Print this pageAAPG ANNUAL CONFERENCE AND EXHIBITION
Making the Next Giant Leap in Geosciences
April 10-13, 2011, Houston, Texas, USA
3-D Geologic Modeling Toward a Site-Specific CO2 Injection Simulation
(1) Earth & Atmospheric Sciences, University of Houston, Houston,, TX.
A solid geological model at reservoir scale is the key starting point toward a site-specific characterization of a CO2 sequestration target. In the Dickman Field of Ness County, Kansas, a 3D structure and property model was built for depleted reservoirs of carbonates and clastic rocks through multi-scale data integration. Work flows were designed to handle some of the challenges commonly involved in geological modeling at the reservoir-scale: targeting geological features normally considered as “sub-seismic” and beyond the resolution of conventional seismic stratigraphy; recognizing the lateral heterogeneity in acoustic properties of laterally interwoven clastics and carbonate lithologies on a karst-modified paleo-topography to restore true subsurface geometry; calibrating legacy well logs to obtain reservoir properties with quantified risk assessments; and extracting a fault-fracture framework from multiple seismic attribute volumes to guide the reservoir property gridding.
As a first step, a depth-converted stratgraphic model was established and validated by log interpretations at 17 well sites. Fault and fracture analysis was based on seismic interpretation and volumetric attributes, supported by log and core evidences and understanding of the regional deformation history. A unique set of porosity was assigned to the stratigraphic model through calibrating porosity logs of different types and correlating log to core measurements. Permeability estimation was based on core measurements available in Dickman and the surrounding oil fields. Water saturation measured from flushed cores was calibrated to the in-situ water saturation. The propagation of these reservoir properties through the model was along preferred orientations guided by fracture and acoustic impedance analysis. The resulting property grid was tested by production history-matching simulation. A reasonable match was obtained after two rounds of input parameter adjustment and the inclusion of a capillary zone in the model.
The initial geological model built from heavily drilled reservoirs was extended to deeper saline aquifers with only three well controls, aided by 3D seismic impedance analysis. The grid served as input to CO2 injection simulations for the deep saline aquifer, a potential carbon capture and sequestration target.