Re: Swift Data rules out beamed theory


Craig Markwardt wrote:
"sean" <jaymoseley@xxxxxxxxxxx> writes:
Craig Markwardt wrote:
"sean" <jaymoseley@xxxxxxxxxxx> writes:
George, "Sean's" thinking is quite muddled in this regard. He never
quite got the difference between something that is proportional to
wavelength vs. proportional to the *difference* in wavelength. [ Even
though his "model" depends critically on it, he tried to get me to
define it for him :-]

As usual your dishonesty is offensive. The fact is I claim that
there is no difference between the two. You claim there is but are
unable to supply any argument or proof. My challenge to you still
stands. If you think there is a difference(which there isnt )
between the two, supply examples to the other readers here as
proof that there is. I believe you tried once but your maths
was so appalling and yourunderstanding of scientific concepts
so dim that your proof was as laughable as it was incorrect.

Case in point, you are again asking me to define your theory.
The fact is that an increase proportional to wavelength
and an increase proportional to the difference in wavelength are
both the same .
Both increase in wavelength and both see a time delay increase
proportional to that increase in the wavelength. The longer the
wavelength the longer the delay. That describes both proportional
to wavelength and proportional to difference in wavelength.
Take the first term `proportional to wavelength`...
If you thought about it, if the two wavelengths were the same
the length difference would be 0 and time delay 0.
If one wavelength was 10 times the other and the delay 10 sec
then where the third is 100 times the first the delay will be
100 seconds between the first and third. (100 is 10 times 10)
Thats proportional to wavelength described above.^
If you take the * proportional to difference in wavelength* term
it still works out the same as I show below.
The difference between the first two (1nm and 10nm) is 9nm with a
10 sec delay. Then that means that the difference between the
first and third (100nm) is.... you guessed it 99nm. And the
delay is 100 seconds. (99 is 10 times 9)
Thats the same result mathematically for both
`proportional to wavelength` and `proportional to the difference
in wavelength.` Both terms see the same increase in time delay
with the same increase in wavelength.
So you`re wrong again Craig.


There are other more serious questions that arise. For example,
according to his web page, optical and radio light curves are
stretched versions of the gamma-ray burst. However, from the
photoshopped figure that he shows, he must also be allowing himself
the freedom to shift the time axis as well. With so many freedoms,
it's not surprising that he finds some match-ups. He often confuses a
*detection* of an afterglow with the *peak* of the afterglow, hence he
will claim an IR detection that "comes after" an optical detection as
support for his theory, when in fact one often doesn't really know
what the IR was doing before that time. This accounts for many of his
claimed successes.

To start with t0 in gamma is actually 5 seconds after the burst
is detected in the graph. Technically t0 in gamma is about t-5
as thats where the count rate is measurebly increasing above
background. And if you look at my graph t0 in radio and optical
both are at t-5 in gamma on the graphs time axis. Hence no
artificial alignment in the time axis. All three have t0 at the
same spot on my graph.
So you are wrong again Craig!!

Again, case in point. You have the radio light curve rising at "T-5",
however the radio observations did not start until *after* the burst.
As your photoshopped graph shows, you have not only stretched but also
shifted the light curves (so the T=0 values do not overlap), which
means you allowed additional freedoms, which makes your claimed
association still more dubious.
Ive really got to straighten this out with you before its too late.
Forget the numbers relating to gamma time axis. The 0 is the trigger
time but is not the beginning of the burst nor its peak. Its an
arbitrary point part way up the FRED rise part of the flux
slope used by BATSE to denote when the grb gamma flux increases
above a certain pre set threshold.
t0 is in fact also known as 21:41:47 UT .
And if you go back to the graph and study it then the more
accurate start time in gamma is 8 seconds previously at
about 21:41:39. And if you draw a vertical line on my graph up
from that point in gamma through the radio and optical lightcurves
youll find that the time at those points in those other two lightcurves
are also... 21:41:39 !!! Incredible isnt it? THat means, contrary
to what you claim,that the gamma, radio and optical lightcurves all
start at the same time !!!!
So, wrong again Craig.
George
dishman wrote in an earlier article...


Take care Sean, I'm not saying the spectra
are definitive or unambiguous, I haven't seen
many and I can't remember the papers I saw to
check whether they were SWIFT grism spectra or
ground based folllow-ups. What I am saying is
that Lyman break data are pefectly valid as a
rough indicator which when combined with ground
data, which is also valid, has allowed the
redshift of many GRBs to be deterined. It may
be less than half as you say, but even that
contains many which are definitely at high
redshift so I think that rules out your local
source idea.

George

If I can respond with some of the latest data from
GRB research. Here are 2 recent gcn posts.
The first claims to supply Lyman break evidence in
optical spectra.

No it doesn't. What it says is quite clear:

"We observed the optical transient of GRB 060512
... under POOR OBSERVING CONDITIONS ...

We report a break in the continuum around 4800A
and bluewards which, IF IDENTIFIED as the onset
of the Lyman-alpha forest, indicates a redshift in the
range z ~ 2.7-2.9. The signal-to-noise is TOO LOW to
allow identification of further lines."

(My emphasis)
Im aware of the seeing conditions in that gcn. After all I posted it
here.
But let me put it this way. If they didnt think that their observation
indicated z ~ 2.7-2.9 then why did they post that in their gcn?

As it turns out, it was a data reduction error, as you can read about
in Starling et al. (GCN #5149):

"In both the TNG and the VLT spectra we see no break in the continuum
emission and no obvious features in either absorption or emission."

There is some more discussion about the possible solutions in that GCN.

However, even if there wasn't a data reduction error, there will
occasionally be cases of low significance features where the
interpretation is not clear. That of course does *not* invalidate
other cases with highly significant redshift indicators.
Its true some of the redshift determinations are not ambiguous.
Ive always said that only some are ambiguous . Read my quotes
from other posts.
And usually thats because they are either single spectra, or
spectras where a suspected host galaxy is included or where a
spectra only of a suspected host galaxy is studied. Youll find
that the occasion where more than one spectra of an grb
afterglow is made and where no host galaxy is present that
there are far fewer unambiguous redshift interpretations.

However, this criticism is irrelevant. It's the *HOST* galaxy. The
source is in the galaxy!
Why is it irrelevent?This is the whole argument. I say that
these spectra have been contaminated by galaxies in the FOV
.. And you now you agree with me above! Not only is it relevent
it is the core of the argument.

No Matter what spin you try to put on this the plain fact of it is that

they observed a spectra and felt there was strong enough indication
to merit a gcn posting saying there is a Lyman break at 4800A. If there

was no evidence of a break Im sure they would not have mentioned it.
And that contradicts the other gcn redshift. Thats called ambiguity,

No. See above. As George quite correctly points out, the authors
originally said, "*IF*" the break-like feature could be identified as
the onset of the Ly-alpha forest, then z ~ 2.7-2.9. Please, clear
your head of conspiratorial "spin" theories for just a second and look
at the conditional, "IF." GCN's are for rapid dissemination of new
results, so speculation like that is allowed, and in fact welcomed.
Since afterglows fade quickly, it's better to know about something
interesting (but perhaps wrong) now, than it is to wait three months
for the correct result.
Now we know its been retracted but: my point is this. IF they
didnt think that feature had looked like the onset of the
Lyman A break then they wouldnt have put that in the gcn.


One final time: the claim was *conditional*. The purpose of making
rapid (and possibly incorrect) announcements is to spur rapid
follow-up confirmations. Better to say what something *might* be than
to not say anything and lose the opportunity. So your point is irrelevant.
My point was that they felt, conditional or not, that it warranted
posting on gcn. If it was irrelevent for me to include it in
my analysis then it was an irrelevent gcn. You tell them the bad news,
not me.
In fact it turns out the redshift interpretation they made had nothing
to do with the conditionallity of the seeing conditions after all
as you so wrongly argue. In fact their subsequent gcn admits
it was due to a an *error in the preliminary flux calibration* .
What was once blamed on poor seeing conditions has now
been blamed on... calibration errors!

A further point I would like to make is the assumption
your team make that the white filter shows no
detection below 7500A for grb 0060522. Yet a look at the Swift
UVOT specs page showing the filter area graphs to me indicates that
the white filter has a *very* low sensitivity to light
at anything above 7000A let alone 7500A.
Above 7000A the sensitivity is shown as only about 3% of that of
peak sensitivity and at 7500A about 2%.
Is your team really sugesting that the afterglow
between 7500A and 8000A was so bright that even where the filter
was at only 2% of sensitivity it could still collect enough light
from range 7500A-8000A to produce an afterglow measurement
at 19.65? I very much doubt it. The bat countrate itself shows it
to not be a bright burst. To me the white
filter exposure suggests a detection in optical down to about 6000A
at least to give that sort of afterglow detection .
This implies a more realistic redshift interpretation of 4.1
and clearly conflicts with Kecks observed Lyman break at 7500A
Sean www.gammarayburst.com

.

Relevant Pages

  • Re: Swift grb satelitte
    ... > lightcurve and spectra were not released could be that the SAA ... > this is at odds with the detailed post-burst gcn description. ... > wavelength and a longer delay between wavelengths. ...
    (sci.astro)
  • Re: Swift Data rules out beamed theory
    ... redshift of many GRBs to be deterined. ... optical spectra. ... Im aware of the seeing conditions in that gcn. ... spectra only of a suspected host galaxy is studied. ...
    (sci.astro)
  • Re: Swift Data rules out beamed theory
    ... redshift of many GRBs to be deterined. ... Here are 2 recent gcn posts. ... optical spectra. ... As it turns out, it was a data reduction error, as you can read about ...
    (sci.astro)
  • Re: Swift Data rules out beamed theory
    ... publishing anything else would be misleading ... That's what the GCN system is for. ... here and others of mine outlining how redshift determination ... discernible from the spectra. ...
    (sci.astro)
  • Re: Swift grb satelitte
    ... Subject: Re: Swift grb satelitte ... I also note that you ignored the fact that light curves and spectra ... and in some already-published papers. ... the delay increases proportional to wavelength? ...
    (sci.astro)
Quantcast