Solar Spectroscopy and (Pseudo-)Diagnostics of the Solar Chromosphere
aa r X i v : . [ a s t r o - ph . S R ] J un To appear in “Recent Advances in Spectroscopy: Astrophysical, Theoretical and Ex-perimental Perspectives”, eds. R. K. Chaudhuri, M. V. Mekkaden, A. V. Raveendranand A. Satya Narayanan, Astrophysics and Space Science Proceedings, Springer-Verlag, Heidelberg, Berlin, 2009.
Solar Spectroscopy and (Pseudo-)Diagnosticsof the Solar Chromosphere
Robert J. Rutten
Sterrekundig Instituut, Utrecht University, Utrecht, The NetherlandsInstitutt for Teoretisk Astrofysikk, Oslo University, Oslo, Norway
Summary.
I first review trends in current solar spectrometry and then concentrateon comparing various spectroscopic diagnostics of the solar chromosphere. Someare actually not at all chromospheric but just photospheric or clapotispheric anddo not convey information on chromospheric heating, even though this is oftenassumed. Balmer H α is the principal displayer of the closed-field chromosphere, butit is unclear how chromospheric fibrils gain their large H α opacity. The open-fieldchromosphere seems to harbor most if not all coronal heating and solar wind driving,but is hardly seen in optical diagnostics. During this wide-ranging conference I gave a wide-ranging review coveringspectroscopy of the solar photosphere, chromosphere, and corona includingcondensed tutorials of line formation theory for these disparate domains.Being page-limited, I only summarize some major review points here andthen concentrate on optical diagnostics of the solar chromosphere – arguablythe hardest solar nut to crack (Judge & Peter 1998). Solar atmosphere physics.
A major quest of solar physics is to understandthe intricate interplay between solar magnetism, gas dynamics, and radia-tion that makes the solar atmosphere such an outstanding research laboratoryand also governs our space environment. The photosphere, chromosphere, andcorona represent optically thick, effectively thin, and optically thin interac-tion domains with much variation in plasma- β . They differ much in researchtechniques and in researcher specialisms, but do require holistic synthesis.Spectroscopy is a key research tool in this endeavor. It differs from otherastronomy in that the solar atmosphere displays its interactions in resolvablefine structure that varies on short time scales. Solar spectroscopy therefore ; for a non-condensed syl-labus see . Robert J. Rutten requires high time resolution, both observationally and interpretationally, inaddition to high angular resolution and large diagnostic diversity. Solar physics now.
Solar physics is experiencing a tremendous boost fromthree developments: wavefront correction enabling 0.1 ′′ resolution from meter-class optical telescopes, continuous multi-wavelength high-cadence monitoringfrom space, and increasing realism of numerical MHD simulations of solar-atmosphere fine structure. Photosphere–chromosphere–corona coupling is akey research topic and becomes addressable at the level of detailed under-standing rather than wishful cartoon/mechanism thinking. Classical spectroscopy.
Traditionally, solar spectroscopy concentrated on abun-dance determination using solar spectrum atlases and assuming LTE witha best-fit plane-parallel hydrostatic-equilibrium model photosphere (review:Rutten 2002). This practice and the resulting standards in helioseismologyand stellar evolution theory have been upset by Asplund’s turning to 3Dtime-dependent Nordlund-Stein granulation simulations and obtaining signif-icantly smaller abundance values for key elements such as oxygen, carbon andnitrogen (e.g., Asplund et al. 2004; review: Nordlund et al. 2009).
Trends in observation.
The emphasis of current solar spectrometry is on imag-ing spectroscopy with high image quality, improving spatial and temporalcoverage and resolution by limiting the spectral sampling to only moderatespectral resolution in only specific diagnostic lines. Optically this is done bestwith Fabry-P´erot interferometers (IBIS, Cavallini 2006, Reardon & Cavallini2008; CRISP, Scharmer 2006, Scharmer et al. 2008; the new G¨ottingen one,Bello Gonz´alez & Kneer 2008). They sample selected line profiles sequentiallyin time. Instantaneous integral-field spectrometry using fiber and/or lensletarrays for field reformatting is highly desirable but remains on the horizon(Rutten 1999b; Lin et al. 2004; Sankarasubramanian et al. 2009). Slit spec-trometers are still in use, but only for precision spectropolarimetry and forultraviolet emission-line measurement from space, in both cases for want ofsuited integral-field technology. Slits are no good because they always sam-ple the wrong place at the wrong time. In ground-based observation, theyalso inhibit wavefront restoration by algorithms such as the very effectiveMOMFBD of van Noort et al. (2005), a must in addition to adaptive optics.In space-based observation, multi-layer mirror technology has enabled thefruitful narrow-passband EUV imaging of SOHO, TRACE, STEREO and thecomprehensive monitoring promised by SDO, but, unfortunately, ultravioletintegral-field spectrometry remains a too distant hope.
Trends in interpretation.
Solar-atmosphere radiative transfer modeling ispresently split between data inversions and simulatory forward modeling. In-version codes initiated by Ruiz-Cobo et al. (1990, 1992) replace the classical“semi-empirical” (= best-fit) “one-dimensional” (= plane-parallel hydrostatic-equilibrium) modeling that either assumed LTE and CRD (Holweger 1967;revised into HOLMUL by Holweger & M¨uller 1974) or relaxed these assump-tions to NLTE and PRD (VALC, Vernazza et al. 1981, revised into FALC by olar Spectroscopy and the Solar Chromosphere 3
Fontenla et al. 1993). The data inversions obtain such best-fit stratificationsper pixel on the solar surface and are applied particularly in photosphericspectropolarimetry. Just as for the standard models, they loose credibilityhigher up where clapotispheric (see below) and chromospheric fluctuationsare too wild for smooth spline-function fitting or temporal averaging and ver-tical columns too poor a sample of 3D structure. Chromospheric line formationrequires forward modeling in the form of NLTE line synthesis within numer-ical simulations including non-equilibrium ionization evaluation (see below).A major task is to implement 3D time-dependent radiative transfer in 3Dtime-dependent MHD codes. A major challenge is to make such codes fast. (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) d (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) e (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) f (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) g b (cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1) (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) a c Fig. 1.
Bound-boundprocess pairs contribut-ing line extinction andemissivity.
Pair a : pho-ton destruction.
Pairsb and c : spontaneousand induced scatter-ing out of the beam.
Pairs d and e : photoncreation.
Pairs f andg : scattering into thebeam. From Rutten(2003).
Fig. 1 displays all radiative excitation–deexcitation process combinationsfor idealized two-level atoms. Their competition defines different radiation-physics domains of spectral line formation: – LTE = large collision frequency – up: mostly collisional = thermal creation (d + e);– down: mostly collisional = large destruction probability (a);– photon travel: “honorary gas particles” or negligible leak. – NLTE = statistical equilibrium or time-dependent – up and down: mostly radiative = non-local sources;– two-level scattering: coherent/complete/partial redistribution;– multi-level travel: pumping, suction, sensitivity transcription. – coronal equilibrium = hot tenuous – up: only collisional = thermal creation (only d);– down: only spontaneous (only d);– photon travel: escape / drown / scatter bf H I, He I, He II. Robert J. Rutten that also characterize line formation in solar-atmosphere domains: – Photosphere.
Optical continuum and weak-line photon escape. High den-sity makes LTE (= Saha–Boltzmann population ratios producing S ν = B ν )valid for optical continua and reasonable for most subordinate lines.“Height of formation” is a valid concept. The major trick is the Eddington-Barbier approximation I ν ( τ ν = 0 , µ ) ≈ S ν ( τ ν = µ ) ≈ B ν [ T ( τ ν = µ )]for quick insight and inversion constraining. Upper-photosphere resonancelines such as the Na I D lines, K I 7699 ˚A, and Ba II 4554 ˚A darken fromsimple resonance scattering (Uitenbroek & Bruls 1992). – Chromosphere.
Strong-line photon escape. LTE is invalid, statistical equi-librium is invalid, complete redistribution is invalid, instantaneous ioniza-tion equilibrium is invalid, single-fluid description is likely to be invalid.Height of formation: fibrils somewhere between the photosphere and thetelescope (Rutten 2007). The major trick is the four-parameter ( S ν , τ ν , λ − λ , ∆λ D ) cloud model of Beckers (1964), but all applications derivingtemperature and density using H α have erred in assuming instantaneousionization/recombination balancing (see below). – Corona.
Immediate photon escape. Optical thinness plus the (d)-only equi-librium (but adding dielectronic recombination into the bound-free anal-ogon) give the major trick: P hν ∝ R n ion N e d z = R N (d T / d z ) − d T ≡ “emission measure”. Resonance scattering may affect the strongest lines.Bound-free scattering out of EUV-line filtergraph passbands makes neutral-hydrogen clouds appear thickly black against bright EUV-line backgroundsin EUV filtergrams (see the cartoon in Fig. 10 of Rutten 1999a). The LTE line-fitting HOLMUL model favored by pre-Asplund abundance de-terminers has no chromosphere whatsoever but a temperature stratificationclose to radiative equilibrium to avoid non-observed reversals (emission cores)in strong Fe I lines. Admitting realistic NLTE departures delivers the same ap-parent line-core excitation temperatures when there is a chromospheric tem-perature rise above h ≈
500 km (Rutten & Kostik 1982), and therefore theNLTE VALC/FALC models fudged the ultraviolet line haze (Zwaan 1975)to resemble HOLMUL LTE lines by imposing a gradual transition from pureabsorption to pure scattering over the upper photosphere (Rutten 1988), sothat they don’t feel the VALC/FALC temperature rise which became knownas “the chromosphere”. However, Carlsson & Stein (1997) showed that sucha global temperature rise is unlikely to occur below h ≈ olar Spectroscopy and the Solar Chromosphere 5 s p i c u l e? c anop y w a v e s ? f l u x t ube p-mode interferenceshocksHIF c anop y heatinggranulationacoustic eventfinger Fig. 2.
Cartoon of the sub-canopy clapotisphere, pervaded by three-minuteshocks. HIF stands for horizontal internetwork field (Lites et al. 1996). From Rutten(1999a).
Such cartoons are now being vindicated by the numerical STAGGER sim-ulations performed in Oslo (Hansteen et al. 2007). The STAGGER snapshotin Fig. 3 shows smooth field at heights above the canopies outlined by higharches of exceedingly large internetwork hydrogen n = 1 and n = 2 NLTEoverpopulations. Clapotispheric shocks interact and dominate beneath thesearches. The intershock phases are cool (first panel) and have large ion and n = 2 overpopulations (last two panels). These arise from sluggish recombina-tion. Hydrogen ionizes fast in the hot shocks, but does not recombine quicklyenough in the cool aftermath to settle its ion/atom balancing before the nextshock comes along. The overpopulation of the n = 2 level, which sets the opac-ity of H α , follows the ground-state overpopulations in the arches and the ionoverpopulations in the clapotisphere.So what is the chromosphere? Classically, the name denotes the thin H α -dominated pink crescents around the eclipsed Sun glimpsed just after secondand just before third contact . The H α image in Fig. 4 shows that this pinkemission is dominated by long fibrils that connect various parts of activeregions. In quieter areas, such H α fibrils typically connect network acrosscell interiors. These fibrils constitute the chromosphere. The parallel Ca II Himage maps them only in the active region and only partially even there.Elsewhere it displays clapotispheric sub-canopy shocks and H V grains. Thus,the H α image shows the chromosphere whereas the Ca II H image is partlyclapotispheric and partly chromospheric (but mostly pseudo, see below). As on July 22 this year here in India if the weather suits. Robert J. Rutten
Fig. 3.
Snapshot from a 2D MHD Oslo STAGGER simulation withnon-equilibrium hydrogen ionization. A movie version is available at .Each panel is a vertical plane reaching from the top of the convection zone to thecorona.
Upper row : temperature with selected magnetic field lines, field strength,gas density.
Lower row : NLTE population departure coefficients for the groundstate, ion state, and first excited state of hydrogen. All color coding is logarithmic.Two fluxtube-like magnetic concentrations are connected by high-arching field thatis outlined by very large hydrogen ground-state and excited-state overpopulations(red and green arches in the lower row). The internetwork-like domain under thearches is mostly cool but pervaded by interacting shocks (clouds in the last twopanels). The movie shows dynamic fibrils (De Pontieu et al. 2007a) that jut outfrom the magnetic concentrations and expand and contract in rapid succession.From Leenaarts et al. (2007).
Fig. 4.
Active region AR10786 imaged synchronously in H α and Ca II H byP. S¨utterlin with the Dutch Open Telescope (DOT) on La Palma on July 8, 2005.Field size 182 ×
133 arcsec. The H α imaging used a tunable Lyot filter with pass-band 0.25 ˚A FWHM, the Ca II H imaging an interference filter with passband 1.35 ˚AFWHM. The DOT also takes images synchronously in the G band, blue and red con-tinua, and since recently in the Ba II 455 ˚A line. Many more DOT images, many DOTmovies, and all reduced DOT data are available at http://dot.astro.uu.nl .olar Spectroscopy and the Solar Chromosphere 7 Fig. 5.
Two sets of IBIS images.
Upper row : active network in H α , Ca II 8542 ˚A,Na I D . Lower row : quiet network in H α , Ca II 8542 ˚A, H α core width. The first fivepanels display the intensity of the line-profile minimum per pixel, the last panel theFWHM of the chromospheric H α core per pixel. IBIS is the Italian double Fabry-P´erot imaging spectrometer at the Dunn Solar Telescope of the US National SolarObservatory. More detail in Cauzzi et al. (2009). Courtesy Kevin Reardon. Fig. 5 shows similar comparisons. The active network again shows longnetwork-spanning fibrils in H α , but the quiet region has the clapotisphere pok-ing through away from network even in H α (cf. Rouppe van der Voort et al.2007; Rutten et al. 2008). The upper Ca II 8542 ˚A image shows the fibrilspartially, the lower one is mostly clapotispheric away from the network. TheNa I D image (third panel) also, or not even clapotispheric but just upper-photospheric since the IBIS spectroscopy shows virtually no shock signatures(Kevin Reardon, private communication). The final panel is discussed below. Many spectral features that are commonly supposed to be chromosphericactually are “below-the-surface viewers” for which enhanced brightness hasnothing to do with chromospheric emissivity or heating but with deep photonescape. The viewing pipes are Spruit fluxtubes, slender kilogauss magneticconcentrations that occur only sparsely and intermittently in quiet-Sun su-pergranulation cell interiors (De Wijn et al. 2005) but assemble more stablyin supergranular cell boundaries to form the magnetic network. At yet larger
Robert J. Rutten density they constitute plage. Fig. 6 explains how they appear as “brightpoints” near disk center and as “faculae” towards the limb. total τ = 1 cont τ = 1 τ µ = 1 Fig. 6.
Hole-in-surface fluxtube brightening.
Left : radial viewing. The magnetic-pressure evacuation deepens the photon escape layer to well below the outside surfaceand causes additional deepening for neutral-atom lines through ionization, for molec-ular lines through dissociation, and for strong-line wings through reduced damping.The escape layers have similar temperatures inside and outside, but the tube hasa flatter temperature gradient due to hot-wall irradiation.
Right : slanted near-limbviewing. The tube evacuation causes facular sampling of the hot granule behind thetube, appearing as a bright stalk. From Rutten (1999a), after Spruit (1976).
Fig. 7.
Temperature stratifications in a fluxtube-like magnetic concentration (dot-ted), a granule (dashed) and an intergranular lane (dot-dashed) in a 3D MHD simu-lation.
Left : against geometrical height.
Right : against column mass.
Markers : τ = 1locations for Mn I 5394.7 ˚A (up arrows), Fe I 5395.2 ˚A (down arrows), continuumin between (arrowless). The solid curve is the PLA model of Solanki & Brigljevic(1992) derived from multi-line spectropolarimetry without angular resolution. Thefabulous agreement between this empirical best-fit model and the ab-initio simula-tion mutually vindicates these very different approaches. Both spectral lines nearlyvanish in the magnetic concentration from ionization. The Mn I line is formed atmuch larger height in the granule and lane than the Fe I line which suffers weaken-ing from convective Dopplershifts. From Vitas et al. (2009). Also this old cartoon has been vindicated by recent simulations, for con-tinuum faculae by Keller et al. (2004), for G-band faculae by Carlsson et al.(2004), for bright points in the wings of strong lines by Leenaarts et al. (2006),and for bright points in the unusually wide lines of Mn I by Vitas et al. (2009). olar Spectroscopy and the Solar Chromosphere 9
The latter analysis, summarized in Fig. 7, explains why the brightening seemsless in all other, less wide, photospheric lines: it is not, but they show the out-side granulation brighter from convective Dopplershift wash-out and so loosefluxtube contrast.These fluxtubes have near-radiative-equilibrium stratifications across theupper photosphere, at least up to h ≈
400 km as sampled by the wings of Ca IIH & K (Sheminova et al. 2005), just as the mean outside photosphere abovethe granulation. No significant mechanical heating therefore: photosphericfluxtubes brighten through deeper viewing. All solar irradiance modeling ef-forts that regard faculae as hot stick-up stalks and mimic such with raised-temperature upper photospheres are inherently wrong (e.g., Unruh et al. 1999;Fontenla et al. 2006; Danilovi´c et al. 2007).The Na I D image in Fig. 5 suggests that also the network brighten-ing in this line is mostly hole-in-the-surface viewing. The outside τ = 1surface is formed in the upper photosphere and very dark from scattering(Uitenbroek & Bruls 1992), the tube inside is likely to be as bright as thecontinuum from ionization. This is now verifiable with Oslo STAGGER sim-ulations.The 5th panel of Fig. 5 suggests that also Ca II 8542 ˚A gains networkbrightening from surface-hole viewing when not obscured by fibrils. TheSTAGGER simulation of Leenaarts et al. (2009) supports this notion. Thefluxtube at left in their Fig. 4 obtains large brightness from deep line-centerformation and Doppler brightening in a post-shock downdraft, a likely occur-rence in deep-shocking fluxtubes (Steiner et al. 1998).The Ca II H & K lines have larger opacity than Ca II 8542 ˚A in any imag-inable structure and reach larger opacity than H α in classical LTE modeling and in one-dimensional NLTE modeling (Vernazza et al. 1981). One wouldtherefore expect to observe a denser fibril forest at right in Fig. 4 than atleft. However, the DOT Ca II H filter and all other H & K filtergraphs havefar too wide passbands to isolate the line core, even the narrowest ones at0.3 ˚A (Brandt et al. 1992; Hoekzema et al. 1998; W¨oger et al. 2006). In addi-tion, fibrils are very dark in H & K (Rutten et al. 2008), so that the additionof bright structures in the inner wings dominates the scene. Magnetic con-centrations are likely to again function as hole-in-the-surface viewers, withthe outside surface clapotispherically cool and dark except in shocks, and thetube inside and close surroundings bright by scattering photons from below(cf. Fig. 8 of Rutten 1999a, another old cartoon).Thus, I contend that the apparent network brightness in Na I D and Ca II8542 ˚A imaging spectroscopy and much of the apparent network brightnessin Ca II H & K filtergrams has less to do with chromospheric heating thanwith hole-in-the-surface viewing. Corollaries are that the Mount-Wilson Ca IIH & K stellar activity monitoring (e.g., Rutten 1986; Duncan et al. 1991) de-livers counts of such holes on the stellar surface, not the amount of chro- mospheric heating, and that the good correspondence between the variousdiagnostics of Loukitcheva et al. (2009) follows naturally because they are allhole counters. And I repeat that the basal flux (review: Schrijver 1995) is mostlikely clapotispheric (Rutten 1999a). So how to diagnose chromospheric heating, or rather, the onset of coronalheating taking place in the chromosphere? My contention that most networkemissivity in Na I D , Ca II 8542 ˚A and Ca II H in Figs. 4 and 5 stems fromphotospheric or clapotispheric view-pipe photon escape leads to the sugges-tion that not much chromospheric heating is seen in these images. The H α fibrils do display chromospheric fine-structure morphology, but whether theseappear dark or bright is a matter of fortuitous combination of the four cloudparameters S ν , τ ν , λ − λ , ∆λ D . Only ∆λ D senses the temperature directlybecause the small atomic mass of hydrogen favors thermal over nonthermalbroadening. The final panel of Fig. 5 therefore displays a measure of chromo-spheric temperature. It evidences heating in and near network (Cauzzi et al.2009).It seems likely that this network heating occurs through open-field chro-mospheric structures, i.e., more vertical than H α fibrils and connecting lessclosely than across network cells. They are barely visible in H α but were dis-covered as fast-waving “straws” in DOT Ca II H near-limb movies, appearingbright against the very dark internetwork clapotisphere (Rutten 2006), andsubsequently as fast-waving “spicules-II” in Hinode Ca II H off-limb movies byDe Pontieu et al. (2007b). De Pontieu et al. (2007c) identified Alfv´en wavesas their modulation agent and estimated that these accelerate the solar wind.Being very slender and very fast-moving, straws/spicules-II are very hard toobserve on-disk (Langangen et al. 2008), but ongoing work (De Pontieu et al.2008) suggests that they occur ubiquitously above unipolar network and plageand may supply heating through component reconnection, reminiscent of theseparatrix shear proposed by Van Ballegooijen et al. (1998, 1999) and markedas cross-dashes in my old cartoon in Fig. 2. I suggest that unresolved strawheating causes the network patches of large H α core width in the last panel ofFig. 5, some of the Ca II 8542 ˚A near-network fibril brightness in the secondpanel of Fig. 5, and the diffuse bright aureoles seen in DOT Ca II H moviesaround network that seem to resolve into straws in the very sharpest one .Another speculation is that straw dynamics also contributes to the solar-wind FIP fractionation. http://dot.astro.uu.nl/albums/movies/2006-04-24-NW-ca.avi or .mov olar Spectroscopy and the Solar Chromosphere 11 There are striking and disconcerting similarities between the visibility of thecorona in the form of EUV-line loops and of the chromosphere in the form ofH α fibrils. Both types of structure seem to map the large-scale magnetic fieldtopology, although there is no direct proof of direct correspondence. In bothcases it is a misconception to regard them as magnetic fluxtubes embeddedin field-free plasma (or vacuum). Loops and fibrils chart bundles of field thatfor some reason contain more gas containing atoms and ions in the right statefor emissivity and/or extinction in the pertinent lines than the surrounding,otherwise very similar field bundles. The actual field topology is probablymore uniform than the fine-scaled emissivity/extinction structuring suggests(Judge 2006). The mass loading and dynamics of these narrow bundles arelikely structured on the scales of intergranular fluxtubes by thermodynami-cal processes affecting these magnetic footpoints, but the actual connectiontopology may be as complicated as in Fig. 18 of De Pontieu et al. (2003).A disconcerting similarity of coronal loops and chromospheric fibrils isthat neither structure has yet been produced naturally in MHD simulations.Acoustic oscillations have been shown to drive and probably mass-load shortdynamic fibrils jutting out from network and plage (Hansteen et al. 2006;De Pontieu et al. 2007a), but the mass loading of the long network-spanningfibrils and yet longer coronal loops remains unexplained. The overpopulationarches in Fig. 3 are promisingly similar to actual H α fibrils, but they containinsufficient H α opacity, notwithstanding their enormous n = 2 overpopulation,to produce the τ ≈ α that is characteristic of fibrils. Thissimulation was 2D only, but also the newer 3D Oslo STAGGER simulationshave not produced fibrils or loops so far. Perhaps the simulation volumes arestill too small, or perhaps they are too unipolar, or perhaps longer history isneeded, perhaps to let trickling siphon flows build up appreciable mass loads.Another disconcerting similarity is that both coronal loops and chromo-spheric fibrils delineate closed fields whereas the action lies in the open-fieldcomponents, respectively driving the fast wind and heating the corona. Theirvisibility is meager: coronal holes are indeed holes in emissivity and chromo-spheric straws/spicules-II were discovered only recently. The natural emphasison higher-visibility loops and fibrils seems to target red herrings qua impor-tance and role.Finally, it seems likely that the so-called transition region exists mostly ashot evaporation sheaths around these ephemeral structures (McIntosh et al.2008; Judge & Centeno 2008; Koza et al. 2009), a far cry from the stablespherical shell invoked in one-dimensional modeling or the radial stratificationwishfully assumed in many transition-region oscillation propagation studies. References
Asplund, M., Grevesse, N., Sauval, A. J., Allende Prieto, C., Kiselman, D. 2004,2 Robert J. RuttenA&A, 417, 751Beckers, J. M. 1964, A Study of the Fine Structures in the Solar Chromosphere,AFCRL Environmental Research Paper No. 49, PhD thesis Utrecht UniversityBello Gonz´alez, N., Kneer, F. 2008, A&A, 480, 265Brandt, P. N., Rutten, R. J., Shine, R. A., & Trujillo Bueno, J. 1992, Cool Stars,Stellar Systems, and the Sun, ASP Conf. Ser., 26, 161Carlsson, M., Stein, R. F. 1997, ApJ, 481, 500Carlsson, M., Stein, R. F., Nordlund, ˚A., Scharmer, G. B. 2004, ApJ, 610, L137Cauzzi, G., Reardon, K., Rutten, R., Tritschler, A., Uitenbroek, H. 2009, A&A, inpress, arXiv:0906.2083Cavallini, F. 2006, Solar Phys., 236, 415Danilovi´c, S., Solanki, S. K., Livingston, W., Krivova, N., Vince, I. 2007, in Modernsolar facilities - advanced solar science, eds. F. Kneer, K. G. Puschmann, & A. D.Wittmann, 189De Pontieu, B., Carlsson, M., McIntosh, S., Hansteen, V., Tarbell, T. 2008, 12thEuropean Solar Physics Meeting, http://espm.kis.uni-freiburg.de , 2.15De Pontieu, B., Hansteen, V. H., Rouppe van der Voort, L., van Noort, M., Carlsson,M. 2007a, ApJ, 655, 624De Pontieu, B., McIntosh, S., Hansteen, V. H., et al. 2007b, PASJ, 59, 655De Pontieu, B., McIntosh, S. W., Carlsson, M., et al. 2007c, Science, 318, 1574De Pontieu, B., Tarbell, T., Erd´elyi, R. 2003, ApJ, 590, 502De Wijn, A. G., Rutten, R. J., Haverkamp, E. M. W. P., S¨utterlin, P. 2005, A&A,441, 1183Duncan, D. K., Vaughan, A. H., Wilson, O. C., et al. 1991, ApJS, 76, 383Fontenla, J. M., Avrett, E., Thuillier, G., Harder, J. 2006, ApJ, 639, 441Fontenla, J. M., Avrett, E. H., Loeser, R. 1993, ApJ, 406, 319Hansteen, V. H., Carlsson, M., Gudiksen, B. 2007, in The Physics of ChromosphericPlasmas, eds. P. Heinzel, I. Dorotoviˇc, & R. J. Rutten, ASP Conf. Ser., 368, 107Hansteen, V. H., De Pontieu, B., Rouppe van der Voort, L., van Noort, M., Carlsson,M. 2006, ApJ, 647, L73Hoekzema, N. M., Rutten, R. J., Brandt, P. N., & Shine, R. A. 1998, A&A, 329,276Holweger, H. 1967, Zeitschrift fur Astrophysik, 65, 365Holweger, H., M¨uller, E. A. 1974, Solar Phys., 39, 19Judge, P. 2006, in Solar MHD Theory and Observations: A High Spatial ResolutionPerspective, eds. J. Leibacher, R. F. Stein, & H. Uitenbroek, ASP Conf. Ser., 354,259Judge, P., Centeno, R. 2008, ApJ, 687, 1388Judge, P. G., Peter, H. 1998, Space Science Reviews, 85, 187Keller, C. U., Sch¨ussler, M., V¨ogler, A., Zakharov, V. 2004, ApJ, 607, L59Koza, J., Rutten, R. J., Vourlidas, A. 2009, A&A, 499, 917Langangen, Ø., De Pontieu, B., Carlsson, M., et al. 2008, ApJ, 679, L167Leenaarts, J., Carlsson, M., Hansteen, V., Rouppe van der Voort, L. 2009, ApJ, 694,L128Leenaarts, J., Carlsson, M., Hansteen, V., Rutten, R. J. 2007, A&A, 473, 625Leenaarts, J., Rutten, R. J., Carlsson, M., Uitenbroek, H. 2006, A&A, 452, L15Lin, H., Kuhn, J. R., Coulter, R. 2004, ApJ, 613, L177Lites, B. W., Leka, K. D., Skumanich, A., Mart´ınez Pillet, V., Shimizu, T. 1996,ApJ, 460, 1019olar Spectroscopy and the Solar Chromosphere 13Loukitcheva, M., Solanki, S. K., White, S. M. 2009, A&A, 497, 273McIntosh, S. W., De Pontieu, B., Tarbell, T. D. 2008, ApJ, 673, L219Nordlund, ˚A., Stein, R. F., Asplund, M. 2009, Living Reviews in Solar Physics, 6, 2Reardon, K. P., Cavallini, F. 2008, A&A, 481, 897Rouppe van der Voort, L. H. M., De Pontieu, B., Hansteen, V. H., Carlsson, M.,van Noort, M. 2007, ApJ, 660, L169Ruiz Cobo, B., del Toro Iniesta, J. C. 1992, ApJ, 398, 375Ruiz Cobo, B., del Toro Iniesta, J. C., Collados, M., S´anchez Almeida, J. 1990,Ap&SS, 170, 113Rutten, R. G. M. 1986, A&A, 159, 291Rutten, R. J. 1988, in Physics of Formation of FeII Lines Outside LTE, eds. R. Viotti,A. Vittone, & M. Friedjung, IAU Coll. 94, Reidel, Dordrecht, 185Rutten, R. J. 1995, in Helioseismology, eds. J. T. Hoeksema, V. Domingo, B. Fleck,& B. Battrick, ESA SP–376 Vol. 1, 151Rutten, R. J. 1999a, in Magnetic Fields and Oscillations, eds. B. Schmieder, A. Hof-mann, & J. Staude, ASP Conf. Ser., 184, 181Rutten, R. J. 1999b, in High Resolution Solar Physics: Theory, Observations, andTechniques, eds. T. R. Rimmele, K. S. Balasubramaniam, R. R. Radick, ASPConf. Ser., 183, 296Rutten, R. J. 2002, J. Astron. Data, 8, 8Rutten, R. J. 2003, Radiative Transfer in Stellar Atmospheres,