Skip to content
Switch branches/tags
Go to file

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


Earl Bellinger's Ph.D. repository on forward and inverse problems in asteroseismology.

If any of these programs are useful to you, please consider citing one or more of the following:

See Releases for the versions of this repository corresponding to those papers.




  1. mesa_template/ -- directory containing default instructions for a MESA evolutionary track (copied automatically by dispatch)

  2. python3 -- generate initial conditions varied in a quasi-random fashion, calls the following files: * ./ -- shell script for generating a parameterized MESA evolutionary track (e.g. -M 1 for a 1 solar mass model) * Rscript discontinuity.R -- detect discontinuities in the simulated evolution * Rscript summarize.R -- summarize an evolutionary track into a matrix

  3. Rscript collate.R -- collect nearly-evenly-spaced points from each summarized simulation into one big data file simulations.dat; this facilitates the inverse problem

  4. ./ -- Run with settings that facilitate comparison with the asteroseismic modeling portal (AMP)

Analyze Models


  1. Rscript diffusion.R -- plots the initial and final simulation metallicities as a function of mass and diffusion

  2. Rscript inputs.R -- creates a diagram showing the initial conditions of the grid based off of ../forward/initial_conditions.dat which is generated by

Random Forest Regression


  1. Rscript tagesstern.R -- degrade BiSON solar frequencies to the level of what is observable from the 16 Cyg stars for the sake of fair evaluation & comparison

  2. Rscript hare_compile.R -- turn the Hare & Hound data into a format I can parse

  3. Rscript perturb.R -- make Monte-Carlo perturbations of solar, Tagesstern, 16 Cyg, kages, and hares data to account for uncertainties in observed data

  4. python3 -- determine the number of evolutionary tracks, models per evolutionary track, and trees in the forest that are needed for to work well * Rscript forest_evaluate.R -- visualize the output of

  5. python3 -- learn what relates observable data to model properties from ../forward/simulations.dat and predict the properties of the stars in perturb/ * Rscript importances.R -- plots the feature importances of the random forests obtained in * Rscript cyg.R -- plots the predicted quantities of 16 Cyg from against literature values * Rscript us-vs-them.R-- plots the predicted quantities of the KAGES stars and the Hare-and-Hound exercise against the literature values; also creates the diffusion plot for the KAGES stars

  6. Rscript legacy.R -- plots the cumulative distribution functions for estimate uncertainties for the LEGACY targets

Asteroseismic Inversions


  • Coming soon...




  • -- shell script for submitting jobs to the condor queuing system
  • -- shell script for redistributing a MESA model mesh and calculating adiabatic pulsation frequencies via ADIPLS
  • seismology.R -- R script for making seismological calculations from a frequencies data file
  • utils.R -- R utility script for plotting, constants, etc
  • -- python library for generating Sobol (quasi-random) numbers
  • -- generates kernel functions from stellar models



  • jcd-kasc/
    • Rscript CD_diagram.R -- plot an asteroseismic H-R diagram from a grid of MESA/GYRE models and overplot LEGACY data points on it
  • python3 -- make spherical harmonics plots to visualize the pulsation frequencies of solar-like oscillators
  • python3 -- make plots of linear, random, and quasi-random (Sobol) grids to justify the use of the latter
  • python3 -- make plots of linear and non-linear (in this case, XOR) classification problems to illustrate the limitations and usefulness of basic and advanced ML routines
  • Rscript 16CygB.R -- make an annotated power spectrum of 16 Cyg B
  • matlab animate_sph_harm.m -- create animations of spherical harmonics
  • Rscript interp_vs_reg.R -- plot the difference between linear interpolation and regression
  • Rscript plot_nearly-even-spacing.R -- show the result of the linear transport problem on finding nearly-evenly spaced points