Science Highlights

Radio-quiet type 2 quasars are driving outflows

My postdoctoral research at JHU under the supervision of Prof. Nadia L. Zakamska is focused on observational studies of quasar / black hole feedback. The prevalence and energetics of quasar feedback is a major unresolved problem in galaxy formation theory, yet direct probes of this process in action are scarce and limited to small samples of active nuclei, and the physical detail remains uncertain. Collaborating with my supervisor and our intelligent colleagues including Jenny E. Greene (Princeton), Nicole P.H. Nesvadba (IAS, Orsay, France) and Xin Liu (UCLA), we have conducted an Integral Field Unit (IFU) campaign on Gemini-North to observe 11 objects which are the most luminous (in [O III]λ5007Å) radio-quiet obscured (type 2) quasars found at z~0.5 in the SDSS survey. To summarize our exciting findings, the nebulae of warm ionized gas (~104 K) around these quasars are
Such smooth and round large-scale gas nebulosities are striking. They are inconsistent with illuminated merger debris and may be the signature of accretion energy reaching gas at large scales. These findings are likely the long-sought direct evidence for galaxy-scale quasar feedback. The results of this campaign have been published in a series of at least two papers in MNRAS in 2013 (I. size & morphology: Liu et al. 2013c; II. gas kinematics: Liu et al. 2013d) and reported in a newsletter on GeminiFocus in 2012.

Ubiquitous outflows from luminous z~0.5 quasars

Are type 1 and type 2 quasars in a different evolutionary state? Theoretical models have long predicted an evolutionary scenario in which galaxy mergers induce both star formation and nuclear activity, triggering a violent transition from an obscured accretion and star formation stage to an unobscured phase as a Type 1 quasar, yet direct observational evidence is lacking. Our conducted another Gemini IFU observations of 12 Type 1 radio-quiet SDSS quasars with redshifts and [O III]5007 luminosities matched to the Type 2 sample. Having removed the Fe II emission and continuum, we detect galaxy-wide [O III]-emitting nebulae surrounding all but one quasars. Surprisingly, their sizes (20-36 kpc across), morphology (smooth and round), velocity fields and velocity dispersions are all similar to the Type 2s, signifying similar quasi-spherical outflows that are likely escaping from the host galaxies.

Hence, energetic outflows are ubiquitous in both luminous type 1 and type 2 quasars at z~0.5, and we see no evolutionary evidence in the nebulae of these populations. A paper describing these findings is now under review (Liu et al. 2013e).

Spatially-resolved star formation law

During my graduate career, I was focusing my research on observational studies of the interstellar medium and its relation to the formation of stars. The shape of the molecular S-K law from different groups as yet fall into two categories: super-linear or roughly linear scaling relation, which is still under debate. We quantitatively show that these controversial results is a result of removing or preserving the diffuse stellar and dust emission which is not related to current star formation. Collaborating with Prof. Jin Koda at SUNY Stony Brook, we performed short-spacing correction for the CARMA interferometry data using the on-the-fly mapping with the Nobeyama 45m dish, carefully removed the diffuse emission of dust and stars, and find that in M51 and NGC 3521, both the slope and scatter of the sub-kpc Schmidt-Kennicutt law depend on the sampling scale: the larger the sampling scale, the flatter the slope, and the smaller the scatter. However, the sub-kpc molecular Schmidt-Kennicutt law is clearly super-linear.

This project was the core of my thesis supervised by Prof. Daniela Calzetti, and was published in the Astrophysical Journal in 2011. This paper (Liu et al. 2011) is currently the most cited of mine (once every three weeks on average since its publication).

Also see our interpretation of these observations by utilizing numerical simulations in a follow-up ApJ paper (Calzetti, Liu & Koda 2012). The conclusion is that geometric and statistical effects themselves can explain all that we have observed.

Dust extinction and geometry in M83

Conventionally, dust extinction is calculated based on a "dust screen" postulation where the emitters and absorbers are well separated, while actual dust geometry is highly complicated. As part of the Early Release Science (ERS) developed by the HST/WFC3 Scientific Oversight Committee (SOC), the starburst galaxy M83 was imaged in multiple hydrogen lines, enabling an unprecedented detailed study of dust extinction in a nearby galaxy.

Involved in WFC3 ERS, we derived 6 pc-resolution extinction maps from Hα/Hβ and Hα/Paβ ratios, finding that Hα/Paβ probes AV values larger by at least 1 mag than the shorter wavelengths. A large diversity of absorber/emitter geometric configurations can account for the data, but a "dust screen" and a uniform mixture of dust and emitters bracket most data points. Moreover, when averaged on large scales (>100 pc), the data become consistent with a "dust screen".

The extinction in any region can be described by a combination of the foreground screen and the uniform mixture model with weights of 1/3 and 2/3 in the center (<2 kpc), respectively, and 2/3 and 1/3 for the rest of the disk. This empirical prescription will significantly improve the accuracy of extinction corrections for M83-like galaxies. These results were presented in a short and cute ApJ Letter in 2013 (Liu et al. 2013b).

Statistical properties of H II regions

To date, the luminosity function of HII regions has been studied predominantly in the Hα line and from the ground. For the first time, we carried out a study of this topic with a sample of galaxies observed in the Paα line by HST from the space. Paα in near-IR is virtually extinction free, and the spatial resolution is 26 parsec or better. Collaborating with Prof. Robert C. Kennicutt Jr. (Cambridge), Dr. Eva Schinner (MPIA), Prof. Yoshiaki Sofue (Tokyo & Meisei Univ.) and their groups, we do not observe the "type II" luminosity function in the form of a broken power law often seen from the ground. Instead, the data is consistent with a single power law N(L) dL ∝ L-2 dL in all the 12 nearby galaxies. This ApJ paper was published in 2013 (Liu et al. 2013a).

Cold interstellar dust

The galaxy pair NGC 1512/1510, consisting of a large, high-metallicity spiral and a low-metallicity, blue compact dwarf, offers us a good opportunity to compare similarities and contrast differences. We combined our new ASTE/AzTEC 1.1mm maps of the two galaxies (obtained by Profs. Min Y. Yun, Grant W. Wilson and the AzTEC group) with archival SINGS IRAC and MIPS images covering the wavelength range 3.6-160 μm and derived accurate dust masses through SED fitting by collaborating with Prof. Bruce Draine (Princeton). In both galaxies, a substantial fraction (>93%) of the total dust mass is in a cool dust component, with temperatures about 14-16 K for NGC 1512 and 15-24 K for NGC 1510. This is the first AzTEC paper on extended sources. The results were published in ApJ (Liu et al. 2010).