search

DavidT3 / XCS-Mass-Paper-I-Analysis

Analysis code and samples for the first XCS cluster mass measurement paper, where we demonstrate our method and perform tests on values from literature
1 stars 0 forks source link

Are MEKAL and APEC temperatures compatible? #28

Closed DavidT3 closed 8 months ago

DavidT3 commented 2 years ago

Unfortunately, I don't know whether temperatures measured in like-for-like conditions with MEKAL and APEC are consistent, and unfortunately I don't remember any papers that have a systematic comparison (and can't find any).

Gonna ask Paul tomorrow, but in the meantime I think I might have to include a comparison in the paper, as LoCuSS did their analysis with MEKAL. Fingers crossed this won't be too difficult, as I should be able to add a function to XGA to do this (https://github.com/DavidT3/XGA/issues/783). The annoyance is that LoCuSS used wabs rather than tbabs, so the function may not ever make it into the main version of the module.

DavidT3 commented 2 years ago

As mentioned in the XGA issue in the original comment above, I'm having to bodge the measurement comparison at the moment because of an annoyingly chosen parameter name in MEKAL. However, in the case where everything is frozen bar normalisation and temperature, I get the following comparison:

image

Now I need to check when metallicity is left free to vary

DavidT3 commented 2 years ago

This shows the comparison between APEC and MEKAL temperatures when metallicity is allowed to vary freely, for the LoCuSS sample. All of the comparisons here are for models fitted on the exact same spectra, with the same inputs - the only difference is the model used.

image

And for these measurements I can also show the metallicity comparison:

image

DavidT3 commented 2 years ago

I should also check how the temperature comparisons look between tbabsapec (the XCS standard model) and wabsmekal (the LoCuSS model) - all the above comparisons have been for tbabsapec and tbabsmekal - the different absorption model might make a difference? I wouldn't necessarily expect them to alter the measured temperatures considering where absorption models have the most effect (low energies) and a lot of temperature constraining comes from the slope beyond 1-2keV, but we'll see!

DavidT3 commented 2 years ago

The comparison between measurements with different absorption models looks essentially identical - I'm not entirely sure if this is true or because of some file mix up but at the moment I don't care enough to find out