FALLOUT: The explosive remnant of the SN 1006 supernova can still be seen in a number of wavelengths, as demonstrated in this composite image of optical, x-ray and radio observations.
Image: X-ray: NASA/CXC/Rutgers/G.Cassam-Chenaï, J.Hughes et al.; Radio: NRAO/AUI/NSF/GBT/VLA/Dyer, Maddalena & Cornwell; Optical: Middlebury College/F.Winkler, NOAO/AURA/NSF/CTIO Schmidt & DSS
-
The Best Science Writing Online 2012
Showcasing more than fifty of the most provocative, original, and significant online essays from 2011, The Best Science Writing Online 2012 will change the way...
Read More »
A type Ia supernova is perhaps the ultimate combination of insult and injury—a star steals material from a companion star, reaches critical mass, becomes unstable, and then unleashes a nuclear blast powerful enough to decimate or destroy its already diminished victim.
The culprit in these cases is clear: type Ia supernovae arise from the cataclysmic explosions of small, dense stars known as white dwarfs. But the victim's identity is clouded, limiting the precision of cosmological distance estimates that rely on these luminous beacons as markers. Traditionally, scientists believed that with the victims were sunlike main-sequence stars or swollen giant stars. But recent studies have pointed to a major role for a lesser known mechanism—pairings of two white dwarfs in which one cannibalizes its orbital companion before exploding as a supernova.
Now a study in the September 27 issue of Nature bolsters the latter argument, concluding that only a small minority of type Ia supernovae stem from main-sequence or giant stars. (Scientific American is part of Nature Publishing Group.) Supernovae driven by double white dwarfs, then, could be the rule rather than the exception.
Across the universe, type Ia supernovae are abundant, but few have occurred in the Milky Way in recorded human history. Supernova 1006, so named for the year A.D. in which it was first seen on Earth, was one such rare occurrence. But its origins remain unclear. So Jonay González Hernández, of the Institute of Astrophysics of the Canary Islands and the University of La Laguna in Spain, and his colleagues looked for remains of the companion star from which the ill-fated white dwarf siphoned material before exploding.
Supernova 1006 exploded 7,100 light-years from Earth, so any surviving relic of the blast should lie at about that distance. But the researchers found only a small handful of stars near the supernova site, and all of them were swollen red-giant stars. "Giant stars are not predicted to be companion stars of progenitors of type Ia supernovae, since the impact of the violent explosion would remove all the envelope of the giant star," leaving only a smaller white dwarf–esque star behind, González Hernández explains.
The lack of a surviving companion seems to rule out any large star as a partner, because the core of such a star should have weathered the blast and should remain visible today. So the companion could well have been another white dwarf, which would have left no trace. In conjunction with other, mostly fruitless searches for supernova survivors, the researchers estimate that fewer than 20 percent of type Ia supernovae originate from the classically assumed scenario of a white dwarf sucking matter away from a normal (non–white dwarf) companion star.
Not all researchers agree. Astronomer Andrew Howell of the Las Cumbres Observatory Global Telescope Network in Santa Barbara, Calif., notes that he and his colleagues recently found evidence for just such a "normal" companion for a type Ia supernova discovered in 2011 in a galaxy some 675 million light-years away. He calls the 20 percent claim in the new Nature paper "a vast overstatement," noting that a normal star somewhat smaller than the sun also would not leave any detectable traces of itself and would fit the bill for the companion to supernova 1006. In Howell's view, though, the 20 percent figure might apply to red-giant progenitors specifically.
So the actual percentage of type Ia supernovae produced by various stellar combinations may remain in dispute. But if anything seems clear from the recent studies on the origins of type Ia supernovae, it is that these blasts are hardly homogeneous in terms of their progenitor stars. Howell says that such variety won't compromise the idea of dark energy, the mysterious entity driving the accelerating expansion of the universe, which was first identified in studies that used the glow of distant type Ia supernovae to trace cosmic distance. The evidence for dark energy, he says, depends on the relative brightnesses of supernovae "and being able to calibrate them using the colors of the supernovae and the shapes of their lightcurves." Unraveling the progenitors of these cosmic beacons—and, in turn, their intrinsic brightness—would, however, help researchers improve the precision of their cosmological measurements.
Rights & Permissions
See what we're tweeting about
30 Comments
Add CommentAstronomer Andrew Howell's assessment seems completely unfounded.
Reply | Report Abuse | Link to thisThe value of type Ia supernovae (SNe) for use as 'standard candles' in accurately determining distance is contingent on the presumption that they all accrete material from a companion (red giant) until they reach the Chandrasekhar limit of about 1.38 Solar masses, then they supernova. Since they were thought to all supernova at about 1.38 Solar masses they were presumed to all produce a consistent peak emission luminosity. Because their source luminosity was though to be identical, the difference between the source luminosity and their observed luminosity could be used to very accurately determine the distance that light had traversed.
If the actual source 'peak emission luminosity' is not precisely known because the mass of the supernova is variable then type Ia SNe are not reliable standard candles.
As I understand, the quote attributed to Howell discussing light curves refers to a chart of luminosity over time - used to identify the peak emission period, since, even for the accretion model of type Ia SNe, only the peak emission luminosity is consistent enough to use to estimate distance.
"Unraveling the progenitors of these cosmic beacons—and, in turn, their intrinsic brightness—would, however, help researchers improve the precision of their cosmological measurements" is a gross understatement, since if accretion model type Ia SNe cannot be definitively distinguished from variable mass type Ia SNe then none can be used for standard candles.
At a minimum, this uncertainty in the analyses used to infer that universal expansion began accelerating about 5 billion years ago should require critical reexamination of that research.
@jtdwyer Others can speak to this even more knowledgeably than I, but I thought I would add a comment in response to your concerns. What Andy was trying to highlight is that the astronomy community has developed a more thorough understanding of the properties of these explosions, enabling better understanding of the scope and scale of the uncertainties.
Reply | Report Abuse | Link to thisThe community has collected thousands (I think) of Type Ia supernovae over the last decade+. By looking at the way their brightness changes over time (lightcurves), their maximum brightnesses (both apparent and absolute), and their spectra (brightness as a function of color/wavelength/frequency), there's now a good understanding of the variation in absolute brightness that Type Ias can produce AND the way that those variations are related to other measurable properties of the explosions.
For example, the main way that Type Ia's are "standardizable" (as opposed to standard) brightness measures is that the length of time it takes for the brightness to decline from its peak value to some well-defined fraction of that peak is directly correlated with the absolute brightness of the peak. You might ask, how do we know the absolute values, since we can only directly measure relative values. For that we must rely on the fact that we know the distance (within some uncertainty) to Type Ias that reside in galaxies where we can measure distance via other methods (e.g., by using Cepheid variable stars).
Beyond rate of decline being correlated with brightness, we're also learning to distinguish Type Ias that are outliers in brightness based on their spectra and the shape (beyond just decline rate) of their lightcurves. These things allow astronomers to either (a) justifiably remove outliers from samples used to calculate the distances that lead to our conclusion about the accelerating expansion of the universe, or (b) account for the uncertainty in distance that these outliers will impose on their conclusions. The latter requires the large samples being collected because one needs to know the rate of these outliers relative to "normal" Type Ias. This work is ongoing.
I don't mean to imply there are no uncertainties in using Type Ias as probes for cosmology. There certainly are, and they are a very active area of research. But as Andy points out, uncertainty about the "parents" of Type Ias does not discount the accelerating expansion of the universe because accounting for the uncertainty in each individual distance doesn't change the broader conclusion.
That sounds reassuring. However, much of what is now known about type Ia SNe has been determined since the 1999 type Ia SNe studies concluding that the expansion of the universe is accelerating. I'm not even sure that the quick 'double degenerate' white dwarf merger scenario was known at all in 1999.
Reply | Report Abuse | Link to thisA Nature Letter published today, "No surviving evolved companions of the progenitor of SN 1006", doi:10.1038/nature11447, http://www.nature.com/nature/journal/v489/n7417/full/nature11447.html
concludes its abstract with the statement:
"In combination with previous results, our findings indicate that fewer than 20 per cent of type Ia supernovae occur through the single-degenerate channel."
As I understand, the consistent peak emission luminosity single degenerate accretion model was thought to represent the vast majority of type Ia SNe until very recently.
If the 1999 'accelerating universe' studies falsely presumed that all their sampled type Ia SNe were produced at about 1.4 Solar masses in accordance with the single degenerate, white dwarf/main sequence binary merger and may have included >1.4 Solar mass double degenerate SNe, I strongly suggest that their data should be reanalyzed and their conclusions be reevaluated. Their distance estimates derived from type Ia SNe peak emission luminosity, central to their accelerating universe conclusion, may have been significantly in error.
Why is it markedly flat on one side?
Reply | Report Abuse | Link to thisA SN may result from either of the routes : i) A slow route in which a white dwarf may accrete matter from a companion main sequence star or red giant, attain critical mass and then explode as SN. Under this route, trace of the main sequence star or red giant should be left ii) A fast route under which a white dwarf may accrete matter from a companion white dwarf
Reply | Report Abuse | Link to thisIn either routes, condition for SN to explode is the achievement of critical mass by exploding white dwarf, which is the Chadrashekhar limit of 1.38 times the mass of Sun. In view of this, on attaining of this limit, traces of companion white dwarf ( from which matter is being accreted) should be left after explosion of SN.
Secondly, a white dwarf is an ultra dense object. How a white dwarf can have accretion of matter from a companion white dwarf. Should both white dwarf merge before exploding into a SN?
Recently, astronomers have detected a binary white dwarf system J0651 wherein two white dwarfs located at about 3000 light years from earth have been found orbiting a common center at a distance less than 1/3rd the distance between earth and moon. If at such a short distance, larger white dwarf has not accreted matter from a smaller white dwarf, how scientists can be sure that a SN may explode by accretion of matter by a white dwarf from a companion white dwarf?
Further, a white dwarf being a an ultra dense object should be possessing high gravitational force. As such, apart from accreting matter from discrete bodies like main sequence stars, red giants and white dwarfs, it will be accerting dust and gas from adjoining sky also. if a white dwarf can attain critical mass ( Chandrashekhar limit of 1.38 times mass of sun) from adjoining gas and dust, why it should not be exploding as a SN even before merger with companion white dwarf?
Article states that less 20% of SNs result from accretion of matter by a white dwarf from companion main sequence star or bloated red dwarf and about 80 % result from accretion/merger of a white dwarf with another companion white dwarf. Though this conclusion can be misleading since sample under study comprises a very small sample of of only 5 SNs. Nonetheless, one can conclude that a considerable percentage of SNs result from merger of white dwarfs and subsequent explosion of merged mass.
Reply | Report Abuse | Link to thisMass of merged white dwarfs may vary, depending upon mass of individual white dwarfs, and SNs may explode even from masses exceeding 1.38 times the solar masses. As such, absolute luminosity of SNs IA may not be same for all SNs. Study of SNs in 1998 and prediction of dark energy thereupon was based upon broad presumption of constant absolute luminosity of studied SNs. ( due to the assumption that all SNs explode at critical limit of Chandrashekhar limit).
However, if the basic presumption itself has been clouded, prediction of dark energy shall obviously be clouded. I think jtdwyer is right in this respect
I think you have been confused by the podcast by John Matson that stated:
Reply | Report Abuse | Link to this"In about a million years, the white dwarfs will get so close that the larger one will start to cannibalize its smaller companion."
http://www.scientificamerican.com/podcast/episode.cfm?id=white-dwarf-binary-stars-make-merge-12-09-17
That podcast references the research report
J. J. Hermes et al. 2012 ApJ 757 L21, "RAPID ORBITAL DECAY IN THE 12.75-MINUTE BINARY WHITE DWARF J0651+2844",
http://dx.doi.org/10.1088/2041-8205/757/2/L21
http://arxiv.org/pdf/1208.5051v1.pdf
Searching the arXiv preprint archive version of that report, neither the text 'cannibal' nor 'accret' appear anywhere. Nor could I find any mention in any other source that agreed with John Matson's statement that one white dwarf can under any circumstances (prior to the moment of collision or merger) accrete matter from another white dwarf.
A white dwarf > about 1.38 Solar masses cannot persist, since it will gravitationally collapse and supernova when it reaches that mass.
As I understand, there are now thought to be (only) two common conditions producing type Ia supernova, the established model in which a white dwarf accretes matter from a companion (red giant or even main sequence) star until it reaches the Chandrasekhar limit of about 1.38 Solar masses - at which point it produces a supernova of 1.38 Solar masses. These supernovae produce a peak emission source luminosity of consistent value.
In the fast, double degenerate type Ia supernova, two white dwarfs (each less than 1.38 Solar masses) in a binary system eventually collide, producing a supernova of varying mass up to but less than 2 x 1.38 Solar masses. The peak emission luminosity of such supernovas is variable, up to the maximum produced by the combined mass of two white dwarfs.
I hope this helps. I think John Matson's earlier podcast assertion that one white dwarf can accrete matter from another is incorrect.
If one white dwarf of mass 1 solar mass merges with another white dwarf of 0.9 solar mass, resulting object will have mass of 1.9 solar mass exceeding critical limit of 1.38 solar masses. Therefore, it will explode as a SNs type 1a. In this manner, all the SNs type 1a will have variable masses ranging in the range from 1.38 solar masses to 2.76 solar masses ( 2*1.38). Though scientists have not discovered so far but the likelihood of merger of more than two white dwarfs in some old star cluster, where stars are closely packed, can not be ruled out at all. In such case, resulting SN may have mass of even more than 1.76 times the solar mass.
Reply | Report Abuse | Link to thisArticle states that only 20 % SNs are formed thru first route i.e accretion of matter by a white dwarf from a main sequence star or red giant implying bulk of SNs are formed from merger of white dwarfs.
Above implies that majority of SNs formed from the merger of white dwarfs will have variable masses which will result in lack of consistent peak luminosity. Therefore, estimation of cosmological distances based upon consistent peak luminosity of SNs shall be clouded. This will endanger the very basis upon which dark energy was predicted in 1998 and for which Nobels have also been conferred
Yes that's essentially all correct (there is a typo: 1.76 should be 2.76). Thanks for clarifying.
Reply | Report Abuse | Link to thisSince the minimum peak emission source luminosity would not have changed, there is actually little chance that reevaluation would overturn the previous findings, but since any double degenerate type Ia SNe included in the studies' samples would have produced invalid distance estimates, if any were included the previous results should technically be discarded.
If accretion model type Ia SNe can be reliably distinguished from double degenerate events, I suspect that a new evaluation of properly selected single degenerate type Ia SNe would produce similar results as the previous studies, but that can't be definitively determined without executing a proper analysis.
Of course, if the actual peak emission source luminosity of high-z type Ia SNe cannot be precisely determined they cannot be used to reliably estimate distance, making the original 'accelerating universe' studies' results unreproducible.
"Since the minimum peak emission source luminosity would not have changed,"
Reply | Report Abuse | Link to thisStudy of SNs type Ia leading to prediction of dark energy was conducted in 1997 and 1998. How minimum peak luminosity can remain constant for all over this period of 14-15 years? Is it necessary that light from those SNe may still be in transit for detection in present time?
Study on progenitors of SNe type Ia is comparatively new.Atronomers work on the presumption that SNe resulting from accretion model should leave some trace of companion star. If they find the trace of a companion star ( main sequence or red dwarf) in the vicinity of explosion of SN, they conclude that SN resulted from the accretion model. Otherwise, astronomers conclude that SN resulted from a double degenerate stars which is a theoretical prediction
Other than above theoretical model, have astronomers any empirical evidence regarding merging of two white dwarfs and exploding thereof the merged mass as a SN ? White dwarfs are quite compact ulta-dense objects. When two ultradense objects having high gravitational forces will collide, lot of energy in the form of gravitational waves ( if such waves really exist) should be emitted. But so far none of gravitational waves have ever been detected from any part of space.
One more issue over which adequate attention does not seem to have been paid. MW galaxy is considered one of the large and old galaxy. It is estimated that it has some 200-400 billion stars and its age is put at about 12-13 billion years. It has also more than 1000 star clusters in which stars are closely packed. Obviously, in an old galaxy with age > 12 billion years having about 200-400 billion stars, nos. of white dwarfs should also be in the range of few millions. Out of million of white dwarfs, at least thousands should be in close proximity nearing merger. If double degenerate model of SNe is correct, at least hundred of white dwarfs should have merged and exploded as SNs. But observation of SN from MW are very rare. Last SN from MW was observed about 300 years ago around 1731 AD. Further, it is also not certain if such rare SNe resulted from single accretion model or from double degenerate star model
Such rare observation of SN resulting from double degenerate model brings the very model as cloudy one.
This is difficult to say clearly:
Reply | Report Abuse | Link to this"Since the minimum peak emission source luminosity would not have changed..."
I specified "emission source luminosity" to indicate the brightness of the exploding star one would experience if one were essentially right next to it. In contrast, the apparent or detected luminosity is the brightness seen by the telescopes that detected that light.
The "peak emission source luminosity" indicates the brightest period that occurs within the first few days of the SN. This peak period of hours or days (I forget) is the only period that can be used to identify the 'consistent' luminosity thought to occur for accretion model single degenerate type Ia SNe. Only the apparent (observed) luminosity during that peak period can be used to accurately estimate distance from the observed diminishment of luminosity resulting from the dispersal of propagated light.
Sorry if I've made this overly complicated in trying to be precise.
There are thought to be hundreds of billions of detected galaxies, not just the Milky Way. Those surveys of type Ia SNe in the late 1990s were primarily interested in "high-z" or distant observations, to estimate the universal expansion rate over as much time as possible.
One of the two studies states in its abstract:
"With previous data from our High-Z Supernova Search Team, this expanded set of 16 high-redshift supernovae and 34 nearby supernovae are used to place constraints on..." Please see:
Riess et al, "Observational Evidence from Supernovae for an Accelerating Universe and a Cosmological Constant", http://iopscience.iop.org/1538-3881/116/3/1009/
http://arxiv.org/abs/astro-ph/9805201v1
If I understand correctly, that study's conclusion that universal expansion began accelerating about 5 billion years ago was primarily based on a sampling of 16 distant type Ia SNe representing only the period from around >1 billion years ago to around 5 billion years ago. IMO, If less than 100% of those can be confirmed as consistent luminosity accretion model single degenerate type Ia SNe, the results should be discarded.
The draft doesn't mention cannibalism or accretion because it isn't about the ultimate outcome of the orbiting white dwarfs. It does however mention tidal forces.
Reply | Report Abuse | Link to thisTidal forces will rip apart a white dwarf just as easily as they do main sequence stars and giants, they just need to be somewhat stronger - and they will be before the white dwarfs can collide.
Once one of the two white dwarfs fills its Roche lobe, it will spill over and accrete on the other.
Thanks jtdwyer. I got your point. Since astronomers have no knowledge of the channel i.e single accretion model or double degenerate model which resulted in the formation of SNe type Ia in 16 and 34 galaxies, therefore, they will also be not certain if peak luminosity of all the 16 and 34 SNe was consistent. In view of this whole study can be invalid if it is not proved that all the 16 and 34 SNe occurred thru accretion single star model
Reply | Report Abuse | Link to thisOne thing more. Why SNe formed thru double degenerate route ( merger of two white dwarfs) don't indicate consistent peak luminosity? Could it be that all SNe formed thru single accretion model explode at same mass of 1.38 solar mass but in case of double degenerate model, SNe may occur at any mass between 1.38 to 2.76 times solar mass. Further, any star at mass > 1.38 solar mass is not stable, as such, SNe explosion formed thru merger of two white dwarfs shall takes place very rapidly and no period shall be left over during which SN may demonstrate consistent peak luminosity
Tidal forces do not establish accretion. That a white dwarf can accrete matter shed by a red giant or even the surface of main sequence star does not establish that it can accrete matter from another white dwarf of any size.
Reply | Report Abuse | Link to thisThe question is: How close do binary white dwarf stars have to be before the gravitation of the primary causes material from its companion to expand past its Roche lobe? I suspect it most likely does not occur until the two collide.
I find no evidence that white dwarf binaries accrete matter from one another - except for the podcast by John Matson. I think he has simply confused the accetion that occurs in the single degenerate model with the double degenerate model.
You state: "Tidal forces will rip apart a white dwarf just as easily as they do main sequence stars and giants, they just need to be somewhat stronger - and they will be before the white dwarfs can collide."
Then its not as easy, is it?
Do you have any evidence supporting your assertion?
I contend it is at least misleading, and false if material is not exchanged prior to collision.
Yes, in the colliding double degenerate model the mass of the supernova explosion is variable - thus so is the source luminosity. As I understand, presuming that source luminosity can then not be precisely determined, no accurate distance estimate can be produced from the observed or apparent luminosity by calculating the dispersion of light over distance.
Reply | Report Abuse | Link to thisBTW, this double degenerate accretion question is a crucial point, since it determines whether the resulting supernova will consistently occur at ~1.38 Solar masses. If a white dwarf accretes matter from anywhere, it will supernova when it reaches ~1.38 Solar masses, and all type Ia SNe will be of consistent peak emission source luminosity and can be used as standard candles.
Reply | Report Abuse | Link to thisIf any type Ia SNe do not supernova until their mass reaches some amount greater than 1.38 Solar masses, as in the case of a sudden collision of two white dwarfs whose combined mass exceeds 1.38 Solar masses, and these events cannot be distinguished from accretion events, then no type Ia SNe can be used as a standard candle to accurately determine distance.
There is a small body of work that examines the dynamics of mass transfer/accretion in double degenerate systems, e.g.:
Reply | Report Abuse | Link to thisMartin et al "Critical Mass Transfer in Double-Degenerate Type Ia Supernovae"
http://arxiv.org/abs/astro-ph/0609192
Dan et al, "Mass transfer dynamics in double degenerate binary systems"
http://iopscience.iop.org/1742-6596/172/1/012034
Dan et al, "Prelude to a double degenerate merger: the onset of mass transfer and its impact on gravitational waves and surface detonations"
http://arxiv.org/abs/1101.5132
Indeed recent studies have shown that long lived mass transfer in double degenerate systems can occur over many orbital periods.
In addition to the previous works on accretion in double degenerate systems, you may find a 2009 paper (that includes 3D movies of accretion simulations) interesting. The paper ("Surface Detonations in Double Degenerate Binary Systems Triggered by Accretion Stream Instabilities" Guillochon et al) and movies are available here:
Reply | Report Abuse | Link to thishttp://astrocrash.net/projects/double-white-dwarf-accretion/
Thanks very much for the sampling of the "small body" of white dwarf accretion studies - all analyses of simulation models used to evaluate accretion conditions.
Reply | Report Abuse | Link to thisFYI, a freely accessible IOP download of the second reference, "Mass transfer dynamics in double degenerate binary systems", is available at
http://iopscience.iop.org/1742-6596/172/1/012034/pdf/1742-6596_172_1_012034.pdf
For several reasons, I found the second reference and the last reference, very interesting. They explain that the initial states of conditions represented in the simulations are extremely crucial to the outcome and has been a source of conflict. In addition, differing simulation methods have also produced different results. I can't really assess these and other methodology issues.
I also found a discussion including simulation animations by the authors of the third reference, "Prelude to a double degenerate merger" at
https://teamwork.jacobs-university.de:8443/confluence/x/Do1DAg
In the face of this simulation 'evidence' it seems apparent that I should admit my error. However, I'm not quite convinced by the assumptions of these simulation studies. Especially, they did not evaluate the implications of binary white dwarf system origins.
While white dwarfs at least initially have a thin atmosphere of hydrogen and helium that could seemingly easily be accreted, the surface gravity of white dwarfs far exceeds that of main sequence stars and especially red giants, owing to their tiny radii relative to their mass. While a white dwarf can easily accrete the disperse gasses expelled by a red giant star or even a main sequence star with much, much lower surface gravity, accreting material from another white dwarf could only be accomplished at extreme proximity.
The vast majority of white dwarfs are thought to be the end product of main sequence stars ranging from 0.5 to 8 Solar masses. In these cases, as their fuel is consumed they first form asymptomatic giant branch stars which then expel most of their material, producing a planetary nebula. A binary system composed of two white dwarfs could only exist if the primary star did not accrete sufficient mass from its companion to produce a supernova. This seems to require that they are sufficiently distant that accretion does not occur, or that the combined mass of the primary white dwarf and the secondary main sequence star is less than 1.38 Solar masses (in which case no supernova can ever occur).
(continued)
Reply | Report Abuse | Link to thisHow could an existing primary white dwarf avoid accretion of the majority of stellar mass expelled by an AGB companion star, then accrete enough material from the remnant companion white dwarf to produce a supernova? This seems to require that they initially be quite distant, which in turn implies that a great deal of time must pass before they collide.
Moreover, a significant percentage of the carbon-oxygen plasma in cooled white dwarfs has been found to crystallize. It seems obvious that accretion of solid crystallized matter from a tiny white dwarf would be effectively impossible...
Personally, I suspect that these simulation studies of initially preexisting binary white dwarfs may have been originally performed in hopes of finding a method in which binary white dwarfs could produce consistent luminosity type Ia SNe of 1.38 Solar masses. At any rate, unless observational evidence of periodic nova flashes preceding type Ia SNe events exists, I remain skeptical of the double degenerate accretion models, except nearly at the moment of collision.
It is clear from the references you've provided that the opinion of at least some experts conflict with my own uninformed assessment. Thanks for that correction...
The primary issue here remains: whether type Ia SNe are only produced when a primary white dwarf star reaches 1.38 Solar masses following the accretion of mass from any companion, or whether ANY type Ia SNe indistinguishably occur at varying masses and luminosities, from collision, for example. If so, they cannot be reliably used as standard candles to accurately estimate distance.
Reply | Report Abuse | Link to thisEven the third reference provided by J Hanford, "PRELUDE TO A DOUBLE DEGENERATE MERGER...", concluded:
"... This could possibly be the long sought-after explosion mechanism how to trigger supernova explosion in a double degenerate scenario which, together with the promising rates, could make double degenerate mergers a viable (sub-luminous) type Ia progenitor system."
Specifically, a "sub-luminous" (less than 1.38 Solar mass) type Ia SN would specifically produce the 'dimmer than expected' SNe observations that directly led to the conclusion that universal expansion is accelerating...
Simulation studies incorporate only those parameters in the algoretham as are currently known to researchers. In view of this, simulation studies may not replicate the actual physical mechanism.
Reply | Report Abuse | Link to thisActual physical mechanism around the vicinity of a binary white dwarfs may be known from actual observational studies.
Are there no actual observational studies around some binary white dwarfs to clinch the evidence if a white dwarf could accrete matter from a companion white dwarf and explode as a SN on attaining mass of 1.38 solar masses before collision of the binary pair?
Initially, a white dwarf may contain disperse gases around its periphery, therefore, another white dwarf may accerete those gases. But after the outer envelope of disperse gases is accreted, matter in inner core may be too compact and dense to be accreted. In that case, a SN shall occur at mass exceeding 1.38 solar mass on merger of white dwarfs.
In view of above, it appears to me that if a white dwarf can attain critical mass ( 1.38 solar masses) by accreting matter from outer envelope of another companion white dwarf, it may explode as SN even before merger. As such, accretion model SN in case of a binary white dwarf pair, before merger, can also be not completely ruled out. In such cases also, like single accretion model SNe where traces of companion main sequence or red dwarf survive after collision, companion white dwarfs should also survive. Therefore, in the vicinity of exploding SN, a white dwarf whose outer envelope was accreted should be detectable after the explosion of SN
@jtdwyer I think that you are worrying about some of the right things, but there are a few points being missed in this discussion. First, there is not a strict assumption of a standard luminosity for all Type Ia's, but rather that we can use other observable properties of these events to calibrate their brightness to a standard value. While there is certainly evidence that these correlations are not applicable to ALL Type Ia's, it is relevant for the majority. Furthermore, we can often distinguish the outliers by other properties and remove them from calculations that rely on a standardizable brightness. This work has been confirmed in the subsequent ~ decade, but these studies rarely, if ever, get the same play in the press.
Reply | Report Abuse | Link to thisIt's also important to point out that the accelerating expansion of the universe has multiple independent lines of support. See research on the Cosmic Microwave Background and clusters of galaxies. Each of these methods points to the same general conclusion and each has its own independent uncertainties.
In reference to your worries about how double degenerate systems produce SNe, you are right to worry; those doing these studies worry as well. However, some of the points you've raised are probably not important. Regarding getting through one member going from AGB to planetary nebula, there are several potential scenarios. I encourage you to read about common envelope evolution, wherein a star expands in its giant phase and its binary companion ends up inside the expanded envelope. This brings the stars closer together, making it possible for them to merge before crystallization becomes important (it's worth noting that this process takes a VERY long time). This process is a subject of much current research. It's a hard problem.
In any case, let's say you get two WDs into a close binary such that, through the emission of gravitational waves, they will get closer and closer. Unless they are of equal mass, the lower mass of the two will be tidally ripped apart, as mentioned before. While you're right this requires much stronger tidal forces than in the case of accretion from MS or RG to WD, this just means that the WDs must be closer before the process begins.
The only way I see this double-degenerate work could falsify the accelerating expansion conclusion is if there is a preponderance of these sub-luminous events as compared to those in the closer-by universe. There is no evidence for this.
jhshiode - Thanks for your excellent and well informed discussion.
Reply | Report Abuse | Link to thisMy principle question is whether any 'outlier' observations were included in the population analyzed in the original two analyses because less was known then than is known now. Moreover, if the current research is correct then an astounding 80% of type Ia SNe are produced from double degenerates. I'm not at all convinced yet that this research is definitive, but there seems to be a great deal of uncertainty about which observations might actually be 'outliers'.
As for CMB and large scale structure support for acceleration, I couldn't find much real info. except this discussion:
http://ned.ipac.caltech.edu/level5/March08/Frieman/Frieman4.html
Everything fits together so well, as long as the vast number of interdependent presumptions are correct. For example:
http://ned.ipac.caltech.edu/level5/March08/Frieman/Frieman2.html#2.3
"A striking success of the consensus cosmology is its ability to account for the observed structure in the Universe, provided that the dark matter is composed of slowly moving particles, known as cold dark matter (CDM), and that the initial power spectrum of density perturbations is nearly scale-invariant... with spectral index... as predicted by inflation."
Just don't move anything! My apologies for being facetious - please see my brief essay (or at least the title): "Inappropriate Application of Kepler's Empirical Laws of Planetary Motion to Spiral Galaxies Created the Perceived Galaxy Rotation Problem - Thereby Establishing a Galactic Presence for the Elusive, Inferred Dark Matter",
http://fqxi.org/community/forum/topic/essay-download/1419/__details/Dwyer_FQXi_2012__Questionin_1.pdf
As I understand, cosmologists' justification for their estimates of dark matter are largely based on galactic dark matter estimates. I can't go into (collided) galaxy cluster weak gravitational lensing support for dark matter here except to note that the lensing effects essentially always coincides with the galaxies' locations...
(continued)
Reply | Report Abuse | Link to thisRegarding a primary WD moving in through a planetary nebula to accrete material from the companion (new WD), How could the primary WD have avoided accreting matter from the proximal companion when it was an AGB? Wouldn't it have been more likely to supernova then, before the companion became a WD? You're right - it is a very hard problem...
So I concede that a larger primary WD will eventually accrete mass from a smaller companion WD, but how close does it have to be and how much is accreted? If a significant amount is accreted at some distance then the primary should supernovae at 1.38 Solar masses. However, if the WDs must be very close before accretion begins then the supernova may consume the combined binary mass, which may exceed 1.38 Solar masses. This could also produce a great variety of light curves - some of which might mimic other types.
I doubt that the accelerating universe conclusion can be falsified, but the absence of evidence is not evidence of absence... I think so much has been recently learned about type Ia SNe that the original accelerating universe analyses should be redone and expanded. It's a little early to rest on our Nobel laureates and consider that all is now known - again.
From your fresh comments, it appears to me that you are reviewing your earlier stand and coming around the point that a primary larger WD may accrete matter from a smaller companion WD provided they are at right distance. Thus a primary WD in the vicinity of a companion WD may explode as a SN at 1.38 solar mass before merger. However, if this model of a SN from a pair of binary WDs is true, following issues will also arise:
Reply | Report Abuse | Link to thisi)Can a primary WD 'take out' matter from the ultradense and compact body of another WD?
ii) Have scientists computed any critical distance from observational data ( not from simulations) at which accretion may commence between a binary pair of WDs?
iii) After explosion of primary WD into a SN, other WD of the binary pair should survive. Have astronomers observed the survived WD or its trace in the vicinity of any SN in past? Astronomers also state that about 80% of SNe occur thru double degenerate model. In view of this, observation of the survived pair of binary pair of WDs after happening of SN should not be uncommon.
You may recall that recently astronomers have observed a system J0651 of binary WDs at a distance of about 3000 light years from earth in which members of the binary pair are encircling a common center and the distance between two WDs is about 1/3rd the distance between earth and moon. But even at this distance, which is a quite small distance compared to astronomical standards, there has been no evidence of accretion of matter.
Sorry for any confusion. I now admit that I can't definitively rule out a primary WD accreting sufficient mass from a companion WD at sufficient distance that it could supernova at 1.38 Solar masses, leaving a companion remnant that might be obscured by its planetary nebula.
Reply | Report Abuse | Link to thisHowever, I don't think that is a likely scenario. As I also said, "However, if the WDs must be very close before accretion begins then the supernova may consume the combined binary mass, which may exceed 1.38 Solar masses." I think this is the most likely scenario.
I agree that observational evidence of significant (non-atmospheric) WD-WD accretion would be more convincing than simulation studies in which the initial state is two proximal WDs. I think that getting two WDs close together without the primary already accreting much of earlier companion AGB's peripheral mass would be unlikely - a scenario untested by the described simulations.
@jtdwyer
Reply | Report Abuse | Link to thisI'm glad to add whatever I can to this discussion. I can't claim to be versed on all the particulars as this is not my subfield of astronomy, so I would highly recommend you take a look at this review (which is open access via the arXiv): http://arxiv.org/abs/0803.0982
While it is 4 years out of date, it gives a decent picture of where the field has come since the original Ia work. This review focuses largely on the other evidence (not Ia's), but section 4.2 has a short discussion of the Ia work, and gives many references you should be able to find on the arXiv as well. These should satisfy some of your concerns about re-analysis. Work is definitely ongoing in this field.
As for the binary evolution questions, the references given earlier (Dan et al., etc.) represent the state-of-the-art.
I hope you appreciated the point I was trying to make, that in order for the observations to have led to inappropriately to the accelerating expansion, there would need to be a preponderance of lower luminosity Type Ia's at high redshift/in the early universe/at large distances. There is no reason to expect this to be the case, given current models and observations of Type Ia's.
jhshiode - My comments #24 & 25 were intended to be a response to your comment #23.
Reply | Report Abuse | Link to thisThanks for the summary reference, but before I read any more research reports, please do me the courtesy of reading my brief essay. If you also review the Rubin et al reference (or other contemporary reports) you should find that the method used to infer the existence of galactic dark matter was invalid. If you review other cited works you'll find that appropriate application of either classical or relativistic dynamics and gravitation can explain the observed galactic rotational characteristics without dark matter or modified gravity.
Since the rationale for the Lambda Cold Dark Matter model lies to a large degree in the IMO invalid perceived evidence for galactic dark matter, the foundation may be weak... This is in itself reason to question the cosmological justification for universal acceleration.
As to the Dan et al references, I saw no response to my criticism that the initial state of all the referenced simulation studies was a proximal WD binary, there was no evaluation of how that configuration was attained.
I hope you appreciate that the state-of-the-art in WD binary models are incomplete and in a state of conflict and flux. Then there's the recent assertions about the large percentages of double degenerate type Ia SNe to consider. I suspect no survey has yet been conducted to determine the distribution of various type of type Ia SNe at various distances/ages. The absence of evidence....
jhshiode - I've reviewed your latest reference, "Dark Energy and the Accelerating Universe." Its a very good 53 page overview - thanks. I haven't studied it in great detail, but like dark matter, I must point out that all of the very complex methods of 'independently' confirming the type Ia SNe findings are all subject to many potential sources of error. Specifically, as I mentioned before, if the current estimations of cosmological dark matter are significantly in error the current 'standard' model of cosmology would collapse, along with many interpretations of observations. Perhaps someone can explain how Keplerian rotational relations should apply to spiral galaxies, since it's never been even attempted...
Reply | Report Abuse | Link to thisThanks for your insightful discussion!