More on the Supernova That Wasn't Until It Was

At the end of last week I wrote about a very interesting new paper that appeared on the arXiv: "The Unprecedented Third Outburst of SN 2009ip: A Luminous Blue Variable Becomes a Supernova" by Jon Mauerhan et al. If you didn't read that post, please go ahead and do so now, as I have some new details about the supernova and the super-fast write-up by the authors.

SN 2009ip. Credit: J.Mauerhan et al.

I contacted the first three authors of the paper and asked for comments, and received the following good replies. First, Alex Filippenko from University of California, Berkeley, acknowledged the work of the first two authors by saying:

"Nathan and Jon were amazing in getting all this together quickly,
but we've long had an interest in SN 2009ip and had been monitoring
its latest developments with great interest."

Then Nathan Smith from the University of Arizona added some details about the significance of the result:

"I would just add one thing that Alex alluded to.  There have been several cases (SN1987A and about a dozen others) where astronomers have seen a nearby supernova and then gone back to data archives and identified the star that exploded … after the fact.  This is the first time in the history of astronomy that we were actually studying the star before it exploded and had in fact written papers about it already.  In other words, seeing this thing go supernova could be seen as dumb luck - and admittedly there is luck involved here - but there is also a very good reason why we were interested in this particular star.  For the last 6 or 7 years, I've been going around at conferences and giving colloquium talks about how we think Luminous Blue Variable (LBVs) explode to make Type IIn supernovae (*).  I've gotten a lot of flack for this from theorists with stellar evolution models that don't allow this to be the case."

Now that they appear to have witnessed an LBV destroyed by a supernova, Nathan and the rest of the team have good reason to feel excited by the new observations and vindicated by them.

The light curve of SN 2009ip. The brightness increases as the magnitude decreases, so the 3rd outburst is brighter than the 1st and 2nd outbursts. Credit: J.Mauerhan et al.

Finally, the first author, Jon Mauerhan from the University of Arizona, added some comments about the initial faintness of the supernova and the remarkable speed of the observing team in publishing data that was fresh off the telescope:

"We think the latest brightening of SN 2009ip is the result of the supernova blast wave hitting circumstellar material ejected by the LBV during its prior eruptions. So, it remains to be explained why the supernova remained so faint initially (Mv=-14.5) when we first detected the high velocity spectral features. If circumstellar material were not present to light things up, would SN 2009ip have remained at such a faint level? This is why we brought up the subject of "failed supernovae" in the paper. 

An excellent group effort definitely helped in getting this out fast. We began writing the paper after we noted the high-velocity features in our first bok spectrum on Sept 17th. So the figure-making programs were written, and ready to accept new data as it came in. 

But as you suspected, a tolerance for "lack of sleep" was definitely involved in getting the data reduced and into the paper so quickly, and I appreciate you recognizing that. For example, I reduced the latest bok spectrum at dawn in the observatory dorm kitchen after we closed up, as well as the photometry on the most recent images that were also rapidly reduced and sent to me by Alex's team. After that, I was able to squeeze in a 2 hour nap before waking up to iterate on some new text with Nathan and make the arXiv submission deadline before the weekend. It was a very fun and exciting but rather exhausting couple of days."

In a further comment provided by Jon he explained that the early faintness of SN 2009ip might be explained by shells of material ejected in previous outbursts obscuring the light from the supernova. Much remains to be learned about the event, since it hasn't been monitored for along, as they point out in the paper:

"At the time of writing, this supernova event is just beginning, so we do not know the detailed nature of the supernova event or how bright it will become."

These results for SN 2009ip are one of many exciting results obtained over the last few years in supernova studies, including the discovery of SN 2006gy, another Type IIn event and for a time the most luminous supernova known. In their paper, Nathan Smith and co-authors, including Alex Filippenko, argued that SN 2006gy may have been a "pair instability supernova", where the core of the massive progenitor star produces so much gamma ray radiation that some of the energy from the radiation is converted into particle and anti-particle pairs. The resulting drop in energy causes the star to collapse under its own huge gravity. After this violent collapse, runaway thermonuclear reactions ensue and the star is obliterated, spewing the remains into space. This is a different type of event from most supernovas, which usually occur when massive stars exhaust their fuel and collapse under their own gravity, followed by an explosion.

An illustration explaining the pair instability trigger for a supernova. Credit: NASA/CXC/M.Weiss

A pair instability supernova is a spectacular way for a star to end its life, as they should be the most powerful thermonuclear explosions in the universe, if they exist. One reason they're important is that all of the elements produced inside the massive star are blasted into space, available later to form into stars and planets. If a black hole forms in the collapse, some or all of this material is locked up forever.

The idea of a pair instability supernova for SN 2006gy is testable, because it made specific predictions about how the light curve would behave over several years, based on the radioactive decay of nickel and cobalt.  Further studies showed that the light curve was not dropping at the predicted rate, which means that the pair instability model probably does not apply.

Another supernova, SN 2007bi, has been claimed to be a good candidate for a pair instability supernova, by a team that includes Alex Filippenko as a co-author. However, other models have been suggested by different groups, including a core-collapse model and a model for a strongly magnetic neutron star, so the idea that SN 2007bi was a pair instability supernova is not without controversy.

How is this discussion relevant for SN 2009ip? The authors are not arguing that this supernova was a pair instability event, but they do consider the possibility that the outbursts seen in 2009 and 2010 were driven by the pulsational pair instability. This is triggered by the same pair instability that should lead to a bright supernova, but the explosion isn't powerful enough to tear the star apart and only the outer part is ejected. The mass of the destroyed star is close to that predicted by theory for the range where the pulsational pair instability should cause pre-supernova outbursts.

I am looking forward to hearing about what SN 2009ip does next and to future supernova discoveries.


(*) A Type IIn supernova is caused by the collapse of a massive star. The "II" means that hydrogen was seen in the spectrum and the "n" means that narrow emission lines from hydrogen are present, often indicating that debris from the explosion is interacting strongly with material surrounding the destroyed star.