Modelling how stochastic parameter ranges impact fault seal uncertainty: a review and worked example

Fault seal analysis tries to predict the length of hydrocarbon column that a fault can trap before capillary leakage occurs. Several tools and empirical relationships have been developed in the petroleum industry to aid this exercise. These assess cross-fault juxtapositions and allow an estimation of the fault gouge permeability and its potential seal capacity. This seal capacity is often based on the inferred clay content of the fault zone as this tends to correlate with capillary seal properties of a fault gouge. A plethora of parameters can affect a fault seal evaluation from the quality of seismic imaging and trap mapping through to the fault deformation mechanisms, and the distribution and properties of the resultant fault gouge that help provide the membrane seal. A review of some methods that may help frame the uncertainty in seal prediction is offered via a worked example. Stochastic modelling can help illustrate the impact that key parameters such as the stratigraphic template, Vshale estimate, fault throw and published algorithms may impose on the seal calculation. Predicted fault seal capacity tends to have a negatively skewed distribution and should be quoted as a range. It is recommended that the mode be quoted as the expected value.