Factorial Experimental Design for Quantitative Uncertainty Assessment of Bounding Fault in Rift Basin

This paper focuses on quantitative characterizing the uncertainty of bounding fault in rift basin, especially in the field with seismic dataset of poor quality. An assessment method based on factorial experimental design is presented to analyze the influences of bounding fault uncertainty on gross reservoir volume (GRV).

The new approach is based on a series of flexible 3D geological model cases. A three levels factorial experiment design is used to evaluate the effects of three factors considered in the calculation of GRV. The response surface function obtained to find the best inherent relationship between the fault parameters and GRV by a nonlinear equation.

The high correlation coefficient proves the validity of the model. Results show that GRV has a significant linear relationship with the lateral displacement distance of the bounding fault, and GRV also shows a less strong linear relation dip change of the bounding fault. However, GRV is not correlated with strike change, even the strike change will lead to a slight increase in GRV.

The novelty of this new approach is in the ability to quantitatively assess the uncertainty of bounding fault even the seismic evidence is insufficient to depict fault features.