From this previous Stackexchange question: Is a detailed absorption spectrum available for carbon dioxide from 300-1100nm?

and this paper: https://www.osti.gov/servlets/purl/1613653

as well as this Wiki article: https://en.wikipedia.org/wiki/Absorption_band

I found atmospheric absorption spectra for CO2, H2O, O2+O3, CH4, and N2O.

The first three being shown in this example from the prior question:

(image from ["Cold Facts on Global Warming"] by T.J.Nelson)

And the second two being shown without units in this Wikipedia image.

(image from ["Wikipedia - Absorption Bands"])

However, there are numerous other gases in the atmosphere, namely:

  • Nitrogen, N2
  • Argon, Ar
  • Neon, Ne
  • Helium, He
  • Krypton, Kr
  • Hydrogen, H2
  • Carbon monoxide, CO
  • Xenon, Xe
  • Nitrogen dioxide, NO2
  • Iodine, I2
  • Ammonia, NH3

Question: Are there absorption / transmittance spectra available for all, some, any of these atmospheric gases across a similar wavelength range (0.1 um to 100 um)? Books, papers, databases, whatever has the data are all acceptable, although I'd like to have intensity or % absorption numbers if possible.

I'm aware of https://en.wikipedia.org/wiki/Spectral_line images, such as Argon's example shown below, although they can be rather difficult to evaluate intensity, and often only exist from 0.4um to 0.7um

Best others I've found was H2 listed this way: http://hyperphysics.phy-astr.gsu.edu/hbase/Tables/hydspec.html

and N2 listed relative to a Deuterium lamp here: Absorption spectrum nitrogen

I also saw this former question where someone was asking for an absorption spectrum library: Absorption spectra of molecules library

  • 1
    $\begingroup$ Re "%absorption", note that your graphs don't show simple line spectra, but illustrate the additional temperature- and pressure-broadening that occurs in the atmosphere. So atmospheric absorption for a particular absorber at frequency $\nu$ is really a function of three variables $(u,p,t)$, where $u$ typically denotes absorber amount. And all that is typically evaluated for a particular thin layer of the atmosphere. So if you're specifically interested in the atmosphere, you'll probably have to consider this more complicated parameterization of absorption. $\endgroup$ Apr 25, 2023 at 22:28
  • 1
    $\begingroup$ The hitran.org database would seem to be the place to start. $\endgroup$
    – Jon Custer
    Apr 25, 2023 at 23:00

1 Answer 1


You have to be careful about the meaning of these spectra. A proper "absorption spectrum" for physical purposes would show the extinction coefficient for one set of conditions, which is independent of the depth of the sample. What your atmospheric gas graphs are showing is the total amount of absorption through all layers of atmosphere at many different pressures and temperatures including scattering. What you can see is that CO2 closes an optical window in the spectrum that was left open by water vapor. As the concentrations of these gases are rising, the windows are getting narrower, but there is no straight forward way from physical extinction coefficients measured on the bench to a graph as shown. For that we would have to know the concentrations and atmospheric layering and then run a raytracing code that would calculate all possible paths that light can take through the atmosphere and and use a physical model (or data) for pressure and temperature broadening of spectral lines.

Jon Custer pre-empted me before I could add the suggestion that "real spectra" are better found in databases like... https://hitran.org/. He is absolutely right about that.

  • 1
    $\begingroup$ Thank you for the answer, and also thank you to @Jon Custer in the comment previously. Looking at Hitran it does appear to be what I'm looking for, although far more complex than I initially thought those calculations might be. Appreciate the help both of you. $\endgroup$
    – G. Putnam
    Apr 26, 2023 at 15:53
  • $\begingroup$ You are welcome. Sorry about the bad news that this is not a simple problem. If it helps, I ran into it as a student a long time ago and began thinking about it. I even wanted to write some code to simulate it... and before I knew it I had half a dozen different effects scribbled down that had to be taken into account, at which point I gave up. What might be interesting is a comparison between a simple line of sight approximation and the measured result. Maybe the zeroth order approximation gets us halfway there? $\endgroup$ Apr 26, 2023 at 22:21
  • 1
    $\begingroup$ I generally agree with that take and it was where I was going to start. Simply take a far away camera aiming straight down, with nearly perpendicular lighting, and see if I can work out bulk absorption %'s similar to those above. Also, found this report from the Air Force's Modtran program that is quite detailed on these topics web.gps.caltech.edu/~vijay/pdf/modrept.pdf Lighting, view angles, molecules / distance, ect... $\endgroup$
    – G. Putnam
    Apr 27, 2023 at 16:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.