search

telegraphic / pygdsm

Python interface to Global Diffuse Sky Models (GDSM) for the radio sky between 10 MHz - 5 THz
MIT License
30 stars 7 forks source link

Negative temperatures in results from PyGDSM (2016 model) #18

Closed telegraphic closed 8 months ago

telegraphic commented 8 months ago

R. Francis reports:

I ... have generated a range of celestial maps for the various frequencies, but I am puzzled by what I see there.

Firstly, I am generating the output in Rayleigh-Jeans temperatures, since we are generating our basic engineering products in calibrated brightness temperature. Secondly, I am using the option of including the CMB signal (and I gather it is the 2.725K signal, not the anisotropy) since we are looking for the total sky temperature not only the foreground galactic signal.

But the only map which shows a result I might expect is the one at 1.426 GHz, which has values from 2.729 to 117 K.

At the other extreme, the map at 36.5 GHz produces value from -4.5e-4K up to 0.12K, with CMB included. With the include_cmb flag set to False all the values are negative.

So either my understanding of what the maps represent is wrong or I am making an error in using the package (or both).

telegraphic commented 8 months ago

It appears that contrary to previous discussions / documentation, the GSM2016 model does not include the 3K CMB. Below 1 GHz, it does predict higher sky-averaged temperatures. Here's a plot of all-sky mean temperature:

image