Quality Control and Integration of Multi-Source Seismic Data Using Geostatistics

Deraisme, Jacques R., Nicolas Jeannée, Geovariances, Avon, France

Geostatistics provide consistent approaches for the evaluation of gross rock volumes, reservoir modeling and assessment of the uncertainties across the exploration-appraisal-development cycle. To achieve that task it is of prime importance to collect and process all the data that bring some relevant information and, particularly, all data coming from inter­preted seismic. The counterpart is that a rigorous control of the validity of these data and of their combination has to be carried out. From basic statistics to factorial co-kriging this paper reviews different techniques that are available and can be incorporated in classical workflows. Basics about appropriate geostatistical methods are briefly reviewed (factorial kriging, collocated cokriging, kriging with external drift), putting into evidence the required prerequisites. The paper emphasizes the practical aspects, hints and pitfalls to be aware of at the different stages of the analysis. Illustrations using real examples related to time to depth conversion issues are provided for the following aspects: identification and filtering of acquisition artifacts, detection of outliers that can lead to a partial review of the seismic interpretation, comparison of seismic surveys with different resolution and coverage, com­bination of several seismic attributes into auxiliary co-factors, calibration to wells data.