CoRe database of binary neutron star merger waveforms and its application in waveform development

CoRe database of binary neutron star merger waveforms and its application in waveform development

Tim Dietrich, David Radice, Sebastiano Bernuzzi, Francesco Zappa, Albino Perego, Bernd Brügmann, Swami Vivekanandji Chaurasia, Reetika Dudi, Wolfgang Tichy, Maximiliano Ujevic Nikhef, Science Park, 1098 XG Amsterdam, The Netherlands Max Planck Institute for Gravitational Physics (Albert Einstein Institute), Am Mühlenberg 1, Potsdam 14476, Germany Institute for Advanced Study, 1 Einstein Drive, Princeton, NJ 08540, USA Department of Astrophysical Sciences, Princeton University, 4 Ivy Lane, Princeton, NJ 08544, USA Theoretical Physics Institute, University of Jena, 07743 Jena, Germany Istituto Nazionale di Fisica Nucleare, Sezione Milano Bicocca, gruppo collegato di Parma, I-43124 Parma, Italy Dipartimento di Scienze Matematiche Fisiche ed Informatiche, Universitá di Parma, I-43124 Parma, Italia Dipartimento di Fisica, Università degli Studi di Milano Bicocca, Piazza della Scienza 3, 20126 Milano, Italia Department of Physics, Florida Atlantic University, Boca Raton, FL 33431 USA Centro de Ciências Naturais e Humanas, Universidade Federal do ABC,09210-170, Santo André, São Paulo, Brazil
July 13, 2019

CoRe database of binary neutron star merger waveforms

Tim Dietrich, David Radice, Sebastiano Bernuzzi, Francesco Zappa, Albino Perego, Bernd Brügmann, Swami Vivekanandji Chaurasia, Reetika Dudi, Wolfgang Tichy, Maximiliano Ujevic Nikhef, Science Park, 1098 XG Amsterdam, The Netherlands Max Planck Institute for Gravitational Physics (Albert Einstein Institute), Am Mühlenberg 1, Potsdam 14476, Germany Institute for Advanced Study, 1 Einstein Drive, Princeton, NJ 08540, USA Department of Astrophysical Sciences, Princeton University, 4 Ivy Lane, Princeton, NJ 08544, USA Theoretical Physics Institute, University of Jena, 07743 Jena, Germany Istituto Nazionale di Fisica Nucleare, Sezione Milano Bicocca, gruppo collegato di Parma, I-43124 Parma, Italy Dipartimento di Scienze Matematiche Fisiche ed Informatiche, Universitá di Parma, I-43124 Parma, Italia Dipartimento di Fisica, Università degli Studi di Milano Bicocca, Piazza della Scienza 3, 20126 Milano, Italia Department of Physics, Florida Atlantic University, Boca Raton, FL 33431 USA Centro de Ciências Naturais e Humanas, Universidade Federal do ABC,09210-170, Santo André, São Paulo, Brazil
July 13, 2019

We present the Computational Relativity (CoRe) collaboration’s public database of gravitational waveforms from binary neutron star mergers. The database currently contains 367 waveforms from numerical simulations that are consistent with general relativity and that employ constraint satisfying initial data in hydrodynamical equilibrium. It spans 164 physically distinct configuration with different binary parameters (total binary mass, mass-ratio, initial separation, eccentricity, and stars’ spins) and simulated physics. Waveforms computed at multiple grid resolutions and extraction radii are provided for controlling numerical uncertainties. We also release an exemplary set of 18 hybrid waveforms constructed with a state-of-art effective-one-body model spanning the frequency band of advanced gravitational-wave detectors. We outline present and future applications of the database to gravitational-wave astronomy.

Figure 1: Simulations contained in the CoRe database. We present the total mass , the mass ratio , the individual dimensionless spins , the eccentricity [no eccentricity measurement is given for too short simulations], the individual quadrupolar tidal parameters , the number of orbits [note that for highly eccentric orbits close to head-on, the number of orbits can drop below ], and the employed resolution of the finest level covering the entire NS for different configurations. Different colored markers refer to different EOS, see top color bar. In the last panel we also include simulations with different grid resolutions and numerical methods (fluxes, mesh refinement strategies etc.); simulations of a fixed configuration performed at the same resolutions but using different methods are marked with vertical bars in this panel.

The era of gravitational-wave (GW) astronomy has been inaugurated with the direct detection of GWs from binary black hole (BBH) mergers Abbott et al. (2016a, b, 2017a, 2017b, 2017c) soon followed by the breakthrough observation of GWs and electromagnetic (EM) signals from a binary neutron star (BNS) collision Abbott et al. (2017d, e); GBM (2017); Coulter et al. (2017).

Numerical relativity (NR) is the fundamental tool to study GWs from systems in the strong-field regime, and it has crucially supported the first discoveries. In particular, different NR groups have publicly released BBH simulation data Mroue et al. (2013); Healy et al. (2017); Jani et al. (2016). These catalogs have been the cornerstone of many scientific results. They have been used to improve our understanding of the merger dynamics Lousto and Zlochower (2014); Le Tiec et al. (2013); Nagar et al. (2015); Varma et al. (2014); Keitel et al. (2017); Gerosa and Moore (2016), to develop waveform models Pan et al. (2014); Damour and Nagar (2014); Taracchini et al. (2014); Husa et al. (2016); Pürrer (2016); Babak et al. (2017); Bohé et al. (2017); Huerta et al. (2017) including surrogates Field et al. (2014); Blackman et al. (2015), and to validate LIGO-Virgo parameter estimation pipelines Abbott et al. (2016c, 2017f).

Following the first successful BNS merger simulations in full general relativity Shibata and Uryu (2000); Shibata (1999), the NR community has made tremendous progresses on several aspects of the problem: (i) the exploration of the effect of different equations of state (EOSs), total mass and mass-ratio on the merger dynamics Hotokezaka et al. (2011); Bernuzzi et al. (2014a); Foucart et al. (2016a); Bernuzzi et al. (2016); Dietrich et al. (2017a); Lehner et al. (2016); Sekiguchi et al. (2016); (ii) the development of many-orbits simulations for high-precision GW modeling Bernuzzi et al. (2012a); Hotokezaka et al. (2013a); Radice et al. (2014a); Bernuzzi and Dietrich (2016); Hotokezaka et al. (2015); Dietrich et al. (2017b); Kiuchi et al. (2017a); (iii) the exploration of BNS mergers from eccentric orbits and dynamical collisions Gold et al. (2012); East and Pretorius (2012); East et al. (2016a, b); Radice et al. (2016a); (iv) the inclusion of aligned spins and spin-precession effects Bernuzzi et al. (2014b); Kastaun et al. (2013); Dietrich et al. (2017c); East et al. (2016a); Kastaun et al. (2017); Dietrich et al. (2015a); Tacik et al. (2015); Dietrich et al. (2018a); (v) the simulations of magnetic effects in connection to gamma-ray bursts engines Anderson et al. (2008); Giacomazzo et al. (2011); Rezzolla et al. (2011); Palenzuela et al. (2013); Kiuchi et al. (2014); Palenzuela et al. (2015); Ruiz et al. (2018); Kiuchi et al. (2017b); (vi) the study of finite-temperature and composition effects using a microphysical descriptions of NS matter together with neutrino transport Sekiguchi et al. (2011); Galeazzi et al. (2013); Neilsen et al. (2014); Sekiguchi et al. (2015); Palenzuela et al. (2015); Foucart et al. (2016b); Foucart (2017); (vii) the study of mass ejecta and EM counterparts Hotokezaka et al. (2013b); Bauswein et al. (2013a); Wanajo et al. (2014); Lehner et al. (2016); Radice et al. (2016a); Dietrich et al. (2017a, c); Foucart et al. (2016b); Fujibayashi et al. (2017a); Perego et al. (2017); Hotokezaka et al. (2018); Bovard et al. (2017). New frontiers in BNS merger simulations are the inclusion of general-relativistic radiation hydrodynamics Shibata and Sekiguchi (2012); Foucart et al. (2015); Foucart (2017) and viscous hydrodynamics effects Radice (2017); Shibata et al. (2017a); Shibata and Kiuchi (2017); Shibata et al. (2017b); Fujibayashi et al. (2017b).

Since 2009 our team has contributed to some of the research lines mentioned above. Here, we present the largest-to-date public database of BNS waveforms composed of new simulations and those published in Bernuzzi et al. (2014b, a, 2015a); Dietrich et al. (2015b); Bernuzzi et al. (2015b); Dietrich et al. (2015a); Radice et al. (2016b, 2017); Bernuzzi and Dietrich (2016); Dietrich et al. (2017a, c); Radice (2017); Radice et al. (2018a); Dietrich and Hinderer (2017); Dietrich et al. (2017b, 2018a); Zappa et al. (2018); Dietrich et al. (2018b). The combined set of simulations required about million CPU-hours on supercomputers in Europe and the United States. We publicly release these data with the goal of supporting researchers and further developments in the field of GW astronomy ( We plan to extend the database with waveforms from upcoming simulations and from other groups/codes.

This article describes the simulation methods in NR, summarizes the quality of the computed waveforms and the key parameters that characterize the GW, and concludes outlining some of the many applications. We use geometrized units and express results in terms of solar masses ( g) if not otherwise stated. Conversion factors to CGS are  cm and  s.

Simulation Methods

Initial data.

Initial data are constructed by solving the Einstein constraint equations in the conformal thin sandwich formalism and by imposing hydrodynamical equilibrium for the star fluid Wilson and Mathews (1995); Wilson et al. (1996); York (1999). The fluid’s flow is chosen to be either irrotational Bonazzola et al. (1999), or prescribed according to the constant rotational velocity formalism Tichy (2011, 2012). Binaries in quasi-circular orbits are built imposing a helical Killing vector Gourgoulhon et al. (2001), whereas for eccentric orbits an approximate “helliptical” Killing vector is used Moldenhauer et al. (2014); Dietrich et al. (2015a). We use either the public Lorene Eric Gourgoulhon, Philippe Grandclément, Jean-Alain Marck, Jérôme Novak and Keisuke Taniguchi () or the SGRID Tichy (2009, 2012); Dietrich et al. (2015a) code. Both codes use multi-domain pseudo-spectral methods with surface fitting coordinates Gourgoulhon et al. (2001); Ansorg (2007).


Dynamical simulations are performed using free-evolution schemes for the Einstein equations and general relativistic hydrodynamics (GRHD). For the spacetime, we employ either the BSSNOK Nakamura et al. (1987); Shibata and Nakamura (1995); Baumgarte and Shapiro (1999) formalism or the Z4c formalism Bernuzzi and Hilditch (2010); Ruiz et al. (2011); Weyhausen et al. (2012); Cao and Hilditch (2012); Hilditch et al. (2013). The latter has improved constraint propagation and damping properties with respect to BSSNOK, especially in matter simulations Bernuzzi and Hilditch (2010); Hilditch et al. (2013). We use the moving puncture gauge Bona et al. (1996); Alcubierre et al. (2003); van Meter et al. (2006); Campanelli et al. (2006a); Baker et al. (2006), which can handle automatically the gravitational collapse without the need for excision Baiotti et al. (2007); Thierfelder et al. (2011a); Dietrich and Bernuzzi (2015). GRHD is solved in flux-conservative form Banyuls et al. (1997). Some mergers are simulated with microphysical EOS and neutrino cooling is taken into account with a leakage scheme. Viscous effects in GR are also simulated in a few cases using the large eddy scheme (GRLES) developed in Radice (2017).

We use two different NR codes: BAM Brügmann et al. (2004, 2008); Thierfelder et al. (2011b) and THC Radice and Rezzolla (2012); Radice et al. (2014a, b, 2015). Both codes use a simple mesh refinement scheme whereby the grid hierarchy is composed of nested Cartesian boxes, some of which can be moved to track the orbital motion of the stars Berger and Oliger (1984); Schnetter et al. (2004); Brügmann et al. (2008). The grid setup is controlled by the resolution in the finest levels. The finest refinement levels cover entirely the NSs during the inspiral. The other levels are constructed by progressively coarsening the resolution by factors of two and extend to the wave-extraction zone. Discretization is based on fourth (or higher) order finite-differencing stencils and GRHD is handled with either standard finite-volume or high-order finite-differencing high-resolution shock-capturing methods Radice et al. (2014a); Bernuzzi and Dietrich (2016). The THC code also implements a neutrino leakage scheme and the GRLES Radice et al. (2016a); Radice (2017).

Wave extraction.

GWs are extracted on coordinate spheres with radius using the spin-weighted  spherical harmonics decomposition of the Weyl scalar , e.g., Brügmann et al. (2008). Some of the THC simulations employ the Cauchy characteristic extraction technique to obtain at future null infinity Reisswig et al. (2009). The metric multipoles are reconstructed using the fixed frequency method (Reisswig and Pollney, 2011). We release the metric multipole as a function of the coordinate time and of the retarded time , where is the tortoise coordinate defined by assuming is the isotropic radius and using the binary total mass for the Schwarzschild spacetime. We also release the GW energy and angular momentum emitted during the simulation that are computed as in Damour et al. (2012a); Bernuzzi et al. (2012b). Our waveforms are extracted at different extraction radii and can be further extrapolated to obtain null-infinity estimates, e.g. Scheel et al. (2009); Reisswig et al. (2010); Bernuzzi et al. (2012a).

Input physics

BNS simulations require several assumptions on the NS fluid and input models describing the matter interactions. The yet unknown EOS is among the most important quantities determining the NS properties and the binary dynamics. It determines the tidal deformations and interactions of the stars during the inspiral Damour (1983); Hinderer (2008); Damour and Nagar (2009a); Hinderer et al. (2010); Damour et al. (2012b), the lifetime and rotation frequency of the merger remnant Bauswein and Janka (2012); Hotokezaka et al. (2013c); Takami et al. (2014); Bernuzzi et al. (2016); Radice et al. (2017), and the amount of unbound matter ejected during the merger process, e.g. Hotokezaka et al. (2013b); Bauswein et al. (2013a); Dietrich et al. (2015b); Lehner et al. (2016). We release data for 16 different EOSs. Two EOSs are polytropic models with adiabatic index Bernuzzi et al. (2014b). Nine EOSs are zero-temperature nuclear physics model represented by piecewise polytropic fits Read et al. (2009a). They are augmented with a -law pressure component during the simulation to approximate temperature effects Shibata et al. (2005). Five EOSs are tabulated finite-temperature microphysical models developed in Lattimer and Swesty (1991); Typel et al. (2010); Hempel and Schaffner-Bielich (2010); Steiner et al. (2013a); Banik et al. (2014), which we also release. Finite-temperature effects are crucial during and after merger, when compressional and shock heating are present, e.g. Bauswein et al. (2010); Sekiguchi et al. (2011); Kaplan et al. (2014); Sekiguchi et al. (2015); Lehner et al. (2016).

The role of magnetic fields on the post-merger dynamics is currently a key open question Giacomazzo et al. (2011); Kawamura et al. (2016); Kiuchi et al. (2017b); Radice (2017); Shibata and Kiuchi (2017); Fujibayashi et al. (2017b); Radice et al. (2018b). Large magnetic field instabilities might cause turbulence and induce viscosity, potentially affecting the merger remnant, mass outflows and the GW emission. Also, while not directly relevant for GW emission on the dynamical timescale of our simulations, neutrino transport plays a crucial role in the merger remnant, e.g. Sekiguchi et al. (2011); Galeazzi et al. (2013); Sekiguchi et al. (2015); Foucart et al. (2016a); Palenzuela et al. (2015); Lehner et al. (2016); Radice et al. (2016a); Bovard et al. (2017); Perego et al. (2017). We plan to include more data from simulations with advanced radiation transport schemes and magnetic field effects as robust NR results become available.

Waveform parameters

Figure 2: Waveforms from the database showing, from top to bottom panel, the influence of total mass, mass ratio, spins, eccentricity, and EOS.

Figure 1 summarizes our database in terms of the main parameters that characterize the GWs.

Binary mass.

In contrast to BBHs, BNS dynamics cannot be rescaled by the binary total mass ( label the NSs)


since enters the description of tidal interactions during the inspiral and determines the merger remnant.

Formation scenarios and the constraints from GW170817 indicate that NS masses lie within  Lattimer (2012); Ozel and Freire (2016); Rezzolla et al. (2018); Shibata et al. (2017b); Ruiz et al. (2018); Margalit and Metzger (2017). Current observations range from  Rawls et al. (2011); Ozel et al. (2012) to  Demorest et al. (2010); Antoniadis et al. (2013) or possibly even  Linares et al. (2018), with BNS masses varying in  Lattimer (); Huang et al. (2018). The wide mass range in our database fully covers the observational and a large fraction of the theoretical limits.

GWs from a merger are shown in the top panel of Fig. 2. High-mass mergers likely result in a prompt BH formation; while high-mass BNS emit strong GWs, they are EM faint due to smaller ejecta and disk masses Bauswein et al. (2013a); Hotokezaka et al. (2013c).

Mass ratio.

The mass ratio


has a clear imprint on the GW/EM signals: BNS with larger are less luminous in GWs Hinderer et al. (2010); Dietrich et al. (2017c); Zappa et al. (2018), but their larger mass ejecta can power bright EM transients Dietrich and Ujevic (2017); Dietrich et al. (2017c); Lehner et al. (2016). The isolated NS mass distribution implies mass ratios up to , but population synthesis models predicts lower values  Dominik et al. (2012); Dietrich et al. (2015a). The largest observed mass ratio in BNSs is  Martinez et al. (2015); Lazarus et al. (2016). The CoRe database contains data with mass ratios up to , which is the largest simulated so far Dietrich et al. (2015a, 2017a). In this simulation the companion NS is tidally disrupted during the merger leading to postmerger GWs with small amplitude (second panel in Fig. 2).


The dimensionless spin of a NS in a binary can be defined as


where the angular momentum is computed from the isolated NS with the same EOS, rotational velocity, and baryonic mass as the constituents of the binary Bernuzzi et al. (2014b); Dietrich et al. (2015a, 2017c). The maximum NS spin is not precisely known, since it depends on the EOS, but existing EOS models predict breakup spins below , corresponding to spin periods of less than 1 ms Lo and Lin (2011). The fastest spinning NS in a BNS system is PSR J1946+2052 Stovall et al. (2018) which will have at merger.

For spins parallel to orbital angular momentum (say -direction), the effective spin Buonanno et al. (2003)


is the quantity determining the leading-order spin-orbit effects on the phase evolution of the binary. Spin-orbit interactions quantitatively change the inspiral-merger and remnant dynamics  Damour (2001); Campanelli et al. (2006b); Bernuzzi et al. (2014b). Neglecting their effect can bias the GW parameter estimation Harry and Hinderer (2018); Agathos et al. (2015); Dietrich et al. (2018c). Spin-precession effects in BNS have been first simulated in Refs. Dietrich et al. (2015a, 2018a); the computed GW signal is shown in Fig. 2 (third panel).


The emission of GWs causes field binaries to circularize to eccentricities by the time they enter the LIGO-Virgo band Peters (1964). Therefore, for an accurate modeling of the GWs, it is important to simulate small eccentricity binaries. The residual (numerical) eccentricity of the initial data can be reduced to using an iterative procedure Pfeiffer et al. (2007); Kyutoku et al. (2014); Dietrich et al. (2015a).

On the contrary, dynamically assembled BNSs or those belonging to hierarchical triplets could be highly eccentric even at the time of merger Bonetti et al. (2018); Lee et al. (2010). An example of a highly eccentric merger with is shown in Fig. 2 (forth panel). The bursts in the GW amplitude are caused by the close encounters of the two stars. These encounters also induce -mode oscillations which allow an independent constraint of the EOS for upcoming 3rd generation GW detectors Gold et al. (2012); Chirenti et al. (2017); Chaurasia et al. ().

Tidal parameters.

Tidal interactions in the post-Newtonian (PN) formalism are described by a multipolar set of parameters proportional to the relativistic Love numbers Damour (1983); Hinderer (2008); Damour and Nagar (2009a); Binnington and Poisson (2009). The dominant effect depends on the gravitoelectric quadrupolar Love numbers and the compactness of the NS through the expression . Tidal interactions are attractive and enter at leading PN order in the GW phasing evolution through the combination Flanagan and Hinderer (2008); Damour and Nagar (2009b); Damour et al. (2012b); Favata (2014),


The tidal parameter is a key quantity to characterize the non-perturbative regime of the merger dynamics as shown in Read et al. (2013); Bernuzzi et al. (2014a); Zappa et al. (2018) and discussed below. Furthermore, it provides a simple but effective parameterization of the characteristic GW post-merger frequencies Bernuzzi et al. (2015b); Rezzolla and Takami (2016); Lehner et al. (2016) and of the disk mass Radice et al. (2018a).

The value of for GW170817 is constrained to be on the basis of the analysis of the GW signal alone Abbott et al. (2017d, 2018a). In addition, Refs. Radice et al. (2018a); Coughlin et al. (2018) suggested that the observation of an EM counterpart to GW170817 allows to place a lower bound on the tidal deformability of , Ref. Radice et al. (2018a), or , Ref. Coughlin et al. (2018). Further constraints arise from the theoretical modeling of matter near nuclear density, e.g. Schaffner-Bielich (2008), other astrophysical observations, e.g. Ozel and Psaltis (2009); Neuhauser et al. (2012); Steiner et al. (2013b); Ozel and Freire (2016); Watts et al. (2016), and from the combination of all these constraints by considering a large set of possible nuclear physics EOSs, e.g. Annala et al. (2018); Most et al. (2018).

Data quality

Waveforms’ error budgets based on convergence tests and finite radius extraction have been presented in Bernuzzi et al. (2012a); Radice et al. (2014a, b, 2015); Bernuzzi and Dietrich (2016); Dietrich et al. (2017b, 2018b). Phase convergence is typically observed for about orbits at sufficiently high resolutions, corresponding to about grid points per NS diameter. The error due to finite-radius extraction dominates in the early part of the simulations, but truncation errors increase towards merger and afterwards where the uncertainty is the largest Bernuzzi et al. (2012a). Typical accumulated phase errors up to merger are estimated as  rad for simulations in which convergence can be proven. We stress, however, that the lowest resolutions employed in our runs are not convergent and do not give quantitatively reliable results for multiple orbits. Post-merger GWs are typically less accurate, but monotonic behavior with grid resolution can be observed at sufficiently high resolutions, e.g. Radice et al. (2017). Our post-merger data are sufficiently robust to infer the energy and frequency content, e.g. Bernuzzi et al. (2015b); Radice et al. (2017); Zappa et al. (2018).

We assessed systematic errors due to the use of nonlinear numerical schemes used for GRHD Bernuzzi et al. (2012b); Radice et al. (2014b) with extensive testing of different algorithms and/or extensive code comparisons. We have tested consistency between BAM and THC for datasets: BAM:0097 and THC:0036, BAM:0063 and THC:0029, BAM:0064 and THC:0028, using exactly the same initial data. We found that phase differences are below the estimated uncertainties. A simple polytropic EOS setup has also been compared to results obtained with the SpEC code Scheel et al. (2006); Szilagyi et al. (2009); Buchman et al. (2012) with similar results Haas et al. (2016).

We stress that all of our waveforms are computed using constraint-satisfying initial data in hydrostatic equilibrium. Constraint violating and/or non-hydrostatic initial data exhibits large unphysical fluid oscillations that contaminate the GW signals. These oscillations are significantly reduced and converge to zero if equilibrium is imposed Dietrich et al. (2018b). Systematic errors generated by the initial data were studied by comparing the evolution of a binary produced by SGRID and Lorene using the same evolution setup (BAM:0026, BAM:0027Bernuzzi et al. (2014b). Differences in the GW phase and collapse time to black-hole were found to be compatible with those expected from finite grid resolutions effects.


The CoRe waveform database has wide applicability to the study of strong-field BNS dynamics and for GW astronomy.

Our simulations showed that, despite the complexity of the physics involved, the main quantities characterizing the merger dynamics, like the mass-rescaled GW frequency and the binding energy per unit mass, are determined by parameters like , emerging from perturbative (PN and effective-one-body, EOB) analysis Bernuzzi et al. (2014a, 2015b). About 100 simulations of the CoRe database were used to compute the total GW luminosity in terms of tidal parameters and the mass-ratio for all BNS with aligned spins , and to set upper limits to the total emitted energy Zappa et al. (2018).

A related application is the study of the merger outcome. NR data are crucial to understand the formation of massive NS remnant Giacomazzo and Perna (2013); Foucart et al. (2016a); Radice et al. (2018b) and prompt black hole formation at merger Kiuchi et al. (2010); Bauswein et al. (2013b); Hotokezaka et al. (2013c).

The data we provide can be used to verify and develop inspiral-merger waveform models for LIGO-Virgo analysis. BAM simulations have been already used in the development of the TEOBResum model Bernuzzi et al. (2015a). Further analytical-numerical comparisons showed that state-of-art tidal EOB models might underestimate tidal effects at merger for stiff EOS and small Dietrich and Hinderer (2017). Our spinning BNS are currently used to test the performances of the TEOBResumS model Damour and Nagar (2014); Nagar et al. ().

High-resolution BAM simulations (BAM:0037, BAM:0064, BAM:0095) were employed to construct the tidal phase model NRtidal Dietrich et al. (2017b, 2018b, 2018c). The latter is a closed-form expression fitting the inspiral-merger GW composed of PN, TEOBResum, and NR data used to augment any BBH waveform model with tidal effects Dietrich et al. (2018c). Notably, NRtidal was used in the LIGO-Virgo analysis of GW170817 Abbott et al. (2017d); Chatziioannou et al. (2018); Abbott et al. (2018a, b), and other groups are using similar approaches for GW modeling Kawaguchi et al. (2018).

A main open challenge is the modeling of GWs from merger remnants Hotokezaka et al. (2013c); Bauswein et al. (2012, 2014); Takami et al. (2014, 2015); Bauswein and Stergioulas (2015); Clark et al. (2016); Rezzolla and Takami (2016); Bose et al. (2017); Chatziioannou et al. (2017). Several features of the signal are understood, but quantitative models are missing. We anticipate that CoRe data will be used to develop new post-merger models to be employed for the analysis of current and third-generation detectors. The latter are the most promising observatories to capture high-frequency GW signals, e.g. Clark et al. (2014); Radice et al. (2016b, 2017).

Our data can also be injected in synthetic detector noise to test parameter estimation pipelines, similarly to what was done for BBHs Abbott et al. (2017f). For BNSs, however, complete waveforms spanning thousands of GW cycles during the inspiral and tens of GW post-merger cycles would be needed. To address the problem, we generate hybrid waveforms combining analytical models and NR data and covering the frequency range of ground-based interferometers Dudi et al. () (see also Read et al. (2009b, 2013); Hotokezaka et al. (2016)). We release 18 of these hybrids corresponding to equal, unequal masses and spinning BNSs.

The CoRe database will have a reach beyond the applications we have just discussed. In the future, we plan to include more quantities from our simulations. For example, mass outflows ejected during merger, e.g. Hotokezaka et al. (2013b); Bauswein et al. (2013a); Sekiguchi et al. (2015); Dietrich et al. (2015b); Radice et al. (2016a); Dietrich and Ujevic (2017); Lehner et al. (2016); Fujibayashi et al. (2017b); Bovard et al. (2017), and disk masses and profiles Radice et al. (2018a). These data will be crucial for the interpretation of EM counterparts.


We thank A. Nagar and S. Ossokine for discussions and A. Sternbeck at TPI Jena for technical help.

T.D. acknowledges support by the European Union’s Horizon 2020 research and innovation program under grant agreement No 749145, BNSmergers. D.R. acknowledges support from a Frank and Peggy Taplin Membership at the Institute for Advanced Study and the Max-Planck/Princeton Center (MPPC) for Plasma Physics (NSF PHY-1523261). S.B. acknowledges support by the European Union’s H2020 under ERC Starting Grant, grant agreement no. BinGraSp-714626. A.P. acknowledges support from the INFN initiative “High Performance data Network” funded by CIPE. R.D. and B.B. were supported by DFG grant BR 2176/5-1. S.V.C. and R.D. were supported by the DFG Research Training Group 1523/2 ”Quantum and Gravitational Fields”. W.T. was supported by the National Science Foundation under grants PHY-1305387 and PHY-1707227.

Computations were performed on the supercomputer SuperMUC at the LRZ (Munich) under the project number pr48pu, Jureca (Jülich) under the project number HPO21, on the supercomputers Bridges, Comet, and Stampede (NSF XSEDE allocation TG-PHY160025), on NSF/NCSA Blue Waters (NSF PRAC ACI-1440083), on Marconi (PRACE proposal 2016153522 and ISCRA-B project number HP10B2PL6K), on the Hydra and Draco clusters of the Max Planck Computing and Data Facility, the compute cluster Minerva of the Max-Planck Institute for Gravitational Physics, and on the ARA cluster of the University of Jena.


Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description