We compare multi-wavelength star formation rate (SFR) indicators out to z ∼ 3 in the GOODS-South field. Our analysis uniquely combines U -to-8µm photometry from FIREWORKS, MIPS 24 µm and PACS 70, 100, and 160 µm photometry from the PEP survey, and Hα spectroscopy from the SINS survey. We describe a set of conversions that lead to a continuity across SFR indicators. A luminosityindependent conversion from 24 µm to total infrared luminosity yields estimates of L IR that are in the median consistent with the L IR derived from PACS photometry, albeit with significant scatter. Dust correction methods perform well at low to intermediate levels of star formation. They fail to recover the total amount of star formation in systems with large SF R IR /SF R UV ratios, typically occuring at the highest SFRs (SF R UV +IR 100 M ⊙ /yr) and redshifts (z 2.5) probed. Finally, we confirm that Hα-based SFRs at 1.5 < z < 2.6 are consistent with SF R SED and SF R UV +IR provided extra attenuation towards HII regions is taken into account (A V,neb = A V,continuum /0.44). With the cross-calibrated SFR indicators in hand, we perform a consistency check on the star formation histories inferred from SED modeling. We compare the observed SFR-M relations and mass functions at a range of redshifts to equivalents that are computed by evolving lower redshift galaxies backwards in time. We find evidence for underestimated stellar ages when no stringent constraints on formation epoch are applied in SED modeling. We demonstrate how resolved SED modeling, or alternatively deep UV data, may help to overcome this bias. The age bias is most severe for galaxies with young stellar populations, and reduces towards older systems. Finally, our analysis suggests that SFHs typically vary on timescales that are long (at least several 100 Myr) compared to the galaxies' dynamical time.

Herschel is an ESA space observatory with science instruments provided by European-led Principal Investigator consortia and with important participation from NASA.

Context. Star formation rate (SFR), metallicity, and stellar mass are among the most important parameters of star-forming (SF) galaxies characterizing their formation and evolution. They are known to be related to each other both at low and high redshift in the mass-metallicity, mass-SFR, and metallicity-SFR relations. Aims. We demonstrate the existence of a plane in a 3D parameter space defined by the axes SFR [log (SFR)(M yr −1 )], gas metallicity [12 + log (O/H)], and stellar mass [log (M star /M )] of SF galaxies. Methods. We used SF galaxies from the "main galaxy sample" of the Sloan Digital Sky Survey-Data Release 7 (SDSS-DR7) in the redshift range 0.04 < z < 0.1 and r-magnitudes between 14.5 and 17.77. Metallicities, SFRs, and stellar masses were taken from the Max-Planck-Institute for Astrophysics-John Hopkins University (MPA-JHU) emission-line analysis database. Results. From a final sample of 32 575 galaxies, we find for the first time a fundamental plane for field galaxies relating the SFR, gas metallicity, and stellar mass for SF galaxies in the local universe. One of the applications of this plane would be to estimate stellar masses from SFR and metallicity. High redshift data from the literature at redshift ∼0.85, 2.2, and 3.5, do not show evidence of evolution in this fundamental plane.

We present results from the deepest Herschel-Photodetector Array Camera and Spectrometer (PACS) far-infrared blank field extragalactic survey, obtained by combining observations of the Great Observatories Origins Deep Survey (GOODS) fields from the PACS Evolutionary Probe (PEP) and GOODS-Herschel key programmes. We describe data reduction and the construction of images and catalogues. In the deepest parts of the GOODS-S field, the catalogues reach 3σ depths of 0.9, 0.6 and 1.3 mJy at 70, 100 and 160 μm, respectively, and resolve ∼75% of the cosmic infrared background at 100 μm and 160 μm into individually detected sources. We use these data to estimate the PACS confusion noise, to derive the PACS number counts down to unprecedented depths, and to determine the infrared luminosity function of galaxies down to L IR = 10 11 L at z ∼ 1 and L IR = 10 12 L at z ∼ 2, respectively. For the infrared luminosity function of galaxies, our deep Herschel far-infrared observations are fundamental because they provide more accurate infrared luminosity estimates than those previously obtained from mid-infrared observations. Maps and source catalogues (>3σ) are now publicly released. Combined with the large wealth of multi-wavelength data available for the GOODS fields, these data provide a powerful new tool for studying galaxy evolution over a broad range of redshifts.

We study the evolution of the dust temperature of galaxies in the SFR−M * plane up to z ∼ 2 using far-infrared and submillimetre observations from the Herschel Space Observatory taken as part of the PACS Evolutionary Probe (PEP) and Herschel Multi-tiered Extragalactic Survey (HerMES) guaranteed time key programmes. Starting from a sample of galaxies with reliable star-formation rates (SFRs), stellar masses (M * ) and redshift estimates, we grid the SFR−M * parameter space in several redshift ranges and estimate the mean dust temperature (T dust ) of each SFR-M * −z bin. Dust temperatures are inferred using the stacked far-infrared flux densities (100-500 μm) of our SFR-M * −z bins. At all redshifts, the dust temperature of galaxies smoothly increases with rest-frame infrared luminosities (L IR ), specific SFRs (SSFR; i.e., SFR/M * ), and distances with respect to the main sequence (MS) of the SFR−M * plane (i.e., Δ log (SSFR) MS = log [SSFR(galaxy)/SSFR MS (M * , z)]). The T dust −SSFR and T dust − Δ log (SSFR) MS correlations are statistically much more significant than the T dust −L IR one. While the slopes of these three correlations are redshiftindependent, their normalisations evolve smoothly from z = 0 and z ∼ 2. We convert these results into a recipe to derive T dust from SFR, M * and z, valid out to z ∼ 2 and for the stellar mass and SFR range covered by our stacking analysis. The existence of a strong T dust −Δ log (SSFR) MS correlation provides us with several pieces of information on the dust and gas content of galaxies. Firstly, the slope of the T dust −Δ log (SSFR) MS correlation can be explained by the increase in the star-formation efficiency (SFE; SFR/M gas ) with Δ log (SSFR) MS as found locally by molecular gas studies. Secondly, at fixed Δ log (SSFR) MS , the constant dust temperature observed in galaxies probing wide ranges in SFR and M * can be explained by an increase or decrease in the number of star-forming regions with comparable SFE enclosed in them. And thirdly, at high redshift, the normalisation towards hotter dust temperature of the T dust −Δ log (SSFR) MS correlation can be explained by the decrease in the metallicities of galaxies or by the increase in the SFE of MS galaxies. All these results support the hypothesis that the conditions prevailing in the star-forming regions of MS and far-above-MS galaxies are different. MS galaxies have star-forming regions with low SFEs and thus cold dust, while galaxies situated far above the MS seem to be in a starbursting phase characterised by star-forming regions with high SFEs and thus hot dust.