Zhang, J., Cheng, X.: 2012,
Astrophys. J. , 143, DOI10.1088/0004-637X/756/2/143Pick, M., Vilmer, N.: 2008,
Astron. Astrophys. Rev , 1. DOI 10.1007/s00159-008-0013-xVr˘snak, B., Magdaleni˘c, J., Zlobec, P.: 2004, Astron. Astrophys. , 753. DOI 10.1051/0004-6361:20034060, 753. DOI 10.1051/0004-6361:20034060