Detector Watch

Mahray
Mahray
Joined: 11 Nov 04
Posts: 43
Credit: 4,913,730
RAC: 33,016

Keep up the watch mate, very

Keep up the watch mate, very interesting information.

Chipper Q
Chipper Q
Joined: 20 Feb 05
Posts: 1,540
Credit: 708,571
RAC: 0

I was wondering about the

I was wondering about the gamma-ray burst indicator, seen near the lower right-hand corner of the site overviews. These snippets are from a couple different overviews during 10/16/6:

I'm guessing about the 'on/off' status... I was wondering how important it is to know (in real-time) when a GRB is detected, since there's an indication on the overviews. Also, what specifically triggers the indication? I'm guessing it's tied in with NASA and ESA's GRB detecting satellites. But are there special things to look for (e.g., in the 'BurstMon' chart) when the indication in 'on'?

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,054
Credit: 109,566,635
RAC: 43,131

RE: I was wondering about

Message 47938 in response to message 47937

Quote:
I was wondering about the gamma-ray burst indicator, seen near the lower right-hand corner of the site overviews. These snippets are from a couple different overviews during 10/16/6:

I'm guessing about the 'on/off' status... I was wondering how important it is to know (in real-time) when a GRB is detected, since there's an indication on the overviews. Also, what specifically triggers the indication? I'm guessing it's tied in with NASA and ESA's GRB detecting satellites. But are there special things to look for (e.g., in the 'BurstMon' chart) when the indication in 'on'?


I too have wondered, so with some searching I've found a number of files on this:

- Searching the LIGO data for coincidences with Gamma Ray Bursts
- Search for Short-Duration GW Bursts Coincident with S2/S3/S4 Gamma-Ray Bursts S5 GRB-GWB Search: First Results
- Short Gamma-Ray Bursts and Compact Binary Mergers Predictions for LIGO
- Gravitational radiation from gamma-ray bursts as observational opportunities for LIGO and VIRGO
- Gravitational radiation from gamma-ray bursts
- Searching for Gravitational Wave Bursts Using External Triggers From Gamma-ray Bursts (GRBs)
- Coincident γ-ray and gravitational wave observations

To summarise some points of interest for you, from these sources:

GRB's are basically two sorts.

- Short are less than 2 seconds, and are hypothecated to originate from compact binary mergers that is NS/NS or NS/BH. I don't think BH/BH mergers squirt out much electromagnetic radiation per se, but I guess their accretion discs do. Anyway if a neutron star ( or white dwarf, or .... ) gets too close to the black hole then it breaks up and forms a torus/donut of material outside the innermost stable circular orbit ( ISCO ) of the black hole - as ( by definition ) anything closer must move in to the event horizon - this torus is presumably the source for the gammas. It would be a pretty rowdy neighbourhood! :-)

- Long are more than 2 seconds, and generally thought to be from massive stellar collapse/supernovae - with some sort of relativistic 'fireball' that subsequently emits also in the x-ray, optical and radio bands - and maybe formation of a black hole within. Another rowdy place...... :-)
Here's an image which displays what is meant by a non-spherically symmetric, or quadrupolar pattern:

Careful not to confuse the terminology with GWB's ( Gravitational Wave Bursts ) - I did! :-)

There is uncertainty in the delay between the gravity waves and the gamma rays. The gravity waves are probably first. The GWB detection/trigger essentially marks/timestamps the LIGO data ( or some analysis effort thereon ) for subsequent search before ( say 120 seconds ) and after that time ( say 60 seconds ).

Largely the GRB triggers are from the SWIFT, BEPPOSAX, INTEGRAL, HETE, XTE, ULYSSES etc group of gamma detecting satellites. This is managed through the Gamma Ray Burst Coordinates Network, see GCN.

I'm not sure about 'BurstMon' etc at present.

So crunchers and burners, here's yet another channel of analysis from the LIGO's ... :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Chipper Q
Chipper Q
Joined: 20 Feb 05
Posts: 1,540
Credit: 708,571
RAC: 0

Wow: the GRB triggered

Wow: the GRB triggered searches are in the same data pipeline as the blind inspiral searches, but a lower signal-to-noise ratio can be used for the filter threshold on the data, which increases the effective range; if you can decrease the SNR to 4, then the LIGO search range is increased by a factor of 2. So that could be out to ~30 Mpc, depending on the type of GRB.

If the progenitor of the GRB is the merger between compact objects, and within range of the LIGOs, then it's possible warn the GRB detecting network more than a minute ahead of time (by detecting the final moments of inspiral) for where to look in the sky to capture the entire, uh, is it also referred to as 'light curve' (as with novae) of the GRB ? (Probably properly called a 'burst' if the progenitor is uncertain?)

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,054
Credit: 109,566,635
RAC: 43,131

RE: Wow: the GRB triggered

Message 47940 in response to message 47939

Quote:

Wow: the GRB triggered searches are in the same data pipeline as the blind inspiral searches, but a lower signal-to-noise ratio can be used for the filter threshold on the data, which increases the effective range; if you can decrease the SNR to 4, then the LIGO search range is increased by a factor of 2. So that could be out to ~30 Mpc, depending on the type of GRB.

If the progenitor of the GRB is the merger between compact objects, and within range of the LIGOs, then it's possible warn the GRB detecting network more than a minute ahead of time (by detecting the final moments of inspiral) for where to look in the sky to capture the entire, uh, is it also referred to as 'light curve' (as with novae) of the GRB ? (Probably properly called a 'burst' if the progenitor is uncertain?)


Nice thought! But I don't think the data pipeline on multiple interferometers is, as yet, realtime enough to grant that... :-)
I think 'light curve' is the right idea, I suppose we'll need something similiar for the gravity waves now - err, 'strain curve', 'strain per root Hertz curve' ... :-)

Also, H2 continues to be down, now out of science mode for a week. Focus is on the end test mass on the Y arm ( ETMY ). I think they have broken the vacuum ( close the gates on either side of the short segment containing ETMY, then open the hatch ) to inspect and photograph. Thus H1 was also affected but it is now back in science mode. I recall reading they have some pretty high capacity pumps ( Varian 2500 litres/sec ? ), so I guess they restore the vacuum fairly quick at least. If I read right, at least some of the concern with H2 has/is focussed on the mounting of ETMY - talk of whether the wires it hangs from are seated properly etc......

At L1 they have had the usual trains and seismic activity, more recently a storm running through, to upset lock. However power supply to the site is now an issue, the breakers are err .... breaking. They would like to get full use of cooling equipment and the water temperaure in the chiller is gradually rising. Be thankful for the high specific heat capacity of water.

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,054
Credit: 109,566,635
RAC: 43,131

Time for a closer look at the

Time for a closer look at the seismic records, their importance is vital to the project, as events in recent times, 'nuke' tests and earthquakes especially have shown. Both sites show them in the e-logs, more so for Livingstone. While both are affected by near and far earthquakes, Livingstone is more frequently so by human activities - logging and trains are often mentioned. These are sample plots for Hanford

and Livingstone

The horizontal axes are as defined here. The vertical axes are in units of micrometers per second. This is a velocity, denoting the speed of movement of elements within the siesmic detectors. The larger the velocity, the greater the disturbance. Although the measurements are labelled as seismic, there is no guarantee that 'earth movements' are responsible for any given output from each instrument. An elephant strolling past will trigger a change, say. Activity related to this week's efforts to diagnose and fix the errors in H2 ( Hanford 2km ) have registered and affected H1 ( Hanford 4km ) - as they are nearby/within the same enclosures. So the various locations of sensors around the LIGO's helps with identifying specific issues ( we hope! ).

If you stand at a corner station and face at 45 degrees from both arms ( ie. looking into the area the arms enclose ) then the arm on your right side is the 'X' arm, and the one your left is the 'Y' arm. Alternatively you could look down from on high and note the arrangement is denoted by a standard XY plot. Yet again, if you had an axis arising from the corner station going straight up ( away from the Earth ) and you called it 'Z' then with X/Y/Z you'd have a right handed Cartesian co-ordinate system.

Some helpful acronymns:

LVEA - Laser & Vacuum Equipment Area ( basically the corner station )
ITM - Input Test Mass ( ie. ITMX and ITMY )
ETM - End Test Mass ( ie. ETMX and ETMY )
PEM - Physics Environment Monitor
STS - Signal Track Search
SEI / SEIS - Seismic
H0 - Hanford ( generic )
H1 - Hanford 4km
H2 - Hanford 2km
L0 - Livingstone ( generic )
L1 - Livingstone 4km
Guralp - dunno, any suggestions?

So, say, 'H0:PEM_MX_SEISZ_10_30Hz' would indicate at Hanford, the Physics Environment Monitor at the Mid station in the X arm for Seismic monitoring in the Z direction ( up/down ) for the 10 to 30 Hz frequency range!

Now, say, 'L1:SEI_ETMY_STS2_Y_0p2_0p35Hz' would designate Livingstone Seismic monitor at End Test Mass on the Y arm Signal Track Search channel 2 in the Y direction ( ie. parallel with the beam axis in that arm ) from 0.2 to 0.35 Hz. 'p' is a place holder for the decimal point so it doesn't get lost in the graphics!

I think the 'Anthropogenic Noise' ( ie. man-made ) plot is similiar.

You get the idea ..... :-)

A micrometer is a millionth of a metre. This is one-thousandth of a millimeter, which is in turn a thousandth of a metre. So one micrometre per second is not real fast or even noticeable in everyday human terms.

NB. I've also found another terminology/acronym list here

L1 is climbing back up after the power problems. At Hanford, well .....

Quote:
H1 ....... A cooperative IFO on the whole

Quote:
H2 continues to be persnickety .....

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Chipper Q
Chipper Q
Joined: 20 Feb 05
Posts: 1,540
Credit: 708,571
RAC: 0
Chipper Q
Chipper Q
Joined: 20 Feb 05
Posts: 1,540
Credit: 708,571
RAC: 0

Some more background on the

Some more background on the Guralp seismometer, from this Technical Note (Geophysical Measurements Along the X Arm at LLO {6/26/01}):
Each LIGO site employs the Guralp model CMG 40T triaxial seismometer, used by the PEM system. One of the conclusions reached back in '01, when a linear array of these seismometers was used to check conditions in the X arm at LLO, is that the arm 'appears to act as a seismic wave guide'.

Also found more info on the 'micrometers per root Hertz' units. The following quote is from the Technical Note (link above, to see the figures mentioned):

Quote:
3 Spectral Analysis of the Seismometer Signals
It is interesting, for many studies of the interferometer’s sensitivity to seismic motion, to see what can be learned by looking at the measured data in the frequency domain. The dependence of the propagation velocity and attenuation length on frequency for each mode have been investigated using this data. The MATLAB routine specgram was used to make time versus frequency plots for each of the data series shown in figure 2. Specgram computes the FFT for a window of data which slides over the time series data. Using the output of specgram, the velocity data were converted to displacement power spectral densities and the log of the magnitude of the displacement spectral density in meters/√Hz is plotted in figures 5-9. To make each spectrogram, 256 element time records were successively Fourier analyzed using a Hanning window and an overlap of 252 points. The arrivals of the acoustic signals are easily observable as vertical bands in the spectrograms. The time axis in each figure is in seconds after the hour measured using the GPS antennae within each data recorder corrected for the clock slough between recorders.


Of course, many other things also introduce noise in the same range of frequencies. The wind is coupled both seismically and acoustically. (IIRC power in the wind goes up with the cube of the wind's velocity.) Jeepers, when the temperature changes, the building creaks (1 – 900 Hz, seismic below 20 Hz), and the dewars glitch (16 Hz). See Introduction to PEM, SciMon camp 2006

Quote:
Bear in mind I have no official status here, this is a private reverse engineering effort...


Also got lucky googling the Guralp, sort of feel like I found a version of the Gutenburg Bible for GW detection data analysis – I found an early version of the Users Manual (pdf file, ~600 pages). I've already learned answers from it, to questions I hadn't even thought of yet! I hope this helps the effort :)

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,054
Credit: 109,566,635
RAC: 43,131

Chipper, you are the man!!

Chipper, you are the man!! :-)
You get 'THis wEek's ModeratOr's Data Interrogatory Science Mining AwarD' ( THEMODISMAD ) ........ :-)
Terrific finds there, downloaded and straight to the good room!

I'll have a lot to digest, but already I quite like this from Geophysical Measurements Along the X Arm at LLO:

Quote:
A simple, but effective, impulsive seismic source was made using a 2 liter plastic soda pop bottle partially filled with liquid nitrogen. A steel weight was attached to the bottle with duct tape prior to filling to insure that the bottle would not float. Once a bottle was filled approximately one-third full of liquid nitrogen and the cap was tightly secured, it was thrown into the center of the erosion control pond immediately north of the LVEA. The water depth at this location is approximately eight feet ....... When the soda bottle ruptured at the bottom of the pond, the water above it served as a reaction mass, making it possible to couple a relatively large amount of energy into the soil as a result of the explosion. Also, because the gas explosion is in water, the initial propagation through water is via a compressional wave and vertical “gravity� water wave motion. Vertical and horizontal shear modes result from the coupling at the soil-water interface at the pond boundary.

Now that's my kind of experiment! ( NB. you can't get shear in a fluid )

Yes ...... it may well be the Gutenberg .... ooh I just gotta follow that one :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,054
Credit: 109,566,635
RAC: 43,131

Well H2 is now up and

Well H2 is now up and collecting science data!

Alas, I think my analysis of the time axis on the Figures Of Merit is wrong.

It looks like 'T0' is in fact the time at the left edge of the horizontal axis ( denoted by -12 )! Thus not the right edge....

You see, a Hanford entry at Fri Oct 20 07:09:54 2006 UTC:

Quote:
H2 in SM S5-01780 ~6hours--6.5Mpc.


implying H2 it went to science mode 'around' midnite of 19/10 going into 20/10.

If you look at the State Vector plot from Livingstone ( Fri Oct 20 05:02:21 2006 UTC ):

Then that puts the H2 ( in blue ) science mode transition ( from state 3 to 4 ) at 16:48:50 plus about seven hours, and NOT 16:48:50 minus about five hours.

Hence the '0' mark on the right edge is the time the screenshot was taken, but that is not the T0 as listed below the axis.

It seems weird, but does that sound right?

Reality check please..... :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.