Some recent surveys aimed at detecting Supernovae (SNe) in the optical have found that starburst galaxies do not show evidence for an enhanced SN rate, with respect to quiescent galaxies (Richmond et al. 1998). This result is puzzling, since the presence of a burst of star formation should boost the SN rate. A similar result was obtained by Navasardyan et al. (2001) who looked for SNe in interacting galaxies, finding no evidence for correlations between SN rate and galaxy properties.
A possible explanation for the shortage of SNe in star forming galaxies is that a large fraction of the SNe is obscured at optical wavelenghts by a large amount of dust, which generally affects starburst systems. The effect of dust extinction can be vastly reduced by observing in the near infrared: at 2 µm the extinction is about a factor of ten lower than in the optical V band, allowing for a much deeper penetration into the dusty star-forming regions. Therefore a survey for SNe in the infrared would allow to detect the obscured SNe which are possibly missed in optical monitoring campaigns.

In late 1999 our team started a K'-band (2.1 µm) monitoring campaign of a sample of 46 Luminous Infrared Galaxies (LIRGs, L(FIR) > 10¹¹ L_SUN) aimed at detecting obscured SNe in starburst galaxies. The survey  was carried on with the NTT-ESO, TNG (Galileo National Telescope) and Kuiper/Steward telescopes. The most important differences from the other monitoring programs are the use of 4-m class telescopes with higher resolution and deeper limiting magnitude, and the selection of a sample of galaxies with higher star formation rate, in order to have a larger number of expected events. During the monitoring, using an optimal image subtraction method, we detected 4 SNe, two of which were discovered by our group: SN1999gw and SN2001db, the first SN detected in the near-IR which has recived a spectroscopic confirmation and one of the more extincted events ever observed.

Work in progress

In order to obtain a better resolution in the nuclear region of the galaxies, where the bulk of the starburst activity is supposed to be, we obtain time in Cycle 12 with the NICMOS infrared camera on Hubble Space Telescope to reobserve 37 galaxies of our sample already observed once with the same instruments. We hope to exploiting its sensitivity and angular resolution to detect from space nuclear obscured SNe which might have been missed by ground-based surveys. We will study the behaviour of the detected events also at radio wavelenghts with VLA. Another monitoring of a sample of nearer galaxies with lower star formation rate is started at the TIRGO telescope.

What have we found?

The number of events observed is about an order of magnitude higher than expected from the relation between the B band luminosity of the galaxies and the SN rate, based on the optical surveys. This results highlights the capability of near-IR observation in detecting obscured SNe which are missed by the classical optical surveys. Nevertheless, the inferred SN rate is still four times lower than that estimated from the far-infrared (FIR) luminosity of the galaxies of the sample. A possible explanation for the shortage of near-IR SNe is the presence of a dust extinction A_V > 30 mag, which would make SNe heavily obscured even in the near-IR. See our result page or the papers by our group for further details. Other groups are currently carring on monitoring in the near infrared to detect obscured supernovae. You can find out more following  our links and the list of papers on infrared SNe.

Near-infrared monitoring is needed not only to compare quiescent and starburst galaxies, but also to obtain a complete estimate of the local SN rate to be compared with the rates at high redshift. Finally, near-infrared searches are needed to detect a significant number of SNe in starburst galaxies, events which are  expected to have peculiar properties: the circumstellar medium around them is expected to be very dense because they occur in molecular clouds and the mass-loss rates of the progenitor stars is expected to be higher because of the higher metallicity. The effects of this peculiar environment can be studied only when a significant number of events are discovered.

KAOSS Home Page

A brief introduction on Supernovae and Supernova types

The detected Supernovae

The Infrared Supernova Rate and other Results

The detection technique

Logbook of the Ulirg  ground-based observations and the galaxy sample

NICMOS observations

 Radio Followup with VLA

Logbook of the Lirg TIRGO observations and related galaxy sample

Our publications

Other useful readings and links

Members of our team

Arcetri Home Page