1: %\documentclass[12pt,preprint]{aastex}%
2: %\documentclass[manuscript]{aastex}
3: %\documentclass{emulateapj_new}
4: \documentclass[preprint2]{aastex}
5: \lefthead{Dewangan et al.}
6:
7: \slugcomment{To apear in ApJ}
8:
9: \def\mes{M{\'e}sz{\'a}ros}
10: \newcommand\asca{{\it ASCA}}
11: \newcommand\sax{{\it BeppoSAX}}
12: \newcommand\chandra{{\it Chandra}}
13: \newcommand\rosat{{\it ROSAT}}
14: \newcommand\rxte{{\it RXTE}}
15: \newcommand\xmm{{\it XMM-Newton}}
16: \newcommand\ftools{{\it FTOOLS}}
17: \newcommand\s{{\rm~s}}
18: \newcommand\ks{{\rm~ks}}
19: \newcommand\mhz{{\rm~mHz}}
20: \newcommand\mpc{{\rm~Mpc}}
21: \newcommand\pc{{\rm~pc}}
22: \newcommand\hz{{\rm~Hz}}
23: \newcommand\kev{{\rm~keV}}
24: \newcommand\ev{{\rm~eV}}
25: \newcommand\kms{\ifmmode {\rm~km\ s}^{-1} \else ~km s$^{-1}$\fi}
26: \newcommand\Hunit{\ifmmode {\rm~km\ s}^{-1}\ {\rm Mpc}^{-1}
27: \else ~km s$^{-1}$ Mpc$^{-1}$\fi}
28: \newcommand\ctssec{\ifmmode {\rm~count\ s}^{-1} \else ~count s$^{-1}$\fi}
29: \newcommand\ergsec{\ifmmode {\rm~erg\ s}^{-1} \else
30: ~erg s$^{-1}$\fi}
31: \newcommand\funit{\ifmmode {\rm~erg\ s}^{-1}\;{\rm cm}^{-2} \else
32: ~ergs s$^{-1}$ cm$^{-2}$\fi}
33: \newcommand\phflux{\ifmmode {\rm~photon\ s}^{-1}\;{\rm cm}^{-2}
34: \else ~photon s$^{-1}$ cm$^{-2}$\fi}
35: \newcommand\efluxA{\ifmmode {\rm~erg\ s}^{-1}\;{\rm cm}^{-2}\;{\rm
36: \AA}^{-1} \else ~erg s$^{-1}$ cm$^{-2}$ \AA$^{-1}$\fi}
37: \newcommand\efluxHz{\ifmmode {\rm~erg\ s}^{-1}\;{\rm cm}^{-2}\;{\rm
38: Hz}^{-1} \else ~erg s$^{-1}$ cm$^{-2}$ Hz$^{-1}$\fi}
39: \newcommand\cc{\ifmmode {\rm~cm}^{-3} \else cm$^{-3}$\fi}
40: \newcommand\FWHM{\ifmmode {\rm~FWHM} \else ${\rm~FWHM}$\fi}
41: \newcommand\Msun{\ifmmode M_{\odot} \else $M_{\odot}$\fi}
42: \newcommand\Lsun{\ifmmode L_{\odot} \else $L_{\odot}$\fi}
43: \newcommand\ltsim{\raisebox{-.5ex}{$\;\stackrel{<}{\sim}\;$}}
44: \newcommand\gtsim{\raisebox{-.5ex}{$\;\stackrel{>}{\sim}\;$}}
45: \newcommand\hbeta{\ifmmode {\rm H}\beta \else H$\beta$\fi}
46: \newcommand\Kalpha{\ifmmode {\rm K}\alpha \else K$\alpha$\fi}
47: \newcommand\nh{\ifmmode N_{\rm H} \else N$_{\rm H}$\fi}
48: \usepackage{graphicx}
49: \usepackage{here}
50:
51:
52: %\received{2006 September 7}
53: \begin{document}
54:
55:
56: % \title{Time delay between soft and hard X-ray emission and the
57: % origin of the soft X-ray excess emission from Narrow-line Seyfert
58: % 1 galaxies Akn~564 and Mrk~1044}
59:
60: \title{An investigation of the origin of soft X-ray excess emission
61: from Narrow-line Seyfert 1 galaxies Akn~564 and Mrk~1044}
62:
63: \author{G. C. Dewangan\altaffilmark{1}, R. E.
64: Griffiths\altaffilmark{1}, Surajit Dasgupta\altaffilmark{2} \& A. R.
65: Rao\altaffilmark{2}} \altaffiltext{1}{Department of Physics,
66: Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213
67: USA; {\tt email: gulabd@cmu.edu; griffith@seren.phys.cmu.edu} }
68: \altaffiltext{2}{Department of Astronomy \& Astrophysics, Tata
69: Institute of Fundamental Research, Mumbai, 400005 India; {\tt email:
70: surajit@tifr.res.in; arrao@tifr.res.in}}
71:
72:
73:
74: \begin{abstract}
75: We investigate the origin of the soft X-ray excess emission from
76: narrow-line Seyfert 1 galaxies Akn~564 and Mrk~1044 using \xmm{}
77: observations. We find clear evidence for time delays between the
78: soft and hard X-ray emission from Akn~564 based on a $\sim 100\ks$
79: long observation. The variations in the $4-10\kev$ band lag behind
80: that in the $0.2-0.5\kev$ band by $1768\pm122\s$. The full band
81: power density spectrum (PDS) of Akn~564 has a break at $\sim
82: 1.2\times10^{-3}\hz$ with power-law indices of $\sim 1$ and $\sim 3$
83: below and above the break. The hard ($3-10\kev$) band PDS is
84: stronger and flatter than that in the soft ($0.2-0.5\kev$) band. Based
85: on a short observation of Mrk~1044, we find no correlation between
86: the $0.2-0.3\kev$ and $5-10\kev$ bands at zero lag. These
87: observations imply that the soft excess is not the reprocessed hard
88: X-ray emission. The high resolution spectrum of Akn~564 obtained
89: with the reflection grating spectrometer (RGS) shows evidence for a
90: highly ionized and another weakly ionized warm absorber medium. The
91: smeared wind and blurred ionized reflection models do not describe
92: the EPIC-pn data adequately. The spectrum is consistent with a
93: complex model consisting of optically thick Comptonization in a cool
94: plasma for the soft excess and a steep power-law, modified by two
95: warm absorber media as inferred from the RGS data and the foreground
96: Galactic absorption. The smeared wind and optically thick
97: Comptonization models both describe the spectrum of Mrk~1044
98: satisfactorily, but the ionized reflection model requires extreme
99: parameters. The data suggest two component corona -- a cool,
100: optically thick corona for the soft excess and a hot corona for the
101: power-law component. The existence of a break in the soft band PDS
102: suggests a compact cool corona that can either be an ionized surface
103: of the inner disk or an inner optically thick region coupled to a
104: truncated disk. The steep power-law component is likely arising from
105: an extended region.
106: %The observed
107: % soft X-ray excess is consistent with thermal Comptonization in the
108: % cool corona and the hard power-law extending to high energies may
109: % arise from thermal or nonthermal or hybrid thermal/nonthermal
110: % Comptonization in the hot plasma.
111:
112:
113: \end{abstract}
114:
115: \keywords{accretion, accretion disks -- galaxies: active -- X-rays:
116: galaxies}
117:
118: \section{Introduction}
119: A significant fraction of type 1 active galactic nuclei (AGN) show
120: `soft X-ray excess emission' above a power-law continuum, usually
121: identified as the steepening of the X-ray continuum below $\sim
122: 2\kev$. This soft excess emission was first observed by {\it HEAO-1}
123: (Pravdo et al. 1981), and {\it EXOSAT} (Arnaud et al. 1985; Singh et
124: al. 1985). Boller, Brandt, \& Fink (1996) showed that AGN with the
125: steepest soft X-ray spectra tend to lie at the lower end of the
126: H$\beta$ line width distribution. These AGN with FWHM$_{\rm H\beta}
127: \ltsim 2000\kms$ are classified as the narrow-line Seyfert~1 galaxies
128: (NLS1; Osterbrock \& Pogge 1985), and are distinguished from the bulk
129: of the Seyfert~1 galaxies. In addition to the strong soft X-ray
130: excess below $2\kev$ and narrower permitted lines, NLS1 also show a
131: number of extreme properties e.g., steep $2-10\kev$ power-law
132: continuum, extreme X-ray variability and strong Fe~II emission. These
133: properties are all related and one suggestion is that NLS1 are the
134: supermassive black hole analogue of stellar mass black hole X-ray
135: binaries (BHB) in their `high/soft' state.
136:
137: However, the origin of the soft X-ray excess emission has remained a
138: major problem in AGN research over the past two decades. The soft
139: excess emission is a smooth continuum component rather than a blend of
140: emission/absorption features as revealed by \chandra{} and \xmm{}
141: grating observations (e.g., Turner et al. 2001a; Collinge et al 2001).
142: When described as a thermal component, the soft excess emission has a
143: remarkably constant temperature ($\sim 100-200\ev$) across AGN with a
144: wide range of black hole masses (Czerny et al. 2003; Gierlinski \&
145: Done 2004; Crummy et al. 2006). There are a number of ideas that have
146: been proposed to explain the origin of the soft excess component.
147: Pounds, Done \& Osborne (1995) pointed out the similarity of the X-ray
148: spectrum of NLS1 and Cyg X-1 in its high state (HS) and suggested that
149: the soft excess could be the optically thick emission from an
150: accretion disk. However, the observed soft excess emission is too hot
151: for a disk around a supermassive black hole. One possibility is that
152: there is a cool, optically thick Comptonizing region in addition to a
153: hot ($\sim 100\kev$), optically thin ($\tau\sim1$) region producing
154: the high energy power-law emission. The cool region can Comptonize the
155: disk photons and smoothly connect the disk emission up to soft X-ray
156: energies (Magdziarz et al. 1998). Gierlinski \& Done (2004) tested
157: this model using high quality \xmm{} spectra of 26 radio-quiet PG
158: quasars. Although the spectra are well described by the Comptonization
159: model, the temperature of the cool Comptonizing region is remarkably
160: constant despite a large range in the black hole masses. They
161: proposed that the observed soft excess is an artifact of heavily
162: smeared, strong partially ionized absorption (see also Gierlinski \&
163: Done 2006; Schurch \& Done 2006). Another possibility is
164: that the emission is dominated by ionized Compton reflection. This can
165: happen if the disk is clumpy at high accretion rates, thus hiding the
166: hard X-ray source among many clumps. The ionized reflection can
167: explain the observed soft excess emission (Fabian et al. 2002). The
168: last two models, based on relativistically smeared absorption and
169: ionized reflection, relate the soft excess emission to atomic
170: processes and therefore provide a natural explanation for its constant
171: temperature for AGN with a wide range in their black hole masses.
172:
173: The three models, cool Comptonization, ionized reflection and heavily
174: smeared absorption, all result in statistically equally good fits.
175: The fitting criteria are not sufficient to discriminate among them
176: (Sobolewska \& Done 2005). Variability properties of the soft and
177: hard X-ray emission and the relationship between them may provide
178: strong constraints on the above models for the soft excess emission. A
179: $35{\rm~day}$ long \asca{} monitoring observation of the NLS1 galaxy
180: Akn~564 revealed a distinction in the variability of the soft excess
181: and power-law components on a time scale of weeks, with the soft
182: excess emission varying by a factor of $\sim6$ compared to a factor of
183: $\sim4$ in the power-law component (Turner et al. 2001b). Similarly, a
184: long $\sim 10{\rm~days}$ \asca{} observation of the NLS1 galaxy
185: IRAS~13224-3809 suggested that the soft excess component dominates the
186: observed variability on a time scale of $\sim~a~week$, but on shorter
187: time scales ($\sim 20000\s$), the power-law component appears to
188: dominate the variability (Dewangan et al. 2002). A $12{\rm~day}$ long
189: \asca{} observation of another NLS1 galaxy Ton~S180 suggested that the
190: soft excess and the power-law fluxes are not well correlated on time
191: scales of $\sim 1000\s$ (Romano et al. 2002). These long \asca{}
192: observations suggest two separate continuum component unlike the
193: heavily smeared, ionized absorption model based on a single continuum
194: component. Gallo et al. (2004) found evidence for alternate leads
195: and lags between the $0.3-0.8\kev$ and $3-10\kev$ band emission from
196: the NLS1 galaxy IRAS~13224-3809. Dasgupta \& Rao (2006) reported
197: energy dependent delay between the soft and hard X-ray emission in
198: Mrk~110. These observations suggest different emitting regions for the
199: soft excess and the power-law component from NLS1 galaxies. In this
200: paper, we use \xmm{} observations of two NLS1 galaxies Akn~564 and
201: Mrk~1044 to study the variability properties of the soft and hard
202: X-ray emission and investigate the origin of the soft excess emission.
203: In Section 2, we outline the \xmm{} observations of Akn~564 and
204: Mrk~1044. We describe the temporal analysis in Section 3 and spectral
205: analysis in Section 4 and discuss the results in Section 5.
206:
207:
208:
209: \setlength{\voffset}{0mm}
210: \begin{figure*}
211: \centering \includegraphics[width=8cm]{f1a.ps}
212: \includegraphics[width=8cm]{f1b.ps}
213: \caption{\xmm{} light curves and hardness ratio of Akn~564 and
214: Mrk~1044. {\it Left:} Combined EPIC-pn and MOS light curves of
215: Akn~564 in the bands $0.2-0.3\kev$ and $5-10\kev$ bands and the
216: hardness ratio. The bin sizes are $512\s$. Vertical dotted lines
217: mark the dips and peaks in the $0.2-0.3\kev$ and $5-10\kev$ light
218: curves that show possible evidence for a delay. {\it Right:}
219: Combined EPIC-pn and MOS light curves of Mrk~1044 in the
220: $0.2-0.5\kev$ and $2-10\kev$ bands and the hardness ratio. The bin
221: sizes are $128\s$. The soft and hard band light curves do not
222: appear to be strongly correlated. }
223: \label{f1}
224: \end{figure*}
225:
226: \section{Observation and Data Reduction}
227: Akn~564 was observed by \xmm{} on 2005 January 04 (ObsID: 0206400101)
228: for an exposure time of $100\ks$. Mrk~1044 was observed by \xmm{} on
229: 2002 July 23 (ObsID: 0112600301) for $8\ks$. The EPIC-pn and MOS
230: cameras \citep{Struderetal01,Turneretal01c} were operated in the small
231: window mode using the medium filter during both the observations. The
232: data were reduced using version 6.5.0 of the SAS software.
233: Examination of the background rate above $10\kev$ showed that the
234: observation of Akn~564 was partly affected by the flaring particle
235: background before an elapsed time of $7.5\ks$ and this early period
236: was therefore excluded to obtain a continuous exposure with steady
237: background. For temporal analysis, we used all the pn and MOS events
238: with patterns $\le 12$ and a continuous exposure of $91.3\ks$ during
239: which both the pn and MOS cameras operated simultaneously. There were
240: three short duration ($<1000\s$), small amplitude background flares
241: (maximum count rate $\sim 2{\rm~counts~s^{-1}}$ above $10\kev$) during
242: the observation of Mrk~1044. The varying background contribution to
243: the source was corrected by subtracting the background rate. For
244: spectral analysis, we chose a count rate cut-off criterion to exclude
245: the high particle background. In case of Akn~564, the MOS data are
246: affected with photon pile-up. In order to achieve the best possible
247: spectral resolution and to have the best spectral calibration, we
248: chose the EPIC-pn data and
249: used the good X-ray events (FLAG=0) with pattern zero (single pixel
250: events) for Akn~564. For Mrk~1044, we used the good EPIC-pn events with
251: pattern $0-4$ (single and double pixel events).
252: We used circular regions with radii of $40\arcsec$,
253: centered at the peak position of Akn~564 and Mrk~1044, to extract the
254: source spectra. We also extracted background spectra from appropriate
255: nearby circular regions, free of sources.
256:
257: We also processed the RGS data obtained from the long observation of
258: Akn~564. We extracted the first order source and the background
259: spectra by making the spatial, order selections on the processed event
260: files, and calibrated by applying the most recent calibration data. We
261: also used a temporal filter to exclude the high particle background
262: based on a count rate cut-off criterion. We have used the ISIS
263: (version 1.4.7) spectral fitting environment(Houck \& Denicola 2000;
264: Houck 2002) for power and energy spectral analysis. Unless otherwise
265: stated, all errors are quoted at $90\%$ confidence level.
266:
267:
268:
269:
270: \begin{figure}[H]
271: \centering \includegraphics[width=7cm]{f2.ps}
272: \caption{Cross correlation function between soft and hard X-ray
273: light curves of Akn~564: (a) $0.2-2\kev - 2-10\kev$, (b)
274: $0.2-1\kev$ Vs $3-10\kev$, (c) $0.2-0.5$ Vs $4-10\kev$ and (d)
275: $0.2-0.3\kev$ Vs $5-10\kev$. The time delays refer to the soft
276: band. Positive peaks imply that the variations in the hard band
277: lag that in the soft band. }
278: \label{f2}
279: \end{figure}
280:
281:
282: \section{Temporal analysis}
283: \subsection{Akn~564}
284: For temporal analysis, we combined the pn and MOS data to increase the
285: signal-to-noise ratio. In order to study cross-correlation between
286: different energy bands, we extracted light curves in several energy
287: bands $0.2-0.3$, $0.2-0.5$, $0.2-1$, $0.2-2$, $2-10$, $3-10$, $4-10$
288: and $5-10\kev$ with time bins of $64\s$. Figure~\ref{f1} shows the
289: light curves of Akn~564 in the $0.2-0.3$ and $5-10\kev$ bands rebinned
290: to $512\s$. X-ray emission from Akn~564 is highly variable. The
291: $0.2-0.3\kev$ band light curve shows a trough-to-peak variations by a
292: factor of $\sim3.3$, while the hard band ($5-10\kev$) shows a
293: trough-to-peak variation by a factor of $\sim 3.6$. Although the
294: $0.2-0.3\kev$ and $5-10\kev$ band light curves appear to be
295: correlated, the most rapid variability is seen in the hard band. A
296: visual comparison of the light curves suggests that variations in the
297: hard band appear to lag those in the soft band. To investigate
298: further, we have performed cross-correlation and power spectral
299: analysis.
300:
301:
302: \subsubsection{Cross Correlation Analysis}
303: We have computed cross-correlation function (CCF) using the XRONOS
304: program {\tt crosscor}. We used the slow direct Fourier algorithm to
305: compute the CCF between the soft and hard band light curves with
306: $64\s$ bins. Figure~\ref{f2} shows the CCFs as a function of time
307: delay between the energy bands (a) $0.2-2\kev$ and $2-10\kev$, (b)
308: $0.2-1\kev$ and $3-10\kev$, (c) $0.2-0.5\kev$ and $4-10\kev$ and (d)
309: $0.2-0.3\kev$ and $5-10\kev$. The $0.2-2\kev$ and $2-10\kev$ band
310: light curves are strongly correlated without any time delay as
311: indicated by the sharp peak at zero lag. However, the strength of the
312: sharp peak decreases with the increase in the separation between the
313: soft and hard bands. In the CCF of $0.2-1\kev$ and $3-10\kev$ band
314: light curves, the sharp peak weakens and another broad peak at
315: positive time lags becomes relatively stronger. The broad peak
316: dominates the CCF for the $0.2-0.5\kev$ and $4-10\kev$ band light
317: curves and only a weaker sharp peak is seen. The sharp peak is not
318: clearly seen in the CCF between $0.2-0.3\kev$ and $5-10\kev$ bands,
319: only the broad peak is seen. Thus the hard $4-10\kev$ or $5-10\kev$
320: band clearly lags the soft $0.2-0.5\kev$ or $0.2-0.3$ band.
321:
322:
323: \begin{figure*}
324: \centering
325: \includegraphics[width=6cm]{f3a.ps}
326: \includegraphics[width=6cm]{f3b.ps}
327: \caption{ Distributions of time lags obtained from the simulated
328: soft and hard band lightcurves. {\it Left:} Time-lag distribution
329: between $0.2-0.5$ and $4-10\kev$ bands. {\it Right:} Time-lag
330: distribution between $0.2-0.3$ and $5-10\kev$ bands. }
331: \label{f3}
332: \end{figure*}
333:
334:
335: It is hard to assess the statistical significance of the time lag
336: based on model fitting to the CCF using the traditional $\chi^2$
337: minimization technique. The reasons are ($i$) the shape of the CCF is
338: generally complex, ($ii$) the CCF data points are not independent and
339: ($iii$) the errors are not normally distributed. Therefore, we have
340: adopted a model independent way, following Peterson et al. (1998), to
341: measure the time lag. The principal source of uncertainty in the time
342: lag between two long, continuous light curves is the uncertainty in
343: the observed count rates. To account for this uncertainty, we have
344: performed Monte Carlo simulations using IDL (version 6.3). We
345: simulated 1000 pairs of soft and hard band lightcurves by modifying
346: the observed count rates by random Gaussian deviates assuming that the
347: errors on the observed count rates are normally distributed. We
348: calculated the time lag between each pair of the simulated soft and
349: hard band light curves as the centroid of the CCF based on all points
350: with correlation coefficients in excess of 0.4. Figure~\ref{f3} shows
351: the distribution of time lags between the $0.2-0.5\kev$ and $4-10\kev$
352: bands. We calculated the time lag between the observed soft and hard
353: bands as the centroid of the CCF and the errors on the time lag by
354: directly integrating the time-lag distribution over a range to obtain
355: $68\%$, $90\%$ or $99\%$ confidence level. Thus we measured the time
356: lag between the $0.2-0.5$ and $4-10\kev$ bands to be $1768\pm92\s$
357: ($68\%$), $\pm122\s$ ($90\%$) or $\pm191\s$ ($99\%$). A similar
358: procedure resulted in the time lag between the $0.2-0.3$ and
359: $5-10\kev$ to be $1767\pm196\s$ ($68\%$), $\pm320\s$ ($90\%$) or
360: $\pm462\s$ ($99\%$).
361:
362: We have also searched for variations in the CCF or time delay. For
363: this purpose, we divided both $0.2-0.5\kev$ and $4-10\kev$ light
364: curves into nine equal segments with lengths $10.1\ks$ and computed
365: the CCFs for each segment. The nine CCFs are plotted in
366: Figure~\ref{f4}. The CCFs are not similar. There is no or weak
367: correlation between the soft and hard band light curves in the
368: segments II-VI and VIII. In segment IV, the soft and hard band appear
369: to be weakly anti-correlated with a soft band lead of $\sim 1200\s$.
370: The two bands are strongly correlated without any lag for the segments
371: I \& IX, while for the segments VII and VIII, the hard band lags
372: behind the soft band. For these segments, the time lag varies from
373: $\sim 900-2900\s$. In segment VII, the soft and hard band appear to
374: anti-correlate at a lag of $\sim -3000\s$. Brinkmann, Papadakis \&
375: Raeth (2007) have employed the sliding window technique to study the
376: cross-correlation properties of Akn~564 between the $0.3-1\kev$ and
377: $3-10\kev$ bands using the same data. There results are similar to the
378: presented here if we note that the first $7.5\ks$ of data were
379: excluded from our analysis. The leads and lags in the X-ray emission
380: from Akn~564 are also similar to the alternate leads and lags in
381: $4500\s$ intervals found between the $0.3-0.8\kev$ and $3-10\kev$ band
382: X-ray emission from IRAS~13224-3809 by Gallo et al. (2004).
383:
384:
385:
386:
387: \begin{figure}[H]
388: \centering
389: \includegraphics[width=7cm]{f4.ps}
390: \caption{Cross correlation functions of $0.2-0.5\kev$ and
391: $4-10\kev$ band lightcurves for Akn~564. The CCFs have been
392: derived by splitting both $0.2-0.5\kev$ and $4-10\kev$
393: lightcurves into nine segments. Each segment has a length of
394: $10.1\ks$.}
395: \label{f4}
396: \end{figure}
397:
398:
399:
400: \begin{figure*}
401: \centering \includegraphics[width=6cm,angle=-90]{f5a.ps} \centering
402: \includegraphics[width=6cm,angle=-90]{f5b.ps}
403: \caption{{\it Left:} Power density spectra of Akn~564 in the full
404: band $0.2-10\kev$. The contribution due to the Poisson noise has
405: not been subtracted. {\it Right:} Power density spectra in the
406: soft ($0.2-0.5\kev$) and hard ($3-10\kev$) bands, derived from the
407: combined EPIC-pn and MOS data. The constant power expected from
408: the Poisson errors have been subtracted. The best-fitting broken
409: power-law models are also shown. The soft band power density
410: spectrum is clearly steeper and weaker than the hard band power
411: density spectrum. }
412: \label{f5}
413: \end{figure*}
414:
415:
416:
417: \subsubsection{The power density spectra}
418: To make a further study of the variability properties of Akn~564, we
419: have calculated power density spectra (PDS) of Akn~564 from the
420: combined pn+MOS lightcurves in the full ($0.2-10\kev$), soft
421: ($0.2-0.5\kev$) and hard ($3-10\kev$) bands, sampled at $16\s$. We
422: used the slow algorithm implemented in the XRONOS program `powspec'.
423: Following Papadakis \& Lawrence (1993), we rebinned the PDS in
424: logarithmic space with a binsize of 20 and performed the fitting
425: within the ISIS (version 1.4.5) spectral fitting environment. We
426: converted the PDS to equivalent energy spectra (power $\rightarrow$
427: counts/bin; $\hz \rightarrow \kev$) using the {\tt sitar} timing
428: package. We fitted the PDS with a simple power-law model. We also
429: used a constant to account for the power arising from the Poisson
430: noise. The simple power-law model resulted in an unacceptable fit
431: (minimum $\chi^2 = 225.2$ for $139$ degrees of freedom (dof)). We then
432: replaced the power-law with a broken power-law model. This model
433: resulted in the minimum $\chi^2 = 169.2$ for $137$ dof. The broken
434: power-law model provided an improved fit at a statistical significance
435: level of $>99.99\%$ compared to the power-law model according to the
436: maximum likelihood ratio test. Thus we conclude that the PDS of
437: Akn~564 has a break at $(1.2\pm0.3) \times 10^{-3}\hz$ with power-law
438: indices $1.3_{-0.2}^{+0.1}$ below the break and $3.2_{-0.4}^{+0.5}$
439: above the break. The break frequency reported here is similar to the
440: high frequency break at $\sim 2\times 10^{-3}\hz$ discovered by
441: Papadakis et al. (2002) based on ASCA data (see also Alevaro et al.
442: 2006) The constant power due to Poisson noise is
443: $0.024\pm0.0012{\rm~(rms/mean)^2 Hz^{-1}}$.
444: % Unless otherwise specified, the errors here and below are quoted at
445: % $90\%$ confidence level.
446: The PDS and the best-fit broken power-law plus constant model are
447: plotted in Figure~\ref{f5}. The power arising from the Poisson errors
448: have not been subtracted.
449:
450: We also fitted the power-law and broken power-law models to the soft
451: ($0.2-0.5\kev$) and hard ($3-10\kev$) PDS of Akn~564. The power-law
452: model resulted in an unacceptable fit to the PDS in the soft band
453: ($\chi^2 = 193.2$ for $139$ dof) with power-law index of $1.7\pm0.1$.
454: The broken power-law model provided a better fit ($\chi^2/dof
455: =160.9/137$) with power-law indices $1.4_{-0.2}^{+0.1}$ and
456: $4.0_{-1.3}^{+1.8}$ before and after the break frequency
457: $1.1_{-0.6}^{+0.2}\times 10^{-3}\hz$. This is an improvement over the
458: power-law fit at a significance level of $>99.99\%$ based on the
459: maximum likelihood ratio test. The power-law model provided an
460: acceptable fit to the hard band PDS ($\chi^2 = 138.4$ for $139$ dof)
461: with an index of $1.2\pm0.1$. The broken power-law model resulted in
462: an improved fit ($\chi^2/dof = 127.2/137$) with indices $<1.0$ ($90\%$
463: upper limit) and $1.8_{-0.4}^{+0.5}$ before and after the break
464: frequency $< 1.2\times 10^{-3}\hz$ ($90\%$ upper limit). This is an
465: improvement over the power-law model at a level of $99.6\%$
466: significance level. However, the parameters are not well determined.
467: We have plotted the soft and hard band PDS and the best-fit broken
468: power-law models in Fig.~\ref{f5} ({\it right panel}). Clearly the
469: soft band PDS is steeper than the hard band PDS. This implies that the
470: X-ray emission in the hard band varies more than that in the soft band
471: on shorter time scales.
472:
473:
474:
475: \begin{figure}[H]
476: \centering \includegraphics[width=7cm]{f6.ps}
477: \caption{Cross correlation function between soft and hard X-ray
478: light curves of Mrk~1044: (a) $0.2-2\kev - 2-10\kev$, (b)
479: $0.2-1\kev$ Vs $3-10\kev$, (c) $0.2-0.5$ Vs $4-10\kev$ and (d)
480: $0.2-0.3\kev$ Vs $5-10\kev$. The time delays refer to the soft
481: band. Positive peaks imply that the variations in the hard band
482: lag that in the soft band. }
483: \label{f6}
484: \end{figure}
485:
486:
487: \subsection{Mrk~1044}
488: As stated above, the observation of Mrk~1044 was affected by
489: short-duration, weak flares. Therefore, the source light curves were
490: corrected for background contribution using appropriate background
491: light curves extracted from blank sky regions. The combined pn+MOS and
492: background corrected light curves of Mrk~1044 are shown in
493: Fig.~\ref{f1} (right). X-ray emission from Mrk~1044 is variable with
494: trough-to-peak variations of $\sim 2$ ($0.2-0.5\kev$) and $\sim 3$
495: ($4-10\kev$). However, the $0.2-0.5\kev$ and $4-10\kev$ light curves
496: do not appear to be strongly correlated.
497:
498:
499: The cross correlations between the soft and hard bands are plotted in
500: Figure~\ref{f6}. Similar to Akn~564, the $0.2-2\kev$ and $2-10\kev$
501: bands are strongly correlated without any delay and the strength of
502: this correlation decreases with the interval between the soft and hard
503: bands. There is no correlation between the $0.2-0.3\kev$ and
504: $5-10\kev$ bands at zero lag. There is a weak anti-correlation at a
505: positive delay of $\sim1000\s$ and a weak correlation at positive
506: delays of $3000-5000\s$. As the CCF for Mrk~1044 is based on a short
507: $7.9\ks$ observation, the CCF is likely only one representation of
508: many possible cross correlations similar to that seen for Akn~564.
509:
510:
511: \section{Spectral Analysis}
512: % In order to investigate the soft and hard X-ray spectral components
513: % and the delay between them,
514: We have performed spectral modeling of Akn~564 and Mrk~1044. We
515: present the results based on the high signal-to-noise pn data for both
516: Akn~564 and Mrk~1044 and the high resolution RGS data for Akn~564. We
517: note that Papadakis et al. (2007) have performed a detailed spectral
518: analysis of Akn~564 using the same observation. The purpose here is to
519: investigate the alternative models for the origin of the soft X-ray
520: excess emission. All the spectra were analyzed with the Interactive
521: Spectral Interpretation System ({\tt ISIS, version 1.4.5}). The pn
522: data were grouped to a minimum of $100$ and $20$ counts per spectral
523: channel for Akn~564 and Mrk~1044, respectively. We consider a model
524: fit as statistically unacceptable if the the null hypothesis can be
525: rejected at more than $95\%$ significance. The errors on the best-fit
526: spectral parameters are quoted at a $90\%$ confidence level. Below we
527: use narrow Gaussian line components with their widths
528: fixed at $\sigma = 10\ev$ that are unresolved by the EPIC-pn data.
529:
530:
531:
532: \subsection{The soft X-ray excess emission}
533: NLS1 galaxies are known to show strong soft X-ray excess emission
534: below $\sim 2\kev$. To show this soft excess, first we fitted a simple
535: absorbed power law (PL) model to the EPIC-pn spectra above $3\kev$
536: ($3-10\kev$ for Mrk~1044 and $3-11\kev$ for Akn~564). The simple PL
537: model resulted in minimum $\chi^2 = 527.4$ for $519$ dof and $119.0$
538: for $127$ dof for Akn~564 and Mrk~1044, respectively, thus providing
539: good fits to the $3-10\kev$ pn data. We have plotted the ratio of the
540: pn data and the best-fit PL model in Figure~\ref{f7} after
541: extrapolating the model to low energies. The plots clearly show strong
542: soft excess emission below $\sim 2.5\kev$ and possible iron K$\alpha$
543: lines near $\sim 6.4\kev$ in the spectra of both the NLS1. The soft
544: X-ray excess emission from NLS1 are usually well described by a simple
545: blackbody or a multicolor disk blackbody (MCD). Addition of the MCD
546: component to the absorbed PL improved the fit. The fit was reasonably
547: good for Mrk~1044 ($\chi^2/dof = 628.0/541$). Addition of a narrow
548: Gaussian line at $\sim 6.4\kev$ further improved the fit ($\chi^2/dof
549: = 611.7/539$). The best-fit parameters of the MCD+PL model for
550: Mrk~1044 are listed in Table~\ref{t1} and the EPIC-pn data, the
551: best-fit model and residuals are plotted in Figure~\ref{f9}.
552:
553:
554: \begin{figure}[H]
555: \centering
556: \includegraphics[width=2.5cm,angle=-90]{f7a.ps}
557: \includegraphics[width=2.5cm,angle=-90]{f7b.ps}
558: \caption{Ratios of observed EPIC-pn data for Akn~564 and Mrk~1044
559: and the corresponding best-fit power-law model fitted above
560: $3\kev$ and
561: extrapolated to lower energies.}
562: \label{f7}
563: \end{figure}
564:
565: The MCD+PL model was a poor fit for Akn~564 ($\chi^2/dof =
566: 1494.1/913$). Examination of the residuals showed broad
567: emission/absorption features below $\sim 1\kev$ in the spectrum of
568: Akn~564. To verify and investigate these features, we have analyzed
569: the high resolution soft X-ray spectra of Akn~564 obtained with the
570: two RGS simultaneously with the EPIC-pn data. We used the Cash
571: statistic to fit the low count RGS data. We fitted a simple PL model,
572: modified with the neutral absorption, jointly to RGS1 and RGS2 data in
573: the $0.35-1.9\kev$ band. This resulted in a steep spectrum ($\Gamma
574: \sim 3.0$) and $N_H = 7.6\times10^{20}{\rm~cm^{-2}}$. Addition of an
575: MCD component improved the fit ($C/dof=2575.9/1572$; $\Delta C =
576: -225.4$ for two parameters). The best-fit parameters are $N_H \sim
577: 5.2\times 10^{20}{\rm~cm^{-2}}$, $\Gamma \sim 2.44$ and the MCD
578: $kT_{in}\sim 165\ev$. Examination of the ratio of the data and model
579: revealed a strong absorption feature at $\sim 0.5\kev$ and additional
580: weak and narrow absorption features in the $0.5-1\kev$ band,
581: suggesting the presence of a warm absorber medium in Akn~564.
582:
583: \begin{figure*}
584: \centering
585: \includegraphics[height=15cm,angle=-90]{f8a.ps}
586: \includegraphics[height=15cm,angle=-90]{f8b.ps}
587: \caption{(ab) RGS1 data, the best-fitting joint model and the ratios
588: of the observed data and the model for Akn~564. (b) The same as
589: (a) but for RGS2. The best-fit model was derived by joint spectral
590: fitting to both RGS1 and RGS2 data.}
591: \label{f8}
592: \end{figure*}
593:
594:
595:
596:
597: %\thispagestyle{empty} \setlength{\voffset}{-23mm}
598: \begin{table*}
599: \centering
600: {\small \caption{Spectral model
601: parameters for Akn~564 and Mrk~1044. \label{t1}}
602: \begin{tabular}{lcc} \tableline\tableline Parameter &
603: Akn~564\tablenotemark{1} & Mrk~1044 \\ \tableline & MCD$+$PL &
604: \\ \tableline
605: $N_H$ ($10^{20}{\rm~cm^{-2}}$) & $5.24$(f) & $3.54$(f) \\
606: $kT_{in}$ (eV) & $179\pm2$ & $138\pm3$ \\
607: $n_{MCD}$\tablenotemark{2} & $3002_{-100}^{+104}$ & $2992_{-277}^{+316}$ \\
608: $\Gamma$ & $2.48\pm0.01$ & $2.14\pm0.04$ \\
609: $n_{PL}$\tablenotemark{3} & $(1.35\pm0.02)\times10^{-2}$ & $(3.2\pm0.1)\times10^{-3}$ \\
610: $E_{FeK\alpha}$ (keV) & $6.72\pm0.03$ & $6.42\pm0.04$ \\
611: $f_{FeK\alpha}$\tablenotemark{4} & $(5.1\pm1.8)\times10^{-6}$ & $(1.14\pm0.46)\times 10^{-5}$ \\
612: $f_{0.3-10keV}$\tablenotemark{5} & $6.6\times10^{-11}$ & $2.1\times10^{-11}$ \\
613: $\chi^2/dof$ & $919.3/854$ & $611.7/539$ \\ \tableline & Blurred
614: Reflection \\ \tableline
615: $N_H$ ($10^{20}{\rm~cm^{-2}}$) & $5.24$(f) & $3.54$(f) \\
616: $\Gamma$ & $2.468_{-0.001}^{+0.002}$ & $2.20_{-0.004}^{+0.002}$ \\
617: $n_{PL}$\tablenotemark{3} & $9.29\pm0.03\times10^{-3}$ & $1.43_{-0.03}^{+0.05}\times10^{-3}$ \\
618: $\beta$ & $\ge10.0$ & $\ge10.0$ \\
619: $r_{in}$($r_g$) & $<1.58$ & $<1.25$ \\
620: $i$ & $66\pm1^\circ$ & ${65.7_{-0.2}^{+0.3}}^\circ$ \\
621: $n_{refl}$\tablenotemark{6} & $(4.27\pm0.02)\times10^{-7}$ & $1.7_{-0.01}^{+0.02}\times10^{-6}$ \\
622: $A_{Fe}$ & $0.69\pm0.01$ & $0.35_{-0.03}^{+0.04}$ \\
623: $\xi {\rm~(erg~cm~s^{-1})}$ & $3452_{-114}^{+129}$ & $334.5_{-1.2}^{+2.1}$ \\
624: $E_{FeK\alpha}$ (keV) & $6.72_{0.03}^{+0.05}$ & $6.42\pm0.04$ \\
625: $f_{FeK\alpha}$\tablenotemark{4} & $2.5_{-1.5}^{+2.1}\times10^{-6}$ & $1.0_{-0.5}^{+0.4}\times10^{-5}$ \\
626: $f_{0.3-10keV}$\tablenotemark{5} & $6.4\times10^{-11}$ & $2.0\times10^{-11}$ \\
627: $\chi^2/dof$ & $954.2/850$ & $671.6/535$ \\ \tableline & Smeared
628: absorption \\ \tableline
629: $N_H$ ($10^{20}{\rm~cm^{-2}}$) & $5.24$(f) & $3.54$ (f) \\
630: $\Gamma$ & $2.783_{-0.008}^{+0.006}$ & $2.49\pm0.02$ \\
631: $n_{PL}$\tablenotemark{3} & $(3.67\pm0.05)\times10^{-2}$ & $(7.6\pm0.2)\times10^{-3}$ \\
632: wind $N_H$ ($10^{22}{\rm~cm^{-2}}$) & $19.6_{-0.9}^{+1.8}$ & $15.7_{-1.6}^{+1.8}$ \\
633: $log (\xi/{\rm erg~cm~s^{-1}})$ & $2.9_{-0.03}^{+0.05}$ & $2.89\pm0.04$ \\
634: $\sigma$\tablenotemark{7} & $0.81\pm0.03$ & $0.29\pm0.03$ \\
635: $E_{FeK\alpha}$ (keV) &$6.72\pm0.03$ & $6.42\pm0.04$ \\
636: $f_{FeK\alpha}$\tablenotemark{4} & $(5.1\pm2.6) \times 10^{-6}$ & $(1.2\pm0.5)\times10^{-5}$ \\
637: $f_{0.3-10keV}$\tablenotemark{5} & $7.1\times10^{-11}$ & $2.1\times10^{-11}$ \\
638: $\chi^2/dof$ & $1147.1/853$ & $592.4/538$ \\ \tableline &
639: Thermal Comptonization \\ \tableline
640: $N_H$ ($10^{20}{\rm~cm^{-2}}$) & $5.24(f)$ & $3.54$ (f) \\
641: $\Gamma$ & $2.46\pm0.01$ & $2.11\pm0.05$ \\
642: $n_{PL}$\tablenotemark{3} & $(1.31\pm0.02)\times10^{-2}$ & $(3.1\pm0.15)\times10^{-3}$ \\
643: $kT_{seed}$($\ev$) & $30$(f) & $30$(f) \\
644: $kT_e$($keV$) & $0.176\pm0.006$ & $0.137_{-0.010}^{+0.012}$ \\
645: $\Gamma_{thcomp}$ & $2.1\pm0.1$ & $2.1\pm0.2$ \\
646: %$n_{C}$ & $2.44\pm0.2$ & $1.02\pm0.15$ \\
647: $E_{FeK\alpha}$ (keV) & $6.72\pm0.03$ & $6.42\pm0.04$ \\
648: $f_{FeK\alpha}$\tablenotemark{4} & $(4.9\pm1.8)\times10^{-6}$ & $(1.1\pm0.5)\times10^{-5}$ \\
649: $f_{0.3-10keV}$\tablenotemark{5} & $6.7\times10^{-11}$ & $2.1\times10^{-11}$ \\
650: $\chi^2/dof$ & $900.2/853$ & $603.0/538$ \\ \tableline
651: \end{tabular}
652: \tablenotetext{1}{For
653: Akn~564, all models were modified by the two phase warm absorber medium
654: inferred from the RGS data and the EPIC-pn data were fitted in the
655: $0.6-11\kev$ band.}
656: \tablenotetext{2}{MCD normalization
657: $n_{MCD}=(R_{in}/km)/(D/10{\rm~kpc})$, where $R_{in}$ is the inner
658: radius and $D$ is the distance.}
659: \tablenotetext{3}{Power-law normalization in units of {$\rm photons~keV^{-1}~cm^{-2}~s^{-1}$} at $1\kev$.}
660: \tablenotetext{4}{Line flux in $\rm photons~keV^{-1}~cm^{-2}~s^{-1}$.}
661: \tablenotetext{5}{Observed flux in units of ${\rm~ergs~cm^{-2}~s^{-1}}$
662: in the $0.3-10\kev$ band.}
663: \tablenotetext{6}{Normalization of the reflected spectrum.}
664: \tablenotetext{7}{Gaussian sigma for velocity smearing in units of $v/c$.}
665: }
666: \end{table*}
667:
668:
669: \subsection{Warm absorber model for Akn~564}
670: To describe the weak absorption features seen in the RGS data, we have
671: created a warm absorber model for Akn~564 using the spectral
672: simulation code CLOUDY version 7.02.01 (last described by Ferland et
673: al. 1998). We used a multicomponent ionizing continuum similar to
674: that observed from Akn~564. We used the `agn' continuum available
675: within CLOUDY. This continuum has four parameters: the temperature of
676: the big blue bump emission, the slope of the high energy X-ray
677: continuum, X-ray to UV ratio ($\alpha_{ox}$) and the UV spectral
678: index. We set the temperature of the big blue bump component and the
679: slope of the high energy continuum to be similar to the temperature of
680: the soft X-ray excess emission and the slope of the hard power law,
681: respectively, derived from the EPIC-pn data. For Akn~564,
682: $\alpha_{ox} = 0.941$ (Gallo 2006) and we used a typical value
683: of $-0.5$ for the UV spectral index (Elvis et al. 1994; Francis 1993).
684: The X-ray power law is assumed to fall as $\nu^3$ above $100\kev$ and
685: the big blue bump component is assumed to have an infra-red
686: exponential cut-off at $kT_{IR} = 0.01{\rm~Ryd}$ or $0.136\ev$. We
687: also included the cosmic microwave background radiation so that the
688: incident continuum has nonzero intensity at very long wavelengths. We
689: assumed a plane parallel geometry and calculated grids of models by
690: varying the ionization parameter and the total hydrogen column
691: density. We also included UTA features from Gu et al. (2006) in our
692: calculation. The grid of models were imported to ISIS in the form of
693: an XSPEC-style multiplicative table model as described in Porter et
694: al. (2006).
695:
696:
697:
698:
699: To describe the weak absorption features in the RGS spectra of
700: Akn~564, we multiplied the CLOUDY warm absorber model to the MCD+PL
701: model. The model was also modified by neutral absorption along the
702: line of sight as before. The addition of the warm absorber component
703: improved the fit ($C/dof=2469.3/1569$; $\Delta C = -106.6$ for three
704: additional parameters) with warm absorber column, $N_W \sim 4\times
705: 10^{20}{\rm~cm^{-2}}$, ionization parameter ($log\xi \sim 2$) and the
706: outflow velocity ($v \sim 300\kms$). An additional warm absorber
707: component further improved the fit ($\Delta C = -70.5$ for three
708: parameters) with $N_W \sim 2\times 10^{-20}{\rm~cm^{-2}}$, $log \xi <
709: 0.3$ and $v \sim -1000\kms$. Examination of the residuals showed a
710: strong absorption feature near $0.5\kev$ at the location of the
711: neutral oxygen edge. However, the shape of the absorption feature does
712: not resemble to an absorption edge, it is similar to an absorption
713: line with a very weak emission line at the center. This feature is
714: likely due to possible RGS calibration errors at the oxygen edge
715: and/or from the Galactic oxygen along the line of sight. Addition of a
716: Gaussian absorption line improved the fit ($\Delta C = -97.3$ for
717: three parameters). The parameters of the best-fit model to the RGS
718: data is listed in Table~\ref{t2}. We note that the best-fit neutral
719: absorption column density derived from the RGS data is $N_H =
720: 5.24_{-0.03}^{+0.13}\times10^{20}{\rm~cm^{-2}}$ which is less than the
721: Galactic column of $6.40\times10^{20}{\rm~cm^{-2}}$ derived from the
722: HI map of Dickey \& Lockman (1990). The best-fit column, however, is
723: consistent with the Galactic column of
724: $5.34\times10^{20}{\rm~cm^{-2}}$ derived from the
725: Leiden/Argentine/Bonn (LAB) Survey of Galactic HI (Kalbera et al.
726: 2005). McKernan, Yaqoob, \& Reynolds (2007) derived cold $N_H =
727: 4_{-1}^{+2}\times10^{20}{\rm~cm^{-2}}$ using the Chandra high energy
728: transmission grating observations of Akn~564. Given the above
729: uncertainties in the measurement of cold $N_H$, we fix the cold
730: absorption at the best-fit RGS value of
731: $N_H=5.24\times10^{20}{\rm~cm^{-2}}$ in all the subsequent fits for
732: Akn~564.
733:
734:
735: \begin{table*}
736: \centering
737: {\small \caption{Best-fit model parameters derived from the high resolution
738: RGS observation of Akn~564 \label{t2}}
739: \begin{tabular}{lll} \tableline\tableline
740: Neutral absorption & $N_H$ ($10^{20}{\rm~cm^{-2}}$) &
741: $5.25_{-0.03}^{+0.13}$ \\ \\
742: Warm absorber I & $N_W$ ($10^{20}{\rm~cm^{-2}}$) & $3.9_{-0.3}^{+0.5}$ \\
743: & $log \xi$ & $2.0\pm0.1$ \\
744: & ($v/\kms$)~\tablenotemark{a} & $-280^{+130}_{-180}$ \\ \\
745: Warm absorber II & $N_W$ ($10^{20}{\rm~cm^{-2}}$) & $2.1_{-0.1}^{+0.2}$ \\
746: & $log \xi$ & $<0.32$ \\
747: & ($v/\kms$)~\tablenotemark{a} & $-1080^{+175}_{-155}$ \\ \\
748: Disk blackbody & $kT_{in}$ (eV) & $171\pm2$ \\
749: & $n_{MCD}$\tablenotemark{b} & $2917_{-69}^{+18}$ \\ \\
750: Power law & $\Gamma$ & $2.48\pm0.01$ \\
751: & $n_{PL}$\tablenotemark{c} & $(1.50\pm0.01)\times10^{-2}$
752: \\ \\
753: Absorption line & $E$ (keV) & $0.541\pm0.001$ \\
754: & $f_{line}$\tablenotemark{d} & $-2.6_{-0.3}^{+0.5}\times 10^{-4}$ \\
755: $C/dof$ & & $2301.5/1577$ \\ \tableline
756: \end{tabular}
757: \tablenotetext{a}{Velocity with respect to the systemic
758: velocity. Negative sign indicates outflow.}
759: \tablenotetext{b}{MCD normalization
760: $n_{MCD}=(R_{in}/km)/(D/10{\rm~kpc})$, where $R_{in}$ is the inner
761: radius and $D$ is the distance.}
762: \tablenotetext{c}{Power-law normalization in units of {$\rm photons~keV^{-1}~cm^{-2}~s^{-1}$} at $1\kev$.}
763: \tablenotetext{d}{Line flux in $\rm photons~keV^{-1}~cm^{-2}~s^{-1}$.}
764: }
765: \end{table*}
766:
767: The above analysis clearly shows the presence of a two phase warm
768: absorber medium in Akn~564. Some or all of the emission/absorption
769: features resulting in the poor quality of the fit to the EPIC-pn,
770: described above in section 4.1 are likely due to the presence of warm
771: absorbers. However, it is difficult to determine accurately the
772: parameters of the warm absorber based on the lower resolution EPIC-pn
773: data alone. Therefore, we used the best-fit warm absorber components
774: derived from the RGS data to model the EPIC-pn data. We used the two
775: component warm absorber model with the fixed parameters along with the
776: MCD+PL model and fitted the EPIC-pn data. The addition of the warm
777: absorber components improved the fit, providing $\chi^2/dof =
778: 1654.1/912$. However, the fit is still not acceptable
779: mainly due to a strong absorption feature near $0.5\kev$ and slight
780: excess of emission below $0.5\kev$. The absorption feature is stronger
781: by a factor of $\sim 2.5$ in the EPIC-pn data than that derived from
782: the RGS data. Also the excess emission below $0.5\kev$ is not evident
783: in the RGS data. Thus EPIC-pn and RGS do not agree well below
784: $0.6\kev$. Therefore we ignore the EPIC-pn data below $0.6\kev$ in
785: the fits for Akn~564 described below.
786:
787:
788:
789: After excluding the data below $0.6\kev$, the MCD+PL model modified
790: with the two component warm absorber and neutral absorption along the
791: line of sight provided a much better fit ($\chi^2/dof =
792: 941.2/856$). Adding a narrow Gaussian line at $\sim 6.6\kev$ further
793: improved the fit ($\Delta chi^2/dof = -21.9$ for two parameters).
794: Figure~\ref{f10} shows the EPIC-pn data, the best-fit model and the
795: $\chi$-residuals and the best-fit parameters are listed in
796: Table~\ref{t1}. The parameters of the MCD and PL components are
797: similar to that derived from the RGS data (see Table~\ref{t2} and
798: \ref{t2}).
799: The MCD component used above to describe the soft X-ray excess
800: emission is physically inappropriate. The inner accretion disk
801: temperatures, $kT_{in} = 179\pm2\ev$ for Akn~564 and $138\pm3\ev$ for
802: Mrk~1044 are much higher than that expected from a disk around a black
803: hole of mass $10^6 - 10^7{\rm~M_{\odot}}$. For this reason, the
804: origin of the soft excess emission has remained unsolved for more than
805: two decades.
806:
807: \begin{figure*}
808: \centering
809: \includegraphics[width=4.7cm,angle=-90]{f9a.ps}
810: \includegraphics[width=4.7cm,angle=-90]{f9b.ps}
811: \includegraphics[width=4.7cm,angle=-90]{f9c.ps}
812: \includegraphics[width=4.7cm,angle=-90]{f9d.ps}
813: \caption{Observed EPIC-pn data and the best-fit models for Mrk~1044
814: corresponding to the best-fit models and their parameters listed
815: in Table~\ref{t1}.}
816: \label{f9}
817: \end{figure*}
818:
819: \subsection{Relativistically blurred ionized reflection}
820:
821: One possibility is that the reflected emission from a partially
822: ionized accretion disk could be strong below $1\kev$ and can account
823: for the strong soft excess emission (see e.g., Crummy et al. 2005).
824: Though there is no clear evidence for a broad iron K$\alpha$ line,
825: such a line from an accretion disk could be hidden in the strong X-ray
826: continuum. To test whether the strong soft X-ray excess emission from
827: Akn~564 and Mrk~1044 could be produced by the ionized reflection
828: emission arising from the reprocessing of the primary X-ray emission
829: in the disk, we have fitted the relativistically blurred ionized
830: reflection model to the EPIC-pn spectra of both the NLS1. We used the
831: table model {\tt reflion} (Ross \& Fabian 2005; Ross, Fabian \& Young
832: 1999) for the ionized reflection and blur the ionized reflection
833: emission relativistically by convolving with a {\tt LAOR} kernel ({\tt
834: kdblur}) to obtain the relativistically broaden ionized reflection
835: from the inner regions of an accretion disk. The parameters of {\tt
836: reflion} model are ionization parameter ($\xi$), iron abundance
837: relative to solar ($A_{Fe}$), photon index of the illuminating
838: power-law and the normalization of reflected spectrum. The parameters
839: of the convolution model {\tt kdblur} are the inner and outer radii of
840: the disk ($r_{in}$ and $r_{out}$), emissivity index ($\beta$) and the
841: disk inclination ($i$). We fixed the outer radius at $r_{out} =
842: 400r_g$ where $r_g = GM/c^2$ is the gravitational radius. We used the
843: power-law component for the continuum. We also used a narrow Gaussian
844: to describe the unresolved iron K$\alpha$ line from distant matter. We
845: also used the two component warm absorber model with the parameters
846: fixed at the values derived from the RGS data. We refer the full model
847: as the blurred reflection model. The best-fit parameters of this model
848: are listed in Table~\ref{t1}. The observed data and the best-fit
849: reflection models are shown in Figure~\ref{f9} (Mrk~1044) and
850: Figure~\ref{f10} (Akn~564). The model resulted in poorer fit than the
851: MCD+PL model for both the NLS1 ($\chi^2/dof = 954.2/837$ for Akn564
852: and $671.6/535$ for Mrk~1044). Also the emissivity index is
853: unphysically large ($\beta \ge 10$) and the inner radius is very small
854: for both the AGN, suggesting that the model cannot account for the
855: observed smoothness of the soft excess X-ray emission. The very steep
856: emissivity indices and the small inner radii of the LAOR kernel were
857: required to smooth the recombination features sufficiently to produce
858: the featureless soft X-ray excess emission from both the NLS1s.
859:
860: \begin{figure*}
861: \centering
862: \includegraphics[width=4.5cm,angle=-90]{f10a.ps}
863: \includegraphics[width=4.5cm,angle=-90]{f10b.ps}
864: \includegraphics[width=4.5cm,angle=-90]{f10c.ps}
865: \includegraphics[width=4.5cm,angle=-90]{f10d.ps}
866: \caption{Observed EPIC-pn data and the best-fitting MCD+PL, blurred
867: ionized reflection, smeared absorption from partially ionized wind
868: and optically thick thermal Comptonization models for Akn~564. The
869: best-fit parameters are listed in Table~\ref{t1}.}
870: \label{f10}
871: \end{figure*}
872:
873:
874: \subsection{Smeared absorption from partially ionized material}
875: The observed soft excess emission could be an artifact of smeared
876: absorption from partially ionized material as described by Gierlin'ski
877: \& Done (2004). To test this scenorio, we have fitted the EPIC-pn data
878: with the smeared absorption model. We used the XSPEC local model {\tt
879: swind1} for the smeared absorption from a partially ionized wind
880: from an accretion disk and the simple PL model for the continuum.
881: Again we used a Gaussian for the narrow iron K$\alpha$ line and the
882: two component warm absorber model with fixed parameters derived from
883: the RGS data. The parameters of the smeared wind model are absorption
884: column density, the ionization parameter ($\xi=L/nr^2$) and the
885: Gaussian sigma for velocity smearing in units of $v/c$. The smeared
886: absorption model provided a good fit to the spectrum of
887: Mrk~1044 ($\chi^2/dof = 592.4/538$) but fails to describe the spectrum
888: of Akn~564 ($\chi^2/dof = 1147.1/842$). This model requires
889: unphysically large smearing velocity ($\sim 0.8c$) for Akn~564 to
890: smooth sufficiently the absorption lines in order to produce the
891: observed smooth soft excess emission (see Table~\ref{t1} and
892: Fig.~\ref{f10}).
893:
894:
895:
896: \subsection{Optically-thick, thermal Comptonization}
897: The strong and smooth soft excess emission from NLS1 can also be
898: described as arising from thermal Comptonization in an optically thick
899: corona. To test such a scenario, we used the thermal Comptonization model
900: (ThComp) described by Zdziarski et al (1996). The free parameters of
901: ThComp model are the asymptotic power-law index ($\Gamma_{thcomp}$),
902: electron temperature ($kT_e$) and the seed photon temperature
903: ($kT_{bb}$). The electron scattering optical depth ($\tau$) can be
904: calculated from the asymptotic power-law photon index
905: ($\Gamma_{compth}$) and electron temperature ($kT_e$) as follows
906: \begin{equation}
907: \alpha = \left[\frac{9}{4} + \frac{1}{(kT_e / m_e c^2) \tau (1 + \tau / 3)}\right]^{1/2} - \frac{3}{2}
908: \end{equation}
909: with $\Gamma_{thcomp} = \alpha + 1$ (Sunyaev \& Titarchuk 1980).
910: We also used the
911: PL model to describe the hard PL for the hot Comptonized
912: component. We fixed the temperature of the seed photons for the
913: optically thick Comptonization at $30\ev$. This is equivalent to
914: assuming that the seed photons for the Comptonizations in the warm and
915: hot plasma arise from the accretion disk. Again we used the narrow
916: Gaussian line for the iron K$\alpha$ line.
917: This model provided good fits for both
918: the NLS1s ($\chi^2/dof = 900.2/842$ for Akn~564 and
919: $\chi^2/dof=603.0/538$ for Mrk~1044). The best-fit parameters are
920: listed in Table~\ref{t1} and the EPIC-pn data and the best-fit
921: model are plotted in Fig.~\ref{f9} and Fig.~\ref{f10} . The electron
922: temperatures are low ($\sim 0.18\kev$ for Akn~564 and $\sim0.14\kev$
923: for Mrk~1044) and electron scattering optical depths are very high
924: ($\tau \sim 45$) for both the NLS1s.
925:
926:
927:
928: We note that the thermal Comptonization model best describes the
929: EPIC-pn data in the $0.6-11\kev$ band for Akn~564 (see Table~\ref{t1}).
930: We expect that the most suitable model must provide a reasonably good
931: fit to the full band EPIC-pn data except for the absorption feature at
932: $0.54\kev$, provided that the calibration errors at low energies
933: ($<0.5\kev$) are at a level of a few percent. We find that the thermal
934: Comptonization model best describes the full band data with
935: $\chi^2/dof = 1164.8/911$ without any significant change in the
936: best-fit parameters listed in Table~\ref{t1}. Addition of an
937: absorption line at $0.54\kev$ improved the fit to $\chi^2/dof =
938: 985.5/908$ with line center at $E=0.54\pm0.005$, $f_{line} = -
939: (4.44\pm0.06)\times 10^{-4}{\rm~ergs~cm^{-2}~s^{-1}}$ and width
940: $\sigma < 8\ev$. The ionized reflection and smeared absorption models
941: both fail to describe the data below $0.5\kev$ when extended to lower
942: energies. Refitting the reflection model with the addition of an
943: absorption line at $0.54\kev$ to the full band data resulted in poor
944: fit ($\chi^2/dof = 1392.2/906$) with large emissivity index ($\beta =
945: 9.17_{-0.03}^{+0.20}$) and small inner radius ($r_{in} < 1.239r_g$) as
946: before. The smeared absorption model with the addition of the
947: absorption line provided the worst fit among the four models.
948:
949:
950:
951: \section{Discussion}
952: We have studied the X-ray temporal and spectral characteristics of
953: Akn~564 and Mrk~1044. We found that the $0.2-2\kev$ and $2-10\kev$
954: band lightcurves are strongly correlated without any time delay.
955: However, the strength of this correlation decreases with increasing
956: separation between the soft and hard bands. We have found significant
957: correlation between the soft ($0.2-0.3\kev$, $0.2-0.5\kev$) and hard
958: ($5-10\kev$, $4-10\kev$) band lightcurves at positive time delays. The
959: $4-10\kev$ band lags behind the $0.2-0.5\kev$ band by $1767\pm122\s$
960: (at $90\%$) in Akn~564. We note that Arevalo et al (2006) found time
961: delays that increase both with time scale and energy separation
962: between the energy bands. This means that there is no single, well
963: defined delay. Instead, the measured delay based on the
964: cross-correlation analysis is the averaged delay between the soft and
965: hard bands. We find no correlation at zero time lag between the
966: $0.2-0.3\kev$ and $5-10\kev$ band emission from Mrk~1044. We confirm
967: the presence of a break at $\sim 1.2\times 10^{-3}\hz$ in the power
968: density spectrum of Akn~564, earlier detected by Papadakis et al.
969: (2002) based on \asca{} observations. We also find that the PDS of
970: Akn~564 is energy dependent. The hard band ($3-10\kev$) PDS,
971: consistent with a simple power law, is flatter and stronger than the
972: soft band ($0.2-0.5\kev$) PDS that shows clear evidence for the break.
973: These observations strongly suggest that different variability process
974: dominate the soft and hard band emission. This may suggest physically
975: distinct spectral components for the soft excess and the power-law
976: components.
977:
978: \subsection{Implications of the observed time delays}
979: The observed hard band delays from Akn~564 demonstrate that the soft
980: excess emission cannot be the reprocessed emission of the primary X-ray
981: emission. Fabian et al. (2002) presented the reflection dominated
982: ionized accretion disk model and successfully reproduced the strong
983: soft excess emission from the NLS1 galaxy 1H~0707-495. In this model,
984: the strong soft excess is caused by the multiple reflections of mostly
985: hidden hard X-ray continuum by the clumpy disk and the stronger
986: reflectivity of the partially ionized material below $\sim 0.7\kev$,
987: associated with an abrupt change in the opacity (Ross, Fabian \&
988: Ballantyne 2002; Fabian et al. 2004). The soft X-ray excess emission
989: is the reprocessed emission in this model, hence it is unlikely that
990: the soft X-rays lead the hard X-ray emission. This is, however, not a
991: strong constraint on the ionized reflection model. Many of the soft
992: X-ray lines expected from the accretion disk are excited by soft X-ray
993: photons with energies slightly greater than the line energies as the
994: photoelectric absorption cross section varies as $\sim E^{-3}$. The
995: ionizing soft X-ray photons can lead the hard ($3-10\kev$) band
996: emission due to the Compton delay of hard photons in the
997: Comptonization process. Since the life times of excited states,
998: responsible for the line photons, are very short compared to the
999: expected Compton delay between the soft and hard photons, the line
1000: photons as well as the scattered soft X-ray emission can still lead
1001: the hard band primary emission. However, in this case, the soft X-ray
1002: excess emission is not expected to show rapid and large amplitude
1003: variability as the reflection component and broad iron line are known
1004: to show little variability in MCG-6-30-15 (Miniutti et al. 2007).
1005: Moreover, the blurred ionized reflection model with physically
1006: plausible parameters cannot reproduce the
1007: observed soft X-ray excess emission from Akn~564 and Mrk~1044.
1008:
1009: Gierlinski \& Done (2004) presented an alternative model for the
1010: origin of the soft excess emission.
1011: In this model, the soft excess emission is mainly due to the strong
1012: and smeared absorption near $\sim 0.7\kev$. The soft excess and the
1013: hard power law are part of the same and single continuum component
1014: except for the small contribution from smeared emission lines. This
1015: means the soft and hard band emission arise from the same continuum
1016: component or physical process - Comptonization. The time delay is a
1017: natural outcome of the Comptonization process. A general prediction
1018: of Comptonization models is that the hard X-ray variations should lag
1019: behind those in softer bands (Payne 1980). Therefore it is natural to
1020: attribute the observed time delays between hard and soft photons to
1021: this process. The time lag represents the difference in the photon
1022: escape times between the soft and hard bands. The hard band photons
1023: tend to have undergone more scattering events, therefore travel longer
1024: and escape the corona later than the soft band photons. Thus the
1025: observed time lag is consistent with a single continuum component and
1026: hence with the smeared absorption model. However, there are three
1027: problems with the smeared wind model. ($i$) This model (XSPEC local
1028: model {\tt swind1}; Gierli{\'n}ski \& Done 2004) does not describe the
1029: X-ray spectrum of Akn~564 satisfactorily. ($ii$) The $3-10\kev$ band
1030: shows more variability power on short time scales than the
1031: $0.2-0.5\kev$ band, which is not expected in the simple Comptonization
1032: models in which the variability is caused by the variations in the seed
1033: photons alone. ($iii$) The smearing velocity for Akn~564 suggested by the
1034: {\tt swind1} model is very large ($v \sim 0.8c$) for disk winds.
1035: Hydrodynamical simulations show that the line driven winds do not have
1036: the required smearing velocity (N. J. Schurch, private
1037: communication). It remains to be seen if the magnetically driven
1038: winds can provide the required smearing.
1039:
1040: The two component model, Comptonization in two separate regions, is a
1041: viable option. In this model, the soft excess emission arises from
1042: the Comptonization in the low temperature (warm), optically thick
1043: plasma while the hard power-law is produced by the Comptonization in
1044: the hot plasma. Indeed the EPIC-pn spectra of Akn~564 and Mrk~1044
1045: are both well described by the two component model. The temperature of
1046: the warm plasma is $\sim 175\ev$ for ~Akn~564 and $\sim135\ev$ for
1047: Mrk~1044. This region, giving rise the soft excess emission, is
1048: optically thick ($\tau \sim 45$). Physically, this component could be
1049: the warm skin on the accretion disk surface or a region between the
1050: disk and an optically thin inner flow (Magdziarz et al. 1998) or an
1051: optically thick corona coupled with a truncated disk (Done \& Kubota
1052: 2006). Depending on the geometry of the disk and the two component
1053: corona system, there can be delay between the soft excess and high
1054: energy power-law components.
1055:
1056: However, the two component Comptonization model has difficulty in
1057: explaining the constancy of the temperature of the Comptonizing region
1058: producing the soft excess emission. Gierlinski \& Done (2004) derived
1059: the temperature of the putative Comptonizing region in the range
1060: $0.1-0.2\kev$ which is remarkably constant for 26 different PG
1061: quasars. They concluded that the constancy of the temperature is
1062: inconsistent with that expected from the span of the asymptotic
1063: power-law index ($\Gamma_{thcomp} \sim 1.5-2.5$). However, the
1064: $0.3-10\kev$ \xmm{} band only includes the high energy tail of the
1065: optically thick Comptonized emission. Both the peak of the soft
1066: excess component ($kT\sim 0.1-0.2\kev$) and the asymptotic power-law
1067: component fall below $0.3\kev$. Using such data, it may be difficult
1068: to measure accurately both the temperature and the asymptotic photon
1069: index, particularly in the presence of line of sight neutral
1070: absorption and complex warm absorber, both modifying the soft X-ray
1071: spectrum below $1\kev$. NLS1 galaxies show a large diversity in their
1072: soft X-ray \rosat{} photon indices ($\Gamma \sim 1.5-5.0$), suggesting
1073: varying shape of the soft excess component (Boller, Brandt \& Fink
1074: 1996). The constancy of the temperature for the soft excess emission
1075: is required to be investigated using a sample of NLS1 galaxies similar
1076: to that used by Boller, Brandt \& Fink (1996). Another possibility is
1077: that the disk/corona geometry is different in NLS1 galaxies as they
1078: are accreting close to the Eddington rate. In such situations, the
1079: inner disk is likely thick and hotter. It remains to be seen if such
1080: thick structure can have uniform temperatures across AGN with a large
1081: range in their luminosity and black hole masses.
1082:
1083:
1084: \subsection{The disk-corona geometry}
1085:
1086: Inverse Compton scattering by thermal electrons provides the best
1087: possible explanation for the observed smooth soft X-ray excess
1088: emission from Akn~564 and Mrk~1044. Our spectral analysis favors two
1089: physically distinct corona: high optical depth, low temperature corona
1090: and high temperature, low optical depth corona. There are two possible
1091: geometries with two component corona. The cool, optically thick corona
1092: may be similar to the coupled disk corona geometry proposed by Done \&
1093: Kubota (2006) for the very high state of black hole binaries. In this
1094: geometry, the inner accretion disk is truncated and is coupled to an
1095: optically thick corona, and gravitational energy release powers both
1096: the disk and the optically-thick corona. Comptonization of the disk
1097: photons in this corona may give rise to the broad soft excess
1098: component. In addition to the optically thick corona, a hot corona is
1099: responsible for the power-law component. The hot corona could be the
1100: same as the patchy corona above the disk and is likely powered by
1101: magnetic flares in the disk similar to the solar flares (Haardt,
1102: Maraschi, \& Ghisellini 1994; Stern et al. 1995; Poutanen \& Svensson
1103: 1996; Poutanen \& Fabian 1999; Merloni \& Fabian 2001). The seed
1104: photons could either be the soft photons from the disk or those
1105: arising from the cool, optically thick corona. Since the disk is
1106: coupled to the optically thick corona, appreciable time lag between
1107: the disk photons and the soft excess photons are not expected. The
1108: variability of the disk flux and the soft excess emission is governed
1109: by the variations in the accretion rate, while the variability of the
1110: power-law component is governed both by the variation in the soft
1111: photons as well as any variability process intrinsic to the corona
1112: such as that related to variation in the electron population. The
1113: flatter power density spectrum of Akn~564 in the hard band may be
1114: related to the additional variability of the power-law component due
1115: to the changes in electron population on short time scales e.g.,
1116: caused by magnetic flares above the accretion disk.
1117:
1118: Another possible geometry is the ionized surface of a disk acting as
1119: an optically thick and cool corona with a temperature of $\sim
1120: 0.1-0.3\kev$ (e.g, Magdziarz et al. 1998; see also O'Brian et al.
1121: 2001; Page et al. 2002; Vaughan et al. 2002). In this case, thermal
1122: Comptonization of disk photons by thermal electrons in the hot surface
1123: of the disk will produce the smooth soft excess. The steep power-law
1124: extending to high energies can be produced in another purely hot
1125: thermal/nonthermal or in a hybrid thermal-nonthermal plasma.
1126:
1127: It is difficult to distinguish the two geometries described above. The
1128: different power spectra of the soft and hard band X-ray emission may
1129: constrain the two geometries. Recently McHardy et al. (2006) have
1130: shown a strong correlation between the break frequency, black hole
1131: mass and the bolometric luminosity for black hole binaries and AGN. If
1132: the break frequencies are related to the characteristic radii from the
1133: central source, their result suggests a characteristic size that
1134: appears to decrease with increasing accretion rate or luminosity. The
1135: clear evidence for a break in the soft band PDS of Akn~564 suggests a
1136: characteristic size associated with the soft X-ray excess emission
1137: which can be the size of the optically thick corona or the radius of
1138: the inner truncated disk in the model suggested by Done \& Kubota
1139: (2006). It is also possible that the disk surface below an inner
1140: radius is ionized and the temperature of the inner disk surface is
1141: higher than that expected from the standard disk. The inner disk
1142: surface may then be responsible for the soft excess emission. The lack
1143: of a clear break in the hard band PDS of Akn~564 may suggest that
1144: power-law emission is probably not associated with a characteristic
1145: size and it arises from an extended corona similar to a patchy corona
1146: above an accretion disk. Future long observations of Akn~564 are
1147: required to clearly demonstrate the presence or absence of a break in
1148: the hard band PDS.
1149:
1150: \section{Conclusions}
1151: We have presented temporal and spectral study of \xmm{} observations
1152: of Akn~564 and Mrk~1044. The main results are as follows:
1153: \begin{enumerate}
1154: \item The $0.2-2\kev$ and $2-10\kev$ band X-ray emission from both the
1155: NLS1 galaxies are strongly correlated without any time delay.
1156: However, the variations in the $0.2-0.5\kev$ band are found to lead
1157: those in the $4-10\kev$ band by $1768\pm122\s$ ($90\%$) in Akn~564.
1158: We do not find any correlation between the $0.2-0.3\kev$ and
1159: $5-10\kev$ X-ray emission from Mrk~1044 at zero time delay. The two
1160: bands appear to anti-correlate at a positive lag of $\sim 1000\s$
1161: and likely correlated at a lag of $\sim3000 - 5000\s$. Long
1162: observations of Mrk~1044 are required to confirm the time delay.
1163: \item The full band power density spectrum of Akn~564 has a break at
1164: $(1.2\pm0.3)\times 10^{-3}\hz$, corresponding to a time scale of
1165: $\sim 1000\s$, similar to the delay between $0.2-0.5$ and $4-10\kev$
1166: bands. It is not clear if there is any physical relation between the
1167: two time scales. There is a clear evidence of a break in the soft
1168: band PDS but the hard band does not show a clear evidence for the
1169: break.
1170: \item The soft ($0.2-0.5\kev$) band PDS of Akn~564 is significantly
1171: steeper and weaker than the hard ($3-10\kev$) band PDS. This implies
1172: that the power-law component is more variable than the soft excess
1173: component on shorter time scales. The implication of this result on
1174: the Comptonization models is that the variations in the power-law
1175: component are not only caused by the variations in the seed photons
1176: but also by the variations in the hot electron population. Thus
1177: electron injection on short time scales ($\la 1000\s$) are required.
1178: This could be evidence for magnetic flares thought to power the
1179: corona above the accretion disk.
1180: \item The soft excess emission from Mrk~1044 is featureless in the
1181: EPIC-pn data and is well described by the smeared absorption or
1182: optically thick Comptonization. The RGS data show evidence for warm
1183: absorbers in Akn~564. The EPIC-pn spectrum of Akn~564 is
1184: well described by a complex model consisting of optically thick
1185: thermal Comptonization in a low temperature ($\sim 0.15\kev$) plasma
1186: and a steep power-law, modified by two phase warm absorber medium
1187: and the Galactic absorption. The ionized reflection and smeared wind
1188: models fail to describe the data satisfactorily.
1189: \item The temporal and spectral characteristics of Akn~564 and
1190: Mrk~1044 are consistent with a two component corona -- a compact low
1191: temperature, optically thick corona and an extended hot corona. The
1192: compact corona could be an inner optically thick region coupled to a
1193: truncated disk or the ionized surface of an inner untruncated disk.
1194: The soft excess emission can be produced in the inner, optically
1195: thick corona and the power-law component is produced in the hot
1196: corona.
1197: \end{enumerate}
1198:
1199: \acknowledgements We are grateful to the referee, I. Papadakis, for
1200: his detailed comments and suggestions that improved this paper
1201: significantly. We thank M. Nowak for writing a number of useful ISIS
1202: functions that have been used in this paper. GCD gratefully
1203: ackowledges the support of NASA grants NNX07AE99G and NNX06AE38G. This
1204: work is based on observations obtained with \xmm{}, an ESA science
1205: mission with instruments and contributions directly funded by ESA
1206: Member States and the USA (NASA). This research has made use of data
1207: obtained through the High Energy Astrophysics Science Archive Research
1208: Center Online Service, provided by the NASA/Goddard Space Flight
1209: Center.
1210:
1211: \begin{thebibliography}{}
1212:
1213: %\bibitem[Agrawal \& Misra(2006)]{2006ApJ...638L..83A} Agrawal, V.~K.,
1214: % \& Misra, R.\ 2006, \apjl, 638, L83
1215:
1216: \bibitem[Ar{\'e}valo et al.(2006)]{2006MNRAS.372..401A} Ar{\'e}valo,
1217: P., Papadakis, I.~E., Uttley, P., McHardy, I.~M., \& Brinkmann, W.\
1218: 2006, \mnras, 372, 401
1219:
1220: \bibitem[Arnaud et al.(1985)]{1985MNRAS.217..105A} Arnaud, K.~A., et
1221: al.\ 1985, \mnras, 217, 105
1222:
1223: \bibitem[Boller et al.(1996)]{1996A&A...305...53B} Boller, T., Brandt,
1224: W.~N., \& Fink, H.\ 1996, \aap, 305, 53
1225:
1226: \bibitem[Brinkmann et al.(2007)]{2007A&A...465..107B} Brinkmann, W.,
1227: Papadakis, I.~E., \& Raeth, C.\ 2007, \aap, 465, 107
1228:
1229:
1230: \bibitem[Collinge et al.(2001)]{2001ApJ...557....2C} Collinge,
1231: M.~J.~et al.\ 2001, \apj, 557, 2
1232:
1233: \bibitem[Crummy et al.(2006)]{2006MNRAS.365.1067C} Crummy, J., Fabian,
1234: A.~C., Gallo, L., \& Ross, R.~R.\ 2006, \mnras, 365, 1067
1235:
1236: \bibitem[Czerny et al.(2003)]{2003A&A...412..317C} Czerny, B.,
1237: Niko{\l}ajuk, M., R{\'o}{\.z}a{\'n}ska, A., Dumont, A.-M., Loska,
1238: Z., \& Zycki, P.~T.\ 2003, \aap, 412, 317
1239:
1240: \bibitem[Dasgupta \& Rao(2006)]{2006ApJ...651L..13D} Dasgupta, S., \&
1241: Rao, A.~R.\ 2006, \apjl, 651, L13
1242:
1243: \bibitem[Dewangan et al.(2002)]{2002A&A...390...65D} Dewangan, G.~C.,
1244: Boller, T., Singh, K.~P., \& Leighly, K.~M.\ 2002, \aap, 390, 65
1245:
1246: \bibitem[Dickey \& Lockman(1990)]{1990ARA&A..28..215D} Dickey, J.~M., \&
1247: Lockman, F.~J.\ 1990, \araa, 28, 215
1248:
1249: \bibitem[Done \& Kubota(2006)]{2006MNRAS.tmp..843D} Done, C., \&
1250: Kubota, A.\ 2006, \mnras, 843
1251:
1252: \bibitem[Elvis et al.(1994)]{1994ApJS...95....1E} Elvis, M., et al.\ 1994,
1253: \apjs, 95, 1
1254:
1255: \bibitem[Fabian et al.(2002)]{2002MNRAS.331L..35F} Fabian, A.~C.,
1256: Ballantyne, D.~R., Merloni, A., Vaughan, S., Iwasawa, K., \& Boller,
1257: T.\ 2002, \mnras, 331, L35
1258:
1259: \bibitem[Fabian et al.(2004)]{2004MNRAS.353.1071F} Fabian, A.~C.,
1260: Miniutti, G., Gallo, L., Boller, T., Tanaka, Y., Vaughan, S., \&
1261: Ross, R.~R.\ 2004, \mnras, 353, 1071
1262:
1263: \bibitem[Fabian et al.(2005)]{2005MNRAS.361..795F} Fabian, A.~C.,
1264: Miniutti, G., Iwasawa, K., \& Ross, R.~R.\ 2005, \mnras, 361, 795
1265:
1266: \bibitem[Ferland et al.(1998)]{1998PASP..110..761F} Ferland, G.~J.,
1267: Korista, K.~T., Verner, D.~A., Ferguson, J.~W., Kingdon, J.~B., \& Verner,
1268: E.~M.\ 1998, \pasp, 110, 761
1269:
1270: \bibitem[Francis(1993)]{1993ApJ...407..519F} Francis, P.~J.\ 1993, \apj,
1271: 407, 519
1272:
1273: \bibitem[Gallo et al.(2004)]{2004MNRAS.347..269G} Gallo, L.~C., Boller, T.,
1274: Tanaka, Y., Fabian, A.~C., Brandt, W.~N., Welsh, W.~F., Anabuki, N., \&
1275: Haba, Y.\ 2004, \mnras, 347, 269
1276:
1277: \bibitem[Gallo(2006)]{2006MNRAS.368..479G} Gallo, L.~C.\ 2006, \mnras, 368,
1278: 479
1279: \bibitem[Gierli{\'n}ski \& Done(2004)]{GD04} Gierli{\'n}ski, M., \&
1280: Done, C.\ 2004, \mnras, 349, L7
1281:
1282: \bibitem[Gierli{\'n}ski \& Done(2006)]{2006MNRAS.371L..16G}
1283: Gierli{\'n}ski, M., \& Done, C.\ 2006, \mnras, 371, L16
1284:
1285:
1286: \bibitem[Goad et al.(2006)]{2006MNRAS.365..191G} Goad, M.~R., Roberts,
1287: T.~P., Reeves, J.~N., \& Uttley, P.\ 2006, \mnras, 365, 191
1288:
1289: \bibitem[Gondoin, Orr, Lumb, \&
1290: Santos-Lleo(2002)]{2002A&A...388...74G} Gondoin, P., Orr, A., Lumb,
1291: D., \& Santos-Lleo, M.\ 2002, \aap, 388, 74
1292:
1293: \bibitem[Gu et al.(2006)]{2006ApJ...641.1227G} Gu, M.~F., Holczer, T.,
1294: Behar, E., \& Kahn, S.~M.\ 2006, \apj, 641, 1227
1295:
1296:
1297: \bibitem[Haardt et al.(1994)]{1994ApJ...432L..95H} Haardt, F.,
1298: Maraschi, L., \& Ghisellini, G.\ 1994, \apjl, 432, L95
1299:
1300: \bibitem[Houck \& Denicola(2000)]{2000ASPC..216..591H} Houck, J.~C.,
1301: \& Denicola, L.~A.\ 2000, Astronomical Data Analysis Software and
1302: Systems IX, 216, 591
1303:
1304:
1305: \bibitem[Houck(2002)]{2002hrxs.confE..17H} Houck, J.~C.\ 2002, High
1306: Resolution X-ray Spectroscopy with XMM-Newton and Chandra,
1307: Proceedings of the international workshop held at the Mullard Space
1308: Science Laboratory of University College London, Holmbury St Mary,
1309: Dorking, Surrey, UK, October 24 - 25, 2002, Ed. Branduardi-Raymont,
1310: G., published electronically.
1311:
1312: \bibitem[Kalberla et al.(2005)]{2005A&A...440..775K} Kalberla, P.~M.~W.,
1313: Burton, W.~B., Hartmann, D., Arnal, E.~M., Bajaja, E., Morras, R., {P{\"o}ppel}, W.~G.~L.\ 2005, \aap, 440, 775
1314:
1315: \bibitem[Mason et al.(2003)]{2003ApJ...582...95M} Mason, K.~O., et
1316: al.\ 2003, \apj, 582, 95
1317:
1318:
1319: \bibitem[McHardy et al.(2006)]{2006Natur.444..730M} McHardy, I.~M.,
1320: Koerding, E., Knigge, C., Uttley, P., \& Fender, R.~P.\ 2006, \nat,
1321: 444, 730
1322:
1323: \bibitem[McKernan et al.(2007)]{2007MNRAS.379.1359M} McKernan, B., Yaqoob,
1324: T., \& Reynolds, C.~S.\ 2007, \mnras, 379, 1359
1325:
1326: \bibitem[Magdziarz et al.(1998)]{1998MNRAS.301..179M} Magdziarz, P.,
1327: Blaes, O.~M., Zdziarski, A.~A., Johnson, W.~N., \& Smith, D.~A.\
1328: 1998, \mnras, 301, 179
1329:
1330: \bibitem[Markoff et al.(2005)]{2005ApJ...635.1203M} Markoff, S.,
1331: Nowak, M.~A., \& Wilms, J.\ 2005, \apj, 635, 1203
1332:
1333: \bibitem[Merloni \& Fabian(2001)]{2001MNRAS.328..958M} Merloni, A., \&
1334: Fabian, A.~C.\ 2001, \mnras, 328, 958
1335:
1336: \bibitem[Papadakis \& Lawrence(1993)]{1993MNRAS.261..612P} Papadakis,
1337: I.~E., \& Lawrence, A.\ 1993, \mnras, 261, 612
1338:
1339: \bibitem[Papadakis et al.(2002)]{2002A&A...382L...1P} Papadakis,
1340: I.~E., Brinkmann, W., Negoro, H., \& Gliozzi, M.\ 2002, \aap, 382,
1341: L1
1342:
1343:
1344:
1345: \bibitem[Papadakis et al.(2007)]{2007A&A...461..931P} Papadakis,
1346: I.~E., Brinkmann, W., Page, M.~J., McHardy, I., \& Uttley, P.\ 2007,
1347: \aap, 461, 931
1348:
1349: \bibitem[Payne(1980)]{1980ApJ...237..951P} Payne, D.~G.\ 1980, \apj,
1350: 237, 951
1351:
1352:
1353: \bibitem[Peterson et al.(1998)]{1998PASP..110..660P} Peterson, B.~M.,
1354: Wanders, I., Horne, K., Collier, S., Alexander, T., Kaspi, S., \&
1355: Maoz, D.\ 1998, \pasp, 110, 660
1356:
1357: \bibitem[Porter et al.(2006)]{2006PASP..118..920P} Porter, R.~L., Ferland,
1358: G.~J., Kraemer, S.~B., Armentrout, B.~K., Arnaud, K.~A., \& Turner, T.~J.\
1359: 2006, \pasp, 118, 920
1360:
1361:
1362: \bibitem[Pounds et al.(1995)]{1995MNRAS.277L...5P} Pounds, K.~A.,
1363: Done, C., \& Osborne, J.~P.\ 1995, \mnras, 277, L5
1364:
1365: \bibitem[Pravdo et al.(1981)]{1981ApJ...251..501P} Pravdo, S.~H.,
1366: Nugent, J.~J., Nousek, J.~A., Jensen, K., Wilson, A.~S., \& Becker,
1367: R.~H.\ 1981, \apj, 251, 501
1368:
1369: \bibitem[Poutanen \& Svensson(1996)]{1996ApJ...470..249P} Poutanen,
1370: J., \& Svensson, R.\ 1996, \apj, 470, 249
1371:
1372: \bibitem[Poutanen \& Fabian(1999)]{1999MNRAS.306L..31P} Poutanen, J.,
1373: \& Fabian, A.~C.\ 1999, \mnras, 306, L31
1374:
1375: \bibitem[Romano et al.(2002)]{2002ApJ...564..162R} Romano, P., Turner,
1376: T.~J., Mathur, S., \& George, I.~M.\ 2002, \apj, 564, 162
1377:
1378: \bibitem[Ross et al.(2002)]{2002MNRAS.336..315R} Ross, R.~R., Fabian,
1379: A.~C., \& Ballantyne, D.~R.\ 2002, \mnras, 336, 315
1380:
1381: \bibitem[Schurch \& Done(2006)]{2006MNRAS.tmp..762S} Schurch, N.~J.,
1382: \& Done, C.\ 2006, \mnras, 762
1383:
1384: \bibitem[Singh et al.(1985)]{1985ApJ...297..633S} Singh, K.~P.,
1385: Garmire, G.~P., \& Nousek, J.\ 1985, \apj, 297, 633
1386:
1387: \bibitem[Sobolewska \& Done(2005)]{2005AIPC..774..317S} Sobolewska,
1388: M., \& Done, C.\ 2005, AIP Conf.~Proc.~774: X-ray Diagnostics of
1389: Astrophysical Plasmas: Theory, Experiment, and Observation, 774, 317
1390:
1391: \bibitem[Stern et al.(1995)]{1995ApJ...449L..13S} Stern, B.~E.,
1392: Poutanen, J., Svensson, R., Sikora, M., \& Begelman, M.~C.\ 1995,
1393: \apjl, 449, L13
1394:
1395: \bibitem[Stobbart et al.(2006)]{2006MNRAS.368..397S} Stobbart, A.-M.,
1396: Roberts, T.~P., \& Wilms, J.\ 2006, \mnras, 368, 397
1397:
1398: \bibitem[Sunyaev \& Titarchuk(1980)]{1980A&A....86..121S} Sunyaev,
1399: R.~A., \& Titarchuk, L.~G.\ 1980, \aap, 86, 121
1400:
1401: \bibitem[Str{\" u}der et al.(2001)]{Struderetal01} Str{\" u}der, L.~et
1402: al.\ 2001, \aap, 365, L18
1403:
1404: \bibitem[Turner et al.(2001a)]{Turneretal01c} Turner, M.~J.~L.~et al.\
1405: 2001a, \aap, 365, L27
1406:
1407: \bibitem[Turner et al.(2001c)]{2001ApJ...548L..13T} Turner, T.~J., et
1408: al.\ 2001b, \apjl, 548, L13
1409:
1410: \bibitem[Turner et al.(2001b)]{2001ApJ...561..131T} Turner, T.~J.,
1411: Romano, P., George, I.~M., Edelson, R., Collier, S.~J., Mathur, S.,
1412: \& Peterson, B.~M.\ 2001, \apj, 561, 131
1413:
1414: \bibitem[Zdziarski et al.(1996)]{1996MNRAS.283..193Z} Zdziarski,
1415: A.~A., Johnson, W.~N., \& Magdziarz, P.\ 1996, \mnras, 283, 193
1416:
1417: \bibitem[Zdziarski \& Gierli{\'n}ski(2004)]{2004PThPS.155...99Z}
1418: Zdziarski, A.~A., \& Gierli{\'n}ski, M.\ 2004, Progress of
1419: Theoretical Physics Supplement, 155, 99
1420: \end{thebibliography}
1421:
1422:
1423:
1424: \clearpage
1425:
1426: \clearpage
1427:
1428:
1429: \clearpage
1430:
1431:
1432:
1433: % \begin{figure}
1434: % \centering
1435: % \includegraphics[width=8cm,angle=-90]{f11.ps}
1436: % \caption{EPIC-pn spectrum of Akn~564 in flux space and the
1437: % best-fit model consisting of optically thick thermal
1438: % Comptonization, power-law, four Laor lines, a narrow absorption
1439: % and an edge (see Table~\ref{t2}).}
1440: % \label{f11}
1441: % \end{figure}
1442:
1443: \clearpage
1444:
1445: \end{document}
1446:
1447: