Links to land demonstrate real-time potential
Dave Ridyard
Input/Output Inc.
Grant Byerley
Hardy Hartmann Nielsen
ConocoPhillips Norge
In the fall of 2002, ConocoPhillips installed and tested a new seafloor seismic technology over the Ekofisk field using a prototype system developed by Input/ Output. The objective was to verify that reliable full wave field seismic data could be recorded over the field. This data could in turn be linked to ConocoPhillips' recently installed onshore visualization center and drilling control room. The availability of up-to-date information for decision makers is considered vital to control an oilfield in real time.
The Ekofisk field lies in the southwest corner of the Norwegian sector of the North Sea. It is part of license 18 operated by Con-ocoPhillips. Partners are Total, Eni, Norsk Hydro, Statoil AS, and Petoro AS. After over 30 years of production, Ekofisk still possesses more reserves than any other field in the Norwegian shelf.
The continuing high levels of production achieved on Ekofisk are made all the more remarkable for two reasons. First, the field has been subject to significant subsidence due to compaction of the underlying chalk reservoir. Although the seafloor is now falling at a reduced rate, it has previously fallen at rates as high as 40 cm/yr. Second, the large interconnected production infrastructure makes acquisition of traditional towed-streamer seismic very difficult. Furthermore, differential pressure and shallow gas in the overburden contribute to inadequate P-wave seismic imaging of the reservoir over the crest of the Ekofisk field.
The seismic obscured area (SOA) at Ekofisk is problematic to current reservoir characterization efforts. Approximately one-third of the reservoir is seismically invisible using conventional P-wave seismic data due to the complicated low P-wave velocity field caused by the presence of gas and overpressured shales in the overburden. The application of converted wave imaging may eventually be used to provide an improved image through the SOA at Ekofisk.
Shear wave imaging
It has been understood for many years that shear waves have the ability to travel through gas clouds that obscure reservoirs from imaging with traditional P-wave seismic. Historically, converted wave (PS) imaging has been too expensive for most applications. Furthermore, results have rarely delivered the expected value due to some of the practical problems of shear wave data acquisition.
The re-occurring problems observed in existing PS-data are attributed to three known factors – hardware, processing, and theoretical limitations. The Ekofisk seafloor seismic test was conducted in an attempt to identify new technology that has the potential to overcome some of the current hardware limitations of conventional ocean bottom multi-component systems that are available today.
From an operational perspective, one of the big challenges of ocean bottom seismic data acquisition is maintaining true vertical sensor orientation. This is typically achieved in one of three ways, and all of them have some undesirable consequence:
- Large omnidirectional geophones do not need true vertical alignment, but they exhibit a poor low frequency response and other non-linear characteristics
- ROVs can manually deploy traditional sensors, but this is time-consuming and expensive
- The most common solution has been the use of mechanical gimbaling systems to dynamically maintain vertical orientation.
null
Gimbals are bulky and expensive to manufacture and deploy. Gimbals also lead to potentially significant data quality problems – especially in the area of vector fidelity.
Vector fidelity is a measure of how accurately a sensor can respond to particle motion on a specific axis. This is difficult to achieve with traditional mechanical gimbals. A poor vector fidelity response results in an inconsistent measurement of the vector wave field. The polarized nature of shear wave data relies on good vector fidelity to provide reliable PS data.
Full wave imaging
Input/Output has developed a unique sensor to address many of the traditional operational and data quality issues associated with multi-component data acquisition. I/O's Vec-torSeis sensor is a tiny accelerometer etched out of silicon wafers that provides true digital output. The device is capable of measuring signals from 0 Hz to several kHz. If three mutually orthogonal sensors are deployed, the unit can detect the component of gravity observed by each sensor and derive the true orientation of the whole assembly with a high degree of accuracy.
VectorSeis raw data (left) exhibits higher frequencies, less "ringiness," and dramatically greater inline/crossline data similarity than conventional sensor data (right).
null
This compact sensor technology allows relatively easy deployment in all water depths and eliminates the need for ROVs and gimbals. The technology offers the potential to acquire state-of-the-art, full bandwidth, high-fidelity seismic data at a lower cost. Recognizing the potential of this new technology, in the fall of 2002, Con-ocoPhillips conducted a pilot test survey using a prototype VectorSeis system over the Ekofisk field.
null
An ideal sensor would show a uniform response for all incident azimuths, yielding a uniform dark blue disk. The conventional sensors (right) show significant non-linearity for signals arriving from the 45° azimuths, compared to the VectorSeis sensors (left).
Data acquisition
A cable with 40 sensor stations was deployed on the seafloor. Each station comprised a VectorSeis 3C sensor, plus a hydrophone, for use in removing water-bottom multiple reverberations. The stations were deployed from a shipboard reel at a 50-m station interval. The rugged, flexible design of the cable allowed standard cable handling techniques to be used to keep test costs low. Another cost-effective feature was the remote recording buoy, which eliminated the need for a dedicated recording vessel. Quality control information was available via radio telemetry link aboard the shooting vessel.
Wide azimuth seismic data was recorded using a relatively small source array in order to evaluate the sensor performance for vector fidelity, coupling, and signal quality.
Upon completion of the seismic data acquisition program, the cable was dragged closer to the platform and reconnected directly to the Ekofisk complex where data was fed through the onshore fiber optic network. This test proved the viability of real-time delivery of the data to ConocoPhillips offices in Stavanger. Microseismic data was also monitored to facilitate further research into the compaction mechanisms of the producing reservoir.
The test cable was online and recorded continuously in passive mode for three weeks after deployment. The cable was left in the water for seven months to assist I/O's engineers in understanding issues related to the long-term or permanent deployment of seafloor cables.
At this 2D line intersection the VectorSeis data show improved resolution and amplitude consistency.
null
Experiment results
For a shot line acquired at 45° to the receiver cable, when the data is organized into a common receiver gather, the inline and crossline sensors should exhibit broadly similar character. The data showed this to be the case, but the conventional sensor data shows that the conventional sensors exhibit a dramatically different response between the in-line and crossline vector components. The conventional sensor also lacks the broad bandwidth of the VectorSeis data.
Since the experiment did not record sufficient data to form a full 3D image, data quality can only be assessed by comparing the results with previously acquired 2D data. Towed streamer data over the seismic obscured area is very poor, so the most useful comparison can be made against earlier ocean bottom cable (OBC) seismic data. The stacked VectorSeis data shows significantly improved resolution and more consistent amplitudes when compared to a 2D line acquired in 1999 using conventional OBC technology.
Permanent monitoring
After the success of the 2002 experiment, in August 2003, ConocoPhillips purchased and installed a small permanent monitoring system for the Ekofisk field using VectorSeis technology. This system is now installed and delivering real-time data to ConocoPhillips offices, without any infield operational staff. The mini system provides continuous microseismic monitoring and also functions as an offshore station for recording regional earthquake activity.
The mini system will also provide the experience and data to help design cost effective permanent reservoir monitoring systems for oil fields requiring real-time seismic imaging on demand. Permanent seabed seismic networks offer the potential to provide frequent, high quality images of the reservoir. Combining these images with other emerging technologies, we can expect to improve the quality and timeliness of key production decisions, thus reducing risk, while increasing net present value and total recovery.
The Ekofisk VectorSeis test data demonstrate that the sensor is a new technology that can be used to provide an improved vector fidelity response, which should ultimately result in more reliable seismic data for future reservoir characterization efforts.
Acknowledgements
The authors would like to thank ConocoPhillips Norway and their PL018 coventurers Total Exploration Norge AS, Norsk Agip A/S, Norsk Hydro Produksjon a.s, Petoro AS, Statoil ASA, and Input/Output for perm-ission to publish the results of this study.
Authors
Dave Ridyard is seabed imaging business development manager for Input/Output. Grant Byerley is a geo-physicist working with ConocoPhillips. Hardy Nielsen is the chief geophysicist of Con-ocoPhillips Norge.
For information, contact Dave Ridyard at tel: 281-879-2135 or email: [email protected].
