View Full Version : progenitors...type 1a

2013-Dec-29, 06:38 PM
Observational clues to the progenitors of type 1a supernovae. Although considered the dark horse early in the model race, the DD (double degenerate) model....two white dwarfs, two neutron stars, or a WD and a NS....looks more promising. pete

2014-Jan-05, 10:55 PM
A DD result sure isn't intuitive. Wouldn't they both puff at some point, regardless of which model, thus favoring a blow of the first WD? Do textbook publishers know about this? *wink*

2014-Nov-14, 08:16 PM
http://arxiv.org/abs/1411.1064 The Rising Light Curves of Type Ia Supernovae
It has long been assumed, but never established, that the rise-time of supernova type Ia are proportional to the rate that these same supernova diminish. It now appears that this is not the case, and this very important.

First of all, co-authors on this paper are heavy-hitters in supernova circles: Nugent, Sullivan and Howell have all authored key and seminal papers.

Second, the light-curve templates developed in the late 1980s through the early 2000's all depend upon light- curve rise-times proportionately mirroring the drop-off of the light curve to roughly half the peak magnitude of the event. This fundamental assumption is wrong.

Supernova are not monitored continuously: we get snap shots as the light-curves evolve over time. For many events, there is very little rise-time information. This is because the events are generally only first-observed as they approach peak magnitude; so there has always been liberal extrapolation between the light-curve tail-off; and the interpretation of the light-curve rise-time.

In order to use supernova Ia as standard candles; each light-curve is monitored and then compared with templates to determine which 'stretch factor' is used to gauge the absolute magnitude of the event. But the 'stretching' has always assumed that the slower the evolution of the 'tail', the slower the rise-time; which should correlated with larger primary explosions.

I don't think anyone saw this coming. I didn't. If the rise-times are highly variable; and in a way that does not correlate with the balance of the light curve, the fundamental physical assumption that larger events produce longer and brighter light-curves is edgy. All of our experience with nuclear explosions tells us that the net overall energy is closely related to the event rise-time. Without this correlation, the assumption that the magnitude of the peak correlates with the tail-off time is fundamentally unsound.

Our study of low-redshift supernova do indicate that the slope of certain bands of a light-curve correlate well with magnitude. But this is a phenomena-driven observation: The seemingly random, (possibly anti-correlating) variability in the rise-times flies in-the-face of the 'Chandrasekarian' assumption that a supernova Ia event is driven by mass accretion to a limit. This means there may be no fundamentally sound reason for segregating type Ia supernova events into "single detonation" or "double detonation" scenarios. The lack of correlation is most compatible with the "double-white dwarf" scenario; but frankly, it does not jive well with either.

What does this mean for using supernova Ia used for cosmological purposes? Even if there is no theory behind the science; the calibration of the light curves is phenomenologically derived; so there is still merit in using supernova as standard candles.

There are problems. There have been seriously discrepant observational data in the ultra-violet bandwidths, especially at very high redshifts. Reddening of light-curves has also been discounted. Supernova researches have dismissed the lack of correlation across all bandwidths; and they have justified this on the base of the 'tightness' of the data fits and low residuals using templating methods. But 'tightness' and low residuals can also occur when data is selectively fit to create 'tightness' and low residuals. It is a given that the low correlation of peak magnitude with rise-time documented in this paper smothers the 'tight correlation' as justification for ignoring other systemics: The 'tight correlation' was not there in the first place.

And what about the other systemics? The most distant events generally have brighter UV signatures when compared with more local events. This can no longer be ignored, or thought to be insignificant because of peaceful, theory-compatible derivations of the data: We must figure out when we are comparing apples with apples; and then let the data drive the conclusions, not the theory.

2014-Nov-17, 01:50 PM
A few SN Ia were caught in a monitoring program during the Kepler main mission (wish they'd get them properly written up in papers...). A couple of early-time light curves are shown in this white paper (http://keplerscience.arc.nasa.gov/docs/WhitePapers/Olling_WhitePaper.pdf) making the case to do more such surveys during the K2 extension. Their main message from these few is that they do not see any distinct shock signatures in the light curves.

2014-Nov-17, 04:39 PM
A few SN Ia were caught in a monitoring program during the Kepler main mission (wish they'd get them properly written up in papers...).

That makes two of us. If I'd known, I would have taken pictures of their slides during their presentation at the AAS in January.