Supernova Ia studies have become the single most important facet in the latest iteration of the big bang. In the last ten years, not only have they furnished the first “concrete evidence” time dilation actually occurs, they have been used both to “pinpoint” Hubble’s constant and provide the “acceleration” needed to address the age and CMB crisis. Most recently, the Riess team announced a ‘jerk’ has been observed in the magnitude of the most distant supernova, once again providing a major rewrite for cosmology.

We should expect the analytical techniques used by the supernova data reduction and interpretive routine to pass the highest scientific standard, within the obvious constraints of our limited observation platform.

This is far from reality. There are systemic errors in the analysis of Supernovae Ia that make a correct quantitative assessment virtually impossible.

First, the High redshift supernova teams are looking for supernovae with a predicted spectral signature and light-curve width and not even including in these studies events that do not fall within the expected ranges.

Second, the initial data reduction of supernova Ia includes, in the ‘k corrections’, corrections for time dilation and relativist magnitude reduction. If this assumption is in any degree incorrect, these assumptions impose circular parameters into every subsequent analytical step.

Third, the researchers are ignoring the most recently observe local events, which have longer light-curves and higher magnitudes than the previously observed supernovae Ia. Although these events are classified as "Ic hypernova", the spectral signatures appear to be more of a homogenization of Ic and Ia spectra. There is also compelling evidence these curious events involve the collision of two binary pairs of white dwarf stars, creating an intense gamma ray. More important, the light-curves of these events are of the same width as the light-curves of the highest redshifted supernova events, most of which are identified as supernovae type Ia.

Fourth, the critical data reductions of Permutter (the Stretch Factor) normalize at a midpoint z-shift of 0.48. A similar normalization is used by Humay in calculating the Delta(15)b value. In both cases, if there is a Malmquist bias in the collection of the data, this bias will run parallel with any time dilation trend, and therefore be interpreted as time dilation.

Fifth, the combination of the first and fourth errors in this sequence magnify any Malmquist selection bias.

Sixth, in spite of the “Malmquist Bias” burying routines mentioned above, as the number of supernova characterized at high redshift has increased, an embarrassing trend has crept into the Delta(15)b numbers: they get smaller with increasing distance. This should be interpreted as a failure of supernovae Ia to completely satisfy the Wilson hypothesis: The amount of time dilation is less than predicted. In my opinion, it is because of this obvious and embarrassing trend that supernova researchers quit publishing tables that contain both the z shift and the delta(15)b values.

Sixth, a careful analysis of Permutter’s ‘Stretch Factor’ methodology reveals a similar trend is emerging in the stretch factors calculated for the increasing sample of supernovae Ia observed at high redshift.

Seventh, the multi-colored light curve methodologies assume the high redshift events do not experience the relativistic flows characterized in the local hypernovae. They cannot, because this would nullify the assertion these light-curve color variations are due to time dilation.

Eighth, even though researchers claim error analysis indicates there is no Malmquist bias in the sample, the mid point normalization routines, as pointed out above, hide any potential bias as a dilation factor. If the relativistic distance modulus is correct, there should be a selection bias of at least 4%, this is potentially much greater if the true attenuation factor (of space) is greater than the current distance modulus calculation. Good scientific practice dictates that either the researches provide plausible explanations for these fortuitous sampling quirks, or that they should reevaluate these shaky analytical techniques.

Ninth, simple statistical analyses can demonstrate the biasing trends in these complex light-curve treatments: After correction for time dilation, the rise time of distance supernovae are less than the local population, an indication that they are also smaller supernovae. Why would we see smaller supernovae with increasing distance? The best explanation is that the time dilation correction is much too great.

Everyone in the field of astronomy is well aware of the potential for, and consequences of failure to account for distance selection effects. It was a combination of a selection effect errors that led Hubble to originally conclude the universe was much younger and smaller than the current constraints.

The claim that supernova Ia chart a 'jerk' in the cosmic acceleration should raise suspicious eyebrows throughout the scientific community. The same rigid data sets which proved Hubble flow are now being backed away from in order to make room for a constant of acceleration.

Adam Riess’s disturbing acknowledgement that a potential Ia at high redshift was thrown out because the “light-curve was too small” adds tremendous credence to the claim distant hypernovae are being improperly identified as supernovae Ia, and the jerk is an artifact of this aliasing. Unfortunately, if high redshift supernovae which appear to be fading too fast are not included in long term light-curve studies, the proper statistical comparisons will never be possible.

In Tonry’s paper in late 2003, a severe shortages of supernova Ia observations also adds credence to the claim the attenuation rate of space is greater than calculated, and the distant sample of supernovae Ia are really hypernovae. It is also with noting that the high redshift supernova sighting frequency is in the same magnitude as the probable occurrence of local hypernovae.

If the greater scientific community continues to accept the broad pronouncement of the supernova Ia researchers at face value, if systemically bad practices are allowed to continue while the authors make astounding claims about both the reliability and meaning of the data, this branch of science will remain in the crippling grasp of dark energy. It is already inevitable that the last fifty years will be dubbed the “Dark Ages of Astronomy”.

Edit: Grammer, readability

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

=D>

Unfortunately it does seem that today's science is one of discarding any anomalous results unless they are so prevalent that you have to address them. I was even taught to do this while training as a Chemist. If a result didn't agree with the projected answer it was discarded and only those that did agree were kept thus allowing the results to agree with the expected outcome. Any surprise I tend to be skeptical to many of the ways that some scientists do their work?

__________________
Howling from the Shadows

It must be fun to lead a life completely unburdened by reality. --- JayUtah

You can't reason an irrational person out of an irrational belief. --- Noclevername

Apollo: The History and the Hoax
Enter the World of Athran

You'd think we'd have cold fusion by now...

__________________

Feynman
>~~~~<
Science is a way of trying not to fool yourself. The first principle is that you must not fool yourself, and you are the easiest person to fool.

Religion is a culture of faith; science is a culture of doubt.

Jerry, I see you have nine arrows in your quiver there. But for supernova non-experts like me, could you tell us which one of those arrows is most likely to hit the bullseye? If I were to try firing off those arrows they'd likely fly all over the place. Remember what Einstein said: (something like) people opposed to relativity raised countless arguments against it, when just one good argument would have sufficed.

Phantom Wolf wrote:

I'm looking for a gentle way to say you were taught wrong, but I don't see it. I was taught that an anomalous answer is a red flag, either on the experiment or on the technique.

Now, that doesn't mean that you can never discard anomalous data, but there has to be a good and consistent reason for doing so.

__________________
If they can get you asking the wrong questions, they don't have to worry about the answers.

In general, my confidence in the findings and conclusions of the highly qualified supernova search teams is considerably higher than my confidence in the veracity of your criticisms. You are completely ignoring the "assumption of cleverness." These guys are pros. In their publications, they are putting their necks and their reputations on the line. Anyone is free to produce a contradictory or falsifying finding. They know this, and from experience they are very careful to check and recheck their data and to consider all possible alternative conclusions before laying it on the line. To imagine that they are hiding data or intentionally misrepresenting findings just to prolong the tenure of their own theories is just that -- the misguided use of someone's overactive imagination.

I'll just respond to your point 1, which I believe is telling....

That's right. They are seeking to filter out everything except Ia supernovae. Of course the widths will vary according to intrinsic brightness as well as time dilation effects (and minimally due to some other details), so the measured width of the light curve is a very important variable.

You make a rational and integral part of the program sound like some conspiracy of data picking. I'm not sure why. Well, with your naysaying agenda, I guess I do see why.

__________________
Everyone is entitled to his own opinion, but not his own facts.

After correction for time dilation, supernova light curves get smaller with distance, indicating they are smaller. Why?

Local Hypernova, the brightest of supenova events, have light-curves that lose one magnitude in 20-30 days, the same general length as what are interpreted as being supernova Ia at high redshifts. No high redshift hypernova, with redshifts greater than one and corresponding light curves of 40-60 days have been observed. Why?

Because we are mistaking high redshift hypernova for supernova Ia.

Edit: This is as concise as I can make it.

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

My criticisms stem from months of dissecting the papers, trying to both understand the reasoning and the conclusions. The contraindications I have highlighted are important, but I have found they are usually obscured in the published reports. My emails to the principles, and I did this for more than a year before going public, are never answered if I ask anything more than the most innocent questions.

Everyone weighs the importance of data according to their own preconceptions and prejudices. I have great confidence in my own analytical abilities. My unanswered questions are poignant and germane.

This would normally be ok, but if the selection criteria includes systematically ignoring supernova with Ia spectral signature, but with light curves that are too short to be classified as Ia after correction for time dilation, the implications are obvious, if not intentional. (Especially since the expected very long light-curves signatures of time dilated hypernova are non-existant.)

My patience has worn thin. Perhaps I feel a little guilty for laughing off Arp myself. I know that he and Jacques are in the twilight of their lives, and I would like to see the corner turn for them. Perhaps I do not want to see my own careful assessment washed into the diluted sea of dissonance. But I think mostly, Cougar, I am tired of being ignored. I live and work in a mutiple discipline world, and I am use to exchanging ideas freely with archeologists, ballasticians, chemists, engineers, metalurgists...

You should not have to stand up for them. If I do not have the facts straight, it should be easy for them to sic a grad student at me with a better club in his bag than my misspelling of Olber’s name. I have my teeth in this one, and I am not letting go.

edit - typo

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

Jerry, I read your longer explanation last night and was just going to reply to it, only to find you edited it right out! Can you post it again?

I'm getting your general point, but it seems like a bit of a long shot, at least in the game of changing people's minds. Your critique rests on the two types of supernovae being mistaken for each other. Yet this very confusion will make your task quite difficult. For instance, what are the criteria that we could use to distinguish a very high redshift supernova from a hypernova? More significantly perhaps, the mainstream should be able to trump you by just looking at low redshift supernovae, where it's clear they are SNe and not hypernovae. If they can show time dilation for these SNe, you're beat, aren't you?

Sorry...notice I also posted this topic twice. Binary dislexia.

As we look back in space and time, the further back we look for a certain type of event, the more likely we are to see the brightest and the biggest -if you were looking at our solar system from somewhere out in the galaxy, you would be much more likely to spot Jupiter than the earth. This is known as a “Malmquist bias”. In deep space surveys, this bias is expected to run at about 4%, which means on average, the average size of objects we find increase by 4% every time the distance increases by a factor of a hundred.

In theory, a certain type of supernova explosion, type Ia, are thought to be quite uniform, and therefore, we should see little, if any Malmquist bias looking into the past. The light curves we see from these supernova should always be about the same length, losing one magnitude from peak in about 15 days. However, if the universe is expanding, when we look at very distant objects, these light curves should appear to occur more slowly. At a red shift of one, an ‘average’ supernova Ia should appear to lose one magnitude in thirty days instead of 15. In 1939 Wilson predicted that with powerful telescopes we would be able to see this time dilation in very distant supernovae. In the mid 1990’s, observation of redshifted supernovae appeared to prove this is true.

In the nine years since Liebengut “confirmed” this, we have learned a lot more about supernovae - (I certainly thought he was right, four years ago.) Specifically, locally, some supernovae that are similar to type Ia take up to 28 days to lose one magnitude, just like the distant events. These have been classified as type Ic hypernovae, and they are brighter and more rare than the local type supernova type Ia. (the plural forms of supernova and hypernova end with “ae”, supernovae and hypernovae)

So how do we know whether the light curves Liebundgut observed belong to type Ic hypernovae or type Ia supernovae? This is extremely important because if they were type Ic hypernova Liebundgut observed, the Wilson prediction is false, the universe is not expanding, and there was no big bang.

A large number (> 200) supernova Ia type events have now been observed. The evidence is mounting that astrophyscists are overestimating the time dilation in the distant events:

When time dilation is factored in, the light-curves appear to be smaller with increasing distance. Since a Malmquist bias would predict just the opposite, we should see slightly larger supernova Ia with slightly longer light-curves, the only reasonable explanation for why the light-curves are getting smaller is that the time dilation factor is too great.

However, if time dilation is not factored in, the light-curves are on average too long to be slightly larger supernova type Ia. However, this would not be true if the most distant events are more like the “hypernovae Ic”, rather than supernova Ia.

So who is right? The few of us who say they are “hypernova Ic” and bigger supernova Ia, or the many who say they are on the average, smaller supernova Ia?

There are compelling reasons to assert we are right:

1) There is no reason on heaven or earth to think that the distant supernova Ia should be smaller than the local sample. They should be, on average larger, and if the attenuation of space is greater than we think it is, we should see even fewer of the normal size supernova than we think we should. This is true: the most recent search for distant supernova yielded less than 5% of the predicted number. However, this is close to the number of hypernova Ic we should expect to find if we have underestimated the attenuation of space.

2) There is very little difference in the spectra of hypernovae Ic and supernova Ia, and this is especially difficult to distinguish at great distance.

3) Since the “hypernova type Ic” are the biggest and the brightest of the local events, they should also be listed in the most distant events we find, and if these “hypernova Ic” are time dilated, they should have light curves that lose one magnitude in fifty to sixty days! We have not found any light curves that look like this. Where are they? A much better answer is that we do see them: They are dimmer than we would predict and show no evidence of time dilation. In fact, just the opposite, they nullify the Wilson hypothesis.

The only reason for interpreting the data any other way, is to force the reduced data to be consistent with a big bang scenario. Observations do not work that way. They are real, and they are telling us the theory is wrong.

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

In 1993 Mark Phillips discovered that Type Ia supernovae exhibit a correlation between their peak luminosities and the rates at which their brightnesses decline: More luminous Type Ia's fade more slowly than less luminous ones.

That is the finding that enabled Sne Ia's to be used as standard candles. This is a crucial factor that must be added to the analysis - one you apparently left out (?)

Type I's differ from Type II's in that they lack any detectable amounts of hydrogen in their spectra.

And there is a vast difference between Type Ia's and Types Ib and Ic. Types Ib and Ic represent variants of the basic core-collapse scenario for massive stars, which result in neutron stars or even black holes after they explode. Type Ia's blow completely to bits, leaving nothing behind. In The Runaway Universe Goldsmith goes on for two pages identifying the additional differences.

__________________
Everyone is entitled to his own opinion, but not his own facts.

Filippenko had a nice review of the spectral characteristics of supernova. There are some significant spectral differences between Type Ia and Ic - such as presence or absence of H lines and Si lines among other differences.

Jerry, do the distant supernova claimed to be Ia show the spectral characteristics typical of Ic?

Yes, the magnitude does vary proportional to width - I was trying to simplify. This is also what creates the opportunity for error: As soon as you correct for time dilation, you shorten the light curve and also predict the supernova is smaller. But look what happens if you correct for time dilation that is not there: Now your correction for time dilation makes the supernova look smaller than it actually is, which means you also underestimate the magnitude.

As long as this magnitude curve follows the predicted Hubble curve, everyone is happy. But two alarming things are happening: The average light curve widths appear to be getting smaller, meaning distant supernova are on average smaller. BIG red flag - this should not happen. Secondly, even with this correction, the supernova were dimming too fast to stay on the Hubble curve (between a redshift of 0.3 and 0.5.) This is why Permutter concluded the expansion rate is increasing.

All True statements. The Ics are also much less brilliant, 2-4 magnitudes, EXCEPT for the strange new class of objects first indentified in 1998bw and now classified as Hypernova Ic. These are brighter than Ia and have the hydrogen emission signatures of Ic, but they also have some of the characteristics normally found only in supernova Ia – 2002ap has been exceptionally difficult to assign a class to.

Middleditch has concluded these explosions, which also emit killer vectored gamma rays, are the collision of double binary white dwarfs type stars. If this is true, the hydrogen component can vary from near zero to a hundred, depending upon the atmosphere of each of the four progenitors. This would explain why these events also have many of the elemental signatures of both supernova Ia and Ic.

In a sample of about 11 events observe at very high redshift, Tonry concluded 8 were supernova and 3 were either hypernova or undetermined, even though the light curves of all eleven events are about the same length as local hypernova, if time dilation is not factored in. But if time dilation is factored in, All of these high redshift events would have the same sized light curves (actually slightly smaller) as local supernova Ia, and NONE of them have the longer light curves we should expect to see in the time dilated hypernova. Why? A more logical conclusion is they are all more like local hypernova, there is no time dilation, and the attenuation of space is so great we cannot see the type Ia at these distances at all (except for the ones weeded from the sample because the light curves, once corrected for time dilation, are too small.)

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

Like the local hypernovae, they are quite ambiguous - see my response to Couger, also in my paper I have drawn together high redshift spectra from several sources I can't tell - We really do not have a large enough local sample of hypernova - to draw firm conclusions, which in and of itself is a good reason to criticize the new cosmology. Since local Hypernova are brighter than local supernova, we should have spotted them first at high redshift. Filippenko &amp; co have concluded...I don't know what their reasoning is. Does anybody?

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

Jerry, What I'm looking for here is whether or not there is clear spectral evidence that allows Ia to be distinguished from Ic Hypernova. For example, Filippenko says this:

Can these differences be clearly identified in the high z supernova?
If so, it would seem to be pretty straight forward to test your argument. I think your argument is quite compelling, but I'm looking for clarification that the spectral features in fact back it up. Certainly, the more distant supernova should be brighter. The effect of Malmquist bias is readily evident in samples out to even only 100 Mpc, so certainly at the much greater distances it should be evident.

Why would the time dilation not be there? The universe is expanding; therefore, distant objects will exhibit time dilation.

__________________
Everyone is entitled to his own opinion, but not his own facts.

Unfortunately the spectral identities at high redshift do not reveal the levels of detail you describe: remember, these spectral lines are dragged out of the Infrared and far-infrared and k-corrected, and since these k corrections includes time dilation corrections, the lines are also narrowed, and this could obscure what little spectral detail is available, especially in doublets and broad lines. This is why it is essential to at least start analyzing the data for the possibility of non-Doppler redshifts.

The best supporting evidence is in these “beams from hell” gamma rays, as Middleditch describes them. We know they are linked locally to hypernova events, we also know some of most highly focused of these beamed rays often originate at high redshifts. We also know that the probability of having one of these focused beams aimed right towards us is less than 5%, we see the rays, so we should expect to observe a significant number these brilliant hypernovae events at high redshift as well. Where are they?

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

Are you mixing pre- and post-1991 information? Most of the "hypernovae" or gamma-ray bursts ARE at high redshift. This NASA article explains what had been learned, at least up to 1998. Where are they? They're out there at high redshift, as the article gives evidence for.

__________________
Everyone is entitled to his own opinion, but not his own facts.

That is exactly what Jerry said.

Observing a GRB associated with an apparent Ia supernova at high redshift would support Jerry's theory that it is really a mis-classified hypernova.

Also, 1998 is a bit early, as the major event conclusively linking hypernovae to GRBs was in 2003: http://www.eso.org/outreach/press-re.../pr-16-03.html

Of course, nothing has ruled out GRBs being caused by other sources (particularly the short, &lt; 2 second bursts), which may account for high redshift GRBs that have no associated hypernova. Limitations of detection are more likely the culprit, as linking a GRB to anything takes some serious coordination.

__________________
Do try not to take me too seriously.

I thought he was saying they're mostly local events and only a few are at cosmological distances, which, I believe, would be backwards.

GRBs are not associated with Ia supernovae, which is a very specific type. (I don't know that any have been so associated.) They're associated with core-collapse supernovae from huge stars 30 times more massive than our sun.

Good link. Thanks.

__________________
Everyone is entitled to his own opinion, but not his own facts.

No, what Jerry is saying is that the supernova are at cosmological distances, but that they are actually Type Ic hypernova - which are more luminous than type Ia supernova.

The role of Malmquist bias must be considered to understand his point. Any survey has a magnitude fainter than which objects are not detected.
Since there is a limiting magnitude, at greater distances the objects detected will be increasingly represented by just the brightest objects.

For example, in a sample of galaxies I've been analyzing, the galaxies in the closest distance bin have a mean rotational velocity of 168 km s-1 while the galaxies in the farthest distance bin have a mean rotational velocity of 214 km s-1. The reason for that effect is that the faster rotating galaxies are more luminous than the slower rotating galaxies. The increase of mean rotational velocity (and therefore luminosity) in a sample with distance is one signature of the presence of Malmquist bias in the sample. Because of the magnitude limit of the sample, the fainter galaxies fall out of the sample at greater distances.

What Jerry is pointing out is that the same is expected for supernova. The most distant supernova detected ought to be increasingly represented by the most luminous supernova. However, Jerry argues, when the assumed time dilation correction for expansion is factored into the calculations, the supernova at greater distances come out less luminous with distance. He further argues that if you do not include the time dilation correction, then the most distant supernova have light curves consistent with local hypernova - exactly the type of supernova we ought to be seeing at large distances. His other relevant question is "where are the hypernova at large distances?"

If he is right, then the universe would not be expanding.

Why didn’t I put it that way in the first place? Thanks dg. These supernova trend problems are fairly obvious if you are looking at the new data - 1998 through 2004.

The supernova research teams must be acutely aware of the ambiguities. But they go ahead and make these broad and unqualified announcements about the history of everything on the bases of a handful of high redshift events no one should be certain we understand. Important data. Bad cosmology.

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

That is the point; if the objects identified as high-redshift Ia supernovae are in fact local hypernovae, there should be a GRB associated with them. Observing such a GRB would be conclusive proof that the objects are definitely not type Ia supernovae.

Unfortunately, if such evidence is ever collected it will be by sheer luck, as we have to first detect the GRB (via HETE-II or a similar instrument), then quickly get a large telescope pointed at it for long enough to collect the necessary data.

I’m having fun with the SN catalogs; I was curious to see what type Ic hypernovae have been found. For anyone wanting some light (no pun intended) reading, this site has the spectra and light curves of most SN:
http://cfa-www.harvard.edu/cfa/oir/R.../RecentSN.html

__________________
Do try not to take me too seriously.

Another major problem is the large uncertainty in GRB positions as seen here in column 3 .

cyrek1 comment

I do not consider the S1a's as reliable candles.
White dwarfs come in a variety of sizes and temperaturesof from over 100,000K to 3000K.
I do not see how they can be relied on.
There seems to be a lot of manipulation involved to produce the results one wants.
The BBer's underlying motivation to restore Einstein's lambda or fill in a gap to replace the missing mass needed for a flat space.

__________________
aka Michael Cyrek

Nice evidence of the hypernova - gamma ray - correlation, but I think Middleditches causal explanation is more plausable: http://lanl.arxiv.org/PS_cache/astro...11/0311484.pdf

I've thought a lot about this, and my colleagues have warned me to tone down the rhetoric as well - I blew a gasket when I first read the details of the 'Stretch Factor' method. They - the mainstream - are wholly convinced there is no viable cosmology outside the BB. Sorry, but observational data refuses to collaborate with the theory. The astrophysical world has to quit behaving like addicts. Intervention is the first step: Confront them kindly but firmly: they are in denial...

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

Your lack of knowledge about Ia supernovae is hardly a good reason to dismiss what most of the astronomical community find quite credible.

What, they're waiting for some good ol' fashioned strong evidence? The nerve of that bunch! :roll:

__________________
Everyone is entitled to his own opinion, but not his own facts.

Cougar, it is the mainstream interpretations of the Supernova Ia that show them getting smaller with increasing distance, not mine. This should be considered good strong evidence something is badly haywired.

Once again, you want to have it both ways: The supernova Ia are very good standard candles, but they also evolve?

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?

Show that they evolve, and you'll get your name in the history books. A simple graph (with numerous data points) should suffice.

__________________
Everyone is entitled to his own opinion, but not his own facts.