search

editeodoro / Bbarolo

Bbarolo is a 3D fitting tool to derive the kinematics of galaxies from emission-line observations.
http://editeodoro.github.io/Bbarolo/
GNU General Public License v2.0
32 stars 9 forks source link

Vdisp and Vexp #12

Closed rcooke-ast closed 5 years ago

rcooke-ast commented 5 years ago

Hi,

I'm impressed by the results of this code - thanks! I just have (hopefully!) a minor clarification to ask.

I am currently using Bbarolo to fit a ringmodel to some ALMA data of a T Tauri star. I am getting reasonable results, except for Vdisp. It seems like the returned value for Vdisp is just the input value, and this parameter is not being included in the minimisation. Looking at the following code: ringmodel.cpp, it seems like VDISP is not considered a model parameter that can be minimised for a ringmodel - is that true? Also, I noticed a VEXP variable in the above code, but I could not find what this variable means in the documentation.

Any help on either of these points would be greatly appreciated!

Many thanks, Ryan.

editeodoro commented 5 years ago

Hi Ryan, first of all, thanks for using BBarolo for your research!

The file that you're referring to (ringmodel.cpp) is for the classical 2D tilted-ring. In 2D, the velocity dispersion can not be fitted, while VEXP is the radial component of the velocity (I should add that to the docs...).

However, I guess you're interested in the 3D fit. The relevant source file is galfit.cpp and the other files galfit_XXX.cpp. I could make a guess on why the code is not fitting the VDISP, but to be sure I need to take a look to your input parameter file and data (you can send me those privately via email).

What BBarolo version are you using? Can you try to download the latest master branch and see if the problem with VDISP persist?

rcooke-ast commented 5 years ago

Hi @editeodoro,

Thanks for the quick reply!

You're quite right that I'm interested int he 3D fit. I had been using the Mac OSX binaries, as I had trouble getting FFTW-3 installed locally. So I've been unable to check if the master branch improves the fit.

An update: I had been using VDISP=0.2 as my starting guess, and I get 0.3 as output for all rings. I'll send you the model and data

Thanks again for your help! Ryan.

editeodoro commented 5 years ago

This problem with low-velocity dispersion systems has been fixed in version 1.4.3 (not released yet, but downloadable from the master branch). Anybody who encounters the same problem should use the latest non-stable version of the code.

Thanks for reporting this and I am closing this ticket.