Eqwidt_lu

[Rillck]

Dear all,

Using the equivalent widths, I must calculate the chemical abundances for different spectral lines to derive the stellar parameters. In previous versions of Turbospectrum, I used to use the eqwidt_lu executable for this purpose. However, when migrating to the latest version, I noticed this executable was absent.

To try to get around this problem, I made a modification to the Makefile, as described below:

Original line: all : cubint_module.mod strings_module.mod babsma_lu bsyn_lu

Modified line: all : cubint_module.mod strings_module.mod babsma_lu bsyn_lu eqwidt_lu

When trying to compile the code after this change, I got the following response:

make: *** [Makefile:196: eqwidt_lu] Error 1

Could you please advise me on the possible error or what may be missing for the executable to be correctly generated?

I thank you in advance for your attention and look forward to a possible solution.

Best regards, Rillck G. S. B. de Amorim

[bertrandplez]

Dear Rillck, This feature is no longer supported, as eqwidt.f was not updated along with bsyn.f It needs to be made consistent with a number of subroutine, like bsyn.f is. A more clever update would be to add a flag in bsyn.f directly to allow the computation of eqw. There are several issues to be tackled: 1) read through the line list and compute the eqw of each line, preferably summing isotopic and hyperfine components into a single profile (the latter was never implemented in eqwidt). This must be done instead of summing all line opacities before computing the spectrum. 2) iterate on the abundance to fit an eqwidt. This feature could be left to an external wrapper. However, in order to optimize one should not recompute a lot of things like the chemical equilibrium, but just the line opacity.