Polarization of the Corona Observed During the 2017 and 2019 Total Solar Eclipses

Solar Science Observatory, NAOJ

　The white-light solar corona consists of the K-corona from the million-degree plasma of the Sun and the F-corona from the interplanetary dust. Linear-polarization information enables the separation of the K- and F-corona. Therefore, polarimetry has long been performed in total eclipse observations as well as in coronagraph observations. Total solar eclipses provide us very low sky-background down to just above the solar limb, which cannot be achieved in coronagraph observations. Therefore, the white-light corona has been a particularly important target for the total solar eclipse observations.

　We carried out polarimetric observations of the white-light corona during the total solar eclipses on 2017 August 21 and 2019 July 2 by taking advantage of professional-amateur collaborations, and successfully obtained data at two different sites for both eclipses. After eliminating sky-background, we obtained the brightness (B_K+F) and polarization (p_K+F) of the K+F corona, as presented in Figure 1 (polarization data are available here).

　In Figure 2, comparison of the derived degree of polarization with other measurement results for the 2017 eclipse are presented. The eclipse data by Vorobiev et al. (2020, Pub. Astron. Soc. Pac. 132, 024202) (green lines), show good coincidence with ours. However, the results of the Large Angle Spectrometric Coronagraph (LASCO) C2 of the Solar and Heliospheric Observatory (LASCO C2 Legacy Archive, http://idoc-lasco.ias.u-psud.fr/sitools/client-portal/doc/) (red lines) taken on the day of the eclipse are systematically smaller than ours and those by Vorobiev et al.

　Because the p_K+F represents the brightness of the K-corona, the discrepancy among the p_K+F values leads an error in the estimation of the amount of hot plasma. However, well-calibrated eclipse data, which were taken with a wide field-of-view, enable intercomparison among various data and contribute to the correction of the systematic error in the results from other observations. The eclipse observations provide a standard to study the amount of the hot corona quantitatively.

　Thus, derived correct amount of the hot plasma of the corona is important to study the coronal plasma-producing mechanism and the coronal variation according to the solar activity cycle.

　These results appeared as Hanaoka, Y., Sakai, Y., and Takahashi, K. “Polarization of the Corona Observed During the 2017 and 2019 Total Solar Eclipses” in Solar Physics (2021, 296, 158; doi: 10.1007/s11207-021-01907-0).

Fig. 1. Polarization map of the K+F corona covering 8.2 × 8.2 Rsun area obtained during the 2017 eclipse after the elimination of the sky background. The grayscale image represents the brightness of the K+F corona, and the degree and orientation of the linear polarization signals are depicted with orange ticks. The solar north is to the top.

Fig. 2. Comparison of the degree of polarization (pK+F) of the streamers around the equator (solid lines) and the poles (dashed lines) among the our eclipse observation (black), the results by Vorobiev et al. (green), and those from LASCO C2 observation (red).

November 24, 2021

Internetwork Magnetic Fields Seen in Fe I 1564.8 nm Full-Disk Images

Solar Science Observatory, NAOJ

　 The solar surface is filled with magnetic fields. Active regions and supergranulation network boundaries, which have strong magnetic fields, are their dominant components. However, magnetic fields in internetwork regions inside the networks are also another important component despite their weak field strength. So far many researchers have investigated the internetwork magnetic field mainly using large solar telescopes, but some of its properties have not been sufficiently understood yet.

　Then we studied the internetwork magnetic field using polarization data of the Fe I 1564.8 nm line, which were obtained with the spectropolarimeter of the Solar Flare Telescope at the NAOJ during 2010-2019. Contrary to most of the previous studies, we used full-disk data taken with a synoptic instrument. Therefore, our analysis [1] sheds light on the properties of the inter-network field derived from a quite different angle from those in the previous studies. The Fe I 1564.8 nm line shows particularly large Zeeman splitting, and it is suitable to study weak magnetic fields.

　Figure 1 shows the appearance of the circular polarization signals of the Fe I 1564.8 nm line, which show the distribution of the longitudinal magnetic field. In the left half showing the strong magnetic field (typically 1.1 kG = 110 mT), we can find white and black patches (representing positive and negative magnetic fields) corresponding to active regions and network boundaries. On the other hand, the right half, which presents the magnetic field less than 400 G (= 40 mT), entirely shows grainy appearance. This is the internetwork magnetic field; the small-scale, weak magnetic fields in the internetwork regions spread over the entire solar disk.

　Taking a closer look at the right half, we can find that the polarization signals increase from the disk center toward the limb. This means that the internetwork magnetic fields are considered to be highly inclined, contrary to the magnetic field of the network boundaries, which are mostly vertical to the solar surface; Figure 2 shows a schematic drawing of the magnetic field structure of the supergranulation networks. Although the majority of previous studies derived similar results, they have not been commonly accepted. Our analysis carried out from the quite different viewpoint from the previous ones supports the highly inclined field.

　Furthermore, by analyzing the data during 2010-2019 covering most of solar cycle 24, we found that the properties of internetwork fields do not show notable cycle variation, even though the period includes both the solar maximum and the solar minimum.

　To understand the solar magnetic field, it is required to make the properties of the internetwork magnetic field clear. Such a study contributes to the revealing of an aspect of the solar magnetic field, which is different from the strong magnetic fields, which cause the abrupt events like flares and coronal mass ejections.

[1] Hanaoka and Sakurai 2020, Astrophysical Journal, 904, 63, doi: 10.3847/1538-4357/abbc07

Figure 1. The two faces of the solar magnetic fields. This figure shows a full-disk circular polarization map of the Fe I 1564.8 nm line showing the longitudinal magnetic field taken on May 10, 2014. The solar north is to the top. Black and white represent positive and negative magnetic fields. Magnetic field signals corresponding to 1.1 kG and less than 400 G are shown in the left half and in the right half, respectively.

Figure 2. Schematic drawing of the magnetic field structure of super granulation networks. Small scale, randomly oriented weak horizontal magnetic fields spread over the internetwork.

November 25, 2020

The latest solar minimum came in December 2019 and
the Solar Cycle 25 has started

Solar Science Observatory, NAOJ

　 Sunspot data obtained by the NAOJ show that the sunspot number reached the minimum, which corresponds to the boundary of the Solar Cycles 24 and 25, most probably in December 2019. The Solar Cycle 25 has started. The solar activity, which varies cyclically, has been low for the past few years. Hereafter the Sun will become active year by year toward the solar maximum. Large solar flares and mass ejections, which are harmful to astronauts, satellite operations, and radio communications, will get more frequent with rising solar activity.

　Figure 1 shows the variation of the monthly average (black solid line) and the 13-month moving average (red solid line) of the sunspot number (the relative sunspot number) from January 2000 to August 2020. While the monthly average is not smooth, the 13-month moving average shows the long-term variation of the solar activity distinctly. The latest solar minimum in December 2019 which found in the 13-month moving average is indicated at the right-hand side of the figure with a dotted line. Our result is consistent with that derived by SILSO, the world data center for the international sunspot number at the Royal Observatory of Belgium [1].

　The transition from the Solar Cycle 24 to 25 was also confirmed with the magnetic properties of the sunspot groups. Figure 2 shows the monthly numbers of sunspot groups belonging to the Cycles 24 and 25 appeared between January 2018 and August 2020. The sunspot groups belonging to the Solar Cycle 25 have become the majority after November 2019, which is close to the minimum.

　How will the solar activity develop in the future? It is known that the period of solar activity tend to become longer during the grand solar minimum; the duration of the Solar Cycle 23 was longer than 12 years, and the activity of the Solar Cycle 24 was lower than that of the Solar Cycle 23. However, the Solar Cycle 24 lasted just 11 years. The possibility that the solar activity influences the climate has been discussed, and therefore, the solar activity in the new Solar Cycle 25 draws a particular attention.

[1] Sunspot Index and Long-term Solar Observations (SILSO), “Solar cycle minimum passed in December 2019”, posted on 21-Aug-2020.

Figure 1. Temporal variations in a monthly mean (black solid line) of and a 13-month moving average (red solid line) of relative sunspot numbers from January 2000 to August 2020. The vertical dotted lines indicate two minima of the 13-month moving average.


Figure 2. Monthly numbers of sunspot groups and active regions identified between January 2018 and August 2020. In addition to the numbered NOAA active regions, unnumbered sunspot groups found by the NAOJ are included.

September 15, 2020

Mysterious polarization signal confirmed in a H-alpha solar flare

　We performed statistical and event studies of linear polarization in the H-alpha (Hα) line during solar flares [1]. The statistical study revealed that, among 71 Hα flares analyzed, including 64 GOES flares, only one event shows significant linear polarization signals. Such an infrequent occurrence of significant linear polarization in solar flares is consistent with the result by Bianda et al. (2005), who studied 30 flares and found no polarization signals.

　Figure displays images of the Hα flare showing significant linear polarization signals. In the event showing the significant polarization, the maximum degree of linear polarization was 1.16 ± 0.06 percent, and the average direction of the polarization deviated by -142.5 ± 6.0 degrees from the solar north. The observed polarization degrees and the directions are consistent with the preceding reports. These strong linear polarization signals did not appear at major flare ribbons, nor did they correlate with either hard or soft X-ray emissions temporally or spatially. Instead they appeared at a minor flare kernel, which corresponds to one of the footpoints of a coronal loop. The active region caused coronal dimming after the soft X-ray peak.

　The observed flare shows no direct evidence that the linear polarization is produced by high energy particles, which are often considered to generate the polarization. On the other hand, our study suggests the possibility that coronal mass ejections, which have been often observed in flares showing linear polarization signals, play an important role for exciting linear polarization at Hα flare kernels.

[1] Kawate¹ and Hanaoka², 2019, The Astrophysical Journal, 872, 74, DOI:10.3847/1538-4357/aafe0f (¹Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency; ²National Astronomical Observatory of Japan, National Institutes of Natural Science)

Figure. Maps of Hα intensity, linear polarization (U/I), MDI continuum intensity, and EIT 195 Å intensity for the Hα flare showing significant linear polarization signals. In each map, contours exhibit Hα intensity levels. A green arrow in the U/I map indicates the Hα ribbon with a strong polarization signal. A blue arrow in the EIT image shows the direction of polarization, which corresponds to the tangential direction of the coronal loop structure at the footpoint. MDI and EIT data are provided by courtesy of SOHO consortium and SOHO is a project of international cooperation between ESA and NASA.

February 25, 2019

Solar Coronal Jets Extending to High Altitudes Observed During

the 2017 August 21 Total Eclipse

Solar Science Observatory, NAOJ

　Solar coronal jets have been extensively studied using soft X-ray and extreme-ultraviolet (EUV) data, and now they are understood as common phenomena in the low corona. However, from soft X-ray and EUV observations alone, it is difficult to know how high the jets extend. One reason is that there is a gap in the height coverage of the corona by the space instruments.

　 At the total solar eclipses, we can observe the corona from the limb to several solar radii under the very low sky background level. At the eclipse on 2017 August 21, we organized a multi-site observation program, and succeeded in taking a time-series of wide dynamic range images of the white-light corona at seven sites during a time period of about 70 minutes. Such observations enabled us to trace the time variation of the corona beyond the height coverage by the spaceborne instruments [1].

　In the eclipse data, we found coronal jets, which are seen as narrow structures upwardly ejected with the apparent speed of about 450 km/s in polar plumes. Six jets were found in the polar coronal hole regions, and their positions are shown in Figure 1. They extend from the solar surface to beyond two solar radii. All of the eclipse jets were preceded by EUV jets observed with the Atmospheric Image Assembly (AIA) of the Solar Dynamics Observatory (SDO) of NASA. Figure 2 shows an example of the jets observed in the EUV and the eclipse. Conversely, all the EUV jets in the polar regions with ordinary brightness, which occurred near the eclipse period, were observed as eclipse jets. These results suggest that ordinary polar jets generally reach high altitudes and escape from the Sun as part of the solar wind, as shown in Figure 3. (The EUV images were provided by courtesy of NASA/SDO and the AIA science team.)

[1] Hanaoka et al., 2018, Astrophysical Journal, 860, 142, DOI:10.3847/1538-4357/aac49b

Figure 1. White-light corona and positions of the eclipse jets found in the images taken during the solar eclipse on 2017 August 21. An EUV image at 211 Å taken with the AIA of the SDO is shown instead of the Moon. The solar north is to the top. The white-light image was processed to suppress the steep radial brightness gradient and to enhance the fine structures in the corona.

Figure 2. EUV images at 211 Å taken with the AIA of the SDO and eclipse white-light images before (left) and after (right) the occurrence of a jet (jet 5 in Figure 1). An enlargement of the EUV image in the box is shown at the upper-left corner for each panel. The jet is indicated by arrows. The white-light images were processed to suppress the steep radial brightness gradient and to enhance the jet.

Figure 3. A schematic drawing of the fast solar wind and a polar jet, both of which escape from the Sun. Part of them reaches to the Earth.

June 27, 2018

Magnetic Field Orientation in Solar Filaments Was Revealed

Solar Science Observatory, NAOJ

We have been carrying out full-Sun, full-Stokes spectropolarimetric observations using the He I 1083.0 nm line since 2010. The He I 1083.0 linear polarization signals of solar filaments show the magnetic field in filaments. Figure.1 shows a filament observed in Hα and in He I 1083.0. The linear polarization signals in panel (b) is approximately parallel to the fine structures seen in the Hα image, and they are almost aligned. We can define the average direction of the magnetic field, and it deviates clockwise with respect to the filament axis.

We have carried out a statistical study of the average orientation of the magnetic field in filaments with respect to their axes for 438 samples (Hanaoka & Sakurai 2017, Astrophysical Journal 851, 130). Figure.2 shows the relation between the latitude and the deviation of the average magnetic field of filament samples. This figure shows a clear hemispheric tendency; most filaments in the northern (southern) hemisphere show the clockwise (counterclockwise) deviation of the average magnetic field direction. The deviation angles of the magnetic field from the axes have the peak around 10-30 degrees (the light-green belts in Figure.2). Such a hemispheric tendency has been studied for the chirality of the fine structures in filaments and the magnetic field in prominences, and our results are consistent with them. Therefore, our results confirm the hemispheric tendency of the filament magnetic fields with the direct measurements of the magnetic field in filaments for the first time.

There is an interesting feature in the filaments which violate the hemispheric tendency; the alignment of the background magnetic field of such filaments in many cases opposite to that of active regions following the Hale-Nicholson law.

A filament is considered to be located at the bottom of a magnetic flux rope, which often erupts into the interplanetary space as a part of a coronal mass ejection (CME). Therefore, our observation of the filament magnetic field can be used to predict the magnetic field of CME flux ropes, which is the key to investigate their impact to the geomagnetic field.

Figure.1　A filament on 2014 November 23 (located in the northern hemisphere). The celestial north is to the top. (a) Hα image of the filament and (b) the linear polarization signals (>0.1 %) represented by red lines drawn on the photospheric magnetogram. At the lower-right corner, the approximate direction of the axis of the filament is shown with a solid line, and the average direction of the linear polarization signals is shown with a dashed line.（©NAOJ）

Figure.2　The left panel shows the relation between the latitude of the filaments and their average deviation angle. Colors and symbols show the characteristics of filaments, but the detailed description is omitted here. The right panel schematically shows the deviation of the magnetic field direction for each quadrant in the left panel.（©NAOJ）

Jan.15, 2018

Huge Sunspots and their Magnetic Structure observed by "Hinode"

Hinode Science Center / Naional Astronomical Observatory of Japan (NAOJ)
Solar Observatory / Naional Astronomical Observatory of Japan (NAOJ)
Institute of Space and Astronautical Science / Japan Aerospace Exploration Agency (ISAS/JAXA)

In the latter half of October, huge sunspots were observed on the surface of the Sun. These sunspots appeared at the east limb of the Sun on Oct. 16, and moved to the west as the Sun rotated (Fig. 1 and Fig. 2(a)). They rotated out of view after Oct. 30. On Oct. 26, the total area of these sunspots became almost 66 times larger than the Earth's cross section (*1). This was the largest sunspot area in this solar cycle, and the largest observed in the last 24 years (since Nov. 18, 1990). In the middle of November, these sunspots appeared again at the east limb, as the Sun's rotation brought them back into view (Fig. 2(b)).

White light images of the sunspots taken by the Solar Flare Telescope of the Solar Observatory/NAOJ from Oct. 18 to 28, 2014. (Only the huge sunspots are superimposed.)

Fig. 2(a) White light image of the sunspots on Oct. 24, 2014 taken by the Solar Flare Telescope of the Solar Observatory/NAOJ

Fig. 2(b) White light image of the sunspots on Nov. 15, 2014 taken by the Solar Flare Telescope of the Solar Observatory/NAOJ

Fig.3 and Fig.4 show the sunspots captured by the solar observing satellite "Hinode" on Oct. 24 and Nov. 15, respectively. In these figures, image (a) is a white light image, and image (b) is the magnetic field map (white = positive (N) pole, black = negative (S) pole).

Fig. 3(a) White light image of the sunspots on Oct. 24, 2014 taken by "Hinode"(NAOJ/JAXA) (Field of view) w: aprox.200,000 km × h: aprox.120,000 km	Fig. 4(a) White light image of the sunspots on Nov. 15, 2014 taken by "Hinode"(NAOJ/JAXA) (Field of view) w: aprox. 120,000 km × h: aprox. 120,000 km
Fig. 3(b) Magnetic field map of the sunspots on Oct. 24, 2014 taken by "Hinode"(NAOJ/JAXA) (Field of view) w: aprox. 200,000 km × h: aprox. 120,000 km	Fig. 4(b) Magnetic field map of the sunspots on Nov. 15, 2014 taken by "Hinode"(NAOJ/JAXA) (Field of view) w: aprox. 120,000 km × h: aprox. 120,000 km

Sunspots look darker because their temperature is lower than the surroundings. Strong magnetic fields in the sunspots lower the temperature, because the magnetic fields obstruct the convection that transports the heat generated in the center of the Sun to the solar surface. Strong magnetic fields sometimes cause solar "flares", huge explosions which occur in the solar atmosphere. Therefore, one of the reasons why "Hinode" accurately measures the magnetic fields on the solar surface is to understand the mechanism of solar flares.
In both Fig. 3(b) and Fig. 4(b), the right (preceding) sunspots have N poles and the left (following) sunspots have S poles. In Fig. 4(b), the left side of the preceding sunspots appears as S poles, and the left side of the following sunspots appears as N poles. This is just an artifact caused because we observed the sunspots obliquely while they were located close to the solar limb.
In Fig. 3(b), N poles and S poles are tangled. This type of sunspot structure often causes solar flares. Actually, huge flares occurred 6 times around the end of October. Although a few middle-class flares occurred on November 15 and 16, the magnetic structure doesn't look as complicated in November as it did in October. Will more flares occur? We will continue careful observations.

How do solar flares influence the Earth? Sometimes when flares occur, high-energy charged particles reach the Earth and magnetic storms occur. In October, the terrestrial environment was not disturbed much, although many flares occurred on the solar surface. The reason is under investigation; one possibility is that the magnetic field in the solar upper atmosphere was so strong that it suppressed the eruption of charged particles. In November, even if there are fewer large flares than in October, there could be flares which greatly influence the Earth. We should carefully watch the evolution of the sunspots.

(*1) the area when viewed from above

Nov. 27, 2014

Accurate Measurements of the Brightness of the White-Light Corona at the Total Solar Eclipses on 1 August 2008 and 22 July 2009

We have published the results from the 2008-2009 eclipses occurred near the deep solar minimum in "Solar Physics" Journal (Hanaoka et al. 2012, Solar Phys. 275, 79). The abstract of the paper is as follows.

We measured the brightness of the white light corona at the total solar eclipses on 1 August 2008 and 22 July 2009, when solar activity was at its lowest in one hundred years. After careful calibration, the brightness of the corona in both eclipses was evaluated to be approximately 0.4 x 10^-6 of the total brightness of the Sun, which is the lowest level ever observed. Furthermore, the total brightness of the K + F-corona beyond 3R_sun in both eclipses is lower than some of the previous measurements of the brightness of the F-corona only. Our accurate measurements of the coronal brightness provide not only the K-corona brightness during a period of very low solar activity but also a reliable upper limit of the brightness of the F-corona.

Fig.1 Images of the white-light corona taken in (left) the 1 August 2008 eclipse and (right) the 22 July 2009 eclipse.

Fig.2 Tangential coronal brightness distributions in the 2008 (dashed curves) and the 2009 (solid curves) eclipses from 1.1 to 4.5 R_sun. In panel (a), typical brightness values of the corona at the solar minima by Saito (1970) are shown with diamonds, and in panel (b), brightness values of the F-corona measured by Morgan and Habbal (2007), by Fainshtein, Tsivileva, and Kashapova (2010), and by Durst (1982) are by triangles, plus signs, and stars.

Sep. 14, 2012

Fig.1 Cyclic solar activity variation in the former half of the 20th century shown by Ca K spectroheliograms.

Fig.2 Photographic plates, on which the Ca K spectroheliograms are recorded.

Fig.3 The sidelostat (Left) and the spectroheliograph (Right), which have been working at the Mitaka Campus of the NAOJ (photos provided by the Archive Office, Public Relations Center).