Any complete standardized SNIa data out there?

Robin Whittle wrote:

....

I have a long way to go before I am confident of understanding how the
observations are transformed into the data which are published. But
it seems that Jerry's first charge of assuming time-dilation is correct.

This is bad science if the purpose is to test whether or not there is in
fact time-dilation in high redshift objects. Maybe they think they
settled the matter once and for all in 1996:

Observation of Cosmological Time Dilation using Type Ia Supernovae as
Clocks
G. Goldhaber et al. http://arxiv.org/abs/astro-ph/9602124

I used to do a fair bit of work on the light curves of nearby supernovae,
and I am currently working on the SNAP project, which hopes to put a 2-m
telescope into space to study supernovae out to z=2 or so.

During all this time (which goes back to the nineteen-eighties), everyone
with whom I've worked in the supernova community does assume that
supernova light curves are affected by time dilation. Well, that's not quite
fair, perhaps; it was still an open question in the late 1980s and the very
early 1990s. But after the first observations of supernovae at "high" redshift
(back then, z=0.20 was "high") showed exactly the sort systematic differences
from local supernovae that one would predict with relativity and a standard
model of the expanding universe, astronomers have accepted time dilation
as a given; they have moved on to other things. That's the way it is.

I do agree that people ought to publish their measurements after corrrecting
for simple instrumental effects, though; it would allow people to check
things for themselves much more easily. Sigh.

Michael Richmond

(Robin Whittle) wrote in message ...
Sean wrote:

I believe I *may* have found some uncorrected data in the SCP
database.

I can't find any such .gifs with Google. Can you give the URL of these
files or of wherever you are looking?


Sorry my mistake. It was called Suspect Supernova lightcurves and the
url is....


http://bruford.nhn.ou.edu/~suspect/c...&OBJECT=1999ee

That should take you to one of the SN pages within the site and all
you do then is click on the G to select gif and you should get a
lightcurve made from observation plots only, I believe, and not a
fabricated `adjusted` one like those Knop has supplied.
However it I am not sure from the site whether these are k corrected
or not. There doesnt seem to be any mention of it. It could be that
this site links to many different sites with gif lightcurves and each
lightcurve has different manipulations like k corrections etc. It does
seem pretty good selection though with IRVB lightcurves available for
some SN1a types and if these lightcurves are what you call raw
unadjusted and I call un k corrected then it should be exactly what we
are looking for.
Can I then assume the online SCP lightcurves NOT k corrected?

I don't think anything like this can be assumed. I am struggling to
understand how Knop et al. and the other major papers in this field,
process their data. Here is my current, partial, probably inadequate
and/or faulty understanding:


Referring to Knop et al. http://www.arxiv.org/abs/astro-ph/0309368
the flux (linear brightness) values used in Figs 1 and 2, seem to
directly correspond to the values in Appendix A, which can also be found at:

http://brahms.phy.vanderbilt.edu/deepsearch/hstpaper/

These have times (horizontal) exactly as observed (as I noted in an
earlier post). However the values (vertical) are not the raw observed
fluxes - they are the product of some corrections. I found it it
confusing trying to figure out exactly how they arrived at these "color
corrected" flux values:

(p8) For both high- and low-redshift supernovae, color corrections
and K-corrections are applied . . .

Many paragraphs of details follow - how they figured out the best way of
deciding on a curve to represent the total light curve, as it would be
with a given filter if observed near to the SN, based on their limited
number of observations at various redshifts with various telescopes and
filters. My understanding of "fitting" means using the MINUIT program:

http://wwwasd.web.cern.ch/wwwasd/cernlib/
http://wwwasdoc.web.cern.ch/wwwasdoc...t/minmain.html

to find the values of various variables which optimise (typically
minimise) the output of some Fortran function. By coding up some stuff,
with light-curve tables and corrected observational data, MINUIT will
find how best to place a light-curve, with starting time, width, and
maximum flux to the observed data points. Its a dodgy business, but
they need a light curve to figure out what the maximum flux was, or
would have been if they had observed it at maximum. They also need this
light-curve to do various corrections which involve assumptions about
the spectrum of the light at any particular point in time.

I can understand trying to correct different observations from
different sources to make them all consistent with each other, but
the k correction as I understand it for SN`s seems odd and
misleading.
I think it would be simpler and more accurate to take the data points
observed in one filter (R for instance ) from a high redshift SN1a
then calculate what wavelength these data points were emitted at ( B
for instance)and then compare that lightcurve with a known observed
lightcurve in B from a low redshift SN1a. If the two match with no
obvious time dilation then the simple conclusion can be reached that
there is NO expansion due to BB.
Sean

(Robin Whittle) wrote in message ...
I am struggling to
understand how Knop et al. and the other major papers in this field,
process their data.

Your understanding looks basically correct to me, though I admit I
have not gone through the papers in detail.

Generally, as I understand it, what I described above as color
correction is in fact K-correction:

At some level, the difference is semantics, not science. As you say,
they do the corrections in two steps. You can think of the color
correction as correcting from the actual I (say) filter to the Bessell
I, and the K correction as correcting from Bessell I to rest-frame B.

So here they say that Appendix A, and as far as I can tell, the flux
levels in Figures 1 and 2, have just been subject to this first level of
correction - color correction -

I read the text as saying both corrections are included.

and I couldn't clearly see where Jerry Jensen's charge of
assuming time-dilation in the high redshift observations was justified
in this paper.

I don't know why you call it a "charge," but as Michael says, the
existence of time dilation is considered proved. For the purposes of
Knop et al., it's appropriate to assume time dilation in the analysis
for other parameters.

This is bad science if the purpose is to test whether or not there is in
fact time-dilation in high redshift objects. Maybe they think they
settled the matter once and for all in 1996:

Yes, exactly.

Tell you what: if you want to do a sanity check, why not pick a single
supernova at z=0.8 or so? At that redshift, the K correction from
observed I to rest B will be very nearly zero. So just take the
K-corrected light curve, and see how long it takes for the brightness
to decline by 1, 2, 3, etc. magnitudes from the peak. Do the same for
a nearby supernova measured in B. I'll be surprised if the times
aren't different by a factor of 1.8. This difference should easily be
large enough to see regardless of small corrections.

(Steve Willner) wrote in message ...
(Robin Whittle) wrote in message ...
I am struggling to
understand how Knop et al. and the other major papers in this field,
process their data.

Your understanding looks basically correct to me, though I admit I
have not gone through the papers in detail.

Generally, as I understand it, what I described above as color
correction is in fact K-correction:

At some level, the difference is semantics, not science. As you say,
they do the corrections in two steps. You can think of the color
correction as correcting from the actual I (say) filter to the Bessell
I, and the K correction as correcting from Bessell I to rest-frame B.

So here they say that Appendix A, and as far as I can tell, the flux
levels in Figures 1 and 2, have just been subject to this first level of
correction - color correction -

I read the text as saying both corrections are included.

and I couldn't clearly see where Jerry Jensen's charge of
assuming time-dilation in the high redshift observations was justified
in this paper.

I don't know why you call it a "charge," but as Michael says, the
existence of time dilation is considered proved. For the purposes of
Knop et al., it's appropriate to assume time dilation in the analysis
for other parameters.

This is bad science if the purpose is to test whether or not there is in
fact time-dilation in high redshift objects. Maybe they think they
settled the matter once and for all in 1996:

Yes, exactly.

Tell you what: if you want to do a sanity check, why not pick a single
supernova at z=0.8 or so? At that redshift, the K correction from
observed I to rest B will be very nearly zero. So just take the
K-corrected light curve, and see how long it takes for the brightness
to decline by 1, 2, 3, etc. magnitudes from the peak. Do the same for
a nearby supernova measured in B. I'll be surprised if the times
aren't different by a factor of 1.8. This difference should easily be
large enough to see regardless of small corrections.

Hi Steve

You couldnt have read my last 3 posts as this is *exactly * what I
have done. (andfor more than just 1 SN). And my results show clearly
there is no time dilation .
To reiterate once again: I have taken Knop lightcurves at I corrected
to B and compared with B from Reiss survey and there is no time
dilation especially on the order of 1.8. Furthermore if one overlays
all of knop I and/or R lightcurves from the lowest to highest
redshift and then calculates the corresponding `spread` to rest frame
one sees that the high redshift curves `spread`( ie the change in
lightcurve profile between the lowest and highest redshift SN observed
in one filter)follow exactly the low redshift spread illustrated
clearly in Reiss` paper. I have been saying this time again here and
elsewhere. The only response I have recieved is from one person who
claims that I cant compare the k corrected lightcurves (presumably
because they cant explain why my findings support a no expansion
hypothesis)To take into account his argument is why I now request a
non k corrected source. However I would like to point out , especially
in light of your approval of my approach of using k corrected data,
that at no time have I been given a clear undersatndable argument of
why I cant use k corrected data.

Thanks Steve for your view that the graphs and numbers are the product
of the two stages of correction:

At some level, the difference is semantics, not science. As you say,
they do the corrections in two steps. You can think of the color
correction as correcting from the actual I (say) filter to the Bessell
I, and the K correction as correcting from Bessell I to rest-frame B.

I read the text as saying both corrections are included.

I recognise that the SCP people consider that time dilation is an
established fact, but critics of the Big Bang Theory see too many
problems with the theory to accept this.

For instance, I think there are several serious contradictions in the
conventional understanding of quasars which would be resolved if the
distance to quasars were assumed to be about the same as ordinary
galaxies, rather than the very large distances calculated from a Big
Bang Doppler interpretation of their redshift. (The next two paragraphs
detail some objections.)

One problem is that the conventional distance estimates require that
quasars have prodigious power outputs - sometimes far larger than the
total outputs of entire galaxies. For instance, "as much energy per
second as a thousand or more galaxies, from a region that has a
diameter about one millionth that of the host galaxy."
(http://chandra.harvard.edu/xray_sources/quasars.html) This seems
rather unlikely, but also, since luminosity variation time-scales
place upper limits on the size of the quasar emitting regions, the
consequently high calculated power densities can lead to the "Inverse
Compton Catastrophe". (See footnote 1.)

Another problem the Big Bang interpretation of redshift creates for
our understanding of quasars is the "superluminous" jets which are
common to the point of ubiquity, compared with jets calculated to be
moving at less than c (the speed of light). If all quasars of such
redshifts really are at such vast distances, then the apparently
superluminal jets can only be explained (according to generally
accepted physics) in terms of the jet moving very close to
light-speed (say 98% - see footnote 2) whilst its axis is close
to our line-of-sight. If this is the case, then I think we should
see many more quasars with jets at wider angles from our
line-of-sight where the observed speeds are not subject to this
effect, and are therefore seen more truly, at velocities just under
c. Also, AFAIK in Seyfert galaxies, for which reasonably reliable
distances can be calculated from their moderate redshifts, when we
see a pair of jets with an axis probably close to right-angles to our
line of sight, the jet speeds are typically 0.3c or so. I (and many
others) argue that the apparent over-abundance of superluminal jet
speeds in quasars is a result of the Big Bang interpretation of
redshift. If quasars (luminous black-hole accretion systems without
a clearly observable "host galaxy" - I think they are not in a galaxy
at all) are considered to be closer than conventionally assumed, then
these problems in understanding them are eliminated. This
explanation requires a mechanism by which most or all of the redshift
of the quasar's light is generated intrinsically - in, or close to,
the object itself.

When discussing alternative redshift mechanisms, we bump into the
"established fact" of expansion - which we consider to be a theory well
worth challenging.

I find Jerry Jensen's critique most intriguing (astro-ph/0404207). The
SCP people assume time dilation and are focused entirely on correcting
for the observational and individual idiosyncrasies of SNe in order to
convert the observations into peak observed fluxes as would be observed
if there was no redshift, extinction or variation between different SNe
1a. Big Bang critics are interested in whether or not these are all
genuinely SN 1a - and if so, whether, after proper corrections their
light curves of high-redshift SNe display the predicted time dilation.

I think we have established that the Appendix A figures and points on
the graphs are unaltered in any sense from the observed time-scale - and
that the flux figures are the product of some complex corrections to
cope with observation with different filters, redshift, Milky Way and
host-galaxy extinction etc. But part of these corrections involve the
stretch factor for both flux and time, by which perceived short-time SNe
have their flux levels corrected downwards.

I think the researchers' assumed time dilation is perfectly valid within
their frame of reference - the Big Bang - which they are attempting to
determine the exact parameters of. Jerry's critique of conventional
supernova analysis has several aspects, and I won't try to summarise it
here. I think his focus on the assumed time-dilation in the early
"correction" steps is that it might mask a Malmquist type II bias which
would otherwise be more obvious. If this bias was more obvious, perhaps
it would be more readily apparent that (as he argues) the supposed SN 1a
at high redshift are actually hypernovae instead.

Tell you what: if you want to do a sanity check, why not pick a single
supernova at z=0.8 or so? At that redshift, the K correction from
observed I to rest B will be very nearly zero. So just take the
K-corrected light curve, and see how long it takes for the brightness
to decline by 1, 2, 3, etc. magnitudes from the peak. Do the same for
a nearby supernova measured in B. I'll be surprised if the times
aren't different by a factor of 1.8. This difference should easily be
large enough to see regardless of small corrections.

This brings us back to the need for raw observational data, which does
not seem to be easily found. Ideally, as you write, it would be
infra-red, presumably space-based, observations for the high redshift
SNe. Then the redshift and the I filters remove the need for much
correction in order to compare with V or B observations of closer SNe
1a. However, we need to be able to prove to ourselves that these really
are the same variety of SN 1a as can be found at lower redshifts.


I have a pile of SNe papers printed out and will read them in order,
starting in Feb 1996, before I feel I understand this situation enough
to evaluate Jerry Jensen's critique of the conventional approach. In
response to your challenge, I point out that in the Knop paper
(astro-ph/0309368), the light-curve of 1998be curve (z=0.644) is clearly
shorter than that of 1998as (z=0.355) when 1998be should be 21.3% longer
(assuming that the two SNe had identical light-curves and that all the
corrections had been done properly). The time axis is uncorrected, and
I guess that the total effect of the corrections on the flux axis is not
so high as to account for this. I have the two sets of curves, one
above the other he

http://astroneu.com/misc-files/1998be-as.gif

It would be difficult or impossible to work backwards from these
corrected figures, even with all the internal data, such as the template
set of spectra which (as I understand it) they developed as a model of
how a "typical" SN1a's spectrum changes in time. What we need is the
raw observations. I wouldn't ask for this data until I have read and
understood the main papers in this field. However I think it should be
readily available - published and freely available for download -
together with all the researcher's software, template spectra etc. This
is the only way the scientific process can operate effectively - to
enable other researchers to duplicate the original work, to explore
variations on their techniques and to apply the techniques to new
observations. This observational data should include information about
those SN1a which were rejected from the study for one reason or another.

I am focusing here on the Knop paper, but if they are mistaken in how
they interpret the observations, then so, probably, are the major papers
which precede this. Its a tall order for outsiders to critique such a
well researched (in many respects) project which has such high
credibility - but we would rather do this than accept the notion that it
is right, and that all our concerns about the contradictions of the BBT
with observations are just a problem in our own minds.

I wrote earlier that it was "bad science" for Knop et al. to assume time
dilation. This was on the basis that they were attempting to show that
the Universe is in fact expanding. But looking closely at their paper
now, I think they are not trying to establish this - they are assuming
it and simply trying to quantify the expansion and its history. I
withdraw this accusation.

A lot hangs on whether or not these and other SNe 1a observations
are consistent with the expected time dilation. I don't think its good
enough to supposedly settle this question with a few observations (as
far as I know, with no raw data published - but I haven't read all the
papers yet) and with some necessarily complex and potentially faulty
"corrections" in 1996, and then proceed as if the Big Bang is a fact
thereafter. If Jerry Jenson's critique about mistaking the only
recently recognised hypernovae for SNe 1a survives scrutiny, then it all
this SN 1a work must be completely re-evaluated, going back at least to
Goldhaber in 1996 http://arxiv.org/abs/astro-ph/9602124 to see if these
interpretations really are consistent with time-dilation.

Sean, where are you finding any potentially raw observational data at
all? Of the 11 HST SNe and 53 previously analysed SNe which are
mentioned in Knop et al., the SN database at
http://bruford.nhn.ou.edu/~suspect/ only mentions one: 1990O

- Robin http://astroneu.com


Footnote 1: The Inverse Compton Catastrophe

This is the first paragraph of the section "Synchrotron self-Compton
emission", page 78, of "Quasars and Active Galactic Nuclei", Ajit K.
Kembhavi and Jayant V. Narlikar, Cambridge University Press 1999.
http://www.amazon.com/exec/obidos/tg...l/-/0521479894 See also page
409. They cite "On the nature of quasi-stellar sources", Hoyle,
Burbidge and Sargent, Nature, 209, 751 (1966).

"y" here means gamma - the Lorentz factor, which is commonly used
to describe velocities (in this case of electrons) which are close
to c. According to http://www.softcom.net/users/greebo/lorentz.htm
the Lorentz factor for a velocity v is given by:

1 / sqrt(1 + (v^2 / c^2))

Here is the quote:

Synchrotron photons can be Compton scattered by the ensemble of
electrons which emits them in the first place, boosting photons
of energy E to energy ~(y^2)*E. The electrons can therefore
lose energy either through synchrotron emission or through
Compton scattering of photons. The ratio of the luminosity
generated by an ensemble of electrons in the Compton and
synchrotron channels is proportional to the ratio of the
energy density of the photons and that of the magnetic field,
as shown in Equation 4.14. The radiation density is high in
the case of luminous, compact sources and it is to be expected
that they produce copious high energy photons through Compton
scattering. These photons can again be Compton scattered, to
still higher energies, and so on until the condition
yE mc^2 is violated. [I think m is the electron mass.] If the
energy density of the once-scattered radiation exceeds that of the
magnetic field, the energy density of the twice-scattered photons
exceeds that of the once-scattered photons. As a result of such
multiple scattering the electrons will lose their energy very rapidly
to high energy photons, thus quenching the source; this is known as
the Compton catastrophe (Hoyle, Burbidge and Sargent 1966.)

In other words, there is a physical limit on the power density which can
be achieved with synchrotron radiation mechanism - which is widely (I
think reasonably and almost universally) believed to be the source of
most or all quasar emissions. I understand that some or many
high-redshift quasars, have observed luminosity variation frequencies
which lead to a size limit on the emission area which suffers the
"Compton catastrophe" if its power output is as high as is required by a
distance estimate which does not accept intrinsic redshift, and instead
calculates distance according to the conventional approach of treating
it as Doppler effect in the context of Big Bang cosmology.

Relativistic beaming can increase the energy we observe without causing
a "Compton catastrophe" - the objection is usually to isotropic
radiators with vary high power densities.

Footnote 2: "Superluminal" motion from relativistic jet alignment

See page 53 to 57 of the abovementioned text or:
http://bustard.phys.nd.edu/PH308/AGN/superluminal.html
On page 410, Kembhavi and Narlikar argue that the 10c "observed" jet
speeds (this depends entirely on the Big Bang's high distance estimate)
involves the jet aligning so closely with our line-of-sight that the
probability of this occuring is 10^-5.

In article ,
Robin Whittle wrote:
For instance, I think there are several serious contradictions in the
conventional understanding of quasars which would be resolved if the
distance to quasars were assumed to be about the same as ordinary
galaxies, rather than the very large distances calculated from a Big
Bang Doppler interpretation of their redshift. (The next two paragraphs
detail some objections.)

Without getting into the argument about supernovae, I just want to
address these two points about quasars.

The first thing to note is that it seems to arise from the days when
quasars, being easy to find, typically had higher redshifts than
galaxies. This is no longer the case. We see `ordinary' galaxies, or
objects that appear to be galaxies with plausible properties, at every
redshift at which we see quasars. So any alternative model for
high-redshift quasars has to explain those as well.

One problem is that the conventional distance estimates require that
quasars have prodigious power outputs - sometimes far larger than the
total outputs of entire galaxies. For instance, "as much energy per
second as a thousand or more galaxies, from a region that has a
diameter about one millionth that of the host galaxy."
(http://chandra.harvard.edu/xray_sources/quasars.html) This seems
rather unlikely,

How do you know that, a priori? I can say that it `seems rather
unlikely' that matter in the world around me is made of tiny subatomic
particles. In that case, it doesn't matter if I think it's unlikely,
since it's also demonstrably true. So, the issue is not whether we
think it's unlikely; the issue is whether the observational evidence
supports it. We see objects that look like galaxies with nuclei many
times brighter than the stars in the galaxy. We need a model to
explain that.

but also, since luminosity variation time-scales
place upper limits on the size of the quasar emitting regions, the
consequently high calculated power densities can lead to the "Inverse
Compton Catastrophe". (See footnote 1.)

The high brightness temperatures that lead to the IC catastrophe are
uniformly in radio-loud (i.e. jetted) quasars, and as your comment says

Relativistic beaming can increase the energy we observe without causing
a "Compton catastrophe" - the objection is usually to isotropic
radiators with vary high power densities.

So, there is no problem if we can show that the bright sources are
strongly relativistically beamed. Which leads to the next section.

Another problem the Big Bang interpretation of redshift creates for
our understanding of quasars is the "superluminous" jets which are
common to the point of ubiquity, compared with jets calculated to be
moving at less than c (the speed of light).

1) The term is `superluminal', not `superluminous'. Just in case
anyone wants to do a little googling (-:.

2) It is not at all true that such jets are `ubiquitous'. There are
plenty of cases where superluminal motion is not observed. (Motions
where the apparent speed is c are sometimes called `subluminal
motion'. You can google for that too, or look on ADS.)

3) There are superluminal jet sources *in our own galaxy* (so-called
microquasars -- another term to search for -- driven by accretion
onto a compact stellar remnant). The distance estimates
to these objects of course have nothing to do with redshift or
Hubble's law. So it is not the case, as was once suggested, that
the calculation of jet speeds faster than the speed of light is a
problem in itself for the cosmological interpretation of
redshift. (For readers who aren't sure how these calculated speeds
arise, see the sci.astro FAQ,
http://sciastro.astronomy.net/sci.astro.8.FAQ, point 8.) I point
this out just for completeness.

If all quasars of such
redshifts really are at such vast distances, then the apparently
superluminal jets can only be explained (according to generally
accepted physics) in terms of the jet moving very close to
light-speed (say 98% - see footnote 2) whilst its axis is close
to our line-of-sight. If this is the case, then I think we should
see many more quasars with jets at wider angles from our
line-of-sight where the observed speeds are not subject to this
effect, and are therefore seen more truly, at velocities just under
c.

But we do see these objects. If we select samples of radio sources at
low frequencies, it turns out that we are selecting mostly on unbeamed
emission from the lobes (the wastebaskets of energy supplied by the
jets). In this situation, we find, if we look in an appropriate
luminosity range, a small number of quasars, with bright jets, and a
larger number of radio galaxies (that is, objects identified as
galaxies but with large-scale luminous radio emission) with much
fainter jets. Some of the objects identified as quasars will indeed
show superluminal motion. Most of the objects in the same luminosity
range identified as radio galaxies will not, when measurements are
possible. The radio galaxies are the quasars seen at large angles to
the line of sight. There is a huge literature on comparison of the
properties of the radio galaxies and quasars, going by the general
name of `unified models' (this also encompasses the unification of
other types of active galaxy). But the point here is simply that you
are correct that we should see jets at large angles to the line of
sight -- and we do. If you go on to ask `have the statistics of the
apparent motions in a complete unbiased sample been analysed to show
that they are consistent with the expected distribution for random
orientation with respect to the line of sight, the answers is
unfortunately no, not yet, because the faintest jets in radio galaxies
are too faint to image adequately with the very long baseline
interferometry technique that is needed to measure motions in the jet.
But people are working towards this goal (see e.g. Hough et al 2002 AJ
123 1258). Highly beamed objects are easy to observe, which is why
there was an apparent preponderance of very extreme objects in the
early days of interferometry.

(If we select samples of radio-loud quasars at high radio frequencies,
we are *selecting* on beamed emission from the jet, so we would
*expect* a bias in the sense that we would see more highly beamed
objects than there are in the population at large. In general this
bias is harder to deal with, because we don't know what the unbeamed
population looks like. That's why I've been careful to specify
low-frequency selection above.)

If quasars (luminous black-hole accretion systems without
a clearly observable "host galaxy"

With the advent of HST (and good data analysis) there are no quasars
without a clearly observable host galaxy -- if you look hard enough.

Martin
--
Martin Hardcastle Department of Physics, University of Bristol
A little learning is a dangerous thing; / Drink deep, or taste not the
Pierian spring; / There shallow draughts intoxicate the brain ...

(sean) wrote in message ...
You couldnt have read my last 3 posts as this is *exactly * what I
have done. (andfor more than just 1 SN). And my results show clearly
there is no time dilation .

Sorry to be so slow, both on the uptake and with the reply.

To reiterate once again: I have taken Knop lightcurves at I corrected
to B and compared with B from Reiss survey and there is no time
dilation especially on the order of 1.8.

Could you show your work for one or two SN from each group? Say which
ones you used, the date of maximum in each case, and the time to drop
a certain number of magnitudes from maximum (perhaps a couple of
different values).

Steve Willner wrote in message ...
(sean) wrote in message ...
You couldnt have read my last 3 posts as this is *exactly * what I
have done. (andfor more than just 1 SN). And my results show clearly
there is no time dilation .

Sorry to be so slow, both on the uptake and with the reply.

To reiterate once again: I have taken Knop lightcurves at I corrected
to B and compared with B from Reiss survey and there is no time
dilation especially on the order of 1.8.

Could you show your work for one or two SN from each group? Say which
ones you used, the date of maximum in each case, and the time to drop
a certain number of magnitudes from maximum (perhaps a couple of
different values).

Hi , sorry about the delay. The method I used was to take individual
SN lightcurves from Knops paper pages 11-12( R.A. Knop
arXiv:astro-ph/0309368 vl 12 Seop 2003) and compare them with ones
supplied in Adam G. Reiss` paper pages 23-25 (arXiv:astro-ph/9810291
vl 19 Oct 1998). I assume the Reiss samples to be all low redshift and
nominally the same redshift and all essentially `rest frame` for the
purposes of comparison with the high redshift SN from Knop. My method
was to then take a Knop lightcurve, recalculate what it would have
been in emission (or `rest frame` I believe its called) and then
compare it with the closest comparable wavelength from the Reiss IRVB
samples. So below are some comparisons I have made between high
redshift Knop and a low redshift Reiss SN `s. So for instance in the
second example below I have taken the observed I band (800nm)
lightcurve of the Knop SN SN1997ek and calculated that it was
initially emmited at a 487nm wavelength. I then overlayed it with the
B band lightcurve SN1995D from Reiss making sure that the flux axis
were identical in height (normalized in a sense although done by
sizing in photoshop) and that the horizontal day to day time axis was
the same for both. the result is that there is no obvious time
dilation despite being low redshift compared to high redshift.
Furthermore if one takes the lowest Knop redshift of 0.35 (emission
592 nm) and compares it with the highest of 0.86 (emisssion430nm) the
result is that they,.. emission range 430-592nm,.. have the same
lightcurve progression as a low redshift Reiss sample of 430nm-592nm
which are identical when overlayed on the time axis (with the peak and
vertical flux axis matched and normalized)


Below are a few samples I have notated for you although I find that
all of Knops` compared with Reiss give roughly the same results.

SN1998be I band z=0.64 = 487nm emission wavelength (B-V Reiss)
SN1998ay I band z=0.64 = 487nm emission wavelength (B-V Reiss)
Both good matches to the Reiss equivelent

SN1997ek I band z=0.86 = 430nm emission wavelength (R Reiss)
Matches Reiss

SN1998ax I band z=0.50 = 533nm emission wavelength (V Reiss)
Matches Reiss

SN1998as I band z=0.35 = 592nm emission wavelength (R-I Reiss)
Not a perfect match but rather matches short end of R , more R-RV than
R, which isnt a problem for me as if there were a time dilation it
would err towards I not V

SN1998as R band z=0.35 = 444nm emission wavelength (B-BV Reiss)
This is a good match on the decay but there is one initial early day
or possibly pre SN observation which doesnt fit but once again its
time contracted so its the opposite of what would be expected if there
were a `time dilation due to BB`. However as I say its a perfect fit
otherwise.

Unfortunately the only R band from Knop that can be matched against a
Reiss equivelent is the SN1998as R band as all the others calculate
down to emission wavelengths smaller than B and the smallest Reiss low
redshift to compare with is B band

Its a bit harder for me to compare the information data points as you
requested as I am unfamiliar with how to convert Knops mag data to
that of the Reiss observed mag datapoints. Knops` seem to be presented
in a different scaling format from Reiss` .Anyways , below SN1997ek
and SN1998ax datapoints as compared to Reiss SN1995D are as follows.
(I have only used Hubbles HST from the Knop tables.)



SN 1997ek I band (Knop)

day... 0 28 40 55
mag 3.83 1.54 0.75 0.46
normalized 1(?) 0.4 0.15 0.1
SN1998ax I band (Knop)
day.... 0 20 31 45 70
mag (no data) 1.95 1.62 0.75 0.47
normalized..." 0.6 0.5 0.25 0.15


SN1995D B band (Reiss)
day..... 0 30 40 54
mag.. 14 16.5 17.3 17.5
normalized.. 0 2.5 3 3.25

As you can see the Knop mags are in a different `mag` format from
Reiss`. The peak mag is a higher numerical value the brighter the
observation reading .I cant figure that out but maybe you can.
(The normalized fluxes for these SN`s are also supplied but to get
them I have to read off the graphs so the values I have supplied are
approximate )
So hopefully this is the data you requested. Incidentally anyone can
request a jpeg file attachment showing this `no time dilation` proof
from me at my www.gammarayburst.com or hotmail address



* An extra point; I notice that the table data is different from the
graph data in some cases . for instance - in 1995D the peak at 30 days
is 16.5 on the graph whereas in the table its 15.5? I`m not sure why
but as I have used the graph data to make my initial claims about no
time dilation I have given it precedence.


regards

Sean

(sean) writes:
Steve Willner wrote in message ...
(sean) wrote in message ...
You couldnt have read my last 3 posts as this is *exactly * what I
have done. (andfor more than just 1 SN). And my results show clearly
there is no time dilation .

Sorry to be so slow, both on the uptake and with the reply.

To reiterate once again: I have taken Knop lightcurves at I corrected
to B and compared with B from Reiss survey and there is no time
dilation especially on the order of 1.8.

Could you show your work for one or two SN from each group? Say which
ones you used, the date of maximum in each case, and the time to drop
a certain number of magnitudes from maximum (perhaps a couple of
different values).

Hi , sorry about the delay. The method I used was to take individual
SN lightcurves from Knops paper pages 11-12( R.A. Knop
arXiv:astro-ph/0309368 vl 12 Seop 2003) and compare them with ones
supplied in Adam G. Reiss` paper pages 23-25 (arXiv:astro-ph/9810291
vl 19 Oct 1998). I assume the Reiss samples to be all low redshift and
nominally the same redshift and all essentially `rest frame` for the
purposes of comparison with the high redshift SN from Knop. My method
was to then take a Knop lightcurve, recalculate what it would have
been in emission (or `rest frame` I believe its called) and then
compare it with the closest comparable wavelength from the Reiss IRVB
samples. So below are some comparisons I have made between high
redshift Knop and a low redshift Reiss SN `s. So for instance in the
second example below I have taken the observed I band (800nm)
lightcurve of the Knop SN SN1997ek and calculated that it was
initially emmited at a 487nm wavelength. I then overlayed it with the
B band lightcurve SN1995D from Reiss making sure that the flux axis
were identical in height (normalized in a sense although done by
sizing in photoshop) and that the horizontal day to day time axis was
the same for both. the result is that there is no obvious time
dilation despite being low redshift compared to high redshift.

On what basis do you claim that "photoshop" is an accurate means to
compare light curves? And, relatedly, on what basis can you claim "no
obvious time dilation," since you have not provided error estimates?

If you are really overlaying the Reiss et al. (1998) and Knop et
al. (2003) light curves by eye using photoshop, then how do you
address the fact that the Knop et al light curves are in relative flux
units (i.e. not logarithmic) and the Reiss et al. light curves are in
(logarithmic) magnitudes, and are thus incompatible with direct visual
overlay?

How do you address the fact that the filter bandpasses of the optical
instruments involved are broad, not narrow, and therefore are not
necessarily compatible with associating them with a single wavelength?

CM

(sean) wrote in message ...
The method I used was to take individual
SN lightcurves from Knops paper pages 11-12( R.A. Knop
arXiv:astro-ph/0309368 vl 12 Seop 2003) and compare them with ones
supplied in Adam G. Reiss` paper pages 23-25 (arXiv:astro-ph/9810291
vl 19 Oct 1998).

It's better to use refereed papers instead of these preprints if you
can. Papers are available from the ADS Abstract Service for most
journals more than three years old. For newer papers, you are stuck
with the preprints unless you have a subscription. It probably
doesn't matter in this case, but I happened to notice another Riess et
al. (note spelling) preprint that was missing a bunch of its tables,
whereas the corresponding journal article was complete.

I assume the Reiss samples to be all low redshift and
nominally the same redshift and all essentially `rest frame` for the
purposes of comparison with the high redshift SN from Knop.

Max redshift is 0.12, so that's a fair assumption.

I then overlayed it with the
B band lightcurve SN1995D from Reiss making sure that the flux axis
were identical in height (normalized in a sense although done by
sizing in photoshop)

I think this is what's wrong.

Below are a few samples I have notated for you although I find that
all of Knops` compared with Reiss give roughly the same results.

SN1997ek I band z=0.86 = 430nm emission wavelength (R Reiss)

The Knop et al. curves are linear flux density, not magnitudes. In
Figure 1, you want the time when the plotted "normalized flux" reaches
0.4, which I make about 39 days in the I band, which is better sampled
than R. (Writing "flux" instead of "flux density" is a very common
mistake. Despite having chided plenty of authors about it, I bet I've
done it myself.)

The wavelength of I is about 814 nm, so we want to compare with
814/1.86=438 nm in the Riess data, i.e., B band. I haven't made a
template but just picked 1995E, z=0.01, because it is well sampled.
The time for 1995E to decay by 1 mag looks like about 17 days to me.
The ratio 29/17=1.71, which looks consistent with the expected 1.85
given the uncertainties of reading off the graphs and the limited
sampling of the light curves. Anyway, it is not consistent with 1.

I haven't done any more examples, but the Knop et al. curves all look
to have longer decay times than the Riess et al. curves.

As you can see the Knop mags are in a different `mag` format from
Reiss`.

Yes, as noted above, Knop et al. use linear flux density. One
magnitude is close to a factor of 2.5, as you no doubt know.

* An extra point; I notice that the table data is different from the
graph data in some cases

If you look at the figure caption, it explains that the B, R, and I
data are plotted with a vertical offset to keep the curves separate.
That won't affect the decay times as long as you consistently use
either the plots or the tables.