Seismic characterization of Rotliegend reservoirs: from bright spots to stochastic simulation

A number of investigations within the North German Basin have shown the impact of 3D seismic on reliable reservoir characterization (Budny 1991; Hartung & Schwartz 1993; Trappe et al. 1996; among others). In general reservoir parameters like porosity or porosity thickness are linked to seismic attributes derived from the 3D seismic data. For Rotliegend reservoir characterization the seismic amplitude and acoustic impedance attributes are most commonly chosen. This study uses geostatistical methods to combine reservoir parameters (porosity, porosity thickness) and seismic attributes (e.g. amplitude, impedance, etc.) (Doyen 1988, Schultz et al. 1994). To integrate all information the following procedure was applied: . Analysis of seismic attributes (amplitude, frequency, impedance, etc.) and reservoir parameters (porosity, porosity thickness) and their crosscorrelation. . Processing of input data (declustering, normal distribution transformation, resampling, smoothing) and checking for trueness of parameter distributions. . Geostatistical analysis and variogram modelling of seismic attributes and reservoir parameters. . Kriging of the reservoir parameters (wells only). Cokriging of seismic and well information. . Stochastic simulation to investigate alternatives and extreme realizations. The paper will show that the above steps lead to a detailed quantitative description of pore volume in the producing units. A new reserve estimation can now be used for optimizing the economic output. From a reservoir engineering point the distribution of the reservoir parameters is of great importance. The distribution of porosity and other parameters is needed for dynamic simulations of the reservoir. This can help to define flow barriers and areas that are poorly reached by the wells. The reservoirs of interest are Rotliegend sandstones within the North German Basin (Fig. 1) at depths of 4500±5000 m. The study area is covered by about 2500 km2 of 3D seismic. Quite a few wells have been drilled in the area. For the study about 65 wells have been selected that have tested the reservoir.