Gravity waves interfering with each other
15 Jul 2005 11:04:08 UTC
Topic 189547
(moderation:
Hi
I have been reading up about gravity waves and something has started bothering me. Perhaps it is because I do not fully understand the measurement process but here goes.
My understanding is that gravity waves distort space time in much the same way that you can imagine a pebble thrown into a smooth pond disturbing the surface. Now if you throw a pebble into the smooth pond then you should be able to easily measure the regular waves using a LIGO style configuration of sensors measuring the relative positional shifts of the surface at perpendicular points on the surface. However, if you throw many different sized pebbles into the pond at different points and at slightly different times then the surface distortion becomes irregular and more like "noise". An instrument configured like LIGO will then be unlikely to detect any regular pattern on the surface and the results will simply appear as noise.
Now my concern is that the distortions by gravity waves are very small (the size of an atom relative to the distance between the earth and the sun). Now wouldn't all the gravity wave sources together in the universe produce an effect in space time similar to the many pebbles in the pond analogy making it almost impossible for LIGO to distinguish the real distortions in space time from noise?
I understand that you can "focus the view into the universe, but this would be like focussing the pond instrument in the direction of one pebble - you will not be able to distinguish the slight additional surface distortion from all the other interfering surface waves.
Am I missing something important? Can anyone help clear this up for me?
Regards
Gravity waves interfering with each other
)
I wondered the same thing – should be lots of chop.
Missing things like:
1)All-Sky Survey,
2)Filtering noise, then performing FFTs and pattern matching techniques on signal data,
3)Search restricted to limited bandwidth by design of detector (different frequencies for different detectors),
4)Search refined by limiting types of sources considered (to binary systems with lots of mass),
5)?? (I'm still learning, too)
Hi Hugh. You can think of
)
Hi Hugh.
You can think of gravity waves much like radio waves. Where a radio transmitter will radiate in the electromagnetic energy band, something like a pulsar radiates gravity waves with "gravitic energy".
Gravity waves therefore have a frequency. A gravity wave detector like LIGO receives noise. The noise is mostly non gravitic in nature and is gradually being filtered out by various methods. The noise that is left is maybe thousands of sources of modulated gravity all appearing at the LIGO detector in the audio spectrum from a few hertz up to maybe 1-2 Kilohertz in the case of rapid motion sources.
By using the einstein@home program, the noise "data" we all are currently crunching is filtered mathematically with the FFT analysis so that time domain becomes frequency domain. Much like your shortwave receiver would allow you to tune into a particular frequency.
As a result any strong gravity wave frequencies appear as a "spike" when the data is viewed. Also, by correlating phase differences between LIGO and other GW sites, a location of the GW source can be determined.
What I have written is from reading other peoples comments and watching some of the tutorials so I could be wrong on some points.
I think there is a
)
I think there is a distinction made between gravity waves and gravitational waves. The former refers to waves (caused by gravity) on the surface of a body of fluid (a pond, the ocean, a star), while the latter refers to what Einstein@Home is working on.
Hi Revolution Thanks for
)
Hi Revolution
Thanks for this. The part that bothers me is that the measurements are based on displacement of the masses and the resultant interference of other sources may drive the actual displacement patterns below a measurable threshold distinguishable from noise, irrespective of the frequency separation. In the case of electromagnetic radiation it is easy to detect only signals from a particular direction. If we can do directional measurement on gravity waves then all is good. If not then I am still concerned.
Regards
Hugh your going to give to
)
Hugh your going to give to folks at LIGO nightmares, if they don't already have them worring about the points you raise.
Each detector is moderately directional by it self.The strain depends on the direction the signal comes from and it polorization.Combining results from multiple detectors the allows for high direction selectivity.
There is a presumption that there are some "close" sources that are strong enough to be isolated from the background.
Hugh, The situation you
)
Hugh,
The situation you are talking about with random-looking "chop" coming from multiple sources of waves (as opposed to mundane noise from earthquakes, electronics, etc.) is known as a "stochastic background."
There is actually a separate search devoted to that. (It's not on Einstein@Home because the computational demands aren't too high.) We just submitted a technical paper on the S3 data.
A stochastic background doesn't affect searches for individual sources unless it's much stronger than they are. In the LIGO frequency band, there will be plenty of things (like the closest pulsar, wherever it is) above the stochastic background. The directionality of the pulsar data analysis helps, but it's mainly that a stochastic background will be dominated by a large number of sources (pulsars or whatever) that are far away and thus weak. The reason you can see a lot of chop on a pond is that the waves are not dying off very fast.
LISA people, however, do worry about digging sources out of a very strong stochastic background. Basically, if you're looking at about an hour period (as opposed to milliseconds like LIGO), there are a lot of white dwarf binaries in the galaxy that will combine to produce a stochastic background that shoots far above the instrument noise curve.
Hope this helps,
Ben
RE: We just submitted a
)
Thank you, Ben!
Looks like excellent work and good progress.
Is the spectrum of the energy density of a stochastic background of GW's expected to have a Gaussian distribution, and if so, which potential source models predict which frequency bands that the peak will occur at? Or is it expected to be of uniform distribution in all bands? What is meant by the unlikelihood of the spectrum being “thermal”? Or does it depend entirely on the proximity of different types of sources to the different types of detectors?
I guess what I'm trying to figure out is, why are there such large differences in the respective values for Omega_0 between LIGO, spacecraft Doppler tracking, and radio pulsar timing (less than 44, less than 0.027, and less than 10^7)?
RE: Chipper Q wrote: I
)
The amplitude of the strain noise changes tremendously from one frequency band to the next, i.e. the spectrum is very far from "white" or frequency-independent. LIGO is looking at by far the highest frequencies.
Different source models predict different spectra (signal amplitude as a function of frequency), but so far they're all well below the observational results in each band. LIGO is making progress faster than the other things, though.
Hope this helps,
Ben
RE: RE: I guess what I'm
)
I thought I was familiar with most of the source models for LIGO, until reading the paper you mentioned. After noting the inclusion of variables like H_0, the only thing more amazing to me than how many bases there are to cover, is how well they're being covered.
As I continue to learn “basics” of GW detection, I'm eager to check out source models for the Doppler tracking and pulsar timing methods, as time permits. Your remarks give me a better idea of what to expect, and it's great to hear about the progress of LIGO. Thanks Ben!