dr-guangtou / asap

Modeling the dark matter halo properties across the M100kpc-M10kpc plane

Hi there, this is Song Huang (黄崧) 👋

• I am an assistant professor at the Department of Astronomy of Tsinghua University in Beijing, China.

• I am studying galaxy formation and galaxy-halo connection through observations. In particular, I am now focusing on the life stories of the most massive galaxies in the Universe.

• 🔭 I’m currently a member of the Hyper Suprime-Cam Subaru Strategic Program (HSC SSP; or HSC survey). It is an ambitious imaging survey using the new prime focus camera on the 8.2m Subaru telescope.

• 🔭 I am also leading the scientific preparation of the 6.5-m MUltiplexed Survey Telescope (MUST) for the next generation of spectroscopic surveys.

• 🌱 A few things I am working on ...

  • taotie: A curated collection of resources for astrophysical research.
  • unagi: A tool to help you use the amazing public data from the HSC survey.
  • dsigma: A pure-python galaxy-galaxy lensing framework.
    • Please also see the much improved version 2.0 by Johannes Lange
  • kungpao: Codes I have been using for analysing galaxy images.

• 📫 Contact:

  • Email: shuang [at] tsinghua.edu.cn
  • Website: https://dr-guangtou.github.io/
  • My ORCID: Song Huang
  • My Twitter: @dr_guangtou
    • Also, my BlueSky ID is: dr-guangtou.bsky.social
  • My Github: dr-guangtou
  • My Weibo (微博): @光头怪博士

asap's Issues

Organize the new UniverseMachine data

• Prepare the new catalogs for UM model
  - [x] In situ v,s. ex situ one: python3 um_smdpl_galaxy_catalog.py
  * ~/data/massive/umachine/1806/smdpl_nomergers
  - [x] Galaxy v.s. IHL one: python3 um_smdpl_galaxy_catalog.py
  * ~/data/massive/umachine/1806/smdpl_bestfit
  - [x] MPDL2 Galaxy v.s. IHL one:
  * /Users/song/data/massive/umachine/1806/mdpl2

• Organize the catalogs, make further mass cuts

  • In situ v.s. ex situ one
  • Galaxy v.s. IHL one
  • MPDL2 Galaxy v.s. IHL one

• Precompute the lensing profiles

  • In situ v.s. ex situ one
  • Galaxy v.s. IHL one
  • MPDL2 Galaxy v.s. IHL one

Stellar mass functions for new HSC massive galaxies sample

Measure the stellar mass functions within different apertures for the new S16A massive galaxy samples.

Few notes:

• Limit the redshift range to 0.3 < z < 0.5
• Lower the M100 limit to logM100 > 11.5
• Taking the failed 1-D profiles into accounts.
• Bootstrap for errors.
• Besides the functions for M10 and M100, estimates the SMF for other apertures too, just in case.
• Separate the spec-z/ spec-z+phot-z, and central/satellite galaxies.
• Save the results in both FITS and .npy format.
• Making QA plots.

Pre-compute the deltaSigma signals for S16A massive galaxies

Placeholder...

Prepare the UniverseMachine SMDPL models for forward modeling

We will organize the galaxy catalogs from the UniverseMachine model using the small multidark planck simulation (SMDPL). The results are provided by Peter Behroozi, and the initial catalogs are made by Christopher Bradshaw.

The models are for galaxies at z~0.4. Two models are provided:

1. sfr_catalog_0.712400.txt.gz: old fashion bcg and icl components one.
2. sfr_catalog_insitu_exsitu_0.712400.txt.gz: new in-situ and ex-situ components one.
3. particles_0.712400.txt.gz: DM particles file.

The useful information has been summarized into a smaller catalog, also the central and satellite galaxies are separated now. (.npz) format.

Here, we:

1. Explore the basic properties of these new models.
2. Compare their SMF with observed ones.
3. Explore new M100-M10 models.
4. Select the relevant massive galaxies, and save the value-added catalog.

We also need to downsample the particle tables and pre-compute the DeltaSigma signals around the selected massive galaxies.

Small MultiDark Simulation

• For details, please see: https://www.cosmosim.org/cms/simulations/smdpl/

Clean up the codes

Clean up the old codes from KITP workshops

• Basic functions in hsc_um2 directory
• Make an easy test case under test
• Useful tools under utils

Down-sample the dark matter halo particles for pre-compute

Downsample the dark matter particle table particles_0.712400.txt.gz to 10M particles for pre-compute.

On my laptop, 10 galaxies:

• 10M particles take ~2s;
• 50M particles take ~13s.