Create mock APOGEE data

Standard python setup.py build/install

sudo python setup.py install

This package required numpy, scipy, astropy, galpy, and healpy. All of these are installed via setup.py. Additionally, the package requires apogee (available here) and mwdust (available here) which are not automatically installed. This program requires the dust maps installed with mwdust. The use of the apogee package is limited to selection functions, and actual APOGEE data is not required.

The program generates mock data in a 3-step process

1. Generate mass samples using an IMF
2. Generate position samples using a galpy density profile
3. Apply APOGEE observational model

Application of the APOGEE observational model:

• matching samples to an isochrone based on stellar mass
• removing samples lying outside the APOGEE observational footprint
• Determining H-band extinction using mwdust
• Apply the APOGEE selection function using redenned H magnitudes and unredenned J-Ks color

Mock generation requires an isochrone to provide H, J, and Ks magnitudes. A sample grid of PARSEC v1.2 isochrones can be downloaded here. This grid has metal fraction 0.0001 <= z <= 0.0060 with spacing 0.0001, and 10 Gyr <= age <= 14 Gyr with spacing 0.5 Gyr.

To begin import dependancies:

import apomock
from galpy import potential
from apogee import select
from astropy import units as apu
import mwdust

We will create mock data corresponding to a power law density profile with slope alpha=2

denspot = potential.PowerSphericalPotential(alpha=2.)
mock = apomock.APOGEEMock(denspot=denspot)

We will now load in the isochrone, if the sample PARSEC grid has been downloaded it can be initialized using load_parsec_isochrone(), otherwise use load_isochrone(). We will select metal fraction of 0.0010 and age of 11 Gyr

mock.load_parsec_isochrone(filename=path_to_parsec_isochrone_grid,z=0.0010,log_age=np.log10(1.1e10))

First sample masses from a Chabrier IMF, we'll make it 10^7 solar masses total. Astropy units are supported for most input.

mock.sample_masses(m_tot=1e7*apu.M_sun)

Now we draw position samples, we'll make the samples lie between galactocentric radius 2 kpc and 70 kpc.

mock.sample_positions(r_min=2*apu.kpc, r_max=70*apu.kpc)

Now we load in the dust map and define the APOGEE selection function. We'll use the most recent Combined19 dust map, and we'll use the main APOGEE DR16 selection function (year 7). Calculation of the selection function is time consuming, it may be useful to save the selection function for future use.

dmap = mwdust.Combined19(filter='2MASS H')
aposf = select.apogeeCombinedSelect(sample='main',year=7)

Finally apply the observational program

mock.apply_selection_function(aposf, dmap)

Documentation is in progress, a few additional features are outlined here. Also see docstrings for individual methods for more information.

The phase-space properties of the mock are held in a galpy.orbit.Orbit instance at mock.orbs. The isochrone-based properties of the mock, such as magnitudes, surface gravity, effective temperature, and more, are accessed using mock.iso[mock.iso_match_indx]. The masses of the samples are accessed at mock.masses. The APOGEE Location IDs of each sample are accessed at mock.locid.

A mock APOGEE allstar file can be generated using mock.make_allstar().

The denspot used for sampling must be spherically symmetric, however there are keywords in sample_positions() that allow transformation of a spherically symmetric density profile to a triaxial density profile with arbitrary rotation.

If using non-PARSEC v1.2 isochrones then they must be loaded using load_isochrone() with a a dictionary (supplied to the argument iso_keys) that links the keys used to access the values in the isochrone with a common set of keys used by this program. The keys are listed below.

The following are required for core functionality:

• 'mass_initial' - Initial mass of each point in the isochrone
• 'jmag' - J-band magnitude
• 'hmag' - H-band magnitude
• 'ksmag' - Ks-band magnitude

The following are only required to make a mock APOGEE allstar file using make_allstar():

• 'z_initial' - Initial metal fraction (z) of each point in the isochrone
• 'logg' - Log base 10 surface gravity
• 'logteff' - Log base 10 effective surface temperature

So for example if the initial mass in the isochrone is accessed by calling iso['Mini'], then one element of iso_keys should be {...,'mass_initial':'Mini',...} and so on.

• Sampling positions from disk and SCF-defined density profiles
• Matching mass samples to a grid of isochrones, rather than just a single isochrone