THE ASTROPHYSICAL JOURNAL, 486:L15–L18, 1997 September 1
©1997. The American Astronomical Society. All rights reserved. Printed in U.S.A.
Up:
Issue Table of Contents
Go to: Search Page | Previous Article | Next Article
Other formats: HTML (small files) | PDF (162 kb) | PostScript pages (178 kb)

A Search for H2O Maser Emission Toward Active Galactic Nuclei: Discovery of a Nuclear Maser Source in NGC 3735

L. J. GREENHILL, J. R. HERRNSTEIN, 1 J. M. MORAN, AND K. M. MENTEN 2

Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138

AND

T. VELUSAMY

Jet Propulsion Laboratory, 4800 Oak Grove Drive, Pasadena, CA 91109

Received 1997 March 10; accepted 1997 June 13


ABSTRACT

     We report the results of a survey for H2O maser emission in the 616–523 transition at 1.35 cm wavelength in 29 active galactic nuclei (AGNs). One new maser was detected. The detection rate among objects with recessional velocities <7000 km s-1 is consistent with rates reported elsewhere for similarly nearby objects (about one in 15). The new maser lies in the edge-on Seyfert 2 galaxy NGC 3735 (inclination 77°) within 10 km s-1 of the systemic velocity. No other emission has been identified at velocities within ±500 km s-1 of the systemic velocity. The maser is coincident with the radio continuum peak of the nucleus at 6 and 3.6 cm wavelengths to within the estimated 1 σ astrometric uncertainty of 0&farcs;1 (15 pc at a distance of 30 Mpc).

Subject headings: galaxies: active—galaxies: individual (NGC 3735)—galaxies: nuclei—masers


CONTENTS


FOOTNOTES

     1 Present address: National Radio Astronomy Observatory, Soccoro, NM, 87801.

     2 Present address: Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, D-53121 Bonn, Germany.

§1. INTRODUCTION

     Recent surveys of H2O masers in the 616–523 transition toward active galactic nuclei (AGNs) have been strongly motivated by several cases in which the masers probably trace the structure and dynamics of parsec-scale molecular disks bound by the central engines. The maser in NGC 4258 is the best example (Miyoshi et al. 1995), followed by the suggestive results of interferometric studies of NGC 1068 (Greenhill et al. 1996; Greenhill & Gwinn 1997) and NGC 4945 (Greenhill, Moran, & Herrnstein 1997). In the case of NGC 4258, the maser emission has also provided the means by which to measure a geometric distance (Herrnstein 1997), to pinpoint the location of the central engine, and to measure the position offset between the engine and the nozzle of the associated relativistic jet at 1.35 cm wavelength (Herrnstein et al. 1997). Water masers can also be directly associated with jet activity in AGNs, as demonstrated by the case of NGC 1052 (Braatz et al. 1996a).

     There are 19 H2O masers known in AGNs, including the detection reported here, the recently discovered source in NGC 5793 (Hagiwara et al. 1997), and the weak maser in M51 (Ho et al. 1987), which has been omitted in some lists because of its relatively low luminosity (see Braatz, Wilson, & Henkel 1996b). The largest single survey to date is that of Braatz et al. (1996b), which includes primarily galaxies with recessional velocities <7000 km s-1. A statistical analysis of results from that survey, including all previous detections, indicates a detection rate of ≈7% among 216 Seyfert 2 nuclei and LINERs, with no masers occurring in Seyfert 1 nuclei (Braatz, Wilson, & Henkel 1997). Strong maser emission requires long gain paths in molecular gas, which presumably are not oriented along the line of sight in the putative face-on orientations of the compact disk and toroidal structures in Seyfert 1 nuclei. Otherwise, no strong correlations have yet been found between maser emission and the global properties of the host galaxies, although where X-ray measurements are available, all known H2O masers lie in galaxies with large X-ray obscuring columns, ≳1023 cm-2 (see, e.g., Braatz et al. 1997).

     We present the results of a single-dish survey of AGNs and preliminary follow-up interferometric study, at subarcsecond angular resolution, of the newly detected maser in NGC 3735 and its associated nuclear radio continuum emission.

§2. SURVEY RESULTS

     We observed 29 AGNs with the NASA Deep Space Network 70 m antenna at Goldstone between 1996 August and 1997 April (Table 1). The system temperature at 1.35 cm wavelength was typically 60–100 K, as measured with respect to a calibrated noise source. Antenna sensitivity was typically 1.7–2.0 Jy K-1, estimated from a previously established gain curve. Observations of pointing calibrators within about 40° of the target AGNs were made every 1–2 hr because residuals from the antenna-pointing solution were potentially a significant fraction of the antenna beam (48&arcsec;). Data for observing tracks for which the fluctuations in cumulative pointing corrections were consistently larger in magnitude than about 15&arcsec; were discarded.

     Single-polarization, position-switched observations typically provided spectra with rms noise levels of 20–30 mJy in 39 kHz channels (0.53 km s-1) in ≈50 minutes. AGNs were observed with either a single 40 MHz bandwidth (540 km s-1) centered on the systemic velocity or with two offset, overlapping, 40 MHz bands, spanning 60 MHz (810 km s-1) centered on the systemic velocity. The amplitude calibration of the observations is uncertain by about 20%. This takes into account the elevation-dependent gain response of the antenna. Multiplicative correction factors are applied to the rms noise levels quoted in Table 1 to reflect the potential degredations in signal-to-noise ratios due to average observed pointing errors and estimated losses due to atmospheric opacity, which was always <0.1 at zenith for the data presented here.

     The selected sample consisted of objects with Seyfert 2, LINER, or transition-type optical spectra and recessional velocities cz ≲ 12,000 km s-1 that had not been observed in previous surveys (e.g., Braatz et al. 1996b). Transition-type objects are intermediate between LINERs and H II nuclei, possibly having stellar and nonstellar sources of ionization (Ho, Filippenko, & Sargent 1993). A majority of the observed sources are nearby (cz < 7000 km s-1) and have only recently been classified as AGNs. The identifications used here are based on estimated line ratios in the published spectra (Ho, Filippenko, & Sargent 1995). Sources were also selected from among the Seyfert 2 nuclei and LINERs in the samples of Green, Anderson, & Ward (1992) and Kirhakos & Steiner (1990), which are galaxies that have been detected at both X-ray and infrared wavelengths, and among the AGNs in the CfA redshift survey (J. Huchra, private communication). The sample of Slee et al. (1994) was also used, though these sources were observed mainly in a complementary survey conducted in 1995 and 1996 at the Parkes antenna of the CSIRO in Australia, which will be discussed elsewhere.

     One new maser has been detected, in NGC 3735. The detection of one maser among the 29 AGNs in the sample, and among the 19 AGNs with cz < 7000 km s-1, is consistent with the detection rate reported by Braatz et al. (1997) for a distance-limited sample (15 in 216). Because the sample of observed galaxies is incomplete, it is difficult to draw statistical inferences. However, in the Braatz et al. sample, the detection rate of water masers in highly inclined galaxies is somewhat greater than it is for galaxies at other orientations. (The distribution of maser galaxy inclinations is broad and almost flat between 30° and 90°, but the sample is deficient in high-inclination galaxies by factors of a few.) Both NGC 3735 and the new maser galaxy NGC 5793 (Hagiwara et al. 1997) are nearly edge-on. The high-inclination bias is made more significant by the addition to the Braatz et al. (1996b) distance-limited sample of NGC 3735, the 15 other nearby (cz < 7000 km s-1) AGNs in our sample that have measured inclinations, and NGC 5793. (Conversely, seven of the 15 known maser galaxies have inclinations >70°.) However, it remains reasonably likely that maser emission will be detectable in galaxies with low and moderate inclinations, since substantial misalignments between the rotation axes of nuclear molecular disks and galaxy-scale stellar disks are possible. Large X-ray–obscuring columns toward many maser host AGNs suggest that misalignments on nuclear scales are smaller and that both NGC 3735 and NGC 5793 will also be found to have significant obscuring columns when the relevant X-ray observations are made.

§3. NGC 3735

     NGC 3735 is a nearly edge-on spiral galaxy in the Canes Venatici–Camelopardalis cloud (Nolthenius 1993). It has previously been classified as a starburst because of its large far-IR flux relative to its 6 cm wavelength radio continuum emission (Condon, Anderson, & Broderick 1995) but has recently been reclassified as a Seyfert 2 object by Ho et al. (1995) on the basis of optical spectroscopy. It has been detected in soft X-ray wavebands up to 2.4 keV (Rush et al. 1996) but not yet observed at higher energies. The characteristics of the galaxy are summarized in Table 2.

     The spectral characteristics of the H2O maser in NGC 3735 are summarized in Table 3. At the time of its discovery on 1996 November 10, the peak flux density of the maser was about 0.1 Jy, and its spectrum consisted of three components, all within 10 km s-1 of the systemic velocity of the galaxy (see Fig. 1). The apparent integrated luminosity of the maser emission was 12 L⊙, on the assumption of isotropic emission of radiation. To study the maser further, we made snapshot observations of the maser and the nuclear radio continuum emission with the NRAO 3 Very Large Array in A configuration on 1996 December 2 and 1997 January 12, respectively (see Table 3). The maser features, observed in a single 6.25 MHz bandwidth, were coincident within 20 mas. The peak flux density was about 0.17 Jy, and the shape of the spectrum was relatively unchanged since the discovery observation, although the coarse 1.3 km s-1 channel spacing of the observations made it difficult to discern subtle differences. The position of the maser is &agr;2000 = 11h35m57&fs;23 ± 0&fs;02, &dgr;2000 = 70°32&arcmin;07&farcs;8 ± 0&farcs;1. The continuum observations consisted of two 50 MHz bands observed in dual circular polarization. The continuum source was about 1 mJy at the 6 and 3.6 cm wavelengths and was not spatially resolved (see Table 3). The measured position of the maser is coincident with the positions of continuum emission to within the estimated astrometric uncertainties of about 0&farcs;1.

Fig. 1

     The maser lies in the nucleus of the galaxy, and it probably marks the location of the obscured central engine. If the maser emission arises in an edge-on disk surrounding the central engine, then the systemic maser emission would be expected to drift in velocity as a result of centripetal acceleration, in analogy to the cases of water emission in NGC 4258 (see, e.g., Baan, Haschick, & Peng 1994) and NGC 2639 (Wilson, Braatz, & Henkel 1995). A second spectrum obtained with Goldstone on 1997 February 10 (in which the peak flux density is 0.2–0.3 Jy) shows that over 3 months the velocity of the 2662 km s-1 feature changed by <1 km s-1, or about half of the half-power line width. Hence, the acceleration is <4 km s-1 yr-1 and the central engine mass is ≲9 × 108R$\mathstrut{^{2}_{{\rm pc}}}$ M⊙, where Rpc is the characteristic disk radius in parsecs. The velocities of the other two spectral features are ambiguous because they are more severely affected by blending.

     Maser emission from an edge-on disk may also exhibit spectral features at velocities symmetrically red- and blueshifted from the systemic velocity by the characteristic rotation speed of the disk. With the 37 m NEROC Haystack telescope 4 in 1997 December, we have searched for “high-velocity” emission within 500 km s-1 of the systemic velocity, operating in single polarization position-switched mode. No emission was detected with an rms noise in the spectra of about 80 mJy (1 σ) in 62 kHz spectral channels (0.84 km s-1). If high-velocity maser emission is ultimately discovered in this galaxy, the inferred rotation speed and upper limits (or estimates) of velocity drift may be used to estimate the characteristic radius of the molecular disk in which the masers may lie, or limits thereon. In this case, very long baseline interferometric observations of the emission will provide an estimate of the disk angular radius and, hence, a geometric distance for this galaxy (see, e.g., Miyoshi et al. 1995). Finally, we note that until the background molecular structure traced by the masers can be modeled and shown to be an edge-on disk with a rotation curve appropriate to its being bound by a supermassive object(s), we cannot preclude the possibility that the NGC 3735 maser is associated with jet activity, star formation, or other phenomena within or superposed on the nucleus.

FOOTNOTES

     3 The National Radio Astronomy Observatory is operated by Associated Universities, Inc., under cooperative agreement with the National Science Foundation.

     4 Radio astronomy observations at the Haystack Observatory of the Northeast Radio Observatory Corporation are supported by a grant from the National Science Foundation.

ACKNOWLEDGMENTS

     We thank the staff of the Deep Space Network (DSN) at Goldstone for their assistance in making these observations and Bob Preston for facilitating the project. We are grateful to the director of the Haystack Observatory and to the NRAO for granting director's discretionary time and VLA snapshot time, respectively. This research has made use of both the DSN facilities at Goldstone and the NASA/IPAC Extragalactic Database (NED), which are operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration, and the SIMBAD database, operated at the Centre de Donnés astronomiques de Strasbourg, France.

REFERENCES

FIGURES


Full image (27kb) | Discussion in text

     FIG. 1.—Spectrum of the water maser in NGC 3735. The fitted line profile consisting of three Gaussian components is shown (see Table 3). The systemic velocity of the galaxy from H I measurements and its uncertainty is indicated by the arrow and error bar. The radio astronomical definition of velocity is assumed. The spectrum was obtained in a 30 minute integration on 1996 November 10 with Goldstone. The spectral channel spacing is 0.52 km s-1 after a four-channel boxcar average in velocity.

TABLES

TABLE 1
AGN SAMPLE
Object&agr;1950&dgr;1950ReferenceDate
(year/day)
vsys a
(km s-1)
&Dgr;F&ngr; b(mJy)Band (km s-1)
NGC 266...00° 47&arcmin; 05&farcs;6+32° 00&arcmin; 23&arcsec;31997/082466129 c±375
IRAS 01217+0122...01 21 49.7+01 22 5611996/315514415±250
NGC 547...01 23 27.5-01 36 1921996/315552415±250
IRAS F01356-1307...01 35 38.0-13 07 2511996/3151211225±250
NGC 660...01 40 21.1+13 23 2531997/08285050 c±375
NGC 1058...02 40 23.1+37 07 4531997/04751832 c±375
ESO 549-G040...03 54 56.1-18 55 1641997/053753445 c±375
NGC 1961...05 36 33.9+69 21 1631997/082393427 c±375
IC 526...08 59 58.0+11 02 2351997/082452728 c±375
IRAS 09250+1230...09 25 00.8+12 30 1851996/259869217±250
NGC 3226...10 20 43.1+20 09 0631997/070115127 c±375
NGC 3254...10 26 31.3+29 44 5031997/053135545 c±375
NGC 3486...10 57 40.4+29 14 3731997/05368145 c±375
NGC 3507...11 00 46.3+18 24 2531997/05397945 c±375
IRAS 11186-0242...11 18 39.1-02 42 3751997/047746432 c±375
NGC 3692...11 25 48.9+09 40 5531996/315172627±250
Mrk 176...11 29 55.4+53 13 3541997/091791931, 53 c d±375
NGC 3735...11 33 04.8+70 48 4231996/315269616±250
NGC 4143...12 07 04.8+42 48 4331996/31598516±250
NGC 4374 (M84)...12 22 31.5+13 09 5131996/315100021±250
NGC 4565...12 33 52.0+26 15 4731997/047122732 c±375
NGC 4589...12 35 29.5+74 28 1031997/070198024 c±375
NGC 4939...13 01 37.3-10 04 1821996/259311117±250
IRAS 14110+0912...14 10 58.9+09 12 5551997/082716532 c±375
IRAS 15480-0344...15 48 04.0-03 44 1841997/091902032 c±250
NGC 6323...17 11 46.9+43 50 1941997/091776147, 41 c d±375
NGC 6393...17 29 40.1+59 40 3011997/047, 053842432 c±375
IRAS 23461+0157...23 46 07.7+01 57 4051997/047916832 c±375
IRAS 21116+0158...21 11 39.9+01 58 1151996/259391224±250

     
a Heliocentric velocity assuming the optical astronomical definition of velocity.
     b 1 σ noise (rms) in a 39 kHz spectral channel.
     c Observation within ±65 km s-1 of vsys is more sensitive by a factor of $\mathstrut{\sqrt{2}}$ because of the overlap of component 40 MHz bands.
     d Sensitivities are for (1) velocities shifted 0–375 km s-1 and (2) velocities shifted -375–0 km s-1, with respect to systemic.
REFERENCES.—(1) Seyfert 2 (Green, Anderson, & Ward 1992), (2) E/S0 galaxies with compact radio cores (Slee et al. 1994), (3) low-luminosity AGNs (Ho, Filippenko, & Sargent 1995), (4) AGNs selected from the CfA redshift survey (J. Huchra, private communication), (5) Seyfert 2 (Kirhakos & Steiner 1992).

Image of typeset table | Discussion in text
TABLE 2
NGC 3735 (UGC 06567)
MeasurementReference
Optical nucleus:
     &agr;2000...11°35&arcmin;59&farcs;66 a1
     &dgr;2000...70°32&arcmin;06&farcs;2
Distance...30 Mpc2
Vsys (H I)...2672 ± 7 km s-1 b3
Vsys (optical)...2647 ± 46 km s-1 b3
Position angle...128°
Inclination...77°4
F&ngr; (6 cm)...28 ± 5 mJy c5
F&ngr; (100 &mgr;m)...18.4 ± 0.9 Jy6
F&ngr; (60 &mgr;m)...6.7 ± 0.3 Jy6
F&ngr; (25 &mgr;m)...1.03 ± 0.08 Jy6
F&ngr; (12 &mgr;m)...0.66 ± 0.04 Jy6
log (LFIR/L⊙)...10.5
log (L0.1–2.4 keV/erg s-1)...40.7 ± 0.27

     
a Position uncertainty 10&arcsec; × 10&arcsec; (95% confidence).
     b Heliocentric systemic velocity, assuming the radio definition of velocity.
     c Beam half-power full width 2&farcm;3.
REFERENCES.—(1) Dressel & Condon 1976, (2) Nolthenius 1993, (3) de Vaucouleurs et al. 1991, (4) Guthrie 1992, (5) Gregory & Condon 1991, (6) Moshir et al. 1992, (7) Rush et al. 1996.

Image of typeset table | Discussion in text
TABLE 3
OBSERVED MASER AND RADIO CONTINUUM PROPERTIES IN NGC 3735
F&ngr; a
(Jy)
vhel b
(km s-1)
&Dgr;v c
(km s-1)
Maser (1996 November 10)
Component 1...0.09 ± 0.012662.0 ± 0.12.8 ± 0.2
Component 2...0.05 ± 0.012666.5 ± 0.21.4 ± 0.4
Component 3...0.08 ± 0.012668.8 ± 0.22.9 ± 0.4
Continuum
F&ngr; (2 cm) d...<1 mJy……
F&ngr; (3.6 cm) d...0.6 ± 0.1 mJy……
F&ngr; (6 cm) d...0.7 ± 0.1 mJy……

     
a Fitted peak flux densities at discovery.
     b Heliocentric velocity, assuming the radio astronomical definition of velocity.
     c Full line width at half-power.
     d Synthesized beamwidths: 0&farcs;24 × 0&farcs;14 (2 cm wavelength) 0&farcs;29 × 0&farcs;19 (3.6 cm wavelength), and 0&farcs;48 × 0&farcs;33 (6 cm wavelength). Detected emission unresolved.

Image of typeset table | Discussion in text
Up: Issue Table of Contents
Go to: Top of This Article | Search Page | Previous Article | Next Article