Dorado group of galaxies. Mapping star formation with FUV imaging from UVIT

Context. We are investigating the star formation properties of galaxies in Dorado, an actively evolving group in the southern hemisphere with a large fraction of both early and late-type galaxies showing optical signatures of interaction/merger events. Aims. Our previous Hα+[N II] work suggested that star formation is depressed in spirals while is not yet shut down in early-type galaxies. However, such star formation estimates may be prone to large uncertainties, in particular for early-type galaxies. Here, we aim at mapping the star formation on longer times scales (100 Myrs) than our ha study (10 Myrs) to deep our analysis. Method. We imaged with the filter CAF2-1, 1300-1800 A (FUV.CAF2) of the Ultraviolet Telescope UVIT on board of ASTROSAT satellite, the group backbone previously observed in Hα+[N II]. These observations include NGC 1533, IC 2039, IC 2038, NGC 1536, NGC 1546, NGC 1549, [CMI2001]4136-01, NGC 1553, IC 2058, PGC 75125, NGC 1566, and, the first FUV view of NGC~1596 and NGC 1602. Results. Most of the recent star formation is found at the periphery of the Dorado group where HI-rich galaxies are located. The FUV.CAF2 morphology of our targets shows young stellar populations in rings and tidal distortions. The Sersic index we derive from 1D fit of the luminosity profiles suggests hat the FUV.CaF2 emission lies in a disk also in early-type galaxies. The star formation rate derived ranges from 0.004+-0.001 to 2.455+-0.027 Mo/ yr-1 of [CMI2001]4136-01 and of NGC 1566, respectively. The ratio betweeen FUV and Ha estimates is 1 for late-types, excluding IC 2058 seen edge-on, but it is about 15 times higher, on average, for early-type galaxies. Conclusions. The disk structure of the FUV.CAF2 emission discovered in all the early-types galaxies points toward dissipative processes and wet merging events. We believe that ``rejuvenation'' episodes in ETGs cannot feed a constant star formation for 100 Myr or longer, as required by comparing SFR estimates using different indicators. The systematic dependence of the Ha/UV ratio we derived may be produced by temporal variations in SFRs.