There's more starlight than we thought (nearly twice as much)

[marcus]

http://www.nytimes.com/2008/05/17/s...artner=rssnyt&emc=rss&oref=slogin&oref=slogin===sample quote===
Galaxies Twice as Bright as They Seem, Study Finds
By DENNIS OVERBYE
Published: May 17, 2008

...

The galaxies are actually twice as luminous as they appear to us in the sky, according to a new study by an international team of astronomers led by Simon Driver, of the University of St. Andrews in Scotland. Dust, however, blocks half the light from getting out.

The results, which have just been published in The Astrophysical Journal Letters, resolve a longstanding problem with the energy budget of the cosmos.

Interstellar dust absorbs the visible light emitted by stars and then re-radiates it as infrared, or heat, radiation. But when astronomers measured this heat glow from distant galaxies, the dust appeared to be putting out more energy than the stars.

...

He and his colleagues embarked on a program of comparing a model of galactic dust with measurements of the light from 10,000 nearby galaxies using the Isaac Newton Telescope in the Canary Islands and other instruments. That allowed them to calculate by color what percentage of starlight was escaping the galaxies and getting to telescopes: 20 percent of short-wavelength ultraviolet light, 45 percent of green light, 75 percent of the red light, and so on.

The end result, that the stars in galaxies are actually pumping out twice as much energy as previously thought, was still a shock, Dr. Driver said. For the universe as a whole, they calculated, that amounts to 5 quadrillion watts per cubic light-year from thermonuclear fusion, a nice new number for those concerned about their cosmic carbon footprints. ...
==endquote==

the technical article this is based on is
http://arxiv.org/abs/0803.4164
The energy output of the Universe from 0.1 micron to 1000 micron
Simon P. Driver (St Andrews), Cristina C. Popescu (UCLan), Richard J. Tuffs (MPIK), Alister W. Graham (Swin.), Jochen Liske (ESO), Ivan Baldry (LJMU)
(Submitted on 28 Mar 2008)

"The dominant source of electromagnetic energy in the Universe today (over ultraviolet, optical and near-infrared wavelengths) is starlight. However, quantifying the amount of starlight produced has proven difficult due to interstellar dust grains which attenuate some unknown fraction of the light. Combining a recently calibrated galactic dust model with observations of 10,000 nearby galaxies we find that (integrated over all galaxy types and orientations) only (11 +/- 2)% of the 0.1 micron photons escape their host galaxies; this value rises linearly (with log(lambda)) to (87 +/- 3)% at 2.1 micron. We deduce that the energy output from stars in the nearby Universe is (1.6+/-0.2) x 10^{35} W Mpc^{-3} of which (0.9+/-0.1) x 10^{35} W Mpc^{-3} escapes directly into the inter-galactic medium. Some further ramifications of dust attenuation are discussed, and equations that correct individual galaxy flux measurements for its effect are provided."

Comments: Accepted by ApJ Letters

 

[Wallace]

This is very interesting work. I saw Simon give a talk on this a few months ago and the result is quite compelling, it resolves some long standing problems (disagreements between theory and observations) and suggests that some new apparent problems from very high redshift galaxy measurements are also probably due to mis-estimating dust extinction. The last sentence in the abstract is crucial, the 'equations that correct individual galaxy flux measurements'. Such a formalism hasn't existed before, except ones based on simple models that were known to be inadequate, but were the best available.

 

[yoyoq]

does this affect standard candles?

 

[Wallace]

No. Supernovae type 1a spectra are all very similar and dust extinction leads to reddening which would be noticeable. Reddening is always either corrected for, or if it it too bad, that SN is not used as a standard candle. Note that reddening should not be confused with redshift, reddening occurs since dust scatters bluer light more than red, hence the blue part of the spectrum is reduced in intensity compared to the red, whereas redshift moves the entire spectrum in frequency.

The reason it is not so simple to correct for reddening in a galaxy is that some of the stars making up the light will be reddened and some will not. Since the spectrum of a galaxy is the sum total of all the starlight that gets out, you don't know in advance what the pre-reddened spectrum looks like.

 

[yoyoq]

i think there is more to it than that.
if this causes distances to be off for some
of the standard candles, it might cascade and
require others to be fixed.
i found this after posting, he says we need
to wait to see if this affects much

wont let me post a url, but its on the bad astronomy blogbablog/2008/05/15/the-universe-so-bright-its-gotta-wear-shades/

 

[marcus]

wont let me post a url, but its on the bad astronomy blog


bablog/2008/05/15/the-universe-so-bright-its-gotta-wear-shades/

new member rule. I like that blog a lot. will help
http://www.badastronomy.com/bablog/2008/05/15/the-universe-so-bright-its-gotta-wear-shades/

 

[Wallace]

Quoting from the bad astronomy post linked to:

For example, the whole idea of dark energy was found due to unexpected brightnesses of distant supernovae; they were fainter than expected. However, all manners of dusty modeling were used in those calculations, and the results held up. So I don’t expect this to change much when it comes to the new model.

As I said, this won't affect standard candles.

 

[yoyoq]

dark energy != standard candles

 

[yoyoq]

arXiv:0805.3565 (May 2008)
Inclination- and dust-corrected galaxy parameters: Bulge-to-disc ratios and size-luminosity relations
Alister W. Graham and C. C. Worley

from the abstract
"...Using the bulge magnitude corrections from Driver et al., we additionally derive the average, dust-corrected, bulge-to-disc flux ratio as a function of galaxy type. With values typically less than 1/3, this places somewhat uncomfortable constraints on some current semi-analytic simulations"

 

[Wallace]

I had a quick scan but couldn't find where they expand on the semi-analytic simulation problem you quote from the abstract. Have you read this in more detail or have any more info? Sounds like an interesting paper, thanks for bringing it to our attention.