a question about spectral analysis

Mao yewei wrote on Mar 21, 2008

Dear all,

I am faced with a question about spectral analysis. In order to measure emission lines, one need to correct stellar absorption for observed spectra, or subtract continuum. Can IRAF carry out this work? Which task may I employ?

Thanks a lot!

Owen

---

Mao yewei wrote on Mar 21, 2008

I find a task named CONTINUUM. It seems to fit continuum for spectra.
Does it work for me?
Any one tried?

---

Mike Fitzpatrick wrote on Mar 21, 2008

CONTINUUM is the right task to use for sontinuum subtraction, there is a similar function in SPLOT if you want to do it interactively. You might also look at DEREDDEN if by 'correct stellar absorption' you mean interstellar redenning.

Overall, the REFERENCES task can be used to look for tasks by keyword, e.g.

cl> refer continuum

Cheers,
-Mike

---

Mao yewei wrote on Mar 21, 2008

thank you! it's very helpful to me.

sorry, i didn't describe "stellar absorption" clearly. that is not interstellar reddening, but absorption lines in observed spectra caused by stellar population. these absorption lines exist at the same wavelength as relevant emission lines. so the measurements are an underestimate of the real flux of emission lines owing to the underlying absorption component. that is why i want to correct the underlying stellar absorption.

---

Francisco Valdes wrote on Mar 21, 2008

To subtract a stellar absorption which falls on emission lines would require subtracting a calibration spectrum obtain in some way. This is similar to dealing with telluric lines. So the the task TELLURIC might be useful if you have a spectrum at a different pointing that samples only the stellar absorption without the emission you are interested in. TELLURIC might not be the right task and maybe this is more along the standard process of sky subtraction. Again you need a spectrum that represents the "sky" which is the spectrum in the absence of the emission line. Simple spectrum subtraction (SARITH) might be enough or you might look at SKYTWEAK. The latter is related to TELLURIC and both have to do with using features in other parts of the spectrum that are common to both the target and sky spectra which you want to subtract as perfectly as possible by scaling one spectrum relative to the other. This then assumes that any absorption in the sky spectrum which falls on your emission features will be the correct scaling and subtraction to leave the pure emission feature.

Yours,
Frank Valdes

---

Mao yewei wrote on Mar 21, 2008

Thank you very much, Frank. I already found that task.

---

Last post on Mar 21, 2008