next up previous [pdf]

Next: A real CMP gather Up: T-X-Y DOMAIN PREDICTION FILTERING Previous: A simple synthetic CMP

A more complicated synthetic CMP gather

This synthetic gather is generated by the Thompson-Haskell method, which has been used as an example in Lumley et al. (1994). The gather includes both primary and multiple events. Velocity stacking approach can remove the far offset multiples successfully, but it is not so successful in near offset, since the multiple event is nearly horizontal in the near offset. I show that PEF based approach can handle the near-offset multiples effectively.

Figure 4 illustrates that most of the multiple energy has been well-separated from primary energy, from near-offset to far-offset. I also extract three traces from the near offset and compare the difference between the two approaches, as shown in Figure 5. It is very clear that the PEF scheme has removed most of the multiple energy in the near offset. The signal-to-noise ratio of PEF based approach higher than the other two figures.

hm-nmo-tx
hm-nmo-tx
Figure 4.
A synthetic CMP gather generated by Haskell-Thompson method. After applying the 3-D PEF, both multiple events and random noise have been removed out of the input CMP gather.
[pdf] [png] [scons]

hm-near-offset-tx
hm-near-offset-tx
Figure 5.
Comparison of the PEF scheme and velocity-stacking scheme in the near offset. Three traces are extracted from the near offset. The left one is the original input, the middle one is the PEF result, and the right one is the result of velocity stacking approach. It is very easy to identify the primary events in the middle one.
[pdf] [png] [scons]


next up previous [pdf]

Next: A real CMP gather Up: T-X-Y DOMAIN PREDICTION FILTERING Previous: A simple synthetic CMP

2015-03-27