Glossary

Quick reference for key terms used throughout the brutus documentation.

Stellar Evolution

EEP
Equivalent Evolutionary Point

An index along an evolutionary track that marks the same phase on every track, whatever the mass. Key values: 202 (ZAMS), 454 (TAMS), 605 (tip of the red giant branch). The default grid spans 202 to 808. See Stellar models for the full list.

ZAMS
Zero-Age Main Sequence

The point when a star begins hydrogen fusion in its core (EEP 202). Marks the start of the main sequence phase.

TAMS
Terminal-Age Main Sequence

The point when core hydrogen is exhausted (EEP 454). The star then evolves off the main sequence toward the subgiant and red giant phases.

isochrone

A curve in the HR diagram connecting stars of the same age but different masses. Used for modeling stellar clusters where stars formed together. Compare with evolutionary track.

evolutionary track

The path a star of fixed initial mass traces through the HR diagram as it evolves. Tracks are parameterized by EEP. Perpendicular to isochrones.

initial mass

The mass of a star at formation, in solar masses (M☉). Combined with age and metallicity, this determines all other stellar properties.

MIST

MESA Isochrones and Stellar Tracks - the stellar evolution models used by brutus. Covers 0.1-300 M☉, [Fe/H] from -4.0 to +0.5, all evolutionary phases from pre-MS through AGB.

IMF
Initial Mass Function

The probability distribution of stellar masses at birth. brutus uses the Kroupa IMF: P(M) ∝ M⁻¹·³ for 0.08-0.5 M☉, P(M) ∝ M⁻²·³ above 0.5 M☉ (normalized over 0.08-100 M☉ by default).

Chemical Abundances

metallicity
[Fe/H]

The logarithmic iron abundance relative to solar: [Fe/H] = log₁₀(N_Fe/N_H) - log₁₀(N_Fe/N_H)☉. Solar is [Fe/H] = 0; metal-poor stars have [Fe/H] < 0.

alpha enhancement
[α/Fe]

The abundance of alpha-process elements (O, Mg, Si, Ca, Ti) relative to iron, compared to solar. Old halo and thick disk stars typically have [α/Fe] ~ +0.3. Note: brutus model interfaces accept alpha enhancement (the afe argument to get_seds and a grid label in load_models), but the shipped MIST data files are computed at [α/Fe] = 0, so models are effectively solar-scaled unless you supply alpha-enhanced grids.

Photometry

magnitude

Logarithmic brightness scale: m = -2.5 log₁₀(F/F_ref). Fainter objects have larger (more positive) magnitudes. A difference of 5 magnitudes corresponds to a factor of 100 in brightness.

flux density
maggie

Linear brightness measurement. brutus works in “maggies”: 1 maggie is the flux of a source of magnitude 0 in the survey’s own system, so flux = 10^(-0.4 × mag). brutus.utils.inv_magnitude() converts magnitudes and their errors.

SED
Spectral Energy Distribution

The distribution of flux across wavelengths or photometric bands. brutus predicts SEDs from stellar models and compares them to observed photometry.

parallax

The apparent angular shift of a star due to Earth’s orbital motion, measured in milliarcseconds (mas). Distance in parsecs = 1000 / parallax_mas.

distance modulus

The difference between apparent and absolute magnitude: μ = m - M = 5 log₁₀(d / 10 pc). A star at 1 kpc has μ = 10 mag.

photometric offset

A multiplicative correction to the flux in one band that removes a systematic difference between the models and the data. brutus.data.load_offsets() reads them; BruteForce.fit applies them through phot_offsets. See Photometric offsets.

Extinction & Dust

extinction

Wavelength-dependent attenuation of starlight by interstellar dust, in magnitudes. A_V denotes extinction in the V-band (~550 nm). Extinction is stronger at shorter wavelengths.

reddening

The color change caused by wavelength-dependent extinction. Quantified as color excess E(B-V) = (B-V)_observed - (B-V)_intrinsic.

R_V

The ratio of total-to-selective extinction: R_V = A_V / E(B-V). Characterizes the dust grain size distribution. Typical values: ~3.1 (diffuse ISM), ~2 (dense molecular clouds), ~5 (diffuse high-latitude).

reddening vector

The direction a star moves in color-color or color-magnitude space due to dust extinction. Depends on the extinction law and stellar spectrum.

line of sight

The direction from the observer toward a point on the sky. Extinction accumulates along it with distance, so stars at different distances in one direction trace the dust in 3D.

Bayestar

The 3D dust map of Green et al. (2019), from Pan-STARRS and 2MASS photometry, covering the sky north of declination -30°. brutus can use it as a prior on extinction; see Priors.

Statistical Inference

posterior

The probability distribution of model parameters given the observed data: P(θ|data) ∝ P(data|θ) × P(θ). This is what brutus computes for each star.

prior

Probability distribution encoding knowledge before seeing the data: P(θ). In brutus: IMF, Galactic structure, dust maps. See Priors.

likelihood

The probability of observing the data given model parameters: P(data|θ). Measures how well a model matches observations.

evidence

The likelihood times the prior, integrated over all parameters: P(data). brutus saves an estimate per star (obj_log_evid), summed over the models it keeps and with unnormalized grid weights. Compare it only between fits that use the same grid and priors.

marginalization

Integrating over nuisance parameters to obtain distributions for parameters of interest. For example, marginalizing over stellar mass to get the distance posterior.

brute force

The grid-based inference approach that gives brutus its name: systematically evaluate the likelihood at all pre-computed grid points, apply priors, then marginalize. Avoids MCMC convergence issues and guarantees complete parameter space coverage.

brutus Classes

model grid

A set of stellar models at fixed points in initial mass, EEP and metallicity (and, in some grids, alpha enhancement and binary mass ratio), each with its SED and stellar properties. The default grid_mist_v9 has 613,530 models.

scale factor

The flux ratio between the observed star and its model at 1 kpc: scale = (1 kpc / d)², with d the distance. brutus fits it for each model with the extinction.

StarGrid

A pre-computed model grid optimized for fast fitting. Stores magnitudes at a reference distance of 1 kpc plus reddening coefficients for each model.

BruteForce

The main fitting class for individual stars. Takes a StarGrid and observed photometry, returns posterior samples for distance, extinction, and stellar parameters.

EEPTracks

Class for interpolating stellar parameters along evolutionary tracks. Provides predictions for any combination of mass, EEP, and metallicity.

Isochrone

Class for generating isochrones at specified age and metallicity. Used with StellarPop for cluster modeling.

StellarPop

Class for generating synthetic photometry for stellar populations. Combines Isochrone predictions with bolometric corrections and extinction.

Note

Model Coverage

MIST models in brutus cover: main sequence stars (0.1-300 M☉), subgiants, red giants, horizontal branch, and asymptotic giant branch stars.

Not covered: white dwarfs, brown dwarfs (< 0.08 M☉), neutron stars, black holes, or very young pre-main-sequence stars (< 1 Myr).

See Also