Borrowing ideas from other industries, a seismic solution provider ponders acquisition in a world without cables.

Since the end of World War II, oil and gas companies, equipment manufacturers and acquisition contractors have made continuous progress in developing new and improved land seismic imaging tools and methods. While these have improved exploration success rates and reduced the risks and costs of finding and developing hydrocarbons, our industry's ability to continue the pace of technology advancement is impeded by a single, highly restrictive constraint - cables.
The vast majority of land acquisition systems rely on cable-based architectures. Cables are required to transmit power, command and control signals, and the acquired seismic data. Unfortunately, cables impact land imaging in several unfavorable ways. Some of these adverse impacts are operational in nature and are directly related to the costs to deploy, repair and roll cable-based sensor grids in the field. Other impacts adversely affect the acquired seismic image. These include an inability to cost-effectively sample the subsurface with a very high coverage of single-point sensors.
If we can eliminate cables for onshore acquisition, our industry stands poised to unlock the Holy Grail of land seismic imaging - cost-effective, fully sampled, full-wave surveys.
Cable-based architecture limitations
Cables serve multiple purposes, though the primary one is to transmit data on a real-time basis from the sensors to the central recording system. While the ability to view all shot records in real time may provide some comfort in quality-controlling the acquisition process, the cost to provide this capability is significant in terms of operational efficiency, HSE and image quality. At least six major downsides result from the use of cable-based recording systems:
Significant system weight. On today's standard seismic survey, cables and miscellaneous ground equipment supporting cable-based data transmission (e.g., batteries, battery booster units) weigh 25 tons or more. Because weight directly contributes to the costs of transporting gear and mobilizing a seismic crew, the cables themselves increase the cost of acquisition. The economic modeling work we have completed suggests that the excess weight introduced by cables accounts for up to 20% of the operational cost of a "typical" survey onshore in North America.
Manpower and logistics intensity. Deploying, rolling, troubleshooting and repairing a cable-based system is a manpower-intensive operation. It is estimated that 25% to 50% of the individuals tasked with spread deployment and retrieval are involved in cable-based activity. Additionally, 50% to 75% of troubleshooting personnel are focused on cable problems. This manpower intensity, all driven by cables, has other second-order cost impacts. For instance, as the number of field personnel goes up, the costs to train, mobilize, feed and shelter them goes up as well.
Increased HSE risk. Every individual in the field represents a potential health and safety liability for both the contractor and the oil and gas company. So cables also drive up health and safety risks. Moreover, the process of moving heavy cables, especially in mountainous areas or other difficult terrain, is a hazardous operation.
Cable repair and maintenance downtime. In our recent work with seismic contractors, we have discovered that up to 50% of operational time is spent on cable troubleshooting. This has a direct impact on costs (one of the seismic crews we studied spent nearly US $1,000 a day repairing cables). The even bigger impact is on productivity, with only 50% of the time spent on actual acquisition. In effect, the cables can cause any seismic survey to be only about half as efficient as it could be.
Complex network architectures that impact production. As the number of stations increase, system productivity time decreases. In complex surveys, there is simply more opportunity for line failures, while troubleshooting often involves time-consuming, sequential, trial-and-error approaches. Modern cable systems with redundant data path capabilities improve reliability to a point, but at the cost of adding even more cables.
Undersampling the subsurface. Cable-based architectures impose constraints on how surveys are designed. For instance, sensors are required to be spaced in gridded geometries at intervals that approximate the length of individual cable takeouts. This prevents a survey from being tailored to unique surface, near-surface and subsurface challenges. Moreover, cable-based architectures make increasing station density prohibitively costly. As shown in Figure 1, the operational cost to acquire a fully sampled, high station count survey increases with a cableless system, but not nearly at the rate as the same survey acquired with cables. While many geophysicists we've interviewed see the ideal survey design as having upwards of 50,000 stations, cable-based systems make this goal operationally impractical. As a result, cables force compromises in final image quality.
Full sampling versus partial sampling
Contemporary 3-D seismic acquisition techniques are not meeting the industry's imperative need to rapidly and cost-effectively identify new drilling opportunities and reduce prospect risk. These shortcomings are directly related to what can be done economically and safely given constraints imposed by station counts and the layout of shot-receiver spreads.
As an example, consider a gas-prone basin where the main producing horizon is at a depth of 8,200 ft to 11,500 ft (2,500 m to 3,500 m). Historically, this horizon was the only economic producing interval in the basin except for a few shallow oil fields. In recent years, advancements in drilling and hydraulic fracturing techniques have increased operator interest in developing deeper gas plays, while higher commodity prices have made economic both smaller, shallow oil plays and subtle, higher-risk opportunities in the original producing horizon.
Unfortunately, existing seismic data - even recently acquired, high-fold, wide-azimuth data - is not suitable for the next generation of prospecting and development. Why not? With traditional cable-based recording systems, we simply cannot cost-effectively deploy enough stations to properly image prospective targets at all basin depths.
The history of seismic imaging in this basin is not unique, but instead follows a path very similar to other basins around the world. First-generation seismic data was 2-D. The next generation of data was narrow-azimuth, short-offset 3-D that tended to be undersampled. In Figure 2, we took a contemporary, wider-azimuth, longer-offset 3-D survey and decimated the data in order to simulate the effects of undersampling. Of note is the time and amplitude scalloping along events from migration of undersampled data. This example is a pre-stack migration, but the same effect occurs to a lesser extent on a post-stack migration and creates a footprint in the data. The scalloping is worse in the shallow travel time part of the section because cost-driven acquisition design constraints limited optimal survey imaging to a very narrow time range.
Early-generation 3-D data was also low fold. Using this data, one could map structure and some broad stratigraphic features. However, long offsets were generally lacking, and the narrow-azimuth nature of the data limited illumination and the ability to interpret anisotropy and fracture patterns. Additionally, there tended to be excessive duplication of offsets in the bins along with significant offset gaps.
Interpreting in the deep or shallow sections using this legacy data means the geophysicist is relying on potentially poor-quality amplitudes. Amplitudes showing up on the seismic data may have nothing to do with the reservoir and everything to do with artifacts of how the data was acquired and processed.
None of this addresses the other aspect of undersampling, which is that we have not effectively reduced the bin size; therefore, subtle features in the reservoir are not imaged any better than with the previous generation of seismic except for the noise reduction higher fold provides and for the upside delivered by wide-azimuth attributes.
To properly image all horizons in this basin, the geophysicist needs a new generation of seismic data with full sampling of the subsurface.
Attempts to address the challenge
Over the past several decades, wireless technology has been applied to seismic acquisition systems in different ways to accomplish different objectives. Unfortunately, none have produced productivity breakthroughs that make high station count surveys practical. In order to better understand the productivity limitations of existing wireless systems, we've highlighted below the general architecture and constraints of currently available systems.
Figure 3 shows line segment examples of two radio implementations in modern land acquisition systems. In the uppermost example (shown in Figure 3a), multiple receiver channels are connected to a common field acquisition unit. Each field acquisition unit utilizes radio telemetry to communicate to a recording truck. The radio link provides command and control from the central station to the field units, while the data is stored locally. The second example in Figure 3a illustrates early implementations of radio telemetry to extend cable-based systems where environmental conditions, permitting issues or other obstructions restrict access. These point-to-point radio connections extend cable telemetry spreads, ultimately connecting all stations to a common central recorder.
A third implementation (Figure 3b) of radio telemetry involves multichannel acquisition units that use wireless telemetry to connect to a central node. The nodes interconnect by cable to each other and to the central station, creating a combination radio/cable telemetry system. Depending on the type of wireless protocol that is implemented, there can be severe bandwidth limitations, limiting its practical use in large-scale acquisition projects.
In each of the above implementations, a significant amount of cable remains both for telemetry and for connecting sensors to the acquisition units. The potential benefit from incorporating wireless is minimized unless a significant reduction in the amount of cable can be achieved.
The ideal land acquisition system
Next-generation acquisition systems should be able to leverage numerous advances in wireless transmission, power and data storage from other industries. Figure 4 shows the form such a system might take. In the diagram, a single data acquisition station is connected to a single 3-C digital MEMS sensor.
Each data acquisition station would have an independent, bi-directional communication path to the recording truck. There would be no telemetry cables interconnecting stations, allowing receivers to be deployed without the constraints of a grid infrastructure.
The digital MEMS sensors would measure true 3-D particle motion and record the full seismic wavefield with unsurpassed vector fidelity. As single-point receivers, they would be less susceptible to the intra-array statics problems of geophones and record the broadest bandwidth that the Earth returns. In addition, they allow additional cables associated with receiver arrays to be removed from the system.
Data recording would occur at the station level, with local storage in solid state memory. Intelligent QC features would automatically notify the operator of trace problems and when the condition of field electronics exceeds user-defined limits. The QC system would send back key attributes of selected traces to ensure the spread was functioning as planned. By eliminating the need to transmit all data in nearly real time, power requirements and bandwidth constraints are greatly reduced.
Each field acquisition unit would operate autonomously, thereby eliminating single points of failure that are present in cable-based systems and allowing stations to be undisturbed once deployed until they are moved to the next stage of the survey. The field acquisition units would also have embedded Global Positioning System (GPS) features to determine their position with a high degree of accuracy and with a reduced need for surveying expenditures and cycle time.
Lastly, this next-generation land acquisition system would be supported by the latest in command and control software. Key features would include the ability to rapidly determine the actual spread configuration vs. the original survey plan, testing availability of all the stations in the spread, advanced troubleshooting and QC protocols, and the ability to record "processing-ready" seismic data without geometry and header errors.
Potential system benefits
This ideal land system could deliver significant benefits to the oil and gas companies and contractors, including:
• Improved seismic image quality;
• Increased operational productivity; and
• Enhanced HSE performance.
Seismic image quality would increase since geophysicists would be able to randomize station placement and customize survey designs for specific subsurface imaging objectives, including imaging shallow, intermediate and deep reservoir targets simultaneously. Station density would be increased dramatically, improving spatial resolution and reducing effective bin size. In addition, surveys could be more cost-effectively designed for wide-azimuth, long-offset acquisition to help with AVO analysis, coping with anisotropy and modeling reservoir fracture networks. Image quality could be further enhanced if 3-C digital sensors were used to acquire a broader frequency spectrum of reflected seismic energy, record both compressional (P) wave and shear (S) wave data, and mitigate intra-array statics issues associated with traditional geophone receiver arrays.
A key benefit of a cableless architecture is operational productivity. Once the cables are gone, the weight of the entire system goes down dramatically. Our best estimate is that the weight could be reduced by 80%. In addition, eliminating the cables would mean greater reliability of the entire land acquisition network and less downtime for troubleshooting and cable repair. Detailed models that we have developed of the conventional operational process suggest that a typical 3,000-station survey could be performed at approximately 80% to 85% of the operational cost if an ideal cableless system was used. The improvements with a cableless system are even more dramatic as station counts increase. Compared to conventional recording systems, a cableless system could acquire 12,500 stations worth of data at approximately 50% to 60% of the operational cost.
Finally, HSE performance should improve. Once the cables are gone, we would expect fewer incidents during deployment. Total weight is substantially decreased. The need to move heavy cables is reduced. And fewer personnel are needed to troubleshoot and repair cables. In addition, less acquisition equipment results in a reduced environmental footprint. Cable lines wouldn't need to be cut, nor would surface groundcover be subject to cable deployment operations. Lastly, the number of support vehicles could possibly be reduced, resulting in fewer emissions, fuel spills and collateral damage in the areas adjacent to the acquisition operations.
Summary
Taking land seismic imaging to the next level requires developing a land recording system capable of cost-effectively acquiring fully sampled, full-wave data. Realizing this vision requires eliminating the cables in traditional seismic acquisition systems. Next-generation, cableless acquisition systems will provide geophysicists with the ability to customize survey designs for specific subsurface imaging targets at any depth and to more widely deploy 3-C digital sensors. In addition to delivering improved image quality, cableless systems also promise to improve the productivity of land acquisition operations and to facilitate enhanced HSE performance. Overall, next-generation cableless acquisition systems appear to be the next big thing in land seismic imaging.

Comments