Source rock and reservoir rock properties are often poorly constrained in the exploration for oil and gas. Geological analogues may be used to create source rock and reservoir rock models in areas with only sparse well data coverage. Exploration risks are often influenced by uncertainties in the rock facies models. This paper outlines a methodology for assessing the consequences of using two or more source and reservoir models on hydrocarbon phase distributions in prospects.
Each source rock and reservoir rock scenario is represented by computerized maps (grids) of thickness, porosities and other properties. The uncertainties of each scenario are described using probabilistic properties. The reservoir thickness may be represented by a uniform distribution with the mean described using a map (grid) with a standard deviation of, for example, 50 m. A Monte Carlo simulation is carried out for this part of the petroleum system and the resulting probabilities of oil and gas in prospects are compiled by weighting each run to calibration wells. The results can be plotted as a map of most likely oil and gas column heights. An uncertainty map is also plotted. The results can be used to rank drilling locations with respect to large oil columns and low uncertainties. The oil and gas column probabilities can be estimated for each well location.
As (more) wells are drilled within a petroleum system the best match simulations can be used to further refine predictions as well as to create an updated map of the petroleum system.