Seismic Resolution

Slide 1

• Slide introduces topic: Seismic Resolution
• This shows a simple sediment wedge model and its seismic expression – we’ll talk about it in this lecture

Slide 2

• We need to discuss two components of seismic resolution:
• Vertical resolution
• Lateral resolution
  
  

Slide 3

• Here is an analogy that we can all relate to:
• You are driving at night
• You spot a light in the distance coming towards you
• You wonder, I seem to see only 1 light; is it a car or a motorcycle
• As the vehicle gets never, we realize it is not a single light but two headlamps – so it is a car

• You first detected some light and know there was a vehicle
• It was not until the vehicle was closer that we were able to resolve two headlights and realize it was a car
• This analogy helps explain the difference between
• Detecting something with seismic data, and
• Resolving two closely-spaced objects

Slide 4

• Detection is the ability to identify that some feature exists
• Resolution is the ability to distinguish two features from one another

Slide 5

• As an example of vertical resolution, consider the geology indicated by the gamma ray log
• At a gross scale, there is a thick shale unit on top of a thick sand unit
• But the sand unit has a thin shale layer interfingered with it near the top
• Low resolution seismic data would detect a shaley unit sitting on top of a sandy unit - one interface
• Seismic data with high resolution would resolve 3 interfaces, identifying the thin shale unit within the predominantly sandy unit

Slide 6

• To further explain vertical resolution, let’s begin by considering a thick sand (unit B) sandwiched between shales (units A and C)
• The RC at the top and base of the sand are shown along with the individual wavelets
• Note the pulse duration is less than the thickness of the sand unit
• The wavelet associated with the upper RC is fully represented (going down) before the wavelet associated with the lower RC starts
• There is no wavelet interference
• A thick bed is one in which the bed thickness in units of two-way time is greater than the pulse duration

Slide 7

• Here the thickness of unit B has been decreased to 0.9 times the pulse duration
• The wavelet associated with the upper RC does not complete (going down) before the wavelet associated with the lower RC starts
• There is some wavelet interference – the end of the “upper” wavelet overlaps the top of the “lower” wavelet
• An interpreter still would be able to distinguish two RCs, but the trough is a “doublet”

Slide 8

• On this slide, the thickness of unit B has been decreased to 1/2 the pulse duration
• The second part of the wavelet associated with the upper RC overlaps with the first half of the wavelet associated with the lower RC
• Wavelet interference is at a maximum
• The trough is larger by about a factor of two than if there was only one RC
• It is more difficult for an interpreter to distinguish two RCs

Slide 9

• To determine seismic resolution, there are two parameters we need to know or estimate
• The velocity in the zone we are interested in
• The peak frequency of the pulse in the zone of interest
• We need to calculate the wavelength of the data
• Vertical resolution is ¼ the wavelength
• The calculation is shown in the center of the slide
• We get the period from 1/peak frequency
• We then get the wavelength by multiplying the period by the velocity

• If you prefer, wavelength = velocity / peak frequency (simple substitution)
• Next we divide the calculated wavelength by 4 to get the vertical resolution

Slide 10

• Time for an exercise
• You will calculate the vertical frequency for:
• A shallow zone
• A deep zone

• The next slide has the ANSWER
• Have the students do the exercise before proceeding

Slide 11

• ANSWER
• The shallow zone of interest has a wavelength of 40 meters; a vertical resolution of 10 meters
• The deep zone of interest has a wavelength of 150 meters; a vertical resolution of 37 meters

Slide 12

• To summarize our discussion of vertical resolution:
• Resolution is the ability...
• Thin bed response...
• Short-duration...
• To improve...

Slide 13

• What do we mean by lateral resolution?
• It means how wide does a feature have to be for us to correctly resolve it
• For example, in the upper diagram, there is a narrow horst block in the center
• If this horst is only 10 meters wide, we probably would not resolve the two edges.
• If it was 2 km wide, we would not have any problem resolving the horst
• What is the minimum width for which we could resolve both edges?
• This is why we want to know the lateral resolution of the seismic data
• In the lower diagram, we have three channel deposits of different widths
• Would we resolve all three; or only the widest one
• Again, this is why we want to know the lateral resolution of the seismic data

Slide 14

• Here is a ‘classic’ seismic model presented by Neidell & Poggiaglioimi, 1977
• In the model there is a reflector (upper black line) that has gaps in it of varying width
• On the next slide, we will explain what a Fresnel zone (FZ) is; for now
• Accept that the first gap = 2x the FZ
• The second gap = 1x the FZ
• Etc.
• The lower part of the figure shows the modeled seismic response (unmigrated)
• Looking at the modeled seismic, we would:
• Recognize the first gap
• Probably recognize the second gap
• Would wonder if the third gap is a break in the reflector
• And probably not recognize any break for the fourth gap
• Remember, the model is 'noise-free'

Slide 15

• As promised, we will now explain what a Fresnel zone (FZ) is
• The seismic waves “illuminate” an area of a subsurface boundary – like the cone of light from a flashlight shining on a carpet
• All the information within this “illuminated” area is “lumped together” or averaged
• The size of this “illumination” circle equals the area in which the seismic wave is ¼ the wavelength of the pulse
• The diameter of this circle is called the FZ
• Shallow in the data the FZ is narrow; it gets progressively broader as we go deeper
• Using our flashlight analogy:
• If our flash light is close to the carpet, the circle of light is small
• If our flash light is far from the carpet, the circle of light is large

Slide 16

• Fortunately for us, the data processing step called migration:
• Not only better positions the reflections in 3D space, but
• Also greatly improves lateral resolution
• This slide shows a reflection indicating a strong decrease in impedance (zero phase central trough) on the left and a abrupt change to a moderate increase in impedance (zero phase central peak) on the right
• The ideal response is in the upper figure
• The real-world response is shown in the central figure – a stacked section without migration
• The bottom shows what happens when seismic migration is applied to the data in the central figure
• Note how the abrupt change in the center is “smeared” in the central figure
• The FZ for this example is on the order of 800 m (red arrow)
• Also note how the migration process has “cleaned up” the image and the abrupt change is much better imaged

Slide 17

• Here is a seismic line with two types of migration:
• On the left a standard (fast,cheap) migration algorithm was used
• On the right, a more sophisticated (more time, money, people-hours) algorithm was used
• Note the fault on each image
• The termination of reflections are much sharper on the right; the fault can be more precisely drawn
• On the left the reflection terminations are more “smeared” since the lateral resolution is much lower

Slide 18

• Here are the equations that we use to calculate the Fresnel diameter
• The equation on the left is for data that have not been migrated
• The parameters are the average velocity down to the zone of interest, the time down to the zone of interest, and the frequency at the zone of interest
• The equation on the right is for data that has had a seismic migration process applied to it
• The parameters are the wavelength of the pulse at the zone of interest; or by substitution the average velocity and the frequency

Slide 19

• Let’s do another exercise
• You will be given the necessary parameters for:
• A shallow zone
• A deep zone

• The ANSWER is on the next slide
• Give the students some time to work the exercise

Slide 20

• ANSWER to the exercise
• For the shallow zone – pre-migration, the FD is 282 m; after migration it is reduced to 10 meters – what an improvement
• For the deep zone – pre-migration, the FD is 1900 m – almost 2 km; after migration it is reduced to 48 meters – another substantial improvement

Slide 21

• This shows the area over which the seismic “smears” the geologic information from our last exercise
• Note the 1 km scale bar
• The small green circle in the upper left is the FD for the shallow zone before migration
• There is a white circle in the center which is the FD after migration
• The large circle on the right is the FD for the deep zone
• The white circle in the center is the FD after migration
• Even if the seismic reflections are fairly flat lying (horizontal), this shows the benefit of migrating the data – even though the reflctions are not repositioned very much since dips are very low

Slide 22

• In summary for lateral resolution:
• Migration...
• Large aperture...
• Fine...
• Prestack...
• Depth migration...