1: %------ for platex ----------
2: %\documentclass{article}
3: %\usepackage{emulateapj}
4: %\usepackage{graphicx}
5: %---- for jlatex ---
6: \documentstyle[11pt,aasms4]{article}
7: \setlength{\topmargin}{-0.7cm}
8: \setlength{\textheight}{24cm}
9: \def\baselinestretch{1.5}
10: %\received{}z
11: %\accepted{}
12: %\journalid{}{}
13: %\articleid{}{}
14: %-------------------------
15: \slugcomment{}
16: \lefthead{KUBOTA ET AL.}
17: \righthead{SPECTRAL TRANSITIONS OF TWO X-RAY SOURCES IN IC~342}
18: \slugcomment{}
19: \nonstopmode
20:
21: \begin{document}
22:
23:
24: \title{DISCOVERY OF SPECTRAL TRANSITIONS FROM TWO \\
25: ULTRA-LUMINOUS COMPACT X-RAY SOURCES IN IC~342}
26: \author{A. {\sc Kubota},$^{1}$ T. {\sc Mizuno},$^{2}$ K. {\sc Makishima},$^{1}$
27: Y. {\sc Fukazawa},$^{2}$\\
28: J. {\sc Kotoku},$^{1}$ T. {\sc Ohnishi},$^{1}$ and M. {\sc Tashiro}$^{1}$}
29: \affil{1:Department of Physics, University of Tokyo,
30: 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan 113-0033}
31: \centerline{\it e-mail : aya@amalthea.phys.s.u-tokyo.ac.jp}
32: \affil{2:Department of Physical Science, Hiroshima University,
33: 1-3-1 Kagamiyama,\\ Higashi-Hiroshima, Japan 739-8526}
34:
35: \begin{abstract}
36: Two {\it ASCA} observations were made of two
37: ultra-luminous compact X-ray sources (ULXs),
38: Source~1 and Source~2, in the spiral galaxy IC~342.
39: In the 1993 observation,
40: Source~2 showed a 0.5--10 keV luminosity of
41: $6 \times 10^{39}$ ergs s$^{-1}$ (assuming a distance of 4.0 Mpc),
42: and a hard power-law spectrum of photon index $\sim 1.4$.
43: As already reported,
44: Source~1 was $\sim 3$ times brighter on that occasion,
45: and exhibited a soft spectrum represented by
46: a multi-color disk model of inner-disk temperature $ \sim 1.8$ keV.
47: The second observation made in February 2000 revealed
48: that Source~1 had made a transition into a hard spectral state,
49: while Source~2 into a soft spectral state.
50: The ULXs are therefore inferred to exhibit two distinct spectral states,
51: and sometimes make transitions between them.
52: These results significantly reinforce the scenario
53: which describes ULXs as mass-accreting black holes.
54: \end{abstract}
55:
56: \keywords{
57: black hole physics --- galaxies: spiral --- X-rays: galaxies}
58:
59: %%%%%%%% 1 %%%%%%%%%%%%
60: \section{Introduction}
61: %%%%%%%% 1 %%%%%%%%%%%%
62: Arm regions of nearby spiral galaxies have long been known to harbour
63: luminous point-like X-ray sources (Fabbiano 1989, 1998; Read et al. 1997; Roberts \& Warwick 2000).
64: Except for several identified with young supernova remnants,
65: they remained unidentified in other frequencies.
66: Being often time variable,
67: these sources are thought to be accreting single objects;
68: we hence call them ultra-luminous compact X-ray sources
69: (ULXs; Makishima et al. 2000, hereafter Paper~I).
70: With the X-ray luminosity reaching $10^{39-40}$ ergs s$^{-1}$,
71: the ULXs are suspected to contain stellar black holes (BHs)
72: of mass up to $\sim 100~M_\odot$ or more
73: (e.g., Colbert et al. 1995; Fabbiano 1998; Immler et al. 1999; Roberts \& Warwick 2000),
74: so as not to violate the Eddington limit.
75: However, this much exceeds the mass estimates ($5-15~ M_\odot$; Tanaka \& Lewin 1995)
76: for accreting stellar BHs in the Galaxy and the Large Magellanic Cloud.
77: There is no consensus on the presence of $\sim 100~M_\odot$ BHs,
78: even though such medium-mass BHs have been proposed under slightly different context
79: (Colbert \& Mushotzky 1999; Matsumoto \& Tsuru 1999; Ptak \& Griffiths 1999).
80: Furthermore, we do not find ULXs in the Milky Way or M31.
81: Therefore, the BH scenario of ULXs remained inconclusive,
82: and their nature has remained a big mystery.
83:
84: A great leap has been achieved through spectroscopy of a dozen ULXs with {\it ASCA}
85: (Takano et al. 1994; Okada et al. 1998; Colbert \& Mushotzky 1999;
86: Mizuno et al. 1999; Mizuno 2000; Kotoku et al. 2000; Paper~I),
87: as the spectra of majority of them have been reproduced by so-called
88: multi-color disk model (MCD model; Mitsuda et al. 1984)
89: that describes emission from an optically-thick standard accretion disk
90: around a BH (Shakura \& Sunyaev 1973).
91: These results significantly reinforce the BH scenario of ULXs,
92: since the MCD model can explain the soft X-ray spectra of
93: Galactic/Magellanic black-hole binaries (BHBs) in ``soft (or high) state''
94: (e.g., Makishima et al. 1986; Ebisawa et al. 1993; Dotani et al. 1997;
95: Kubota et al. 1998; Feroci et l. 1999; Paper~I).
96:
97: The scenario is nevertheless still clouded by a self-inconsistency,
98: that the measured inner-disk temperature of ULXs, $T_{\rm in} = 1.0-1.8$ keV,
99: is too high for the implied high BH mass (Colbert \& Mushotzky 1999; Paper~I).
100: Although the scenario may be salvaged by invoking rapid BH rotation
101: (Mizuno et al. 1999; Paper~I; Mizuno 2000; also Zhang et al. 1997b)
102: and/or ``slim disk'' concept (Abramowicz et al. 1988; Watarai et al. 2000),
103: it is important to assemble further observational clues.
104:
105: Here, we report further evidence supporting the BH interpretation of ULXs,
106: i.e., transitions of some ULXs between two distinct spectral states.
107: These results are based on two observations with {\it ASCA} (Tanaka et al. 1994) of two ULXs,
108: both in the spiral galaxy IC~342 at a distance of 4.0 Mpc (Tully 1988).
109:
110: \placefigure{fig:images}
111:
112: %=====================
113: \section{Observation}
114: %=====================
115: The first observation of IC~342 was performed on 1993 September 19,
116: as reported by Okada et al. (1998).
117: The obtained X-ray image is reproduced in Figure~\ref{fig:images}a.
118: In addition to unresolved emission from the nuclear region,
119: we see two ULXs, the brighter Source~1 and the fainter Source~2,
120: both known from the {\it Einstein} era (Fabbiano \& Trinchieri 1987).
121: On this occasion, Source~1 exhibited clear short-term variability (Okada et al. 1998).
122:
123: We observed IC~342 with {\it ASCA} again on 2000 February 24--March 1.
124: The GIS was operated in the normal PH mode,
125: while the SIS in the Faint mode.
126: Although the acquired X-ray image, Figure~\ref{fig:images}b,
127: again reveals the two sources,
128: their relative intensities have reversed from those in 1993.
129: As quantified in Table~1,
130: the Source~1 flux decreased to $\sim 30\%$ of its 1993 value,
131: while that of Source~2 increased by a factor of 1.8 meantime.
132:
133: \placetable{tbl:specfits}
134:
135: %==================================
136: \section{Data analysis and results}
137: %==================================
138: %--------------------------
139: \subsection{Data in 1993}
140: %--------------------------
141: Spectral results on Source~1 from the 1993 observation are already
142: described in Okada et al. (1998), Paper~I, and Mizuno (2000).
143: Referring to Paper~I, we simply quote here
144: that the 1993 spectra of Source~1 can be reproduced successfully
145: by an absorbed MCD model of $T_{\rm in} \sim 1.8$ keV.
146: In Table~1, we reproduce essence of these results.
147:
148: We here analyze the GIS/SIS spectra of Source~2 after Mizuno (2000).
149: We selected good data in the same manner as in Okada et al. (1998),
150: and accumulated photons around the Source~2 centroid, separately for the four detectors.
151: We subtracted background using blank-sky data.
152: We then added the SIS0 and SIS1 spectra into a single SIS spectrum,
153: and those from GIS2 and GIS3 into a single GIS spectrum.
154: The net exposure became 38 ks for the GIS and 36 ks for the SIS.
155: The derived Source~2 spectra, shown in Figure~\ref{fig:specfits}a,
156: are much harder than those of Source~1,
157: and a joint GIS/SIS fit with an absorbed power-law model of
158: photon index $\Gamma \sim 1.4$ has been successful (Table~1).
159: Although the absorbed MCD fit is also acceptable,
160: it requires an unrealistically high disk temperature of $T_{\rm in} \sim 3.0$ keV (Table 1).
161: An absorbed Bremsstrahlung model of temperature $>23$ keV is also acceptable ($\chi^2/\nu = 101.1/87$).
162: Considering typical spectra of BHBs,
163: we consider the power-law fit to be most appropriate for Source~2.
164:
165: \placefigure{fig:specfits}
166:
167: %--------------------------
168: \subsection{Data in 2000}
169: %--------------------------
170: We use only the GIS data for our spectral study of the 2000 observation,
171: because the long-term SIS degradation has made its low energy response rather uncertain.
172: We selected good GIS data using criteria of geomagnetic cutoff rigidity $>6$ GeV,
173: and the target elevation above the earth rim $> 5^\circ$.
174: This has yielded 276 ks of good exposure.
175:
176: Through the long observation,
177: Source~1 and Source~2 both showed a relatively steady 0.7--10 keV intensity of
178: 0.027 c s$^{-1}$ and 0.045 c s$^{-1}$ per GIS detector respectively,
179: with mild variation.
180: We here utilize the whole length of GIS data for spectral evaluation,
181: by accumulating events over two circular regions of radius
182: $2.'5$ around Source~1 and $3.'0$ around Source~2.
183: By subtracting background utilizing blank-sky data,
184: and adding data from the two GIS detectors,
185: we have obtained the Source~1 and Source~2 spectra,
186: shown in Figures \ref{fig:specfits}b and \ref{fig:specfits}c, respectively.
187:
188: The Source~1 spectrum has become much harder and less convex than in 1993,
189: and fitted approximately ($\chi^2/\nu = 119.8/88$)
190: with an absorbed power-law of $\Gamma \sim 1.8$.
191: However at 7--10 keV, we observe negative residuals suggestive of an Fe-K edge.
192: By introducing an edge absorption at $\sim 8.4$ keV,
193: the fit became acceptable (Table~1);
194: an $F$-test indicates that the edge feature is significant at 99.5\% confidence.
195: Although the MCD fit is not too bad (Table~1),
196: the absorption it requires falls below the Galactic value of $\sim 3 \times 10^{21}$ cm$^{-2}$.
197: We therefore consider the power-law fit with an Fe-K edge appropriate.
198:
199: In contrast, the Source~2 spectrum has become significantly more convex than in 1993.
200: It can now be expressed adequately with an MCD model of $T_{\rm in}=1.6$ keV,
201: whereas the power-law fit is unacceptable (Table 1).
202: Thus, the Source~2 spectrum in 2000 is close in shape to the Source~1 spectrum in 1993,
203: except for a higher photoelectric absorption.
204: An inclusion of a separate power-law component, with photon index fixed at 2.5,
205: yields only an upper limit of 15\% in terms of its 0.5--10 keV flux,
206: and the MCD parameters are not affected beyond the fitting errors (see also \S~3.3 of Paper~I).
207: This indicates dominance of the optically-thick disk emission in the {\it ASCA} band.
208:
209: To examine possible changes in the absorption,
210: we fitted the Source~1 spectra in 1993 and that in 2000
211: with the MCD model and the power-law model respectively, under a common absorption.
212: We obtained an acceptable ($\chi_\nu^2=1.13$) joint fit,
213: with $N_{\rm H} = 5.3 \pm 0.5$ cm$^{-2}$ and little changes in the other parameters.
214: Similarly, the 1993 power-law fit and the 2000 MCD fit to the Source~2 spectra
215: accepted ($\chi_\nu^2=0.94$) a common absorption of $N_{\rm H} = 18.1 \pm 0.8$ cm$^{-2}$.
216: Therefore, the transition of neither source was accompanied by measurable changes in the absorption.
217:
218: %==================================
219: \section{Discussion}
220: %==================================
221: In the first {\it ASCA} observation of IC~342,
222: Source~1 exhibited a high luminosity exceeding $10^{40}$ ergs s$^{-1}$ (Table~1),
223: prominent short-term variation,
224: and a convex spectrum expressed with an MCD model
225: of a high disk temperature, $T_{\rm in} \sim 1.8$ keV.
226: In Paper~1, these features have been identified as prototypical ULX properties.
227:
228: In the same observation, Source~2 showed a distinct hard spectrum
229: expressed with a power-law of $\Gamma \sim 1.4$.
230: Among the ULXs so far studied with {\it ASCA}, another object,
231: i.e., Source~A in NGC~1313 observed in 1993, exhibited a similarly hard spectrum
232: with $\Gamma \sim 1.7$ (Petre et al. 1994; Colbert \& Mushotzky 1999; Mizuno 2000);
233: the remaining objects showed the MCD-type spectra (Paper~I).
234: We hence presume that there are two subtypes of ULXs,
235: a majority exhibiting the soft MCD-type spectra,
236: and a minority showing the hard power-law spectra.
237: However, from these results alone, we cannot tell
238: whether or not these subtypes represent intrinsically distinct two source populations.
239:
240: The second {\it ASCA} observation made in 2000 revealed
241: that Source~1 had made a soft-to-hard state transition,
242: whereas Source~2 a hard-to-soft one,
243: although their reciprocal behavior must be a chance coincidence.
244: To visualize this, we present in Figure~\ref{fig:deconvolved} their deconvolved spectra.
245: These results immediately imply that;
246: (1) the two ULXs must be accreting compact objects,
247: to make such drastic spectral changes in 7 years;
248: (2) the two ULX subtypes represent two well-defined states
249: (soft state and hard state) of the same source population;
250: (3) some (if not all) ULXs make transitions between the two spectral states;
251: and
252: (4) at least in the {\it ASCA} band, the source is less luminous while in the hard state.
253: These properties give a conclusive support to the BH interpretation of ULXs,
254: because they are just what characterize the known BHBs
255: (e.g., Maejima et al. 1984; Tanaka \& Lewin 1995; Tanaka \& Shibazaki 1996;
256: Zhang et al. 1997a; Wilms et al. 2000).
257: A similar (though less convincing) example may be the aforementioned Source~A in NGC~1313,
258: because its spectrum turned very soft in 1995
259: ($\Gamma \sim 2.8$; Colbert \& Mushotzky 1999; Mizuno 2000).
260:
261: \placefigure{fig:deconvolved}
262:
263: Thus, the long-lasting puzzle of ULXs are being unveiled,
264: but a series of new questions are arising in turn.
265: Which is the right solution to the
266: ``too high $T_{\rm in}$'' problem?
267: Are the inferred high BH masses of ULXs consistent with the current
268: understanding of stellar BH formation?
269: Why are ULXs absent in Our Galaxy and M31?
270: Do they form a population distinct from the Galactic/Magellanic BHBs?
271: In short, the ULX formation scenario is yet to be clarified.
272:
273: The authors thank the {\it ASCA} team members.
274: The present work is supported in part by
275: the Grant-in-Aid for Center-of-Excellence, No. 07CE2002,
276: from Ministry of Education, Science, Sports and Culture of Japan.
277:
278: \begin{thebibliography}{99}
279: \setlength{\baselineskip}{3mm}
280: \setlength{\itemsep}{0.5mm}
281: \bibitem[]{} Abramowicz, M. A., Czerny, B., Lasota, P., \& Szuszkiewicz, E. 1988,
282: ApJ, 332, 646
283: \bibitem[]{} Colbert, E. J. M., \& Mushotzky, R. F. 1999, ApJ, 519, 89
284: \bibitem[]{} Colbert, E. J. M., Petre, R., Schlegel, E., \& Ryder, S. D. 1995, ApJ, 446, 177
285: \bibitem[]{} Dotani, T., et al. 1997, ApJ, 485, L87
286: \bibitem[]{} Ebisawa, K., Makino, F., Mitsuda, K., Belloni, T., Cowley, A.,
287: Schmidke, P., \& Treves, A. 1993, ApJ, 403, 684
288: \bibitem[]{} Fabbiano, G. 1989, ARA\&A, 27, 87
289: \bibitem[]{} Fabbiano, G., \& Trinchieri, G. 1987, ApJ, 315, 46
290: \bibitem[]{} Fabbiano, G. 1998, in The Hot Universe, eds. K. Koyama, S. Kitamoto and M. Itoh
291: (Kluwer Academic, Dordrechit), 93
292: \bibitem[]{} Feroci, M., Matt, G., Pooley, G., Costa,, E., Tavani, M., \& Belloni, T. 1999,
293: A\&A, 35, 985
294: \bibitem[]{} Immler, S., Vogler, A., Ehle, M., \& Pietsch, W. 1999, A\&A, 352, 415
295: \bibitem[]{} Kotoku, J., Mizuno, T., Kubota, A., \& Makishima, K. 2000, PASJ, in press
296: \bibitem[]{} Kubota, A., Tanaka, Y., Makishima, K., Ueda, Y., Dotani, T., Inoue, H.,
297: \& Yamaoka, K. 1998, PASJ, 50, 667
298: \bibitem[]{} Maejima, Y., Makishima, K., Matsuoka, M., Ogawara, Y., Oda, M.,
299: Tawara, Y., \& Doi, K. 1984, ApJ 285, 712
300: \bibitem[]{} Makishima, K., et al. 1986, ApJ, 308, 635
301: \bibitem[]{} Makishima, K., et al. 2000, ApJ, 535, 632 (Paper~I)
302: \bibitem[]{} Matsumoto, H., \& Tsuru, T. G. 1999, PASJ, 51, 321
303: \bibitem[]{} Mitsuda, K., et al. 1984, PASJ, 36, 741
304: \bibitem[]{} Mizuno, T. 2000, PhD Thesis, University of Tokyo
305: \bibitem[]{} Mizuno, T., Ohnishi, T., Kubota, A., Makishima, K., \& Tashiro, M. 1999, PASJ, 51, 663
306: \bibitem[]{} Okada, K., Dotani, T., Makishima, K., Mitsuda, K., \& Mihara, T. 1998, PASJ, 50, 25
307: \bibitem[]{} Petre, R., Okada, K., Mihara, T., Makishima, K., \& Colbert, E. J. M. 1994, PASJ, 46, L115
308: \bibitem[]{} Ptak, A., \& Griffiths, R. 1999, ApJ, 517, 85
309: \bibitem[]{} Read, A. M., Ponman, T. J., \& Strickland, D. K. 1997, MNRAS, 286, 626
310: \bibitem[]{} Roberts, T. P., \& Warwick, R. S. 2000, MNRAS 315, 98
311: \bibitem[]{} Shakura, N. I., \& Sunyaev, R. A. 1973, A\&A, 24, 337
312: \bibitem[]{} Takano, M., Mitsuda, K., Fukazawa, Y., \& Nagase, F. 1994, ApJ, 436, L47
313: \bibitem[]{} Tanaka, Y., Inoue, H., \& Holt, S. S. 1994, PASJ, 46, L37
314: \bibitem[]{} Tanaka, Y., \& Lewin, W. H. G. 1995, in X-ray Binaries,
315: eds. W. H. G. Lewin, J. van paradijs, and W. P. J. van den Heuvel
316: (Cambridge University Press, Cambridge), p126
317: \bibitem[]{} Tanaka, Y., \& Shibazaki, N. 1996, ARA\&A, 34, 607
318: \bibitem[]{} Tully, B. R. 1988, Nearby Galaxy Catalog, Cambridge University Press
319: (Cambridge University Press, Cambridge)
320: \bibitem[]{} Watarai, K., Fukue, J., Takeuchi, M., \& Mineshige, S. 2000, PASJ, 52, 133
321: \bibitem[]{} Wilms, J., Nowak, M. A., Pottschmidt, K., Heindl, W. A., Dove, J. B.,
322: \& Begelman, M. C. 2000, astro-ph/0005489
323: \bibitem[]{} Zhang, S. N., Cui, W., Harmon, B. A., Paciesas, W., Remillard, R. E.,
324: \& van Paradijs, J. 1997a, ApJ, 477, L95
325: \bibitem[]{} Zhang, S. N., Cui, W., \& Chen, W. 1997b, ApJ, 482, L155
326:
327: \end{thebibliography}
328:
329:
330:
331: %%%%%%%%%%%%%%%% Table %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
332: \begin{table}[h]
333: \caption{The spectral parameters of IC~342 Source~1 and Source~2, with 90\% confidence limits.}
334: \begin{center}
335: \begin{small}
336: \begin{tabular}{lcccccccc}
337: \hline %--------------------------------------------------------
338: Epoch & $f_{\rm x}^{a)}$ & $L_{\rm x}^{b)}$ &
339: \multicolumn{3}{c}{Fit with the MCD model}&
340: \multicolumn{3}{c}{Fit with the power-law model} \\
341: & & &$T_{\rm in}$ (keV) &$N_{\rm H}^{c)}$& $\chi^2/\nu$
342: &Photon index &$N_{\rm H}^{c)}$& $\chi^2/\nu$\\
343: \hline %---------------------------------------------------------
344: \hline %---------------------------------------------------------
345: Source 1 \\
346: ~~1993 $^{d)}$&10.2 & 1.9 & $1.77 \pm 0.05$ & $4.7 \pm 0.3$ & 137.4/135 &
347: $1.90 \pm 0.05$ & $9.3 \pm 0.6$ & 266.5/135\\
348: ~~2000 & 3.1 & 0.6 & $2.06 \pm 0.08$ & $1.9 \pm 0.4$ & 123.9/89 &
349: $1.73 \pm 0.06^{e)}$ & $6.4 \pm 0.7^{e)}$ & 101.1/86$^{e)}$\\
350: \hline %---------------------------------------------------------
351: Source 2 \\
352: ~~1993 & 4.1 & 0.8 &$3.03 \pm 0.30$ & $9.9 \pm 0.9$ & 96.4/87 &
353: $1.39 \pm 0.10$ &$14.3 \pm 1.6$ & 102.8/87 \\
354: ~~2000 & 7.2 & 1.4 &$1.62 \pm 0.04$ &$18.2 \pm 0.8$ & 90.5/89 &
355: 2.48$^{f)}$ & 31.8$^{f)}$ & 232.0/89 \\
356: \hline %---------------------------------------------------------
357: \end{tabular}
358: \end{small}
359: \end{center}
360: \begin{footnotesize}
361: \begin{itemize}
362: \setlength{\baselineskip}{3mm}
363: \setlength{\itemsep}{-2mm}
364: \item[$^{a)}$] The 0.5--10 keV source flux at the top of atmosphere,
365: in unit of $10^{-12}$ ergs s$^{-1}$ cm$^{-2}$.
366: \item[$^{b)}$] The absorption-uncorrected 0.5--10 keV luminosity
367: in $10^{40}$ ergs s$^{-1}$,
368: assuming 4.0 Mpc distance and an isotropic emission.
369: \item[$^{c)}$] Column density for absorption assuming solar abundances,
370: in unit of $10^{21}$ cm$^{-2}$.
371: \item[$^{d)}$] Results taken from Paper~I.
372: \item[$^{e)}$] An ionized Fe-K edge (at $8.4 \pm 0.3$ keV) is applied,
373: with an optical depth of $0.9 \pm 0.5$.
374: \item[$^{f)}$] Errors are not shown since the fit is highly unacceptable.
375: \end{itemize}
376: \end{footnotesize}
377: \label{tbl:specfits}
378: \end{table}
379: %%%%%%%%%%%%%%%% Table %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
380:
381: \clearpage
382:
383: %%%%%%%%%%%%%%%% Figure 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
384: \vspace*{-3cm}
385: \epsscale{0.8}
386: \plotone{f1.eps}
387: \figcaption{
388: The 0.7--10 keV X-ray images of IC~342 taken with the {\it ASCA} GIS,
389: shown after smoothing with a Gaussian kernel of $\sigma = 0'.5$
390: and superposed on the optical image.
391: The background has not been subtracted,
392: and the X-ray contour levels are linear.
393: The source to the lower right of the optical nucleus is Source~1,
394: while that to the upper right is Source~2.
395: (a) The data taken on 1993 September.
396: (b) Those of 2000 February.
397: \label{fig:images}
398: }
399: %%%%%%%%%%%%%%%% Figure 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
400:
401: \vspace{1cm}
402: %%%%%%%%%%%%%%%% Figure 2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
403: \epsscale{0.4}
404: \plotone{f2.eps}
405: \figcaption{
406: The {\it ASCA} spectra of IC~342 Source~1 and Source~2,
407: together with prediction of the best-fit model (Table~1).
408: The model parameters are given in Table~1.
409: Fit residuals are shown for both the power-law and MCD models.
410: (a) The SIS (open circles) and GIS (filled circles) spectra of Source~2 in 1993,
411: jointly fitted with an absorbed power-law model.
412: (b) The GIS spectrum of Source~1 in 2000,
413: fitted with an absorbed power-law model incorporating the ionized Fe-K edge.
414: (c) The GIS spectrum of Source~2 obtained in 2000,
415: fitted with an absorbed MCD model.
416: \label{fig:specfits}
417: }
418: %%%%%%%%%%%%%%%% Figure 2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
419:
420: %\clearpage
421: \vspace*{1cm}
422: %%%%%%%%%%%%%%%% Figure 3 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
423: \epsscale{0.4}
424: \plotone{f3.eps}
425: \figcaption{
426: The {\it ASCA} GIS spectra of IC~342
427: Source~1 (panel a) and Source~2 (panel b),
428: presented after removing the instrumental responses and the photoelectric absorption.
429: Open and filled circles indicate the data taken in 1993 and 2000, respectively.
430: The Source~1 spectrum in 1993 and that of Source~2 in 2000
431: are deconvolved using the MCD-model fits,
432: whereas the others using the power-law fits.
433: The solid lines indicate these best-fit models in their incident forms.
434: \label{fig:deconvolved}
435: }
436: %%%%%%%%%%%%%%%% Figure 2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
437:
438:
439: \end{document}
440:
441:
442:
443:
444:
445:
446:
447:
448:
449:
450:
451: