1: %\documentclass[openbib]{article}
2: \documentclass[12pt,preprint]{aastex}
3: %\documentclass[manuscript]{aastex}
4: %\usepackage{psfig}
5: %\usepackage{lineno}
6: \usepackage{epsfig}
7:
8: \def\references{\bibliographystyle{/net/star/Users/schryver/book/references/swanbib}\bibliography{/net/star/Users/schryver/book/references/ref_karel}}
9:
10:
11: %\def\citet#1{\citeauthor{#1} (\citeyear{#1})}
12: %\def\citep#1{\citeauthor{#1}, \citeyear{#1}}
13:
14: \hyphenation{Schrij-ver}
15:
16: %\def\referee#1{{\sc #1}}
17: %\def\referee#1{{#1}}
18:
19: %%%%%%%%%%%%%%%%% END OF PREAMBLE %%%%%%%%%%%%%%%%
20:
21: \begin{document}
22: %\linenumbers
23:
24: \title{Non-linear force-free field modeling of a solar active region
25: around the time of a major flare and coronal mass ejection}
26:
27: \author{C.J.\ Schrijver\altaffilmark{1}, M.L.\ DeRosa\altaffilmark{1}, T.\ Metcalf\altaffilmark{2}, G.\ Barnes\altaffilmark{2},
28: B.\ Lites\altaffilmark{3}, T.\ Tarbell\altaffilmark{1}, J.\ McTiernan\altaffilmark{4}, G.\ Valori\altaffilmark{5},
29: T.\ Wiegelmann\altaffilmark{6}, M.S.\ Wheatland\altaffilmark{7}, T.\ Amari\altaffilmark{8}, G.\ Aulanier\altaffilmark{9},
30: P.\ D{\'e}moulin\altaffilmark{9}, M.\ Fuhrmann\altaffilmark{10}, K.\ Kusano\altaffilmark{11},
31: S.\ R{\'e}gnier\altaffilmark{12}, J.K.\ Thalmann\altaffilmark{6}}
32:
33: \altaffiltext{1}{ Lockheed Martin Advanced Technology Center, Palo Alto, Ca., USA;}
34: \altaffiltext{2}{ Colorado Research Associates, Boulder, Co., USA; }
35: \altaffiltext{3}{ High Altitude Observatory, National Center for Atmospheric Research, Boulder, Co., USA; }
36: \altaffiltext{4}{ Space Sciences Laboratory, University of California, Berkeley, USA; }
37: \altaffiltext{5}{ Astrophysikalisches Institut Potsdam, Potsdam, Germany; }
38: \altaffiltext{6}{ Max-Planck Institut f{\"u}r Sonnensystemforschung, Katlenburg-Lindau Germany; }
39: \altaffiltext{7}{ School of Physics, University of Sydney, Australia; }
40: \altaffiltext{8}{ CNRS, Centre de Physique Theorique, Ecole Polytechnique, Palaiseau, France; }
41: \altaffiltext{9}{ LESIA, Observatoire de Paris, CNRS, UPMC, Universit{\'e} Paris Diderot, Meudon, France; }
42: \altaffiltext{10}{ Universit{\"a}t Potsdam, Institut f{\"u}r Physik, Potsdam, Germany;}
43: \altaffiltext{11}{ Earth Simulator Center, Japan Agency for Marine-Earth Science and Technology, Yokohama, Japan; }
44: \altaffiltext{12}{ School of Mathematics and Statistics, University of St. Andrews, Fife, UK.}
45:
46: \begin{abstract}
47: Solar flares and coronal mass ejections are associated with rapid
48: changes in field connectivity and powered by the partial dissipation
49: of electrical currents in the solar atmosphere. A critical unanswered
50: question is whether the currents involved are induced by the motion of
51: pre-existing atmospheric magnetic flux subject to surface plasma
52: flows, or whether these currents are associated with the emergence of
53: flux from within the solar convective zone. We address this problem
54: by applying state-of-the-art nonlinear force-free field (NLFFF)
55: modeling to the highest resolution and quality vector-magnetographic
56: data observed by the recently launched Hinode satellite on NOAA Active
57: Region 10930 around the time of a powerful X3.4 flare. We compute 14
58: NLFFF models with 4 different codes and a variety of boundary
59: conditions. We find that the model fields differ markedly in geometry,
60: energy content, and force-freeness. We discuss the relative merits of
61: these models in a general critique of present abilities to model the
62: coronal magnetic field based on surface vector field measurements. For
63: our application in particular, we find a fair agreement of the
64: best-fit model field with the observed coronal configuration, and
65: argue (1) that strong electrical currents emerge together with
66: magnetic flux preceding the flare, (2) that these currents are carried
67: in an ensemble of thin strands, (3) that the global pattern of these
68: currents and of field lines are compatible with a large-scale twisted
69: flux rope topology, and (4) that the $\sim 10^{32}$\,erg change in
70: energy associated with the coronal electrical currents suffices to
71: power the flare and its associated coronal mass ejection.
72: \end{abstract}
73:
74: \keywords{Sun: activity, Sun: magnetic fields, Sun: flares, Sun: corona}
75:
76: \section{Introduction}
77: Solar flares and coronal mass ejections derive their energy from
78: electrical currents that run through the solar outer atmosphere. There
79: is growing evidence that the strong electrical currents involved in
80: major flaring tend to emerge embedded within magnetic flux after being
81: generated within the solar convection zone, rather than being induced
82: by the displacement of pre-existing magnetic flux by plasma flows on
83: the surface
84: \citep[e.g.][]{leka+etal1996,wheatland2000,demoulin+etal2002,demoulin+etal2002a,falconer+etal2002,leka+barnes2003a,leka+barnes2003,schrijver+etal2005a,wiegelmann+etal2005,regnier+canfield2005,jing+etal2006,schrijver2007}. The
85: high-resolution vector-magnetographic capabilities of the recently
86: launched Hinode satellite, and the advances in computational
87: capabilities to model and analyze the atmospheric magnetic fields
88: based on these surface field measurements, should enable us to make
89: significant advances in addressing this problem.
90:
91: Aside from flares and eruptive events, the magnetic field in the solar
92: corona evolves slowly as it responds to changes in the surface field,
93: implying that the electromagnetic Lorentz forces in this low-$\beta$
94: environment are relatively weak and that any electrical currents that
95: exist must be essentially parallel or anti-parallel to the magnetic
96: field wherever the field is not negligible. The problem of
97: determining the coronal field and its embedded electrical currents
98: thus leads to the problem of reconstructing the 3D magnetic field from
99: the observed boundary conditions, without having to deal with the
100: effects of plasma forces on that field.
101:
102: The vertical component of coronal electrical currents entering
103: the corona from below the photosphere can in principle
104: be deduced from the tangential components of the
105: vector-magnetic field at the base of the
106: corona. When combined with the vertical magnetic field component and
107: the condition that there are at most small
108: Lorentz forces (i.e., that the field is ``force free''), the resulting
109: model for the current-carrying coronal field is commonly referred to
110: as a nonlinear force-free (NLFF) field. Modeling such a field is in itself
111: a difficult problem that requires that a number of steps be taken successfully,
112: as outlined below \citep[see also, e.g.,][]{sakurai1989,mcclymont+etal1997,amari+etal1997}.
113:
114: First, the measured polarization signals need to be inverted to form a
115: vector-magnetogram, which requires detailed models of radiative
116: transport of polarized light through the solar atmosphere.
117:
118: Second, the procedure involves resolving an intrinsic 180$^\circ$
119: ambiguity in the components perpendicular to the line of sight, which
120: result from a degeneracy of the polarization properties. As the
121: electrical currents that penetrate the photosphere are carried by
122: compact flux tubes or potentially by fine structures within sunspots,
123: knowledge of the small-scale gradients in the field components
124: tangential to the solar surface is critical. Several procedures have
125: been developed to address this problem (see Section~2 for select
126: references), all of which involve a subjective choice, either on how
127: to deal with discontinuities interactively, or on what functional to
128: use in an automated iterative procedure.
129:
130: Third, the Lorentz forces at the base of the corona (caused by
131: buoyancy forces and drag forces from surface plasma flows) must be
132: dealt with, because formally the assumption that currents and field
133: are collinear is valid only above the lower chromosphere
134: \citep{metcalf+etal95}, where the plasma $\beta$ rapidly lies well
135: below unity at least within the strong-field core of active
136: regions. This step, often referred to as preprocessing, also involves
137: subjective choices about how the field may be modified to remove net
138: forces and torques while smoothing and tilting the observed vector
139: field. A parallel study by \citet{wiegelmann+etal2008} confirms the
140: expectation by \citet{metcalf+etal2007} based on a model test case
141: that successful preprocessing distorts the observed photospheric field to
142: an approximation of the chromospheric field.
143:
144: Fourth, the NLFF field computation requires a numerical code to
145: determine a coronal field that is compatible with the observed
146: boundary condition. This, too, involves choices about the iteration
147: scheme itself and about the application of boundary and initial
148: conditions. All of these affect the outcome, even in the case of
149: ``perfect knowledge,'' as shown by the tests performed by
150: \citet{schrijver+etal2005b} and \citet{metcalf+etal2007}.
151:
152: In a series of precursor studies, we have addressed the above set of
153: problems \citep[e.g.,][and references therein]{metcalf1994,wiegelmann+etal2005c,schrijver+etal2005b,amari+etal2005,metcalf+etal2007}.
154: We here proceed with an application of the developed methodology to
155: state-of-the-art solar observations.
156:
157: \section{Observations}
158: The Solar Optical Telescope (SOT) on board the Hinode spacecraft
159: \citep{kosugi+etl2007} observed AR\,10930 in the chromospheric
160: Ca\,II\,H channel and in the near-photospheric G band, while also
161: obtaining magnetogram sequences for over a week with near-continuous
162: coverage \citep[see Figs.\,\ref{fig:1} and
163: \ref{fig:2}; the region's evolution is also
164: described by, e.g.,][]{zhang+etal2007}. These and other observations
165: show that AR\,10930 exhibited only B- and C-class flares from
166: 2006/12/08 through 2006/12/13, when an X3.4 flare, peaking in soft
167: X-rays at 2:40\,UT, ended this relatively quiet period. At the time
168: of the flare, the region was at 23$^\circ$\,west and 5$^\circ$\,south
169: of disk center, and thus well positioned for vector-magnetographic
170: observations.
171:
172: If we look at this earliest phase in the flare, the very first
173: brightenings in the chromosphere (in Ca\,II\,H) are visible at
174: 02:04\,UT over a pair of converging concentrations of opposite sign in
175: the line-of-sight magnetic field (at position A in
176: Fig.\,\ref{fig:2}d). Twelve minutes later, three flare ribbons are
177: evident over opposite line-of-sight magnetic polarity
178: (Fig.\,\ref{fig:1}b): at position E and at the penumbral edges at
179: positions B and D in Fig.~\ref{fig:2}c). Such flare ribbons are
180: commonly interpreted as the sites where energetic flare particles
181: impact the lower atmosphere, thereby identifying the photospheric end
182: points of the field lines on which these particles are accelerated
183: during the energization of the flare. Hinode/XRT observed an early
184: bright soft X-ray feature over the first chromospheric brightening
185: (near E in Fig.\,\ref{fig:2}c). The east-west bright X-ray ridge
186: (below the center in Fig.\,\ref{fig:1}a) straddles the emerging
187: flux ends on the northern early flare ribbon near position B in
188: Fig.\,\ref{fig:2}c and on the southern one near position D around
189: 02:15\,UT. The overlying higher arched loops do not exhibit bright
190: ribbons until approximately 2:30\,UT (at positions C and near D in
191: Fig.\,\ref{fig:2}c); conversion of its excess energy into particle
192: kinetic energy apparently starts only some 25\,min after the first
193: impulsive energy conversion.
194:
195: In the days leading up to the X3.4 flare, the main changes in
196: AR\,10930 comprised a strong eastward motion of the smaller, southern
197: sunspot relative to its larger neighbor to the north.
198: Flux emergence between the two spots, as well as in the area
199: west of that, continued strongly from the early hours of
200: 2006/12/10 through the second half of 2006/12/14. Towards the end of
201: that period, the southern spot was moving rapidly eastward, while
202: multiple ridges of both polarities showed up between the northern and
203: southern spots, some even forming interpenumbral connections. Snapshots
204: of four Hinode magnetograms taken with the Narrow-band Filter Imager (NFI)
205: are shown in Fig.~\ref{fig:2}. Note that the NFI signal is highly
206: non-linear within the spot umbrae, where the very strong field causes the
207: signal to fade back to zero; this does not happen in the spectrum-based
208: vector-magnetographic SP data
209: (described below) that we use as input for our NLFFF methods.
210: We show a selection of
211: the NFI images here, because they are part of a movie (see below) with
212: magnetograms taken at a cadence of
213: two minutes, whereas there is an interval of 8\,h between the SP maps
214: before and after the flare.
215:
216: Figure\,\ref{fig:2} shows only a small area around the sunspots at
217: 4\,h intervals up to the start of the X3.4 flare. These panels suggest
218: relatively little change over time. The overall
219: appearance, in fact, changes so little
220: that we use the lower two panels to show the
221: detailed positions of the early flare ribbons and the modeled
222: electrical currents described in detail below. But the field is, in fact,
223: very dynamic: successive generations of ridges and concentrations of
224: flux form and disappear along the region between the spots and in the spots'
225: adjacent penumbrae. This evolution can be seen in an 8\,d movie in the
226: electronic addenda, with 2\,min between successive magnetograms (shown
227: rebinned to a 1,000 by 500 pixel movie at 1/4 of the full instrumental
228: resolution).\footnote{The movie is also available at \hbox{http://www.lmsal.com/$\sim$schryver/NLFFF/HinodeNFI$\_$X3.4.mov}. A smaller version covering a 2h\,h interval around the flare is also available.}
229:
230: The evolution of the emerging field between the spots is characterized
231: by the frequent occurrence of opposite-polarity ridges, either next to
232: each other, offset along their length, or separated by a strip of
233: weaker line-of-sight field. Such nearly parallel strands are a
234: characteristic signature of emerging flux bundles that carry currents
235: along their core \citep[see, e.g., Fig.\,2 in ][for simulations of an
236: emerging flux rope]{magara2006}. Such fibril electrical
237: currents cause the magnetic field to spiral about the axis of a flux
238: rope, so that when this flux rope breaches the solar surface, two
239: ridges are observed where the spiraling field points upward and
240: downward in two mostly parallel ridges in close proximity on either
241: side of the rope's axis. We return to this point in the discussion of
242: the model field.
243:
244: In addition to the filtergram sequences, the SOT Spectro-Polarimeter
245: (SP) obtained maps of the central regions of AR\,10930 before
246: and after the X3.4 flare. A pre-flare map was obtained between
247: 20:30\,UT and 21:15\,UT on 2006/12/12, and a subsequent post-flare
248: map between 3:40\,UT and 4:40\,UT on the next day; both have
249: 0.3\,arcsec pixels, and span 1024 steps with 512 pixels along the
250: north-south oriented slit.
251:
252:
253: These Hinode SP data were prepared by the standard ``SP$\_$PREP''
254: \citep{lites+etal2007a} available through SolarSoft, and the resulting
255: polarization spectra inverted to a vector magnetic field map
256: using an Unno-Rachkovsky inversion with a Milne-Eddington atmosphere
257: \citep{skumanich+lites1987,lites+skumanich1990,lites+etal1993}.
258: Pixels with a net polarization below
259: the threshold required for full inversion, but still with measureable Stokes
260: $V$, are treated in the following manner: a longitudinal flux density
261: is derived \citep[see, for example,][]{lites+etal2007b}, and assuming that the
262: field in these regions is in fact radially directed, the observing angle
263: allows the azimuths and inclinations to be determined in the observer's frame.
264: In this manner, less than 0.5\%\ of the pixels in the map are undetermined
265: and discontinuities due to the thresholds are minimized.
266:
267: The vector magnetic field maps are then subjected to an ambiguity
268: resolution approach which uses simulated annealing to minimize a
269: functional of the electric current density and divergence-free
270: condition \citep{metcalf1994}. The algorithm applied includes the
271: enhancements described in \citet{metcalf+etal2006}, and was the
272: top-performing automated method amongst those compared in that same
273: study. The ambiguity resolution appears to work successfully for most
274: of the field of view, and leaves only a few
275: very small patches where the transverse field changes
276: discontinuously. We cannot tell whether these patches are artifacts or
277: real, and consequently do not attempt to remove them from the
278: processed vector field. The vertical components of the SP
279: vector-magnetic field before and after the flare are shown in
280: Figs.\,\ref{fig:1}c and
281: \ref{fig:1}d, respectively. Maps of the vertical current density, $j_z$,
282: are shown in Figs.\,\ref{fig:1}e, and\,f; we note that the overall
283: currents are balanced to within 0.7\%\ in the field of view, while the
284: currents within each polarity reveal a net current between the field
285: polarities of approximately 20\%\ for positive $B_z$ and 15\%\ in the
286: negative polarity. The line-of-sight components of the central area as
287: measured by the NFI are shown enlarged in Fig.\,\ref{fig:2}. We note
288: that the polarity patterns seen in the filter-based NFI data match
289: those in the spectrum-based SP vector magnetograms (the latter are
290: available as addenda to the electronic edition of this
291: paper\footnote{FITS files are also available at
292: http://www.lmsal.com/$\sim$schryver/NLFFF/; the file contents is described
293: in the FITS header comment field.}), as expected for subsonic
294: flows affecting fields observed at 120\,m\AA\ away from the line
295: center position at 6302\,\AA.
296:
297: We embedded the Hinode/SP maps in a much larger,
298: lower-resolution SOHO/MDI \citep{soho} line-of-sight magnetogram in order
299: to incorporate information on flux outside the SP
300: map, subject to the current-free (potential) approximation.
301: In order to be able to apply the NLFF codes with
302: present computational resources, the data are rebinned $2\times 2$, to
303: 0.63\,arcsec pixels. A Green's function potential field
304: \citep{metcalf+etal2007} is computed for the entire expanded area
305: to serve as initial condition, and as side- and upper-boundary
306: conditions (where applicable for the methods; see \citet{metcalf+etal2007}).
307: The central area of $150 \times 150$\,Mm
308: ($203\times 203$\,arcsec, or $320\times 320$ pixels) is extracted for
309: modeling, together with the corresponding potential-field cube over
310: 256 vertical pixels.
311:
312: \section{NLFFF modeling}
313: We apply the four NLFF field algorithms described in
314: \citet{schrijver+etal2005b} and, where modified, by
315: \citet{metcalf+etal2007}: the weighted optimization algorithm by
316: \citet{wiegelmann2004}, the uniformly weighted optimization model as
317: implemented by McTiernan \citep{wheatland+etal2000}, the
318: magneto-frictional code by \citet{valori+etal2005}, and the
319: current-field interaction method by \citet[][and references
320: therein]{wheatland2006,wheatland2007}. For the Wheatland method,
321: three solutions are computed with different boundary conditions on
322: vertical current: currents are chosen from regions with positive
323: vertical field $B_z$; from regions with negative vertical field; and
324: based on an average of those two.
325:
326: Apart from working with the observed (disambiguated) vector fields, we
327: also apply
328: preprocessing to the lower boundary vector field. This
329: removes net magnetic forces and torques which should not exist in
330: the model. The preprocessing is performed with a method devised by
331: \citet{metcalf+etal2007}, which leaves $B_z$ unchanged, and
332: one by \citet{wiegelmann+etal2005c}, which allows all field components
333: to change. Both of these are applied with and without spatial
334: smoothing. The characteristic value of the rms difference between
335: observed and preprocessed values for $B_{x,y}$ are $\sim 15$\%\ of
336: the standard deviation in $B_{x,y}$.
337: A summary of these preprocessing algorithms
338: is included in \citet{metcalf+etal2007}.
339:
340: The work by \citet{metcalf+etal2007} showed that the additional step
341: of preprocessing the observed photospheric vector-magnetic data
342: resulted in a marked improvement in the agreement between the
343: resulting NLFFF extrapolations and their model reference field. They
344: argue that this
345: result likely stems from the fact that the preprocessed
346: photospheric field is a good approximation
347: of the corresponding chromospheric field, where the Lorentz forces
348: are much weaker. This is confirmed by \citet{wiegelmann+etal2008} in
349: a detailed evaluation of the
350: preprocessing process. As in the earlier study,
351: we find here that the best fits are obtained for preprocessed data.
352: We return to this topic in Section~4.
353:
354: Using these algorithms, we obtain 14 distinct model fields,
355: summarized in Table~I. We use
356: two different measures to identify the model that best fits the
357: observed corona, and one
358: that identifies the most internally consistent force-free field;
359: all to these identify the same model field as ``best.''
360:
361: A subjective goodness-of-fit is provided visually by comparing the
362: TRACE and Hinode/XRT images with the computed field lines, and
363: assessing the match for five characteristic signatures of the field,
364: labeled in Fig.\,\ref{fig:1}a as follows: 1) the sheared arcade
365: between the spots, 2) the eastern arch of loops around the southern
366: spot, 3) the low, nearly horizontal field west of the southern spot,
367: 4) the arcade high over that horizontal field, and 5) the absence of
368: shear around the emerging flux northwest of the northern spot. Table~I
369: shows the model fields ordered by the resulting metric $Q_{\rm m}$:
370: each good or poor correspondence contributes a bonus or penalty to the
371: metric of $+1$ or $-1$, respectively, while an ambiguous
372: correspondence is not weighed in the value of $Q_{\rm m}$. Only the
373: Wheatland positive-field solution ($Wh^+_{pp}$) applied to a
374: preprocessed lower boundary (including spatial smoothing) successfully
375: reproduces all five of these features.\footnote{Renderings of seven of
376: the NLFFF models and of the potential-field extrapolation
377: for the preprocessed pre-flare fields are shown in
378: the electronic addenda. These images are also available at
379: \hbox{http://www.lmsal.com/$\sim$schryver/NLFFF/}.}
380:
381: A second, objective measure for the goodness of fit is based on
382: finding field lines in the models that best match a set
383: of identified coronal loops in the Hinode/XRT and TRACE coronal
384: images, and computing the deviation between these in projection
385: against the solar disk. Coronal observations of AR\,10930 were made
386: both with Hinode's XRT using its thin-Be/open filter wheel setting,
387: and with TRACE \citep{traceinstrument} in its 195\,\AA\ passband, both
388: with 1\,arcsec pixels. In order to increase the signal-to-noise level,
389: we use the geometric mean of sets of exposures, as shown in
390: Fig.\,\ref{fig:1}a, for loop tracing. These loop traces are then
391: compared to 100 field lines computed for each of the model fields with
392: starting points distributed within the modeled volume along the line
393: of sight through the midpoint of each traced loop $i$. For each of these
394: field lines, we compute the area contained between the corresponding
395: loop trace and the field line projected against the plane of the sky;
396: this area $A_i$ is bounded by line segments that connect the ends of
397: the traced loops to the nearest points on the projected field
398: lines. The field lines with lowest value for $A_i$ are selected as the
399: best-fit field lines for each traced loop. The model field with the
400: lowest total value $\Sigma_i \min(A_i)$ for the set of traced loops is
401: identified as the best model. Formally, the best fitting model field
402: is again the $Wh^+_{pp}$ solution, but Fig.\,\ref{fig:1}a shows that
403: the correspondence between model field and corona is far from perfect
404: - we return to this issue in Section~4.
405:
406: A final criterion is based on the consistency of the model field with
407: the properties of a truly force-free field: the $Wh^+_{pp}$ solution
408: also has the lowest residual Lorentz forces (as measured by a
409: current-weighted angle between magnetic field and electrical current),
410: while being among the solutions with the lowest average absolute
411: divergence of the model field (both of these metrics should equal zero for a
412: perfect field). Table~I lists the values of these metrics. Note that
413: the metrics for field divergence are naturally grouped by NLFFF
414: algorithm type.
415:
416: In the present study, the Wheatland $Wh^+_{pp}$ model outperforms that of the
417: weighted-optimization algorithm by Wiegelmann that provided the best
418: solutions in our earlier two trial studies
419: \citep{schrijver+etal2005b,metcalf+etal2007}.
420: Based on this single application, we cannot identify why the
421: $Wh^+_{pp}$ model performs best. In fact, we note that the
422: current-field iteration procedure does not converge to a fixed
423: field for any of the various boundary conditions, but rather oscillates.
424: The code was run for 20 iterations in each case, at which stage the
425: field is still changing at successive iterations (on average by 2\%\
426: in the vector norm), although the energy was found to vary only slowly
427: over the final iterations. We note that this was also seen for
428: an analysis of test fields for a similar Grad-Rubin-based method
429: by \citet{amari+etal2005}. Despite this, the $Wh^+_{pp}$ model
430: is still the best model field. We expect that it performs best among
431: the other Wheatland solutions owing to the following property:
432: The Wheatland code requires that no electrical currents leave the
433: volume. As the lower-boundary vector field over-determines the
434: solution, the code uses the boundary current information either where
435: $B_z$ is positive or where it is negative, and then only where the
436: field strength exceeds 5\%\ of the maximum value and where the
437: horizontal field changes its azimuth by less than 120$^\circ$ between
438: neighboring pixels (this criterion is intended to limit the spurious
439: currents associated with incorrect disambiguations of the
440: perpendicular field component). The field lines with $B_z >0$ primarily emanate
441: from the smaller, southern spot, which mostly close within the volume
442: without major problems. This allows the $Wh^+_{pp}$ solution to
443: achieve a highest-fidelity representation of the real coronal field,
444: in marked contrast to the $Wh^-_{\rm pp}$ solution for $B_z < 0$, in
445: which the field is highly distorted.
446:
447: In a comparison with other methods, we note that the Wiegelmann and
448: McTiernan algorithms attempt to find a solution by minimizing a
449: functional that includes a non-zero divergence of the magnetic field
450: (and thus indirectly a source for electrical currents). Thus, magnetic
451: flux and electrical currents can originate within the model's volume,
452: in particular near the lower boundary where many of the currents of
453: interest run. Similarly, the Valori code can allow for higher values
454: in the field's divergence in the attempt of attain lower residual
455: Lorentz forces. The Wheatland code makes use of a vector potential,
456: and thus has an intrinsically low divergence (compare the values in
457: columns 5 and 8 in Table~I, see also \citet{wheatland2007}). Whereas
458: this yields a model field that best meets the formal requirement of
459: being force free, one might have expected that the artificial sources
460: would provide other methods additional freedom to find a better
461: matching field. But it apparently does not.
462:
463: We defer further discussion of the relative performance of the various
464: methods to a later study, in which the effects of different base
465: vector-magnetic fields and instrumental fields of view need to be
466: compared for a much larger sample of observations which, at present,
467: does not yet exist. We can add one observation on the potential effect
468: of the field of view. The $Wh_{\rm pp}^+$ algorithm was also applied
469: to a vector-field footprint extended westward to include the
470: relatively small amount of flux in the leading part of the region that
471: appears to be involved in the post-eruption arcade. This extension
472: makes the footprint twice as wide, and required lowering the model
473: volume by a factor of two to maintain the demand on computer
474: memory. The $Wh_{\rm pp}^+$ pre-flare model field for this extended
475: footprint is very similar to the model field discussed above, both in
476: terms of field lines and in terms of electrical currents, although the
477: free energy is lower by some 5\%. We conclude that our inferences
478: discussed below about the pre-flare field configuration are not
479: affected significantly by our choice of the field of view. The
480: $Wh_{\rm pp}^+$ model for the extended post-flare configuration did
481: not converge; that aspect will require substantial further study.
482:
483: \section{Discussion and conclusions}
484: The best-fit $Wh^+_{pp}$ model NLFF field suggests that, prior to the
485: flare, the strongest electrical currents run between the main spot
486: groups. These currents connect the penumbrae through filamentary
487: currents that arch up to $h\approx 12$\,Mm over
488: low-lying current strands that lie within $h\approx 6$\,Mm over
489: the emerging field between the spots. Not surprisingly, the bulk of
490: the free energy (i.e., the energy in excess of that of the
491: minimum-energy state given by the purely potential field) is
492: associated with these electrical currents (Figs.\,\ref{fig:1}c,\,d,
493: and~\ref{fig:2}). The main current strands connect footpoints that are
494: well separated, and that traverse a substantial distance over
495: and along the region of
496: emerging flux between the spots. They connect the emerged and emerging
497: flux adjacent to the southern spot to the umbra of the northern spot
498: (Fig.\,\ref{fig:2}c, and the top panels in Fig.\,\ref{fig:3}).
499:
500: In the best-fit $Wh^+_{pp}$ model, a low-lying, compact structure of
501: current-carrying emerging flux lies below the main current arcade in
502: the region containing opposite-polarity ridges of vertical field (see
503: the perspective volume renderings in Fig.\,\ref{fig:3}a, and also
504: Fig.~\ref{fig:4}c). This is the site at which the energy release in the
505: X3.4 flare starts its impulsive phase (see Fig.~\ref{fig:2}c,\,d).
506:
507: The flare appears to tap energy from most of the flux system that has
508: emerged over the preceding days: the currents in the relatively high
509: arching arcade are drastically weakened after the flare
510: (Fig.~\ref{fig:4}a,\,b), whereas the long fibril of concentrated
511: current low above the solar surface completely disappears
512: (Fig.~\ref{fig:4}c,\,e). We interpret the flare/CME as a cataclysmic
513: energy drain from a current-carrying flux rope that emerged from below
514: the photosphere. The complex field evolution with many mixed-polarity
515: ridges that is observed in the Hinode/NFI magnetogram sequence
516: suggests that the flux rope is either a bundle of smaller strands
517: wrapped about each other, or a single rope with internal structure
518: both in terms of field strength and current density. The 8\,h interval
519: between the SOT-SP vector magnetograms is far too long to address this
520: issue.
521:
522: With the $Wh^+_{pp}$ NLFFF model in hand, what can we say about
523: the topology of the pre-eruptive magnetic field?
524: The field line plots in Fig.~\ref{fig:3} can be interpreted as
525: a low altitude sheared arcade between the spots underneath an
526: essentially potential field that is nearly orthogonal to the arcade;
527: such a configuration has been described by, e.g.,
528: \citet{devore+antiochos2000}. When considering
529: the counterclockwise rotation of the southern, positive-polarity
530: sunspot evident in the NFI movie (see the electronic addenda) one may
531: alternatively infer an overall twisting of the magnetic field, as
532: has been observed, for example, by TRACE for the corona over some rotating
533: sunspots \citep[see, e.g.,][]{brown+etal2002a}. Several distinct
534: conceptual models exist for
535: such flux ropes. One of these is for ``included'' flux
536: ropes which have strong internal axial fields (that carry
537: at most weak currents) separated from
538: an external field by a current layer (this concept was used in the
539: study by \citet{metcalf+etal2007}. Another is for ``circuit'' flux
540: ropes that are produced by (thick or thin) net currents along their
541: axes surrounded by field lines that spiral about the current
542: \citep[e.g.,][]{titov+demoulin1999,torok+kliem2007}.
543: Then there are ``twisted'' flux ropes, that either may emerge as such
544: \citep[e.g.,][]{fan+gibson2004,amari+etal2004} or may form from photospheric
545: vortical motions
546: \citep[e.g.,][]{amari+etal1996,torok+kliem2003,aulanier+etal2005}.
547: These should show a large-scale bipolar direct current pattern
548: surrounded by external return currents (i.e., currents with opposite
549: direction within the same magnetic polarity); the existence of such
550: patterns continues to be debated \citep[see][and references
551: therein]{wheatland2000}.
552:
553: A potential discriminator for the applicability of any of these
554: concepts to AR\,10930 is the distribution of pre-eruptive photospheric
555: and coronal electric currents as deduced from the vector magnetograms
556: (Fig.~\ref{fig:1}e) and the NLFF field model (Fig.~\ref{fig:4}a, c, and\,e).
557: On the scale of the entire active region, the current pattern in
558: AR10930 bears a striking resemblance to that of ``twisted'' flux
559: ropes. Comparing the pre-flare column of Fig.\,\ref{fig:4}
560: with those of Figs.~9 and 10
561: in \citet{aulanier+etal2005}, we see (1) a swirling bipolar pattern
562: in $j_z$ in and immediately above the
563: photosphere, with
564: concentrations at the sunspots, and arching around the spots to form
565: parallel lanes of vertical
566: current located on both sides of the large-scale
567: neutral line between the spots; (2) narrower and fainter return currents
568: away from the main
569: neutral line on the edges of the current concentrations;
570: (3) an elongated strong patch in the coronal currents
571: right at the neutral line with essentially horizontal currents
572: and field. In contrast, the current pattern of
573: AR\,10930 does not match that expected for the other types of flux
574: ropes: sheared arcades are associated with parallel lanes of direct
575: and return currents, ``included'' ropes mostly show a single current
576: shell whose photospheric footpoints result in narrow parallel lanes,
577: and ``circuit'' flux ropes show unidirectional currents only.
578:
579: In this picture, the pre-eruptive compact current carrying structure,
580: at the location of the strong soft X-ray emission observed before and
581: during the flare, would simply trace the shortest twisted field lines
582: of a large-scale twisted flux rope. This is based on the expectation
583: that a given end-to-end twist applied to short, low field lines
584: results in stronger currents than when the same twist is applied to
585: longer, higher field lines
586: \citep{aulanier+etal2005}.
587:
588: The interesting similarity of our model field to theoretical models of
589: twisted flux ropes, while consistent with the evolution seen in the
590: NFI magnetograms, is encouraging, but will require confirmation beyond
591: this single example. The relative roles of pre-existing twist below
592: the photosphere and any added twist during the emergence associated
593: with plasma flows will need to be assessed. Moreover, the current pattern
594: shown in Fig.\,{fig:4} clearly shows a strong degree of filamentation,
595: which might have non-negligible consequences in terms of field
596: topology. This filamentation may be a consequence of a process that
597: is conducive to generating smaller scales (including, e.g., the
598: convective collapse upon emergence), or may in fact hold the key to
599: another formation process for the overall configuration.
600:
601: The best-fit $Wh^+_{pp}$ solutions for the pre- and post-flare
602: observations show a decrease in the free energy from 32\%\ of the
603: potential-field energy to a post-flare configuration with only 14\%\
604: in excess of the potential field model for the post-flare state
605: (Table~I). This corresponds to a drop in free energy of $3\times
606: 10^{32}$\,ergs, even as the total field energy in the potential field
607: extrapolation shows an increase by $\sim 10^{32}$\,ergs due to the
608: continuing emergence of flux in the 8\,h interval between the SOT-SP
609: maps. This decrease in free energy is adequate to power an X-class
610: flare (with energies of $10^{31-32}$\,ergs;
611: \citet{hudson1991,bleybel+etal2002}) associated with a coronal
612: mass ejection \citep[with energies up to about
613: $10^{32}$\,ergs;][]{hundhausen97}.
614:
615: We note that the continuing emergence of flux between the pre- and
616: post-flare SP maps not only causes the energy in the post-flare
617: potential field to be larger than the pre-flare potential field (see
618: last line in Table~I), but also causes most model fields to have
619: energy ratios $E/E_{\rm p,pre}$ that increase from before to after the
620: flare. Only the best-fit $Wh_{pp}^+$ model shows a drop in energy,
621: likely because its strong pre-flare currents lose more energy than the
622: flux emergence adds in this particular model field. Note that some
623: model fields have energies below that of the potential field; causes
624: for this anomaly were discussed for a test field by
625: \citet{metcalf+etal2007}.
626:
627: This exercise illustrates several problematic issues with the NLFFF
628: extrapolation process. First, the ``preprocessing'' of the observed
629: vector field yields a marked improvement in the quality of most of the
630: NLFF field models. The preprocessing is one way of reducing the
631: effect of the Lorentz forces acting in the photosphere; further
632: studies of how best to deal with these forces should be undertaken
633: \citep[along the lines of that for a model field by][]{wiegelmann+etal2008}.
634: Second, although provided with the same boundary conditions, only one
635: among seven model fields based on preprocessed boundary conditions
636: matches the observed coronal field acceptably by visual inspection
637: based on a limited set of key features (as measured by $Q_{\rm m}$ in
638: Table~I). This reveals the sensitivity to details in model
639: implementation and boundary and initial conditions also seen in our
640: two earlier studies \citep{schrijver+etal2005b,metcalf+etal2007}. An
641: integral part of the latter problem is, of course, the resolution of
642: the 180$^\circ$ ambiguity intrinsic to the measurement of the field
643: component perpendicular to the line of sight \citep[see the tests of
644: various methods based on known model fields described
645: by][]{metcalf+etal2006,li+etal2007}. It is clear that a systematic
646: study of these differences between the algorithms ranging from the
647: observations to the final NLFF field models, and their relative merits
648: and problems, is needed to identify one or a
649: few of the most successful strategies.
650:
651: And, finally, we note that even the best-fit $Wh^+_{pp}$ model
652: provides a rather poor match to the observed coronal loops. This may
653: be a consequence of the global nature of such loops: even if the field
654: locally would be modeled to within a few degrees, the integration
655: along the path of a field line could still lead to a markedly
656: different path relative to those in the true field and to observed
657: coronal loops. On the other hand, we note that the
658: poor match may be a consequence of an intrinsic problem of tracing
659: field lines by using coronal brightenings: \citet{aulanier+etal2005}, for
660: example, argue that coronal S-shaped structures associated with
661: twisted flux ropes may not trace individual field lines, but rather
662: are formed by an ensemble of partially outlined loops subject to
663: line-of-sight integration.
664:
665: Although we can measure the divergence of the field line
666: relative to the observed loops, there is unfortunately no known method
667: that quantifies how significant such path differences are in terms of
668: the field's total energy or helicity. Both of the latter quantities
669: are of interest to, say, space weather forecasters, but we have yet to
670: learn how the comparison of observed loops to model field lines can
671: be used to improve estimates of energy and helicity.
672:
673: On the positive side, (1) the best-fit $Wh^+_{pp}$ model field
674: compares relatively well with the observed corona, and (2) the energy
675: estimates and the pre-to-post flare energy difference suffices to
676: power the flare. Consequently, we think that the best available
677: vector-magnetographic data and NLFFF modeling techniques support our
678: main finding:
679:
680: We conclude that the filamentary electrical currents that emerge with
681: the magnetic flux between the main sunspots in AR\,10930 over a period
682: of up to several days (1) carry enough energy to power the observed
683: major X3.4 flare and associated coronal mass ejection, (2) are
684: involved in earliest impulsive phase of the flare, and (3) show a
685: substantial decrease in magnitude even as their associated field lines
686: connect over shorter distances when comparing pre- and post-flare
687: states. Thus, our application of nonlinear force-free field modeling
688: to state-of-the art vector-field data on a complex active region
689: provides strong evidence in support of the growing notion that major
690: solar flares are directly associated with the energy carried by
691: electrical currents that emerge from below the solar surface, and
692: is suggestive, at least in this case, of an emerging twisted flux rope.
693:
694: \acknowledgements
695: Tom Metcalf died in a skiing accident before this manuscript was
696: completed; we dedicate it to the memory of his friendship and
697: collegiality. Hinode is a Japanese mission developed and launched by
698: ISAS/JAXA, collaborating with NOAJ as domestic partner, NASA (USA) and
699: STFC (UK) as international partners. CJS, MLD, TRM, and GB were
700: supported by by Lockheed Martin Independent Research Funds. The work
701: of TW was supported by DLR grant 50 OC 0501 and that of JKT by DFG
702: grant WI 3211/1-1. GV was supported by DFG grant HO1424/9-1. GA and
703: PD gratefully acknowledge financial support by the European Commission
704: through the SOLAIRE Network (MTRN-CT-2006-035484). MSW acknowledges
705: generous support provided by the Observatoire de Paris enabling travel
706: to the workshop. The team thanks the CIAS of Observatoire de Paris
707: for hosting the NLFFF4 workshop.
708:
709: %\references
710: \begin{thebibliography}{}
711:
712: \bibitem[\protect\citeauthoryear{{Amari} {\em et~al.\/}}{1997}]{amari+etal1997}
713: {Amari}, T., {Aly}, J.~J., {Luciani}, J.~F., {Boulmezaoud}, T.~Z., {\&}
714: {Miki{\'c}}, Z.: 1997,
715: \newblock Sol. Phys. 174, 129
716:
717: \bibitem[\protect\citeauthoryear{{Amari} {\em et~al.\/}}{2006}]{amari+etal2005}
718: {Amari}, T., {Boulmezaoud}, T.~Z., {\&} {Aly}, J.~J.: 2006,
719: \newblock A{\&}A 446, 691
720:
721: \bibitem[\protect\citeauthoryear{{Amari} {\em et~al.\/}}{2004}]{amari+etal2004}
722: {Amari}, T., {Luciani}, J.~F., {\&} {Aly}, J.~J.: 2004,
723: \newblock ApJL 615, 165
724:
725: \bibitem[\protect\citeauthoryear{{Amari} {\em et~al.\/}}{1996}]{amari+etal1996}
726: {Amari}, T., {Luciani}, J.~F., {Aly}, J.~J., {\&} {Tagger}, M.: 1996,
727: \newblock ApJL 466, 39
728:
729: \bibitem[\protect\citeauthoryear{{Aulanier} {\em
730: et~al.\/}}{2005}]{aulanier+etal2005}
731: {Aulanier}, G., {D{\'e}moulin}, P., {\&} {Grappin}, R.: 2005,
732: \newblock A{\&}A 430, 1067
733:
734: \bibitem[\protect\citeauthoryear{{Bleybel} {\em
735: et~al.\/}}{2002}]{bleybel+etal2002}
736: {Bleybel}, A., {Amari}, T., {van Driel-Gesztelyi}, L., {\&} {Leka}, K.~D.:
737: 2002,
738: \newblock A{\&}A 395, 685
739:
740: \bibitem[\protect\citeauthoryear{Brown {\em et~al.\/}}{2002}]{brown+etal2002a}
741: Brown, D.~S., Nightingale, R.~W., Alexander, D., Schrijver, C.~J., Metcalf,
742: T.~R., Shine, R.~A., Title, A.~M., {\&} Wolfson, C.~J.: 2002,
743: \newblock in G. Tsiropoula, U. Sch{\"u}hle, and H. Sawaya-Lacoste (Eds.), {\em
744: Proceedings of the workshop on Magnetic Coupling of the Solar Atmosphere
745: (Santorini)\/}, ESA SP-505, ESA Publications Division, Noordwijk, The
746: Netherlands, p.~261
747:
748: \bibitem[\protect\citeauthoryear{{D{\' e}moulin} {\em
749: et~al.\/}}{2002a}]{demoulin+etal2002a}
750: {D{\' e}moulin}, P., {Mandrini}, C.~H., {Van Driel-Gesztelyi}, L., {Lopez
751: Fuentes}, M.~C., {\&} {Aulanier}, G.: 2002a,
752: \newblock Sol. Phys. 207, 87
753:
754: \bibitem[\protect\citeauthoryear{{D{\'e}moulin} {\em
755: et~al.\/}}{2002b}]{demoulin+etal2002}
756: {D{\'e}moulin}, P., {Mandrini}, C.~H., {Van Driel-Gesztelyi}, L., {Thompson},
757: B.~J., {Plunkett}, S., {Kov{\' a}ri}, Z., {Aulanier}, G., {\&} {Young}, A.:
758: 2002b,
759: \newblock A{\&}A 382, 650
760:
761: \bibitem[\protect\citeauthoryear{{DeVore} and
762: {Antiochos}}{2000}]{devore+antiochos2000}
763: {DeVore}, C.~R. {\&} {Antiochos}, S.~K.: 2000,
764: \newblock ApJ 539, 954
765:
766: \bibitem[\protect\citeauthoryear{{Falconer} {\em
767: et~al.\/}}{2002}]{falconer+etal2002}
768: {Falconer}, D.~A., {Moore}, R.~L., {\&} {Gary}, G.~A.: 2002,
769: \newblock ApJ 569, 1016
770:
771: \bibitem[\protect\citeauthoryear{{Fan} and {Gibson}}{2004}]{fan+gibson2004}
772: {Fan}, Y. {\&} {Gibson}, S.~E.: 2004,
773: \newblock ApJ 609, 1123
774:
775: \bibitem[\protect\citeauthoryear{Handy {\em et~al.\/}}{1999}]{traceinstrument}
776: Handy, B.~N., Acton, L.~W., Kankelborg, C.~C., Wolfson, C.~J., Akin, D.~J.,
777: Bruner, M.~E., Carvalho, R., Catura, R.~C., Chevalier, R., Duncan, D.~W.,
778: Edwards, C.~G., Feinstein, C.~N., Freeland, S.~L., Friedlander, F.~M.,
779: Hoffman, C.~H., Hurlburt, N.~E., Jurcevich, B.~K., Katz, N.~L., Kelly, G.~A.,
780: Lemen, J.~R., Levay, M., Lindgren, R.~W., Mathur, D.~P., Meyer, S.~B.,
781: Morrison, S.~J., Morrison, M.~D., Nightingale, R.~W., Pope, T.~P., Rehse,
782: R.~A., Schrijver, C.~J., Shine, R.~A., Shing, L., Strong, K.~T., Tarbell,
783: T.~D., Title, A.~M., Torgerson, D.~D., Golub, L., Bookbinder, J.~A.,
784: Caldwell, D., Cheimets, P.~N., Davis, W.~N., Deluca, E.~E., McMullen, R.~A.,
785: Amato, D., Fisher, R., Maldonado, H., {\&} Parkinson, C.: 1999,
786: \newblock Sol. Phys. 187, 229
787:
788: \bibitem[\protect\citeauthoryear{{Hudson}}{1991}]{hudson1991}
789: {Hudson}, H.~S.: 1991,
790: \newblock Sol. Phys. 133, 357
791:
792: \bibitem[\protect\citeauthoryear{Hundhausen}{1997}]{hundhausen97}
793: Hundhausen, A.~J.: 1997,
794: \newblock in J.~R. Jokipii, C.~P. Sonnett, and M.~S. Giampapa (Eds.), {\em
795: Cosmic winds and the heliosphere\/}, University of Arizona Press, Tucson,
796: Arizona, p.~259
797:
798: \bibitem[\protect\citeauthoryear{{Jing} {\em et~al.\/}}{2006}]{jing+etal2006}
799: {Jing}, J., {Song}, H., {Abramenko}, V., {Tan}, C., {\&} {Wang}, H.: 2006,
800: \newblock ApJ 644, 1273
801:
802: \bibitem[\protect\citeauthoryear{Kosugi {\em et~al.\/}}{2007}]{kosugi+etl2007}
803: Kosugi, T., Matsuzaki, K., Sakao, T., Shimizu, T., Sone, Y., Tachikawa, S.,
804: Hashimoto, T., Minesugi, K., Ohnishi, A., Yamada, T., Tsuneta, S., Hara, H.,
805: Ichimoto, K., Suematsu, Y., Shimojo, M., Watanabe, T., Shimada, S., Davis,
806: J.~M., Hill, L.~D., Owens, J.~K., Title, A.~M., Culhane, J.~L., Harra, L.~K.,
807: Doschek, G.~A., {\&} Golub, L.: 2007,
808: \newblock sp 243, 1
809:
810: \bibitem[\protect\citeauthoryear{{Leka} and {Barnes}}{2003a}]{leka+barnes2003a}
811: {Leka}, K.~D. {\&} {Barnes}, G.: 2003a,
812: \newblock ApJ 595, 1277
813:
814: \bibitem[\protect\citeauthoryear{{Leka} and {Barnes}}{2003b}]{leka+barnes2003}
815: {Leka}, K.~D. {\&} {Barnes}, G.: 2003b,
816: \newblock ApJ 595, 1296
817:
818: \bibitem[\protect\citeauthoryear{{Leka} {\em et~al.\/}}{1996}]{leka+etal1996}
819: {Leka}, K.~D., {Canfield}, R.~C., {McClymont}, A.~N., {\&} {van
820: Driel-Gesztelyi}, L.: 1996,
821: \newblock ApJ 462, 547
822:
823: \bibitem[\protect\citeauthoryear{{Li} {\em et~al.\/}}{2007}]{li+etal2007}
824: {Li}, J., {Amari}, T., {\&} {Fan}, Y.: 2007,
825: \newblock ApJ 654, 675
826:
827: \bibitem[\protect\citeauthoryear{{Lites} {\em et~al.\/}}{1993}]{lites+etal1993}
828: {Lites}, B.~W., {Elmore}, D.~F., {Seagraves}, P., {\&} {Skumanich}, A.~P.:
829: 1993,
830: \newblock ApJ 418, 928
831:
832: \bibitem[\protect\citeauthoryear{Lites {\em et~al.\/}}{2007a}]{lites+etal2007a}
833: Lites, B.~W., Ichimoto, K., Katsukawa, Y., Kubo, M., Shine, R.~A., Tarbell,
834: T.~D., Kamio, S., {\&} Shimojo, M.: 2007a,
835: \newblock in preparation
836:
837: \bibitem[\protect\citeauthoryear{{Lites} and
838: {Skumanich}}{1990}]{lites+skumanich1990}
839: {Lites}, B.~W. {\&} {Skumanich}, A.: 1990,
840: \newblock ApJ 348, 747
841:
842: \bibitem[\protect\citeauthoryear{Lites {\em et~al.\/}}{2007b}]{lites+etal2007b}
843: Lites, B.~W., Socas-Navarro, H., Berger, T., Frank, Z., Shine, R.~A., Tarbell,
844: T.~D., Title, A.~M., Ichimoto, K., Katsukawa, Y., {\&} Tsuneta, S.: 2007b,
845: \newblock ApJ,
846: \newblock in press
847:
848: \bibitem[\protect\citeauthoryear{{Magara}}{2006}]{magara2006}
849: {Magara}, T.: 2006,
850: \newblock ApJ 653, 1499
851:
852: \bibitem[\protect\citeauthoryear{{McClymont} {\em
853: et~al.\/}}{1997}]{mcclymont+etal1997}
854: {McClymont}, A.~N., {Jiao}, L., {\&} {Miki{\'c}}, Z.: 1997,
855: \newblock Sol. Phys. 174, 191
856:
857: \bibitem[\protect\citeauthoryear{{Metcalf}}{1994}]{metcalf1994}
858: {Metcalf}, T.~R.: 1994,
859: \newblock Sol. Phys. 155, 235
860:
861: \bibitem[\protect\citeauthoryear{Metcalf {\em
862: et~al.\/}}{2007}]{metcalf+etal2007}
863: Metcalf, T.~R., DeRosa, M.~L., {Schrijver}, C.~J., Barnes, G., {Van
864: Ballegooijen}, A.~A., Wiegelmann, T., Wheatland, M.~S., , Valori, G., {\&}
865: McTiernan, J.: 2007,
866: \newblock Sol. Phys.,
867: \newblock submitted
868:
869: \bibitem[\protect\citeauthoryear{Metcalf {\em et~al.\/}}{1995}]{metcalf+etal95}
870: Metcalf, T.~R., Jiao, L., McClymont, A.~N., Alexander, N., Canfield, R.~C.,
871: {\&} Uitenbroek, H.: 1995,
872: \newblock ApJ 439, 474
873:
874: \bibitem[\protect\citeauthoryear{{Metcalf} {\em
875: et~al.\/}}{2006}]{metcalf+etal2006}
876: {Metcalf}, T.~R., {Leka}, K.~D., {Barnes}, G., {Lites}, B.~W., {Georgoulis},
877: M.~K., {Pevtsov}, A.~A., {Balasubramaniam}, K.~S., {Gary}, G.~A., {Jing}, J.,
878: {Li}, J., {Liu}, Y., {Wang}, H.~N., {Abramenko}, V., {Yurchyshyn}, V., {\&}
879: {Moon}, Y.-J.: 2006,
880: \newblock Sol. Phys. 237, 267
881:
882: \bibitem[\protect\citeauthoryear{{R{\'e}gnier} and
883: {Canfield}}{2006}]{regnier+canfield2005}
884: {R{\'e}gnier}, S. {\&} {Canfield}, R.~C.: 2006,
885: \newblock A{\&}A 451, 319
886:
887: \bibitem[\protect\citeauthoryear{{Sakurai}}{1989}]{sakurai1989}
888: {Sakurai}, T.: 1989,
889: \newblock Sol. Phys. 121, 347
890:
891: \bibitem[\protect\citeauthoryear{Scherrer {\em et~al.\/}}{1995}]{soho}
892: Scherrer, P.~H., Bogart, R.~S., Bush, R.~I., Hoeksema, J.~T., Kosovichev,
893: A.~G., Schou, J., Rosenberg, W., Springer, L., Tarbell, T.~D., Title, A.,
894: Wolfson, C.~J., Zayer, I., {\&} {The MDI Engineering Team}: 1995,
895: \newblock Sol. Phys. 162, 129
896:
897: \bibitem[\protect\citeauthoryear{{Schrijver}}{2007}]{schrijver2007}
898: {Schrijver}, C.~J.: 2007,
899: \newblock ApJL 655, 117
900:
901: \bibitem[\protect\citeauthoryear{{Schrijver} {\em
902: et~al.\/}}{2005}]{schrijver+etal2005a}
903: {Schrijver}, C.~J., {De Rosa}, M.~L., Title, A.M., {\&} {Metcalf}, T.~R.: 2005,
904: \newblock ApJ 628, 501
905:
906: \bibitem[\protect\citeauthoryear{{Schrijver} {\em
907: et~al.\/}}{2006}]{schrijver+etal2005b}
908: {Schrijver}, C.~J., DeRosa, M.~L., Metcalf, T.~R., Liu, Y., McTiernan, J.,
909: R{\'e}gnier, S, Valori, G., Wheatland, M.~S., {\&} Wiegelmann, T.: 2006,
910: \newblock Sol. Phys. 235, 161
911:
912: \bibitem[\protect\citeauthoryear{{Skumanich} and
913: {Lites}}{1987}]{skumanich+lites1987}
914: {Skumanich}, A. {\&} {Lites}, B.~W.: 1987,
915: \newblock ApJ 322, 483
916:
917: \bibitem[\protect\citeauthoryear{{Titov} and
918: {D{\'e}moulin}}{1999}]{titov+demoulin1999}
919: {Titov}, V.~S. {\&} {D{\'e}moulin}, P.: 1999,
920: \newblock A{\&}A 351, 707
921:
922: \bibitem[\protect\citeauthoryear{{T{\"o}r{\"o}k} and
923: {Kliem}}{2003}]{torok+kliem2003}
924: {T{\"o}r{\"o}k}, T. {\&} {Kliem}, B.: 2003,
925: \newblock A{\&}A 406, 1043
926:
927: \bibitem[\protect\citeauthoryear{{T{\"o}r{\"o}k} and
928: {Kliem}}{2007}]{torok+kliem2007}
929: {T{\"o}r{\"o}k}, T. {\&} {Kliem}, B.: 2007,
930: \newblock Astronomische Nachrichten 328, 743
931:
932: \bibitem[\protect\citeauthoryear{{Valori} {\em
933: et~al.\/}}{2005}]{valori+etal2005}
934: {Valori}, G., {Kliem}, B., {\&} {Keppens}, R.: 2005,
935: \newblock A{\&}A 433, 335
936:
937: \bibitem[\protect\citeauthoryear{{Wheatland}}{2000}]{wheatland2000}
938: {Wheatland}, M.~S.: 2000,
939: \newblock ApJ 532, 616
940:
941: \bibitem[\protect\citeauthoryear{{Wheatland}}{2006}]{wheatland2006}
942: {Wheatland}, M.~S.: 2006,
943: \newblock Sol. Phys. 238, 29
944:
945: \bibitem[\protect\citeauthoryear{{Wheatland}}{2007}]{wheatland2007}
946: {Wheatland}, M.~S.: 2007,
947: \newblock Sol. Phys.,
948: \newblock in press
949:
950: \bibitem[\protect\citeauthoryear{{Wheatland} {\em
951: et~al.\/}}{2000}]{wheatland+etal2000}
952: {Wheatland}, M.~S., {Sturrock}, P.~A., {\&} {Roumeliotis}, G.: 2000,
953: \newblock ApJ 540, 1150
954:
955: \bibitem[\protect\citeauthoryear{{Wiegelmann}}{2004}]{wiegelmann2004}
956: {Wiegelmann}, T.: 2004,
957: \newblock Sol. Phys. 219, 87
958:
959: \bibitem[\protect\citeauthoryear{{Wiegelmann} {\em
960: et~al.\/}}{2006}]{wiegelmann+etal2005c}
961: {Wiegelmann}, T., {Inhester}, B., {\&} {Sakurai}, T.: 2006,
962: \newblock Sol. Phys. 233, 215
963:
964: \bibitem[\protect\citeauthoryear{{Wiegelmann} {\em
965: et~al.\/}}{2005}]{wiegelmann+etal2005}
966: {Wiegelmann}, T., {Lagg}, A., {Solanki}, S.~K., {Inhester}, B., {\&} {Woch},
967: J.: 2005,
968: \newblock A{\&}A 433, 701
969:
970: \bibitem[\protect\citeauthoryear{{Wiegelmann} {\em
971: et~al.\/}}{2007}]{wiegelmann+etal2008}
972: {Wiegelmann}, T., Thalmann, J.~K., Schrijver, C.~J., DeRosa, M.~L., {\&}
973: Metcalf, T.~R.: 2007,
974: \newblock Sol. Phys., submitted
975:
976: \bibitem[\protect\citeauthoryear{{Zhang} {\em et~al.\/}}{2007}]{zhang+etal2007}
977: {Zhang}, J., {Li}, L., {\&} {Song}, Q.: 2007,
978: \newblock ApJL 662, 35
979:
980: \end{thebibliography}
981:
982: \clearpage
983: \begin{table}
984: \caption{Metrics for the field extrapolations, in order of quality Q
985: based on the visual correspondence to the coronal pre-flare image.}
986: \begin{tabular}{l|r|lcr|lcr}\hline
987: \label{table:metrics}
988: &&\multicolumn{3}{c}{pre-flare: 2006/12/12} & \multicolumn{3}{c}{post-flare: 2006/12/13} \\
989: Model\tablenotemark{1} &$Q_{\rm m}$ \tablenotemark{2}& $E/E_{\rm p,pre}$ \tablenotemark{3} & ${\rm CW}\sin$ \tablenotemark{4}& $\langle |f_i|\rangle\times10^8$ \tablenotemark{5}&
990: $E/E_{\rm p,pre}$& ${\rm CW}\sin$& $\langle |f_i|\rangle\times10^8$\\
991: \hline
992:
993: 1: $Wh^+_{pp}$&5& 1.32& 0.24& 3.6 & 1.19& 0.18& 2.0\\
994:
995: 2: $Wh^+_{np}$&3& 1.10& 0.27& 3.9 & 1.23& 0.27& 4.6\\
996: 3: $Wie_{wp}$&3& 1.09& 0.35& 19. & 1.18& 0.32& 13.\\
997: 4: $Val_{pp}$&3& 1.10& 0.28& 230.& 1.27& 0.31& 190.\\
998:
999: 5: $Wh^0_{pp}$&2& 1.04& 0.28& 3.0 & 1.53& 0.27& 3.7\\
1000: 6: $Wie_{ns}$&2& 1.04& 0.43& 22. & 1.13& 0.39& 30.\\
1001: 7: $Val_{np}$&2& 0.88& 0.29& 220.& 0.99& 0.34& 210.\\
1002:
1003: 8: $Wie_{np}$&1& 0.95& 0.43& 24. & 1.04& 0.39& 27.\\
1004:
1005: 9: $Wie_{pp}$&0& 1.05& 0.44& 18. & 1.15& 0.39& 21.\\
1006: 10: $McT_{pp}$&0& 1.01& 0.61& 29. & 1.07& 0.59& 25.\\
1007:
1008: 11: $Wh^0_{np}$&-1& 1.03& 0.27& 2.5 & 1.12& 0.23& 2.6\\
1009: 12: $Wh^-_{np}$&-1& 1.04& 0.25& 2.9 & 1.11& 0.24& 2.9\\
1010: 13: $Wh^-_{pp}$&-1& 1.05& 0.27& 3.2 & 1.16& 0.19& 2.2\\
1011:
1012: 14: $McT_{np}$&-2& 0.95& 0.64& 26. & 1.00& 0.61& 24.\\
1013:
1014: 15: pot'l &-3& 1& - & 0.8 & 1.04& -& 0.8\\
1015: \hline
1016: \end{tabular}
1017: \tablenotetext{1}{Models: Wh: Wheatland, Wie: Wiegelmann, Val: Valori, McT:
1018: McTiernan; $+,-,0$: based on positive or negative $B_z$, or both,
1019: respectively; np: no preprocessing; ns: preprocessed without
1020: smoothing, pp: preprocessing including smoothing; wp: Wiegelmann's
1021: preprocessing and smoothing. }
1022: \tablenotetext{2}{Quality of fit by visual
1023: inspection for five features: a good/poor correspondence for each
1024: feature adds $+1,-1$, respectively, to the total value; an ambiguous
1025: correspondence adds $0$. }
1026: \tablenotetext{3}{Energy, relative to the energy in the
1027: pre-flare potential field model. }
1028: \tablenotetext{4}{Current-weighted value of
1029: $\sin{\theta}$, where $\theta$ is the angle between the electrical
1030: current and the magnetic field in the model solution. }
1031: \tablenotetext{5}{The
1032: unsigned mean over all pixels $i$ in the comparison volume of the
1033: absolute fractional flux change $|f_i| = {{|(\nabla\cdot {\bf
1034: B})_i}|/{(6|{\bf B}|_i/\Delta x) }}$, where $\Delta x$ is the grid
1035: spacing
1036: \citep[compare][]{wheatland+etal2000}.}
1037: \end{table}
1038: \clearpage
1039:
1040:
1041: \figcaption{The chromosphere, corona, and magnetic field of NOAA
1042: AR\,10930. Panel {\em a} shows the $320 \times 320$ pixel footprint of
1043: the full model volume for the NLFFF codes (itself surrounded by a much
1044: larger skirt of a line-of-sight magnetic map). The largest square is
1045: the $224\times 224$ pixel area (with sides of 101\,Mm) used for the
1046: energy estimates, and shown in panels {\em b--d} and in
1047: Figs.~\ref{fig:3} and\,\ref{fig:4}. The smaller square shows the
1048: footprint shown in Fig.~\ref{fig:2}. Panels: {\em a)} time-averaged
1049: Hinode/XRT soft X-ray image. Individual loops traced on this image or
1050: on a TRACE 195\,\AA\ image (not shown here) are represented in green;
1051: the best-matching model field lines for the $Wh^+_{\rm pp}$ model
1052: field are shown in purple. Numbers identify the characteristic
1053: signatures in the field that were used in the subjective assessment of
1054: the goodness-of-fit, as discussed in Section~3. {\em b)}
1055: Chromospheric Ca\,II\,H image for the onset phase of the flare. {\em
1056: c)} Pre-flare Hinode/SOT-SP $B_z$ on 2006/12/12 around
1057: 21\,UT. Contours show the vertically integrated energy density in the
1058: $Wh^+_{\rm pp}$ field minus that of the potential field; the white
1059: contour lies near the maximum value and the green contour at half
1060: that. {\em d)} Post-flare Hinode/SOT-SP $B_z$ on 2006/12/13 around
1061: 4\,UT. Contours as in panel {\em b}. {\em e, f)} Maps of the vertical
1062: current density $j_z$ corresponding to the pre- and post-flare maps
1063: shown in panels {\em c} and {\em d}, respectively. The color scale for
1064: the lower four panels runs from blue (negative) to red (positive),
1065: saturating into black or white, respectively. \label{fig:1}}
1066:
1067: \figcaption{Time series of magnetograms showing the evolution of
1068: the line-of-sight magnetic field as observed by the HINODE Solar
1069: Optical Telescope with the Narrow-Band Filter Imager (NFI) at 4\,h
1070: intervals prior to the X3.4 flare. Coordinates (with north up and
1071: -~by solar-physics convention~- west towards the right) are in pixels
1072: of 0.16\,arcsec each; the area shown covers the central 10\%\ of the
1073: footprint of the NLFF field model volume (compare the smallest square
1074: in Fig.\,1a to its full field of view). Note that the NFI magnetograph
1075: signal is non-monotonic, disappearing in the umbrae of the two spots.
1076: Panel {\em d} shows in red the brightest segments of the flare ribbons
1077: seen in the Ca\,II\,H channel at 02:16:39\,UT (cf.,
1078: Fig.\,\ref{fig:1}d); it also outlines the brightest quiescent kernel
1079: in soft X-rays seen by the Hinode X-ray telescope at the same time
1080: (green contour) and the brightest coronal structure (black contour;
1081: cf., Fig.\,\ref{fig:1}a). The latter two are repeated in panel {\em
1082: c}, taken close to the time of the pre-flare vector-magnetogram
1083: obtained by the spectro-polarimeter. The magnetic information in that
1084: panel has been replaced by the line-of-sight integral of the
1085: vector norm of the
1086: electrical currents (cf., Fig.~\ref{fig:4}a) within the blue contour,
1087: which is where these currents are strongest.\label{fig:2}}
1088:
1089: \figcaption{Visualizations of the magnetic field over NOAA
1090: Active Region 10930 before (top) and after (bottom) the X3.4 flare,
1091: shown against the corresponding map of $B_z$. Sample field lines
1092: outline the field; white field lines close within the NLFF model
1093: volume, while colored field lines (purple or green for the two
1094: polarities of $B_z$ at their base) leave that volume to connect to
1095: more distant regions. The rendered volumes (red) show where the
1096: electrical current densities are highest, using the same threshold
1097: level in both panels (cf., Fig.\,\ref{fig:2}c). The compact, low
1098: current system below the large, high-arching currents in the top panel
1099: corresponds to the site (position A in Fig.\,\ref{fig:2}d) of the
1100: initial brightenings of the X3.4 flare and associated coronal mass
1101: ejection.\label{fig:3}}
1102:
1103: \figcaption{{\em a,\,b)} Vertically integrated electrical
1104: currents in the $Wh^+_{\rm pp}$ pre-flare (left) and post-flare
1105: (right) model fields, on the same color scale. {\em c,\,d)} As panels
1106: {\em a,\,b}, but showing the horizontal currents integrated over the
1107: lowest 6,100\,km (12 pixels), using the same grey scale. {\em e,\,f)}
1108: As panels {\em c,\,d} for $j_z$, with white/black for
1109: positive/negative $j_z$, respectively.\label{fig:4}}
1110:
1111:
1112: \clearpage
1113:
1114: \begin{figure}
1115: \epsscale{.8}
1116: \plotone{f1.eps}
1117: \epsscale{1}
1118: \centerline{f1.eps}
1119: \end{figure}
1120: \clearpage
1121:
1122: \begin{figure}
1123: \plotone{f2.ps}
1124: \centerline{f2.ps}
1125: \end{figure}
1126: \clearpage
1127:
1128: \begin{figure}
1129: \centerline{\psfig{figure=f3a.eps,width=7.9truecm,bbllx=58bp,bblly=63bp,bburx=357bp,bbury=302bp,clip=} \psfig{figure=f3b.eps,width=7.9truecm,bbllx=58bp,bblly=5bp,bburx=357bp,bbury=244bp,clip=}}
1130: \vskip 0.1truecm
1131: \centerline{\psfig{figure=f3c.eps,width=7.9truecm,bbllx=49bp,bblly=63bp,bburx=348bp,bbury=302bp,clip=} \psfig{figure=f3d.eps,width=7.9truecm,bbllx=58bp,bblly=5bp,bburx=357bp,bbury=244bp,clip=}}
1132: \centerline{f3[a,b,c,d].eps}
1133: \end{figure}
1134: \clearpage
1135:
1136: \begin{figure}
1137: \plotone{f4.ps}
1138: \centerline{f4.ps}
1139: \end{figure}
1140:
1141: \end{document}
1142:
1143: \end{document}
1144: \begin{figure}[ht!]
1145: %\centerline{\psfig{figure=sixpanel2.eps,width=15cm,clip=}}
1146: \centerline{\psfig{figure=f1.eps,width=15cm,clip=}}
1147: \caption{\null}\label{fig:1}
1148: \end{figure}
1149:
1150: \begin{figure}[ht!]
1151: %\centerline{\psfig{figure=sotcomposite.ps,width=\textwidth,clip=}}
1152: \centerline{\psfig{figure=f2.ps,width=\textwidth,clip=}}
1153: \caption{\null}\label{fig:2}
1154: \end{figure}
1155:
1156: \begin{figure}[ht!]
1157: %\centerline{\psfig{figure=currvol_snapshot_karel_preflare_top.eps,width=7.9truecm,bbllx=58bp,bblly=63bp,bburx=357bp,bbury=302bp,clip=} \psfig{figure=currvol_snapshot_karel_preflare.eps,width=7.9truecm,bbllx=58bp,bblly=5bp,bburx=357bp,bbury=244bp,clip=}}
1158: %\vskip 0.1truecm
1159: %\centerline{\psfig{figure=currvol_snapshot_karel_postflare_top.eps,width=7.9truecm,bbllx=49bp,bblly=63bp,bburx=348bp,bbury=302bp,clip=} \psfig{figure=currvol_snapshot_karel_postflare.eps,width=7.9truecm,bbllx=58bp,bblly=5bp,bburx=357bp,bbury=244bp,clip=}}
1160: \centerline{\psfig{figure=f3a.eps,width=7.9truecm,bbllx=58bp,bblly=63bp,bburx=357bp,bbury=302bp,clip=} \psfig{figure=f3b.eps,width=7.9truecm,bbllx=58bp,bblly=5bp,bburx=357bp,bbury=244bp,clip=}}
1161: \vskip 0.1truecm
1162: \centerline{\psfig{figure=f3c.eps,width=7.9truecm,bbllx=49bp,bblly=63bp,bburx=348bp,bbury=302bp,clip=} \psfig{figure=f3d.eps,width=7.9truecm,bbllx=58bp,bblly=5bp,bburx=357bp,bbury=244bp,clip=}}
1163: \caption{\null}\label{fig:3}
1164: \end{figure}
1165:
1166: \begin{figure}[ht!]
1167: %\centerline{\psfig{figure=compare_currents.ps,width=\textwidth,bbllx=194bp,bblly=394bp,bburx=428bp,bbury=700bp,clip=}}
1168: \centerline{\psfig{figure=f4.ps,width=\textwidth,bbllx=194bp,bblly=394bp,bburx=428bp,bbury=700bp,clip=}}
1169: \caption{\null}\label{fig:4}
1170: \end{figure}
1171:
1172: \end{document}
1173: