Methodology for Quantifying Uncertainty Versus Data Spacing Applied to the Oil Sands

ABSTRACT The inevitable uncertainty in modelling spatial variables decreases as more data become available. This reduction depends on the properties of the variable being modelled. Spatially continuous variables show less uncertainty for a given spacing than erratic variables. A geostatistical simulation based approach is developed to quantify the relationship between uncertainty and data spacing. Reference realizations are sampled to condition additional realizations and quantify uncertainty. The procedure is automated, making it straightforward to consider sensitivity and different geological domains. A case study of the relationship between uncertainty and data spacing for net oil sands thickness data from northern Alberta is presented. _ KEYWORDS Simulation, Sampling,