This paper demonstrates the value of seismic data to define heterogeneous earth models for tight shales, improve reservoir quality and completion quality mapping, and to guide decisions of wellbore placement and completion design. The process includes quantitative classification of rock units at log-scale, definition of sample locations for laboratory testing, characterization of all dominant rock classes, and development of core-to-log transforms for all measured properties. The process also includes classification of the seismic data, based on the log-derived model, and development of class by class rock physics models for populating measured properties across the seismic volume. The quantitative relationship between core, log and seismic measurements allows population of the seismic model with complex properties. The result is a Heterogeneous Earth Model with rock classes representing the large-scale heterogeneity of the geologic system, and rock properties that characterize the behavior of each of the classes. By defining the variability of groups of properties representing reservoir quality and completion quality across the seismic cube, one can map changes in them and use this information to define reservoir potential, landing points and completion design. A case study from the Haynesville field is presented.
Conventional seismic analysis methods generally lack the ability to account for the heterogeneous nature of tight shale systems. Consequently they can provide predictions of properties that while representing the mean behavior of the system, may have a significant error at a local level. Additionally, conventional analysis is often based on rock physics models that are limited to acoustic stimuli, and fail to relate many of the complex factors that influence unconventional reservoir and well performance.
The method presented here is applicable to heterogeneous, organic-rich, oil and gas mudstone systems. Results lead to a well characterized Heterogeneous Earth Model, significantly improved mapping of variability in reservoir quality and completion quality, and better assessment of the opportunities and problems for exploration and long-term field development. This has already resulted in successful application by an operator in the Haynesville/Bossier formations in NW Louisiana, USA.