astro-ph0409416/ms.tex
1: %\documentclass{aastex}
2: \documentclass[12pt,preprint]{aastex}
3: \slugcomment{Accepted for publication 
4: in {\it the Astrophysical Journal Letters}  and scheduled for v. 614, 2004, October 20 issue} 
5: \def\lax {\ifmmode{_<\atop^{\sim}}\else{${_<\atop^{\sim}}$}\fi} 
6: \def\gax {\ifmmode{_>\atop^{\sim}}\else{${_>\atop^{\sim}}$}\fi} 
7: \def\gtorder{\mathrel{\raise.3ex\hbox{$>$}\mkern-14mu
8:              \lower0.6ex\hbox{$\sim$}}}
9: \begin{document}
10: 
11: \title{Is M82 X-1 Really An Intermediate-Mass Black Hole? 
12: X-ray Spectral and Timing  Evidence}
13: 
14: 
15: \author{Ralph  Fiorito\altaffilmark{1} and Lev Titarchuk\altaffilmark{2}}
16: 
17: \altaffiltext{1}{University of Maryland, College
18: Park and NASA/GSFC, Greenbelt MD 20771; rfiorito@milkyway.gsfc.nasa.gov; 
19: rfiorito@UMD.edu}
20: \altaffiltext{2}{George Mason University/CEOSR,
21: %Center for Earth
22: %Observing and Space Research, 
23: Fairfax, VA 22030 and US Naval Research
24: Laboratory, Code 7655, Washington, DC 20375; ltitarchuk@ssd5.nrl.navy.mil and 
25: lev@lheapop.gsfc.nasa.gov}
26: %\altaffiltext{3}{NASA/ Goddard Space Flight Center, code 660, Greenbelt 
27: %MD 20771,  rfiorito@milkyway.gsfc.nasa.gov; lev@lheapop.gsfc.nasa.gov}
28: 
29: 
30: 
31: \begin{abstract}
32: 
33: Ultra-luminous X-ray sources (ULXs) with apparent luminosities
34: up to 100's of times  the Eddington luminosity for a neutron star have been discovered
35: in external galaxies. The existence of intermediate mass black holes has been
36: proposed to explain these sources. We present evidence for an
37: intermediate-mass black hole in the ULX M82 X-1 based on the spectral features and
38: timing (QPO) properties of the X-radiation from this source. We revisited XMM
39: Newton and RXTE data for M82 X-1 obtained in 2001 and 1997 for XMM and RXTE
40: respectively. We show for these observations that the source is either in
41: transition or in a high/soft state with  photon spectral indices 2.1 and 2.7
42: respectively. We confirm the early determination of the QPO frequency
43: $\nu\approx 55$ mHz  in this source by Strohmayer \& Mushotzky  and identify this as
44: the low frequency QPO for the source. We apply a new method to determine the BH
45: mass of M82 X-1. The method uses the index-QPO low frequency correlation that
46: has been recently been established in galactic black hole candidates GRS
47: 1915+105, XTE J1550-564, 4U 1630-47 and others. Using scaling arguments and
48: the correlation derived from consideration of galactic BHs, we conclude that
49: M82 X-1 is an intermediate BH with a mass of the order of 1000 solar masses.
50: 
51: % if 
52: % we will be far more precise in the determination of a
53: %black mass than previously published determinations.
54: \end{abstract}
55: \keywords{accretion, accretion disks---black hole physics---stars: radiation mechanisms: nonthermal---physical data and processes}
56: 
57: \section{Introduction}
58: A number of external galaxies, notably ones with active star formation
59: regions, have revealed the presence of so called ultra-luminous X-ray sources
60: (ULXs) which one can operationally define as an X-ray source, which is not
61: coincident with the nucleus of its host galaxy and which has an apparent
62: luminosity in excess of an order of magnitude above the Eddington luminosity
63: for a neutron star ($L_{X}\gax 10^{39}$ erg~s$^{-1}$). The high luminosity of ULXs
64: have led to speculation that these objects may be intermediate ($10^{2}-10^{4})M_{\odot}$ BHs 
65: [e.g. Colbert \& Mushotzky 1999; Strohmayer \& Mushotzky 2003, hereafter SM03; Shrader \&
66: Titarchuk 2003, hereafter ST03; Miller et al. 2003] or, more conservatively, beamed sources of lower mass
67: (King et al. 2001). However, due to their distances the
68: prospects for dynamical measurement of the mass of ULXs are poor and other
69: means must be sought to identify their masses.
70: 
71: %Shrader \& Titarchuk (2003), hereafter ST03, show (see their Tables 1 and 3)
72: %that the blackbody color temperatures in GBHs are an order magnitude higher
73: %than those seen in ULXs. One can speculate that that this ratio (10:1) can be
74: %translated to $1:10^{4}$ for the ratio of the mass of a GBH to that of an ULX
75: %(see Eq. 5 in Shrader \& Titarchuk 1999, hereafter ST99). But this estimate is
76: %valid only if one assumes the same mass accretion rate $\dot{m},$ measured in
77: %Eddington units, occuring in these sources (that may differ by order of
78: %magnitude depending on the source spectral state), the same inner disk radius
79: %$r_{in}$ (which can vary from one state to another and from one source to
80: %another source) and the same color factor $T_{h}$ (which ultimately depends on
81: %the viscosity in a particular source (see Titarchuk and Fiorito 2004,
82: %hereafter TF04).  
83:  ST03
84: %ST03 
85: present 
86: the analysis of observational data for
87: representative objects of several classes: GBHs, narrow-line Seyfert galaxies
88: (NLS1s) and ULXs. They apply a methodology, which uses the spectral
89: characteristics of the X-ray source radiation, to determine the mass. 
90: %This method
91: %relies on the successful fitting of the observed spectrum with a physically
92: %meaningful model disk-corona model, i.e. one which correctly accounts for  and
93: %includes the Comptonization of soft thermal disk photons by a corona of hot
94: %electrons surounding the black hole. The bulk motion Comptonization  (BMC) model  
95: %is one such generic model which has
96: %been used with success for such purposes and which gives a physically
97: %meaningful value for the color temperature of the disk and the spectral index of the
98: %power law component (see ST99).
99:  Using a color
100: temperature from the bulk motion Comptonization  (BMC) model  
101: %the BMC model 
102: %Shrader and Titarchuk (2003) 
103: ST03 calculate the mass of
104: several GBHs. Their results are in agreement with the masses obtained by
105: optical or other methods when available. 
106: %Using this method, 
107: ST03 also
108: calculate masses in ULXs for seven sources. The most compelling cases for the
109: presence of intermediate-mass objects are  NGC 1313 and NGC 5408 objects
110: for which $\log(M/M_{\odot})=2.2,~3.0$ respectively.
111: 
112: There is  suggestions in the literature (see  King et al. 2001) that the radiation
113: from some of ULXs sources is beamed and thus the real luminosity is much smaller
114: than the Eddington luminosity for a solar mass BH (or neutron star). 
115: In fact, King et al. do not exclude the possibility that  {\it individual} ULX's may contain 
116: extremely massive BH. But they emphasize that the formation of large number of ultramassive BHs is problematic in
117: terms of evolutionary scenario of  binary systems.
118: The recent discovery of the fast variability [quasiperiodic oscillation (QPO)] of the X-ray radiation in
119: ULX M82 by SM03,  suggests that  
120: the X-ray emission area is quite compact in this particular source.
121: %presents a strong
122: %argument  against beaming presumably ruling out the jet interpretation
123: %for the high luminosity. 
124: In fact, for a jet of a typical size about
125: $10^{13-14}$ cm one cannot expect the QPO frequency $\nu\simeq50$ mHz that is
126: observed by SM03 in M82 X-1. Thus, if the radiation from these variable sources
127: is isotropic, then the high inferred luminosity of M82-X1 requires an
128: intermediate-mass BH, i.e. a BH more massive than one that can be formed in
129: the collapse of a single normal star. 
130: 
131: If ULXs are indeed accreting BHs with properties presumably similar to GBHs,
132: similar observable properties, in particular, correlations of the X-ray spectral
133: indices and QPO frequencies in ULXs should be seen [see Vignarca, et.al.
134: (2003); Titarchuk and Fiorito (2004),
135: hereafter TF04; Titarchuk, Lapidus \& Muslimov (1998)].
136: %Such a correlation is predicted by the transition layer model 
137: %[see Titarchuk, Lapidus \& Muslimov (1998), for details of the model],
138: %which has been successfully applied to GBHs by TF04. 
139: %They identify the nature
140: %of low and high frequency QPOs seen in GBH sources and relate them to the size
141: %of a compact coronal region, which is primarily responsible for the
142: %Comptonization of soft disk photons,  the observed power law component and
143: %hence the particular spectral state (low/hard or high/soft) seen in the
144: %majority of GBH observations (cf. Vignarca, et. al. 2003 and Belloni 2003).
145: %In particuraly, the analysis of TF04 (1) identifies the low frequency QPO as the fundamental
146: %eigenfrequency of the corona, (2) predicts the strong observed correlation
147: %between the low frequency QPO and the spectral index, 
148: %(3) explains the
149: %spectral phase transition between low hard and high soft states observed in
150: %GBHs and the tracking of the low frequency QPO with spectral index with
151: %spectral state and 
152: %(3) shows how the measured values of the low frequency QPO
153: %and the spectral index can be used to measure the mass of a BH.
154: TF04 
155: %further 
156: show how the mass of one GBH source can be used to determine
157: that of another, e.g. GRS 1915+105 and XTE J1550-564 which exhibit remarkably
158: similar QPO frequency-index correlation curves (shown here as Fig. 1). 
159: %Note the
160: %saturation of the index $\Gamma$ at low ($\sim 1.7$) and high values
161: %($\sim2.7$). 
162: Once the mass of one object (e.g.
163: GRS 1915+105) is determined by a fit of theory to the measured QPO frequency-index
164: correlation curve, the mass of the other (XTE J1550-564) can be  found
165: by simply scaling, i.e. sliding the correlation curve for GRS 1915+105 along
166: the frequency axis until it coincides with the correlation curved of XTE
167: J1550-564, in this case by the factor 12/10. Thus the inferred mass of XTE
168: J1550-564 is 10/12 times less than that of GRS 1915+105. The mass
169: determination using the QPO frequency-index correlation fit is consistent with
170:  X-ray spectroscopic and dynamical mass determinations for these sources (see references in ST03).
171: %Borozdin et al. 1999, hereafter BRT99; Greiner et al. 2001; Orosz et al. 2002; ST03). 
172: %BRT99
173: %estimated the central object mass in GRS 1915+105, $m=M/M_{\odot}
174: %=(13.5\pm1.5)/(\cos{i})^{1/2}$ (where $\cos i$ is the cosine of the disk inclination angle), and recent dynamical measurements of Greiner
175: %et al. (2001) find a reduced mass of $14\pm4$ for the BH in GRS 1915+105,
176: %corroborating the BRT99 result. Applying the same X-ray spectroscopic method
177: %of BRT99  Titarchuk \& Shrader (2002)
178: %(see also ST03) obtain the BH mass of XTE J1550-564 as $9.4\pm2.1$ solar masses.
179: %Orosz et al. (2002) made a dynamical mass determination for the compact object
180: %in XTE J1550-564. These authors find a most probable BH mass of 10.6
181: %$M_{\odot}$ and an allowable range of 9.7-11.6 $M_{\odot}$. The consistency
182: %between these various measurements gives us great confidence that the
183: %QPO-index method  and the method of ST99 are generally applicable to other BH sources.
184: Since QPO frequencies are inversely proportional to the mass of the
185: central object one can, in principle, determine the mass of a ULX using the
186: same kind of scaling as described above for GBHs. 
187: %However, this method relies
188: %on  a accurate determination of the spectral index along with a
189: %measured detection and the identification of the low QPO frequency (see TF04). 
190: It is important to note that the BH mass determination method using the QPO-index correlation
191: is independent of the orientation of source.
192: % with respect to the Earth observer (of the  inclination angle).
193: %One can question, for example  that the photon index may have an orientation dependence.
194: In fact, the photon index of the Comptonization spectrum is a function of the Comptonization parameter only which
195:  is  a product of the  mean energy change at any   photon scattering and the mean number of
196: scatterings in the Compton cloud (corona) (see e.g. Sunyaev \& Titarchuk 1980).  
197: 
198: In \S 2 we present the results of our spectral and timing analysis of archival XMM and
199: RXTE data from the ULX source M82 X-1. In \S 3 we discuss our results of the
200: data, compare it to that presented by SM03 and employ our results to estimate
201: the mass of M82 X-1. Conclusions also follow in \S 3.
202: 
203: 
204: 
205: 
206: 
207: \section{DATA  ANALYSIS  and RESULTS}
208: 
209: We have revisited archival spectral and timing XMM and RXTE data from the ULX
210: source M82 X-1, which has been previously investigated by SM03  and Rephaeli and Gruber (2002)
211:  hereafter RG02.
212: We analysed four archival observations: (1) the XMM EPIC PN and MOS data from
213: the 30 ks observation May 5, 2001 (OBSIDS 0112290201) and (2) RXTE PCA data
214: from 1997 archival observations of RG03 \{OBSIDS: 20303-02-02(2.5 ks),
215: 20303-02-03(4.76ks) and 20303-02-04(2.5 ks)\}. The latter two observations
216: show the presence of QPOs and a power law (PL) component in their energy
217: spectra, which we have reanalyzed in terms of the transition layer model
218: previously applied by us to the study of galactic black hole sources (TF04).
219: We were not able to identify the QPO reported by SM03 for OBSID
220: 20303-02-02(2.5 ks).
221: 
222: 
223: \subsection{Energy spectra}
224: %\centerline{\it XMM Data}
225: %\subsubsection{XMM data}
226: {\it XMM data.--} We extracted the spectrum of M82 X-1 from both the EPIC MOS and PN images of M82
227: in a circle of 18$^{\prime\prime}$  in radius  around the bright source (M82 X-1), using the latest version of the SAS data reduction software
228: and response matrices. We focused our attention on the energy range 3.3
229: keV~-~10 keV where interference from the soft diffuse component surrounding
230: the bright source is minimized. We find that this energy range gives more
231: consistent results for the spectral modeling and timing analysis than that
232: used by SM03 (2-10 keV), i.e. increasing the lower energy value from 2 to 3.3 keV produced 
233: the lowest value of resuduals in our energy spectral fits ($\chi_{red}^{2}\leq2$), and had virtually no effect on 
234: the $\chi_{red}^{2}$ and parameter values of our fits to the power spectrum when compared those using the energy
235: range 2-10 keV.
236: Using XSPEC (version 11.3), we find that an absorbed
237: power law model with a Gaussian iron line component and the Bulk
238: Comptonization model (BMC) (Titarchuk et al. 1997) with a Gaussian iron line
239: component fixed at 6.5 keV and a line width of 0.4 keV, give equally good fits
240: ($\chi_{red}^{2}\leq1$), when the column density is left as a
241: free parameter. Since the real column density is unknown, because of the
242: obscuration effect of the surrounding diffuse emission surrounding M82 X-1, we
243: allow this parameter to be free. 
244: %We tested the evolution of the fit $\chi_{red}^{2}$
245: %for the BMC model parameter disk temperature (kT) as this parameter
246: %systematically varied from 0.1 to 0.5 keV. 
247: The $\chi_{red}^{2}$, and fitted
248: parameters showed little change with temperature up until  $kT=0.5$  keV. 
249: In order to obtain the
250: index, normalization and column density, 
251: we therefore fixed the value of kT
252: below this value, i.e. $kT=0.1$ keV. 
253: We found that for the best fit spectrum 
254:  the contribution of the Comptonized  component in the BMC spectrum is dominant. 
255: The BMC model spectrum  is a sum of
256: the (disk) black-body component and Comptonized black-body component, 
257: where $A/(1+A)$ is a relative weight of the Comptonized component. 
258: Because all values of parameter $A\gg 1$ are consistent with observations  we fixed $\log A$ to 5.
259:  %was also fixed to be much greater than 1, namely $\log A=5$, which is a reasonable assumption  for
260: %this quantity. 
261: With these constraints we find that the power law index and
262: column densities from the PL and BMC fits give close to the same results for
263: all the observations investigated by us. We also also noticed that an absorbed
264: multi-temperature thermal disk model (diskBB) with a Gaussian, provides a
265: statistically good fit but that a high disk temperature $(T_{i}\sim3$ keV) is
266: required in agreement with the results of SM03. An example of a fit to our
267: reduced PN data using the BMC-Gaussian Fe Line model is given in Figure 2. A
268: comparison of fits to our reduced data for the PL and BMC with that produced
269: by SM03 for the energy band (3.3-10 keV) produces nearly identical results.
270: Fits to the MOS data [similar data is mentioned but not presented in (SM03)]
271: gives close to the same fitted parameters as the PN data fits.
272: %\textbf{Figure D1. BMC fit to the XMM PN energy spectrum in the range 3.3 to
273: %10 keV.}
274: 
275: \smallskip
276: %\centerline{\it RXTE Data}
277: %\subsubsection{RXTE data}
278: {\it RXTE data.--}Spectra from M82 was also obtained from the archival observations of RG02
279: (using PCA Standard Products and the Epoch 3 response matrix).
280: A comparison of the free spectral fit parameters for the XMM EPIC PN, MOS and
281: RXTE PCA data is given in Table 1. Based on the photon indices obtained from
282: the XMM PN observation we conclude that the source was in a transition from
283: hard to soft state , i.e. characterized by an index between 1.5 and 2.5, for
284: the time period of the XMM observation and that for the period of the RXTE
285: data reported here the source was in a soft state (i.e spectral index $\sim2.5$).
286: 
287: \subsection{Timing Analysis}
288: 
289: %\centerline{\it XMM Data}
290: %\subsubsection{XMM}
291: {\it XMM data.}--We extracted a light curve from the EPIC\ PN instrument selected, from the same area A described in 
292: \S 2.1
293: %\S 2.1.1 
294: in the energy band 3.3-10 keV with a 1 sec time bin and one 30 ks window which includes all the PN data. 
295: From this light curve we computed the power
296: spectra using the current release of the XRONOS program. The power spectrum
297: was rebinned a factor of 256 to obtain the results shown in Figure 3 (left panel) which
298: exhibits a clear QPO.
299: % at about 50 mHz.
300: The high error in the red noise portion of the power spectra below 30mHz does
301: not allow a determination of the expected break frequency below the QPO frequency. Thus we
302: can not reliably identify the break frequency which is usually found with the
303: QPO low frequency in GBHs source from the XMM data (see Wijnands \& van der
304: Klis 1999). The data suggests but does not confirm the presence of a peak at approximately
305: 100mHz, but the statistics are too poor to positively identify this frequency.
306: (see also the XMM data of SM03). Because of the severity of the red noise below 30 mHz we have, following SM03 chosen
307: to fit the power spectrum with a model consisting of a constant to account for the Poisson noise, a simple power law
308: to account for the red noise and  a Lorentzian to account for the presence of the QPO. Our fit to the data using this
309: model gives  a good fit $\chi_{red}^{2}=47$ for 58 dof, $\nu_{QPO}=58.5\pm 1.7$ mHz, $\Delta \nu_{QPO}/2=5.8\pm 1.5$ mHz and 
310: $A_{QPO}=0.011\pm 0.003$ which are in good agreement with the more accurate results obtained by SM03 using all three EPIC
311: instruments. Here $A_{QPO}$ represents the total rms power in the QPO,
312: and $\Delta\nu_{QPO}$ is the FWHM of the QPO. We checked the significance of the QPO by observing that $\chi_{red}^{2}=70$ when the QPO was excluded from the
313: model. For this value of $\chi_{red}^{2}=70$ using the F test  one  finds that   a probability $\approx 3\times10^{-5}$ 
314: for a random occurance of the QPO feature. These
315: results give us great confidence that the QPO is observed by the PN instrument (cf. SMO3).
316: 
317: %\centerline{\it RXTE Data}
318: %\subsubsection{RXTE}
319: {\it RXTE data.--}We extracted a light curve using good time intervals and 3-10 keV energy
320: selected power density spectra from the three RXTE OBSIDs listed above,
321: employing a 128 Hz sampling rate and only the top xenon layers of the
322: operating PCU's for each observation time interval. The IDL program RADPS,
323: written by C. Markwardt was used to produce the power density spectrum (PDS) averaged over several 1024 second
324: intervals for each light curve. 
325: %Since the duration of the light curves were
326: %typically 3ks in contrast to 30ks for the XMM data, the statistics were poorer
327: %for the RXTE PDS. Nevertheless, 
328: We clearly identified single QPO's in the PDS
329: for two of the three OBSIDs studied by SM03. 
330: %Table 2 shows our results. 
331: We fit
332: our PDS data to a sum of two Lorentzian peaks, the lowest or zero order
333: Lorentzian was picked to identify the red noise break frequency of the PDS. A
334: clearly identifiable break frequency at $26\pm 2.5$ mHz with $A_{br}=4.8\pm 0.4$ confirmed by a plot of
335: frequency time power vs frequency, coincident with a QPO at $106\pm 2$ mHz and $A_{QPO}=0.095\pm 0.018$ was
336: observed for OBSID 20303-02-04. A QPO at  $48.9\pm 1$ mHz and $A_{QPO}=0.04\pm 0.013$ was also observed for OBSID 20303-02-03,  
337: but a clearly  definable  break frequency was not observed for this OBSID.
338: Lorentzian line fits to the PDS for the former case is presented in Figure 3 (right panel).
339: We also looked for variation of the total variability as a function of energy
340: by observing the variability in two energy bands: 3-6 keV and 6-10 keV, but no
341: variation was observed.
342: 
343: %Table 2 give the properties of the QPO's observed for the RXTE OBSIDs where
344: %they are clearly observed. $A$ is the total area under the corresponding
345: %Lorentzian. For example $A_{QPO}$ represents the total rms power in the QPO,
346: %and $\Delta\nu_{QPO}$ is the FWHM of the QPO.
347: 
348: 
349: \section{Discussion and Conclusions}
350: The spectral data for M82 X-1 definitely show the spectral features of
351: high/soft spectral phase when the photon indices $\Gamma$ are in the range of
352: $2-2.7$. For XMM PN OBSID and RXTE 20303-02-04 OBSID we find $\Gamma
353: =2.07\pm0.07$ and $\Gamma=2.67\pm0.1$ respectively; the former value is
354: characteristic of transition to the high/soft state, and the later to the
355: high/soft state. In these observations two QPO frequencies $\sim50$ mHz and
356: $\sim100$ mHz have been identified. 
357: %Because of it vicinity to the break
358: %frequency in the PDS, 
359: We interpret and identify the observed $\nu_{QPO}\sim50$
360: mHz as the fundamental low frequency and $\nu_{2QPO}\sim100$ mHz as its first
361: harmonic. We note that the presence of one predominant QPO, i.e. 50 or 100 mHz
362: in different observations is not an unusual occurrence (see e.g. TF04) and can
363: be explained as the result of the local driving frequency conditions in the
364: coronal region; i.e. a resonance condition is established for one
365: particular eigenmode of the compact coronal region so that this mode is
366: predominantly observed. In other cases, for example  in GRS 1915+105 
367: (Fiorito et al. 2003)   the fundamental and first harmonics of
368: the low frequency QPO (along with break frequencies) are simultaneously observed. Such features are seen
369: in a variety of other GBH sources as well. Also the proximity of the inferred fundamental $\nu_{QPO}\sim50$ mHz to the observed break frequency,
370: $\nu_b\sim26$ mHz is similar to what is observed in GBH's. 
371: 
372: %We obtain that the BH mass $M$ is order of $10^{3}$ solar masses in M82 X-1 
373: If we identify $\nu_{QPO}\sim 50$ mHz as the low frequency QPO frequency $\nu_{low}$ keeping in
374: mind that $\Gamma\sim2.7$ and $\nu_{low}\sim5$ Hz for 10 solar masses (see
375: Fig. 1, for the index-QPO frequency correlation in XTE 1550-564) and the fact
376: that $\nu_{low}$ is inversely proportional to $M$, we calculate $M\sim
377: 5~(\mathrm{Hz})/0.050~(\mathrm{Hz})\times10=10^{3}$ solar masses. As we show
378: above (see also TF04) this scaling have been observed for galactic black holes
379: but this is the first time it has been applied to a ULX source (M82 X-1) to
380: estimate the BH mass.
381: This value of BH mass is consistent with  mass evaluations obtained using
382: the absolute normalization and color temperature of other ULXs (NGC 253, NGC
383: 1399 X-2, X-4 and IC 342 X-1) 
384: %(see ST03).
385:  which have been analyzed by ST03.
386: We note the possibility that the 50 mHz QPO may not be the lowest intrinsic QPO frequency and that a lower one, which may be
387: obscured by the red noise at frequencies $\nu_{QPO}< 50$ mHz may exist. 
388: Therefore the frequency $\nu_{QPO}\approx 50$ should be interpreted 
389: as an upper limit and therefore the inferred mass as a lower limit.
390: 
391: %It is important to note that in order to be meaningful and useful the spectral
392: %index obtained in spectral fits must arise from a from physically realistic
393: %model. 
394: %As noted by Titarchuk \& Shrader (2002) a simple absorbed power law to
395: %model the spectrum from an accreting BH is physically unrealistic, since at
396: %low energies the presence of the disk and its soft photons (presumably a black
397: %body) is entirely neglected. The correct physical model must include this
398: %source of soft photons as well as the upscattered (Comptonized) photons.  The
399: %BMC model used here is a general model of Comptonization which is valid for
400: %thermal or bulk motion upscattering of disk photons. The free parameters of
401: %the model include the characteristic color temperature of the accretion disk,
402: %the spectral index, which is related to the Comptonization efficiency, the
403: %column density and the thermal component normalization. All of these
404: %parameters relate to physically measurable or estimable quantities. It is
405: % important to  note that the high energy tail predicted by
406: %this model is effectively the same as a purely absorbed power law.
407: % and
408: %therefore phenomenologically should produce the same index and column density
409: %as that of BMC. 
410: %Thus the simple PL model will presumably give  physically
411: %meaningful values for the spectral index and the column density for the high
412: %energy photons. In other words the BMC model is effectively a power law at
413: %high energies. Our fitted results verify that this is indeed the case for
414: %M82-X1. Furthermore, 
415: %In cases, 
416: %such as the present one, 
417: %where obscuration of the source does not allow a well defined value for BMC parameters such as the
418: %color temperature we must rely on estimated values for this quantity, which we
419: %have done by setting an upper bound of $kT=0.5$ keV, which is consistent with
420: %expected temperature of a 1000 solar mass BH accretion disk.
421: %On the other hand we cannot ascribe any physical meaning to any fitted
422: %parameters derived from a model obtained from a purely thermal model such as
423: %DISKBB which entirely neglects any Comptonization and is known to give
424: %unphysically high disk temperatures, cf.  $kT=3$ keV obtained by us and SM03 using
425: %this model.
426: %\section{Conclusions}
427: To conclude, we have presented a reanalysis and new interpretation of XMM Newton and RXTE
428: data obtained from M82 X-1. This analysis presents via the BMC model spectral
429: photon indices 2.1-2.7 which are seen in the high/soft states of extragalactic
430: and galactic black holes and identification of $\nu_{QPO}\approx 50$ mHz as
431: an observable upper limit on the  low frequency QPO for M82-X1. Using this value for the  low QPO frequency,
432: the predominantly observed spectral index $\sim 2.7$ and the index-QPO
433: frequency correlation recently obtained for GBHs establishes  a lower limit the
434: mass of M82 X-1 of the order 1000 solar masses.
435: The demonstrated application of our method  which uses the low QPO frequency and the index of the power-law component of the
436: spectra  presents a new,
437: potentially powerful tool for determing the nature and mass of  ULXs.
438: However, the confirmation of a ULX as an intermediate BH awaits either the
439: simultaneous application of the two independent methods we have described
440: above, i.e. the ST03 method and  QPO low frequency-index correlation or
441: direct dynamical evidence. 
442: 
443: 
444: R.F. gratefully acknowledges helpful discussions with T. Strohmayer, C.
445: Markwardt and L. Angelini.
446: % of the NASA Goddard Space Flight Center. 
447: L.T. acknowledges the support of this
448: work by the Center for Earth Observing and Space Research of George Mason University.
449: % George Mason
450: %University.
451: 
452:  
453: %We also acknowledge the thorough analysis 
454: %of this paper by the referee and his/her constructive and interesting suggestions.  
455:  
456: \begin{thebibliography}{}
457: %\bibitem[Akhiezer, et al. (1975)]{ash}
458: %Akhiezer, A.I., Akhiezer, A.I., Polovin, R.V., Sitenko, A.G., \&
459: % Stepanov, K.N. 1975, Plasma Electrodynamics (New York: Pergamon)
460: 
461: %\bibitem[Alfv\'en \& Falthammar (1963)]{afl63} Alfv\'en, H., \&
462: %Falthammar, C-G. 1963,Cosmical Electrodynamics.  Fundamental Principles (London: Qxford
463: %University Press)
464: 
465: %\bibitem[Alpar, \& Shaham, 1985]{ash}
466: %Alpar, M. \& Shaham, J. 1985, Nature, 316, 239
467: 
468: %\bibitem[Barret,  \&  Grindlay 1995]{bar}
469: %Barret, D., \&  Grindlay, J.E. 1999, ApJ, 440, 841  
470: 
471: %\bibitem[Belloni et al. (2002)]{BPK}
472: %Belloni, T., Psaltis, D., \& van der Klis, M. 2002, ApJ, 572, 392 (BPK)
473: 
474: %\bibitem[Bjerkness et al. (1933)]{bj}
475: %Bjerkness, V. Bjerkness, J., Solberberg, H. \& Bergron,
476: %T. 1933, {\it Physikalish Hydrodynamik}, Berlin: Springer
477: 
478: %\bibitemloniBoris, \& Book (1975)]{bb}
479: %Boris, J.P. \& Book, D.L. (1975) J. Comp. Phys., 11. 38
480: 
481: %\bibitem[Blandford \& Znajek (1977)]{bz}
482: %Blandford, R., \& Znajek, R.L. 1977, MNRAS, 179, 433
483: 
484: %\bibitem[Borozdin et al. (1999)]{bor}
485: %Borozdin, K., et al.
486: % Revnivtsev, M., Trudolyubov, S., Shrader, C, \& 
487: %Titarchuk, L.  
488: %1999, ApJ, 517, 367 (BRT99) 
489: 
490: %\bibitem[Chakrabarti (1990)]{ch61}
491: %Chakrabarti S. K. {\it Theory of transonic astrophysical flows}
492: %1990, World Scientific Publishing Co.
493: 
494: %\bibitem[Chandrasekhar (1961)]{c61} 
495: %Chandrasekhar, S. 1961, {\it Hydrodynamics and
496: %Hydromagnetic Stability}, Oxford: Oxford at the Caredon
497: %Press (C61)
498: 
499: %\bibitem[Cui et al. (2000)]{cui}
500: %Cui, W., Shrader, C., Haswell, C., \& Hynes, R. 2000, ApJL, 535, L123.
501: 
502: %\bibitem[Hide  (1956)]{hi}
503: %Hide, R. 1956,  Quart. J. Math. Appl. Math. 9, 35
504: 
505: 
506: %\bibitem[Burlaga 1995]{bur}
507: %Burlaga, L.F. 1995, Interplanetary Magnetohydrodynamics.
508: %(New York: Qxford University Press) 
509: 
510: %\bibitem[Chandrasekhar  (1956)]{chand}
511: %Chandrasekhar, S.  1956, Proc. Indian Acad. Sci. 42, 1 
512: 
513: 
514: 
515: %\bibitem[Chakrabarti \& Titarchuk (1995)]{ct95}
516: %Chakrabarti S.K. \& Titarchuk, L. G. 1995, ApJ,  455, 623 (CT95)
517: 
518: %\bibitem[Cheng et al. (1992)]{ebi}
519: %Cheng, K., et al. 1992, ApJ, 397, 664 
520: 
521: \bibitem[Colbert, \& Mushotzky, (1999)]{CM}
522: Colbert, E. J. M., \& Mushotzky, R. F. 1999, ApJ, 519, 89
523: 
524: %\bibitem[Di Salvo  et al. (1999)]{di}
525: %Di Salvo T., Mendez, M., van der Klis, M., Ford, E., 
526: %\& Robba, N.R. (1999) \apj, submitted
527: 
528: %\bibitem[Della Valle et al. (1991)]{del}
529: %Della Valle, M., Jarvis, B.J., \& West, R.M. 1991, Nature, 353, 50 
530: 
531: %\bibitem[Ebisawa et al. (1994)]{ebi}
532: %Ebisawa, K., et al. 1994, PASJ, 46, 375 
533:   
534: %\bibitem[Esin et al. (1997)]{esin}
535: %Esin, A.A., McClintock, J.E. \& Narayan, R. 1997, ApJ, 489, 865
536: 
537: %\bibitem[Gierlinski et al. (1999)]{gier}
538: % Gierlinski, M., et al. 1999, MNRAS, 309, 496
539:  
540: %\bibitem[Jonker et al. (2002)]{jon02}
541: %Jonker, P.G.,  et al.  2002, MNRAS, 333, 665
542: 
543: %\bibitem[Jonker et al. (1999)]{jon}
544: %Jonker, P.G.,  van der Klis, M. \& Wijnands, R. 1999, \apj , 511, L41
545: 
546: %\bibitem[Jonker et al. (2000)]{jon}
547: %Jonker, P.G.,  et al. 2000, \apj , 537, 374
548: 
549: %\bibitem[Ford \& van der Klis (1998)]{fvk}
550: %Ford, E., et al. 1998, \apj , 508, L155 
551: 
552: %\bibitem[Ford \& van der Klis (1998)]{fvk}
553: %Ford, E., \& van der Klis, M. 1998, \apj , 506, L39 (FVK)
554: 
555: %\bibitem[Galanin (1995)]{gal}
556: %Galanin, M.P. (1995), Quasi-stationary Electromagnetic Fields in
557: %Inhomogeneous Mediums, Moscow: Nauka 
558: 
559: %\bibitem[Ghosh  (1998)]{g98}
560: %Ghosh, P. 1998, \apj , 506, L109 
561: 
562: %\bibitem[Greiner et al. (2001)]{gre}
563: %Greiner, J., Cuby, J., \& McCaughrean, M.J.  2001, Nature, 414, 522
564: 
565: \bibitem[Fiorito et al. (2003)]{fio}
566: Fiorito, R., Markwardt, C., \&  Swank, J. 2003, HEAD meeting of AAS
567: 
568: %\bibitem[Griffiths et al. (2000)]{grif}
569: % Griffiths, R.E. et al. 2000, Science, 290, 1325
570:  
571: %\bibitem[Haberl  \& Titarchuk (1995)]{HT}
572: %Haberl, F., \& Titarchuk, L.G. 1995, A\&A , 299, 414
573: 
574: %\bibitem [Hoeksema \&  Scherrer] {HS}
575: %Hoeksema, J.T.,  \&  Scherrer, P.H. 1987,\apj, 318, 428   
576: 
577: %\bibitem[Hutchings  et al. (1987)]{Hut}
578: %Hutchings J.B., et al. 1987, AJ, 94, 340
579: %\bibitem[Hubeny et al. 2001]{HBKA}
580: %Hubeny, I., Blaes, O., Krolik, J.H., \& Agol, E. 2001, ApJ., 559, 680
581: 
582: %\bibitem[Humphrey et al. 2003]{}
583: % Humphrey, P.J., Fabbiano, G., Elvis, M., Church, M.J., \& Balucinska-Church, M.
584: % 2003, MNRAS, 344, 144
585:  
586: %\bibitem[Kaaret et al.  (2001) ]{kaa01}
587: %Kaaret, P., et al., 2001,  ApJ, 480, L27
588: 
589: %\bibitem[Kaaret et al.  (1997) ]{kaa97}
590: %Kaaret, P., Ford, E., \& Chen, K., 1997,  ApJ, 480, L27
591: 
592: %\bibitem[Klein et al.  (1996) ]{kl96}
593: %Kuznetsov, O.A., et al. 1999,  \apj, 514, 691
594: 
595: \bibitem[King et al. (2001)]{}King, A.R., et al. 
596: %A.R., Davies, M.B., Ward, M.J., Fabbiano, G., \& Elvis, M.
597:  2001, ApJ, 552, L109
598:   
599: %\bibitem[Kuznetsov \& Titarchuk (2002)]{kt02}
600: %Kuznetsov, S.I. \& Titarchuk, L.  2002,  \apj, 571, L137 (KT02)
601: %\bibitem[Laurent \& Titarchuk  (1998) ]{lt98}
602: %Laurent, Ph., \& Titarchuk, L.G., 1998 ApJ, 511,  (LT98)
603: 
604: %\bibitem[Lai (1998) ]{l98}
605: %Lai, D. 1998 ApJ, 502, 721 
606: 
607: %\bibitem[Lai (1998) ]{l99}
608: %Lai, D. 1999 ApJ, 524, 1031 
609:  
610: %\bibitem[Lamb \& Miller (2001)]{lm}
611: %Lamb, F.K., \& Miller, M.C. 2001, \apj,  544, 1210
612: 
613: %\bibitem[Laurent \& Titarchuk (1999)]{lt99}
614: %Laurent, P. \& Titarchuk, L.   1999,  \apj, 511, 289 
615: 
616: %\bibitem[Laurent \& Titarchuk (2001)]{lt01}
617: %Laurent, P. \& Titarchuk, L.   2001,  \apj, 562, L67 
618: 
619: %\bibitem[Liang (1998)]{liang98}
620: %Liang, E.P. 1998, Phys. Rep. 302, 66
621: 
622: %\bibitem[Lewin et al. (1995)]{lew}
623: %Lewin, W.H.G., Paradijs, J.V. \& Taam, R.E. 1995, in X-ray Binaries
624: %(Eds, Lewin, W.H.G., Paradijs, J.V. \& van den Heuvel,E.P.J) Cambridge:
625: %University Press
626: 
627: %\bibitem[Mauche  (2002)]{mau}
628: %Mauche, C.W. 2002 \apj,  580, 423
629: 
630: %\bibitem[Markwardt  et al. (1999)]{mss}
631: %Markwardt, C.B.,  Strohmayer, T.E. \& Swank, J.H. 1999, \apj,  512, L125
632: 
633: %\bibitem[Mendez  \& van der Klis (1998)]{men99}
634: %Mendez, M., \& van der Klis, M.  1998, \apj , 505, L23
635: 
636: %\bibitem[Merloni et al. (2000)]{} Merloni, A., Fabian, A.C., \& Ross, R.R.  2000, MNRAS,%
637: %313, 193
638: %\bibitem[McClintock (1998)]{mcc98}
639: %McClintock, J.E., 1998, Harward-Smithsonian Center for Astrophysics,
640: %Preprint Ser., 4643
641: %\bibitem[Miller (2000) ]{ml00}
642: %Miller, M.C. 2000, astro-ph/0007287 
643: \bibitem[Miller  et al.  (2003)]{ml}
644: Miller, J. et al.,  2003, \apj,  585, L37
645: 
646: %\bibitem[Miller  \& Lamb (1998)]{ml}
647: %Miller, M.C., \&  Lamb, F.K. 1998, \apj,  499, L37
648: 
649: %\bibitem[Miller  et al. (1998)]{mlp}
650: %Miller, M.C.,  Lamb, F.K. \& Psaltis, D. 1998, \apj,  508, 791
651: 
652: %\bibitem[Moon \& Eikenberry  (2001a)]{me+a}
653: %Moon, D.S.,  Eikenberry, S.S. 2001a, \apj,  552, L137
654: 
655: %\bibitem[Moon \& Eikenberry  (2001b)]{me+b}
656: %Moon, D.S.,  Eikenberry, S.S. 2001b, \apj,  549, L225
657: 
658: %\bibitem[Narayan \&  Yi  (1994)]{narayan94}
659: %Narayan, R., \& Yi, I.  1994, ApJ, 428, L13
660: 
661: %\bibitem[Oran \& Boris (1987)]{ash}
662: %Oran, E.S. \& Boris, J.P.  1987, Numerical Simulations of Reactive Flow, 
663: %New York: Elsevier
664: %\bibitem[Molteni, et al. (2001)]{mol}
665: %Molteni, D. et al., 2001,  ApJ., 563, L57
666: 
667: %\bibitem[Molteni, Toth, \& Kuznetsov (1999)]{mtk}
668: %Molteni D., Toth G., \& Kuznetsov, O. 1999, ApJ, 516, 411
669: 
670: %\bibitem[Molteni, Lanzafame \&  Chakrabarti (1994)]{mlc94}
671: %Molteni D.,  Lanzafame, G., \&  Chakrabarti, S. 1994, ApJ, 425, 161
672: 
673: %\bibitem[Orosz et al. (2002)]{}
674: % Orosz, J.,  et al. 2002, 
675:  %in" A Massive Star Odyssey, from Main Sequence to Supernova", 
676: % IAU Symp. 212, 
677:  %eds. K. A. van der Hucht, A. Herraro, \& C. Esteban, (San Francisco: ASP), 
678: %(astro-ph/0209041)
679: %\bibitem[Osherovich \& Titarchuk (1999a)]{ot1}
680: %Osherovich, V.A. \& Titarchuk, L.G. 1999a,  ApJ, 522, L113 (OT99a)
681: 
682: %\bibitem[Osherovich \& Titarchuk (1999b)]{ot2}
683: %Osherovich, V.A. \& Titarchuk, L.G. 1999b, ApJ, 523, L73 (OT99b)
684: 
685: %\bibitem[Osherovich \& Gliner, E.B. (2000)]{og}
686: %Osherovich, V.A. \& Gliner, E.B. 2000, in preparation
687: 
688: %\bibitem[Osherovich et al (1984)]{osh}
689: %Osherovich, V.A., Tzur, I., \& Gliner, E.B. 1984, \apj, 284, 412
690: 
691: %\bibitem[Novikov \& Thorne (1972)]{nt72}
692: % Novikov, I.D., \& Thorne K. 1972,  In Black Holes, (De Witt and B.S. De
693: %Witt (eds.) New York:Gordon and Breach, 1972, p. 935
694: 
695: %\bibitem[ Nowak et al. (1997)]{now97}
696: %Nowak, M.A., Wagoner, R.V., Begelman, M.C. \& Lehr, D.E. 1997, \apj, 284, L91
697: 
698: %\bibitem [Orosz et al.(1996)]{orosz96}
699: %Orosz, J.,  Bailyn, C.D., McClintock, J.E. \& Remillard, R. 1996, ApJ, 468, 
700: %330
701: %\bibitem[Perez et al. (1997)]{per97}
702: %Perez, C.A., Silbergleit, A.S.,  Wagoner, R.V., Begelman, M.C. \& Lehr, D.E. 
703: %1997, \apj, 476, 589
704: 
705: %\bibitem[Press et al. 1992]{press}
706: %Press, W.H., Teukolsky, S.A., Vetterling, W.T., \& Flannery, B.P.
707: %1992, Numerical Recipes (New York: Cambridge University Press) 
708: 
709: %\bibitem[Panagia, et al. (1998)]{pan}
710: %Panagia, N. et al.. 1991, ApJ, 380, L23 
711: %\bibitem[Psaltis, et al. (1999)]{ps99}
712: %Psaltis, D., Belloni \& van der Klis, M. 1998, ApJ, 520, 262 
713: 
714: %\bibitem[Psaltis, et al. (1998)]{ps}
715: %Psaltis, D., et al. 1998, ApJ, 501, L95 
716: 
717: %\bibitem[Ptak \& Colbert (2002)]{PM}
718: %Ptak, A., \& Colbert, E. 2002, ApJS, 143, 25
719: 
720: %\bibitem[Ray et al.  1999]{rwf+99}
721: %Ray, P.~S. et al.  1999, in X-ray Astronomy 1999, Bologna, Italy,
722: %(astro-ph/9911236)
723: 
724: %\bibitem[Rayleigh (1883)]{ray88}
725: %Rayleigh, L.  1883, Proc. London Math. Soc. 14, 170
726: \bibitem[Rephaeli \& Gruber (2002)]{RG}Rephaeli, Y. \& Gruber, D.  2002, A\&, 389, 752 
727: 
728: \bibitem[Remillard]{Rem02a}Remillard, R. et al. 2002a, ApJ, 580, 1030 
729: 
730: \bibitem[Remillard]{Rem02b}Remillard, R. et al. 2002b, ApJ, 564, 962
731: 
732: %\bibitem[Roberts \& Warwick (2000)]{RW00}
733: %Roberts, T.P., \& Warwick, R. et al. 2000, MMRAS, 315, 98
734: %\bibitem[Remillard,  et al. (2002)]{rem02}
735: %Remillard, R.A., et al., 2002, astro-ph/0208402
736: 
737: %\bibitem[Remillard,  et al. (1999)]{rem}
738: %Remillard, R.A., et al., 1999, ApJ, 522, 397
739: 
740: %\bibitem[Quinlan, \& Shapiro, (1990)]{qS}
741: %Quinlan, G., \& Shapiro, S. 1990, ApJ, 356, 483 
742: 
743: %\bibitem[Ryu, Chakrabarti, \& Molteni (1997)]{rcm97}
744: %Ryu, D., Chakrabarti, S.K. \& Molteni, D. (1997)
745: %ApJ.,474, 378, 1997
746: 
747: %\bibitem[Shahbaz, T., Naylor, T., \& Charles, P.A. (1997)]{}
748: %Shahbaz, T., Naylor, T., \& Charles, P.A. 1997, MNRAS, 285, 607
749: 
750: %\bibitem[ Shakura \& Sunyaev (1973)]{SS73}
751: %Shakura, N.~I. \& Sunyaev, R.~A. 1973, A\&A, 24, 337 (SS73)
752: 
753: %\bibitem[Shaposhnikov, Titarchuk \& Haberl (2002)]{sth02}
754: %Shaposhnikov, N. \& Titarchuk, L. \& Haberl, F.    2003,  \apj, 593
755: 
756: %\bibitem[Shimura \& Takahara (1995)]{shimura95}
757: % Shimura, T., \& Takahara, F. 1995, ApJ, 445, 780 (ShT95)
758: 
759: %\bibitem[Shrader \& Titarchuk (2000)]{shT00}
760: %Shrader, C., \& Titarchuk, L.G. 2000, in proc. 5th Compton Symposium, 
761: %AIP-510
762: 
763: \bibitem[Shrader \& Titarchuk (2003)]{shT03}
764:  Shrader, C., \& Titarchuk, L.G. 2003,  ApJ, 598, 168 (ST03)
765:  
766: %\bibitem[Shrader \& Titarchuk (1999)]{shT99}
767: % Shrader, C., \& Titarchuk, L.G. 1999,  ApJ, 521, L121 (ST99)
768: 
769: %\bibitem[Shrader \& Titarchuk (1998)]{shT}
770: %Shrader, C., \& Titarchuk, L. 1998, ApJ, 499, L31 (ShT98)
771: 
772: \bibitem[Sobzcak]{Sob99b}Sobczak, G.J., et al. 1999, ApJ, 517, L121
773: 
774: \bibitem[Sobzcak]{Sob00a}Sobczak, G.J., et al. 2000, ApJ, 531, 537
775: 
776: %\bibitem[Swartz et al. (2004)]{swar}Swartz, D.A., et al. 2004, ApJ, in press (astro-ph/0405498)
777: %\bibitem[Stella \& Vietri 1999] {sv} 
778: %Stella, L., \& Vietri, M.  1999, Phys. Rev. Letters, 82, 17 
779: 
780: %\bibitem[Stella et al.  1999] {sv} 
781: %Stella, L., \& Vietri, M. \& Morsink  1999, Astrophys. Lett. Comm., 38, 57 
782: 
783: \bibitem[Strohmayer et al. (2001)]{str01}
784: Strohmayer, T.~E., \& Mushotzky, R.F.  2003, ApJ, 586, L61 (SM03)
785: 
786: \bibitem[Sunyaev \& Titarchuk (1980)]{ST80}
787: Sunyaev, R.A. \& Titarchuk, L.G.  1980,  A\&A, 86, 121 
788: 
789: %\bibitem[Strohmayer et al. (2001)]{str01}
790: %Strohmayer, T.~E. 2001a, ApJ, 552, L49  
791: 
792: %\bibitem[Strohmayer et al. (2000)]{str00}
793: %Strohmayer, T.~E.  2001b, ApJ, 554, L169  
794: 
795: %\bibitem[Strohmayer et al. (1998a)]{str98}
796: %Strohmayer, T.~E., Zhang, W., Swank, J.~H., \& Lapidus, I.I. 1998a, \apj, 
797: %503, L147 
798: 
799: %\bibitem[Strohmayer et al. (1998b)]{str98}
800: %Strohmayer, T.~E., Zhang, W., Swank, J.~H., White, N.E. 
801: %\& Lapidus, I.I. 1998b, \apj , 498, L135 
802: 
803: %\bibitem[Strohmayer et al. (1997)]{str97}
804: %Strohmayer, T.~E., Zhang, W., Swank, J. 1997, \apj , 487, L77 
805: 
806: %\bibitem[Strohmayer et al. (1996)]{str96}
807: %Strohmayer, T.~E., Zhang, W., Swank, J.~H., Smale, A.,
808: %Titarchuk, L., Day, C., \&  Lee, U. 1996, \apj , 469, L9 (S96)
809: 
810: %\bibitem[Thorne  (1986) ]{tit94}
811: %Thorne, K., Price, R.H., \& Macdonald, D.A. 1986,  Black Holes: the Membrane 
812: %Paradigm,  New Haven: Yale University Press
813: %\bibitem[Sunyaev \& Titarchuk (1980)]{ST80}
814: %Sunyaev, R.A. \& Titarchuk, L.G. 1980,  A\&A, 86, 121
815: 
816: %\bibitem[Titarchuk (2002)]{t02}
817: %Titarchuk, L.  2002,  \apj, 578, L71 (T02)
818: 
819: %\bibitem[Titarchuk (2003)]{t03}
820: %Titarchuk, L.  2003,  \apj, 591, 354
821: 
822: %\bibitem[Titarchuk, \& Wood (2002)]{tw}
823: %Titarchuk, L.G., \& Wood, K.S. 2002,  
824: %\apj, 577, L (TW02)
825: 
826: %\bibitem[Titarchuk, \& Shaposhnikov (2002)]{ts02}
827: %Titarchuk, L. \&  Shaposhnikov, N.  2002,  \apj, 570, L25 (TS02) 
828: 
829: %\bibitem[Titarchuk, \& Shrader (2002)]{tsh02}
830: %Titarchuk, L. \&  Shrader, C.R.  2002,  \apj, 567, 1057 (TSh02)
831: 
832: %\bibitem[Titarchuk, et al. (2001)]{toa}
833: %Titarchuk, L.G., Bradshaw, C.F., \& Wood, K.S. 2001,  
834: %\apj, 560, L45 (TBW)
835: 
836: %\bibitem[Titarchuk, et al. (2001)]{toa}
837: %Titarchuk, L.G., Bradshaw, C.F., Geldzahler, B.J., \& Fomalont, E.B.  2001,  
838: %\apj, 555, L45 (TBGF)
839: 
840: %\bibitem[Titarchuk, Mastichiadis \&  Kylafis, (1996)]{tmk96}
841: % Titarchuk, L. G., Mastichiadis, A., \& Kylafis, N. D. 1996, A\&As, 120, 171
842: %(TMK96)
843: \bibitem[Titarchuk, Mastichiadis \&  Kylafis, (1997)]{tmk97}
844:  Titarchuk, L. G., Mastichiadis, A., \& Kylafis, N. D. 1997, ApJ, 487, 834
845: %(TMK97)
846: \bibitem[Titarchuk, \& Fiorito (2004)]{tF}
847: Titarchuk, L.G. \& Fiorito, R.  2004,  \apj, 612,  (astro-ph/0405360) (TF04) 
848: 
849: %\bibitem[Titarchuk, \& Osherovich (2001)]{to01}
850: %Titarchuk, L.G. \& Osherovich, V.A.  2001,  \apj, 555, L55 (TO01)
851: 
852: %\bibitem[Titarchuk, \& Osherovich (2000a)]{toa}
853: %Titarchuk, L.G. \& Osherovich, V.A.  2000a,  \apj, 537, L39 (TO00a)
854: 
855: %\bibitem[Titarchuk, \& Osherovich (2000b)]{tob}
856: %Titarchuk, L.G. \& Osherovich, V.A.  2000b,  \apj, 542, L111 (TO00b) 
857: 
858: %\bibitem[Titarchuk, \& Osherovich (1999)]{to}
859: %Titarchuk, L.G. \& Osherovich, V.A.  1999,  \apj, 518, L95 (TO99)
860: 
861: %\bibitem[Titarchuk, et al. (1999)]{tok}
862: %Titarchuk, L.G., Osherovich, V.A. \& Kuznetsov, S.I. 1999,  
863: %ApJ, 525, L129 (TOK) 
864: 
865: \bibitem[Titarchuk, Lapidus \& Muslimov (1998)]{tlm98}
866: Titarchuk, L., Lapidus, I.I., \& Muslimov, A. 1998, \apj,  499, 315 
867: (TLM98)
868: 
869: %\bibitem[Titarchuk \& Zannias (1998)]{tz98}
870: %Titarchuk, L., \& Zannias. T., 1998, ApJ, 493, 863 (TZ98)
871: 
872: %\bibitem[Titarchuk  (1994) ]{tit94}
873: %Titarchuk, L.G. 1994, ApJ, 434, 570 
874: 
875: %\bibitem[Titarchuk and Wood (2002)]{}Titarchuk, L. and Wood, K. 2002, ApJ,
876: %577, L23 (TW02)
877: 
878: %\bibitem[Titarchuk \& Zannias (1998)]{TZ98} Titarchuk, L.G. \& Zannias, T. 1998, 493, 863 
879: %\bibitem[Ustugova (1999)]{u}
880: %Ustugova, G.V. et al.  1999, \apj, 516, 221
881: 
882: %\bibitem[van der Klis  2000]{vk00}
883: %van der Klis, M. 2000, Annual Rev. of Astronomy and Astrophysics, 
884: %38, 717 (VDK) 
885: 
886: %\bibitem[van der Klis et al. 1997]{vdk97}
887: %van der Klis, M., Wijnands, A.D., Horne, K. \& Chen, W. 1997,
888: %\apj , 481, L97 
889: 
890: 
891: %\bibitem[van der Klis et al. 1996]{vdk96}
892: %van der Klis, M. et al. 1996 \apj , 469, L1
893: 
894: %\bibitem[van Straaten et al. 2000]{VS}
895: %van Straaten, S., Ford, E., van der Klis, M., Mendez, M., Kaaret, P. 2000,
896: %\apj, 540, 1049
897: 
898: \bibitem[Vignarca et al. (2003)]{vig}Vignarca, F., et al.
899: %Migliari, S., Belloni, T.,
900: %Psaltis, D., \& van der Klis, M.   
901: 2003   A\&A, 397, 729 (V03)
902: 
903: %\bibitem[Wagoner  (1999)]{wag99}
904: %Wagoner, R.V.  1999, Phys. Rep. 311, 259
905: 
906: \bibitem[Wijnands \& van der Klis  1999]{wV99}
907: Wijnands, A.D. \& van der Klis, M.  1999,
908: \apj, 514, 939
909: 
910: %\bibitem[Wood et al.  2001]{wo01}
911: %Wood, K. S., Titarchuk, L., Ray, P.S., et al.  2001, \apj, 563, 246
912: 
913: %\bibitem[Wood et al.  2000]{wo}
914: %Wood, K. S. et al.  2000, \apj, 544, L45
915: 
916: %\bibitem[Woudt \& Warner (2001)]{ww}
917: %Woudt, P.A., \& Warner, B.  2002, MNRAS, 433, 411 
918: 
919: %\bibitem[Wu   (2001)]{wu}
920: %Wu, X-B.   2001, \apj , 552, 227
921: 
922: %\bibitem[Zdziarski et al. (2001)]{z01}
923: %Zdziarski, A., et al. 2001, ApJ, 554, L45
924: 
925: %\bibitem[Zhang et al. (1997)]{zh97}
926: %Zhang, W., Strohmayer, T.E., \& Swank, J.H. 1997, ApJ, 482, L167
927: 
928: \end{thebibliography}
929: %\begin{table*}
930: %\caption{The best-fit spectral parameters.}
931: %\begin{tabular}
932: %[c]{lllllll}%
933: %Obsid & Instrument & $n_{H(PL)}\;\times10^{22}$ & $n_{H(BMC)}\;\times10^{22}$ &
934: %$\Gamma_{PL}$ & $\Gamma_{BMC}$ & $N_{BMC}\;\times(10^{-3})$\\
935: %011201 & XMM-PN & $7.50\pm0.69$ & $7.50\pm0.69$ & $2.07\pm0.07$ &
936: %$2.07\pm0.07$ & $0.26\pm0.08$\\
937: %011201 & XMM-MOS & $6.98\pm1.17$ & $7.42\pm1.21$ & $1.71\pm0.14$ &
938: %$1.82\pm0.16$ & $0.21\pm0.13$\\
939: %20303-02-02 & XTE-PCA & $3.73\pm0.76$ & $4.43\pm1.53$ & $2.62\pm0.09$ &
940: %$2.65\pm0.13$ & $1.13\pm0.67$\\
941: %20303-02-03 & XTE-PCA & $4.09\pm0.78$ & $5.26\pm1.93$ & $2.46\pm0.09$ &
942: %$2.52\pm0.13$ & $1.11\pm0.67$\\
943: %20303-02-04 & XTE-PCA & $6.06\pm0.81$ & $5.21\pm0.85$ & $2.67\pm0.09$ &
944: %$2.63\pm0.96$ & $1.35\pm0.55$%
945: %\end{tabular}
946: %\end{table*}
947: 
948: \begin{deluxetable}{rrrrrrrrrr} 
949: \tablecolumns{7} 
950: \tablewidth{0pc} 
951: \tablecaption{The best-fit spectral parameters.}
952: \tablehead{ 
953: \colhead{OBSID}&\colhead{Instrument}
954:   & \colhead{$n_{H(PL)}\times10^{22}$}  & \colhead{$n_{H(BMC)}\times10^{22}$}  & \colhead{$\Gamma_{PL}$}  & \colhead{$\Gamma_{BMC}$}  
955: & \colhead{$N_{BMC}\times(10^{-3})$} }
956: \startdata 
957: 011201 & XMM-PN & $7.50\pm0.69$ & $7.50\pm0.69$ & $2.07\pm0.07$ & $2.07\pm0.07$ & $0.26\pm0.08$\\
958: 011201 & XMM-MOS & $6.98\pm1.17$ & $7.42\pm1.21$ & $1.71\pm0.14$ & $1.82\pm0.16$ & $0.21\pm0.13$\\
959: 20303-02-02 & XTE-PCA & $3.73\pm0.76$ & $4.43\pm1.53$ & $2.62\pm0.09$ &
960: $2.65\pm0.13$ & $1.13\pm0.67$\\
961: 20303-02-03 & XTE-PCA & $4.09\pm0.78$ & $5.26\pm1.93$ & $2.46\pm0.09$ &
962: $2.52\pm0.13$ & $1.11\pm0.67$\\
963: 20303-02-04 & XTE-PCA & $6.06\pm0.81$ & $5.21\pm0.85$ & $2.67\pm0.09$ &
964: $2.63\pm0.1$ & $1.35\pm0.55$
965: \enddata 
966: %\tablenotetext{a}{$M_i=M(\cos{i})^{1/2}$}
967: \end{deluxetable}
968: 
969: %\begin{deluxetable}{rrrrrrrrrr} 
970: %\tablecolumns{6} 
971: %\tablewidth{0pc} 
972: %\tablecaption{Timing Feature of RXTE PDS}
973: %\tablehead{ 
974: %\colhead{OBSID}&\colhead{Instrument}
975: %  & \colhead{$\nu_{brk}(mHz)$}  & \colhead{$A_{brk}$}  & \colhead{$\nu_{QPO}(mHz)$} & \colhead{$A_{QPO}$}}
976: %\startdata 
977: %$20303-02-04$ & $XTE-PCA-Xe$ & $26\pm2.5$ & $4.8\pm0.4$ & $106\pm2$ &  $0.095\pm0.018$\\
978: %$20303-02-03$ & $XTE-PCA-Xe$ & $$ & $$ & $48.9\pm1$ & $0.040\pm0.013$
979: %\enddata 
980: %\tablenotetext{a}{$M_i=M(\cos{i})^{1/2}$}
981: %\end{deluxetable}
982: %\begin{table*}
983: %\caption{Timing Feature of RXTE PDS.}
984: %\begin{tabular}
985: %[c]{lllllll}%
986: %instrument & obsid & $f_{brk}(mHz)$ & $A_{brk}$ & $f_{QPO}(mHz)$ & $\Delta
987: %f_{QPO}(mHz)$ & $A_{QPO}$\\
988: %$XTE-PCA-Xe$ & $20303-02-04$ & $26\pm2.5$ & $4.8\pm0.4$ & $106\pm2$ & $24\pm7$
989: %& $0.095\pm0.018$\\
990: %$XTE-PCA-Xe$ & $20303-02-03$ & $12\pm1.1$ & $2.1\pm0.3$ & $48.9\pm1$ &
991: %$12\pm4.4$ & $0.040\pm0.013$%
992: %\end{tabular}
993: %\end{table*}
994: 
995: \newpage
996: \begin{figure}[ptbptbptb]
997: \includegraphics[width=5.1in,height=4.5in,angle=0]{f1.eps}
998: \caption{
999: Comparison of the observed (points) and the  theoretical correlations
1000: (solid lines) of photon index vs QPO low frequency between GRS 1915+105 (observations by Vignarca et al. 2003) and
1001: XTE J1550-564 [observations by Sobczak et al. (1999), (2000); Remillard et al. (2002a,b), see also 
1002:   Fig. 6, 8 in Vignarca et al.]. Black points and line for GRS 1915+105 and red points and line
1003: for XTE J1550-564. The XTE J1550-564 curve is produced by sliding the GRS
1004: 1915+105 curve along the frequency axis with factor $12/10$ (see text for
1005: details). }
1006: \label{1915_1550}
1007: \end{figure}
1008: \newpage
1009: \begin{figure}[ptbptbptb]
1010: \includegraphics[width=4.5in,height=5.1in,angle=-90]{f2.eps}
1011: \caption{
1012: BMC fit to the XMM PN energy spectrum of M82-X1 in the range 3.3 to
1013: 10 keV.
1014: }
1015: \label{M82-X1}
1016: \end{figure}
1017: \newpage
1018: \begin{figure}[ptbptbptb]
1019: \includegraphics[width=3.4 in,height=3.3in,angle=90]{f3a.eps}
1020: \includegraphics[width=2.6in,height=3.in,angle=90]{f3b.eps}
1021: \caption{
1022: Left panel: Power spectrum from XMM EPIC PN instrument for the source 
1023: M82-X1  for the  energy range 3.3-10keV showing  a QPO at 57 mHz.
1024: Right panel:Same for RXTE OBSID 20303-02-04 showing a break
1025: frequency at 26 mHz and QPO frequency at 106 mHz. 
1026: }
1027: \label{PDS}
1028: %\label{M82-X1_XMM_QPO}
1029: \end{figure}
1030: %\newpage
1031: %\begin{figure}[ptbptbptb]
1032: %\includegraphics[width=4.5in,height=5.1in,angle=90]{f4.eps}
1033: %\caption{
1034: %Power vs Frequency for RXTE OBSID 20303-02-04 showing a break
1035: %frequency at 26 mHz and QPO frequency at 106 mHz. 
1036: %}
1037: %\label{M82-X1_RXTE_QPO}
1038: %\end{figure}
1039: 
1040: 
1041: \end{document}
1042:                                                                                  
1043: