A&A 465, L47?L50 (2001)
DOI: 10.1051/0004-6361:20077126
c? ESO 2007
Astronomy
&
Astrophysics
Letter to the Editor
Are solar maximum fan streamers a consequence
of twisting sheet structures?
H. Morgan and S. R. Habbal
Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu, HI 96822, USA
e-mail: hmorgan@ifa.hawaii.edu
Received 18 January 2007/Accepted 24 February 2007
ABSTRACT
Context. Fan streamers are often observed at low to mid latitudes in the corona at solar maximum, appearing narrow in latitudinal
extent near the Sun, and fanning out with height, adopting an approximately linear, but not necessarily radial, configuration above
?3 R
circledot
.
Aims. We o?er arguments to support the conjecture that such structures may sometimes consist of high density, non-uniform sheets,
viewed edge-on near the Sun, and twisting to a more face-on alignment by 3 R
circledot
.
Methods. EUV and white light observations of a fan streamer observed on 2000/12/05 are analyzed. A simple 3D density model is
used to recreate the streamer structure.
Results. EIT images show a thin bright sheet at the base of the streamer. The continuation of this structure through the EIT, MLSO
MKIV coronameter, and LASCO C2 fields of view, suggests that this sheet is formed mostly of open magnetic field lines. The overall
large-scale appearance of the streamer is well simulated by a simple model of a twisting high-density sheet. If the twisting-sheet
conjecture is valid, there is a correlation between the distribution of enhanced rays within the streamer viewed in white light, and the
distribution of small regions of enhanced brightness seen on the disk in EIT 171 ? at the position of the streamer base.
Conclusions. We suggest that the apparent poleward divergence of equatorial coronal rays, or threads, seen during solar maximum
above active regions, may sometimes be a consequence of such a twisting sheet topology.
Key words. Sun: corona
1. Introduction
The solar maximum corona is extremely complex, and higher
resolution observations in the future may reveal increased com-
plexity at smaller scales. Calculating the three-dimensional den-
sity structure of such a complex corona is challenging. At the
height of solar maximum, bright and filamentary structures are
seen at all latitudes, often filling most of the coronal image,
with few regions of low brightness, and large reconfigurations
in structure can occur from one solar rotation to the next. Large
helmet streamers are rare, although many smaller helmet stream-
ers are seen at all latitudes. At low to mid latitudes, fan-shaped
structures often appear to emerge from active regions. These
structures are highly filamentary, and contain bright rays which,
in contrast to helmet streamers, seem to diverge super-radially.
Fan streamers have been seen in eclipse observations for many
decades (see references in Koutchmy & Nikoghossian 2002).
Observations of equatorial poleward-diverging rays during so-
lar maximum have been described by Eselevich & Eselevich
(2006) as an unexpected phenomenon. Coronal linear threads
were studied in detail by Koutchmy & Nikoghossian (2002)
from eclipse and other observations, who show a collection of
bright, diverging threads rising from flaring active regions. In
this Letter, we provide observational evidence to support the
conjecture that fan streamers are a consequence of twisting high
density sheet-like structures.
2. Examples of fan streamers
The images of Fig. 1 are processed using a recently devel-
oped technique, the Normalizing Radial-Graded-Filter (NRGF)
(Morgan et al. 2006). Viewing NRGF-processed images is
equivalent to viewing a large number of normalized latitudinal
profiles simultaenously, thus providing continuity of structure
between the fields of view of various instruments. This is the
basic means we use to identify fan streamers, and to view their
diverging structure, which occurs mostly at heights below the
LASCO C2 field of view. The highly complex appearance of
the solar maximum corona is apparent from Fig. 1. It is hard to
define boundaries between streamers and surrounding structure.
Except for a few dark regions of limited latitudinal extent, the
whole corona contains filamentary structure of varied brightness.
It is di?cult to equate the type of bright structure seen here with
the more clearly defined helmet streamers of solar minimum.
This work presents an interpretation to explain the appear-
ance of fan streamers, a term we interpret as bright structures
which expand from a narrow latitudinal extent at low heights to
fill a more significant angular extent at larger heights. Since it
is di?cult to define boundaries between various structures, fan
streamers are not always easy to identify. The white arrows in
the images of Fig. 1 point to the bases of fan streamers. They
rise from low to mid latitudes, and are associated with active re-
gions at the limb. The brightness across the fan streamers is non-
uniform, that is, there are narrow bands of enhanced brightness
Article published by EDP Sciences and available at http://www.aanda.org or http://dx.doi.org/10.1051/0004-6361:20077126
L48 H. Morgan and S. R. Habbal: Are fan streamers twisting sheet structures?
Fig.1. Fan streamers observed during solar maximum on dates a) 2000/02/13, b) 2000/11/20, and c) 2000/02/24. White arrows point to the bases
of the fan streamers. The images are composed of observations by the Solar and Heliospheric Observatory?s (SOHO) Extreme Ultraviolet Imaging
Telescope (EIT) in the Fe IX/X 171 ? bandpass at heights (<1.3 R
circledot
) (Delaboudiniere et al. 1995), the Mauna Loa Solar Observatory?s (MLSO)
MKIV coronameter (1.3?2.25 R
circledot
) (Fisher et al. 1981) and by the Large Angle and Spectrometric Coronagraph (LASCO) C2 instrument (>2.25 R
circledot
)
(Brueckner et al. 1995).
Fig.2. The north-west corona observed on 2000/02/24 by EIT 171 ?,
MLSO MKIV, and LASCO C2. a) Image processed by the NRGF.
b) Same image processed with the madmax filter (Koutchmy et al. 1988,
1992). Rays seen in the NRGF image contain finer threads which are
barely discernible, but are revealed by the madmax filter.
which extend continuously through the field of view of MLSO
MKIV and LASCO C2. These rays are close to linear (but not
necessarily radial) at heights above ?3 R
circledot
, but are often bent at
lower heights. It is not entirely clear whether these rays are the
same features described by Koutchmy & Nikoghossian (2002) as
coronal linear threads, since the eclipse white light images given
in that work are processed in a di?erent manner to the images
of Fig. 1. Figure 2a is processed with the NRGF, and Fig. 2b
with the NRGF and an edge-enhancement routine. The threads
seen in the edge-enhanced image are very similar to the threads
described by Koutchmy & Nikoghossian (2002). Figure 2 shows
clearly that the base of these rays, or threads, extends down to the
solar surface, as seen in the EIT field of view. That fan streamers
are not necessarily associated with closed field regions, and are a
common feature during solar maximum, is evident from Figs. 1
and 2.
3. Fan streamers as twisting sheets
The top image of Fig. 3 shows an observation of a fan streamer
made during 2000/12/05. It is apparent at the limb for ?3days,
and appears half a solar rotation later at the west limb, altered,
but still fan-shaped. The streamer shows a super-radial diver-
gence at heights below ?3 R
circledot
. This expansion is symmetrical
around the central axis of the streamer. We suggest that the gen-
eral structure of this streamer is a high-density twisting sheet,
Fig.3. a) Equatorial fan streamer observed during 2000/12/05. The im-
age is a composite of EIT 171 ?, MKIV and LASCO C2 pB obser-
vations. The two dotted lines are at heights of 3.5 and 4.5 R
circledot
,andare
of relevance to Fig. 4. b), c) Detail of the equatorial streamer at low
heights, observed by EIT in 171 ? at a) 19:00 on 2000/12/05, and b)
19:00 on 2000/12/06.
with threads of higher density embedded in the sheet forming
the rays. The sheet at low heights in the EIT image is seen close
to edge-on, and appears bright and narrow. Between the Sun and
H. Morgan and S. R. Habbal: Are fan streamers twisting sheet structures? L49
Fig.4. a) EIT 171 ? map of the source region of the fan streamer
of Fig. 3, observed during 2000/12/16. The brightness is reversed
(black regions have the highest intensity). b) Brightness of the EIT
map integrated over latitudes 12
?
?18
?
, given in arbitrary units. c) Fine-
scale structure (or enhanced rays) observed in pB across the equatorial
streamer observed by LASCO C2 during 2000/12/05. To increase the
signal to noise, many latitudinal profiles of pB between heights of 3.5?
4.5 R
circledot
are normalized to a mean of zero and a standard deviation of one
and summed (see Fig. 3a). The resulting latitudinal profile is fitted to a
2-degree polynomial. This smooth fit is subtracted from the original to
reveal the smaller scale variations across the streamer.
?3 R
circledot
, the sheet twists to a more face-on alignment. This is a
simple topology which can explain the apparent non-linear di-
vergence or expansion of the fan streamer.
More details of the fan streamer as seen by EIT 171 ? are
shown in Figs. 3b, c. A collection of rays rise from an ac-
tive region complex labeled S, which we call the source re-
gion. In Fig. 3b, the source region is mostly hidden behind the
limb, although we can see the rays at heights above the limb.
Figure 3c shows that part of the source region has become vis-
ible on the disk. Also apparent in Fig. 3c is a small system of
closed loops, existing alongside the open rays above the source
region. By 2000/12/16, the source region is fully revealed in EIT,
centered at the meridian. This region is shown in the longitude-
latitude map of Fig. 4a. A chain of active regions extends from
Carrington longitude ?30
?
?100
?
, limited in latitude between
?10
?
?20
?
.
Figure 4b shows the profile of intensity across the EIT source
region map, integrated across pixels between latitudes of 12
?
?
18
?
. This plot serves to show the position of the most active
spots across the source region, plotted against Carrington lon-
gitude. Figure 4c shows the position of the enhanced rays across
the fan streamer as seen by LASCO C2, plotted against position
angle. Given that some unknown proportion of the active region
brightness seen on the disk in EIT arises from closed field struc-
tures, there is still reasonable correlation between the EIT and
LASCO profiles to the right of these plots (between longitudes
55
?
?100
?
in Fig. 4b, or position angles 65
?
?95
?
in Fig. 4c),
and this is supportive of the twisting-sheet interpretation of the
streamer structure. The correlation between the EIT and LASCO
profiles across this section of the streamer is close to 70%. The
sheet emerging from the westward side of the source region
twists southwards, forming the edge of the fan streamer near the
equator. This is the part which shows the strongest correlation.
The eastward part of the sheet twists to the north, forming the
north-east edge of the fan streamer.
To lend support to this interpretation we construct a model
where we place a twisted, high-density sheet, of thickness 4
?
,
within a uniform 3D grid. The model is very similar to that de-
scribed in detail by Morgan et al. (2007), and is conceptually
similar to other works aimed at interpreting structures within the
corona (Vibert et al. 1997; Thernisien & Howard 2006, for ex-
ample). The base of the sheet at the solar surface extends from
longitudes 30
?
to 100
?
at a constant latitude of 15
?
, correspond-
ing to the observed source region shown in Fig. 4a. With in-
creasing height, the sheet is twisted, until a final radial configu-
ration is reached at 3 R
circledot
, with an inclination of 60
?
to east-west.
Figure 5 shows the surface (red) and final (yellow) configuration
of the model streamer overlaid on a Wilcox Solar Observatory
(WSO) photospheric magnetic field map. The density outside
of the streamer is zero. Within the streamer, the density pro-
file across the sheet (in arbitrary units) is fixed by the profile
of observed LASCO pB variation across the active region (as
the normalized profile shown in the middle plot of Fig. 4c, but
increased to a minimum value of 1 to avoid negative values).
We obtain pB images (again in arbitrary units) from the density
model by integrating along lines of sight within the uniform grid
(van de Hulst 1950; Qu?merais & Lamy 2002), with geometri-
cal alignments (or coordinates within the grid) equivalent to the
observations made on 2000/12/05. The radial gradient of bright-
ness within the image is then removed using the NRGF. It is
this step which allows a qualitative comparison of modeled and
observed structures, which are also processed with the NRGF.
The NRGF, which flattens a coronal image into a set of normal-
ized latitudinal profiles, allows a non-rigorous study of structure
without the calculation of absolute density.
The comparison between the modeled and observed images
is shown in Fig. 6. The complicated diverging shape of the ob-
served fan streamer is well replicated by the simple twisting
sheet structure. Although the overall shape and the distribution
of rays within the model streamer are not matched exactly with
the data, this could be easily improved by fine-tuning the model
parameters, such as the position of the rotation axis of the twist-
ing sheet, the relative densities of the rays within the streamer,
the rate of twist at lower heights, and the shape of the mod-
eled sheet. For the sake of simplicity, we have assumed that the
sheet follows a straight line in longitude-latitude space at a given
height. However, from the discrepancies between the model and
observation, the sheet is likely to have some curvature. Given the
deliberate simplicity of the model, the agreement is good, show-
ing that this fan streamer is very likely to be a large twisting
sheet structure.
4. Discussion
The existence of high-density thin sheets in the corona is well
known (Sturrock & Smith 1968; Thernisien & Howard 2006,
for example). That they emerge from active regions is also not
L50 H. Morgan and S. R. Habbal: Are fan streamers twisting sheet structures?
Fig.5. Modeled configuration of the equatorial streamer, overlaid on a
synoptic map of the photospheric magnetic field measured by WSO.
The streamer base at the Sun is shown as a red region, and the final ra-
dial configuration shown as the yellow region. Brighter regions in the
photospheric field map show a stronger positive field. Neutral lines are
shown as solid bold contours. The dashed bold contour shows the neu-
tral line at a height of 2.5 R
circledot
, as calculated by WSO using a PFSS
extrapolation of the photospheric field.
Fig.6. a) Observed and b) modeled images of the equatorial streamer
for date 2000/12/05. The central meridian is at a longitude of 203
?
, with
the east limb at a longitude of 113
?
. The source region of the equatorial
streamer is at longitude ?30
?
?100
?
, hidden just behind the east limb.
surprising. These structures are usually associated with, or con-
tain, magnetic neutral sheets, which are expected to form above
the cusps of large closed magnetic field regions, where streams
of open field emerging from surrounding regions of opposite po-
larity converge into a stalk (Newkirk 1967, for example). The
present work provides evidence that a high-density sheet need
not always form above an arcade of closed loops, but may form
very low in the corona, or even below. It is unclear wether this
high-density sheet traces a boundary between regions of oppo-
site magnetic polarity. The position of the neutral line in the
corona, calculated by a potential field source surface (PFSS)
extrapolation of the observed photospheric field (Altschuler &
Newkirk 1969; Wang & Sheeley 1992) is shown in Fig. 5 as
a dashed bold line. The approximate configuration of the fan
streamer sheet in the corona is given by the yellow band, which
bears no relation to the PFSS neutral line. This is not entirely
surprising, since PFSS calculations cannot account for rapid
changes in the photosphere during solar maximum (Wang et al.
2000). Liewer et al. (2001) considered that during times outside
solar minimum, more filamentary structure, not predicted by the
uniform heliospheric current sheet of PFSS models, is produced
by newly opened field lines created by reconnection processes in
active regions.
One possible mechanism for the creation of large-scale twist-
ing sheets is the likely rigid rotation of the corona above the dif-
ferential rotation of the photosphere (Howard 1984; Lewis et al.
1999), which would force the deformation, or twisting, of struc-
ture. The di?erential rotation would demand a time-dependent
deformity, with twists increasing or decreasing with time, de-
pending on the particular geometry (alignment and latitude) of
the structure.
Fan streamers may be more common at low latitudes be-
cause chains of active regions in narrow latitudinal bands, but
extended in longitude, are more common in these regions . There
are also geometrical line of sight considerations. To achieve a
fluted shape with diverging rays, a twisted sheet must appear nar-
row, or must be seen edge-on, near the Sun. A high-density sheet
rising from a constant latitude from the Sun is a flatter surface
when near to the equator, and would appear bright and narrow
when seen from Earth. At increasing latitudes, such a surface be-
comes curved like a section of a cone, and would not be apparent
as a thin structure in observations.
5. Summary
Fan streamers are a common feature of the solar maximum
corona, and they extend all the way to the solar surface. We
interpret the structure of the fan streamer of 2000/12/05 as a
sheet of high density material, emerging directly from a ex-
tended chain of active regions at low latitude. The twisting of the
high-density sheet with height recreates the diverging fan shape
of the streamer at heights below 3 R
circledot
. If we accept this interpre-
tation of structure, the distribution of enhanced brightness rays
within the sheet correlate with the position of spots of enhanced
brightness along the source region on the disk.
Acknowledgements. The pB data used in this work are courtesy of the MLSO
and the LASCO/SOHO consortium. The MLSO coronagraph is operated by
HAO, a division of the NCAR, which is sponsored by the NSF. SOHO is a mis-
sion of international cooperation between ESA and NASA. The WSO data were
obtained from the Stanford University website courtesy of J. T. Hoeksema. WSO
is supported by NASA, the NSF and ONR.
References
Altschuler, M. D., & Newkirk, G., Jr. 1969, Sol. Phys., 9, 131
Brueckner, G. E., Howard, R. A., Koomen, M. J., et al. 1995, Sol. Phys., 162,
357
Delaboudiniere, J.-P., et al. 1995, Sol. Phys., 162, 291
Eselevich, M. V., & Eselevich, V. G. 2006, Sol. Phys., 235, 331
Fisher, R. R., Lee, R. H., MacQueen, R. M., & Poland, A. I. 1981, Appl. Opt.
20, 1094
Howard, R. 1984, ARA&A, 22, 131
Koutchmy, O., Koutchmy, S., Nitschelm, C., Sykora, J., & Smartt, R. N. 1988,
Solar and Stellar Coronal Structure and Dynamics, 256
Koutchmy, S., & Nikoghossian, A. G. 2002, A&A, 395, 983
Koutchmy, S., Altrock, R. C., Darvann, T. A., et al. 1992, A&AS, 96, 169
Lewis, D. J., Simnett, G. M., Brueckner, G. E., et al. 1999, Sol. Phys., 184, 297
Liewer, P. C., Hall, J. R., De Jong, M., et al. 2001, J. Geophys. Res., 106, 15903
Morgan, H., & Habbal, S. R. 2007, A&A, 464, 357
Morgan, H., Habbal, S. R., & Woo, R. 2006, Sol. Phys., 236, 263
Newkirk, G. J. 1967, ARA&A, 5, 213
Qu?merais, E., & Lamy, P. 2002, A&A, 393, 295
Sturrock, P. A., & Smith, S. M. 1968, Sol. Phys., 5, 87
Thernisien, A. F., & Howard, R. A. 2006, ApJ, 642, 523
van de Hulst, H. C. 1950, Bull. Astron. Inst. Netherl., 11, 135
Vibert, D., Lamy, P., & Liebaria, A. 1997, ESA SP-404: Fifth SOHO Workshop:
The Corona and Solar Wind Near Minimum Activity, 713
Wang, Y.-M., & Sheeley, N. R., Jr. 1992, ApJ, 392, 310
Wang, Y.-M., Sheeley, N. R., Jr., & Rich, N. B. 2000, Geophys. Res. Lett., 27,
149