1: %% AASTeX v5.x LaTeX 2e macros.
2:
3: %% The command below calls the preprint style
4: %% which will produce a one-column, single-spaced document.
5: %% \documentclass[preprint]{aastex}
6:
7: %% manuscript produces a one-column, double-spaced document:
8: %%\documentclass[manuscript]{aastex}
9:
10: %% preprint2 produces a double-column, single-spaced document:
11:
12: \documentclass[preprint2]{aastex}
13:
14: %% If you want to create your own macros, you can do so
15: %% using \newcommand. Your macros should appear before
16: %% the \begin{document} command, e.g. \newcommand{\vdag}{(v)^\dagger}
17:
18: \def\msun{$M_{\odot}$}
19: \def\lsun{$L_{\odot}$}
20: \def\rsun{$R_{\odot}$}
21: \def\mdot{$\dot M$}
22: \def\ergsec{\hbox{erg s$^{-1}$ }}
23: \def\ergcm{\hbox{erg cm$^{-2}$ s$^{-1}$ }}
24: \def\Rg{$R_{\rm g}$}
25: \def\xte{{\it RXTE~}}
26: \def\h1743{H1743--322}
27:
28: %% You can insert a short comment on the title page using the command below.
29:
30: \slugcomment{}
31:
32: %% If you wish, you may supply running head information, although
33: %% this information may be modified by the editorial offices.
34: %% The left head contains a list of authors, (max. 3 or et al.). The right
35: %% head is a modified title of up to roughly 44 characters. Running heads
36: %% will not print in the manuscript style.
37:
38: %\shorttitle{HFQPOs in \h1743}
39: %\shortauthors{Remillard et al.}
40:
41: \begin{document}
42:
43: \title{The X-ray Outburst of \h1743: High-Frequency QPOs with a 3:2
44: Frequency Ratio}
45:
46: %% Use \author, \affil, and the \and command to format
47: %% author and affiliation information.
48:
49: \author{Ronald A. Remillard}
50: \affil{Center for Space Research, MIT, Cambridge, MA 02139-4307 }
51: \email{rr@space.mit.edu}
52:
53: \and
54:
55: \author{Jeffrey E. McClintock}
56: \affil{Harvard-Smithsonian Center for Astrophysics, 60
57: Garden St. MS-3, \\ Cambridge, MA 02138}
58: \email{jem@cfa.harvard.edu}
59:
60: \and
61:
62: \author{Jerome A. Orosz}
63: \affil{Dept. of Astronomy, San Diego State University, 5500 Campanile
64: Drive, \\ San Diego, CA 82182-1221}
65: \email{orosz@sciences.sdsu.edu}
66:
67: \and
68: \author{Alan M. Levine}
69: \affil{Center for Space Research, MIT, Cambridge, MA 02139-4307 }
70: \email{aml@space.mit.edu}
71:
72: \begin{abstract}
73:
74: The 2003 X-ray outburst of the candidate black-hole binary, \h1743,
75: was investigated in frequent pointed observations (2--250 keV) with
76: the {\it Rossi} X-ray Timing Explorer. We consider one particular
77: program of 130 observations. We organized these data into 111 time
78: intervals and conducted a search for the presence of high-frequency
79: quasiperiodic oscillations (HFQPOs) in the range 50--2000 Hz. Only a
80: single observation (2003 June 13) yielded a detection above $4
81: \sigma$; the central frequency of $239 \pm 4$ Hz is consistent with
82: the 240 Hz QPO reported for this source on 2003 May 28 \citep{hom03}.
83: We next grouped the observations in several different ways and
84: computed the average power-density spectra (PDS) in a search for
85: further evidence of HFQPOs. This effort yielded two significant
86: results for those observations defined by the presence of
87: low-frequency QPOs (0.1-20 Hz) and an absence of ``band-limited''
88: power continua: (1) The 9 time intervals with the highest X-ray flux
89: yielded an average PDS with a QPO at $166 \pm 5$ Hz. ($4.1 \sigma$;
90: 3--35 keV); and (2) a second group with lower X-ray flux (24 time
91: intervals) produced an average PDS with a QPO at $242 \pm 3$ Hz ($6.0
92: \sigma$; 7--35 keV). The ratio of these two frequencies is $1.46 \pm
93: 0.05$. This finding is consistent with results obtained for three
94: other black hole systems that exhibit commensurate HFQPOs in a 3:2
95: ratio. Furthermore, the occurrence of H1743-322's slower HFQPO at
96: times of higher X-ray luminosity closely resembles the behavior of
97: XTE~J1550-564 and GRO~J1655-40. We discuss our results in terms of a
98: resonance model that invokes frequencies set by general relativity for
99: orbital motions near a black-hole event horizon.
100:
101: \end{abstract}
102:
103: \keywords{black hole physics --- general relativity --- stars:
104: individual
105: (\h1743) --- stars: oscillations --- X-rays: stars}
106:
107: \section{Introduction}
108:
109:
110: A key question in black hole binary (BHB) research is whether
111: high-frequency QPOs (HFQPOs), observed in the range of 40--450 Hz,
112: represent a unique timing signature that may constrain a black hole's
113: mass and spin via a model rooted in general relativity (GR; see
114: McClintock \& Remillard 2003). \nocite{mcrem03} The HFQPO signals are
115: weak, and an attack on this problem requires observations of bright
116: X-ray transients during their rare and short-lived ($\sim$ months)
117: outbursts.
118:
119: HFQPOs have been detected in 7 confirmed or candidate BHB systems,
120: including \h1743 (see below). In each of three sources (GRO~J1655--40,
121: XTE~J1550--564, and GRS~1915+105), transient HFQPOs have appeared at a
122: pair of frequencies that are commensurate with a 3:2 ratio
123: \citep{rem02a,rem03}. Furthermore, when the HFQPOs were compared
124: among the three sources, it was found that the frequencies scale
125: inversely with black hole mass. This result is consistent with
126: expectations for oscillations produced at some characteristic radius
127: in a strong-gravity environment described by GR theory, but
128: only in the special case where the values of the dimensionless black
129: hole spin parameter are similar for the three sources \citep{mcrem03}.
130:
131: Commensurate HFQPO frequencies can be interpreted as a signature of an
132: oscillation driven by some type of resonance condition. Abramowicz \&
133: Kluzniak (2001) \nocite{abr01} had earlier proposed that QPOs could
134: represent enhanced emission from a particular radius where there is a
135: resonance in the GR coordinate frequencies for orbital/epicyclic
136: motions in strong gravity \nocite{mer99} (see Merloni et al. 1999).
137: Resonances in some form may be applicable to both BH and NS systems
138: \citep{abr03}. We note that GR coordinate frequencies and associated
139: beat frequencies in the inner accretion disk were invoked in earlier
140: work on variable-frequency QPOs in both neutron-star and some BHB
141: systems (Stella, Vietri, \& Morsink 1999). \nocite{ste99}
142:
143: Herein, we report on a general search for HFQPOs in the source \h1743.
144: The 2003 outburst of this source was first detected on March 21 in
145: hard X-rays (15--200 keV) by INTEGRAL (IGR~J17464--3213; Revnivtsev et
146: al. 2003). \nocite{rev03} Follow-up observations with \xte led to the
147: recognition that the source is a recurrent X-ray nova first observed
148: with HEAO1 in 1977--1978 \citep{mar03,gur78}. The X-ray spectral and
149: temporal properties exhibited during the first outburst had
150: established \h1743 as a black-hole candidate; its X-ray spectrum
151: contained both a soft component (1--10 keV) and a hard X-ray tail
152: (10--100 keV; Cooke et al. 1984), \nocite{coo84} and there were no
153: pulsations or X-ray bursts that would identify the source as a
154: neutron-star system \citep{tan95}.
155:
156: An X-ray light curve (1.5--12 keV) for the recent outburst is
157: displayed in the top panel of Fig. \ref{fig:asm}, using data from the
158: \xte All-Sky Monitor (ASM; Levine et al. 1996). \nocite{lev96} The
159: bottom panel shows measurements of the spectral hardness ratio,
160: defined as the ratio of the ASM count rate at 5--12 keV to that at
161: 3--5 keV. These data show that the initially hard spectrum rapidly
162: evolves to a softer one. A number of bright flares are seen
163: before the source settles into a relatively steady and soft X-ray
164: state. Several bright flares reach intensities near 1.5 Crab
165: (i.e. $3.6 \times 10^{-8}$ \ergcm at 2--10 keV) during 2003 April
166: 18-24.
167:
168: Radio detections of \h1743 were obtained during the first few weeks of
169: the 2003 outburst, yielding a precise celestial position and evidence
170: for a transient radio jet (Rupen, Mioduszewski, \& Dhawan 2003a;
171: 2003b; 2003c). \nocite{rup03a,rup03b,rup03c} Soon thereafter, a faint
172: optical counterpart was detected (R=21.9), despite $\sim 13$ mag of
173: dust extinction \citep{ste03}.
174:
175: \xte made 227 pointed observations of \h1743 (=XTE~J1746--326) during
176: the 2003--2004 outburst, in the course of carrying out five different
177: observing programs. In this paper we report on the results of a
178: search for HFQPOs (50--2000 Hz) from this source using the 130
179: observations from program ID 80146. The times of these observations
180: are indicated in the row of vertical lines in the top panel of
181: Fig. \ref{fig:asm}. An HFQPO was reported for \h1743 after a 25 ks
182: observation with \xte on 2003 May 28 for program ID 80135
183: \citep{hom03}. This observations yielded an HFQPO detection ($4.5
184: \sigma$) at 240 Hz, with a possible second feature ($2.5 \sigma$) at
185: 160 Hz \citep{hom03}.
186:
187: In the following sections we describe our analysis techniques and
188: methods for data grouping that lead to significant detections of
189: HFQPOs at 242 and 166 Hz. Assuming \h1743 is a genuine BH, our
190: discovery establishes this source as the fourth BH where a pair of
191: HFQPOs appear to have commensurate frequencies in a 3:2 ratio. It
192: is further be shown that additional properties of these HFQPOs
193: bear a striking resemblance to those seen in GRO~J1655--40 and
194: XTE~J1550--564 \citep{rem02a}.
195:
196: \section{Observations and Data Analysis}
197:
198: We consider all of the 130 observations of \h1743 conducted by \xte
199: under program ID 80146. This data set contains occasional sequences
200: of short observations, and we therefore chose to collect and organize
201: the observations into the 111 time intervals that are listed in Table
202: \ref{tab:obs}. The number of observations combined for each time
203: interval is given in column 4. The maximum time span for any interval
204: is 0.48 days, and the total exposure time is 498 ks.
205:
206: Our X-ray timing analyses utilize data from the \xte~Proportional
207: Counter Array (PCA). The data telemetry modes chosen for this program
208: allow us to construct PCA light curves with $125 \mu$s time resolution
209: in 3 energy bands: 3-7, 7-14, and 14-35 keV. The PCA consists of 5
210: detector units (PCUs). However, the observations of \h1743 utilized a
211: variable number of PCUs (usually 3) because some of the PCUs are
212: cycled off periodically to avoid problems with detector breakdown.
213: For each observation interval, the telemetry modes for high-speed data
214: merged all of the good events from the PCUs that were in operation.
215:
216: Table \ref{tab:obs} also provides, for each observation interval, the
217: mean count rate (c/s/PCU) at 2-35 keV (col. 6) and at 7--35 keV
218: (col. 7). The latter quantity is useful in evaluating the strength of
219: nonthermal radiation in \h1743 and was used to group observations in
220: the search for HFQPOs. The PCA count rates are derived from
221: background-subtracted energy spectra processed from the `Standard 2'
222: data mode using PCU \#2, which was in use (along with PCU \#0) during
223: every observation of program 80146. All of the spectral reductions
224: were performed via the `ftools' routines in the HEAsoft software
225: package distributed by NASA's HEASARC.
226:
227: The procedures used to compute power density spectra (PDS) and to
228: search for QPOs are described in Remillard et al. (2002).
229: \nocite{rem02a} Briefly, for each of the 111 time intervals (Table
230: \ref{tab:obs}), we compute Fourier transforms for each 256-s segment
231: in the PCA light curve, and we output the average PDS for the
232: available number of segments. We then subtract the deadtime-corrected
233: Poisson noise and normalize the PDS to units of (rms deviation /
234: mean)$^2$ Hz$^{-1}$. In each frequency bin, we compute the
235: uncertainty as the larger of either the statistical error ($2/N^{0.5}$
236: prior to rms normalization) or the empirical standard deviation of the
237: mean power, where $N$ is the number of transforms (of 256 s duration)
238: in a given observation interval. We note that this conservative
239: approach can yield error bars that appear large, compared to
240: bin-by-bin fluctuations, when the broad power continuum varies
241: significantly during an observation. Finally, the power densities and
242: uncertainties are re-binned in logarithmic intervals of frequency
243: ($\nu$), maintaining (for this study) a minimum $\Delta \nu / \nu =
244: 0.04$. When we average the PDS for groups of observations, we
245: compute the weighted mean (using $\sigma^{-2}$) and the net uncertainty
246: for each frequency bin.
247:
248: We search for HFQPOs using statistical tests to detect a Lorentzian
249: peak rising above the local power continuum. The continuum, in terms
250: of log $P_{\nu}$, is modeled with a second order polynomial in log
251: $\nu$, which presumes a power-law function with allowances for broad
252: curvature. QPOs are distinguished from broad power peaks using a
253: coherence parameter, $Q = \nu / FWHM \ga 2$. For each PDS, we use
254: $\chi^2$ minimization to obtain the best fit for the QPO profile and
255: the local power continuum. Finally, in our general QPO searches, we
256: seek results that have acceptable values of $\chi_{\nu}^2$ while
257: surpassing a significance threshold for the integrated QPO power
258: ($P$), relative to the uncertainty ($\sigma_P$): $P/\sigma_P \ga 4$.
259: This elevated threshold compensates for the large number of PDSs and
260: trial frequencies considered in the analysis of data sets obtained for
261: typical bright X-ray transients.
262:
263: \section{Results}
264:
265: We searched for QPOs in the range 50--2000 Hz for each of the 111
266: observing intervals (Table \ref{tab:obs}). For each interval, we
267: considered separately the average PDS for the energy ranges 3-35,
268: 7-35, and 14-35 keV. Only one interval (2003 June 13 = MJD 52803)
269: yields a detection above $4 \sigma$. In this case the energy range is
270: 7--35 keV, the central frequency is $239 \pm 4$ Hz, the integrated rms
271: amplitude is $r = 2.0 \pm 0.2$~\%, $Q \sim 9$, and the detection
272: significance is 4.3 $\sigma$. The central frequency is consistent
273: with the 240 Hz QPO detected for this source on 2003 May 28 (MJD
274: 52787) in observations from a different \xte program \citep{hom03}.
275: There is one additional noteworthy finding from this initial search
276: for HFQPOs. For the interval with highest PCA count rate (1.5 Crab on
277: MJD 52765), an HFQPO candidate ($3.8 \sigma$) appears at $162 \pm 7$
278: Hz with $Q \sim 4$ and $r = 1.1 \pm 0.2$~\% (3-35 keV).
279:
280: Most of the HFQPOs detections in BHB outbursts have been found in PDS
281: averaged over a number of \xte observations \citep{rem99,cui00,rem02c}
282: because high statistical precision is required in order to detect a
283: faint signal spread over a bandwidth of 20 to 50 Hz. Furthermore,
284: judicious data selection is required, since HFQPOs are usually
285: detected only in the ``steep power-law state'' \citep{mcrem03}, and
286: within this state further considerations may be required to deal with
287: amplitude variability, frequency switching, and choice of PDS energy
288: range. We therefore investigated several strategies for grouping the
289: observations of \h1743 prior to further searches for HFQPO detections.
290:
291: In the case of XTE~J1550-564, it was shown that the phase lags (13-30
292: keV vs. 2-13 keV) in low-frequency QPOs (LFQPOs; 0.1-20 Hz) could be
293: used to define three LFQPO types that are well correlated with HFQPO
294: properties \citep{rem02c}. These LFQPO types predicted, respectively,
295: the absence of HFQPOs or the appearance of HFQPOs at either 185 or 276
296: Hz. We investigated this method for the case of \h1743, but we found
297: the results to be statistically unsatisfactory because the source is
298: fainter than XTE~J1550-564 and the LFQPO amplitudes are
299: comparatively low.
300:
301: As an alternative approach, one can capitalize on correlations found
302: between HFQPOs and the properties of the power continuum and the
303: energy spectrum. The LFQPO type (``C'') that forecasts an absence of
304: HFQPOs in XTE~J1550-564 can also be recognized for its association
305: with a ``band-limited'' power continuum in which the power density is
306: flat at low frequencies and then drops abruptly at frequencies above the
307: LFQPO and its harmonic. Such a PDS can be recognized via the
308: integrated $rms$ power derived from the PDS (0.1 to 10 Hz), and those values
309: are given in col. 8 of Table \ref{tab:obs}. Observation intervals for
310: \h1743 with integrated $rms > 0.11$ all show LFQPOs and band-limited
311: power continua, and we label these as PDS type ``q-bl'' in col. 9 of
312: Table \ref{tab:obs}. Intervals with PDS that are devoid of LFQPOs are
313: noted as PDS type ``0''; these cases also exhibit the lowest
314: values of the PCA hardness ratio ($HR = $col. 6 / col. 7 $ < 0.017$,
315: with one exception). These properties suggest that PDS type 0
316: designations identify the times when the source is in a
317: thermal-dominant state \citep{mcrem03}. The remaining intervals
318: contain LFQPOs without band-limited power continua and are labeled
319: ``q'' in col. 9 of Table \ref{tab:obs}.
320:
321: We computed average PDS for groups ``q-bl'' (23 observation intervals)
322: and ``0'' (44 intervals), excluding the
323: last 9 intervals where the 2-35 keV count rate drops below 550
324: c/s/PCU ($\sim 0.2$ Crab) and the PDS become statistically weak. We
325: then divided the ``q'' type PDS into 2 groups distinguished by the
326: source brightness above or below 430 c/s/PCU at 7-35 keV (col. 8 of
327: Table \ref{tab:obs}). This latter criterion attempts to exploit the
328: fact that the strength of the steep power-law component distinguishes
329: the observations that yield each of the QPOs with frequencies that scale
330: in a 3:2 ratio (hereafter referred to as the $2 \nu_0$ QPO and the $3 \nu_0$
331: QPO) in both XTE~J1550-564 and GRO~J1655-40 \citep{rem02a}.
332:
333: The PDS for the 4 groups, plotted in units of log($\nu \times
334: P_{\nu}$) vs. log $\nu$, are shown in Fig. \ref{fig:pdstypes}. The
335: PDS fro the ``q-bl'' group shows a strong and very broad feature that
336: peaks near 3 Hz. The PDS for the ``0'' group appears to show the same
337: feature, although reduced in strength by a factor $\sim 30$. HFQPO
338: features are apparent only in the PDS for the ``q'' groups, and they
339: occur at different frequencies. We note that all of the four PDS show
340: a residual continuum above 100 Hz that is an artifact of an imperfect
341: deadtime model for the PCA instrument.
342:
343: In Fig. \ref{fig:pdsmod} we show the HFQPO detections for the two
344: ``q'' groups at higher frequency resolution, along with the fitted
345: models for the QPO profiles. The 9 ``q'' intervals with the higher
346: X-ray fluxes produce an average PDS (2-35 keV) with a QPO ($4.1
347: \sigma$) at $166 \pm 5$ Hz, with $r = 0.60 \pm 0.08$ \% and $Q = 5.7
348: \pm 1.6$. The ``q'' group comprising 26 intervals at lower X-ray flux
349: produces an average PDS (7-35 keV) with a QPO ($6.0 \sigma$) at $242
350: \pm 3$ Hz, with $r = 1.1 \pm 0.1$ \% and $Q = 11 \pm 1$. Note that
351: the transition in \h1743 from a broader HFQPO at $2 \nu_0$ in a broad
352: bandwidth to a narrower feature at $3 \nu_0$ in a harder energy band
353: occurs with decreasing 7-35 keV count rate, which is presumably
354: dominated by a non-thermal spectral component. This is precisely the
355: behavior exhibited by the HFQPOs from XTE~J1550-564 and GRO~J1655-40
356: \citep{rem02a}.
357:
358: \section{Discussion}
359:
360: The X-ray light curve and variability characteristics of \h1743 during
361: its 2003 outburst fully support the identification of this source as a
362: black hole candidate (see \S 1). The behavior of \h1743 resembles the
363: BHBs XTE~J1550-564 and GRO~J1655-40 in many ways, although it was not
364: as bright as those sources (by a factor of 3--4), comparing the
365: brightest 4-month intervals of each outburst. This is the likely
366: reason why we do not detect HFQPOs from \h1743 during individual
367: observations, other than those on MJD 52787 and possibly MJD 52765.
368:
369: The effort to group the PDS in order to gain statistically significant
370: HFQPO detections for \h1743 was guided by results for XTE~J1550-564
371: and GRO~J1655-40 \citep{rem02a}. Our best results were found simply by
372: selecting observations with LFQPOs present, excluding cases with q-bl
373: type power continua, and then grouping the PDS according to the 7-35
374: keV count rate, which represents the strength of the nonthermal X-ray
375: flux. This effort yielded two HFQPOs with properties that closely
376: resemble the HFQPOs of XTE~J1550-564 and GRO~J1655-40, i.e. the
377: central frequencies are consistent with a 3:2 ratio, the amplitudes
378: have $rms \sim $ 1\%, and the lower frequency QPO (166 Hz for \h1743)
379: is seen at times of highest non-thermal flux. In addition, the higher
380: frequency QPO (242 Hz) has a narrower profile (i.e. higher $Q$ value)
381: and the detection bandwidth has a relatively higher mean photon
382: energy. We note that the one remaining BH system with HFQPOs in a 3:2
383: ratio (GRS~1915+105) was investigated using a different analysis
384: method in which each QPO was extracted from a portion of a particular
385: type of violently variable light curve \citep{rem02b,rem03}.
386:
387: HFQPOs with frequencies at $2 \nu_0$ and $3 \nu_0$ can be considered
388: as expressions of a single frequency system. As with the other BHBs
389: discussed above, the frequency system in \h1743 appears to be invariant
390: through changes in luminosity or interruptions due to the source's
391: evolution through a thermal-dominant or other states. The observations
392: that correspond with the two groups that yield HFQPO detections
393: (i.e. PDS type ``q'' in Table 1) span a range 204--1083 c/s/PCU at
394: 7--35 keV and 1234--3775 c/s/PCU at 2--35 keV. The temporal order of
395: these observations is shown with small arrows in the top panel of
396: Fig. \ref{fig:asm}, where the upper/lower row represents the 242/166
397: Hz group, respectively. There are three gaps in the series of q-type
398: observations (i.e. combined arrows) that occur when
399: the PDS types are either 0 or q-bl (see Table 1).
400:
401: In summary, black hole HFQPOs appear at stable pairs of frequencies
402: with 3:2 ratio, and they exhibit common behavior patterns linked to
403: the properties of the energy spectra. These results support the view
404: that HFQPOs convey a distinct temporal signature for each accreting
405: black hole. The detection bandwidth and high frequencies suggest that
406: HFQPOs originate near the black hole event horizon. Furthermore, the
407: stability of HFQPO pairs with changing luminosity suggests that the
408: frequencies may depend only on the inherent properties of the black
409: hole, viz. its mass and spin.
410:
411: We have shown that the HFQPOs in \h1743 appear to originate from the
412: same mechanism as the HFQPOs with 3:2 frequency ratio in the other
413: BHBs. An empirical relationship has been derived (only 3 sources)
414: between the HFQPO frequency systems and black hole mass
415: \citep{mcrem03}, and we may apply this to the case of \h1743. We
416: assume that HFQPO frequencies depend only on the black hole mass
417: ($M_x$) and spin, which is frequently evaluated in terms of the
418: dimensionless spin parameter: $a_* = cJ/GM_x^2$, where $J$ is the
419: angular momentum of the black hole. If the black hole in \h1743 has a
420: similar value of $a_*$ as the other BHBs, then the HFQPOs (with $\nu_0
421: \sim 81$ Hz) suggests: $M_x = 931 / \nu_0 \sim$ 11.5 \msun.
422:
423: Detailed models of the HFQPO properties and behavior patterns must
424: confront the problem that we do not understand the origin of the steep
425: power-law spectrum. This radiation component is always fairly strong
426: when the 3:2 oscillations appear. Furthermore, the strength of the
427: steep power-law appears to regulate the frequency switching from $2
428: \nu_0$ to $3 \nu_0$.
429:
430: As noted in \S 1, commensurate HFQPOs in BHBs
431: have been interpreted with a ``parametric resonance'' concept that
432: hypothesizes enhanced emissivity from accreting matter at a radius
433: where two of the three coordinate frequencies (i.e. azimuthal, radial,
434: and polar) have commensurate values that match (either directly or via
435: beats) the observed QPO frequencies. For the cases with optical
436: determinations of the black hole mass, the value of $a_*$ can be
437: determined via the application of this resonance model if the correct
438: pair of coordinate frequencies can be identified. Reasonable values
439: ($0.25 < a_* < 0.95$) are derived from the observed HFQPOs for either
440: 2:1 or 3:1 ratios in either orbital:radial or polar:radial coordinate
441: frequencies \citep{abr01,rem02a}. The driving mechanism that would
442: allow accretion blobs to grow and survive at the resonance radius has
443: not been specified, and it is known that there are severe damping
444: forces in the inner accretion disk \citep{mark98}. On the other hand,
445: MHD accretion simulations under GR do show transient condensations in
446: the inner disk \citep{haw01}. Ray-tracing calculations under GR
447: \citep{schn04} show that the putative blobs could indeed produce the
448: HFQPO patterns, and that the appearance of $3~\nu_0$ versus
449: $2~\nu_0$ for the stronger QPO is governed by the angular width
450: of the blob. Clearly, more work is needed to
451: investigate this resonance model.
452:
453: An alternative scenario is to extend the models for ``diskoseismic''
454: oscillations to include non-linear effects that might drive some type
455: of resonant oscillation. Diskoseismology treats the inner disk as a
456: resonance cavity in the Kerr metric \citep{kat01,wag99}. Normal modes
457: have been derived for linear perturbations, and the extension of this
458: theory would be both very interesting and difficult. There are also
459: models that invoke oscillations from geometries other than a thin
460: disk, e.g. the accretion torus of Rezzolla et al. (2003).
461: \nocite{rez03} In any case, BHB HFQPOs deserve careful study as a
462: potential opportunity to derive primary information about black holes
463: while developing astrophysical applications for the strong-field regime
464: of GR theory.
465:
466: \acknowledgements Partial support for R.R. was provided by the NASA
467: contract to MIT for RXTE instruments. J.M. acknowledges partial
468: support from NASA grant NAG5-10813. We thank Jeroen Homan for
469: productive discussions about H1743-322.
470:
471:
472: \begin{thebibliography}{}
473:
474: \bibitem[Abramowicz \& Kluzniak 2001]{abr01}Abramowicz, M. A., \&
475: Kluzniak, W., 2001, A\&A, 374, L19
476:
477: \bibitem[Abramowicz et al. 2003]{abr03}Abramowicz, M. A., Bulik, T.,
478: Bursa, M., \& Kluzniak, W., 2003 A\&A, 404, L21
479:
480: \bibitem[Cooke et al. 1984]{coo84}Cooke, B. A., Levine, A. M., Lang, F.
481: L., Primini, F. A., \& Lewin, W. G. H. 1984, ApJ, 285, 258
482:
483: \bibitem[Cui et al. 2000]{cui00}Cui, W., Shrader, C. R., Haswell, C. A.,
484: \& Hynes, R. I. 2000, \apj, 535, L123
485:
486: \bibitem[Gursky et al. 1978]{gur78}Gursky et al. 1978, ApJ, 223, 973
487:
488: \bibitem[Hawley \& Krolik 2001]{haw01}
489: Hawley, J.F. \& Krolik, J.H. 2001, ApJ, 548, 348
490:
491: \bibitem[Homan et al. 2003]{hom03}Homan, J., Miller, J. M., Wijnands, R.,
492: Steeghs, D., Belloni, T., van der Klis, M., \& Lewin, W. G. H. 2003, ATEL, \#162
493:
494: \bibitem[Kato 2001]{kat01}Kato, S. 2001, PASJ, 53, 1
495:
496: \bibitem[Levine et al.]{lev96}
497: Levine, A. M., et al. 1996, \apjl, 469, L33
498:
499: \bibitem[Markovic \& Lamb 1998]{mark98}
500: Markovic, D. \& Lamb, F.K. 1998, ApJ, 507, 316
501:
502: \bibitem[Markwardt \& Swank 2003]{mar03}Markwardt, C. B., \& Swank, J.
503: H. 2003, ATEL \#136
504:
505: \bibitem[McClintock \& Remillard 2003]{mcrem03}McClintock, J. E. \&
506: Remillard, R. A. 2003, in "Compact Stellar X-ray Sources," eds. W.H.G. Lewin \& M. van
507: der Klis, (Cambridge: Cambridge U. Press), in press; astro-ph/0306213
508:
509: \bibitem[Merloni et al. 1999]{mer99}Merloni, A., Vietri, M., Stella, L.,
510: \& Bini, D. 1999, \mnras, 304, 155
511:
512: \bibitem[Orosz et al. 2004]{oro04}Orosz, J. A., McClintock, J. E.,
513: Remillard, R. A., \& Corbel, S. 2004, ApJ, submitted; astro-ph/0404343
514:
515: \bibitem[Parmar et al. 2003]{par03}Parmar, A. N., Kuulkers, E.,
516: Oosterbrock, T., Barr, P., Much, R., Orr, A., Williams, O. R., \&
517: Winkler, C. 2003, A\&A, 411, L421
518:
519: \bibitem[Remillard et al. 1999]{rem99}Remillard, R. A., Morgan, E. H.,
520: McClintock, J. E., Bailyn, C. D., \& Orosz, J. A. 1999, \apj, 522, 397
521:
522: \bibitem[Remillard et al. 2002a]{rem02a}Remillard, R. A., Muno, M. P.,
523: McClintock, J. E. \& Orosz, J. A. 2002a, ApJ, 580, 1030
524:
525: \bibitem[Remillard et al. 2002b]{rem02b}Remillard, R. A., Muno, M. P.,
526: McClintock, J. E. \& Orosz, J. A. 2002b, in ``New Views on Microquasars'',
527: Procs. of the 4th Microquasar Workshop, eds. Durouchoux, Fuchs, \&
528: Rodriguez, (Center for Physics: Kolkata), 57 (astro-ph/0208402)
529:
530: \bibitem[Remillard et al. 2002c]{rem02c}Remillard, R. A., Sobczak, G. J.,
531: Muno, M., \& McClintock, J. E. 2002c, \apj, 564, 962
532:
533: \bibitem[Remillard et al. 2003]{rem03}Remillard, R. A., Muno, M. P.,
534: McClintock, J. E. \& Orosz, J. A. 2003, BAAS, 35, 648
535:
536: \bibitem[Revnivtsev et al. 2003]{rev03}Revnivtsev, M., Chernyakova, M.,
537: Capitanio, F., Westergaard, N. J., Shoenfelder, V., Gehrels, N., \&
538: Winkler, C. 2003, ATEL, \#132
539:
540: \bibitem[Rupen, Mioduszewski, \& Dhawan 2003a]{rup03a} Rupen, M. P.,
541: Mioduszewski, A. J., \& Dhawan, V. 2003, ATEL, \#137
542:
543: \bibitem[Rupen, Mioduszewski, \& Dhawan 2003b]{rup03b} Rupen, M. P.,
544: Mioduszewski, A. J., \& Dhawan, V. 2003, ATEL, \#139
545:
546: \bibitem[Rupen, Mioduszewski, \& Dhawan 2003c]{rup03c} Rupen, M. P.,
547: Mioduszewski, A. J., \& Dhawan, V. 2003, ATEL, \#142
548:
549: \bibitem[Rezzolla et al. 2003]{rez03}
550: Rezzolla, L., Yoshida, S'i, Maccarone, T. J., \& Zanotti, O. 2003,
551: MNRAS, 344, L37
552:
553: \bibitem[Schnittman \& Bertschinger 2004]{schn04}
554: Schnittman, J. D. \& Bertschinger, E. 2004, ApJ, in press;
555: astro-ph/0309458
556:
557: \bibitem[Steeghs et al. 2003]{ste03}Steeghs, D., Miller, J. M., Kaplan,
558: D., \& Rupen, M. 2003, ATEL \#146
559:
560: \bibitem[Stella, Vietri, \& Morsink 1999]{ste99}Stella, L., Vietri, M.
561: \& Morsink, S. M. 1999, ApJ, 524, L63
562:
563: \bibitem[Strohmayer 2001]{str01}
564: Strohmayer, T.E. 2001, ApJ, 552, L49
565:
566: \bibitem[Tagger et al. 2004]{tag04}Tagger, M., Varniere, P., Rodriguez,
567: J. \& Pellat, R. 2004, ApJ, in press; astro-ph/0401539
568:
569: \bibitem[Tanaka \& Lewin 1995]{tan95}
570: Tanaka, Y. \& Lewin, W.H.G. 1995, in ``X--ray Binaries'',
571: eds. W.H.G. Lewin, J. van Paradijs \& E.P.J. van den Heuvel,
572: (Cambridge:Cambridge U. Press), 126
573:
574: \bibitem[Wagoner 1999]{wag99}Wagoner, R. V. 1999, Physics Reports, 311,
575: 259
576:
577: \bibitem[Wagoner, Silbergleit, \& Ortega-Rodriguez 2001]{wag01}Wagoner,
578: R. V., Silbergleit, A. S., \& Ortega-Rodriguez, M. 2001, \apj, 559, L25
579:
580: \end{thebibliography}
581:
582:
583: \clearpage
584:
585: \begin{deluxetable}{rllcrrccc}
586: \tablenum{1}
587: %\tabletypesize \small
588: %\tabletypesize \footnotesize
589: \tabletypesize \scriptsize
590: \tablecaption{RXTE Observations of H1743-322: Program 80146}
591: \tablehead{
592: \colhead{\#} & \colhead{RXTE\tablenotemark{a}} & \colhead{MJD\tablenotemark{b}} &
593: \colhead{Num.} & \colhead{Exposure} & \colhead{PCA Rate\tablenotemark{c}} &
594: \colhead{PCA Rate\tablenotemark{d}} & \colhead{rms\tablenotemark{e}} &
595: \colhead{PDS} \\
596: \colhead{} & \colhead{Day} & \colhead{Time} & \colhead{Obs.} &
597: \colhead{(s)} &
598: \colhead{2-35 keV} & \colhead{7-35 keV} & \colhead{(0.1-10 Hz)} &
599: \colhead{Type} \\
600: }
601:
602: \startdata
603: 1 & 3390 & 52743.24 & 1 & 2679 & 1940.2 & 461.5 & 0.112 & q-bl \\
604: 2 & 3391 & 52744.22 & 1 & 3059 & 1550.1 & 398.4 & 0.137 & q-bl \\
605: 3 & 3393 & 52746.20 & 1 & 2910 & 708.8 & 193.0 & 0.151 & q-bl \\
606: 4 & 3394 & 52747.63 & 1 & 3206 & 1024.2 & 247.6 & 0.117 & q-bl \\
607: 5 & 3397a & 52750.31 & 2 & 5783 & 2320.3 & 608.2 & 0.050 & q \\
608: 6 & 3397b & 52750.81 & 1 & 15520 & 2061.4 & 495.2 & 0.101 & q \\
609: 7 & 3398a & 52751.09 & 2 & 6899 & 1660.9 & 384.3 & 0.101 & q \\
610: 8 & 3398b & 52751.32 & 2 & 8406 & 3478.3 & 882.2 & 0.085 & q \\
611: 9 & 3398c & 52751.71 & 1 & 3282 & 1763.3 & 365.6 & 0.104 & q \\
612: 10 & 3398d & 52751.99 & 1 & 3653 & 1694.8 & 326.9 & 0.089 & q \\
613: 11 & 3399 & 52752.97 & 1 & 10764 & 2281.3 & 575.7 & 0.077 & q \\
614: 12 & 3400 & 52753.17 & 1 & 2978 & 2255.7 & 596.4 & 0.048 & 0 \\
615: 13 & 3401 & 52754.58 & 2 & 3542 & 1743.5 & 383.4 & 0.117 & q-bl \\
616: 14 & 3402 & 52755.96 & 1 & 7168 & 2637.8 & 414.6 & 0.073 & q \\
617: 15 & 3403a & 52756.23 & 1 & 3541 & 2303.8 & 364.1 & 0.069 & q \\
618: 16 & 3403b & 52756.71 & 1 & 3158 & 1234.5 & 204.2 & 0.074 & q \\
619: 17 & 3404 & 52757.80 & 1 & 6977 & 1108.4 & 136.7 & 0.070 & 0 \\
620: 18 & 3405 & 52758.99 & 1 & 6999 & 2021.7 & 194.9 & 0.055 & 0 \\
621: 19 & 3407a & 52760.08 & 1 & 3439 & 1130.0 & 145.7 & 0.063 & 0 \\
622: 20 & 3407b & 52760.56 & 2 & 3809 & 1152.9 & 145.1 & 0.071 & 0 \\
623: 21 & 3408 & 52761.61 & 1 & 5214 & 1694.6 & 216.0 & 0.066 & 0 \\
624: 22 & 3409 & 52762.75 & 1 & 6533 & 1733.9 & 256.1 & 0.067 & 0 \\
625: 23 & 3410a & 52763.10 & 1 & 3418 & 1985.9 & 300.3 & 0.059 & q \\
626: 24 & 3410b & 52763.62 & 1 & 2769 & 2219.5 & 390.0 & 0.061 & q \\
627: 25 & 3411 & 52764.91 & 1 & 6755 & 2228.6 & 392.5 & 0.062 & q \\
628: 26 & 3412 & 52765.87 & 1 & 3398 & 3775.2 & 1083.3 & 0.047 & q \\
629: 27 & 3413 & 52766.57 & 1 & 2558 & 1569.8 & 392.9 & 0.143 & q-bl \\
630: 28 & 3414 & 52767.87 & 1 & 6784 & 1292.2 & 352.2 & 0.159 & q-bl \\
631: 29 & 3415 & 52768.55 & 1 & 2154 & 1307.5 & 336.5 & 0.149 & q-bl \\
632: 30 & 3416 & 52769.75 & 1 & 2433 & 1227.6 & 326.0 & 0.159 & q-bl \\
633: 31 & 3417 & 52770.53 & 2 & 3729 & 1533.1 & 387.0 & 0.140 & q-bl \\
634: 32 & 3418 & 52771.87 & 2 & 7672 & 1066.5 & 347.5 & 0.192 & q-bl \\
635: 33 & 3419 & 52772.74 & 1 & 6068 & 1000.1 & 368.6 & 0.214 & q-bl \\
636: 34 & 3420 & 52773.73 & 1 & 6757 & 994.0 & 365.3 & 0.212 & q-bl \\
637: 35 & 3421 & 52774.58 & 1 & 6702 & 1036.4 & 330.9 & 0.194 & q-bl \\
638: 36 & 3422 & 52775.63 & 1 & 6708 & 953.6 & 339.1 & 0.212 & q-bl \\
639: 37 & 3423 & 52776.68 & 1 & 6607 & 712.7 & 280.1 & 0.227 & q-bl \\
640: 38 & 3424 & 52777.67 & 1 & 6588 & 979.7 & 333.6 & 0.207 & q-bl \\
641: 39 & 3425 & 52778.53 & 1 & 6550 & 1012.4 & 319.8 & 0.192 & q-bl \\
642: 40 & 3426 & 52779.58 & 2 & 4873 & 1073.7 & 316.4 & 0.186 & q-bl \\
643: 41 & 3427 & 52780.60 & 1 & 3395 & 1079.9 & 317.4 & 0.184 & q-bl \\
644: 42 & 3428 & 52781.62 & 1 & 6805 & 1046.5 & 314.7 & 0.191 & q-bl \\
645: 43 & 3429 & 52782.70 & 1 & 3265 & 1093.2 & 299.9 & 0.173 & q-bl \\
646: 44 & 3430 & 52783.51 & 1 & 6529 & 1430.5 & 357.0 & 0.129 & q-bl \\
647: 45 & 3431 & 52784.57 & 1 & 6721 & 2082.5 & 409.6 & 0.078 & q \\
648: 46 & 3432 & 52785.48 & 1 & 6711 & 2158.8 & 432.0 & 0.083 & q \\
649: 47 & 3433 & 52786.35 & 1 & 6433 & 2517.7 & 597.7 & 0.067 & q \\
650: 48 & 3435 & 52788.51 & 2 & 7095 & 1606.6 & 240.0 & 0.056 & q \\
651: 49 & 3436 & 52789.26 & 1 & 3477 & 1908.4 & 327.2 & 0.057 & q \\
652: 50 & 3437 & 52790.24 & 2 & 3990 & 1781.1 & 296.6 & 0.055 & q \\
653: 51 & 3438 & 52791.63 & 1 & 5796 & 1717.7 & 272.1 & 0.061 & q \\
654: 52 & 3439 & 52792.39 & 2 & 6737 & 1755.4 & 296.6 & 0.056 & q \\
655: 53 & 3440 & 52793.61 & 1 & 3138 & 1785.1 & 286.8 & 0.052 & q \\
656: 54 & 3441 & 52794.56 & 1 & 7321 & 2558.6 & 636.5 & 0.030 & q \\
657: 55 & 3442 & 52795.38 & 2 & 3741 & 1587.5 & 236.5 & 0.050 & q \\
658: 56 & 3443 & 52796.27 & 1 & 7018 & 1785.1 & 307.2 & 0.060 & q \\
659: 57 & 3444 & 52797.58 & 1 & 7159 & 1806.0 & 325.6 & 0.066 & q \\
660: 58 & 3445 & 52798.54 & 1 & 3297 & 2647.5 & 675.5 & 0.029 & q \\
661: 59 & 3446 & 52799.46 & 1 & 2998 & 1561.3 & 233.7 & 0.048 & q \\
662: 60 & 3447 & 52800.73 & 2 & 11651 & 1528.1 & 215.4 & 0.035 & q \\
663: 61 & 3448 & 52801.91 & 1 & 4856 & 1619.8 & 260.7 & 0.055 & q \\
664: 62 & 3449 & 52802.96 & 1 & 4563 & 1614.5 & 272.1 & 0.057 & q \\
665: 63 & 3450 & 52803.57 & 1 & 6352 & 1539.1 & 259.6 & 0.058 & q \\
666: 64 & 3451 & 52804.62 & 1 & 6803 & 1607.7 & 298.8 & 0.069 & q \\
667: 65 & 3452 & 52805.47 & 1 & 5030 & 1577.6 & 278.3 & 0.063 & q \\
668: 66 & 3453 & 52806.63 & 1 & 3321 & 1361.0 & 187.3 & 0.032 & 0 \\
669: 67 & 3454 & 52807.62 & 1 & 2431 & 1366.3 & 170.1 & 0.033 & 0 \\
670: 68 & 3455 & 52808.64 & 1 & 5880 & 1311.8 & 162.4 & 0.032 & 0 \\
671: 69 & 3456 & 52809.62 & 1 & 5203 & 1261.5 & 145.1 & 0.033 & 0 \\
672: 70 & 3457 & 52810.54 & 1 & 5768 & 1137.5 & 101.8 & 0.033 & 0 \\
673: 71 & 3458 & 52811.46 & 2 & 4235 & 1160.8 & 129.1 & 0.035 & 0 \\
674: 72 & 3459 & 52812.67 & 1 & 4844 & 1132.5 & 120.9 & 0.038 & 0 \\
675: 73 & 3461 & 52814.47 & 1 & 3930 & 1112.8 & 102.3 & 0.027 & 0 \\
676: 74 & 3462 & 52815.84 & 1 & 3303 & 1003.0 & 77.0 & 0.030 & 0 \\
677: 75 & 3463 & 52816.58 & 1 & 4889 & 978.9 & 75.7 & 0.030 & 0 \\
678: 76 & 3464 & 52817.51 & 1 & 5704 & 1007.9 & 80.0 & 0.029 & 0 \\
679: 77 & 3465 & 52818.55 & 1 & 4621 & 994.1 & 74.4 & 0.028 & 0 \\
680: 78 & 3466 & 52819.34 & 2 & 2731 & 970.2 & 73.2 & 0.029 & 0 \\
681: 79 & 3467 & 52820.43 & 1 & 2418 & 943.9 & 63.6 & 0.029 & 0 \\
682: 80 & 3468 & 52821.42 & 1 & 2331 & 927.6 & 58.1 & 0.028 & 0 \\
683: 81 & 3469 & 52822.53 & 1 & 4410 & 922.1 & 57.1 & 0.025 & 0 \\
684: 82 & 3470 & 52823.48 & 1 & 4759 & 873.8 & 53.2 & 0.024 & 0 \\
685: 83 & 3471 & 52824.45 & 1 & 3329 & 881.8 & 52.1 & 0.027 & 0 \\
686: 84 & 3472 & 52825.37 & 1 & 2640 & 870.8 & 51.0 & 0.030 & 0 \\
687: 85 & 3473 & 52826.42 & 1 & 3370 & 893.7 & 54.8 & 0.025 & 0 \\
688: 86 & 3474 & 52827.41 & 1 & 3274 & 864.8 & 52.6 & 0.028 & 0 \\
689: 87 & 3475 & 52828.37 & 1 & 6712 & 869.0 & 51.7 & 0.026 & 0 \\
690: 88 & 3476 & 52829.28 & 2 & 5002 & 839.2 & 49.3 & 0.025 & 0 \\
691: 89 & 3477 & 52830.19 & 2 & 949 & 863.9 & 52.8 & 0.027 & 0 \\
692: 90 & 3478 & 52831.48 & 1 & 3371 & 860.5 & 51.2 & 0.025 & 0 \\
693: 91 & 3481 & 52834.21 & 2 & 4408 & 838.0 & 49.3 & 0.023 & 0 \\
694: 92 & 3484 & 52837.20 & 1 & 3492 & 834.4 & 47.4 & 0.024 & 0 \\
695: 93 & 3487 & 52840.56 & 1 & 2137 & 837.5 & 49.9 & 0.027 & 0 \\
696: 94 & 3490 & 52843.51 & 1 & 3179 & 816.3 & 52.9 & 0.026 & 0 \\
697: 95 & 3493 & 52846.17 & 2 & 4941 & 792.2 & 45.0 & 0.024 & 0 \\
698: 96 & 3499 & 52852.99 & 1 & 1770 & 749.2 & 41.0 & 0.027 & 0 \\
699: 97 & 3502 & 52855.54 & 1 & 3696 & 722.0 & 37.5 & 0.022 & 0 \\
700: 98 & 3509 & 52862.38 & 1 & 1766 & 668.4 & 35.1 & 0.029 & 0 \\
701: 99 & 3512 & 52865.41 & 1 & 1952 & 630.1 & 31.9 & 0.031 & 0 \\
702: 100 & 3516 & 52869.35 & 1 & 1463 & 600.6 & 31.1 & 0.032 & 0 \\
703: 101 & 3522 & 52875.12 & 1 & 1655 & 569.6 & 28.3 & 0.028 & 0 \\
704: 102 & 3524 & 52877.70 & 1 & 676 & 557.9 & 26.7 & 0.028 & 0 \\
705: 103 & 3530 & 52883.29 & 1 & 3043 & 511.4 & 24.8 & 0.029 & 0 \\
706: 104 & 3533 & 52886.57 & 1 & 1919 & 498.1 & 20.5 & 0.030 & 0 \\
707: 105 & 3535 & 52888.93 & 1 & 2342 & 493.6 & 23.3 & 0.023 & 0 \\
708: 106 & 3538 & 52891.43 & 1 & 2801 & 504.4 & 32.3 & 0.025 & 0 \\
709: 107 & 3541 & 52894.05 & 1 & 1062 & 489.2 & 31.4 & 0.028 & 0 \\
710: 108 & 3545 & 52898.72 & 1 & 953 & 426.2 & 26.3 & 0.032 & 0 \\
711: 109 & 3547 & 52900.88 & 1 & 926 & 419.9 & 19.5 & 0.033 & 0 \\
712: 110 & 3550 & 52903.99 & 1 & 769 & 392.3 & 19.6 & 0.033 & 0 \\
713: 111 & 3552 & 52905.96 & 1 & 537 & 385.0 & 22.8 & 0.036 & 0 \\
714:
715: \enddata
716:
717: \tablenotetext{a}{RXTE day is the mission MET time in units of truncated days.}
718: \tablenotetext{b}{Midpoint of observation, $MJD=JD-2,400,000.5$.}
719: \tablenotetext{c}{Source count rate in PCA at 2--35 keV, using PCU\#2.}
720: \tablenotetext{d}{Source count rate in PCA at 7--35 keV, using PCU\#2.}
721: \tablenotetext{e}{RMS source fluctuations at 2-35 keV, integrated over the range 0.1--10 Hz, and expressed as a fraction of the average source count rate.}
722:
723: \label{tab:obs}
724:
725: \end{deluxetable}
726:
727:
728: \clearpage
729:
730:
731: \begin{figure}
732:
733: \figurenum{1} \plotone{h1743_asm.ps}
734: \caption{\xte ASM light curve (1.5-12 keV) and hardness ratio (HR2)
735: covering the 2003 outburst of H1743-322. For reference, the count rate
736: for the Crab Nebula is 75.5 ASM c/s. The tick marks in the top panel
737: indicate the times of RXTE pointed observations considered in this
738: paper. The arrows show the times of observations contributing
739: to QPO detections at 166 Hz (upper row) and 242 Hz (lower row),
740: as explained in \S 3. \label{fig:asm}}
741:
742: \end{figure}
743:
744:
745: \newpage
746:
747:
748: \begin{figure}
749:
750: \figurenum{2} \plotone{pds_types.ps}
751: \caption{Average PDS, in units of log ($\nu \times P_{\nu}$), for
752: observations of H1743-322 grouped by the properties of the power
753: continua and the X-ray flux. The power spectra are computed for
754: photons in the range 2-35 keV, except for the group in the
755: bottom-right panel, where the PDS at 7--35 keV is shown. HFQPOs are
756: not seen when there is a band-limited power-continuum (group ``Q-BL'';
757: see text), nor when LFQPOs are absent (during times that correlate
758: with very soft spectra). However, HFQPOs are seen (right panels) when
759: LFQPOs are present and the band-limited power continuum is diminished.
760: Furthermore, the HFQPOs appears to switch between two commensurate
761: frequencies when these cases are divided into two ranges of luminosity
762: above 7 keV. \label{fig:pdstypes}}
763:
764: \end{figure}
765:
766:
767: \newpage
768:
769:
770: \begin{figure}
771:
772: \figurenum{3} \plotone{hfqpo_mdl.ps}
773: \caption{Magnified view of the high-frequency QPOs for the same two
774: groups that have their average PDS displayed in the right panels of Fig. 2.
775: The solid lines show the QPO profile fits, which assume a Lorentzian
776: profile above the local power continuum. The fits are computed and
777: displayed in units of $P_{\nu}$, rather than the units of $\nu \times P_{\nu}$
778: used in Fig. 2. The central frequencies are $166 \pm 5$ Hz for the high
779: $L_x$ group at 2-35 keV, and $242 \pm 3$ Hz for the lower $L_x$ group at 7-35
780: keV. These results are statistically consistent with an interpretation
781: as commensurate frequencies scaled in a 3:2 ratio. \label{fig:pdsmod} }
782:
783: \end{figure}
784:
785:
786: \end{document}
787:
788:
789: