Skip to main content

Utilities for interacting and interpolating stellar evolution models

Project description

Kīauhōkū

Python utilities for stellar model grid interpolation.

If you find this package useful, please cite Claytor et al. (2020).

Download the model grids from Zenodo.

(C) Zachary R. Claytor
Institute for Astronomy
University of Hawaiʻi
2020 November 23

Kīauhōkū
From Hawaiian:

  1. vt. To sense the span of a star's existence (i.e., its age).
  2. n. The speed of a star (in this case, its rotational speed).

This name was created in partnership with Dr. Larry Kimura and Bruce Torres Fischer, a student participant in A Hua He Inoa, a program to bring Hawaiian naming practices to new astronomical discoveries. We are grateful for their collaboration.

Kīauhōkū is a suite of Python tools to interact with, manipulate, and interpolate between stellar evolutionary tracks in a model grid. It was designed to work with the model grid used in Claytor et al. (2020), which was generated using YREC with the magnetic braking law of van Saders et al. (2013), but other stellar evolution model grids are available.

Installation

Kīauhōkū requires the use of Python 3 and uses the following Python packages:

  • numpy
  • scipy
  • pandas
  • matplotlib
  • miniutils
  • pyarrow (or some package that supports parquet files)
  • numba
  • emcee

Personally, I create a conda environment for this. In this example I'll call it "stars".

conda create -n stars numpy scipy pandas matplotlib pyarrow numba emcee
conda activate stars
pip install git+https://github.com/zclaytor/kiauhoku

Brand new: Kīauhōkū is on PyPI! It requires Python 3.8 or less, but you can finally do this:

pip install kiauhoku

You still need to download the grids from Zenodo and follow grid-specific install instructions. I'm working on including the basic files with the pip install.

I don't care about the documentation. Just let me get started!

  1. Download the model grids from Zenodo.

  2. Open an interactive Python session in the directory with the install script. Here we'll install the "fastlaunch" grid, which is a YREC grid that's been processed with the Rotevol rotation tracer code:

from kiauhoku import rotevol
path_to_grid = wherever_you_installed_grids + '/grids/fastlaunch'
rotevol.install(path_to_grid)
  1. You're ready to go! You can import and interpolate away.
import kiauhoku as kh
grid = kh.load_interpolator('fastlaunch')

How it works

We start with output evolution tracks from your favorite stellar modeling software. For rotevol output, these are the *.out files. Each *.out file has, for one specific initial metallicity and alpha-abundance, a series of evolution tracks for a range of initial masses. The "fastlaunch" grid for kiauhoku has eight *.out files, corresponding to
[M/H] ~ [-1.0, -0.5, 0.0, 0.5] and
[alpha/M] ~ [0.0, 0.4].
Each file contains 171 evolution tracks for 0.30 <= M/Msun <= 2.00 in steps of 0.01*Msun.

  1. First we load the tracks into a pandas MultiIndexed DataFrame and save to a parquet file.

  2. Age is not an optimal dimension for comparing consecutive evolution tracks. For this reason we condense each evolution track in the time domain to a series of Equivalent Evolutionary Phases (EEPs) after the method of Dotter (2016). The EEP-based tracks are packaged into a MultiIndexed DataFrame and saved to parquet.

  3. We finally load the EEP-based tracks into a kiauhoku.stargrid.StarGridInterpolator object. The StarGridInterpolator is based on the DataFrameInterpolator (DFInterpolator) from Tim Morton's isochrones package. It performs linear interpolation between consecutive evolution tracks for an input mass, metallicity, alpha-abundance, and either age or EEP-index. We then pickle the interpolator so it can be accessed quickly and easily.

Basic Usage

Once you have everything running, try doing this:

import kiauhoku as kh
grid = kh.load_interpolator('fastlaunch')
star = grid.get_star_eep((1, 0, 0, 330))

If it works, you should get something close to the sun. The argument to get_star_eep is a tuple containing the model grid indices. In this case, those are mass (in solar units), metallicity, alpha-abundance, and EEP index. See the documentation for more details.

Kīauhōkū comes with MCMC functionality through emcee. See the jupyter notebook mcmc.ipynb for an example.

Installing Custom Model Grids

To install your own custom grid, you will want to create a setup script (see custom_install.py for an example). The only requirements are that your setup file contains (1) a function called setup that returns a pandas MultiIndexed DataFrame containing all your evolution tracks, (2) a variable name that is set to whatever you want your installed grid to be named, and (3) a variable raw_grids_path that sets the path to wherever your custom raw grid is downloaded.

The index for this DataFrame is what all the "get" functions will use to get and interpolate tracks and EEPs. Thus, if you want to access your grid using mass and metallicity, you'll want the DataFrame returned by setup to have mass and metallicity, as well as a column to represent the time/EEP step.

You can also use the setup file to define custom EEP functions (see custom_install.my_RGBump) for an example) and to tell kiauhoku which columns to use in its default EEP functions.

Once your setup file is ready, you can install your custom grid using

import kiauhoku as kh
kh.install_grid('custom_install')

If you create a setup file for your favorite model grid and you'd like it to be public, create a pull request and I'll add you as a contributor!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kiauhoku-1.2.0.tar.gz (1.1 MB view hashes)

Uploaded Source

Built Distribution

kiauhoku-1.2.0-py3-none-any.whl (34.0 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page