The Core-Collapse Supernova Rate in Arp299 Revisited

We present a study of the CCSN rate in nuclei A and B1 of the luminous infrared galaxy Arp299, based on 11 years of Very Large Array monitoring of their radio emission at 8.4 GHz. Significant variations in the nuclear radio flux density can be used to identify the CCSN activity in the absence of high-resolution very long baseline interferometry observations. In the case of the B1-nucleus, the small variations in its measured diffuse radio emission are below the fluxes expected from radio supernovae, thus making it well-suited to detect RSNe through flux density variability. In fact, we find strong evidence for at least three RSNe this way, which results in a lower limit for the CCSN rate of 0.28 +/- 0.16 per year. In the A-nucleus, we did not detect any significant variability and found a SN detection threshold luminosity which allows only the detection of the most luminous RSNe known. Our method is basically blind to normal CCSN explosions occurring within the A-nucleus, which result in too small variations in the nuclear flux density, remaining diluted by the strong diffuse emission of the nucleus itself. Additionally, we have attempted to find near-infrared counterparts for the earlier reported RSNe in the Arp299 nucleus A, by comparing NIR adaptive optics images from the Gemini-N telescope with contemporaneous observations from the European VLBI Network. However, we were not able to detect NIR counterparts for the reported radio SNe within the innermost regions of nucleus A. While our NIR observations were sensitive to typical CCSNe at 300 mas from the centre of the nucleus A, suffering from extinction up to A_v~15 mag, they were not sensitive to such highly obscured SNe within the innermost nuclear regions where most of the EVN sources were detected. (abridged)
Comment: 12 pages, 4 figures and 7 tables. Accepted for publication in MNRAS