A short talk on GW170817 at UoC

A few weeks ago couple of months ago I gave a small talk during the journal club of the Astronomy group at the University of Crete (on 20/Oct/2017) about the most recent gravitational wave detection (GW170817) corresponding to a neutron star merger. As I have been involved in a couple of papers [1] it was the prime time to present it after the official announcement on Oct 16th. However, I thought that this event deserves a more detailed post (that took a couple of months to finalize).

Before Aug 17th, 2017, there have been only 4 confirmed detections: GW150914 (the first one [2]), GW151226, GW170104, and GW170814 (plus LVT151012 as a tentative detection). The last one (GW170814) is the first detected by both LIGO and VIRGO, which joined the collaboration on Aug 1st, 2017. So, it was only a few days after that that they detected another very important event.

On Aug 17th, 2017, at 12:41:04 UTC, a clear signal was detected by LIGO-Hanford (just at the end of the second LIGO cycle – 9 days later and LIGO would have have been off!). The raw data from LIGO-Livingston detector included a glitch (see Christofer Berry’s post [3]). After reprocessing the data to remove this artifact there was a another clear signal. The VIRGO didn’t manage to show any significant signal but that was due to its antenna orientation and sensitivity (which is important to constrain the sky positions though). The duration of the signal was approximately 60 s allowing for about 3000 cycles. This is the longer and the stronger (at SNR~32.4) signal detected so far (Abbott et al. (LIGO Scientific Collaboration and Virgo Collaboration) 2017, [4]).

Time-frequency representations of data containing the gravitational-wave event GW170817, observed by the LIGO-Hanford (top), LIGO-Livingston (middle), and Virgo (bottom) detectors. Times are shown relative to August 17, 2017 12∶41:04 UTC. The amplitude scale in each detector is normalized to that detector’s noise amplitude spectral density. In the LIGO data, independently observable noise sources and a glitch that occurred in the LIGO-Livingston detector have been subtracted. (Text and Fig. 1 from
Abbott et al. (LIGO Scientific Collaboration and Virgo Collaboration) 2017, [4])

However, such a long signal is expected to be produced by the merging of neutron stars. And indeed the total mass estimate was 2.73–3.29 M☉ with a mass ratio of 0.4–1.0. This means that the individual masses of the sources that merged were 1.36–2.26 M☉ and 0.86–1.36 M☉, well within the limits of neutron star masses.

GW170817 Localization and Triangulation Annuli. We can pinpoint sources like GW170817 much more accurately now that we can triangulate the signal between Hanford, Livingston, and Virgo. The rapid Hanford-Livingston localization is shown in blue, and the final Hanford-Livingston-Virgo localization is in green. The gray rings are one-sigma triangulation constraints from the three detector pairs. (Text and figure from LIGO page on GW170817 [5])

Almost at the same time (~2s after the GW detection) FERMI (and INTEGRAL) detected a short Gamma-Ray Burst (GRB170817A). The detection of such a contemporaneous signal triggered the community that something exceptional was going on. It is interesting to note that the detection of the GRB only (without any GW detection) wouldn’t be sufficient to mobilize a follow-up campaign, especially because the error of sky position from FERMI is huge. The localization obtained from the GW detection and the information that an electromagnetic (EM) counterpart may exist increased the significance of this source.

Left: GRB data (top) collected by Fermi and gravitational wave data (bottom) collected by LIGO. Right: Source localization on the sky from the INTEGRAL GRB satellite (light blue band), Fermi (the dark blue disk), LIGO alone (green ovals), and LIGO and VIRGO data combined (dark green oval). Notice that all sources identified the LIGO-Virgo area. [Credit: LIGO, Virgo, Fermi, Swope, DLT40] [Text and image (slightly modified) from LivingLigo post [6])

Very soon after the detection all collaborators of LIGO/VIRGO were informed about the possible EM counterpart. Around 70 teams around the world started using almost all available (both ground and space) telescopes to detect and study it. The first group that managed to discover the EM counterpart was the One-Meter, Two-Hemisphere collaboration (Coulter et al. 2017, [7]) that uses the Nickel telescope at Lick Observatory in California and the Swope telescope at Las Campanas Observatory in Chile. Due to the sky position of the GW source (southern hemisphere and close to the Sun) only the Swope telescope could actually observe around that region, and only for a couple of hours before it set (a month later and this sky region would have been totally hidden behind the Sun). At the time of LIGO trigger it was daylight in Chile they had to wait at least 10 hours before any observation initiates. In the meantime though the collaboration had time to prepare the observing strategy. As they aim to observe the best candidate galaxies according to the properties derived from the GW detection, they built a prioritized list of galaxies based on the position (within the locus identified by LIGO/VIRGO), the distance (around 40 Mpc as estimated by the GW detection), the stellar mass and star formation of the galaxy (for more details see Coulter et al. 2017). The compiled list included 100 galaxies and they start observing them at 23:23 UT.

Sky region covering the 90th-percentile confidence region for the location of GW170817, along with the 50th, 70th, and 90th-percentile contours (indicating the probability to find the host galaxy). Grey circles represent the locations of galaxies observed by the Swope telescope on 2017 August 17-18 to search for the EM counterpart to GW170817. The size of the circle indicates the probability of a particular galaxy being the host galaxy for GW170817. The square regions represent individual Swope pointings with the solid squares specifically chosen to contain multiple galaxies (and labeled in the order that they were observed) and the dotted squares being pointings which contained individual galaxies. The blue square labeled ’9’ contains NGC 4993, whose location is marked by the blue circle, and SSS17a. (Text and figure from Coulter et al. 2017, [7])

After only 20 min (at 23:33 UT equivalent to 10.9 h after LVC trigger), on the 9th image they obtained they identified a new source in the galaxy NGC 4993. An S0 galaxy at 40 Mpc, it was the 12th most probable galaxy to host the GW source with a probability of 2.2%. The new transient, named SSS17a, was detected at i=17.476±0.018 mag (impressively, its V magnitude was at 17.35 mag well within the range of equipment used by amateur astronomers).

3×3 arcmin images centered on NGC 4993 with North up and East left. Panel A: Hubble Space Telescope F606W-band (broad V) image from 4 months before the GW trigger (25, 35). Panel B: Swope image of SSS17a. The i-band image was obtained on 2017 August 17 at 23:33 UT by the Swope telescope at Las Campanas Observatory. SSS17a is marked with the red arrow. No object is present in the Hubble image at the position of SSS17a. (Text and figure from Coulter et al. 2017, [7])

After the discovery of the new transient an intensive spectroscopic and photometric campaign at the Las Campanad Observatory initiated. I was fortunate enough to find myself there (with some colleagues) for an observing run. At the time we were asked to obtain spectra and images of a source without any more information regarding what it was (apart from the fact that these observations were triggered by a LIGO event). Actually one of our images (taken on Aug 21) shows the dramatic change in color of SSS17a, only a few days after its detection.

Pseudo-color images of SSS17a in the galaxy NGC 4993. Images are 1×1 arcmin and centered on NGC 4993; SSS17a is indicated by a blue arrow in each panel. The red, green, and blue channels correspond to the H-band, i-band, and g-band images. (A) Images taken on the night of 2017 August 17, 0.5 days after the merger. (B) Images taken on the night of 2017 August 21, 4.5 days after the merger. Over four days SSS17a both faded and became redder. (Text and figure from Drout et al. 2017, [8])

Both the light curves and the spectra display fast changes in the temperature of the expelled material. Even within the first hour of observations the spectra show a drop of temperature from ~11000K to 9300K, which is indicative of a material expansion at ~0.3c.

Top: Evolution of the ultraviolet to near-infrared spectral energy distribution (SED) of SSS17a. (A) The vertical axis is the logarithm of the observed flux. Fluxes have been corrected for foreground Milky Way extinction. Detections are plotted as filled symbols and upper limits for the third epoch (1.0 days post-merger) as downward pointing arrows. Less-constraining upper limits at other epochs are not plotted for clarity. Between 0.5 and 8.5 days after the merger, the peak of the SED shifts from the near-UV (<4500 A) to the near-IR (>1 μm), and fades by a factor >70. The SED is broadly consistent with a thermal distribution and the colored curves represent best-fitting blackbody models at each epoch. In 24 hours after the discovery of SSS17a, the observed color temperature falls from ≳10,000 K to ∼5,000 K. The epoch and best-fitting blackbody temperature (rounded to 100 K) are listed. (B) Filter transmission functions for the observed photometric bands. (Text and figure from Drout et al. 2017, [8])
Bottom: Spectroscopic time series of SSS17a. The vertical axis is observed flux. Observations began ∼0.5 days after the merger and were obtained with the LDSS-3, MagE, and IMACS spectrographs on the Magellan telescopes. These spectra have been calibrated to the photometric observations. Colored bands indicate the wavelength ranges of the g, r, i, z, and Y photometric filters. (Text and figure from Shappee et al. 2017, [9])

In total, its evolution has been faster and unlike anything else we have observed so far. This very fast expansion of the material and its cooling could be attributed to the formation of lanthanide elements through the r-process. All elements up to Fe can be produced within the massive stars and some of the heavier elements during their supernovae explosions. Theory predicted that the majority of the heaviest elements (such as gold, platinum, uranium, etc) should be produced during the merging process of neutron stars, where a large number of neutrons are available within very short times. But we lacked observations up to now.

Spectra of SSS17a compared with a broad range of other astronomical transients at several evolutionary phases. While the ∼0.5 day spectrum of SSS17a has few features and is potentially an extreme version of some other hot and/or fast transients, it evolves rapidly in comparison. Within 3 days of the LIGO trigger, the optical spectrum of SSS17a is no longer similar to other known transients. Dates listed are relative to the time of explosion for all objects. All spectra are divided by their median value and displayed with arbitrary additive offsets for clarity. (A) SSS17a compared to the Type Ia supernova (SN Ia) SN2011fe and the afterglow spectrum of the short gamma-ray burst GRB130603B. Few observations of other transients within 1 day of explosion are available. (B) SSS17a at 3.46 d after explosion compared to the SN Ia ASASSN-14lp, the Type II supernova SN2006bp, and the long gamma-ray burst and its associated afterglow and broad-lined Type Ic supernova GRB030329/SN2003dh at similar times relative to explosion. (C) SSS17a at 7.45 d after explosion compared to SN2011fe, the rapid blue transient PS1-12bv, the fast Type Ic supernova SN2005ek, and the GRB/SN GRB980425/SN1998bw. (Text and figure from Shappee et al. 2017, [9])

Apart from the gravitational waves, a neutron star merger would become visible as a transient event, known also as a kilonova [10]. The current models are able to fit only partly the observed data, but help to derive important conclusions regarding the nature of these explosions. In particular, SSS17a could be described better by a two-component mass ejection, with each component being responsible mainly for the the early and later behavior. Estimates of the released material reach up to a few Earth masses for gold and platinum (and up to 16000 Earth masses for heavier elements in total).

Comparison of SSS17a to theoretical models. The vertical axis is observed flux. The models shown are for three possible physical interpretations of SSS17a. While the red kilonova model provides a reasonable likeness to the data at late times, the early time spectra and kinematics require lanthanide-free relativistic material. No single model shown here or described in the current literature can self-consistently reproduce the full spectroscopic time series of SSS17a. (A) Lanthanide-rich red kilonova model from a neutron star merger and dynamical ejection. At each epoch, the absolute luminosity is scaled to match the data. (B) Disk-wind model with a neutron star that immediately collapses after the merger. The absolute luminosity is scaled by a factor of ≳10 at each epoch to match the data. (C) Lanthanide-poor blue kilonova model from a neutron star merger and dynamical ejection. The model has been crafted to match the observations at early times. (Text and figure from Shappee et al. 2017, [9])

It took SSS17a almost 9 days before CHANDRA managed to detect a X-ray faint source, and 16 days from the discovery to become visible in radio wavelengths also, producing finally light across the whole EM spectrum (see the corresponding infographic).

Luck was definitely on our side, as the position and the timings were marginal: the GRB was not exciting by itself, SSS17a was setting within one hour and one month later it would have been hidden behind the Sun, 9 days later LIGO would be down. Despite these constraints, the astronomical community and the whole collaboration proved to be prepared enough not to lose this unique opportunity. The event mobilized almost half the astronomers around the globe (approximately 4000 people in a community of 10000). This led to a an impressive number of publications counting ~250 GCN circulars and ~80 papers during the “first wave” of its announcement only (with the most striking cases of the ApJ Letter of 60 pages containing all obtained observations; LIGO et al. 2017 [11]). The new era of multi-messenger astronomy is here and will routinely discover and study new events aftert the future upgrades of the GW detectors.

–> The captions of the figures have been slightly modified to include most important information (references have been removed for example).

Interesting links
— LIGO GW170817 infographic: https://dcc.ligo.org/public/0146/G1701993/008/infographic.png
— LIGO GW170817 factsheet: www.ligo.org/detections/GW170817/images-GW170817/GW170817_Factsheet.pdf
— GCN circular archive: https://gcn.gsfc.nasa.gov/gcn3_archive.html
— List of “first-wave” papers: https://lco.global/~iarcavi/kilonovae.html
— ApJL focus publications: http://iopscience.iop.org/journal/2041-8205/page/Focus_on_GW170817?utm_medium=email&utm_source=iop&utm_term=&utm_campaign=12734-36073&utm_content=FI
— Official reddit.com page: https://www.reddit.com/r/IAmA/comments/76yu54/we_are_the_ligo_scientific_collaboration_the/
— Christofer Berry’s site: https://cplberry.com/2017/10/16/gw170817/
— LivingLigo: https://stuver.blogspot.com/2017/10/GW170817.html

[1] http://maravelias.info/2017/10/the-first-neutron-star-merger
[2] http://maravelias.info/2016/02/gravitational-waves-detected
[3] https://cplberry.com/2017/10/16/gw170817, accessed on 19/Oct/2017
[4] Abbott et al. (LIGO Scientific Collaboration and Virgo Collaboration) 2017, Phys. Rev. Lett. 119, 161101, https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.119.161101
[5] http://www.ligo.org/detections/GW170817.php, accessed on 19/Oct/2017
[6] https://stuver.blogspot.gr/2017/10/GW170817.html, accessed on 19/Oct/2017
[7] Coulter et al. 2017, Science/eaap9811, http://science.sciencemag.org/content/early/2017/10/13/science.aap9811
[8] Drout et al. 2017, Science/eaaq0049, http://science.sciencemag.org/content/early/2017/10/16/science.aaq0049
[9] Shappee et al. 2017, Science/eaaq0186, http://science.sciencemag.org/content/early/2017/10/13/science.aaq0186
[10] https://en.wikipedia.org/wiki/Kilonova, accessed on 14/Nov/2017
[11] LIGO et al. 2017, “Multi-messenger Observations of a Binary Neutron Star Merger”, arXiv:1710.05833, https://arxiv.org/abs/1710.05833

Leave a Reply

Your email address will not be published. Required fields are marked *