Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://hea-www.harvard.edu/PINTofALE/papers/d_MM_palermo.ps.gz Äàòà èçìåíåíèÿ: Mon Feb 12 23:15:08 2001 Äàòà èíäåêñèðîâàíèÿ: Sat Dec 22 01:38:16 2007 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: extreme ultraviolet

**TITLE**
ASP Conference Series, Vol. **VOLUME**, **PUBLICATION YEAR**
**EDITORS**
Stellar Coronae: The First 25 Years
Jeremy J. Drake
Smithsonian Astrophysical Observatory, MS­3, 60 Garden Street,
Cambridge, MA 02138, USA
Abstract.
``We are going to have to learn from the Sun''
---A. Fabian, referring to heating mechanisms for accretion disk coro­
nae, COSPAR, Warsaw, 2000.
Hot X­ray emitting coronae were detected on stars other than the Sun
about twenty five years ago. This review describes some of the obser­
vational advances that have improved our knowledge of stellar coronae,
with an emphasis toward understanding the more active stars.
It was a quarter of a century ago that reports appeared in the literature of
the first significant detections of X­rays emitted by the coronae of a stars other
than the Sun. The first was the pseudo­serendipitous detection of the Capella
system during pointing calibration of a rocket flight carrying a proportional
counter experiment for observing Puppis A and the Perseus cluster on 5 April
1974 (Catura et al. 1975 1 ). Very shortly afterward, X­rays were detected during
planned observations made by the joint Dutch­US Astronomische Nederlandse
Satelliet (ANS) of the flare stars UV Ceti and YZ CMi, and of Capella (Heise
et al. 1975; Mewe et al. 1975). Capella turned out to be the brightest corona
in the X­ray sky after the Sun when not outshone by comparatively rare large
flares on other bright candidates.
This very brief review touches on some results of ``X­ray'' observations of
stars in the intervening 25 years that I see as important steps in reaching our
current level of understanding, albeit still rather primitive, of the nature of
stellar coronae. I use the quotes here because much progress has also been
made based on extreme ultraviolet (EUV) observations (¸ 100 \Gamma 900 š A) taken
with the Extreme Ultraviolet Explorer (EUVE) 2 : distinction between science
based on arbitrary wavelength regimes such as ``X­ray'' or ``EUV'' is a bit silly
in this context---it is the energy of the phenomenon being observed that is the
distinguishing characteristic of the science. It is sometimes not realised that hot
plasmas (eg several 10 6 ­10 7 K) emit prominent lines in the 100 š A regime from
the same Fe XVIII­XXIV ions that can dominate spectra in the ¸ 10 š A regime
(the former being transitions of the type \Deltan = 0 and the latter \Deltan ? 0).
1
The PI, Dick Catura, said they chose Capella only for is proximity to the Perseus cluster
2
EUVE ceased operations at 00:00 GMT January 31 2001 and was placed in safe­hold; it will
probably re­enter the atmosphere in 2002­2003.
1

2 J.J. Drake
The Cosmic
Sexiness Ladder
Information
Quasars,
AGN
Milky Way,
CV's etc
Stars
The Sun
Cosmology,
Black
Holes
Sexiness
Planet
Hunting
Galaxies
J.J. Drake
Figure 1. Model of the relative visceral appeal of the different fields
of astrophysics. As pointed out by Ehud Behar, atomic physics lies
slightly below solar physics. Next down would, I think, be the weather.
A review on stellar coronae in Palermo is a daunting task, faced with the
prestigious group of scientists built up here by Vaiana, who could, if anyone
can, be called the father of stellar coronae for his drive of the subject and
role in the development of instrumentation. I use the speaking time and page
limit as an excuse. Notable omissions from this paper are the copious X­ray
emitting pre­main sequence stars, and the ``normal'' (ie non­interacting) giants.
Micela (this volume) offers an excellent review of studies of open clusters. I
try to aim at building an understanding of the stars most commonly observed
in X­ray surveys---stars with significantly higher X­ray luminosities than the
Sun. I apologise in advance for omission of many important references---a good
fraction of the 12 page limit could easily be taken up by the reference list alone,
in addition to silly figures.
1. ``The widespread astrophysical practice of declaring the nature of
unresolved celestial objects is more entertainment than science''
---Eugene Parker (1996)
I have outlined the main ingredients that go into a solar­like corona, together
with the energy loss terms that tend to dissipate it, in Figure 2. The ingredients
are: an input plasma source, that we know can be compositionally fractionated
relative to the underlying photosphere (x5); magnetic fields of sufficient strength
to confine the plasma; and heating mechanisms. I have intentionally divided it
this way because without any one of these ingredients there is no corona. Energy
loss terms are kinetic energy, in a wind or in flows back to the photosphere;
thermal conduction back toward the photosphere; and radiation. In the solar

APS Conf. Ser. Style 3
case, we can, through the right observations, obtain some information on all
these loss terms. More obviously, through direct EUV and X­ray imaging, using
eg Yohkoh, SoHO and TRACE, we can actually see the gist of what is going on.
In the stellar case, we have no immediate hope of imaging coronae (however,
see x8). Very little is currently know about stellar winds in ``coronal'' stars and
mass loss rates generally remain unknown (see x7). Looking at the other loss
terms, radiative loss is proportional to the square of the plasma density (though
one has to know the chemical composition---see x5), which can be determined
from observation. In order to ascertain conductive losses, one must really have
some idea of the morphology of the emitting plasma, which we can only guess
at. The only practical approach involves making strong assumptions, such that
any radiative losses one could observe in the UV from cooler plasma (10 4 ­10 5 K)
arise from conductive heating by hotter coronal plasma, and then modelling
the system. An example of this type of modelling can be found in Griffiths
(1999). Conductive losses will dominate over radiative losses for temperatures
above a few million K, unless the plasma density is quite high---10 12 cm \Gamma3 or
more---when radiation dominates.
The job of the observer, then, is to attempt to fathom the physics of all the
different processes based only on observations of the radiative losses. Contem­
plating this, Parker's (1996) outlook tends to appear less skeptical.
2. ``Stars along the entire main sequence, of all luminosity classes,
pre­main sequence stars as well as very evolved stars, have been
detected'' ---Vaiana et al. (1981)
A skeletal historical summary of some of the major advances in coronal physics
from a stellar perspective is listed in Table 1. The greatest single observational
advance prior to the first detections of stellar coronae in 1974­75 was arguably
from the Skylab mission, whose battery of UV­X­ray telescopes monitored the
Sun over the nine month period of the mission and revealed in great detail the
magnetic loop structure of the solar corona (see, eg, Golub & Pasachoff 1997).
On the stellar front, it was really the Einstein observatory that produced
the single largest breakthrough (see, eg, Sciortino 1993 for an insightful sum­
mary of the Einstein era). The stellar survey of Vaiana et al. (1981) revealed 143
sources and showed that X­ray emission occurred over nearly the entire main­
sequence (some detections of spectral types late B and early A were likely due
to UV leaks), and in giant stars earlier than mid­K. The observed luminosities
covered a range of 2­3 orders of magnitude for each spectral type, a result that
had profound consequences for coronal heating theories. The prevailing view up
to the mid­late 1970's was that coronae were heated by acoustic waves generated
by the convection zone, a scenario which places the Sun near the most X­ray
luminous of late­type stars. The observations instead showed the Sun to be
comparatively faint in X­rays, and, in the words of Vaiana in 1979 3 , ``both the
levels of X­ray emission and the general behaviour of the median luminosities
disagree with theoretical predictions''. The Einstein results were a remarkable
3 Reported by Giacconi 1993

4 J.J. Drake
Table 1. Milestones in the Study of Stellar Coronae
1940 Coronal lines Temperatures of ¸million K diagnosed from
identified forbidden Fe lines
1949 Rocket Flight First solar X­rays detected
1960 Rocket Flight Pinhole camera produces first X­ray image of
solar corona
1965 Rocket Flights First images of solar coronal structure with
grazing incidence optics
1973 Skylab High resolution EUV/X­ray images of coronal
structure
1974­5 Rocket Flight/ First detection of X­rays from Stellar Coronae
ANS (Capella) followed by UV Ceti and YZ CMi
1978 EINSTEIN Discovery of hot X­ray emitting coronae on a
wide range of stars; link with rotation rate and
magnetic dynamo
1983 EXOSAT Dispersed soft X­ray spectra (precursor to
Chandra LETG); Detailed light curves
1990 ROSAT First sensitive all­sky survey; confirmation of
ubiquity of hot coronae on main­sequence stars
F­M,giants
1992 EUVE First ``high'' resolution (–=\Delta– ¸ 200) stellar
coronal spectra: temperature structure; plasma
density; abundances
1993,6 ASCA, SAX Extensive board­band low resolution observations
1999 Chandra, XMM First sensitive high resolution (–=\Delta– ¸ 1000) soft
X­ray spectra: first detailed X­ray plasma
diagnostics
demonstration of the power of sacrificing the spatial resolution and detail af­
forded by studies of the Sun for the chance to observe stars with a wide range
of fundamental parameters. One further discovery was that stellar X­ray lumi­
nosities were highly correlated with rotation velocity (Walter & Bowyer 1981;
Pallavicini et al. 1981), as had been found for chromospheric activity indicators
in the 1960's. Combined with evidence from solar observations that acoustic
waves did not propagate up through the transition region, the Einstein results
were a significant factor in making the paradigm shift from acoustically heated
coronae to coronae heated magnetically as a result of dynamo action excited by
stellar rotation. An exception to this was the case of the O­B stars, whose X­ray
emission is thought most likely to result from shock­excited radiatively driven
winds, although a more solar­like coronal component is not yet ruled out (see,
eg, Kahn et al. 2001; Waldron & Cassinelli 2001).
One further important discovery from the Einstein survey was that spectral
hardness increased with X­ray luminosity, indicating that the more active stars
have hotter coronae (Schmitt et al. 1990).
Following a decade after the Einstein survey, the more sensitive and exten­
sive ROSAT mission was to confirm in great detail that all late­type dwarfs (F­
M) and giants down to mid­K are X­ray sources and have hot coronae (Schmitt

APS Conf. Ser. Style 5
????
Fractionation
F rad ji
= AK ji
Z
\DeltaT ji
G ji (T )N 2
e (T )
dV (T )
dT
dT F con / T
5
2
e
dT e
ds
Source
Heating
Mechanism(s)
Confinement
Mechanism(s)
Fields)
(X-ray, EUV...)
Kinetic Energy
(Gravity, Magnetic
Radiation
(wind...)
Conduction
Warm-Up Act
Input Plasma
"Corona"
Figure 2. Illustration of the ingredients required to make a solar­like
corona and the energy loss terms that tend to dissipate it.
1997). Observations of larger numbers of field stars and open clusters (see also
Micela, this volume) enabled a more detailed examination of the connection be­
tween coronal activity and dynamos. A natural ``dynamo number'' that drops
out of mean field dynamo theory is the ratio of rotation period to convective
turnover time, P r =Ü c (also referred to as the Rossby number). It was found that
when chromospheric activity indicators where compared to this Rossby number
the correlation became much tighter than for the rotation period alone (Noyes
et al. 1984). This same tight correlation is also manifest in the X­ray regime,
as is nicely illustrated in the relation between the ratio of X­ray to bolometric
luminosity LX =L bol (a measure of the efficiency of turning the available energy
into coronal losses) and P r =Ü c for open clusters and field stars shown in Pattern
& Simon (1996). Looking back at Figure 2 to what I have termed the three
essential coronal ingredients of an input plasma source, magnetic confinement,
and heating, to me it is remarkable that we can describe with reasonable success
the level of coronal activity over a wide range of spectral types using only the
two parameters rotation period and convective turn over time. I will return
briefly to this in section x6.
The sensitive ROSAT survey enabled significant volume­limited studies of
stellar X­ray emission. From Schmitt's (1997) study of F­K dwarfs, I note two
results in particular: (i) the minimum surface X­ray flux level found is similar to
that of solar coronal holes, indicating that this non­zero flux represents a basal
activity minimum; (ii) the maximum surface flux found exceeds the surface flux

6 J.J. Drake
level in solar active regions. This latter result was not really new in that, eg,
Vaiana & Rosner (1978) had already pointed out that the Sun completely covered
with active regions would have an X­ray luminosity of ¸ 2 \Theta 10 29 erg s \Gamma1 ---two
or so orders of magnitude lower than the very brightest stellar coronae. How,
then, are the very active stars explained?
3. ``If George Field with a snap of his theoretical fingers can extend
the range of EUV observations by 1,000 times, who knows what
may be possible in real life?'' ---S. Bowyer
The Extreme Ultraviolet Explorer (EUVE) enabled for the first time ``high''
resolution spectroscopy (–=\Delta– ¸ 200) of stellar coronae and the application
of spectroscopic plasma diagnostics that hitherto had only been feasible for
the Sun. The EUVE spectrometers covered the wavelength range 70­750 š A,
within which spectral lines from ions formed at temperatures ranging from a
few 10 5 K to 2 10 7 K can be found. EUVE opened up the determination of
plasma temperature, density and element abundances; at a similar time in the
early 1990's the low resolution CCD pulse height spectra of ASCA enabled some
temperature discrimination beyond that of ROSAT, and also promised access to
element abundances (see x5). A summary of EUV stellar physics can be found
in Bowyer, Drake & Vennes (2000), and Drake (1996).
To my mind, the most important contribution of EUVE was in diagnosing
plasma densities. Through the equation for radiative loss from a single spectral
line (Figure 2), it can be seen that the emitted line flux (and continua) for a
plasma of given composition is essentially proportional to the emitting volume
and the square of the density; for a given flux, the plasma density then defines
the emitting volume. Observations of density­sensitive lines of Fe IX­XIV in
stars with activity levels similar to the Sun (based on LX and rotation rate)
such as the ff Cen AB system (unresolved at EUVE resolution) and Procyon
(F5IV), yielded values of N e ¸ 10 9 ­10 10 cm \Gamma3 (eg review by Laming 1998).
Coupled with the emission measures as a function of temperature, derived using
Fe lines from ions formed at a range of different temperatures, which appeared
similar to typical solar coronal emission measure distributions (EM / T 3=2 with
a peak EM at T ¸ 2 10 6 K), to first order it was clear that these coronae were
likely quite similar to the solar corona.
In contrast, in order to explain coronae 100 times brighter than a Sun
covered in active regions, either the plasma density must be higher, or the
corona much more extended and voluminous---both notable departures from so­
lar coronal structure. Eclipsing systems offer potentially valuable information on
whether active stellar coronae are extended or compact, as Walter et al. (1983)
exploited in carefully­timed observations of the RS CVn­type binary AR Lac
at primary and secondary eclipse. Eclipses were seen, indicating compact coro­
nae, but with a hint that extended components might be present. However,
eclipses tended to evade some later observations of both AR Lac and TY Pyx
by EXOSAT (eg Pallavicini 1993), whose high orbit enabled long periods of con­
tinuous viewing. Bolstered by additional evidence from radio observations that
the radio­emitting (gyro­synchrotron) regions of these coronae appeared to be
several stellar radii in size, some concluded that the X­ray (thermal) coronae

APS Conf. Ser. Style 7
must be very large. There is no solid basis for the connection, since the elec­
tron population responsible for the radio emission is completely different to that
which excites the line and continuum emission and they need not be cospatial.
It appears (to me) that the EXOSAT observations must have been somewhat
of an unlikely fluke---eclipses have been seen in just about all eclipsing systems
observed since (eg Schmitt 1998), indicating that these very active coronae must
in fact be quite compact (see also Favata, this volume, for recent developments).
One also has to remember that these systems can be quite variable on timescales
similar to the durations of eclipses, and eclipse signatures can get wiped out.
Also fashionable (and still today) were suggestions that the emission origi­
nated from the region between the two stars of an active binary, where magnetic
fields might interact. Recent Chandra HETG observations of the RS CVn­type
binary V711 Tau (K1IV+G5V) have sufficient resolving power (¸ 1000) to see
the orbitally­induced Doppler shift that locates the majority of the emission on
K1 subgiant, and not in between the two stars (Ayres et al. 2001).
Just how compact the active coronae are could in principle be ascertained
from EUVE density determinations. Initial estimates using lines of Fe XXI
and Fe XXII formed at 10 7 K or so yielded N e ¸ 10 12 ­10 13 cm \Gamma3 for active
systems such as V711 Tau, AU Mic, oe Gem and Capella (eg review by Drake
1996). Such high densities imply tiny surface filling factors of 1­10 % and scale
heights based on simple plasma loop models of only 100­10,000 km. At cooler
temperatures, Fe XII­XIV lines indicated densities of 10 10 cm \Gamma3 , again similar to
solar active regions. These results suggested that the 10 7 K plasma is structured
quite differently to the 10 6 K material. However, one problem with the EUVE
analyses is that, at the resolution of the EUVE spectrometers, the Fe XXI and
Fe XXII diagnostics are blended. The often used Fe XXI 102.35 š A, for example,
is blended with lines of O VIII and when compared with Fe XXI 128.73 š A can
yield a spuriously high density. While the densities are likely still ``high'' (10 11
or higher), good quality Chandra spectra are still needed for confirming these
density estimates for the hot plasma. Densities of 10 10 cm \Gamma3 have very recently
been confirmed for the cooler 10 6 K plasma in the Capella coronae based the
He­like ions of O (see, eg, the contributions of Brinkman and Canizares, this
volume), and for V711 Tau (Drake, as yet unpublished).
4. ``We envision a model wherein continual flaring associated with
the complex of star spots provides sufficient energy injection to
power a 10 7 K corona'' ---Walter et al. (1978)
While the 10 6 K plasma on the active stars appears similar to solar active regions,
and the hot plasma appears to be of higher density and relatively compact,
the nature of the hotter plasma remains a puzzle. I think part of the answer
can be gleaned from looking at the ``intermediate activity'' stars---stars with
LX 10 or 100 times that of the active Sun. Plasma densities based on EUVE
observations of Fe XIII and Fe XIV lines in ffl Eri (K2V) and ¸ Boo A are N e ¸
10 9 ­10 10 cm \Gamma3 ---similar again to solar active regions. Drake et al. (2001) have
recently shown that, in addition to plasma densities, the coronal temperature
structures of ¸ Boo A and ffl Eri are also very similar to that of the brightest
solar active regions. Moreover, the X­ray luminosities of these stars can be

8 J.J. Drake
understood if the stars are entirely covered in these active regions. However, the
more active of the two, ¸ Boo A, shows a ``hot extension'' to its EM distribution,
up to 10 7 K---a hint of the hot plasma that dominates the very active stars.
This ties in with some important earlier work of G¨udel and co­workers based on
rough EM distributions derived from ASCA and ROSAT observations of solar
analogues of different ages (``The Sun in Time''). G¨udel et al. (1997) found in
the young and more rapid rotators a considerable portion of the EM residing
at temperatures of ¸ 10 7 K, but with a significant EM at solar active region
temperatures of a few 10 6 K, and that the hot component gradually disappeared
toward older, more slowly rotating stars. This hot component is the same plasma
Drake et al. (2001) see in ¸ Boo A. G¨udel (1997) succeeded in mimicing these
EM distributions with a distribution of hydrodynamic flare events, suggesting
that flare heating plays a dominant role in these stars.
EUVE Deep Survey (DS) observations are very useful for investigating time­
variability of coronal emission because of the long (days­weeks) observing times
spent on some stellar targets. Audard et al. (2000 in references therein) and
Kashyap et al. (in preparation) have investigated the frequency distributions
of flare energies in different stars using long EUVE DS observations. Using
both flare detection techniques and monte carlo simulations of flare distributions
theese authors conclude that the variability observed in EUV fluxes of active
stellar coronae is largely consistent with this emission being due primarily to a
superposition of flare events. If correct, the conclusion is that the hot plasma
on active stars is flaring plasma, with a similar plasma density to that found in
solar flares (¸ 10 11 \Gamma 10 12 cm \Gamma3 ). As Walter et al. (1978) discussed, this is not a
new suggestion, though the grounds for the conclusion are now somewhat more
substantial.
5. ``One may then ask whether the chemical composition of the low
corona actually differs from that of the photosphere'' ---Pottasch
(1963)
The insight of Pottasch (1963) was aided by some oscillator strengths of the
day for neutral Fe lines being too high by about an order of magnitude, lead­
ing to underestimates of the photospheric abundances compared to Pottasch's
coronal values. Nevertheless, two or so decades later the review of Meyer (1985)
summarised a growing body of evidence confirming that Pottasch's suspicions
were correct. In the solar corona, the compositional differences appear to be
related primarily to the element first ionization potentials (FIPs) in that the
abundances of elements with FIP ¸ ! 10 eV (e.g. Mg, Si, Fe) are enhanced com­
pared to abundances of high FIP ( ¸ ? 10 eV; e.g. O, Ne, Ar) elements by an as
yet unidentified mechanism that preferentially feeds the corona with ions from
material at chromospheric temperatures. In other stars, examples of this ``FIP
effect'' have been found in stellar coronae with solar­like and intermediate activ­
ity levels, based on analyses of EUVE spectra. In contrast, analyses of EUVE,
ASCA and BeppoSAX spectra of more active stars have shown their coronae to
be metal poor, though in some cases it is possible that the metal paucity is also
shared by the parent photospheres. These results are reviewed and discussed
by, e.g., Feldman & Laming (2000), S.A. Drake (1996) and references therein.

APS Conf. Ser. Style 9
Starting with Ginga observations of the RS CVn­like binary UX Ari (Stern
et al. 1992), apparent increases in coronal abundances during large flare events
have been reported for some active stars, based mostly on analyses of low res­
olution X­ray spectra but also including joint EUVE and ASCA observations
(eg Mewe et al. 1997; G¨udel et al. 1999). While I caution that apparent abun­
dance changes accompanying changes in plasma temperature, as occurs in the
flare observations, could be the result of errors in adopted ion populations, it
is tempting to invoke chromospheric evaporation of compositionally unadulter­
ated plasma into the metal­depleted corona to explain these results, as various
authors have done.
Adding further confusion to the gamut of coronal abundance anomalies is
the recent observation of large Ne abundances in the XMM­Newton and Chandra
spectra of V711 Tau (Brinkman et al., 2001; Drake et al. 2001). The former
also claim a linear relation between logarithmic abundance and element FIP,
though the latter did not find convincing evidence for this. Based on a literature
survey of published ASCA analyses, Drake (2001) pointed out that enhanced Ne
abundances appear to be a common feature of all active stars. The reasons for
the Ne enhancements are not yet clear, though Brinkman et al. (2001) and Drake
et al. (2001) both point out that some solar flare events have been observed to
be Ne rich.
While difficult to understand at present, I remain convinced that coronal
abundance anomalies will yield valuable insights into the physical processes that
sustain stellar coronae.
6. ``Why are there no order­of­magnitude X­ray activity cycles in
these active stars?'' ---Stern (1999)
Stern (1999; see also Kashyap & Drake 1999, Drake et al. 1996) reviewed evi­
dence that long­term soft X­ray variability in active stars as judged by different
observations made by X­ray instruments such as the ROSAT and Einstein pro­
portional counters over periods of a decade or more, amounts typically to only a
factor of two or so---much less than the order of magnitude or more that the solar
corona shows from solar minimum to maximum. These authors suggested that
a turbulent dynamo instead of a large scale field ff \Gamma ! dynamo might dominate
the activity of these stars.
Related to this I think is the observation that there appears to be no change
in LX =L bol in the late main sequence, going from stars that have radiative cores
to stars that are supposedly fully convective (Fleming et al. 1995; though this
is still not entirely without controversy). This is important because standard
mean field ff­! dynamo theory relies on a differential rotation located at the
boundary between the radiative core and convective envelope to generate the
seed toroidal field which is amplified and transformed into a poloidal field (and
vice­versa) by the helicity in convective motions (the ff­effect).
It appears to me most likely that if the coronal X­ray loss efficiency is un­
affected by lack of a radiative core in relatively active and fully convective stars,
then the convective core is irrelevant in higher mass stars where it is present---at
least in late K and early M dwarfs. We are left then with a turbulent (ff 2 ) dy­
namo in active stars, such as that discussed by Durney, De Young & Roxburgh

10 J.J. Drake
(1993). Such a dynamo would not be prone to the periodic field reversal as hap­
pens with the large scale solar field, and, provided stellar rotation is sufficiently
rapid to influence convection and give it a definite handedness, large scale fields
could result from turbulence­driven dynamo processes (D. Hughes, private com­
munication). I suggest that the turbulent dynamo also underscores the tightness
of the LX =L bol vs Rossby number correlation: the only relevant parameters for
the underlying dynamo are then the rotation velocity to drive the convective
helicity through the Coriolis force, and the convection zone depth.
7. ``Using physical parameters which are reliably determined from
X­ray data, we show that thermal winds from cool dwarfs cause
—
M to exceed the solar value by factors of 10 3 or more'' ---Mullan
(1996)
While the Sun provides us with a nearby example of a late­type main sequence
mass­losing star, its mass loss rate as measured in situ by different spacecraft
amounts to only about ¸2\Theta10 \Gamma14 M fi /yr---orders of magnitude lower than any
direct non­solar measurements or upper limits for mass loss rates. Some theo­
retical estimates of mass loss rates for active stars (eg Mullan 1996) are as high
as a few 10 \Gamma11 M fi yr \Gamma1 . Such mass loss rates would be a significant source of
ISM enrichment, and would have important effects on the coronal energy bal­
ance (Figure 2). Very recent indirect estimates of mass loss rates for ff Cen and
Proxima by Wood et al. (2001) based on observations of the Ly­ff absorption
from heliospheric H I heated by the stellar wind­ISM interaction have yielded
rates of ¸ 2 —
M fi and ¸ 0:2 —
M fi , respectively.
Wargelin & Drake (2001) have shown that X­ray emission due to charge
exchange interactions of the ionized wind with the neutral ISM should be de­
tectable for nearby late­type dwarfs, and for more distant stars if the mass loss
rates are indeed substantially higher than the solar value. The signature would
be diffuse X­ray emission surrounding the star on spatial scales of up to a few
arcsec and should be detectable in Chandra ACIS­S images with exposures of
a few tens of ks. If observed, such emission could provide hitherto unobtain­
able information on wind geometry, ion composition, mass­loss rates, and the
distribution of neutral gas in the ISM.
8. ``A perfect 10 on the Drake scale of sexiness'' ---W. Cash
Webster Cash's remark and kind indulgence was aimed at the proposed future
micro­arcsecond X­ray Imaging Mission (MAXIM; see Cash's contribution to
this volume) view of matter around the supposed black hole in the core of M87.
However, at a level of about 8 on the same scale would surely be the resolved
disk of an active star. It hardly need be said that such images would enable
substantial advances in the field of stellar coronae.
Nearer home, one important step is higher spectral resolution to better
understand the dynamics of hot plasma (again, Figure 2). While we relish for
some fleeting moments the resolving powers of Chandra and XMM­Newton, it
should not go unnoticed that these instruments are really only of ``moderate''
resolution because the intrinsic spectral line profiles are not resolved. To resolve

APS Conf. Ser. Style 11
fully the Doppler widths of spectral lines at temperatures of 10 6 ­10 7 K requires a
resolving power not much short of about 10,000---a figure not even approached
by any slated future missions. As the Chandra and XMM­Newton missions
demonstrate, it is very difficult to reach resolving powers of more than 1000
with a reasonable effective area in the soft X­ray range where most the line
diagnostics are (5­25 š A) at a reasonable cost. However, to do so in the EUV
near 100 š A where similar plasma diagnostics can be found is relatively easy
and cheap using multi­layer, normal incidence optics. Such missions have been
proposed but not yet accepted. The downside to this wavelength regime is that
the number of accessible extragalactic objects higher on the Cosmic Sexiness
Ladder than things like stellar coronae is small. Unfortunately, the scientific
gain in this remaining unexplored window of true high resolution high energy
spectroscopy tends to have less influence on mission development than position
on the ladder. I don't think Parker would disagree with that.
Acknowledgments. JJD was supported by NASA contract NAS8­39073
to the Chandra X­ray Center. I extend warm thanks to Vinay Kashyap, Brad
Wargelin, Bob Rosner and Martin Zombeck for useful discussions.
References
Audard, M., G¨udel, M., Drake, J.J., Kashyap, V.L., 2000, ApJ, 541, 396
Ayres, T., Brown, A., Osten, R.A., Huenemoerder, D.P., Drake, J.J., Brick­
house, N.S., Linsky, J.L., 2001, ApJ, in press
Bowyer, S., Drake, J.J., Vennes, S., 2000, Ann.Rev.A&A, 38, 231
Brinkman, A.C., et al., 2001, A&A, 365, L324
Catura, R.C., Acton, L.W., Johnson, H.M., 1975, ApJ, 196, L47
Drake, J.J., 1996, in Cool Stars, Stellar Systems and the Sun, 9th Cambridge
Workshop, eds. R. Pallavicini and A. Dupree, ASP Conf. Ser. (ASP: San
Francisco), 203
Drake, J.J., Brickhouse, N.S., Kashyap, V., Laming, J.M., Huenemoerder, D.P.,
Smith, R., & Wargelin, B.J., 2001, ApJ, in press
Drake, J.J., Peres, G., Orlando, S., Laming, J.M., Maggio, A., 2000, ApJ, 545,
1074
Drake, J.J., Stern, R.A., Stringfellow, G., Mathioudakis, M., Laming, J.M.,
Lambert, D.L., 1996, ApJ, 469, 828­833
Drake, S.A., 1996, in Cosmic Abundances, ASP Conf. Ser., 99, eds. S.S. Holt &
G. Sonneborn, 215
Durney, B.R., De Young, D.S., Roxburgh, I.W., 1993, Sol. Phys., 145, 207
Feldman, U., Laming, J.M., 2000, Phys. Scripta, 61, 222
Fleming, T.A., Schmitt, J.H.M.M., Giampapa, M.S., 1995, ApJ, 450, 401
Giacconi, R., 1993, in Physics of Solar and Stellar Coronae, eds. J.F. Linsky &
S. Serio, Kluwer, Dordrecht, 3
G¨udel, M., 1997, ApJ, 480, L121
G¨udel, M., Guinan, E.F., Skinner, S.L., 1997, ApJ, 483, 947
G¨udel, M., Linsky, J.L., Brown, A., Nagase, F., 1999, ApJ, 511, 405­421

12 J.J. Drake
Golub, L., Pasachoff, J.M., 1997, ``The Solar Corona'', CUP, Cambridge, Eng­
land
Heise, J., Brinkman, A.C., Schrijver, J., Mewe, R., Gronenschild, E.H.B.M., den
Boggende, A.J.F., Grindlay, J., 1975, ApJ, 202, L73
Kahn, S.M. et al., 2001, A&A, 365, L312
Kashyap, V., Drake, J.J., 1999, ApJ, 524, 988
Laming, J.M., 1998, in Cool Stars, Stellar Systems & the Sun, 10th Cambridge
Workshop, eds. R.A. Donahue & J.A. Bookbinder, ASP Conf. Ser., 154,
447
Mewe, R., Heise, J., Gronenschild, E.H.B.M., Brinkman, A.C., Schrijver, J., den
Boggende, A.J.F., 1975, ApJ, 202, L67
Mewe, R., Kaastra, J.S., van den Oord, G.H.J., Vink, J., Tawara, Y., 1997,
A&A, 320, 147­158
Meyer, J.­P. 1985, ApJS, 57, 172
Mullan, D.J., 1996, in Cool Stars, Stellar Systems and the Sun, 9th Cambridge
Workshop, eds. R. Pallavicini and A. Dupree, ASP Conf. Ser. (ASP: San
Francisco), 461
Noyes, R.W., Hartmann, L.W., Baliunas, S.L., Duncan, D.K., Vaughan, A.H.,
1984, ApJ, 279, 763
Pallavicini, R., 1996, in Physics of Solar and Stellar Coronae, eds. J.F. Linsky
& S. Serio, Kluwer, Dordrecht, 237
Pallavicini, R., Golub, L., Rosner, R., Vaiana, G.S., Ayres, T., Linsky, J.L.,
1981, ApJ, 248, 279
Parker, E., 1996, in Magnetodynamic Phenomena in the Solar Atmosphere:
Prototypes of Stellar Magnetic Activity, eds. Y. Uchida, T. Kosugi and
H.S. Hudson, Kluwer (Dordrecht)
Pottasch, S.R., 1963, ApJ, 137, 945
Sciortino, S., 1993, in Physics of Solar and Stellar Coronae, eds. J.F. Linsky &
S. Serio, Kluwer, Dordrecht, 211
Schmitt, J.H.M.M., 1997, A&A, 318, 215
Schmitt, J.H.M.M., 1998, Cool Stars, Stellar Systems & the Sun, 10th Cam­
bridge Workshop, eds. R.A. Donahue & J.A. Bookbinder, ASP Conf.
Ser., 154, 463
Schmitt, J.H.M.M., Collura, A. Sciortino, S., Vaiana, G. S., Harnden, F.R., Jr.,
Rosner, R., 1990, ApJ, 365, 704
Stern, R.A., 1998, in Solar and Stellar Activity: Similarities and Differences,
Eds. C.J. Butler & J.G. Doyle, ASP Conf. Ser., 158, 47
Stern, R.A., Uchida, Y., Tsuneta, S., Nagase, F., 1992, ApJ, 400, 321­329
Vaiana, G.S., et al., 1981, ApJ, 244, 163
Waldron, W.L., Cassinelli, J.P., 2001, ApJL, in press
Walter, F.M., Bowyer, C.S., 1981, ApJ, 245, 671
Walter, F., Charles, P., Bowyer, S., 1978, ApJ, 225, L119
Wargelin, B.J., Drake, J.J., 2001, ApJ, 546, L57
Wood, B.E., Linsky, J.L., M¨uller, H­R., Zank, G.P., 2001, ApJ, 547, L49