Supernova Models
POSYDON supports different supernova models for the formation of compact objects. The supernova model can be set in the configuration file of the population synthesis model.
Overview
The supernova treatment in POSYDON is determined by two key aspects:
Core-collapse mechanism: The physical prescription for computing remnant properties
Computational approach: How and when the collapse outcome is calculated
Core-Collapse Mechanisms
POSYDON supports the following core-collapse mechanisms:
Mechanism |
Engine |
Description |
|---|---|---|
Fryer+12-delayed |
|
Classical rapid and delayed supernova engines from Fryer et al. (2012) |
Fryer+12-rapid |
|
Classical rapid and delayed supernova engines from Fryer et al. (2012) |
Sukhbold+16-engine |
|
Uses pre-computed results from Sukhbold et al. (2016) based on neutrino-powered explosions |
Patton&Sukhbold20-engine |
|
Advanced engine combining Patton & Sukhbold (2020) results for realistic explosion landscapes |
Maltsev+25-engine |
|
Maltsev et al. (2025) with updated explodability criteria and neutrino-driven physics. Only valid M_CO < 10 Msun! |
Couch+20-engine |
|
Simulations of turbulence-aided neutrino-driven core-collapse supernovae from Couch et al. (2020) |
direct |
|
Simplified prescriptions that directly collapse the pre-supernova to the baryonic mass |
direct_he_core |
|
Simplified prescriptions that directly collapse the pre-supernova to the baryonic mass of the helium core |
Pulsational pair-instability supernova prescriptions
POSYDON also supports pulsational pair-instability supernova (PPISN) prescriptions, which can be set in the configuration file:
Computational Approaches
The outcome of core collapse can be computed using three different approaches, controlled by configuration flags:
Pre-Trained Interpolators
use_interp_values = True
use_profiles = True
use_core_masses = True
This is the default and recommended approach for population synthesis. The pre-trained interpolators use outcomes that were computed offline from the detailed MESA stellar profiles. Because the on-the-fly calculations use the down-sampled profiles, the interpolators are generally more accurate and faster.
On-the-Fly Calculations with Downsampled Profiles
use_interp_values = False
use_profiles = True
use_core_masses = False
This approach recalculates the core-collapse outcome during the simulation using stellar profiles. However, unlike the detailed profiles used to train the interpolators, these profiles are downsampled (reduced resolution) for computational efficiency.
Advantages: More flexible, can handle evolutionary scenarios not covered by pre-computed grids, allows for model exploration
Disadvantages: Slower than interpolators, potential accuracy loss due to profile downsampling
When to use: For detailed model comparisons, sensitivity studies, or cases requiring custom physics
Core-Masses Approach (Classical Population Synthesis)
use_interp_values = False
use_profiles = False
use_core_masses = True
This classical approach uses only the core masses at carbon depletion, without requiring detailed stellar profiles.
Pre-Defined Supernova Models
POSYDON provides a set of pre-defined supernova models (SN_MODEL_v2_XX) for which
the initial-final interpolator has been trained.
This requires the user to correctly set the supernova mechanism and engine in the configuration file of the population synthesis model.
Model Overview
The following table lists all pre-defined supernova models and their key characteristics:
Model |
Mechanism |
Engine |
Conserve H-env |
PPI Mass Loss |
|---|---|---|---|---|
SN_MODEL_v2_01 |
Fryer+12-delayed |
- |
No |
-20.0 |
SN_MODEL_v2_02 |
Fryer+12-delayed |
- |
Yes |
-20.0 |
SN_MODEL_v2_03 |
Fryer+12-delayed |
- |
No |
0.0 |
SN_MODEL_v2_04 |
Fryer+12-delayed |
- |
Yes |
0.0 |
SN_MODEL_v2_05 |
Fryer+12-rapid |
- |
No |
-20.0 |
SN_MODEL_v2_06 |
Fryer+12-rapid |
- |
Yes |
-20.0 |
SN_MODEL_v2_07 |
Fryer+12-rapid |
- |
No |
0.0 |
SN_MODEL_v2_08 |
Fryer+12-rapid |
- |
Yes |
0.0 |
SN_MODEL_v2_09 |
Sukhbold+16-engine |
N20 |
No |
-20.0 |
SN_MODEL_v2_10 |
Sukhbold+16-engine |
N20 |
Yes |
-20.0 |
SN_MODEL_v2_11 |
Sukhbold+16-engine |
N20 |
No |
0.0 |
SN_MODEL_v2_12 |
Sukhbold+16-engine |
N20 |
Yes |
0.0 |
SN_MODEL_v2_13 |
Patton&Sukhbold20-engine |
N20 |
No |
-20.0 |
SN_MODEL_v2_14 |
Patton&Sukhbold20-engine |
N20 |
Yes |
-20.0 |
SN_MODEL_v2_15 |
Patton&Sukhbold20-engine |
N20 |
No |
0.0 |
SN_MODEL_v2_16 |
Patton&Sukhbold20-engine |
N20 |
Yes |
0.0 |
SN_MODEL_v2_17 |
Sukhbold+16-engine |
W20 |
No |
-20.0 |
SN_MODEL_v2_18 |
Sukhbold+16-engine |
W20 |
Yes |
-20.0 |
SN_MODEL_v2_19 |
Sukhbold+16-engine |
W20 |
No |
0.0 |
SN_MODEL_v2_20 |
Sukhbold+16-engine |
W20 |
Yes |
0.0 |
SN_MODEL_v2_21 |
Patton&Sukhbold20-engine |
W20 |
No |
-20.0 |
SN_MODEL_v2_22 |
Patton&Sukhbold20-engine |
W20 |
Yes |
-20.0 |
SN_MODEL_v2_23 |
Patton&Sukhbold20-engine |
W20 |
No |
0.0 |
SN_MODEL_v2_24 |
Patton&Sukhbold20-engine |
W20 |
Yes |
0.0 |
SN_MODEL_v2_25 |
Maltsev+25-engine |
M16 |
No |
-20.0 |
SN_MODEL_v2_26 |
Maltsev+25-engine |
M16 |
Yes |
-20.0 |
SN_MODEL_v2_27 |
Maltsev+25-engine |
M16 |
No |
0.0 |
SN_MODEL_v2_28 |
Maltsev+25-engine |
M16 |
Yes |
0.0 |
All pre-defined models use:
use_interp_values = False(on-the-fly calculations, not pre-trained interpolators)use_profiles = True(detailed MESA profiles, downsampled during calculation)use_core_masses = False
Configuration
To use a pre-defined model, specify it in your population synthesis configuration file:
[supernova]
model = SN_MODEL_v2_13
This will automatically load all parameters for that model. Alternatively, you can customize individual parameters:
[supernova]
mechanism = Patton&Sukhbold20-engine
engine = N20
ECSN = Tauris+15
conserve_hydrogen_envelope = False
PPI_extra_mass_loss = -20.0
use_interp_values = True
use_profiles = False
use_core_masses = False