With exploration and production (E&P) trends towards smaller and more complex reservoirs, there is an urgent need for technology supporting more detailed reservoir understanding. The integration dilemma for reservoir modeling has historically been how to reconcile well-based measurements, having high vertical resolution and poor spatial resolution, with seismic data, having good spatial resolution but poorer vertical resolution. Recent advances in seismic technology are changing the dynamics of this relationship, providing hope for increasing the fidelity of our reservoir models.

Better seismic technology

Over the past several years seismic technology has evolved forward on a number of fronts spanning data acquisition, data processing and interpretation. While seismic surveys have steadily increased in number of spatial samples acquired, recent improvements in acquisition equipment and techniques have focused upon boosting vertical resolution. Particularly exciting are the emerging full-wave seismic technologies that are providing more types of subsurface measurements in addition to increasing vertical resolution. Wellbore seismic has also made some important recent advances, with passive microseismic monitoring emerging as a powerful new tool.

Significant developments are also occurring in seismic imaging, as algorithms move to prestack wave equation migration, time and depth anisotropic corrections, and automated velocity estimation. Signal enhancement for seismic multiples and other contaminants, coupled with imaging improvements, are contributing to better seismic resolution from the processing stage.

Improved seismic attributes and processes are also available to adapt seismic data for specific framework and reservoir property modeling tasks. Seismic interpretation and reservoir modeling tools, however, have experienced little recent innovation, with most software tools tracing to beginnings 10 or even 20 years ago. With so much surrounding innovation and clear business needs, a new generation of software tools are emerging to integrate and fully extract reservoir value from all data sources.

Improved reservoir framework and grid modeling

A wave of seismic processing enhancements have emerged to create better imaged seismic. Prestack wave equation depth migration, anisotropic depth and time migration, automated velocity estimation and model-based multiple removal are some of the "state-of-the-art" techniques in the seismic processor's toolkit. At the same time, capabilities like structure-oriented filtering and spectral enhancement are allowing interpreters to create high-resolution and more noise-free seismic images. These enhanced products, in turn, are making it possible to create reservoir frameworks that are more accurate subsurface representations. In addition, increased seismic resolution reveals greater stratigraphic character that can guide detailed reservoir grid modeling.

Seismic attributes are also playing an increasingly important role in both reservoir framework and grid modeling. Attributes such as coherency, spectral decomposition, colored inversion and curvature, amongst others, offer a range of perspectives on the seismic character of the reservoir.

In essence, modern seismic attributes offer the opportunity to tailor the seismic character for specific interpretation and reservoir modeling objectives. For example, while coherency attributes reveal much about faulting, curvature can provide more subtle fracturing details, and spectral decomposition can allow mapping of subtle channels and other stratigraphic details.

The multi-scene image in Figure 1 illustrates the value of a number of seismic enhancements and attributes available in modern reservoir interpretation systems. On the left panel, a high-frequency time image is intersected by a structure-oriented filtered vertical section and a combined coherency and amplitude horizontal slice, all aiding in the creation of a more accurate faulted surface interpretation. The overall benefit of this multi-attribute workflow is the construction of more accurate reservoir frameworks and detailed stratigraphic grids, as illustrated in the right panel of Figure 1.

Improved reservoir property modeling

Amplitude versus offset (AVO) and related approaches have advanced seismic analysis well beyond framework and grid modeling to the estimation of reservoir properties. Exploiting differing seismic reflection characteristics, with validation from lithology and fluid modeling of well log data, AVO workflow practitioners attempt to estimate reservoir fluid content and lithology.

Modern seismic inversion techniques attempt to streamline the property modeling process by automatically inferring major rock and fluid effects. Techniques ranging from "colored inversion" through to model-based and stochastic inversion approaches all provide means to extract information more tuned to rock and fluid properties. New techniques for simultaneous inversion of "traditional" or full-wave seismic data offer further opportunities to refine estimates of velocities and density associated with rock and fluid properties.

Building more realistic and high-resolution property models that directly honor the seismic information is accomplished by calibrating the seismic information with the well information. Forward modeling of model-derived impedance, density or sonic information from well logs produce synthetic seismograms that optimally tie the 3-D seismic volume, thus producing highly accurate property models at the well log scale for facies modeling.

High-resolution fluid and/or lithologic facies models then use these seismically tied property models to accurately distribute facies trends in 3-D, choosing from a broad set of geostatistical algorithms

Improved dynamic reservoir understanding

Time-lapse or 4-D seismic technology has proven to be an effective tool for measuring dynamic reservoir changes in certain types of reservoirs. This technology is particularly applicable in situations where AVO techniques have proven effective. With improved repeatability of seismic acquisition and processing, 4-D seismic will continue to grow in effectiveness and quantitative accuracy. Full-wave seismic, as previously described, offers a further level of effectiveness for repeated seismic surveys.

However, the greatest potential for improved dynamic reservoir modeling likely lies with better integration of 4-D seismic data with both historic and predicted reservoir production. The significance of this transition is to move workflows from a largely qualitative analysis of reservoir changes to more detailed, quantitative monitoring of fluid movements and phases within the reservoir.

Emerging software tools are leading the way, simplifying multivolume comparisons and bringing together data measuring natural and induced fracture patterns, as illustrated in the left panel of Figure 2. Curvature seismic attributes, as discussed previously, can support discrete fracture network modeling and provide valuable information for planning hydraulic fracturing, both leading to a better understanding of local and global permeability. Composite display of different seismic data, using primary colors, is also illustrated and is an essential technique for multivintage seismic comparisons.

Seismic of a very different form is beginning to have an impact upon the understanding of fracture flow patterns and well drainage extents within certain reservoirs. Passive seismic monitoring of reservoirs, from both surface and downhole receivers, has proven to be useful in mapping natural and induced fracture patterns associated with phenomena like reservoir compression in the North Sea and hydraulic fracturing operations in tight reservoirs. Passive seismic technology may offer a powerful complement to 4-D seismic techniques by measuring local porosity and permeability changes within the reservoir. Figure 2 illustrates ways in which microseismic events can be integrated with both seismic attributes and reservoir simulation models, to infer fracture patterns and related permeability effects.

The road ahead

Significant technology advances continue to be made in the seismic arena which should translate into better reservoir models. However, despite having acquisition and processing techniques that dramatically increase seismic resolution, legacy software tools are lacking in capabilities to fully support new workflows. New software innovations are required to integrate and interpret data ranging from 4-D seismic through multi-attribute volumes to microseismic and other information. To address a range of E&P trends, this new generation of software tools will have key attributes, including connectivity to multiple data sources, built-in learning support, extensibility for third parties, automation of tedious tasks and rapid performance. As the E&P industry retools, traditional workflow barriers will be removed, allowing easy movement across: data access, seismic attribute creation, time/depth correlation, structural framework and 3-D grid generation, property modeling, and dynamic reservoir modeling.

For more information, visit www.transformsw.com