geoeffective solar event of june 21, 2015: effects on the ...€¦ · 12371, a few minutes before...

Comprehensive analysis of the Geoeffective Solar Event of June 21, 2015: Effects on the Magnetosphere,

Plasmasphere and Ionosphere Systems - part 1.

Francesca Zuccarello (1) et al. (1)Department of Physics and Astronomy, University of Catania, Italy.

14° European Space Weather Week November 27 – December 1 2017: Session 3

•  Active Region NOAA 12371 1.  Main characteristics; 2.  Magnetic configuration; 3.  Fractal analysis; 4.  Flare forecasting

•  The flare SOL2015-06-21T01:02 1.  Flare evolution; 2.  The associated CME

•  SEP event

1.  General Features; 2.  The SEP forecast method;

•  Summary and Conclusions (Part 1)

Comprehensive analysis of the Geoeffective Solar Event of June 21, 2015: Effects on the Magnetosphere, Plasmasphere and Ionosphere Systems - part 1.

Characteristics of NOAA 12371

u The AR exhibites a δ configuration (middle panel).

u At chromospheric heights, a sigmoidal-like structure is visible along the PIL

o  A full-halo CME left the Sun on June 21, 2015 from NOAA 12371, encountering Earth on June 22, 2015 and generating a strong geomagnetic storm (max Dst value: -204 nT).

o  The CME was associated with an M2 class flare observed at 01:42 UT, located at N12E16.

o  Using satellite data from solar, heliospheric, magnetospheric missions and ground-based instruments, we performed a comprehensive Sun-to-Earth analysis.

Solar event of June 21, 2015

Figure 2. Top: Map of the photospheric continuum of AR NOAA 12371, acquired bySDO/HMI some minutes before SOL2015-06-21T01:02. The region indicated with a solidline shows the FOV used for the analysis of SOT/SP data. Middle: Simultaneous SDO/HMImagnetogram. The values of the longitudinal field are saturated at ±2000 G (white/black).Bottom: Simultaneous SDO/HMI magnetogram. Red (blue) areas indicate positive (negative)polarity. SDO/AIA emission at 304 A passband is superimposed on the magnetogram map.

SOLA: SOLA-S-16-00286_R2_Gug.tex; 7 September 2017; 11:56; p. 9







NOAA 12371 in the photosphere (SDO/HMI). Simultaneous SDO/HMI magnetogram. 304 Å image superimposed on the

magnetogram. Red (blue) areas indicate positive (negative) polarity,

Fractal and multi-fractal analysisSolar event of June 21, 2015

Figure 3. Time series of the fractal and multifractal parameters measured on the AR 12371,by considering both unsigned (black circles) and signed (positive and negative, red diamondsand blue crosses, respectively) flux data. Top: fractal parameters D0 (left) and D8 (right).Bottom: Cdiv (left) and Ddiv (right). Time 0 corresponds to 00:00 UT on June 20, 2015.Vertical thin-solid (thin-dashed) lines indicate the time of occurrence of M-class (C-class) flareshosted by the AR. Flares associated with the CME occurred on June 21, 2015 are indicated bythick-solid line. Error bars show the uncertainty associated with the measured values, detailsare given in the text. For clarity, the error bars are only shown for the results from unsignedflux data.

opposite temporal evolution. Indeed, the time series of the fractal (multifractal)parameters measured on the AR 12371 look rather similar and flat over time,but for the results of the D0 and D8 (Cdiv and Ddiv) measurements derived fromthe positive flux data that show a net decrease (increase) during the analyzedperiod. The trends of the values estimated for the same quantities from unsignedand negative flux data are rather unvaried over time. We conclude that, while theabove measurements point out the eruptive potential of the AR NOAA 12371ahead of the events occurred on June 21, 2015, they also suggest the lack ofclear effects of these events in the photospheric configuration of the AR NOAA12371 magnetic field.

Figure 2 (top panel) shows the photospheric configuration of AR NOAA12371, a few minutes before the start of SOL2015-06-21T01:02. The AR ex-hibited a central part with opposite polarities in contact to each other, sharingsome penumbral filaments (δ configuration, see Figure 2, middle panel). Atchromospheric heights, a sigmoidal-like structure is visible along the polarityinversion line (PIL) present in the region (bottom panel).

Along the PIL, peculiar flows of upflows/downflows of about ∓1.5 km s−1

were found, which are not related to the classical Evershed flow observed in


Time series of the fractal and multifractal parameters measured on the AR 12371, by considering both unsigned (black circles) and signed (positive and negative, red diamonds and blue crosses, respectively) flux data. Top: fractal parameters D0 (left) and D8 (right). Bottom: Cdiv (left) and Ddiv (right). Time 0 corresponds to 00:00 UT on June 20, 2015.

Blue (red) symbols display results for negative (positive) magnetic flux. Thin-solid (thin-dashed) lines: time of M-class (C-class) flares. Thick-solid line: flares associated with the CME occurred on 21/06 Error bars are shown for unsigned flux data.

§  Eruptive potential of NOAA 12371 ahead of the events occurred on June 21, §  Lack of clear effects of these events in the photospheric magnetic field configuration

Flare forecasting parameters ��from SDO/HMI magnetograms

Log(R) parameter as a function of time, indicating the probability to have a flare > M1 in the next 24 hours.


Figure 15. Rescaled parameters as a function of time. We rescale to unity all the parametersin order to compare the trends. Shaded yellow area: solar longitude > 60◦. Shaded grey andred: flares > M1 produced by AR12371.

as those reported in figure 3 for the multifractal parameters. This supports theconclusions reported in section 2.1, stating that there is little or no evidenceat all of a change of configuration of the magnetic field at photospheric levelassociated to the flare.

4. Halo CME

As we mentioned, during the June 21, 2015 event none of the space-basedcoronagraphs on-board STEREO spacecraft was acquiring data. Nevertheless,the LASCO-C2 and -C3 visible light coronagraphs on-board SOHO acquired avery nice sequence of images showing the halo-CME and the CME-driven shockexpanding towards the Earth. In particular during the event the LASCO-C2coronagraph (with a projected field of view going from 2.1 to 6.0 solar radii)acquired images with the ”Open” filter at 02:36 UT (the last frame just beforethe CME enters in the LASCO-C2 field of view) and at 02:48, 03:12, 03:24 and03:36 UT. This sequence shows nicely the early expansion of the halo-CME, aswell as the propagation of the CME-driven shock ahead of the CME front. Thesubsequent expansion of the CME was captured higher up by the LASCO-C3coronagraph (with a projected field of view going from 3.6 to 33 solar radii), thatacquired images with the ”Open” filter at 03:06 UT (the last frame just beforethe CME enters in the LASCO-C3 field of view) and at 03:18, 03:12, 03:24 and03:36 UT. This sequence shows very well the interplanetary expansion of thehalo-CME.


Rescaled parameters as a function of time. Shaded yellow area: solar longitude > 60◦. Shaded grey and red: flares > M1 produced by AR12371.

q  Log(R) is a proxy of the photospheric electrical currents (Schrijver 2007) and is a measure of the maximum energy available in the AR.

q The log(R) values are high during all the examined period, indicating a high probability of flare occurrence.

Piersanti et al.

Figure 13. The log(R) parameter as a function of time. We report the probability to have aflare > M1 in the next 24 hours as from Schrijver (2007). Shaded yellow area: solar longitude> 60◦. Shaded grey and red: flares > M1 produced by AR12371.

Figure 14. The log(R) parameter as a function of time. We here concentrate on the firsthours of June 21, 2015. We report the probability to have a flare > M1 in the next 24 hoursas from Schrijver (2007). Shaded areas: flares > M1 produced by AR12371, in red the flareinvestigated in this paper.


Velocity maps and magnetic field extrapolationPiersanti et al.

sunspots. These flows are reminiscent of the velocity field configuration found inδ complexes by Shimizu, Lites, and Bamba (2014) and Cristaldi et al. (2014),which has been attributed to shear accumulation.

Figure 4. Top: Map of the Doppler velocity of AR NOAA 12371, acquired by SDO/HMI someminutes before SOL2015-06-21T01:02. Bottom: Same at the time of flare peak. The values ofthe Doppler velocity are saturated at ∓1.5 km s−1 (blue/red).

Taking advantage of the resolving power of the New Solar Telescope at BigBear Solar Observatory (BBSO, see also Jing et al. (2016)), we can imagethe fine details of the photospheric configuration of AR NOAA 12371. InFig. 5 (left) we show a continuum HMI image displaying the photosphericconfiguration of AR NOAA 12371 marked with a red box indicating theIRIS satellite FoV, while the blue box indicates the BBSO FoV centered in theδ complex. Fig. 5 (right) shows an image acquired by BBSO in the TiO band,


Piersanti et al.

Figure 10. Linear force-free field extrapolation of the photospheric magnetic fieldof the AR NOAA 12371.

Table 2. Evolution of the mean value of shear angle θ, dip angle ∆γ, current|jz|, and gradient |∇Bsz |, along the PIL of AR NOAA 12371.

Time < θ > < ∆γ > < |jz | > < |∇Bsz | >

(UT) (degrees) (degrees) (mA/m2) (G/m)

2015-06-20 15:10:48 42.7 0.94 16.0 14.4

2015-06-20 20:03:52 43.9 -2.19 17.9 25.2

2015-06-21 01:00:29 67.6 -1.23 18.2 12.4

2015-06-21 06:22:26 64.1 -0.93 13.4 9.8

angle, larger than 45◦. Note that small patches in the FOV far from thePIL showing a large shear angle, near regions with Bst less than 200G (white background) may be affected by errors in the 180◦ azimuthambiguity resolution. The shear angle exhibits a slight decrease afterthe flare.

We also used the results obtained with the NPFC code to estimate the electriccurrent in the vertical direction, |jz|, and the gradient of the vertical componentof the magnetic field, |∇Bsz|, following Georgoulis and LaBonte (2004).

In Table 2 we report the mean (unsigned) values of the shear angle, dip angle,|jz|, and |∇Bsz | calculated along the PIL. We see that the shear angle increasesuntil the flares occur, and decreases at the end. The dip angle exhibits a similarbehavior. Also |jz | values grow until the eruptive event occurs and diminish afterthe flares, while |∇Bsz| begins to lessen before the events. This trend indicatesthat a dynamical process of energy storage is taking place during hours beforethe eruptive phenomena, through shear accumulation. Then, after the energyrelease events, a relaxed state is reached.


Top: Map of the Doppler velocity of before the flare (SDO/HMI).

Bottom: Same at the time of flare peak. Doppler velocity is saturated at ∓1.5 km s−1 (blue/red).

Evidence of flows not related to the Evershed flow.

Similar to the velocity f i e l d c on f i gura t i on found in δ complexes, a t t r ibuted to shear accumulation.


Figure 11. Potential field extrapolation of the full disk magnetic field on June 21 at 00:04UT. On the solar surface the longitudinal component of the photospheric magnetic field takenby SDO/HMI is shown.


P o t e n t i a l f i e l d extrapolation of the full disk magnetic field on June 21 at 00:04 UT. (Potential Field Source Surface - PFSS)

Linear force-free field extrapolation of the photospheric magnetic field of the AR NOAA 12371.

Characteristics of the flaresPiersanti et al.

Figure 8. Sequence of AIA 211 A images showing the evolution of the flare that occurredin AR NOAA 12371. The two ribbons of the flare are clearly visible at [-300:-180], [80:300] inall the images. The destabilization and later eruption of a filament can be observed startingat 01:38 UT at coordinates [-200:-100], [-50:50]. An animation of this figure is available in theonline journal.

(1981). As a proxy of this shear, we used the horizontal shear angle θ, as definedin Romano et al. (2014); Gosain and Venkatakrishnan (2010).

We computed the dip angle, which measures the difference between the incli-nation angle of the observed field and that of the potential field (see, e.g., Gosainand Venkatakrishnan, 2010; Petrie, 2012; Romano et al., 2014). This quantity is


Piersanti et al.

Table 1. Characteristics of the two flares observed by theGOES-15 satellite in AR NOAA 12371, before the CME.

Flare Class Time (UT)

begin peak end

SOL2015-06-21T01:02 M2.0 01:02 01:42 02:00

SOL2015-06-21T02:06 M2.6 02:06 02:36 03:02

and validated by Alberti et al. (2017a), in order to make a prediction of theassociated SEP event. Furthermore, to investigate the influence of the disturbedelectric field on the low latitude ionosphere induced by geomagnetic storms(Muella et al., 2010; Alfonsi et al., 2013; Tulasi Ram et al., 2016; Spogli et al.,2016), we focused on the morphology of the crests of the EIA. To do that, weconcentrate on the ionospheric characterization provided by the simultaneoususe of GNSS receivers, ionosondes and Langmuir probes on board the SWARMconstellation. In addition, we analyzed the response of the different magneto-spheric current systems to the ICME arrival by a comparison between TS04model (Tsyganenko and Sitnov, 2005) predictions, magnetospheric observationsand geomagnetic measurements during the Sudden Impulse (SI). In particular,using ground based observations from low to high latitudes, we reconstructedthe ionospheric current system associated to the SI. Moreover, we investigatedthe dynamics of the plasmasphere during the different phases of the geomagneticstorm by examining the time evolution of the radial profiles of the equatorialplasma mass density as inferred from field line resonances detected by the EMMAnetwork (1.5 < L < 6.5). We present the general features of the geomagneticresponse to the ICME, by applying innovative data analysis tools that allowto investigate the time variation of ground-based observations of the Earth’smagnetic field during the associated geomagnetic storm and a description ofthe polar ionospheric convection is also presented. Finally, using SuperDARNmeasurements, we analyzed the polar ionospheric convection during the SI, themain phase and the recovery phase of the GS.

2. Solar Data

The CME that encountered the Earth and generated the geomagnetic stormon June 22, 2015 originated in active region (AR) NOAA 12371. This ap-peared on the eastern limb of the solar disk on June 16, 2015. At that time,its magnetic configuration was classified as β, evolving into βγδ in the followingdays. On June 21 two subsequent flares were observed in the AR and their X-ray flux was measured by the GOES-15 satellite: SOL2015-06-21T01:02 andSOL2015-06-21T02:06, classified as M2.0 and M2.6, respectively. At 02:36 UTthe LASCO coronagraphs aboard the SOHO satellite first observed thehalo CME expanding into the heliosphere.

A number of solar facilities observed AR NOAA 12371 during its passageacross the solar disk, and during time intervals close to the CME as well.


Piersanti et al.

Figure 1. GOES X-rays flux curves in the 18 A channel (solid line) and in the 0.54 A channel(dotted line). The vertical line indicates the time of the first detection of the halo CME.


GOES X-rays flux curves .Vertical line: time of the first detection

of the halo CME.

Sequence of AIA 211 A images showing the evolution of the flare


Figure 5. Left: continuum SDO/HMI image showing the photospheric configuration of ARNOAA 12371. The red box indicates the FoV observed by the IRIS satellite; right: BBSOimage acquired in the TiO band.

Figure 6. SDO/HMI magnetogram at the peak of SOL2015-06-21T01:02. Red (blue) areasindicate positive (negative) polarity. A composite image of SDO/AIA emission at 94 A and335 A passbands is superimposed on the magnetogram map.

centered at 750.7 nm, which shows the details of the δ complex. We can seethat the eastern umbra is characterized by the presence of light bridges and thatthe penumbral filaments located between the two opposite polarity umbrae arehighly sheared.

The M2.0 flare is located along the PIL, as shown in Figure 6. Figure 2.1displays the morphology of the coronal regions of AR NOAA 12371 close to theflare peak, as visible in SDO/AIA images. The online movies in the various pass-bands show that, actually, the evolution between the two M2.0 and M2.6 flaresoccurs without interruption. During the event, several coronal structures are


The M2.0 flare is located along the PIL. Red (blue) areas indicate positive (negative) polarity. A composite image of SDO/AIA emission at 94 A and 335 A passbands is superimposed on the magnetogram map.

The evolution between the two M2.0 and M2.6 flares occurs without interruption. During the e v ent , s e v e ra l c o rona l s t ruc ture s are destabilized in a succession reminiscent of a domino-like effect triggered by the an activation process occurring in the δ complex.

Piersanti et al.

Figure 10. Linear force-free field extrapolation of the photospheric magnetic fieldof the AR NOAA 12371.

Table 2. Evolution of the mean value of shear angle θ, dip angle ∆γ, current|jz|, and gradient |∇Bsz |, along the PIL of AR NOAA 12371.

Time < θ > < ∆γ > < |jz | > < |∇Bsz | >

(UT) (degrees) (degrees) (mA/m2) (G/m)

2015-06-20 15:10:48 42.7 0.94 16.0 14.4

2015-06-20 20:03:52 43.9 -2.19 17.9 25.2

2015-06-21 01:00:29 67.6 -1.23 18.2 12.4

2015-06-21 06:22:26 64.1 -0.93 13.4 9.8

angle, larger than 45◦. Note that small patches in the FOV far from thePIL showing a large shear angle, near regions with Bst less than 200G (white background) may be affected by errors in the 180◦ azimuthambiguity resolution. The shear angle exhibits a slight decrease afterthe flare.

We also used the results obtained with the NPFC code to estimate the electriccurrent in the vertical direction, |jz|, and the gradient of the vertical componentof the magnetic field, |∇Bsz|, following Georgoulis and LaBonte (2004).

In Table 2 we report the mean (unsigned) values of the shear angle, dip angle,|jz|, and |∇Bsz | calculated along the PIL. We see that the shear angle increasesuntil the flares occur, and decreases at the end. The dip angle exhibits a similarbehavior. Also |jz | values grow until the eruptive event occurs and diminish afterthe flares, while |∇Bsz| begins to lessen before the events. This trend indicatesthat a dynamical process of energy storage is taking place during hours beforethe eruptive phenomena, through shear accumulation. Then, after the energyrelease events, a relaxed state is reached.


Piersanti et al.

Figure 12. Top: map of the vertical component Bsz some minutes before the startof the flaring activity in AR NOAA 12371; middle: simultaneous map of the shearangle; bottom: map of the shear angle three hours after the flares. The solid red lineindicates the PIL.


Map of the vertical component of the magnetic field some minutes before the flare. Solid red line: PIL.

Simultaneous map of the shear angle

Map of the shear angle three hours after the flares.

The analysis of the shear angle, of the gradient of the vertical magnetic field and of the electric current indicate that an energy storage mechanism, compatible with shear accumulation, is active before the eruption. After the flares, the region of the δ complex achieves a more relaxed state.

Evolution of the shear angle and of the dip angle

Left: difference between the pB** image acquired during the halo CME and the last pB image available before the eruption. White pixels indicate density increase. Right: map of the position along the LOS of the density increases associated with the CME.


pB difference

-6 -4 -2 0 2 4 6X (Rsun)

-6

-4

-2

0

2

4

6

Y (R

sun)

-6 -4 -2 0 2 4 6-6

-4

-2

0

2

4

6LOS position map

-6 -4 -2 0 2 4 6X (Rsun)

-6

-4

-2

0

2

4

6

Y (R

sun)

-6 -4 -2 0 2 4 6-6

-4

-2

0

2

4

6

0

1

2

3

4

5

6

LOS

dist

ance

(Rsu

n)

Figure 16. Left panel: the difference between the pB image acquired during the halo CME(polarized sequence acquired on June 21 between 02:54 and 03:02 UT) and the last pB imageavailable before the eruption (polarized sequence acquired on June 20 between 21:00 and 21:08UT). Negative values (black) have been excluded in the polarization ratio analysis to consideronly pixels (white) where the CME transit leads to a density increase. Right panel: map ofthe position along the LOS of the density increases associated with the CME as obtained withpolarzation ratio technique (see text).

The resulting map of z values is shown in Figure 16 (right panel); this mapsuggests a correlation between distances ρ from the Sun projected on the plane ofthe sky and distances z along the line of sight, indicating that the reconstructedcloud of 3D points has a distribution similar to the surface of a cone with vertexlocated on the CME source region on the Sun and axis parallel to the line ofsight. In order to better understand the resulting 3D structure of the halo CME,we built bar-plots (Figure 17) showing the distribution of POS distances ρ (topleft panel), LOS distances z (top right), as well as the distribution of polar anglesφ on the POS (bottom left) and of angles θ from the POS. These plots showthat points where the polarization ratio technique is successful are distributedquite homogeneously in projected distance on the POS and less homogeneouslyin polar angle; moreover, the bulk of reconstructed points is located at a distanceof about 2 solar radii from the POS and that are expanding at an angle fromthat plane of about 25◦. We point out that a big source of uncertainty is relatedwith the total time required to acquire the whole polarized sequence by about7m 20s; during this time any CME feature with projected speed of 1000 km s−1

moved by ∼ 600 arcsecs, corresponding to ∼ 25 pixels (for a 512 × 512 pixelsLASCO-C2 image).

All the above information derived from white light images are crucial topredict the CME arrival time at 1 AU and to study the CME interplanetarypropagation. For instance, a simple estimate of the ICME arrival time at 1


The associated Halo Coronal Mass Ejection

At 02:36 UT the LASCO-C2* coronagraph aboard the SOHO satellite first observed the halo CME expanding into the heliosphere.

* No STEREO observations

** pB: polarized white light brightness

The CORIMP Catalogue indicates t h a t t h e C M E i s s l i g h t l y accelerating (a ≃ 150 m s−2) during the early expansion phase (between ∼ 3 and ∼ 6 UT), and then slightly decelerating (a ≃ −150 m s−2) higher up in the LASCO-C3 field-of-view.

The reconstructed cloud of 3D points has a distribution similar to the surface of a cone with vertex located on the CME source region on the Sun and axis parallel to the line of sight.

The SEP event of 21 June

Piersanti et al.

10-3

10-2

10-1

1

101

102

103

104

(p/cm2ssr)

Proton

flux

> 10 MeV> 30 MeV> 60 MeV

20 21 22 23 24 25 26 27 28 29 301

101

102

103

104

105

106

107

(p/cm2ssrMeV)

June 2015

Proton

flux

P1 (0.047 - 0.066 MeV)P2 (0.066 - 0.114 MeV)P3 (0.114 - 0.190 MeV)P4 (0.190 - 0.310 MeV)P5 (0.310 - 0.580 MeV)P6 (0.580 - 1.05 MeV)P7 (1.05 - 1.89 MeV)P8 (1.89 - 4.75 MeV)

Figure 18. Temporal behavior of the proton integral (top) and differential (bottom) flux asrecorded in different energy channels (energy reported in the legend) by EPAD/GOES andEPAM/ACE, respectively, during the June 21 SEP event (http://omniweb.gsfc.nasa.gov). Thecyan, dashed black and solid black lines mark the time of the associated flare maximum, June19 CME-driven shock and June 21 CME-driven shock at ACE, respectively.

5.1. HIGH ENERGY OBSERVATIONS and the PAMELAapparatus

The PAMELA instrument provides the opportunity to extend the analysis of theSEP event to higher energies. PAMELA was launched on board the Resurs-DK1Russian satellite by a Soyuz rocket from the Baikonur space centre on the 15th

of June 2006 with an inclination of 70◦ and in a elliptical orbit (now circularwith an almost stable altitude of ∼570 km). The apparatus core is composed of


Temporal behavior of the proton integral (top) and differential (bottom) flux as recorded in different energy channels by EP A D / G O E S a n d E P A M / A C E , respectively, during the June 21 SEP event.

The cyan, dashed black and solid black lines mark the time of the associated flare maximum, June 19 CME-driven shock and June 21 CME-driven shock at ACE, respectively.

Piersanti et al.

]-1

s-1

sr

-2In

tegr

ated

Flu

x [p

roto

ns x

cm

1

10

210

310

410

510

GOES-15 x 100 (8-15 MeV)GOES-15 x 10 (15-40 MeV)GOES-15 (40-80 MeV)

Days of June 201521 22 23 24 25 26 27 28 29 30

Cou

nts/

s

2−10

1−10

1

10

210

310

Preliminary PAMELA x 10000 (450 MV)

PAMELA x 200 (650 MV)PAMELA (1500 MV)

Figure 19. In the top panel, the GOES proton fluxes as a function of time in three energyintervals is presented. In the bottom panel PAMELA counts per second are shown for threedifferent rigidity channels. The vertical line indicates the maximum time of the M2.6 flare onthe Sun, while the horizontal one highlights the almost undisturbed ∼ 1500 MV count rateplus the Forbush decrease created by the Halo CME associated to the flare. The longer datasampling for PAMELA (3 hours) with respect to the GOES one (only 32 seconds) is due toboth statistical and orbital limitations. The latter are caused by the magnetic cut-off thresholdwhich blocks the arrival of very low energy particles in specific regions of the Earth. Data fromPAMELA are preliminary.

Cane (2000)) which is due to the interplanetary counterpart of the full halo CMEleaving the solar surface at about 02:30 UT of June 21.

The Forbush decrease was also observed by the worldwide neutron monitor(NM) network. For instance, the Rome NM (geographic coordinates: 41.86◦N,12.47◦E,sea level; effective vertical cutoff rigidity - Epoch 1995: 6.27 GV) registeredabout a 5% variation in the cosmic ray intensity, as displayed in Figure 20 (fromhttp://webusers.fis.uniroma3.it/svirco/Dati).


The PAMELA instrument provides the opportunity to extend the analysis of the SEP event to higher energies.

Top panel: GOES proton fluxes as a function of time in three energy intervals.

Bottom panel: PAMELA counts per second for three different rigidity channels.

Vertical line: maximum time of the M2.6 flare;

Horizontal line highlights the almost undisturbed ∼ 1500 MV count rate plus the Forbush decrease created by the Halo CME associated to the flare.


Figure 20. Time history of the cosmic ray intensity recorded at the Rome NM (SVIRCOObservatory) for June 2015.

Figure 21 shows the event-integrated differential proton flux as a function ofrigidity measured by PAMELA in the time interval June 22-23 with respect tothe galactic flux measured in the first 20 days of June. Both fluxes are scaled tobetter show the amount of the increase due to the June 21 SEP event.

5.2. June 21, 2015 SEP event forecasting

The forecast of the June 21, 2015 SEP event is provided by using the ESPERTAmodel (Laurenza et al., 2009; Alberti et al., 2017a). The inputs of the modelare three solar parameters, i.e., the associated flare location, the 1 − 8A SXRintegrated intensity and ∼ 1 MHz Type III time-integrated intensity to give awarning for the occurrence an SEP event, within 10 minutes following the flaremaximum. The time-integrated SXR intensity is performed between the 1/3power point before the X-ray peak and the 1/3 power point after it, while, dueto the less regularity of the radio emission, the radio time-integration starts 10minutes before the time of the SXR integration until 10 minutes after the X-raypeak (see Laurenza et al. (2009); Alberti et al. (2017a) for more details).

Figure 22 shows the probability contours (solid lines) for SEP forecastingobtained by Laurenza et al. (2009); Alberti et al. (2017a) as function of thetime-integrated radio intensity at 1 MHz and the time-integrated X-ray flareintensity, for the flare longitude range E40 - W19. The dashed line represents athreshold for the occurrence of an SEP event: if the values of the associated flareparameters are located above the curve, an SEP event is predicted to occur; ifthey are under the curve, no SEP event is expected. The values obtained for theM2.6 flare (having longitude W00) associated with the June 21 SEP event are:0.16 J/m2 for the SXR fluence and 7.8 × 106 sfu × min for the ∼ 1 MHz TypeIII time-integrated intensity. It can be seen in Figure 20 that they are higher(see magenta asterisk) than the probability threshold. Hence, a positive forecastis issued at 02:46 UT (10 minutes after the SXR peak) for the June 21, 2015SEP event, with a leading time of ∼ 19 hours before the actual occurrence ofthe SEP event at 21:35 UT.


Time history of the cosmic ray intensity recorded at the Rome NM (SVIRCO Observatory) for June 2015. A Forbush decrease is observed, with a 5% of variation in the cosmic ray intensity.

SEP event forecastingThe forecast of the June 21, 2015 SEP event is provided by the ESPERTA (Empirical model for Solar Proton Events Real Time Alert) model (Laurenza et al., 2009; Alberti et al., 2017a), which allows to give a warning within 10 minutes following the flare maximum.

Inputs of the model:§  the flare location, §  the 1 − 8 Å SXR integrated intensity §  ∼ 1 MHz Type III time-integrated

intensity


104

105

106

107

108

0.30.3

0.60.6

0.90.9

10-3 10-2 10-1 1 10103

104

105

106

107

108

Integrated X-ray flare intensity (J/m2)

Integrated

radiofluxat1MHz(SFU

min)

E 40o - W 19o

Figure 22. Integrated 1 MHz radio intensity versus integrated 1-8 A soft X-ray intensityfor > M2 soft X-ray flares located in the longitude range E40 - W19: solid lines representthe probability contours; the dashed line is the probability threshold; the magenta asteriskcorresponds to the values obtained for the x-ray flare associated with the June 21 SEP event.

7. Magnetospheric Response

The impact of the the two magnetic clouds produces several effects on theMagnetosphere-Plasmasphere system by generating magnetic field variations,the destabilization of magnetospheric current systems, particle injection and pre-cipitation. These effects can be investigated by using different data sets relatedto in-situ measurements of fields and particles.

7.1. Geosynchronous analysis

Figure 24 shows the SW and the IMF observations by WIND (box a) and themagnetospheric field observations at geosynchronous orbit (box b) by GOES13(LT=UT-5) and GOES15 spacecrafts (LT=UT-9). The IP3 shock was observedby WIND on June 22, 2015, ∼18:07 UT, located at XSE ∼203.0 RE , YSE ∼-34.1RE , and ZSE ∼-11.0 RE ; it was characterized by remarkable variation of the SWpressure (∆PSW ∼31.5 nPa) and IMF strength (∆BIMF ∼22.3 nT), associatedwith a relevant increase of the southward IMF component (Bz,IMF ∼-20.0 nT),


Probability contours (solid lines) for SEP forecasting as function of the time-integrated radio intensity at 1 MHz and the time-integrated X-ray flare intensity, for the flare longitude range E40 - W19. If the values of the associated flare parameters are located above the dashed curve, a SEP event is predicted to occur.

The values for the M2.6 flare (having longitude W00) are indicated by the magenta asterisk, i.e., above the probability threshold. Hence, a positive forecast is issued at 02:46 UT (10 minutes after the SXR peak), with a leading time of ∼ 19 hours (the SEP event occurred at 21:35 UT).

Conclusions (Part 1)v  AR with a complex δ configuration and a sigmoidal-like structure along the PIL

v  Fractal analysis: eruptive potential of the AR, but lack of clear effects on Bphot

v  Log(R) à high probability of M-class flare occurrence

v  Velocity field map à shear accumulation

v  Shear angle, ∇Bz and Iz à energy storage (due to shear accumulation)

v  The evolution between the two M2.0 and M2.6 flares occurs without interruption (domino-like effect)

v  The M2.0 flare is located along the PIL and involves a filament eruption. After the flares, the δ complex achieves a more relaxed state.

v  Halo CME: slightly accelerating during the early expansion and then slightly decelerating. Reconstructed cloud of 3D points à surface of a cone with vertex located on the CME source region.

v  The EXPERTA model provides a SEP event forecasting ~19 hours before its occurrence.

Continues …..

geoeffective solar event of june 21, 2015: effects on the ...€¦ · 12371, a few minutes before...

Documents