1

Is there an equation that describes the Sun's emitted power on the surface [in $\frac{W}{m^2}$] over a selected wavelength range (from $\lambda_1$ to $\lambda_2$) ?

I am guessing this can be calculated using Planck's law, but I just can't find the right equation to integrate it and get the result.

EDIT:

http://www.wikiwand.com/en/Stefan%E2%80%93Boltzmann_law#/Derivation_from_Planck.27s_law

Could I integrate this equation from $\nu_1$ to $\nu_2$ intead of 0 to $\infty$:

$\frac{P}{A} = \frac{2 \pi h}{c^2} \int_0^\infty \frac{\nu^3}{ e^{\frac{h\nu}{kT}}-1} d\nu$

Like this:

$\frac{P}{A} = \frac{2 \pi h}{c^2} \int_{\nu_1}^{\nu_2} \frac{\nu^3}{ e^{\frac{h\nu}{kT}}-1} d\nu$

Where $\nu_1$ and $\nu_2$ are the frequencies of the light.

And then I would use Simpsons rule to numerically integrate and get the result?

And just a quick question, how inaccurate is Planck's law for the Sun at UV spectrum?

user1806687
  • 157
  • 1
  • 9
  • there exists a measurement with the black body curve in wikipedia. http://en.wikipedia.org/wiki/Sunlight . the Planck formula is in the appropriate paragraph here http://en.wikipedia.org/wiki/Black-body_radiation – anna v Nov 14 '14 at 20:32
  • @annav Please see edit. – user1806687 Nov 14 '14 at 20:46
  • Many forms here: http://en.wikipedia.org/wiki/Planck%27s_law#Different_forms – Asaf Nov 14 '14 at 20:52
  • Planck's law gives W/sr/m$^3$. You would need to integrate over solid angle and wavelength to get W/$m^2$. – Kyle Kanos Nov 14 '14 at 21:08
  • See http://www.wikiwand.com/en/Stefan%E2%80%93Boltzmann_law#/Derivation_from_Planck.27s_law . I believe this calculation is very similar to what you would like to do. – Jonathan Gleason Nov 14 '14 at 21:26

1 Answers1

2

The Planck function is tricky to integrate. I found the following http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19680008986.pdf might help. In particular, Appendix C contains numerical tables of integrals for the Planck function between some frequency $\nu$ and infinity. Obviously, by subtracting one of these results using $\nu_1$ and another using $\nu_2$ will give the integral between $\nu_1$ and $\nu_2$.

It also contains the (short) Fortran code to do the integrals yourself.

The flux at the solar surface is $\pi B_{\nu}$ in W m$^{-2}$ Hz$^{-1}$ using the appropriate temperature for the solar photosphere.

NB: The Sun is not a blackbody to any significant precision near the peak of the distribution and instead you should integrate a spectrophotometric solar atlas or possibly a synthetic spectrum generated from a model atmosphere that is appropriately tuned for the Sun. Many such models exist - e.g. http://kurucz.harvard.edu/sun.html It is probably reasonable to treat the Sun as a blackbody from the near infrared and at longer wavelengths.

Here's a rough sketch of the intrinsic solar spectrum (and one after it has passed through the Earth's atmosphere), that I found at http://www.crisp.nus.edu.sg/~research/tutorial/optical.htm and which clearly shows the non-blackbody nature of the spectrum below 1 micron (including in the UV region).

enter image description here

ProfRob
  • 130,455
  • @KyleKanos Yes, this is good. I have added a picture for clarity. – ProfRob Nov 14 '14 at 21:58
  • @RobJeffries Since I cannot figure out, does any of those models contain calculations for UV light? From around 200nm to 500nm. How would I go about integrating those? I am sorry if I am asking for to much, but I am lost and I really need your help. – user1806687 Nov 15 '14 at 10:54
  • @user1806687 I think there are UV solar atlases - the one I linked to stops at 300nm. You will need to do some research. How would you go about integrating them? Well you need to write some software that numerically integrates the data in the wavelength ranges you want. All this depends on how accurately you need the result... There are no ready-made solutions that I know of. – ProfRob Nov 15 '14 at 12:04
  • @RobJeffries I just have no clue what which column means. – user1806687 Nov 15 '14 at 12:25