Jacques Merland
Cie. Generale de Geophysique
Massy, France
Three-dimensional (3D) seismic has without doubt been the most significant development in geophysical exploration over recent years. Its advantage over traditional two-dimensional (2D) seismic is that it provides a clearer, continuous image of the subsurface, and thus pinpoints structural and stratigraphic features.
Contrary to popular belief, 3D seismic is also more cost-effective than 2D seismic. Indeed, for each shot or vibrator point, 3D seismic uses all the energy reflected within a given volume, limited only by the maximum acceptable distance (offsets) for emergent reflections, whereas 2D seismic utilizes only the energy propagated along a thin section of this block, (Fig. 1). As a result, modern 3D technology makes it possible to record 3D seismic surveys with comparable time and cost to a 2D seismic survey.
3D operations must satisfy two major requirements. First, controlled recording of a sufficient number of simultaneous channels must be performed at each energy source point, in order to provide continuous sampling within the 3D volume, while respecting economic and time constraints. Second, both seismic and supporting data (such as topographic co-ordinates, recording geometry and field statistics) must be acquired and managed simultaneously in order to arrive at a coherent data set. This thorough integration of field operations is achieved today by means of field management systems, which run a central database on field workstations (many systems operate, for example, on Microvax computers). These systems monitor, control and continuously update the flow of supporting information which accompanies seismic data acquisition, (Fig. 2).
CREATION OF THE CENTRAL DATABASE
The central database is created during the planning phase of a 3D survey by integrating surveying and geodetic parameters. A planned receiver and source point grid is generated by coverage simulation-certain systems have the capacity to perform simulations over 200,000 "bins" (elementary subsurface cells) in one pass. All acquisition parameters and control functions required for field operations and quality control are also fed into the database, which is loaded into the field computer at the start of operations.
GRID LAYOUT AND SURVEYING
Inertial positioning systems, (Fig. 3), and differential satellite positioning systems make simultaneous layout and surveying of land seismic grids possible. Because x, y and z co-ordinates are available in real time, the x-y coordinates applied in the field are also checked against the planned co-ordinates from the database. Layout thus becomes a virtually error-free process. Any departure from the planned grid, imposed by local field conditions, is automatically registered. The final field co-ordinates are entered into the central database immediately after each surveying party returns from the field, so that updated location maps, grid displays and simulations are available to the crew on a daily basis.
ACQUISITION MANAGEMENT AND MONITORING
Spread geometry and acquisition parameters for each day's work are extracted from the database before each shift and loaded onto diskette for direct transfer to the recording truck's computer. This process can be performed routinely by a field management system combined with the SN 368 seismic recorder.
Throughout his shift, the observer receives computerized assistance in programming spread patterns and in checking field equipment, (Fig. 4). Field acquisition parameters and quality control test results are recorded on another diskette, which updates the central database at the end of the day. Further quality control checks, such as coverage update plots, are readily available from the central database, for use in case modifications to the acquisition program are required.
FIELD STATIC CORRECTIONS AND
DATA INTEGRATION
Supplementary surveys performed for the purposes of static corrections are handled by the central database in much the same way as the main data flow. The static corrections are computed by the field data management system and assigned in the central database to the corresponding stations of the seismic grid, (Fig. 5).
When a given acquisition sequence has been completed, two sets of data are available: seismic data recorded on magnetic tape, and a coherent set of supporting data, which can be transferred from the central database to a diskette. These magnetic tapes and diskettes are then delivered to a data processing center, where a simple file transfer program combines the two data sets at the formatting stage. No further data input is required prior to processing.
Presenting a coherent data set in this form results in a useful reduction in 3D processing time. Processing on supercomputers reduces turnaround even further. As many operating companies know, this can be of vital importance when petroleum engineers are awaiting well sites from seismic interpretation.
FIELD EXAMPLES
Example 1: 3D survey performed on a proven field, (Fig. 6). In this survey, which lasted several months, an average of 61 km2 were covered per month, with 16-fold, 25 m x 25 m bins. The acquisition crew recorded 7700 vibrated points (VPs) per month, with 256 stations per VP. Stations were laid out in 4 lines of 64 traces each; lines were 200 m apart; the station interval was 50 m. Acquisition costs were approximately $500,000 per month. With the same 50 m trace and VP interval, typical 2D costs in this area average $2000 per line km. Investing $500,000 a month on a 2D survey would then yield about 250 km of seismic lines, which would cover the same area as the 3D survey with a 500 m spaced square grid. Obviously, the resulting data set would only be a small fraction of the information gathered by the 3D survey. In other zones, depending upon surface conditions, the equivalence of costs between 2D and 3D surveys corresponds to 2D grid spacing ranging from 400 to 800 meters.
Example 2: Mini 3D survey carried out in a production zone, (Fig. 7). In this case, a comprehensive 3D image of the area around a producing well was obtained in a short time by surveying and laying out a mini 3D grid.
An area of 3 x 1.8 km was covered by a total of 480 traces organized in 8 parallel lines of 60 traces each. Lines were placed 200 m apart with a 50 m station interval. 296 shot points were staked, avoiding pipes and other obstacles, so as to yield 16-fold, 25 m x 25 m bins. The shot point array consisted of 12 2m-deep holes. Approximately 10 days were required for survey preparation, including reconnaissance, surveying, spread layout and checks. For the actual survey, seven three-man drilling crews, one observer and his assistants took three days to perform simultaneous drilling and recording with no preloading. Another three days were required for clean-up operations.
3D sections were processed and delivered to the interpreter 19 days after the last shot was recorded. Surface and well seismic ties were added to the same sections and delivered five days later. The total cost of the survey was in the order of $250,000; it thus cost less than a single development well and provided a complete and accurate seismic image of the area in question without interrupting drilling operations.
Once the crew has been mobilized, a series of such mini 3D surveys can be performed in order to obtain, within a short period of time, a complete and calibrated 3D image of the field at very reasonable cost.
CONCLUSIONS
Continuous progress in the efficiency of 3D land acquisition is part of a general trend which has already seen enormous improvements in 3D seismic data processing, with the advent of supercomputers, and in 3D interpretation, with universal use of interactive interpretation and mapping systems.
Current 3D seismic technology has proved a highly efficient and cost-effective exploration tool, delivering continuous and accurate information far surpassing that of traditional 2D seismic methods.
Indeed, some explorationists are even predicting that 3D seismic can effectively replace initial reconnaissance surveys. The time gained by reducing two traditional steps-reconnaissance and detail surveys-to one, added to the time gained in processing turnaround, could then allow drilling to begin a year earlier, with greater chances of success.
Copyright 1990 Oil & Gas Journal. All Rights Reserved.
