LIGO-VIRGO-KAGRA

Product: Gravitational Waves Transients Catalog (GWTC) Skymaps Search

This product looks for gravitational waves events around a given set of (ra,dec) localization coordinates. Optionally one can identify the galaxies present in the density probability area of the matching GW events. The catalogs come from the LIGO, Virgo, and KAGRA (LVK) Collaborations. They consist in GWTC-4 catalog and previous ones (GWTC-1, GWTC-2, GWTC-2.1, GWTC-3 ).

The events are binary mergers between black holes and neutron stars. Both black holes and neutron stars are classified as compact objects. The LVK detectors are sensitive to three different types of compact binary mergers:

  • Binary black hole (BBH) merger
  • Binary neutron star (BNS) merger
  • Neutron star - black hole (NSBH) merger

All the data used in this product can be found at Gravitational Wave Open Science Centre (GWOSC)

Input parameters (MMODA)

Here we define the input parameters for the product:

  • RA (deg): the input right ascension (on top common panel)
  • Dec (deg) : the input declination (on top common panel)
  • rastd : the ra standard deviation (used for galaxy and grb catalogs search)
  • decstd : the dec standard deviation (used for galaxy and grb catalogs search)
  • prob : the probability density envelope [0,1] (used for gw source search)
  • waveform : Waveform approximant used for generating skymap [Mixed;IMRPhenomXPHM;SEOBNRv4PHM]
  • use_glade : Use GLADE+ Catalog
  • use_grb : Use GRBweb Catalog

Searching the GW Catalogs skymaps matching ra, dec and prob

In this product, we use the S3 repo hosted by MMODA platform.

We loop over the catalogs skymap fits files to find any event that matches the set of inputs (ra,dec,rastd,decstd,prob). If we find one matching event, we plot the skymap with a zoom on the region of interest and give the mean luminosity distance from the event.

If use_glade is true, we search for crossmatch between the requested region and the event probability density where galaxies are identified from GLADE+ and give a table of the results.

If use_grb is true, we search for crossmatch between the requested region and the event probability density where GRB events are identified from GRBweb and give a table of the results.

Outputs (MMODA)

Here we define the outputs for the module:

  • skymap : the matching skymap
  • galaxy_in_prob : the table of nearby galaxies to GW event
  • grb_in_prob : the table of nearby GRB events to GW event

Product: GW Events Parameters Estimation

This product shows the results of the parameters estimation for all events catalogs. It shows you how to read and plot parameter estimation results. For example, GWTC-3 is the "third gravitational wave transient catalog" from the LIGO, Virgo, and KAGRA (LVK) Collaborations. GWTC-4 catalog updates previous ones (GWTC-3, GWTC-2.1) with merger events observed during the first part of Observing Run O4a. The events listed in the catalogs are binary mergers between black holes and neutron stars. Both black holes and neutron stars are classified as compact objects. The LVK detectors are sensitive to three different types of compact binary mergers:

  • Binary black hole (BBH) merger
  • Binary neutron star (BNS) merger
  • Neutron star - black hole (NSBH) merger

What is parameter estimation?

After detecting a gravitational wave signal, we would like to learn about what kind of compact object produced the signal. We can then answer questions like:

  • What type of source was it likely to be? BBH, BNS, NSBH or MassGap.
  • How massive were the black holes or neutron stars that collided?
  • Which direction on the sky did the signal come from?

The properties (or parameters) of the merging objects decide the shape of the gravitational waveform signal.

Parameter estimation is a detailed analysis of the shape of the signal waveform. We compare many possible waveforms to the observed signal and find credible ranges for the parameters of the source.

There are two types of parameters that describe a gravitational wave signal.

  • Intrinsic parameters are properties of the binary itself, including the masses and spins of the compact objects.
  • Extrinsic parameters are properties which describe how we view the merger from the Earth. This includes the position of the source in the sky, the distance to the source, and its orientation.

More information on parameter estimation can be found in the GWOSC Open Data Workshops.

The result of parameter estimation analysis is a list of posterior samples for each parameter. The samples represents the posterior distribution of the parameter, which tells us the most credible values of the signal parameters. This product shows you how to read and plot parameter estimation samples as a starting point for use in your own analysis.

For more information on gravitational wave data and GWTC have a look at:

Input parameters (MMODA)

Here we define the input parameters for the product:

  • src_name is the Object name (on top common panel). For example GW150914. You can retrieve other source names using the skymap event search module.
  • delta_t (float) is the spacing between frequency samples
  • f_low (Hz) is the frequency to start evaluating the waveform
  • method is the approximant used to produce posterior samples
  • approximant is the approximant wafeform used to produce the template fit
  • start (seconds) is the lower bound of time window
  • stop (seconds) is the higher bound of time window
  • fs_low (Hz) id the strain bandpass low frequency cut-off
  • fs_high (Hz) is the strain bandpass high frequency cut-off

Downloading the data

We download data directly with pesummary.gw.fetch.fetch_open_samples. This will download the specified event data and unpack into the local directory.

The cumulated different analysis configurations used in the catalogs are:

  • IMRPhenomXPHM: Samples produced using the IMRPhenomXPHM waveform, a phenomenological waveform including precession and higher modes
  • SEOBNRv4PHM: Samples produced using the SEOBNRv4PHM waveform, an effective one-body waveform including precession and higher modes
  • Mixed: A mixed set of samples created by combining results from both waveforms into one set of samples
  • IMRPhenomXO4a: Samples produced using the IMRPhenomXO4a waveform, a phenomenological waveform including precession and higher modes
  • SEOBNRv5PHM: Samples produced using the SEOBNRv5PHM waveform, an effective one-body waveform including precession and higher modes
  • NRSur7dq4: Samples produced using the NRSur7dq4 waveform, a numerical relativity surrogate waveform including precession and higher modes
  • Mixed: A mixed set of samples created by combining results from IMRPhenomXPHM-SpinTaylor, SEOBNRv5PHM, and NRSur7dq4 into one set of samples
  • Mixed+XO4a: A mixed set of samples created by combining results from the four waveforms included in Mixed as well as IMRPhenomXO4a into one set of samples

C01 or C00 labels refer to the data quality level that the parameter estimation runs were performed on. C01 is data that has gone through a more detailed process of cleaning and conditioning.

We can access the samples through the data.samples_dict property. This is a MultiAnalysisSamplesDict object, containing samples for all runs listed above via data.labels. We can then access each individual set of samples within the samples_dict as a SamplesDict object. Both SamplesDict and MultiAnalysisSamplesDict are nested Python dictionaries.

Plotting posterior distributions

pesummary can be used to make plots of the posterior distributions. We use here simply matplotlib by chosing parameters and plot the posterior distribution as a simple histogram. We choose to show only the plot for mass 1, mass 2, final mass, and luminosity distance posteriors only, to avoid saturating the outputs. However, a complete list of posterior contained in the PEDataRelease file is shown in the tool_log, for information.

Detector Strain Data

Raw detector data for each event can be found via the Gravitational Wave Open Science Centre (GWOSC). We can then check this data for signals and compare to our waveforms to see how consistent the results of our parameter estimation are with the true data. We can find the segments containing the data directly using the gwosc python package. It is possible to load the strain data directly from GWOSC, using Gwpy Timeseries object. The GWOSC website, linked above, will also allow to directly download the strain data at different sampling rates.

The expected merger time is given by the [H-L-V]1_time parameter from the approximant sample set. This parameter is the sampler calculated time of merger in the various detectors.

The data is given with only the frequencies between fs_low and fs_high cut-offs to represent detectors sensitive region.

Outputs (MMODA)

Here we define the outputs of the product:

  • fig_mass1 : the mass1 distribution
  • fig_mass2 : the mass2 distribution
  • fig_finalmass : the final mass distribution
  • fig_luminositydistance : the luminosity distance distribution
  • fig_strainH1 : The event strain fitted of Handford detector
  • fig_strainL1 : The event strain fitted of Livingston detector
  • fig_strainV1 : The event strain fitted of Virgo detector

Product: Catalogs Event Selection

The gwosc package allows users to search for publicly released events. In this product we query gwosc database events using some selection criteria like mass1, mass2 lower and highest bounds, and luminosity distance. Optionally, one can restrict the search in a time window as given in the main common window. In output we give the result of the query as a list of event names with corresponding gps detection time. You can use directly the result events names as an input source name to be used in the parameter estimation product.

Input Parameters (MMODA)

Here we define the input parameters for the product:

  • m1_low (solar mass) : the lowest bound of mass 1 source (heavier)
  • m1_high (solar mass) : the highest bound of mass 1 source
  • m2_low (solar mass) : the lowest bound of mass 2 source (lighter)
  • m2_high (solar mass) : the highest bound of mass 2 source
  • dist_low (Mpc) : the lowest bound of luminosity distance
  • dist_high (Mpc) : the highest bound of luminosity distance
  • use_time (bool) : if True use the time interval given in the main common window

Outputs (MMODA)

Here we define the outputs for the product:

  • eventselection_List : the table of result events from selection criteria
  • gpsselection_List : the table of result gps corresponding to events

Product: Catalogs statistics

This product loads one or several confident GWTC catalog from GWOSC EventAPI, retrieves detector networks distribution (H1/L1/V1) from GWOSC, then computes the 90% credible sky area for (almost) all events using the official Zenodo PE skymap tarballs.

Input Parameters (MMODA)

Here we define the input parameters for the product:

  • use_gwtc4 : "GWTC-4.0 catalog"
  • use_gwtc3 : "GWTC-3-confident catalog"
  • use_gwtc2 : "GWTC-2.1-confident catalog"

Outputs (MMODA)

Here we define the outputs for the product:

  • fig_network : catalogs distribution PIE of detectors network (LIGO - VIRGO)
  • fig_pie : catalogs distribution PIE of BBH, BH-NS, NS-NS
  • fig_m1m2 : source frame M1 vs source frame M2 masses
  • fig_histograms : total mass, luminosity distance and SNR catalogs distribution histograms
  • fig_skylocalization : Compare localization for 1, 2 or 3 detectors (BBH only)

Python libraries used and code repository

Following python modules used in this module are developed byt the LVK collaboration:

Acknowledgments

This work was made possible thanks to the ACME program, the LVK collaboration, the GWOSC team, and AstroOrdas project team. Particular thanks to Denys Savchenko who helped me to set-up the module. For any help or questions related to this module, please contact Daniel Sentenac