Diagnostics of the Coronal Hole and the adjacent Quiet Sun by The Hinode/EUV Imaging Spectrometer (EIS)
aa r X i v : . [ a s t r o - ph . S R ] S e p Solar PhysicsDOI: 10.1007/ ••••• - ••• - ••• - •••• - • Diagnostics of the Coronal Hole and the adjacentQuiet Sun by
The Hinode/EUV Imaging Spectrometer (EIS)
P. Kayshap · D. Banerjee , · A.K. Srivastava c (cid:13) Springer ••••
Abstract
A comparison between a Coronal Hole (CH) and the adjacent Quiet-Sun (QS) has been performed using the spectroscopic diagnostics of
Hinode/ theEUV Imaging Spectrometer (EIS). Coronal funnels play an important role in theformation and propagation of the nascent fast solar wind. Applying Gaussianfitting procedures to the observed line profiles, Doppler velocity, intensity, linewidth (FWHM) and electron density have been estimated over CH and adjacentQS region of a North Polar Coronal Hole (NPCH). The aim of this study is toidentify the coronal funnels based on spectral signatures. Excess width regions(excess FWHM above a threshold level) have been identified in QS and CH. Theplasma flow inversion (average red-shifts changing to blue-shifts at a specificheight) in CH and excess width regions of QS take place at ∼ × K.Furthermore, high density concentration in excess width regions of QS providesan indication that these regions are the footprints of coronal funnels. We havealso found that non-thermal velocities of CH are higher in comparison to QSconfirming that the CHs are the source regions of the fast solar wind. Dopplerand non-thermal velocities as recorded by different temperature lines have beenalso compared with previously published results. As we go from lower to uppersolar atmosphere, down-flows are dominated in lower atmosphere while coronallines are dominated by up-flows with a maximum value of ∼ − inQS. Non-thermal velocity increases first but after Log T e = 5.47 it decreasesfurther in QS. This trend can be interpreted as a signature of the dissipationof Alfv´en waves, while increasing trend as reported earlier may attribute to Aryabhatta Research Institute of Observational Sciences(ARIES), Manora Peak, Nainital-263002, Indiaemail: [email protected] Indian Institute of Astrophysics (IIA), Bangalore-560034,Indiaemail: [email protected] CESSI, Indian Institute of Science Education andResearch, Kolkata, Mohanpur - 741252, Indiaemail: [email protected] Department of Physics, Indian Institute of Technology(Banaras Hindu University), Varanasi-221005email: [email protected]
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 1 ayshap et al. the signature of the growth of Alfv´en waves at lower heights. Predominance ofoccurrence of nano-flares around O vi formation temperature could also explainthe non-thermal velocity trend. Keywords:
Coronal Holes; Alfv´en waves; Spectral Line, Diagnostics
1. Introduction
Coronal Holes (CHs; Waldmeier 1975) are the regions on the Sun, which appeardark in comparison to the quiet-Sun (QS) because they emit less in UV & X-rays and maintained at a lower temperature than the surroundings. The differentmagnetic structures of QS and CH are responsible for their different appearancein coronal lines (Wiegelmann and Solanki, 2004). CHs are dominated by coronalfunnels, which originate from the adjoining locations of super-granule boundariesat the photosphere and expand abruptly towards higher atmosphere (Gabriel,1976; Axford and McKenzie, 1997; Marsch and Tu, 1997; Hackenberg, Mann,and Marsch, 1999). QS regions are dominated by closed magnetic filed lines.Although, QS has also the coronal funnels with less expansion. The filling factorsof these coronal funnels are less in QS compared to CHs. Coronal funnels playan important role in the formation as well as propagation of nascent fast solarwind in the CHs (Tu et al. , 2005; He, Tu, and Marsch, 2008; Tian et al. , 2010).Spectroscopic analysis is important to understand the dynamics of the QS aswell as CH. The Doppler velocity, which provides important information to anymodel of solar coronal heating, has a tendency to invert from pre-dominantred-shifts ( i.e. , downflows) to blue-shifts ( i.e. , upflows) as we go higher fromchromosphere to corona in CHs (Warren, Mariska, and Wilhelm, 1997; Hassler et al. , 1999; Peter and Judge, 1999; Peter, 1999; Xia, Marsch, and Curdt, 2003)as well as in QS (Chae, Yun, and Poland, 1998; Peter and Judge, 1999; Teriaca,Banerjee, and Doyle, 1999a; Dadashi, Teriaca, and Solanki, 2011). Several pos-sible mechanisms have been proposed to explain the observed red-shifts of thesolar transition region (TR) and coronal lines, e.g. , return of spicular material(Pneuman and Kopp, 1978; Athay, 1984), downward propagating acoustic waves(Hansteen, 1993; Hansteen, Maltby, and Malagoli, 1997; Teriaca, Banerjee, andDoyle, 1999a). The turnover temperature from red-shift to blue-shift is a crucialobservational parameter for future modelling.Full-width-at-half-maximum (FWHM) is also an important parameter to un-derstand the dynamics of QS and CH. Some observed solar Extreme-ultraviolet(EUV) and Far-ltraviolet (FUV) spectral line profiles are found to be broaderthan those expected from thermal broadening (Boland et al. , 1975; Doschek etal. , 1976; Mariska, Feldman, and Doschek, 1978; Banerjee et al. , 1998). These ex-cess widths, after subtracting thermal and instrumental widths from the observedwidths, are called as non-thermal widths. Non-thermal velocities as derived fromthese widths reveal the presence of non-thermal motions/unresolved flows orthe presence of waves in the solar atmosphere. Early studies show that thenon-thermal velocities increase within a narrow range of temperature (fromchromosphere up to TR temperatures) in the QS near disk centre (Doschek et SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 2 omparison Between Coronal Holes and Quiet-Sun al. , 1976; Mariska, Feldman, and Doschek, 1978; Dere and Mason, 1993)). Afterthe launch of Solar Ultraviolet Measurements of Emitted Radiation (SUMER)on board SoHO, the non-thermal velocity has been investigated for a quite largerange of solar atmosphere ( i.e. , 10 K to 2 × K). It is found that initiallynon-thermal velocity increases but after a certain temperature decreases corre-sponding to coronal heights in QS as well as active regions (Chae, Sch¨uhle, andLemaire, 1998; Teriaca, Banerjee, and Doyle, 1999a; Doschek and Feldman, 2000;Peter, 2001; Patsourakos and Klimchuk, 2006; Doschek et al. , 2007)). Severalmechanisms have been proposed to interpret the pattern of non-thermal velocitythrough the solar atmosphere, e.g. , MHD wave models or magnetic reconnectiongenerated turbulence (Doschek et al. , 1976; Mariska, Feldman, and Doschek,1978), unresolved laminar flows, waves and turbulent flows (Chae, Yun, andPoland, 1998), Alfv´en waves (Peter, 2001), etc . Recently, Coyner and Davila(2011) have shown that the non-thermal velocities represent strong distributionfrom 19 km s − to 22 km s − in the various parts of the solar atmosphere.In the present study, we have performed a comparison between QS and CHson the basis of various spectral parameters ( e.g. , Doppler velocity, non-thermalwidth, etc .). On the basis of the variations of these parameters and densitycontrast, we have located the most probable regions of coronal funnels in QS.Variation of the non-thermal velocity and average Doppler velocity with tem-perature have also been investigated in QS by including some previous relevantresults with our estimated values. The present work is organized as follows.In Section 2, we describe the details about the observation and data reduction.Results related to the comparison between QS and CH as well as identification ofthe coronal funnels are presented in Section 3. Section 4 outlines the temperaturedependent behaviour of non-thermal velocity in QS and CH. Discussion andConclusions are presented in the last section.
2. Observations and Data Reduction
Extreme Ultraviolet Imaging Spectrometer (EIS) on board
Hinode spacecraft is anormal-incidence EUV spectrometer. EIS has spatial resolution of ∼ i.e. , 170 - 211 ˚A) and long wavelength band( i.e. , 246 - 292 ˚A). Four types of slits/slots ( e.g. ,1-,2-,40-,256-arcsecond) areavailable in the EIS observations (Culhane et al. , 2007). A North Polar CoronalHole (NPCH) observation captured on 10 October 2007 by Hinode /EIS, hasbeen used in the present work. The observed region (FOV) is marked as a whiterectangular box on the SoHO/EIT 195 ˚A image ( cf. , Figure 1). NPCH wasobserved using the 2-arcsecond slit in the raster mode for more than four hours( i.e. , from 14:03 UT to 18:17 UT) by capturing 101 exposures with 155 s exposuretime on each scanning step. This very long exposure ( i.e. , 155 second) enhancesthe count statics, therefore, this data set is reliable for deriving various lineparameters (e.g., Doppler velocity, FWHM, etc.) from the line profiles and alsoideal for density diagnostics. Standard Hinode/EIS routine eis prep.pro, whichis available in SolarSoft (SSW) package, has been used for the calibration of raw
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 3 ayshap et al.
Figure 1.
White rectangular box marks the FOV of the observed region of
Hinode /EIS,over-plotted on the SoHO/EIT 195 ˚A image. This North Polar Coronal Hole (NPCH) wasobserved by
Hinode /EIS from 14:13 UT to 18:17 UT on 10 October 2007.
Table 1.
This table shows all the four lines, which areused in the present analysis. Formation temperaturesas well as the standard wavelengths are also listed.Sr. No. Ion Wavelength (˚A) Log T (K)1 Si VII 275.35 5.82 Fe X 184.54 6.03 Fe XII 195.12 6.14 Fe XIII 202.04 6.2 data. This routine converts original raw data into physical data after processingvarious necessary steps, e.g. , subtraction of the dark current, removal of cosmicrays and hot pixels and radiometric calibration, etc.
Four spectral lines havebeen used in the present analysis as listed in Table 1. Si vii x xiii et al. , 2007), while Fe xii xii et al. , 2009). Although, Young et al. (2009) haveshown that Fe xii xii e.g. , slit tilt, thermal variation, EIS CCD offset) must be addressedbefore the Gaussian fitting to derive the basis parameters from the observed lineprofiles. Hinode /EIS has single grating that disperse the light onto two CCDs
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 4 omparison Between Coronal Holes and Quiet-Sun
Figure 2.
Two samples of Gaussian fitting (shown by blue solid line) in Si vii x xii xiii σ error is also overplotted by red color in each sample fit. SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 5 ayshap et al. and spatial offset is present between these two CCDs of the
Hinode /EIS. Wehave used standard
Hinode /EIS routine, eis ccd offset.pro, to find out the pixeloffset and direction of the offset ( i.e. , up or down) in all lines relative to theHe ii et al. (2010) have used in his analysis. After calculating slit tilt byusing eis slit tilt.pro standard Hinode /EIS routine, we have built a wavelengthcorrection offset array ( i.e. , for wavelength correction) for all four lines. Thesewavelength offset arrays have the combined effect of slit tilt as well as thermaldrift of respective spectral line. We have fitted a single Gaussian function on thethree line profiles ( i.e. , Si vii , Fe x and Fe xiii ). Double Gaussian has been fittedon the Fe xii line profile to separate the Fe xii xii i.e. , in case of Fe xii et al. (2009). Figure 2 represents the two samplesof Gaussian fit in each line profile at two different locations. From the Gaus-sian fitting, we have extracted the basic parameters (i.e., intensity, Gaussiansigma, line centeroid, etc.) at each location (pixels) of the commonly observedregion. Rest wavelength measurement is difficult task as the absolute wavelengthcalibration is not available for Hinode /EIS. Unfortunately, any low tempera-ture chromospheric line is not present in our wavelength window, therefore,in the present case we have used the Limb method (Peter and Judge, 1999)for measuring the rest wavelengths. After measuring the rest wavelengths ineach line profile, we have estimated the Doppler velocity at each location ofthe observed region. Similarly, we have estimated the FWHM from Gaussiansigma ( i.e. , FWHM=2.3548 σ ) over commonly observed region. The intensity,Doppler velocity and FWHM maps are shown in Figure 3 corresponding to allthe four lines. Apart from the Doppler velocity, intensity and FWHM, electrondensity can also be estimated with the help of intensity ratio of two densitysensitive lines and CHIANTI database (Dere et al. , 1997; Landi et al. , 2006).In the present analysis, we have used Fe xii xii et al. (2009) have shown that both density sensitive Fe xii lines sufferfrom the blending problem ( i.e. , Fe xii xi xii xii xi xii xi xi et al. (2009) haveshown that S xi xii inthe regime of moderate densities while S xi xii xi transitions is not present in the presentdata-set, therefore, we have assumed that S xi SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 6 omparison Between Coronal Holes and Quiet-Sun
Figure 3.
Intensity (top-row), Doppler velocity (middle-row) and FWHM (bottom-row) mapscorresponding to the NPCH (as marked by a rectangular box in Figure 1) as observed by fourlines, Si vii x xii xiii SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 7 ayshap et al. Fe xii xi xii xii xi et al. , 2009). To remove the blending fromFe xii et al. (2009). After calculating the intensity ratio from these two densitysensitive lines, we have derived the density over the observed common region( cf. , left panel of Figure 7). The density map (Figure 7) also represents theoverplotted small boxes, which we have used for the present work and will bedescribed in the upcoming section. We concentrated primarily on three regionsin this work, which is shown by three dashed black boxes on the Doppler velocitymap (Figure 4). These three regions represent quiet-Sun (QS; lower box), quiet-Sun with coronal hole boundary (QSCH; middle box) and coronal hole (CH;upper box) respectively. Measurement of the various plasma parameters havebeen evaluated for various localised regions that correspond to the three broadlyclassified physical regions as QS, QSCH and CH.
3. Comparison Between QS and CH Parameters
Ion temperature and unresolved non-thermal motions and/or presence of MHDwaves are responsible for the spectral broadening of optically thin coronal emis-sion lines in the solar atmosphere. Generally, it is assumed that ions and electronsare in the thermal equilibrium, although the temperature of the ions and elec-trons may be different from each other, particularly at extended part of thecorona. But for the inner corona, we have assumed that ion temperature isidentical to the electron temperature. FWHM can be expressed as follows,
F W HM = (cid:20) W inst + 4 ln 2 (cid:18) λc (cid:19) (cid:18) kT i M i + ξ (cid:19)(cid:21) / , (1)where, T i , M i , ξ and W inst are the ion-temperature, ion-mass, non-thermal veloc-ity, and instrumental-width respectively. Instrumental, thermal and non-thermalwidths are the three components of the observed FWHM. Instrumental as wellas thermal ( i.e. , at a particular ion formation temperature) width componentsare constant, therefore, the variations in the FWHM arise due to the variationin the non-thermal width. The non-thermal width provides signature of variousunresolved small-scale dynamics in the solar atmosphere, which do not normallyrecorded by imaging instruments. Figure 4 shows the Doppler velocity map inFe xii SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 8 omparison Between Coronal Holes and Quiet-Sun
Figure 4.
The image shows Doppler velocity map of the observed region, which is overlaidby intensity contours as well as excess width locations (marked by yellow plus symbols). Allselected regions have been over-plotted on the Doppler velocity map by red rectangular boxes. values in the neighbouring regions may allow us to locate these locations wherevarious transients/explosive events are occurring. Furthermore, we have searchedthe excess width locations in all three regions by setting different threshold levelin each box. Widths above the threshold levels have been considered as theexcess widths. The average widths as well as the threshold levels in all the threeregions ( i.e. , QS, QSCH and CH) have been listed in Table 2. After finding theexcess width locations, we have marked these locations over the Doppler velocitymap by yellow plus signs ( cf. , Figure 4). Finally, we have selected eight differentboxes of different sizes in these three regions. Some boxes, out of these eightboxes, correspond to the excess width regions while rest of the boxes are locatedin the average width regions. All the eight selected boxes are overplotted on thesame Doppler velocity map ( cf. , Figure 4) by dark red color and are numbered.The description of all the selected regions ( i.e. , three big boxes as well as eightsmall boxes) is given in Table 3. We have chosen these small eight regions to
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 9 ayshap et al.
Table 2.
This table shows the average observed widths of all three region (i.e., QS,QSCH and CH). Similarly, the adopted threshold level of widths in each box to locatethe excess widths in these regions have also been shown in this table.Sr. No. Big Black Dashed Box Average Width (m˚A) Threshold Width (m˚A)1 QS 33.65 1.20 × Average Width2 QSCH 33.47 1.30 × Average Width3 CH 39.42 1.35 × Average Width compare various parameters corresponding to the excess width locations withthe locations having average width below the threshold values. On the basis ofthis comparison, we have performed a comparison between QS and CH as wellas tried to locate the coronal funnels in QS, QSCH and CH.
Table 3.
The table gives the details about the all selected boxes,which is used in the present analysis.Description of different selected regionsBig Black Dashed Boxes Small Boxes DescriptionQS (Lower Box) Box 1 Excess WidthBox 2 Excess WidthBox 3 Average WidthQSCH (Middle Box) Box4 Excess WidthBox 5 Average WidthBox 6 Excess WidthCoronal Hole (Upper Box) Box 7 Excess WidthBox 8 Average Width i.e. , up-flows/down-flows) in the solar atmosphere. Thelines, which we have used in the present study, cover the TR (Si vii e = 5.8 K) up to the solar corona (Fe xiii e = 6.2 K),corresponding to the QS (upper left, Figure 5), QSCH (upper right, Figure 5)and CH (bottom left, Figure 5). The overall trend is shown in the right panelof Figure 5. The Doppler velocities corresponding to the observed excess widthregions of QS ( i.e. , Box 1 and 2) are ∼ ± ∼ -2.44 ± − at Si vii formation temperature, therefore, these excess width regions showmarginal up-flows in upper TR. The plasma up-flows in these excess widthregions increases with temperature up to the corona, which have maximumDoppler velocity of ∼ -9.42 ± − and ∼ -10.07 ± − in both SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 10 omparison Between Coronal Holes and Quiet-Sun
Figure 5.
Doppler velocity variations with temperature corresponding to different regions QS(top-left), QSCH (top-right) and CH (bottom-left). The overall trend is shown in bottom-rightpanel. excess width regions ( i.e. , Box 1 and 2) at the temperature Log T e = 6.2 K ( cf. ,Figure 5). The Doppler velocity of Box 3, which is located in the average widthregion of QS, is ∼ -3.24 ± − at Si vii ion formation temperature, whichincreases further in the solar atmosphere up to Fe xii formation temperaturewith the blue-shift value ∼ -8.05 ± − . However, at the Fe xiii i.e. , -7.7 ± − ). The reduction in blue-shift at Fe xiii xii cf. , Table 3). Box 4, which is located at the boundary of CH, shows down-flows( i.e. , red-shifts) at TR and as the temperature increases the red-shift invertsinto blue-shift with the maximum value of ∼ -18.93 ± − at Log T e =6.2 K. On contrary, the Doppler velocity of Box 6, which is also an excess widthregion, does not change very much from Si vii ± − ) to Fe xiii ± − ). Therefore, the Doppler velocityis almost constant in this box from TR up to the solar corona. The Doppler SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 11 ayshap et al. velocity of Box 5, which is an average width QS region near CH boundary, isweakly red-shifted (0.13 ± − ) at Log T e = 5.8 K. After that the up-flowspeed increases with temperature with a maximum value of -10.43 ± − at Log T e = 6.2 K. Therefore, it is very difficult to make distinction betweenthe excess and average width regions on the basis of the net Doppler velocitiesin the QSCH regions.Excess width (Box 7) and average width (Box 8) regions of CH ( cf ., Table 3)have the Doppler velocities of -0.92 ± − and -2.93 ± − inSi vii at Log e = 5.8, respectively. The blue-shifts increase with temperature inboth the regions and the maximum values of blue shifts are -13.30 ± − and -12.29 ± − in these excess as well as average width boxes atLog T e = 6.2 K. Again, it is very difficult to distinguish excess width location(Box 7) from average width location (Box 8) because both types of regionsshow similar pattern of the Doppler velocity. One can conjecture that it may bethat corresponding to CHs the funnels have expanded so much at these heights,therefore, one can not distinguish a funnel and inter funnel regions.The overall comparison of Doppler velocity of each emission line in all theeight boxes (Bottom right panel of Figure 5) shows that coolest line( i.e. , Si vii e =5.8 K) is slightly blue-shifted in the QS, QSCH and CH.Although, the Box 5, which is located at the boundary of CH and QS (seeFigure 4), shows larger red-shift at this temperature. The higher temperaturelines ( i.e. , Fe x xii xiii et al. , 1999) to calculate the accurate rest wavelengths of the usedEIS lines. In the limb method, it is assumed that the average Doppler velocityis almost zero around the solar limb ( i.e. , motions along the line of sight cancelout on average in an optically thin plasma). For the 4 th order polynomial fit(solid black line) form chromosphere to the corona, we have include results fromTeriaca, Banerjee, and Doyle 1999a.3.2. Electron Density VariationsThe best available density sensitive Fe xii line pair has been used for the densitymeasurement (Section 2). The left panel of Figure 7 represents the density map of SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 12 omparison Between Coronal Holes and Quiet-Sun
Figure 6.
Average Doppler shift in the QS at disk center corresponding to various TR andCH ions as measured by SUMER and EIS spectra and as reported by earlier works as labeled.Our estimated Doppler velocity results of QS are marked by red-star symbols. Positive valuesindicate red-shifts (down-flows), while negative values indicate blue-shifts (up-flows). The solidline represents a fourth order polynomial fit to the data. the observed region while the right panel shows averaged density correspondingto the three QS boxes, three QSCH boxes and two CH boxes (for box description;see Section 3). The density decreases progressively from QS to CH ( cf. , rightpanel; Figure 7), which is expected. In the QS region, the electron densities ofthe excess width boxes (i.e., Box 1 and 2) are Log N e = 8.63 cm − and Log N e = 8.60 cm − respectively, while the electron density of the average width box( i.e. , Box 3) is Log N e = 8.47 cm − . Therefore, the excess width boxes havethe higher electron densities in comparison to the density of average width boxwithin QS.In QSCH region ( cf. , QS+CH dashed box of right panel of Figure 7), theelectron density of the excess width boxes (i.e., Box 4 and 6) are Log N e = 8.30cm − and Log N e = 8.33 cm − respectively, while the electron density of averagewidth box ( i.e. , Box 5) is Log N e = 8.45 cm − . The electron density of averagewidth box is high in comparison to the excess width boxes because the averagewidth box is located in the quiet-Sun while the excess width boxes are locatedin the Qs-CH and CH regions. Coronal hole boundaries are the most probablelocations of various explosive events (Madjarska, Doyle, and van Driel-Gesztelyi,2004; Madjarska and Wiegelmann, 2009; Madjarska et al. , 2012), therefore, weexpect that the excess width as well as high density locations lie along the coronalhole boundaries. Excess widths are present along the coronal hole boundary( cf. , yellow plus signs in Figure4). Although, the density of Box 4 that covers SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 13 ayshap et al.
Figure 7.
The electron density map of the observed (left panel). The small selected regionsare overplotted on the electron density map by white rectangular boxes. The averaged electrondensities of all boxes are shown in the right panel. the coronal hole boundary along with large portion of the coronal hole, doesnot represent high density compared to the density of the box completely lyingwithin coronal hole ( i.e. , Box 6). Presence of the higher density (left panel ofFigure 7) as well as maximum number of excess width pixels within Box 6 areto be noted. In CH, the density of the excess (Box 7) and average (Box 8) widthlocations are Log N e = 8.50 cm − and Log N e = 8.34 cm − respectively. Thehigher electron density corresponding to the excess width box (Box 7) may beinfluenced by the presence of a bright point (see left panel of Figure 7). Therefore,if we neglect the presence of the high density regions from the Box 5 and Box7, then we can assume that the electron density of all excess and average widthboxes located in coronal hole ( i.e. , Box 4, 6, 7 and 8) are almost same. On thebasis of the electron density, it is not possible to make distinction between excessand average width regions of CH and QSCH.
4. Temperature Dependent Behaviour of Non-Thermal Velocity inQS
Non-thermal velocities can provide information about the unresolved motions/MHDwaves in the solar atmosphere. Instrumental, thermal and non-thermal widthsare the three component of the observed line width. Therefore, to get non-thermal width, the removal of the instrumental and thermal widths from the
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 14 omparison Between Coronal Holes and Quiet-Sun line width has been performed by using following formula, W nt = [ W obs − W inst − W th ] / , (2)where, W nt , W obs , W inst and W nt are non-thermal, observed, instrumental andthermal widths respectively. The on-average instrumental width (i.e., W inst ) is ∼
66 m˚A while the thermal width can be calculated using the relation, W th = 4 ln 2 (cid:18) λc (cid:19) (cid:20) kT i M i (cid:21) . (3)After estimating non-thermal widths, we have calculated the non-thermal veloc-ities of QS and CH for Si vii xii xiii ξ = (cid:18) W nt c λ (cid:19) / (4)QS region non-thermal velocities of Si vii , Fe xii and Fe xiii lines have been Table 4.
Quiet Sun Non-thermal Velocity ξ (km s − )Box number Si VII Fe XII Fe XIII1 23.64 ± ± ± ± ± ± ± ± ± listed in Table 4 while the CH non-thermal velocities have been listed in theTable 5. The non-thermal velocities follow the same pattern in all selected boxes( i.e. , located in QS and CH regions) within the narrow temperature range (fromSi vii up to Fe xiii ion temperature) in the solar atmosphere. The non-thermal Table 5.
Coronal Hole Non-thermal Velocity ξ (km s − )Box number Si VII Fe XII Fe XIII7 33.28 ± ± ± ± ± ± velocity increase from Si vii up to the Fe xii and after that it decreases towardsFe xiii ion in the all selected boxes. We have included previous results as reportedin Doschek et al. (1976); Mariska, Feldman, and Doschek (1978); Dere and Mason(1993); Chae, Yun, and Poland (1998). This plot covers a large portion of thesolar atmosphere and allows us to compare our results with previous numbers.It also provides a new unified physical scenario in the solar atmosphere whichshows how the non-thermal velocity behave between 4.0 K ≤ Log T e ≥ SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 15 ayshap et al.
Figure 8.
Variation of the non-thermal velocities with temperature corresponding to the QS.To cover a wide range of temperatures, we have included previous results as labeled. Usingnon-thermal velocities reported by Teriaca, Banerjee, and Doyle 1999a and derived from Box3, we have fitted a 3 rd order polynomial as shown by solid black line. Our estimated resultsof QS and CH boxes are plotted to the right side of the red vertical dashed line. K, i.e. , covering from chromosphere, TR, corona above QS. The behaviour ofnon-thermal velocity with temperature in all selected QS ( i.e. , 1 st , 2 nd and 3 rd ) boxes along with previous results has been shown in Figure 8. We have alsoplotted the non-thermal velocities of CH boxes ( i.e. , 7 th and 8 th boxes) in thesame plot. We found that non-thermal velocities first rise up, attains a maximumvalue, and thereafter decreases further in the solar atmosphere ( i.e., quiet-Sunhere). A 3 rd order polynomial, using our QS non-thermal velocity ( i.e. , Box 3)and non-thermal velocity by Teriaca, Banerjee, and Doyle (1999a), has beenfitted. This shows that the non-thermal velocity peaks at Log T e ∼
5. Discussion and Conclusions
In the present work, we have studied the polar coronal hole and adjacent quiet-Sun, as observed on 10 October 2007 by
Hinode /EIS. Using
Hinode /EIS spectra,a comparison between QS and CH has been performed. In the present study, wefound that on average the line width of CH (39.42m ˚A) is higher in comparisonto the line width of QS (33.65 m˚A, cf. , Table 2). Even, after setting a higherthreshold limit of the line widths in CH ( i.e. , 1.35 × CH average line width) incomparison to the QS threshold line width limit ( i.e. , 1.20 × QS average width)to locate excess width locations in CH and QS, we found that CH has larger
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 16 omparison Between Coronal Holes and Quiet-Sun number of excess width locations in comparison to the excess width locationsin QS (see Figure 4). We assumed that excess width locations may correspondto regions where various type of transients/explosive events may occur in solaratmosphere. Variations in the line width is directly related to the variations ofthe non-thermal width component (Section 3), therefore, the CH has higher non-thermal width as well as large number of excess non-thermal width locations incomparison to the QS. As we know that CHs are the source regions of the fastsolar wind (Krieger, Timothy, and Roelof, 1973; Tu et al. , 2005; Kohl et al. , 2006;He, Tu, and Marsch, 2008; Tian et al. , 2010) as well as the sites for oscillationand various type of MHD waves (Banerjee et al. , 2001a,b; O’Shea, Banerjee,and Doyle, 2006, 2007; Banerjee, P´erez-Su´arez, and Doyle, 2009; Gupta et al. ,2010; Krishna Prasad, Banerjee, and Gupta, 2011). Therefore, the presence ofthe larger non-thermal width and large number of excess width locations inthe CH in comparison to the QS is quite obvious due to the presence of theseflows/waves in the CH.It is now well established that the lower temperature lines are dominated bydown-flows while higher temperature lines are dominated by blue-shifts in QSand CH (Chae, Yun, and Poland, 1998; Peter and Judge, 1999; Teriaca, Banerjee,and Doyle, 1999a; Tian et al. , 2008a,c; Hansteen et al. , 2010; Tian et al. , 2010;Dadashi, Teriaca, and Solanki, 2011). Therefore, there is transition from red-shift to blue-shift at some temperature, which is an important observationalparameter from the point of view of the TR models. Excess width regions ofQS are slightly blue-shifted at Log T e = 5.8 K ( i.e. ,Si vii e = 5.8. On the basis of observed blue-shift (-1.9 ± − ) of theNe viii vii line, Peter and Judge (1999) have already beeninferred that such transition occur at Log T e = 5.7 K in QS. Therefore, we canalso conclude that transition ( i.e. , from red-shift to blue-shift) also takes placeat the same temperature in these excess width regions of the QS. Similar toexcess width regions of QS, the average and excess width regions ( i.e. , box, 6,7 and 8) of CH also show marginal blue-shifts at Log T e = 5.8 K. Therefore,we can infer that transition temperature in CH is similar to the QS. Although,one CH box ( i.e. , Box 4), which is located at the coronal hole boundary, showsstrong red-shift at Log T e = 5.8 K. This region might have been influenced bylarge number of explosive events. The properties of the excess width regionsof the QS are similar to CH ( i.e. , excess and average width regions), but, thedensities of the excess width regions of QS are higher in comparison to the CH.In contrast, one average width box (Box 3) of QS does not show the continuousplasma up-flows while another average width box (Box 5) shows continuousplasma up-flows. Similarly, Box 3 and 5 show strong blue-shift (-3.24 km s − )and weak red-shift (+0.13 km s − ) at Log T e = 5.8, therefore, we can not deducea common transition temperature for these average width regions of QS. As weknow that the coronal funnels as well as small and large magnetic loops are thebasic building blocks of the solar atmosphere. The upper TR line blue-shiftspatches are associated with coronal funnels (Tu et al. , 2005), although, theseblue-shift patches may be associated with the magnetic loops as well. As we SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 17 ayshap et al. have shown earlier that QS excess width boxes show continuous plasma up-flowswhile QS average width region does not represent the continuous plasma upflows. Similarly, the averaged densities of QS excess width regions are higher incomparison to the average width region of QS, although, the density differenceis not very much. We should note that we are measuring the electron densityon the basis of the intensity ratio of Fe xii line, which represents the electrondensity at coronal heights. At the coronal heights, due the large expansion of thecoronal funnels, the electron density in the coronal funnels may not be differentsignificantly from the surrounding region. Therefore, slightly higher densities aswell as presence of continuous plasma up-flows in the excess width regions ofQS provide sufficient signature to identify that these excess width locations inQS are associated with the coronal funnels. On the basis of Doppler velocity,excess width and density, we can not distinguish the excess width boxes fromthe average width boxes in QSCH and CH. As we know that QS are mostlydominated by coronal loops and only a small fraction of the QS area is occupiedby coronal funnels while the CH are mostly dominated by coronal funnels (Peter,2001). Therefore, we have easily located the coronal funnels in QS but it is veryhard to isolate the footprints of the coronal funnels within coronal holes. It maybe possible that all selected boxes in CH ( i.e. , excess and average) are coronalfunnel occupied regions because of large filling factor. We can propose that theplasma flow inversion ( i.e. , inversion from down-flows to up flows) takes place atLog T e = 5.7 K in CH. Similarly, the inversion of plasma flow also takes place atthe same temperature (Log T e = 5.7 K) in excess width regions of QS. However,the average width regions of the QS do not show similar pattern of the Dopplervelocity, therefore, it is hard to get conclusive plasma flow inversion temperaturein these average width regions of QS. Finally, we found that excess width regionsof QS show continuous plasma up-flows as well as high density concentrationwhile the average width region of QS does not follow continuous plasma flowpattern and the density of the average width region is lower in comparison tothe excess width regions of QS. Therefore, the excess width regions are the mostprobable locations of coronal funnels in QS, however, we could not locate thefootprints of coronal funnels in the QSCH and CH using the same procedure.As we know that the coronal holes are dominated by open magnetic filed lines,therefore, it may be possible that all the selected regions of QSCH and CH(except Box 5) at these heights are within coronal funnels because in CH theparameters are similar within funnel and the adjacent background region.Apart from this, Doppler shift of TR and coronal lines has been also in-vestigated while combining our results with previously reported numbers fromSUMER ( cf. , Figure 6). This new plot now covers a wide range of temperaturefrom chromosphere up to high corona. Doppler velocity inversion from red-shiftto blue-shift takes place around Log T e = 5.7 K (see the peak of 4 th orderpolynomial fit in Figure 6), which is in good agreement with earlier results (Peterand Judge, 1999; Teriaca, Banerjee, and Doyle, 1999a). Recently Fu et al . (2014)have measured the Doppler velocity variations with temperature in the sameNPCH area, however, they have focused on the on-disk plume structures. Wefeel that these plume structures could be also used as tracers of coronal funnels.Some of their results complement results presented here. Our focus has been SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 18 omparison Between Coronal Holes and Quiet-Sun on the line widths rather than the outflow velocities alone. Earlier, 1-D modelswhich assumes energy release in magnetic loops had successfully managed toreproduce the observed red-shift at TR temperatures and blue-shifts at coronaltemperatures (Teriaca et al. , 1999b; Spadaro et al. , 2006). However, these modelsdo not explicitly consider the heating mechanisms. More recently, Srivastava et al . (2014) presented numerical model in 2-D realistic solar atmosphere, thegeneration of the blue-shifts (out-flows) due to heating pulses, as observed byHinode/EIS at coronal temperature. Peter, Gudiksen, and Nordlund (2004, 2006)presented, 3D models assuming that the coronal heating is caused by jouledissipation of currents produced by stressing and braiding of the magnetic fieldsproduce red-shifts at all temperatures. Episodic injection of emerging magneticflux, which reconnects with the existing field, produces rapid, episodic heatingof the upper chromospheric plasma to coronal temperatures (Hansteen et al. ,2010). Dadashi, Teriaca, and Solanki (2011) measured the averaged Dopplershifts from 0.01 up to 2.1 MK and they have shown that 3D coronal models aremore appropriate for the explanation of the observed Doppler velocity patternfrom chromosphere up to the corona.The inhomogeneous solar plasma should support different MHD modes aswell as acoustic wave and this has been confirmed by observations and the-ory (Ofman and Davila, 1997; DeForest and Gurman, 1998; Banerjee et al. ,2001a,b; O’Shea, Banerjee, and Doyle, 2006; Dwivedi and Srivastava, 2006;Srivastava and Dwivedi, 2007; O’Shea, Banerjee, and Doyle, 2007; Banerjee,P´erez-Su´arez, and Doyle, 2009; Gupta et al. , 2010; Krishna Prasad, Banerjee,and Gupta, 2011; Chmielewski et al. , 2013). It has been proposed that severalmechanisms are responsible for the non-thermal broadening, e.g. , propagationof waves (Mariska, Feldman, and Doschek, 1978; Banerjee, P´erez-Su´arez, andDoyle, 2009), non-thermal motions (Doschek et al. , 1976; Athay and Dere, 1991;Chae, Yun, and Poland, 1998), MHD turbulence (Gomez and Ferro Fontan,1988, 1992; Heyvaerts and Priest, 1992) and nano-flare heating (Patsourakosand Klimchuk, 2006). In-spite of all these physical processes, the nature ofthe non-thermal broadening is not fully understood. We confirm here a morecomplete picture of the non-thermal velocity in the wide temperature rangefrom ∼ × K to 1.58 × (Figure 8). Initially, the non-thermal velocityincreases with temperature but after certain TR temperature (Log T e =5.8 K)non-thermal velocity decreases further upto inner corona (Log T e =6.2 K). Un-damped Alfv´en waves are responsible for non-thermal broadening (Hassler et al. ,1990; Banerjee et al. , 1998; Wilhelm et al. , 2004, 2005; Banerjee, P´erez-Su´arez,and Doyle, 2009), while narrowing of spectra after a certain height in the solaratmosphere is most likely signature of the Alfv´en wave dissipation (Doyle et al. ,1997; Roberts, 2000; Pek¨unl¨u et al. , 2002; O’Shea, Banerjee, and Doyle, 2005).Therefore, propagation of Alfv´en waves increases the non-thermal velocity upto the inversion point in QS (Log T e = 5.47 K) and after the inversion point,which lies in the upper TR/lower corona, Alfv´en waves may dissipate throughvarious mechanisms, e.g. , viscous & ohmic (Roberts, 2000), resonant absorption(Ionson, 1978; Erdelyi and Goossens, 1996; Doyle et al. , 1997), phase mixing, etc .Although, which mechanism is responsible for the Alfv´en wave dissipation can SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 19 ayshap et al. not be speculated here. This is one possible explanation for our results related tothe non-thermal velocity pattern in QS. Apart from the Alfv´en wave propagationand dissipation mechanism, another possible explanation may lie in terms ofthe nano-flares, which is frequently occurring at O vi temperature (Peter andJudge, 1999). Teriaca et al. vi formation temperature ( i.e. , 3 × K ) and they concluded thatnon-thermal velocities arise due to the prevalent occurrence of the nano-flaresin this region. The presence of the lower values of non-thermal velocities aboveand below this region is quite obvious due to the energy loss. In the present case,the non-thermal velocity peaks at Log T e = 5.47 K (2.95 × K) in the QS.Therefore, our peak value of the non-thermal velocity is also very close to theO vi formation temperature and it may be probable justification for our non-thermal results. In conclusion, Alfv´en wave propagation and dissipation as wellas prevalent occurrence of the nano-flares around at O vi formation temperaturecan explain the variation of non-thermal velocity with temperature. These twomechanisms are the viable source but that does not rule out other possibilities.Our results should help in constraining the atmospheric models. We hope thatour attempt of identifying the footprints of coronal funnels based on densitycontrast and excess widths shed new light on the overall complexity and topologyof the polar regions in general. This approach of searching for additional widthsas precursors for transients can be further improved with future better resolutionspectrographs. Acknowledgements
We acknowledge the Hinode/EIS observation for this study. Hinodeis Japanese mission developed and launched by ISAS/JAXA, with NAOJ as domestic partnerand NASA and UKSA as international partners. P. Kayshap acknowledges the support fromIndian Institute of Astrophysics, Bangalore for his visit.
References
Aiouaz, T., Peter, H., and Lemaire, P.: 2005,
Astron. Astroph. , 713.Athay, R.G.: 1984,
Astrophys. J. , 412.Athay, R.G. and Dere, K.P.: 1991,
Astrophys. J. , 323.Axford, W.I. and McKenzie, J.F.: 1997,
Cosmic Winds and the Heliosphere , 31.Banerjee, D., Teriaca, L., Doyle, J.G., and Wilhelm, K.: 1998,
Astron. Astroph. , 208.Banerjee, D., O’Shea, E., Doyle, J.G., and Goossens, M.: 2001a,
Astron. Astroph. , 691.Banerjee, D., O’Shea, E., Doyle, J.G., and Goossens, M.: 2001b,
Astron. Astroph. , L39.Banerjee, D., P´erez-Su´arez, D., and Doyle, J.G.: 2009,
Astron. Astroph. , L15.Boland, B.C., Dyer, E.P., Firth, J.G., Gabriel, A.H., Jones, B.B., Jordan, C., McWhirter,R.W.P., Monk, P., and Turner, R.F.: 1975,
Monthly Notices of the Royal AstronomicalSociety , 697.Chae, J., Sch¨uhle, U., and Lemaire, P.: 1998,
Astrophys. J. , 957.Chae, J., Yun, H.S., and Poland, A.I.: 1998,
The Astrophysical Journal Supplement Series , 151.Chmielewski, P., Srivastava, A.K., Murawski, K., and Musielak, Z.E.: 2013,
Monthly Noticesof the Royal Astronomical Society , 40.Coyner, A.J. and Davila, J.M.: 2011,
Astrophys. J. , 115.Culhane, J.L., Harra, L.K., James, A.M., Al-Janabi, K., Bradley, L.J., Chaudry, R.A., Rees, K.,Tandy, J.A., Thomas, P., Whillock, M.C.R., Winter, B., Doschek, G.A., Korendyke, C.M.,Brown, C.M., Myers, S., Mariska, J., Seely, J., Lang, J., Kent, B.J., Shaughnessy, B.M.,Young, P.R., Simnett, G.M., Castelli, C.M., Mahmoud, S., Mapson-Menard, H., Probyn,
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 20 omparison Between Coronal Holes and Quiet-SunB.J., Thomas, R.J., Davila, J., Dere, K., Windt, D., Shea, J., Hagood, R., Moye, R., Hara,H., Watanabe, T., Matsuzaki, K., Kosugi, T., Hansteen, V., and Wikstol, Ø.: 2007,
SolarPhys. , 19.Dadashi, N., Teriaca, L., and Solanki, S.K.: 2011,
Astron. Astroph. , A90.Dammasch, I.E., Wilhelm, K., Curdt, W., and Hassler, D.M.: 1999,
Astron. Astroph. , 285DeForest, C.E. and Gurman, J.B.: 1998,
Astrophys. J. , L217.Dere, K.P. and Mason, H.E.: 1993,
Solar Phys. , 217.Dere, K.P., Landi, E., Mason, H.E., Monsignori Fossi, B.C., and Young, P.R.: 1997,
Astronomyand Astrophysics Supplement Series , 149.Doschek, G.A., Vanhoosier, M.E., Bartoe, J.-D.F., and Feldman, U.: 1976,
The AstrophysicalJournal Supplement Series , 417.Doschek, G.A. and Feldman, U.: 2000, Astrophys. J. , 599.Doschek, G.A., Mariska, J.T., Warren, H.P., Brown, C.M., Culhane, J.L., Hara, H., Watanabe,T., Young, P.R., and Mason, H.E.: 2007,
Astrophys. J. , L109.Doyle, J.G., O’Shea, E., Erdelyi, R., Dere, K.P., Socker, D.G., and Keenan, F.P.: 1997,
SolarPhys. , 243.Dwivedi, B.N. and Srivastava, A.K.: 2006,
Solar Phys. , 143.Erdelyi, R. and Goossens, M.: 1996,
Astron. Astroph. , 664.Fu, H., Xia, L., Li, B., Huang, Z., Jiao, F., and Mou, C.: 2014,
ArXiv e-prints , arXiv:1408.5473.Gabriel, A.H.: 1976,
Royal Society of London Philosophical Transactions Series A , 339.Gomez, D. and Ferro Fontan, C.: 1988,
Solar Phys. , 33.Gomez, D.O. and Ferro Fontan, C.: 1992,
Astrophys. J. , 662.Gupta, G.R., Banerjee, D., Teriaca, L., Imada, S., and Solanki, S.: 2010,
Astrophys. J. ,11.Hackenberg, P., Mann, G., and Marsch, E.: 1999,
Space Science Reviews , 207.Hansteen, V.: 1993, Astrophys. J. , 741.Hansteen, V., Maltby, P., and Malagoli, A.: 1997,
Magnetic Reconnection in the SolarAtmosphere , 116.Hansteen, V.H., Hara, H., De Pontieu, B., and Carlsson, M.: 2010,
Astrophys. J. , 1070.Hassler, D.M., Dammasch, I.E., Lemaire, P., Brekke, P., Curdt, W., Mason, H.E., Vial, J.-C.,and Wilhelm, K.: 1999,
Science , 810.Hassler, D.M., Rottman, G.J., Shoub, E.C., and Holzer, T.E.: 1990,
Astrophys. J. , L77.He, J.-S., Tu, C.-Y., and Marsch, E.: 2008,
Solar Phys. , 147.Heyvaerts, J. and Priest, E.R.: 1992,
Astrophys. J. , 297.Ionson, J.A.: 1978,
Astrophys. J. , 650.Kohl, J.L., Noci, G., Cranmer, S.R., and Raymond, J.C.: 2006,
Astronomy and AstrophysicsReview , 31.Krieger, A.S., Timothy, A.F., and Roelof, E.C.: 1973, Solar Phys. , 505.Krishna Prasad, S., Banerjee, D., and Gupta, G.R.: 2011, Astron. Astroph. , L4.Landi, E., Del Zanna, G., Young, P.R., Dere, K.P., Mason, H.E., and Landini, M.: 2006,
TheAstrophysical Journal Supplement Series , 261.Madjarska, M.S., Doyle, J.G., and van Driel-Gesztelyi, L.: 2004,
Astrophys. J. , L57.Madjarska, M.S. and Wiegelmann, T.: 2009,
Astron. Astroph. , 991.Madjarska, M.S., Huang, Z., Doyle, J.G., and Subramanian, S.: 2012,
Astron. Astroph. ,A67.Mariska, J.T., Feldman, U., and Doschek, G.A.: 1978,
Astrophys. J. , 698Marsch, E. and Tu, C.-Y.: 1997,
Solar Phys. , 87.McIntosh, S.W., Davey, A.R., Hassler, D.M., Armstrong, J.D., Curdt, W., Wilhelm, K., andLin, G.: 2007,
Astrophys. J. , 650.Ofman, L. and Davila, J.M.: 1997,
Astrophys. J. , L51.O’Shea, E., Banerjee, D., and Doyle, J.G.: 2005,
Astron. Astroph. , L35.O’Shea, E., Banerjee, D., and Doyle, J.G.: 2006,
Astron. Astroph. , 1059.O’Shea, E., Banerjee, D., and Doyle, J.G.: 2007,
Astron. Astroph. , 713Patsourakos, S. and Klimchuk, J.A.: 2006,
Astrophys. J. , 1452.Pek¨unl¨u, E.R., Bozkurt, Z., Afsar, M., Soydugan, E., and Soydugan, F.: 2002,
Monthly Noticesof the Royal Astronomical Society , 1195.Peter, H. and Judge, P.G.: 1999,
Astrophys. J. , 1148.Peter, H.: 1999,
Astrophys. J. , L77.Peter, H.: 2001,
Astron. Astroph. , 1108.Peter, H., Gudiksen, B.V., and Nordlund, ˚A.: 2004,
Astrophys. J. , L85Peter, H., Gudiksen, B.V., and Nordlund, ˚A.: 2006,
Astrophys. J. , 1086.
SOLA: PK_DB_AKS_Funnel-V2.tex; 13 September 2018; 13:13; p. 21 ayshap et al.Pneuman, G.W. and Kopp, R.A.: 1978,
Solar Phys. , 49.Roberts, B.: 2000, Solar Phys. , 139.Spadaro, D., Lanza, A.F., Karpen, J.T., and Antiochos, S.K.: 2006,
Astrophys. J. , 579.Srivastava, A.K. and Dwivedi, B.N.: 2007,
Journal of Astrophysics and Astronomy , 1.Srivastava, A.K., Konkol, P., Murawski, K., Dwivedi, B.N., and Mohan, A.: 2014, ArXiv e-prints , arXiv:1407.7124.Teriaca, L., Banerjee, D., and Doyle, J.G.: 1999a,
Astron. Astroph. , 636.Teriaca, L., Doyle, J.G., Erd´elyi, R., and Sarro, L.M.: 1999b,
Astron. Astroph. , L99.Teriaca, L., Banerjee, D., Falchi, A., Doyle, J.G., and Madjarska, M.S.: 2004,
Astron. Astroph. , 1065.Tian, H., Marsch, E., Tu, C.-Y., Xia, L.-D., and He, J.-S.: 2008a,
Astron. Astroph. , 267.Tian, H., Xia, L.-D., He, J.-S., Tan, B., and Yao, S.: 2008c,
Chinese Journal of Astronomyand Astrophysics , 732.Tian, H., Tu, C., Marsch, E., He, J., and Kamio, S.: 2010, Astrophys. J. , L88.Tu, C.-Y., Zhou, C., Marsch, E., Xia, L.-D., Zhao, L., Wang, J.-X., and Wilhelm, K.: 2005,
Science , 519.Waldmeier, M.: 1975,
Solar Phys. , 351.Warren, H.P., Mariska, J.T., and Wilhelm, K.: 1997, Astrophys. J. , L187.Wiegelmann, T. and Solanki, S.K.: 2004,
Solar Phys. , 227.Wilhelm, K., Dammasch, I.E., Marsch, E., and Hassler, D.M.: 2000,
Astron. Astroph. ,749.Wilhelm, K., Dwivedi, B.N., Marsch, E., and Feldman, U.: 2004,
Space Science Reviews ,415.Wilhelm, K., Fludra, A., Teriaca, L., Harrison, R.A., Dwivedi, B.N., and Pike, C.D.: 2005,
Astron. Astroph. , 733Xia, L.D., Marsch, E., and Curdt, W.: 2003,
Astron. Astroph. , L5.Young, P.R., Del Zanna, G., Mason, H.E., Dere, K.P., Landi, E., Landini, M., Doschek, G.A.,Brown, C.M., Culhane, L., Harra, L.K., Watanabe, T., and Hara, H.: 2007,
Publications ofthe Astronomical Society of Japan , 857Young, P.R., Watanabe, T., Hara, H., and Mariska, J.T.: 2009, Astron. Astroph. , 587., 587.