description of magnorm in documentation is confusing
Issue #30
resolved
The documentation wiki defines magnorm as "The normalization of the flux of the object. We use AB magnitudes at (500 nm)/(1+z) (which is roughly equivalent to V (AB) or g (AB))". The factor of 1/(1+z) in the wavelength definition makes it sound like magnorm is applied to the restframe wavelength that is redshifted to 500nm in the observer's frame. Experimenting with PhoSim has led me to believe this is not the case; that magnorm is applied to the flux at 500nm in the restframe of the source. Could you please clarify this definition so that it does not confuse future users.
Comments (11)
-
-
closing. reopen if its still confusing.
-
- changed status to resolved
-
reporter
I'm sorry, but I don't understand your explanation. In whose frame is that 500nm reckoned: the observer's frame, or the source's restframe?
-
its in the source's rest frame. i think i misspoke in the comments, but not in the original wiki text. since the whole SED is in the source's frame then it is at 500 nm in that frame. but according to us, it is at 500/(1+z). make sense?
-
reporter
Isn't it 500nm * (1+z) according to us? A source at redshift z = 1 will have the wavelengths of its light stretched by a factor of 2 so what the source perceives as 500nm, we perceive at 1000nm.
-
i mean it is normalizing the spectrum at shorter wavelength than 500, since it will be at 500 after redshifting. i see the source of the confusion though. maybe better to just say in the "source frame"?
-
reporter
That has not been my experience. I did some tests comparing PhoSim and the other lsst_sims routines designed to integrate the SED over the LSST bandpasses. Agreement was best when I applied normalization to the SED at 500nm before redshifting the spectrum. I will try to draw up an example code and send it to you so that you can see what I mean.
-
yes, that is correct. just having a hard time stating it correct.
-
just updated the text again. now it doesn't talk about 1+z factors. it just says in the source's rest frame (=SEDs rest frame).
-
reporter
Thanks. I, at least, have an easier time understanding that.
- Log in to comment
i think the confusion may come from the fact that the SED itself that you give it for a particular sources is in the rest frame of the source. Then when PhoSim redshifts the spectrum, i think what is written on the wiki makes sense. It doesn't use 500 nm in the SED to normalize things, and instead it uses 500/(1+z). I'll update the documentation in a second.