Ghost Reflections
Ghost Reflections
One of the commonest form of undesirable ray associated with marine seismic acquisition is the ghost reflection. The sea surface is an almost perfect reflector. Energy traveling from the shot (under the water) is traveling through a medium with a velocity of about 1500 meters per second, and a density close to 1.025 grams/cc. The air above the sea surface has an acoustic velocity of about 350 meters per second, with a density of about 0.0013 grams/cc. Plugging these into the "RC" equation gives a reflection coefficient from below the sea surface of about -0.9994, or almost perfect reflection (with sign reversal).
Here's the ray paths showing the possible routes for ghost reflectors:
The energy travels from the shot, up to the surface, reflects almost perfectly (but sign reversed) and then travels a similar path to the direct ray from the shot to the reflector.
The ray returning from the reflector does exactly the same thing at the receiver. There can be four different rays produced:-
1- Direct ray.
2- Ghosted only at the shot.
3- Ghosted only at the receiver.
4- Ghosted at both shot & receiver.
Given that the actual ray paths will have more curvature than those shown in the diagram, the ray paths from the shot to the surface and back down again will be almost vertical. If we assume that they are vertical, we can calculate the time difference (in milliseconds) between the direct ray and the ghosted ray as approximately 4/3 times the depth of the shot in meters (we're assuming a water velocity of 1500 m/s). The same calculation can be done for the receiver.
If, for example, our shot depth is 6 meters, then the time difference between the direct and ghosted rays will be about 8 milliseconds. If we assume that the shot energy consists of a set of constant frequency cosine waves added together (we'll see later that this is a valid assumption), we can consider the effect of the ghost on the individual frequency components.
You can see that the component frequencies of 40 and 80 Hz are actually increased by the ghost, whereas the 125 Hz component (in this case) is completely removed! The direct ray and the ghost are exactly 180 degrees out of phase, and one removes the other.
In general terms, a ghost will cause "notches" in the frequency spectrum at frequencies that are a multiple of (roughly) 750/(Shot Depth) and 750/(Receiver Depth) Hz.
In practice, the Shot and Receiver depths are usually chosen so that these frequency notches are beyond the range of frequencies required for the seismic data.
Typically, depths of less than 10 meters are used, making the first notch > 75 Hz.
Here's the ray paths showing the possible routes for ghost reflectors:
The energy travels from the shot, up to the surface, reflects almost perfectly (but sign reversed) and then travels a similar path to the direct ray from the shot to the reflector.
The ray returning from the reflector does exactly the same thing at the receiver. There can be four different rays produced:-
1- Direct ray.
2- Ghosted only at the shot.
3- Ghosted only at the receiver.
4- Ghosted at both shot & receiver.
Given that the actual ray paths will have more curvature than those shown in the diagram, the ray paths from the shot to the surface and back down again will be almost vertical. If we assume that they are vertical, we can calculate the time difference (in milliseconds) between the direct ray and the ghosted ray as approximately 4/3 times the depth of the shot in meters (we're assuming a water velocity of 1500 m/s). The same calculation can be done for the receiver.
If, for example, our shot depth is 6 meters, then the time difference between the direct and ghosted rays will be about 8 milliseconds. If we assume that the shot energy consists of a set of constant frequency cosine waves added together (we'll see later that this is a valid assumption), we can consider the effect of the ghost on the individual frequency components.
You can see that the component frequencies of 40 and 80 Hz are actually increased by the ghost, whereas the 125 Hz component (in this case) is completely removed! The direct ray and the ghost are exactly 180 degrees out of phase, and one removes the other.
In general terms, a ghost will cause "notches" in the frequency spectrum at frequencies that are a multiple of (roughly) 750/(Shot Depth) and 750/(Receiver Depth) Hz.
In practice, the Shot and Receiver depths are usually chosen so that these frequency notches are beyond the range of frequencies required for the seismic data.
Typically, depths of less than 10 meters are used, making the first notch > 75 Hz.
Re: Ghost Reflections
Never knew this. Does this apply to boomer or sparker systems?
Re: Ghost Reflections
The depths for the boomers and sparkers are usually shallow ~1m enough and the main range of the frequencies should be fine.
Re: Ghost Reflections
Is there any chance you could expand on these ghost 'notches' in the frequency spectrum. Is it because like the 125hz diagram you used, they are out of phase by 180 degrees and then cancel when added together. As such this reveals itself as a 'notch'.
Re: Ghost Reflections
That's pretty much the case - its constructive and destructive interference between the upgoing and downgoing wavefield at the source and receiver end.BarryR wrote:Is there any chance you could expand on these ghost 'notches' in the frequency spectrum. Is it because like the 125hz diagram you used, they are out of phase by 180 degrees and then cancel when added together. As such this reveals itself as a 'notch'.
You have the same effect in FM radio transmissions with two transmission sources- "beat frequencies" - or to some extent noise cancelling headphones.
There's similar effects in terms of path-length and spatial resolution too - Fresnel Zones are related to ray-path lengths and signals becoming out-of-phase.