GRoupEd sTar formAtion: converting sink particles to stars that are sampled from observational initial mass function (IMF).
Paper: Liow K. Y., Rieder S., Dobbs C. L., Jaffa S. E., 2022, MNRAS, 510, 2657 (ADS link of the paper here).
Grouped star formation is developed to form stars dynamically in parsec-scale simulations, such that that the sink particles are first grouped together, then the total group masses are used to sample the IMF and form stars. The current version is not designed to be used to generate initial conditions.
greta
is built upon the AMUSE framework.
First, install AMUSE
:
pip install amuse-framework
pip install amuse
Then, install my version of AMUSE MASC
(c.f. Steven Rieder's MASC) and create a symbolic link:
git clone https://github.com/kyliow/MASC.git masc
cd {amuse directory}/src/amuse/ext
ln -s ../../../../masc/src/amuse/ext/masc
cd ../../../../
Lastly, install greta
:
git clone https://github.com/kyliow/greta.git greta
The command line arguments that can be parsed to greta.py
are:
-i
: Input AMUSE sink particle set file name.-o
: Output AMUSE star particle set file name. Default isstars.amuse
.-d
: Grouping distance parameter in pc. Default is 0.-v
: Grouping speed parameter in km/s. Default is 0.-t
: Grouping age parameter in Myr. Default is 0.--lower-limit
: Lower limit of stellar mass in MSun. Default is 0.5.--upper-limit
: Upper limit of stellar mass in MSun. Default is 100.-r
: Random seed. Default isNone
.
To 'turn off' any of the three grouping parameters, simply set them to unrealistically high values. See Liow et al. (2022) for more information.
An example command to run greta.py
with grouping parameters 1 pc, 3 km/s and 1 Myr:
python greta.py -i sinkfilename.amuse -o output.amuse -d 1 -v 3 -t 1
Acronym was generated using ACRONYM
(Cook 2019).