1: % First draft June 5 2001

2: %Version 1, submitted Aug. 22, 2002

3: %Version 2, submitted Sep. 26, 2002

4: %\documentclass[12pt,preprint]{aastex} % uncomment this for referee's ver.

5: \documentclass[manuscript]{aastex}   % comment this line for referee's ver.

6: \usepackage{emulateapj5}             % comment this line for referee's ver.

7: \usepackage{natbib}

8: \shorttitle{The Structure of Stellar Coronae in Active Binary Systems}

9: \shortauthors{Sanz-Forcada, Brickhouse \& Dupree}

10: 

11: \newcommand{\gl}{$\lambda$}

12: 

13: \begin{document}

14: 

15: \title{The Structure of Stellar Coronae in Active Binary Systems}

16: \author{J. Sanz-Forcada\altaffilmark{1,2},

17:   N. S. Brickhouse\altaffilmark{1}, and A. K. Dupree\altaffilmark{1}}

18: \altaffiltext{1}{Harvard-Smithsonian Center for Astrophysics; 60 Garden St.,

19: Cambridge, MA 02138 (USA)}

20: \altaffiltext{2}{INAF -- Osservatorio Astronomico di Palermo;

21:   Piazza del Parlamento, 1; Palermo, I-90134 (Italy)}

22: \email{jsanz@astropa.unipa.it, nbrickhouse@cfa.harvard.edu,

23:   adupree@cfa.harvard.edu}

24: 

25: 

26: 

27: \begin{abstract}

28: 

29: A survey of 28 stars (22 active binary systems, plus 6 single stars or

30: wide binaries for comparison) using extreme ultraviolet spectra has been conducted to

31: establish the structure of stellar coronae in active binary systems

32: from the emission measure distribution (EMD), electron densities, and

33: scale sizes. Observations obtained by the {\it Extreme Ultraviolet

34: Explorer} satellite (EUVE) during 9 years of operation are included

35: for the stars in the sample.  EUVE data allow a continuous EMD to be

36: constructed in the range log~T$_e(K)\sim$5.6--7.4, using iron emission

37: lines. These data are complemented with IUE observations to model

38: the lower temperature range (log~T$_e[K]\sim$4.0--5.6).

39: Inspection of the EMD shows an outstanding narrow enhancement, or

40: ``bump'' peaking around log~T$_e(K)\sim$6.9 in  25 of the stars,

41: defining a fundamental coronal structure.  The emission measure per

42: unit stellar area decreases with increasing orbital (or photometric)

43: periods of the target stars; stars in binaries generally have more material at coronal

44: temperatures than slowly rotating single stars.  High electron densities ($N_e\ga

45: 10^{12}~cm^{-3}$) are derived at $\sim$log~T$_e(K)\sim$7.0 for some targets, implying

46: small emitting volumes.

47: 

48: The observations suggest the magnetic stellar coronae of these stars

49: are consistent with two basic classes of magnetic loops: solar-like

50: loops with maximum temperature around log~T$_e(K)\sim$6.3 and lower

51: electron densities ($N_e\ga 10^{9}-10^{10.5}~cm^{-3}$), and hotter

52: loops peaking around log~T$_e(K)\sim$6.9 with higher electron

53: densities ($N_e\ga 10^{12}~cm^{-3}$).  For the most active stars,

54: material exists at much higher temperatures (log~T$_e[K]\ge$6.9) as well.

55: However, current {\it ab initio}

56: stellar loop models cannot reproduce such a configuration.  Analysis

57: of the light curves of these systems reveals signatures of rotation of

58: coronal material, as well as apparent seasonal (i.e. year-to-year) changes in the activity

59: levels.

60: 

61: \end{abstract}

62: 

63: \keywords{stars: coronae --- stars: flares --- stars: individual --- 

64: x-rays: stars}   

65: 

66: 

67: 

68: %------------------------------------

69: \section{Introduction}

70: The study of coronal structure from early X-ray and EUV satellites has

71: generally been limited to 2 or 3 temperature emission measure

72: fits. After the launch of the {\it Extreme Ultraviolet Explorer}

73: satellite (EUVE) a continuous emission measure distribution (EMD) in

74: the coronal region has been obtained for a few objects, but no

75: systematic study has been carried out to date in a substantial set of

76: stars.  Early EUVE observations have shown a quite different coronal

77: structure in active stars from that of the solar corona

78: \citep{dup93}. After nine years of EUVE data collection, many cool

79: stars have been observed \citep[see, for instance,][]{crai97}, some of

80: them several times, allowing the 

81: acquisition of good spectra for many stars so that reliable EMDs can

82: be calculated.  A survey of 28 stars has been conducted \citep{tesis}

83: to find stellar parameters that can be related to the observed coronal

84: emission. In this study a total of 22 active binary systems (in

85: particular RS~CVn and BY~Dra systems), and 6 single stars or wide

86: binaries has been included.  This sample covers a wide range of

87: luminosity class, spectral type, and rotational period (see

88: Table~\ref{tabparam}), and hence differences might be expected to

89: occur in their coronae that can be related to stellar parameters.  In

90: this paper we present the results for 21 stars, complementing our

91: previous analysis \citep*{paper1}, which includes the stars V711~Tau,

92: UX~Ari, $\sigma$~Gem, II~Peg, $\beta$~Cet (also studied here with

93: recent observations), and AB~Dor. Two other binaries, Capella and

94: $\lambda$~And 

95: (Dupree et~al. 1993; Dupree, Brickhouse, \& Sanz-Forcada 2002; 

96: {Sanz-Forcada}, {Brickhouse}, \& {Dupree} 2002)

97: %\citep{dup93,dup02,sanz01} 

98: complete the set of 28 stars.

99: 

100: The observed EMD of Capella \citep{dup93} revealed the presence of a

101: narrow enhancement or ``bump'' at log~T(K)$\sim$6.8, that varies

102: little in observations taken at different epochs \citep{dup02}. Other

103: stars show similar structure including $\lambda$~And \citep{sanz01},

104: and the six stars in \citet{paper1}, some of which are also studied by

105: \citet{gri98}.  Analysis of the changes observed in the EMD during

106: large flares has also been carried out for 6 stars: $\lambda$~And

107: \citep{sanz01}, V711~Tau, UX~Ari, $\sigma$~Gem, and II~Peg

108: \citep{paper1}, and AR~Lac (this work), showing that the bump remains

109: and is stable in temperature, while the emission measure increases

110: during flares.

111: 

112: In this paper we describe the detailed results for the remaining 21

113: stars of the sample and the global conclusions from the entire sample.

114: The stars have been grouped according to their observed coronal

115: spectra: a first group of low and intermediate activity stars,

116: dominated by lines formed at log~T(K)$\sim$5.8--6.5, with different

117: levels of flux in lines formed at higher temperatures; a second group

118: of ``active'' stars, with spectra dominated by lines formed at

119: log~T(K)$\sim$6.7--7.1; and a third group with a very significant

120: presence of even hotter material (indicated by Fe XXIII-XXIV).

121: 

122: In \S~\ref{sec:observations} we discuss the EUVE and IUE observations,

123: followed by the techniques employed in the analysis of the data

124: (\S~\ref{sec:analysis}). Individual results, following the

125: classification of the three groups, are described in

126: \S~\ref{sec:results}. A general discussion of these results and a

127: comparison between the different degrees of activity are made in

128: \S~\ref{sec:discussion}, and the Conclusions are summarized in

129: \S~\ref{sec:conclusion}.

130: 

131: 

132: %------------------------------------

133: \section{Observations}\label{sec:observations}

134: 

135: EUVE observations taken between 1993 January and 2000 September are

136: used (Table~\ref{euvetimes}).  Some of the observations were awarded

137: to us through the Guest Observer program, while most of them were made

138: available through the Multimission Archive at Space Telescope (MAST).

139: EUVE spectrographs cover the spectral range 70--180~\AA, 170--370~\AA, 

140: and 300--750~\AA\ for the short-wavelength (SW), medium-wavelength

141: (MW), and long-wavelength (LW) spectrometers respectively, with

142: corresponding spectral dispersion of $\Delta\lambda\sim$ 0.067, 0.135,

143: and 0.270~\AA/pixel, and an effective spectral resolution of

144: $\lambda/\Delta\lambda$$\sim$200--400. The Deep (DS) Survey Imager has

145: a band pass of 80--180~\AA\ \citep*{hai93}.

146: 

147: EUVE light curves (Fig.~\ref{plc}) were built from the DS image, by

148: taking a circle centered on the source, and subtracting the sky

149: background within an annulus around the center.  Standard procedures

150: were used in the IRAF package EUV v.1.9.  Time bins are 600~s. Points

151: affected by the ``dead spot'' are marked as open circles in the light

152: curves, while filled circles mark the corrected points. The ``dead

153: spot'' is a low gain area of the DS detector that affects some of the

154: observations taken in 1993 and 1994, resulting in variable levels of

155: contamination of the signal \citep[see][]{mill95}. We corrected

156: the effects of the dead spot contamination by ratioing unaffected DS

157: flux to the flux from the integrated SW spectrum measured

158: simultaneously.  This 

159: gives a correction factor that can be applied to the

160: DS points affected by the dead spot.  When all the DS flux

161: points are contaminated, we normalized using a star 

162: with unaffected DS flux, and a similar emission 

163: measure distribution. The error bar 

164: included in the figures indicates the average of the non-contaminated

165: DS fluxes.  Spectra for each star, extracted

166: from each spectrograph, are binned over the total observation 

167: and then summed for the whole set of

168: observations. Fig.~\ref{specs} shows the SW and MW spectra for all the 

169: targets, and Fig.~\ref{specesp} contains the LW spectra for 5 of the 

170: stars.  The addition

171: of spectra from different observations is not of concern since no

172: significant degradation was reported in the performance of the EUVE

173: spectrographs during the mission \citep{abb96}.\footnote{Results of

174: the final EUVE calibration observations made during the last month of

175: EUVE science operations in January 2001 demonstrated: (1) no

176: significant degradation in the SW spectrometer; (2) possibly up to

177: 15\% loss of sensitivity non-uniformly in the MW spectrometer; (3) no

178: change (to $<$10\%) in the LW detector when comparison of 1993 and

179: 2001 spectra are made.  The LW detector showed some degradation in

180: 1999, but the detector recovered by the end of the mission.  The DS

181: count rates are in agreement within 10\% of previous

182: measurements. (See {\it

183: ftp://legacy.gsfc.nasa.gov/euve/doc/final\_calib.report}.)}  Lines

184: identified in the summed EUVE spectra are given in

185: Table~\ref{euveflux} and Table 4.  Our primary goal is to identify Fe diagnostic

186: lines, but we also include strong lines from other elements.

187: 

188: Spectra from the {\it International Ultraviolet Explorer} (IUE)

189: archive (NEWSIPS extractions) have also been used to construct the EMD

190: curve of the stars by providing lines formed at lower temperatures

191: than those occurring in the EUVE spectra. Low resolution spectra

192: ($\sim$6 \AA) covering $\lambda\lambda$~1100--1950 were employed. In

193: the case of AR~Lac, quiescent and active spectra have been selected,

194: with discrimination based on the changes observed in line fluxes found

195: in different spectra. IUE line fluxes used to determine the EMD are

196: listed in Table~\ref{tabiue}.

197: 

198: 

199: %------------------------------------

200: \section{Data Analysis}\label{sec:analysis}

201: 

202: The DS light curves of the targets are shown in Fig.\ref{plc} where 

203: they are compared with the orbital phase for the binaries.  For

204: $\epsilon$~Eri and LQ Hya, a photometric phase is displayed; no

205: periods are considered for $\alpha$~Cen, $\beta$~Cet and Procyon.

206: 

207: To obtain fluxes of the individual EUV emission lines we first

208: performed optimized extractions from the summed two-dimensional images

209: by removing an averaged background evaluated on either side of the

210: spectrum, using the software provided in IRAF and the EGODATA 1.17

211: reference data set.  A local continuum in the spectrum itself,

212: determined by visual inspection, was subtracted from each line where

213: necessary.  The error in the line flux is defined as

214: $\sigma=1/[S+B(1+1/n)]^{1/2}$, where S is the net signal, B is the

215: estimated average background, and n is the oversampling ratio (i.e.,

216: the ratio of total background pixels to the number of total source

217: spectral pixels in the image), having a value n$\sim$10--15 in our

218: extraction.

219: 

220: To correct the observed fluxes for interstellar hydrogen and helium

221: continuum absorption, we used a ratio \ion{He}{1}/\ion{H}{1}=0.09

222: \citep{kimb93}, and values for the hydrogen column density obtained in

223: different ways for each star.  For some targets, direct measurements

224: of the column density were available from Lyman series absorption

225: features. Frequently the observed ratios of the \ion{Fe}{16} \gl335

226: and \gl361 lines can indicate the amount of interstellar absorption

227: because the theoretical ratio (1.94 in photon units) is determined

228: from fundamental atomic physics.  When these line fluxes are

229: available, they have been used to establish or corroborate hydrogen

230: column densities to the targets (see Fig.~\ref{ismcalc}). When those

231: values were not accurate enough, we deduced the column density from

232: tabulations \citep{fru94} of stars nearby in the sky, and these were

233: the adopted values if no additional references are given.

234: Table~\ref{tabparam} lists the values assumed. Further discussion

235: follows in the sections for the individual stars.

236: 

237: The electron density in the corona of the stars at

238: log~T$_e(K)\sim6.9-7.0$ has been inferred from ratios of the observed

239: fluxes (corrected for interstellar absorption) of

240: \ion{Fe}{19}~$\lambda$91.02/$\lambda$108.37, \ion{Fe}{20}

241: $\lambda$110.63/($\lambda$118.66+$\lambda$121.83),

242: \ion{Fe}{21}~$\lambda$102.22/$\lambda$128.73,\\

243: \ion{Fe}{21}~($\lambda$142.16+$\lambda$142.27)/$\lambda$102.22, and

244: \ion{Fe}{22} $\lambda$114.41/$\lambda$117.17 in the summed spectra.

245: Iron line emissivities were generally computed for the densities

246: derived in each spectrum, when available.  Atomic models for

247: \ion{Fe}{20}--{\small\sc XXII} were taken from \citet*{bri95}, with

248: \ion{Fe}{19} from Liedahl's HULLAC calculations

249: \citep[see][]{bri98}. These models have recently been compared with

250: measured tokamak spectra at different densities in the range $10^{12}$

251: -- $10^{14}~cm^{-3}$ at $\sim 10^7$~K and show good agreement

252: \citep{four01}. Table~\ref{tabne} shows the results for each

253: star.

254: 

255: 

256: %\subsection{EMD}

257: We performed a line-based analysis of the emission spectra in order to

258: calculate the EMD ($\int N_e N_H dV$ cm$^{-3}$, where $N_e$ and $N_H$

259: are electron and hydrogen densities, in cm$^{-3}$) corresponding to

260: the observed fluxes.  In contrast to ROSAT and ASCA measurements,

261: which assume coronal models with only 2 or 3 temperatures, EUVE

262: gives information on a continuous set of ionization states.  In fact,

263: all stages of iron ionization are represented from \ion{Fe}{9} through

264: \ion{Fe}{24} except for \ion{Fe}{17}, which has no strong transitions

265: in the EUV spectral range.  We used the line emissivities calculated

266: from \citet{bri95} for the EUVE iron lines, based on a solar iron

267: abundance\footnote{The solar iron abundance is defined as (12. $+$

268: log~$Fe \over H$), where $Fe \over H$ represents the ratio of iron to

269: hydrogen by number.}  of 7.67 \citep{anders}. Line emissivities from

270: \citet{raym88} are used for the (non-iron) lines formed in the UV

271: region.  Theoretical fluxes were calculated using the assumed EMDs

272: \citep[see][ and references therein]{dup93,bri98} which were then

273: iterated to obtain the EMD that best matches the observed

274: fluxes. Generally agreement is better than a factor of two. It is

275: important to note that we have integrated over the entire atomic

276: emissivity function for each line in order to predict the model

277: fluxes, and do not simply assume formation at a single temperature or

278: temperature range.  Iron lines severely contaminated by lines of other

279: elements and some complex blends have been estimated by using the

280: Astrophysical Plasma Emission Code (APEC) v1.10 \citep{smith01}.

281: These are excluded from the EMD analysis as marked in

282: Table~\ref{euveflux}. Figure~\ref{emdfigs} shows the EMD of summed

283: spectra of the stars in the sample.  Values used for the EMD are given

284: in Tables~\ref{tabemd} and \ref{tabemd2}, with a simple

285: characterization suggested in Table~\ref{slopes}.

286: 

287: 

288: \section{Results}\label{sec:results}

289: 

290: \subsection{Low activity levels}\label{sec:lowactive}

291: 

292: In this group we include stars showing a spectrum dominated by lines

293: formed at\\ log~T(K)$\sim$5.8--6.5 (\ion{Fe}{9}--{\small \sc XVI}),

294: although with different contributions from the lines formed at higher

295: temperatures (see Figs.~\ref{specs}a, \ref{specesp}).   In

296: the EMD derived for these stars, $\alpha$~Cen represents the lowest

297: activity level observed in the sample, both in the transition 

298: region and in the corona. Procyon also has low levels 

299: in the corona, but the transition region EMD is comparable to 

300: that observed in more active stars (see \S~\ref{sec:discussion}). 

301: Finally, $\epsilon$~Eri and $\xi$~UMa have high emission measure 

302: in the two temperature

303: ranges, such that their EMDs represent an intermediate step towards

304: the second group included in the sample.

305: 

306: 

307: \subsubsection{$\epsilon$~Eri}

308: $\epsilon$~Eri (HD 22049, HR 1084) is a relatively young star

309: ($\sim$1~Gyr) showing high levels of activity.  The effects of the

310: dead spot prevent an analysis of the seasonal (i.e. year-to-year)

311: variations of the light curves, but small scale variations can be seen

312: with a frequency of $\sim$1--1.5 days, and some flaring activity could

313: be present in the second part of the 1995 observations

314: (Fig.~\ref{plc}).

315: 

316: Studies of the EUVE observations have been carried out by

317: \citet*{lami96} and \citet{sch96}. \citet{lami96} derived the emission

318: measure of each individual line as if it were emitting only at its

319: maximum temperature thus obtaining an upper limit to the EMD

320: value. This model peaks at log~T(K)$\sim$6.5. Densities derived from

321: \ion{Fe}{14} line ratios \citep{lami96} yield a value of

322: log~N$_e$(cm$^{-3}$)$\sim$9.5 at log~T(K)$\sim$6.2, similar to results

323: found by \citet{sch96}.

324: 

325: We added all the available EUVE data of $\epsilon$~Eri in order to

326: improve the statistics of the spectra, and the accuracy of the EMD.

327: Although the EMD calculated (Fig.~\ref{emdfigs}) indicates a peak

328: around log~T(K)$\sim$6.4, a range of values of the EMD 

329: predicts very similar line

330: fluxes.  It is difficult to distinguish a peak around

331: log~T(K)$\sim$6.5, from two peaks at log~T(K)$\sim$6.4 and

332: log~T(K)$\sim$6.8.  In order to identify a preferred EMD model, it

333: would be optimum to use lines with a maximum of emission around

334: log~T(K)$\sim$6.5--6.7 such as Fe XVII which are not available in the

335: EUVE spectral range.

336: 

337: The electron density calculated from the \ion{Fe}{21} lines

338: (3$\times$10$^{13}$ cm$^{-3}$, see Table~\ref{tabne}) points towards

339: the presence of two kinds of structures at log~T(K)$\sim$6.2 and

340: log~T(K)$\sim$7.0, because the inferred densities differ by $\sim$4

341: orders of magnitude, although some caution must be exercised in the

342: interpretation of the measured electron density at high temperatures

343: (see \S~\ref{sec:discussion}) in this star (the adopted value in the EMD 

344: was log~N$_e$[cm$^{-3}$]$\sim$13.0).  The observed levels of

345: EUV/flaring activity are higher than the flux levels of the quiet Sun,

346: creating different conditions for the atmosphere of the planet found

347: by \citet{hat00} associated with $\epsilon$~Eri.

348: 

349: 

350: \subsubsection{Procyon}

351: 

352: Procyon ($\alpha$~CMi, HD 61421, HR 2943) is an F5~IV-V star

353: frequently used for comparison with the Sun.  Its rotational period of

354: 9.06~d is estimated from a v$\cdot$sin~i=6.1 $km\ s^{-1}$

355: \citep{med99}.  The DS light curve of Procyon shows the presence of

356: short-term modulation in a non-periodic time pattern of $\sim$14~hr in

357: 1993 (Fig.~\ref{plc}).  The observations taken in 1999 show higher

358: dispersion in flux values than in 1993 and 1994; however, this effect can be due to the

359: increase of activity in the Sun, as has been found in other

360: observations taken in 1999 and 2000. In any case it does not seem

361: possible to attribute these variations to intrinsic changes in the

362: star.

363: 

364: During the EMD fitting process in Procyon, some of the lines have

365: shown larger discrepancies between the observed fluxes and those

366: predicted by the atomic model of \citet{bri95}. In particular, the

367: \ion{Fe}{12} $\lambda$364.4 line seems too strong, as if blended,

368: though the APEC model does not show an obvious candidate for

369: blending.

370: 

371: 

372: \subsubsection{$\xi$~UMa}

373: $\xi$ UMa is a multiple system formed by four stars grouped in two

374: spectroscopic binaries \citep{griffin98} in a visual orbit of

375: $\sim$60~yr.  EUVE can not resolve the components.  $\xi$~UMa A (HD

376: 98231) is a spectroscopic binary with a P$_{orb}$=670.24 days

377: \citep{griffin98}.  A G0V star is the only star observed, and the

378: companion is unknown -- perhaps an M star \citep{griffin98}.

379: $\xi$~UMa Aa shows low activity levels in the chromosphere, having

380: very weak emission in the \ion{Ca}{2} H \& K lines \citep{mon95}.

381: $\xi$~UMa~B (HD 98230) shows an active spectrum, with stronger

382: \ion{Ca}{2} H \& K emission lines coming from the observed G5V star,

383: and a possible contribution from an unseen late-K dwarf

384: \citep{griffin98}.  The low value of the mass function

385: (f[m]=0.000046M$_{\odot}$) found by \citet{griffin98} points towards a

386: very small value of the inclination ($\la 11\arcdeg$), meaning a

387: system observed almost at the pole. From the effective temperature

388: \citep[T$_{eff}$=5650~K,][]{cayr94}, and the typical values of stellar

389: radii available in \citet{gray92} relating radii and mass with

390: spectral type, we estimate R= 0.95~R$_\odot$ for $\xi$~UMa~B.  IUE

391: spectra from both components (Table~\ref{tabiue}) show that most lines

392: are stronger in the B component of the system by a factor

393: $\sim$2:1. Hence it is expected that the B component will be the main

394: source of flux in the EUV wavelengths, but the A component is likely

395: to have a non-negligible contribution of EUV light.

396: 

397: The light curve (Fig.~\ref{plc}) has been phased using the ephemeris

398: given by \citet{griffin98} for the A and B components, and adapted to

399: follow the criteria of $\phi_{orb}$=0 corresponding to the primary

400: star behind, resulting in T$_0$(HJD)=2,442,442.916, P$_{orb}$=3.980507

401: d.  The DS light curves of $\xi$~UMa show some flaring activity, as

402: well as semi-periodic fluctuations similar to those observed in other

403: systems.

404: 

405: The \ion{Fe}{16} line ratio in the MW spectrum (2.25$\pm$0.15)

406: provides an accurate value of the hydrogen column density of

407: $N_H=8.\pm3\times10^{17}$~cm$^{-2}$ (Fig.~\ref{ismcalc}), lower than

408: the value of $N_H=1.5\times10^{18}$~cm$^{-2}$ estimated by

409: \citet{sch95} from nearby stars.  Fig.~\ref{emdfigs} shows the EMD

410: calculated for this system. This is the most outstanding example of

411: two peaks in the emission measure, with very similar values of the

412: emission measure for each peak. This result fits the observed fluxes

413: better than an EMD with only one peak at intermediate temperatures.

414: 

415: 

416: \subsubsection{$\alpha$~Cen}

417: As in the case of $\xi$~UMa, EUVE can not separate the light coming

418: from $\alpha$~Cen A (HD 128620, G2V) and B (HD 128621, K1V).  Hence,

419: the observed light curve and spectra correspond to both stars. IUE

420: spectra show higher flux levels in the A component (see

421: Table~\ref{tabiue}) than in the B component, but {\em EINSTEIN}

422: satellite x-ray observations show that the K1 star is the predominant

423: source in the range 0.2--4~keV \citep{gol82}, with an approximate

424: ratio 2:1.

425: 

426: The DS light curve of $\alpha$~Cen does not show any short-term

427: change, and no flares are registered. Analysis of seasonal changes

428: shows small variations, counting an increase in the flux by $\sim$15\%

429: from 1995 to 1997, while the 1993 observations are compromised by the

430: dead spot.  The hydrogen column density adopted is $N_H=6\times

431: 10^{17}$~cm$^{-2}$, calculated by \citet{lin96} from Lyman~$\alpha$

432: and Mg~II~h~\&~k lines.  As in the case of Procyon, disagreements

433: occur between the observed and predicted fluxes in the \ion{Fe}{12}

434: $\lambda$364.4.

435: 

436: The general shape of the EMD is similar to that of the Sun in the

437: absence of flares, up to log~T(K)$\sim$6.5 \citep*[see][]{rea01}.  The

438: electron density calculated from \ion{Fe}{10}, \ion{}{12}, \ion{}{13},

439: and \ion{}{14} lines in the range log~T(K)$\sim$6.0--6.5 gives a value

440: of log~$N_e(cm^{-3})\sim$9.5 (Mewe et~al. 1995; Drake, Laming, \&

441: Widing 1997).

442: %\citep{mew95,drake97}.

443: 

444: 

445: \subsection{Active stars}

446: %beta Cet, AY Cet, AR Psc, CC Eri, YY Gem, BF Lyn, LQ Hya, DH Leo, BH

447: %CVn, V824 Ara, ER Vul, BY Dra

448: 

449: This group of stars includes those for which the EMD 

450: is clearly dominated by

451: the emitting material at log~T(K)$\sim$6.9 signaled by strong

452: lines of \ion{Fe}{18} and \ion{Fe}{19}, but with relatively small

453: contribution from material at higher temperatures (compared to the

454: most active stars in the third group).  Capella \citep{dup02} is 

455: included in this group, and was also present in the sample of 28 stars

456: studied in \citet{tesis}.   

457: 

458: 

459: \subsection{$\beta$ Cet}

460: $\beta$~Cet (HD 4128, HR 188, K0 III) is a single star with an

461: apparent low rotational velocity \citep*[v$\cdot$sin~i= 3.5 $km\

462: s^{-1}$][]{melo01}, but with surprisingly high levels of activity in

463: X-rays. An estimate of the rotational period of 189.1~d can be deduced

464: from the radius and inclination in Table~\ref{tabparam} and

465: v$\cdot$sin~i \citep{gray89}.  Two sets of EUVE observations are

466: available for $\beta$~Cet. An observation of 6 days during 1994

467: September shows a light curve with no significant variation, and a

468: spectrum and EMD similar to that of active binary systems like Capella

469: \citep{paper1}.  A second set of observations taken during 2000 August

470: shows flaring events in the DS light curve, and a level of emission

471: much higher than in the 1994 campaign suggesting seasonal changes in

472: the level of coronal activity of this star \citep*{ayr01}. The

473: accumulated 808~ks of exposure time in the SW spectrum makes it the

474: longest stellar observation with EUVE, and allows the analysis of a

475: high-quality spectrum.\footnote{ Unfortunately detectors for the MW

476: and LW spectrometer were not turned on for this exposure.}

477: \citet{ayr01} applied models based on upper-limit EMDs to the observed

478: spectra of quiescence and flaring stages during this set of

479: observations\footnote{The observations in the EUVE archive contain one

480: set of corrupted data, and the first $\sim$4 days of observation

481: included in \citet{ayr01} are not available in the archive.}. These

482: authors report a total exposure time of 645~ks, well below the

483: exposure time calculated from archival data.  This affects the fluxes

484: reported by \citet{ayr01}, which are larger than fluxes reported here

485: by $\sim$15-35~\%.  No details are given by these authors on how well

486: their model predicts the line fluxes, and during flaring stages their

487: model spectrum seems not to accurately match some of the observed

488: lines, including the density sensitive lines.

489: 

490: The EMD derived from  the summed 2000 spectrum yields  one of the best

491: fits of the sample (see Fig.~\ref{emdfigs}), although the lack

492: of MW and LW spectra containing the \ion{Fe}{15},

493: \ion{Fe}{16} and \ion{Fe}{24} lines, weakens the constraint on the

494: EMD at log~T(K)$\sim$6.5 and log~T(K)$\sim$7.3.

495: The resulting EMD shows clear differences with respect to the

496: 1994 observations in the temperature range available in this

497: observation, exhibiting a higher EMD mainly for temperatures higher than the

498: bump at log T(K) $\sim$ 6.8 \citep[see][]{paper1}. 

499: 

500: \subsubsection{AY Cet}

501: AY~Cet (HD 7672) is an RS CVn system with the EUVE flux dominated by a

502: G5III active star, and a faint white dwarf companion with negligible

503: flux contribution in this band. 

504: \citet{sch95} had difficulties applying a global fit to the noisy

505: spectrum of AY~Cet, and only the assumption of a low iron abundance 

506: allowed a result

507: without a false ``hot tail'' in the EMD. The global

508: fit yielded  an EMD

509: peaking at log~T(K)$\sim$7.0. Our EMD modeling

510: indicates a peak at  log~T(K)$\sim$6.9 (Fig.~\ref{emdfigs}) with no

511: ``hot tail.'' The shape of our EMD does not depend on abundances,

512: since we use only Fe lines.

513: 

514: \subsubsection{AR Psc}

515: AR~Psc (HD 8357, G7 V + K1 IV) has a shorter photometric period

516: \citep*[P$_{phot}$=12.38 days;][]{cuti01} than the orbital period of

517: 14.3023 days \citep{fek96}, something unusual among RS~CVn

518: stars. \citet{fek96} proposed that AR Psc has not yet arrived on the

519: main sequence to explain the lack of synchronization between the two

520: periods. The EUVE observations of AR~Psc with the DS (Fig.~\ref{plc})

521: reveal the persistence of an active region during a full rotational

522: period.  This active region is visible around JD$\sim$2,450,690.2 and

523: JD$\sim$2,450,702.5, approximately the duration of the optical

524: photometric period.  Enhanced emission is also observed after the

525: second appearance of this active region; this feature can not be

526: unambiguously considered as a flare since the low enhancement of light

527: ($\sim$50\%) does not match the duration of the event ($\sim$1 day)

528: when compared to other flares of similar duration

529: \citep{ost99,paper1}.\footnote{\citet{ost99} report the presence of two

530: flares in the AR~Psc light curve, with the first flare starting at

531: phase $\phi_{orb}$=1.082.  This is an orbital phase when no data were

532: taken and does not correspond to the date reported by these authors.

533: In the second flare, at $\phi_{orb}$=1.378, 

534: the beginning of the second enhancement that we identify,

535: a rise time of 51.3~hr and a decay time of 42.1~hr are reported

536: as e-folding times, but these times do not match the duration of the

537: observed rise and decay of the flare.

538: }

539:   

540: There is much uncertainty in the hydrogen column density used for this

541: star. A value of $N_H=2\times10^{18}$~cm$^{-2}$ is assumed, estimated

542: from column densities measured in nearby stars \citep{fru94}.  AR Psc

543: shows a fairly small bump in the EMD (Fig. \ref{emdfigs}), but peaking

544: at a higher temperature than usual, at log~T(K)$\sim$7.1. A better

545: knowledge of the hydrogen column density is needed to specify the EMD

546: particularly in the lower T regions defined by the iron lines at long

547: wavelengths.

548: 

549: 

550: \subsubsection{CC Eri}

551: Similar to other systems observed in this sample, small-scale

552: variability is  present in the light curve of 

553: CC~Eri (HD 16157, K7 V + M3 V),

554: with non-periodic variations of 10--14~hr (Fig.~\ref{plc}).   

555: A short-duration

556: flare could be present at JD$\sim$2,449,977.5.

557: A value of $N_H=2.6\times10^{18}$~cm$^{-2}$ was estimated by 

558: \citet{pan95}

559: from the distance to the star and the average hydrogen column density.

560: \citet{ama00}, using this value, derived an

561: upper limit to the  EMD based on the   

562: temperature of maximum emissivity of IUE and EUVE spectral lines. Their

563: derived  EMD has a minimum around log~T(K)$\sim$5.0 and a peak at

564: log~T(K)$\sim$6.8.   

565: 

566: In the EMD we calculate with the whole emissivity function, the 

567: resulting   

568: distribution (see Fig.~\ref{emdfigs}) shows a minimum  at

569: higher temperatures (between 5.5 -- 5.8 dex), and 

570: an apparent  overabundance of

571: nitrogen is indicated by the \ion{N}{5} $\lambda$1240 line, similar to

572: the cases mentioned in \citet{paper1}.  The local enhancement in

573: the EMD occurs at log T(K)= 6.8.

574: 

575: 

576: \subsubsection{YY Gem}

577: YY~Gem (Castor C, HD 60179C) is a well known active system with two M

578: dwarf stars in the double-lined spectrum.  The DS light curves are

579: characterized by the presence of many short flaring events (at least

580: 7), including an enhancement by a factor of at least $\sim$9 at the

581: very end of the observation (Fig.~\ref{plc}).  Lower flux levels are

582: displayed in the right panel to show the small-scale variability. This

583: system has an orbital inclination of 86.3$\arcdeg$ (see

584: Table~\ref{tabparam}), making it one of the best targets to search for

585: rotational modulation and eclipses, as have been found with ROSAT

586: \citep{schm98}. But the presence of frequent flaring, along with

587: the noise found in the data, makes it difficult to find such evidence

588: from the present observations.

589: 

590: The hydrogen column density towards nearby stars \citep{fru94}

591: suggests a column density of $N_H \sim 2.5\times 10^{18}$~cm$^{-2}$ to

592: YY Gem, but this value seems to be high compared to nearby stars in

593: our sample.  The use of the \ion{Fe}{16} line ratio

594: ($\lambda$335/$\lambda$361) in the MW (1.8$\pm$0.3) yields an upper

595: limit to the column density of $N_H \la 6\times 10^{17}$~cm$^{-2}$

596: (Fig.~\ref{ismcalc}). In view of these divergent results, we use the

597: upper limit given by the \ion{Fe}{16} line ratio, since it represents

598: a direct measurement towards this star.

599: 

600: The EMD of the system (see Fig.~\ref{emdfigs}) can be used to predict

601: the \ion{Ar}{15} $\lambda$221.15 line using APEC and good agreement is

602: found, assuming [Ar/Fe]~0.8 (a noble gas enhancement) 

603: and  a solar oxygen abundance \citep{anders}.

604: Also, the complex blend at $\lambda$192~\AA\ includes the

605: \ion{Fe}{12} $\lambda$193.51 line, which may contribute $\sim$40\% of

606: the observed flux. These lines give some information regarding the

607: stellar EMD near log~T$_e(K)\sim$6.3, although a better determination

608: of the ISM absorption would be helpful.

609: 

610: 

611: \subsubsection{BF Lyn}

612: 

613: Variations by up to $\sim$20\% with respect to the average value are

614: observed in the light curve of BF~Lyn (HD 80715, K2 V + dK),

615: suggesting the presence of small-scale modulation in a semi-periodic

616: pattern.  As in the case of YY~Gem, the value of the hydrogen column

617: density towards BF~Lyn is very uncertain, although the presence of the

618: \ion{Fe}{16} lines makes possible an estimate from their ratio

619: (1.54$\pm$0.63) of $N_H \la 5\times 10^{17}$~cm$^{-2}$, as shown in

620: Fig.~\ref{ismcalc}. Since the S/N of these lines is quite low, an

621: intermediate value to that reported in nearby stars (similar to those

622: in the case of YY~Gem) has been adopted.  The assumed value in the EMD

623: calculations is $N_H=1.5\times 10^{18}$~cm$^{-2}$, more consistent

624: with the EMD shape estimated when lines at long wavelengths are

625: excluded from the fit.

626: 

627: 

628: \subsubsection{LQ Hya}

629: LQ Hya (HD 82558) is one of the single stars included in the

630: sample. The youth of this star appears to be the main cause of the

631: high levels of activity observed \citep[see][ and references

632: therein]{mon99}. The EUVE light curve (Fig.~\ref{plc}) shows

633: variations by a factor of 2 during the ``quiescent'' state of the

634: corona, and also two impulsive flares (intense flares of short

635: duration) are present.  The variations of the ``quiescent'' corona do

636: not follow a clear periodic pattern, and are not related to the

637: photometric period of 1.63~days \citep{cuti01}, but optical modulation

638: is present on a scale of $\sim$1.2 days.

639: 

640: \citet{woo00} find an abnormally high value of the hydrogen column

641: density of $N_H= 1.1^{+0.5}_{-3.1}\times 10^{19}$~cm$^{-2}$, from an

642: analysis of Lyman~$\alpha$ and Mg~II~h~\&~k lines.  Since the Mg lines

643: could be affected by stellar activity and and circumstellar gas, and

644: data from nearby stars seems to disagree strongly with these values,

645: we used the conservative value given by the lower limit, $N_H=8\times

646: 10^{18}$~cm$^{-2}$.  Given the lack of lines at longer wavelengths in

647: the spectrum of LQ Hya (that could be affected even more by

648: uncertainties in the adopted value of $N_H$), changes in the shape of

649: the EMD due to such uncertainties will be minimal, leading mainly to a

650: vertical displacement of the EMD.  Good agreement is obtained between

651: the predicted and observed fluxes (see Fig.~\ref{emdfigs}).

652: 

653: 

654: \subsubsection{DH Leo}

655: The observed 1995 light curve of DH~Leo (HD 86590) shows much

656: variability (changes by up to $\sim$37\% from the average value,

657: Fig.~\ref{plc}), and some short flare-like events

658: occur as well. \citet{ste96} proposed that rotational modulation could be

659: present in these observations.  The \ion{Fe}{16} line ratio does not

660: provide an accurate value of the column density in DH~Leo since the

661: 361\AA\ line has poor statistics. But the low flux observed in this

662: line points to a rather high column density. Values of nearby stars in

663: \citet{fru94} suggest $N_H \sim 1\times 10^{18}$~cm$^{-2}$. On the

664: other hand, \citet*{diam95} deduced a wide range of values of

665: $N_H=4.^{+28}_{-N/A}\times 10^{18}$~cm$^{-2}$ obtained from a fitting

666: to ROSAT spectra. An intermediate value with those of nearby stars has

667: been adopted, $N_H\sim 2\times 10^{18}$~cm$^{-2}$.

668: 

669: 

670: \subsubsection{BH CVn}

671: BH CVn (HD 118216, HR 5110) shows a remarkable pattern of variability

672: in the EUVE light curve (Fig.~\ref{plc}), with semi-periodic

673: variations of $\sim$20--30~hr.

674: 

675: There is much uncertainty in the determination of the hydrogen column density

676: in the direction of BH~CVn. Spectral fits to ROSAT data yield very

677: high column densities 

678: (over $N_H \ga 1\times 10^{19}$~cm$^{-2}$, Diamond et~al. 1995;

679: {Graffagnino}, {Wonnacott}, \& {Schaeidt} 1995),

680: %\citep[over $N_H \ga 1\times 10^{19}$~cm$^{-2}$,][]{diam95,graf95}, 

681: although this is not a reliable

682: method to determine $N_H$. The \ion{Fe}{16} line ratio of

683: 1.45$\pm$1.17 does not point to an accurate value either, due to the

684: low S/N of the lines, but the upper limit of this ratio corresponds to

685: $N_H \sim 3\times 10^{18}$~cm$^{-2}$ (Fig.~\ref{ismcalc}). On the

686: other hand, the values derived from nearby stars \citep{fru94} yield a

687: lower estimate of $N_H \sim 7\times 10^{17}$~cm$^{-2}$. As a

688: compromise among these values we have chosen the upper limit given by

689: the \ion{Fe}{16} line ratio, $N_H= 3\times 10^{18}$~cm$^{-2}$, 

690: also consistent with \citet{mit97}.  The

691: calculated EMD (Fig.~\ref{emdfigs}) shows a bump similar to other

692: stars, with a decreasing EMD at higher temperatures.

693: 

694: 

695: \subsubsection{V824 Ara}

696: About 2.5 days of DS light curve observations were obtained for

697: V824~Ara (HD 155555) in 1996. The low flux (note the 3000~s binning)

698: gives quite large error bars (see Fig.~\ref{plc}). Nevertheless, it is

699: possible to identify some modulation coincident with the orbital and

700: photometric periods of the system ($\sim$1.68 days).  Local maxima

701: appear at phases 1.0, 1.5, and 2.0, consistent with orbital

702: modulation, although flare-like variability can not be excluded.

703: 

704: The small number of lines present in the spectrum of this young active

705: binary system makes the estimate of the EMD less accurate.

706: \citet{aira98} made an analysis based on the peak of the emissivity

707: function in order to calculate minimum values of the emission

708: measure. The resulting EMD (Fig.~\ref{emdfigs}) shows a poorly

709: constrained bump near log~T(K)$\sim$6.9. The electron density

710: (Table~\ref{tabne}) derived from the ratio \ion{Fe}{21}

711: $\lambda$102.22/$\lambda$128.73 is not very reliable, since the

712: $\lambda$102.22 line flux was measured including a blend with the

713: \ion{Fe}{19} $\lambda$101.55 line. The flux of the \ion{Fe}{19} line

714: was estimated from the EMD to account for its contribution to the

715: blend.

716:  

717: 

718: \subsubsection{ER Vul}

719: The DS light curve of the partially eclipsing system ER~Vul (HD

720: 200391) shows some variability, as well as flare-like activity at a

721: low level. \citet{ost99} suggested the variations resulted from

722: small-scale flares, and found no eclipses or periodicities in this

723: BY~Dra-type system.   The EUVE observations span

724: $\sim$10 epochs of the binary enhancing detection of

725: periodicity.  A power spectrum of  the DS light curve 

726: (with flaring portions omitted) shows a maximum corresponding to the

727: optical period.  When the DS light curve is phased to this period,

728: there is modulation at most by a factor of 5 with a suggestion of 

729: absorption near phase 0.5  when the G0V component is partially

730: occulted by the G5V star.  Thus the corona may be

731: more compact in the hotter star of ER~Vul.

732: 

733: There is much uncertainty in the value of the hydrogen column density

734: towards ER Vul, since the values towards nearby stars reported in

735: \citet{fru94} show discrepancies.  The value adopted by \citet{ruc98}

736: of $N_H=3\times 10^{18}$~cm$^{-2}$ has also been used in this

737: work. This uncertainty could affect the determination of the slope at

738: temperatures below the bump in the EMD because that region is defined

739: by lines at longer wavelengths where the impact of interstellar

740: absorption is largest.

741: 

742: 

743: \subsubsection{BY Dra}

744: The DS light curve of BY Dra (HD 234677) covered an orbital

745: period of this system (Fig.~\ref{plc}). 

746: Although some variation occurs that could 

747: correspond to the rotational period of the system of $\sim$3.83 days (see

748: Table~\ref{tabparam}), several flares  prevent a clear 

749: identification of rotational modulation.

750: 

751: Values from nearby stars suggest a quite high hydrogen column

752: density ($N_H=5\times 10^{18}$~cm$^{-2}$), 

753: consistent with the lack of flux detected in the MW spectrum,

754: probably due to ISM absorption. We have used this 

755: value in the absence of other evidence.  Unless large differences 

756: in the hydrogen column are present with respect to this 

757: assumed value, there will be only a minimal influence on the 

758: shape of the EMD (Fig.~\ref{emdfigs}).

759: 

760: 

761: The electron density calculated for BY Dra makes use of lines

762: with S/N lower than 3, and hence these values are not very reliable.

763: The EMD is fit to emissivities at log~N$_e$(cm$^{-3}$)$\sim$12, which

764: gives a good fit to the density-sensitive resonance line \ion{Fe}{21}

765: $\lambda$128.7. 

766: 

767: 

768: \subsection{Very active stars}

769: %VY Ari, sigma2 CrB, V478 Lyr

770: Some stars in the sample contain substantial amounts of material at

771: temperatures beyond log~T(K)$\sim$7.0, sometimes with 

772: emission measures even larger than the

773: values at log~T(K)$\sim$6.9.  Most of those stars 

774: included in \citet{paper1}, UX~Ari, V711~Tau, $\sigma$~Gem, II~Peg, and

775: AB~Dor belong in this category of high activity. The 

776: presence of material at these temperatures is 

777: determined by the \ion{Fe}{23} and \ion{Fe}{24} lines.

778: \ion{Fe}{24} occurs in the MW spectrum which frequently suffers

779: from insufficient exposure time in current 

780: data sets.  Future observations, for instance with

781: Chandra or XMM-Newton,  could reveal high temperature

782: emission that was not detected in short exposures with EUVE. 

783: 

784: \subsubsection{VY Ari}

785: The DS measurements of  VY~Ari (HD 17433)  are compromised 

786: by the dead spot of the DS detector during the observation 

787: (Fig.~\ref{plc}). 

788: However, the SW light curve supports the modulation observed in

789:   the DS light curve,

790:   with a decrease of flux of $\sim$60\% during the observation.

791: %Although the light curve suggests  some

792: %modulation, the contamination due to the dead spot effect prevents

793: %a reliable conclusion.

794: 

795: We have used the same value for the  hydrogen column density as in the

796: case of the nearby star UX~Ari

797: \citep[$N_H=1.5\times10^{18}$~cm$^{-2}$,][]{paper1}.

798: In constructing the EMD, we find that

799: the predicted fluxes

800: in two of the \ion{Fe}{20} lines do not well match the measured values.

801: While the observed flux of $\lambda$110.63 is too strong,

802: the $\lambda$121.83 transition  

803: is too weak. These lines could be affected by other blends

804: not included in the analysis.

805: In any case, this makes  the electron 

806: density (cm$^{-3}$) of 13.8 dex deduced from the 

807: ratio \ion{Fe}{20} $\lambda$110.63/($\lambda$118.66+$\lambda$121.83) 

808: less certain, and  points towards a lower

809: value, probably closer to that deduced from

810: \ion{Fe}{21} ratios of $\sim$12.5 dex (see Table~\ref{tabne}).

811: The resulting EMD (Fig.~\ref{emdfigs}) is not very sensitive to the

812: use of a different electron density in this range (the adopted value

813: was log~N$_e$[cm$^{-3}$]$\sim$13.0).

814: 

815: 

816: 

817: \subsubsection{$\sigma ^2$~CrB}

818: 

819: The DS light curve of $\sigma ^2$ CrB (HD 146361) shows intense flares

820: (Fig.~\ref{plc}), and also small-scale variability on the order of

821: several hours ($\sim$6--10~hr). The flare timing of this observation

822: was analyzed in detail by \citet{ost99} and \citet{ost00}.

823: 

824: The good S/N achieved in the \ion{Fe}{16} lines in the MW spectrum

825: allows an accurate determination of the hydrogen column density,

826: resulting in a value of $N_H=2.5^{+1.5}_{-0.9}\times

827: 10^{18}$~cm$^{-2}$ (Fig.~\ref{ismcalc}), consistent also with

828:   \citet{mit97}.  Fig.~\ref{emdfigs} shows an 

829: EMD that reflects a very hot corona, with an increasing value at

830: temperatures higher than the ``bump.''  Also the general level of the

831: EMD is comparable to that of RS~CVn systems composed of subgiants,

832: whereas the components of $\sigma^2$ CrB are dwarf stars,

833: demonstrating their unusually high activity levels.

834: 

835: 

836: \subsubsection{V478~Lyr}

837: The lack of good statistics in the DS light curve (Fig.~\ref{plc}) of

838: V478~Lyr (HD 178450) masks possible low level variability.  Variations

839: of $\sim$25\% with respect to the average flux are found in the data,

840: with no clear evidence of DS eclipses. The system has an inclination

841: of 83$\arcdeg$ and undergoes partial eclipses \citep{cabs}.

842: 

843: An intermediate value of the hydrogen column density, $N_H=4\times

844: 10^{18}$~cm$^{-2}$ has been used from estimates for stars nearby in

845: the sky \citep{fru94}.  The  EMD is not affected

846: significantly by the uncertainties in $N_H$ (see

847: Table~\ref{tabparam}), since all the emission line fluxes are taken

848: from the SW spectrum, where interstellar optical depths are small.

849: Although the number of lines measurable with reliable statistics in

850: this system is small, it was possible to construct an EMD similar to

851: other stars in the very active group.

852: 

853: 

854: \subsubsection{AR Lac}\label{arlac}

855: AR~Lac (HD 210334) is an eclipsing binary with two active stars

856: (G2IV/K0IV) that shows chromospheric emission originating from the

857: K0IV star \citep{mon97}. The measurement of the primary eclipse depth

858: in the 1993 and 1997 observations (Fig.~\ref{plc}) reveals a

859: contribution by the K0IV star of at least a $\sim$37\% (measured at

860: the center of the primary eclipse) of the total EUV light.  The

861: partial contamination of the light curve by the dead spot in 1993, and

862: the presence of flares around phases x.5 in 1993 and 1997 prevent a

863: reliable measurement of the secondary eclipse.  The light curve in the

864: 2000 observations, reported by \citet{pea01}, is dominated by a large

865: flare.  The EUV flux increases by at least a factor of $\sim$17 from

866: the quiescent level. A total net energy release of 2.0$\times

867: 10^{35}$erg is found in the range 80--170~\AA, after the subtraction

868: of the ``quiescent'' contribution of the 1993 and 1997 summed

869: observations, following the method explained in \citet{paper1}.  The

870: flux obtained before the subtraction of the quiescent contribution was

871: 4.4$\times 10^{35}$erg. A complementary analysis of the observations

872: in 2000, with partially simultaneous Chandra data can be found in

873: \citet{huen01}.

874: 

875: During the 2000 flare, AR~Lac shows an increase in flux in the hottest

876: lines of the EUVE spectrum, and some increase also in the continuum

877: level.  There may be an increase of the average density measured

878: during the flaring observation as indicated by \ion{Fe}{22} but not

879: confirmed by other ratios (see Table~\ref{tabne}).

880: 

881: 

882: Although there is some uncertainty in the measurement of the

883: \ion{Fe}{16} lines (especially the $\lambda$360.8 line), we use the

884: hydrogen column density derived from their flux ratio (2.5$\pm$0.8)

885: found in the MW spectrum corresponding to the 1993 and 1997

886: observations co-added (Fig.~\ref{ismcalc}).  The resulting value

887: $N_H(cm^{-2})=1.8\times 10^{18}$ is consistent with the column density

888: of $N_H(cm^{-2})=2\times 10^{18}$ assumed by \citet{gri98}, calculated

889: from nearby stars.

890: 

891: \citet{kaa96} and \citet{gri98} used the 1993 data to obtain EMDs with

892: a clear peak around log~T$_e(K)\sim$6.9, although they differ in the

893: distribution at other temperatures. A preliminary analysis of the 1997

894: data alone was made by \citet{bri99}. The addition of the 1993 and

895: 1997 data allows the analysis to be extended to almost the whole range

896: of log~T$_e(K)\sim$4.0--7.4.  The 2000 data are used to compare a

897: flaring sequence to the ``quiescent'' observations of 1993 and

898: 1997\footnote{Although these observations contain some flaring

899: activity, they are dominated by the quiescent state.}.

900: 

901: Fig.~\ref{emdfigs} displays the EMD calculated from the 1993 and 1997

902: summed data.  The height of the bump is less prominent during the

903: flare because the material at higher and lower temperatures increases.

904: However the value of the EMD at the bump remains constant

905: demonstrating its stability in the corona (Fig.~\ref{stagesarlac}).

906: This reinforces the hypothesis of large flares as phenomena unrelated

907: to the bump at log~T$_e(K)\sim$6.9, suggested previously by

908: \citet{paper1}.

909: 

910: 

911: \subsubsection{FK Aqr}

912: The nearby active binary system FK Aqr (HD 214479) is composed of two M

913: dwarf stars orbiting with a $\sim$4 day period. 

914: The DS light curves in the 1997 campaign show frequent 

915: flaring activity  (Fig.~\ref{plc}), with impulsive flares of 

916: short-duration ($\sim$15~hr) that are quite strong 

917: (increases by up to a factor of 5). The presence of these 

918: flares blurs any possible

919: modulation related to rotation. In contrast, the 1994 observations

920: show a quiet corona, and only a small

921: flare is present. Some enhancement of flux arises during the

922: second half of the orbital period in 1994, probably due to the presence of an

923: active region in the line of sight, as \citet{ste96} proposed.

924: Values of inclination and photospheric radii of this system can be

925: inferred in a first approximation by assuming that the

926: pair of M2V/M3V dwarfs follows the relations derived by \citet{gray92}. 

927: This would imply stellar radii of 0.55/0.52~R$_{\odot}$, 

928: and with the typical

929: mass of 0.42 and $m\, sin^3 i=0.27$ for the primary, an inclination of

930: i$\sim$60$\arcdeg$ can be estimated. 

931: 

932: There are no direct measurements of

933: the interstellar column density towards this system, 

934: but stars nearby in the sky \citep{fru94}

935: suggest

936: a value of $N_H=7-10\times 10^{17}$~cm$^{-2}$. The \ion{Fe}{16} line

937: ratio in the MW spectrum (1.96$\pm$0.20) provides an upper limit to

938: the hydrogen column density of $N_H=7\times 10^{17}$~cm$^{-2}$

939: (Fig.~\ref{ismcalc}). This 

940: value was assumed here.

941: Fig.~\ref{emdfigs} shows a remarkably well-fitted EMD, with a first

942: bump in temperature similar to the solar EMD \citep[see][]{rea01},

943: and the typical second bump found in most of the stars in

944: this sample around log~T$_e(K)\sim$6.9.  

945: 

946: 

947: %------------------------------------

948: \section{Discussion}\label{sec:discussion}

949: 

950: One goal of this research is to identify the stellar 

951: parameters that influence the coronal structures.

952: Systematics of the light curves, EMD, and densities are

953: discussed below. We then compare the  underlying patterns of coronal

954: structure to the properties of the stars.

955: 

956: \subsection{Light Curves}

957: The intrinsic variability found in the light curves of these systems

958: makes it difficult to confirm seasonal (year-to-year) changes in the EUV

959: emission. Among the quiet stars, only $\alpha$~Cen shows some small

960: decrease in the light curve flux (by $\sim$15\%) between 1995 and

961: 1997, but the observations are too short to attribute these changes to

962: seasonal variations. The fluctuations found in the quiescent emission

963: of II~Peg are of order $\sim$50\% \citep{paper1}, although in this

964: case the effects of intrinsic variations could be more important. The

965: clearest case for seasonal fluctuations is found in the single giant

966: star $\beta$~Cet, which in 1994 is very quiet for the 6.4 day

967: observation. The DS flux is about a factor of 2 lower than in 2000

968: \citep{paper1}, and \citet{ayr01} identify at least 5 flaring events

969: during the $\sim$34 days of monitoring in 2000 (see Fig. 1).  Some

970: line fluxes are enhanced by more than 50\% (a factor of 5 for

971: \ion{Fe}{23}/{\sc \small XX} $\lambda$132.85!). The intrinsic

972: variability found in stars like $\sigma$~Gem, LQ Hya or FK Aqr, even

973: in the absence of large flares, makes it difficult to confirm seasonal

974: changes.

975: 

976: Signatures of rotation are clearly found in the eclipses of AR~Lac,

977: and might be present in the light curves of AR~Psc, ER~Vul,

978:   VY~Ari and BY~Dra.

979: Marginal evidence for modulation was found previously in II Peg, UX

980: Ari, $\sigma$ Gem, and AB Dor \citep{paper1}.  Additional short-term

981: variations are found in many stars in this sample, indicating

982: non-periodic changes on time scales between 0.3 and 1.5 days.

983: Instrumental effects can be rejected in the case of V711~Tau, and

984: these variations are likely to be real for many stars in the sample,

985: including V711~Tau, LQ~Hya, $\sigma^2$~CrB and BF Lyn (see

986: Fig.~\ref{plc}).  This variability can be attributed to flare-like

987: phenomena on a small scale, perhaps reflecting the existence of many

988: solar-like flares (shorter duration and intensity than the large

989: flares which are well observed in these stars).

990: 

991: 

992: \subsection{Emission Measure Distribution}

993: 

994: A continuous distribution of material occurs in these coronas,

995: spanning 3 decades of temperature or more.  It is now clear that the

996: earlier simplifications of 1-T or 2-T coronal models resulted from

997: insufficient spectral resolution and/or incomplete global models

998: preventing identification and analysis of the wide range of excitation

999: and ion stages naturally present in the coronae of cool stars.

1000: Results assembled in Fig. 4 show the distribution of material to be

1001: generally similar among active cool stars, whether single or binary.

1002: A decrease in the EMD between 10$^4$K and 10$^{\sim 5.5}$K and then an

1003: increase, with one or more local enhancements between 10$^6$ and

1004: 10$^7$K represent the structure of the stellar chromosphere,

1005: transition region, and corona.

1006: 

1007: However, a detailed comparison shows that clear differences appear at

1008: temperatures above log~T(K)$\sim$5.2.  Only $\alpha$~Cen, with a

1009: solar-like EMD, has a distribution that begins increasing above

1010: temperatures as low as log~T(K)$\sim$5.2.  The remaining stars in the

1011: sample have the minima in the EMD at higher temperatures

1012: [log~T(K)$\sim$5.8]. This difference between the minima of the Sun and

1013: Capella was noted from the first EMD derived from EUVE \citep{dup93}.

1014:   We note that the Sun shows minima at different

1015: temperatures for different types of structures. For example, solar

1016: coronal holes frequently have a higher temperature EMD minimum than do

1017: active regions.

1018: 

1019: 

1020: At temperatures above 10$^6$ K, additional structures are found,

1021: characterized by local enhancements of the EMD over narrow temperature

1022: ranges.  Such a feature was also discovered in Capella and labeled the

1023: ``bump'' \citep{dup93,bri00}, and have been

1024: identified with high latitude coronal features \citep{bri98}.  Several

1025: stars show a local enhancement at log T(K)=6.2 

1026: reminiscent of the temperature of the solar corona in addition to a

1027: bump at log T(K) =6.8.  The presence of this second high temperature

1028: bump is unambiguously found in 25 out of 28 stars in the sample (all

1029: except $\epsilon$~Eri, $\alpha$~Cen, and Procyon).  Clearly, these

1030: represent a fundamental coronal structural feature.  Finally, the

1031: stars with an increasing EMD beyond log~T(K)$\sim$6.9 represent the

1032: very active classification.  The progressive addition of hotter

1033: material marks the increase in activity level.

1034: 

1035: Given the different physical sizes of the stars in the sample, it is

1036: useful to evaluate the emission measure weighted by the emitting

1037: surface of the stars (4$\pi$[R$_1^2$+R$_2^2$]), so that the

1038: emission measure per unit area defines an effective column density of coronal

1039: material. 

1040: Fig.~\ref{sevstars} shows 6 cases representative of

1041: different degrees of activity, weighted by the size of the emitting

1042: stars -- the ``column'' EMD.   Procyon  shows the same amounts of

1043: material as $\alpha$~Cen at temperatures above log~T(K)$\sim$6. 

1044: Except for $\alpha$~Cen, 

1045: the chromospheric structures are similar, while the greatest

1046: divergence occurs at the highest temperatures. 

1047: 

1048: 

1049: \subsection{Electron Density}

1050: Electron densities derived from line ratios at log~T(K)$\sim$7.0

1051: indicate values of\\ log~$N_e(cm^{-3})\ga$12. There is considerable

1052: dispersion in the data for values of log~$N_e(cm^{-3})\ga$13, as noted

1053: previously for Capella \citep{bri96}, as well as for densities at

1054: lower temperatures in solar active regions \citep{bri95}. Numerous

1055: observational issues compromise these results. The presence of blends

1056: not well evaluated in the models and uncertainties in the placement of

1057: the continuum used as a base for the line flux measurements clearly

1058: influence some of these measurements.  In particular, the \ion{Fe}{20}

1059: line ratios show a systematically higher density than other iron line

1060: ratios at similar temperatures, as well as a poorer fit of the

1061: emission measure; therefore more caution must be taken with the

1062: results from this ion. While the atomic models for the diagnostic line

1063: ratios have been benchmarked under controlled conditions

1064: \citep{four01}, higher resolution spectra with good signal to noise,

1065: are needed.

1066: 

1067: Nevertheless, density diagnostics from several stars in this sample

1068: have good statistics and consistency among several diagnostics,

1069: e.g. VY~Ari and CC~Eri discussed here and V711~Tau and $\sigma$~Gem

1070: \citep{paper1}.  Given these results, a ``conservative'' value of

1071: log~$N_e(cm^{-3})\sim$12 seems plausible.  We note that lower values

1072: of density at log~T(K)$\sim$6.2 found by several authors in different

1073: stars \citep[see \S~\ref{sec:lowactive} and, e.g.][]{can00} are not

1074: necessarily inconsistent. \citet{bri96} used similar results for

1075: Capella from Fe M-shell and L-shell diagnostics, and suggested the

1076: presence of two different types of structures.

1077: 

1078: The densities calculated at log~T(K)$\sim$7.0, in combination with the

1079: emission measure values at that temperature, can be used as a first

1080: approximation to estimate the scale size of the emitting 

1081: structures \citep{sanz01}, with the caveat that there is no

1082: information on their geometry and filling factors. The

1083: calculations show in all cases that such structures 

1084: are small ($\la$0.02~R$_*$), both for dwarfs and giant stars.

1085: Small structures have been found also by several authors, e.g.,

1086:   \citet{dup93},  \citet*{bow00}, \citet{phi01} and references

1087:   therein.

1088: 

1089: Current loop models cannot accommodate the presence of the high electron

1090: densities found here at coronal temperatures. Observations of the 

1091: Sun only detect such high 

1092: values during solar flares \citep{phil96}.  Light curves from EUVE 

1093: do not have enough signal in short integration periods

1094: to detect the presence of fast solar-like flares in stars, so it is 

1095: not possible to differentiate

1096: such flares from the emission in quiescence. The presence  of frequent solar-like 

1097: flares in these stars can not be ruled out or confirmed with

1098: the present data.  Such flaring could account for both the

1099: shape of the EMD and the high electron densities. 

1100: A model of (continuous) nano-flare heating in the Sun can predict the shape

1101: of the EMD with a bump at log T(K)=6.8 \citep{kli01},

1102: but with densities two orders of magnitude lower than observed here.

1103: 

1104: \subsection{Comparison with Stellar Properties}

1105: 

1106: For a quantitative comparison of these EMDs, we extract

1107: parameters of the distribution for comparison to physical

1108: properties of the systems.  To define the temperature of

1109: the peak of the EMD, we consider

1110: the three largest values of the emission measure in

1111: the range log~T(K)=5.8--7.3.

1112: The temperature defined for the peak and its emission measure

1113: were compared to the orbital

1114: periods\footnote{The optical photometric period was employed in the absence

1115: of orbital period.} of 30 stars, as shown in

1116: Fig.~\ref{periods}. This sample includes also the data from

1117: Capella \citep{dup93}, 44~Boo \citep{bri98}, $\lambda$~And

1118: \citep{sanz01}, the 6 stars (V711 Tau, II Peg, $\sigma$ Gem,

1119: UX Ari, AB Dor, and $\beta$ Cet) in \citet{paper1}, and the Sun

1120: during  solar maximum \citep*{orl00}. 

1121: 

1122: The temperature of the peak of the EMD (Fig.~\ref{periods}a) remains

1123: relatively constant at log T(K)= 6.9 for the binary stars

1124: and 3 single stars (AB Dor, LQ Hya, and $\beta$ Ceti).

1125: AB Dor and LQ Hya are young rapidly rotating effectively

1126: single stars. Beta Ceti appears as an anomaly with strong

1127: emission  - an apparently 

1128: single slowly rotating star - consistent

1129: with its classification as  K0 III giant. It may be a 

1130: clump star that experiences a regeneration of its magnetic

1131: dynamo, or it may be oriented pole-on to our line of sight, 

1132: and the rapid rotation is not observable.

1133: 

1134: The mean electron densities at log T(K)=6.9 for the sample

1135: are shown in Fig.~\ref{periods}b.  Density values range between

1136: 10$^{12}$--10$^{13.5}$ cm$^{-3}$ with no systematic dependences

1137: on binarity or orbital period for the systems with P$_{orb}<$20~d.

1138: There is no evidence for the highest densities in longer period

1139: systems, but only 3 objects are in that group. 

1140: 

1141: Values of the EMD at the peak temperatures are shown in

1142: Fig.~\ref{periods}c and \ref{periods}d.  For binaries, the value of

1143: the EMD increases with increasing 

1144: orbital period, as dwarf stars tend to have shorter periods

1145: in our binary sample, whereas the larger RS~CVn subgiants and

1146: giants have longer periods. 

1147: Scaling the emission measure at its peak value

1148: by the areas of the stars reveals the decrease in ``column'' 

1149: EMD with increasing orbital period.  There may be a saturation

1150: of the ``column'' EMD at periods less than 2.3 days where a constant

1151: value appears consistent with the data.

1152: 

1153: 

1154: Line fits to the data are 

1155: superimposed, indicating the best fit for all the objects,

1156: \begin{displaymath}

1157: log (EM_{peak}/area)=50.4-0.85\ log P_{orb}

1158: \end{displaymath}

1159: 

1160: \noindent

1161: and only those with period longer  than 2 days,  

1162: 

1163: \begin{displaymath}

1164: log (EM_{peak}/area)=50.5-0.89\ log P_{orb}

1165: \end{displaymath}

1166: 

1167: \noindent

1168: where P$_{orb}$ is given in days, EM in units of cm$^{-3}$, and area

1169: is defined by 4$\pi$[R$_1^2$+R$_2^2$] , with radius 

1170: in solar units.  In all

1171: cases a solar photospheric abundance was assumed. Since the absolute

1172: value of the emission measure peak depends on the iron to hydrogen

1173: abundance, this assumption requires testing. Furthermore, 

1174: if there are substantial iron enhancements or depletions as a function

1175: of coronal temperature, the EMDs will need to be reconstructed

1176: accordingly.

1177: 

1178: We also plot in Fig.~\ref{lumperiod} the orbital periods of the

1179: systems against the EUV (80--170~\AA) luminosity weighted

1180: by the bolometric luminosity (L$_{EUV}$/L$_{bol}$). The EUV luminosity

1181: was calculated 

1182: from the integrated SW spectrum of the stars, corrected for the

1183: instrumental effective area and interstellar absorption. The bolometric

1184: luminosity results from the application of the bolometric corrections

1185: available in \citet{flo96}. 

1186: This plot may be affected by

1187: uncertainties in the calculation of the bolometric luminosity. The

1188: EUV flux may arise from one star in a binary, yet the magnitude ($V$)

1189: and color ($B-V$) of the system are used to calculate the 

1190: bolometric luminosity.  The results in Fig.~\ref{lumperiod}

1191: show much dispersion.  However, the general behavior of

1192: increasing $L_{euv}/L_{bol}$ with shorter period  and a 

1193: possible saturation at periods $\sim$ 1 day or less

1194: are consistent with studies of L$_X$/L$_{bol}$ of

1195: active cool stars 

1196: (cf. Walter \& Bowyer 1981; Pallavicini et al. 1981; {Fleming},

1197: {Gioia}, \& {Maccacaro} 1989).

1198: %\citep[cf.][]{wal81,pal81,fle89}.

1199: An increase of flux in X-rays is generally found for faster 

1200: rotators, with the relation

1201: becoming ``flat'' at some point at rotational periods between 1 and

1202: 10 days, marking a so-called ``saturation limit''.  In the sample

1203: considered here, a simple linear relation can fit  the

1204: data, although the stars with fastest rotation ($\la$ 2.3 d) 

1205: follow a flatter distribution with  period. 

1206: 

1207: Our results  suggest the

1208: presence of three kinds of structures,  at

1209: temperatures of\\ log~T(K)$\sim$6.3, $\sim$6.8, and  log~T(K)$\ga$7.2,

1210: that dominate the emitting coronae of cool stars. Loop 

1211: models predict a shape

1212: roughly similar to that deduced here from the combined UV and EUV

1213: analysis \citep[cf.][]{paper1}. These models generally balance

1214: radiative losses by a heating function, and conduction redistributes

1215: energy along the loop.  An emission measure would increase

1216: until the peak temperature of the loop, beyond which the amount

1217: of material would drop drastically. The addition of loops at higher

1218: temperatures can compensate for this drastic fall in the loop

1219: emission measure in order to reproduce the observed stellar EMD.  

1220: 

1221: The classic view of static loops with fixed

1222: cross-section gives an EMD that increases linearly 

1223: in the high temperature region, with a predicted 

1224: slope of 1.5. But more complex loops, with

1225: expanding cross-section 

1226: ({Schrijver}, {Lemen}, \& {Mewe} 1989; {Griffiths}, N.~W. 1999; 

1227: Hussain et~al. 2002)

1228: %\citep{schr89,gri99,hus01} 

1229: can account for

1230: larger slopes near the peak temperature of the loop, although

1231: not with high electron densities. 

1232: Low activity stars, like the Sun, Procyon, or $\alpha$~Cen,

1233: would be dominated by solar-like loops, peaking at log~T(K)$\sim$6.3,

1234: and with electron densities in the range

1235: log~$N_e(cm^{-3})\sim$9--10.5 (Drake, Laming, \& Widing 1995; Mewe

1236: et~al. 1995; Drake et~al. 1997).

1237: %\citep{dra95,mew95,drake97}. 

1238: Solar-like flares can

1239: produce a bump in the EMD at temperatures around log~T(K)$\sim$7.1

1240: \citep{orl00,rea01}. Stars such as Capella \citep{dup93} and FK~Aqr 

1241: are dominated by structures 

1242: with maximum temperature around log~T(K)$\sim$6.9 and

1243: log~$N_e(cm^{-3})\ga$12.  Finally, only the most active stars show the

1244: possible presence (not well constrained with EUVE data) of hotter loops

1245: that could explain the observed emission of the hottest lines. These

1246: hot loops may be directly related to the existence of large flares in

1247: stars like 

1248: UX~Ari, $\sigma$~Gem, V711~Tau and II~Peg \citep{paper1}. For the case

1249: of AR~Lac presented here, the value of the  emission measure increases

1250: during the 

1251: flares at all temperatures, with only a slight

1252: increase in the EMD slope at the hottest temperatures (see Table~\ref{slopes}).

1253: 

1254: It is significant that the high temperature enhancement

1255: (the ``bump'') appears ubiquitous in the coronae of cool

1256: stars.  Moreover the temperature of this enhancement is

1257: almost the same in a wide variety of cool stars.  

1258: It is not clear why this happens.  \citet{geh93} 

1259: noted a very small ($<$3\%) 

1260: inflection in a theoretical

1261: radiative cooling curve near 6$\times$10$^6$K and suggested

1262: it might account for stable coronal structures.  However these

1263: considerations apply when radiation dominates over conductive losses and when the 

1264: abundances are photospheric.  Both constraints may

1265: not apply  in stellar coronae.  Additionally, changes in the

1266: ionization equilibrium and atomic physics will impact such

1267: small details of the cooling curves.  It is fair to conclude

1268: that current theoretical models can not reproduce the observed

1269: emission measure distributions with high densities.

1270: 

1271: 

1272: 

1273: %------------------------------------

1274: \section{Conclusions}\label{sec:conclusion}

1275: 

1276: 

1277: \begin{enumerate}

1278: 

1279: \item Emission measure distributions (EMD) were derived from EUVE

1280: spectra of 22 active binary systems and 6 single stars. The

1281: overwhelming majority (25) of the

1282: stars in the sample  show an outstanding ``bump'' -- a local

1283: enhancement of the EMD -- over a restricted temperature range. This

1284: bump occurs near log~T$_e$~(K)$\sim$6.8--7.0. Its presence does not

1285: depend on the luminosity class of the star or the activity levels

1286: present,

1287: and confirms a fundamentally new coronal structure.

1288: 

1289: \item The emission measure per unit area (``column'' EMD) increases 

1290: towards shorter orbital periods, with a possible ``saturation'' effect

1291: at periods less than ~2.3 days.

1292: 

1293: \item \ion{Fe}{19}--{\small\sc XXII} line flux

1294:   ratios, formed at log~T$_e(K)\sim$7 and measured in the summed

1295:   spectra for each star indicate high electron 

1296:   densities (log~$N_e[cm^{-3}]\ga$12).  In conjunction with

1297:   lower densities found previously at lower temperatures, these

1298:   results provide additional evidence for different structures in 

1299:   stellar coronae.

1300: 

1301: \item A  second local enhancement of the EMD 

1302: peaking at log~T$_e(K)\sim$6.3 could

1303: reflect the presence of solar-like loops in the corona of some stars

1304: in the sample, and dominates the EMD in  $\alpha$~Cen AB,

1305: $\epsilon$~Eri, and Procyon.

1306: 

1307: \item \ion{Fe}{22}--{\small\sc XXIV} line fluxes indicate the presence

1308: of much hotter material at temperatures log~T$_e(K)\ga$7 in some stars

1309: (VY Ari, $\sigma^2$ CrB, V478 Lyr, and AR Lac).

1310: 

1311: \item The derived EMDs suggest these stellar coronae  are 

1312: composed of solar-like magnetic loops (peaking at

1313: log~T$_e[K]\sim$6.3). The loops at log~T$_e(K)\sim $6.9 are not yet

1314: understood from loop models. Loops peaking at

1315: log~T$_e(K)\ga$7.2 may be related to large flares.

1316: 

1317: \item Fluctuations in the EUV light curve of many stars in

1318: the sample are observed in a non-periodic 8--36~hr pattern, indicating

1319: the existence of frequent low-level variability.

1320: 

1321: \end{enumerate}

1322: 

1323: 

1324: \acknowledgments

1325: 

1326: This research is supported in part by NASA grants NAG5-7224,

1327: NAG5-11093, 

1328: and NAG5-3550 and CXC Contract NAS8-39073 to the 

1329: Smithsonian Astrophysical Observatory.   JSF is

1330: grateful for support in part by the Predoctoral Program of the 

1331: Smithsonian Astrophysical Observatory, and the Real Colegio 

1332: Complutense at Harvard University. 

1333: This research has made use of the SIMBAD database, operated 

1334: at CDS, Strasbourg, France, and of data obtained through 

1335: the High Energy Astrophysics Science Archive Research Center Online

1336: Service, provided by the NASA/Goddard Space Flight Center.  This

1337: research has also made use of NASA's Astrophysics Data System Abstract

1338: Service. The authors want to acknowledge D.~J. Christian

1339: (CEA/Berkeley) for his aid with the management of the EUVE archival

1340: data. 

1341: 

1342: \begin{thebibliography}{}

1343: %\bibliographystyle{apj}

1344: %\bibliography{paper3}

1345: \bibitem[{{Abbott} {et~al.}(1996){Abbott}, {Boyd}, {Jelinsky}, {Christian},

1346:   {Miller-Bagwell}, {Lampton}, {Malina}, \& {Vallerga}}]{abb96}

1347: {Abbott}, M.~J., {Boyd}, W.~T., {Jelinsky}, P., {Christian}, C.,

1348:   {Miller-Bagwell}, A., {Lampton}, M., {Malina}, R.~F., \& {Vallerga}, J.~V.

1349:   1996, \apjs, 107, 451

1350: 

1351: \bibitem[{{Airapetian} \& {Dempsey}(1998)}]{aira98}

1352: {Airapetian}, V.~S. \& {Dempsey}, R.~C. 1998, in ASP Conf. Ser. 154: Cool

1353:   Stars, Stellar Systems, and the Sun, Vol.~10, 1377

1354: 

1355: \bibitem[{{Amado} {et~al.}(2000){Amado}, {Doyle}, {Byrne}, {Cutispoto},

1356:   {Kilkenny}, {Mathioudakis}, \& {Neff}}]{ama00}

1357: {Amado}, P.~J., {Doyle}, J.~G., {Byrne}, P.~B., {Cutispoto}, G., {Kilkenny},

1358:   D., {Mathioudakis}, M., \& {Neff}, J.~E. 2000, \aap, 359, 159

1359: 

1360: \bibitem[{{Anders} \& {Grevesse}(1989)}]{anders}

1361: {Anders}, E. \& {Grevesse}, N. 1989, \gca, 53, 197

1362: 

1363: \bibitem[{{Ayres} {et~al.}(2001){Ayres}, {Osten}, \& {Brown}}]{ayr01}

1364: {Ayres}, T.~R., {Osten}, R.~A., \& {Brown}, A. 2001, \apjl, 562, L83

1365: 

1366: \bibitem[{{Baliunas} {et~al.}(1983){Baliunas}, {et~al.}}]{bali83}

1367: {Baliunas}, S.~L., {et~al.}  1983, \apj, 275, 752

1368: 

1369: \bibitem[{{Barden} \& {Nations}(1986)}]{bar86}

1370: {Barden}, S.~C. \& {Nations}, H.~L. 1986, in LNP Vol. 254: Cool Stars, Stellar

1371:   Systems and the Sun, Vol.~4, 262

1372: 

1373: \bibitem[Bowyer, Drake, \& Vennes(2000)]{bow00} Bowyer, S., 

1374: Drake, J.~J., \& Vennes, S.~;.\ 2000, \araa, 38, 231 

1375: 

1376: \bibitem[{{Brickhouse}(1996)}]{bri96}

1377: {Brickhouse}, N.~S. 1996, in IAU Colloq.~152: Astrophysics in the

1378: Extreme Ultraviolet, ed. by  S. Bowyer and R.~F. Malina, 105 

1379: 

1380: \bibitem[{{Brickhouse} \& {Dupree}(1998)}]{bri98}

1381: {Brickhouse}, N.~S. \& {Dupree}, A.~K. 1998, \apj, 502, 918

1382: 

1383: \bibitem[{{Brickhouse} {et~al.}(1999){Brickhouse}, {Dupree}, {Sanz-Forcada},

1384:   {Drake}, {White}, \& {Singh}}]{bri99}

1385: {Brickhouse}, N.~S., {Dupree}, A.~K., {Sanz-Forcada}, J., {Drake}, S.~A.,

1386:   {White}, N.~E., \& {Singh}, K.~P. 1999, AAS/High Energy Astrophysics

1387:   Division, 31, 0909

1388: 

1389: \bibitem[{{Brickhouse}{et~al.}(2000){Brickhouse},{Dupree}, {Edgar},

1390:   {Liedahl}, {Drake},{White},\& {Singh}}]{bri00}

1391:   {Brickhouse}, N.~S., {Dupree}, A.~K., {Edgar}, R.~J.,

1392:   {Liedahl}, D.~A., {Drake}, S.~A., {White}, N.~E., \& {Singh}, K.~P.

1393:   2000, \apj, 530, 387

1394: 

1395: \bibitem[{{Brickhouse} {et~al.}(1995){Brickhouse}, {Raymond}, \&

1396:   {Smith}}]{bri95}

1397: {Brickhouse}, N.~S., {Raymond}, J.~C., \& {Smith}, B.~W. 1995, \apjs, 97, 551

1398: 

1399: \bibitem[{{Canizares} {et~al.}(2000){Canizares}, {et~al.}}]{can00}

1400: {Canizares}, C. R., {et~al.} 2000, \apj, 539, L41

1401: 

1402: \bibitem[{{Cayrel de Strobel} {et~al.}(1994){Cayrel de Strobel}, {Cayrel},

1403:   {Friel}, {Zahn}, \& {Bentolila}}]{cayr94}

1404: {Cayrel de Strobel}, G., {Cayrel}, R., {Friel}, E., {Zahn}, J.~., \&

1405:   {Bentolila}, C. 1994, \aap, 291, 505

1406: 

1407: \bibitem[Craig et al.(1997)]{crai97} Craig, N.~et al.\ 1997, 

1408: \apjs, 113, 131 

1409: 

1410: \bibitem[{{Cutispoto}(1998)}]{cuti98}

1411: {Cutispoto}, G. 1998, \aaps, 131, 321

1412: 

1413: \bibitem[{{Cutispoto} {et~al.}(2001){Cutispoto}, {Messina}, \&

1414:   {Rodon{\`o}}}]{cuti01}

1415: {Cutispoto}, G., {Messina}, S., \& {Rodon{\`o}}, M. 2001, \aap, 367, 910

1416: 

1417: \bibitem[{{Diamond} {et~al.}(1995){Diamond}, {Jewell}, \& {Ponman}}]{diam95}

1418: {Diamond}, C.~J., {Jewell}, S.~J., \& {Ponman}, T.~J. 1995, \mnras, 274, 589

1419: 

1420: \bibitem[{{Donati}(1999)}]{don99}

1421: {Donati}, J.~. 1999, \mnras, 302, 457

1422: 

1423: \bibitem[{{Drake} {et~al.}(1995){Drake}, {Laming}, \& {Widing}}]{dra95}

1424: {Drake}, J.~J., {Laming}, J.~M., \& {Widing}, K.~G. 1995, \apj, 443, 393

1425: 

1426: \bibitem[{{Drake} {et~al.}(1997){Drake}, {Laming}, \& {Widing}}]{drake97}

1427: ---. 1997, \apj, 478, 403

1428: 

1429: \bibitem[{{Dupree} {et~al.}(1993){Dupree}, {Brickhouse}, {Doschek}, {Green}, \&

1430:   {Raymond}}]{dup93}

1431: {Dupree}, A.~K., {Brickhouse}, N.~S., {Doschek}, G.~A., {Green}, J.~C., \&

1432:   {Raymond}, J.~C. 1993, \apjl, 418, L41

1433: 

1434: \bibitem[{{Dupree} {et~al.}(2002){Dupree}, {Brickhouse}, \&

1435:   {Sanz-Forcada}}]{dup02}

1436: {Dupree}, A.~K., {Brickhouse}, N.~S., \& {Sanz-Forcada}, J. 2002, \apj,

1437:   submitted

1438: 

1439: \bibitem[{{Fekel}(1996)}]{fek96}

1440: {Fekel}, F.~C. 1996, \aj, 112, 269

1441: 

1442: \bibitem[{{Fekel}(1997)}]{fek97}

1443: ---. 1997, \pasp, 109, 514

1444: 

1445: \bibitem[{{Fleming} {et~al.}(1989){Fleming}, {Gioia}, \& {Maccacaro}}]{fle89}

1446: {Fleming}, T.~A., {Gioia}, I.~M., \& {Maccacaro}, T. 1989, \apj, 340, 1011

1447: 

1448: \bibitem[{{Flower}(1996)}]{flo96}

1449: {Flower}, P.~J. 1996, \apj, 469, 355

1450: 

1451: \bibitem[{{Fournier} {et~al.}(2001){Fournier}, {May}, {Liedahl},

1452: {Pacella}, {Finkenthal}, {Leigheb}, {Mattiolli}, \& {Goldstein}}]{four01}

1453: {Fournier}, K. B., {May}, M. B., {Liedahl}, D. A.,

1454: {Pacella}, D., {Finkenthal}, M., {Leigheb}, M., {Mattiolli}, M., \&

1455: {Goldstein}, W. H. 2001, \apj, 561, 1144

1456: 

1457: \bibitem[{{Fruscione} {et~al.}(1994){Fruscione}, {Hawkins}, {Jelinsky}, \&

1458:   {Wiercigroch}}]{fru94}

1459: {Fruscione}, A., {Hawkins}, I., {Jelinsky}, P., \& {Wiercigroch}, A. 1994,

1460:   \apjs, 94, 127

1461: 

1462: \bibitem[{{Gehrels and Williams} (1993){Gehrels},\&

1463: {Williams}}]{geh93}

1464: {Gehrels}, N., \& {Williams}, E.~D. 1993, \apj, 418, L25

1465: 

1466: \bibitem[{{Golub} {et~al.}(1982){Golub}, {Harnden}, {Pallavicini}, {Rosner}, \&

1467:   {Vaiana}}]{gol82}

1468: {Golub}, L., {Harnden}, F.~R., {Pallavicini}, R., {Rosner}, R., \& {Vaiana},

1469:   G.~S. 1982, \apj, 253, 242

1470: 

1471: \bibitem[{{Graffagnino} {et~al.}(1995){Graffagnino}, {Wonnacott}, \&

1472:   {Schaeidt}}]{graf95}

1473: {Graffagnino}, V.~G., {Wonnacott}, D., \& {Schaeidt}, S. 1995, \mnras, 275, 129

1474: 

1475: \bibitem[{{Gray}(1989)}]{gray89}

1476: {Gray}, D.~F. 1989, \pasp, 101, 1126

1477: 

1478: \bibitem[{{Gray}(1992)}]{gray92}

1479: ---. 1992, The observation and analysis of stellar photospheres (Cambridge

1480:   Astrophysics Series, Cambridge: Cambridge University Press, 1992, 2nd ed.)

1481: 

1482: \bibitem[{{Griffin}(1998)}]{griffin98}

1483: {Griffin}, R.~F. 1998, The Observatory, 118, 273

1484: 

1485: \bibitem[{{Griffiths}(1999)}]{gri99}

1486: {Griffiths}, N.~W. 1999, \apj, 518, 873

1487: 

1488: \bibitem[{{Griffiths} \& {Jordan}(1998)}]{gri98}

1489: {Griffiths}, N.~W. \& {Jordan}, C. 1998, \apj, 497, 883

1490: 

1491: \bibitem[{{Haisch} {et~al.}(1993){Haisch}, {Bowyer}, \& {Malina}}]{hai93}

1492: {Haisch}, B., {Bowyer}, S., \& {Malina}, R.~F. 1993, Journal of the British

1493:   Interplanetary Society, 46, 331

1494: 

1495: \bibitem[{{Hatzes} {et~al.}(2000){Hatzes}, {et~al.}}]{hat00}

1496: {Hatzes}, A.~P., {et~al.} 2000, \apjl, 544, L145

1497: 

1498: \bibitem[{{Huenemoerder} {et~al.}(2002){Huenemoerder}, {Canizares}, \&

1499:   {Sanz-Forcada}}]{huen01}

1500: {Huenemoerder}, D.~P., {Canizares}, C.~R., {Drake}, J.~J., \&

1501:   {Sanz-Forcada}, J. 2002, \apj, in preparation

1502: 

1503: \bibitem[{{Hussain} {et~al.}(2001){Hussain}, {van Ballegooijen}, {Jardine}, \&

1504:   {Collier Cameron}}]{hus01}

1505: {Hussain}, G. A.~J., {van Ballegooijen}, A.~A., {Jardine}, M., \& {Collier

1506:   Cameron}, A. 2002, \mnras, in preparation

1507: 

1508: \bibitem[{{Irwin} {et~al.}(1992){Irwin}, {Fletcher}, {Yang}, {Walker}, \&

1509:   {Goodenough}}]{irw92}

1510: {Irwin}, A.~W., {Fletcher}, J.~M., {Yang}, S. L.~S., {Walker}, G. A.~H., \&

1511:   {Goodenough}, C. 1992, \pasp, 104, 489

1512: 

1513: \bibitem[{{Kaastra} {et~al.}(1996){Kaastra}, {Mewe}, {Liedahl}, {Singh},

1514:   {White}, \& {Drake}}]{kaa96}

1515: {Kaastra}, J.~S., {Mewe}, R., {Liedahl}, D.~A., {Singh}, K.~P., {White}, N.~E.,

1516:   \& {Drake}, S.~A. 1996, \aap, 314, 547

1517: 

1518: \bibitem[{{Kimble} {et~al.}(1993){Kimble}, {Davidsen}, {Long}, \&

1519:   {Feldman}}]{kimb93}

1520: {Kimble}, R.~A., {Davidsen}, A.~F., {Long}, K.~S., \& {Feldman}, P.~D. 1993,

1521:   \apjl, 408, L41

1522: 

1523: \bibitem[{{Klimchuk} \& {Cargill}(2001)}]{kli01}

1524: {Klimchuk}, J.~A., \& {Cargill}, P.~J. 2001, \apj, 553, 440

1525: 

1526: \bibitem[{{Laming} {et~al.}(1996){Laming}, {Drake}, \& {Widing}}]{lami96}

1527: {Laming}, J.~M., {Drake}, J.~J., \& {Widing}, K.~G. 1996, \apj, 462, 948

1528: 

1529: \bibitem[{{Linsky} \& {Wood}(1996)}]{lin96}

1530: {Linsky}, J.~L. \& {Wood}, B.~E. 1996, \apj, 463, 254

1531: 

1532: \bibitem[{{Marino} {et~al.}(1998){Marino}, {Catalano}, {Frasca}, \&

1533:   {Marilli}}]{mar98}

1534: {Marino}, G., {Catalano}, S., {Frasca}, A., \& {Marilli}, E. 1998,

1535:   Informational Bulletin on Variable Stars, 4599, 1

1536: 

1537: \bibitem[{{de Medeiros} \& {Mayor}(1999)}]{med99}

1538: {de Medeiros}, J.~R. \& {Mayor}, M. 1999, \aaps, 139, 433

1539: 

1540: \bibitem[{{Melo} {et~al.}(2001){Melo}, {Pasquini}, \& {De Medeiros}}]{melo01}

1541: {Melo}, C.~H.~F., {Pasquini}, L., \& {de Medeiros}, J.~R. 2001, \aap, 375, 851

1542: 

1543: \bibitem[{{Mewe} {et~al.}(1995){Mewe}, {Kaastra}, {Schrijver}, {van den Oord},

1544:   \& {Alkemade}}]{mew95}

1545: {Mewe}, R., {Kaastra}, J.~S., {Schrijver}, C.~J., {van den Oord}, G. H.~J., \&

1546:   {Alkemade}, F. J.~M. 1995, \aap, 296, 477

1547: 

1548: \bibitem[{{Miller-Bagwell} \& {Abbott}(1995)}]{mill95}

1549: {Miller-Bagwell}, A. \& {Abbott}, M. 1995, EUVE Guest Observer Data Products

1550:   Guide

1551: 

1552: \bibitem[{{Mitrou} {et al.}(1997){Mitrou}, {Mathioudakis}, {Doyle}, \& {Antonopoulou}}]{mit97}

1553:   Mitrou, C.~K., Mathioudakis, M., Doyle, J.~G., \& Antonopoulou, E.\

1554:   1997, \aap, 317, 776  

1555: 

1556: \bibitem[{{Montes} {et~al.}(1995){Montes}, {Fernandez-Figueroa}, {de Castro},

1557:   \& {Cornide}}]{mon95}

1558: {Montes}, D., {Fernandez-Figueroa}, M.~J., {de Castro}, E., \& {Cornide}, M.

1559:   1995, \aap, 294, 165

1560: 

1561: \bibitem[{{Montes} {et~al.}(1997){Montes}, {Fernandez-Figueroa}, {de Castro},

1562:   \& {Sanz-Forcada}}]{mon97}

1563: {Montes}, D., {Fernandez-Figueroa}, M.~J., {de Castro}, E., \& {Sanz-Forcada},

1564:   J. 1997, \aaps, 125, 263

1565: 

1566: \bibitem[{{Montes} {et~al.}(1999){Montes}, {Saar}, {Collier Cameron}, \&

1567:   {Unruh}}]{mon99}

1568: {Montes}, D., {Saar}, S.~H., {Collier Cameron}, A., \& {Unruh}, Y.~C. 1999,

1569:   \mnras, 305, 45

1570: 

1571: \bibitem[{{Morel} {et~al.}(2000){Morel}, {Provost}, {Lebreton}, {Th{\'e}venin},

1572:   \& {Berthomieu}}]{mor00}

1573: {Morel}, P., {Provost}, J., {Lebreton}, Y., {Th{\'e}venin}, F., \&

1574:   {Berthomieu}, G. 2000, \aap, 363, 675

1575: 

1576: \bibitem[{{Orlando} {et~al.}(2000){Orlando}, {Peres}, \& {Reale}}]{orl00}

1577: {Orlando}, S., {Peres}, G., \& {Reale}, F. 2000, \apj, 528, 524

1578: 

1579: \bibitem[{{Osten} \& {Brown}(1999)}]{ost99}

1580: {Osten}, R.~A. \& {Brown}, A. 1999, \apj, 515, 746

1581: 

1582: \bibitem[{{Osten} {et~al.}(2000){Osten}, {Brown}, {Ayres}, {Linsky}, {Drake},

1583:   {Gagn{\'e}}, \& {Stern}}]{ost00}

1584: {Osten}, R.~A., {Brown}, A., {Ayres}, T.~R., {Linsky}, J.~L., {Drake}, S.~A.,

1585:   {Gagn{\'e}}, M., \& {Stern}, R.~A. 2000, \apj, 544, 953

1586: 

1587: \bibitem[{{Pallavicini} {et~al.}(1981){Pallavicini}, {Golub}, {Rosner},

1588:   {Vaiana}, {Ayres}, \& {Linsky}}]{pal81}

1589: {Pallavicini}, R., {Golub}, L., {Rosner}, R., {Vaiana}, G.~S., {Ayres}, T., \&

1590:   {Linsky}, J.~L. 1981, \apj, 248, 279

1591: 

1592: \bibitem[{{Pan} \& {Jordan}(1995)}]{pan95}

1593: {Pan}, H.~C. \& {Jordan}, C. 1995, \mnras, 272, 11

1594: 

1595: \bibitem[{{Pasquini} {et~al.}(1991){Pasquini}, {Cutispoto}, {Gratton}, \&

1596:   {Mayor}}]{pas91}

1597: {Pasquini}, L., {Cutispoto}, G., {Gratton}, R., \& {Mayor}, M. 1991, \aap, 248,

1598:   72

1599: 

1600: \bibitem[{{Pease} {et~al.}(2001){Pease}, {Drake}, {Kashyap}, {Ratzlaff},

1601:   {Saar}, {Dobrzycki}, {Adams}, \& {Wolk}}]{pea01}

1602: {Pease}, D., {Drake}, J.~J., {Kashyap}, V., {Ratzlaff}, P.~W., {Saar}, S.~H.,

1603:   {Dobrzycki}, A., {Adams}, N.~R., \& {Wolk}, S.~J. 2001, in Stellar Coronae in

1604:   the Chandra and XMM-Newton era, ed. by J. Drake, \& F. Favata (Noordwijk:

1605:   ASP)

1606: 

1607: \bibitem[{{Perryman} {et~al.}(1997){Perryman} {et~al.}}]{hippa}

1608: {Perryman}, M. A.~C., {et~al.} 1997, \aap, 323, L49

1609: 

1610: \bibitem[{{Phillips} {et~al.}(1996){Phillips}, {Bhatia}, {Mason}, \&

1611:   {Zarro}}]{phil96}

1612: {Phillips}, K. J.~H., {Bhatia}, A.~K., {Mason}, H.~E., \& {Zarro}, D.~M. 1996,

1613:   \apj, 466, 549

1614: 

1615: \bibitem[Phillips et al.(2001)]{phi01} Phillips, K.~J.~H., 

1616: Mathioudakis, M., Huenemoerder, D.~P., Williams, D.~R., Phillips, M.~E., \& 

1617: Keenan, F.~P.\ 2001, \mnras, 325, 1500 

1618: 

1619: \bibitem[{{Raymond}(1988)}]{raym88}

1620: {Raymond}, J.~C. 1988, in NATO ASIC Proc. 249: Hot Thin Plasmas in

1621:   Astrophysics, ed. R.~Pallavicini, 3

1622: 

1623: \bibitem[Reale, Peres, \& Orlando(2001)]{rea01} {Reale}, F., 

1624: {Peres}, G., \& {Orlando}, S.\ 2001, \apj, 557, 906 

1625: 

1626: \bibitem[{{Rucinski}(1998)}]{ruc98}

1627: {Rucinski}, S.~M. 1998, \aj, 115, 303

1628: 

1629: \bibitem[{{Saar} \& {Osten}(1997)}]{saa97}

1630: {Saar}, S.~H. \& {Osten}, R.~A. 1997, \mnras, 284, 803

1631: 

1632: \bibitem[{{Sanz-Forcada}(2001)}]{tesis}

1633: {Sanz-Forcada}, J. 2001, PhD thesis, University Complutense of Madrid

1634: 

1635: \bibitem[{{Sanz-Forcada} {et~al.}(2001){Sanz-Forcada}, {Brickhouse}, \&

1636:   {Dupree}}]{sanz01}

1637: {Sanz-Forcada}, J., {Brickhouse}, N.~S., \& {Dupree}, A.~K. 2001, \apj, 554,

1638:   1079

1639: 

1640: \bibitem[{{Sanz-Forcada} {et~al.}(2002){Sanz-Forcada}, {Brickhouse}, \&

1641:   {Dupree}}]{paper1}

1642: ---. 2002, \apj, 570, 799

1643: 

1644: \bibitem[{{Schmitt}(1998)}]{schm98}

1645: {Schmitt}, J. H. M.~M. 1998, in ASP Conf. Ser. 154: Cool Stars, Stellar

1646:   Systems, and the Sun, Vol.~10, 463

1647: 

1648: \bibitem[{{Schmitt} {et~al.}(1996){Schmitt}, {Drake}, {Stern}, \&

1649:   {Haisch}}]{sch96}

1650: {Schmitt}, J. H. M.~M., {Drake}, J.~J., {Stern}, R.~A., \& {Haisch}, B.~M.

1651:   1996, \apj, 457, 882

1652: 

1653: \bibitem[{{Schrijver} {et~al.}(1989){Schrijver}, {Lemen}, \& {Mewe}}]{schr89}

1654: {Schrijver}, C.~J., {Lemen}, J.~R., \& {Mewe}, R. 1989, \apj, 341, 484

1655: 

1656: \bibitem[{{Schrijver} {et~al.}(1995){Schrijver}, {Mewe}, {van den Oord}, \&

1657:   {Kaastra}}]{sch95}

1658: {Schrijver}, C.~J., {Mewe}, R., {van den Oord}, G. H.~J., \& {Kaastra}, J.~S.

1659:   1995, \aap, 302, 438

1660: 

1661: \bibitem[{{Smith} {et~al.}(2001){Smith}, {Brickhouse}, {Liedahl}, \&

1662:   {Raymond}}]{smith01}

1663: {Smith}, R.~K., {Brickhouse}, N.~S., {Liedahl}, D.~A., \& {Raymond}, J.~C.

1664:   2001, \apj, 556, 91

1665: 

1666: \bibitem[{{Stern} \& {Drake}(1996)}]{ste96}

1667: {Stern}, R.~A. \& {Drake}, J.~J. 1996, in IAU Colloq. 152: Astrophysics in the

1668:   Extreme Ultraviolet, 135

1669: 

1670: \bibitem[{{Strassmeier} {et~al.}(1993){Strassmeier}, {Hall}, {Fekel}, \&

1671:   {Scheck}}]{cabs}

1672: {Strassmeier}, K.~G., {Hall}, D.~S., {Fekel}, F.~C., \& {Scheck}, M. 1993,

1673:   \aaps, 100, 173

1674: 

1675: \bibitem[{{Strassmeier} {et~al.}(1999){Strassmeier}, {Serkowitsch}, \&

1676:   {Granzer}}]{str99}

1677: {Strassmeier}, K.~G., {Serkowitsch}, E., \& {Granzer}, T. 1999, \aaps, 140, 29

1678: 

1679: \bibitem[{{Torres} \& {Ribas}(2002)}]{torres01}

1680: {Torres}, G. \& {Ribas}, I. 2002, \apj, 567, 1140

1681: 

1682: \bibitem[{{Walter} \& {Bowyer}(1981)}]{wal81}

1683: {Walter}, F.~M. \& {Bowyer}, S. 1981, \apj, 245, 671

1684: 

1685: \bibitem[{{Wood} {et~al.}(2000){Wood}, {Ambruster}, {Brown}, \&

1686:   {Linsky}}]{woo00}

1687: {Wood}, B.~E., {Ambruster}, C.~W., {Brown}, A., \& {Linsky}, J.~L. 2000, \apj,

1688:   542, 411

1689: 

1690: \end{thebibliography}

1691: 

1692: \clearpage

1693: %----------------------------------------------

1694: %   Figures

1695: %

1696: 

1697: %-------------------

1698: %Fig. 1a

1699: \begin{figure}\

1700: \epsscale{.80}

1701: \plotone{f1a.eps}

1702:   \caption{DS light curves as a function of Julian Date (lower axis)

1703:   and orbital phase (upper axis). We use the convention that at

1704:   orbital phase 1.0 the primary  star is located behind the secondary

1705:   star (see Table~\ref{tabparam}); photometric phase is used in the

1706:   upper axis for $\epsilon$~Eri and LQ~Hya, and no periods are

1707:   available for

1708:   $\alpha$~Cen, Procyon, and $\beta$~Cet. Open circles mark data affected

1709:   by the dead spot (and corrected by the DS to SW flux ratio,

1710:   see \S~\ref{sec:observations}), while solid circles represent unaffected

1711:   data. An average one-sigma error bar is shown as

1712:   reference on the left side of each plot. Only points with S/N

1713:   higher than 5 are plotted. The bin size is 600~s. 

1714:   \label{plc}}

1715: \end{figure}

1716: 

1717: %-------------------

1718: %Fig. 1b

1719: \setcounter{figure}{0}

1720: \begin{figure}

1721: \plotone{f1b.eps}

1722:   \caption{(b) continued.} 

1723: \end{figure}

1724: 

1725: %-------------------

1726: %Fig. 1c

1727: \setcounter{figure}{0}

1728: \begin{figure}

1729: \plotone{f1c.eps}

1730:   \caption{(c) continued.}

1731: \end{figure}

1732: 

1733: %-------------------

1734: %Fig. 1d

1735: \setcounter{figure}{0}

1736: \begin{figure}

1737: \plotone{f1d.eps}

1738:   \caption{(d) continued.}

1739: \end{figure}

1740: 

1741: %-------------------

1742: %Fig. 2a

1743: \begin{figure}

1744:   \epsscale{0.9}

1745: \plotone{f2a.eps}

1746:   \caption{(a) EUVE SW and MW spectra of the stars in the sample. Ion

1747:   stages of iron are marked in the top panel. Spectra are smoothed by

1748:   5 pixels. Dotted lines indicate the zero flux level of each

1749:   spectrum.  \label{specs}}

1750: \end{figure}

1751: 

1752: %-------------------

1753: %Fig. 2b

1754: \setcounter{figure}{1}

1755: \begin{figure}

1756:   \epsscale{0.9}

1757: \plotone{f2b.eps}

1758:   \caption{(b) continued.}

1759: \end{figure}

1760: 

1761: %-------------------

1762: %Fig. 2c

1763: \setcounter{figure}{1}

1764: \begin{figure}

1765:   \epsscale{0.9}

1766: \plotone{f2c.eps}

1767:   \caption{(c) continued.}

1768: \end{figure}

1769: 

1770: %-------------------

1771: %Fig. 2d

1772: \setcounter{figure}{1}

1773: \begin{figure}

1774:   \epsscale{0.9}

1775: \plotone{f2d.eps}

1776:   \caption{(d) continued.}

1777: \end{figure}

1778: 

1779: %-------------------

1780: %Fig. 3

1781: \begin{figure}

1782:   \epsscale{0.9}

1783: \plotone{f3.eps}

1784:   \caption{AR Lac, $\beta$~Cet and LW spectra of 5 of the stars. Note

1785:     that 2$^{nd}$ order of the \ion{He}{1} line is detected in the LW

1786:     spectra at 608~\AA.\label{specesp}}

1787: \end{figure}

1788: 

1789: \clearpage

1790: 

1791: %-------------------

1792: %Fig. 4

1793: \begin{figure}

1794:   \epsscale{1.}

1795: \plotone{f4.eps}

1796:   \caption{\ion{Fe}{16} \gl335 and \gl361 line flux ratios (in photon

1797:   units) expected for different values of interstellar hydrogen column

1798:   density (N$_H$), and observed ratios for several stars in the

1799:   sample, with 1-$\sigma$ observational error bars. 

1800:   \label{ismcalc}}

1801: \end{figure}

1802: 

1803: 

1804: %-------------------

1805: %Fig. 5a

1806: \begin{figure}

1807: \epsscale{0.80}

1808: %\plotone{f5acol.eps}

1809: \plotone{f5a.eps}

1810:   \caption{(a){\it Upper panels}: EMD for the summed EUVE 

1811: spectrum combined with the IUE spectrum for each star. Thin lines 

1812: represent the relative contribution function for each ion (the 

1813: emissivity function multiplied by the EMD at each point).

1814: {\it Lower panels}: Observed-to-predicted line ratios for the ion 

1815: stages in top figure with S/N greater than 3. The dotted line 

1816: denotes a factor of 2. Symbols used are open circles for N, diamonds

1817: for C, and open triangles for Si. Fe lines with S/N

1818: higher than 4 are denoted with filled circles, solid triangles for

1819: those with S/N between 3 and 4, and the plus sign (+) for S/N less

1820: than 3. Electron densities used in the calculation of the EMD are shown; if a value

1821: is not given, then the density was taken as 10$^{12}$ cm$^{-3}$.

1822:   \label{emdfigs}}

1823: \end{figure}

1824: \clearpage

1825: 

1826: %-------------------

1827: %Fig. 5b

1828: \setcounter{figure}{4}

1829: \begin{figure}

1830: %\plotone{f5bcol.eps}

1831: \plotone{f5b.eps}

1832:   \caption{(b) continued.} 

1833: \end{figure}

1834: 

1835: %-------------------

1836: %Fig. 5c

1837: \setcounter{figure}{4}

1838: \begin{figure}

1839: %\plotone{f5ccol.eps}

1840: \plotone{f5c.eps}

1841:   \caption{(c) continued.} 

1842: \end{figure}

1843: 

1844: %-------------------

1845: %Fig. 5d

1846: \setcounter{figure}{4}

1847: \begin{figure}

1848: %\plotone{f5dcol.eps}

1849: \plotone{f5d.eps}

1850:   \caption{(d) continued.} 

1851: \end{figure}

1852: 

1853: %-------------------

1854: %Fig. 5e

1855: \setcounter{figure}{4}

1856: \begin{figure}

1857: %\plotone{f5ecol.eps}

1858: \plotone{f5e.eps}

1859:   \caption{(e) continued.} 

1860: \end{figure}

1861: 

1862: \clearpage

1863: 

1864: %-------------------

1865: %Fig. 5f

1866: \setcounter{figure}{4}

1867: \begin{figure}

1868:   \epsscale{0.5}

1869: %\plotone{f5fcol.eps}

1870: \plotone{f5f.eps}

1871:   \caption{(f) continued.} 

1872:   \epsscale{1.}

1873: \end{figure}

1874: 

1875: %-------------------

1876: %Fig. 6

1877: \begin{figure}

1878: \epsscale{1.0}

1879: \plotone{f6.eps}

1880:   \caption{Comparison between the EMD calculated from the observations

1881:     on AR~Lac in 1993+1997 and those in 2000.} 

1882:   \label{stagesarlac}

1883: \end{figure}

1884: 

1885: %-------------------

1886: %Fig. 7

1887: \begin{figure}

1888: %\plotone{f7col.eps}

1889: \plotone{f7.eps}

1890:   \caption{EMD of selected stars normalized to the solar photospheric

1891:     radius [EMD/$4 \pi (R_1^2+R_2^2)$= ``column'' EMD, 

1892:     assuming that both stars in binary systems 

1893:     contribute to the observed emission, with radii measured in

1894:     R$_\odot$]}

1895:     \label{sevstars}

1896: \end{figure}

1897: 

1898: %-------------------

1899: %Fig. 8

1900: \begin{figure}

1901: \plotone{f8.eps}

1902:   \caption{Orbital (or photometric) period relations with several features for 30

1903:     single and binary

1904:     stars: (a) Temperature at the peak [3 largest

1905:     values of the emission measure (EM)], (b) density at

1906:     log~T(K)$\sim$6.9, (c) EM at 

1907:     the peak, (d) EM at the peak per unit area [EM/$4 * \pi

1908:   *(R_1^2+R_2^2)$], with radii in solar units. A solid line with the

1909:     best fit to all the data has been plotted in (d), along with a dashed

1910:     line representing the fit including only objects with period over

1911:     1 day.  Filled circles represent binary systems, diamonds are

1912:     single stars, and an open circle represents the Sun during the

1913:     solar maximum \citep[from][]{orl00}.} 

1914:     \label{periods}

1915: \end{figure}

1916: 

1917: %-------------------

1918: %Fig. 9

1919: \begin{figure}

1920:   \epsscale{0.5}

1921: \plotone{f9.eps}

1922:   \caption{Orbital period versus EUV (80--170~\AA) luminosity

1923:     weighted by the bolometric luminosity (see text). A solid line with the

1924:     best fit to all the data has been plotted in (d), along with a dashed

1925:     line representing the fit including only objects with period

1926:   longer than 

1927:     2 days.  Filled circles represent binary systems and diamonds are

1928:     single stars.} 

1929:     \label{lumperiod}

1930:   \epsscale{1.}

1931: \end{figure}

1932: 

1933: \clearpage

1934: %------------ TABLES ---------------------------

1935: 

1936: %------------------

1937: % Table 1

1938: \input{tab1}

1939: 

1940: %------------------------------------------------- begin table

1941: % Table 2

1942: %

1943: % Observations times

1944: \begin{deluxetable}{llrrr}

1945: \tablewidth{0pt}

1946: \tablecolumns{5}

1947: \tablecaption{Exposure Times for EUVE Spectrographs \label{euvetimes}}

1948: \tablehead{

1949:     \colhead{}& \colhead{}& \multicolumn{3}{c}{Exposure time (ks)} \\ 

1950:     \cline{3-5}

1951:     \colhead{Name} &\colhead{Start date}  & \colhead{SW} &

1952: \colhead{MW} & \colhead{LW}     }

1953: \startdata

1954:     $\beta$ Cet     & {2000 Aug  5} & 808   & \nodata & \nodata \\

1955:     AY Cet          & {1993 Sep 28} & 116   & \nodata & \nodata \\

1956:     {AR Psc}        & {1997 Aug 26} & 388   & \nodata & \nodata \\

1957:     {CC Eri}        & {1995 Sep 13} & {257} & 144 & 132 \\

1958:     {VY Ari}        & {1994 Oct 6}  & {244} & 159 & 156 \\

1959:     {$\epsilon$ Eri}& {1993 Oct 22} & { 87} &  62 &  60 \\

1960:     {$\epsilon$ Eri}& {1995 Aug 31} & {495} & 240 & 214 \\

1961:     {Capella}       & {2001 Jan 14} & 49    &  48 &  48 \\

1962:     {YY Gem}        & {1995 Feb  2} & 425   & 295 & 292 \\

1963:     {Procyon}       & {1993 Jan 11} &  91   &  96 & 92  \\

1964:     {Procyon}       & {1994 Mar 12} &  227  & 137 & 136 \\

1965:     {Procyon}       & {1999 Nov  6} & 108   &  92 &  90 \\

1966:     {BF Lyn}        & {1994 Apr 14} & {110} & 69  & \nodata \\

1967:     {LQ Hya}        & {1993 Dec 10} & {359} & \nodata & \nodata \\

1968:     {DH Leo}        & {1995 Feb 12} & {311} & 171 & 176 \\

1969:     {$\xi$ UMa}     & {1993 Mar 28} & { 55} &  56 & 52  \\

1970:     {$\xi$ UMa}     & {1997 May 14} & {316} & 310 & 307 \\

1971:     {BH CVn}        & {1996 Feb 12} & {430} & 191 & 187 \\

1972:     {$\alpha$ Cen}  & {1993 May 29} & {124} & 107 & 104 \\

1973:     {$\alpha$ Cen}  & {1995 Mar 03} & {173} & 67  & 63  \\

1974:     {$\alpha$ Cen}  & {1997 Mar 10} & {121} & 117 & 121 \\

1975:     {$\sigma^2$ CrB}& {1994 Feb 16} & {213} &  97 &  89 \\

1976:     {V824 Ara}      & {1996 Apr 30} &  40   & \nodata & \nodata \\

1977:     {V478 Lyr}      & {1998 May 18} & {240} & \nodata & \nodata \\

1978:     {ER Vul}        & {1995 Sep 20} & {286} & 181 & 183 \\

1979:     {AR Lac}        & {1993 Oct 12} & { 96} &  94 &  94 \\

1980:     {AR Lac}        & {1997 Jul  3} & { 74} &  73 &  73 \\

1981:     {AR Lac}        & {2000 Sep 14} &  {63} &  61 & {65}\\

1982:     {FK Aqr}        & {1994 Sep 11} & {134} & 130 & 128 \\

1983:     {FK Aqr}        & {1997 Oct  9} & {334} & 321 & 330 \\

1984:     {BY Dra}        & {1997 Sep 22} & {194} & \nodata & \nodata \\

1985: \enddata

1986: \end{deluxetable}

1987: %--------------------------------------------- end table

1988: 

1989: %------------------

1990: % Table 3_4

1991: \input{tab3_4}

1992: %------------------

1993: % Table 5

1994: \input{tab5}

1995: %------------------

1996: % Table 6

1997: \input{tab6}

1998: %------------------

1999: % Table 7

2000: \input{tab7}

2001: %------------------

2002: % Table 8

2003: \input{tab8}

2004: %------------------

2005: % Table 9

2006: \input{tab9}

2007: %------------------

2008: % Table 10

2009: \input{tab10}

2010: 

2011: 

2012: \end{document}

2013: 

2014: 

2015: 

2016: 

2017: 

2018: 

2019: