Exploration Geophysics - Volume 20, Issue 1-2, 1989

Multiple attenuation techniques have to be based on some difference between the multiples and the primary reflections. The two major differences that are exploited are firstly velocity, and secondly the fact that multiples are periodic sequences of events, and hence are predictable, while the primaries are non-periodic. The widely used frequency-wavenumber (F-K) domain techniques rely on velocity difference only, but a recent variation of this method also makes use of any difference in dip between primaries and multiples to give significantly greater multiple attenuation. For short period multiples, velocity differences may be insufficient for much attenuation, and the process has to be based on the multiple's periodicity, using some type of long predictive deconvolution operator. One problem with this approach is that the multiple period varies with time, particularly at long offsets. Transforming the record to the tau-p domain removes this variation however, allowing more effective deconvolution of the multiples. Another recent approach is to model a multiples-only record by wave equation methods, and subtract it from the recorded data. At present however, this is limited to well defined multiple generators, such as the water layer. With the variety of multiple attenuation processes available today, the geophysicist needs to understand the types of multiple problem to which each is most suited, in order to select the technique most applicable to his data.

Constant velocity FK domain migration, as developed by Stolt in 1978, has significant performance advantages over all other commercially available migration algorithms.

However, even with Stolt’s pre-migration time stretching (or psuedo-depth conversion), this algorithm suffers serious accuracy problems in the presence of any significant velocity variations. This paper develops a method of extending the accuracy of FK domain migration to steep dips in arbitrary velocity fields.

The basis of the new method is the partitioning of the unmigrated data set into dip ranges and computing unique time stretching factors for each. Stolt’s original time stretching is shown to be the zero-dip special case of this more general algorithm. Following the separate migration of the partitions, the data is merged to form a correctly migrated data set. As few as four partitions is shown to produce very significant improvements.

Even though Extended FK Domain Migration is a few times slower than conventional FK domain migration, it can still produce accuracy comparable to phase shift migration at a fraction of the cost.

Late Cenozoic deformation along the Australian/Pacific plate boundary is seen in onshore New Zealand as zones characterised by extension- or transcurrent- or contraction-related structures. High-resolution multichannel seismic reflection data were acquired in several of these tectonic zones and successfully reveal the shallow structures within them.

Thirty kilometres of dynamite reflection data in the Rangitaiki Plains, eastern Bay of Plenty, define a series of NE-trending normal faults within this extensional back-arc volcanic region. The data cross surface ruptures activated during the 1987 Edgecumbe earthquake.

In the southern North Island, a 20 km Mini-Sosie^TM seismic profile details the Quaternary sedimentation history and reveals the structure of the active strike-slip and thrust fault systems that form the western and eastern edges of the Wairarapa basin, respectively. This basin is considered to sit astride the boundary between a zone of distributed strike-slip faults and an active accretionary prism.

In the Nelson area, northwestern South Island, previously unrecognised low-angle thrust faults of Neogene or Quaternary age are seen from Mini-Sosie data to occur at very shallow depths. Crustal shortening here was previously thought to arise from movement on high-angle reverse faults, and the identification of these low-angle faults has prompted a reassessment of that model.

A grid of 18 km of Mini-Sosie seismic data from the central eastern South Island delineates Neogene or Quaternary thrust faults in Cenozoic sediments. The thrusts are interpreted as reactivated Early Eocene normal faults, and the thrust fault geometry is dominated by these older structures.

Seismic refraction data were obtained for the Bass and Gippsland Basins during the 1988 cruise of the BMR research vessell“Rig Seismic”. Seismic recorders were deployed on land by BMR and Monash University to record long-offset wide-angle reflection and refraction data using the ship’s air-guns as the energy source. Preliminary results have now been obtained from these data providing information on deep crustal structure related to the basin formation. Two crustal layers have been detected with velocities of 4.5 km/s increasing to 7.4 km/s (unreversed) at depths exceeding 20 km.

Additional data have now been obtained over a traverse length of 170 km to provide constraints on the deep structure of Bass Strait and the Lachlan Fold Belt in Victoria and Tasmania.

Key words:

Before oil volumes and economics can be calculated for an oil field, the seismic time map must be converted to a depth map. The Skua Field, located in Permit AC/P2 in the Timor Sea, has proved particularly difficult to depth convert. Velocity anomalies and inconsistencies in the seismic times, termed ‘lags’, have created distortions in the seismic time map which require compensation.

Beneath a velocity anomaly, both seismic undershoot and increased velocity, which are difficult to determine, must be compensated for during depth conversion. The current depth map was produced by smoothing through the pull-up regions on the time map, which required judgement, then depth, converting using a regional average velocity field.

The seismic lag, which is the difference between the seismic time and an ideal vertical path travel time, can only be measured at the wells and appears to be unpredictable. The seismic lag between Skua-4 and Skua-5 was assumed to change linearly in order to produce the depth map. Large lags can be introduced into the data in the common depth point stack stage of data processing. The stacking velocity with the largest stack response does not necessarily result in the smallest lag error.

Key words:.

The Puffin Field is located within the Vulcan Sub-basin of the Timor sea, off the Northwest Coast of Australia. It lies within the offshore exploration permit AC/P2, operated by BHP Petroleum and its co-venturers. It is situated on the Ashmore Platform, an old Triassic horst which is normal faulted against the Swan Graben, a major Mesozoic depocentre and the regional source area.

Three wells were drilled in the 1970’s. Puffin-1 and Puffin-3 encountered oil in ‘FIT’ tests from within the Maastrichtian 100 ft sand, and Puffin-2 flowed over 4000 barrels of oil per day from a slightly younger 4 m sand.

On examination of the results of the Puffin wells, it was evident that there were severe velocity anomalies and differing oil water contacts in the Puffin field. The top of the 100 ft reservoir sand is at 2031.4 m subsea in Puffin-1, 2045 m subsea at Puffin-2 and 2074 m subsea at Puffin-3. The two way times to these events were 1392 ms, 1328 ms and 1398 ms respectively. The interpreted oil water contacts in Puffin-1 and Puffin-3 were 2033 and 2077 ms subsea respectively with no contact seen at Puffin-2.

In an attempt to resolve these anomalies the AC/P2 joint venture undertook a detailed seismic reprocessing project of the 1980 data with special emphasis on detailed velocity analysis. This 1987 reprocessing effort involved two passes of velocity filtering and velocity analysis at every 600 m. Velocity analyses were picked on a horizon-consistent basis, such that variations in interval velocity for key horizons could be established for later use in depth conversion. Although sceptical in using stacking functions as the input velocities to depth conversion, they were used, as no viable alternative was feasible.

Data quality was reliable to the top of the Palaeocene Calcilutite, and six horizons were picked with their respective velocities to this level. Analysis of the data indicated that the two major units exhibiting interval velocity variation were the Pliocene ‘low velocity layer’ and the Eocene carbonates. Using the smoothed stacking velocity down to the Top Palaeocene Calcilutite the three wells tied the depth conversion with an accuracy of 0.5%. Below this horizon two constant interval velocities were used from well data as the quality of the seismic pick were not as reliable.

To verify this model BHPP also undertook a ‘layer-cake’ velocity approach which, although confirming the anomalous zones, could not be used laterally away from the three wells, which unfortunately all lay in a straight line.

Two wells, Puffin-4 and Parry-1 were drilled in 1988 to test the resultant interpretation. The wells intersected the Top Palaeocene Calcilutite within 1% of prognosis at Puffin-4 and within 2.2% of prognosis at Parry-1, therefore confirming the stacking velocity model used in depth conversion. However, both wells came in deep to prognosis at the deeper, objective level as a result, in the case of Puffin-4, of being on the downthrown side of a small fault, and at Parry-1 due to a thickening of the Paleocene section and seismic mispicking of the Top Palaeocene Calcilutite. Had the mispick at Parry-1 been avoided then the tie would have been less than 1.0%. Both these mis-interpretations were made in the part of the section where the quality of seismic was poorest. These two results suggest that even though the depth conversion to the Top Paleocene Calcilutite is accurate to within 1%, the magnitude of the velocity variation is larger than the magnitude of the independent depth closure.

The Puffin Field requires both better quality seismic below the Base Palaeocene Calcilutite, or the means to resolve the lateral extent and possible thickness of a 4 m sand away from Puffin-2.

Until such a method of obtaining either better quality seismic to the objective level, or to be able to define the seismic resolution of the differing sand bodies of a minimum size of 4 m, the Puffin Field will remain a Geophysical enigma.

The Carpentaria Basin in the west/central portion of Cape York Peninsula is largely unexplored for petroleum, and there is an apparent ambiguity in the basement depths interpreted from gravity and aeromagnetic data. It was decided that deep seismic refraction surveys at a variety of sites should prove cost-effective in defining the geologic model for the basin. Of particular interest is the possible existence of a north-south trending elongate infrabasin inferred qualitatively from a strong gravity low shown Figure 1.

Results of the refraction work indicate that the magnetic and gravity data suggestive of the presence of an infrabasin are probably related to lithological variations within basement. Furthermore, it is improbable that the thickness of the sedimentary pile anywhere within the area of investigation exceeds 1100 metres.

Basement velocities are high, from 5500m/sec to 6200m/sec, typical of fresh igneous and/or metamorphic lithologies. Carbonates could not be totally excluded on the basis of these velocities alone, but are improbable in view of the gravity and magnetic data.

At some locations there is evidence for the presence of an intermediate section of higher velocity within the sedimentary sequence. This is thought to be quite thin, and possibly representative of the Toolebuc Formation.

Structure contour maps compiled and synthesised from interpretive company data for top of Strzelecki Group/Basement, intra-Latrobe Group, and top of Latrobe Group seismic horizons give an indication of the formation and evolution of the Gippsland Basin during Early Cretaceous, and Late Cretaceous to Tertiary tectonic events. Comparison of the intra- and top of Latrobe Group mapping illustrates a change from a largely extensional to a compressive tectonic regime during the Eocene.

One of the Victorian Government’s policies in the oil and gas area is to enhance the benefits to the State in the energy sector by assessing the nature and extent of the petroleum resources.

To evaluate the production capacity of developed and undeveloped gas fields, a comprehensive study was commissioned by the DITR in 1988.

The first step in a study of this type is to check the accuracy of the depth maps, to see if they adequately describe the reservoir geometry.

Raypath modelling, using the Advanced Interpretation Mapping System (AIMS – Version III), was carried out by Geophysical Services International (GSI), Sydney, on a number of selected profiles over the Snapper, Marlin-Turrum, Barracouta, Kipper and Emperor fields for the DITR. Input data for the models were extracted from the operator’s maps. The software simulates the normal incidence raypaths (or wave theory solution) for all shotpoints, and from this information it generates gather records and/or synthetic seismic profiles. By comparing the model data with those from data acquisition, processing and interpretation, it was possible to check the validity of the interpretation of the reservoir’s geometry.

This modelling work showed that the synthetic data were comparable with the acquisition and processing data, confirming that the depth maps (tied to well control) produced by the operator using its proprietary software are adequate and most likely to represent subsurface configuration of the reservoirs.

Artificial Intelligence (Al) systems have been used with some success in the areas of dipmeter interpretation, quantitative log interpretation and well-to-well correlation. A prototype expert system has been developed using a rule-based approach to lithology identification. Extensions of the system are now being considered to do mineral identification for the problem of mineral model construction for multi-mineral log interpretation algorithms.

Magnetotelluric data obtained in the Eromanga Basin can be interpreted using one-dimensional models to describe plane layers consistent with geological mapping. Interpretations are based on the results of non-linear inversions generating a minimum least- squares error between the observations and the model. However there is no statistical justification for selecting highly complex starting models. In particular adequate solutions can be generated using models based on 2, 3 or 4 layers over basement; additional layers defining fine structure can only be retained using external geological constraints. Solutions based on random starting models suggest gradations in resistivity (1-4 ohm m) consistent with sediment compaction. Major discontinuities in all models (4-30 ohm m) are assumed to indicate a basement contact at depths of 7-8 km.

A survey was carried out over the Tallyabra anticline in the Eromanga Basin Queensland, using both controlled source and passive magnetotelluric techniques to investigate the structure in the top 2 km. The results from the controlled source system were very encouraging and, using the resistivity log from the well Kenmore 7 as a control, the top of the oil bearing Hutton Sandstone was resolved at a depth of about 1400 m. The magnetotelluric results were consistent with those from the controlled source, but the data accuracy was so much lower that in these circumstances they added little information.

Vertical Seismic Profiling (VSP) is often used to provide high resolution seismic images near a wellbore. A new borehole seismic technique, the TOMEX® survey (Rector, et al., 1988), uses the vibrations produced by a drill bit as a downhole seismic energy source to produce inverse VSP data. No downhole instrumentation is required to acquire the data, and the data recording does not interfere with or delay the drilling process. Hence, there is no loss of rig time in performing the survey. These characteristics offer a method to acquire SWD (seismic-while-drilling) borehole seismic surveys. In addition, 3-D imaging around a well can be obtained at significant savings compared to conventional offset VSP imaging.

The continuous signals generated by the bit during drilling are monitored with a reference sensor attached to the top of the drillstring, and the reference sensor signals are crosscorrelated with signals from surface-positioned geophones to produce inverse VSP data. Deconvolution and time shifts are then performed to remove the effects of recording the source reference trace at a location that is a considerable distance from the source. Results from tests demonstrate that these processed drill-bit source data are virtually identical to conventional forward VSP data.

In using the drill bit as a downhole seismic source for inverse VSP, many of the limitations of conventional VSP are overcome. Several applications for VSP that were previously considered to be prohibitively expensive are now feasible. Furthermore, this seismic-while-drilling technique offers the potential for the explorationist to make real-time drilling decisions at the well site.