Autonomous 4C Nodes used in infill areas to complement streamer data, deepwater case study. Pierre-Yves Granger and Michel Manin, Compagnie Générale de Géophysique, 91341 Massy, France Jean-Luc Boelle*, Enrico Ceragioli and Frédéric Lefeuvre, TOTAL S.A., avenue Larribau, 64018 Pau, France Emmanuel Crouzy, TOTAL E&P ANGOLA, Luanda, Angola Summary During the summer of 2004 an experimental OBS survey was acquired and processed by CGG over the Girassol field operated by Total Angola, offshore West Africa. Five ARMSS nodes (Autonomous Reservoir Monitoring Seismic System) were deployed in this area at a water depth of 1300 meters. The primary objective of this trial was to verify the operational sequence and performance of this new generation of 4C recording equipment. There were also a number of secondary objectives which were to benchmark the recorded data with a view to infilling the streamer acquisition in this difficult environment and to evaluate the added value of recording four components. In this paper, after a short description of the recording system and the acquisition layout, the results of the processing are compared with streamer data acquired previously in the same area. The 4D capabilities of the node technology are assessed. A comparison of the results proves that nodes could be used in infill areas to complement streamer data. Introduction Deepwater hotspots, such as Angola, Brazil and the Gulf of Mexico, are becoming increasingly congested with production infrastructure. Operators wishing to monitor production at these sites are faced with the impracticability of a time-lapse seismic survey (4D) of the whole area using the conventional surface streamer method. Ocean Bottom Seismic technologies offer an attractive solution for acquiring infill time-lapse seismic data as a complement to streamer acquisition to give a full picture of the target zone. As an experiment, data was acquired in the deep water of Angola with five autonomous nodes and analysed with respect to potential future industrial applications. Description of the autonomous nodes Figure 1 shows the node in operating mode. The sensor head incorporates three fixed geophones in a Galperin arrangement. Tilt measurements are performed by three accelerometers. A vibrating system is included to ensure optimum coupling with the seafloor at the layout stage and a hydrophone is in direct contact with the water to complete the 4C system. A separate case contains the full recording system including batteries and the data storage unit, which can operate non stop for forty-five days at water depths of 3,000 meters. The system provides a high degree of vector fidelity through a decoupling of the sensors to the case unit via a fit-for-purpose umbilical wire. The compactness of

the autonomous system compared to the cable system allows placement adjacent to existing seabed facilities while its innovative design removes the need to plant the nodes, thereby reducing layout time and cost. Once deployed, the nodes remain in place and can be positioned quite close to production infrastructure. Repeatability is ensured as the nodes can be repositioned at the same location.

Girassol Field acquisition overview The node pilot project conducted over the Girassol field in the summer of 2004 included layout, acquisition and

Figure 1: The upper picture shows the sensor head firmly coupled on the sea-floor sand. The following picture (bottom) shows the entire system, two nodes are located few meters apart for a repeatability test purposes.

Autonomous 4C Nodes used in infill

recovery. The aims were to validate the new node technology with a view to its future deployment for industrial applications and to assess its 4D capabilities with respect to the fidelity of its seismic signal compared with previous high-resolution streamer surveys. ROV facilities were used to lay out five nodes 400 meters apart in a straight line at a water depth of 1,300m. A basket was used to descend the five nodes to the seabed, where the ROV took them to their final position (figure 2), the maximum positioning error was below 3 meters. The exact positioning was cross-checked at the processing stage by travel-time analysis using a water velocity profile acquired with Sippican probes. A seismic source boat made a series of passes above the nodes to acquire a patch of 21 shooting lines in a 2 x 6 km rectangle. A flip-flop shooting technique produced a full grid with a 50 x 37.5 shot interval. The total operation turnaround took less than 72 hours to lay out the nodes, shoot and retrieve them.

The continuous record of the raw data was uploaded from the node disk and split into constant length shot-records. The three geophones were rotated from a Galperin arrangement to X, Y and Z components using the accelerometer measurements. The navigation was merged before the data was sent to the processing centre. Figure 3 shows an example of the hydrophone and the three rotated components for the central shooting line passing over the nodes. Data processing The preliminary QCs confirmed the high level of vector fidelity delivered by the nodes. 3D particle motion analysis was used to confirm the re-orientation of the X, Y and Z components. The three components delivered from the

Galperin system were also compared using the CGG method (Gratacos & Granger, 2004), where a composite vertical geophone is derived using as a reference the hydrophone data after spectral calibration of the three geophones.

Figure 2: In order to reduce the turnaround time of the remotely operated vehile (ROV) operations, all the necessary nodes are descended in a basket to the seabed.

Hydrophone Amplitude Geophone spectra Hydrophone Vertical geophone

Horizontal X geophone Horizontal Y geophone

Figure 3: Displays of the four components of a receiver gather before the rotation of the horizontal componenst into the source-receiver direction. This figure illustrates the quality of the recorded data. Hydrophone and vertical geophone show a similar frequency content. The first pegleg indicates the water column of 1300 meters.

Pegleg

Autonomous 4C Nodes used in infill

For a reliable comparison with the previous 3D HR streamer volume (Beydoun et al, 2002), shot points were interpolated in the two grid directions prior to migration of each 3D node volume separately with the full Kirchhoff pre-stack time method (figure 4). The five volumes were then sorted into 6.25 x 6.25 bin sizes. Finally, the full 3D dataset was migrated in one pass for both hydrophone and geophone data (figure 5). To complete the comparison, a single matching operator was computed to match the hydrophone node results with the streamer results. A test, involving one operator per trace, did not improve the merge and showed a remarkable stability of the various operators. Added advantage of full wavefield recording Obviously the node system allowed for simultaneous recording of the full wavefield. Although this was not the

primary aim of the project, it seemed appropriate to evaluate the added value of recording pressure and velocities for both compression and shear waves. Hydrophone/geophone summation In this particular area, the target zone is not contaminated with the peglegs. Nevertheless, with a view to future applications in a deepwater context, the hydrophone/geophone combination was evaluated. Figure 6 illustrates this operation and results are obtained after spectral calibration of the geophone toward the hydrophone. The geophone is then subtracted from the hydrophone according to a simple compensation of its energy versus incidence angle. The sum provides the down-going waves (the source response).

Figure 5: The 3D migration results of hydrophone and geophone data from the five nodes are compared with the previous 3D HR streamer data.

Figure 4: Single node 3D migration of hydrophone data compared to the previous 3D HR streamer migration.

3D Common Receiver Gather Hydrophone volume

Single node Migration results

Reference: 3D HR streamer

migration Fold = 52

Single fold

3D HR streamer migration; Fold = 52 5 Nodes - Hydrophone – 3D Migration 5 Nodes – Geophone – 3D Migration

Autonomous 4C Nodes used in infill

Shear waves A preliminary processing sequence was applied to the horizontal component. For each source location, the data was rotated into the radial-transverse coordinates, producing maximum energy of the direct arrival on the radial component and minimum energy on the transverse component as expected in the presence of vector fidelity. A simple velocity function was derived and applied for comparison with the vertical component. Wide bandwidth and a high shear wave signal-to-noise ratio with a reasonable compression wave residual were observed. A preliminary test of event matching was run (figure 7), producing very promising results in terms of estimating the

compression/shear wave velocity ratio. These results demonstrated the usefulness of shear wave recording for either calibration of P-wave AVO or even more sophisticated dual-wave inversion. Conclusions The results of the Girassol project confirmed that high-quality data could be acquired from a sparse distribution of autonomous nodes. Despite the very low coverage provided, the quality of the migrated infill volumes (hydrophone and vertical geophone data) compared well with the previous 3D HR streamer results at the target level. The quality of the data pleased the Total interpreters and use of node technology for future time-lapse survey in congested places is envisaged. Results indicate that a single patch of 50 nodes could easily infill a surface streamer acquisition in an obstructed area of about 4 km2. References Beydoun, W., Kerdraon, Y., Lefeuvre, F., Bancelin, J-P., Medina, S. and Bleines, B., 2002, Benefits of a 3DHR survey for Girassol Field appraisal and development, Angola : The Leading Edge, Vol. 21, N°. 11, pp. 1152-1155. B. Gratacos and PY. Granger, 2004, Calibration of multicomponent sensors, 66th Mtg. EAGE Acknowledgements The authors would like to thank Total E&P Angola, Total S.A., Sociedade Nacional de Combustíveis de Angola (SONANGOL), Esso Exploration Angola (Block 17) Ltd., BP Exploration (Angola) Ltd., Statoil Angola Block 17 A.S., and Norsk Hydro for permission to present the information contained in this paper.

Figure 6: The reference hydrophone (left) exhibits source bubble effect and huge multiples after the first pegleg. The hydrophone-geophone combination (centre) exhibits mainly the primaries. The hydrophone+geophone combination exhibits mainly the down-going waves (right).

Figure 7: Half migrated hydrophone image (left) stretched to compare with half radial component receiver gather (right) after PS wave NMO corrections.

Hydrophone Hydrophone - Geophone Hydrophone + Geophone

Pegleg

Compression Waves

Shear Waves