Most basin modeling programs apply strictly deterministic numeric methods to quantify physical and chemical processes during the evolution of a sedimentary basin. This leaves the user at present with the manual performance of systematic sensitivity and uncertainty studies, which is usually a time-consuming, user-dependent, and practically impossible procedure.

In this chapter, the use of stochastic techniques in basin modeling is evaluated. As an example, stochastic simulation of hydrocarbon expulsion is demonstrated using Norsk Hydros QuickVol3D. Hydrocarbon quantities that are expelled from mature source rocks are calculated given the spatial uncertainties of the initial thickness and quality of the source rock as well as the temporal and spatial uncertainty of the source rock maturity. A synthetic case study highlights the dependency of uncertainties of critical input parameters and geologic situations.

Fast stochastic algorithms are needed to perform quantitative sensitivity analysis and risking of petroleum system models. Results of such studies identify the most critical input parameters much more systematically than those of other methods. These results may ultimately be used to redefine further work flow and redirect data gathering in exploration for hydrocarbons, namely to focus on the most critical parameters for a given prospect.