Broad Grid of 2-component kilonova models

Abstract for data in kn_sim_cube_v1.tar.gz:

This dataset is a grid of multi-angle, multi-wavelength spectra from 900 2D axisymmetric, 2-component kilonova models.  The masses and velocities of each component are 0.001, 0.003, 0.01, 0.03, or 0.1 M_⊙ and 0.05, 0.15, or 0.3 c, respectively. The models in this dataset have been described by Wollaeger et al (2021), "A Broad Grid of 2D Kilonova Emission Models" (NASA ADS link here).  This dataset is also posted on the LANL CTA website.  The upload is a tarball with 3 files per model: spectra (_spec_), luminosity (_lums_), and broadband magnitudes (_mags_).  The model name has model properties, for example: Run_TS_dyn_all_lanth_wind1_all_md0.03_vd0.3_mw0.003_vw0.05_lums_2020-04-25.dat is a luminosity file produced on 4/25/2020 with toroidal (T) low-Ye ejecta, spherical (S) high-Ye ejecta, "Wind 1" high-Ye composition (wind1), 0.03 M_⊙ and 0.3c low-Ye ejecta mass and velocity, and 0.003 M_⊙ and 0.05c high-Ye ejecta mass and velocity.  The data format is described in data_format.pdf.

Abstract for data in active_learning_sims.tar.gz:

This dataset is also of multi-angle, multi-wavelength spectra from 2D axisymmetric, 2-component kilonova models, but is generated by automated placement (active learning) in mass-velocity of each component, rather than being restricted to the above grid values.  The models and AL placement methods have been described by Ristic et al (2022), "Interpolating Detailed Simulations of Kilonovae: Adaptive Learning and Parameter Inference Applications" (NASA ADS link here).  This dataset is also posted on the LANL CTA website.  The data contained in active_learning_sims.tar.gz is formatted the same way as in kn_sim_cube_v1.tar.gz, following the description in data_format.pdf.

Abstract for data in active_learning_sims_v2.tgz

This dataset is an updated version of active_learning_sims.tar.gz, with additional placed simulations using the active learning procedure described by Ristic et al (2022) (see NASA/ADS link above).  The main difference with respect to the previous active learning set is that this release contains active learning simulations placed for all four morphologies/compositions used in the broad grid.  The models are also described by Kedia et al (2022), "Surrogate light curve models for kilonovae with comprehensive wind ejecta outflows and parameter estimation for AT2017gfo" (NASA ADS link here).

Abstract for surrogate_models_and_jupyter_notebooks.tgz

This tarball provides four surrogate kilonova models and sample jupyter notebooks, developed by Marko Ristic, tuned to reproduce kilonova the simulation models. Each model corresponds to each of the four morphology/composition combinations. Surrogates provide multi-band output versus time and viewing angle. The models are associated with two companion papers discussing methodology of generation and application, Ristic et al (2022) and Kedia et al (2022).

The sample notebooks provided can be used as a starting point to understand the use of these surrogates for quickly generating kilonova light curves for a desired set of input model parameters. The usual runtime of each of these jupyter notebooks is 5 to 10mins and may require installing of 2 or 3 python packages. The surrogates and sample jupyter notebooks can also be accessed at the following GitHub repository: https://github.com/markoris/surrogate_kne