Limlam Mocker: A package for Line Intensity Mocks

Limlam mocker is a barebones code to create line intensity maps from a given halo catalogue. It only requires NumPy and SciPy - no special astronomy packages. Most was written in python 2, but python 3 should work as well.

To Run

all parameters are in ./params.py and are very self explanatory. Set them up as you wish and run with

>>> ./lim_mocker.py

This will load in the halo catalogue, assign luminosities to each halo, and bin them up into a 3D intensity map data cube of size (npix_x, npix_y, nmaps). This map will then be saved using the npz format - eg https://docs.scipy.org/doc/numpy-1.12.0/reference/generated/numpy.savez.html . This file contains all required info for the maps - field of view, pixel size, frequency of maps, etc...

A small sample halo catalogue is included, but many more of larger sizes are available by contacting the author.

the basic workflow

The example script provided in lim_mocker.py shows the basic workflow:

• import params initialises the parameters for the mock map.
• load_peakpatch_catalogue loads the halos from the lightcone specified.
• cull_peakpatch_catalogue implements a mass and field of view cutoff.
• Mhalo_to_Lco calculates luminosities for all halos.
• params_to_mapinst generates the map instance from given parameters.
• Lco_to_map populates this map with flux (units of T) from the halos.
• save_maps saves the maps to the npz file as described above.
• map_to_pspec then also calculates a 3D spherically averaged power spectrum for this map.

To add your own L_CO(M,z,...) function

add it to halos_to_luminosity.py, following the ones already there eg:

>>> dict = {'Li':          Mhalo_to_Lco_Li,
>>>        'Padmanabhan': Mhalo_to_Lco_Padmanabhan}

>>> def Mhalo_to_Lco_Li(halos, scatter):
>>>        ...

For testing, you could also make use of the 'arbitrary' model, but this is strongly discouraged, and adding new prescriptions to halos_to_luminosity.py is strongly encouraged.

To add your own halo catalogue

use the template in load_halos.py

Advanced usage: using limlam_mocker directly in your own code

This repo is a complete program in itself, designed so that you can painlessly run lim_mocker.py and not have to worry about the guts. If you want to use code other than lim_mocker.py with this library, the limlam_mocker folder inside this repository will function as a self-contained Python package.

Depending on the OS and the Python version, what may work is to simply place this folder in your local site-packages directory or create a symlink there. So if the contents of e.g. ~/.local/lib/python3.5/site-packages/limlam_mocker (the local site-packages directory for Python 3.5 in Linux) are equal to the contents of the limlam_mocker folder in this repo (not the root contents of this repo!), you should be able to take advantage of limlam_mocker functions from any Python script on your computer, regardless of where it is placed.

Extended usage: using modules from limlam_mocker.extensions in your code

The basic limlam_mocker module allows for calculation of line-intensity auto-correlation spectra. To work with uncertainties on this power spectrum or scenarios like cross-correlation with galaxy surveys (see arXiv:1809.04550), you can also do this:

>>> from limlam_mocker.extensions import llm_xcorr as llmx
>>> from limlam_mocker.extensions import llm_error as llme

llm_xcorr provides an extended version of the default map_to_pspec.py code, reducing redundant definitions of k-space parameters and allowing calculations of auto and cross spectra. llm_error automatically allows for calculation of noise power spectra for heterodyne receivers and all-k signal-to-noise for auto and cross spectra. Documentation on both is currently sparse, but should be improved in future.

This code was written by George Stein - gstein@cita.utoronto.ca

with many additions by Dongwoo Chung - dongwooc@stanford.edu

A version with many more options and functions (useful power spectrum calculation too!) can be found at https://github.com/dongwooc/imapper2, written by Tony Li and Dongwoo Chung.

License

limlam_mocker is licensed under the GNU General Public License v3.0