Evidence of Impulsive Heating in Active Region Core Loops

Abstract

Using a full spectral scan of an active region from the Extreme-Ultraviolet Imaging Spectrometer (EIS) we have obtained Emission Measure EM$(T)$ distributions in two different moss regions within the same active region. We have compared these with theoretical transition region EMs derived for three limiting cases, namely \textit{static equilibrium}, \textit{strong condensation} and \textit{strong evaporation} from \cite{ebtel}. The EM distributions in both the moss regions are strikingly similar and show a monotonically increasing trend from $\log T[\mathrm{K}]=5.15 -6.3$. Using photospheric abundances we obtain a consistent EM distribution for all ions. Comparing the observed and theoretical EM distributions, we find that the observed EM distribution is best explained by the \textit{strong condensation} case (EM$_{con}$), suggesting that a downward enthalpy flux plays an important and possibly dominant role in powering the transition region moss emission. The downflows could be due to unresolved coronal plasma that is cooling and draining after having been impulsively heated. This supports the idea that the hot loops (with temperatures of 3{-}5 MK) seen in the core of active regions are heated by nanoflares. Comment: 17 pages, 4 figures, Accepted for publication in The Astrophysical Journal

Full text

Page 1

arXiv:1009.0663v1 [astro-ph.SR] 3 Sep 2010
Evidence of Impulsive Heating in Active Region Core Loops
Durgesh Tripathi, Helen E. Mason
Department of Applied Mathematics and Theoretical Physics, University of Cambridge,
Wilberforce Road, Cambridge CB3 0WA, UK
and
James A. Klimchuk
NASA Goddard Space Flight Center, Greenbelt, MD20771, USA
d.tripathi@damtp.cam.ac.uk
ABSTRACT
Using a full spectral scan of an active region from the Extreme-Ultraviolet
Imaging Spectrometer (EIS) we have obtained Emission Measure EM(T) distri-
butions in two different moss regions within the same active region. We have
compared these with theoretical transition region EMs derived for three limiting
cases, namely static equilibrium, strong condensation and strong evaporation from
Klimchuk et al. (2008). The EM distributions in both the moss regions are strik-
ingly similar and show a monotonically increasing trend from logT[K] = 5.15–6.3.
Using photospheric abundances we obtain a consistent EM distribution for all
ions. Comparing the observed and theoretical EM distributions, we find that
the observed EM distribution is best explained by the strong condensation case
(EMcon), suggesting that a downward enthalpy flux plays an important and pos-
sibly dominant role in powering the transition region moss emission. The down-
flows could be due to unresolved coronal plasma that is cooling and draining
after having been impulsively heated. This supports the idea that the hot loops
(with temperatures of 3–5 MK) seen in the core of active regions are heated by
nanoflares.
Subject headings: Sun: corona — Sun: atmosphere — Sun: transition region —
Sun: UV radiation

Page 2

– 2 –
1. Introduction
Solar coronal heating has been one of the most intensely studied problems in solar
physics over the past 60 years. Despite major advances in the observational and theoretical
tools, the solution remains elusive (see e.g., Klimchuk 2006). Active regions provide a target
of opportunity to study the dominant heating mechanisms for the corona. The basic building
blocks of active regions are coronal loops. Because of the lack of cross-field transport of mass
and thermal energy in the corona, the problem of coronal heating can effectively be reduced
to a clear understanding of the heating mechanism in a single isolated coronal loop or loop
strand.
The large-scale 1 MK loops and fan-like loops (which appear at the periphery of active re-
gions) are seen clearly in observations recorded using the Transition Region and Coronal Ex-
plorer (TRACE; Handy et al. 1999) 171˚ A passband. Their properties appear to be consis-
tent with impulsive heating models (see e.g., Tripathi et al. 2009; Warren, Winebarger, and Mariska
2003; Klimchuk 2006, 2009, and references cited therein). However, the 3–5 MK loops in the
core of the active regions are more diffuse and more difficult to isolate even with present day
observatories. Hence it is more difficult to measure the physical plasma parameters along
these hot core loops and to compare those properties with simulated results (although see
L´ opez Fuentes and Klimchuk 2010). The ”moss” regions are located at footpoints of hot
(3–5 MK) loops (Martens et al. 2000; Antiochos et al. 2003; Tripathi et al. 2010) and they
map the photospheric magnetic flux very closely (Tripathi et al. 2008). The moss regions,
being the footpoints of the hot loops, provide a unique opportunity to study the dominant
heating mechanism at work in active region core loops.
A number of publications involving observational results, analytical and 1D numerical
simulations have proposed that steady heating is at play in the hot loops associated with
moss regions. The main argument is that minimal variability is observed in brightness of the
moss emission or in the Doppler shifts and line widths of spectral lines. Antiochos et al.
(2003) pointed out that impulsive heating would produce transient loops in the cores of
active regions that would be seen by TRACE as the plasma cools through 1 MK. Such
loops are present but not ubiquitous. The authors also noted that the brightness of moss is
quite steady when averaged over several TRACE pixels. Brooks and Warren (2009) later
showed that Doppler shifts and line widths observed by the Extreme ultraviolet Imaging
Spectrometer (EIS) on Hinode are also quite steady. Finally, Tripathi et al. (2010) measured
the density and thermal properties of moss observed by EIS and found that they do not
change significantly over periods ranging from an hour to 5 days.
These observations provide compelling evidence that heating in the cores of active re-
gions must be quasi-steady if the cross-field scale of the heating is comparable to the 2 arcsec

Page 3

– 3 –
(1.5 Mm) resolution of EIS or larger. That is, we can conclude that the heating is steady, but
only if the plasma does not have unresolved sub-structure. There is good reason to believe
that such sub-structure may exist, however (e.g., Klimchuk 2006). Individual thin strands
could be highly time variable, and yet an unresolved bundle would appear steady as long as
the strands are not in phase. This is true for both the coronal portions of the strands and
their moss footpoints.
We therefore seek a diagnostic of heating that does not require that the fundamen-
tal structures be resolved. Doppler shifts are one possibility.
produce no Doppler shift because the loop evolves to a static equilibrium (although see
Mariska and Boris 1983; Klimchuk et al. 2010). Impulsive heating would produce a net
red shift on the other hand. In an unresolved bundle of strands, some strands would have
upflows due to evaporation and others would have downflows due to condensation (cooling
and draining). Simulations show that the upflows are faster, fainter, and shorter lived than
the downflows (Patsourakos and Klimchuk 2006). Hence, the composite line profile would
be red shifted by several kilometers per second, and very hot lines would also have a faint
blue wing enhancement. The red shifts are predicted to decrease with temperature and
may even become blue shifts in the hottest lines (Bradshaw and Klimchuk 2010, in prepa-
ration). The observational picture concerning Doppler shifts in moss regions is unclear.
Some published measurements indicate minimal red-shifts consistent with steady heating
(e.g., Brooks and Warren 2009), while other measurements indicate larger red-shifts con-
sistent with impulsive heating (e.g., Doschek et al. 1976; Brueckner 1981; Klimchuk 1987;
Doschek et al. 2008; Del Zanna 2008). This is an important area of future work.
Steady heating tends to
Another diagnostic of coronal heating that works when there is unresolved sub-structure
is the emission measure distribution, EM(T), or the closely related differential emission mea-
sure distribution, DEM(T). Since the plasma is roughly isothermal along the coronal portion
of a strand, EM(T) in the corona is determined primarily by the distribution of strand tem-
peratures. An isothermal corona with all strands having the same temperature implies steady
heating (Warren 1999; Landi et al. 2002), but some coronal EM(T) support a nanoflare
interpretation (e.g., Patsourakos and Klimchuk 2009; O’Dwyer et al.
2009). Warren et al. (2010) suggested that the lack of a strong correlation between hot and
warm emission in the core of an active region is evidence that the heating must be steady.
We disagree that this is the only interpretation. The relative intensities of coronal emission
formed at, say, logT = 6.0 and 6.5 can vary greatly depending on the magnitude of the
nanoflares (e.g., Klimchuk et al. 2008, Figs. 4 and 7, dashed curves).
2010; Reale et al.
Moss is the transition region footpoints of hot strands, and the EM(T) there is deter-
mined largely by the variation of temperature along the strands. Klimchuk et al. (2008) have

Page 4

– 4 –
shown that the form of EM(T) is different under three limiting conditions: static equilibrium,
strong condensation and strong evaporation. We make use of these differences in the work
reported here. We measure the EM distribution of two moss regions observed by EIS and
compare them to the theoretical predictions. In this way we distinguish between steady and
impulsive heating.
The rest of the paper is structured as follows. In section 2 we present our observations.
In section 3 we provide the emission measure expressions following Klimchuk et al. (2008).
Section 4 provides results and we summarise and discuss our results in section 5.
2.Observations
In this study, we have analysed observations recorded by the EIS aboard Hinode. The
EIS instrument observes the spectral ranges 170–210 and 250–290˚ A, providing observations
in a broad range of temperatures (logT[K] = 4.7–7.3). For detailed information on EIS see
Culhane et al. (2007). For most of the EIS studies only a limited number of spectral lines
were chosen due to the limitations on telemetry. However, occasionally full spectral scans of
specific regions have been telemetered. Here we use one such raster of an active region AR
10961 recorded on July 01, 2007 at 03:18 UT. The active region first appeared on the east
limb on June 26th, 2007. The data analysed were obtained from a single raster that scanned
a field of view (FOV) of 128′′by 128′′with the 1′′slit, moving from west to east with an
exposure time of 25 sec. The over-plotted box on the TRACE 171˚ A image displayed in the
left panel of Fig. 1 marks the region which was rastered by EIS. An EIS image obtained in
Fe XII λ195.12˚ A is shown in the right panel of Fig. 1.
The full spectral scan allowed us to choose spectral lines formed over a broad range of
temperature, which is essential in order to obtain a good EM distribution of the plasma.
This is the first time a full spectral scan is being used to derive an EM distribution in active
region moss. In our study we have used relatively clean spectral lines (see Table 1) with tem-
peratures of formation logT[K] = 5.1–6.5 based on EIS spectral atlases (Landi and Young
2009; Young and Landi 2009). The standard EIS software provided in SolarSoft were used
to process the data and eis auto fit, also provided in SolarSoft, was used for Gaussian fitting
the spectral lines. To de-blend some of the blended lines labelled ’b’ in Table 1 we have used
the same procedure as in Tripathi et al. (2010) based on the suggestions of Young et al.
(2007, 2009).
We have derived EM following the approach of Pottasch (1963) whereby individual
emission lines yield estimates of the emission measure at the lines’ temperatures of formation.

Page 5

– 5 –
Table 1: Spectral lines used to study the emission measure distribution in moss regions.
IonWavelengthlog T
[K]
5.15
5.15
5.35
5.45
5.45
5.45
5.60
5.60
5.65
5.80
5.80
5.85
6.05
6.05
6.15
6.15
6.15
6.20
6.20
6.20
6.25
6.30
6.35
6.45
6.55
[˚ A]
279.6
279.9
248.5
183.9
184.1
276.6
246.0
185.2
269.0
278.4
275.3
171.1
184.5
258.1
261.0
180.4
188.3
192.4
193.5
195.1
202.0
274.2
284.2
263.0
193.8
O IV
O IV
O V
O VI
O VI
Mg V
Si VI
Fe VIII
Mg VI
Mg VIIb
Si VII
Fe IX
Fe X
Si IX
Si X
Fe XI
Fe XI
Fe XII
Fe XII
Fe XIIb
Fe XIII
Fe XIVb
Fe XV
Fe XVI
Ca XIV