This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| IMRPhenom (phenomenological waveforms) | |
|---|---|
| Name | IMRPhenom (phenomenological waveforms) |
| Caption | Phenomenological waveform family used in gravitational-wave data analysis |
| Developer | Various collaborations and research groups |
| Released | 2000s–present |
| Programming language | C, C++, Python, Fortran |
| Platform | High-performance computing clusters |
| Genre | Gravitational-wave models |
IMRPhenom (phenomenological waveforms) are a family of phenomenological models used to describe the inspiral, merger, and ringdown phases of compact binary coalescences in gravitational-wave astronomy. They synthesize analytical post-Newtonian results, black-hole perturbation theory, and numerical relativity data to produce fast, frequency-domain and time-domain approximants for data analysis by collaborations and institutions. These models have supported detection, parameter estimation, and tests of general relativity in observatories and projects worldwide.
IMRPhenom models emerged from efforts by research groups associated with LIGO Scientific Collaboration, Virgo Collaboration, KAGRA Observatory, Max Planck Society, and university teams to create computationally efficient templates for searches by detectors such as LIGO, Virgo, and KAGRA. Early work involved partnerships with groups at Caltech, MIT, Cardiff University, University of Birmingham, Cornell University, Montana State University, and University of Jena. Phenomenological approaches complemented waveform families from groups including Effective One Body (EOB), Numerical Relativity (NR), and contributors like Frans Pretorius, Mark Scheel, Luis Lehner, Bernd Brügmann, Manuela Campanelli, and Carlos Lousto.
The evolution of phenomenological models traces milestones associated with specific releases and papers by teams tied to Indian Institute of Science Education and Research, University of Wisconsin–Milwaukee, RRI (Raman Research Institute), University of Mississippi, Rutherford Appleton Laboratory, University of Glasgow, Swinburne University of Technology, and Monash University. Notable versions include early frequency-domain models employed in searches around the GW150914 event, later extensions to precessing systems, and multimodal ringdown-inclusive variants used in analyses related to GWTC-1, GWTC-2, and subsequent catalogues compiled by Gravitational Wave Open Science Center. Contributors such as Alessandra Buonanno, Baiotti, Ajith, Hannam, Santamaría, Baker, and Rezzolla influenced successive parameter-space coverage, spin alignment, and mass-ratio ranges.
Phenomenological waveforms integrate analytical building blocks from sources associated with Albert Einstein's theory, drawing on post-Newtonian expansions developed by researchers at Potsdam Institute for Climate Impact Research groups and formalism advanced by teams including Luc Blanchet, Thibault Damour, Alessandra Buonanno, and Gustavo Lousto. Ringdown modeling uses perturbation theory rooted in work by Teukolsky and studies on quasi-normal modes by Kokkotas and Berti. Phenomenological fits parametrize amplitude and phase across frequency bands using calibration targets supplied by numerical relativity simulations from centers like Simulating eXtreme Spacetimes (SXS), RIT (Rochester Institute of Technology), AEI (Albert Einstein Institute), GT (Georgia Tech), and LBIE (Leiden Binaries Initiative). The framework leverages waveform synthesis strategies compatible with parameter-estimation pipelines used by LALSuite, Bilby, PyCBC, and RIFT.
Calibration of IMRPhenom models relies on cross-comparisons with numerical relativity catalogs produced by groups at Caltech/Cornell, SXS Collaboration, RIT, AEI, and Cardiff University. Validation exercises have appeared in collaborative studies involving LIGO Laboratory, Virgo Collaboration, NINJA Project, and Mock Data Challenges run by teams at Max Planck Institute for Gravitational Physics, University of Birmingham, Cardiff University, University of Amsterdam, and Perimeter Institute for Theoretical Physics. Systematic error assessments reference observational campaigns such as the O1, O2, and O3 observing runs and employ statistical frameworks developed in association with George D. Christensen-affiliated groups, Scott A. Hughes collaborators, and Bayesian toolchains created at Imperial College London.
Phenomenological waveforms have been used in detection pipelines that found events like GW150914, GW151226, GW170814, and further entries in GWTC-1 and GWTC-2. They support parameter estimation for sources involving partners at LIGO Scientific Collaboration, Virgo Collaboration, KAGRA, NINJA, and astronomers at Harvard University, Princeton University, University of Chicago, Columbia University, University of Tokyo, and Caltech. Phenomenological templates facilitate rapid scans for transient signals in searches coordinated with electromagnetic partners such as Fermi Gamma-ray Space Telescope, Swift Observatory, Pan-STARRS, Zwicky Transient Facility, and neutrino observatories like IceCube and ANTARES. They also enable tests of strong-field predictions inspired by work at Perimeter Institute, CERN, Institute for Advanced Study, and Kavli Institute for Theoretical Physics.
Limitations stem from reduced accuracy in high-mass-ratio, high-spin, and strongly precessing regimes where numerical relativity coverage is sparse; these challenges are the focus of groups at SXS Collaboration, RIT, AEI, GT, Simons Foundation, NSF, European Research Council, and university consortia. Ongoing improvements include hybridization techniques, surrogate modeling advanced by researchers at Caltech, Cornell University, Flatiron Institute, CWI, and Microsoft Research, and inclusion of higher-order modes and eccentricity addressed by teams at University of Birmingham, Cardiff University, University of Glasgow, and University of Southampton. Cross-validation with EOBNR models and development of reduced-order models involve collaborations with Max Planck Institute for Gravitational Physics, University of Maryland, Northwestern University, and Yale University.
Implementations appear in software ecosystems maintained by projects like LALSuite, PyCBC, Bilby, RIFT, GWINC, and the Gravitational Wave Open Science Center. Code contributions and performance tuning have been carried out by researchers at Caltech, MIT, Cardiff University, University of Birmingham, RIT, AEI, SXS Collaboration, Flatiron Institute, Perimeter Institute, and Max Planck Society. High-performance deployments utilize computing resources at XSEDE, CERN IT, NCSA, Lawrence Livermore National Laboratory, NERSC, and national supercomputing centers affiliated with NSF and PRACE.