search

yusuke-takase / grasp2alm

Package supporting conversion from GRASP beam format to spherical harmonic coefficients for CMB experiments
GNU General Public License v3.0
0 stars 2 forks source link

Try to avoid intermediate conversion to Healpix grid #5

Open mreineck opened 2 months ago

mreineck commented 2 months ago

The step of going to a Healpix grid before converting to a_lm should probably be avoided for accuracy reasons. The original beam2alm code did actually produce a Healpix map, but it was only used for diagnostic purposes; the actual SHTs were performed on an equidistant (theta, phi) grid using the function beam_transform%bm_polar2alm.

I'll try to provide the necessary ducc0.sht calls in a pull request soon.

yusuke-takase commented 2 months ago

Thank you for the information. The fortran module beam_transform seems to be worked if there is the function that corresponds sharpf_do_job_d as you say. After the PR I can work on the fortran translation to python!

mreineck commented 2 months ago

Yes, I have exactly the required function in ducc0, and I plan to integrate it directly into my PR, so hopefully no extra work should be necessary.

Actually, maybe I even have a way to extract a_lm directly from the rectangular grid of a GRD beam. That would be very handy, but I cannot promise this yet...