J-M Kendall

Fracture characterization at Valhall: Application of P-wave amplitude variation with offset and azimuth (AVOA) analysis to a 3D ocean-bottom data set

The delineation and characterization of fracturing is important in the successful exploitation of many hydrocarbon reservoirs. Such fracturing often occurs in preferentially aligned sets; if the fractures are of subseismic scale, this may result in seismic anisotropy. Thus, measurements of anisotropy from seismic data may be used to delineate fracture patterns and investigate their properties. Here fracture-induced anisotropy is investigated in the Valhall field, which lies in the Norwegian sector of the North Sea. This field is a chalk reservoir with good porosity but variable permeability, where fractures may significantly impact production, e.g., during waterflooding. To investigate the nature of fracturing in this reservoir, P-wave amplitude variation with offset and azimuth (AVOA) is analyzed in a 3D ocean-bottom cable (OBC) data set. In general, 3D ocean-bottom seismic (OBS) acquisition leads to patchy coverage in offset and azimuth, and this must be addressed when considering such data. To overcome this challenge and others associated with 3D OBS acquisition, a new method for processing and analysis is presented. For example, a surface fitting approach, which involves analyzing azimuthal variations in AVO gradients, is used to estimate the orientation and magnitude of the fracture-induced anisotropy. This approach is also more widely applicable to offset-azimuth analysis of other attributes (e.g., traveltimes) and any data set where there has been true 3D data acquisition, land or marine. Using this new methodology, we derive high-resolution maps of P-wave anisotropy from the AVOA analysis for the top-chalk reflection at Valhall. These anisotropy maps show coherent but laterally varying trends. Synthetic AVOA modeling, using effective medium models, indicates that if this anisotropy is from aligned fracturing, the fractures are likely liquid filled with small aspect ratios and the fracture density must be high. Furthermore, we show that the fracture-normal direction is parallel to the direction of most positive AVO gradient. In other situations the reverse can be true, i.e., the fracture-normal direction can be parallel to the direction of the most negative AVO gradient. Effective medium modeling or comparisons with anisotropy estimates from other approaches (e.g., azimuthal variations in velocity) must therefore be used to resolve this ambiguity. The inferred fracture orientations and anisotropy magnitudes show a degree of correlation with the positions and alignments of larger scale faults, which are estimated from 3D coherency analysis. Overall, this work demonstrates that significant insight may be gained into the alignment and character of fracturing and the stress field variations throughout a field using this high-resolution AVOA method.

Imaging induced seismicity with the LoFS permanent sensor surface array

Here we use data from the Life of Field Seismic (LoFS) surface array in conjunction with a migration style approach to locate microseismic events in the Valhall field during 6.5 hours of hydraulic fracturing. The method is fully automated and requires minimal user input. This makes it ideal for application to data from surface arrays such as the LoFS, where signals from individual microseismic events are not visible in recordings from a single geophone. We initially identified 1111 events in the migration results including a major group consisting of sources both within and above the reservoir, forming a planar feature with near vertical dip. Cluster analysis was used to separate the 361 events forming this feature from the parent dataset. Events were recorded throughout the injection period and reveal a large peak in seismicity after about 4.75 hours, which is synchronous with an increase in the downhole pressure. Using a variety of synthetic tests we examined the resolution and robustness of our results. Modest perturbations (±5%) to the 3D velocity model used to locate events primarily cause a static shift in the event depths. However, using a homogenous model resulted in a considerable reduction in the imaging power. We also performed a series of semi-synthetic tests (synthetic data with real noise) to establish the minium SNR requirements for event detection. Events with SNR’s greater than 0.25 are detectable by our imaging approach, this limit is related to the nature of the coherent noise in the data. Experiments with double-couple sources showed that movements on faults consistent with the planar event distribution are unlikely to be imaged by our procedure. Hence we believe that the sources observed do not have mechanisms consistent with their planar distribution and/or they contain a significant nondouble couple component. One possible interpretation of the events is that elastic or pressure wavefronts emanating from the well trigger activity on aligned faults in and around the reservoir where the in situ stress is close to that required for failure.

Influence of fault transmissibility on seismic attributes based on coupled fluid‐flow and geomechanical simulation

In this study, we present results from linked fluid–flow, geomechanical and seismic modeling to examine the influence of fault transmissibility on seismic attributes. The model is a graben structure with two normal faults subdividing a sandstone reservoir into three compartments. The predicted seismic traveltime differences are consistent with reservoir compaction and overburden extension. For the case of high transmissibility we observe a large spatial extent in traveltime anomalies as well as a fault related stress guide effect. For lower transmissibilities, the influence of production on seismic attributes becomes more localized around the well reservoir compartment. Anisotropy predictions show perturbations associated with the faults as well as the production induced stress redistribution in the overburden.

Seismic Anisotropy as an Indicator of Reservoir Quality in Siliciclastic Rocks

F028 Seismic Anisotropy as an Indicator of Reservoir Quality in Siliciclastic Rocks J-M. Kendall* (University of Bristol) J. Maddock (University of Leeds) S.A. Hall (3S Laboratoire Grenoble) & Q.J. Fisher (University of Leeds) SUMMARY We present results from an integrated project that explores the potential to use observations of seismic anisotropy to interpret lithologic and fluid properties (the SAIL project). Our approach links detailed petrofabric analyses laboratory measurements of seismic velocities in core samples and reservoir-scale seismic observations. We consider a suite of sliciclastic reservoir rocks that exhibit anisotropy on three length-scales: the crystal grain and fracture scale. Crystal-preferred-orientation (CPO)