Supernova Casts Doubt on "Standard Candle"

Krishna Dagli writes, "A supernova more than twice as bright as others of its type has been observed, suggesting it arose from a star that managed to grow more massive than theoretically thought possible. The observation suggests that Type 1a supernovae may not be 'standard candles' — all having the same intrinsic luminosity — as previously thought. This could affect their use as probes of dark energy, the mysterious force causing the expansion of the universe to accelerate."

Comments

Oh really...

[tttonyyy]

"twice as bright as others of its type"

Obviously not a /. reader then. ;)

Re:Oh really...

[Lissajous]

"twice as bright as others of its type"

Obviously not a /. reader then. ;)

I don't get it.

The universe will out

[Oligonicella]

Models are just that, models. Change them when the universe shoves reality down your throat. Far too many people think that math defines the universe instead of describing it.

Re:The universe will out

[rocketman768]

I think you are exactly right. I am a mathematician. People should understand that all of mathematics is an abstract concept created by humans. Why does 2+3 = 5? Because we said it does...not because it is universally true. Sometimes (in the case of 'models'), we put some math together to attempt to explain what we see. As we discover new behaviors in whatever system we're looking at, we have to change the math. So, this article is about one of those instances.

Re:The universe will out

[KutuluWare]

I think his point would be more accurately expressed as this:

"Why is 2 + 3 = 5?"

Because the arbitrary definitions which we assigned to the symbols 2, 3, 5, +, and = happen to represent real-world concepts that exhibit the behavior that 2 + 3 = 5, and not because there is any abstract universal rule that "2 + 3 = 5" and we simply need to find real-world behavior to prove it. That is, the real-world behavior has always existed, but the mathematical language used to express it was invented by us and assigned to those behaviors specifically to make the mathematics true.

(Or something, it's early.)

--K

Re:The universe will out

[inviolet]

Because the arbitrary definitions which we assigned to the symbols 2, 3, 5, +, and = happen to represent real-world concepts that exhibit the behavior that 2 + 3 = 5, and not because there is any abstract universal rule that "2 + 3 = 5" and we simply need to find real-world behavior to prove it.

Quoted for truth. I want to elaborate (i.e. ramble) on it a bit . . .

Numbers are indeed a deductive system: they are true because they are defined to be true. They are true in all conceivable universes. This makes them useful but also hollow: they contain no empirical content, and hence are immune to all conceivable experimental results.

Nevertheless, they (and all other deductive symbols) can participate in inductive statements, such as "2 algae cells will combine with 3 fungi cells to produce 1 lichen".

Re:Canada-France-Hawaii telescope?

[Silver Sloth]

because

CFHT is a joint facility of:

* National Research Council of Canada (see also Herzberg Institute of Astrophysics),
* Centre National de la Recherche Scientifique of France (see also CNRS/INSU ), and
* University of Hawaii (see also UH/IfA ).

i.e. two national bodies and one local body. This is all on their website http://www.cfht.hawaii.edu/

Re:Canada-France-Hawaii telescope?

[$RANDOMLUSER]

That's how we Americans do geography. We know where Hawaii is, and we know that Canada is (vaguely) north of us, and France is somewhere in Asia.

This is a GOOD thing....

[trip11]

...as supernova are not well understood. First off I am not an astrophysicist, though I am a high energy physicist (and have taken some astro classes). One thing that has been discussed in nuclear classes I have taken is how little we understand just how a supernova functions at the atomic level. The number of competing effects going on during the collapse of a star is just amazing. You have gravitational pull, thermal pressure, rotational 'pressure', electromagnetic forces in a regular star. Now you start to collapse the star and you have to add in the transition of millions of individual nuclei becoming in effect one large nucleous as they all mearge. (not to mention the energy output from this). In effect the strong force comes into play along with the standard EM and gravitational forces. It gets much more complicated than that, but it has been several years since those classes.

So why do I think this is a 'good thing'? As the article speculates, it is likely that this supernova was different because of some rotational process or perhaps colliding stars, or some other exotic combination. This is exactly the sort of process that can be used as a test of supernova models to see how well they do. Over all I find this a very exciting observation and hopefully it produces more new science!

The observational evidence is surprisingly scanty

[StupendousMan]

I study supernovae for a living.

The Nature paper in which this work is published has a figure showing all the measurements of this supernova's brightness; you can see it on Nature's web site at

http://www.nature.com/nature/journal/v443/n7109/fi g_tab/nature05103_F1.html

There are four measurements near time of maximum light, in the red (r) and near-infrared (i) passbands. There are many more measurements starting about 15 days after maximum light in the rest frame, including some in a blue-green (g) passband. Here's what the researchers did to find the maximum brightness of this supernova, so that they could compare it to others:

    a) fit models based on the light curves of other supernovae to the r and i measurements,
                  and the late-time g measurements

    b) choose a different passband -- the greenish V passband of the Johnson-Cousins system,
                  which is closest to their own g passband (the one with no data at max light)

    c) use their models to estimate what the light curve in the V filter would have been

This can be a tricky business. Their major conclusion, that this supernova was more luminous than typical ones, is probably correct, but their claim that they can measure the peak magnitude in the V-band to an uncertainty of 6 percent seems a bit bold.

As the press release states, if atypical SNe are very rare, then this probably doesn't have any major impact on the use of Type Ia SNe in cosmology.

Re:more important than 'probes of dark energy'

[The_Wilschon]

I doubt it. Our actual measurements of dark energy won't come under much increased doubt. Although Type IA supernovae the first (IIRC) indicator of dark energy, we still have a number of other indicators. I was just as PASCOS 2006, and saw several talks on dark energy, where various quantities related to the acceleration of the universe were really overconstrained by about 4-5 different measurements. The only one I can recall at the moment is gravitational lensing. The neat thing is that although overconstraint has the possibility to show an inconsistency, it doesn't do so here. The measurements all line up at one point (well, a distribution around one point, but that distribution is quite nicely peaked in one location, indicating consistency.).

Similarly, Type IA SN are not the only mechanism by which we measure the age of the universe, so I'm not too concerned. The other reason I'm not too concerned is that the age of the universe was already in doubt. Another talk at PASCOS dealt with something else that I can't recall at the moment (curse my memory in the morning!) that cast into simultaneous doubt all or nearly all of our universe age indicators. IIRC, according to his talk, the universe could well be 20% older than our current best estimate.

Of course, since all these are not quite my field (I was at PASCOS for the particle physics), I can't answer for whether or not these guys were just crazies and all the cosmologists were ignoring them, or if these are serious problems that will be dealt with in the next few years. I'd be inclined, however, to assume that they were quite legit.