A lower limit to the accretion disc radius in the low-luminosity AGNNGC 1052 derived from high-angular resolution data
A lower limit to the accretion disc radius in the low-luminosity AGNNGC 1052 derived from...
Reb, Lennart; Fernández-Ontiveros, Juan A; Prieto, M Almudena; Dolag, Klaus
2018-05-05 00:00:00
A lower limit to the accretion disc radius L1 A lower limit to the accretion disc radius in the low-luminosity AGN NGC 1052 derived from high-angular resolution data 1;2;3? 2;3 2;3 Lennart Reb, Juan A. Fernández-Ontiveros, M. Almudena Prieto, 1;4 and Klaus Dolag Universitäts-Sternwarte München, Scheinerstraße 1, D-81679 München, Germany Instituto de Astrofísica de Canarias, C/Vía Láctea, E-38205 La Laguna, Tenerife, Spain Universidad de La Laguna, Dept. Astrofísica, Avd. Astrofísico Fco. Sánchez s/n, E-38206 La Laguna, Tenerife, Spain Max-Planck Institute for Astrophysics, Karl-Schwarzschild-Straße 1, D-85741 Garching, Germany Accepted XXX. Received YYY; in original form ZZZ ABSTRACT We investigate the central sub-arcsec region of the low-luminosity active galactic nucleus NGC 1052, using a high-angular resolution dataset that covers 10 orders of magnitude in frequency. This allows us to infer the continuum emission within the innermost 17 pc around the black hole to be of non-thermal, synchrotron origin and to set a limit to the maximum contribution of a standard accretion disc. Assuming the canonical 10 per cent mass-light conversion eciency for the standard accretion disc, its inferred accretion power would be too low by one order of magnitude to account for the observed continuum luminosity. We thus introduce a truncated accretion disc and derive a truncation radius to mass-light conversion eciency relation, which we use to reconcile the inferred accretion power with the continuum luminosity. As a result we find that a truncated disc providing the necessary accretion power must be truncated at r & 26 r , consistent with the tr g inner radius derived from the observations of the Fe K line in the X-ray spectrum of this nucleus. This is the first time to derive a limit on the truncation radius of the accretion disc from high-angular resolution data only. Key words: accretion, accretion discs – black hole physics – galaxies: individual: NGC 1052 – galaxies: jets – galaxies: nuclei 1 INTRODUCTION cessible with adaptive optics facilities at the Very Large Telescope (VLT). The use of subarcsec resolution observations minimises the About 1=3 of all galaxies in the Local Universe harbour active contamination with host galaxy light in the critical infrared (IR) galactic nuclei (AGNs) with low accretion rates and modest lu- regime, allowing us to sample the nuclear continuum of the inner- minosities (Ho 2008). These are known as low-luminosity AGNs most few parsecs. X-ray observations of this source indicate a nu- (LLAGNs) and typically share a number of observational character- 22 23 2 clear hydrogen column density of N 10 10 cm (e.g. istics, which dier from their bright counterparts, Seyfert galaxies Hernández-García et al. 2014). A similar column density has been and quasars: they miss the blue bump, i.e. the footprint of the accre- derived from free-free absorption at radio wavelengths, suggesting tion disc, while Fe K line measurements indicate the absence of that ionised gas is responsible for the absorption seen in the X- optically thick material close to the black hole; the radio loudness, rays (e.g. Kadler et al. 2004b). Indeed, the silicate emission feature associated to outflows or (compact) jets, tends to increase towards at 9:7 m and our parsec-scale dust maps indicate very thin dust lower luminosities or Eddington ratios (Ho 1999, 2008; Nagar et al. around the nucleus with A < 1 mag (Tang et al. 2009; Nadolny et 2005). al. in prep.), rendering a free view down to the nucleus. The LLAGN in the nearby elliptical galaxy NGC 1052 is an ideal case to study the connection between the accretion flow and the jet emission at low luminosities. A twin-jet system emanates In this letter we apply a conservative, model independent ap- from the central black hole of M 1:55 10 M (Woo & Urry proach to investigate the power balance in the LLAGN NGC 1052: 2002), with its axis oriented close to the plane of the sky (e.g. we estimate the non-thermal luminosity from the sub-arcsec flux Baczko et al. 2016). It is located in the southern hemisphere at a distribution, which we compare with the predictions for the mass distance of D 18 Mpc (1 = 86 pc; Jensen et al. 2003), thus ac- flow through the standard accretion disc. Under the assumption that this mass flow provides all the (accretion) power required to supply the non-thermal emission, we investigate a possible truncation of E-mail: lennartreb@usm.lmu.de the thin accretion disc. Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly079/4993252 by Ed 'DeepDyve' Gillespie user on 08 June 2018 L2 L. Reb et al. λ (μm) cent on all the measured fluxes to account for possible variability 1m 10cm 1cm 1mm 0.5 2 10 100keV 100 10 1 0.1 (Maoz et al. 2005; Anderson & Ulvestad 2005). NGC 1052 Sub-arcsec resolution The flux distribution compiled in this way samples the nu- Low resolution clear region over 10 orders of magnitude in frequency at sub-arcsec Jet–disc model 00 00 scales ( 0: 2-0: 4), i.e. it represents the same physical scale around Thermalised plasma − 1 the black hole (. 35 pc) over nearly the entire accessible electro- Inverse Compton magnetic spectrum. − 2 Non-thermal electrons Standard disc − 3 − 4 3 RESULTS − 5 3.1 The nature of the continuum emission − 6 The continuum flux distribution shown in Fig. 1 reveals a constant − 7 flux from radio wavelengths up to the mid-IR. The spectral energy − 8 distribution (SED) in Fig. 2 shows that the mid-IR emission domi- 8 9 10 11 12 13 14 15 16 17 18 19 nates the bolometric luminosity below 100 keV. We isolate a strong log (ν / Hz) point-like source all the way from the mid-IR to the extreme UV, Figure 1. NGC 1052’s flux distribution. Black dots represent the sub-arcsec following an inverted power law shape over 2 orders of magnitude resolution fluxes (. 0: 4), grey spikes are low-angular resolution measure- in frequency ( 2:58 0:02, F 13 0:78 Jy, F / ; 1:3110 Hz ments. The errorbars of the dots (and the size of the spike symbols) span a Koljonen et al. 2015). There is no evidence for a noticeable amount relative logarithmic error of10 per cent. The compact jet–disc model rep- of dust extinction in the optical/UV, neither for the contribution of resentation (black thin line) with its individual components: synchrotron the blue bump, the footprint of the accretion disc. emission (blue dashed line) and synchrotron self Comptonisation (cyan Yu et al. (2011) modelled the continuum emission of NGC double-dot-dashed line) of the thermalised plasma, synchrotron emission 1052 with a model including a radiatively inecient accretion flow of the non-thermal electrons (green dash-dotted line), and the maximum (RIAF; see Yuan & Narayan 2014 for a review) plus a jet. The sub- contribution of a standard accretion disc (orange dotted line). arcsec resolution mid-IR to UV continuum departs largely from the RIAF prediction: the RIAF overestimates the optical/UV flux measurements by more than one order of magnitude and underesti- 2 HIGH-ANGULAR RESOLUTION FLUX mates the mid-IR flux by roughly one order of magnitude (see fig. DISTRIBUTION 3 in Fernández-Ontiveros et al. 2013). This work is based on sub-arcsec resolution flux measurements of The observed flat spectrum at radio wavelengths followed the nuclear region of NGC 1052 in multiple spectral ranges (see by a steep decay at higher frequencies (i.e. flat-inverted) is typ- Fernández-Ontiveros et al. 2012; Koljonen et al. 2015). In the IR, ically produced by the superposition of several self-absorbed observations are acquired with the VLT using the Nasmyth adaptive synchrotron components, as those produced in a compact jet optics system and the near-IR imager and spectrograph (NaCo), and (Blandford & Königl 1979). Yu et al. (2011) included a jet in their the VLT imager and spectrometer for mid-IR (VISIR; Asmus et al. model to account for the radio emission, but the model is incompat- 2014). In the optical/ultra-violet (UV), nuclear fluxes are measured ible with the IR and optical/UV SED, especially when the new sub- in archival Hubble Space Telescope (HST) images. All fluxes in the arcsec IR measurements are included. The reason is that their jet- IR and optical/UV range are measured using aperture photometry dominated model assumes 0:7 for the optically thin jet emis- centred at the unresolved central component and subtracting the lo- sion, while a much steeper index is required to fit the high-angular cal background from a surrounding annulus, and are corrected for resolution data. A possible explanation is that the steep IR to UV galactic reddening (A = 0:073 mag; Schlafly & Finkbeiner 2011). emission is associated to synchrotron emission by leptons in a ther- 00 00 The apertures are 0: 2, radius, except in VISIR ( 0: 4). How- malised plasma at the base of the jet (e.g. Shahbaz et al. 2013). For ever, VISIR data do not have a significant contribution from the instance, this configuration can be accomodated by a detailed com- host galaxy (Asmus et al. 2014). Both, high and low-angular reso- pact jet model (Fig. 1; Marko et al. 2005; Fernández-Ontiveros in lution measurements are in agreement above 10 m, and are thus prep.). considered as true nuclear emission. In the framework of the Unified Model for AGNs (Antonucci The continuum flux distribution is completed with radio mea- 1993; Urry & Padovani 1995) the IR emission is ascribed to ther- surements, mainly from Very Large Array (VLA) and Very Long mal emission from dust in a torus, due to the reprocessing of the Baseline Interferometry (VLBI), and with X-ray flux measure- UV radiation emitted by the accretion disc. In Fig. 2 the average ments, which have been collected after an extensive and careful Seyfert 2 sub-arcsec resolution template of Prieto et al. (2010) is search in the literature. Since the contribution of the stellar pop- overlaid, which is based on a physical region of similar size. This ulation is expected to be negligible at high energies, X-ray fluxes flux distribution is dominated by thermal emission from dust and are considered to be of nuclear origin above 2 keV and can be com- departs significantly from the shape of the IR continuum in NGC pared consistently with sub-arcsec measurements at other frequen- 1052. Therefore, the IR emission in NGC 1052 does not show a sig- cies. nificant contribution from a torus as observed in Seyfert 2 AGNs. The photometric error of the measurements is within a range The detection of 4:5 per cent polarisation degree in the mid- of 3 5 per cent, however, we assume a minimum error of 10 per IR continuum further argues against its thermal origin (Rieke et al. 1982). The shape of the observed continuum emission does not seem to be a product of thermal or RIAF emission. Instead the flat- The high and low-angular flux measurements are available in separate tables on the journal website alongside the paper. inverted spectrum is likely the signature of dominant synchrotron Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly079/4993252 by Ed 'DeepDyve' Gillespie user on 08 June 2018 log (F / Jy) A lower limit to the accretion disc radius L3 emission from a jet. Therefore, we assume that non-thermal mech- which can be considered as a conservative lower limit to the non- anisms primarily produce the overall continuum emission and con- thermal rest-frame luminosity. sequently the continuum luminosity. 4 TRUNCATION OF THE ACCRETION DISC 3.2 Accretion power and conservative continuum luminosity A similar procedure was carried out in a previous work for the The geometrically thin, optically thick standard accretion disc LLAGN M87 (Prieto et al. 2016). In M87, the maximum luminos- (Shakura & Sunyaev 1973) is used to describe the big blue bump 41 1 ity obtained for a standard accretion disc was 3:4 10 erg s observed at high accretion rates. It is an integral component in 42 1 and the observed continuum luminosity 2:7 10 erg s . While the framework of the Unified Model, required to feed the inner- in this case the rest-frame luminosity could be reconciled with most processes. We model the geometrically thin and optically the inferred accretion power through a standard disc, the jet ki- thick multi colour accretion disc following Mitsuda et al. (1984). netic power estimated from X-ray cavities was significantly higher The disc emission is independent of the micro-physics and depends 43 1 (Q & 10 erg s ; e.g. Allen et al. 2006). This comparison cast mainly on the inner edge radius and temperature. Throughout the doubt on the existence of ecient accretion in M87, as it is the letter we fix the disc outer edge to be at r = 2000 r , gravitational out g case for bright, radiatively ecient AGNs (e.g. Merloni 2016). radius r = MG c , where G is the gravitational constant and c In the case of NGC 1052 the accretion power P of a standard the speed of light, and the inclination to be i = 72 with respect to disc is even below the rest-frame luminosity L – it is too low RF the line of sight (Kadler et al. 2004a; Baczko et al. 2016). In sec- by one order of magnitude. Any hotter standard accretion disc, i.e. tion 4.2 we will discuss the impact of dierent inclinations. We fix with a higher accretion power, violates the spectral limits of the the inner edge boundary of the standard accretion disc to r = 6 r , b g measurements. Therefore, we investigate in the following the sce- the innermost stable circular orbit around a Schwarzschild black nario of a truncated accretion disc to explore the possibility of a hole (Shakura & Sunyaev 1973). The maximum contribution of a significantly higher accretion rate with a much lower radiative e- standard accretion disc matching the optical/UV fluxes imposed by ciency. the power law (Fig. 2; Koljonen et al. 2015) is shown in Figs. 1 & 2 40 1 and corresponds to a disc luminosity of L 4:0 10 erg s . SD We estimate the maximum accretion power P, i.e. the power 4.1 Radiative eciency of the truncated accretion disc gained by accretion through the disc, by (i) assuming the canonical We begin with the basic equation defining the spectrum of the clas- 10 per cent mass-light conversion eciency of the standard disc, ˙ ˙ sical Shakura & Sunyaev (1973) standard accretion disc to derive L = Mc , where M is the accretion rate and = 0:1, and (ii) SD SD SD ˙ the mass-light conversion eciency as a function of the truncation that all the accreted mass is available in form of energy, P = Mc . radius r . The luminosity dL(r) released at distance r from the cen- tr The latter means that we assume complete mass-energy conversion tral mass M per surface unit dA is for the accreted mass and neglect advection into the black hole. For further simplification we assume a constant M through the accre- p dL(r) 3 GM = M 1 r =r : (1) tion process, i.e. we neglect mass-loss due to radiation. This yields 3 dA 8 r 6 1 a mass accretion rate of M 7:1 10 M yr and an accretion p The term r =r lowers the luminosity originating from the in- 41 1 b power of P = 10 L 4:0 10 erg s . Under the assumption SD nermost regions of the disc, since gravitationally released energy of steady accretion, i.e. no extraction of additional rotational power is mechanically transported outwards before being converted into from the black hole spin, P must account for all subsequent pro- heat (Shakura & Sunyaev 1973). Integrating the luminosity radi- cesses, i.e. for the production of the non-thermal continuum lumi- ated within an annulus yields nosity and the jet’s kinetic power output. out To estimate the non-thermal continuum luminosity L we in- dL(r) RF L(r ; r ) = 4 r dr in out terpolate a selection of the measurements using a broken power dA in 2 0 13 law, as indicated in Fig. 2, integrate this flux distribution from p 6 B C7 2 r 3 6 1 1 B 1 1 C7 42 1 4 b 6 B C7 ˙ 6 B C7 radio to UV (5:26 10 erg s 2:70 10 L ), and sub- edd = MGM 6 B C7 : (2) 4 @ A5 3=2 3=2 40 1 2 r r 3 in out r r tract the maximum disc contribution (2:5 10 erg s ). We ex- out in cluded the X-ray regime from the integration, since reprocessing In order to derive the luminosity of a truncated accretion disc we set by the inverse Compton eect might result in a double-counting of the inner boundary radius and lower integration limit to the trunca- photons. In the integration we included the sub-arcsec resolution tion radius r = r = r . This is applying the zero-torque condition b in tr measurements and low-angular resolution measurements between at the truncation radius, which gives us a lower, hence a more con- 10 m and 3 mm, where no sub-arcsec resolution measurements servative luminosity (Quataert & Narayan 1999). We thus obtain are available. The low-angular resolution fluxes are consistent with MGM sub-arcsec fluxes at the edges of this range, suggesting that they are L(r ; r ) = [1 C(r =r )] ; (3) tr out out tr 2r dominated by nuclear emission. tr This inferred jet luminosity L might be boosted to a cer- 1 3=2 where C(x) = 3x 2x . Equalising Eq. 3 with the direct mass- tain extent. We calculate the maximum boosting factor for a jet light conversion L = Mc yields tr inclination of 72 on the basis of an optically thin spectral in- r r g g 2+ dex of 2:58 and a conical jet shape, L = L , where = [1 C(r =r )] : (4) J RF tr out tr 1 1 2r 2r tr tr the relativistic Doppler factor is defined as =
(1 cos i) (e.g. Urry & Padovani 1995). For the range of intrinsic velocities C(r =r ) becomes negligible for r r . Therefore, the mass- out tr tr out 0:21 . . 0:64 (Baczko et al. 2016) and a fixed inclination of light conversion eciency of the truncated accretion disc , i.e. tr 4:58 72 the maximum boosting factor 1:26 occurs for 0:31. the fraction of the energy gained through accretion that is radiated 42 1 Thus, we obtain a de-boosted luminosity of L 4:210 erg s , by the disc, scales inversely with the truncation radius. RF Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly079/4993252 by Ed 'DeepDyve' Gillespie user on 08 June 2018 L4 L. Reb et al. λ (μm) the inferred truncation radius to r & 30 r , but would not alter our 1m 10cm 1cm 1mm 0.5 2 10 100keV tr g 100 10 1 0.1 conclusions. NGC 1052 We assume in this work an inclination of 72 , which is the highest value of the overlap between the inclination ranges of Kadler et al. (2004b) and Baczko et al. (2016). For a higher incli- nation the flux constraints would allow for a hotter standard disc with a higher P. For a lower inclination the maximum boosting factor would increase, resulting in a lower L . To assess the im- Sub-arcsec res. RF Low resolution pact of the geometry in our estimate of r we explored the range tr Interpolation of accessible inclinations 64 . i . 87 of Baczko et al. (2016). In Power law particular, we compared the maximum possible accretion power of Average Seyfert 2 the standard disc (i = 87 ) with the minimum rest-frame luminos- Standard disc ity (de-boosted by the maximum possible boosting factor, which Hot standard disc Truncated disc occurs for i = 64 ). Even in this extreme case the accretion power would be still lower than the rest-frame luminosity by more than a 8 9 10 11 12 13 14 15 16 17 18 19 factor of 2, further arguing against a standard disc transporting the log (ν / Hz) accretion power. Figure 2. NGC 1052’s spectral energy distribution (SED). The measure- Possible alternatives to a truncated accretion disc are the ex- ments and the maximum contribution of the standard disc (orange dotted traction of rotational energy via the Blandford & Znajek (1977) line) correspond to those shown in Fig. 1. The broken power law interpola- mechanism or the accretion of mass through a geometrically thick tion connecting the selected measurements (light blue line), and the power and optically thin corona embedding the standard accretion disc law with = 2:58 (dark-green dotted line) are also shown. The Seyfert 2 (Churazov et al. 2001). These mechanisms could provide addi- sub-arcsec resolution template of Prieto et al. (2010), scaled to match the tional power to supply the continuum emission. However, the ex- sub-arcsec mid-IR flux, is overlaid (grey line). The hot standard disc (ma- traction of rotational energy cannot provide enough power accord- genta dashed line) and the truncated disc (red line) provide each the accre- ing to current simulations (e.g. Tchekhovskoy et al. 2011) and this tion power required to supply the continuum emission. process requires a high M to be ecient (e.g. Yuan & Narayan 2014), which does not seem to be the case for NGC 1052. The optically thin corona could in principle provide a significant mass flow. In that case 90 per cent of the mass flow would have to be 4.2 Constraints to the truncation radius accreted through the corona, which is unlikely. Previously, an alternative method has been used to A (hot) standard disc providing the accretion power required to sup- claim truncation of the accretion disc in other targets (e.g. ply the non-thermal continuum luminosity measured, P = L , cor- RF 5 1 ˙ Quataert & Narayan 1999; Yuan & Narayan 2004). The application responding to an accretion rate of M 7:3 10 M yr , exceeds of RIAF-based models to the continuum emission yields the mass the limits of optical/UV flux measurements (see Fig. 2). Therefore, flow of the RIAF. Since the spectrum of a standard disc providing the inner hotter part of such a disc must be truncated at a minimum this mass flow ( = 0:1) exceeds the limits of optical/UV flux radius to avoid exceeding the continuum flux observed. SD measurements, the disc must be truncated and is usually assumed To estimate the minimum truncation radius r we apply an ap- tr to transition into the RIAF at its inner edge. However, the pre- proximate method. First, following Quataert & Narayan (1999) the dicted mass flow relies on the detailed RIAF modelling and could disc luminosity L depends mainly on the inner edge temperature T in 2 4 be reduced, if additional processes, e.g. jet emission, contribute to and r , L 4r T . L is connected to P via the mass-light con- tr tr in the continuum emission. Therefore, this argument should be inter- version eciency . We adopt / r according to Eq. 4, which tr tr tr 3=4 preted carefully. sets for P = L a first relation between r and T , T / r . RF tr in in tr Second, the peak of the disc spectrum shifts linearly in frequency with T according to Wien’s displacement law, and the flux at the in 2 3 peak primarily depends on r and T as F / r T (insert- tr in peak tr in 4.3 Comparison with X-ray observations ing Wien’s displacement law into Planck’s law). We approximate the optical/UV continuum with the power law of Koljonen et al. Hard X-rays are likely emitted in the innermost region around (2015), where we use the (cold) standard disc as a reference point the black hole and allow to investigate the accretion flow. Above (T 7400 K). Equating this flux constraint with F defines a 10 keV, Brenneman et al. (2009) found no significant Compton re- in peak 2=(3+) second scaling relation between both parameters, T / r . flection, which is a signature of the inner region of the accretion in tr The intersection of both relations yields a lower limit for r and an disc. However, their detection of the relativistically broadened flu- tr upper limit for T . In other words, this is the solution for a trun- orescent Fe K line at 6:4 keV, which is usually associated to the in cated disc of P = L with a spectrum being ‘tangent’ to the power inner edge of an accretion disc, indicates an origin of the emis- RF law continuum observed. We find that this disc must be truncated sion within . 45 r of the black hole. A possible explanation is that at least at r & 26 r (T . 4400 K, . 2:3 per cent), the corre- Compton reflection might not be the dominant process in creating tr g in tr sponding spectrum is shown in Fig. 2 (red solid line). the X-ray emission in this source. Alternatively, the broadened iron The scaling of r with P depends on the spectral shape of the line component might not be connected to the inner edge of the ac- tr measurements, log r / (3 + )=(1 + 3) log P 0:64 log P. This cretion disc, but produced by optically thick material close to the tr means that in the present case a disc with a larger truncation radius base of the jets, which would be also the production site of hard could supply a substantially higher accretion power without being X-rays. In this case the inner edge of the disc could be located even detected in the SED. Including the X-rays (L 1:29 further outside. If the iron line traces the inner edge of the accretion 20 100 keV 42 1 10 erg s ; Beckmann et al. 2009) in L would slightly increase disc, our evidence for a truncated disc at r & 26 r would also be J tr g Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly079/4993252 by Ed 'DeepDyve' Gillespie user on 08 June 2018 log ( L / erg s ) A lower limit to the accretion disc radius L5 consistent with the limits of Brenneman et al. (2009) and the trun- Spanish Ministry of Economy and Competitiveness (MINECO) un- cation would help to explain the low Compton reflection observed. der grant number MEC-AYA2015-53753-P. KD acknowledges the support of the DFG Cluster of Excellence “Origin and Structure of the Universe” and the Transregio programme TR33 “The Dark Universe”. This research has made use of NASA’s Astrophysics Data System, the NASA/IPAC Infrared Science Archive, and of the 5 SUMMARY AND OUTLOOK NASA/IPAC Extragalactic Database (NED). The sub-arcsec resolution SED for the nucleus of NGC 1052, a pro- totypical LLAGN, does not show the big blue bump associated to the presence of an accretion disc. In this work, we took advantage of high-angular resolution observations to estimate the maximum REFERENCES possible accretion power of a standard accretion disc, correspond- Allen S. W., Dunn R. J. H., Fabian A. C., Taylor G. B., Reynolds C. S., 6 1 ing to a mass accretion rate of 7:110 M yr . In addition, the 2006, MNRAS, 372, 21 sub-arcsec measurements with the VLT reveal a SED dominated by Anderson J. M., Ulvestad J. S., 2005, ApJ, 627, 674 non-thermal emission, of synchrotron origin. From the sub-arcsec Antonucci R., 1993, ARA&A, 31, 473 continuum emission we derive a conservative estimate for the rest- Asmus D., Hönig S. F., Gandhi P., Smette A., Duschl W. J., 2014, MNRAS, frame luminosity. This is higher than the maximum accretion power 439, 1648 Baczko A.-K., et al., 2016, A&A, 593, A47 of a standard accretion disc by one order of magnitude – a standard Beckmann V., et al., 2009, A&A, 505, 417 accretion disc cannot provide the required power to supply the con- Blandford R. D., Königl A., 1979, ApJ, 232, 34 tinuum emission. Blandford R. D., Znajek R. L., 1977, MNRAS, 179, 433 Therefore, we explore the possibility of a truncated accretion Brenneman L. W., et al., 2009, ApJ, 698, 528 disc in order to reconcile the inferred accretion power with the rest- Churazov E., Gilfanov M., Revnivtsev M., 2001, MNRAS, 321, 759 frame luminosity, and find that the maximum mass-light conver- Fender R. P., Gallo E., Jonker P. G., 2003, MNRAS, 343, L99 sion eciency of the accretion disc scales inversely with the trun- Fernández-Ontiveros J. A., Prieto M. A., Acosta-Pulido J. A., Montes cation radius. For the first time, a truncation of the accretion disc M., 2012, in Journal of Physics Conference Series. p. 012006 is inferred from high-angular resolution data only. In order to pro- (arXiv:1206.0777), doi:10.1088/1742-6596/372/1/012006 vide the necessary accretion power in NGC 1052, a thin accretion Fernández-Ontiveros J. A., Almudena Prieto M., Acosta-Pulido J. A., 5 1 Marko S., González-Martín O., 2013, in European Physical Journal disc with an accretion rate of 7:3 10 M yr truncated at Web of Conferences. p. 04005, doi:10.1051/epjconf/20136104005 r & 26 r is required. This estimate is consistent with the results tr g Hernández-García L., González-Martín O., Masegosa J., Márquez I., 2014, of Brenneman et al. (2009) obtained from X-ray observations, who A&A, 569, A26 find no significant disc reflection in the hard X-ray spectrum and Ho L. C., 1999, ApJ, 516, 672 confine the broad Fe K line emission region to within . 45 r of Ho L. C., 2008, ARA&A, 46, 475 the black hole. Jensen J. B., Tonry J. L., Barris B. J., Thompson R. I., Liu M. C., Rieke The sub-arcsec resolution sampling of the LLAGN NGC 1052 M. J., Ajhar E. A., Blakeslee J. P., 2003, ApJ, 583, 712 reveals a continuum flux distribution that can be described by a Kadler M., Kerp J., Ros E., Falcke H., Pogge R. W., Zensus J. A., 2004a, broken power law plus an inverse Compton emission component. A&A, 420, 467 With these means, NGC 1052 does not represent a unique case, Kadler M., Ros E., Lobanov A. P., Falcke H., Zensus J. A., 2004b, A&A, 426, 481 instead this seems to be a common characteristic within the sub- Kolehmainen M., Done C., Díaz Trigo M., 2014, MNRAS, 437, 316 arcsec resolution LLAGN sample of Fernández-Ontiveros et al. Koljonen K. I. I., et al., 2015, ApJ, 814, 139 (2012). These LLAGNs with presumably jet-dominated spectra Maoz D., Nagar N. M., Falcke H., Wilson A. S., 2005, ApJ, 625, 699 (Mason et al. 2012; Koljonen et al. 2015) will be discussed in a Marko S., Nowak M. A., Wilms J., 2005, ApJ, 635, 1203 forthcoming paper (Fernández-Ontiveros et al. in prep.). Mason R. E., et al., 2012, AJ, 144, 11 Up to now the search for a direct link between the Fe K Merloni A., 2016, in Haardt F., Gorini V., Moschella U., Treves A., line broadening and the truncation radius in X-ray binaries has not Colpi M., eds, Lecture Notes in Physics, Berlin Springer Verlag revealed an unambiguous picture (e.g. Kolehmainen et al. 2014; Vol. 905, Lecture Notes in Physics, Berlin Springer Verlag. p. 101 Plant et al. 2014). The importance and eciency of RIAFs and (arXiv:1505.04940), doi:10.1007/978-3-319-19416-5_4 jets are subject to an ongoing discussion (e.g. Fender et al. 2003; Mitsuda K., et al., 1984, PASJ, 36, 741 Yuan & Narayan 2014; Koljonen et al. 2015). The influence of ad- Nagar N. M., Falcke H., Wilson A. S., 2005, A&A, 435, 521 Plant D. S., O’Brien K., Fender R. P., 2014, preprint, (arXiv:1411.7411) ditional mechanisms, e.g. accretion via a corona or advection into Prieto M. A., Reunanen J., Tristram K. R. W., Neumayer N., Fernandez- the black hole, is still not well understood. Therefore, the new ap- Ontiveros J. A., Orienti M., Meisenheimer K., 2010, MNRAS, 402, 724 proach presented in this letter can be applied to other LLAGNs to Prieto M. A., Fernández-Ontiveros J. A., Marko S., Espada D., González- shed some light on the changes in the internal structure of the ac- Martín O., 2016, MNRAS, 457, 3801 cretion disc that are predicted at low accretion rates. Quataert E., Narayan R., 1999, ApJ, 520, 298 Rieke G. H., Lebofsky M. J., Kemp J. C., 1982, ApJ, 252, L53 Schlafly E. F., Finkbeiner D. P., 2011, ApJ, 737, 103 Shahbaz T., Russell D. M., Zurita C., Casares J., Corral-Santana J. M., ACKNOWLEDGEMENTS Dhillon V. S., Marsh T. R., 2013, MNRAS, 434, 2696 Shakura N. I., Sunyaev R. A., 1973, A&A, 24, 337 The authors would like to thank Sera Marko for providing ac- Tang Y., Gu Q.-S., Huang J.-S., Wang Y.-P., 2009, MNRAS, 397, 1966 cess to the compact jet–disc model. The authors would like to Tchekhovskoy A., Narayan R., McKinney J. C., 2011, MNRAS, 418, L79 thank the referee for valuable comments which helped to improve Urry C. M., Padovani P., 1995, PASP, 107, 803 the manuscript. LR is grateful to Eugene Churazov for useful dis- Woo J.-H., Urry C. M., 2002, ApJ, 581, L5 cussions. JAFO and AP acknowledge financial support from the Yu Z., Yuan F., Ho L. C., 2011, ApJ, 726, 87 Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly079/4993252 by Ed 'DeepDyve' Gillespie user on 08 June 2018 L6 L. Reb et al. Yuan F., Narayan R., 2004, ApJ, 612, 724 Yuan F., Narayan R., 2014, ARA&A, 52, 529 Downloaded from https://academic.oup.com/mnrasl/advance-article-abstract/doi/10.1093/mnrasl/sly079/4993252 by Ed 'DeepDyve' Gillespie user on 08 June 2018
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.pngMonthly Notices of the Royal Astronomical Society LettersOxford University Presshttp://www.deepdyve.com/lp/oxford-university-press/a-lower-limit-to-the-accretion-disc-radius-in-the-low-luminosity-3waCilVMlH
A lower limit to the accretion disc radius in the low-luminosity AGNNGC 1052 derived from high-angular resolution data