Andreas Kammer

Fractured reservoirs in the Eastern Foothills, Colombia, and their relationship with fold kinematics

Analysis of fracture systems in subsurface structures is limited by the amount and uncertainty of available data. With the aim of analyzing the distribution of fracture systems, we studied surface structures as analogs for oil fields in the fractured reservoirs of the Llanos foothills of Colombia. Here, we document the presence of four widespread fracture systems whose distribution is related to fold geometry and folding mechanism. At surface, in the Tierranegra and Silbadero anticlines, the principal fracture systems are symmetrical with respect to northeast- and northwest-trending fold axes, showing higher fracture intensities in the forelimbs of the structures. In the Guavio anticline, higher fracture intensities are located in the backlimb, with principal east–west and northwest–southeast directions. In contrast, we document northeast–southwest fractures near the hinge zones in the adjacent synclines. This distribution suggests that in the Guavio anticline, fractures respond to movement of the hanging-wall above a ramp, consistent with a fault-bend-fold model. Whereas, in the Tierranegra and Silbadero anticlines, fractures respond to limb rotation and hinge migration consistent with detachment fold models. Comparing these with subsurface structures, we identified that El Morro anticline has fracture distributions like those in the Tierranegra and Silbadero anticlines, but have higher fracture intensities. In the case of the Cusiana Structure, fracture intensities are higher in the crest but not in the limbs, and intensities differ from the ones found in the Guavio anticline, showing that these structures are not appropriate analogs. The results show how fracture distribution depends on structural position and fold evolution, and is controlled in part by folding mechanism. This suggests that models based on Holocene fold geometry cannot accurately predict the observed fracture distributions and should not be used to construct discrete fracture network models. Instead, the patterns we describe can be used as a guide for similar structures. Our work illustrates the possibility of having different fracture patterns and fracture abundances in adjacent folds in the same fold-thrust belt.