First X-ray observations of Low-Power Compact Steep Spectrum Sources
M.Kunert-Bajraszewska, A.Labiano, A.Siemiginowska, M.Guainazzi
MMon. Not. R. Astron. Soc. , 1– ?? (2013) Printed 14 June 2018 (MN LaTEX style file v2.2) First X-ray observations of Low-Power Compact Steep SpectrumSources
M. Kunert-Bajraszewska , A. Labiano, , , A. Siemiginowska , M. Guainazzi, Toru´n Centre for Astronomy, Faculty of Physics, Astronomy and Informatics, NCU, Grudziacka 5, 87-100 Toru´n, Poland European Space Astronomy Centre of ESA, PO Box 78, Villanueva de la Ca˜nada, 28691, Madrid, Spain Harvard-Smithsonian Center for Astrophysics, Cambridge, MA 02138, USA
ABSTRACT
We report first X-ray
Chandra observations of a sample of seven low luminosity compact(LLC) sources. They belong to a class of young compact steep spectrum (CSS) radio sources.Four of them have been detected, the other three have upper limit estimations for X-ray flux,one CSS galaxy is associated with an X-ray cluster. We have used the new observationstogether with the observational data for known strong CSS and gigahertz-peaked spectrum(GPS) objects and large scale FR Is and FR IIs to study the relation between morphology, X-ray properties and excitation modes in radio-loud AGNs. We found that: (1) The low powerobjects fit well to the already established X-ray - radio luminosity correlation for AGNs andoccupy the space among, weaker in the X-rays, FR I objects. (2) The high excitation galaxies(HEG) and low excitation galaxies (LEG) occupy distinct locus in the radio / X-ray luminos-ity plane, notwithstanding their evolutionary stage. This is in agreement with the postulateddi ff erent origin of the X-ray emission in these two group of objects. (3) We have tested theAGN evolution models by comparing the radio / X-ray luminosity ratio with the size of thesources, and indirectly, with their age. We conclude that the division for two di ff erent X-rayemission modes, namely originate in the base of the relativistic jet (FR Is) or in the accretiondisk (FR IIs) is already present among the younger compact AGNs. (4) Finally, we found thatthe CSS sources are less obscured than the more compact GPSs in X-rays. However, the anti-correlation between X-ray column density and radio size does not hold for the whole sampleof GPS and CSS objects. Key words: galaxies-active, galaxies-evolution, X-rays-galaxies
We still know little about how radio galaxies are born and howthey subsequently evolve, but it is generally accepted that the GHzPeaked Spectrum (GPS) and Compact Steep Spectrum (CSS) radiosources are young, smaller versions of the large-scale powerful ra-dio sources (O’Dea et al. 1991; Fanti et al. 1990, 1995; Readheadet al. 1996a; O’Dea & Baum 1997). Recently, the High FrequencyPeakers have been added to the sequence, as possible progenitorsof GPS sources (e.g., Orienti et al. 2007, and references therein).The GPS and CSS sources are powerful but compact radiosources whose spectra are generally simple and convex with peaksnear 1 GHz and 100 MHz respectively. The GPS sources are con-tained within the extent of the optical narrow emission line region( (cid:46) (cid:46)
15 kpc, see O’Dea 1998, for a review).In the general scenario of the evolution of powerful radio-loud AGNs, GPS sources evolve into CSS sources and these intosupergalactic-size FR I or FR II objects (Fanaro ff &Riley 1974).The dynamic evolution of the double-lobed radio sources, char-acterized by the total extent of the source, advance speed of the hotspots and the dependence of the density distribution of the inter-stellar and intergalactic medium along the way of the propagatingjets and lobes, predicts the increase of the radio power with the lin-ear size of the source in the GPS and CSS phase until they reach the1-3 kpc size. Then the larger CSS objects should start to slowly de-crease their luminosity but the sharp radio power decrease is visibleonly in the FR I and FR II phase of evolution (Begelman & Cio ffi ff of the material supply to the central engine of thegalaxy, the sources begin their fading phase. They can come backon the main evolutionary sequence after the re-ignition of the radioactivity (e.g., Kozieł-Wierzbowska et al. 2012; Konar et al. 2012).However, population studies have drawn attention to the ex-istence of far too many compact sources compared to the numberof large-scale objects (O’Dea & Baum 1997). It has been proposedthen that some of the young radio-loud AGNs, namely the GPSand CSS sources, can be short-lived objects (Reynolds & Begel-man 1997; Czerny et al. 2009; Kawakatu et al. 2009b; Kunert-- c (cid:13) a r X i v : . [ a s t r o - ph . C O ] N ov M. Kunert-Bajraszewska et al.
Table 1.
Basic properties of the sample and X-ray models. Redshifts followed by a “p” are photometric. Fluxes are in 10 − erg s − cm − . Limits are 3 σ . Thenumbers in parentheses indicate the errors calculated as √ counts . Source RA(J2000) Dec (J2000) ID z Counts F . − F − N H , Gal N H Γ Chandra . − cm − ) (10 cm − ) Obs ID0810 +
077 08:13:23.76 07:34:05.80 Q 0.112 119 (11) 32 + − + − + . − . + . − . +
049 09:09:51.13 04:44:22.13 G 0.640p 3 a < < − b +
355 09:45:25.89 35:21:03.50 G 0.208 103 (10) 32 + . − . + − <
23 1.6 + . − . +
045 13:24:19.70 04:19:07.20 G 0.263 53 (7) 14 + . − . + . − . − + . − . +
390 15:43:49.49 38:56:01.40 G 0.553 0 a < < − b +
536 15:59:27.66 53:30:54.70 G 0.179 9 (3) 2.8 + . − . + . − . − b +
049 16:26:50.30 04:48:50.50 G 0.040p 4 a < < − b a means no detection, only upper limit for flux. b means Γ= Figure 1.
Chandra
X-ray (color) and MERLIN 1.6 GHz (contours) emission from 0810 +
077 (left panel) and 0942 +
355 (right panel). The radio maps are takenfrom (Kunert-Bajraszewska et al. 2010). We have used 0.15 pixel blocking for the X-ray images. The cross indicates the position of an optical counterparttaken from the SDSS. The radio contours increase by factor of 2, the first contour level corresponds to ≈ σ and amounts 1.7 mJy / beam (0810 + / beam (0942 + Bajraszewska et al. 2010) and that not one but a few evolutionarypaths exist (Marecki et al. 2003; Kunert-Bajraszewska et al. 2010;An & Baan 2012). Detection of several candidates for dying com-pact sources (Giroletti et al. 2005; Kunert-Bajraszewska et al. 2006,2010; Orienti et al. 2010) supports this view. The determining fac-tors for the further evolution of compact radio objects could occurat sub-galactic (or even nuclear) scales, or they could be related tothe radio jet-ISM interactions and evolution. Our previous studiessuggest that the evolutionary track could be related to the interac-tion, strength of the radio source, and excitation levels of the ion-ized gas (Kunert-Bajraszewska et al. 2010; Kunert-Bajraszewska &Labiano 2010), instead of the radio morphology of the young radiosource.The characteristics (size, radio power and young age) of GPSand CSS sources make them excellent probes of interaction (andtherefore evolution) of radio sources. Furthermore, they have notcompletely broken through the ISM, so these interactions are ex-pected to be more important than in the larger sources. Observa-tions of UV, HI and, especially, of the ionized gas in GPS and CSSsources suggest the presence of such interactions (Labiano 2008a;Labiano et al. 2008b; Holt et al. 2006; Labiano et al. 2005; Axon etal. 2000; de Vries et al. 1999, 1997; Gelderman & Whittle 1994).Additional clues on the evolution of compact GPS and CSSsources may come from the X-ray band, but still little is known about the nature of the X-ray emission in these young sources. The-oretical models predict strong X-ray emission from young radiosources, due to the recent triggering of the nuclear activity and / orthe expansion through the ISM (e.g., Siemiginowska et al. 2012;Siemiginowska 2009, and references therein). The Chandra and
XMM-Newton observations of GPS and CSS objects made so farhave focused on sources with high radio emission (e.g., Siemigi-nowska et al. 2008; Vink et al. 2006; Guainazzi et al. 2004, 2006;Tengstrand et al. 2009). These sources, when included in the L − keV versus L GHz diagram, group in the region occupied by power-ful FR II sources (Tengstrand et al. 2009). Therefore, the locationof GPS and CSS sources in the radio to X-ray luminosity diagramis consistent with them being powered by accretion, and thereforeevolving onto a track of constant X-ray, accretion-driven luminos-ity to FR IIs, as well as with the correlation between radio and X-ray luminosity observed in FR Is, which would point to a commonorigin for the emission in these two bands.In this paper, we present the first X-ray observations of lowpower radio sources, starting to fill the gap in the L − keV ver-sus L GHz diagram, and shedding some light on the origin of high-energy emission of young radio sources and their evolution. c (cid:13) , 1– ?? irst X-ray observations of Low-Power Compact Steep Spectrum Sources The current sample consists of 7 sources (0810 + + + + + + < GHz < × erg s − . The ra-dio and optical properties of the LLC were discussed and analyzedin Kunert-Bajraszewska et al. (2010) and Kunert-Bajraszewska &Labiano (2010) respectively.The 7 current sources form the so called pilot sample and wereselected to represent di ff erent stages of the radio source evolutionwithin the ISM: weak or undetected radio core and strong lobesor breaking up radio lobes with bright radio core, and linear sizesranging from 2 to 17 kpc . The sample was observed using
Chandra
ACIS-S3 with 1 / ∼ Chandra data were reduced using CIAO 4.5 (Fr-uscione et al. 2006) with the calibration files from CALDB 4.4.5.All our sources are contained within the FWHM of the PSF. Weused a circular extraction region for each source, with radius 2 (cid:48)(cid:48) ,which also contains all the radio emission. The background regionsconsist of four circular regions of radius 10 (cid:48)(cid:48) around the source. TheCIAO default tools were used to extract the spectra and associatedrmf and arf files. The total counts detected for each source are listedin Table 1.We used Sherpa (Freeman et al. 2001) to fit the spectra, usingan absorbed power-law in the 0.5-7 keV energy range: N ( E ) = e − N GalH σ ( E ) × e − N zobsH σ [ E (1 + z obs )] × AE − Γ (1)where N(E) is in photons cm − s − , A is the normalization at1 keV, Γ is the photon index of the power law, σ (E) and σ [E(1 + z obs )] are the absorption cross-sections (Morrison & McCammon1983; Wilms et al. 2000), and N GalH and N z obs H are the column den-sities of the Milky Way (Kalberla et al. 2005; Dickey & Lockman1990) and the source. The Galactic absorption was kept constantduring fitting. The second absorption component is assumed to beintrinsic to the quasar and located at the redshift of the source. Themodel was applied to all sources. However, 0907 + + +
049 do not have enough counts to produce a reasonablefit. The results are summarized in Table 1.We use H = Ω M = . , Ω Λ = .
73 (Spergel et al. 2003)throughout the paper.
We have observed a pilot sample of Low Luminosity Compact(LLC) sources (7 out of 44 objects) with
Chandra . Four of themhave been detected, the other three have upper limit estimations forX-ray flux (see Table 1). One of the objects, 1321 + Chandra
Table 2.
Spectral type and luminosities of the sources.Source log L − log L log L [OIII] Spectraltype0810 +
077 42.8 41.4 40.8 LEG0907 + < < +
355 42.9 41.4 42.0 HEG1321 +
045 42.3 41.6 40.3 LEG1542 + < + < + < − . Limits are 3 σ . Spectral type and radio morphologyaccording to Kunert-Bajraszewska et al. (2010); Kunert-Bajraszewska &Labiano (2010). Table 3.
Number of ionizing photons.Source Distance Log N H β Log N
Nuc N Nuc / N H β +
077 513.6 52.7 54.7 100.00942 +
355 1014.3 53.5 55.3 63.01321 +
045 1323.9 53.0 55.1 126.0Source name, luminosity distance in Mpc, N H β - number of photons / sneeded to ionize H β , N Nuc - number of ionizing photons / s produced by thenucleus, and the ratio between the last two. If the ratio is (cid:62)
1, the nucleusis producing enough photons to ionize H β , if the ratio is lower than one,another source of ionization, such as shocks, is required. ACIS-S images of two of the sources discussed here with the largestnumber of X-ray photons are shown in Fig. 1. We also overlayedthe radio MERLIN 1.6 GHz contours on the X-ray emission withthe indications of radio components.0810 +
077 is a quasar classified as LEG. Its radio morphol-ogy consist of three components: the weak central one (C) and twojets / lobes (E and W). The optical counterpart is coincident with thecomponent C and we suggested this could be a radio core (Kunert-Bajraszewska et al. 2010). However, this is based on observationsat only one radio frequency so it should be treated as tentative. Thebrightest part of the X-ray emission lies between the components Cand W. The potential o ff set between the centroid of the X-ray emis-sion and component C or W can be consistent with the astrometricuncertainty of Chandra .0942 +
355 is a galaxy classified as HEG, larger than 0810 + +
355 thebrightest part of the X-ray source is right in the center of the source,between the two jets.1558 +
536 is a galaxy classified as LEG with di ff use, double-like morphology (Kunert-Bajraszewska et al. 2010). Only ninecounts were detected in Chandra observations of this source andwe did not produced an image of it.The radio and optical properties of the whole sample of LLCsources have been discussed and analyzed by us (Kunert-Bajra-szewska et al. 2010; Kunert-Bajraszewska & Labiano 2010). Wesuggested that they can represent a population of short-lived ob-jects and undergo the CSS phase of activity many times beforethey become large scale FR I or FR II (Kunert-Bajraszewska et al.2010; Kunert-Bajraszewska & Labiano 2010). What is more, the c (cid:13) , 1– ?? M. Kunert-Bajraszewska et al. evolution of the radio source seems to be independent from its ra-dio morphology but rather determined by the properties of the cen-tral engine: strength, accretion mode, excitation level of the ionizedgas. In some objects the surrounding environment could be also animportant factor influencing the evolution (Cegłowski et al. 2013).Optically many of the LLC sources belong to the class of lowexcitation galaxies (LEGs), which are thought to be powered by theaccretion of hot gas (Hardcastle et al. 2007; Buttiglione et al. 2010)and can be progenitors of large scale LEGs (Kunert-Bajraszewska& Labiano 2010).
Labiano (2008a) found that compact AGNs show a strong corre-lation between [O III] λ ∼
10 times higher [O III] λ / or indicates di ff erences in the environmentof HEG and LEG objects.In the pilot sample of LLC sources observed with Chandra wehave found that two sources, 0810 +
077 and 0942 + + +
077 (LEG). If wecompare the OII / OIII y OIII / Hb of the detections with photoioniza-tion and shock models (MAPPINGS, Allen et al. 2008), 0810 + +
355 with 80%photoionization and 20% shocks (Kunert-Bajraszewska & Labiano2010).
We compared the number of ionizing photons produced by the nu-cleus of the source, with the number of photons needed to producethe observed emission line luminosity (see e.g. Wilson et al. 1988;Baum & Heckman 1989; Axon et al. 2000; O’Dea et al. 2000). As-suming radiative recombination under case B conditions, the num-ber of ionizing photons / s, N H β , needed to produce the observed H β luminosity L H β is: N H β = . × ( L H β / erg s − ) (2)We use the integrated [OIII] λ H β/ [ OIII ] λ = . ± . / s in the continuum, between frequen-cies ν and ν is given by: N Nuc = π D S ( α h ) − ( ν − α − ν − α ) (3)where D is the luminosity distance, the flux density spectrumis given by F ν = S ν − α (we adopt α =
1, e.g. O’Dea et al. 2000)and h is Planck’s constant. We are only interested in the photonswith enough energy to ionize Hydrogen, i.e. those between ν = Hz (912Å or 13.6eV) and ν = Hz (2 keV). Forour spectral index, α =
1, higher frequencies do not add a significantnumber of photons. Note that this analysis is subject to the caveat that the continuum emission may not be emitted isotropically, andthe extended nebulae may see a di ff erent luminosity than we do(e.g. Penston et al. 1990)The results are shown in Table 3. We find that the nucleus ap-parently produces enough ionizing photons to power the emissionline luminosity in 0810 +
077 and 0942 + / X-ray correlations
The
Chandra and
XMM-Newton studies of GPS / CSS sources per-formed so far show they are strong X-ray emitters (Guainazziet al. 2004, 2006; Vink et al. 2006; Siemiginowska et al. 2008;Tengstrand et al. 2009). The X-ray emission of CSS and GPS ob-jects is probably a result of a recent triggering of the nuclear activ-ity and can be characterized by an absorbed power law model withhigh ( > cm − ) column densities (Guainazzi et al. 2006; Vinket al. 2006; Siemiginowska et al. 2008; Tengstrand et al. 2009). Butthere are also several detections of X-ray morphology in these com-pact objects. Extended hot 0.5-1 keV interstellar medium (ISM) hasbeen detected in the case of two CSS sources, 3C303.1 (O’Dea etal. 2006) and 3C305 (Massaro et al. 2009), and it is interpreted asshock-heated environment gas. The X-ray jets in GPS sources andlarge scale X-ray emission associated with some of them have beenreported by (Siemiginowska et al. 2008). These objects are classi-fied as ’GPS sources with extended emission’ and discussed in theframe of the theory of intermittent radio activity (Stanghellini et al.2005). However, the radio structures of GPS and CSS sources aremuch smaller than the spatial resolution of the current X-ray in-struments in most cases, what prevents us from identification of theorigin of their X-ray emission. There are several theoretical predic-tions of X-ray emission from evolving radio sources: i) as thermalemission emitted by the ISM of the host galaxy shock heated by theexpanding radio structure (Heinz et al. 1998; O’Dea et al. 2006), orii) that produced in the accretion disk’s hot corona (Guainazzi etal. 2004, 2006; Vink et al. 2006; Siemiginowska et al. 2008), andfinally iii) as non-thermal radiation produced through IC scatteringof the local thermal radiation fields o ff the lobe electron population(Stawarz et al. 2008; Ostorero et al. 2010) or by mini shells (Kinoet al. 2013). Tengstrand et al. (2009) show that the radio versusX-ray luminosity plane can be a useful tool to derive constraintson the evolution of compact radio sources. Studies of compact ra-dio AGN so far have been biased towards high-luminosity objects( L GHz > erg s − ). In this Section we extend these studies tothe low-luminosity regime that our pilot Chandra study probes forthe first time.Our goal is to compare the X-ray properties of di ff erent groupsof radio objects, GPS, CSS and large-scale FR I and FR II sourcesas well as the X-ray properties of low and high power com-pact AGNs. For this purpose we have built the control sample ofGPS / CSS sources (Siemiginowska et al. 2008; Tengstrand et al.2009; Massaro et al. 2010, 2012) and FR I and FR II objects (Sam-bruna et al. 1999; Donato et al. 2004; Grandi et al. 2006; Evans etal. 2006; Balmaverde et al. 2006; Belsole et al. 2006; Hardcastle etal. 2006; Massaro et al. 2010, 2012) from results recently publishedin the literature. Our pilot sample of low luminosity CSS sourcesconsist of only seven objects. That is why we have also includedin it the low luminosity 3C305 described by Massaro et al. (2009).The total number of GPS / CSS sources is 40 objects, and FR I and c (cid:13) , 1– ?? irst X-ray observations of Low-Power Compact Steep Spectrum Sources l og ( L - k e V / e r g / s ) log (L
365 MHz / erg/s) FRIFRIIstrong GPS/CSS_G strong_GPS/CSS_QSOs weak_CSSs 3839404142
38 40 42 44 46 l og ( L - k e V / e r g / s ) log (L / erg/s) FRIFRIIstrong GPS/CSS_Gstrong_GPS/CSS_QSOsweak_CSSs Figure 2.
Luminosity diagrams for AGNs: 2-10 keV - 365 MHz (left) and 2-10 keV - 5 GHz (right). Open red circles indicate FR II sources and black opensquares - FR Is. Strong GPS and CSS galaxies and quasars were plotted separately as black circles and blue squares respectively. Weak CSS sources areindicated with red squares.
FR IIs - 34 and 85 sources, respectively. The samples are, how-ever, biased in terms of their redshift distribution. The GPS / CSSand FR II samples are well matched in the redshift, but the FR Isare generally at lower redshift. There are 6 GPS / CSS objects withredshift in the range 1 > z <
2. All other sources from all groupshave redshit z < / CSS sources is also unresolved in X-rays. The ex-ceptions from the above-stated rule are a few GPS sources with ex-tended structures (Stanghellini et al. 2005). In the case of them theradio and X-rays values used in this paper refers to their milliarc-seconds VLBA structures as reported in Tengstrand et al. (2009)and Siemiginowska et al. (2008). / X-ray luminosity plane
We have compared the X-ray luminosity of the sources from thepilot sample with their radio properties at 5 GHz and 365 MHz(Fig. 2). We have included also the control sample of GPS / CSSand FR I and FR II objects as described above.The low power objects fit well to the already established X-ray - radio luminosity correlation for AGNs and occupy the spaceamong, weaker in the X-rays, FR I objects. This trend is visible onboth plots, X-ray vs. 356 MHz and 5 GHz (Fig.2), and is indepen-dent of radio frequency. However, the 356 MHz radio luminosityversus X-ray luminosity plot shows larger scatter among observ-able data than in the case of 5 GHz luminosity. This is caused bythe fact that the X-ray emission is mostly associated with the com-pact central regions of AGNs while the low frequency flux densityis dominated by the extended radio structures. As has been alsoshown by Hardcastle et al. (1999) much of the dispersion in 5 GHzluminosity originate in beaming. Future X-ray observations of thewhole sample of LLC sources would give us a definitive informa-tion about their place on the radio / X-ray luminosity plane.We have then plotted all groups of AGNs on the 5 GHz / X-ray luminosity plane (Fig. 3) with a division for high excitation l og ( L - k e V / e r g / s ) log (L / erg/s) FRI_LEGFRII_HEG/BLOGPS/CSS_LEGFRII_LEGGPS/CSS_HEG Figure 3. / BLO sources and GPS / CSS HEG objectsare indicated as open red circles and red circles respectively. FR II LEGs areindicate with black crosses and FR I LEGs and GPS / CSS LEGs with openand black squares respectively.
Table 4.
Correlation and Regression Analysis for Fig. 3Sample N r-Pearson Linear Regressioncoe ffi cient Slope InterceptHEG 82 0.68 0 . ± .
09 11 . ± . . ± .
12 5 . ± . galaxies (HEG) and low excitation galaxies (LEG). We took theoptical identification from Buttiglione et al. (2010) in the case ofFR I and FR II and indicated them as LEG and HEG / BLO. Ac-cording to Buttiglione et al. (2010) the broad line objects (BLO)can be considered as members of the HEG class. Identifications ofGPS / CSS objects were taken from (Kunert-Bajraszewska & Labi-ano 2010) (see also Table 2 in this paper) and Table A1. We haveonly four LEGs among the GPS / CSS class and actually all of themhave been classified as CSS sources. HEGs are found among strongGPS and CSS objects. The HEG / LEG plot confirms what we have c (cid:13) , 1– ?? M. Kunert-Bajraszewska et al. previously found in Kunert-Bajraszewska & Labiano (2010). TheHEG and LEG AGNs group in two di ff erent parts of the plot.A Pearson correlation analysis applied to both sub-samples re-vealed a significant X-ray / radio correlation (Table 4). In the radioversus X-ray luminosity plane (Fig. 2), objects with a di ff erent mor-phology are aligned along the same correlation, with an increasingfraction of large-scale FR II morphologies at higher luminosities.Compact sources are well aligned along this correlation, with weakCSS (strong CSS / GPS) closer to the parameter space occupied byFR I (FR II). However, the ionization mechanism seems to discrim-inate more neatly radio sources in this plane. Low versus high ion-ization sources occupy distinct locus in this plane, notwithstand-ing their evolutionary stage. This evidence agrees with a scenariowhereby the X-ray emission in large-scale HEG sources is dom-inated by spectral components due to (obscured) accretion as op-posed to LEG objects where the X-ray emission should be domi-nated by non-thermal synchrotron jets (Hardcastle et al. 2009; An-tonucci 2012; Son at al. 2012). As seen on the radio / X-ray lumi-nosity diagram (Fig. 3), there are two branches, each being drivenby di ff erent excitation mode and each containing compact sources.The FR morphology, as well as the GPS and CSS division, seemsto be independent on the excitation modes (Buttiglione et al. 2010;Kunert-Bajraszewska & Labiano 2010; Gendre et al. 2013). / X-ray luminosity ratio
Another test for the AGN evolution models is a comparison of ra-dio to the X-ray luminosity ratio with the size of the sources, andindirectly, with their age (Fig. 4). The long-term evolution of ex-tragalactic radio sources have been investigated by a number of au-thors in di ff erent ways: i) as a variation of the radio power versusthe total linear size (O’Dea & Baum 1997; Kunert-Bajraszewskaet al. 2010; An & Baan 2012), or ii) dynamic evolution of FR II-like double radio sources characterized by the advance speed ofthe hot spots, total extent of the source and depending on the den-sity distribution in the host galaxy along the path of the jets andlobes (Begelman & Cio ffi / Compact Sym-metric Object (CSO) stage, the radio power of the sources increaseswith time and the source size. The increase rate of radio power di-minishes in the transition region (1-3 kpc from the center of thehost galaxy) where the balance between adiabatic losses and syn-chrotron losses have been achieved. After this short period the ra-dio power of CSS sources starts to slowly decreases with the sourcesize. The sharp decrease in the radio power versus the total extentof the source occurs only in the large FR I and FR II objects withthe FR Is being below the luminosity threshold, L MHz ∼ . W H z − sr − , on the radio power / linear size plane. If the X-rayemission in radio-loud AGNs is due to accretion only, the evolu-tion of the radio and X-ray wavebands could be totally decoupledand the radio / X-ray luminosity ratio should reproduced the radiopower evolution with the linear size of the source. The Fig. 4 showsthat the above assumption is not true not only in the case of largeFR Is and FR IIs but probably also in the case of young GPS andCCS sources. The less radio powerful FR Is have the radio / X-rayluminosity ratio higher than many FR IIs what may imply higherX-ray luminosity decrease with radio power in FR Is than FR IIs.This can be explained by the idea that the X-ray emission in FR Is originates from the base of a relativistic jet (Evans et al. 2006) andare thought to be synchrotron emission (Sambruna et al. 2004; Wor-rall et al. 2009). However, the X-ray emission of the FR IIs comesmostly from the obscured X-ray component probably associatedwith the accretion and in less part from the relativistic jet producedin an inverse Compton process (Balmaverde et al. 2006; Belsole etal. 2006; Evans et al. 2006). When incorporating a di ff erent divi-sion among large scale objects we notice that, on average, the lowexcitation radio galaxies (LERG) and narrow line radio galaxies(NLRG) have higher radio to X-ray luminosity ratio than quasars(Q) and broad line radio galaxies (BLRG). According to Hardcastleet al. (2009) the common correlation of FR II NLRGs, LERGs andFR Is indicates that the X-ray and radio emission comes from thesame jet-related component. The interpretation of the place of GPSand CSS sources on the radio / X-ray luminosity ratio versus linearsize plane is even more di ffi cult because of the large scatter of theobservable values. As we have already mentioned the beaming candisrupt the radio / X-ray correlations. However, at least in the caseof compact steep spectrum objects this e ff ect should be small (Wuet al. 2013). We then suggest that what we observe in the groupof young GPS and CSS sources is a mix of two di ff erent types ofX-ray / radio relation. We conclude that at some radio power levelthe compact AGNs starts to resemble the FR Is, where the X-rayemission is a synchrotron type associated with the jet. It has beenalready proposed by the SED modelling of two strong CSS objectsthat their X-ray emission can be a sum of X-ray emission fromthe accretion disk and non-thermal X-ray emission from the par-sec scale radio jet (Kunert-Bajraszewska et al. 2009; Migliori et al.2012).Recently, Stawarz et al. (2008) and Ostorero et al. (2010)have discussed an alternative evolutionary model for GPS sources,which predicts the dependency of the broadband Spectral EnergyDistribution on the source linear size. In their model high-energyemission is produced by upscattering of various photon fields bythe lobes’ electrons of the < / X-ray luminosity ratio and linear size ofthe sources. H - linear size relation Finally, we have drawn a relation between the measured columndensity and the total extent of the radio source for GPS and CSSsources (Fig. 5). It has been already reported (Pihlstr¨om et al. 2003;Vermeulen et al. 2003) that the small sources ( < > / CSO objects evolve in a disk distribution of gaswith a power-law radial density dependence. The same explana-tion could lay behind the (tentative) anti-correlation between X-raycolumn density and radio size in GPS galaxies (Tengstrand et al.2009). Ostorero et al. (2010) reported a positive correlation be-tween the radio and X-ray hydrogen column densities what canpoints toward the cospatiality of the radio and X-ray emission re-gions. We extended the discussion about the relation of the N H ver-sus the linear size to the CSS sources. We have noticed that the N H value of the CSS sources is on average lower than that of GPS ob-jects. However, the correlation between the X-ray column densityand radio size does not hold when including larger CSS objects inthe sample of small and young AGNs. c (cid:13) , 1– ?? irst X-ray observations of Low-Power Compact Steep Spectrum Sources -4-3-2 -1012 -3 -2 -1 0 1 2 3 4 l og ( L G H z / L - k e V ) log (linear size / kpc) GPS/CSOCSSFRII
FRI -4 -3-2-1012 -3 -2 -1 0 1 2 3 4 l og ( L G H z / L - k e V ) log (linear size / kpc) GPS/CSOCSSBLRGNLRGQLERG Figure 4.
The 5 GHz luminosity / -2,5 -2 -1,5 -1 -0,5 0 0,5 1 1,5 2 2,5 l og ( N H ( c m - )) log (linear size / kpc) GPS/CSOCSS Figure 5.
X-ray column density versus linear size. GPS and CSS sourcesare indicated as black squares and blue circles respectively.
In this paper we presented the X-ray
Chandra observations ofa pilot sample of low luminosity compact steep spectrum (CSS)sources. Four of them have been detected, the other three have up-per limits estimations for the X-ray flux. Only for two CSS objectswe were able to estimate the X-ray column density which is ofthe order of 10 cm − . We then expanded the sample of compactAGNs with other GPS / CSS sources with X-ray detections foundin the literature and used it, together with a sample of large FR Iand FR II sources, to determine the nature of the relation betweenmorphology, X-ray properties and excitation modes in radio-loudAGNs. We found the following results: • We have compared the X-ray luminosity of the radio sourcesfrom all above mentioned groups with their radio properties. Thelarge-scale FR II sources and strong GPS and CSS objects settleat higher X-ray and radio luminosities. While the low power CSSsoccupy the space among, weaker in the X-rays, FR I objects. Thistrend is visible independently of radio frequency. • The HEG and LEG sources occupy distinct locus in theradio / X-ray luminosity plane, notwithstanding their evolutionarystage. This is in agreement with the postulated di ff erent origin of the X-ray emission in low and high ionization objects. Compactsources can be found in both excitation modes driven branches. • The less radio powerful FR Is have higher radio / X-ray lu-minosity ratio than many FR IIs what may imply higher X-ray lu-minosity decrease with radio power in FR Is than FR IIs. This is inagreement with the previous findings saying that the X-ray emis-sion in FR Is originates from the base of a relativistic jet while theX-ray emission of FR IIs has accretion origin. The same can betrue in the case of smaller radio AGNs, namely the GPS and CSSsources. The result of this study hints toward the fact that belowsome radio power level the compact GPS and CSS sources start toresemble the FR Is, or to be more specific, the LERG and NLRGobjects. • The X-ray hydrogen column density of the CSS sources is onaverage lower than that of GPS objects. But the correlation betweenthe X-ray column density and radio size does not hold for the wholesample of GPS and CSS objects.
ACKNOWLEDGEMENTS
AL wishes to thank CfA for their hospitality and support during thevisit.This research has made use of data obtained by the ChandraX-ray Observatory, and
Chandra
X-ray Center (CXC) in the appli-cation packages CIAO, ChIPS, and Sherpa. This research is fundedin part by (NASA) contract NAS8-03060. Partial support for thiswork was provided by the
Chandra grants GO1-12124X and GO1-12145X.This research has made use of NASA’s Astrophysics Data Sys-tem Bibliographic Services and of the NASA / IPAC ExtragalacticDatabase (NED) which is operated by the Jet Propulsion Labora-tory, California Institute of Technology, under contract with the Na-tional Aeronautics and Space Administration.This publication makes use of data products from the SDSS.Funding for the SDSS and SDSS-II has been provided by the Al-fred P. Sloan Foundation, the Participating Institutions, the Na-tional Science Foundation, the U.S. Department of Energy, theNational Aeronautics and Space Administration, the JapaneseMonbukagakusho, the Max Planck Society, and the Higher Ed- c (cid:13) , 1– ?? M. Kunert-Bajraszewska et al. ucation Funding Council for England. The SDSS Web Site ishttp: // / .The SDSS is managed by the Astrophysical Research Con-sortium for the Participating Institutions. The Participating Institu-tions are the American Museum of Natural History, AstrophysicalInstitute Potsdam, University of Basel, University of Cambridge,Case Western Reserve University, University of Chicago, DrexelUniversity, Fermilab, the Institute for Advanced Study, the JapanParticipation Group, Johns Hopkins University, the Joint Institutefor Nuclear Astrophysics, the Kavli Institute for Particle Astro-physics and Cosmology, the Korean Scientist Group, the ChineseAcademy of Sciences (LAMOST), Los Alamos National Labora-tory, the Max-Planck-Institute for Astronomy (MPIA), the Max-Planck-Institute for Astrophysics (MPA), New Mexico State Uni-versity, Ohio State University, University of Pittsburgh, Universityof Portsmouth, Princeton University, the United States Naval Ob-servatory, and the University of Washington. REFERENCES
Allen, M. G., Groves, B. A., Dopita, M. A., Sutherland, R. S.,Kewley, L. J., 2008, ApJS, 178, 20.An, T., Baan, W. A., 2012, ApJ, 760, 77Antonucci, R., 2012, A&AT, 27, 557Axon, D. J., Capetti, A., Fanti, R., Morganti, R., Robinson, A.,Spencer, R., 2000, AJ, 120, 2284Balmaverde, B., Capetti, A., & Grandi, P. 2006, A&A, 451, 35Baum, S. A. & Heckman, T., 1989, ApJ, 336, 702Begelman, M. C. & Cio ffi , D. F. 1989, ApJ, 345, L21Begelman, M. C. 1996, in Proceedings of the Greenbank Work-shop, Cygnus A A Study of a Radio Galaxy, ed. C. L. Carilli, &D. E. Harris (Cambridge: CUP), 209Belsole, E., Worrall, D. M., & Hardcastle, M. J. 2006, MNRAS,366, 339Buttiglione S., Capetti, A., Celotti, A., Axon, D.J., Chiaberge, M.,Macchetto, F.D., Sparks, W.B., 2010, A&A, 509, 6Carvalho, J. C., 1994, A&A, 292, 392Carvalho, J. C., 1998, A&A, 329, 845Cegłowski, M., Gawro´nski, M. P., Kunert-Bajraszewska, M.,2013, A&A, 557, 75Czerny, B., Siemiginowska, A., Janiuk, A., Nikiel-Wroczy´nski,B., Stawarz, Ł., 2009, ApJ, 698, 840Dickey, J. M. & Lockman, F. J., 1990, ARA&A, 28, 215Donato, D., Sambruna, R. M., & Gliozzi, M. 2004, ApJ, 617, 915Evans, D. A., Worrall, D. M., Hardcastle, M. J., Kraft, R. P., &Birkinshaw, M. 2006, ApJ, 642, 96Fanaro ff , B. L., & Riley, J. M.: 1974, MNRAS 167, 31Fanti, R., Fanti, C., Schilizzi, R., et al., 1990, A&A, 231, 333Fanti, C., Fanti, R., Dallacasa, D., et al. 1995, A&A 302, 317Freeman, P., Doe, S., & Siemiginowska, A. 2001, Proc. SPIE,4477, 76Fruscione, A., et al., 2006, SPIE, 6270Gelderman, R. & Whittle, M., ApJS, 1994, 91, 491Gendre, M. A., Best, P. N., Wall, J. V., Ker, L. M., 2013, MNRAS,arXiv:1301.1526Giroletti, M., Giovannini, G., & Taylor, G. B.: 2005, A&A 441,89Grandi, P., Malaguti, G., & Fiocchi, M. 2006, ApJ, 642, 113Gregory, P. C., Scott, W. K., Douglas, K., & Condon, J. J. 1996,ApJS, 103, 427 Guainazzi, M., Siemiginowska, A., Rodriquez-Pascual, P.,Stanghellini, C., 2004, A&A, 421,461Guainazzi, M., Siemiginowska, A., Stanghellini, C., Grandi, P.,Piconcelli, E., Azubike Ugwoke, C., 2006, A&A, 446, 87Hardcastle, M. J. & Worrall, D. M., 1999, MNRAS, 309, 969Hardcastle, M. J., Evans, D. A., & Croston, J. H. 2006, MNRAS,370, 1893Hardcastle, M. J., Evans, D. A., Croston, J. H., 2007, MNRAS,376, 1849Hardcastle, M. J., Evans, D. A., Croston, J. H., 2009, MNRAS,396, 1929Heinz, S., Reynolds, C. S., Begelman, M. C., 1998, ApJ, 501, 126Holt, J., Tadhunter, C. N., & Morganti, R., 2006, AN, 327, 147Kaiser, C. R. & Best, P. N. 2007, MNRAS, 381, 1548Kaiser, C. R. & Alexander, P., 1997, MNRAS, 286, 215Kalberla, P. M. W. et al., 2005, A&A, 440, 775Kawakatu, N. & Kino, M. 2006, MNRAS, 370, 1513Kawakatu, N, Kino, M., Nagai, H. 2009, ApJ, 697, 173LKawakatu, N, Nagai, H., Kino, M. 2009, AN, 330, 283Kino, M., Ito, H., Kawakatu, N., Orienti, M. 2013, ApJ, 764, 134Kino, M. & Kawakatu, N. 2005, MNRAS, 364, 659Konar, C., Hardcastle, M. J., Jamrozy, M., Croston, J. H., Nandi,S., 2012, MNRAS, 424, 1061Kozieł-Wierzbowska, D., Jamrozy, M., Zola, S., Stachowski, G.,Ku´zmicz, A., 2012, MNRAS, 422, 1546Kunert-Bajraszewska, M., Marecki, A., & Thomasson, P., 2006,A&A 450, 945Kunert-Bajraszewska, M. & Thomasson, P., 2009, AN, 330, 210Kunert-Bajraszewska, M., Siemiginowska, A., KatarzyÅski, K.,Janiuk, A., 2009, ApJ, 705, 1356Kunert-Bajraszewska, M.,Gawro´nski, M. P., Labiano, A.,Siemiginowska, A., 2010a, MNRAS, 408, 2261Kunert-Bajraszewska, M., Labiano, A., 2010b, MNRAS, 408,2279Kunert-Bajraszewska, M., Siemiginowska, A., Labiano, A., 2013,ApJ, 772, L7Labiano, A., O’Dea, C.P., Gelderman, R., et al., 2005, A&A, 436,493Labiano, A., 2008a, A&A, 488, 59Labiano, A., O’Dea, C. P., Barthel, P. D., et al. 2008b, A&A, 477,491Marecki, A., Spencer, R. E., & Kunert, M.: 2003, PASA 20, 46Massaro, F., et al., 2009, ApJL, 692, L123Massaro, F., et al., 2010, ApJ, 714, 589Massaro, F., et al., 2012, ApJSS, 203, 31Migliori, G., Siemiginowska, A., Celotti, A., 2012, ApJ, 749, 107Morrison, R., & McCammon, D., 1983, ApJ, 270, 119O’Dea, C. P., Baum, S. A., Stanghellini, C., 1991, ApJ, 380, 66O’Dea, C. P., & Baum, S. A., 1997, AJ 113, 148O’Dea, C. P., 1998, PASP, 110, 493O’Dea, C. P., De Vries, W. H., Worrall, D. M., Baum, S. A.,Koekemoer, A., 2000, AJ, 119, 478O’Dea, C. P., Mu, B., Worrall, D. M., Kastner, J., Baum, S., deVries, W. H., 2006, ApJ, 653, 1115Orienti, M., Dallacasa, D., Stanghellini, C., 2007, A&A, 475, 813Orienti, M., Murgia, M., Dallacasa, D., 2010, MNRAS, 402, 1892Ostorero, L., et al., 2010, ApJ, 715, 1071Penston, M. V., et al., 1990, A&A, 236, 53Pihlstrm, Y. M., Conway, J. E., & Vermeulen, R. C. 2003, A&A,404, 871Readhead, A. C. S., Taylor, G. B., Xu, W., et al., 1996a, ApJ 460,612 c (cid:13) , 1– ?? irst X-ray observations of Low-Power Compact Steep Spectrum Sources Reynolds, C. S., & Begelman, M. C., 1997, ApJ, 487, L135Sambruna, R., Eracleous, M., & Mushotzky, R. 1999, ApJ, 526,60Sambruna, R. M., Gambill, J. K., & Maraschi, L. et al. 2004, ApJ,608, 698Siemiginowska, A., Stawarz, Ł., Cheung, C. C., et al. 2012, ApJ,750, 124Siemiginowska, A., 2009, AN, 330, 264Siemiginowska, A., LaMassa, S., Aldcroft, T. L., Bechtold, Jill;Elvis, M., 2008, ApJ, 684, 811Snellen, I. A. G., Schilizzi, R. T., Miley, G. K., de Bruyn, A. G.,Bremer, M. N., R¨ottgering, H. J. A., 2000, MNRAS, 319, 445Son, D., et al., 2012, ApJ, 757, 140Spergel, D. N., et al., 2003, ApJS, 148, 175Stanghellini, C., O’Dea, C. P., Dallacasa, D., et al. 2005, A&A,443, 891Stawarz, Ł., Ostorero, L., Begelman, M. C., Moderski, R.,Kataoka, J., Wagner, S., 2008, ApJ, 680, 911Tengstrand, O., Guainazzi, M., Siemiginowska, A., FonsecaBonilla, N., Labiano, A., Worrall, D. M., Grandi, P., Piconcelli,E., 2009, A&A, 501, 89Vermeulen, R. C., Pihlstrm, Y. M., Tschager, W., et al. 2003,A&A, 404, 861Vink, J., et al., 2006, MNRAS, 367, 928Worrall, D. M. 2009, A&AR, 17, 1de Vries, W. H., O’Dea, C. P., Baum, S. A., Barthel, P. D. 1999,ApJ, 526, 27.de Vries, W. H., O’Dea, C. P., Baum, S. A., Sparks, W. B., Biretta,J., de Ko ff , S., Golombek, D., Lehnert, M. D., Macchetto, F.,McCarthy, P., Miley, G. K., ApJS 110, 191.White, R.L., Becker, R.H., Helfand, D. J., Gregg, M.D., 1997,ApJ, 475, 479Wilms, J., Allen, A., McCray, R., 2000, ApJ, 542, 914Wilson, A. S., Ward, M. J. & Hani ff , C. A., 1988, ApJ, 334, 121Wu, F., et al., 2013, A&A, 550, 113De Young, D. S., 1993, ApJ, 402, 95De Young, D. S., 1997, ApJL, 490, 55 APPENDIX A: SPECTROSCOPIC CLASSIFICATION OFTHE OTHER CSS / GPS SOURCES WITH X-RAY DATA
The spectroscopic data are available for 6 CSS sources from thesamples of CSS / GPS sources of Tengstrand et al. (2009) andSiemiginowska et al. (2008) (Table A1). The HEG / LEG classifi-cation was based on the line ratios observed in the SDSS spectraaccording to the description by Kunert-Bajraszewska & Labiano(2010). c (cid:13) , 1– ?? M. Kunert-Bajraszewska et al.
Table A1.
Emission lines measurements and spectroscopic classification of compact sources. The columns show the following : (1) source name, (2)-(3)coordinates, (4) redshift, (5)-(7) flux in 10 − erg s − cm − , O II - total flux of the O II doublet, (8)-(10) log of luminosities; luminosities are in erg s − , (11)spectroscopic classification. Wavelengths are in Angstroms. ’a’ means the classification is based on the [O II] λλ / [O III] λ z [O II] [O III] H β log L − log L log L [OIII] Spectralhh mm ss dd mm ss λ λ λ λ +
31 07:41:10.7 31:12:00 0.63 445.7 1490.6 3675.6 44.9 44.3 43.4 HEG a Q1250 +
568 12:52:26.3 56:34:20 0.32 1066.0 3564.8 1653.3 44.2 43.2 43.1 HEG b +
125 13:47:33.3 12:17:24 0.12 1038.9 4682.3 623.9 43.8 42.7 42.2 HEGOQ +
208 14:07:00.4 28:27:15 0.07 37.8 1068.0 18749.0 > a + +
26 16:09:13.3 26:41:29 0.47 63.6 200.0 44.6 44.5 43.8 42.2 HEGc (cid:13) , 1–, 1–