--> Abstract: Sub-basaltic Imaging Using Converted Waves - A Pilot Study from the North Atlantic Margin, by X-Y. Li, P. Hanssen, and C. Macbeth; #90923 (1999)
[First Hit]

Datapages, Inc.Print this page

LI XIANG-YANG, PETER HANSSEN and COLIN MACBETH, British Geological Survey

Abstract: Sub-basaltic imaging using converted waves - A pilot study from the North Atlantic Margin

Introduction

Over recent years, hydrocarbon exploration activities have increased significantly on the North-East Atlantic Margin between northern Norway and western Ireland. A problem for exploration on the Atlantic Margin is the widespread occurrence of Early Tertiary basalt lavas. This high-velocity interval causes serious degradation of the seismic image of the deeper subsurface. There have been several attempts to address this problem with new acquisition methods such as long offset and wide-angle Previous HitrefractionNext Hit (Kiorboe and Petersen 1995).

Recent work has shown the potential of imaging structure beneath basalts using sub-basaltic converted-waves (SBCs) due to local conversions as demonstrated by Purnell (1992) and Li and MacBeth (1997) with laboratory data. This technique was also subsequently adapted for imaging structure inside salt bodies by Wang et al. (1994). Despite this, there appear to be few, if any, field and numerical modelling studies. Here we fill this gap using seismic data from the North Atlantic Margin.

Basic concept

When a P-wave is incident upon the top of a basalt layer with high P-wave velocity, most of the wave energy is Previous HitreflectedNext Hit. However, beyond a certain critical angle, an efficient conversion can occur from the incident P-wave to an S-wave travelling inside the basalt layer. This converted S-wave will in turn transmit through as the S-wave, or reconvert to the P-wave at the bottom of the basalt layer. When these waves reflect at an interface beneath the basalt, the conversion, or re-conversion may occur again (Figure 1a). These wavetrains due to local conversions containing S-wave segments within or below the basalt are referred to, generically, as sub-basaltic converted-waves (SBCs).These waves can also be named according to their raypaths as a PSPPSP, and PSSSSP wave, respectively. To verify the SBC arrivals, we calculate the effective reflection coefficients and full-wave synthetic shot records from the basic model in Figure 1a. This consists of 900m of water overlying 910m of sediments (sludge), and beneath 200m of basalt is underlain by two sands, each 800m thick.The SBC arrivals are possibly the only waves that have sufficient energy to penetrate the basalt (Figure 1b).

A real data example

A test line from the Atlantic Margin was selected for investigating the converted waves. A 6km cable was used for acquisition, and recorded for ten seconds with two millisecond sampling interval. We select 441 shot records from shot point 1071 to 1511 for test processing. The thickness of the sediments between the seabed and the basalt remains roughly constant with 900m from left (1500) to right (1100) with a gentle dip towards the right. The water depth increases from about 600m to 1300m in the area of interest.

Primary P-wave reflections from beneath the basalt can hardly be observed. The records are dominated by the seabed multiples and multiples from the top basalt (Figure 2a). Any method aiming to stack primary reflections at near offset is unlikely to be successful. However, at far offset, there is a converted wavetrain ahead of the primary reflection from the seabed. This wavetrain consists of several coherent arrivals with different moveouts. As the water depth increases from 600m to 1300m, this wavetrain gradually moves downwards, and is eventually obscured completely by the seabed reflections and associated multiples. This wavetrain contains the SBC arrivals as indicated by full-wave modelling.

Test processing and results

The key step in converted-wave processing is the determination of stacking velocity. Problems arise when attempting to pick stacking velocities beneath the basalt, which are not unique (Figure 2b), so that several velocity curves are possible. The choice of velocity is largely determined by the picking of stacking velocity for events from the bottom of the basalt. A sharp increase in stacking velocity is expected for the PP event from the bottom of the basalt. However, the stacking velocity for the PSSP event from the bottom of the basalt will remain almost the same as the velocity for the PP event from the top of the basalt as the interval shear-wave velocity of the basalt is similar to the P-wave velocity of surrounding rocks. Thus, the velocity maxima for the PP events from the top of the basalt, and for the PP and the PSSP events from the bottom of the basalt will form a triangular shape on the contour velocity spectra (Figure 2b); the separation of these three points will depend on the Vp/Vs ratio of the basalt. This feature can then help in velocity analysis for SBC arrivals.

Once the stacking velocity is determined, the rest of the processing is the same as for P-waves. The final stacking section for SBC arrivals beneath the basalt is shown in Figure 3a. Apart from the top of the basalt, the PP bottom of the basalt and the PSSP bottom of the basalt are also identified. The average thickness of the basalt is about 100ms in P-wave two-way time, and about 200ms in shear-wave two-way time, giving rise to about a thickness of 200m and Vp/Vs ratio of 2 for the basalt. Furthermore, continuous horizons can be identified beneath the basalt, indicating the presence of sediments beneath the basalt. In contrast, the original section for stacking the PP waves (Figure 3b) shows little information below the basalt and the basement was thought to be very near the basalt. A comparison has also been made for the velocity models used for stacking the test line. The velocity horizons show good agreement and correlation with the stacked sections.

Discussion and conclusions

We have carried out a quantitative study to examine the feasibility of using converted shear-waves for imaging structure beneath high velocity basalt layers. The field data examined show clear Previous HitreflectedTop and refracted arrivals which consist of raypaths with S-wave segments through the basalt. Numerical modelling confirms these observations. Processing tests based on the SBCs has improved the structural imaging below the basalt. The processing also indicates that converted wavefield extrapolation and separation, and multiple suppression, two-boat geometry for surface-to-surface shooting, or for surface-to-seabed shooting will be key issues for fully utilizing the SBCs at all offsets.

AAPG Search and Discovery Article #90923@1999 International Conference and Exhibition, Birmingham, England