EAGE/SPG Workshop on Broadband Seismic

Variable-depth streamer acquisition is a key solution for broadband marine seismic allowing for a drastic increase in the available frequency bandwidth, in both the low and high ends of the frequency spectrum, from 2.5 Hz to the source ghost notch. During acquisition, the direct and ghost arrivals at the receiver side creates a well-known interference pattern, which includes the receiver ghost notch, where frequency loss at the notch frequency depends directly on the receiver depths. As a consequence of this, a clear strong notch will always occur with a conventional flat-streamer acquisition. On the contrary, varying the receiver depths along a streamer allows recording a wide diversity of receiver ghosts or notches. Additionally, the sea-state noise level decreases as the cable is towed deeper: this technique thus benefits from towing solid streamers at what is generally considered as extreme depths (up to 50 meters), which allows recording superior signals to noise ratio at low frequencies. Rather than using a linear increase in streamer depth with offset (original slant streamer geometry), a custom profile is designed in order to provide the optimum receiver ghost diversity, particularly for shallow events, and can be tuned to provide the maximum possible bandwidth for a given geological setting and water depth.

Broad bandwidth data is becoming increasingly more desirable driven by the needs of seismic reservoir characterisation. There are now a range of solutions on offer to the free-surface ghost problem and its effects on seismic bandwidth. We present a new de-ghosting processing technology (DUG Broad) and provide examples of its application on a range of real data sets.

In the past years, numerous efforts have been made to increase the bandwidth of seismic data, both for the low and the high frequencies. Seismic data has gained two octaves in the low frequencies (from 10 Hz down to 2.5 Hz) and more than one octave in the high frequencies (from 80 Hz to 160 Hz and up to 200 Hz). This has the potential to fill the information gap described by Claerbout (1985). The aim of this paper is to discuss this gap and present two real data examples of variable-depth streamer acquisition where this goal has been achieved.

The seismic industry is constantly seeking ways of improving the contribution of seismic data to the upstream E&P workflow from seismic acquisition to reservoir modeling. We review recent developments in broadband seismic and illustrate how these add value for seismic interpreters and geoscientists involved in reservoir characterization or quantitative interpretation projects. In 2007 a dual-sensor streamer acquisition system was developed with the objective of providing broader seismic frequency bandwidth without any compromise in pre-stack data quality or acquisition efficiency. The increase in bandwidth is achieved by removing the sea-surface ghost at the receiver end via the principle of wavefield separation. Results over the last five years have demonstrated the benefits of this system in processing, seismic interpretation and reservoir geophysics. Case studies from different geological settings illustrate the benefits to end-user practitioners in seismic interpretation and seismic reservoir characterization across a range of E&P asset development phases from exploration to appraisal and field development and optimization.

Expanding the bandwidth of surface seismic data, particularly towards low frequencies, is essential for many exploration and production objectives. Broader band signals, both in land and marine environments have marked benefits for imaging deeper targets, imaging through absorptive overburdens, and especially inversion for rock properties. Various methods have been proposed and implemented to expand seismic bandwidth; these include both acquisition and signal processing methods. A question that is often asked is how much difference does changing the acquisition geometry make? In this paper, we present a case study of a consistent, experimental offshore dataset in Southeast Asia. This data consists of a single boat pass of different cable depth configurations. These data were then processed with their appropriate deghosting methods and results compared. In addition, we examine methods for evaluating the success of these methods and their potential pitfalls. The key determinant of the eventual bandwidth of surface seismic data is the convolution of the source, near surface effects (free surface ghosts in the marine situation), the overall earth attenuation, and the level of additive environmental noise. Some of these effects can be modified by changes in the field acquisition geometry or at least, deterministically compensated for. The noise level may constrain or limit the capacity of signal processing tools to compensate for the “field” effects. When evaluating the raw and processed data it is wise to use various types of analysis and displays. Simple seismic amplitude stack sections and associated spectra can be misleading. Spectral “split” plots and inversion of the seismic data is often more indicative of success. In marine seismic acquisition, the free-surface ghost effect is one factor that can strongly impact the data bandwidth characteristics according to streamer and source depth. It is also a factor that could easily be adjusted in the survey design. Shallow towing favors the higher frequency response at the expense of low frequencies, whilst deeper towing favors the lower frequencies, at the expense of higher frequencies. Moreover, a deeper tow typically has lower levels of swell noise. At very low frequencies, the critical two octaves between two to eight Hertz, the level of ambient towing noise rises, the “DC notch” strengthens, and the power output of airgun sources declines. This combination provides both the biggest challenge and the opportunity of bandwidth expansion.

A set of land broadband seismic data from Oman is presented. They belongto a few dense wide azimuth surveys which were recently recorded using a 1.5Hz-86Hz sweep.We show how the quality of the broadband signal can be effectively controlled at each processing step in such surveys. The aspects in which the low frequencies differ from standard frequencies are presented and the benefits of applying octave dependent processes are discussed. The ability to acquire and image high quality low frequencies on marine data sets has been demonstrated for sometime. We illustrate via these examples that this goal can also be achieved on land data.

Broadband marine is not new - the original method for broadband marine data acquisition by combining vector (vertical geophone) and scalar (hydrophone) data was proposed by Milo Backus in 1958! (Water Reverberations--Their Nature and Elimination, Geophysics, 1958). This ocean bottom dual sensor approach has been applied, primarily for appraisal and development applications, for a many years worldwide but the scale of the surveys, by dint of their focus on field specific imaging objectives, has been limited compared to towed streamer surveys in both size and duration. One of the challenges set by the oil companies has been to reduce the unit costs of ocean bottom data – “If only the square kilometer rates were lower we would shoot more data” is a common mantra. The difficulty in doing this has been the inherent technical downtime experienced by all the contractors operating ocean bottom systems – the terminations, connectors, power distribution and data telemetry components within a traditional ocean bottom cable (OBC) system are inherently prone to failure due to the intrinsic nature of the cable deployment/recovery cycle where the cables are stressed and de-stressed every time they are laid onto/recovered from the seabed. It is akin to recovering and deploying the full streamer spread every line change for towed streamer operations.

As the hydrocarbon E&P industry is venturing into ever deeper and more complex areas, the challenge of exploration and exploitation of oil and gas in such settings is growing all the time. Innovative, alternative approaches are needed to deal with ever increasing needs of better imaging, interpretation and understanding of subsurface. In this context there has been more and more expectations and deliverability from the development in seismic API. The main driver to these expectations being seismic resolution improvement through enhanced low and high frequencies. In the recent past advancement in marine seismic acquisition has evolved around improving the bandwidth of seismic signal by enhancing low and high frequencies information under the Broadband API. Structural imaging though was/is the main objective, reservoir characterization improvements has been the key requirement/expectations from broadband seismic applications. Broad band may prove to be very effective and vital, not only in exploring the frontier areas but it could be vital to maximize recovery from existing fields and also to devote attention to the vast potential of unconventional hydrocarbon resources like shale gas. In general, it is evident that success comes from an integrated effort between acquisition and processing innovations.