engleski

The Importance of Selected Geological Variables in Characterization of Mature Hydrocarbon Fields

Selection of geological variables for characterisation of the mature hydrocarbon fields is crucial for later outputs. The very first task is selected appropriate values for variable, removing at least (a) outliers, (b) data with known measurements error. The next is collected at least 20-30 point data for selected variable or two variables that can be correlated. The collected dataset of geological variables can be analysed independently for each variable dataset, making its own statistics and maps. However, sometimes variables can be significantly correlated and data applied together (like in cokriging mapping). How geological variables are measured on different geographical and physical scales, often is only appropriate correlations of ranks. It means that linearity and regression can not be always “key tool” for establishing connections between reservoir geological variables. The mapping is crucial process in reservoir characterisation, what means that each selected geological variable would need to be mapped. For variable with enough point data, deterministical geostatistical approach is recommended (kriging). If correlation is observed between two variables deterministic approach, numerically “dependant”, is advised (cokriging). However there are some variables, which in reality do not have enough large sampled dataset to be characterised with “representative” statistics (variance, average, variogram…), i.e. statistics valid for population of variable. Those are usually variable that representative petrophysical or depositional properties of reservoir (like porosity or thickness). In such case, geostatistical stochastical mapping (sequential Gaussian simulation, sequential indicator simulation) is welcome improved of deterministical geological model.

Geological variables; mature hydrocarbon fields; sandstone reservoirs; depositional environment; geological model

nije evidentirano