The redshifted network contrast of transition region emission
aa r X i v : . [ a s t r o - ph . S R ] J a n Astronomy&Astrophysicsmanuscript no. 0490˙ms c (cid:13)
ESO 2018October 25, 2018
The redshifted network contrast of transition region emission
W. Curdt , H. Tian , , B. N. Dwivedi , , and E. Marsch Max-Planck-Institut f¨ur Sonnensystemforschung (MPS), Max-Planck-Str.2, 37191 Katlenburg-Lindau, Germanye-mail: [email protected] School of Earth and Space Sciences, Peking University, China Department of Applied Physics, Institute of Technology, Banaras Hindu University, Varanasi-221005, IndiaReceived July 1, 2008; accepted October 1, 2008
ABSTRACT
Aims.
We study the VUV emission of the quiet Sun and the net redshift of transition region lines in the SUMER spectral range. Weaim at establishing a link with atmospheric processes and interpreting the observed downflow as the most evident part of the prevailingglobal coronal mass transport.
Methods.
We rank and arrange all pixels of a monochromatic raster scan by radiance and define equally-sized bins of bright, faint, andmedium-bright pixels. Comparing the bright pixels with the faint pixels, we determine the spectrally-resolved network contrast for 19emission lines. We then compare the contrast centroids of these lines with the position of the line itself. We establish a relationshipbetween the observed redshift of the network contrast with the line formation temperature.
Results.
We find that the network contrast is o ff set in wavelength compared to the emission line itself. This o ff set, if interpreted asredshift, peaks at middle transition region temperatures and is 10 times higher than the previously reported net redshift of transitionregion emission lines. We demonstrate that the brighter pixels are more redshifted, causing both a significant shift of the networkcontrast profile and the well-known net redshift. We show that this e ff ect can be reconstructed from the radiance distribution. Thisresult is compatible with loop models, which assume downflows near both footpoints. Key words.
Sun: UV radiation – Sun: transition region – Line: formation – Line: profile
1. Introduction
Observations and interpretations of red- and / or blueshifted emis-sion lines from cosmic objects are crucial to understand thephysical processes at work there. The net redshift in the so-lar transition region (TR) emission lines has been knownsince the Skylab era (e.g., Doschek et al., 1976, and refer-ences therein). Redshifts have also been recorded in stellarspectra (Ayres et al., 1983; Wood et al., 1996). More recently,Brekke et al. (1997) and Chae et al. (1998) independently ver-ified this result, analysing high spectral resolution observationsfrom the Solar Ultraviolet Measurements of Emitted Radiation(SUMER) instrument on S oHO (Wilhelm et al., 1995). Boththese groups found similar results for the quantitative depen-dence of the net redshift on line formation temperature in therange from 10 K to 10 K. The reported peak downflow of 6to 8 km / s at a temperature of around 10 K is four times higherthan the detection limit of the instrument. However, both groupsadopted incorrect literature values for the rest wavelengths forthe Ne viii and Mg x emission. Dammasch et al. (1999) andPeter & Judge (1999) have established more realistic rest wave-length values for these species and have demonstrated the dis-appearance of the net redshift in coronal emission lines. To ourknowledge, a satisfactory physical explanation of the net red-shift has not yet been found. Peter (2006) has recently reportedthe first full-Sun VUV emission line profile, showing enhancedemission in the wings.We present a new method to investigate and explain theTR redshift using the network contrast. A spectrally resolvedcontrast curve has been included in the SUMER disk atlas ofCurdt et al. (2001, hereafter referred to as SDA). Here, the net-
780 782 784 786 788 790 792wavelength / Å0.01 0.1110 r a d i a n ce / W s r − m − Å − . N e V III . M g V III . S X I . F e V II . M n V II . S V . O I V . M g V III . N a V III . O I V . F e V II n e t w o r k c on t r a s t Fig. 1.
Enlarged portion of the SUMER spectral atlas(Curdt et al., 2001) showing radiances of average QS (black),sunspot (red), and coronal hole (blue) regions. The network con-trast (ratio bright network / cell interior in green), which is nor-mally around 3, increases to values of 6 to 8 in TR lines, and thecentroids of the contrast profiles are clearly redshifted.work contrast – defined as the radiance ratio of pixels in thebright network and pixels in the cell interior – has values ofabout 3 in the continua, and rises to values of 6 to 8 in TRemission lines. A similar result was reported by Reeves (1976).In coronal lines such as Ne viii and Mg x the contrast is belowthe background value, a finding which is equivalent to the result W. Curdt et al.: The redshifted network contrast of transition region emission of Doschek (2006), who reported a low correlation between theemission of the TR and corona.An enlarged cutout of the SDA quiet-Sun profile – domi-nated by the strong Ne viii , S v , and O iv emission lines around785 Å – is displayed in Fig.1. It is also obvious, although notexplicitly mentioned in the SDA, that the network profiles areredshifted compared to the emission lines themselves. Our goalis to give a physical explanation for this o ff set.In this paper, we extend the earlier work of Reeves (1976)and the work reported in the SDA by a comprehensive inves-tigation of the contrast employing a much larger data set. Weshow that our result is a direct consequence of the redshift-to-brightness relationship and can be reconstructed by a simplemodel using multi-component contributions to the line profile.A full discussion of the implications for loop models is beyondthe scope of this work. We only present here some salient fea-tures, and a detailed study taking account of atmospheric modelswill be covered in a separate paper.In contrast to the earlier work of Brekke et al. (1997) andChae et al. (1998), our new indirect method is unique in severalways, namely(i) it does not require an accurate wavelength calibration,(ii) it is independent of an exact knowledge of the rest wave-length,(iii) it closely relies on physical processes in the solar atmo-sphere.
2. Method
The signal in each pixel is a mixture of emission from di ff erentplasmas along the LOS and from unresolved fine structures. Wehave decomposed them by a statistical method based on the as-sumption that statistically di ff erent components can be separatedin the radiance distribution. In our new method, we di ff erentiateamong di ff erent classes of brightness in the radiance distribu-tion instead of simply averaging the brightness over all pixels.We assume that individual bins of di ff erent brightness do behavedi ff erently. In particular, we make use of the fact that brighterpixels have a tendency to appear redshifted in many emissionlines. Such a redshift-to-brightness relationship was also notedby Dammasch et al. (2008) in a di ff erent data set, and must con-sequently have an imprint on the network contrast. This is thecore of our method: the strong redshift observed in the networkcontrast as compared to the position of the line itself. For a col-lection of prominent, blend-free emission lines, we determinedthe network contrast (ratio of 33% of the brighter pixels as com-pared to 33% of the dim pixels) and compared the contrast curveto the emission line itself.A rather crude and empirical method was used in the SDA todisplay the network contrast; the radiance of a few bright pixelsout of 300 along the slit, which were thought to represent the net-work, was compared to the radiance of the remaining pixels. Thepixel selection was made in a static way for all 36 individual ex-posures of the data set, a procedure that may not be appropriatein view of the temporal evolution of the Sun. However, the shiftof the slit image caused by the wavelength scan (due to misalign-ment of the grating) had been compensated for by employing thestandard delta pixel routine.In the present analysis, we measure and compare the posi-tion of the line centroids (the line center determined by spectralcentroiding) relative to the position of the contrast maximum orminimum. We find that the centroid of the contrast profile is nor-mally shifted by several pixels towards longer wavelengths. Fig. 2.
Raster scan in the emission of N iv , Ne viii and the contin-uum around 780 Å. In the continuum map brightness contours at33% and 67% levels are overlaid. These have been used to definebright, faint and medium-bright pixels.
3. Observations
The data set in the SDA is a snapshot of 300 pixels along theslit. Thus, the given spectral radiances still have significant un-certainties and can only approximately represent the quiet Sun.Also, better values for the network contrast could have beenachieved if a better statistical basis were available like the onewe report here, where rasters are observed instead of a single ex-posure. Our data set consists of raster scans of size 51 ′′ × ′′ in14 di ff erent wavelength windows of ≈
44 Å covering the entirewavelength range from 670 Å to 1490 Å with only few insignifi-cant gaps. We employed a slit of size 1 ′′ × ′′ . The raster incre-ment was 1.5 ′′ and the exposure time was 90 s. The rasters wereobtained in the so-called ’Schmierschritt’ mode, which meansthat each transmitted spectrum is composed of four elementaryexposures with a 22.5 s dwell time and a 0.375 ′′ step incre-ment (Wilhelm et al., 1995). We already compensated on boardfor the parasitic movement of the slit image mentioned earlier.Therefore, we can safely assume that the North-South o ff set be-tween the individual rasters is negligible. We have compensatedfor the solar rotation after each raster scan. Therefore, we alsoassume that all rasters do map the same portion of the Sun in theEast-West direction.Our data set was obtained during an observation on 5 April2007 running from 00:51 UTC to 13:10 UTC. The initial point-ing – centre of the first raster – was x = y =
0. Fifty min-utes are needed for each raster. Standard procedures from theSUMERsoft library were applied for the data reduction.The major improvement of this observation (called the ’su-per atlas’) as compared to a normal reference spectrum is theincrease of the number of pixels by more than an order of mag-nitude. This data set allows us to produce monochromatic rasterscans for all emission lines and all continua in the SUMER spec-tral range. As an example, we show in Fig. 2 the maps obtainedsimultaneously in the emission of N iv , Ne viii and in the contin-uum around 780 Å. SUMERsoft is a software library, which constitutes the inte-grated experience with SUMER data analysis tools. It is available athttp: // / projects / soho / sumer / text / list sumer soft.htm. Curdt et al.: The redshifted network contrast of transition region emission 3 Fig. 3.
The shift of the network contrastrelative to the position of the emissionline as a function of formation tempera-ture. Individual o ff sets have been deter-mined for comparison with the profileof the bright pixels ( △ ), the faint pix-els ( ◦ ), and the average pixels ( (cid:3) ). Theshift is scaled as Doppler flow. This isto make it comparable to earlier, directmeasurements and should not be con-fused with a real flow. We have selected 19 prominent and blend-free emission linesto produce monochromatic maps. For each raster we also pro-duce a map of the continuum. We use this continuum map, wherethe elements of the chromospheric network are well-structuredand at instrument resolution, to rank all pixels by radiance. Wedefine equally-sized bins of bright network pixels, of faint cell-interior pixels and of pixels with medium brightness. The samebin definition was used for all emission lines in a raster forthe determination of the spectrally resolved contrast profile. Foreach spectral pixel the ratio of the radiance found in the bright-pixel bin over that of the faint pixel bin was determined. Like inthe SDA, a significant increase of the contrast is observed for alllines except for N i , Ni ii , Si ii , Ne viii , and Mg ix , which show acontrast minimum. To account for temporal variations, we haverepeated the definition of radiance bins for each raster.All selected lines are listed in Table 1 with wavelength andformation temperature and with the o ff set results and timing in-formation of the relevant raster. In order to compare our resultwith previous work, we have converted the o ff sets to Dopplerflows, applying v / c = ∆ λ/λ . Fig. 3 displays the main result ofour work.Individual o ff sets have been determined for a comparisonwith the profile of the bright pixels ( △ ), the faint pixels ( ◦ ), andthe average pixels ( (cid:3) ). In TR emission, the data points for thebright and for the faint pixel profiles deviate, as can be expectedfrom the earlier work of Brekke et al. (1997) and Chae et al.(1998). This interesting secondary result is a direct consequenceof the redshift-to-brightness relationship.
4. Discussion and summary
Fig. 3 has similarities to the respective figures in Brekke et al.(1997) and Chae et al. (1998). We have expressed the o ff set ∆ λ in speed units of Doppler flow and arrived at values of about60 km / s. This is about 10 times higher than the average netdownflow found in earlier work. This finding demonstrates thatour new method enhances the visibility of the Doppler shift,which leads to a significant increase of the sensitivity, but shouldnot be confused with a real flow. Table 1.
Observed o ff set of the network contrast for 19 emissionlines. Lines from the same raster are co-temporal, raster start andstop time are in UTC. line λ / Å log T / K ∆ λ / Å timeMg ix iv iv viii v iv v ii iii iii ii vi ii iii v ii ii i iv We now use a simple model to reconstruct the observed red-shift of the network contrast. We use the fact that the radiancedistribution of the average quiet Sun follows a single lognormaldistribution function and that a redshift-to-brightness relation-ship exists. Pauluhn et al. (2000) have shown in great detail thatthe emission of the quiet Sun including the network and the in-tranetwork is better described by a lognormal distribution thanby two Gaussians. This is in agreement with the brightness dis-tribution of the λ
765 N iv TR line, as shown in Fig.4a.For this distribution, we have defined 40 equally sized ra-diance bins. For each bin we determined the line position. Wefound a linear relationship between the logarithm of the spec-tral radiance, L λ , and the redshift of the line centroid, as dis-played in Fig.4b. Dammasch et al. (2008) applied a di ff erent bindefinition, but they arrive at a similar result of 3 km / s per radi- W. Curdt et al.: The redshifted network contrast of transition region emission
Fig. 4.
Reconstruction of the observednetwork contrast in the radiance of the λ
765 N iv TR line (a) employing thealso observed redshift-to-brightness re-lationship (b) to rescale the abscissa ofthe lognormal radiance distribution (c)and a convolution of three separategroups of pixels with a Gaussian pro-file on a continuum (d). ance decade. We now use the redshift-to-brightness relationshipto rescale the abscissa in the histogram in Fig. 4c and defineseparate groups of bright pixels, of faint pixels and the pixels ofmedium brightness. The pixels in each group are convolved witha Gaussian line profile typical for TR emission on a continuumbackground. Thus we arrive at three di ff erent profiles for thosegroups of pixels. As a consequence of the redshift-to-brightnessrelationship shown in Fig.4b, the reconstructed profiles are o ff -set from each other by a few km / s (cf., Fig. 4d). The contrastprofile – the ratio of the bright profile to the faint profile – peaks,however, near a redshift value of ≈
70 km / s, which is close to ourempirical result in Fig. 3. This demonstrates that the observedo ff set of the network contrast can be reconstructed in a quan-titative manner with some basic assumptions. We note that thecontrast profile, which we use in our new method, is much moresensitive to shifts than the line profile itself. We also note thatthe observed skewness of the contrast profile in lines like λ viii or λ vi seems to be a real result of the reconstruc-tion (cf., Fig. 4d).We emphasize that our results cannot be used to make anystatement about systematic flows in coronal emission. The con-trast depression simply indicates that the corona is decoupledfrom the chromosphere, and therefore no deviation from the restwavelength can be expected for the contrast minimum. Coolplasma with a low degree of ionization will not be guided bythe magnetic field and will also not participate in the concentra-tion process in the downflow near loop footpoints. The redshift-to-brightness relationship as a direct consequence of this mostevident part of the prevailing global coronal mass transport(Marsch et al., 2008) may be the physical explanation for thewell-known net redshift in TR emission and of the o ff set in thespectrally resolved network contrast.Our new result corroborates the recent work ofDammasch et al. (2008) and Marsch et al. (2008). In actu-ality, blueshifts and redshifts respectively correspond to upflowsand downflows of the plasma on open and closed field lines asnoted by Marsch et al. (2008). The redshifted network contrastof TR emission, ubiquitous redshifts and sporadic blueshifts in the solar atmosphere show the physical characteristics of masstransport which we may term as ’coronal convection’.Dammasch et al. (2008) argue that the downflow concen-trated near both footpoints of coronal loops is powered bya quasi-continuous heating process. They follow the sugges-tion of Feldman et al. (2001) and infer that unresolved brightfeatures in the network are tiny loops, which are also red-shifted at both legs. Pauluhn & Solanki (2007) developed aheuristic nanoflare model showing that continuous small-scalebrightenings could produce the observed radiance distribution.Fontenla et al. (2007, 2008) assume that the Farley-Bunemanninstability could be responsible for the heating process. It is dif-ficult to use our observation in favour of or against any nanoflareheating model reported in the literature (see the review of Innes,2004, and references cited therein). These questions can cer-tainly not be answered from our observation alone. More the-oretical work is needed, which is beyond the scope of this com-munication. Acknowledgements.
The SUMER project is financially supported by DLR,CNES, NASA, and the ESA PRODEX Programme (Swiss contribution).SUMER is part of
SOHO of ESA and NASA. The work of BND was supportedby a grant from the MPS. HT is supported by China Scholarship Council forhis stay at MPS. We thank the referee for critical comments which improved theclarity of this Letter.
References
Ayres, T. R., Stencel, R. E., Linsky, J. L., Simon, T. E., Jordan, C., Brown, A., &Engvold, O., 1983, ApJ 274, 801.Brekke, P., Hassler, D. M., & Wilhelm, K.,1997, Sol. Phys. 175, 349.Chae, J., Yun, H. S., & Poland, A. I., 1998, ApJS 114, 151.Curdt, W., Brekke, P., Feldman, U., Wilhelm, K., Dwivedi, B.N., Sch¨uhle, U., &Lemaire, P. 2001, A&A 375, 591.Dammasch, I. E., Wilhelm, K., Curdt, W., & Hassler, D. M., 1999, A&A 346,285.Dammasch, I. E., Curdt, W., Dwivedi, B. N., & Parenti, S., 2008, Ann. Geophys.,in press.Doschek, G. A., Feldman, U., & Bohlin, J. D., 1976 ApJ 205, L177.Doschek, G. A., 2006, ApJ 649, 515.Feldman, U., Dammasch, I. E., & Wilhelm, K., 2001, ApJ 558, 423.Fontenla, J M., Curdt, W., Avrett, E. H., & Harder, J. 2007, A&A 468, 695.Fontenla, J M, Peterson, W. K., & Harder, J. 2008, A&A 480, 839. . Curdt et al.: The redshifted network contrast of transition region emission 5
Innes, D. E., 2004, in ed. H. Lacoste