No abstract

A study of [S III]λλ9096, 9532 emitters at z = 1.34 and 1.23 is presented using our deep narrow-band H 2 S 1 (centered at 2.13 µm) imaging survey of the Extended Chandra Deep Field South (ECDFS). We combine our data with multi-wavelength data of ECDFS to build up spectral energy distributions (SEDs) from the U to the K s -band for emitter candidates selected with strong excess in emitters in our sample are dominated by star-forming galaxies (SFGs) with stellar mass 8.7 < log(M/M ⊙ ) < 9.9.

It remains challenging to detect the low surface brightness structures of faint high-z galaxies, which is key to understanding the structural evolution of galaxies. The technique of image stacking allows us to measure the averaged light profile beneath the detection limit and probe the extended structure of a group of galaxies. We carry out simulations to examine the recovery of the averaged surface brightness profile through stacking model HST/ACS images of a set of galaxies as functions of Sérsic index (n), effective radius (R e ) and axis ratio (AR). The Sérsic profile best fitting the radial profile of the stacked image is taken as the recovered profile, in comparison with the intrinsic mean profile of the model galaxies. Our results show that, in general, the structural parameters of the mean profile can be properly determined through stacking, although systematic biases need to be corrected when spreads of R e and AR are counted. We find that Sérsic index is slightly overestimated and R e is underestimated at AR < 0.5 as the stacked image appears to be more compact due to the presence of inclined galaxies; the spread of R e biases the stacked profile to have a higher Sérsic index. We stress that the measurements of structural parameters through stacking should take these biases into account. We estimate the biases in the recovered structural parameters from stacks of galaxies when the samples have distributions of R e , AR and n seen in local galaxies.

We present the first attempt at measuring the production rate of tidal dwarf galaxies (TDGs) and estimating their contribution to the overall dwarf population. Using HST/ACS deep imaging data from GOODS and GEMS surveys in conjunction with photometric redshifts from COMBO-17 survey, we performed a morphological analysis for a sample of merging/interacting galaxies in the Extended Chandra Deep Field South and identified tidal dwarf candidates in the rest-frame optical bands. We estimated a production rate about 1.4 × 10 −5 per Gyr per comoving volume for long-lived TDGs with stellar mass 3 × 10 8−9 M ⊙ at 0.5 < z < 1.1. Together with galaxy merger rates and TDG survival rate from the literature, our results suggest that only a marginal fraction (less than 10%) of dwarf galaxies in the local universe could be tidallyoriginated. TDGs in our sample are on average bluer than their host galaxies in the optical. Stellar population modelling of optical to near-infrared spectral energy distributions (SEDs) for two TDGs favors a burst component with age 400/200 Myr and stellar mass 40%/26% of the total, indicating that a young stellar population newly formed in TDGs. This is consistent with the episodic star formation histories found for nearby TDGs.