Seismic Data Processing part 2 Flashcards
How is constant velocity stack done and what does it aim to achieve?
Computer stacks seismic traces in a CMP gather using a range of possible stacking velocities
The idea is that any particualr reflection will show the highest amplitude after stacking when NMO has been corrected
How does the velocity stack function differ from the constant velocity stack?
Velocity stack function estimates the trend of velocity against time and corrects the data with these values instead of computerised range of velocities
Why might muting be required?
- If F-K filtering is not sufficient to remove the effects of ground roll and refracted arrivals
- NMO stretching of the signal offsets is most exaggerated at shallow depths
These effects can be removed by muting
How is muting done?
Systematic zeroing of the early sample values on a trace with increasing muting zone with increasing offset
What is the advantage of CMP stacking? How are corrected and stacked used to create a seismic line?
Stacking will cause true reflections to be enhanced while random noise tends to be cancelled out
Corrected and stacked traces displayed side by side to create seismic section
Why does normal stacking generate multiples?
Different NMO behaviour from primary reflectors
How is a multiple stack produced?
- Stacking velocities of multipes found
- Multiple moveout correction then applied and gather is stacked to produce trace where multiples are strongly enhanced
- Trace then scaled and subtracted from CMP record
- Multiple moveout is removed and NMO for primary reflections is applied and re-stacked
How does deconvolution enhance the data?
Reduces blurring of reflected wavelets by recompressing them to the clean spike emitted from the source, result is clearer reflections and reduced multiples
How does migration enhance data?
Corrects distortions caused by plotting inclined reflectors as horizontally and vertically below the CMP, also collapses defractions
How does filtering enhance data?
Removes undesirable noise
How can we reproduce the Earth’s filter?
Cross-correlating the inverse of the signal with the stacked traces
Name and describe three types of deconvolution filters
Spiking: to shape the signal so that will then only contain sharp min phase wavelets
Predictive: remove multiples, especially short period reverberations
Wavelet shaping: convert min phase data to zero phase data
Why is wavelet shaping not always feasible?
Dependent on ability to synthesise min phase wavelet by mathematically modelling in frequency domain
What is the processing technique inversion and what should it give you? When can it be particularly useful?
Inversion is applied to zero phase data and involves integration to produce seismic impedance trace which represents acoustic properties of rocks
What can inversion be used for?
Predicting lateral variations in porosity/fluid content away from a well
