FUKA: Frankfurt Uni./Kadath Initial Data Solvers

More info:  external link

This week marks the public release of the Frankfurt University/Kadath initial data solvers, FUKA (https://kadath.obspm.fr/fuka/). The details of the analytics and the numerical implementation are included in our recent submission on arxiv: https://arxiv.org/abs/2103.09911

To summarize, our collection of solvers provide spectral solutions to the eXtended Conformal Thin Sandwich formulation of the Einstein field equations and are able explore a vast range of the parameter space available when assuming conformal flatness including eccentricity reduced initial data of binaries composed of highly compact, asymmetric, and highly spinning objects.

In the case of BHs, excision conditions are imposed to generate isolated BHs, binary black holes, and black hole neutron star binary ID. Specifically related to binary configurations, FUKA can reliably generate binaries with varying mass ratio as well as mixed spins with the restriction that the spin axis is parallel with the orbital axis, i.e. (anti-)aligned.

Additionally, FUKA includes an isolated TOV solver as well as a binary neutron star solver. In the case of a BNS configuration, we are able to explore the very far reaches of the parameter space with regards to mixed spins near the mass shedding limit and mass ratio (q < 0.5), again, with the restriction of the spin axis being parallel with the orbital axis. All solvers involving Neutron stars are compatible with cold equations of state in LORENE format as well as piecewise polytropic equations of state.

Finally, FUKA includes the capabilities to enable eccentricity reduction either by using Post Newtonian estimates or through iterative reduction as detailed in our work for all the binaries configurations mentioned.

A few final remarks:
– Each initial data type just described has an associated exporter that can be used easily by any evolution code to gain access to the relevant quantities. To verify as such, each ID type has been successfully evolved using the Einstein Toolkit with a minimal interface thorn to intermediate between the relevant exporter and the toolkit.
– These codes are hosted on the Kadath website and git repository as a feature branch.
The website, https://kadath.obspm.fr/fuka/, includes a link to the git repository as well as installation instructions to compile the Kadath library and the initial data solvers.
– The Doxygen documentation for this feature branch can be compiled locally once the git repository has been cloned. This includes a friendly interface for viewing documentation related to the base Kadath as well as the additional features not included in the main Kadath branch.
– Low resolution initial data can be easily obtained with small scale systems (~128 cores) within a couple of hours in the case of binary ID.
– Development will continue with a focus on optimization and user experience based on community feedback

Copyright (c) 2021 Samuel Tootle (tootle[AT]itp.uni-frankfurt.de). Licensed under GPL3 license.