Map reveals strange cosmos.

[Nicolaas Vroom]

There are two ways to image the CMB radiation: As two circles or combined as an ellipse. M Tagmark in 2003 used the circle appoarch in the BBC article.
In order to show the CMB radiation as a rotating sphere I wrote a small program in Visual Basic. To get a copy select:
http://users.telenet.be/nicvroom/M.%20Tegmark%20-%20Cosmos.htm
The .exe file is almost at the end of the url.

My current activity is to calculate the Power Spectrum only by using the data (Temperature fluctuations) of this image. The range shown is between -100 micro Kelvin and 100 micro Kelvin.

Nicolaas Vroom

[jacob navia]

Le 02/02/13 23:26, Nicolaas Vroom a �crit :

> M Tagmark in 2003 used the circle appoarch in the BBC article.

Any reference a bit more specific than that?

Thanks

[Phillip Helbig---undress to reply]

In article <mt2.0-23031...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> There are two ways to image the CMB radiation: As two circles or
> combined as an ellipse. M Tagmark in 2003 used the circle appoarch in
> the BBC article.

It might not be clear to all readers: I think what you are discussing
are different projections. The "two circles" would be views of a sphere
from opposite sides (where, of course, we are outside the sphere while
we are inside the sphere of the CMB---in this case, it is like a
celestial globe with stars) while "combined in an ellipse" would be a
projection which shows the whole sky in one image, such as also exist
for maps of the Earth. Of course, there are other projections, but
these two are the only common ones in discussing the CMB (and the
ellipse is much more common).

> My current activity is to calculate the Power Spectrum only by using the
> data (Temperature fluctuations) of this image. The range shown is
> between -100 micro Kelvin and 100 micro Kelvin.

Is this just an exercise in order to understand how it is done or do you
hope to do something new?

[Eric Flesch]

On Sat, 02 Feb, Nicolaas Vroom <nicolaa...@pandora.be> wrote:
>In order to show the CMB radiation as a rotating sphere I wrote a small program in Visual Basic.

I had a look and it is pretty, but rotating spheres do not
traditionally present the same lateral motion on the limb as at the
centre.

[Jos Bergervoet]

On 2/3/2013 10:19 AM, Phillip Helbig---undress to reply wrote:
> In article <mt2.0-23031...@hydra.herts.ac.uk>, Nicolaas Vroom
> <nicolaa...@pandora.be> writes:
>
>> There are two ways to image the CMB radiation: As two circles or
>> combined as an ellipse. M Tagmark in 2003 used the circle appoarch in
>> the BBC article.
>
> It might not be clear to all readers: I think what you are discussing
> are different projections. The "two circles" would be views of a sphere
> from opposite sides (where, of course, we are outside the sphere while
> we are inside the sphere of the CMB---in this case, it is like a
> celestial globe with stars) while "combined in an ellipse" would be a
> projection which shows the whole sky in one image, such as also exist
> for maps of the Earth. Of course, there are other projections, but
> these two are the only common ones in discussing the CMB (and the
> ellipse is much more common).

But if it comes to programming a visualization, then
the sphere with control by mouse (or by giving Euler
angles or something similar) would probably be a good
choice. In the example here, however, control is missing.

>> My current activity is to calculate the Power Spectrum only by using the
>> data (Temperature fluctuations) of this image. The range shown is
>> between -100 micro Kelvin and 100 micro Kelvin.
>
> Is this just an exercise in order to understand how it is done or do you
> hope to do something new?

And, for others who might want to play with it, is
there a link to the raw data? (The ASCII file with
the numbers that give rise to these images?)

--
Jos

[Nicolaas Vroom]

Op zondag 3 februari 2013 10:19:35 UTC+1 schreef Phillip Helbig-het volgende:

It is mainly an exercise to understand and the best way is IMO to do it yourself.

Please go to the document:
http://www.ualberta.ca/~pogosyan/teaching/ASTRO_122/lect31/lecture31.html
And goto the section: "Why are the Boomerang and WMAP Data Important"
Near the left image is written:
"If the universe is closed, light rays from opposite sides of a hot spot
bend toward each other and as a result the hot spot appears to us to be
larger than it actually is."
IMO the following sentence should be added:
" and the cold spots appears to us SMALLER than it actually is "
Question: Why this preference of hot spots while what we are are discusing
here are photon intensities.

This whole issue is tricky because bending towards hotspots for a closed universe implies bending away for coldspots. Does this mean that the universe can be both closed and open?

The deeper issue is: How is image "a" calculated. Does image a, b and c represent the same section of the Universe.

Nicolaas Vroom

[Phillip Helbig---undress to reply]

In article <mt2.0-11107...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> "If the universe is closed, light rays from opposite sides of a hot spot
> bend toward each other and as a result the hot spot appears to us to be
> larger than it actually is."
> IMO the following sentence should be added:
> " and the cold spots appears to us SMALLER than it actually is "

No. The hot spot is simply some observable feature. One could just as
well consider a cold spot. This is not unique to the CMB; it applies to
all angular-size distances in cosmology.

> Question: Why this preference of hot spots while what we are are discusing
> here are photon intensities.

The map of the CMB is a map of temperature.

> This whole issue is tricky because bending towards hotspots for a
> closed universe implies bending away for coldspots.

No, it doesn't.

[Nicolaas Vroom]

Op zondag 3 februari 2013 20:07:55 UTC+1 schreef Jos Bergervoet het volgende:

> On 2/3/2013 10:19 AM, Phillip Helbig---undress to reply wrote:
> > In article <mt2.0-23031...@hydra.herts.ac.uk>, Nicolaas Vroom
> > <nicolaa...@pandora.be> writes:
>
> But if it comes to programming a visualization, then
> the sphere with control by mouse (or by giving Euler
> angles or something similar) would probably be a good
> choice. In the example here, however, control is missing.

Be my guest. But I do not think this is of much scientific value.

>
> >> My current activity is to calculate the Power Spectrum only by using
> >> the data (Temperature fluctuations) of this image. The range shown
> >> is between -100 micro Kelvin and 100 micro Kelvin.

The range is from -200mK to 200 mK.
In reality the coldest (delta) temperatures are lower than -200 mK
and the highest (delta) temperatures are higher than 200 mK

What is worthwhile is to calculate the Power Spectrum from
the surface of the Sun and from the surface of our Earth.

> And, for others who might want to play with it, is
> there a link to the raw data? (The ASCII file with
> the numbers that give rise to these images?)
>

For raw data goto: http://map.gsfc.nasa.gov/media/121238/index.html
Each value is a combination of 4 parameters:
alpha (=255), red, green and blue.
However for some reason NASA used tricky colors.
When you go to: http://space.mit.edu/home/tegmark/saskmap.html
they use for the color scheme from -100 mK to 100 mK the following scheme:
Red ---> X 0 0 0 X X
Green -> 0 0 X X X 0
Blue --> X X X 0 0 0

X stands for the value 255.
The colors are from left to right: magenta, blue, cyan, green, yellow, and red.
In between magenta and blue there are 254 colors:
(255,0,255) next (254,0,255) next (253,0,255) finally (1,0,255) and (0,0,255)
In between all blue and cyan there are also 254 colors etc.
This scheme is simple and allows you to calculate the Temperature
in a straight forward manner.

The Nasa scheme is much more complicated:
The scheme is from -200 mK to 200 mK .
The color code starting with the coldest is:
(33,6 98) (32,7,96) (32,9,95) (32,9,91) (32,10,86) (32,11,77) (32,12,75) (27,20,65) etc
Maybe there are more intermediate values used.
The color code ending with the higest temperature is:
(243,52,17) (241,43,16) (236,34,17) (227,25,18) and (213,19,19)
When you consider the value green: The lowest value is 6, slowly
increases to 255 and than decreases back to 19.
It is more complicated to calculate the temperature using this scheme (?)
IMO what NASA should have done is to use the following scheme:
Red ---> 0 X 0 0 0 X X X
Green -> 0 0 0 X X X 0 X
Blue --> 0 X X X 0 0 0 X
0,0,0 stands for black and X,X,X stands for white with X = 255.

Nicolaas Vroom

[Phillip Helbig---undress to reply]

In article <mt2.0-4804...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> What is worthwhile is to calculate the Power Spectrum from
> the surface of the Sun and from the surface of our Earth.

Can you explain why? Consider also that the data were taken while the
satellite revolved (with the Earth) around the Sun.

[Nicolaas Vroom]

Op maandag 4 februari 2013 08:39:56 UTC+1 schreef Phillip Helbig---undress to reply het volgende:

> In article <mt2.0-11107...@hydra.herts.ac.uk>, Nicolaas Vroom
>
> <nicolaa...@pandora.be> writes:
>
> > "If the universe is closed, light rays from opposite sides of a hot spot
> > bend toward each other and as a result the hot spot appears to us to be
> > larger than it actually is."
> > IMO the following sentence should be added:
> > " and the cold spots appears to us SMALLER than it actually is "
>
> No. The hot spot is simply some observable feature. One could just as
> well consider a cold spot. This is not unique to the CMB; it applies to
> all angular-size distances in cosmology.
>

This whole issue is discussed in more detail here:
http://users.telenet.be/nicvroom/M.%20Tegmark%20-%20Cosmos.htm#hotspots

> > Question: Why this preference of hot spots while what we are are discusing
> > here are photon intensities.
>
> The map of the CMB is a map of temperature.

Maybe it is a matter of both.
When you study the following you can see that the "equator" consists
of 4096 pixels. The question is what is measured.
I do not know the details but I expect that the number of photons
for each pixel is counted for a fixed time period. However that is not
enough you have to do that for each pixel for different frequencies
or wavelengths. The (raw) value for each pixel is than set
with the frequency with the highest count.
However I expect that that is not the end.
I expect that a certain smoothing is taken into account. That means
that the final frequency is calculated based on a couple of surrounding
pixels.
Based on the final frequency you can calculate a color and the temperature
for each picture.

Nicolaas Vroom

[Nicolaas Vroom]

Op dinsdag 5 februari 2013 21:40:48 UTC+1 schreef Phillip Helbig---undress to reply het volgende:

For the Sun what I propose is the measurement of the radiation (temperatures)
of the surface of the Sun.
For the Earth it is the height above and below sea level.
(Of course you can also try Earth surface temperatures)
My prediction is that the three power spectra will look rather similar.

My understanding is, that the calculation Power Spectrum based on the data
in http://map.gsfc.nasa.gov/media/121238/index.html
is purely a mathematical operation. That means it does not matter
if the colors represent temperatures, pressures or flows the power spectrum
calculated will be the same using the same values.
In short the calculation should not take any physics into account.

Nicolaas Vroom.

[Phillip Helbig---undress to reply]

In article <mt2.0-23897...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> > No. The hot spot is simply some observable feature. One could just as
> > well consider a cold spot. This is not unique to the CMB; it applies to
> > all angular-size distances in cosmology.
> >
> This whole issue is discussed in more detail here:
> http://users.telenet.be/nicvroom/M.%20Tegmark%20-%20Cosmos.htm#hotspots

It should be mentioned that the observed sphere of photons is not a
physical object, it has no distance. The observed sphere is something
like the night sky. The same with observable Universe.

Not really. The CMB is not a hard surface, but one can speak of a
distance to an optical depth of 1 or whatever, just like with the "
surface" of the Sun.

This is an article from 2003 which uses both the words decoupled and
recombination. IMO we should use the word decoupling era, because that
was what, assumed, has happened. Recombination reflects going back in
time, which did not happen.

Two different concepts. "Decoupling" refers to matter becoming
transparent to radiation. "Recombination" is something different; it
refers to electrons and nuclei combining to form atoms. The two are
related but not simultaneous. (As you note, "recombination" is a
misnomer. Here, it should be "combination". Generally, in physics,
when this happens, it happens to something which was previously ionized,
hence REcombination.)

What I want to understand is how are those vertical lines calculated?
Why are they not evenly spaced?

This is probably due to the graph being presented with a different axis
to that which was used internally, i.e. angle and multipole moment.

When you test all the images in that same document this is the same:
Always one color is zero, always one value is 255.
When you consider this spherical image this is not true: Max Tegmark's
Home. The picture is blurred. One color that is added is black to make
it look round. That means the original data is modified. Why?

Just to give it the appearance of 3-dimensionality.

> When you study the following you can see that the "equator" consists
> of 4096 pixels. The question is what is measured.
> I do not know the details but I expect that the number of photons
> for each pixel is counted for a fixed time period. However that is not
> enough you have to do that for each pixel for different frequencies
> or wavelengths. The (raw) value for each pixel is than set
> with the frequency with the highest count.
> However I expect that that is not the end.
> I expect that a certain smoothing is taken into account. That means
> that the final frequency is calculated based on a couple of surrounding
> pixels.
> Based on the final frequency you can calculate a color and the temperature
> for each picture.

CMB observations are made with radio-astronomy techniques, i.e. there is
no counting of photons. Several frequencies are measured.

[Phillip Helbig---undress to reply]

In article <mt2.0-16732...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> > In article <mt2.0-4804...@hydra.herts.ac.uk>, Nicolaas Vroom
> > <nicolaa...@pandora.be> writes:
> >
> > > What is worthwhile is to calculate the Power Spectrum from
> > > the surface of the Sun and from the surface of our Earth.
> >
> > Can you explain why? Consider also that the data were taken while the
> > satellite revolved (with the Earth) around the Sun.
>
> For the Sun what I propose is the measurement of the radiation (temperatures)
> of the surface of the Sun.
> For the Earth it is the height above and below sea level.
> (Of course you can also try Earth surface temperatures)
> My prediction is that the three power spectra will look rather similar.

They don't.

> My understanding is, that the calculation Power Spectrum based on the data
> in http://map.gsfc.nasa.gov/media/121238/index.html
> is purely a mathematical operation. That means it does not matter
> if the colors represent temperatures, pressures or flows the power spectrum
> calculated will be the same using the same values.

Yes, but the Sun looks different than the CMB, so the power spectrum
will look different.

> In short the calculation should not take any physics into account.

Yes, but that doesn't mean that everything looks the same.

[Jos Bergervoet]

On 2/5/2013 3:28 PM, Nicolaas Vroom wrote:
...

>> And, for others who might want to play with it, is
>> there a link to the raw data? (The ASCII file with
>> the numbers that give rise to these images?)
>>
> For raw data goto: http://map.gsfc.nasa.gov/media/121238/index.html

There may be a misunderstanding here, but I don't
see any (link to) data on that page. Also if I surf
around over the links given, no data ever come within
reach..

> Each value is a combination of 4 parameters:
> alpha (=255), red, green and blue.
> However for some reason NASA used tricky colors.
> When you go to: http://space.mit.edu/home/tegmark/saskmap.html
> they use for the color scheme from -100 mK to 100 mK the following scheme:
> Red ---> X 0 0 0 X X
> Green -> 0 0 X X X 0
> Blue --> X X X 0 0 0

I get the impression that you are trying to obtain
data here by reverse-engineering some visualized
representation of the data.

That may be the only way if the original data set
is not publicly available.

> X stands for the value 255.
> The colors are from left to right: magenta, blue, cyan, green, yellow, and red.
> In between magenta and blue there are 254 colors:
> (255,0,255) next (254,0,255) next (253,0,255) finally (1,0,255) and (0,0,255)
> In between all blue and cyan there are also 254 colors etc.
> This scheme is simple and allows you to calculate the Temperature
> in a straight forward manner.

You mean: backwards interpreting a color picture
that is created by someone else from the actual
data set. (As far as I can follow you. Correct me
if I'm wrong!)

> The Nasa scheme is much more complicated:
> The scheme is from -200 mK to 200 mK .
> The color code starting with the coldest is:
> (33,6 98) (32,7,96) (32,9,95) (32,9,91) (32,10,86) (32,11,77) (32,12,75) (27,20,65) etc
> Maybe there are more intermediate values used.

Well, if they want to keep the data secret, they
should of course use a complicated color scheme.
If not, then my question is still: where are the
data? (Simple black-on-white ascii text will do!)

[Mod. note: of course the COBE and WMAP data are not secret. The first
hit on Googling for 'COBE data' should get you somewhere useful. The
WMAP data are at
http://lambda.gsfc.nasa.gov/product/map/dr5/m_products.cfm . I am one
of many non-WMAP-affiliated scientists who has written papers using
the WMAP data --mjh]

--
Jos

[Steve Willner]

In article <mt2.0-21352...@hydra.herts.ac.uk>,

Phillip Helbig---undress to reply <hel...@astro.multiCLOTHESvax.de> writes:
> Two different concepts. "Decoupling" refers to matter becoming
> transparent to radiation. "Recombination" is something different; it
> refers to electrons and nuclei combining to form atoms. The two are
> related but not simultaneous.

There's a nomenclature problem in that "decoupling" is sometimes used
to refer to moment when matter and radiation no longer have the same
energy density. That's some time in the first few minutes after the
Big Bang. I personally prefer to stick to that usage, but I'm
probably in the minority of astronomers in that view. The takeaway
point is that if you use 'decoupling', you better be sure context
makes clear which one you mean.

I don't see how recombination and decoupling (in the second sense)
can be other than simultaneous. The transformation from plasma to
normal atoms is what made the universe transparent.

> (As you note, "recombination" is a
> misnomer. Here, it should be "combination". Generally, in physics,
> when this happens, it happens to something which was previously ionized,
> hence REcombination.)

I think you meant "previously neutral." In normal physics, neutral
atoms are ionized, and then they recombine. The _process_ of going
from separate nuclei+electrons to neutral atoms is called
"recombination," and it's natural to use the same word (even though
it's a misnomer) for cosmic recombination, even though there were no
neutral atoms prior to then.

> CMB observations are made with radio-astronomy techniques, i.e. there is
> no counting of photons.

Some are made that way. Others, e.g., DIRBE, use photo-detectors,
and still others use bolometers. (DIRBE had two bolometer channels.)
None of this matters once the data are calibrated into physical
surface brightness units. Because the CMB spectrum is an exact
blackbody, the surface brightness units can be expressed as a
temperature. That last conversion may fail where the spectrum has
been affected by foregrounds, e.g., Sunyaev-Zeldovich clusters.

The really tricky part is subtracting non-cosmological foregrounds:
the Zodiacal light, Galactic sources, and bright local galaxies. I
believe that's the dominant uncertainty in the whole-sky data, though
not necessarily in smaller regions (higher multipoles).

There's a huge amount of information at
http://lambda.gsfc.nasa.gov/
I think all the raw and processed data from both COBE and WMAP are
there, but I haven't checked for certain.

--
Help keep our newsgroup healthy; please don't feed the trolls.
Steve Willner Phone 617-495-7123 swil...@cfa.harvard.edu
Cambridge, MA 02138 USA

[Nicolaas Vroom]

Op donderdag 7 februari 2013 00:27:16 UTC+1 schreef Jos Bergervoet het volgende:

> On 2/5/2013 3:28 PM, Nicolaas Vroom wrote:
> ...
> >> And, for others who might want to play with it, is
> >> there a link to the raw data? (The ASCII file with
> >> the numbers that give rise to these images?)
> >>
> > For raw data goto: http://map.gsfc.nasa.gov/media/121238/index.html
> There may be a misunderstanding here, but I don't
> see any (link to) data on that page.

The information is on the right hand side:
The url for the largest display is at:
http://map.gsfc.nasa.gov/media/121238/ilc_9yr_moll4096.png
I use both Corel Photo Paint and Visual Studio 2010 to process the data
and to calculate the temperatures.

> I get the impression that you are trying to obtain
> data here by reverse-engineering some visualized
> representation of the data.

That is correct. I calculate the temperatures based on the color
values red, green and blue. I wish there was an easier way.

> Well, if they want to keep the data secret, they
> should of course use a complicated color scheme.
> If not, then my question is still: where are the
> data? (Simple black-on-white ascii text will do!)
>
> [Mod. note: of course the COBE and WMAP data are not secret. The first
> hit on Googling for 'COBE data' should get you somewhere useful. The
> WMAP data are at
> http://lambda.gsfc.nasa.gov/product/map/dr5/m_products.cfm . I am one
> of many non-WMAP-affiliated scientists who has written papers using
> the WMAP data --mjh]

In the file:
http://lambda.gsfc.nasa.gov/product/map/dr5/pow_tt_spec_get.cfm
shows the calculated power spectrum using LCDM model
(To read the .tar.gz files I used 7 Zip)
In the text is written:
# Column 5 = portion of column3 error attributed to cosmic variance,
# assuming the best-fit LCDM model
My impression is that if you want to retrieve temperature data
you have to study FITS and HEALPIX.
For example:
http://lambda.gsfc.nasa.gov/common/fits_header.cfm?fitsfile=%2Fdata%2Fmap%2Fdr5%2Fskymaps%2F9yr%2Fraw%2Fwmap_band_imap_r9_9yr_V_v5.fits

[Mod. note: correct. You will not get anything scientifically useful
by messing around with PNG files -- mjh]

Nicolaas Vroom

[Phillip Helbig---undress to reply]

In article <mt2.0-22227...@hydra.herts.ac.uk>, Steve Willner

<wil...@cfa.harvard.edu> writes:

> Phillip Helbig---undress to reply <hel...@astro.multiCLOTHESvax.de> writes:
> > Two different concepts. "Decoupling" refers to matter becoming
> > transparent to radiation. "Recombination" is something different; it
> > refers to electrons and nuclei combining to form atoms. The two are
> > related but not simultaneous.
>
> There's a nomenclature problem in that "decoupling" is sometimes used
> to refer to moment when matter and radiation no longer have the same
> energy density. That's some time in the first few minutes after the
> Big Bang. I personally prefer to stick to that usage, but I'm
> probably in the minority of astronomers in that view. The takeaway
> point is that if you use 'decoupling', you better be sure context
> makes clear which one you mean.

Right. There are three things which happen around the same time:
combination (usually called recombination), matter becoming transparent
to radiation, and the energy density of radiation dropping below that of
matter. The three are related, but distinct processes. As you say, the
nomenclature can be confusing here.

> > (As you note, "recombination" is a
> > misnomer. Here, it should be "combination". Generally, in physics,
> > when this happens, it happens to something which was previously ionized,
> > hence REcombination.)
>
> I think you meant "previously neutral."

No, I meant "previously ionized". That is, a neutral sustance is
ionized then it recombines. So, for it to REcombine, it had to have
been ionized previously. However, in the scenario here, before (re)
combination there was the ionized state, but before the ionized state
there was no neutral state, hence it can combine, but not recombine.

> In normal physics, neutral
> atoms are ionized, and then they recombine. The _process_ of going
> from separate nuclei+electrons to neutral atoms is called
> "recombination," and it's natural to use the same word (even though
> it's a misnomer) for cosmic recombination, even though there were no
> neutral atoms prior to then.

Right.

> > CMB observations are made with radio-astronomy techniques, i.e. there is
> > no counting of photons.
>
> Some are made that way. Others, e.g., DIRBE, use photo-detectors,
> and still others use bolometers. (DIRBE had two bolometer channels.)

Right. IIRC, Planck has "traditional" receivers for the lower
frequencies and bolometers for the higher ones, but nothing like a CCD
such as those used for higher frequencies (e.g. optical).

[Steve Willner]

In article <mt2.0-23319...@hydra.herts.ac.uk>,

Phillip Helbig---undress to reply <hel...@astro.multiCLOTHESvax.de> writes:
> There are three things which happen around the same time:
> combination (usually called recombination), matter becoming transparent
> to radiation, and the energy density of radiation dropping below that of
> matter. The three are related, but distinct processes. As you say, the
> nomenclature can be confusing here.

I thought there was a time when nucleons decoupled from gamma rays,
but apparently "decoupling" is not used for that. There is "neutrino
decoupling," which occurs about 1 s after the Big Bang. Dark matter
decoupling should also occur, though when depends on the properties
of the dark matter.

About the best reference I could find quickly is
http://www.oxfordreference.com/view/10.1093/acref/9780198608585.001.0001/acref-9780198608585-e-101

> No, I meant "previously ionized". That is, a neutral sustance is
> ionized then it recombines.

The second sentence is the point I was making about the terminology.
We don't disagree on the physics or even very much on the
terminology.

[Phillip Helbig---undress to reply]

In article <mt2.0-25241...@hydra.herts.ac.uk>, Steve Willner

<wil...@cfa.harvard.edu> writes:

> > There are three things which happen around the same time:
> > combination (usually called recombination), matter becoming transparent
> > to radiation, and the energy density of radiation dropping below that of
> > matter. The three are related, but distinct processes. As you say, the
> > nomenclature can be confusing here.
>

> I thought there was a time when nucleons decoupled from gamma rays,
> but apparently "decoupling" is not used for that. There is "neutrino
> decoupling," which occurs about 1 s after the Big Bang. Dark matter
> decoupling should also occur, though when depends on the properties
> of the dark matter.

In the wider sense, everything decouples at some point, associated with
some (postulated) symmetry breaking in some GUT.

[Nicolaas Vroom]

Op maandag 11 februari 2013 22:34:02 UTC+1 schreef Steve Willner het volgende:

> In article <mt2.0-23319...@hydra.herts.ac.uk>,
> Phillip Helbig---undress to reply <hel...@astro.multiCLOTHESvax.de> writes:
> > There are three things which happen around the same time:
> > combination (usually called recombination), matter becoming transparent
> > to radiation, and the energy density of radiation dropping below that of
> > matter. The three are related, but distinct processes. As you say, the
> > nomenclature can be confusing here.

The most confusing part is a clear description of the processes that
took place where and when. (with an indication how sure we are)

> About the best reference I could find quickly is
> http://www.oxfordreference.com/view/10.1093/acref/9780198608585.001.0001/acref-9780198608585-e-101

1) This document claims:
" Eventually, however, with the plasma at around 3000K, even these
photons become too feeble to prevent atoms forming.
With no free electrons left, photons have nothing to interact with and
travel freely through the Universe - they are said to have decoupled etc".
The question is what means freely? Does this imply undisturbed?

2) My understanding of radiation (photons) is that they are created when
electrons move from a higher band to a lower band.
3) At page 287 of the Book "Astronomy and Cosmology" by Fred Hoyle 1975
below Figure 6.21 is written:
"Because of absorption and reemission and because of scattering inside
a (proto) star, radiation leaks out of the interior only very slowly"
At page 288 below Figure 6.22 is written:
When the temperature near the surface of a newly forming star falls below
4000 K the gases are no longer able to block the escape of radiation
in an effective way"
4) From the document http://arxiv.org/abs/1212.5225 (9 Year Bennett)
At page 83 is written:
"5.3.7.3. ILC Considerations
The primary difficulty with any method of extracting the CMB from the data
is determining how much of the temperature in each pixel is foreground
and how much is CMB. The data only constrain the sum of these two, and
we must make other assumptions in order to separate them.
The ILC specifically assumes that the CMB has a black body spectrum"

IMO what they should have added in #4 is:
How much from foreground, how much from intermediate (proto stars) and
how much from CMB.

Op vrijdag 8 februari 2013 11:41:48 UTC+1 schreef Phillip Helbig---undress to reply het volgende:
> In article <mt2.0-22227...@hydra.herts.ac.uk>,

>
> > > CMB observations are made with radio-astronomy techniques, i.e. there is
> > > no counting of photons.
> >
> > Some are made that way. Others, e.g., DIRBE, use photo-detectors,
> > and still others use bolometers. (DIRBE had two bolometer channels.)
>
> Right. IIRC, Planck has "traditional" receivers for the lower
> frequencies and bolometers for the higher ones, but nothing like a CCD
> such as those used for higher frequencies (e.g. optical).

When you study page 14 of document in #4 above you will see that they use
the word intensity a lot. This indirectly IMO implies photon count.
The document also shows that (only?) 5 frequency bands are measured
(K, Ka, Q, V and W) which indirectly implies that not all
CMB photons are not taken into account

Nicolaas Vroom

[Nicolaas Vroom]

Op vrijdag 8 februari 2013 11:09:43 UTC+1 schreef Nicolaas Vroom het volgende:

> Op donderdag 7 februari 2013 00:27:16 UTC+1 schreef Jos Bergervoet het volgende:
>

> [Mod. note: correct. You will not get anything scientifically useful
> by messing around with PNG files -- mjh]

PNG files are almost the same as BMP files. When you store and read those
files the accuracy stays the same. With JPG files that is not the case.
The biggest problem is to calculate the frequency (temperature)
from the color scheme.
With the fits file (a "text" file) I have the same problem.
The length of each record is 80 characters.
Each color is stored as alpha,red,green of blue.

Only a limited # of alpha values are possible (except 0):
all the values between 52 and 66. Highest is 62
all the values between 181 and 191. Highest is 189
For red all the values are possible
There is a small preference:
the values 0,1,2 and 3 have a very high chance.
Low chances are with 125,126 and 127.
128, 129 130 again have a very high chance
253, 254, 255 have again a low chance
For green and blue all the values have the same chance.
What this mean is that the accuracy IMO is not high.
(For the fits file tested) Of course I can be wrong.

Nicolaas Vroom

[Phillip Helbig---undress to reply]

In article <mt2.0-4647...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> > > There are three things which happen around the same time:
> > > combination (usually called recombination), matter becoming transparent
> > > to radiation, and the energy density of radiation dropping below that of
> > > matter. The three are related, but distinct processes. As you say, the
> > > nomenclature can be confusing here.
>

> The most confusing part is a clear description of the processes that
> took place where and when. (with an indication how sure we are)

Sometimes, even people who should know better get it wrong and/or their
books suffer from typographical errors. Check out
http://www.jb.man.ac.uk/~jpl/cosmo/raine.html for example. Also, check
out http://www.jb.man.ac.uk/~jpl/cosmo/bad.html and in particular the
last item which probably more or less directly answers your questions in
this thread.

> 1) This document claims:
> " Eventually, however, with the plasma at around 3000K, even these
> photons become too feeble to prevent atoms forming.
> With no free electrons left, photons have nothing to interact with and
> travel freely through the Universe - they are said to have decoupled etc".
> The question is what means freely? Does this imply undisturbed?

It means that they probably won't get re-absorbed.

> 2) My understanding of radiation (photons) is that they are created when
> electrons move from a higher band to a lower band.

That's one way, but there are many others.

> IMO what they should have added in #4 is:
> How much from foreground, how much from intermediate (proto stars) and
> how much from CMB.

These days, one observes the CMB at many different frequencies. The CMB
has the same structure at all frequencies (well, almost) whereas
foregrounds have different intensities at different frequencies, so
multi-frequency observations can help remove the foregrounds.

> When you study page 14 of document in #4 above you will see that they use
> the word intensity a lot. This indirectly IMO implies photon count.

Not in the sense in which this term is normally used in physics.

> The document also shows that (only?) 5 frequency bands are measured
> (K, Ka, Q, V and W) which indirectly implies that not all
> CMB photons are not taken into account

One can extrapolate from the observed wave bands.

[Steve Willner]

In article <mt2.0-4647...@hydra.herts.ac.uk>,

Nicolaas Vroom <nicolaa...@pandora.be> writes:
> 1) This document claims:
> " Eventually, however, with the plasma at around 3000K, even these
> photons become too feeble to prevent atoms forming.
> With no free electrons left, photons have nothing to interact with and
> travel freely through the Universe - they are said to have decoupled etc".
> The question is what means freely? Does this imply undisturbed?

Most CMB photons arrive undisturbed. Some are affected by hot gas in
galaxy clusters (Sunyaev-Zeldovich effect), others by interaction
with high-energy particles (inverse Compton effect), and a few others
simply absorbed by one thing or another (such as ionized gas in
galaxies). CMB measurements have to account for these effects.

> 2) My understanding of radiation (photons) is that they are created when
> electrons move from a higher band to a lower band.

In general, electromagnetic radiation is emitted any time a charged
particle accelerates. See Maxwell's equations. For individual
photons, you have to use quantum mechanics, but the basic idea is the
same. The classical emission and absorption formula isn't wildly
wrong for astrophysical plasmas. (My memory is that it's off by a
factor of 5 or so for radio frequency of 1 GHz and typical
temperatures and densities.)

> 3) At page 287 of the Book "Astronomy and Cosmology" by Fred Hoyle 1975
> below Figure 6.21 is written:
> "Because of absorption and reemission and because of scattering inside
> a (proto) star, radiation leaks out of the interior only very slowly"

Yes, that's when the protostar is neutral. Most of the absorption
comes from metals, not hydrogen or helium, but that's a detail.

> At page 288 below Figure 6.22 is written:
> When the temperature near the surface of a newly forming star falls below
> 4000 K the gases are no longer able to block the escape of radiation
> in an effective way"

Yes, they become ionized. Notice that 4000 K is almost the same as
the 3000 K people talk about for the CMB. I haven't worked out the
numbers, but I expect the difference is because protostars are denser
than the CMB plasma.

> 4) From the document http://arxiv.org/abs/1212.5225 (9 Year Bennett)
> At page 83 is written:
> "5.3.7.3. ILC Considerations
> The primary difficulty with any method of extracting the CMB from the data
> is determining how much of the temperature in each pixel is foreground
> and how much is CMB.
> The data only constrain the sum of these two, and
> we must make other assumptions in order to separate them.
> The ILC specifically assumes that the CMB has a black body spectrum"

That's essentially what I wrote a few days ago. These facts are well
known.

> IMO what they should have added in #4 is:
> How much from foreground, how much from intermediate (proto stars) and
> how much from CMB.

Protostars are part of the foreground. They aren't a very big part,
though, and they are confined to specific regions, mostly near the
Galactic plane.

> When you study page 14 of document in #4 above you will see that they use
> the word intensity a lot.

My guess is that they mean "specific intensity," though some people
shorten it. (The "specific" means per unit bandwidth of the
detector.) "Surface brightness" is another term. If you want to do
physical interpretations, you have to keep the units straight, but
that's not difficult.

> This indirectly IMO implies photon count.

Any measurement in physical units implies photon count. The energy
of a photon is Planck's constant times its frequency, so converting
from energy units to photon units is trivial. The actual measurement
can come from any kind of detector.

> The document also shows that (only?) 5 frequency bands are measured
> (K, Ka, Q, V and W) which indirectly implies that not all
> CMB photons are not taken into account

If the CMB were the only source in the sky, one frequency band would
suffice to measure it. More bands are used in order to separate the
foreground contributions, which have different temperatures, from the
desired CMB signal. For example, protostars have temperatures of a
few hundred kelvins and therefore will produce stronger signals at
higher frequencies.

[Nicolaas Vroom]

Op donderdag 14 februari 2013 08:30:43 UTC+1 schreef Phillip Helbig---undress to reply het volgende:

> In article <mt2.0-4647...@hydra.herts.ac.uk>, Nicolaas Vroom

> > IMO what they should have added in #4 is:
> > How much from foreground, how much from intermediate (proto stars) and
> > how much from CMB.
>
> These days, one observes the CMB at many different frequencies. The CMB
> has the same structure at all frequencies (well, almost) whereas
> foregrounds have different intensities at different frequencies, so
> multi-frequency observations can help remove the foregrounds.

When you study http://arxiv.org/pdf/1212.5225v2.pdf at page 45 you can see
that the planet Saturn generates photons which include the same
5 frequency bands which are characteristic for the CMB radiation.
All this noise has to be removed from the observed intensities.
The amount of noise per pixel can be "easily" estimated because Saturn
is a moving target.

When you study http://www.nasa.gov/mission_pages/hubble/science/xdf.html
you can see how much intermediate radiation there is
This image covers an area of approximate one pixel.
The problem is that the galaxies them self (like Saturn) also generate
a lot of CMB radiation. This noise has to be subtracted for all the visible
objects in this one pixel. To do that accurately I expect is very difficult.
Each galaxy in this image is surrounded by small black areas.
When you select a black spot inbetween two galaxies you can claim
that such a spot represents true CMB radiation which comes
from a source immediate behind that point at further distance.
However if you move towards the right (but still left of the galaxy)
this is not true anymore because also CMB radiated is bended.
That means that the source can come from behind the galaxy or
even from almost any place on the right.
When you move over the galaxy towards the right rim the reverse
starts to happen: The source of the CMB radiation can come
from almost any place towards the left.

For galaxies near us, this disturbance is more severe.

What I want to say because all stars at all distances generate
the same frequecies as CMB radiation it is very difficult
to establish which is which (which is real CMB)
Secondly, where the origin is of the CMB radiation. That means
the origin of the CMB radiation (a certain percentage) is
not the position of the pixel measured.

Nicolaas Vroom
http://users.pandora.be/nicvroom/

[Nicolaas Vroom]

Op zaterdag 16 februari 2013 09:45:45 UTC+1 schreef Steve Willner het volgende:

> In article <mt2.0-4647...@hydra.herts.ac.uk>,
>
> Nicolaas Vroom <nicolaa...@pandora.be> writes:

> > 1) This document claims: etc
>
> Most CMB photons arrive undisturbed. etc

This is part of the issue. The major issue is which percentage of
the photons in the 5 frequency bands have their origin very close after
the Big Bang. The problem is that (proto) stars also generate the
same type of photons.
The second issue is that all stars influence the path of the background
radiation. That means the true place of birth becomes obscure.

> > The document also shows that (only?) 5 frequency bands are measured
> > (K, Ka, Q, V and W) which indirectly implies that not all
> > CMB photons are not taken into account
>
> If the CMB were the only source in the sky, one frequency band would
> suffice to measure it. More bands are used in order to separate the
> foreground contributions, which have different temperatures, from the
> desired CMB signal. For example, protostars have temperatures of a
> few hundred kelvins and therefore will produce stronger signals at
> higher frequencies.

The issue is not temperatures but intensities at all frequencies
which all the intermediate stars (galaxies) produce within the
frequency range that is characteristic for the CMB radiation.
In reality this is not done: only a certain # of frequencies
are taken into account.
For the 5 frequencies considered the result is 5 corrected
intensities for each pixel (Healpix)
The second issue is calculate one result for each pixel.
This result could be the frequency of the maximum intensity
but I'am not sure.
The major issue is, as you seem to indicate, that it maybe is
impossible for certain pixels, to calculate any reliable intensity
for certain frequencies which represent CMB radiation.
The final issue is to give a physical interpretation for all
results calculated.

In summary the origin of the photons in the range from 23 to 94 Ghz
is partly from shortly after the Big Bang (500 million years) and
partly from (proto) stars born 1 billion years after the BB until
the present. The problem is how much is each.

Nicolaas Vroom
http://users.pandora.be/nicvroom/

[Richard D. Saam]

On 2/18/13 8:47 AM, Nicolaas Vroom wrote:
>
> In summary the origin of the photons in the range from 23 to 94 Ghz
> is partly from shortly after the Big Bang (500 million years) and
> partly from (proto) stars born 1 billion years after the BB until
> the present. The problem is how much is each.
>

The peak frequency of Black Body radiation at 2.73 K is 160.4 GHz
Isn't it safe to assume that the CMB Black Body curve maintains itself
through 23 to 94 GHz
and can be used as a baseline for other component contribution?

Richard D. Saam

[Phillip Helbig---undress to reply]

In article <mt2.0-29847...@hydra.herts.ac.uk>, "Richard D. Saam"

<rds...@att.net> writes:

> > In summary the origin of the photons in the range from 23 to 94 Ghz
> > is partly from shortly after the Big Bang (500 million years) and
> > partly from (proto) stars born 1 billion years after the BB until
> > the present. The problem is how much is each.
> >
>

> The peak frequency of Black Body radiation at 2.73 K is 160.4 GHz
> Isn't it safe to assume that the CMB Black Body curve maintains itself
> through 23 to 94 GHz
> and can be used as a baseline for other component contribution?

Of course, though historically one had to remove foreground sources in
order to observe this. Once it is established, then one can use this.
Other sources are not black bodies, hence observing in several frequency
bands allows one to remove foreground sources.

Note that almost all photons are CMB photons.

[Richard D. Saam]

I would assume that this statement is true
in the context that there are
no theoretical lower or upper frequency limits
to the CMB Black Body Spectrum.
But there must be limits of some kind.

[Nicolaas Vroom]

Op dinsdag 19 februari 2013 21:52:30 UTC+1 schreef Phillip Helbig---undress to reply het volgende:

>
> Of course, though historically one had to remove foreground sources in
> order to observe this.

One has to remove all contamination i.e. all photons in the
frequency band considered that are non CMB photons.

Figure 12 page 44 (http://arxiv.org/pdf/1212.5225v2.pdf 9 year) shows
what is partly involved. This part is relatif simple.
http://www.nasa.gov/mission_pages/hubble/science/xdf.html shows more.
This part is IMO extremly difficult because only one pixel
(out of many) is considered which contains many galaxies

An additional problem is gravitational lensing, that means the
bending of star light.

Document (http://arxiv.org/pdf/1212.5226v2.pdf 9 years) at page 23
in paragraph 5.3 explains that gravitational lensing can be used
to calculate cosmological parameters. (to our advantage)

The problem is that also CMB photons are bended. This works to our
disadvantage and makes a physical interpretation difficult.

> Once it is established, then one can use this.

This is true in theory. In practice it is difficult to know for sure.

> Other sources are not black bodies, hence observing in several frequency
> bands allows one to remove foreground sources.
>
> Note that almost all photons are CMB photons.

What do you mean ?
CMB photons originated shortly after the BB. Many are captured by intervening
stars which inturn also create photons at the same frequency.
I expect that from certain frequency bands for certain pixels maybe 90%
has to be removed, because of intermediate stars and proto stars.
See Figure 12 mentioned above.

Nicolaas Vroom
http://users.pandora.be/nicvroom/

[Phillip Helbig---undress to reply]

In article <mt2.0-4038...@hydra.herts.ac.uk>, "Richard D. Saam"

<rds...@att.net> writes:

> > Note that almost all photons are CMB photons.
> >

> I would assume that this statement is true
> in the context that there are
> no theoretical lower or upper frequency limits
> to the CMB Black Body Spectrum.
> But there must be limits of some kind.

Most photons in the universe are CMB photons. That is, they originate
there, and not in stars, planetary nebulae, disco lasers etc.

I don't see what limits have to do with this. Of course, at very high
and very low frequencies the intensity of the CMB (or any black body) is
low.

[Phillip Helbig---undress to reply]

In article <mt2.0-18838...@hydra.herts.ac.uk>, Nicolaas Vroom

<nicolaa...@pandora.be> writes:

> An additional problem is gravitational lensing, that means the
> bending of star light.

Of any light, not just star light.

> The problem is that also CMB photons are bended. This works to our
> disadvantage and makes a physical interpretation difficult.

Gravitational lensing has been taken into account in standard CMB
analysis for a long time now.

> > Note that almost all photons are CMB photons.

> What do you mean ?

There is a certain number of photons in any volume of the universe at
any one time. If that volume is large enough to be representative, then
most of the photons originate in the CMB.

> CMB photons originated shortly after the BB. Many are captured by intervening
> stars which inturn also create photons at the same frequency.

Many but not most.