how to set up the file ".par"

[Couleyr]

I noticed that there are many parameters settings in the file ".par".

First, what is the principle of "BMAJ" and "BMIN" settings? Their changes have a significant impact on the results!

And, how can I set up a mask according to my own needs? If I mask the value of the unwanted place, and then give the mask fits file to ".par" file, I will get the following error: Creating mask (SMOOTH) ... SMOOTH error: new beam smaller than old beam old beam: mm x nn new beam: mm x nn terminate called without an active exception Aborted (core dumped)

[editeodoro]

Hello, BMAJ and BMIN are the major and minor axis of the elliptical PSF in arcsecs. So you need to give the right values for your observations!

For the mask, if you want to give a file that you have made, the correct syntax is:

MASK file(mask.fits)

Cheers,

[Couleyr]

Okay, that is how I set it, but when I add the mask file, I still get the above error message

[editeodoro]

Please, send me your cube and parameter files by email and I'll check them out to figure out what the problem might be.

[Couleyr]

files.zip

[editeodoro]

Ok, you had not changed the syntax for the mask parameter (see above), that's why it was trying to build a different mask. In any case, your mask will not work, because it is 2D, while BBarolo wants a 3D mask because the data are 3D. Don't worry, just let the code making its own mask.

I attach an Archive.zip with a parameter file that will return a quite nice model. I have also slightly modified your FITS file removing a linear baseline in each spectrum (because the noise mean was not zero).

Archive.zip

I should note that this is a nearly edge-on galaxy and BBarolo is known to have problems with these systems (see DiTeodoro&Fraternali2015), so you need to be careful with the interpretation of the output parameters, depending also on your scientific goals. For example the external parts of the rotation curve will be mostly ok, but in the internal parts of the galaxy the rotation velocity may be underestimated. Of course, it is possible to build more sophisticated models, starting from BB's best fit and modify them by hand, but I guess it depends on what's your science case.

Cheers, Enrico

[Couleyr]

Oh, it is a 2D format that I really ignored, and we did not notice the baseline problem! Thank you vey much for your help!

As for what you said about dealing with edge-on galaxy, I basically do not rely on the parameters of the model output at present. I will take a look at the article you recommended later!

Thanks again! PS: By the way, how did you remove the linear baseline? Using the continuum?

Cheers, PB Chen