An Integrated Approach to Reservoir Typing and Characterization of a Complex Deepwater System, Stampede Field, Green Canyon, Gulf of Mexico

Abstract

The main reservoir interval in the middle Miocene Stampede Field is both unique and complicated: unique, as it's one of the deepest fields in the GOM (~27,000’ BML) and complicated because of its 2 orders of magnitude range in permeability, mineralogical influences on wire-line logs, elevated water saturations (>50%) and cored wells that display significantly better quality than poorer quality non-cored wells. In addition, state-of-the-art dual coil 3D seismic data provides no uplift to define reservoir architecture in this single cycle seismic reservoir. The prior attempt to type facies was done by training facies relationships in the cored wells with the multi-resolution graphical clustering method and then propagating relationships to the un-cored wells. The previous method was improved upon when the team began using a new multi-disciplinary tool, Geo2Flow. Originally employed to identify reservoir compartments using J-functions, we detected trends that represented facies types. By plotting well data in J-function versus Sw space it's immediately clear that there was a poorly understood rock type that was 1. Under-represented in the cored intervals 2. Abundant in the non-cored wells and 3. Not well defined with the existing facies scheme. By integrating all scales and types of the core datasets, we hypothesized that the abundant, poorer quality facies was deposited in a setting that allowed the flow to quickly collapse, trapping a higher proportion of fines, lithics and feldspars that were transformed during diagenesis into authegenic clays, further deteriorating reservoir quality and contributing to the higher water saturations and affecting log response.

AAPG Datapages/Search and Discovery Article #90216 ©2015 AAPG Annual Convention and Exhibition, Denver, CO., May 31 - June 3, 2015