Pressure and Fluid Flow Systems in the Permian Rotliegend in the Netherlands Onshore and Offshore
Published:January 01, 2011
- PDF LinkChapter PDF
Hanneke Verweij, Erik Simmelink, Jim Underschultz, 2011. "Pressure and Fluid Flow Systems in the Permian Rotliegend in the Netherlands Onshore and Offshore", The Permian Rotliegend of the Netherlands, Jürgen Grötsch, Reinhard Gaupp
Download citation file:
Hydrodynamics-based approaches were used to characterise and analyse the present-day pressure and fluid-flow conditions in the Permian Rotliegend reservoirs in the Netherlands. These approaches involve the use of multi-well pressure– depth plots, regional all fluid overpressure maps, salinity maps, and hydraulic-head maps. The maps and plots revealed a general regional trend of, often stepwise, decreasing fluid overpressures from northeast towards the south. Values of fluid overpressure vary between hard geopressures (Pexcess > 40 MPa in block L2) and near-hydrostatic pressures (Pexcess < 1 MPa in southern offshore). The highest overpressures occur in a zone following the northern limit of the Permian Rotliegend reservoirs. The width of the zone of high overpressures extends southward into the onshore Netherlands in the area of the Lauwerszee Trough. The hydraulic-head map of the Rotliegend reservoir demonstrates the potential for a general southward dewatering direction.
The hydrodynamic evaluation identified that there are distinct regional differences between the southern and the northern part of the area with respect to important factors influencing both pressure generation (such as sedimentary loading and gas generation) and dissipation (by fluid flow) in the Rotliegend reservoir. The distribution of observed overpressures and hydraulic heads reflect these regional differences. We show that because the vertical and lateral dewatering of the Rotliegend reservoirs is controlled by the permeability framework, the regional variations therein exert a major influence on the observed distribution of fluid overpressure. Relatively high fluid overpressures are maintained in zones where dewatering of the Rotliegend is severely restricted. This is especially apparent in the southern part of the Dutch Central Graben and also in the northern part of the Lauwerszee Trough.
Knowledge of present-day pressure and fluid-flow systems in the Permian Rotliegend is of critical importance for optimising exploration, appraisal, and production of its contained natural resources (hydrocarbon gases and geothermal energy), subsurface storage of hydrocarbon gases and CO2, and safe drilling. This is because pore fluids and pore-fluid pressures influence the physical and chemical properties of the subsurface (Ingebritsen et al., 2006; Neuzil, 2003; Tóth, 2009; Verweij, 1993, 2003; Zoback, 2010; and references therein). For example, pore pressures in the context of effective stress and poro-elasticity affects the strength of both intact and faulted rock; it influences the leakage potential of top seals and reservoir-bounding faults and as a consequence maximum hydrocarbon column heights and gas storage capacity in some reservoirs (Underschultz, 2007, 2009; Zoback, 2010). Process-based understanding and prediction of present-day distributions and physicochemical properties of rocks and fluids of the Permian Rotliegend, such as diagenetic cement composition, porosity–permeability, and excess fluid pressures, require fundamental knowledge of the evolution of fluid flow systems on geological timescales (Gaupp et al., 1993; Gaupp et al., 2004; Ingebritsen et al., 2006; Lanson et al., 1995; Lanson et al., 1996; Verweij, 2003).
Research on pressure and fluid-flow systems in sedimentary basins and the development and application of hydrodynamic analysis for purposes of oil and gas exploration in the Netherlands started in the late eighties (Verweij, 1990a, 1990b; Bloch et al., 1993). Continuation of such research was initially hampered by the lack of publicly available data on the deep subsurface of onshore and offshore Netherlands. This situation improved during the late nineties (Verweij, 2003; Verweij and Simmelink, 2002). Since 2002 a quality-controlled pressure, water-chemistry, and temperature database using public information from oil and gas wells was developed during joint industry projects (Simmelink et al., 2003; Simmelink et al., 2008). The increase in data availability stimulated more detailed analysis and characterisation of the pressure and fluid-flow conditions in the subsurface of the Netherlands.
Here we present and discuss the distribution of pre-production fluid overpressure, salinity, and hydraulic head in the Permian Rotliegend in the offshore and northeastern part of the onshore Netherlands.
The dominant features of the present-day structural framework and stratigraphic buildup of onshore and offshore Netherlands are summarised in Figure 1. Figure 2 shows the main structural elements in the area. The basins, platforms, and highs reflect the complex history of extension and compression related to world-wide reorganisations of lithospheric plates (Ziegler, 1990; Duin et al., 2006; De Jager, 2007).
A major Variscan transpressive deformation event in Early to Middle Permian induced strong uplift and erosion of Carboniferous strata. Rotliegend sedimentary sequences now overlie Carboniferous sequences separated by the base Permian unconformity, which is generally known as the Saalian unconformity. The Upper Carboniferous Westpha-lian coals and carbonaceous shales, the principal source rocks for gas in the Dutch subsurface, are preserved in a large part of the area. The post-Variscan history has been influenced by the older Variscan geological and structural configuration of the area. Many of the Variscan fault sytems have been reactivated multiple times during the Mesozoic and Cenozoic (Dirkzwager et al., 2000, Dronkers and Mrozek, 1991; Nalpas et al., 1995; Van Wijhe 1987, Ziegler, 1990). The Late Jurassic–Early Cretaceous, Kimmerian, ex-tensional tectonic phases shaped the Mesozoic basinal structures, and Late Cretaceous, Sub-Hercynian, compressional phases induced inversion of the Mesozoic basins. Figure 1 reveals the main unconformities in the Carboniferous to present sedimentary sequence that are related to the Saalian, Mid-Late Kimmerian, and Sub-Hercynian tectonic phases.
After the Saalian uplift and erosion, the Carboniferous was covered by Rotliegend, Zechstein, Triassic, and Early Jurassic sediments. The Slochteren Formation and the Silverpit Formation of the Upper Rotliegend Group are each other’s lateral equivalent. The Slochteren Formation comprises conglomerates and sandstones of fluvial and aeolian origin. Figure 3 shows the generalised facies distribution of the Upper Slochteren Member. The transition and interfingering of the Slochteren Formation and the Silverpit Formation (claystones, siltstones, and evaporites) occur in a relatively narrow east–west zone. In this zone, the Slochteren Formation splits up into the two Slochteren sandstone members that are separated by claystones and siltstones of the Ameland Member. The Lower Slochteren Member extends farther north into the Dutch Central Gra-ben than does the Upper Slochteren Member (Mijnlieff and Geluk, this volume) (Fig. 4). Eventually both Slochteren members grade northward into the siltstones and claystones of the Silverpit Formation. The marine Zechstein evapor-ites, carbonates, and clays overlie the Rotliegend. Large parts of the Dutch area have been covered by Zechstein evaporites (Fig. 5). The successive structural development in this region has been strongly influenced by the Zechstein salt deposits (Figs. 1, 5).
The Kimmerian tectonic phases induced a differential dynamic development of the main structural elements in the area: while sediments accumulated in the subsiding rift basins during Late Jurassic–Early Cretaceous times, the adjacent platforms and highs were uplifted and eroded (Figs. 1, 2).
During the Late Cretaceous, regional subsidence and widespread deposition of chalk occurred. Chalk deposition was interrupted in the Mesozoic basins by the initiation of inversion movements related to the Sub-Hercynian tectonic phase. The central parts of the Broad Fourteens Basin and the West Netherlands Basin experienced severe inversion. Inversion of the Dutch Central Graben was strongest in its southern half. Deposition of chalk did not cease completely in the inverted Vlieland Basin. While the sedimentary fill of the basins became folded, uplifted, and subject to erosion, deposition of chalk continued in adjacent areas (such as Noord Holland Platform, Winterton High, Central Offshore Platform, Cleaver Bank High, Friesland Platform, and Schill Grund High).
The rise of the land masses surrounding the North Sea in Mid-Palaeocene times in combination with the Mid-Palaeocene low stand of sea level induced a regional phase of erosion and terminated the deposition of chalk. This erosion phase was followed by regional subsidence and deposition of predominantly marine clays during Late Palaeocene and Eocene.
The regional subsidence was interrupted by the Late Eocene–Early Oligocene uplift, related to the Pyrenean tectonic phase. Pyrenean compressive tectonics involved uplift and erosion of the West Netherlands Basin but also extended offshore and was responsible for erosion in the southern part of the Broad Fourteens Basin. During the Oligocene, the sea transgressed across the previously eroded basins. This resulted in marine deposits of limited thickness. Sedimentation was interrupted at the end of the Oligocene, related to the Savian phase of erosion. In the Neogene, deltas prograded from the south and southeast of the Netherlands and from the Fennoscandia border zone into the North Sea Basin (Sørenson et al., 1997; Overeem et al., 2001). After Miocene times the main depocentres developed in the northern offshore area. Sedimentation rates started to increase during the Pliocene and remained high during the Quaternary. Figure 6 shows the net result of the Tertiary and Quaternary burial history.
The burial history of Upper Carboniferous and Rotliegend stratigraphic units located in different structural parts of offshore Netherlands, outside the strongly inverted southern part of offshore Netherlands, indicates that these units are at their maximum depth of burial at present. The burial history of the Broad Fourteens Basin and the Noord Holland Platform illustrate the differences in burial histories between inverted basins and platforms (Fig. 1).
Westphalian–Upper Rotliegend Gas System
The three main elements of the Westphalian–Upper Rotliegend gas system are the Westphalian source rocks, the Permian Rotliegend reservoirs, and the Zechstein evaporites as the ultimate top seal. The Slochteren Formation of the Upper Rotliegend Group is the most important gas reservoir in the Netherlands (Lokhorst, 1998; Gerling et al. 1999, De Jager and Geluk, 2007). In general, hydrocarbon generation was widespread until the Middle Jurassic. Hydrocarbon generation continued in the basins until the Late Cretaceous inversion. On the platforms and highs hydrocarbon generation was interrupted due to uplift and erosion in Late Jurassic–Early Cretaceous times. Reburial and associated heating of the Westphalian source rocks in basins, platforms, and highs beyond the temperatures previously experienced by the rocks induced a later phase of hydrocarbon generation in Cenozoic times (e.g., along the edges of the West Netherlands Basin; De Jager et al., 1996; Noord–Holland Platform, Central Offshore Platform, Cleaverbank High, Friesland Platform; Abdul Fattah et al., 2010; Verweij, 2003; Verweij et al., 2009a; Verweij et al., 2009b; Verweij et al., 2010; Verweij et al., in press).
In two successive joint-industry projects executed by TNO and CSIRO from 2002 to 2007, an Oracle relational database of quality-controlled data on pressure, water chemistry, and temperature was developed for the Dutch offshore and the northeastern part of the onshore. The database was filled with public information from about 700 oil and gas wells. In the projects, an existing database model and quality-control system, initially developed by CSIRO Petroleum for the Australian Petroleum Basins, was extended and modified for specific application to the Netherlands. At the end of the projects the quality-controlled database included about 10,000 pre-production pressures (90% derived from wireline formation tests), about 900 “leak-off” pressures, more than 9000 temperatures, and about 950 water salinity values. The leak-off pressure derived from real leak-off tests corresponds to fracture initiation. Most data in the database are from formation integrity tests and from reported values of leak-off pressure (assumed to represent the highest pressure achieved during testing before actual leak-off) and may underestimate the strength of the tested formations. The temperature database includes bottom-hole temperatures, drill-stem test temperatures, and Horner-corrected bottom-hole and drill-stem test temperatures. Salinity data were obtained from three sources: formation-water analyses, petrophysical log analyses, and analyses of formation pressure gradients.
The pressure and fluid-flow characterisation of the Permian Rotliegend reservoirs include the presentation and discussion of the wireline formation test pressures available for the Rotliegend, and the calculated overpressures, hydraulic heads, and salinities derived from a selection of data from offshore wells.
Hydrodynamics-based approaches were used to characterise and analyze the present-day pressure distributions in the Permian Rotliegend reservoirs. These approaches involve the use of single-well and multi-well pressure– depth plots, regional overpressure maps, and hydraulic-head maps. Bachu (1995), Dahlberg (1995), Verweij (1993), Otto et al. (2001), and Bachu and Michael (2002) provide an overview of hydrodynamic-analysis techniques.
Calculating and Mapping Fluid Overpressures
Overpressure values were calculated for “all fluids” (water, gas, condensate, oil) in the Permian Rotliegend reservoirs. The overpressure, or excess pressure, of a fluid at a certain depth is the difference between the measured pore-fluid pressure and the hydrostatic pressure at that depth. In the calculations we used a hydrostatic gradient of 0.01 MPa/m that represents seawater density (1020 kg/m3) (Fig. 7). In reality, the hydrostatic gradient is determined by the actual density of the formation water. The overpressure calculation based on the assumed seawater density can be considered to represent the lower bound of hydrostatic pressure in the Rotliegend reservoirs. Fluid overpressures incorporate the effect of the density contrast between a petroleum fluid and pore water (Fig. 7). For example, the gas pressure at the crest of a structure for a gas column of 100 m is Pgas = (Pgwc – 0.002 × 100) MPa (assuming a gas gradient of 0.002 MPa/m; Pgwc = fluid pressure at the gas– water contact), and as a consequence exceeds an assumed hydrostatic water pressure at that depth by 0.8 MPa (for a water gradient of 0.01 MPa/m).
Based on the calculated overpressure values at well locations, two types of maps were constructed. The first type displays overpressures at the wells for all fluid types, gas, oil, condensates, and formation water (Fig. 8). The second type of map is the hydraulic-head map, which is derived from the overpressures in the formation water only (see below).
The all-fluids overpressure map depicts a mixture of pressure points for several fluids. If two fluid types (gas and water) were observed in a well and both have pressure data, two overpressure values were registered and mapped. Although the separate overpressure points from different fluid types have no direct relation to each other, it was thought to be useful to plot the hydrocarbon overpressures on a map as well. This is justified especially in the wells where reliable data on gradients and fluid contacts are missing, which prevents calculation of a formation-water overpressure in such cases.
Calculating and Mapping Salinity
Analysis of data on formation-water chemistry and water resistivity from the database in combination with pressure-gradient analysis allowed construction of a map of formation-water salinity (Fig. 9). The water–salinity maps aid in the evaluation of the pressure and flow system.
Compilation of hydraulic-head maps requires knowledge on formation-water density. While formation-water analyses give the most information of the three data sources, the process of drilling introduces foreign material and drilling fluids to the subsurface prior to sampling the formation water, which often contaminates the formation-water sample (Gonzalez et al., 1998). For this reason culling criteria are required to determine if the formation-water sample truly represents the in situ formation water. Culling criteria described by Hitchon and Brulotte (1994), Hitchon et al. (1999), and Underschultz et al. (2002) were examined and modified to suit the conditions and lithologies of the Dutch subsurface. The culling process and removal of multiple analyses of sequential water samples from the same test interval resulted in a final data set of water analyses (of 106 values) to be used for mapping and formation-water analyses.
Wireline electric logs utilise the ability of an aqueous solution to carry an electrical charge that is dependent on the presence of ions in solution, their total concentration and valence, and temperature. The relations most commonly used account for the conductivity factors of Na+ and Cl− ions, resulting in a NaCl concentration equivalent (Pickett, 1966). While the NaCl-equivalent salinity obtained from analysis of electric logs can be an excellent estimate of formation-water salinity, one must be aware of the potential pitfalls. These include situations where the formation water is not NaCl-type water, where water resistivity (Rw) is estimated across variable lithologies, the assumption of a constant Rw over a large depth interval is invalid, the formation temperature is poorly constrained, or drilling fluid has invaded the formation to the point where it affects the log response (Underschultz et al., 2002). The source of data for this study was primarily well completion reports; therefore, prior knowledge of formation-water type is not known and a standard empirical relation for NaCl type waters was assumed.
If sufficient formation-water pressure data are available in vertical profile within a single reservoir unit, the vertical pressure gradient can be converted to formation-water density, and if the formation temperature is known, it can be converted to formation-water salinity (Rowe and Chou, 1970; Underschultz et al., 2002). In this way values of 84 formation-water salinity were obtained using a pressure-gradient method.
The most reliable salinity value was selected at each well in the reservoir from each of the three data types (water sample, log, or pressure-gradient analysis). The data were then posted on the map. Contours were constructed manually. All well data were honoured by the contours. In the event of more than one data type at a single well, a decision was made on which value to honour. In addition the contours were allowed to be influenced by faults. When contouring hydrodynamic data that represent a continuum (such as water salinity or hydraulic head) in faulted strata, the trend of the data within each unfaulted block should first be examined separately. If the trend in an unfaulted block is parallel or subparallel to the structural trend, then the fault can be represented as a discontinuity in the overall contour map for that surface (Underschultz et al., 2005). From the dataset of formation-water analyses, if all major ions were determined, a water type was defined using STIFF diagrams (Stiff, 1951). These water types are posted on the water-salinity map (Fig. 9).
Calculation and Mapping of Hydraulic Head
The distribution of hydraulic head in a certain unit represents the distribution of energy of the formation water in that unit and is directly related to the direction and magnitude of potentiometric driving forces for formation water flow within the unit. In contrast to fluid overpressure, which is based on measured pressures in oil, gas, or water, the hydraulic head is based on water pressures only.
The potential of formation water is determined by its place in the earth’s gravity field and by the pressure and density of the water. The water potential is more commonly described by hydraulic head—the elevation relative to datum—to which a water column would rise if an open well were drilled down to that unit. The use of hydraulic head implicitly assumes that water flow is approximately horizontal and parallel to bedding, and density variations are small and can be ignored. When these assumptions are not valid, the flow direction and flow magnitude can no longer be accurately obtained from a hydraulic-head map. To determine if density-related gravity effects can be ignored in a density-variable sloping stratigraphic unit, one can calculate the dimensionless driving-force ratio (DFR; Davis, 1987).
Values of hydraulic head were calculated for formation-water pressure data in the Rotliegend reservoirs using an average formation-water density of 1081 kg/m3. The average formation-water density was derived from data on water-sample analyses and from pressure-gradient analysis.
The mapped values of hydraulic head were contoured manually (Fig. 10). While the contours of hydraulic head honour the selected value at each well, they were also allowed to be influenced by faults.
In reality there would be a very steep gradient within the plane of the fault, but this is not possible to represent on a regional map. DFRs were calculated and mapped. The calculation requires information about the density contrast between the formation water and the reference density (1081 kg/m3), the slope of the structure overlying the Rotliegend reservoir, and the slope of the potentiometric surface. Locations where the contour map of DFR is greater than 0.5 are marked by a hatched polygon on the associated hydraulic-head map (Fig. 10). In these areas density-related effects cannot be ignored and flow directions cannot be derived from the hydraulic-head map.
Characterisation of Pressure and Fluid-Flow Conditions
Figure 9 shows the salinity distribution mapped for the Permian Rotliegend. While there is considerable tortuosity to the contours, there remains a freshening trend towards the onshore. In the Broad Fourteens Basin the salinity ranges from more than 300,000 mg/L in the north to less than 150,000 mg/L in the southern part of Q block. There is a rapid increase in salinity between the Broad Fourteens Basin and the Noord–Holland Platform, where values are generally more than 250,000 mg/L. Within the Rotliegend strata there is an interesting shift across the boundary between the north end of the Broad Fourteens Basin and the Cleaver Bank High, which is marked by a drop in salinity from more than 300,000 mg/L to less than 200,000 mg/L. This suggests that a hydraulic barrier may separate the Rotliegend strata in this area. The lower salinities on the Cleaver Bank High are mainly constrained by wireline-log-derived values, which are considered less reliable. With no reliable formation-water analyses to confirm the lower salinity, it leaves open the possibility of poor assumptions on the log analysis.
The hydrochemical composition of formation water in the Permian Rotliegend belongs to three basic water types: CaCl–SO4, NaCl–SO4, and CaMg–Cl. The anion chemistry is dominated by Cl, with a small proportion of SO4 and low HCO3. The cation chemistry is more variable, with either Na or Ca forming the dominant cation.
The NaCl–SO4 type of waters might be considered typical “oil-field waters”, where the dominant cation– anion pair is sodium chloride. It is this type of water that standard petrophysical-log analysis and pressure-gradient analysis that estimate “NaCl-equivalent salinity” assume to be present. Figure 9 shows that CaCl–SO4 and CaCl–Mg type waters are present in the Rotliegend strata, hence there is a risk of error in using standard petrophysi-cal analysis and pressure-gradient analysis to determine salinity and Rw.
The distribution of all fluid overpressures (Fig. 8) reveals a general regional trend of decreasing fluid overpressures from northeast towards the south. The values of fluid overpressure vary between hard geopressures in the north to nearly hydrostatic pressures in the southern offshore. Very high fluid overpresses were observed in block L2, in the southern part of the Dutch Central Graben (Pexcess > 40 MPa), and in the northern part of the Lauwerszee Trough (Pexcess >20 MPa). In general the highest overpressures occur in a zone following the northern limit of the Permian Rotliegend reservoirs. The observed maximum fluid overpressures in the northwestern part of the area (J and western K blocks, Cleaverbank High) are less than those observed in the northern parts of the Central Offshore Platform. The width of the zone of high overpressures extends southward into the onshore Netherlands in the area of the Lauwerszee Trough.
The southward decrease of fluid overpressures develops often in a stepwise manner. This is illustrated by Figure 11, showing the lateral change of pore-water overpressure on the Central Offshore Platform. The stepwise character of the change is also apparent from the large difference of overpressure values between those observed in the Permian Rotliegend reservoirs in the Lauwerszee Trough (reaching values of > 20 MPa) and the ones on the adjacent onshore part of the Friesland Platform (< 10 MPa). Corona (2005) described in detail the characteristics of the pressure differentials in the Rotliegend reservoirs that occur cross-fault in the Lauwerszee Trough itself.
Multiwell cross-plots of pore-fluid pressure versus depth further illustrate this regional variation of southward-decreasing fluid pressures (Figs. 12, 13). The cross-plots show the change of fluid pressure with depth for the main structural elements. In the northern structural elements, fluid pressures are well above hydrostatic (Fig. 12). Large variations in pressure occur for the same depth of measurement in each structural element (Fig. 12). In contrast, the pressures in the southern structural elements reach nearly normal values and show only minor variations in pressure for the same depth of measurement (Fig. 13).
The hydraulic-head distribution in offshore Netherlands (Fig. 10) confirms the general trend shown by the overpressure map (Fig. 8). Hydraulic-head gradients in the north (Central Offshore Platform and Noord Holland Platform) are higher than the gradients in the southern Broad Four-teens Basin and the West Netherlands Basin. The hydraulic heads in the northwestern part of the area (Cleaverbank High) are less than the heads on the Central Offshore Platform. In the Broad Fourteens Basin the hydraulic heads decrease towards the southwestern thrust zone.
Factors Influencing Conditions of Pressure and Fluid Flow
The observed conditions of fluid overpressures and hydrodynamics in the Permian Rotliegend reservoirs are controlled by the combined influence of a) past and/or ongoing pressure-influencing mechanisms, and b) characteristics of the geological framework. The observed departure from hydrostatic conditions in the Permian Rotliegend reservoirs indicates that there is disequilibrium between a and b in the sense that the fluid-pressure regime has not adjusted to the evolving geologic framework and is in a transient state. The time required to propagate or dissipate pore-water overpressures by water flow through a water-saturated permeable rock framework is proportional to its hydraulic diffusivity (related to properties such as porosity, permeability, compressibility of the rock, and the viscosity and compressibility of the pore fluid). The propagation and dissipation of fluid overpressures by multiphase flow is in addition influenced by differences in capillary pressures of the rock framework.
A variety of mechanisms have been proposed to explain the generation of pore-water overpressures in sedimentary basins (Grauls, 1998; Mann and Mackenzie, 1990; Neuzil, 1995, 2003; Osborne and Swarbrick, 1997; Yardley and Swarbrick, 2000):
An increase in compressive stress, by processes such as sedimentary loading, vertical or lateral tectonic loading, and glacial loading;
An increase in fluid volume, induced by processes such as temperature increase (aquathermal pressuring), production of water by diagenetic and metamorphic processes (for example mineral dehydration), and hydrocarbon generation (gas generation from late-stage source-rock maturation and from oil cracking);
Fluid movement and associated redistribution of overpressures.
Stress-related mechanisms are considered to be the most likely causes of overpressure in many sedimentary basins (e.g., Osborne and Swarbrick, 1997). An important stress-related mechanism is sedimentary loading. In reality, different pressure-influencing mechanisms may operate simultaneously during continuous burial of sedimentary units subject to sedimentary loading: not only the sedimentary loading itself, but also aquathermal pressuring, dehydration of minerals, and/or generation of gas. As outlined above, local fluid overpressures in the subsurface include the effect of the density contrast between petroleum fluid and pore water.
The competition between a stress-related pressure-generating mechanism and a pressure-dissipating mechanism operating during a certain time interval controls the overpressure distribution. For example, at a given depth in a water-saturated rock pore, considering a certain time interval, ongoing sedimentation increases the overburden stress which, in turn, tends to cause compaction and porosity loss. In a hydraulically open system, the compaction and porosity loss can be accommodated by fluid flow from the pore without buildup of overpressure. This can be observed in permeable reservoir rocks in hydraulic communication with the ground surface, allowing fluid flow induced by the sedimentary loading and equilibrium to be maintained in the reservoir. When the subsiding reservoir rock is unable to dewater in response to loading, the pore fluids bear part of the increase in overburden stress and become overpres-sured. The rate of dissipation of the overpressures in time after the sedimentary loading has stopped and depends on the hydraulic characteristics of the reservoir rock itself in relation to the regional hydraulic framework.
Moss et al. (2003) indicated that stress-related mechanisms, such as sedimentary loading and gas generation are the two dominant causes thought to contribute to overpressures in Central North Sea basins. In order to evaluate these mechanisms for the Dutch part of the North Sea, we first look at the burial histories. The burial histories of the Rotliegend reservoirs and the underlying gas-prone Coal Measures of the Limburg Group show regional variations that depend on their geographical location and structural position (Fig. 1). Figure 1 shows a difference in burial history of the Permian Rotliegend reservoirs in the strongly inverted basins in the southern offshore (Broad Fourteens and West Netherlands Basins) and at the platforms and highs in the north. The pre-Tertiary units in the southern basins are all largely above their maximum burial depth today. More to the north (Central Offshore Platform, Noord– Holland Platform, Cleaverbank High, northern part of Friesland Platform), the stratigraphic units are largely at maximum burial depth today. The generalised spatial variation in sedimentary loading of the Rotliegend reservoir during Tertiary and Quaternary times is illustrated by the thickness maps given in Figure 6. The overall sedimentary loading during the Tertiary as well as during Quaternary times increases from south to north.
Several studies have shown that the relatively rapid Late Tertiary and Quaternary sedimentary loading plays an important role in explaining present-day overpressure distributions in the subsurface of the Netherlands offshore, especially in Tertiary mudstones, Cretaceous Chalk and Rijnland Group, and Jurassic units in the northern offshore (Verweij, 2006; Verweij et al., 2009b). This is in accordance with findings concerning the origin of overpressures in the North Sea outside the Dutch territory (e.g., Gaarenstroom et al., 1993; Moss et al., 2003; Winefield et al., 2005).
The fluid overpressures observed in the Rotliegend reservoirs in the Dutch Central Graben and the Lauwerszee Trough are too high to be explained by this recent sedimentary loading alone. Two possible mechanisms that played a role in generating these overpressures are long-time sedimentary loading, starting already in pre-Tertiary times, and gas generation. With respect to gas generation, recent 3D basin-modelling studies demonstrated that Carboniferous source rocks generate gas during Cenozoic times in the northern parts of the study area (i.e., Cleaver Bank High, Central Offshore Platform, Noord–Holland Platform, the offshore part of the Friesland Platform, and the southern part of the Dutch Central Graben; Verweij et al., 2009a; Verweij et al., 2010; Verweij et al., in press; Abdul Fattah et al., 2010). In general, sedimentary loading and gas generation are ongoing processes in these offshore areas, and as such are pressure-generating processes operating today.
The conditions of pressure and fluid flow described in this paper are based on observed pre-production fluid pressures. The actual present-day conditions of pressure and fluid flow in the Rotliegend reservoirs are also influenced to a greater or lesser extent by fifty years of gas production from the reservoirs.
Irrespective of the mechanisms of generation of fluid overpressures, preservation of the observed overpressures (Fig. 8) is influenced by the hydraulic characteristics of the Permian Rotliegend reservoir and its location in the geologic framework (primarily the permeability framework). A regional permeability framework is characterised by the distribution, interconnectivity, thickness, and dip of porous and permeable and less permeable units (reservoirs and seals), and by the location of geological structures and tectonic elements of importance for fluid flow, such as fault and fracture zones, unconformities, and salt structures.
Figures 3, 4, and 11 show the distribution and generalised facies of the Rotliegend reservoir. The best porosities and permeabilities are encountered in dry aeolian sandstones (Gaupp and Okkerman, this volume). Advanced mechanical compaction and intense illitization in reservoirs that were deeply buried during a long time decreased the original porosity and permeability significantly. In addition, the decreasing thickness of the reservoirs and the increasing intercalation of clay and siltstones in the northern transition zone reduced its hydraulic conductivity. The lateral hydraulic continuity of the Rotliegend reservoirs is affected by the numerous faults cross-cutting the reservoirs (Figs. 8, 11). Van Hulten (2007) observed that the northern zone of the Rotliegend reservoir is especially prone to compartmentalisation by faults, because the reservoir in this zone is thin and characterised by many shale layers. Low-permeability fault zones may slow down the dewatering.
The evaporites of the Zechstein Group provide a regional top seal for vertical fluid flow from the Upper Rotliegend Group in the northern part of the area (Fig. 5).
Regional Differences in Factors of Influence on Pressure and Fluid-Flow Conditions
The outline above shows that there are distinct regional differences between the southern offshore and the more northern part of the area with respect to important factors influencing both pressure generation and dissipation in the Rotliegend reservoirs. Figure 1 summarises the regional variation in burial history of the Rotliegend and in the geological framework along a schematised SW–NE cross section through offshore Netherlands. The distribution of observed overpressures and hydraulic head (Figs. 8, 10, 11, and the cross-plots given in Figures 12 and 13) reflect these regional differences:
The Zechstein evaporites restrict vertical bleed-off of pressures in large parts of the area (Fig. 5). Outside the northern zone of high overpressures, the southward-increasing thickness of the Rotliegend reservoir seem to provide a lateral escape pathway to bleed off overpressures by fluid flow to areas with a more permeable geologic framework (such as areas where the vertical Zechstein evaporites are missing). The hydraulic-head map shows the potential for such lateral formation-water flow (Fig. 10).
The salinity and the hydraulic-head map show the influence of large boundary fault zones on the characteristics of the water flow system. The hydraulic-head map further suggests that, on a geological timescale, the numerous faults cross-cutting the Rotliegend reservoir may hamper the flow of water but do not completely restrict it.
Relatively high overpressures are maintained in the northern transitional zones of the Rotliegend reservoir where a) intercalation of clay and siltstones have reduced its hydraulic reservoir properties, and b) vertical and lateral dewatering of the Permian Rotliegend reservoir is severely restricted (Figs. 3, 4, 8, 11);
The rather abrupt changes in overpressure along the northern transition zone are probably related to fault-related restricted lateral continuity of the Rotliegend reservoir in combination with its decreasing thickness and permeability close to the northern sand limit (Figs. 3, 4, 11). Severely restricted conditions are indicated by the very high fluid overpressures in the southern part of the Dutch Central Graben (block L2) and also in the northern part of the Lauwerszee Trough. The very high overpressures in the northern part of the trough are maintained in a pressure compartment bounded by faults (Corona, 2005). The formation water in the trough is overpres-sured; pressure differentials occur across faults, and overpressures generally decrease from NW to SE in the trough, consistent with the direction of gas fill-and-spill across fault-juxtaposition leak points (Corona, 2005).
The distribution of fluid overpressures in the Permian Rotliegend reservoir in offshore and northeastern onshore of the Netherlands reveal a general regional trend of decreasing fluid overpressures from northeast towards the south. The fluid-overpressure values vary between hard geopressures (Pexcess > 40 MPa in block L2) and nearly hydrostatic pressures (Pexcess < 1 MPa in southern offshore). The highest overpressures occur in a zone following the northern limit of the Permian Rotliegend reservoirs, where Zechstein evaporites restrict vertical bleed-off of pressures. The rather abrupt changes in overpressure along the northern limit are probably related to fault-related restricted lateral continuity of the Rotliegend reservoir in combination with its decreasing thickness and permeability close to the northern sand limit. The hydraulic-head map of the Rotliegend reservoir illustrates the potential for a general southward dewatering direction.
We have demonstrated that there are distinct regional differences between the southern and the northern part of the area with respect to important factors influencing both pressure generation (such as sedimentary loading and gas generation) and dissipation (by fluid flow) in the Rotliegend reservoir. The distribution of observed overpressures and hydraulic heads reflect these regional differences. The permeability framework, that is, the distribution, intercon-nectivity, thickness, and dip of seals (such as Zechstein evaporites and sealing faults) and permeable pathways, controls the vertical and lateral dewatering of the Rotliegend reservoirs. The regional variations in the permeability framework exert a major influence on the observed distributions of fluid overpressures. Relatively high fluid overpressures can be maintained in zones where dewatering of the Rotliegend is severely restricted. This is especially apparent in the southern part of the Dutch Central Graben, and also in the northern part of the Lauwerszee Trough.
The authors are grateful to Total E&P Nederland and Energie Beheer Nederland B.V. (EBN) for permission to publish the regional pressure map (Fig. 8). We thank the reviewers for their constructive comments and suggestions.
Figures & Tables
The Permian Rotliegend of the Netherlands
More than 50 years ago, the discovery of the giant Groningen Gas Field in the subsurface of the Netherlands by NAM B.V. marked a turning point inthe Dutch and European energy market initiating the replacement of coal by gas. Despite the fact that the Rotliegend dryland deposits in the Southern Permian Basin are one of Europe's most important georesources, no sedimentological overview is available to date for the subsurface of the Netherlands. This SEPM Special Publication presents for the first time such a summary of the present-day knowledge, including a comprehensive core atlas from on- and offshore wells.