LEU, WERNER, Geoform Ltd., Switzerland, BERND KLUG, Preussag Energie GmbH, Germany, ROLAND SCHEGG, Petroconsultants SA, Switzerland
Abstract: 2-D Basin Modeling in the North West German Basin - Influence of Reservoir Model on Gas Distribution
With a 2-D basin modeling study along a 60 km long north-south trending geoseismic section in the North German Basin, the gas distribution in Rotliegend reservoirs is analysed. The modeling includes detailed burial, thermal and migration history reconstruction in a structurally complex area. Gas charging of the northward dipping reservoir unit by a 2 km thick upper Carboniferous source interval is tested with several scenarios concerning the thermal evolution and the lithology distribution in the undrilled section. It will be shown that a detailed reservoir is not only essential for reservoir prediction but also a key parameter for a successful simulation of gas migration.
The greater study area belongs to the Pompeckj Block of the Northwest German Basin that was marginally affected by the Variscan thrust front. During the Namurian and early Westphalian thick coal bearing source intervals were deposited, that charged most of the known gas accumulations in this basin. A Lower Permian volcano-clastic episode was followed by fluvial and aeolian sedimentation of the prolific reservoir rocks of the Havel Subgroup (upper Rotliegend). The upper Permian Zechstein evaporites form the major seal of this clastic play. Enhanced subsidence during the Triassic and Early Jurassic trigged the major gas generation phase. Lower Triassic Bunter sediments form the second important reservoir interval. Throughout the Jurassic major traps formed in the overburden, related to doming and deformation of the Zechstein salt.The development of the regional unconformity in the Lower Cretaceous (Albian) was followed by renewed subsidence, culminating in the Tertiary. Maximum burial depth are reached at present day.
Numerical basin model
The modeling was carried out using the finite difference program BasinMod® 1D and -2D. Burial history modeling provides the basic framework for subsequent thermal, hydrocarbon generation and migration simulation. Pore pressure and fluid flow from compaction combined with topography driven water flow are used in the program through effective stress/porosity relations. Migration is calculated as a 3-phase flow of water, oil and gas. First-order kinetic models were used for the generation of oil, methane and nitrogen.
Input data and calibration
The present day geometry of the modeled transect is mainly based on depth converted seismic sections. Stratigraphic data of nearby wells was used to define the lithological model and to extrapolate the section to a common base at the bottom of the Namurian. Missing section in major unconformities (base Cretaceous and base Permian) was reconstructed from seismic data in the greater study area.The deformation of the salt dome in the central part of the cross section (Fig. 1) was simulated in a simplified way, by assuming normal slip along major boundary faults. In a first step the pore pressures were calibrated using test data of Wells A and B in the southern part of the section. Numerous bottom hole temperatures and top-to-.bottom vitrinite maturity profiles provided means to calibrate the thermal history. A sensitivity analysis demonstrated that variations in the heat flow history, as well as the amount of missing section below major unconformities are only of limited importance for the final gas saturation distribution.
Gas migration and accumulation
Main gas sourcing phases from Westphalian/Namurian coals are modeled in the Triassic/Early Jurassic and Late Cretaceous/early Tertiary. Moderate overpressures of up to ~100 bars result at present day in the southern part of the section throughout the Rotliegend reservoir unit below the main Zechstein seal. Migration direction is mainly vertical with subordinate horizontal components in distinct carrier beds. By the end of the Jurassic the main trap below the salt dome is already filled to a maximum with gas. Thereafter gas saturation is locally decreasing due to seal leakage. After the Cenozoic generation phase the entire Rotliegend section shows a gas saturation of 60 - 70 %. This is mainly a result of the homogenous reservoir properties assumed in this simple reservoir model (Fig. 2).
Implications of detailed reservoir model
Based on drilled sections combined with high resolution sequence stratigraphic concepts the Rotliegend lithology model of the target reservoir zone was refined for a second simulation series. This more realistic model included individual sandstone units, bedding parallel evaporite stringers acting as regional vertical migration barriers and a general increase of shale percentage towards the North. The petrophysical properties of the defined lithology mixes were further adjusted to match modeled and measured porosity's and pore pressures near wells.
The resulting present day gas saturation along the section shows distinct differences to the simple reservoir model (Fig. 2). High values are only modeled in well defined and good quality reservoir intervals in the southern part of the section. The lithological variations strongly influence the trap filling evolution. Especially the intra-Rotliegend evaporite seals have a positive effect on gas quality in this area characterized by a generally N2 prone gas charge.
The results demonstrate that meaningful simulations of fluid migration and accumulation to predict gas quality in undrilled structures have to be based on detailed lithology models of the main carrier and reservoir beds.
This publication and the
presentation of the results was made possible by the generous release of
proprietary data by Mobil Erdgas-Erdöl GmbH (Celle), BEB Erdgas GmbH
(Hanover) and Preussag Energie GmbH (Lingen).
Figure 1. Stratigraphic input cross section used for the numerical simulations.The section is based on seismic, well data and high resolution sequence stratigraphic concepts.
Figure 2: Modeled present day gas saturation resulting with detailed reservoir facies model for Rotliegend interval. Compare the difference to the simple reservoir model case (inset below).
AAPG Search and Discovery Article #[email protected] International Conference and Exhibition, Birmingham, England