Assessing the value of 4C and 4D seismic data
Stephen A. Hall and Colin MacBeth
Reservoir Geophysics Group, Heriot-Watt University, Scotland
Two of the most significant, recent developments in seismic acquisition and processing technologies are repeat (time-lapse) surveys and multi-component ocean bottom seismic (OBS). These advances have given rise to many new avenues for interpretation of reservoir properties throughout the reservoir region and over time. Here we assess the value of these new technologies with respect to the trade-off between the additional information that may be gained and the new challenges that are introduced.
4D-4C seismic
Time-lapse reservoir characterisation is the comparison of measurements taken at different times, usually before, during and after production of a reservoir; this often referred to as 4D. However, 4D is strictly 3 spatial dimensions and one in time. Such 4D analysis is only possible in reservoir characterisation with 3D seismic data (providing a full 3-dimensional image of the reservoir region away from areas characterised by wells), which is repeated over time. 4D measurements have primarily been utilised to provide assessment of fluid movements in a reservoir and more recently workers have been investigating other production-related changes such as pressure and saturation. Four-component (4C), is associated with ocean-bottom seismic acquisition and describes acquisition using three orthogonal geophones (mutually perpendicular with one vertical and two horizontal phones) and a vertical hydrophone. This acquisition allows shear-wave energy to be recorded, e.g, from converted reflection or transmission. Multi-component data acquisition originally gained popularity in hydrocarbon exploration onshore during the mid 1980’s but this interest has tailed off due to the inability of the data to deliver the promised enhanced images and reservoir characterisation. During the second half of the 1990’s multi-component acquisition re-emerged in the marine environment, with the advent of ocean bottom seismic in the form of cable (OBC) or node acquisition (OBN). The first 3D-4C surveys were acquired around 1997 and have proven successful in imaging in difficult areas e.g., where there is shallow gas such as at Valhall. However we are now at a critical time where the 3D-4C technologies must prove themselves; the avenue for this lies primarily with the need for improved resolution of reservoir properties that is required in time-lapse monitoring of reservoir production.
Applications of 4C data
The primary objective of 4C acquisition has been imaging in difficult situations, e.g., using converted waves in the presence of shallow gas (e.g., Valhall and Tomalieten) or for low P-wave impedance-contrast reservoirs (e.g., Alba); P-wave image improvement (“true-3D” data; P-Z summation, for multiple attenuation); sub-salt and sub-basalt imaging. Many of these primary objectives are relevant to specific reservoirs where imaging is difficult. Secondary objectives of 4C acquisition are: lithology/fluid prediction, e.g., discrimination between sand and shale and quantification of P-P bright spot anomalies; determining reservoir properties such as saturation; characterising fracturing (e.g., orientation and intensity). These secondary goals are those that are a primary concern in time-lapse monitoring and are thus rapidly becoming a more significant motivation for the application of 4D-4C.
Positive aspects of 4D-4C: reduction of uncertainty
Rich and un-tapped source of information for lithology and fluid discrimination from both post-stack and pre-stack data - for example, changes in density and VS may be more directly accessible from P-S AVO. Furthermore, combining P-S data with P-P data (and using two possible shear modes in conjunction) could lead to a reduction in the non-uniqueness of many characterisation challenges. Likewise repeat surveys can reduce the uncertainty in many reservoir properties. For example combining 4D and 4C and could help to distinguish pressure from saturation, better quantify possible gas zones and assess fluid flow. Sand-shale discrimination is a good example of this potential. 4D-4C data also allows better quantification and characterisation of seismic anisotropy through azimuth-offset analyses and shear-wave data (e.g., using shear-wave splitting).
“True 3D” data – OBS acquisition strategies, e.g., orthogonal shooting, allow true 3D imaging plus correct AVO or velocity analysis. For example, if spatially dependent azimuthal anisotropy exists streamer data will have a bias (inaccuracy) due to the sail-line azimuth; 3D OBS allows correct analysis by considering the full 3D data.
Good noise reduction techniques – e.g., dual sensor suppression for receiver-related multiples, up/down separation, repeat surveys.
Permanent receivers may make repeated surveying economically possible but must be carefully designed; could reduce noise levels.
Negative aspects of 4D-4C: sources of uncertainty
Lack of repeatability and vector infidelity - sediment-instrument coupling is unpredictable and uncertain, waves are not easy to separate due to infidelity. Deployment of receivers is a problem.
Cost of acquisition - the best receiver deployment procedure is still being evaluated being a trade-off between data quality/repeatability and cost reduction. However, analysis of early 3D OBS surveys indicates a need for more channels and reduced spacing. Currently the sparse acquisition strategies, due to cost issues, reduces the value and potential gains plus increases the processing difficulties (see below).
New processingchallenges - Sparse receiver distribution requires careful planning of the positions to avoid acquisition footprint and produces new challenges even in standard processing. Converted wave processing requires a more complicated set of procedures than P-P data, e.g., rotations, receiver statics, and common conversion point binning that varies with depth and is updated based on revised velocity analysis. Furthermore non-hyperbolic terms in the moveout equation are more significant and there is a greater susceptibility to anisotropy effects. However, these factors also highlight the potential of P-S data due to the enhanced sensitivity to rock properties. DMO and migration is often problematic for P-S data and may also be challenging for P-P data with current 3D OBS acquisition strategies.
Interpretation - how to match P-P and P-S images to gain the benefits of a joint interpretation; development of new techniques and algorithms is needed to understand the P-S post-stack and pre-stack results.
Conclusion
On balance we should move towards 4C acquisition and combined processing of P-wave and converted shear waves because in the near future there will be:
Increased need for constrained reservoir information only available from 4C seismic – ongoing advances in acquisition and processing (improved noise levels, better-designed experiments, smart wells, high-resolution velocity measurements, anisotropy) will allow this.
More cost-effective multi-component acquisitions and permanent installations – permanent installations and repeat acquisition reduces the overall cost of 4D (and 4C) and could reduce uncertainty through improved signal:noise.
More time-lapse activity, with a new 4C-4D era being possible over the next few years (precursors from consortia - Vacuum field, Teal South). Time-lapse analysis will require increasingly more accurate information with reduced (and quantified) uncertainty; multicomponent processing and analysis could provide the additional data to allow this to happen.
However many challenges still exist in the interpretation of multicomponent data (P-P and P-S) not least of which is calibration of P-P and P-S data/analysis.