1: %\documentclass[traditabstract]{aa}
2: \documentclass[twocolumn,oldversion]{aa}
3: %\documentclass[referee,traditabstract]{aa} % for a referee version
4: %\documentclass[onecolumn]{aa}
5: %\documentclass[twocolumn]{aa}
6: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
7: \usepackage{graphicx}
8: \usepackage{txfonts}
9: \usepackage{natbib}
10: \bibpunct{(}{)}{;}{a}{}{,}
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12: \newcommand{\vcas} {4U 0115+63}
13: \newcommand{\exo} {EXO 2030+375}
14: \newcommand{\bq} {V0332+53}
15: \newcommand{\ks} {KS 1947+300}
16:
17: \def\simless{\mathbin{\lower 3pt\hbox
18: {$\rlap{\raise 5pt\hbox{$\char'074$}}\mathchar"7218$}}} %< or of order
19: \def\simmore{\mathbin{\lower 3pt\hbox
20: {$\rlap{\raise 5pt\hbox{$\char'076$}}\mathchar"7218$}}} %> or of order
21:
22: \def\msun{~{\rm M}_\odot}
23: \def\rsun{~{\rm R}_\odot}
24:
25: \begin{document}
26:
27: \title{Rapid spectral and timing variability of Be/X-ray binaries during type II
28: outbursts \\
29: }
30:
31: \subtitle{}
32:
33: \author{P. Reig
34: \inst{1,2}
35: }
36:
37: \authorrunning{Reig et~al.}
38: \titlerunning{Rapid variability of Be/X-ray binaries}
39:
40: \offprints{P. Reig}
41:
42: \institute{IESL, Foundation for Research and Technology-Hellas, 71110,
43: Heraklion, Greece
44: \and
45: Physics Department, University of Crete, 71003, Heraklion, Greece
46: \email{pau@physics.uoc.gr}
47: }
48:
49: \date{Received ; accepted}
50:
51:
52:
53: \abstract{
54: X-ray colour-colour (CD) and colour-intensity (HID) diagrams are powerful
55: tools that allow the investigation of spectral variability without the
56: assumption of any spectral model. These diagrams have been used extensively
57: in low-mass X-ray binaries and black-hole candidates but very few
58: applications are found for high-mass X-ray binaries. We have investigated
59: the spectral and timing variability of four accreting X-ray pulsars with
60: Be-type companions during major X-ray outbursts. The aim is to define
61: source states based on the properties (noise components) of the aperiodic
62: variability in correlation with the position in the colour-colour diagram.
63: Different spectral states were defined according to the value of the X-ray
64: colours and flux. Transient Be/X-ray binaries exhibit two branches in their
65: colour-colour and colour-intensity diagrams: the horizontal branch
66: corresponds to a low-intensity state and shows the larger fractional $rms$,
67: similar to the the island state in atolls and horizontal branch in Z
68: sources; the diagonal branch corresponds to a high-intensity state, in
69: which the source spends about 75\% of the total duration of the outburst.
70: Despite the complexity of the power spectra due to the peaks of the pulse
71: period and its harmonics, the aperiodic variability of Be/X-ray binaries
72: can be described with a relatively low number of Lorentzian components.
73: Some of these components can be associated with the same type of noise as
74: that seen in low-mass X-ray binaries, although the characteristic
75: frequencies are about one order of magnitude lower. The analysis of
76: the CD/HID and power spectra results in two different types of Be/X. While
77: in \vcas, \ks\ and \exo\ the hard colour decreases as the count rate
78: decreases, in \bq\ it increases. The pattern traced by \bq\ then results
79: in a Z shaped track, similar to the low-mass $Z$ sources, without the
80: flaring branch. In contrast, the horizontal branch in \vcas, \ks\ and \exo\
81: corresponds to a low/soft state, not seen in other types of X-ray binaries.
82: The noise at very low frequencies follows a power law in \bq\ (like in LMXB
83: Z) and it is flat-topped in \vcas, \ks\ and \exo\ (like in LMXB atoll). \bq\
84: shows a noise component coupled with the periodic variability that it is
85: not seen in any of the other three sources.
86:
87: }
88:
89: \keywords{stars: individual: 4U 0115+63, V 0332+53, KS 1947+300, EXO
90: 2030+375 -- X-rays: binaries -- stars: neutron -- stars: binaries close
91: --stars: emission line, Be
92: }
93:
94:
95: \maketitle
96:
97: %-------------------------------------------------------------
98: \begin{table*}
99: \caption{Optical and X-ray information of the systems and the outbursts analysed
100: in this work.}
101: \label{info}
102: \centering
103: \begin{tabular}{lcccccccc}
104: \hline\hline
105: Source &Spectral&P$_{\rm spin}$ &P$_{\rm orb}$ &$e$ &Distance &Outburst &$L_{\rm X,peak}$ &On-source \\
106: name &type &(s) &(days) & &(kpc) &duration (d) &3-30 keV (erg s$^{-1}$) &time (ks) \\
107: \hline
108: 4U 0115+63 &B0.2V &3.6 &24.3 &0.34 &8.1 &55 &$1.2\times10^{38}$ &91 \\
109: KS 1947+300 &B0V &18.7 &40.4 &0.03 &10 &165 &$6.2\times10^{37}$ &132 \\
110: EXO 2030+375 &B0.5III-V&41.8 &46.0 &0.41 &7.1 &155 &$1.4\times10^{38}$ &166 \\
111: V 0332+53 &O8-9V &4.4 &34.2 &0.30 &7 &105 &$3.2\times10^{38}$ &198 \\
112: \hline\hline
113: %\multicolumn{9}{l}{$a$: 3-30 keV in erg s$^{-1}$} \\
114: \end{tabular}
115: \end{table*}
116: %-------------------------------------------------------------
117:
118: \section{Introduction}
119:
120:
121: The colour-colour (CD) and hardness-intensity (HID) diagrams are very
122: useful tools to investigate the spectral variations of an X-ray source.
123: They are model- and instrument-independent and reflect the intrinsic
124: properties of the system. The first application of this type of plots was
125: the separation of different types of galactic X-ray sources and the
126: classification of X-ray populations. The spectral hardness was defined by
127: the ratio of counting rates in each of the proportional counters of the
128: {\em MIT OSO-7} experiment \citep{mark77,rapp77}. \citet{whit84} made use
129: of the X-ray colour-colour diagrams, obtained from {\em HEAO 1}
130: observations, to study the spectral states among the different types of
131: X-ray binaries, including black-hole candidates.
132:
133: CDs became widely used during the EXOSAT era, which also marks the
134: introduction of HIDs \citep{pried86,klis87a,hasi87}, where "hardness"
135: refers to the ratio of the count rate in two energy bands. Before EXOSAT,
136: CDs contained sources of different origin. The novelty of the EXOSAT CDs
137: and HIDs was that they were applied to individual sources. CD/HIDs became a
138: useful way to investigate the rapid aperiodic variability in correlation
139: with spectral states, as defined by the spectral hardness, of individual
140: low-mass X-ray binaries and black-hole candidates \citep{schu89,hasi89}.
141:
142: With the advent of RXTE the use of CDs and HIDs have become part of the
143: standard data analysis. RXTE has provided an unprecedentedly large database
144: of X-ray observations with exceptionally good timing and moderate spectral
145: resolution. Together with the functional form of the variability
146: components ("noise"), CD and HID have become an essential tool in
147: developing a whole new phenomenology on the spectral and timing properties
148: of X-ray binaries by introducing the notion of source states. A state is
149: defined by the appearance of a spectral (i.e. power-law, blackbody) or
150: variability component (i.e. Lorentzian) associated with a particular and
151: well-defined position of the source in the CD/HID. A transition between
152: states is assumed when the relative strength of the spectral or variability
153: components varies and the source motion in the CD/HID changes direction
154: \citep{klis06}.
155:
156: While there are numerous references in the literature on the application of
157: CD/HID analysis on low-mass X-ray binaries \citep[][and references
158: therein]{klis06} and black-hole systems \citep[see e.g.][]{bell05}, very
159: little work of this type has been done on high-mass X-ray binaries.
160: \citet{bell90} performed an aperiodic variability study of 12 high-mass
161: X-ray binaries from EXOSAT archive data. The sample included \exo\ and \bq.
162: However, the few observations analysed of \bq\ did not correspond to a
163: major X-ray outburst, while the data of \exo\ covered the decay of the 1985
164: outburst only partially. The other 10 sources were supergiant X-ray
165: binaries in bright states. The vast majority of the rapid aperiodic
166: variability studies in accreting X-ray pulsars have concentrated on the
167: detection of QPOs \citep{ange89,jern00,qu05};
168: see also \citet[][and references therein]{shir02}
169:
170: Massive X-ray binaries are classified according to the luminosity class of
171: the optical component into Be/X-ray binaries (dwarf or subgiant) and
172: supergiant (luminosity class I-II) X-ray binaries. The former tend to be
173: transient systems while the latter are persistent sources. Be stars
174: are non-supergiant fast-rotating B-type and luminosity class III-V stars
175: which at some point of their lives have shown spectral lines in emission.
176: In the infrared they are brighter than their non-emitting counterparts of
177: the same spectral type. The line emission and infrared excess
178: originate in extended circumstellar envelopes of ionized gas surrounding
179: the equator of the B star. When the Be star takes part on a binary system
180: where the companion is a neutron star, then the system is referred to as a
181: Be/X-ray binary.
182:
183: Be/X-ray binaries are variable on time scales from seconds to years. The
184: fastest variability is found in the X-ray band. Virtually all the Be/X-ray
185: binaries with identified optical counterparts are X-ray pulsars. Pulse
186: periods cover the range 1--10$^3$ s. On longer time scales (months to
187: years) the variability is also apparent in the optical and infrared bands
188: and it is attributed to structural changes of the circumstellar disc. The
189: long-term X-ray variability of the transient Be/X-ray binaries is
190: characterised by two type of outbursting activity
191:
192: \begin{itemize}
193:
194: \item Type I outbursts. These are regular and (quasi)periodic outbursts,
195: normally peaking at or close to periastron passage of the neutron star.
196: They are short-lived, i.e., tend to cover a relatively small fraction of
197: the orbital period (typically 0.2-0.3 $P_{\rm orb}$). The X-ray flux
198: increases by about one or two order of magnitude with respect to the pre-outburst
199: state, reaching peak luminosities $L_x \leq 10^{37}$ erg s$^{-1}$.
200:
201: \item Type II outbursts represent major increases of the X-ray flux,
202: $10^{3}-10^{4}$ times that at quiescence. They reach the Eddington
203: luminosity for a neutron star ($\sim 10^{38}$ erg s$^{-1}$) and become the
204: brightest objects of the X-ray sky. They do not show any preferred orbital
205: phase and last for a large fraction of an orbital period or even for
206: several orbital periods. The formation of an accretion disc during Type II
207: outbursts \citep{kris83,motc91,haya04,wils08} may occur. The discovery of
208: quasi-periodic oscillations in some systems \citep{ange89,take94,fing96} would
209: support this scenario. The presence of an accretion disk also helps explain
210: the large and steady spin-up rates seen during the giant outbursts, which
211: are difficult to account for by means of direct accretion.
212:
213: \end{itemize}
214:
215: In this work we present a systematic study of the aperiodic variability of
216: four Be/X-ray binaries during type II outbursts. First, we give a summary
217: of the relevant information of the sources. Sect.~\ref{obs} summarises the
218: observations and the characteristics of the instruments used in the
219: analysis. Sect.~\ref{red} refers to the reduction of the data and describes
220: how the CD/HID and power spectra were obtained. A detailed description of
221: the results is given in Sect.~\ref{res}. This section is divided into three
222: subsections describing the outburst profiles, the colour analysis and the
223: aperiodic variability. Our interpretation of the results and the
224: implications of this work are discussed in Sect.~\ref{disc}. Final
225: conclusions are drawn in Sect.~\ref{con}.
226:
227:
228: %-------------------------------------------------------------
229: \begin{table*}
230: \caption{Observations and definition of source states.}
231: \label{reg}
232: \centering
233: \begin{tabular}{lcccccccc}
234: \hline\hline
235: Source&Branch$^a$ &Time range &Proposal &On-source &$<SC>^b$ &$<HC>^b$&Flux$^{c}$ &Luminosity$^{d}$ \\
236: state & &MJD &ID &time (s) & & & & \\
237: \hline \hline \noalign{\smallskip}
238: \multicolumn{9}{c}{{\bf \vcas}}\\
239: \hline \noalign{\smallskip}
240: Rise1 &DB &53254.0676-53256.4660 &90089 &3872 &2.08$\pm$0.05 &0.71$\pm$0.03 &0.95 &7.5 \\
241: Rise2 &DB &53260.1495-53265.0921 &90089 &3536 &1.91$\pm$0.05 &0.81$\pm$0.04 &1.5 &11.8 \\
242: Peak &DB &53267.1686-53272.9721 &90089 &29744 &1.87$\pm$0.01 &0.89$\pm$0.01 &1.4 &11.0 \\
243: Decay1 &DB &53274.8282-53280.7687 &90089 &9712 &1.95$\pm$0.02 &0.85$\pm$0.02 &1.0 &7.9 \\
244: Decay2 &DB &53278.6230-53280.7687 &90089 &6128 &2.03$\pm$0.04 &0.70$\pm$0.06 &0.8 &6.0 \\
245: Decay3 &DB &53282.5715-53282.6673 &90014 &5536 &2.09$\pm$0.01 &0.62$\pm$0.01 &0.6 &5.0 \\
246: Decay4 &HB &53284.6063-53304.5730 &90014 &32500 &1.99$\pm$0.04 &0.57$\pm$0.03 &0.11 &0.9 \\
247: \hline \noalign{\smallskip}
248: \multicolumn{9}{c}{{\bf \ks}}\\
249: \hline \noalign{\smallskip}
250: Rise0 &HB &51874.1502-51878.3595 &50425 &16720 &1.57$\pm$0.07 &1.24$\pm$0.02 &0.01 &0.12\\
251: Rise1 &DB &51883.8286-51892.2963 &50425 &14060 &1.68$\pm$0.04 &1.13$\pm$0.05 &0.09 &1.0 \\
252: Rise2 &DB &51896.1987-51910.9382 &50425 &17744 &1.55$\pm$0.03 &1.14$\pm$0.02 &0.25 &2.3 \\
253: Rise3 &DB &51914.1750-51932.3484 &50425 &10080 &1.49$\pm$0.03 &1.16$\pm$0.03 &0.36 &4.3 \\
254: Peak &DB &51941.6795-51968.5845 &50425 &23344 &1.40$\pm$0.02 &1.20$\pm$0.02 &0.53 &6.3 \\
255: Decay1 &DB &51970.2073-51990.1352 &60402 &15536 &1.51$\pm$0.03 &1.19$\pm$0.02 &0.39 &4.7 \\
256: Decay2 &DB &51992.1719-52010.8263 &60402 &9888 &1.59$\pm$0.02 &1.16$\pm$0.02 &0.22 &2.6 \\
257: Decay3 &DB &52014.3856-52050.1873 &60402 &9984 &1.65$\pm$0.04 &1.05$\pm$0.07 &0.11 &1.3 \\
258: Decay4 &HB &52052.9510-52078.1324 &60402 &14720 &1.32$\pm$0.08 &1.11$\pm$0.04 &0.008 &0.1 \\
259: \hline \noalign{\smallskip}
260: \multicolumn{9}{c}{{\bf \exo}}\\
261: \hline \noalign{\smallskip}
262: Rise1 &DB &53908.6156-53924.7247 &91089 \& 92067&43200 &1.39$\pm$0.07 &1.09$\pm$0.02 &1.4 &8.4 \\
263: Rise2 &DB &53926.6282-53943.2576 &91089 &26512 &1.26$\pm$0.02 &1.12$\pm$0.01 &2.1 &12.7 \\
264: Peak &DB &53944.0532-53975.4312 &91089 &35536 &1.23$\pm$0.02 &1.13$\pm$0.02 &2.4 &14.5 \\
265: Decay1 &DB &53976.1186-53998.6782 &91089 &25168 &1.32$\pm$0.05 &1.09$\pm$0.03 &1.6 &9.7 \\
266: Decay2 &DB &53999.5902-54015.4534 &92067 &8656 &1.47$\pm$0.05 &0.99$\pm$0.02 &0.57 &3.4 \\
267: Decay3 &DB &54016.5928-54030.9436 &92067 &10912 &1.43$\pm$0.03 &1.02$\pm$0.04 &0.11 &0.69 \\
268: Decay4 &HB &54031.7791-54047.9337 &92067 &16144 &1.41$\pm$0.03 &0.96$\pm$0.03 &0.06 &0.37 \\
269: \hline \noalign{\smallskip}
270: \multicolumn{9}{c}{{\bf \bq}}\\
271: \hline \noalign{\smallskip}
272: Rise1 &DB &53340.2869-53346.7436 &90089 &48752 &2.18$\pm$0.02 &0.89$\pm$0.03 &2.3 &13.5 \\
273: Rise2 &DB &53352.7706-53361.1469 &90089 &32720 &2.06$\pm$0.03 &0.73$\pm$0.02 &5.0 &29.3 \\
274: Peak &DB &53364.9113-53369.5965 &90089 \& 90427&13040 &1.95$\pm$0.02 &0.66$\pm$0.01 &5.5&32.2 \\
275: Decay1 &DB &53374.0267-53381.6610 &90427 \& 90014&23504 &2.04$\pm$0.02 &0.74$\pm$0.01 &4.2&24.6 \\
276: Decay2 &DB &53385.8724-53395.4413 &90014 &38672 &2.11$\pm$0.03 &0.83$\pm$0.03 &2.6 &15.2 \\
277: Decay3 &DB &53398.4560-53414.0726 &90014 &21296 &2.20$\pm$0.04 &0.94$\pm$0.03 &1.5 &8.8 \\
278: Decay4 &HB &53416.0726-53424.4139 &90014 &20160 &2.05$\pm$0.15 &1.13$\pm$0.05 &0.29 &1.7 \\
279: \hline\hline \noalign{\smallskip}
280: \multicolumn{9}{l}{$a$: HB: Low-intensity (Horizontal) Branch; DB: high-intensity (Diagonal) branch}\\
281: \multicolumn{9}{l}{$b$: Normalised to the Crab. Average Crab values are
282: $<SC>_{\rm Crab}=0.34$ and $<HC>_{\rm Crab}=0.47$} \\
283: \multicolumn{9}{l}{$c$: $\times 10^{-8}$ erg cm$^{-2}$ s$^{-1}$ in the 3--30 keV band} \\
284: \multicolumn{9}{l}{$d$: $\times 10^{37}$ erg s$^{-1}$ in the 3--30 keV band} \\
285: \end{tabular}
286: \end{table*}
287: %-------------------------------------------------------------
288:
289: %---------------------------------outbprof.eps----------------------------
290: \begin{figure*}
291: \centering
292: % \includegraphics[width=8cm]{./figures/lcoutb.eps} \\
293: \includegraphics[width=16cm]{./10021f1.eps}
294: \caption{The outburst profiles as seen by ASM RXTE. Vertical
295: lines mark the time of the RXTE PCA observations. Different colours
296: distinguish the spectral regions. Bin size is 1 day. 1 Crab corresponds
297: to $\sim$75 c/s. {\em See the electronic edition of the Journal for a
298: colour version of this figure}.
299: }
300: \label{outbprof}
301: \end{figure*}
302: %-------------------------------------------------------------
303: %----rate_hr_1947.eps-----rate_hr_2030.eps-----rate_hr_0115.eps----rate_hr_0332.eps----------
304: \begin{figure*}
305: \begin{tabular}{cc}
306: \includegraphics[width=8cm]{./10021f2a.eps} &
307: \includegraphics[width=8cm]{./10021f2b.eps} \\
308: \includegraphics[width=8cm]{./10021f2c.eps} &
309: \includegraphics[width=8cm]{./10021f2d.eps} \\
310: \end{tabular}
311: \caption[]{PCA light curves and hardness ratio evolution. Each point represents
312: a 512-s integration. Symbols are as follow: pre-outburst (open black
313: circles, only in \ks), rise1 (red circles), rise2
314: (orange squares), rise3 (pink crosses, only in \ks), peak (black triangle-up),
315: decay1 (violet stars), decay2 (brown diamonds), decay3 (green triangle-down)
316: and decay4 (blue pluses).
317: {\em See the electronic edition of the Journal for a colour version of
318: this figure}.}
319: \label{rate_hr}
320: \end{figure*}
321: %------------------------------------------------------------------------------
322:
323: \section{The sources}
324: \subsection{4U 0115+63}
325:
326: \vcas\ is one of the most active and best studied Be/X-ray transients as it
327: was one of the first Be/X-ray binaries to be discovered. The oldest
328: available X-ray observations date back to August 1969 when the {\em Vela
329: 5B} satellite detected the source as three small outbursts separated by 180
330: days \citep{whit89}. Since then about 15 outbursts have been reported
331: \citep{reta07}. During the RXTE's life time 3 major outbursts have been
332: detected: in 1999, 2000 and 2004. Here we analyse observations from the
333: 2004 event. The neutron star orbits a B0.2Ve in a moderately eccentric orbit
334: with $P_{\rm orb}=24.3$ d, $e=0.34$ and $a_x \sin i=140.1$ lt-s
335: \citep{rapp78} and rotates with $P_{\rm spin}=3.6$ s, as inferred from the
336: pulsed X-ray emission. The distance to the source is estimated to be
337: $\sim$8 kpc \citep{reta07}.
338:
339: The X-ray spectrum shows up to five cyclotron resonance scattering features
340: (fundamental and four harmonics) and constitutes the accreting neutron star
341: with the highest number of cyclotron lines \citep{hein04}. The optical and
342: infrared emission is characterised by cyclic changes with a period of
343: $\sim$ 5 years. Another peculiarity is that X-ray outbursts in \vcas\
344: appear to come in pairs, i.e., two in every cycle. However, sometimes the
345: second outburst is missing \citep{reta07}. mHZ QPOs were detected during
346: the 1999 and 2004 outbursts \citep{hein99,cobu04}.
347:
348:
349: \subsection{\ks}
350:
351: The Be/X-ray binary \ks\ was first detected in the X-ray band on June 8,
352: 1989 by the TTM coded mask X-ray spectrograph on the Kvant module of the
353: Mir space station \citep{boro90} with a peak flux of 70 mCrab. About 35
354: days later the flux was at the limit of sensitivity of the instrument with
355: a 3$\sigma$ upper limit of 10 mCrab. X-ray pulsations with a pulse period
356: of 18.7 s were found in the X-ray flux of GRO J1948+32 in April 1994 by the
357: all-sky monitor BATSE aboard CGRO \citep{chak95}. The source reached a peak
358: pulse flux of 50 mCrab in the energy range 20-75 keV. About 25 days later
359: the flux decayed bellow the detection threshold of BATSE. Given the very
360: large error box of GRO J1948+32 no connection with this source and \ks\ was
361: realised. The source remained in quiescence until the beginning of 2001
362: \citep{galo04}. In this occasion the source reached a maximum flux of
363: 120mCrab, the largest in its history. This outburst is the object of the
364: present work. Prior to this major outburst, weak emission ($\sim$ 20
365: mCrab) and X-ray pulsations (18.7579$\pm$0.0005 s) from \ks\ had been
366: detected in 2000 November \citep{levi00,swan00}. The coincidence in the
367: value of the pulse period and the location of \ks\ inside the GRO J1948+32
368: error circle suggest that the two sources are in fact the same object.
369: Pulse timing analysis allowed to solve for the orbital parameters resulting
370: in orbital period $P_{\rm orb}=40.415\pm0.010$ d, eccentricity
371: $e=0.033\pm0.013$ and projected radius $a_x \sin i=137\pm3$ lt-s
372: \citep{galo04}. A broad-band spectral study of \ks\ using {\it BeppoSAX}
373: observations \citep{naik06} showed that the energy spectrum in the 0.1--100 keV energy
374: band has three components: a Comptonized component, a $\sim$0.6 keV
375: blackbody component, and a narrow, weak iron emission line at 6.7 keV with
376: a low column density of material in the line of sight. Although the
377: optical counterpart to \ks\ was correctly identified soon after the Kvant
378: detection \citep{gora91,gran91} its identification as a Be/X-ray binary was
379: suggested by \citet{negu03}. \ks\ is associated with a moderately reddened
380: V=14.2 B0Ve star located at $\sim$ 10 kpc.
381:
382: %--ccd_pcu2_norm_1947.eps---ccd_pcu2_norm_2030.eps----ccd_pcu2_norm_0115.eps---ccd_pcu2_norm_0332.eps---
383: \begin{figure*}
384: \begin{tabular}{cc}
385: \includegraphics[width=8cm]{./10021f3a.eps} &
386: \includegraphics[width=8cm]{./10021f3b.eps} \\
387: \includegraphics[width=8cm]{./10021f3c.eps} &
388: \includegraphics[width=8cm]{./10021f3d.eps} \\
389: \end{tabular}
390: \caption[]{Colour-colour diagrams of the four sources
391: investigated in this work. Each point represents a 512-s integration.
392: SC=7-10 keV/4-7 keV, HC=15-30 kev/10-15 keV. Symbols are as follows:
393: rise0 (open black circles, only in \ks), rise1 (red circles), rise2
394: (orange squares), rise3 (pink crosses, only in \ks), peak (black triangle-up),
395: decay1 (violet stars), decay2 (brown diamonds), decay3 (green triangle-down)
396: and decay4 (blue pluses). {\em See the electronic edition
397: of the Journal for a colour version of this figure}.}
398: \label{cd}
399: \end{figure*}
400: %------------------------------------------------------------------------------
401: %--int_sc_pcu2_1947----int_sc_pcu2_2030---int_sc_pcu2_0115---int_sc_pcu2_0332----------
402: \begin{figure*}
403: \begin{tabular}{cc}
404: \includegraphics[width=8cm]{./10021f4a.eps} &
405: \includegraphics[width=8cm]{./10021f4b.eps}\\
406: \includegraphics[width=8cm]{./10021f4c.eps} &
407: \includegraphics[width=8cm]{./10021f4d.eps}\\
408: %\includegraphics[width=8cm]{./figures/int_sc_pcu2_0115.eps} &
409: %\includegraphics[width=8cm]{./figures/int_hc_pcu2_0115.eps}\\
410: %\includegraphics[width=8cm]{./figures/int_sc_pcu2_0332.eps} &
411: %\includegraphics[width=8cm]{./figures/int_hc_pcu2_0332.eps}\\
412: \end{tabular}
413: \caption[]{Hardness-intensity diagrams. Each point represents a 512-s integration.
414: Intensity corresponds to the energy range 4-30 keV. Symbols are as follows:
415: rise0 (open black circles, only in \ks), rise1 (red circles), rise2
416: (orange squares), rise3 (pink crosses, only in \ks), peak (black triangle-up),
417: decay1 (violet stars), decay2 (brown diamonds), decay3 (green triangle-down)
418: and decay4 (blue pluses).
419: {\em See the electronic edition of the Journal for a colour version of this
420: figure}.}
421: \label{hid}
422: \end{figure*}
423: %------------------------------------------------------------------------------
424:
425:
426: \subsection{\exo}
427:
428: \exo\ was discovered by EXOSAT in 1985 \citep{parm89}. Since then it has
429: been extensively observed in the X/$\gamma$-ray band by various missions:
430: EXOSAT \citep{reyn93}, ROSAT \citep{mavr94}, RXTE \citep{reig98,reig99},
431: CGRO \citep{stol99}, INTEGRAL \citep{came05} and SWIFT \citep{kloc07}.
432: \exo\ is the prototype of X-ray variability in Be/X-ray binaries. It shows
433: a regular pattern of type I outbursts ($L_{\rm x}\approx 10^{36}$ erg
434: s$^{-1}$) during each periastron passage \citep{wils02}. In addition, \exo\
435: has shown two major outbursts ($L_{\rm x}\simmore 10^{38}$ erg s$^{-1}$):
436: the first one in 1985 that led to its discovery as an X-ray transient and
437: the second one in 2006. In this work we analyse RXTE observations of the
438: latest event.
439:
440: \exo\ contains a rotating neutron star ($P_{\rm spin}=41.7$ s) orbiting a
441: Be primary \citep{motc87,coe88,reigal98} in a moderately eccentric
442: ($e=0.4$) wide ($P_{\rm orb}=46$ d) orbit and is located at about 7.1 kpc
443: \citep{wils02}. Quasi-periodic oscillations with a frequency of 0.2 Hz were
444: detected from this source when the luminosity was close to its maximum
445: during the 1985 outburst \citep{ange89}. In common with other high-mass
446: X-ray binaries, the continuum spectral shape in the range 1-30 keV can be
447: represented by a power law with an exponential cutoff \citep{reyn93}. A
448: blackbody component has been reported to give good fits at very high
449: ($L_{\rm x}\sim 10^{38}$ erg s$^{-1}$) luminosity \citep{sun94} and
450: relatively low ($L_{\rm x}\approx 10^{36}$ erg s$^{-1}$) luminosity
451: \citep{reig99}.
452:
453:
454: \subsection{\bq}
455:
456: A similar analysis to the one presented here has already been published in
457: \citet{reig06}. For the sake of clarity and in order to apply the same
458: reduction procedure to all sources, we have re-analysed the observations in
459: \citet{reig06}. Note also that the present work includes data during the
460: rise of the outburst that were not previously analysed and that the
461: frequency range of the power spectra is two orders of magnitude larger than
462: in \citet{reig06}.
463:
464: \bq\ is a hard X-ray transient that spends most of its life in a quiescent
465: X-ray state. After more than 15 years of quiescence the Be/X-ray binary
466: \bq\ underwent a giant outburst in December 2004. The outburst was
467: predicted about 9 months before its occurrence from the brightening of the
468: optical companion \citep{gora04}. Prior to the event reported here, \bq\ had
469: went into outburst is three other occasions. The 1973 outburst led to its
470: discovery as a bright X-ray source by the {\em Vela 5B} satellite
471: \citep{terr84}. In 1983 it reappeared in the form of three small
472: outbursts \citep{tana83}. EXOSAT observations of this activity period
473: resulted in the discovery of X-ray pulsations with $P_{\rm spin}=4.4$ s and
474: the determination of the orbital parameters, $P_{\rm orb}=34.25$ d and
475: $e=0.31$ \citep{stel85}. In 1989 {\em Ginga} detected \bq\ again
476: and allowed the discovery of a cyclotron resonant scattering feature at
477: 28.5 keV \citep{maki90} and QPOs at 0.051 Hz \citep{take94}. The optical
478: counterpart to \bq\ is an O8-9Ve star at a distance of $\sim 7$ kpc,
479: showing H$\alpha$ in emission and strong and variable infrared emission
480: \citep{bern84,corb86,coe87,negu99}.
481:
482:
483: \section{Observations}
484: \label{obs}
485:
486: We have analysed data obtained by the Proportional Counter Array (PCA) and
487: the All Sky Monitor (ASM) onboard the Rossi X-ray Timing Explorer (RXTE).
488: The ASM data consist of daily flux averages in the energy range 1.3-12.1
489: keV. The PCA covers the lower part of the energy range 2--60 keV, and
490: consists of five identical coaligned gas-filled proportional units.
491: Due to RXTE's low-Earth orbit, the data consist of a number of contiguous
492: data intervals (typically 1 hr long) interspersed with observational gaps
493: produced by Earth occultations of the source and passages of the satellite
494: through the South Atlantic Anomaly. We shall refer to each of these data
495: intervals as "pointings". Data taken during satellite slews, passage
496: through the South Atlantic Anomaly and Earth occultation were removed.
497:
498: During the RXTE life time the response of the detectors varied due to
499: ageing. Also, gain changes are applied occasionally, making the channel
500: boundaries for a given energy range change with time, and also slightly
501: affecting the effective areas of the detectors. Each gain change is the
502: start of a new "gain epoch". To minimize this effect, data from the
503: same "epoch" was used. All the data analysed in this work correspond to
504: "epoch" 5. In addition, the source X-ray colours were normalised to those
505: of the Crab that are closest in time. Table~\ref{reg} gives relevant
506: information about the observations.
507:
508: \section{Data reduction and analysis}
509: \label{red}
510:
511: In this section we describe the different tools (CD/HID and power
512: spectra) used in the data analysis. ASM light curves were obtained to study
513: the outburst profiles. PCA data can be collected and telemetered to the
514: ground in many different ways depending on the intensity of the source and
515: the spectral and timing resolution desired. In this work we used {\em
516: Standard2} to obtain colour-colour and hardness-intensity diagrams and
517: time-average energy spectra. In addition, various high-time resolution
518: modes ({\em good\_xenon, Event\_125$\mu$s\_64M\_0\_1s and
519: SB\_62$\mu$s\_0\_49}) were used to extract the light curves with which the
520: power spectra were produced.
521:
522: \subsection{Outburst profiles}
523:
524: The outburst profiles were obtained from the 1-day binned ASM light curves
525: and are shown in Figure~\ref{outbprof}. Table~\ref{info} gives the duration
526: and peak intensity of each individual outburst together with some relevant
527: information about the systems. The time of the PCA observations are marked
528: in Fig.~\ref{outbprof} with different colours indicating different spectral
529: regions (see the electronic edition of the Journal for a colour version of
530: this figure). Each vertical mark corresponds to a pointing. In the energy
531: range of the ASM, 1 Crab corresponds to $\sim$75 c s$^{-1}$. In
532: general, the PCA sampling of the outbursts is excellent. The only part of
533: the outburst not covered by the PCA data is the 7-10 first days, which can
534: be attributed to the transient nature of the sources (it takes some time to
535: issue an alert warning and reschedule the observations). The only source
536: with available data covering the entire outburst is \ks.
537:
538: \subsection{Colour-colour and hardness-intensity diagrams}
539: \label{col}
540:
541: Colour-colour (CD) and hardness-intensity (HID) diagrams are powerful
542: tools that allow the investigation of spectral variability without the
543: assumption of any spectral model.
544: In order to study the evolution of the spectral shape as the
545: outburst evolves, the RXTE light curve was divided into different parts
546: depending on count rate and whether the outburst was on the rise, peak or
547: decay. Given the good sampling of the PCA observations, these
548: three phases were further subdivided into shorter intervals. Thus each
549: interval defines a spectral region characterised by count rate, soft and
550: hard colours. Table~\ref{reg} gives relevant information about each region
551: and source.
552:
553: The CD and HID were obtained from the background-subtracted {\em Standard
554: 2} PCA light curves in the energy ranges $c_1=4-7$ keV, $c_2=7-10$ keV,
555: $c_3=10-15$ keV and $c_4=15-30$ keV. The soft colour (SC) was defined as
556: the ratio $c_2/c_1$ and the hard colour (HC) as the ratio $c_4/c_3$. These
557: ratios are expected to be insensitive to interstellar absorption effects.
558: The hydrogen column densities to the systems, obtained from model fits to
559: the X-ray spectra, are in the range $0.6-3\times 10^{22}$ cm$^{-2}$. The
560: evolution with time of the PCU2 count rate and the soft and hard colours is
561: shown in Fig.~\ref{rate_hr}.
562:
563: The CD was constructed by plotting the hard colour as function of the soft
564: colour (Fig.~\ref{cd}) and the HID by plotting the count rate in the 4-30
565: keV band as a function of one of the colours (Fig.~\ref{hid}). To recover
566: the true values of the colours of the sources, the soft colour should be
567: multiplied by 0.34 and the hard color by 0.47 (quoted values are averages
568: of the Crab colours during the observations). The variation of the Crab
569: colours computed as the root-mean-square, i.e. the standard deviation over
570: the mean colour, was, on average, 2.1\% and 1.6\% for the soft and hard
571: colours, respectively.
572:
573: To reduce the uncertainty in the individual 16-s points (default
574: resolution of the {\em Standard 2} mode) the light curves were rebinned to
575: 512-s bins. Also, data for which the resulting relative errors were larger
576: than 5\% were excluded. When the count rate per PCU was lower than 10 c
577: s$^{-1}$, relative errors up to 10\% were allowed.
578:
579: %--------------qpo_0115.eps-----------------------------------------------
580: \begin{figure}
581: \centering
582: \includegraphics[width=8cm]{./10021f5.eps}
583: \caption{QPOs near the peak of the outburst in \vcas. Time zero is
584: indicated in MJD in each panel. The time resolution is 4 s.}
585: \label{qpo_0115}
586: \end{figure}
587: %-------------------------------------------------------------
588:
589: %-------------------------------------------------------------
590: \begin{table*}
591: \caption{Power spectral parameters (centroid frequency, FWHM and $rms$) for the
592: $L_b$, $L_l$ and $L_u$ components. Errors are 90\% confidence level. The
593: centroid frequency of the zero-centred Lorentzian was fixed during the fit.
594: HB stands for horizontal branch.}
595: \label{specres1}
596: \centering
597: \begin{tabular}{lc|ccc|ccc|cccc}
598: \hline\hline
599: Spectral&$L_x$ &\multicolumn{3}{c}{$L_b$} &\multicolumn{3}{c}{$L_l$} &\multicolumn{3}{c}{$L_u$} & \\
600: region &$\times 10^{37}$&$\nu_b$ &$FWHM_b$ &$rms_b$ &$\nu_l$ &$FWHM_l$ &$rms_l$ &$\nu_u$ &$FWHM_u$ &$rms_u$&$\chi^2$/ \\
601: &erg s$^{-1}$ & &$10^{-3}$ (Hz) &(\%) &(Hz) &(Hz) &(\%) &(Hz) &(Hz) &(\%) &dof$^*$ \\
602: \hline \hline \noalign{\smallskip}
603: \multicolumn{12}{c}{{\bf \vcas}} \\
604: \hline \noalign{\smallskip}
605: rise1 &7.5 &0 &2.0$^{fixed}$ &19$\pm$3 &0 &1.6$\pm$0.3 &8.7$\pm$0.6 &-- &-- &-- &1.4/78 \\
606: rise2 &11.8 &0 &2.0$^{fixed}$ &12$\pm$2 &0 &0.87$\pm$0.06 &10.2$\pm$0.2 &0.1$\pm$0.1 &4.1$\pm$0.3 &8.2$\pm$0.3 &1.3/76 \\
607: peak &11.0 &** &** &** &0 &0.78$\pm$0.08 &9.5$\pm$0.3 &0.4$\pm$0.3 &4.9$\pm$0.7 &4.0$\pm$0.7 &2.0/76 \\
608: decay1 &7.9 &** &** &** &0 &0.72$\pm$0.07 &10.0$\pm$0.3 &1.5$\pm$0.7 &4.3$\pm$0.9 &3.9$\pm$0.4 &1.3/76 \\
609: decay2 &6.0 &0 &3.0$\pm$2.0 &13$\pm$3 &0 &0.76$\pm$0.10 &9.7$\pm$0.3 &0 &5.3$\pm$1.0 &3.7$\pm$0.7 &1.9/74 \\
610: decay3 &5.0 &** &** &** &0 &0.73$\pm$0.04 &11.0$\pm$0.2 &-- &-- &-- &1.3/75 \\
611: decay4(HB)&0.9 &-- &-- &-- &0.10$\pm$0.01 &0.34$\pm$0.01 &20.0$\pm$0.3 &-- &-- &-- &1.1/78 \\
612: \hline \noalign{\smallskip}
613: \multicolumn{12}{c}{{\bf \ks}} \\
614: \hline \noalign{\smallskip}
615: rise0(HB)&0.1 &-- &-- &-- &0 &0.087$\pm$0.005 &35.4$\pm$0.7 &-- &-- &-- &2.1/83 \\
616: rise1 &0.8 &-- &-- &-- &0 &0.14$\pm$0.01 &17.0$\pm$0.4 &-- &-- &-- &1.6/86 \\
617: rise2 &2.3 &0 &9$\pm$3 &9.7$\pm$0.5 &0 &0.40$\pm$0.04 &14.0$\pm$0.5 &0 &2.3$\pm$0.8 &6.3$\pm$1.2 &0.9/82 \\
618: rise3 &4.3 &0 &15$\pm$10 &8.7$\pm$0.9 &0 &0.57$\pm$0.04 &13.2$\pm$0.2 &0 &4.0$\pm$0.8 &6.2$\pm$0.4 &1.7/82 \\
619: peak &6.3 &0 &12$\pm$2 &9.5$\pm$0.5 &0 &0.65$\pm$0.04 &14.3$\pm$0.2 &1.1$\pm$0.3 &3.6$\pm$0.2 &6.5$\pm$0.5 &1.3/81 \\
620: decay1 &4.7 &0 &9$\pm$2 &10.2$\pm$0.7 &0 &0.48$\pm$0.05 &13.8$\pm$0.5 &0 &2.7$\pm$0.4 &7.7$\pm$0.6 &1.0/82 \\
621: decay2 &2.6 &0 &14$\pm$3 &9.2$\pm$0.8 &0 &0.61$\pm$0.04 &14.0$\pm$0.4 &-- &-- &-- &1.5/84 \\
622: decay3 &1.3 &0 &11$\pm$9 &8.4$\pm$0.9 &0 &0.52$\pm$0.06 &12.0$\pm$0.6 &-- &-- &-- &1.0/84 \\
623: decay4(HB)&0.01 &-- &-- &-- &0 &0.10$\pm$0.02 &15.8$\pm$0.6 &-- &-- &-- &1.4/87 \\
624: \hline \noalign{\smallskip}
625: \multicolumn{12}{c}{{\bf \exo}} \\
626: \hline \noalign{\smallskip}
627: rise1 &8.4 &0 &44$\pm$6 &15.2$\pm$0.5 &0 &0.50$\pm$0.02 &21.7$\pm$0.3 &0 &2.9$\pm$0.2 &10.5$\pm$0.2 &1.7/81 \\
628: rise2 &12.7 &0 &15$\pm$6 &6.9$\pm$0.4 &0.07$\pm$0.01 &0.55$\pm$0.02 &24.2$\pm$0.2 &0.79$\pm$0.15 &1.8$\pm$0.1 &9.8$\pm$0.7 &1.8/79 \\
629: peak &14.5 &0 &3.3$\pm$2.0 &9.7$\pm$2.1 &0.20$\pm$0.01 &0.50$\pm$0.02 &20.9$\pm$0.7 &0.79$\pm$0.15 &2.4$\pm$0.2 &12.3$\pm$1.4 &2.5/75 \\
630: decay1 &9.7 &0 &3.4$\pm$1.0 &10.2$\pm$1.0 &0.17$\pm$0.02 &0.64$\pm$0.09 &22.0$\pm$1.5 &0.74$\pm$0.60 &2.3$\pm$0.8 &10.4$\pm$4.9 &1.4/76 \\
631: decay2 &3.4 &0 &12$\pm$5 &8.1$\pm$0.9 &0.11$\pm$0.04 &0.83$\pm$0.07 &21.7$\pm$1.3 &0.54$\pm$0.60 &5.3$\pm$1.2 &9.2$\pm$0.9 &1.3/80 \\
632: decay3 &0.7 &0 &42$\pm$20 &9.2$\pm$0.8 &0.20$\pm$0.09 &0.64$\pm$0.07 &15.6$\pm$0.7 &0.95$\pm$0.80 &4.3$\pm$0.8 &7.8$\pm$0.4 &1.1/78 \\
633: decay4(HB)&0.4 &-- &-- &-- &0 &0.046$\pm$0.005&22.1$\pm$0.9 &-- &-- &-- &2.0/85 \\
634: \hline \noalign{\smallskip}
635: \multicolumn{12}{c}{{\bf \bq}} \\
636: \hline \noalign{\smallskip}
637: rise1 &13.5 &0 &2.5$\pm$0.4 &6.1$\pm$0.3 &$<0.007$ &0.70$\pm$0.02 &8.9$\pm$0.1 &-- &-- &-- &2.1/80 \\
638: rise2 &29.3 &0 &7.0$\pm$2.5 &3.5$\pm$0.3 &0.33$\pm$0.04 &1.54$\pm$0.06 &7.6$\pm$0.2 &-- &-- &-- &1.5/79 \\
639: peak &32.2 &0 &7.6$\pm$4.0 &3.8$\pm$0.5 &0.8$\pm$0.1 &2.0$\pm$0.09 &6.0$\pm$0.3 & -- &-- &-- &2.5/76 \\
640: decay1 &24.6 &0 &3.7$\pm$2.0 &3.1$\pm$0.3 &0.04$\pm$0.02 &0.85$\pm$0.05 &8.6$\pm$0.1 &-- &-- &-- &1.2/80 \\
641: decay2 &15.2 &0 &2.5$\pm$0.3 &6.8$\pm$1.1 &0.08$\pm$0.01 &0.86$\pm$0.01 &10.2$\pm$0.2 &-- &-- &-- &2.3/81 \\
642: decay3 &8.8 &0 &1.9$\pm$0.9 &8.1$\pm$0.6 &0.04$\pm$0.01 &0.54$\pm$0.01 &13.1$\pm$0.3 &-- &-- &-- &1.3/81 \\
643: decay4(HB)&1.7 &0 &7.2$\pm$4.0 &9.5$\pm$1.1 &0.09$\pm$0.01 &0.21$\pm$0.01 &20.9$\pm$0.3 &-- &-- &-- &1.5/85 \\
644: \hline \hline \noalign{\smallskip}
645: %\multicolumn{12}{l}{$a$: $\times 10^{37}$} \\
646: \multicolumn{12}{l}{$*$: including components in Table~\ref{specres2} and \ref{specres3}} \\
647: \multicolumn{12}{l}{$**$: affected by the mHz QPO. See Fig.~\ref{qpo_0115} and Table~\ref{specres3}. See also \citet{cobu04}}\\
648: %\multicolumn{12}{l}{HB: Horizontal branch (see text)} \\
649: \end{tabular}
650: \end{table*}
651: %-------------------------------------------------------------
652: %-------------------------------------------------------------
653: \begin{table}
654: \caption{Power spectral parameters (centroid frequency, FWHM and $rms$) for the
655: $L_{u'}$ component in \exo. Errors are 90\% confidence level.}
656: \label{specres2}
657: \centering
658: \begin{tabular}{lccc}
659: \hline\hline
660: Spectral &$\nu_{u'}$ &$FWHM_{u'}$ &$rms_{u'}$ \\
661: region & (Hz) & (Hz) &(\%) \\
662: \hline \noalign{\smallskip}
663: rise1 &-- &-- &-- \\
664: rise2 &2.5$\pm$0.1 &5.7$\pm$0.2 &7.2$\pm$0.3 \\
665: peak &3.1$\pm$0.6 &6.6$\pm$0.4 &7.1$\pm$1.0 \\
666: decay1 &2.3$^{+1.1}_{-0.4}$ &6.7$\pm$0.7 &7.8$\pm$2.2 \\
667: decay2 &-- &-- &-- \\
668: decay3 &-- &-- &-- \\
669: decay4 &-- &-- &-- \\
670: \hline \hline \noalign{\smallskip}
671: \end{tabular}
672: \end{table}
673: %-------------------------------------------------------------
674:
675: %-------------------------------------------------------------
676: \begin{table*}
677: \caption{Power spectral parameters (centroid frequency, FWHM and $rms$) for the
678: $L_{mQPO}$, $L_{LF}$ and $L_s$ components. Errors are 90\% confidence level. The
679: centroid frequency of the zero-centred Lorentzian was fixed during the fit.
680: HB stands for horizontal branch.}
681: \label{specres3}
682: \centering
683: \begin{tabular}{lc|ccc|ccc|ccc}
684: \hline\hline
685: Spectral&$L_x$ &\multicolumn{3}{c}{$L_{mQPO}$} &\multicolumn{3}{c}{$L_{LF}$} &\multicolumn{3}{c}{$L_s$} \\
686: region &$\times 10^{37}$&$\nu_{mQPO}$ &$FWHM_{mQPO}$ &$rms_{mQPO}$&$\nu_{LF}$ &$FWHM_{LF}$ &$rms_{LF}$ &$\nu_s$ &$FWHM_s$ &$rms_s$ \\
687: &erg s$^{-1}$ & (Hz) & (Hz) &(\%) & (Hz) & (Hz) &(\%) &(Hz) &(Hz) &(\%) \\
688: \hline \hline \noalign{\smallskip}
689: \multicolumn{11}{c}{{\bf \vcas}} \\
690: \hline \noalign{\smallskip}
691: rise1 &7.5 &-- &-- &-- &0 &0.14$\pm$0.03 &10.7$\pm$0.7 &-- &-- &-- \\
692: rise2 &11.8 &-- &-- &-- &0.038$\pm$0.008 &0.06$\pm$0.01 &6.3$\pm$0.5 &-- &-- &-- \\
693: peak &11.0 &3.7$\pm$0.4 &4.3$\pm$0.7 &8.8$\pm$0.4 &0.006$\pm$0.003 &0.09$\pm$0.01 &8.0$\pm$0.3 &-- &-- &-- \\
694: decay1 &7.9 &2.6$\pm$0.4 &5.5$\pm$1.0 &8.0$\pm$0.8 &0.049$\pm$0.006 &0.06$\pm$0.01 &7.3$\pm$0.6 &-- &-- &-- \\
695: decay2 &6.0 &-- &-- &-- &0.059$\pm$0.005 &0.06$\pm$0.01 &6.3$\pm$0.6 &-- &-- &-- \\
696: decay3 &5.0 &3.6$\pm$0.6 &4.2$\pm$1.0 &5.4$\pm$0.3 &0.065$\pm$0.003 &0.05$\pm$0.01 &6.8$\pm$0.3 &-- &-- &-- \\
697: decay4(HB)&0.9 &-- &-- &-- &0 &0.03$\pm$0.01 &12.8$\pm$1.0 &-- &-- &-- \\
698:
699: \hline \noalign{\smallskip}
700: \multicolumn{11}{c}{{\bf \bq}} \\
701: \hline \noalign{\smallskip}
702: rise1 &13.5 &-- &-- &-- &0.038$\pm$0.003 &0.059$\pm$0.006 &4.7$\pm$0.1 &0.22$\pm$0.01 &0.12$\pm$0.01 &4.1$\pm$0.1 \\
703: rise2 &29.3 &-- &-- &-- &0.029$\pm$0.003 &0.079$\pm$0.009 &4.4$\pm$0.1 &0.25$\pm$0.01 &0.26$\pm$0.03 &5.8$\pm$0.3 \\
704: peak &32.2 &-- &-- &-- &0.010$\pm$0.009 &0.094$\pm$0.020 &4.1$\pm$0.6 &0.23$\pm$0.01 &0.57$\pm$0.03 &8.5$\pm$0.4 \\
705: decay1 &24.6 &-- &-- &-- &0.022$\pm$0.007 &0.078$\pm$0.009 &4.5$\pm$0.4 &0.24$\pm$0.01 &0.15$\pm$0.03 &4.0$\pm$0.3 \\
706: decay2 &15.2 &-- &-- &-- &0.046$\pm$0.002 &0.043$\pm$0.003 &5.3$\pm$0.2 &0.23$\pm$0.01 &0.13$\pm$0.01 &6.2$\pm$0.1 \\
707: decay3 &8.8 &-- &-- &-- &0.045$\pm$0.002 &0.053$\pm$0.005 &9.4$\pm$0.4 &0.23$\pm$0.01 &0.06$\pm$0.02 &3.2$\pm$0.9 \\
708: decay4(HB)&1.7 &-- &-- &-- &0.043$\pm$0.005 &0.050$\pm$0.030 &14.4$\pm$0.6 &-- &-- &-- \\
709: \hline \hline \noalign{\smallskip}
710: %\multicolumn{11}{l}{HB: Horizontal branch} \\
711: %\multicolumn{8}{l}{$a$: $\times 10^{37}$} \\
712: %\multicolumn{5}{l}{$a$: 95\% confidence level} \\
713: \end{tabular}
714: \end{table*}
715: %-------------------------------------------------------------
716:
717:
718: %------------------------psd_1947-------------------------------------
719: \begin{figure*}
720: \centering
721: \includegraphics[width=17cm]{./10021f6.eps}
722: \caption{Power spectra of \ks\ and fit functions of various spectral
723: regions.
724: The number at the top left of each panel is the X-ray luminosity in
725: units of $10^{37}$ erg s$^{-1}$. HB stands for horizontal branch.
726: The different lines indicate the individual Lorentzian components of
727: the fit: $L_b$ (dotted), $L_l$
728: (thick solid black),
729: $L_u$ (thin solid black). The pulse peaks
730: (fundamental and harmonics) are represented by thin dashed orange
731: lines.}
732: \label{psd2}
733: \end{figure*}
734: %-------------------------------------------------------------
735: %----------------------psd_2030---------------------------------------
736: \begin{figure*}
737: \centering
738: \includegraphics[width=17cm]{./10021f7.eps}
739: \caption{Power spectra of \exo\ and fit functions of various spectral
740: regions. The number at the top left of each panel is the X-ray luminosity
741: in units of $10^{37}$ erg s$^{-1}$. HB stands for horizontal branch.
742: The different lines indicate the individual Lorentzian components of
743: the fit: $L_b$ (dotted), $L_l$
744: (thick solid black), $L_u$ (thin solid black),
745: $L_{u'}$ (thick solid grey). The pulse peaks
746: (fundamental and harmonics) are represented by thin dashed orange
747: lines.}
748: \label{psd3}
749: \end{figure*}
750: %-------------------------------------------------------------
751: %--------------------psd_0115-----------------------------------------
752: \begin{figure*}
753: \centering
754: \includegraphics[width=17cm]{./10021f8.eps}
755: \caption{Power spectra of \vcas\ and fit functions of various spectral
756: regions. The number at the top left of each panel is the X-ray luminosity in
757: units of $10^{37}$ erg s$^{-1}$. HB stands for horizontal branch.
758: The different lines indicate the individual Lorentzian components of
759: the fit: $L_b$ (dotted), $L_l$
760: (thick solid black), $L_{LF}$ (thick dashed),
761: $L_u$ (thin solid black). The pulse peaks
762: (fundamental and harmonics) are represented by thin dashed orange
763: lines. In "peak" and "decay1" the $L_b$ component is affected by the
764: 3 mHz QPO, $L_{\rm VLF}$ (see text and Fig.~\ref{qpo_0115}).}
765: \label{psd1}
766: \end{figure*}
767: %-------------------------------------------------------------
768: %-----------------------psd_0332--------------------------------------
769: \begin{figure*}
770: \centering
771: \includegraphics[width=17cm]{./10021f9.eps}
772: \caption{Power spectra of \bq\ and fit functions of various spectral
773: regions. The number at the top left of each panel is the X-ray luminosity
774: in units of $10^{37}$ erg s$^{-1}$. HB stands for horizontal branch.
775: The different lines indicate the individual Lorentzian components of
776: the fit: $L_b$ (dotted), $L_l$
777: (thick solid black), $L_{LF}$ (thick dashed),
778: $L_{s}$ (dashed-dotted), $L_u$
779: (thin solid black). The pulse peaks (fundamental and harmonics) are
780: represented by thin dashed orange lines.}
781: \label{psd4}
782: \end{figure*}
783: %-------------------------------------------------------------
784:
785:
786:
787: \subsection{Power spectral analysis}
788: \label{powspec}
789:
790: Power spectra were created by performing a fast Fourier transform on the
791: light curves according to the following procedure: one light curve in
792: the energy range 2-20 keV (channels 0--49) was extracted for each spectral
793: region with a time resolution of $2^{-8}$ seconds. Each light curve was
794: divided into 512-s segments and an FFT was calculated for each segment. The
795: final power spectra resulted after averaging all the individual power
796: spectra and rebinning logarithmically in frequency. All power spectra are
797: presented in the $rms$ normalisation, where the power integrated over a
798: certain frequency interval equals the squared $rms$ of the source
799: \citep{bell90,miya91}.
800:
801:
802: The plots of the power spectra are shown in the $\nu \times P_{\nu}$
803: representation, where each power is multiplied by the corresponding
804: frequency. This representation helps visualize at which frequency the
805: contribution to the total $rms$ variability is maximum
806: \citep{bell02}. Note that the frequency at which maximum power is attained
807: in the $\nu \times P_{\nu}$ representation does not equal the centroid
808: frequency of the Lorentzian, $\nu_0$, but $\nu_{\rm
809: max}=(\nu_0^2+(FWHM/2)^2)^{1/2}$, where FWHM is the Lorentzian full-width
810: at half maximum.
811:
812: The statistical Poisson noise ($P_{N}=2/R_T$, where $R_T$ is the source
813: count rate) was modified by dead time effects. Dead-time effects are
814: expected to affect the power spectra at high frequencies, especially during
815: the high-flux states, where the count rate is over $\sim
816: 1000$ c s$^{-1}$. One way of checking whether the dead-time effects are
817: important is by measuring the power at high frequencies in a Leahy
818: normalised power spectrum. The expected power density should be at a level
819: of 2 \citep{leah83,klis89,jern00}. As an example, the power density of \exo\
820: during the peak of the outburst ("peak"), calculated for $\nu > 100$ Hz, is
821: 1.935, while that obtained during the end of the outburst ("decay3") is 1.995.
822: The correction of the Poisson noise $P_N(\nu)$ for dead-time effects was
823: calculated as described in \citet{nowa99}.
824:
825:
826: The study of the rapid aperiodic variability in X-ray pulsars is hampered
827: by the peaks due to the periodic pulsations. The spin period and its
828: harmonics show up as narrow peaks in the power spectrum that distort the
829: continuum. The width of these peaks depends on the frequency resolution,
830: which in turn, depends on the length of the segments ($\Delta \nu=1/T$,
831: where $T$ is the total duration of the segment). On one hand, a high
832: frequency resolution is required in order to have well defined peaks. On
833: the other hand, higher frequency resolution implies less number of power
834: spectra to average and worse signal-to-noise, especially at higher
835: frequencies. As mentioned above we found a good compromise with $T=512$
836: seconds.
837:
838: Multiple Lorentzian profiles provide a good description of the power
839: spectra in low-mass accreting neutron star binaries \citep[see
840: e.g.][]{stra02,stra03,reig04} and black-hole systems \citep[see
841: e.g.][]{nowa00,bell02,pott03}. To facilitate the comparison between the
842: noise components and timing rapid variability of all types of X-ray
843: binaries we have followed this approach here too. The peaks from the
844: pulsations were fitted to Lorentzian functions with fixed frequency (at the
845: expected value according to the spin period and its harmonics) and width
846: (=0.001 Hz $\approx 1/T$). Tables~\ref{specres1} and ~\ref{specres2} give
847: the results of the power spectral analysis. Figures \ref{psd1}--\ref{psd4}
848: show the power spectra and noise components of various spectral regions.
849: Any given power spectrum can be fitted with the sum of two to four broad
850: Lorentzian profiles (excluding the peaks), although we identify seven
851: different types of noise components.
852:
853: \subsection{Energy spectra}
854:
855: The main objective of this work is to investigate the aperiodic
856: variability of high-mass X-ray pulsars in correlation with the spectral
857: states as defined in CD/HIDs, in the same way as it has been extensively
858: done in low-mass X-ray binaries. A detailed spectral analysis is out of
859: the scope of this work and will be performed in a future paper. The reader
860: is referred to \citet{wils08} (\exo), \citet{mowl06} (\bq), \citet{naka06}
861: (\vcas) and \citet{naik06} (\ks) for a detailed analysis of the energy
862: spectra during type II outbursts. However, for the sake of clarity in the
863: discussion, it is illustrative to know the X-ray luminosity of the different
864: spectral regions. Thus we obtained a time-averaged spectrum for each
865: spectral region. The 3-30 keV spectrum was fitted with an absorbed
866: power-law model modified at high energies by an exponential cutoff. In
867: addition, a Gaussian profile at $\sim$ 6.4 keV, that accounts for iron
868: emission line, was required in most cases. The X-ray flux and luminosity in
869: the 3-30 keV energy range is given in Table~\ref{reg}. The distance to the
870: sources is given in Table~\ref{info}.
871:
872: \section{Results}
873: \label{res}
874:
875:
876: \subsection{Outburst profiles}
877:
878:
879: The total duration of the \vcas\ outburst was $\sim$55 days ($\sim$ MJD
880: 53245--53300) but the rise was shorter ($\sim$11 days) than the decay
881: ($\sim$27 days). The brightest phase of the outburst displays a multi-peak
882: profile and lasted for at least 15 days. The source spent about one third
883: of the time in this bright state. The maximum flux was achieved on
884: $\sim$MJD53264 and amounted to $\sim$280 mCrab. The PCA observations of
885: \vcas\ began at the mid point of the rising part of the outburst on
886: September 6, 2004 (MJD 53254) and finished on October 25, 2004 (MJD
887: 53304).
888:
889: The ASM light curves of \ks\ and \exo\ show very similar characteristics,
890: despite the difference in overall luminosity (the \exo\ outburst was a
891: factor of 4 brighter). Not only the outburst profile and duration
892: ($\sim$165 days and $\sim$155 days, respectively) are similar but also the
893: pre- and post-outburst X-ray activity. In these two sources the main
894: outburst is preceded by a minor one, which started one orbital period
895: before the onset of the major event. Likewise, the major outburst was
896: followed by a series of smaller outbursts separated by a time interval
897: consistent again with the orbital period (40 days in \ks\ and 46 days in
898: \exo). The outburst profiles are quite symmetric with the rise and decay
899: lasting for about 60-70 days. The \ks\ ASM light curve contains a gap
900: during the rise due to its proximity to the Sun at that time.
901:
902: \bq\ showed the most intense outburst. The maximum flux recorded in the
903: energy range 1.3-12.1 keV reached $\sim$1.5 Crab. The outburst began in
904: 2004 November (MJD 53330) and reached maximum flux about one month later
905: ($\sim$ MJD 53368). The total duration of the outburst was $\sim 105$ days.
906: As in \vcas, the decay was slower than the rise. The difference appeared at
907: the end of the decay, where a longer tail can be seen
908: (Fig.~\ref{outbprof}).
909:
910:
911: There appears to be a relationship between the shape and duration of the
912: outbursts and the orbital period of the system. \ks\ and \exo\ have the
913: wider orbits and show the longer and more symmetric outbursts. The outburst
914: profiles of \vcas\ and \bq\ have positive skewed (elongated tail at the
915: right) and last shorter. The skewness statistics (the third moment
916: about the mean normalised to the standard deviation,
917: $\frac{1}{N}\sum{\left(x_i-\bar{x}\right)^3}/\sigma^3$)
918: gives the following
919: values for \vcas, \bq, \ks\ and \exo: 0.24, 0.58, 0.16 and --0.035,
920: respectively. In Be/X-ray binaries, the reservoir of matter available for
921: accretion onto the neutron star companion comes from the circumstellar disk
922: around the Be star's equator. The duration of the X-ray outbursts is
923: expected to be proportional to the amount of accreted material. Thus longer
924: outburst duration implies larger disks. This result agrees with the
925: correlation between the orbital period and strength of the H$\alpha$ line
926: \citep{reig97,reig07}. The longer the orbital period, the wider the orbit,
927: the larger the disk. This correlation is explained by the truncation of the
928: disk by the neutron star, which will be more efficient for narrower orbits.
929: \citet{okaz01} estimated that the Be discs in \vcas, \bq\ and \exo\ are
930: truncated at the 4:1 resonance radius, while that of \ks\ at the 3:1
931: resonance. In terms of stellar radii, assuming the stellar parameters given
932: in \citet{okaz01}, these resonance radii translate into truncation radii
933: $\sim$ 4.7 R$_*$, $\sim$ 5.6 R$_*$, $\sim$ 6.9 R$_*$ and $\sim$ 8 R$_*$
934: for \vcas, \bq, \exo\ and \ks, respectively. This numbers should be taken
935: with caution since the calculations are strongly dependent on the disc
936: viscosity and assume the same disc physical parameters in all systems.
937:
938: \subsection{Colour analysis}
939:
940: Be/X-ray binaries show two spectral branches in the CD/HID. We shall
941: refer to these branches as the low-intensity or horizontal branch and the
942: high-intensity or diagonal branch. However, not all sources display the
943: entire pattern of variability. Sometimes only one branch is visible in the
944: CD. In \ks, the scatter of the points that populate the low-intensity
945: branch is too large to clearly define the branch (see Fig.~\ref{cd}). In
946: \exo, only the hardest (in terms of SC) part of this branch is visible.
947:
948: Unless the motion in the horizontal branch is faster than $\sim$1
949: day$^{-1}$, it seems unlikely that the lack of a complete branch in \ks\
950: and \exo, at the end of the outburst, is due to observational gaps. At
951: least in \exo, the frequency of the PCA observations was $\sim 1$ day$^{-1}$.
952: More likely, the lack of a clear horizontal branch in the CD of these two
953: sources can be attributed to the fact that they did not return to
954: quiescence. Both \ks\ and \exo\ showed smaller outbursts (type I) after the
955: major outburst (type II) analysed in this work. The X-ray emission level
956: did not return to zero in these two systems (i.e. below the sensitivity
957: threshold of the PCA) as can be seen in Fig.~\ref{rate_hr}. The average
958: count rate of the last observations was 30 c s$^{-1}$ PCU$^{-1}$ in \exo\
959: and 5 c s$^{-1}$ PCU$^{-1}$ in \ks. In contrast, \vcas\ seems to have gone
960: under the detectability limit of the instruments. It is reasonable to think
961: that if the intensity had gone lower, then the horizontal branch in \ks\
962: and \exo\ would have shown up more clearly. Note that \bq\ also went back
963: to $\sim$ 0 c s$^{-1}$ and also displays two distinct branches in the CD.
964:
965: In LMXBs, the patterns observed in a CD are often also recognizable in the
966: corresponding HID. Sometimes, however, spectral branches are more distinct
967: in one of the two diagrams. Whether CD or HID presents the cleanest pattern
968: depends on source and on the quality of the data. Figure~\ref{hid}
969: displays the HID of the four X-ray pulsars analysed in this work. In
970: contrast to the CD, the two branches are clearly seen for all four sources
971: when the HC is replaced by the intensity. The junction of the two
972: branches occur when the count rate is at $\sim$10\% of the peak value
973: ($\sim$30\% in \vcas). The second column in Table~\ref{reg} indicates
974: the branch in which the source lay during those particular observations.
975: For practical reasons, the name ``horizontal branch" is also used if
976: intensity is plotted on a logarithmic scale in the HID, even though in this
977: case this branch does not appear horizontal.
978:
979: %If a logarithmic scale is used in the Y-axis (intensity) then the
980: %horizontal branch appears also as a diagonal branch. In this case, the
981: %shape of the HID resembles the "larger than" symbol ($>$). The branches
982: %join at the right-hand side of the HID, i.e. at the highest value of the
983: %SC.
984:
985: The low-intensity (horizontal) branch is populated by points with a low
986: count rate. Since the PCA monitoring began when the outbursts were already
987: in progress, the low-intensity branch contains points from the end of the
988: outbursts. The only source for which data prior to the onset of the
989: outburst are available is \ks. As can be seen in Fig.~\ref{hid}, the data
990: points (open black circles) that corresponds to the very beginning of the
991: outburst also lie in the low-intensity branch. Thus we conclude that the
992: horizontal branch represents the start and end point of the source in
993: its journey through the CD/HID. In terms of X-ray luminosity, this branch
994: appears when $L_X \simless 10^{37}$ erg s$^{-1}$.
995:
996: The evolution of the X-ray colours as the outburst evolves, particularly
997: the hard colour (HC), clearly distinguishes \bq\ from the other three
998: systems. Unlike, \ks, \vcas\ and \exo, both the SC and HC in \bq\ decrease
999: as the count rate increases (see Fig.\ref{rate_hr}). This different
1000: behaviour of the HC implies that the motion in the diagonal branch, as the
1001: count rate increases, is from the bottom right part of the CD to top left
1002: part in \vcas, \ks\ and \exo\ and from bottom left to top right in \bq.
1003: Another important difference is that the low-intensity branch is harder
1004: than the high-intensity branch in \bq, while it is softer in \exo, \ks\ and
1005: \vcas\ (Fig.~\ref{cd}). In this latter case, the horizontal branch
1006: corresponds to a low/soft (in terms of the hard colour) state that it is
1007: not seen in low-mass X-ray binaries or black-hole systems.
1008:
1009: After the peak, the source returns following the reverse track. The
1010: sources spend most of the outburst in the diagonal branch. The sources stay
1011: in this branch about 70\%--80\% of the duration of the outburst. As can be
1012: inferred from Table~\ref{reg}, typical timescales are months in the
1013: diagonal branch and weeks in the horizontal branch.
1014:
1015: \vcas\ exhibits hysteresis, that is, the X-ray colours adopt different
1016: values depending on whether the source is on the rise or the decay of the
1017: outburst, despite that the count rate is similar. This effect is also
1018: present in \bq\ and \exo, although not as significant as in \vcas. It is
1019: absent in \ks\ (see Fig.~\ref{hid}). In \exo\ and \bq\ both X-ray colours
1020: are larger during the rise than during the decay, that is, the sources
1021: present a harder spectrum during the rise. In \vcas, the changes are more
1022: complex. During the rise SC is larger and HC is smaller than during the
1023: decay. Although the larger error bars in \ks\ might prevent us from
1024: detecting hysteresis in this source, an effect with the amplitude seen in
1025: \vcas\ should also be seen in the SC-intensity diagram of \ks.
1026:
1027: Interestingly, cyclotron resonance scattering features have been observed
1028: for all sources except \ks. In particular, \vcas\ shows up to five
1029: cyclotron line harmonics \citep{cobu04}. However, while the energy of the
1030: fundamental line is found at 11 keV in \vcas\ \citep{sant99}, it appears at
1031: 27 keV in \bq\ \citep{pott05}. The cyclotron line in \exo, at 11 keV, is
1032: less significant and lacks harmonic content \citep{wils08}. According to
1033: the definition of the X-ray colours given in Sect.~\ref{col}, the HC in
1034: \vcas\ is expected to be more strongly affected by the presence of
1035: the cyclotron line and its harmonics, while in \bq\ and \exo\ its
1036: contribution to the variability in HC would be insignificant. Moreover, the
1037: energy of the fundamental cyclotron line in \vcas\ has been seen to
1038: increase from $\sim$11 kev to 16 keV as the X-ray luminosity decreased
1039: below $5\times10^{37}$ erg s$^{-1}$ \citep{naka06}.
1040: The relationship between hysteresis, as seen in the HID, and cyclotron
1041: lines needs to be verified through detailed spectral analysis.
1042:
1043:
1044: \subsection{Aperiodic variability}
1045:
1046:
1047: The power spectra of Be/X-ray binaries is dominated by the peaks of
1048: the pulse component and its harmonics. In order to study the aperiodic
1049: variability (band-limited noise and possible QPOs), the pulse noise should
1050: be removed. However, this is not an easy task. Some authors substituted the
1051: points contaminated by the pulses and their harmonics by a spline fit to
1052: the remaining points or to a power law connecting the average of a few
1053: points before and after the pulse gap \citep{bell90}. This procedure does
1054: not give satisfactory results if the peaks are too broad. Others simply
1055: remove the points from the power spectra \citep{reig06}. \citet{ange89}
1056: removed the pulse noise after fitting it with a model consisting of the
1057: first 10 harmonics of the coherent pulsar signal and a power law. In this
1058: work we have opted to maintain the peaks and fit them with Lorentzians. In
1059: addition to the distortion introduced by the pulse noise, the power spectra
1060: continuum of Be/X-ray binaries contain significant substructures only
1061: visible in certain states. Sometime the continuum shows bumps and wiggles,
1062: making it difficult to obtain good fits. This difficulty was also found by
1063: \citet{bell90}, who could not fit the power spectra of a number of
1064: high-mass X-ray binaries with simple analytical laws, such as power law and
1065: cutoff, as in low-mass X-ray binaries.
1066:
1067: In general, the power spectra of Be/X-ray binaries is characterised
1068: by band-limited noise (BLN), that is, a steeping of the power density
1069: toward high frequency and a flattening toward low frequency. Although a
1070: double broken power law may give good fits in some cases (\ks) it cannot
1071: fit all sources and all states (\exo). In addition, all sources present
1072: peaked noise in certain spectral states.
1073:
1074: Despite this complexity, we find that the power spectra of Be/X-ray
1075: binaries can be fitted acceptably with the sum of a small number of
1076: Lorentzian functions. Ignoring the peaks of the pulse component, 2--4
1077: Lorentzian profiles are needed to obtain acceptable fits. However, we
1078: identify 7 different types of noise components. 4 of these components are
1079: band-limited (BLN) and 3 QPO. The results of the power spectral fitting
1080: are given in Tables~\ref{specres1}--\ref{specres3}. Figures
1081: \ref{psd1}--\ref{psd4} show the power spectra of the four X-ray pulsars for
1082: various regions. Different line types indicate different noise components
1083: as explained below.
1084:
1085: $L_b$ (dotted line) is a zero-centred Lorentzian that describes the noise
1086: below 0.01 Hz. Due to the very low characteristic frequency of this
1087: component (often peaks outside the available frequency range) it is not
1088: always well constrained. In some states of \vcas, $L_{b}$ is modified by
1089: the appearance of a slow QPO. $L_{b}$ disappears in the horizontal branch
1090: of all sources except in \bq.
1091:
1092: $L_{l}$ (thick solid line) is also a broad component with $Q<0.4$ or
1093: centred at zero. Its characteristic frequency shifts to low frequencies
1094: when the count rate is low. This component dominates most of the power
1095: spectrum, covering the range from $\sim$0.1--10 Hz. In some cases (\bq), it
1096: is the only component fitting the high-frequency end of the power spectrum.
1097:
1098: In \vcas, \ks\ and \exo\ one or two more Lorentzians are needed to account
1099: for the high-frequency noise. We shall call these component $L_{u}$ (black
1100: thin solid line) and $L_{u'}$ (grey thick solid line), with characteristic
1101: frequencies $\nu_{u}\approx 1-3$ Hz, $\nu_{u'}\approx 4$ Hz, respectively
1102: and $Q\approx 0-0.5$. These components only appear when the count rate is
1103: high, i.e., near and at the peak the outburst.
1104:
1105: In general, $L_{b}$ accounts for the noise short-ward of the main pulse
1106: peak, while $L_{l}$ and $L_{u}$/$L_{u'}$ fit the power spectrum above the
1107: spin frequency. Together they account for the band-limited noise.
1108:
1109: The faster pulsars (\bq\ and \vcas) display peaked noise components, which
1110: we shall refer to as QPO noise, even though the $Q$ value is not always
1111: larger than 2. Table~\ref{specres3} gives the best-fit parameters of these
1112: components. $L_{\rm LF}$ (dashed line in Fig~\ref{psd1} and \ref{psd4}) is
1113: common to the two sources and has a characteristic frequency $\nu_{\rm LF}
1114: \approx 0.05$ Hz and $0.5 \simmore Q_{\rm LF}\simmore 2$. It becomes
1115: stronger as the count rate decreases. Note that although the centroid
1116: frequency of this Lorentzian component increases as the count rate
1117: decreases, its width decreases. The characteristic frequency (i.e. maximum
1118: frequency) remains fairly constant at the value indicated above ($\sim$0.05
1119: Hz). \bq\ shows, in addition, another QPO (that we shall call $L_{s}$,
1120: dotted-dashed line) whose centroid frequency coincides with the frequency
1121: of the fundamental peak (0.23 Hz) of the pulse period and remains constant
1122: throughout the outburst. The strength of $L_{s}$ increases with count rate,
1123: reaching maximum during the peak of the outburst. However, it disappears in
1124: low-intensity states, when $L_X$ goes below $\sim 10^{37}$ erg s$^{-1}$.
1125: \vcas\ also shows another QPO at $\sim$ 3 mHz. This QPO is even visible in
1126: the light curve (Fig.~\ref{qpo_0115}) and has also been reported in
1127: previous outbursts of the source \citep{hein99,cobu04}. We have called this
1128: component $L_{\rm mQPO}$.
1129:
1130: %--------------------------freqcor-----------------------------------
1131: \begin{figure}
1132: \centering
1133: \includegraphics[width=8cm]{./10021f10.eps}
1134: \caption{Correlation between the characteristic frequencies of the
1135: noise components. Be/X exhibit lower characteristic frequencies than
1136: LMXBs.
1137: }
1138: \label{freqcor}
1139: \end{figure}
1140: %-------------------------------------------------------------
1141:
1142: \section{Discussion}
1143: \label{disc}
1144:
1145: In this section, we compare the correlated spectral and timing
1146: variability of high-mass X-ray binaries (HMXBs) and low-mass X-ray
1147: binaries (LMXBs). We comment on the similarities and differences of
1148: these two types of X-ray binaries.
1149:
1150: \subsection{Comparison with low-mass X-ray binaries}
1151: \label{lmxb}
1152:
1153: The Be/X-ray binaries that display a wide dynamic range in intensity
1154: (Type II outbursts) trace two-branch patterns in their CD/HID. As in
1155: low-mass systems, the sources do not jump through the diagram but moves
1156: smoothly, following the pattern. The horizontal branch (HB) corresponds to
1157: a low-intensity state and shows the larger fractional $rms$, similar to the
1158: the island state in atolls and horizontal branch in Z sources. This branch
1159: represents the start and end states of the source through the outburst. In
1160: the high-intensity diagonal branch (DB), the noise components display higher
1161: characteristic frequencies and lower $rms$ than in the HB.
1162:
1163: The power spectra of LMXBs in the frequency range below 100 Hz are
1164: characterised by \citep{bell02,stra02,stra03,oliv03,klis06}: {\em i)} a
1165: power-law red noise component in the lowest frequency range, traditionally
1166: known as very-low frequency noise (VLFN). This component is weak or absent
1167: in the island and horizontal branches of atoll/Z sources and increases in
1168: strength as the source moves along the branches; {\em ii)} strong
1169: band-limited noise (BLN), consisting of flat-topped noise that breaks at a
1170: frequency $\nu_b \sim 0.01-50$ Hz. This component can be represented by a
1171: zero-centred Lorentzian and fits the low-frequency end of the power
1172: spectrum; {\em iii)} peaked noise at frequency $\sim 5 \nu_b$. When this
1173: peaked noise appears as a QPO then it is called $L_{LF}$ and its
1174: characteristic frequency $\nu_{LF}$. Otherwise, it is normally referred as
1175: $L_h$. Sometimes, both components are present, i.e, a narrow $L_{LF}$ and a
1176: broad $L_h$; {\em iv)} two zero-centred Lorentzians fitting the
1177: high-frequency part of the power spectrum. They normally appear when
1178: $\nu_b \simless 1$ Hz and are given the names $L_l$ and $L_u$,
1179: respectively. It is always observed that $\nu_b < \nu_{LF} < \nu_l <
1180: \nu_u$.
1181:
1182: In Be/X-ray binaries the power spectra can also be well fitted by a
1183: relatively small number of Lorentzian components. Some of these components
1184: can be identified in the power spectra of LMXBs. As in LMXBs, the
1185: band-limited noise in Be/X-ray binaries is fitted by three broad
1186: Lorentzians: $L_b$ at low frequencies and $L_l$ and $L_u$ at high
1187: frequencies. In Fig.~\ref{freqcor} we have plotted the relationship
1188: between the characteristic frequencies, i.e. $\nu_{\rm
1189: max}=(\nu_0^2+(FWHM/2)^2)^{1/2}$, of $L_b$, $L_{LF,h}$ and $L_l$ for the
1190: HMXBs \vcas\ (filled squares) and \bq\ (open squares) and for the LMXBs
1191: \citep{bell02,stra02,stra03,reig04} 4U 1728-34 (black filled circles), 4U
1192: 1608-52 (cyan filled triangles), GS 1826-24 (orange filled circles), 1E
1193: 1724-3045 (open circles), 4U 0614+09 (diamonds) and Aql X-1 (stars). For
1194: Aql X-1, $L_h$ was plotted. They are known as the WK relation (after
1195: \citet{wijn99a}, upper panel) and the PBK relation (after \citet{psal99},
1196: lower panel). The fact that the data points corresponding to the Be/X-ray
1197: binaries fall on the global correlations found for LMXBs, if extrapolated
1198: at low frequencies, supports the identification of the noise components.
1199:
1200:
1201: There are some important differences in the rapid aperiodic noise
1202: components between HMXBs and LMXBs, namely: {\em i)} the lower
1203: characteristic frequencies, {\em ii)} the lack of a clear correlation of
1204: the noise component parameters with mass accretion rate and {\em iii)} the
1205: noise component, $L_s$, associated with the peaks of the periodic
1206: modulation.
1207:
1208: In accretion-powered X-ray pulsars, QPOs fall in the milliHz range, that is
1209: few order of magnitudes lower than in LMXBs. There are about 12 high-mass
1210: X-ray pulsars that show QPOs in their power spectra. The frequencies of
1211: these QPOs range from 1--400 mHz \citep{shir02,inam04,mukh06}. Recently,
1212: \citet{kaur07} reported the discovery of a 1.3 Hz QPO in XTE J0111.2--7317,
1213: a Be/X-ray binary located in the SMC. Many of these QPOs have $Q$ values
1214: lower than 2, hence strictly speaking, they should be referred as peaked
1215: noise (conventionally, peaked noise is considered as a QPO when the quality
1216: factor $Q=\nu/FWHM > 2$). However, since these features (also $L_{LF}$ and
1217: $L_s$ in this work) represent a substantial concentration of power in a
1218: limited frequency range they are referred to as QPOs.
1219:
1220: If it is assumed that QPOs are formed as a result of Keplerian motion
1221: of inhomogeneities in an accretion disk then the longer implied timescales
1222: of the QPOs in HMXB is somehow expected as the inner radius of the
1223: accretion disk must be larger than the magnetospheric radius ($r_m\sim10^8$
1224: cm). 10-400 mHz
1225: QPOs imply disk radius of the order of $10^8-10^9$ cm ($R_d\sim
1226: (GM/4\pi^2\nu^2_{QPO})^{1/2}$). The origin of the mHz QPO is, however,
1227: unclear. Neither the Keplerian frequency model \citep{klis87b} nor the beat
1228: frequency model \citep{alpa85} can explain the origin of these QPO in all
1229: systems. QPOs below the neutron star spin frequency cannot be explained by
1230: the Keplerian frequency model because if the Keplerian frequency at the
1231: magnetosphere is less than the spin frequency then the propeller mechanism
1232: would inhibit accretion \citep{stel86}. This is the case of \vcas\ and \bq.
1233: Likewise, the beat frequency model predicts a centrifugal inhibition
1234: threshold that it is at variance with the observations \citep{fing98}. A
1235: third model, the magnetic disk precession model \citep{shir02}, attributes
1236: the mHz QPO in X-ray pulsars to warping/precession modes induced by
1237: magnetic torques near the inner edge of the accretion disk. The
1238: applicability of this model to all sources is not as yet certain
1239: \citep{mukh06}.
1240:
1241: Unlike LMXBs where the power spectral parameters and strength of the
1242: various noise components (anti)correlate with the mass accretion rate,
1243: $\dot{M}$, HMXB noise appears to depend weakly on $\dot{M}$ (we assume that
1244: the accretion rate increases/decreases during the rise/decay of the
1245: outburst), at least at high and intermediate accretion rates. Neither the
1246: characteristic frequencies nor the $rms$ of the Lorentzians that account
1247: for the band-limited noise show a clear relationship with X-ray flux in
1248: the diagonal branch, despite the change of about 1.5 order of magnitude in
1249: luminosity (see Tables~\ref{specres1}--\ref{specres3}). The lack of such
1250: correlations were also noticed in the 1985 outburst of \exo\
1251: \citep{ange89,bell90}.
1252:
1253: $L_s$ is exclusive of accreting X-ray pulsars and it
1254: is associated with the pulse noise. Its central frequency coincides with
1255: the frequency of the fundamental peak of the pulse period and suggests a
1256: strong coupling between the periodic and aperiodic noise components. This
1257: coupling implies that the instabilities in the accretion flow that give
1258: rise to the aperiodic variability must travel all the way down to the
1259: neutron star surface. The reader is referred to \citet{lazz97} and
1260: \citet{burd97} for a detailed study of the coupling between the periodic
1261: and aperiodic variability in X-ray pulsars and to \citet{reig06} for the
1262: particular case of \bq. $L_s$ is absent in the horizontal branch. In the
1263: case of \bq, the disappearance of the pulse peaks in the horizontal branch
1264: is accompanied by the disappearance of $L_s$, which supports the idea of a
1265: common physical origin.
1266:
1267: %-----------------------psdexam2--------------------------------------
1268: \begin{figure}
1269: \centering
1270: \includegraphics[width=8cm]{./10021f11.eps}
1271: \caption{Representative power spectra of high- and low-intensity
1272: states. Note the different shape of the noise at low frequencies.
1273: Flat-topped noise is seen in the HB of \ks\ (and \exo\ and \vcas)
1274: and power-law noise in \bq. Confront this figure with Fig.4 of
1275: \citet{klis94} and note the similarities between \ks \ and atoll
1276: sources and between \bq\ and Z sources.
1277: The "peak" power spectrum was shifted by $\times$100 for easy display.
1278: }
1279: \label{psdexam}
1280: \end{figure}
1281: %-------------------------------------------------------------
1282:
1283:
1284: \subsection{Are there atoll and Z sources in HMXBs?}
1285:
1286: The correlated timing and spectral analysis defines two different types of
1287: Be/X-ray binaries. The differences are not only seen in the shape of the CD
1288: but also in the number and properties of the noise components.
1289:
1290: The pattern traced by \bq\ in the CD reminds that of the Z-sources, with
1291: the diagonal branch being the analogue to the normal branch and the HMXB
1292: horizontal branch branch being the counterpart of the LMXB horizontal
1293: branch. The flaring branch would be missing in \bq. As in Z sources, the
1294: hardest spectrum corresponds to a lower count rate state. Also, the
1295: fractional rms is higher and the characteristic frequencies are lower in
1296: the horizontal branch than in the diagonal branch.
1297:
1298: In \vcas, \ks\ and \exo, the HB also corresponds to a
1299: low-intensity state and also displays the larger $rms$ of the noise
1300: components. However, the HB in these three sources is softer than the
1301: diagonal branch. Likewise, the overall shape and motion along the branches
1302: differs from that seen in \bq\ and LMXBs atoll sources. The HB in \vcas,
1303: \ks\ and \exo\ corresponds to a low-intensity spectrally soft state not
1304: seen in other types of X-ray binaries.
1305:
1306: Further similarities between \bq\ and Z sources and \vcas, \ks\ and \exo\
1307: and atolls sources are found in the shape of the very-low frequency noise
1308: ($\simless 0.1$ Hz). Fig.~\ref{psdexam} shows the power spectra of the HB
1309: and peak of the outburst for \bq\ and \ks. This figure should be compared
1310: to, e.g., Fig.4 in \citet{klis94}. The noise below $\sim$0.1 Hz in the HB
1311: is flat-topped in \ks\ (and in \exo\ and \vcas), as in the island state of
1312: atoll sources, and power-law like in \bq, as in the HB of Z sources.
1313:
1314: Initially, the differences between LMXB $Z$ and atoll sources were ascribed
1315: to a higher magnetic-field strength and higher accretion rates in $Z$
1316: sources. However, the detection in atoll sources of QPOs that are
1317: reminiscent of HBO and NBO \citep{ford98,wijn99b} and the existence of sources that
1318: display the two types of behaviour, $Z$ {\em and} atoll
1319: \citep{homa07a,homa07b},
1320: question the validity of the proposed higher magnetic field in the $Z$
1321: sources. Nevertheless, it is worth noting that \bq\ displays the
1322: most intense outburst and contains the stronger magnetic field of the four
1323: sources. The peak luminosity in \bq\ is about a factor 2, 3 and 5 that of
1324: \exo, \vcas\ and \ks, respectively. The strength of the magnetic field can
1325: be estimated from the energy of cyclotron resonant scattering features
1326: (CRSF) through the relation
1327:
1328: \[E_{\rm cyc}=11.6\frac{B}{10^{12} {\rm G}} (1+z)^{-1} \,\, {\rm keV}\]
1329:
1330: \noindent where $B$ is the magnetic field and $z$ is the gravitational
1331: redshift. As mentioned above, \ks\ is the only source for which no CRSF has
1332: been reported, while for \vcas\ $E_{\rm cyc}=12$ keV \citep{sant99}, for
1333: \exo\ $E_{\rm cyc}=11$ keV \citep{wils08}, and for \bq\ $E_{\rm cyc}=27$
1334: keV \citep{pott05}. Thus the magnetic field in \bq\ is about 2--3 times
1335: stronger than in the other three systems.
1336:
1337:
1338: It is tempting to attribute the differences between Be/X-ray binaries to
1339: the source brightness and magnetic field strength. The power-law noise at
1340: low frequencies in the HB of \bq\ could be due to the larger intensity and
1341: the presence of $L_S$ noise to a higher magnetic field. \vcas, \ks\ and
1342: \exo\ show flat-topped noise and lack $L_s$ component. However, with only
1343: one source of this type no definitive conclusion can be drawn about whether
1344: the source brightness and magnetic field strength allows the distinction
1345: between different types of systems among HMXB. Future observations will
1346: clarify this issue.
1347:
1348:
1349: \section{Conclusions}
1350: \label{con}
1351:
1352: We have performed, for the first time, a systematic analysis of the spectral
1353: (through the study of hardness ratios) and rapid aperiodic variability of
1354: four Be/X-ray pulsars during type II outbursts. Our aim was to characterise the
1355: rapid aperiodic variability in correlation with the position in the
1356: colour-colour diagram as it has been done for black-hole systems and
1357: low-mass X-ray binaries.
1358:
1359: The correlated spectral-timing behaviour in Be/X-ray binaries shares a
1360: number of similarities with low-mass X-ray binaries, namely, {\em i)}
1361: existence of spectral branches in the CD/HID. At high and intermediate flux
1362: the sources move along the diagonal branch; at very low count rate the soft
1363: colour decreases, while the hard colour remains fairly constant defining a
1364: horizontal branch {\em ii)} smooth motion, i.e. without jumps, in the
1365: CD/HID, {\em iii)} the low-intensity states are more variable (in terms of
1366: fractional rms), {\em iv)} description of power spectra in terms of a small
1367: number of Lorentzian components, {\em v)} similar BLN namely, broad
1368: Lorentzians $Q \sim 0-0.3$ describing the low-frequency and high-frequency
1369: noise ($L_b$, $L_l$, $L_u$) and peaked noise in between ($L_{LF}$) and {\em
1370: vi)} flat-topped noise at the lower frequencies in the horizontal branch
1371: that turns into power-law noise in the diagonal branch.
1372:
1373: There are also important differences: {\em i)} different patterns in the
1374: CD/HID. Unlike low-mass X-ray binary and black hole systems, the
1375: low-intensity state is associated with a soft state (except for \bq), {\em
1376: ii)} slower motion along the spectral branches, hours to days in LMXBs,
1377: weeks to months in HMXBs, {\em iii)} the characteristic time scales implied
1378: by the noise components are about one order of magnitude longer in HMXBs
1379: (e.g. mHz QPOs), {\em iv)} $L_S$ noise in HMXBs and {\em v)} no apparent
1380: correlation between the power spectral parameters (characteristic
1381: frequencies, $rms$) and mass accretion rate.
1382:
1383: Although we have analysed all Be/X-ray binaries that showed type II
1384: outbursts in the period 1996--2007 and had good data coverage and
1385: sampling, the number of sources is not large enough to draw definite
1386: conclusions on the underlying physical parameters that may explain the
1387: differences between the two subgroups. Increasing the number statistics is,
1388: however, difficult given the transient nature of their X-ray emission and
1389: unpredictability of the type II outbursts.
1390:
1391: Even though all four sources analysed in this work are X-ray pulsars, it is
1392: not clear whether the high-mass X-ray binaries with supergiant companions
1393: show a similar behaviour. In these systems accretion occurs via the strong
1394: stellar wind of the massive companion, which contrasts with accretion via a
1395: disc, as it is believed to occur in Be/X-ray binaries during outburst.
1396: The problem of this type of studies in supergiant X-ray binaries is that
1397: they are rather stable in long time scales and do not show a wide range in
1398: X-ray flux.
1399:
1400:
1401: \begin{acknowledgements}
1402:
1403: This research was supported by the European Union Marie Curie grant
1404: MTKD-CT-2006-039965. This research has made use of NASA's Astrophysics Data
1405: System Bibliographic Services and of the SIMBAD database, operated at the
1406: CDS, Strasbourg, France. The ASM light curve was obtained from the
1407: quick-look results provided by the ASM/RXTE team.
1408:
1409: \end{acknowledgements}
1410:
1411: \begin{thebibliography}{}
1412:
1413: \bibitem[Angelini et al.(1989)]{ange89}
1414: Angelini, L.; Stella, L.; Parmar, A. N., 1989, ApJ, 346, 906
1415: \bibitem[Alpar \& Shaham(1985)]{alpa85}
1416: Alpar, M. A. \& Shaham, J. 1985, Nature, 316, 239
1417: \bibitem[Belloni \& Hasinger(1990)]{bell90}
1418: Belloni, T., \& Hasinger, G. 1990, 230, 103
1419: \bibitem[Belloni et al.(2002)]{bell02}
1420: Belloni, T., Psaltis, D. \& van der Klis, M. 2002, ApJ, 572, 392
1421: \bibitem[Belloni et al.(2005)]{bell05}
1422: Belloni, T., Homan, J., Casella, P. et al. 2005, A\&A, 440, 207
1423: \bibitem[Bernacca et al.(1984)]{bern84}
1424: Bernacca P.L., Iijima T., Stagni R. 1984, A\&A, 132, L8
1425: \bibitem[Borozdin et al.(1990)]{boro90}
1426: Borozdin, K., Gilfanov, M., Sunyaev, R., et al., 1990, Sov. Astron. Lett.,
1427: 16, 345
1428: \bibitem[Burderi et al.(1997)]{burd97}
1429: Burderi, L., Robba, N.R., La Barbera, N., Guainazzi, M. 1997, ApJ, 481, 943
1430: \bibitem[Camero-Arranz et al(2005)]{came05}
1431: Camero Arranz, A., Wilson, C. A., Connell, P., et al. 2005, A\&A, 441, 261
1432: \bibitem[Chakrabarty et al.(1995)]{chak95}
1433: Chakrabarty, D., Koh, T., Bildsten, L., et al. 1995, ApJ, 446, 826
1434: \bibitem[Coburn et al.(2004)]{cobu04}
1435: Coburn, W., Kalemci, E., Kretschmar, P., et al. 2004, ATel 337
1436: \bibitem[Coe et al.(1987)]{coe87}
1437: Coe M.J., Longmore A.J., Payne B.J., Hanson C.G. 1987, MNRAS, 226, 455
1438: \bibitem[Coe et al.(1988)]{coe88}
1439: Coe, M. J., Payne, B. J., Longmore, A., Hanson, C. G., 1988, MNRAS, 232, 865
1440: %\bibitem[Coe et al.(2006)]{coe06}
1441: %Coe M.J., Reig, P. McBride, V.A., Galache, J.L., Fabregat, J., 2006, MNRAS,
1442: %368, 447
1443: \bibitem[Corbet et al.(1986)]{corb86}
1444: Corbet R.H.D., Charles P.A., van der Klis M. 1986, A\&A, 162, 117
1445: %\bibitem[di Salvo et al.(2000)]{salv00}
1446: %Di Salvo, T., Stella, L., Robba, N. R., et al. 2000, ApJ, 544, L119
1447: %\bibitem[di Salvo et al.(2002)]{salv02}
1448: %Di Salvo, T., Farinelli, R., Burderi, L. et al. 2002, A\&A, 386, 535
1449: %\bibitem[Gierlinski \& Done(2002)]{gier02}
1450: %Gierlinski, M., Done, C. 2002, MNRAS, 331, L47
1451: \bibitem[Finger et al.(1996)]{fing96}
1452: Finger, M. H., Wilson, R. B., Harmon, B. A. 1996, ApJ, 459, 288
1453: \bibitem[Finger(1998)]{fing98}
1454: Finger, M. H., 1998, AdSpR, 22, 1007
1455: \bibitem[Ford \& van der Klis(1998)]{ford98}
1456: Ford, E. \& van der Klis, M. 1998, ApJ, 506, L39
1457: \bibitem[Galloway et al.(2004)]{galo04}
1458: Galloway, D.K., Morgan, E.H., Levine, A.M., 2004, ApJ, 613, 1164
1459: \bibitem[Goranskii et al.(1991)]{gora91}
1460: Goranskii, V.P., Esipov, V.F., Lyutyi, V.M., \& Shugarov, S.Yu., 1991, Sov
1461: Astron. Lett., 17, 399
1462: \bibitem[Goranskij \& Barsukova(2004)]{gora04}
1463: Goranskij, V. \& Barsukova, E. 2004, ATel 245.
1464: \bibitem[Grankin et al.(1991)]{gran91}
1465: Grankin, K.N., Shevchenko, V.S., Yakubov, S.D., 1991, Sov. Astron. Lett,
1466: 17, 415
1467: \bibitem[Hasinger(1987)]{hasi87}
1468: Hasinger, G., 1987, A\&A, 186, 153
1469: \bibitem[Hasinger \& van der Klis(1989)]{hasi89}
1470: Hasinger, G., van der Klis, M., 1989, A\&A, 225, 79
1471: \bibitem[Hayasaki \& Ozakaki(2004)]{haya04}
1472: Hayasaki, K. \& Ozakaki, A.T. 2004, MNRAS, 350, 971
1473: %\bibitem[Heindl \& Coburn(1999)]{hein99}
1474: %Heindl W.A. \& Coburn W., 1999, IAUC 7126
1475: \bibitem[Heindl et al.(1999)]{hein99}
1476: Heindl, W. A., Coburn, W., Gruber, D. E., Pelling, M. R., Rothschild, R. E.
1477: et al. 1999, ApJ, 521, L49
1478: \bibitem[Heindl et al.(2004)]{hein04}
1479: Heindl, W. A., Rothschild, R. E., Coburn, W. et al., 2004, in {\it X-ray Timing 2003:
1480: Rossi and Beyond}. AIP Conference Proceedings, Vol. 714, held 3-5 November, 2003
1481: in Cambridge, MA. Edited by Philip Kaaret, Frederick K. Lamb, and Jean H. Swank.
1482: Melville, NY: American Institute of Physics, p.323-330
1483: \bibitem[Homan et al.(2007a)]{homa07a}
1484: Homan, J., van der Klis, M. Wijnands, R. Belloni, T., Fender, R. 2007a,
1485: ApJ, 656, 420
1486: \bibitem[Homan et al.(2007b)]{homa07b}
1487: Homan, J., Wijnands, R., Altamirano, D. Belloni, T. 2007b, ATel 1165
1488: \bibitem[Inam et al.(2004)]{inam04}
1489: Inam, S. Ç., Baykal, A., Swank, J., Stark, M. J., 2004, ApJ, 616, 463
1490: \bibitem[Jernigan et al.(2000)]{jern00}
1491: Jernigan, J.G., Klein, R.I., Arons, J., 2000, ApJ, 530, 975
1492: \bibitem[Kaur et al.(2007)]{kaur07}
1493: Kaur, R., Paul, B., Raichur, H., Sagar, R., 2007, ApJ, 660, 1409
1494: \bibitem[Klochkov et al.(2007)]{kloc07}
1495: Klochkov, D., Horns, D., Santangelo, A., et al. 2007, A\&A, 464, L45
1496: %\bibitem[Kreykenbohm et al.(2005)]{krey05}
1497: %Kreykenbohm, I., Mowlavi, N., Produit, N. et al. 2005, A\&A, 433, L45
1498: \bibitem[Kriss et al.(1983)]{kris83}
1499: Kriss, G. A., Cioffi, D. F., Canizares, C. R. 1983, ApJ, 272, 439
1500: \bibitem[Lazzati \& Stella(1997)]{lazz97}
1501: Lazzati, D. \& Stella, L., 1997, ApJ, 476, 267
1502: \bibitem[Leahy et al.(1983)]{leah83}
1503: Leahy, D.A., Darbro, W., Elsner, R.F. et al., 1983, ApJ, 266, 160
1504: \bibitem[Levine \& Corbet(2000)]{levi00}
1505: Levine, A., \& Corbet, R., 2000, IAUC No 7524
1506: \bibitem[Makishima et al.(1990)]{maki90}
1507: Makishima K., Mihara T., Ishida M., et al. 1990, ApJ, 365, L59
1508: \bibitem[Markert et al.(1977)]{mark77}
1509: Markert, T.H, Canizares, C.R., Clark, G.W., et al. 1977, ApJ, 218, 801
1510: \bibitem[Mavromatakis(1994)]{mavr94}
1511: Mavromatakis, F., 1994, A\&A, 285, 209
1512: %\bibitem[McClintock \& Remillard(2006)]{clin06}
1513: %McClintock, J.E., Remillard, R.A., 2006, in {\em Compact Stellar X-ray sources},
1514: %eds. W.H.G. Lewin \& M. van der Klis, Cambridge University Press, p.157
1515: \bibitem[Miyamoto et al.(1991)]{miya91}
1516: Miyamoto, S., Kimura, K., Kitamoto, S., Dotani, T., Ebisawa, K. 1991, ApJ,
1517: 383, 784
1518: \bibitem[Mowlavi et al.(2006)]{mowl06}
1519: Mowlavi, N.; Kreykenbohm, I.; Shaw, S. E. et al. 2006, A\&A, 451, 187
1520: \bibitem[Motch \& Janot-Pacheco(1987)]{motc87}
1521: Motch, C., Janot-Pacheco, E., 1987, A\&A, 182, L55
1522: \bibitem[Motch etal(1991)]{motc91}
1523: Motch, C., Stella, L., Janot-Pacheco, E., \& Mouchet, M. 1991, ApJ, 369, 490
1524: %\bibitem[Muno et al.(2002)]{muno02}
1525: %Muno, M.P., Remillard, R.A., Chakrabarty, D. 2002, ApJ, 568, L35
1526: \bibitem[Mukherjee et al.(2006)]{mukh06}
1527: Mukherjee, U., Bapna, S., Raichur, H., Paul, B., Jaaffrey, S.N.A. 2006,
1528: JApA, 27, 25
1529: \bibitem[Nakajima et al.(2006)]{naka06}
1530: Nakajima, M., Mihara, T., Makishima, K., Niko, H., 2006, ApJ, 646, 1125
1531: \bibitem[Nowak et al.(1999)]{nowa99}
1532: Nowak, M.A., Vaughan, B.A., Wilms, J., Dove, J.B., Begelman, C. 1999, ApJ,
1533: 510, 874
1534: \bibitem[Nowak(2000)]{nowa00}
1535: Nowak, M.A. 2000, MNRAS, 318, 361
1536: \bibitem[Naik et al.(2006)]{naik06}
1537: Naik, S., Callanan, P.J., Paul, B., Dotani, T., 2006, ApJ, 647, 1293
1538: \bibitem[Negueruela et al.(1999)]{negu99}
1539: Negueruela I., Roche P., Fabregat J., Coe M.J. 1999, MNRAS, 307, 695
1540: \bibitem[Negueruela et al.(2003)]{negu03}
1541: Negueruela, I., Israel, G.L., Marco, A., Norton, A.J., Speziali, R.,
1542: 2003, A\&A, 397, 739
1543: \bibitem[Okazaki \& Negueruela(2001)]{okaz01}
1544: Okazaki, A., Negueruela, I., 2001, A\&A, 377, 161
1545: \bibitem[Olive et al.(2003)]{oliv03}
1546: Olive, J., Barret, D., Gierlinski, M., 2003, ApJ, 583, 416
1547: \bibitem[Parmar et al.(1989)]{parm89}
1548: Parmar, A. N., White, N. E., Stella, L., Izzo, C., Ferri, P., 1989, ApJ,
1549: 338, 359
1550: \bibitem[Pottschmidt et al.(2003)]{pott03}
1551: Pottschmidt, K., Wilms, J., Nowak, M. et al. 2003, A\&A, 407, 1039
1552: \bibitem[Pottschmidt et al.(2005)]{pott05}
1553: Pottschmidt, K., Kreykenbohm, I., Wilms, J., et al., 2005, ApJ, 634, L97
1554: \bibitem[Priedhorsky et al.(1986)]{pried86}
1555: Priedhorsky, W, Hasinger, G, Lewin, W.H.G, 1986, ApJ 306, L91
1556: \bibitem[Psaltis et al.(1999)]{psal99}
1557: Psaltis, D, Belloni, T., van der Klis, M. 1999, 520, 262
1558: \bibitem[Qu et al.(2005)]{qu05}
1559: Qu J.L., Zhang S., Song L.M., Falanga, M. 2005, ApJ, 629, L33
1560: \bibitem[Rappaport et al.(1977)]{rapp77}
1561: Rappaport, S., Markert, T., Li, F.K., et al. 1977, ApJ, 217, L29
1562: \bibitem[Rappaport et al.(1978)]{rapp78}
1563: Rappaport, S., Clark, G. W., Cominsky, L., Li, F., \& Joss, P. C. 1978,
1564: ApJ, 224, L1
1565: \bibitem[Reig et al.(1997)]{reig97}
1566: Reig, P., Fabregat, J., Coe, M.J., 1997, A\&A, 322, 193
1567: \bibitem[Reig \& Coe(1998)]{reig98}
1568: Reig, P., Coe, M. J., 1998, MNRAS, 294, 118
1569: \bibitem[Reig et al.(1998)]{reigal98}
1570: Reig, P., Stevens, J. B., Coe, M. J., Fabregat, J., 1998, MNRAS, 301, 42
1571: \bibitem[Reig \& Coe(1999)]{reig99}
1572: Reig, P., Coe, M. J., 1999, MNRAS, 302, 700
1573: \bibitem[Reig et al.(2004)]{reig04}
1574: Reig, P., van Straaten, S. \& van der Klis, M. 2004, ApJ, 602, 918
1575: \bibitem[Reig et al.(2006)]{reig06}
1576: Reig, P., Mart\'{\i}nez-N\'u\~nez, S. Reglero, V., 2006, A\&A, 449, 703
1577: \bibitem[Reig(2007)]{reig07}
1578: Reig, P., 2007, MNRAS, 377, 867
1579: \bibitem[Reig et al.(2007)]{reta07}
1580: Reig, P.,Larionov, V., Negueruela, I., Arkharov, A. A., Kudryavtseva, N.
1581: A., 2007, A\&A, 462, 1081
1582: \bibitem[Reynolds et al. (1993)]{reyn93}
1583: Reynolds, A. P., Parmar, A. N., White, N. E., 1993, ApJ, 414, 302
1584: \bibitem[Santangelo et al.(1999)]{sant99}
1585: Santangelo, A., Segreto, A., \& Giarrusso, S. 1999, ApJ, 523, L85
1586: \bibitem[Schulz et al.(1989)]{schu89}
1587: Schulz, N.S., Hasinger, G. Truemper, J., 1989, A\&A, 225, 48
1588: %\bibitem[Soong \& Swank(1989)]{soon89}
1589: %Soong, Y.\& Swank, J. H., in ESA, The 23rd ESLAB Symposium on Two Topics in X Ray
1590: %Astronomy. Vol. 1: X Ray Binaries p. 617
1591: \bibitem[Shirakawa \& Lai(2002)]{shir02}
1592: Shirakawa, A. \& Lai, D. 2002, ApJ, 565, 1134
1593: \bibitem[Stella et al.(1985)]{stel85}
1594: Stella L., White N.E., Davelaar J., et al. 1985, ApJ, 288, L45
1595: \bibitem[Stella et al.(1986)]{stel86}
1596: Stella, L., White, N. E., Rosner, R. 1986, ApJ, 308, 669
1597: \bibitem[Stollber et al.(1999)]{stol99}
1598: Stollberg, M.T., Finger, M.H., Wilson, R. B., et al. 1999, ApJ, 512, 313
1599: \bibitem[Sun et al.(1994)]{sun94}
1600: Sun, X.J., Li, T.P., Wu, M., Cheng, L.X., 1994, A\&A, 289, 127
1601: %\bibitem[Sunyaev \& Revnivtsev(2000)]{suny00}
1602: %Sunyaev, R. \& Revnivtsev, M., 2000, A\&A, 358, 617
1603: \bibitem[Swank \& Morgan(2000)]{swan00}
1604: Swank, J., Morgan, E., 2000, IAUC No 7532
1605: \bibitem[Takeshima et al.(1994)]{take94}
1606: Takeshima T., Dotani T., Mitsuda K., Nagase F. 1994, ApJ, 436, 871
1607: \bibitem[Tanaka(1983)]{tana83}
1608: Tanaka, Y. 1983, IAUC 3891
1609: \bibitem[Terrell \& Priedhorsky(1984)]{terr84}
1610: Terrell J., \& Priedhorsky W.C. 1984, ApJ, 285, L15
1611: \bibitem[van der Klis et al.(1987a)]{klis87a}
1612: van der Klis, M., Jansen, F., van Paradijs, J. et al., 1987a, ApJ, 313, L19
1613: \bibitem[van der Klis et al.(1987b)]{klis87b}
1614: van der Klis, M., Stella, L., White, N., Jansen, F., Parmar, A. N. 1987b,
1615: ApJ, 316, 411
1616: \bibitem[van der Klis(1989)]{klis89}
1617: van der Klis, M., 1989, in Timing Neutron stars, NATO ASI ser. C vol. 262,
1618: ed. H.\"Ogelman,\& E.P.J. van den Heuvel, p.27
1619: \bibitem[van der Klis(1994)]{klis94}
1620: van der Klis, M. 1994, ApJSS, 92, 511
1621: \bibitem[van der Klis(2006)]{klis06}
1622: van der Klis, M. 2006, in {\em Compact Stellar X-ray sources},
1623: eds. W.H.G. Lewin \& M. van der Klis, Cambridge University Press, p.39
1624: \bibitem[van Straaten et al.(2002)]{stra02}
1625: van Straaten, S., van der Klis, M., di Salvo, T. \& Belloni, T. 2002, ApJ,
1626: 568, 912
1627: \bibitem[van Straaten et al.(2003)]{stra03}
1628: van Straaten, S., van der Klis, M., \& M\'endez, M. 2003, ApJ, 596, 1155
1629: \bibitem[Wilson et al.(2002)]{wils02}
1630: Wilson, C.A., Finger, M.H., Coe, M.J., Laycock, S., Fabregat, J. 2002, ApJ,
1631: 570, 287
1632: %\bibitem[Wilson \& Finger(2006)]{wils06}
1633: %Wilson, C.A., Finger, M.H., 2006, ATel 877
1634: \bibitem[Wilson et al.(2008)]{wils08}
1635: Wilson, C.A., Finger, M.H., Camero Arranz, A. 2008, ApJ, 678, 1263
1636: \bibitem[White \& Marshall(1984)]{whit84}
1637: White, N.E., Marshall, F.E., 1984, ApJ, 281, 354
1638: \bibitem[Whitlock(1989)]{whit89}
1639: Whitlock L. 1989, ApJ, 344, 371
1640: \bibitem[Wijnands \& van der Klis(1999a)]{wijn99a}
1641: Wijnands,R. \& van der Klis, M. 1999a, ApJ, 514, 939
1642: \bibitem[Wijnands \& van der Klis(1999b)]{wijn99b}
1643: Wijnands, R. \& van der Klis, M. 1999b, ApJ, 522, 965
1644: %\bibitem[Zhang et al.(1995)]{zhan95}
1645: %Zhang, W., Jahoda, K., Swank J., Morgan, E., Giles, A., 1995, ApJ, 449, 930
1646: %\bibitem[Zhang et al.(1996)]{zhan96}
1647: %Zhang, W., Morgan, E., Jahoda, K., et al. 1996, ApJ, 469, L29
1648: \end{thebibliography}
1649:
1650: \end{document}
1651: