Medium Resolution Near-Infrared Spectra of the Host Galaxies of Nearby Quasars
Huynh Anh N. Le, Soojong Pak, Myungshin Im, Minjin Kim, Chae Kyung Sim, Luis C. Ho
aa r X i v : . [ a s t r o - ph . GA ] J un Medium Resolution Near-Infrared Spectra of the HostGalaxies of Nearby Quasars
Huynh Anh Nguyen Le a , Soojong Pak a, ∗ , Myungshin Im b , Minjin Kim c ,Chae Kyung Sim a , Luis C. Ho d,e a School of Space Research, Kyung Hee University1732 Deogyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do, 446-701, Korea b Department of Physics and Astronomy, Seoul National UniversityCenter for the Exploration of the Origin of the Universe (CEOU), Seoul, Republic ofKorea c Korea Astronomy and Space Science Institute, Daejeon, Republic of Korea d Kavli Institute for Astronomy and Astrophysics, Peking University, Beijing 100871,China e Department of Astronomy, Peking University, Beijing 100871, China
Abstract
We present medium resolution near-infrared host galaxy spectra of lowredshift quasars, PG 0844 + 349 (z=0.064), PG 1226 + 023 (z=0.158), andPG 1426 + 015 (z=0.086). The observations were done by using the InfraredCamera and Spectrograph (IRCS) at the Subaru 8.2 m telescope. The fullwidth at half maximum of the point spread function was about 0.3 arcsecby operations of an adaptive optics system, which can effectively resolve thequasar spectra from the host galaxy spectra. We spent up to several hoursper target and developed data reduction methods to reduce the systematicnoises of the telluric emissions and absorptions. From the obtained spectra,we identified absorption features of Mg I (1.503 µ m), Si I (1.589 µ m) and CO(6-3) (1.619 µ m), and measured the velocity dispersions of PG 0844 + 349to be 132 ±
110 km s − and PG 1426 + 015 to be 264 ±
215 km s − . Byusing an M BH − σ relation of elliptical galaxies, we derived the black hole(BH) mass of PG 0844 + 349, log ( M BH /M ⊙ ) = 7.7 ± log ( M BH /M ⊙ ) = 9.0 ± ∗ Corresponding author
Email addresses: [email protected] (Huynh Anh Nguyen Le), [email protected] (Soojong Pak)
Preprint submitted to Advances in Space Research November 14, 2018 alues from broad emission lines with an assumption of a virial factor of 5.5.
Keywords: galaxies: active-galaxies: kinematics and dynamics
1. Introduction
Nearby galaxies have bulge with supermassive black holes (Richstone,1998). Understanding the link between the supermassive black holes andtheir host galaxies is important in studying the formation and evolution ofthe galaxies. The relation of M BH − σ has been discovered, in which M BH is the mass of supermassive black hole and σ is the stellar velocity disper-sion of the bugle (e.g., Ferrarese, Pogge, & Peterson et al., 2001; Gebhardt,Kormendy & Ho et al., 2000; Gebhardt, Bender & Bower et al., 2000).Nevertheless, the measurements of stellar velocity dispersion of host galaxyare difficult in optical bands because of the presence of young stars in thehost galaxy. Absorption lines in optical bands such as Mg b at 517 nm andCa triplet at 850 nm are diluted by continuum. Therefore, it is necessaryto use stellar lines in other wavebands in measuring velocity dispersion. CObandheads in near-infrared (NIR) have been suggested to be the best instudying the velocity dispersion of nearby galaxies (McConnell, 2011). Inaddition, NIR stellar lines have the potential of explaining for the relationbetween supermassive back holes and their host galaxies.In this paper, we present the medium resolution host galaxy spectra ofnearby quasars in H-band obtained at the Subaru telescope. Thanks tothe advantages of using adaptive optics technology, we can isolate the quasarspectra from the host galaxy spectra. The obtained spectra with medium res-olution can be used to determine the stellar velocity dispersions in the bulgeof the host galaxies, and to estimate the supermassive black hole masses.Section 2 of this paper shows the observation processes. The detailed datareduction processes of NIR quasar spectra are presented in section 3. Resultsand discussions are shown in section 4. Section 5 is the conclusion.
2. Observations
The observations were performed at the Subaru 8.2 m telescope usingthe IRCS (Kobayashi et al., 2000) operated with the Adaptive Optics (AO),AO36 (Hayano, Takami, & Guyon et al., 2008), on 2003 February 11 and2004 April 3 and 4. The average AO-assisted point spread function was 0.3arcsec. 2 .1. Observation of Quasars
We observed three nearby quasars, PG 0844 + 349, PG 1226 + 023, andPG 1426 + 015. Table 1 shows the log of the observations. In 2003, weobserved PG 0844 + 349 only. The slit width was 0.3 arcsec with R = 10 ,and the position angle of the slit was 0 deg. The echelle setting of thespectrograph was in H + setting (1 . − . µ m), and the total integrationtime was about two hours with each exposure of 180 sec. The observationswere done in an Nod-off-slit mode. We first observed the target and thenmoved the telescope to the nearby background sky. The sequences of theobservations were object → sky → sky → object .In 2004, we observed three targets: PG 0844 + 349, PG 1226 + 023, andPG 1426 + 015. The slit width was 0.6 arcsec with R = 5 × . The echellesettings of the spectrograph were in H − setting (1 . − . µ m) and H +setting, and the total integration time was one hour for each target. Otherinstrument settings and the observation modes were the same as in 2003. We observed A0 V type standard stars to correct the telluric absorptionlines in the target spectra. In addition, bright template stars (H <
3. Data Reduction
Data reduction was done by using
IRAF tasks following the methodsdescribed in Pyo (2002). The details of the data reduction for standard starsand template stars can be found in Le, Kang, & Pak et al. (2011). Thehost galaxy spectra were reduced by using similar procedures as that of thetemplate stars. Fig. 1 shows the detailed data reduction processes.The host galaxy spectra within the radius from 0.24 to 1.89 arcsec areextracted for PG 0844 + 039, and from 0.24 to 2.34 arcsec for PG 1226 + 023and PG 1426 + 015. We chose the minimum radius to be 0.24 arcsec toensure that the extracted host galaxy spectra are not affected by emissionfrom QSOs. We confined the maximum radius to extract the host galaxy IRAF (Image Reduction and Analysis Facility) is distributed by the National OpticalAstronomy Observatories (NOAO). R e = 1.89 arcsec. The slit length of Subaru/IRCS, L = 5.17arcsec, however, is shorter than the diameters of PG 1226 + 023 ( D = 12.56arcsec) and of PG 1426 + 015 ( D = 8.22 arcsec). Therefore, the maximumradius to be extracted should be smaller than half of the slit length, L = 2.59arcsec.The effects of the residual OH sky-lines could cause one of the problemsof our obtained spectra. The emission lines of OH cannot be completelycorrected by the sky-background subtraction processes. We masked out thedata points which contain noises from the OH sky-lines.
4. Results and Discussion
Fig. 2 shows the reduced spectra of PG 0844 + 349, PG 1226 + 023, andPG 1426 + 015. In the figure, the spectrum of K2 III (HD52071) is alsoplotted to be compared with the observed host galaxy spectra. In addition,the spectra of PG 1426 + 015 in Watson, Martini, & Dasyra et al. (2008) andDasyra, Taconi, & Davies et al. (2007) are shown in the figure for compar-isons.From Fig. 2, we identify prominent features, e.g., Mg I (1.488 µ m), Mg I(1.503 µ m), Si I (1.589 µ m), and CO (6-3) (1.619 µ m) in the spectrum ofPG 0844 + 349. But the absorption features such as CO (3-0) (1.558 µ m)and CO (4-1) (1.578 µ m) cannot be seen due to the effects of the remainingOH sky-lines.After the redshift correction, the host galaxy spectrum of PG 1226 +023 in Fig. 2 has a limited wavelength coverage to be compared with themolecular lines of the stellar template spectrum. Unfortunately, these arenot stellar spectra in the literature that overlap with the observed spectrumof PG 1226 + 023 to identify the molecular lines.In the case of PG 1426 + 015 spectrum, we could detect Mg I (1.503 µ m),Si I (1.589 µ m) and CO (6-3) (1.619 µ m) lines comparing the K2 III stellartemplate spectrum and the host galaxy spectra of Watson, Martini, & Dasyraet al. (2008) and Dasyra, Taconi, & Davies et al. (2007). But the CO (3-0) http: // / Observing / Instruments µ m) and CO (4-1) (1.578 µ m) lines are hard to confirm because ofthe effects of the remaining OH sky-lines. The signal-to-noise (S/N) ratios ofMg I and Si I absorption lines are 3. The S/N ratio of CO (6-3) absorptionline is 5. From the obtained spectra, we identified a few stellar absorption lines ofthe host galaxy and measured the velocity dispersion of the host galaxy usingthe direct fitting method of Barth, Ho, & Sargent (2002).We assume that the host galaxy spectrum follows that of K2 III type stars.We define the model of host galaxy spectrum from the equation as M ( λ ) = A + T G ( λ, σ ) (1)where A is arbitrary constant value from the unknown contribution fromthe quasar continuum; T G is convolution of the stellar spectrum with theline-of-sight velocity distribution; σ is the velocity dispersion; and λ is therest wavelength. The best-fit convolved stellar spectrum to the host galaxyspectrum is found based on the minimum of chi-square values. The chi-squarevalue is calculated by χ = X λ (cid:18) M λ − O λ ǫ λ (cid:19) (2)where M λ is the model spectrum from the equation (1); O λ is the observedhost galaxy spectrum; and ǫ λ is the error of the host galaxy spectrum whichhas typical value of 0.037 and is determined by adding the standard deviationof the spectrum to the root-mean-square of each data point.From Fig. 2, due to the redshift correction, we could not measure thevelocity dispersion of host galaxy spectrum of PG 1226 + 023. In case ofPG0844 + 349, we identified some prominent features since the effects of theremaining OH sky-lines make it hard to calculate the velocity dispersion ofthe host galaxy. We have corrected the remaining OH sly-lines effects byremoving those data points of the host galaxy spectra which are affectedby sky-lines. We have calculated the velocity dispersion of the host galaxyPG 1426+015 to be σ = 264 ±
215 km s − from the fitting with K2 III stellarspectrum at Mg I (1.503 µ m) line (Fig. 3) and Si I (1.589 µ m) line (Fig.4). The reduced chi-square value is 0.8. In Fig. 5, from the measurementof CO (6-3) (1.619 µ m), the velocity dispersion of PG 0844 + 349 is 1325
110 km s − with the reduced chi-square of 0.4. Due to the low S/Nratio of the data, the errors of velocity dispersions are very large. But thebest-fit sigma values which are calculated from our method are consistentwith others. From the measurements of Watson, Martini, & Dasyra et al.(2008) and Dasyra, Taconi, & Davies et al. (2007), the velocity dispersionsof host galaxy of PG 1426 + 015 are 217 ±
15 km s − and 185 ±
67 km s − ,respectively, which are similar to our results. The data obtained by ISAAClong-slit spectrometer on the 8m Antu unit of the Very Large Telescope(Dasyra, Taconi, & Davies et al., 2007) has higher S/N ratio compared toour data obtained by IRCS, Subaru telescope.From the velocity dispersion measurements, we derived the black holemasses using the M BH − σ relation of elliptical galaxies (Kormendy & Ho,2013). Table 2 shows the measured values of velocity dispersion and blackhole mass estimates of these quasars. The obtained black hole masses ofPG 0844 + 349 and PG 1426 + 015 are log ( M BH /M ⊙ ) = 7.7 ± log ( M BH /M ⊙ ) = 9.0 ± f = 5 . log ( M BH /M ⊙ ) = 8.0 ± log ( M BH /M ⊙ ) = 9.1 ±
5. Conclusions
We obtained NIR medium resolution host galaxy spectra of nearby quasars,PG 0844 + 349, PG 1226 + 023, and PG 1426 + 015 in H-band, using theIRCS instrument and the AO of the Subaru telescope. The data analysismethod of the NIR spectra is presented.From the spectra, we derived the stellar velocity dispersion of the hostgalaxy and its relation to the central super-massive BH. From the identifiedstellar absorption lines, we have obtained the velocity dispersion of PG 1426+015 to be 264 ±
215 km s − based on the measurement of Mg I (1.503 µ m)and Si I (1.589 µ m). In the case of PG 0844 + 349, we have measured thevelocity dispersion of the host galaxy to be 132 ±
110 km s − based on thecalculation of CO (6-3) (1.619 µ m). 6y using an M BH − σ relation of elliptical galaxies, the BH masses ofPG 0844 + 349 and PG 1426 + 015 are estimated to be log ( M BH /M ⊙ ) =7.7 ± log ( M BH /M ⊙ ) = 9.0 ± eferences Barth, A. J. , Ho, L. C., & Sargent, W. L. W. 2002, A Study of the DirectFitting Method for Measurement of Galaxy Velocity Dispersions, AJ, 124,2607Dasyra, K. M., Taconi, L. J., Davies, I. R., Genzel, R., Lutz D., Peterson,B. M., Veilleux, S., Baker, A. J., Schweitzer, M., & Sturm, E. 2007, HostDynamics and Origin of Palomar-Green QSOs, ApJ, 657, 102Ferrarese, L., Pogge, R. W., Peterson, B. M., Merritt, D., Wandel, A., &Joseph, C. L. 2001, Supermassive Black Holes in Active Galactic Nuclei. I.The Consistency of Black Hole Masses in Quiescent and Active Galaxies,ApJ, 555, L79Gebhardt, K., Kormendy, J., & Ho, L. C., et al. 2000, A Relationship betweenNuclear Black Hole Mass and Galaxy Velocity Dispersion, ApJL, 539, L13Gebhardt, K., Bender, R., & Bower, G., et al. 2000, Black Hole Mass Es-timates from Reverberation Mapping and from Spatially Resolved Kine-matics, ApJL, 543, L5Hayano, Y., Takami, H., Guyon, O., Oya, S., Hattori, M., Saito, Y., Watan-abe, M., Murakami, N., Minowa, Y., Ito, M., Colley, S., Eldred, M., Golota,G., Dinkins, M., Kashikawa, N., & Iye, M. 2008, ”Current status of thelaser guide star adaptive optics system for Subaru Telescope,” Proc. SPIE7015, 701510Kaspi, S., Smith, P. S., Netzer, H., Maoz, D., Jannuzi, B. T., & Giveon,U. 2000, Reverberation Measurements for 17 Quasars and the Size-Mass-Luminosity Relations in Active Galactic Nuclei, ApJ, 533, 631Kim, D., Im, M., & Kim, M. 2010, New Estimators of Black Hole Mass inActive Galactic Nuclei with Hydrogen Paschen Lines, ApJ, 724, 386Kobayashi, N., et al. 2000, IRCS: Infrared Camera and Spectrograph for theSubaru Telescope, SPIE, 4008, 1056Kormendy, J., & Ho, L. C. 2013, Coevolution (Or Not) of SupermassiveBlack Holes and Host Galaxies, ARAA, 51, 5118e, H. A. N., Kang, W., Pak, S., Im, M., Lee, J. E., Ho, L. C., Pyo, T. S.,& Jaffe, D. T. 2011, Medium Resolution Spectral Library of Late-TypeStellar Templates in Near-Infrared Band, JKAS, 44, 125McConnell, N. J., et al. 2011, The Black Hole Mass in Brightest ClusterGalaxy NGC 6086, AJ, 728, 100Onken, C. A., Ferrarese, L., Merritt, D., Peterson, B. M., Pogge, R. W.,Vestergaard, M., & Wandel, A. 2004, Supermassive Black Holes in ActiveGalactic Nuclei. II. Calibration of the Black Hole Mass-Velocity DispersionRelationship for Active Galactic Nuclei, ApJ, 615, 645Peng, C. Y., Ho, L. C., Impey, C. D., & Rix, H. W. 2002, Detailed StructuralDecomposition of Galaxy Images, AJ, 124, 266Peterson, B. M., Ferrarese, L., & Gilbert, K. M., et al. 2004, Central Massesand Broad-Line Region Sizes of Active Galactic Nuclei. II. A HomogeneousAnalysis of a Large Reverberation-Mapping Database, ApJ, 613, 682Pyo, T. S. 2002, Near Infrared [Fe II] Spectroscopy of Jets and Winds Ema-nating from Young Stellar Objects, Ph.D. thesis, Tokyo Univ.Richstone, D., et al. 1998, Supermassive Black Holes Then and Now, Nature,395, A14Watson, L. C., Martini, P., Dasyra, K. M., Bentz, M. C., Ferrarese, L.,Peterson, B. M., Pogge, R. W., & Tacconi, L. J. 2008, First Stellar VelocityDispersion Measurement of a Luminous QSO Host with Gemini NorthLaser Guide Star Adaptive Optics, ApJ, 682, L21Woo, J. H., Schulze, A., Park, D., Kang, W. R., Kim, S. C., & Riechers,D. A. 2013, Do Quiescent and Active Galaxies Have Different M BH − σ ∗ Relations?, ApJ, 772, 49 9 able 1: Observation log
Date Quasars z R EchelleSetting Slit Width Total Exposure(UT) (arcsec) (sec)2003 Feb 11 PG 0844 + 349 0.064 10000 H + 0.3 37 × H − × H − × H + 0.6 8 × H − × H + 0.6 5 × H + 0.6 8 × H − × H + 0.6 15 × able 2: Velocity dispersion measurements and black hole masses Quasars σ a σ b log ( M BH /M ⊙ ) c log ( M BH /M ⊙ ) d (km s − ) (km s − )PG 0844 + 349 132 ± e ± ± ± f ±
15 9.0 ± ± a This work. b Velocity dispersion values from Watson, Martini, & Dasyra et al. (2008). c This work. Black hole mass values are derived by using the formula (7) of Kormendy & Ho (2013). d Black hole mass values from Peterson, Ferrarese, & Gilbert et al. (2004). e Velocity dispersion value based on the measurement of CO (6-3) (1.619 µ m) line. f Velocity dispersion value based on the measurements of Mg I (1.508 µ m) and Si I (1.589 µ m) lines. igure 1: Data reduction processes. igure 2: Spectra of host galaxy PG 0844+349 (shifted with z = 0 . z = 0 . z = 0 . igure 3: Spectra of K2 III (HD52071) (blue-line) and host galaxy PG 1426 + 015 (shiftedwith z = 0 . igure 4: Spectra of K2 III (HD52071) (blue-line) and host galaxy PG 1426 + 015 (shiftedwith z = 0 . igure 5: Spectra of K2 III (HD52071) (blue-line) and spectra of host galaxy PG 0844+349(shifted with z = 0 .064). The best fit of the velocity convolved K2 III spectrum is shown inred-line. The dashed lines show the regions used for measurements of velocity dispersions.The A0 V and OH sky-lines spectra are shown in the lower plot.