PDA

View Full Version : help explain the distribution of narrow band emission in nebulae?



whwiii
2009-Dec-09, 08:05 AM
hello
i recently shot my first tricolor narrow band image:
http://whwastro.homestead.com/files/big/ngc281-HST-1280.jpg
FS102@f/6, astrodon 5nm Ha, Oiii, 3nm Sii filters, SX H9 camera.
~30x20 min each channel
10/5-11/8/09 Los Alamitos CA. bortle white skies.

i'm curious about the distribution of the signal in this emission nebula. my understanding is that the central star illuminates the surrounding hydrogen, explaining the Ha signal. dense areas of molecular gas form pillars with ionization fronts caused by the stellar wind of the hot stars.

here's a blink of the Ha, Sii, and Oiii emissions (i've attempted to balance the nebula signal as the Sii and Oiii are quite weak):
http://tinyurl.com/y96rkdg

why is the Sii signal so strong in the pillars and weak around the central stars
while the Oiii signal seems to have the opposite distribution, almost the compliment of the Sii in the Ha image?

is the Sii more susceptible to stellar wind, or does Sii ionize preferentially at the ionization front of the pillar? does the Oiii somehow change states at the ionization front, it seems to just disappear? could the Oiii be simply picking up a reflection component?

any insight into this would be appreciated

thanks

-bill warden
astro.whwiii.net

StupendousMan
2009-Dec-09, 01:41 PM
The strength of an emission line depends on a number
of different factors:

- the overall density of atoms (and, more
specifically, free electrons)

- the abundance of the ion in question

- the temperature of the local gas

- the ionizing flux from nearby stars

- properties of the energy levels in the ion

If you want to explain the relative strengths of
emission lines from different species in your photograph,
you need to include all these factors -- which means
you need to have a model of the distribution of gas
in the nebula, the relative abundances in different
places, etc.

One of the classic textbooks on this subject is
Osterbrock's "Astrophysics Of Gaseous Nebulae And
Active Galactic Nuclei"

http://www.amazon.com/Astrophysics-Gaseous-Nebulae-Active-Galactic/dp/1891389343/ref=sr_1_1?ie=UTF8&s=books&qid=1260366069&sr=8-1

Spaceman Spiff
2009-Dec-09, 04:41 PM
why is the Sii signal so strong in the pillars and weak around the central stars
while the Oiii signal seems to have the opposite distribution, almost the compliment of the Sii in the Ha image?

is the Sii more susceptible to stellar wind, or does Sii ionize preferentially at the ionization front of the pillar? does the Oiii somehow change states at the ionization front, it seems to just disappear? could the Oiii be simply picking up a reflection component?

any insight into this would be appreciated

thanks

-bill warden
astro.whwiii.net

Hi Bill,

As StupendousMan has noted the answers to your questions depend on a number of physical factors. But maybe I can fill in a few holes.

1) The gas you see glowing in most emission line nebula is gas that is ionized mostly by the energetic UV photons of nearby hot (and thus, massive) stars, often associated with active star forming regions (or planetary nebula -- a dying star).

2) Elements and ions of these elements all absorb and emit at preferred photon energies (wavelengths). The more electrons are that missing from the element, the more energetic the photons must be to create and maintain that state of ionization.

3) As the ionizing radiation passes through a gas column, electrons are knocked off. Depending on the intensity and energy distribution of the incident radiation, the gas column will form an ionization structure, which will in large part determine where the various emission lines of the ions will form.

Crudely illustrated with a few key ions as:

HotStar ---->B-He++----He+(O++)-----(O+,H Balmer)-H/He-(O,S+)--neutrals----molecules.

The source of energetic UV light is noted as "HotStar". This light is incident on the nearby gas column at starting at "B" (and its depth moves toward the right). I then note as a function of depth** within the gas column the key zones of ionization: He++ (helium doubly stripped), He+ (helium missing 1 electron), and H/He (the location of the ionization front beyond which the gas has absorbed most of the ionizing photons). In parentheses are species that have significant emission.

In the first zone the gas is highly ionized because the intensity of the UV light is high and little of the energetic light has yet been absorbed. Depending on the temperature of the star, you might not have a He++ zone (typical HII regions do not have well-formed He++ zones, but PNe surrounding hot white dwarfs do). It is here that the Ne+4 emission lines and then those of Ne++ and He+ form. Hydrogen is highly ionized, and hydrogen line (Balmer) emission remains weak.

In the second zone, nearly all of the photons capable of ionizing He+ have been absorbed (w/in the He++ zone). It is here that the O++ forbidden line doublet appears in emission. Hydrogen remains largely ionized, and so the hydrogen emission lines (formed via recombination: H+ captures an electron e- --> H) are weak but begin to pick up in strength.

Going a bit deeper in depth within the cloud, the doubly-ionized oxygen is replaced by singly ionized oxygen, and so the O+ emission lines begin to pop up. The emission lines of hydrogen and neutral helium rise in strength.

Then WHAM --- an ionization front (or boundary) forms whereupon nearly all photons capable of ionizing H and neutral He have been absorbed. It is right along this boundary that hydrogen and helium recombine and become neutral. The singly ionized heavy elements and the H, He emission lines emit strongly along the ionized side of this boundary, while S+ and O emit strongly right up against the neutral side of the boundary.

Deeper in still, molecules form and molecular emission lines pop up.

Thus to a large degree, the answer to your question boils down to "how thick is the cloud that is emitting the lines? If some portion of it is not thick enough to support a H ionization front, O++ might well emit strongly (and He+ in PNe), but O+ and the hydrogen Balmer lines will be weak, and the lines of S+ and O will be ~zilch. That's why the structures delineated by the emission lines from higher ionized species species appear diffusely voluminous, while those from N+, O+ and the hydrogen Balmer lines -- and especially S+ and O -- appear sharp and distinct, even globule-like in regions of dense gas globules.

But if you really want to understand what is going on, Osterbrock's textbook is the place to go.


** the "depth" of interest here is actually a measure of the product of the gas number density and the physical depth. So a small dense globule might have the same "thickness" in this sense as lower density diffuse gas of a larger physical depth.

parejkoj
2009-Dec-09, 05:45 PM
Dang it... the Calvin fans win again!

Osterbrock & Ferland explain all it quite well, though it's a very technical description (Spaceman Spiff's summary is a great one, though). If you can qualitatively understand chapters 2 and 3 (danger: thar be integrals!), you'll have a good start. In particular, the plots in those chapters show how the ionization fraction changes throughout a cloud, given changes in temperature, density, etc.

In fact, significant portions of chapters 2 and 3 are available via Google Books (http://books.google.com/books?id=HgfrkDjBD98C&dq=osterbrock+%26+ferland&printsec=frontcover&source=bl&ots=4RckEN8IQu&sig=aG7WsyP80tk8LWQW6iD3zGzIFTs&hl=en&ei=V-AfS7mBCdOOlAfWy8XoBQ&sa=X&oi=book_result&ct=result&resnum=2&ved=0CAwQ6AEwAQ#v=onepage&q=&f=false)! Take a look at figure 2.4 on page 32, and figure 2.7 on page 42, and the accompanying text. But keep in mind that each nebula is different, and if you want to really understand your image (nice shots, by the way!), you'll have to account for everything in StupendousMan's post (and maybe more!).

If you have a computing bent, Cloudy (http://www.nublado.org/) is the standard (as far as I'm aware -- I'd love to learn of other major projects) for numerically computing the physical properties of gas clouds.

ngc3314
2009-Dec-09, 06:04 PM
Ah, if I wait long enough someone else will do most of the typing! One more bit to add - many of these nebulae are dynamic systems, with the ionization eating gradually into a surrounding dense cloud. This means that species seen at high density or with low ionization potential (such as [S II]) are especially enhanced at the interface.

Parejkoj - there are a few additional codes out there aside from the old standby Cloudy (and its extension Cloudy-3D). Mappings (http://www.ifa.hawaii.edu/~kewley/Mappings/
), XStar ( http://heasarc.nasa.gov/lheasoft/xstar/xstar.html
), and Mocassin (http://hea-www.harvard.edu/~bercolano/) turned up in a list I found quickly. I thought Grazyna Stasinska might have a released version of her code as well, but if so I can't fid it.

parejkoj
2009-Dec-09, 06:37 PM
ngc3314: have you used any of those, or do you know how they compare to Cloudy? I might need to make use of such a code in the future, and I'm looking for suggestions on which to start with.

(don't mean to derail the thread, whwiii!)

Spaceman Spiff
2009-Dec-09, 11:21 PM
ngc3314: have you used any of those, or do you know how they compare to Cloudy? I might need to make use of such a code in the future, and I'm looking for suggestions on which to start with.

(don't mean to derail the thread, whwiii!)

They each have their strengths and weaknesses, depending on the application. Mappings (or one version of it) includes the option for forming shock structures and their emission. Also if you're interested in modeling galactic spectra from stellar populations in combination with the HII region emission line spectra which accompany hot stars in young populations, then the Mappings-III Starburst99 combination package is probably the way to go (see this latest paper (http://arxiv.org/PS_cache/arxiv/pdf/0908/0908.0460v2.pdf)). Otherwise, Cloudy is probably more robust with regards to modeling HII region spectra in detail. XSTAR is like Cloudy, but is optimized for X-ray photoionized plasmas (and so X-ray spectra). I would guess that XSTAR (Tim Kallman is the author) does better with X-ray spectra, but otherwise Cloudy is possibly your choice between the two (IMO). Mocassin is a Monte Carlo radiation transfer code, operating in 3D (Barbara Ercolano is the author), and optimized around photon/grain interactions in dusty photoionized gas (protoplanetary disks, PNe). Morisset & Stasinska have a cool code optimized around PNe and HII blisters, called NEBU-3D (it is a 3D emission geometry mock-up, but cool nonetheless), and Morisset has configured Cloudy into it (http://xxx.lanl.gov/abs/astro-ph/0605400).

See here (http://adsabs.harvard.edu/abs/2003MNRAS.340.1136E) and here (http://adsabs.harvard.edu/abs/2005MNRAS.362.1038E) for working links to papers describing Mocassin. See here (http://adsabs.harvard.edu/abs/2005MNRAS.360..499M), here (http://adsabs.harvard.edu/abs/2006RMxAA..42..153M) and here (http://adsabs.harvard.edu/abs/2008RMxAA..44..171M) for info on Morisset and Stasinska's NEBU-(pseudo)3D code. Stasinska's code is often used in modeling general nebular emission from distant galaxies. Finally, there is Elitzur's code DUSTY (http://www.pa.uky.edu/%7Emoshe/dusty/), which does 3D radiation transfer for grains.

If geometry isn't crucial to the spectrum of interest, and you're most interested in modeling UV to mid-IR spectra, then Cloudy is probably the most versatile (IMO).

(sorry for the temporary side-railing of this train).


disclaimer: by trade, I am an avid consumer of Cloudy. I am definitely not an unbiased critic as I have not extensively used the other codes, but I did try to convey some useful information based on my knowledge of the literature.

whwiii
2009-Dec-10, 05:48 AM
what a great forum
thanks for the helpful replies
will check out the book
btw forgot to mention it's ngc 281, the pacman nebula
here it is in a more traditional pallet :lol:

http://whwastro.homestead.com/files/temp/ngc281-pacman-Ha-600.jpg

-bill w

parejkoj
2009-Dec-10, 06:52 AM
Thanks that's just the kind of summary I was looking for.