{
"subject": "LIGO/Virgo G184098: Refined localizations from CBC parameter estimation",
"submitter": "Leo Singer at NASA/GSFC ",
"circularId": 18858,
"email": "leo.p.singer@nasa.gov",
"createdOn": 1452716104000,
"body": "The LIGO Scientific Collaboration and Virgo report:\n\nWe have completed a Bayesian parameter estimation analysis of the GW\ncandidate G184098 (GCN 18330) under the assumption that the signal\narises from a compact binary coalescence (CBC) and using the latest\noffline calibration of the GW strain data (GCN 18851). The data is\nstill found to be most consistent with a binary black hole merger\n(GCN 18388).\n\nTwo refined sky maps are now available and can be retrieved from\nGraceDB (https://gracedb.ligo.org/events/G184098):\n\n * bayestar_gstlal_C01.fits.gz, produced by rapid triangulation of\n times, phases, and amplitudes on arrival. It was not available in\n low latency because at the time of the event our online CBC template\n banks were configured for neutron star but not binary black hole\n mergers. This localization is based upon the recovery of the CBC\n candidate from the offline GSTLAL search (GCN 18851), though it is\n practically interchangeable with the BAYESTAR sky map based on the\n PyCBC recovery.\n\n * LALInference_skymap.fits.gz, using Bayesian Markov-chain Monte Carlo\n and nested sampling to perform forward modeling of the full GW\n signal including spin precession and regression of systematic\n calibration errors. We regard the LALInference sky map as the most\n accurate and *authoritative* localization for this event. Though\n additional parameter estimation runs with different waveform\n approximants are ongoing, we expect that they will agree closely\n with this one.\n\nBoth of the sky maps above agree with the initial LALinference Burst\n(LIB) localization (GCN 18330, LIB_skymap.fits.gz) on favoring the\nsouthern portion of the annulus determined by an arrival time\ndifference between LIGO Hanford and LIGO Livingston of about 7 ms.\n\nThe table below presents a quantitative comparison of the available\nlocalizations along the lines of Sec. 4.5 of Essick et al. (2015,\nhttp://adsabs.harvard.edu/abs/2015ApJ...800...81E). The first column\ngives the area in deg2 of the 90% credible region, and the second\ncolumn gives the area in deg2 of the overlap with the LALInference 90%\ncredible region.\n\n Area Overlap Algorithm (filename) \n-------------------------------------------------------------\n 310 230 cWB (skyprobcc_cWB_complete.fits)\n 750 270 LIB (LIB_skymap.fits.gz)\n 400 350 BAYESTAR (bayestar_gstlal_C01.fits.gz)\n 600 - LALInference (LALInference_skymap.fits.gz)"
}