French's Primordial Study

This post continues the discussion of 'Ned' Wright's website "Errors in The
Big Bang Never Happened."

Several people have independently pointed me to Ned Wright's website "Errors
in The Big Bang Never Happened." One of those was the sci.physics.research
moderator -- who used this website as justification for refusing any mention
of TBBNH -- or any references contained therein. The crank.dot.net site
lists "Ed (sic) Wright's invaluable page detailing the errors in ... Eric
Lerner's arguments from The Big Bang Never Happened." So I guess it's time
to discuss "Ned Wright's TBBNH Page. Last modified 4-May-2000, © 1997-2000"
http://www.astro.ucla.edu/~wright/lerner_errors.html


While dealing with the issue of elemental abundances (see the thread "Ned
Wright's TBBNH Page (B3)"), Ned Wright referenced the following study:
"Galaxies with the spectra of giant H II regions", French, ApJ, Part 1, vol.
240, Aug. 15, 1980, p. 41-59. The details of this study deserve a close
look. However, I felt that focusing on "French" instead of TBBNH required a
separate thread. This is it.

{Excerpt from Ned Wright's page}
================================================
Lerner wants to make helium in stars. This presents a problem because the
stars that actually release helium back into the interstellar medium make a
lot of heavier elements too. Observations of galaxies with different helium
abundances show that for every 3.2 grams of helium produced, stars produce 1
gram of heavier elements (French, 1980, ApJ, 240, 41). Thus it is not even
possible to make the 28% helium fraction in the Sun without making four
times more than the observed 2% heavier elements fraction,
=============================================== =

I'd like to deal with the study proffered by Ned in some detail.

The full reference is "Galaxies with the spectra of giant H II regions",
Howard B. French, ApJ, Part 1, vol. 240, Aug. 15, 1980, p. 41-59. The
abstract begins: "The properties and nature of 14 galaxies which exhibit
strong, sharp, H II region-like emission lines are investigated." This
study is of a selected set of galaxies with the unusual property of strong,
sharp, H II regions. The study notes explicitly that "The nature of these
galaxies is uncertain. All of the available data are consistent with their
being young objects, at least in the sense of their having only recently
formed most of their stars."

But the value of 3.2 for the ratio of (delta Y / delta Z), generated in the
study (and quoted by Ned) is based on a *theoretical* calculation for the
standard big bang. Specifically, the value is obtained by interpolating
between current local (final) heavy element abundances in the Solar
neighborhood and the 'primordial' He value, calculated in French.

French obtains his He 'primoridal' value from applying equation 6, where 'B'
is (delta He / delta O):

(He/H) = (He/H)_0 + B (O/H)

Figure 6 is the denouement of the study. The plot has 'significant
scatter,' as noted by French. But (as French notes) a trend is visible. The
He/H elemental ratios range from a low of .05 to a high of .12. The O/H
ratios range from .2E-4 to 4E-4. French used a chi squared, linear fit to
the 14 data points to obtain a 'primordial' He value of .069 +- .006. This
is the linear intercept with the ordinate axis at (O/H) = 0. French
provides a 'primordial' He mass fraction of 21.6% +- 1.5%.

But French did NOT use his data fit (which -- by necessity -- calculated B)
in order to determine (delta Y / delta Z). French changed methods. French
explains: "It is difficult to derive delta Y / delta Z from the slope found
from equation (6) because the overall scale of the oxygen abundance might be
affected by such considerations as temperature fluctuations, and because it
is necessary to know the cosmic abundance of oxygen to metals by mass.
Instead, delta Y / delta Z was computed from the derived Y_0 {primoridal
Helium abundance} and assuming that the Orion nebula has Y = .28 and Z =
..02. This gives delta Y / delta Z = 3.2 +- 0.7, in good agreement with
previous results." (The Orion nebula values were taken from the results of
a Peimbert and Torres-Peimbert paper, discussed below.)

Instead of basing the ratio on the data from the 14 galaxies, French took a
one-significant figure estimate of the 'local' Orion Nebula of the Milky
Way, and -- based on the theoretical presumption that the Big Bang was
correct -- ASSUMED that the heavy-element generation would result in the
'correct' modern observed values. French works backward from the local,
currently observed value. In effect assuring that the answer comes out
right in the region we can actually measure well. (And I do not imply that
this was the 'purpose' of the method.)


Lets actually take a look at French's Figure 6 and do a back of the envelope
calculation. According to French, this has two drawbacks. The first is
"temperature fluctuations." This is a valid experimental problem. However
it is ALWAYS a valid experimental problem for every measurement ever done
for elemental abundances. It was dealt with -- for example -- in the
studies that French used for the source of his assumed 2% heavy element
ratio. In fact, French used temperature corrections in order to get the
primordial helium ratio -- which is itself an elemental ratio. Hence we
need not consider the 'temperarure fluctuation' argument a valid complaint
against using the actual data collected and plotted by French.

French does not provide the value of 'B' (delta Y/ delta O) obtained from
the his chi squared fit to the data that was used to generate his value of
(He/H)_0. Unfortunately, my calculator with the chi squared function went
to that big sandpile in the sky. So, I'm doing this first pass linear fit
by eye. This is not so bad, because French gave one endpoint of the line.
The linear fit must pass through (0,.069), as this was the 'primordial'
point reported by French. The linear fit also appears to pass through the
point (3.0E-4, .10). The value of 'B' is therefore (.10-.069) / (3.0E-4 -
0) = 103 {(He/H) / (O/H)}. Or, to put it another way, 103 helium nuclei are
produced for every nucleus of oxygen. Using rough atomic weights,
this is approximately 103 x (4 / 16) = 26. grams of He released for every
gram of oxygen.


The second problem mentioned by French is the need for the abundance of
oxygen as a part of the cosmic 'heavy element' fraction. Contrary to the
implication in French's claim, this is not hard to come by. The very papers
used by French for his 'endpoint' total value of 'heavy elements' provide
this data in detail. Indeed, it is the source of the 'total' that French
borrows.

The source of French's Orion neblua heavy element abundance totals is given
as Peimbert, M. and Torres-Peimbert, S., 1974, ApJ 193, 327. "Chemical
composition of H II regions in the Large Magellanic Cloud and its
cosmological implications." ApJ, v 193, Oct. 15, 1974, pt. 1, p. 327-333.
The source for French's Orion nebula Helium abundance is Peimbert and
Torres-Peimbert, MN RAS 179, 217. Although French used the latter paper for
He, it also provides twelve measurements of the abundance of oxygen.

Table 10 in the second paper (1976) also provides the details of other
elemental abundances measured in the Orion nebula. The numbers are not
significantly different than for the three values provided in the first
(1974) paper. They oxygen values average about 8.79 on the same scale. The
values do not change significantly with source, and are highly repeatable.

From Table 10, Peimbert and Torres-Peimbert, MN RAS 179, 217

Element Log N(x) E/H (number %) E/H (mass %)
------- -------- -------------- ------------
He 11.0 0.10 28.%
C 8.52 3.3E-4 0.40%
N 7.76 5.8E-5 0.08%
O 8.75 5.6E-4 0.90%
Ne 7.90 7.9E-5 0.16%
S 7.41 2.6E-5 0.08%
Cl 5.15 1.4E-7 0.01%
Ar 6.7 5.0E-6 0.02%

Total heavy element mass abundance: 1.6%.

Both of the Peimbert and Torres-Peimbert papers indicated that only 75% of
the heavy elements are in gas form. The other 25% is contained in dust --
which does not show up in the above surveys. Hence, the actual abundances
are increased by a factor of 4/3, and the total heavy element mass abundance
is 2.1%. The Peimbert and Torres-Peimbertat papers both report only a
single significant figure estimate of 2% heavy elements.

However, the above data allows a direct estimate of the ratio (O/Z). Which
is 0.56 (.9/1.6). Oxygen is obviously the best marker for Z. Carbon is in
a poor 2nd place, and the carbon values are more uncertain (see both papers
for details).

So, now that we have a repeatable factor for (O/Z), we can cross-check
French's results by the following equation:

(delta Y/ delta Z) = (delta Y/ delta O) * (O/Z)

where the last ratio is both observationally and theoretically a constant:
(O/Z) = (delta O/delta Z). Hence, we have:

(delta Y/ delta Z) = (26.) * (0.56) = 14.5. Which is four times higher than
the value of 3.2 found by French's back-calculation based on the Big Bang
theory. But this ratio of 14.5 is taken solely from the data used by
French -- but without the theoretical constraints of the Big Bang theory.


If we now repeat Ned Wright's effort, we find that to reach the Solar value
of 28% helium by mass, we need 6.4% helium, starting from French's
'primordial' Helium of 21.6%. We therefore make 6.4% / 14.5 =0.44% heavy
elements. NOT the 2% heavy elements needed by the Big Bang. The Big Bang
is off by a factor of four from the heavy element data, well outside
observational error bands.


How does Lerner's theory look when compared to observational data (instead
of the BB theory results)? Lerner's theory basically states that stars
generate the helium and heavy elements. So, presuming that we need to
generate 28% helium, we will also generate 28% / 14.5 = 1.9% heavy elements.
Well within the 20% error band, estimated earlier. And completely based
upon French's data, without back-calculating from what is theoretically
'needed' to get to current Solar-vicinity values.

It should also be noted that French states that the He values in his study
should be considered 'upper bounds.' The actual He abundances displayed in
Figure 6 may therefore be overestimates. French's 21.6% 'primordial' value
may be high. French observed young galaxies with He abundances as low as
11%, and provided reasoning that these abundances were 'real'. Even though
they are far below the theoretical Big Bang 'primordial' values.


OK, folks. Now I've got some numbers, but I suspect an error. It's TOO
good a 'fit,' and reproduces Lerner's model TOO easily. On which step did I
make an error? (If any.)

1) Using French's 'primordial' value of 0.069 as one point on the line.
2) Determing the value of slope 'B' (103 He nuclei per O nucleus) from the
data plot & 0.069.
3) Converting the above to 26 gm He per gm O.
4) Copying the Log N(x) values from Table 10.
5) Calculating the number percents in Table 10.
6) Calculating the mass percents in Table 10.
7) Adding the mass percents to get the total mass percent.
8) Calculating the O/Z ratio.
9) Adding the dust fraction.
Someplace else.


I'll be spending some more time on this, soon. But I wanted to get out the
'Ned Wright page' posts before some new projects come in later today. So
I'm sending this out more tentatively than I'd like.


A courtesy copy of this post is provided to Ned Wright.

greywolf42
ubi dubium ibi libertas

"greywolf42" wrote in message ...
This post continues the discussion of 'Ned' Wright's website "Errors in The
Big Bang Never Happened."


[...]

Lerner was wrong, and greywolf42's defense of Lerner is wrong. The data
clearly reject the idea that helium and oxygen were both produced in the
same process. Systematic effects make the interpretation of line ratios
in terms of Z and Y more uncertain than the raw data. But the observed
effect is definite: nebulae with very different O/H ratios have very
similar He/H ratios.


While dealing with the issue of elemental abundances (see the thread "Ned
Wright's TBBNH Page (B3)"), Ned Wright referenced the following study:
"Galaxies with the spectra of giant H II regions", French, ApJ, Part 1, vol.
240, Aug. 15, 1980, p. 41-59. The details of this study deserve a close
look. However, I felt that focusing on "French" instead of TBBNH required a
separate thread. This is it.

{Excerpt from Ned Wright's page}
================================================
Lerner wants to make helium in stars. This presents a problem because the
stars that actually release helium back into the interstellar medium make a
lot of heavier elements too. Observations of galaxies with different helium
abundances show that for every 3.2 grams of helium produced, stars produce 1
gram of heavier elements (French, 1980, ApJ, 240, 41). Thus it is not even
possible to make the 28% helium fraction in the Sun without making four
times more than the observed 2% heavier elements fraction,
=============================================== =

I'd like to deal with the study proffered by Ned in some detail.


[...]

French obtains his He 'primoridal' value from applying equation 6, where 'B'
is (delta He / delta O):

(He/H) = (He/H)_0 + B (O/H)

Figure 6 is the denouement of the study. The plot has 'significant
scatter,' as noted by French. But (as French notes) a trend is visible. The
He/H elemental ratios range from a low of .05 to a high of .12. The O/H
ratios range from .2E-4 to 4E-4. French used a chi squared, linear fit to
the 14 data points to obtain a 'primordial' He value of .069 +- .006. This
is the linear intercept with the ordinate axis at (O/H) = 0. French
provides a 'primordial' He mass fraction of 21.6% +- 1.5%.


[...]

The only important issue is whether the points in Figure 6 are consistent
with a line through the origin. If both helium and oxygen are both produced
stars then one would expect that fit to work. But the answer is clearly no.
In fact, if French has done his analysis correctly, this line is rejected
by more than 10 standard deviations. I cited this work not because it is
the best current data, but because it was available to Lerner when he wrote
TBBNH. The current data is much more definitive. For example, see
Figure 4 of Schramm and Turner, http://arXiv.org/abs/astro-ph/9706069.
One dataset gives Y_p = 0.232+/-0.003(stat)+/-0.005(sys); while another
gives Y_p = 0.243+/-0.003(stat). The systematic errors affect the scaling
of Y_p but not the level at which the zero-intercept model is rejected.
With this data the helium proportional to oxygen model is rejected by more
than 80 standard deviations.

Most helium in the Universe was not made in stars.


A courtesy copy of this post is provided to Ned Wright.

greywolf42
ubi dubium ibi libertas

Ned Wright wrote in message
om...
"greywolf42" wrote in message
...
This post continues the discussion of 'Ned' Wright's website "Errors in
The
Big Bang Never Happened."


[...]

Lerner was wrong, and greywolf42's defense of Lerner is wrong.

I'm looking forward to seeing your evidence. However, *THIS* post is
specifically about French's study.

The data
clearly reject the idea that helium and oxygen were both produced in the
same process.

This statement conflicts the original argument you levelled at TBBNH --
which was "...for every 3.2 grams of helium produced, stars produce 1 gram
of heavier elements." Quite clearly, your focus is on the *value* 3.2 that
is 'produced by stars' via the same process (otherwise it wouldn't be
constant). But French's DATA cannot support the value of 3.2 reported by
French. The DATA provides a slope of 26., based upon French's 21.6%
'primordial' He value taken from the same data.

There are only 14 data points in French, based upon 'unusual' galaxies,
which are presumed to be very young. And to quote French, there is
'significant scatter' in the data. Hardly any basis to 'clearly reject'
anything. Stellar theory predicts that helium and oxygen are both produced
and emitted by stars. Though not necessarily in the same manner. Which was
one of my points.

Systematic effects make the interpretation of line ratios
in terms of Z and Y more uncertain than the raw data.

To what specific systematic effects are you referring? A similar claim was
made by French. However this statment contradicts the data in French's
references (Peimbert and Torres-Peimbert). In the P&TP references, the
oxygen-to-hydrogen ratio and the helium to hydrogen ratio in gas clouds did
not vary significantly in the local regions of the Milky way, regardless of
source or observer.

But the observed
effect is definite: nebulae with very different O/H ratios have very
similar He/H ratios.

Except for the nebula that French observed with helium levels far below the
BB predictions. Which French described as 'real'. The point is not whether
they were 'similar', but what the trend is. And whether the data support
French's claim about a ratio of 3.2 He/O generation by stars.

{snip the intro, which appears out-of-order from Ned's top-post.}

French obtains his He 'primoridal' value from applying equation 6, where
'B'
is (delta He / delta O):

(He/H) = (He/H)_0 + B (O/H)

Figure 6 is the denouement of the study. The plot has 'significant
scatter,' as noted by French. But (as French notes) a trend is visible.
The
He/H elemental ratios range from a low of .05 to a high of .12. The O/H
ratios range from .2E-4 to 4E-4. French used a chi squared, linear fit
to
the 14 data points to obtain a 'primordial' He value of .069 +- .006.
This
is the linear intercept with the ordinate axis at (O/H) = 0. French
provides a 'primordial' He mass fraction of 21.6% +- 1.5%.


[...]

The only important issue is whether the points in Figure 6 are consistent
with a line through the origin.

There are two theories being tested here. The Big Bang and Lerner's
stellar-generation. So "a line through the origin" is not the 'only
important issue.' If the slope of the line is incorrect, then the Big Bang
is contradicted. And the actual slope in Figure 6 is a factor of 5 too high
for the big bang.

Your focus in your criticism of TBBNH was based on the specific value of
3.2 reported by French. The 3.2 value nicely reproduces the Solar (and
Orion) values of He and O in the BB theory, working from French's 21.6% He
'primordial' value. However, the value of 3.2 was calculated by French by
back-calculating from precisely this local value -- and ignoring his own
data. His own data show a ratio of 26., not 3.2.

If both helium and oxygen are both produced
stars then one would expect that fit to work.

Both He and O *are* produced by stars. I think you mean if they are 'only'
produced by stars (no BB). And, no, that is not correct in Lerner's theory.
Lerner's theory postulates different generations of stars (Outrageously
massive, very massive, massive, and the current galactic mix). These
different generational types of stars each release different ratios of He to
heavy elements. Hence, a straight line is not expected by Lerner (unless
they are only the very first, outrageously massive stars).

But the answer is clearly no.

And that is why the question is -- what is the ratio of helium to oxygen
observed by French? The point is not 'just' to disprove Lerner's theory.
But to validate the BB, as well. French's data *clearly* is too steep to
support the big bang. Which is why I asked for comments as to where my
calculation went wrong -- and which you snipped. So I repeat the list of
questions:

On which step did I make an error? (If any.)

1) Using French's 'primordial' value of 0.069 as one point on the line.
2) Determing the value of slope 'B' (103 He nuclei per O nucleus) from the
data plot & 0.069.
3) Converting the above to 26 gm He per gm O.
4) Copying the Log N(x) values from Table 10.
5) Calculating the number percents in Table 10.
6) Calculating the mass percents in Table 10.
7) Adding the mass percents to get the total mass percent.
8) Calculating the O/Z ratio.
9) Adding the dust fraction.
Someplace else.


In fact, if French has done his analysis correctly, this line is rejected
by more than 10 standard deviations.

And that -- of course -- was precisely my point. 'Did French do his
analysis correctly?' But French's reported value of 3.2, combined with his
0.69% (number) 'primordial' value misses all the data points in figure 6!

I cited this work not because it is
the best current data, but because it was available to Lerner when he
wrote TBBNH.

Precisely why we need to look at French's study. And not 'current' studies.
*You* excoriate Lerner on the basis of French. Either "French" supports
your point, or it doesn't.

The current data is much more definitive. For example, see
Figure 4 of Schramm and Turner, http://arXiv.org/abs/astro-ph/9706069.

I'll be happy to start another thread on the history and current
observational support for He/heavy element generation. But it's not related
to what was available to TBBNH. (It will have to wait a week or so, for
some other projects on my end.)

One dataset gives Y_p = 0.232+/-0.003(stat)+/-0.005(sys); while another
gives Y_p = 0.243+/-0.003(stat). The systematic errors affect the scaling
of Y_p but not the level at which the zero-intercept model is rejected.
With this data the helium proportional to oxygen model is rejected by more
than 80 standard deviations.

And I'll be happy to examine this in a non-TBBNH thread.

Most helium in the Universe was not made in stars.

That -- of course -- is the question. However, your own reference (French)
states that the He ratios should be considered 'upper bounds', not exact
values. Hence there is a very easy solution to your claim that the values
of He are 'too high' to support Lerner's view.

greywolf42
ubi dubium ibi libertas

greywolf42 wrote in message
...

Ned Wright wrote in message
om...
"greywolf42" wrote in message
...
This post continues the discussion of 'Ned' Wright's website "Errors
in
The
Big Bang Never Happened."


[...]

{snip to unclear statement}

The data
clearly reject the idea that helium and oxygen were both produced in the
same process.

This statement conflicts the original argument you levelled at TBBNH --
which was "...for every 3.2 grams of helium produced, stars produce 1 gram
of heavier elements." Quite clearly, your focus is on the *value* 3.2
that
is 'produced by stars' via the same process (otherwise it wouldn't be
constant). But French's DATA cannot support the value of 3.2 reported by
French. The DATA provides a slope of 26., based upon French's 21.6%
'primordial' He value taken from the same data.

Yes, I know that you believe that *most* of the He was formed in the big
bang, not in stars. However, helium and oxygen are most definitely both
produced by stars.

{snip}

greywolf42
ubi dubium ibi libertas

greywolf42 wrote in message
...


{snip}

Keep those hands over the ears and keep humming, cosmologists! You can
ignore the data if you just bury your head in the sand. Don't irritate the
folks who award research grants.

greywolf42
ubi dubium ibi libertas

greywolf42 wrote:

Keep those hands over the ears and keep humming, cosmologists!

Are you aware you are an idiot? Most idiots aren't, I find.

Paul

Paul F. Dietz wrote in message
...
greywolf42 wrote:

Keep those hands over the ears and keep humming, cosmologists!

Are you aware you are an idiot? Most idiots aren't, I find.

I'm curious. Why do you think I'm an idiot? Just because I poke fun at
cosmologists playing at science -- but actively running away from data that
contradicts their pet theories?

Why did you feel the need to post a pure ad hominem attack, instead of
addressing the issues, or showing me where I made an error of substance?

greywolf42
ubi dubium ibi libertas

greywolf42 wrote:

I'm curious. Why do you think I'm an idiot? Just because I poke fun at
cosmologists playing at science -- but actively running away from data that
contradicts their pet theories?

Because you spout manifestly idiotic crap? Because your 'poking'
is intellectually ludicrous?

Why did you feel the need to post a pure ad hominem attack

Because you so richly deserve it, of course. You have not presented
even the tiniest shred of a logical argument against the big bang.

Paul

Paul F. Dietz wrote in message
...
greywolf42 wrote:

I'm curious. Why do you think I'm an idiot? Just because I poke fun at
cosmologists playing at science -- but actively running away from data
that
contradicts their pet theories?

Because you spout manifestly idiotic crap? Because your 'poking'
is intellectually ludicrous?

The 'manifestly' gives your game away.

Why did you feel the need to post a pure ad hominem attack

Because you so richly deserve it, of course. You have not presented
even the tiniest shred of a logical argument against the big bang.

This thread is not on the big bang, but on French's paper. The references
to this page are not even on the big bang -- but on Ned Wright's critique of
a book.

You need to elevate your reading comprehension skills or lower you religious
notions.

greywolf42
ubi dubium ibi libertas