An Erupting Filament and Associated CME Observed by Hinode, STEREO and SOHO

Transcription

1 **FULL TITLE** ASP Conference Series, Vol. **VOLUME**, **YEAR OF PUBLICATION** **NAMES OF EDITORS** An Erupting Filament and Associated CME Observed by Hinode, STEREO and SOHO A. Bemporad INAF-Torino Astronomical Obs., via Osservatorio 20, 25 Pino Torinese (TO), Italy G. Del Zanna Mullard Space Science Lab., University College London, Holmbury St. Mary, Dorking, Surrey RH5 6NT, UK V. Andretta and M. Magrí INAF-Capodimonte Astronomical Obs., salita Moiariello 16, Napoli, Italy G. Poletto INAF-Arcetri Astrophysical Obs., L.go E. Fermi 5, Firenze, Italy Y.-K. Ko Naval Research Lab., 4555 Overlook Ave. S.W., Washington, DC Abstract. A multi-spacecraft campaign was set up in May 7 to observe the off-limb corona with Hinode, STEREO and SOHO instruments (Hinode HOP 7). During this campaign, a filament eruption and a coronal mass ejection (CME) occurred on May 9 from NOAA at the West limb. The filament eruption starts around 13:40 UT and results in a CME at 4 SW latitude. Remarkably, the event was observed by STEREO (EUVI and COR1) and by the Hinode/EIS and SOHO/UVCS spectrometers. We present results from all these instruments. High-cadence data from Stereo/EUVI A and B in the He ii λ304 line were used to study the 3-D expansion of the filament. A slow rising phase, during which the filament moved southward, was followed by an impulsive phase during which the filament appeared to change direction and then contribute to the westward-expanding CME as seen in STEREO/COR 1. Hinode/EIS was scanning with the 2 slit the region where the filament erupted. The EIS spectra show remarkable non-thermal broadening in lines emitted at different temperatures at the location of the filament eruption. The CME was also observed by the SOHO/UVCS instrument: the spectrograph slit was centered at 1.7 solar radii, at a latitude of 5 SW and recorded a sudden increase in the O VI λλ and Si xii λ520 spectral line intensities. We discuss the overall morphology of this interesting eruptive event, and provide a preliminary assessment of its temperature and density structure. 1

2 2 Bemporad et al. Figure 1. Left: the May 9, 7 eruption as seen by the EUVI (He ii λ304 line) and COR1 (H-α λ6565 line) telescopes aboard the STEREO B spacecraft; the dashed box marks the FOV covered by the EIS raster, while the solid line shows the position of the UVCS slit FOV centered at 1.7 R, at a latitude of 5 SW. Right: a comparison between two EUVI He ii λ304 images of the erupting prominence acquired at the same time (13:52 UT) from the STEREO A and B spacecrafts. These two images have been used to derive information on the 3-D structure of the filament (see text). 1. Introduction Between May 7 10, 7 a Hinode TRACE SOHO campaign (Hinode HOP 7) was running, aimed at measuring electron temperatures, densities, and elemental abundances in the low corona at different altitudes and latitudes above the Active Region (AR) NOAA 10953, that was crossing the West solar limb in these days (see also Del Zanna et al. 8, in this volume for more detailed information). This campaign involved in particular the Hinode EUV Imaging Spectrometer (EIS; see Culhane et al. 7) and the SOHO Ultraviolet Coronagraph Spectrometer (UVCS; see Kohl et al. 1995). On May 9, 7 a prominence eruption occurred: the ejected plasma crossed, during its propagation, both the EIS and UVCS fields of view (FOV; see Fig. 1.) and resulted finally in a slow Coronal Mass Ejection (CME). Unfortunately, at this time the source AR was located 14 behind the West solar limb, so there is no information on its pre-eruptive evolution. The present study focuses on the physical parameters of the erupting plasma as derived from EIS and UVCS spectra. The eruption has been also observed by the two Solar Terrestrial Relations Observatory (STEREO; see Kaiser et al. 8) A and B SECCHI EUV Imagers (EUVI; see Wuelser et al. 4): these data have been used to derive information on the 3-D structure of the erupting prominence. After illustrating the May 9, 7 event ( 2), we describe the first results we derived from EUVI, EIS and UVCS data ( 3); conclusions are summarized in 4.

3 Erupting Filament Observed by Hinode, STEREO and SOHO 3 Figure 2. The erupting filament material and the surrounding corona as seen by EIS in the He ii λ (left) and Fe xii λ (right) spectral line intensities. 2. The May 9, 7 Filament Eruption Between April 25 and May 8, 7 the βγδ AR NOAA 10953, located at a latitude of 11 S, crosses the disk dragged by solar rotation. Images acquired in the He ii λ304 line by the two EUVI telescopes show that, even if no major flares occur, the coronal loops above the AR are highly unstable. Chromospheric material is continuously ejected from the AR in a sequence of small homologous eruptions that were unable to result in a CME. On May 9, 7, when the AR is already behind the solar limb, one of these eruptions results in a CME: snapshots of this event are shown in Fig. 1. The high cadence ( 37s) images acquired on that day by the EUVI instruments in the HeII λ304 line (see Fig. 1) show a tongue of plasma, anchored at an approximate latitude of 24 S, that starts to expand southward around 13:40 UT. Over the following minutes the prominence progressively accelerates and changes its direction of propagation resulting in a slow (v CME 310 km s 1 ), decelerating (a CME 7.4 m s 2 ) CME (see the LASCO CME catalog, list/) that propagates around a latitude of 26 S (Fig. 1, panels a d). The eruption crossed both the EIS and the UVCS FOV and was observed also by EUVI in the He ii λ304 line. 3. Analysis of EUVI, EIS and UVCS Data Thanks to the very high spatial resolution ( 1.5 /pixel) of the EUVI telescopes it is possible to perform triangulation studies by identifying the same features (e.g. bright He ii knots) in pair of frames acquired at the same time (Fig. 1) and to study the 3-D structure and expansion of the filament. It turns out that, at 13:52 UT, the filament is mostly a 2-D, hook shaped structure with an average thickness along the line of sight of 0.098R and a length of 0.43R ; there is no evidence for any 3-D flux rope shape. By identifying the same EUV features in successive frame pairs, it has been possible also to study the 3-D prominence expansion. Results show that over the following 20 minutes the filament expands in 3-D undergoing not only a strong radial acceleration a CME = (180 ± 80)m s 2, but also a 3 times larger tangential acceleration, leading to the mentioned deflection and change in the direction of propagation.

4 4 Bemporad et al. The EIS raster acquired starting from 13:15 UT sampled the final part of the eruption, i.e. just a few minutes after the transit of the main part of the filament. The resulting EIS intensity maps (Fig. 2) show enhanced emission of the chromospheric erupting material only in the cooler spectral lines such as He ii λ and Fe viii λ (with maximum formation temperature of T K and T K, respectively), while no emission was detected in the hotter lines such as Fe xii λ and Fe xv λ (T K and T K, respectively), implying that, as expected, the prominence plasma is at most a temperature of 10 5 K. Interestingly, at the position of the ejected material, all the observed spectral lines show strong non-thermal line broadenings. In particular, once the instrumental and thermal broadenings are subtracted, it turns out that for instance the plasma emitting in the Fe xii λ line has non-thermal velocities up to 120 km s 1 (i.e. a kinetic temperature K, far in excess of the ion kinetic temperature). This effect seems to decrease for increasing iron ionization stages. Because the erupting material is not emitting in the Fe xii line, this non-thermal broadening has to be ascribed to the surrounding corona. The UVCS data show, during the CME transit, a relatively faint ( 20%) increase in the O vi λλ doublet lines and a smaller increase also in the Si xii λ520.6 line; the O vi λ intensity map shows that this increase is due to the transit of the CME front across the slit. This corresponds to an electron density increase by 60% with respect to the surrounding corona whose density is around cm 3. The h vs t curves extrapolated at the UVCS FOV altitude show a good agreement with the observed CME transit time. 4. Conclusions We presented here results from a preliminary analysis of a unique data set which allowed us to study a prominence eruption observed from STEREO/EUVI, Hinode/EIS and SOHO/UVCS data. The information we derived from STEREO on the 3-D structure of the erupting filament led us to rule out the presence of a flux rope, even if this kind of structures is envisaged in many flare CME models. The spectra acquired by EIS showed unexpected non-thermal velocities in the corona along the eruption path. These could be ascribed to magnetic reconnection occurring behind the eruption and closing the opened field lines, but also to plasma turbulence or other effects. To our knowledge, this is the first time that a similar study has been conducted on an erupting prominence; even if the data analysis and interpretation are still in progress, this initial study demonstrates the great potentiality of STEREO and Hinode for the study of solar eruptions. References Culhane, J.L. et al. 7, Solar Phys., 179, 279 Del Zanna, G. et al. 8, Proc. of the 2nd Hinode Science Meeting, this volume Kaiser, M.L. et al. 8, Space Sci. Rev., 136, 5 Kohl, J.L. et al. 1995, Solar Phys., 162, 313 Thompson, W.T. et al. 3, Proceedings of the SPIE, 4853, 1 Wuelser, J.-P. et al. 4, Proceedings of the SPIE, 5171, 111

5 **FULL TITLE** ASP Conference Series, Vol. **VOLUME**, **YEAR OF PUBLICATION** **NAMES OF EDITORS** Study of Quiet Sun Through the Solar Atmosphere: From the Chromosphere Up to Coronal Layers Lucia Abbo INAF-Osservatorio Astronomico di Torino, I-25 Pino Torinese, Torino, Italy Alan Gabriel Institut d Astrophysique Spatiale, F-914 Orsay, France Louise Harra UCL- Mullard Space Science Laboratory, Holmbury St. Mary, Dorking, Surrey RH5 6NT United Kingdom Abstract. We analyze intensity maps over a range of temperatures covering the chromosphere to the solar corona, near a polar coronal hole. Using observations from EIS spectrometer on Hinode, we examine the width of the network boundary as a function of temperature. Very preliminary results show that there is a gradual increasing of the network boundary width through the transition region up to coronal layers. Existing observations are being studied and newer observation plans are currently under way. 5. Introduction The solar transition region is characterized as a discontinuity in the temperaturedensity structure of the outer solar atmosphere: the temperature jumps from to K in only a few thousand kilometers and the density decreases of a factor of 40 over the same height range. Images in spectral lines formed through the transition region show the boundaries of the supergranular cells in enhanced intensity, up to coronal temperatures, at which point the cell visibility disappears (Huber et al. 1974). This was interpreted as due to the magnetic field lines, concentrated at the cell boundaries, spreading out at the higher level and lower pressure of the corona, as illustrated by a theoretical model of Gabriel (1976). Although it is now understood that these nearly unipolar magnetic funnels are also mixed with small bipolar magnetic loops (e.g. Dowdy et al. 1986), the expansion rate of the funnels remains important for the physics of these regions. The spreading out leads to near horizontal fields, the famous canopy known to chromospheric workers. Theory cannot predict the precise temperature of this spreading, which must depend on observations. Our purpose is to study the network of supergranular cells through the analysis of spectral lines observed with the EUV Imaging Spectrometer (EIS) on board Hinode in order to explore the expansion rate with height of the magnetic funnels. A similar study was published by Patsourakos et al. 1999: they analyzed data from the CDS spectrometer on SOHO and they found out that network boundaries have an almost constant width up to a temperature of about K and then fan 5

6 6 Abbo, Gabriel, and Harra Figure 3. Left panel: EIT image of He II at 304 Å with superposed the field of view of EIS. Central and right panels: intensity maps of Fe VIII and Fe X lines. out rapidly as coronal temperatures are reached, in agreement with earlier theoretical models of the transition region (Gabriel 1976). Our results are compared with those of Patsourakos et al. We point out that Hinode/EIS, with its higher angular resolution and availability of lines in the 0.5 to 1 MK temperature range offers an exceptional possibility to study this problem. 6. Observations and Data The EUV Imaging Spectrometer (EIS) on Hinode performs high resolution spectra in the two wavelength bands Å and Å, corresponding to lines with formation temperature from up to about K, covering the upper part of the solar atmosphere. Observations of this analysis consist of a series of 1 x 256 exposures forming a total raster of obtained on April 15 8, starting at 11:45:28 UT. Exposure time is about 45 s for each slit position and it took about 109 minutes to observe the entire region. The field of view of EIS, shown in the left panel of Figure 1 superposed to the EIT He II 304 Å image, is near the coronal hole at the South Pole. For our purpose, six spectral lines have been considered (Table 1), comprising the range of formation Table 1. Selected spectral lines for this study (ion, wavelength and log of formation temperature derived from CHIANTI database). Ion Wavelength (Å) Log(T) He II O IV O V Fe VIII Mg VII Fe X

7 Quiet Sun Through Solar Atmosphere 7 temperatures from the chromosphere to the corona (logt varies from 4.7 to 6.0). The intensity maps of the Fe VIII and Fe X spectral lines are shown in the central and right panels of Figure 1: in the transition region line, it is quite evident the presence of several supergranular cells while network boundaries have almost disappeared in the Fe X image. Moreover, in the upper part of the EIS field of view there is an active region very bright in Fe X line corresponding to coronal temperatures. 7. Autocorrelation Analysis In order to study characteristic scale lengths in the two dimensional monochromatic images, we have applied 2D autocorrelation function based on Fourier analysis. Figure 2 illustrates an example of autocorrelation image for the Fe VIII line with contour levels (top panel) and the plots of the autocorrelation function for half of the central row and column of the image (bottom panel). The initial decrease corresponds to the scale of the smallest persistent structures shown in the intensity maps which is interpreted as the width of the cells boundary. The contour levels show an almost isotropic distribution of the structures. The network boundary size has been calculated as the width at half maximum Figure 4. Top: an example of the image of the autocorrelation functions calculated for Fe VIII line with contour levels. Bottom: plots of the autocorrelation functions for Fe VIII for half of the central row (left) and column (right) of the image.

8 8 Abbo, Gabriel, and Harra Figure 5. Network boundary width as a function of temperature: the full dots show the results of the present study, triangles are the values from Patsourakos et al of the autocorrelation function for each spectral line. In particular, we have calculated the average of the values obtained along half of the central row and column of the autocorrelation functions. In order to reduce the effect of bright points found at the network boundaries, the maximum of intensity has been limited in the original images to a value depending on the spectral line. 8. Results and Comments From the study of the autocorrelation functions, we derived the width of the network boundary through the solar atmosphere. The results are shown in Figure 3 as full points. The solid line represents the trend of the increasing width of the network which has been extrapolated to lower temperature as shown by dotted line. The first point corresponds to He II line which shows an higher value than expected, likely due to the large optical thickness of this line. The plot shows a gradual increasing of the network width through the transition region up to coronal layers. These results can be compared with those published by Patsourakos et al (triangles in Fig. 3). They found values much higher in the transition region temperature which they claimed to be in agreement with the model of Gabriel (1976). We would like to stress that our study is very preliminary and, moreover, we are planning newer observations of EIS in order to improve the quality of data sets. References Dowdy, J.F., Jr., Rabin, D. and Moore, R. L. 1986, Solar Phys., 1, 35 Gabriel, A.H. 1976, Philos. Trans. R. Soc. London A, 281, 575 Huber, M.C.E., et al. 1974, ApJ, 194, L115 Patsourakos, S., Vial, J.-C., Gabriel A.H. and Bellamine N. 1999, ApJ, 522, 540

9 **FULL TITLE** ASP Conference Series, Vol. **VOLUME**, **YEAR OF PUBLICATION** **NAMES OF EDITORS** More of the Inconvenient Truth About Coronal Dimmings Scott W. McIntosh 1, Joan Burkepile 1, Robert J. Leamon 2 Abstract. We continue the investigation of a CME-driven coronal dimming from December 14 6 using unique high resolution imaging of the chromosphere and corona from the Hinode spacecraft. Over the course of the dimming event we observe the dynamic increase of non-thermal line broadening of multiple emission lines as the CME is released and the corona opens; reaching levels seen in coronal holes. As the corona begins to close, refill and brighten, we see a reduction of the non-thermal broadening towards the pre-eruption level. The dynamic evolution of non-thermal broadening is consistent with the expected change of Alfvén wave amplitudes in the magnetically open rarefied dimming region, compared to the dense closed corona prior to the CME. The presented data reinforce the belief that coronal dimmings must be temporary sources of the fast solar wind. It is unclear if such a rapid transition in the thermodynamics of the corona to a solar wind state has an effect on the CME itself. Establishing the poorly understood physical connection between Coronal Mass Ejections (CMEs) and transient coronal holes (e.g., Kahler and Hudson 1) using detailed spectroscopic measurement is a must. Since their initial observation with Skylab (Rust and Hildner 1976) they have come to be viewed as the residual footprint of the CME in the corona (e.g., Thompson et al. 0), the radio and plasma signatures of which are observed in interplanetary space (e.g., Neugebauer et al. 1997; Attrill et al. 8). In this short contributed paper we continue the analysis of McIntosh (9) (hereafter SWM9) extending the analysis of Hinode (Kosugi et al. 7) Extreme-ultraviolet Imaging Spectrometer (EIS; Culhane et al. 7) data and close by presenting the previously unpublished indicators of CME-forced evolution in the chromosphere and photosphere that are provided by a study of co-teporal Solar Optical Telescope (SOT; Tsuneta et al. 8) data. 1. Further Observations The EIS dataset comprises of three spectroheliogram raster observations (19:20-21:34UT, 01:15-03:30, 04:10-06:24UT) in multiple spectral windows, targeted at the following edge of NOAA AR The spectra belonging to the Fe XIII Å and Fe XV Å lines are fitted and the non-thermal line widths (hereafter v nt ) are determined using the formalism of SWM9. In addition, the active region complex was continuously monitored by the focal 1 High Altitude Observatory, National Center for Atmospheric Research, P.O. Box 0, Boulder, CO 80307, USA 2 Adnet Systems Inc., NASA Goddard Space Flight Center, Code 671.1, Greenbelt, MD USA 9

10 10 McIntosh, Burkepile, and Leamon A) 14-Dec-6 19:20: D) Intensity [(B - A) / A] B) 15-Dec-6 01:15: E) Intensity [(C - A) / A] C) 15-Dec-6 04:10: F) Intensity [(C - B) / B] Hinode/EIS Å Peak Intensity [log 10 erg cm -2 s -1 sr -1 Å -1 ] 2.00 A) 14-Dec-6 19:20: D) Intensity [(B - A) / A] -50 B) 15-Dec-6 01:15: E) Intensity [(C - A) / A] C) 15-Dec-6 04:10: F) Intensity [(C - B) / B] Hinode/EIS Å Peak Intensity [log 10 erg cm -2 s -1 sr -1 Å -1 ] Hinode/EIS Å Percentage Intensity Change Hinode/EIS Å Percentage Intensity Change G) 14-Dec-6 19:20: H) 15-Dec-6 01:15: I) 15-Dec-6 04:10: G) 14-Dec-6 19:20: H) 15-Dec-6 01:15: I) 15-Dec-6 04:10: Hinode/EIS Å Non-Thermal Line Width [km/s] Hinode/EIS Å Non-Thermal Line Width [km/s] J) Line Width [(H - G) / G] K) Line Width [(I - G) / G] L) Line Width [(I - H) / H] J) Line Width [(G - H) / G] K) Line Width [(I - G) / G] L) Line Width [(I - H) / H] Hinode/EIS Å Percentage Change Hinode/EIS Å Percentage Change X (arcsec) X (arcsec) X (arcsec) X (arcsec) X (arcsec) X (arcsec) Figure 1. Phases of the coronal dimming observed in the intensity and nonthermal line width of the Fe XIII (left) and Fe XV (right) emission lines. Figure 2. SOT observations of Stokes V/I in NFI - Fe I 6302Å (panel A) the BFI G-band (panel B) and Ca IIH (panel C) channels of the active region nearest to 12/14/6 22:00UT. Movies are available to accompany the panels of this movie at

11 More of the Inconvenient Truth 11 plane package over the course of the event. The Broadband Filter Imager (BFI) alternately observed Ca IIH (3968Å) and G-Band (43Å) from 12/14/6 19:13-12/15/8 15:58UT with a spatial scale of 0.11 at a cadence ranging from 3 seconds to 5 minutes, but with most frames taken at a 2 minute cadence. The Narrowband Filter Imager (NFI) took shuttered Stokes I and V measurements at -172mÅ from line center of the Fe I Å with a mean cadence of 2 minutes and a spatial scale of Figure 1 shows the three phases of the coronal dimming observed in the intensity and v nt of the Fe XIII Å (left) and Fe XV Å (right) emission lines. In each case panels A through C show the line intensity before, at the peak and near the end of the dimming event, panels D through F show the percentage change in intensity between panels A and B, C and A and, C and B respectively. Panels G through I show v nt before, at the peak and near the end of the dimming event while panels J through L show the percentage change in v nt between panels H and G, I and G, and, I and H. The solid white contours in panels D and E and the black contours in panels H, I, J and K indicate a 75% reduction in the Fe XII Å line intensity (SWM9). Like the former analysis we see a large ( 15%) increase in v nt that is concentrated in and around the perimeter of the 75% intensity reduction contour. As the dimming event progresses (comparing panels C, A and E), there continue to be regions of difference in v nt late in the lifetime of the dimming, becoming more spatially compact and following the contraction of the intensity decrease contour. In panel F, we see that v nt appears to be slowly recovering to pre-eruption levels Chicken or Egg? - Evolution in the Deeper Atmosphere Close inspection of the movies associated with Fig. 2 show that something quite unexpected has happened to the southern portion of the active region during the dimming event. In the region around position 670, -120 (marked with diamonds in all figures) between 22:37UT and 22:44UT the dim penumbra-like emission that we would have commonly associated with cool plasma lying on nearly horizontal magnetic fields completely disappears. This change is most clearly visible in the G-band images, but mirrored in those of Ca II 1. The penumbral emission disappears as the post-flare/cme brightenings/ribbons move southward over that region. Gibson and Fan (8) indicate that these ribbons are associated with the changing morphology of the magnetic field lines over the active region as the CME lifts off. Interestingly, in panels B and D of Fig. 1, we see that this location shows a significant increase in v nt over the pre-event values that is almost immediately followed by a significant reversal. Unfortunately, there is not enough space in this short article for a detailed discussion of this point or the SOT Spectro-Polarimeter observed the vector field in the vicinity of the active region complex before (12/14/6 22:00-23:03UT), and after (12/15/6 :45-06:48UT) the dimming (see online folder for a graphic of the SP observations). 1 We also notice that by the end of the timeframe considered that a considerable portion of the penumbral structure to the South and East of the region has also gone.

12 12 McIntosh, Burkepile, and Leamon Comparing the inverted 2 SP observations (Lites et al. 7) of the active region is not straightforward (a considerable amount of time, and subsequent small-scale evolution has occurred between the SP rasters - see, e.g., the movie of Fig. 2). There has been a considerable flux cancellation and decrease of field inclination in the region around (680, -120 ) leaving a weaker, predominantly positive, polarity mean field. These factors would contribute to the apparent disappearance of the penumbral emission. A significantly more detailed investigation of the small-scale evolution and resulting global magnetic topology changes is required to address this issue. However, we should consider if the change in magnetic topology was instigated by the CME (tied to the dimming) or if it simply related to the gross restructuring of this very complex active region in its effort to find a lower energy state after the eruption (e.g., Low 1). 2. Discussion and Conclusion The spectral information presented reinforces the conclusion of SWM9 that the changes of the coronal non-thermal line widths are tied to the thermodynamic evolution of the dimming region - and are consistent with the growth and recovery of Alfvén wave amplitudes in the varying magnetic topology. Again, we emphasize that, as temporally open magnetic regions, coronal dimmings must be temporary sources of the fast solar wind as the values of v nt reached are equivalent to those observed in coronal holes (McIntosh et al. 9). It remains to be seen what impact that fast wind stream, and it s momentum, will have on the CME itself. Clearly, including the rapid thermodynamic variation of the coronal plasma (neglecting changes observed deeper in the atmosphere, Sect. 1.1.) make the understanding of a CME (an intrinsically self-consistent eruption affecting all of the atmospheric layers) much more complex than is currently considered. Acknowledgments. SWM acknowledges support from NASA grants NNX08AL22G, NNX08AU30G. The National Center for Atmospheric Research is sponsored by the NSF. References Attrill, G. D. R., et al. 8, Solar Phys., 252, 349 Culhane, J. L., et al. 7, Solar Phys., 243, 19 Gibson, S. E., Fan, Y. 8, J. Geophys. Res., 113 (A9), Kahler, S. W., and Hudson, H. S. 1, J. Geophys. Res., 106 (A12), Kosugi, T., et al. 7, Solar Phys., 243, 3 Lites, B. W., Casini, R., Garcia, J., Socas-Navarro, H. 7, Mem. S. A. It., 78, 148 Low, B. C. 1, J. Geophys. Res., 106 (A11), Neugebauer, M., Goldstein, R., and Goldstein, B. E. 1997, J. Geophys. Res., 102, McIntosh, S. W. 9, ApJ (in press) McIntosh, S. W., Leamon, R. J., and De Pontieu, B. 9, ApJ (submitted) Rust, D. M., Hildner, E. 1976, Solar Phys., 48, 381 Thompson, B. J. et al. 0, Geophys. Res. Lett., 27, 1431 Tsuneta, S., et al. 8, Solar Phys., 249, Developed at NCAR under the framework of the Community Spectro-polarimetric Analysis Center - CSAC;