Detection of a possible X-ray Quasi-periodic Oscillation in the Active Galactic Nucleus 1H~0707-495

0.3 MB·

by  Hai-Wu Pan

Detection of a possible X-ray Quasi-periodic Oscillation in the
Active Galactic Nucleus 1H 0707-495
6 Hai-Wu Pan1,2, Weimin Yuan1,2, Su Yao1,2, Xin-Lin Zhou1, Bifang Liu1,2, Hongyan Zhou3,4
1
0 & Shuang-Nan Zhang5,1
2

n
a
J
ABSTRACT
8
2
Quasi-periodicoscillation(QPO)detected in the X-rayradiationof blackhole X-raybinaries

(BHXBs)is thoughttooriginatefromdynamicalprocessesinthe closevicinityofthe blackholes
]
E (BHs), andthus carriesimportant physicalinformationtherein. Such a feature is extremely rare
H in active galactic nuclei (AGNs) with supermassive BHs. Here we report on the detection of a
. possible X-ray QPO signal with a period of 3800s at a confidence level >99.99%in the narrow-
h
lineSeyfert1galaxy(NLS1)1H0707-495inonedatasetin0.2-10keVtakenwithXMM-Newton.
p
- The statistical significance is higher than that of most previously reported QPOs in AGNs. The
o QPO is highly coherent (quality factor Q = ν/∆ν > 15) with a high rms fractional variability
r
t (∼15%). AcomprehensiveanalysisoftheopticalspectraofthisAGNisalsoperformed,yielding
as a central BH mass 5.2×106M⊙ from the broad emission lines based on the scaling relation.
[ The QPO follows closely the known frequency-BH mass relation, which spans from stellar-mass

to supermassive BHs. The absence of the QPO in other observations of the object suggests it
1
a transient phenomenon. We suggest that the (high-frequency) QPOs tend to occur in highly
v
9 accreting BH systems, from BHXBs to supermassive BHs. Future precise estimation of the BH
3 mass may be used to infer the BH spin from the QPO frequency.
6
7 Subjectheadings: galaxies: active—galaxies: nuclei—X-rays: galaxies—galaxies: individual(1H0707-
0 495)
.
1
1. INTRODUCTION 2012; Markoff et al. 2015; Scaringi et al. 2015).
0
6 A compelling line of evidence for this postula-
1 Active galactic nuclei (AGNs), powered by tion is the striking similarity in the variability of
: black hole (BH) accretion with BHs of 105 − the X-ray radiation between AGNs and BHXBs
v
i 109M⊙ at the center of galaxies, are thought to (e.g., Uttley et al. 2002; Markowitz et al. 2003;
X be scaled-up versions of black hole X-ray binaries
McHardy et al.2006;McHardy2010;Vaughan et al.
r (BHXBs)(McHardy2010;Gonz´alez-Mart´ın& Vaughan 2003a, 2011). One characteristic and enigmatic
a
feature of the variable X-rays is Quasi-Periodic
1Key Laboratory of Space Astronomy and Technology, Oscillation (QPO), which has been observed in
NationalAstronomicalObservatories,ChineseAcademyof
the X-ray light curves of dozens of BHXBs (e.g.,
Sciences, 20A Datun Road, Chaoyang District, Beijing,
China;[email protected], [email protected] Strohmayer 2001; Remillard et al. 2002). X-
2University of Chinese Academy of Sciences, School of ray QPOs were found to have two types, low-
AstronomyandSpaceScience, Beijing100049, China frequency QPOs (LFQPOs) and high-frequency
3Polar Research Institute of China, 451 Jinqiao Road, QPOs(HFQPOs)(Remillard & McClintock2006,
Pudong,Shanghai 200136, China
hereafter RM06). HFQPOs sometimes occur
4KeyLaboratoryforResearchinGalaxiesandCosmol-
in pairs with a constant frequency ratio of 3:2,
ogy, The University of Sciences and Technology of China,
ChineseAcademyofSciences, Hefei,Anhui230026, China and unlike LFQPOs, their frequencies (fQPO)
5Key Laboratory of Particle Astrophysics, Institute of are stable despite significant variations in fluxes
HighEnergyPhysics,ChineseAcademyofSciences,Beijing (Remillard et al. 2002). This indicates that
100049, China
1
HFQPO is likely a fundamental property that to reduce the data and extract science products
might be linked to the mass (and spin) of the from the observation data files (ODFs) using the
BH. Based on only three BHXBs with mea- XMM-NewtonScienceAnalysisSystem(SAS,ver-
sured BH masses, RM06 suggested an inverse lin- sion 13.0.0). Only good events (single and double
ear relation between f and BH mass, which pixel events, i.e. PATTERN ≤ 4 for PN or ≤ 12
QPO
was later extended to intermediate-mass BH for MOS) are used. Source events are extracted
(IMBH) in ultraluminous X-ray sources (ULXs) from a 40-arcsec circular region, and background
by Abramowicz et al. (2004). A number of events from a source-free circle with the same ra-
HFQPO models have been proposed in the liter- dius on the same CCD chip. X-ray light curves
ature (see Lai & Tsang 2009 and Belloni & Stella are constructed for all the three EPIC detectors
2014 for reviews and references therein). in the 0.2-10keV energy band with a binsize of
In AGNs with supermassive BHs (SMBHs), 100s, and are corrected for instrumental factors
however, QPOs are rarely detected. So far, using EPICLCCORR.
there is only one widely accepted case in which The combined PN+MOS1+MOS2 light curve
a significant QPO was unambiguously detected is shown in Figure 1. Even by eye, a periodic-
in the Narrow-Line Seyfert 1 (NLS1) galaxy ity can be clearly seen. We search for periodicity
RE J1034+396 with f ≃ 2.7 × 10−4Hz by calculating the root-mean-square (rms) ampli-
QPO
(Gierlin´ski et al.2008;Alston et al.2014;Hu et al. tudesofthelightcurvefoldedwithvariousperiods
2014). Recently, A ∼ 200s X-ray QPO was de- as a function of period. A strong peak appears at
tectedinthetransientX-raysourceSwiftJ164449.3+57345a1r,ound 3800s. It is also found that this peak is
thoughttobeoftidaldisruptionofastarbyador- the strongest if the latter half of the light curve
mant SMBH (Reis et al. 2012). By extending BH is used only, with a quasi-period of 3800±170s
masses to the SMBH range, Zhou et al. (2015) (full-width at half-maximum). We thus take an
suggestedthatthe f −M scalingrelationis operationalcutasindicatedinFigure1(solidver-
QPO BH
universal spanning ∼ 6 orders of magnitude from tical line) and consider the last ∼ 55ks segment
stellar-mass BHs to SMBHs. onlyhereafter,inordertoachievethehighestcon-
Besides RE J1034+396, possible detections of fidence level for the periodic signal. This segment
X-ray QPOs were also reported in a few other has a mean count rate of 6.1countss−1 and frac-
AGNs, however, at relatively low levels of signifi- tional rms variability 41%, while the whole light
cance. A∼3.8hrQPOin2XMMJ123103.2+110648 curve gives 6.7countss−1 and 37%, respectively.
was reported and suggested to be a LFQPO The folded light curve with a period 3800s is also
(Lin et al. 2013), and a ∼ 2hr HFQPO was also showed in Figure 1 (inset), with the best-fit sinu-
reported in MS 2254.9-3712(Alston et al. 2015). soid model.
In this letter, we report on the discovery of a Thepowerspectrumdensity (PSD)ofthe light
significant QPO signal in one XMM-Newton ob- curve is computed as the modulus-squared of
servationof1H0707-495,anearby(redshift0.04), the discrete Fourier transform (DFT), and the
typical NLS1 (V´eron-Cetty & V´eron 2010). We (rms/mean)2normalisationischosen(Vaughan et al.
also re-examine the BH mass of 1H 0707-495 by 2003b). The PSD for the last 55ks segment
analysing its available optical spectroscopic data, is plotted in Figure 2, and a strong peak at
and compare the QPO with the f −M re- 2.6 × 10−4Hz is evident. The continuum (red
QPO BH
lation. A cosmology with H = 70kms−1Mpc−1, noise) is well fitted with a power-law of a slope
0
Ω =0.3 and Ω =0.7 is adopted. −1.90 (χ2 = 301/282 dof, in the log-log space).
m Λ
This signal is even more prominent in the resid-
2. X-RAYQUASI-PERIODICOSCILLA- uals plot (data/model). To test its statistical
TION significance, a Monte Carlo technique is applied
following Reis et al. (2012) and Lin et al. (2013).
1H 0707-495 was observed with XMM-Newton We simulate one million light curves using the
with an exposure of ∼ 100ks on February 6 method of Timmer & Koenig (1995) from a PSD,
2008 in the full frame imaging mode (Obs ID: which is assumed to be the above best-fit power-
0511580401). We follow the standard procedure law. Thusthedistributionofthevariabilitypower
2
ateachfourierfrequencycanbeobtainedfromthe
PSDs of the simulated light curves. The dashed
line inFigure 2 representsthe 99.99%significance
level,indicatingasignificantQPO.Usingthedata
ofPNcameraandthecombinedMOScamerasin-
dividually leads to the same result with the QPO
significance levels >99.99% in both cases.
TotesttheQPOinREJ1034+396,Gierlin´ski et al.
(2008) used a method developed by Vaughan
(2005). This method was later improved signif-
icantly by Vaughan (2010) to be an even more
stringent test based on Bayesian statistics, in Fig. 1.— XMM-Newton light curve of 1H 0707-
combination with PSD fitting. We also adopt 495, which is extracted and combined from the
this improved method here. Instead of a simple PN,MOS1andMOS2detectorsin0.2-10keVwith
power-law (H0 model), a bending power-law (H1 a binsize of 100s. The light curve is divided into
model) is used to fit the PSD continuum. The twosegmentsbythesolidline,andonlythesecond
low-frequency slope is fixed at -1, while the high- segment is used in this letter. The dotted verti-
frequency slope is fitted to be −2.4±0.22, with a cal lines show the expected periodicity of 3800s,
bendingfrequencyof1.4×10−4Hz. ThenMarkov whichcanbeseenfromthelightcurveevenbyeye.
Chain Monte Carlo simulations are performed, Upper Right Inset: the folded light curve with a
resulting a small posterior predictive p-value of period3800s. Errorsarepropagatedfromthe un-
2.5(±0.5)×10−3 for the significant outlier at the foldedcurve. Thebest-fitsinusoidisshowedasthe
frequency ∼2.6×10−4Hz [p=3.9(±0.62)×10−3 solid line and the mean count rate as the dotted
for the PN light curve and p=4.0(±0.63)×10−3 line. Two cycles are plotted for clarity.
for the combined MOS light curve, respectively].
This strongly indicates the presence of the QPO.
This periodicity of ∼ 2.6× 10−4Hz is highly
coherent and confined to only one frequency bin
in our analysis. Considering the frequency resolu-
tion for the observation (1/T = 1.75×10−5Hz),
the quality factor (Q = ν/∆ν > 15) is high. The
rms fractional variability in the QPO is ∼ 15%.
Compared to the rare previously reported QPOs
in AGNs in the literature, the QPO in 1H 0707-
495 has a significance level at least similar to
or even higher than most of the others but only
somewhat lower than that in RE J1034+396 1.
The significanceofthis QPOwouldbe reducedto
< 90% if the whole observation duration is con-
sidered. We alsoanalyzeddataofother 14XMM-
Newtonobservationsof1H0707-495withexposure
time longer than 40ks, but no significant QPO is Fig. 2.— Power spectrum density of the X-ray
found. Thus the QPO appears to be a transient light curve of 1H 0707-495. The solid line repre-
feature; similar to that found in RE J1034+396 sents the best-fit power-law, and the dashed line
(Gierlin´ski et al. 2008; Middleton et al. 2011) represents the 99.99% confidence level. The dot-
ted horizontalline showsthe expected levelof the
1The p-value for RE J1034+396 is ∼ 5× 10−4, if only Poissonian noise. The data/model residuals are
the second segment of its light curve is used, following
shown in the lower panel. A statistically signifi-
Gierlin´skietal.(2008).
cant peak is clearly present at ∼ 2.6×10−4Hz.
3
3. BLACK HOLE MASS ESTIMATION
AND THE f −M RELATION
QPO BH
Using the empiricalvirialmethod basedon op-
ticalspectroscopicdata,the BHmass of1H0707-
495 has been estimated in previous studies, as
2+−41×106M⊙ froma6dFspectrum(Bian & Zhao
2003) and 4 × 106M⊙ (Done & Jin 2015) from
CTIOspectrumtakenbyLeighly & Moore(2004).
Were-analyzeboththespectra,followingthepro-
cedure as described in Yao et al. (2015) to fit
the optical spetra. Only the CTIO spectrum is
shown in Figure 3 for demonstration. The flux of
the 6dF spectrum is calibrated using the CTIO
spectrum. The continuum is modeled with a Fig. 3.— CTIO spectrum of 1H0707−495. The
power law and the Fe ii emission is modeled us- continuummodeledwithapower-lawisplottedin
ing the V´eron-Cetty et al. (2004) templates. The green. Thebest-fitFe iiemissionisrepresentedby
Balmer lines are deblended into a broad and a the lower black curve. The residual emission line
narrow component, and the former is modeled spectra (black line) after subtracting the power-
with a Lorentzian profile. The narrow compo- law continuum and the Fe ii model is shown in
nent is modeled by a Gaussian with FWHM fixed the insets for the Hβ+[Oiii] and Hα regions, re-
at the CTIO spectral resolution (∼330kms−1, spectively (red: broad lines; blue: narrow lines).
Leighly & Moore 2004). The flux ratio of the See text for details.
[Oiii]λλ4959,5007doubletisfixedatthetheoret-
ical value of 1:3 while each of them is fitted with
stellar-/intermediate-mass BH to the SMBH
two Gaussians, one for the line core with FWHM
regime, Zhou et al. (2015) suggested it to be a
fixed to the narrow Balmar line widths, the other
universal relationship for astrophysical BHs at
for the blueshifted wing. The widths of the broad
all mass scales. In Figure 4, we reproduce the
Hβ line are fitted to be 1002kms−1 (CTIO) and
f −M relation derived by RM06 based on
1054kms−1 (6dF), which agree with each other QPO BH
three BHXBs (solid line) and overplot 1H 0707-
withinmutualerrors. Thenarrownessofthebroad
495, as well as those BH systems with known
Hβ line, as well as the strong Fe ii emission are
QPOs (see Table 1 in Zhou et al. 2015). Clearly,
characteristic of NLS1 AGN. The BH mass is es-
the QPO in 1H 0707-495 conforms with the
timated using the broad Hβ line width and the
f − M relation within the BH mass un-
monochromatic luminosity at 5100˚A (λL = QPO BH
5100 certainty range.
4.0×1043ergs−1)(Vestergaard & Peterson2006),
giving MBH = 5.2× 106M⊙ (CTIO) and 5.7× 4. DISCUSSION
106M⊙ (6dF), respectively.
Given the large uncertainty ∼ 0.5dex of this 4.1. Reliability of the QPO and Compari-
method (Vestergaard & Peterson 2006), the esti- son with RE J1034+396
mated BH mass is consistent with previous re-
A significant QPO signal is detected at a
sults. Weconsiderourspectralanalysistobemore
> 99.99% confidence level in one of the X-ray
comprehensive and rigorous compared to those in
light curves of the NLS1 AGN 1H 0707-495
previous work, and thus M = 5.2(±0.5dex)×
BH taken with XMM-Newton. The QPO frequency
106M⊙ to be the best-estimate of the BH mass (∼ 2.6 × 10−4Hz) and the estimated mass of
in 1H0707−495. Adopting a bolometric luminos-
ity of 3.6×1044ergs−1 estimated from9×λL the central BH (MBH = 5.2 × 106M⊙) con-
5100 form with the previous f − M relation.
(Kaspi et al. 2000), the Eddington ratio is 0.5, QPO BH
We note that the same data set was also ana-
which is typical of NLS1.
lyzed by Gonz´alez-Mart´ın& Vaughan (2012) in
By extending the f − M relation from
QPO BH a study of the X-ray timing properties of a large
4
searches. Consideringallthenon-detectionswould
lower the overall significance of the detection in
the statistical sense, giving only a null probabil-
ity p ∼ 22×(2.5×10−3) ∼ 0.05. However, this
is the most conservative case and should only be
considered as a limit, since it is known that a
QPO does not always repeat in every observa-
tion. QPOsinAGNsmaywellbeatransientphe-
nomenon(e.g.,thedatasuggestadutycycle∼5%
inthecaseof1H0707-495). Furthermore,thefact
that the QPO in 1H 0707-495 follows closely the
f −M relation supports its genuineness.
QPO BH
Fig. 4.— Relation between QPO frequency and
The QPO frequency in 1H 0707-495 is very
BH mass. This is an updated version of Figure 1
close to that in RE J1034+396(∼2.7×10−4Hz),
from Zhou et al. (2015), where objects with sig-
which was argued to be a HFQPO (Zhou et al.
nificant QPO detections are plotted (dots), to-
2010, 2015). This is not surprising given their
gether with the newly detected QPO in 1H 0707-
495 (star). The solid line represents the extrapo- similar BH masses (MBH = 4+−32 × 106M⊙ for
RE J1034+396 and 5.2 × 106M⊙ for 1H 0707-
lation of the relation derived in RM06, based on
495). Moreover, the two AGNs are similar
the three BHXBs. The dotted line and dashed
in several ways, e.g., the NLS1 classification
linerepresenttherelationsderivedfromthemodel
(V´eron-Cetty & V´eron 2010), the high Edding-
of 3:2 resonance (Kluzniak & Abramowicz 2002)
ton ratios (L /L = 1.25 for RE J1034+396
withthespinparametera=0.998(dotted)anda=0 Bol Edd
and 0.5 for 1H 0707-495), the X-ray spectra with
(dashed),respectively. Upper rightinset: Azoom-
a strong soft excess below 1keV, as well as the
in comparison of 1H 0707-495 with the model
PSD shapes (Gonz´alez-Mart´ın& Vaughan 2012).
predictions, with various available BH mass esti-
This suggests that the HFQPOs tend to occur in
mates. aandbdenotethemassfromBian & Zhao
this type of AGNs.
(2003) and Done & Jin (2015) respectively, while
c denotes our best-estimated mass derived in this
4.2. Do HFQPOs Tend to Occur at the
paper. Lower left inset: the inferred BH spin for
Highest Accretion State?
a range of BH mass values for 1H 0707-495 from
the 3:2 resonance model, assuming that the QPO HFQPOs in BHXBs are observed only when
represents the 2×f peak. the systems are at the very high state with high
0
. accretion rates (Remillard & McClintock 2006;
Lai & Tsang 2009). Interestingly, the two AGNs
havingsignificantQPOdetections,REJ1034+396
sample of AGN, but no significant QPO was re-
and 1H 0707-495, are both NLS1s, which are
ported. Thismightbeascribedtothatthedataof
thought to be an AGN analogue of BHXB at
the whole observation duration was used in their
the very high state. Figure 5 shows the distribu-
work, which led to reduced significance of the
tion of the Eddington ratios for all the BH ac-
signal (< 90%, most likely due to a shift of the
cretion systems with reliable QPO detections (as
QPO phase). The same situation also occurred
shown in Figure 4) from BHXBs to AGNs. For
in the case of RE J1034+396, in which the QPO
the three BHXBs, the Eddington ratios are cal-
was significant in only part of the light curve,
culated from the average bolometric luminosities
and the significance became much lowerwhen the
at which the QPO occurred from Remillard et al.
whole light curve was used (Gierlin´ski et al. 2008;
(2002) and Belloni et al. (2006). For the tidal
Middleton et al.2011). For1H0707-495thereare
disruption event (TDE) Swift J164449.3+573451
15 XMM-Newton observations with good quality
in which a QPO was found, the Eddington ra-
light curves, totaling 1200ks exposure time. This
tio is believed to be around unity or even higher
may be equivalent to that the QPO is detected
since its observed luminosity is highly super-
in only one out of effectively ∼1200ks/55ks⋍22
Eddington(Reis et al.2012). TheEddingtonratio
5
corotational instability of non-axisymmetric g-
mode or p-mode trapped in the innermost re-
gion of the accretion disk (g-/p-mode models;
Li et al. 2003). All these models can explain the
3:2 frequency ratio of the twin-peak QPOs. The
f −M relationisalsopredictedinallexcept
QPO BH
the g-mode model (Silbergleit & Wagoner 2008).
OnlytheAEIandp-modemodelscanexplainwhy
theHFQPOsoccurexclusivelyattheveryhighac-
cretion state. A common issue for the RPM, RM
and AOM models is a lack of underlying physical
mechanisms at work, e.g., how orbiting hot spots
survive in a differentially rotating disk, how the
resonances are produced, etc (see Lai & Tsang
Fig. 5.— Distribution of the Eddington ratios of 2009; Belloni & Stella 2014 for brief reviews and
the BH accretion systems with HFQPO detected references therein). Discussion concerning these
(blue), including BHXBs, ULX (M82 X-1), TDE models is beyond the scope of this paper. Nev-
(Swift J164449.3+573451) and AGNs (1H 0707- ertheless, here we briefly address the explanation
495 and RE J1034+396). As a comparison, also andimplication ofour results in the frameworkof
plotted is the distribution ofthe Eddingtonratios the resonance model.
of AGNs from the Gonz´alez-Mart´ın & Vaughan
In the resonance model, the HFQPOs are sug-
(2012)sampleinwhichQPOwassearchedforbut
gested to be associated with the frequencies of
notdetected. TheKolmogorov-Smirnovtestgives
three fundamental oscillation modes of disk fluid
a probability p = 0.2% that the two distributions
elements around a BH: the Kepler orbital mo-
are drawn from the same population.
tion, the radial and vertical (to the orbital plane)
.
epicyclicoscillationmodes(Abramowicz & Klu´zniak
2001). These frequencies naturally scale inversely
ofM82X-1wassuggestedtobe0.8(Pasham et al. andlinearlywiththeBHmass,withadependence
2014). As a comparison, we overplot the Ed- on the BH spin (Nowak & Lehr 1998). Assuming
dington ratio distribution for the AGN sample that the observed 3:2 twin-peak QPO frequen-
of Gonz´alez-Mart´ın& Vaughan (2012), in which ciescorrespondto theverticalandradialepicyclic
QPOs were searched for but not found. Clearly frequencies as in Kluzniak & Abramowicz (2002),
the two distributions differ significantly; a two- the predicted fQPO − MBH relation is well con-
sided Kolmogorov-Smirnov test yields a small p- sistent with the fitted relation from RM06 (solid
value of0.2%. This resultsuggeststhatthe HFQ- line in Figure 4), which corresponds to a high
POs tend to occur at the highest accretion state BH spin close to the maximum value (Zhou et al.
of BH accretion systems. 2015). To illustrate the effect of BH spin, the
f −M relation assuming two extreme spin
QPO BH
4.3. Origins of HFQPO and Possible Link values a=0 (no spin; dashed) and a=0.998(maxi-
with BH Spin mum spin; dotted) are overplottedin Figure 4 for
the2×f frequency(thelowerfrequencyofthe3:2
0
The origin of the HFQPO is unclear, though
twin-peak QPO). A zoom-in comparisonwith the
a number of models have been proposed in the
resultsof1H0707-495is showninthe upper-right
literature, e.g., the relativistic precession model
inset of Figure 4, where the BH mass estimates
(RPM, Stella et al. 1999), resonance model (RM,
in this and previous work are over-plotted. As
Abramowicz & Klu´zniak 2001), acoustic oscil-
can be seen, the model can well reproduce the
lation modes in pressure-supported accretion
observed QPO frequency for a wide range of BH
tori (AOM, Rezzolla et al. 2003), instability at
spin, given the relatively large uncertainty range
disk-magnetosphere interface (IDI, Li & Narayan
of M . The inferred spin values from a range
BH
2004), accretion-ejectioninstability inmagnetized
of BH masses are shown in the lower-left inset
disks (AEI, Tagger & Pellat 1999), and global
6
in Figure 4. As one immediate inference, the al- Belloni, T. M., & Stella, L. 2014,Space Sci. Rev.,
lowed BH mass in 1H 0707-495 is in the range 183, 43
(2−8)×106M⊙ (corresponding to from a=0 to
a=0.998). Ourbest-estimateMBH =5.2×106M⊙ Belloni, T., Soleri, P., Casella, P., M´endez, M., &
indicates a moderately high BH spin ∼ 0.8. It Migliari, S. 2006, MNRAS, 369, 305
should be noted that Fabian et al. (2009) sug-
Done, C., & Jin, C. 2015, arXiv:1506.04547
gested a high BH spin for 1H 0707-495 based on
its relativistic Fe line andreflectionspectra,while Fabian,A.C.,Zoghbi,A.,Ross,R.R.,etal.2009,
Done & Jin (2015) argued for a low spin. In gen- Nature, 459, 540
eral, NLS1 AGN are suggested to have averagely
Gierlin´ski,M.,Middleton,M.,Ward,M.,&Done,
low or moderate BH spins as a population, as
C. 2008, Nature, 455, 369
found by Liu et al. (2015) from the relativistic
Fe line profile of the stacked spectra. Clearly, a
Gonz´alez-Mart´ın, O., & Vaughan, S. 2012, A&A,
preciseBH massmeasurement,e.g.,via the rever-
544, AA80
berationmappingmethodwithwhichanaccuracy
∼30% could be reached (Peterson et al. 2004), is Hu, C.-P., Chou, Y., Yang, T.-C., & Su, Y.-H.
neededforconstrainingtheBHspininsuchAGNs 2014,ApJ, 788, 31
with detected QPOs.
Kaspi, S., Smith, P. S., Netzer, H., et al. 2000,
ApJ, 533, 631
W.Y. thanks S. Vaughan, P. Uttley and C.
Jin for helpful discussion. S. Vaughan is also Kluzniak, W., & Abramowicz, M. A. 2002,
thanked for making the Bayesian test code avail- arXiv:astro-ph/0203314
able, and C. Jin and K. Leighly for providing the
Lai, D., & Tsang, D. 2009, MNRAS, 393, 979
CTIO spectral data. This work is supported by
theNationalNaturalScienceFoundationofChina
Leighly, K. M., & Moore, J. R. 2004, ApJ, 611,
(Grant No.11473035), and the Strategic Priority
107
Research Program“The Emergence of Cosmolog-
ical Structures” of the Chinese Academy of Sci- Li,L.-X.,Goodman,J.,&Narayan,R.2003,ApJ,
ences (Grant No. XDB09000000). This work 593, 980
is based on observations obtained with XMM-
Newton,anESAsciencemissionwithinstruments Li, L.-X., & Narayan,R. 2004, ApJ, 601, 414
andcontributionsdirectlyfundedbyESAMember
Lin, D., Irwin, J. A., Godet, O., Webb, N. A., &
States and NASA. This research has made use of
Barret, D. 2013, ApJ, 776, LL10
the NASA/IPAC Extragalactic Database (NED).
Liu, Z., Yuan, W., Lu, Y., & Zhou, X. 2015,MN-
REFERENCES RAS, 447, 517
Abramowicz, M. A., & Klu´zniak, W. 2001, A&A,
Markoff, S., Nowak, M. A., Gallo, E., et al. 2015,
374, L19
ApJ, 812, L25
Abramowicz, M. A., Klu´zniak, W., McClintock,
Markowitz, A., Edelson, R., Vaughan, S., et al.
J. E., & Remillard, R. A. 2004,ApJ, 609, L63
2003,ApJ, 593, 96
Alston,W.N.,Markeviˇciu¯te˙,J.,Kara,E.,Fabian,
McHardy,I.M., Koerding,E.,Knigge,C., Uttley,
A. C., & Middleton, M. 2014, MNRAS, 445,
P., & Fender, R. P. 2006, Nature, 444, 730
L16
McHardy,I.2010,LectureNotesinPhysics,Berlin
Alston, W. N., Parker,M. L., Markeviˇciu¯te˙,J., et
Springer Verlag, 794, 203
al. 2015, MNRAS, 449, 467
Middleton, M., Uttley, P., & Done, C. 2011, MN-
Bian, W., & Zhao, Y. 2003, MNRAS, 343, 164
RAS, 417, 250
7
Nowak, M. A., & Lehr, D. E. 1998, Theory of Vestergaard, M., & Peterson, B. M. 2006, ApJ,
Black Hole Accretion Disks, 233 641, 689
Pasham, D. R., Strohmayer,T. E., & Mushotzky, Yao,S.,Yuan,W.,Zhou,H.,etal.2015,MNRAS,
R. F. 2014, Nature, 513, 74 454, L16
Peterson, B. M., Ferrarese, L., Gilbert, K. M., et Zhou, X.-L., Zhang, S.-N., Wang, D.-X., & Zhu,
al. 2004, ApJ, 613, 682 L. 2010, ApJ, 710, 16
Reis, R. C., Miller, J. M., Reynolds, M. T., et al. Zhou, X.-L., Yuan, W., Pan, H.-W., & Liu, Z.
2012,Science, 337, 949 2015,ApJ, 798, LL5
Remillard,R.A.,Muno,M.P.,McClintock,J.E.,
& Orosz, J. A. 2002, ApJ, 580, 1030
Remillard, R. A., & McClintock, J. E. 2006,
ARA&A, 44, 49
Rezzolla,L.,Yoshida,S.,Maccarone,T.J.,&Zan-
otti, O. 2003, MNRAS, 344, L37
Scaringi,S.,Maccarone,T.J.,Koerding,E.,etal.
2015,arXiv:1510.02471
Silbergleit, A. S., & Wagoner, R. V. 2008, ApJ,
680, 1319
Stella,L.,Vietri,M.,&Morsink,S.M.1999,ApJ,
524, L63
Strohmayer, T. E. 2001,ApJ, 552, L49
Tagger, M., & Pellat, R. 1999, A&A, 349, 1003
Timmer, J., & Koenig, M. 1995, A&A, 300, 707
Uttley, P., McHardy, I. M., & Papadakis, I. E.
2002,MNRAS, 332, 231
Vaughan, S., Fabian, A. C., & Nandra, K. 2003,
MNRAS, 339, 1237
Vaughan, S., Edelson, R., Warwick, R. S., & Utt-
ley, P. 2003,MNRAS, 345, 1271
Vaughan, S. 2005, A&A, 431, 391
Vaughan, S. 2010, MNRAS, 402, 307
Vaughan, S., Uttley, P., Pounds, K. A., Nandra,
K., & Strohmayer, T. E. 2011, MNRAS, 413,
2489
V´eron-Cetty, M.-P., Joly, M., & V´eron, P. 2004,
A&A, 417, 515
V´eron-Cetty,M.-P., & V´eron,P. 2010,A&A, 518, This2-columnpreprintwaspreparedwiththeAASLATEX
A10 macrosv5.2.
8