Observing Run O4 for the LIGO/Virgo/Kagra detector network ist set to begin on May 24th 2023. For amateurs of the High Energy Network this aboth could be an interesting time, The hope is that the gravitational wave detectors will detect two neutron stars merging during the next ca 18 months, hopefully in a galxy not too far away. Only one such event has even been observed in gravitational waves and in electromagnetic wavelengths, a second event would be extremely useful for astrophysicists. The problem: the gravitational wave detectors have a very bad angular resolution, we will know that an event has happened, but we will initially only get a very imprecise sky localization of the event. Optical telescopes will then have to survey the region of the sky hinted to by the GW detectors and look for the optical counterpart of the event, focusing on the galaxies in the region. All alerts will be public, if you are interested in participating in the hunt you should use GCN or a smart phone up like https://apps.apple.com/us/app/gravitational-wave-events/id1441897107 Joinming collaborations like the KilonovaCatchers is a good way to coordinate with others, see the threads in this forum for details.
I think it's a good idea to put discussions concerning the preparations for O4 and actual alerts into one thread in this forum. This is by no means intended to be blog or a monologue :-.) , let's have a joint discussion.
Some last-minute news before O4 begins:
The Italian detector, VIRGO, has failed to reach the target sensitivity during the commissioning phase leading up to O4. To meet the goals, it is necessary to carry out some rather extensive repairs, e.g. part of the vacuum system needs to be vented :-(, parts exchanged, and then te vacuum needs to be restored.
https://www.ego-gw.it/blog/2023/05/11/virgo-postpones-entry-into-o4-observing-run/
In a virtual "townhall" meeting with EM observers yesterday May 11th 2023, a spokesman for Virgo mentioned "Fall 2023" as the expected date when Virgo will be able to join O4 with something close to the target detector performance.
This has significant impact for us amateur observers: without VIRGO, only the two LIGO detectors will be available to "triangulate" the sky position of a detected gravitational wave event such as a neutron star merger. Or in other words: the sky maps that will come with alerts to indicate where to point our telescopes to look for optical counterparts will have much bigger uncertainties. Now if you want to see something positive in this: it means that more eyes are needed on the sky and the contributions of many amateur astronomers could be even more valuable.
CS
HBE
How does an EM observer sign up for GW alerts?
Ray
There are several options:
Via GCN: Even tho the "G" in GCN stands for Gamma (Ray Burst), the network is now also used fro things like neutrino and Gravitational Wave alerts. GCN recently got a new user interface (the legacy system is still also up, tho), read more at https://gcn.nasa.gov/
Via a smart phone app: For Apple devices, there is this here : https://apps.apple.com/us/app/gravitational-wave-events/id1441897107 Unfortunately not ported to Android
Both these services will give you access to the raw alerts issued by the detector collaboration, but a) these are quite technical and b) once an interesting event happens you will want to have additional information like the galaxies contained in the sky localization regions of the alert, because the kilonova we are hunting for will happen inside a host galaxy, not some random spot in the sky. This is something you can either do yourself, or you can join a Pro-Am collaboration that offers this as a service along with the raw alerts:
Join KilonovaCatchers (KNC): Highly recommended!! http://kilonovacatcher.in2p3.fr/ This gives you a lot of advantages in the hunt for a kilonova:
So my recommendation is: Join KNC, and then use the phone app (and possibly GCN) to know when to rapidly get to your scope and get going observing, with the help of observation recommendations that KNC will give you shortly after the alert.
CS
HB
You might also be…
Ray,
You might also be interested in the Gravitational Wave Open Science Center (gwosc.org). They just finished (yesterday 5/17) a series of lectures on this topic. They also have the several years of archived lectures. This is pretty detailed technical stuff, but a good way to understand GWs. Check it out.
Dave H.
In an interesting coincidence, several GCNs (33813-33817) have reported a possible compact binary merger with one of the objects potentially being a neutron star just detected by LIGO. You can find the GCNs on the HEN webpage miscellaneous links page.
Dave
So today saw the first "real" (this is not a drill!) public alert in O4, even tho the LIGO detectors are still in the "Engineering Run" mode and the alert generation is still somewhat manual. Spoiler alert: nothing to get exceited about, but as an educational tool, let's go thru this alert and see why you should not spend all your money on observing this via iTelescope yet.
As a text GCN notice, the alert history can be seen here: https://gcn.gsfc.nasa.gov/notices_l/S230518h.lvc
Let's look at it line by line:
There's a lot to unpack here.
TRIGGER_NUM: S230518h
The name of the event. For every event there will be more than one notice message, updating the information on the event as analysis progresses, and events can also be retracted. As more than one event might see messages being updated during any given time, you need to be clear about which message belongs to which event, and whether a new message refers to a new event or is just an update/retraction. See "NOTICE_TYPE" and "SEQUENCE_NUM" for this.
GROUP_TYPE: 1 = CBC
CBC stands for compact binary coalescence. So this event is about either neutron stars or black holes merging. Another type could be "Burst" which is about supernovae and similar events. Most alerts will be about CBCs, so let's focus on this for the moment.
FAR: 3.219e-10 [Hz] (one per 35957.2 days) (one per 98.51 years)
FAR="False Alarm Probablity" , an estimation about how credible this event is. Roughly speaking you would expect an alert of this "quality" to be a false alert (not actually associated to a real astrophysical event) once every 98.51 years. SO this alert seems pretty solid. Good. The Threshold for public alerts in O4 is , IIRC, around only a FAR of 2 per day. So yes, you should expect 2 alerts per day (!) but the FAR value will tell you just how excited (confident) you should be about each alert.
PROB_NS: 1.00 [range is 0.0-1.0]
The probability that a neutron star was involved in the merger, IF we assume this is actually a real merger event. Here the pobability is 1.0 (100%). If this is real , it's certain that a neutron star was involved. Good!!! You need at least one neutron star to create an optical signal. Two black holes merging are probably not making a light show.
PROB_REMNANT: 0.00 [range is 0.0-1.0]
Probability that the merger (assuming it is a real event) left a remnant that can be seen in electromagnetic radiation (somewhat simplified expnanation). Here the pobability is 0%. BUMMER! How come?
PROB_BNS: 0.00 [range is 0.0-1.0]
PROB_NSBH: 0.86 [range is 0.0-1.0]
PROB_BBH: 0.03 [range is 0.0-1.0]
PROB_MassGap: -1 [range is 0.0-1.0] VALUE NOT ASSIGNED!
PROB_TERRES: 0.09 [range is 0.0-1.0]
These numbers should add to 1.0. Here we learn that the event was a neutron-star-neutron-star merger with probaility 0%. That's too bad. NS-NS collisions are the best way to create a kilonova that we are hunting for. With probability 86%, it was a merger between a black hole and a neutron star. It depends on the mass of the bodies whether the NS gets "eaten" as a whole (and once the NS is inside the black hole we won't see any electromagnetic radiation from this because what happens in a black hole stays in the black hole) or whether the neutron star is torn into pieces while it inspirals into the black hole, which could create a light signal for us to detect. But in this case (PROB_REMNANT=0%) the alert is certain that the former is the case. BUMMER
PROB_MassGap = -1 must be some kind of bug. It refers to events where at least one of the bodies seems to have a mass that is somewhere between that of a typical neutron star and that of a black hole.
PROB_TERRES: 0.09
means there is a 9% chance that this is just caused by some terrestrial effect on the detectors, not a real astrophysical event. Intuitively this is a bit at odds with the low false alarm rate of once per almost 100 years. For the detailed statistical interpretation I refer you to https://emfollow.docs.ligo.org/userguide/index.html
SKYMAP_FITS_URL: https://gracedb.ligo.org/api/superevents/S230518h/files/bayestar.multiorder.fits
This is the URL where you can download a skymap that gives a hint where the event came from. This is not so much an image for human consumption but a data file that encodes for every region in the sky how likely the event happend in that direction and if so, what is the likely distance to the event if it happend in that direction. So this map carries information about 3 dimensions: RA, DEC, and distance (!) which makes it useful for finding matching potential host galaxies.
At the moment the link seems broken with a permission problem on the server, so let's discuss the sky maps on another day. If you could download the map, you would see that the potential region that the event comes from forms a huge ring in the sky, with the most interesting parts covering an area of > 1000 square degrees! That's a lot.
Unfortunately, until the third detector in Italy is repaired and ready to join O4 (see message above), we cannot hope for better sky maps most of the time which is really sad.
CS
HB
Heinz,
Thanks for going into the details of the alert.
Mike F
Seems like I heard that more detectors are in the works. How soon do India and Japan come online?
LIGO India is at an early stage of construction and isn't even mentioned in the long term planning timeline published here: https://observing.docs.ligo.org/plan/
KAGRA in Japan has an ambitious design (cryogenic mirrors, underground tunnels and more) which turned out to be quite challenging to implement. It will join O4 for short periods but the detector's range is such that it would spot a BNS event in the Andromeda galaxy but not much more distant...so it's not that useful for O4 and will become scientifically more relevant only in O5.
CS
HB
Meanwhile the alert system has been switched to fully automatic mode, and the human rapid response team that will generate GCN circulars are also in action. There was an alert yesterday that probably is more representative of the majority of alerts that we can expect to come, and you can read all about it in a GCN circular.
https://gcn.nasa.gov/circulars/33838
As you can see, there can be a delay of several hours between the event being detected (which will trigger the more machine readable GCN notices) and the circular being issued, so you will want to listen to the notices either directly or thru phone apps like astro-colibri. Still, the circulars basically do the same thing that I tried above: explain the alert data in a human-readable form and it is instructive to read the circulars.
The key sentences from the circular linked above are:
"S230520ae is an event of interest because its false alarm rate, as estimated by the online analysis, is 3.1e-09 Hz, or about one in 10 years. The event's properties can be found at this URL: https://gracedb.ligo.org/superevents/S230520ae"
Ok, this should very likely be a real event and not just a detector fluke. Good. But...
"Assuming the candidate is astrophysical in origin, the probability that the lighter compact object has a mass < 3 solar masses (HasNS) is <1%. [6] Using the masses and spins inferred from the signal, the probability of matter outside the final compact object (HasRemnant) is <1%. [6] "
Bad for us optical astronomers. We would like to see higher probabilities for these metrics in order to get excited.
"The classification of the GW signal, in order of descending probability, is BBH (>99%), Terrestrial (<1%), BNS (<1%), or NSBH (<1%). "
Too bad, BBHs are not we are looking for. It's BNS events that most likely can be seen in optical EM.
"The preferred sky map at this time is bayestar.multiorder.fits,1. For the bayestar.multiorder.fits,1 sky map, the 90% credible region is 1716 deg2. Marginalized over the whole sky, the a posteriori luminosity distance estimate is 1974 +/- 583 Mpc (a posteriori mean +/- standard deviation). "
For some weird reason, the distance estimate is given in the circulars but is not given as a dedicated field in the machine readable notices. Instead, the estimate is provided as meta-data in the headers of the FITS files containing the sky-map. Let's discuss this on another day when we have a somewhat more relevant alert.
CS
HB