The physical nature of the 8 o'clock arc based on near-IR IFU spectroscopy with SINFONI★

Abstract

We present an analysis of near-infrared integral field unit spectroscopy of the 8 o'clock arc, a gravitationally lensed Lyman break galaxy, taken with SINFONI. We explore the shape of the spatially-resolved Hβ profile and demonstrate that we can decompose it into three components that partially overlap (spatially) but are distinguishable when we include dynamical information. We use existing B and H imaging from the Hubble Space Telescope to construct a rigorous lens model using a Bayesian grid based lens modelling technique. We apply this lens model to the SINFONI data cube to construct the de-lensed Hβ line-flux velocity and velocity dispersion maps of the galaxy. We find that the 8 o'clock arc has a complex velocity field that is not simply explained by a single rotating disk. The Hβ profile of the galaxy shows a blue-shifted wing suggesting gas outflows of ∼ 200 km s −1 . We confirm that the 8 o'clock arc lies on the stellar mass-oxygen abundance-SFR plane found locally, but it has nevertheless significantly different gas surface density (a factor of 2-4 higher) and electron density in the ionized gas (five times higher) from those in similar nearby galaxies, possibly indicating a higher density interstellar medium for this galaxy.

LBT/ARGOS adaptive optics observations of z ∼ 2 lensed galaxies

“…1). The Hβ flux map of the arc obtained by Shirazi et al (2014) using the IFU instrument SINFONI shows relative fluxes between the three images which are consistent with those in the LBT/ARGOS spectrum. Therefore, we can confirm that differential slit losses are not at the origin of the different fluxes associated with the three knots.…”

“…Gravitational lens models of the arc have been presented by Dessauges-Zavadsky et al (2011) and Shirazi et al (2014); the three images are related to a unique emitting region in the source plane, with A2 inverted with respect to the other two images. For consistency with the analysis of J1038 and J1958, for this galaxy we also compared the integrated spectra extracted from the three different knots, deriving, as expected, consistent nonparametric measurements (within 1σ), with W80 ≈ 280 km/s and V10 ≈ −200 km/s (see Table 2).…”

“…In support of the outflow picture, Finkelstein et al (2009) andDessauges-Zavadsky et al (2010) reported that ISM absorption lines in the rest-frame UV spectrum of J0022 are blueshifted with respect to the stellar photospheric features of about 120 − 160 km/s. Moreover, Shirazi et al (2014), modelling the Hβ spatially resolved emission line velocity map in the source plane, showed that the velocity field cannot be fitted by a simple rotating disk and that more complex kinematics, possibly associated with outflows, are present.…”

“…All stellar masses are corrected for the lensing magnification factors reported in Column (4). Column (6): SFR from the Hα flux corrected for extinction and lensing magnification using Kennicutt 1998relation (J1958 and J1343 from Leethochawalit et al 2016; J0022 from Shirazi et al 2014, using the Hβ flux). Lensing-corrected SFR of J1110 from Johnson et al (2017) and obtained by SED fitting analysis, nominally corrected from a Chabrier to a Salpeter IMF.…”

“…Column (13): estimate of the spatial resolution in the source plane. For J1038, J0022 and J1110 we refer to the A2 knots, using the magnifications derived by Jones et al (2013), Shirazi et al (2014) and Johnson et al (2017), respectively. fects they may have on the global baryon cycle, host galaxy and outflow properties have been studied looking for causal relationships.…”