1: %\documentclass[onecolumn]{emulateapj}
2: %\documentclass[twocolumn]{emulateapj}
3: \documentclass[12pt,preprint]{aastex}
4: %\usepackage[twocolumn]{emulateapj5}
5:
6: %% manuscript produces a one-column, double-spaced document:
7:
8: %% \documentclass[manuscript]{aastex}
9:
10: %% preprint2 produces a double-column, single-spaced document:
11:
12: %% \documentclass[preprint2]{aastex}
13:
14: %% Sometimes a paper's abstract is too long to fit on the
15: %% title page in preprint2 mode. When that is the case,
16: %% use the longabstract style option.
17:
18: %% \documentclass[preprint2,longabstract]{aastex}
19:
20: %% If you want to create your own macros, you can do so
21: %% using \newcommand. Your macros should appear before
22: %% the \begin{document} command.
23: %%
24: %% If you are submitting to a journal that translates manuscripts
25: %% into SGML, you need to follow certain guidelines when preparing
26: %% your macros. See the AASTeX v5.x Author Guide
27: %% for information.
28:
29: %\newcommand{\vdag}{(v)^\dagger}
30: %\newcommand{\myemail}{skywalker@galaxy.far.far.away}
31:
32: %% You can insert a short comment on the title page using the command below.
33:
34: %\slugcomment{Not to appear in Nonlearned J., 45.}
35:
36:
37: \shorttitle{X-ray and optical variability of NGC 1313 X-2}
38: \shortauthors{Mucciarelli et al.}
39:
40: \begin{document}
41:
42: \title{X-ray and optical variability of the ultraluminous X-ray source NGC 1313 X-2}
43:
44: \author{P. Mucciarelli\altaffilmark{1,2}, L. Zampieri\altaffilmark{2},
45: A. Treves\altaffilmark{3}, R. Turolla\altaffilmark{4} and R. Falomo\altaffilmark{2}}
46: \altaffiltext{1}{Department of Astronomy, University of Padova, Vicolo
47: dell'Osservatorio 2, I-35122 Padova,
48: Italy; paola.mucciarelli@oapd.inaf.it}
49: \altaffiltext{2}{INAF-Astronomical Observatory of Padova, Vicolo dell'Osservatorio 5,
50: I-35122 Padova, Italy; zampieri@pd.astro.it, falomo@pd.astro.it}
51: \altaffiltext{3}{Department of Physics and Mathematics, University of Insubria,
52: Via Valleggio 11, I-22100 Como, Italy; treves@mib.infn.it}
53: \altaffiltext{4}{Department of Physics, University of Padova, Via Marzolo 8, I-35131
54: Padova, Italy; turolla@pd.infn.it}
55:
56:
57: \begin{abstract}
58:
59: We present an analysis of recent {\it XMM-Newton} and {\it HST}
60: archive data of the ultraluminous X-ray source NGC 1313
61: X-2. Quasi-simultaneous observations taken with {\it XMM-Newton}, {\it HST}
62: and VLT allow us to study both the X-ray light curve and its
63: correlation with the optical emission of the two proposed ULX
64: counterparts. At the end of December 2003 the source experienced a
65: short, but intense flare, reaching a maximum luminosity of
66: $\sim\,10^{40}$ erg/s. At the same time, the optical flux of both the suggested
67: counterparts did not show pronounced variations ($\la 30$\%).
68: Assuming that the ULX emission is isotropic and taking X-ray
69: reprocessing into account, the optical data for one of the proposed
70: counterparts are consistent with it being an early type, main sequence
71: star of $\sim 10-18 M_\odot$ losing matter through Roche-lobe
72: overflow onto a $\sim 120 M_\odot$ black hole at an orbital separation
73: corresponding to a period of $\sim 2$ days.
74:
75: \end{abstract}
76:
77: %% Keywords should appear after the \end{abstract} command. The uncommented
78: %% example has been keyed in ApJ style. See the instructions to authors
79: %% for the journal to which you are submitting your paper to determine
80: %% what keyword punctuation is appropriate.
81:
82:
83: \keywords{galaxies: individual (\objectname{NGC 1313}) --- stars: individual
84: (\objectname{NGC 1313 X-2}) --- X-rays: binaries --- X-rays: galaxies}
85:
86:
87: \section{Introduction}
88:
89: When, at the beginning of the 1980s, point-like, off-nuclear X-ray
90: sources in the field of nearby galaxies were first detected (see,
91: e.g., \citealt{fab89}), it was immediately recognized that their
92: luminosity was unusually large. If physically associated with their
93: host galaxies, these sources would have an isotropic luminosity
94: in excess of the Eddington limit for a $10 M_\odot$ object. Nowadays,
95: more than 150 ultraluminous X-ray sources (ULXs) are known (see
96: e.g. \citealt{rob00,col02,swa04,liu05}).
97:
98: It is estimated that a significant fraction of ULXs are interacting
99: supernovae or background AGNs ($\sim$ 50\%; see
100: \citealt{fosc02,mas03,swa04}). However, the X-ray variability
101: of many of them is similar to that observed in Galatic X-ray
102: binaries (see e.g. \citealt{lapar01,col02,swa04,zamp04}, hereafter Z04).
103: The recent detection of a
104: 62 days modulation in the light curve of M 82 X-1, interpreted as the
105: orbital period of the system, provided a direct confirmation of the
106: binary nature of at least some ULXs \citep{ksm06a,ksm06b}. Moreover, ULX
107: spectral properties share similarities with those of Galactic black hole X-ray
108: binaries (BHXRBs, e.g. \citealt{fosc02a}). In several cases the spectrum can be well
109: reproduced by a multicolor disk (MCD) blackbody plus a power law (PL),
110: although the temperature of the MCD component is often much lower than
111: that observed in BHXRBs (e.g. \citealt{mi03,mi04,fk05}). For the
112: brightest ULXs, a possible curvature above 2-3 keV has been recently
113: reported and more sophisticated spectral models appear to give better
114: agreement with observations \citep{stob06}.
115:
116: All these properties along, in some cases, with the detection of
117: stellar-like optical counterparts (\citealt{rob01,goad02,liu02,liu04,kaa04}; Z04;
118: \citealt{kaa05,muc05}, hereafter M05; \citealt{sor05}), strongly suggest that a
119: sizeable fraction of ULXs are accreting X-ray binaries. The present debate is focussed on
120: understanding what type of binaries they are. Many of the ULX
121: properties can be explained if they do not emit isotropically
122: \citep{kin01,kin02,kp03} or are dominated by emission from a relativistic jet
123: \citep{kfm02,gak02,kaa03}. Another possibility is that they are truly
124: emitting above the Eddington limit for $10 M_\odot$, either because
125: accretion proceeds through a slim disk \citep{ebi03,kaw03} or because
126: the compact object is an Intermediate Mass Black Hole (IMBH) with a
127: mass in excess of $100 M_{\odot}$ (e.g. \citealt{cm99,mi03,pat05,pat06}).
128: Despite the inherent difficulties due to the low counting statistics, X-ray timing
129: analysis has been attempted in some ULXs and led to the detection of a quasi
130: periodic oscillation in the power density spectrum of M 82 X-1. This may
131: represent a powerful, independent method to measure the black hole mass
132: \citep{sm03,ft04,dew06,muc06}.
133:
134: Multiwavelength observations are an unvaluable tool to investigate the
135: nature of ULXs. Radio emission, when present, gives important clues about
136: the geometry, energetics and lifetime of ULXs \citep{kaa03,miln05}.
137: Optical observations are crucial to identify ULX counterparts and
138: to study the properties of putative ULX binary systems. Up to now only a
139: very
140: small number of ULXs have been convincingly associated with stellar
141: objects of known spectral type (e.g. \citealt{liu02,liu04,kaa04};
142: M05). All these ULXs are hosted in young stellar environments or
143: star-forming regions and their optical counterparts have properties
144: consistent with those of massive stars. Some ULXs are also associated
145: with extended optical emission nebulae \citep{pak02,pak05}.
146:
147: NGC 1313 X-2 is a well studied ULX. The X-ray variability, high (isotropic)
148: luminosity and presence of a soft X-ray spectral component make it a
149: prototypical object. Furthermore, the presence of an emission nebula
150: (\citealt{pak02,ramsey06}; Z04) and the detection of optical counterpart(s)
151: (Z04; M05) provide a considerable amount of information on the ULX environment,
152: available only for a very limited number of objects. Here we present a
153: systematic study of the X-ray and optical variability of NGC 1313 X-2 based on
154: archive data of the {\it XMM-Newton} satellite, and the ESO VLT and {\it HST}
155: telescopes. Observations are reported in \S~\ref{obs} and results in
156: \S~\ref{variability}, where also a model for the optical emission is presented.
157: Discussion follows.
158:
159:
160: \section{Observations}\label{obs}
161:
162: \subsection{X-ray observations}\label{x-ray}
163:
164: {\it XMM-Newton} observed NGC 1313 in 9 exposures taken between 2000
165: and 2004. All the observations are listed in Table \ref{tab:olxmm}. A
166: detailed analysis of the October 2000 data of NGC 1313 X-2 have been
167: performed by Z04 (see also \citealt{tur04}). Here we report results from an
168: analysis of the EPIC pn exposures of the 8 more recent
169: observations. Data reduction and extraction have been carried out with
170: standard software ({\sc XMM-SAS} v. 6.0.0). All the observations are
171: affected by solar flares. The good time intervals left after
172: subtraction of the high background periods (when the total off-source
173: count rate above 10 keV is $>$ 1.0 counts s$^{-1}$ for EPIC pn) are
174: reported in Table
175: \ref{tab:olxmm}.
176: For the analysis we consider all the exposures with a good time
177: interval longer than 1 ks.
178: %The only exception is observation 2 that,
179: %despite being shorter, was also included in the analysis to increase
180: %the actual time span sampled.
181: After performing standard cleaning of the
182: event lists, we extracted source counts from a circle of 40$\arcsec$
183: centered on the position of NGC 1313 X-2 (Z04). The background
184: counts were extracted from a circle of 50$\arcsec$ on the same CCD.
185:
186: The spectral analysis was carried out within {\sc XSPEC} (v.
187: 11.2.0). A two-component model consisting of an absorbed
188: multicolor disk blackbody (MCD) plus a power law (PL) has been employed
189: throughout, as it is routinely done for BHXRBs. Previous applications of
190: the same spectral model to a number
191: of ULXs, including NGC 1313 X-2 (\citealt{mi03,mi04,cr04,kong05}; Z04) gave
192: a satisfactory fit to the data. The absorbing column
193: density inferred from the different datasets is consistent with a
194: constant value. We then performed again the fits fixing $N_{H}$
195: to the average value weighted by the exposure time
196: ($N_{H} = 4.02 \times 10^{21}$cm$^{-2}$). For consistency, we also repeated the
197: analysis of the 2000 EPIC pn spectrum of NGC
198: 1313 X-2 following the procedure described above. The results from the
199: spectral fits are
200: reported in Table \ref{tab:x2fit}. The {\it XMM-Newton} fluxes were
201: consistently derived from the parameters of the spectral fits and are
202: reported, with the corresponding luminosities, in Table \ref{tab:x2fit}
203: (a distance of 3.7 Mpc was assumed for the host galaxy;
204: \citealt{tul88}). The errors have been estimated from the maximum and
205: minimum values of the flux, obtained varying the fit
206: parameters systematically.
207:
208: \subsection{Optical observations}
209: \label{optical}
210:
211: B, V and R images and spectra of NGC 1313 X-2 were taken with
212: VLT+FORS1 in December 2003. The results were presented by M05. {\it
213: HST} images of this field were also obtained with ACS in two epochs
214: (see Figure \ref{fig:bv}). The optical observations were performed in
215: parallel with the {\it XMM-Newton} pointings. The observation log of the VLT
216: and {\it HST} images considered here are reported in Table
217: \ref{tab:olxmm}. Aperture photometry was performed on the drizzled
218: calibrated data (reduced by the {\it HST} pipeline) and transformed to
219: the Cousins system following \cite{sir05} (see Table
220: \ref{tab:lmag}).
221:
222: The uncertainty on the {\it HST} magnitudes are dominated by the calibration
223: error ($\sim$ 0.03 mag for the {\it HST} photometry of point sources),
224: including filter transformation.
225: As a further check of the internal consistency of the {\it HST}
226: photometry, we compared the magnitudes of thirteen field stars
227: obtained in the two epochs. The difference is significant only for one
228: source in the sample ($\sim$ 0.3 mag). Excluding this source, the
229: variability of which is probably intrinsic, the magnitude changes are
230: randomly scattered around zero, with a mean absolute deviation of 0.04
231: mag.
232:
233:
234: The {\it HST} images clearly confirm that two distinct objects are
235: present inside the X-ray error box of NGC 1313 X-2 (see Figure
236: \ref{fig:bv}), as first shown by M05 on the basis of VLT data.
237: If we assume $A_V \simeq 0.3$ (Z04), taking Galactic absorption into
238: account (the \citealt{card89} extinction law with $R_V=A_V/E(B-V)=3.1$
239: has been adopted throughout), the unreddened colors inferred from the
240: 1st {\it HST} epoch are (B-V)$_0 \sim -0.13$ and $\sim$1.46 for C1 and
241: C2, respectively (see Table \ref{tab:lmag}). The color for object C1
242: is consistent with that derived from VLT data ((B-V)$_0 \sim -0.2$;
243: M05). On the basis of the {\it HST} photometry, the color of object C2
244: is close to that of a K3-K4 supergiant. For both objects, there is
245: evidence of variability in the V band between the two
246: {\it HST} epochs ($\sim 0.1$ mag, Figure \ref{fig:x2cl}; see also \citealt{ramsey06}).
247:
248:
249: %**********************************************
250:
251: \section{X-ray and optical variability}\label{variability}
252:
253: \subsection{X-ray light curve}\label{lightcurve}
254:
255: Figure \ref{fig:x2fl} shows the (unabsorbed) X-ray flux
256: for all the available observations of NGC 1313 X-2. The {\it XMM-Newton} data
257: were derived from the best fitting spectral models (see Table
258: \ref{tab:x2fit}), while the {\it Einstein},
259: {\it ROSAT} and {\it ASCA} data are taken from Z04.
260:
261: Until 2000 NGC 1313 X-2 exhibited variability up to a factor of two on a
262: timescale of months, with a maximum luminosity of $\sim 4 \times
263: 10^{39}$ erg s$^{-1}$ (Z04).
264: %The more recent data reported here show a
265: %quite intense flare lasting a few days occurring at the end of 2003.
266: Around December 25, 2003 (observation 6 in Table
267: \ref{tab:olxmm}), the source produced an intense
268: flare, reaching a maximum unabsorbed flux of $\sim 10.0 \times
269: 10^{-12}$erg cm$^{-2}$ s$^{-1}$ (Table \ref{tab:x2fit}). At the
270: distance of NGC 1313 this corresponds to an intrinsic luminosity of
271: $\sim 10^{40}$erg s$^{-1}$. Clearly this value depends on the adopted
272: spectral model and hence should be taken with care (see e.g. the
273: slighty smaller values recently obtained for the October 2000 observation
274: by \citealt{stob06} adopting a more sophisticated
275: spectral model). We also measured the fluxes of another ULX in the
276: field (NGC 1313 X-3), known to be an interacting supernova (SN 1978K),
277: in order to check if the significant luminosity increase was real
278: or artificially produced by residual systematic effects between the
279: two observations. The flux of the supernova ($\sim 8.2 \times
280: 10^{-13}$erg cm$^{-2}$ s$^{-1}$) is consistent
281: with a constant, within the uncertainties (the variation is $\la$
282: 20\%). Hence we conclude that the luminosity increase of NGC 1313 X-2
283: is significant and fully qualifies NGC 1313 X-2 as a bright
284: ULX.
285:
286: From the observed maximum luminosity ($L_{max} \sim 1.5 \times
287: 10^{40}$erg s$^{-1}$) and assuming isotropic emission, the black hole
288: mass $M_{BH}$ obtained setting $L_{max}=L_{Edd}$ ($L_{Edd}$ is Eddington
289: luminosity) is $\simeq 120 M_\odot$, about 2 times larger than that
290: previously estimated by Z04. Sub-Eddington accretion would imply an
291: even larger mass.
292:
293:
294: \subsection{X-ray spectral changes}\label{spectchange}
295:
296: The spectra of numerous ULXs, including NGC 1313 X-2, are well described by a
297: MCD+PL model, similarly to those observed in Galactic BHXRBs. In fact, there is
298: also some evidence of a closer similarity, inasmuch as some ULXs appear to show
299: state transitions \citep{mak04,win05}. In the following we summarize the main
300: results of an analysis of the X-ray spectral variability of NGC 1313 X-2. The
301: most significant result is that the slope of the PL component seems to
302: correlate with the flux, i.e. at higher fluxes the spectrum hardens (see Table
303: \ref{tab:x2fit}). This behavior was already noticed by Z04 on the basis of a
304: comparison between two {\it ASCA} observations and is opposite to that
305: usually shown by Galactic BHXRBs. A similar correlation was also
306: observed in a few ULXs in the Antennae galaxy by \cite{fab03}. The MCD
307: component is important in the Oct 2000 and in the 2003 pointings with
308: higher counting statistics (observations 4 and 5 in Table
309: \ref{tab:olxmm}). Although there is some evidence of intrinsic
310: variability of the thermal component, no definite conclusion can be
311: reached at present because of the insufficient statistics. Finally, we note
312: that the flux of the MCD component is comparable to that of the PL
313: component (see Table \ref{tab:x2fit}).
314:
315:
316: \subsection{Modeling the optical emission}\label{model}
317:
318: In order to study the optical emission properties of objects C1 and C2 and
319: compare them with the {\it HST}+VLT photometry, we implemented a model to
320: compute the optical spectrum of a binary system with an IMBH taking irradiation
321: effects into account. Our calculation relies on the same assumptions discussed
322: in \cite{cop06}, who recently presented a thorough investigation of the
323: infrared-through-optical emission properties of X-ray binaries with IMBHs. More
324: specifically, we assume that accretion onto the IMBH is fuelled by a massive
325: companion filling its Roche lobe and that the X-ray emission is isotropic; the
326: consequences of introducing some degree of beaming are discussed later on. A
327: standard Shakura-Sunyaev disk (e.g. \citealt{fkr02}) is assumed and both the
328: X-ray irradiation of the companion (including the effects of disk shadowing)
329: and the self-irradiation of the disk are accounted for. Following
330: \cite{cop06}, radiative transfer at the donor and disk surfaces is treated
331: assuming a plane-parallel atmosphere in radiative equilibrium, illuminated by
332: the X-ray flux emitted from the innermost part of the accretion disk (see also
333: \citealt{wu01}). In order to keep our treatment simple, the companion
334: star is taken to be spherical, neglecting the effects produced by the Roche
335: lobe geometry and also those related to the (possible) deformation induced by
336: radiation pressure. Limb and gravity darkening were not included.
337:
338: The model depends on the masses of the two components, the binary
339: period (which, in turn, fixes the orbital separation), the accretion
340: rate and the (unirradiated) temperature of the donor, in addition to
341: the inclination angle and the orbital phase. The accretion efficiency
342: and the albedo of the surface layers were chosen to be
343: $\eta=GM_{BH}/c^2 r_{in}=r_S/2 r_{in}=0.17$ (where $r_S$ is the
344: Schwarzschild radius and $r_{in}=3 r_S$ is the inner disk radius) and
345: $\alpha=0.9$ (e.g. \citealt{jpa96}). Following
346: \cite{cop06}, we took the hardness ratio $\xi=F_X(<1.5\, {\rm keV})/F_X(>1.5\,
347: {\rm keV})=0.1$. The absorption parameters in the same two spectral
348: bands were selected as $k_s=2.5$ and $k_h=0.01$. The V and B
349: magnitudes of the (irradiated) disk plus donor have been computed for
350: several values of the parameters of the binary. Each sequence of
351: models, at fixed inclination angle $i$, accretion rate $\dot M$ and
352: donor mass $M$, corresponds to a track in the color-magnitude diagram
353: (CMD; B-V vs. V here) along which the BH mass varies. Only inclination
354: angles smaller than $\sim 70 ^{\circ}$ are considered because eclipsing
355: effects of the accretion disk on the donor (and viceversa) have not been
356: taken into account. On the other hand, from the available data there is no positive
357: evidence for eclipses of the X-ray source or the donor in NGC 1313 X-2.
358: The mass and luminosity class of the donor fix its (unirradiated) surface
359: temperature $T_{eff}$.
360: Different tracks have been obtained varying the orbital period $P_{orb}$
361: which, in turn, determines the Roche lobe radius. The maximum allowed
362: period is that for which the Roche lobe radius is equal to the donor radius.
363: The computed tracks are compared with the optical (unreddened) magnitudes
364: and colors of objects C1 and C2 for both the VLT and {\it HST} observations in
365: \S~\ref{application}.
366: %Results are reported in Section \ref{var}.
367:
368: \subsection{Objects C1 and C2}\label{application}
369:
370: In order to constrain the parameters of the binary, we used the
371: optical binary emission model introduced in Section \ref{model}. One
372: of the 2003 {\it XMM-Newton} pointings of NGC 1313 X-2 is within 2-3
373: days from the 1st {\it HST} epoch, while another is
374: close to the VLT observation. These two {\it XMM-Newton}
375: observations are those of 2003 November 25
376: and December 23; the latter was preferred to the observation of December 25 because
377: of the higher statistics. It is therefore of interest to compare the V
378: and B magnitudes of objects C1 and C2 in these two epochs. The variation of the unabsorbed
379: X-ray flux between the same epochs is $\sim$80\% (see Table \ref{tab:x2fit}). At the same
380: time, however, the V and B band magnitudes of C1 do not show significant evidence of
381: variability. The relative photometric error between the {\it HST} and VLT data as measured
382: on a sample of field stars is in fact $\simeq 0.3$ mag (see Figure \ref{fig:x2cl}). A
383: similar conclusion is reached also for object C2. Again, the magnitude change between
384: the {\it HST} and VLT epochs is always smaller than the relative photometric error
385: ($\simeq 0.5$ mag for objects fainter than V=24; see again Figure \ref{fig:x2cl}).
386:
387: As an initial guess for the donor parameters in our model we use the values inferred on
388: the basis of VLT photometry (M05): an O9-B0 V star of $\sim 20 M_\odot$, $T_{eff}\sim
389: 30000$ K for C1, and a G-K I star of $\sim 10 M_\odot$, $T_{eff}\sim 4500$ K for C2. As
390: mentioned above, this is consistent with what inferred from the {\it HST} photometry.
391: The donor star in the case of of object C1 is assumed to be on the zero age main sequence
392: and in contact with the Roche lobe (i.e. its radius is equal to the Roche lobe radius).
393: Possible evolutionary effects or disturbances caused by the intense mass transfer are not
394: accounted for.
395: Results for object C1 are shown in Figure~\ref{fig:model} for two different values of $\dot
396: M$, chosen in such a way to match the
397: {\it XMM-Newton} flux measured in the two observations of
398: 2003 November 25 and December 23. The tracks on the CMD diagram are in agreement with
399: the observed V band magnitude and (B-V) color of object C1 for
400: $P_{orb}\simeq 1.7$~d, $M\simeq 15 M_\odot$ and $T_{eff}\simeq 25000$~K
401: (corresponding to an early B main sequence star). Taking into account for
402: current uncertainties on both color and magnitude, the companion mass and
403: temperature, and the orbital period may vary in the ranges $10\lesssim M/M_\odot\lesssim 18$,
404: $20000\, {\rm K}\lesssim T_{eff} \lesssim 30000\, {\rm K}$, and $1.5\lesssim P_{orb}
405: \lesssim 2$ d, respectively.
406: If the donor makes contact with its Roche lobe and $0.1\lesssim\,M/M_{BH}\, \lesssim 0.8$,
407: the orbital period
408: becomes a function only of the donor radius (or mass). In fact, when combining
409: together the III Kepler's law and the expression for the Roche lobe
410: radius \citep{fkr02}, the dependence on $M_{BH}$ disappears.
411: As shown in Figure~\ref{fig:model}, the VLT
412: and {\it HST} photometric points intersect the corresponding tracks at
413: about the same value of the IMBH mass. This value increases with increasing inclination angle.
414: Results shown in Figure~\ref{fig:model} refer to orbital phase zero (superior conjunction).
415: The variation in the V (B) band between
416: the 1st {\it HST} epoch and the VLT one is $\simeq 0.23$ ($\simeq 0.25$),
417: consistent (within the errors) with what observed. Thus, although
418: in these systems X-ray irradiation is very intense, the induced optical variability
419: is not very large owing to the high intrinsic emission of the massive B donor.
420: The calculation for phases 0.25 and 0.5 gives results similar
421: to those obtained for phase 0, typically within 0.15 mag. Thus, this is the
422: expected amplitude of the modulation possibly induced by the orbital motion.
423: It is interesting to note that this result is consistent with the degree
424: of variability observed in the V band between the two {\it HST} observations
425: ($\sim 0.1$ mag; see Figure \ref{fig:x2cl}).
426:
427: A direct comparison of the three cases illustrated in Figure~\ref{fig:model}
428: shows that relatively large values of the inclination angle ($i\gtrsim
429: 50^\circ-60^\circ$) are required in order to obtain the correct optical flux
430: for a BH mass $M_{BH}\sim 120 M_\odot$. At lower inclination angles (top and
431: middle panels of Figure~\ref{fig:model}), the black hole masses needed to
432: reproduce the optical magnitudes and color of object C1 are too small for the
433: X-ray flux to be below the Eddington limit (if the emission is isotropic).
434: Therefore, unless the Eddington limit can be circumvented, the binary system is
435: expected to have a significant inclination angle. We note that, for any value
436: of $i$, the BH mass inferred from Figure~\ref{fig:model} is always $\gtrsim 70
437: M_{\odot}$. This absolute lower limits for $M_{BH}$ is still compatible with a
438: very small beaming ($\gtrsim 0.6$). In this case the general picture
439: discussed above reasonably continues to hold. On the other hand, if the ULX
440: emission is more seriously beamed, the BH mass could be smaller: for a beaming
441: factor $\sim$ 1/6, $M_{BH}$ can be as small as $20 M_\odot$ without exceeding
442: the Eddington limit. In this case, we expect no X-ray irradiation of both the
443: disk and the companion, being the emission collimated away from the orbital
444: plane. To test this possibility we computed a new sequence of models, following
445: the same procedure outlined above, but switching off the disk/donor
446: irradiation. It turns out that it is possible to reproduce the correct
447: magnitude and color, although the donor is now less massive and cooler.
448: However, this implies that the star is too small to fill its Roche lobe and
449: thus accretion can not proceed through Roche lobe overflow. Wind accretion may
450: still be possible, although it seems unlikely that it can produce the required
451: value of $\dot M$.
452:
453: The situation for object C2 is reversed. We explored the
454: parameter space by varying the donor mass and orbital period, but did
455: not find any combination of values which could reproduce the data in the framework
456: of isotropic emission. In particular, X-ray irradiation causes the (B-V) color to always
457: exceed the observed one. On the other hand, a massive and very extended K-type supergiant
458: ($M \sim 16 M_\odot$, $T_{eff}\sim 4000$~K, $P_{orb}\sim 800$ days) would have
459: properties consistent with those of object C2 if the black hole mass is $\sim 20
460: M_\odot$. This of course requires a (moderate) beaming. We checked that a beaming
461: factor of $\sim 1/6$ is enough and that the companion fills its Roche lobe.
462: The optical magnitudes are correctly reproduced
463: because the (unirradiated) disk contribution becomes negligible in comparison with
464: the star intrinsic luminosity. However, in this case practically no variation in the optical
465: is expected in response to an increase of the accretion rate, and the predicted magnitudes
466: of C2 are constant. This is in contrast with the evidence of variability
467: observed in the V band between the two {\it HST} observations ($\sim 0.1$ mag;
468: see Section \ref{optical} and Figure \ref{fig:x2cl}), although some variations may be induced also by the
469: orbital ellipsoidal modulation of the donor (which we did not take into account).
470: %However, the possibility that
471: %C2 is the companion star poses also some theoretical problems. In fact,
472: %unless the orbital separation is exactly what is needed to
473: %reach contact with the Roche lobe during the supergiant phase or the
474: %system is caught in a very short ascending phase to the giant branch,
475: %the mass transfer from such a massive, supergiant star is huge. Accretion
476: %at such large (super-Eddington) rates would lead to a transient behavior
477: %with peak luminosities in excess of $10^{41}$ ergs$^{-1}$, inconsistent with
478: %observations \citep{pz06}.
479: %However, the possibility that C2 is the companion star poses also some
480: %theoretical problems. Evolutionary models of IMBH binary systems with X-ray
481: %irradiation show that the most likely counterpart is likely to be object C1
482: %\citep{pz06}.
483:
484:
485: %************************************************
486:
487: \section{Discussion}\label{concl}
488:
489: Although present data do not allow to reach a definite conclusion on the actual
490: counterpart of the ULX NCG 1313 X-2, some firm points may be derived from
491: the analysis presented in the preceeding sections. If C1 is the counterpart, as
492: it seems more likely, our model indicates that this is
493: an IMBH X-ray binary with a relatively massive main sequence donor which
494: fills its Roche lobe. Taking a
495: black hole mass of $\sim 120 M_\odot$, as required to account for the observed X-ray flux
496: in terms of isotropic emission, the donor mass is in the interval $10-18 M_\odot$
497: (taking photometric uncertainties into account, \S~\ref{application}). This is
498: larger than the maximum main sequence mass of the parent stellar association,
499: $\sim 8-9 M_\odot$, estimated using multicolor photometry and isochrone fitting by
500: \cite{pak05} and \cite{ramsey06}.
501: %We note however, that this discrepancy sensibly reduces if one takes a donor
502: %mass close to the lower limit, $\sim 10 M_\odot$, allowed by present
503: %uncertainties on the photometry (\S~\ref{application}).
504: However, considering that the lower bound for the donor mass is $10
505: M_\odot$, the difference is small. We note also that, if C1 is the counterpart
506: and C2 belongs to the same stellar association, the estimated masses of the two
507: stars correctly places them on (or close to) and out of the main sequence,
508: respectively.
509: If the counterpart is C2 then the source is a binary formed by a late
510: type, massive supergiant and a stellar mass black hole with beamed X-ray emission.
511: However, this scenario has some shortcomings. First, it predicts little if no optical
512: variability, and this is in apparent contrast with the variations seen in
513: the two {\it HST} observations.
514: %We caution, however, that more accurate photometry
515: %needs to be performed in order to confirm the optical variability.
516: Second, the duration of the supergiant phase for a $\sim 15 M_\odot$ star is
517: very short (a few$\times 10^5 \, {\rm yr}$), making the possiblity of catching
518: the binary is such an evolutionary stage not very likely \citep{pz06}.
519: Third, the accurate astrometry of the field made possible by the cross-identification on the {\it
520: Chandra} and {\it HST} images of a background galaxy appears to strengthen the association of NGC 1313
521: X-2 with object C1, ruling out object C2 (J.-F. Liu, private communication).
522:
523: For object C1, an orbital modulation of amplitude $\Delta V\sim 0.15$ is expected
524: because of orbital inclination and X-ray irradiation effects. This modulation is superimposed
525: to a comparable variation caused by changes in the irradiating X-ray flux
526: ($\sim 0.2$ mag). In this respect, it is interesting to note that similar variations
527: in the observed B band VLT+Subaru photometry of object C1 have been recently reported also
528: by \cite{pak05}, consistent with our findings. In principle, with a sufficient and
529: suitably spaced number of observations,
530: the orbital modulation can be singled out and measured with large area ground telescopes or
531: {\it HST}. The detection of this modulation would lead
532: to the unambiguous determination of the orbital period of the binary. This, in turn,
533: would allow us to definitely discriminate between C1 and C2 and, most importantly, to
534: constrain the mass ratio of NGC 1313 X-2 and, eventually, the mass of the black hole.
535:
536:
537: \acknowledgements
538: We are grateful to Claudio German\`a for his help with the binary X-ray
539: irradiation code. We also thank an anonymous referee for useful comments that improved a
540: previous version of this paper. We acknowledge financial contribution from contract
541: ASI-INAF I/023/05/0 and MURST under grant PRIN-2004-023189. This paper is based on
542: observations obtained with {\it XMM-Newton}, an ESA science mission with instruments and
543: contributions directly funded by ESA Member States and NASA and on observations made with
544: the NASA/ESA Hubble Space Telescope, obtained from the data archive at the Space Telescope
545: Institute. STScI is operated by the association of Universities for Research in Astronomy,
546: Inc. under the NASA contract NAS 5-26555.
547:
548: \begin{thebibliography}{}
549:
550: \bibitem[\protect\citeauthoryear{Cardelli et al.}{1989}]{card89} Cardelli, J.A., Clayton, G.C.
551: \& Mathis, J.S. 1989, \apj, 345, 245
552: %\bibitem[\protect\citeauthoryear{Chakrabarty}{1998}]{cha98}
553: %Chakrabarty, D. 1998, \apj, 492, 342
554: \bibitem[\protect\citeauthoryear{Colbert \& Mushotzky}{1999}]{cm99}
555: Colbert, E.J.M., \& Mushotzky, R.F. 1999, \apj, 519, 89
556: \bibitem[\protect\citeauthoryear{Colbert \& Ptak}{2002}]{col02} Colbert, E.J.M.,
557: \& Ptak, A.F. 2002, \apjs, 143, 25
558: \bibitem[\protect\citeauthoryear{Copperwheat et
559: al.}{2005}]{cop06}Copperwheat, C., Cropper, M., Soria, R., \& Wu,
560: K. 2005, \mnras, 362, 79
561: \bibitem[\protect\citeauthoryear{Cropper et al.}{2004}]{cr04}
562: Cropper, M. et al. 2004, \mnras, 349, 39
563: \bibitem[\protect\citeauthoryear{de Jong, van Paradijs \& Augusteijn}{1996}]{jpa96}
564: de Jong, J.A., van Paradijs, J., \& Augusteijn, T. 1996, \aap, 314, 484
565: \bibitem[\protect\citeauthoryear{Dewangan et al.}{2006}]{dew06}
566: Dewangan, G.C., Titarchuk, L., \& Griffiths, R.E. 2006, \apjl, 637, L21
567: \bibitem[\protect\citeauthoryear{Ebisawa et al.}{2003}]{ebi03}
568: Ebisawa, K. et al. 2003, \apj, 597, 780
569: \bibitem[\protect\citeauthoryear{Fabbiano}{1989}]{fab89} Fabbiano, G. 1989, \araa, 27, 87
570: \bibitem[\protect\citeauthoryear{Fabbiano et al.}{2003}]{fab03}
571: Fabbiano, G. et al., 2003, \apj, 584, 5
572: \bibitem[\protect\citeauthoryear{Feng \& Kaaret}{2005}]{fk05}
573: Feng, H., \& Kaaret, P. 2005, \apj, 633, 1052
574: \bibitem[\protect\citeauthoryear{Fiorito \& Titarchuk}{2004}]{ft04}
575: Fiorito, R., \& Titarchuk, L., 2004, \apjl, 614, L113
576: \bibitem[\protect\citeauthoryear{Foschini et al.}{2002a}]{fosc02a} Foschini, L.
577: et al. 2002a, \aap, 392, 817
578: \bibitem[\protect\citeauthoryear{Foschini et al.}{2002b}]{fosc02} Foschini, L.,
579: Ho, L.C. \& Masetti, N. 2002b, \aap, 396, 787
580: \bibitem[\protect\citeauthoryear{Frank, King \& Raine}{2002}]{fkr02}
581: Frank, J., King, A., \& Raine, D.J. 2002, Accretion Power in Astrophysics
582: (Cambridge, Cambridge University Press)
583: \bibitem[\protect\citeauthoryear{Georganopoulos et al.}{2002}]{gak02}
584: Georganopoulos, M., Aharonian, F.A., \& Kirk, J.G. 2002, \aap, 388, L25
585: \bibitem[\protect\citeauthoryear{Goad et al.}{2002}]{goad02}
586: Goad, M.R. et al. 2002, \mnras, 335, L67
587: \bibitem[\protect\citeauthoryear{Kaaret et al.}{2003}]{kaa03}
588: Kaaret, P., Corbel, S., Prestwich, A.H., \& Zezas, A. 2003, Science, 299, 365
589: \bibitem[\protect\citeauthoryear{Kaaret et al.}{2004}]{kaa04}
590: Kaaret, P., Ward, M.J., \& Zezas, A. 2004, \mnras, 351, 83
591: \bibitem[\protect\citeauthoryear{Kaaret}{2005}]{kaa05}
592: Kaaret, P. 2005, \apj, 629, 233
593: \bibitem[\protect\citeauthoryear{Kaaret, Simet \& Lang}{2006a}]{ksm06a}
594: Kaaret, P., Simet, M.G., \& Lang, C.C. 2006a, Science, 311, 491
595: \bibitem[\protect\citeauthoryear{Kaaret, Simet \& Lang}{2006b}]{ksm06b}
596: Kaaret, P., Simet, M.G., \& Lang, C.C. 2006b, \apj, 646, 174
597: \bibitem[\protect\citeauthoryear{Kawaguchi}{2003}]{kaw03}
598: Kawaguchi, T. 2003, \apj, 593, 69
599: \bibitem[\protect\citeauthoryear{King et al.}{2001}]{kin01} King, A.R.
600: et al. 2001, \apjl, 552, 109
601: \bibitem[\protect\citeauthoryear{King}{2002}]{kin02}
602: King, A.R. 2002, \mnras, 335, L13
603: \bibitem[\protect\citeauthoryear{King \& Pounds}{2003}]{kp03}
604: King, A.R., \& Pounds, K.A. 2003, \mnras, 345, 657
605: \bibitem[\protect\citeauthoryear{Kong et al.}{2005}]{kong05}
606: Kong, A.K.H., Rupen, M.P., Sjouwerman, L.O. \& Di Stefano, R. 2005,
607: in Proceedings of the XXII Texas Symposium on Relativistic Astrophysics, Stanford University,
608: in press (astro-ph/0503465)
609: \bibitem[\protect\citeauthoryear{K\"ording et al.}{2002}]{kfm02}
610: K\"ording, E., Falcke, H. \& Markoff, S. 2002, \aap, 382, L13
611: \bibitem[\protect\citeauthoryear{La Parola et al.}{2001}]{lapar01}
612: La Parola, V. et al. 2001, \apj, 556, 47
613: \bibitem[\protect\citeauthoryear{Liu et al}{2002}]{liu02} Liu, J., Bregman,
614: J.N., \& Seitzer, P. 2002, \apjl, 580, 31
615: \bibitem[\protect\citeauthoryear{Liu et al}{2004}]{liu04} Liu, J., Bregman,
616: J.N., \& Seitzer, P. 2004, \apj, 602, 249
617: \bibitem[\protect\citeauthoryear{Liu \& Bregman}{2005}]{liu05} Liu, J., \& Bregman,
618: J.N. 2005, \apjs, 157, 59
619: \bibitem[\protect\citeauthoryear{Makishima et al.}{2004}]{mak04} Makishima, K.
620: et al. 2004, AAS, 204
621: \bibitem[\protect\citeauthoryear{Masetti et al.}{2003}]{mas03}
622: Masetti, N. et al. 2003, \aap, 406, L27
623: \bibitem[\protect\citeauthoryear{Miller et al.}{2003}]{mi03}
624: Miller, J.M., Fabbiano, G., Miller, M.C. \& Fabian, A.C. 2003, \apj, 585, L37
625: \bibitem[\protect\citeauthoryear{Miller et al.}{2004}]{mi04}
626: Miller, J.M., Fabian, A.C. \& Miller, M.C. 2004, \apj, 607, 931
627: \bibitem[\protect\citeauthoryear{Miller et al.}{2005}]{miln05}
628: Miller, N.A., Mushotzky, R.F., \& Neff, S.G. 2005, \apjl, 623, L109
629: \bibitem[\protect\citeauthoryear{Mucciarelli et al.}{2005}]{muc05}
630: Mucciarelli P. et al. 2005, \apjl, 633, 101 (M05)
631: \bibitem[\protect\citeauthoryear{Mucciarelli et al.}{2006}]{muc06}
632: Mucciarelli, P. et al. 2006, \mnras, 365, 1123
633: \bibitem[\protect\citeauthoryear{Pakull \& Mirioni}{2002}]{pak02} Pakull, M.W.,
634: \& Mirioni, L. 2002, in Proc. ESA Symp., New Visions of the X-ray
635: Universe in the {\it XMM-Newton\/} and {\it Chandra\/} Era, eds.
636: F. Jansen et al. (ESA SP-488) (astro-ph/0202488)
637: \bibitem[\protect\citeauthoryear{Pakull, Gris\'e \& Motch}{2006}]{pak05}
638: Pakull, M.W., Gris\'e, F., \& Motch, C. 2006, IAUS, 230, 293
639: \bibitem[\protect\citeauthoryear{Patruno et al.}{2005}]{pat05}
640: Patruno, A., Colpi, M., Faulkner, A., Possenti, A. 2005, \mnras, 364, 344
641: \bibitem[\protect\citeauthoryear{Patruno et al.}{2006}]{pat06}
642: Patruno, A., Portegies Zwart, S., Dewi, J. \& Hopman, C. 2006, \mnras, 370. L6
643: \bibitem[\protect\citeauthoryear{Patruno \& Zampieri}{2006}]{pz06}
644: Patruno, A., \& Zampieri, L. 2006, MNRAS, submitted
645: \bibitem[\protect\citeauthoryear{Ramsey et al.}{2006}]{ramsey06}
646: Ramsey, C.J. et al. 2006, \apj, 641, 241
647: \bibitem[\protect\citeauthoryear{Roberts \& Warwick}{2000}]{rob00}
648: Roberts, T.P., \& Warwick, R.S. 2000, \mnras, 315, 98
649: \bibitem[\protect\citeauthoryear{Roberts et al.}{2001}]{rob01}
650: Roberts, T.P. et al. 2001, \mnras, 325, L7
651: \bibitem[\protect\citeauthoryear{Sirianni et al.}{2005}]{sir05}
652: Sirianni, M. et al. 2005, \pasp, 117, 1049
653: \bibitem[\protect\citeauthoryear{Soria et al.}{2005}]{sor05}
654: Soria, R. et al. 2005, \mnras, 356, 12
655: \bibitem[\protect\citeauthoryear{Stobbart et al.}{2006}]{stob06}
656: Stobbart, A.M., Roberts, T.P., Wilms, J. 2006, \mnras, 368, 397
657: \bibitem[\protect\citeauthoryear{Strohmayer \& Mushotzky}{2003}]{sm03}
658: Strohmayer, T.E., Mushotzky, R.F., 2003, \apjl, 586, L61
659: \bibitem[\protect\citeauthoryear{Swartz et al.}{2004}]{swa04} Swartz, D.A.,
660: Ghosh, K.K., Tennant, A.F., \& Wu, K. 2004, \apjs, 154, 519
661: \bibitem[\protect\citeauthoryear{Tully}{1988}]{tul88} Tully, R.B. 1988, Nearby
662: Galaxies Catalog (Cambridge: Cambridge University Press)
663: \bibitem[\protect\citeauthoryear{Turolla et al.}{2006}]{tur04}
664: Turolla, R. et al. 2006, in AdSpR, 38, 1374
665: \bibitem[\protect\citeauthoryear{Winter et al.}{2005}]{win05} Winter, L.M.,
666: R.F. Mushotzky and C.S.Reynolds Collaboration, 2005, AAS, 207, 90.05
667: \bibitem[\protect\citeauthoryear{Wu et al.}{2001}]{wu01}
668: Wu, K., Soria, R., Hunstead, R.W., \& Johnston, H.M. 2001, MNRAS, 320, 177
669: \bibitem[\protect\citeauthoryear{Zampieri et al.}{2004}]{zamp04}
670: Zampieri, L. et al. 2004, \apj, 603, 523 (Z04)
671: \end{thebibliography}
672:
673: \clearpage
674:
675: \begin{deluxetable}{llccrrc}
676: \tablecolumns{7}
677: \tabletypesize{\scriptsize}
678: \tablecaption{Observation log of the {\it XMM-Newton} EPIC pn pointings and of the
679: VLT+FORS1 and {\it HST}+ACS photometric observations of NGC 1313
680: X-2.\label{tab:olxmm}}
681: \tablewidth{0pt}
682: \tablehead{ &\colhead{Instrument} & \colhead{Obs. Id.} & \colhead{Date}
683: &
684: \colhead{Exposure} & \colhead{GTI\tablenotemark{a}} & \colhead{Filter}
685: }
686: \startdata
687: 1 & {\it XMM-Newton} EPIC pn & 0106860101 & 2000-10-17 & 31637s & 20600s & Medium \\
688: 2 & {\it XMM-Newton} EPIC pn & 0150280101 & 2003-11-25 & 8365s & 1087s & Thin
689: \\
690: 3 & {\it XMM-Newton} EPIC pn & 0150280201 & 2003-12-09 & 5620s & 0s & Thin
691: \\
692: 4 & {\it XMM-Newton} EPIC pn & 0150280301 & 2003-12-21 & 10334s & 8272s & Thin
693: \\
694: 5 & {\it XMM-Newton} EPIC pn & 0150280401 & 2003-12-23 & 14094s & 3600s & Thin
695: \\
696: 6 & {\it XMM-Newton} EPIC pn & 0150280501 & 2003-12-25 & 15282s & 1668s & Thin
697: \\
698: 7 & {\it XMM-Newton} EPIC pn & 0150280701 & 2003-12-27 & 16666s & 0s & Thin
699: \\
700: 8 & {\it XMM-Newton} EPIC pn & 0150280601 & 2004-01-08 & 14756s & 7696s & Thin
701: \\
702: 9 & {\it XMM-Newton} EPIC pn & 0150281101 & 2004-01-16 & 7034s & 4208s & Thin
703: \\
704: & VLT+FORS1 & & 2003-12-24 &840$\times$2 & &B
705: \\
706: & VLT+FORS1 & & 2003-12-25 &600$\times$2 & &V
707: \\
708: & {\it HST}+ACS(I epoch) & & 2003-11-22 &580$\times$2 & &F555w \\
709: & {\it HST}+ACS & & 2003-11-22 &630$\times$4 & &F435w
710: \\
711: & {\it HST}+ACS(II epoch) & & 2004-02-22 &600$\times$4 &
712: &F555w \\
713: \enddata
714: \tablenotetext{a}{Good Time Intervals in which the total off-source count rate above 10 keV is $<$ 1.0 counts s$^{-1}$.}
715: \end{deluxetable}
716:
717:
718: \begin{deluxetable}{ccccccccc}
719: \tablecolumns{9}
720: \tabletypesize{\scriptsize}
721: \tablecaption{Spectral analysis of {\it XMM-Newton} EPIC pn data of NGC 1313
722: X-2.\tablenotemark{a}
723: \label{tab:x2fit}}
724: \tablewidth{0pt}
725: \tablehead{
726: & \colhead{Date} & \colhead{Count rate} & \colhead{Flux\tablenotemark{b}}
727: &
728: \colhead{$F_{MCD}/F_{PL}$\tablenotemark{c}} &
729: \colhead{$L$\tablenotemark{d}} & \colhead{$kT_{MCD}$~[keV]} &
730: \colhead{$\Gamma$} & \colhead{$\chi^{2}_{red}/d.o.f.$}
731: }
732: \startdata
733: 1 & 2000-10-17 & 0.24 & 4.00$^{+0.82}_{-0.62}$ &0.93 &
734: 6.52$^{+1.34}_{-1.02}$ &
735: 0.16$^{+0.02}_{-0.01}$ & 2.3$^{+0.1}_{-0.1}$ & 1.24/75
736: \\
737: 2 & 2003-11-25 & 0.67 & 5.07$^{+0.28}_{-0.18}$ &-- &
738: 8.30$^{+0.04}_{-22.0}$
739: & -- & 2.3$^{+0.2}_{-0.2}$ & 1.06/44
740: \\
741: 4 & 2003-12-21 & 0.80 & 9.46$^{+1.55}_{-1.65}$ &0.63 &
742: 15.43$^{+2.53}_{-2.69}$ &
743: 0.13$^{+0.01}_{-0.02}$ & 1.9$^{+0.1}_{-0.1}$ & 1.11/144
744: \\
745: 5 & 2003-12-23 & 0.89 & 9.19$^{+3.10}_{-1.74}$ &0.94 &
746: 14.99$^{+5.06}_{-2.84}$
747: & 0.15$^{+0.03}_{-0.03}$ & 1.8$^{+0.1}_{-0.1}$ & 0.99/140
748: \\
749: 6 & 2003-12-25 & 0.56 & 10.0$^{+8.80}_{-3.85}$ &0.87 &
750: 16.31$^{+14.36}_{-6.28}$ &
751: 0.15$^{+0.04}_{-0.04}$ & 2.1$^{+0.2}_{-0.2}$ & 0.99/42
752: \\
753: 8 & 2004-01-08 & 0.39 & 5.63$^{+2.97}_{-1.00}$ &0.47 &
754: 9.18$^{+4.84}_{-1.63}$ &
755: 0.13$^{+0.03}_{-0.02}$ & 2.5$^{+0.1}_{-0.1}$ & 0.82/67
756: \\
757: 9 & 2004-01-16 & 0.34 & 4.38$^{+1.01}_{-1.12}$ &0.45 &
758: 7.14$^{+1.65}_{-1.83}$ &
759: 0.17$^{+0.02}_{-0.03}$ & 2.3$^{+0.1}_{-0.1}$ & 1.01/66
760: \\
761: \enddata
762: \tablenotetext{a}{$N_{H}$ has been frozen at $4.02 \times
763: 10^{21}$cm$^{-2}$}
764: \tablenotetext{b}{Unabsorbed flux in units of $10^{-12}$ erg cm$^{-2}$
765: s$^{-1}$}
766: \tablenotetext{c}{Ratio of the (unabsorbed) fluxes of the MCD
767: and PL components.}
768: \tablenotetext{d}{Unabsorbed luminosity, in units of $10^{39}$ erg s$^{-1}$,
769: for a distance of 3.7 Mpc}
770: \end{deluxetable}
771:
772:
773:
774: \begin{deluxetable}{lccc}
775: \tablecolumns{4}
776: \tabletypesize{\scriptsize}
777: \tablecaption{Observed magnitudes and colors of the two candidate optical
778: counterparts of NGC 1313 X-2.\label{tab:lmag}}
779: \tablewidth{0pt}
780: \tablehead{
781: \colhead{ } &\colhead{Filter} &\colhead{C1}& \colhead{C2}
782: }
783: \startdata
784: {\it HST} I epoch & B & 23.72$\pm$0.04 & 26.02$\pm$0.04 \\
785: VLT & B & 23.50$\pm$0.15 & $\geq$ 25.2\\
786: {\it HST} I epoch & V & 23.75$\pm$0.04 & 24.46$\pm$0.04\\
787: VLT & V & 23.60$\pm$0.15 & 24.10$\pm$0.15\\
788: {\it HST} II epoch& V & 23.61$\pm$0.04 & 24.57$\pm$0.04\\
789: {\it HST} I epoch & B-V & -0.03$\pm$0.06 & 1.56$\pm$0.06 \\
790: VLT & B-V & -0.1$\pm$0.2 & $\geq$ 1.1\\
791: \enddata
792: \end{deluxetable}
793:
794: \clearpage
795:
796: \begin{figure}
797: \plottwo{f1a.ps}{f1b.ps}
798: \caption{{\it HST}+ACS images of NGC 1313 X-2: left F435W (B) band, right
799: F555W (V) band. The {\it Chandra} error box, the candidate optical counterparts C1 and C2 and the field sources A
800: and D are shown (following Z04 and M05).
801: \label{fig:bv}}
802: \end{figure}
803:
804: \begin{figure}
805: \epsscale{.70}
806: \plotone{f2_color.eps}
807: \caption{{\it Top and middle panels}: $\Delta V=V-V(HST\; \rm{I\; epoch})$ and $\Delta
808: B=B-B(HST\; \rm{I \;epoch})$
809: for objects C1 and C2 for the available epochs. The error bars
810: correspond to 0.3 and 0.5 mag for V and B respectively (see text for
811: details). {\it Lower
812: panel}: {\it XMM-Newton} count rates of NGC 1313 X-2 in the
813: [0.2-10.0] keV range. \label{fig:x2cl}}
814: \end{figure}
815:
816: \begin{figure}
817: \begin{center}
818: \includegraphics[angle=-90,scale=.50]{f3_color.eps}
819: \end{center}
820: \caption{The X-ray lightcurve of NGC 1313 X-2. Fluxes are unabsorbed and refer to the
821: [0.2-10.0] keV energy interval (see Table \ref{tab:x2fit}). {\it
822: Einstein}, {\it ROSAT} and {\it ASCA} points are taken from Z04. The insert refers
823: to the more recent {\it XMM-Newton} data (Table \ref{tab:x2fit})
824: and the arrows mark the time of the {\it HST} and VLT observations (Table \ref{tab:olxmm}).
825: \label{fig:x2fl}}
826: \end{figure}
827:
828:
829: \begin{figure}
830: \epsscale{.8}
831: \plotone{f4_color.eps}
832: \caption{Color-magnitude diagram for the (irradiated) disk plus donor model for
833: $P_{orb}\simeq 1.7$~d, $M\simeq 15 M_\odot$ and $T_{eff}\simeq
834: 25000$~K (object C1). Each panel refers to a different inclination
835: angle $i$. The two tracks correspond to $\dot M=3.5$ and $6.7\, \dot
836: M_{Edd}$. These values are chosen in such a way to match the {\it XMM-Newton} flux
837: measured in the two observations of 2003 November 25 and December
838: 23, that are quasi-simultaneous to the {\it HST} I
839: epoch and VLT observations, respectively. The labels indicate the BH mass.
840: The V magnitude and
841: B-V color as obtained from the VLT and the {\it HST} observations are also
842: shown ({\it open squares}).\label{fig:model}}
843: \end{figure}
844:
845:
846: \end{document}
847:
848:
849: