タービダイト層厚分布の統計学的解析 その貯留岩キャラクタリゼーションへの応用の可能性

Abstract

Future targets of petroleum exploration will shift to stratigraphic traps, specifically turbidite reservoirs which are often distributed in deep-water environments. Therefore the demand for reducing exploratory risks will increase in exploring such subtle traps. Quantitative and stochastic models will be necessary for reservoir characterization and distribution prediction of such traps. Statistical analysis of bed thicknesses of turbidites is one of the most important methods for this purpose.In order to quantify bed thicknesses, it is essential to know how the bed thicknesses are distributed. Bed thickness distributions of turbidites have been interpreted as log-normal, power-law, exponential or truncated normal distributions. The identification of distribution laws, however, has been often difficult to interpret due to scattering of data and lack of quantitative tests in previous studies. In addition, the origins of distribution laws have been controversial and there has been no satisfactory interpretation. Although quantitative assessments of erosion in turbidite successions may give a useful parameter for reservoir characterization, their methods have yet been clarified. The identification and quantification of segments observed in the bed thickness distributions may give important insights for understanding depositional systems and reservoir characterization. If both of the volume distribution and the bed thickness distribution of turbidites are assumed to obey power law, the volume distribution can be estimated from the bed thickness distribution. This result would have a significant implication for stochastic modeling of reservoirs.Nevertheless the incompleteness of statistical methods, the bed thickness statistics can give useful information for 3-D reservoir modeling and predicting reserves.