The Impact of Stellar Populations on the Dynamics of Merger Remnants
aa r X i v : . [ a s t r o - ph . C O ] A ug Galaxy Wars: Star Formation and Stellar Populations in Interacting GalaxiesASP Conference Series, Vol. **VOLUME**, c (cid:13) **YEAR OF PUBLICATION**B. Smith, J. Higdon, S. Higdon, and N. Bastian The Impact of Stellar Populations on the Dynamics ofMerger Remnants
B. Rothberg, J. Fischer
Naval Research Laboratory, Remote Sensing Division, Code 7211, 4555Overlook Ave SW, Washington D.C. 20375
Abstract.
Many studies and simulations suggest gas-rich mergers do notcontribute significantly to the overall star-formation rate and total mass func-tion of galaxies. The velocity dispersions ( σ ) of Luminous & UltraluminousInfrared Galaxies measured using the 1.62 or 2.29 µ m CO bandheads imply theywill form m < m ∗ ellipticals. Yet, σ ’s obtained with the Calcium II triplet (CaT)at 0.85 µ m suggest all types of mergers will form m > m ∗ ellipticals. Presentedhere are recent results, based on high-resolution imaging and multi-wavelengthspectroscopy, which demonstrate the dominance of a nuclear disk of Red Su-pergiants (RSG) or Asymptotic Giant Branch (AGB) stars in the near-infraredbands, where dust obscuration does not sufficiently block their signatures. Thepresence of these stars severely biases the dynamical mass. At I -band, wheredust can sufficiently block RSG or AGB stars, LIRGs populate the Fundamen-tal Plane over a large dynamic range and are virtually indistinguishable fromelliptical galaxies. The “Toomre Hypothesis” (Toomre & Toomre 1972; Toomre 1977), proposesthat the merger of two-gas rich spiral galaxies will form an elliptical galaxy,often with a final stellar mass larger than the sum of the progenitors. In thelocal universe, Luminous and Ultraluminous Infrared Galaxies are ideal candi-dates for forming massive elliptical galaxies (Kormendy & Sanders 1992). Theseare objects with L IR > L ⊙ between 8-1000 µ m (Sanders & Mirabel 1996),contain vast quantities of molecular gas, and show strong evidence of recent orongoing merging activity. Radio recombination line observations of the nearestULIRG Arp 220 imply a formation rate of 10 M ⊙ yr − (Anantharamaiah et al.2000), while CO interferometric data indicate that 0.15-0.46 of the dynamicalmass of this system is gaseous (Downes & Solomon 1998; Greve et al. 2009).The star-formation rates and vast quantities of gas in LIRGs/ULIRGs couldadd a significant stellar component to the total mass of the merger.However, a number of studies, all using infrared CO bandheads to mea-sure central velocity dispersions ( σ ◦ ), have shown that LIRGs have massesconsistent with low-moderate luminosity elliptical galaxies ( L ∼ L ∗ )(Shier & Fischer 1998) and ULIRGs have masses consistent with L ≤ L ∗ (e.g.Genzel et al. (2001)). These results have raised significant doubts as to whethergas-rich mergers contribute significantly to the formation of elliptical galaxies.Yet, σ ◦ measured from the Calcium II Triplet absorption lines (CaT), sug-gest gas-rich mergers, including LIRGs, have masses which span nearly the en-1 B. Rothberg & J. Fischer tire mass range of elliptical galaxies (Lake & Dressler 1986; Rothberg & Joseph2006) (hereafter RJ06). This difference in σ , or σ -mismatch, is counter-intuitive.Namely, that LIRGs/ULIRGs, which are undergoing intense star-formation andpossess large quantities of dust, should show smaller σ ◦ at longer wavelengths.The use of infrared stellar lines to measure σ ◦ was initially motivated by theneed to pierce the veil of extinction in starburst galaxies and measure their“true” dynamical masses. However, the results presented here show that IR-luminous mergers are Janus-like, that is, they reveal two different dynamicalfaces depending on the wavelength observed. The obscuring characteristics ofdust in the optical in IR-luminous galaxies behaves in a manner beneficial fordetermining the true mass of merger remnants. The dynamical properties of a sample of 14 advanced (single-nuclei) mergerremnants are compared with a sample of 23 elliptical galaxies. The mergerremnants are a subsample of the 51 merger remnants discussed in detail inRothberg & Joseph (2004) (hereafter RJ04). The photometric data for themerger remnants include
F814W ( ∼ I -band) imaging from the Wide-Field Plane-tary Camera 2 ( WFPC2 ) or the Advanced Camera for Surveys Wide-field Cam-era (
ACS/WFC ) on
HST and K -band imaging from Quick Infrared Camera(QUIRC) on the University of Hawaii 2.2m telescope. The kinematic data forthe merger remnants include CaT observations from ESI on Keck-2, and CO ob-servations from either NIRSPEC on Keck-2 or GNIRS on Gemini South (Pro-gram GS-2007A-Q-17, P.I. Rothberg). Additional kinematic and photometricdata were obtained from the literature for several merger remnants and thecomparison sample of ellipticals. A similar σ -mismatch was reported for a sample of 25 nearby early-type (predom-inantly S0) galaxies by Silge & Gebhardt (2003). They also found that σ ◦ ,optical > σ ◦ ,CO , and suggested that the kinematics of the cold stellar component in S0galaxies was obscured by dust, and detectable only in the IR while the kinemat-ics of the hot spheroid dominated optical wavelengths. However, it remainedunclear whether bonafide ellipticals produced the same discrepancy. As notedin Rothberg (2009), no discernible difference between σ ◦ ,optical and σ ◦ ,CO wasfound for the comparison sample of 23 elliptical galaxies. On the other hand,LIRGs showed a large discrepancy, as first noted in RJ06. The Fundamen-tal Plane (FP) is a two-dimensional plane embedded in the three-dimensionalparameter space of σ ◦ , the half-light (effective) radius ( R eff ), and the surfacebrightness within the effective radius ( <µ> eff ). All elliptical galaxies lie on theFundamental Plane. The LIRG/ULIRG studies noted earlier found that whilethese mergers were overly luminous in the infrared, they would eventually evolveonto the FP, but the small σ ◦ ,CO meant they could not be the progenitors ofellipticals with L > L ∗ . RJ06, however, found that the observed range of σ ◦ ,CaT meant that gas-rich mergers could populate nearly all of the mass-range of ellip-ticals, including L > L ∗ . Figure 1 is a two panel figure which shows the I -band he Impact of Stellar Populations on the Dynamics of Merger Remnants σ ο (km s -1 ) + 0.25 < µ I > eff (mag arcsec -2 ) -1.0-0.500.51.01.5 l o g R e ff ( k p c ) LIRG mergersmergersEllipticals I -band FP 4.5 5.0 5.5 6.0 6.5log σ ο (km s -1 ) + 0.205 < µ K > eff (mag arcsec -2 ) LIRG mergersmergersEllipticals K -band FP Figure 1. Shown are the I -band ( left ) and K -band ( right ) Fundamen-tal Planes from Scodeggio et al. (1997) and Pahre et al. (1998) respectively.Overplotted in both panels are LIRG merger remnants (open circles), non-LIRG merger remnants (filled circles) and elliptical galaxies (open diamonds).All 6 LIRGs, 20/23 ellipticals, and 3/8 non-LIRG merger remnants are over-plotted on the I -band FP. and K -band FPs, with LIRG merger remnants (open circles), non-LIRG mergerremnants (filled circles) and ellipticals (open diamonds).As expected, the ellipticals show little difference between the I and K -bandFPs. The difference in the location of (primarily) the LIRGs in the I -band and K -band is striking. Figure 1 explains the apparent contradictory results betweenearlier LIRG/ULIRG studies, which used “pure” H or K -band FPs and thosefrom RJ06, which used a “hybrid” FP (CaT σ ◦ and K -band photometry). Inthe I -band, the dynamical properties of LIRGs are indistinguishable from el-lipticals. Figure 2 shows a comparison between the M dyn /L and M dyn in the I -band ( left ) and K -band ( right ) for the merger remnants and elliptical galaxies(same symbols as Figure 1). Overplotted is the evolution of M/L over time fora burst population from Maraston (2005) (hereafter M05). Figure 2 shows thatin the I -band, the measured dynamical masses and stellar ages of the LIRGsare nearly the same as elliptical galaxies. However, the K -band measurementsimply LIRGs have smaller M dyn and young ages.The K -band is dominated by the presence of young stars. Numerical simula-tions have long predicted that gaseous dissipation in the merging event funnelsthe gas into the barycenter of the merger (e.g. Barnes & Hernquist (1991);Barnes (2002)). This forms a rotating gaseous disk in the central 1-2 kpc ofthe merger, which then undergoes a strong starburst, forming a rotating disk ofyoung stars. One observational signature of this starburst is the presence of “ex-cess light” in the surface brightness profiles of mergers (e.g. Mihos & Hernquist(1994)), first detected in the K -band by RJ04. The σ -mismatch detected inIR-luminous galaxies is another observational signature of these rotating centralstarbursts. Dust associated with these nuclear starbursts blocks most of their B. Rothberg & J. Fischer light at λ < µ m, while allowing the random motions of the nearly virializedolder stars to dominate the σ ◦ measurement at I -band. This functions in a sim-ilar manner to an occulting mask in a coronograph. At H and K -band, the RSGand AGB stars can account for 60-90% of the light (M05), therefore the diskkinematics overwhelms the σ ◦ ,CO measurement. dyn /M O • -1.5-1.0-0.500.51.01.5 l o g ( M / L I ) dyn /M O • -2.0-1.5-1.0-0.50 l o g ( M / L K ) Figure 2. Two panel figure showing pure I -band ( left ) and pure K -band( right ) M / L vs. M dyn . The overplotted vector (solid line) in each panel is theevolution of M / L for a single-burst stellar population with solar metallicityand a Salpeter IMF as computed from Maraston (2005). The dotted verticalline in each panel indicates m ∗ . References
Anantharamaiah, K. R., Viallefond, F., Mohan, N. R., Goss, W. M., & Zhao, J. H.2000, ApJ, 537, 613Barnes, J. E. & Hernquist, L. E. 1991, ApJ, 370, L65Barnes, J. E. 2002, MNRAS, 333, 481Downes, D. & Solomon, P. M. 1998, ApJ, 507, 615Genzel, R., Tacconi, L. J., Rigopoulou, D., Lutz, D., & Tecza, M. 2001, ApJ, 563, 527Greve, T. R., Papadopoulos, P. P., Gao, Y., & Radford, S. J. E. 2009, ApJ, 692, 1432Kormendy, J. & Sanders, D. B. 1992, ApJ, 390, L53Lake, G. & Dressler, A. 1986, ApJ, 310, 605Maraston, C. 2005, MNRAS, 362, 799 (M05)Mihos, J. C. & Hernquist, L. 1994, ApJ, 437, L47Pahre, M. A., Djorgovski, S. G., & de Carvalho, R. R. 1998b, AJ, 116, 1591Rothberg, B. & Joseph, R. D. 2004, AJ, 128, 2098 (RJ04)—. 2006, AJ, 131, 185 (RJ06)Rothberg, B. 2009, in Galaxy Evolution: Emerging Insights and Future ChallengesSanders, D. B. & Mirabel, I. F. 1996, ARA&A, 34, 749Scodeggio, M., Giovanelli, R., & Haynes, M. P. 1997, AJ, 113, 101Shier, L. M. & Fischer, J. 1998, ApJ, 497, 163Silge, J. D. & Gebhardt, K. 2003, AJ, 125, 2809Toomre, A. 1977, in Evolution of Galaxies and Stellar Populations, p. 401 he Impact of Stellar Populations on the Dynamics of Merger Remnants5