Testing HOS on Sims

[jharno]

Testing HOS on new sims

In this hack we'll run our HOS on suites of simulations: SkySim5000, as well as a few scenarios for the new HACC sims

Contacts: Joachim, Cyrille & Jia Day/Time: Fully remote & Asynchronous. Main communication channel: Slack #desc-hos-topical-team GitHub repo: TBD Zoom room (if applicable):

Goals and deliverable

Run your HOS on SkySim5000 mocks and a few scenarios for the new HACC sims to test convergence at the level of summary statistics..

Resources and skills needed

Access to mocks on NERSC, access to Healsparse to read the new mocks.

Detailed description

This hack is meant to test convergence of the new simulation ran by the HACC team. It has a baseline volume and particle count which we will later vary. At the moment we are varying the redshift sampling only (see instructions further below), and the goal is to compare your statistics against the SkySim5000 mocks for each case.

[jharno]

SkySim5000 mock catalogues:

The SkySim5000 catalogues are organised in 5 tomographic bins, which can be found here: /global/homes/j/jharno/IA-infusion/SkySim5000/GalCat/StageIV_nz/V0/

Columns are: [ra_rand, dec_rand, shear1, shear2, weight, z]

[jharno]

SkySim5000 kappa maps:

TBD

[xuod]

The convergence/shear maps from the new HACC simulation are to be found as healsparse files (nside=4096, nside_coverage=32) in /global/cscratch1/sd/xuod/HOS_sims/L845/HACC150/shells_z{nshells}_subsampleauto_groupiso/kappa_Euclid_dndz_fu08_bin1-5/{what}_hacc_nz{iz}_nside4096_seed{seed}.fits where

• the number of snapshots nshells used in one of 19, 50 and 106
• what is one of kappa, gamma and gamma2
• seed is one of 0, 1 and 2, corresponding to different random rotations/flips/translations (consistent across the different choices of nshells).

There are also convergence shear maps for source planes located after each shell. Simply replace kappa_Euclid_dndz_fu08_bin1-5 by kappa_zsnapshots.