Grid Types
A POSYDON grid is a collection of MESA simulations of stellar or binary systems that share a common set of physical assumptions and differ in their initial conditions. Understanding how grids are organized is key to knowing which data you are using and how you can extend it with your own runs.
What grids are there?
POSYDON ships with different grid types:
HMS-HMSgrids: hydrogen-rich main-sequence primaries with hydrogen-rich main-sequence secondariesHMS-HMS_RLOgrids: hydrogen-rich main-sequence primaries with hydrogen-rich main-sequence secondaries, starting from Roche-lobe overflow (RLO)HMS-CO_RLOgrids: hydrogen-rich main-sequence primaries with a compact object companion (CO), starting from Roche-lobe overflow (RLO)HeMS-COgrids: helium-rich main-sequence primaries with a compact object companion (CO)HeMS-CO_RLOgrids: helium-rich main-sequence primaries with a compact object companion (CO), starting from Roche-lobe overflow (RLO)single_HMSgrids: single hydrogen-rich main-sequence starssingle_HeMSgrids: single helium-rich main-sequence stars
What a grid contains
Each grid contains a set of MESA simulations that are down-sampled to save disk space.
For each run, the grid contains the initial_values, final_values, and history
of the simulation. The history is split between the binary_history and
the histories of the individual stars (history1 and history2).
The grid therefore spans a defined region of this parameter space, sampled on a rectangular (grid) of points. Each run type has its own file, so within a given metallicity you will find one dataset per type.
Fixed grids
A fixed (or rectilinearly sampled) grid is one in which the initial conditions are chosen on a fixed rectangular lattice of values ahead of time. Every point in the lattice is simulated, giving a uniform and predictable coverage of the parameter space. The published data releases (DR1, DR2) are fixed grids. You can run your own fixed grids with the POSYDON MESA grid API — see Run a fixed MESA grid and MESA-grids.
Example organization of downloaded data
A typical data directory is organized by run type and metallicity, e.g.:
POSYDON_data/
HMS-HMS/
1e+00_Zsun.h5
...
HMS-HMS_RLO/
...
single_HMS/
...
where each .h5 file contains one PSyGrid for a given run type and
metallicity.
Summary
See also
Data Releases for the published datasets and their storage costs.
MESA-grids for the tutorials on running your own grids.