Units of Spectrum: counts or photons?

[alexkolo]

In general counts don't equal photons. Hence, I'm confused with the given units in the documentation.

In the description of make_spectrum the unit of the energy spectrum is given as:

Returns: Emissivity in counts cm^3 s^-1 bin^-1.

while in the corresponding example it states:

a = pyatomdb.spectrum.make_spectrum(ebins, ite,dummyfirst=True)
ax.loglog(ebins, a, drawstyle='steps', label='Unbroadened')
ax.set_ylabel('Emissivity (ph cm$^{3}$ s$^{-1}$ bin$^{-1}$)')

So, one time it's counts and in another time it's photons. Which one is correct?
I would naively assume counts, if the response had been applied. However, in the example only a dummy response was used. Hence, I'm not sure. How would I make sure that I really get counts and not photons?

The description of the class spectrum.Session may also be incorrect. For its method return_spectra it states that:

Returns: The spectrum in photons cm^5 s^-1 bin^-1, with the response, or photons cm^3 s^-1 bin^-1 if raw is set.

I understand the unit for the raw case. However, if the response is applied, shouldn't the unit be counts cm^5 s^-1 bin^-1 ?

[jagophile]

Hi,

You are correct that there is a distinction between counts and photons, and that the documentation should be updated to reflect this.

The gist of it is, however, that the moment you apply a response, you are dealing with counts, since counts = photons * effective area [from the response]. If you have not applied a response, you have photons.

I will clean up the documentation to make this clearer, but for now does that rule of thumb help?

[alexkolo]

Yes, it does. Thank you.

[alexkolo]

I reopened this since it's hasn't been corrected yet in the documentation.
Feel free to close it, once it was done ;)