The Rationale for an Integrated Stratigraphic Framework of the Upper Rotliegend II Depositional System in the Netherlands
Published:January 01, 2011
- PDF LinkChapter PDF
Kees Van Ojik, Alain R. Böhm, Holger Cremer, Mark C. Geluk, Mat G.G. De Jong, Harmen F. Mijnlieff, S. Djin Nio, 2011. "The Rationale for an Integrated Stratigraphic Framework of the Upper Rotliegend II Depositional System in the Netherlands", The Permian Rotliegend of the Netherlands, Jürgen Grötsch, Reinhard Gaupp
Download citation file:
Due to the nature of the depositional environment and most importantly the lack of (bio) stratigraphic control, it remains difficult to establish a robust and reliable stratigraphic framework for the Upper Rotliegend which can be used as a guideline to better understand the internal architecture. The most important challenges encountered are the identification of the large-scale basin architecture and basin fill, the identification of a sequence stratigraphic model for (semi-) arid continental deposits, and the relationship between the distribution of Upper Rotliegend sediments and the underlying Variscan structural framework.
Based on the present-day knowledge and status of particular stratigraphic aspects, it can be concluded that a single “silver bullet” solution does not exist for providing a more profound understanding of the depositional model. It is evident that none of the methods available should be used in isolation but rather integrated in the framework of sequence stratigraphy providing the petroleum geologist a genetic and predictive geological model.
The Middle to Upper Permian Upper Rotliegend II (Gast, 1995; Geluk and Van Adrichem Boogert, 1995; Upper Rotliegend Group in the Netherlands of Van Adrichem Boogaert and Kouwe, 1993) is one of the most studied stratigraphic units in northwestern Europe, holding a significant proportion of the northwestern European hydrocarbon reserves. Various aspects of these deposits have been studied in detail, including the sedimentology, stratigraphy, diagenesis, structural geology, etc., with the prime purpose to improve the understanding and prediction of the reservoir quality and its vertical and lateral distribution.
Due to the nature of the Upper Rotliegend sediments, some stratigraphic challenges remain difficult to resolve. The most important challenges identified are:
Internal basin-fill geometry and architecture on a regional scale.
Major sequence stratigraphic key surfaces in the various depositional environments and the lateral extent of these key surfaces.
Generation and distribution of accommodation space during the Late Permian in close relationship to the underlying basement controlled by the Variscan structural framework.
In this paper a short description is provided of the present-day knowledge of particular stratigraphic aspects of the Upper Rotliegend, in close relationship to the challenges described above.
In general, the sediments of the Upper Rotliegend are confined between two regionally important horizons (Fig. 1):
Base Zechstein Coppershale (“Kupferschiefer”) at the top and
Base Permian Unconformity (BPU) at its base.
The Coppershale is a pronounced zone of homogeneous black, vaguely laminated claystone some 20 to 50 cm thick, and rich in organic matter (Appendices B1a, B1b). It is classified as the basal unit of the Zechstein Group and occurs across almost the entire Southern Permian Basin. It is considered to represent an important synchronous event in the basin; the age dating in the Harz Mountains area of 257.3 ± 1.6 Ma based on Re/Os isotopes by Brauns et al. (2003) therefore is of significant importance.
In Germany copper-rich deposits of the Coppershale were already mined during the times of the Roman Empire. The term Rotliegend (“Underlying Red”) originates from medieval German mining industry where bedrocks below the Coppershale were identified by their red colours.
The base of the Upper Rotliegend II represents an erosive event picked at the boundary between rocks of Carboniferous age, or locally volcanics of the Early Permian Lower Rotliegend and red-bed rocks of the Upper Rotliegend I (App. B1c-B1i). This boundary is informally referred to as the Base Permian Unconformity (BPU), or Hercynian unconformity, which was formed by the amalgamation of a series of erosive events due to the Variscan Orogeny and Early Permian rift tectonics (Ziegler, 1990; Glennie, 1998; Gast and Gundlach, 2006; Geluk, 2007). The name Saalian Unconformity used in the Netherlands (e.g., Van Wijhe, 1987) is not comprehensive because it refers just to one of the unconformities involved (Geluk, 2005). The term Hercynian Unconformity should be discarded as well, inasmuch as it refers to the Hercynian orogeny hence to the unconformity between folded Hercynian basement and the earliest molasse. The term BPU is not correct in a stratigraphic sense but is generally accepted as an informal work term.
The Rotliegend was first encountered in wells in the northern Netherlands (e.g., Haren-1, 1955). Initial descriptions were provided by Thiadens (1963) and Stäuble and Milius (1970), who referred to the Rotliegend Formation (Lower Permian). The early subdivision into the Ten Boer and Slochteren members was designed around the southern part of the Groningen field but would still form the base of the later subdivisions into reservoir and non-reservoir units. The first formal definition was given by Van Adrichem Boogaert (1976), who proposed the Upper Rotliegend Subgroup, subdivided into the Slochteren and Silverpit formations. Herewith the foundation was laid for the present-day subdivision in the Netherlands. In a common lithostratigraphic nomenclature between the Nederlandse Aardolie Maatschappij (NAM) and the Rijks Geologische Dienst (RGD) 1980, the volcaniclastic sediments were classified as part of the Lower Rotliegend Group and the red beds were redefined to the Upper Rotliegend Group. Based on the increasing amount of well penetrations, the complexity of the interfingering Slochteren and Silverpit formations was better understood; next to the Ten Boer also the Ameland and Hollum claystones were introduced. This definition was modified only slightly by Van Adrichem Boogaert and Kouwe (1993). Finally, minor revisions of the nomenclature were proposed by Geluk (2005).
For detailed correlations of genetic reservoir units, however, a lithostratigraphic framework in barren dry-land deposits offers insufficient detail and resolution; hence a sequence stratigraphic approach has received much attention during the last decades. This approach is based on the recognition that:
some main depositional environments can be distinguished,
climate exerts a major control on the distribution of lithofacies zones formed in these environments,
climatic variations result in shifting of the relative positions of these facies belts, and
the development of drying- and wetting-upward sequences and possible recognition of timelines.
Several attempts have been made to correlate Upper Rotliegend sediments in the Netherlands based on a sequence stratigraphic or cyclostratigraphic approach (Yang and Baumfalk, 1991; Gast, 1991; Yang and Nio, 1994; Yang and Kouwe, 1995; George and Berry, 1994). The quality and resolution of these sequence stratigraphic frameworks remains, however, hypothetical due to the lack of a supporting and independent means of dating the relative age. Other methods such as chemostratigraphy have been tested to support stratigraphic correlations, but these methods are still rather underexplored, and with some exceptions (e.g., Schuurman, 1998) results of these techniques have mostly remained restricted to proprietary company reports. A possibly powerful biostratigraphic control based on the recognition of phytoliths may add value in the future.
The Rotliegend in the Netherlands can be correlated lithostratigraphically without major problems to the adjacent countries (Glennie, 1972, 1998; Gast, 1991; Glennie et al. 2003). It represents only the youngest post-rift part of the German Rotliegend succession (Geluk, 2005, his Fig. 1; Gast and Gundlach, 2006).
The Upper Rotliegend Group of the Netherlands correlates with the Silverpit and Leman formations of the UK Southern North Sea (Johnson et al., 1994). The equivalent in Germany is the Elbe Subgroup (upper part of Upper Rotliegend II; Plein et al,, 1995; Verdier, 1996; Doornenbal and Stevenson, 2010). Equivalents of the Upper Rotliegend II have been described as the so-called Zechstein conglomerate in northeast Belgium (De Craen and Swennen, 1992; Dusar et al., 2001) and in the Lower Rhine coal-mining area in Germany (Hilden, 1988).
Initially the nature of the depositional environment of the Rotliegend sediments in the Netherlands was subject of discussion (K.W. Glennie, personal communication). Based on outcrop descriptions of Rotliegend rocks in Germany and elsewhere it was, however, soon realised that these sediments were deposited in a mixed fluvial and aeolian deposi-tional environment (Pannekoek, 1956; Thiadens, 1963).
The time of initiation of sedimentation varied across the Netherlands, but it is assumed that in the Netherlands the first Rotliegend sediments are approximately some 263 Ma old (Guadalupian). Unfortunately, in the Netherlands no absolute age dating of the classic Rotliegend series could be derived so far, mainly due to the lack of suitable material. A K/Ar date obtained from the volcanic and volcaniclastic series assigned to the Lower Rotliegend in southeast Drenthe, onshore Netherlands, resulted in a minimum age of 258.6 ± 6 Ma (Sissingh, 2004), which, if proven to be correct, leads to the conclusion that at least part of the Upper Rotliegend was deposited contemporaneously with the Lower Rotliegend volcaniclastic series. Datings of Rotliegend volcanics outside the Netherlands indicate ages around 299 Ma (Breitkreuz and Kennedy, 1999).
Rotliegend sedimentation ended quite abruptly when the SPB was flooded and drowned at 257.3 ± 1.6 Ma (Lopingian), resulting in the deposition of the Coppershale (Brauns et al., 2003).
The intracratonic Southern Permian Basin is a large sag basin that developed upon a series of Early Permian rift basins. The main depocentre of the Upper Rotliegend in the Southern Permian Basin was located in the German and northern Dutch offshore areas. The fine-grained evaporitic sediments in this depocentre are interpreted to represent a desert lake complex. Two major fluvial fairways, one located in the eastern Netherlands and one in the western offshore, spilled sediment into this basin, with source areas located mainly towards the south. In the areas of relatively quiescent conditions between these fluvial systems, wind-blown sediment settled down in extensive erg plains. Based on the proportional amount of evaporites and fine-grained material versus coarser-grained sediments, a twofold lithostratigraphical subdivision has been made into the Silverpit and Slochteren formations (Appendices A.1–A.5, B2a–B2e).
These formations are divided further into a number of members based on their relative vertical position (Van Adrichem Boogaert and Kouwe, 1993). Figure 2 presents an overview of the lithostratigraphic succession types of the Upper Rotliegend in the Netherlands and its subdivision into members and units.
Red-brown sandy clay and silt intervals of the Ten Boer rest conformably on the Main Slochteren or Upper Slochteren members (Appendices B2a, B2b). Towards the north it merges into the Upper Silverpit, and to the south it passes laterally into the sandstones of the Slochteren Formation.
The red-brown sandy claystones and siltstones of the Ameland are restricted to the configuration where Upper and Lower Slochteren can be identified. Both the Upper and Lower Slochteren pinch out gradually towards the north, as does the Ameland towards the south. In those areas where the Lower Slochteren is not present the fine-grained deposits below the Upper Slochteren are classified as the Buren Member. In those areas where both Upper and Lower Slochteren are absent, the Ameland is an indistinct part of the Silverpit Formation.
The Upper and/or Lower Slochteren are defined only north of the southern pinch-out line of the Ameland. Due to the lack of diagnostic criteria south of this line, the Slochteren Formation does not allow any further formal subdivision.
The informal Akkrum sandstone unit is reserved for a local sandy development of the Ten Boer in the northern onshore part of the Netherlands.
In the Netherlands the term “Weissliegend” is an informal classification of white to grey usually structureless sandstones at the top of the Rotliegend immediately below the Coppershale (Appendix B5c). The “Weissliegend” is explained as reworked and fluidised aeolian dunes caused by the sudden flooding of the SPB at the end of the Permian Rotliegend (Glennie and Buller, 1983). Subsequent bleaching of the red desert sediments was caused by reducing bottom water and/or formation water.
A truly seismostratigraphic analysis of the Upper Rotliegend strata is hampered due to several causes which result in a relatively poor imaging of seismic data. Firstly, the significant depth (generally beyond 3 km depth) limits the seismic resolution of Rotliegend strata. Secondly, the overlying Zechstein evaporite sequence absorbs and distorts the seismic signal, especially in areas with strong halokinesis. Thirdly, the Upper Rotliegend is broken up into numerous fault blocks of different size and orientation. However, recent seismic acquisition and processing techniques have improved the resolution tremendously, resulting in an increasingly better imaging of the Rotliegend strata.
The upper boundary of the Rotliegend is generally picked easily on the prominent high-amplitude and very continuous set of reflections at the base of the Zechstein. This seismic response is the result of the abrupt change from the high-impedance carbonate and anhydrite litholo-gies to the relatively low-impedance clastic sediments (predominantly claystone) of the Upper Rotliegend II.
The lower boundary of the Rotliegend is represented by an unconformity, which may be present as an angular unconformity, or as a disconformity, in which case the Rotliegend and Carboniferous beds are (approximately) parallel. The impedance contrast between the Rotliegend and the underlying Carboniferous is generally too low to generate a clear reflection and is therefore in many cases hard to pick. Generally, only in case of a prominent high-angle unconformity can the BPU be interpreted with confidence on seismic sections.
The internal seismostratigraphic architecture of the Upper Rotliegend II package is difficult to ascertain. Some large-scale reflections representing stratification in concordant configuration with the overlying Zechstein reflections are present. Occasionally, very subtle toplap patterns of seismic events in the upper part of the Rotliegend terminating against the overlying Zechstein deposits can be observed. This may indicate a very gentle unconformity locally present.
The seismic section in Figure 3 is flattened on the base of the Zechstein and demonstrates that the Rotliegend is clearly present in an onlapping and discordant configuration with respect to the underlying Carboniferous sediments. This onlap pattern supports an aggradational onlap model for the large-scale basin-fill geometry
Intra–Upper Rotliegend II seismic facies are predominantly fairly transparent and nondiagnostic as a result of the lack of significant and consistent impedance contrasts. However, in those areas where a lithostratigraphic subdivision in the four main members (ROSLL–ROCLA–ROSLU– ROCLT) and a consistent, well-defined reflection corresponding to the Ameland Member can be recognised (Fig. 3). The Ameland Member represents a series of claystones deposited during times of high water runoff and/or high base level interpreted to represent a maximum wetting phase. This event occurred on a basinwide scale; hence, the Ameland reflector and its correlative equivalent in Germany (Ebstorf Shale at the base of the Upper Hannover Subgroup) represent a persistent time line. As can be seen in the north–south seismic section of Figure 3, the Ameland Member reflection onlaps southward and terminates against the underlying Carboniferous. This supports the conclusion that towards the south the Upper Rotliegend II sediments represent a series deposited time-equivalent to the Upper Slochteren and Ten Boer members.
From this north–south section from north of Groningen to SE Drenthe (Fig. 3) some further observations can be made:
The two-way-time seismic thickness of the combined Upper Slochteren and the Ten Boer members above the “Ameland” reflector is remarkably constant, in contrast to the two-way-time seismic thickness of the unit below.
Towards the southern part of the section the character of the seismic facies in the top part of the Upper Rotliegend II (upper 50 ms) changes laterally from a horizontally layered facies with strong vertical variations in contrast to a more transparent seismic facies with little vertical variations supporting the presence of (diachronous) lateral facies changes.
Careful evaluation and interpretation of ordinary seismic reflection data in combination with impedance seismic data can aid in resolving the stratigraphic issues. However, it should be kept in mind that on a local block or field scale the internal architecture of the Upper Rotliegend is more complicated and that it does not necessarily follow a simple stratigraphic layer-cake model (Parry et al., 2009).
Analysis Of Depositional Trends
An unconventional method for the stratigraphic analysis of wireline logs has been developed in recent years by Nio et al. (2005) and Nio et al. (2006). The method focuses on the identification and correlation of (time-) equivalent vertical lithofacies changes and trends in wireline-log data. The methodology comprises two main elements. First, a facies-sensitive log (usually the GR log) is transformed into a spectral trend (or INPEFA) curve showing uphole changes in the waveform properties of the data. Second, the curves are interpreted and a stratigraphic model is developed from them using the concepts of global cyclostratigraphy— the theory that climate change is a fundamental control on lithofacies succession (Perlmutter and Matthews, 1990; Perlmutter et al., 1998).
The frequency analysis routinely applied in CycloLog® (software developed by ENRES International) is based on the principles of the maximum entropy method (MEM). In a first step, a statistical probabilistic model is constructed, which holds the characteristics of the input log data (Nio et al., 2006). Then, a prediction error filter (PEF) is generated, which, by means of a sliding window, is used to compute changes in the spectral composition of the data series, i.e., differences between the predicted and actual log properties. Output from this analysis is a PEFA (prediction error filter analysis) curve, which, in turn, by mathematical integration, is converted into a so-called spectral trend curve, also called INPEFA (Integrated PEFA). The shape of the INPEFA curve is a function of the interval over which it is calculated. Typically, the shorter the interval, the more detail is revealed. Examples of these INPEFA curves are shown in Figure 4.
Extensive work on wireline-log data from many strati-graphic intervals and areas has shown that the INPEFA curve reveals trends and trend changes in the lithological succession which are often not immediately obvious from the unprocessed data and, occasionally, even overlooked (De Jong et al., 2006; De Jong et al., 2007). Through careful matching of the patterns of the INPEFA curves—and the underlying lithofacies patterns—correlations are made between wells. The equivalent patterns are considered nearly synchronous based on the concepts of a global cyclostratig-raphy. The succession of orbitally controlled climate change is considered to have a fundamental, ultimate control on the pattern of vertical lithofacies change, which is revealed by the INPEFA curve.
Characteristically, the basic motif of the INPEFA curve is the C shape. In clastic deposits, it consists of a lower “limb” representing relatively coarse-grained deposits and an upper “limb” corresponding to relatively fine-grained deposits. The lithofacies are genetically linked: the lower “limb” represents a trend with a progradational or related component, whereas the upper “limb” represents a period of retrogradation or related processes. C shapes often occur in a hierarchical pattern. Detailed correlations in the Rotliegend, involving large numbers of wells with in part close spacing, show that the salt deposits of the so-called Silverpit Lake are equivalent to the relatively coarse-grained deposits which are represented in the INPEFA curves by the lower ‘“limbs” of the C shapes. This relationship is shown in the correlations of Figure 5.
Chemostratigraphy is a correlation tool involving the use of inorganic geochemical data based on the analysis of the vertical distribution of key elements and their proportional amounts. Several proprietary chemostratigraphic surveys of Rotliegend sediments have been carried out on behalf of the hydrocarbon E&P industry, yet none of these have been released to the public domain yet.
Establishing a chemostratigraphic zonation usually follows a methodology involving a thorough statistical evaluation of the dataset and the introduction of certain element ratios. Prior to any interpretation of the data it is, however, essential to establish the mineralogical affinities of elements, particularly with respect to the “key” elements used for chemostratigraphic correlation. The elements employed as key elements generally vary between geographical areas and stratigraphic intervals due to variations in:
detrital mineralogy and provenance: different source areas feeding sediments into the basin potentially producing differing mineralogical compositions;
depositional environment: certain elements, particularly the relatively mobile ones (e.g., Cu, Ni, Mo) may be sensitive to changes in depositional conditions; and
diagenesis: under different diagenetic regimes, including weathering, some elements may become remobilised, resulting in the detrital geochemical signatures of the sediments becoming either modified or obliterated.
In any given study, it is important to understand the mineralogical affinities of elements, particularly the key elements. The public information and knowledge of these affinities is, however, very restricted due to the limited number of chemostratigraphic surveys carried out, or not available publicly.
Important key elements and their ratios that are typically used in the chemostratigraphic surveys are, amongst others:
strontium (SR)/yttrium (Y): Sr-bearing clay minerals and feldspars vs. Y-bearing heavy minerals;
scandium (Sc)/strontium (Sr): Sc-bearing zircon vs. Sr-bearing clay minerals and feldspars
rubidium (Rb)/potassium (K): Rb-bearing clay minerals, micas, and feldspars vs. K-bearing clay minerals, micas, and K-feldspar;
uranium (U)/aluminium (Al): U-bearing heavy minerals; Al is concentrated, almost exclusively, in clay minerals;
calcium (Ca)/potassium (K): carbonate minerals, bio-genic phosphate, and/or anhydrite vs. K feldspars, micas, and clay minerals; and
uranium (U)/potassium (K): U-bearing heavy minerals vs. K feldspars, micas, and clay minerals.
Principal conclusions from chemostratigraphic surveys (internal reports) carried out on core material recovered from the Upper Rotliegend II in wells located in the northern Dutch offshore are as follows:
A close relationship exists between sediment geochemistry and grain-size fractions and lithofacies. Several fluvio-aeolian lithofacies can be differentiated quite clearly based on their geochemical composition.
Some important stratigraphic differences independent of facies in sediment geochemistry are apparent. The geochemistry of the Ameland is distinct from the geochemistry of similar grain-size classes of the overlying Upper Slochteren and Ten Boer, suggesting marked variations over time in the background geochemical signature and, hence, provenance of sediments.
Some lateral changes and variations in the geochemical signature are observed which are attributed to local development of Mg chlorites due to the alteration of volcanic detritus as a function of lateral variations in the burial and diagenetic history.
Magnetostratigraphy is a powerful tool for the correlation of absolute time zones in sedimentary sequences, relying upon the identification of normal (north-seeking, downward-dipping) and reverse (south-seeking, upward-dipping) periods of polarity (magnetozones) which reflect the direction of the Earth’s ambient geomagnetic field at the time of deposition. Studies have helped to develop the Permian geomagnetic polarity timescale (GPTS), resulting in the recognition of a long period of reversed polarity known as the Kiaman Superchron during the Early Permian, and a series of polarity reversals during the Late Permian Permo-Triassic Superchron (Gradstein et al., 2004; Menning et al., 2004; Steiner, 2006). Proprietary magneto-stratigraphic studies carried out on Rotliegend material have indicated the presence of at least five normal and four reversed magnetozones in the Upper Slochteren from the offshore K-blocks and two normal and three reversed magnetozones in the Lower Slochteren from the offshore L-blocks. This would indicate that all of the analysed sections postdate the Kiaman Superchron, and in combination with absolute age dating of the Zechstein Coppershale it would indicate the deposition of the Upper Rotliegend in the northern Dutch offshore during the Capitanian. However, it is clear that these magnetostrati-graphic results do not match the formal GPTS as documented by Gradstein et al. (2004) and revised correlations by Steiner (2006). Further selective infill is required to establish magnetozone boundaries more reliably and accurately to allow an improved chronostratigraphic zona-tion of the Upper Rotliegend. It should, however, be realised that the recognition of individual magnetozones will remain severely hampered by the presence of many intraformational erosive events and time gaps associated with continental red-bed deposits.
The application of a nonmarine biostratigraphy to Permian deposits was synthesised recently by Lucas et al. (2006). They demonstrated the inadequate time-based definition of Permo-Triassic deposits. Some macrofloras and microfloras and faunas have long been important in non-marine Permian correlations, including the presence of marine faunas in sporadic marine incursions identified in Germany (Legler et al., 2005). On top of the limitations referred to by Lucas et al. (2006), it should be realised that material from some of these fossil groups (e.g., tetrapod footprints) are inaccessible in the Netherlands due to the limitations of sampling methods and techniques commonly used in the upstream hydrocarbon industry.
Biogenic Silica Particles (bsps)
Most Rotliegend sandstones, particularly the Slochteren deposits, do not contain any organic-walled microfossils (sporomorphs, pollen) (Amthor and Okkerman, 1998). This hampers a biostratigraphic subdivision of the different Rotliegend formations and members, and the cross-correlation of different wells. However, recent reanalysis of a number of Rotliegend cores from various offshore and onshore sites brought a species-rich assemblage of siliceous microfossils to light (Garmin et al., 2010). These microfos-sils are referred to as biogenic silica particles (BSPs) inasmuch as their origin has not been entirely clear until today (Fig. 6). Since the extraction of the first BSPs in 2006, their potential for a biostratigraphic application on Carboniferous–Permian deposits in the Netherlands has been demonstrated in a number of empirical studies.
The examination of almost 150 Rotliegend samples from five wells (Table 1) revealed the presence of BSPs in all studied samples. The observations provided evidence for 22 different BSP morphs, which were assigned as BSP001 to BSP022 (Fig. 6). The down-core distribution of the identified BSPs shows a distinct pattern indicating that at least some of them have stratigraphic significance (Fig. 7). The tentative Rotliegend zonation in Figure 7 suggests, for example, that the boundary between the Ameland and Upper Slochteren members can be well characterised by BSP occurrence. However, studying more Rotliegend sedimentary sequences in sufficient resolution must corroborate evidence presented. Moreover, the occurrence of the BSPs in the Coppershale and the Carboniferous has not yet been studied. The possibility of characterising the Base Permian Unconformity by BSPs is currently being explored. This new technique, however, may open valuable opportunities to biostratigraphic characterisation of the Rotliegend sediments.
The record of BSP-bearing Palaeozoic continental deposits is relatively sparse. A single evidence of continental biogenic silica microfossils, referred to so-called phytoliths, is reported in the literature from Mesozoic and Palaeozoic deposits in Antarctica (Carter, 1999). Phytoliths (literally meaning “plant rocks”) are composed mainly of hydrated silica (SiO2 with 5–15% H2O) and are usually formed by land plants which store phytoliths as important structural bodies in their leafs, stems, or roots (Piperno, 2006). The presence of gymnosperms, lycopods, and ferns in the fossil record since the Late Devonian (Briggs and Crowther, 2001) supports the assumption that BSPs were already formed by Palaeozoic land plants, and hence BSPs should be preserved in the fossil record. Analysis of fresh material from fossils of various still-living plants, among them horsetail, ginkgo, and club moss, revealed a number of phytoliths that strongly resemble those found in the Rotliegend samples (Fig. 6). However, the possibility that the Rotliegend BSPs are of a different origin, for example are skeletons of siliceous aquatic algae, cannot be excluded at this point in time.
Some of the most challenging stratigraphic questions are:
(1) Large scale basin-fill geometry and architecture: Various regional models have been established over time trying to explain the lateral and vertical thickness distributions of the Upper Rotliegend in relation to the basin-fill type. Generally two extreme end members can be recognised: the retrogradational (onlap) model (representing the expanding Rotliegend “lake”), and the aggradational (wedge) model.
Lithostratigraphic correlations based on the formal definitions by Van Adrichem Boogaert et al. (1994) may work locally on a field or block scale but pose problems when establishing a more regional correlation of lithostratigraphic units. Due to the waning supply of clastic material towards the basin centre and subsequent lateral variations in depo-sitional environment, it is clear that many of the lithostrati-graphic boundaries are diachronous. The absence of distinct regional intra–Upper Rotliegend II markers prevents a reliable correlation of strata from the basin centre towards its margins, and hence the identification of lateral variations of the depositional history in the basin.
(2) Regional conceptual depositional model: Based on the assumption that climate exerts a major control on the lateral and vertical variations in depositional environments, it is generally accepted that systematic trends and/or cycles can be observed in these depositional environments. This model is a foundation for the description of Rotliegend sediments and input into an allostratigraphic descriptive methodology. During wet periods, the playa lake (Appendices B2a, B2b) and the fluvial system (Appendices B3b, c, B3e, B4d–g, B4m) expand, both at the expense of the dry-sandflat–dune facies (Appendices B3a, B4h, B4k, l, B4o, p). The opposite occurs during dry periods, leading to a broadening of the dry-sandflat–dune facies. In relation to variations in groundwater level, drying- and wetting-upwards cycles can be identified which can be correlated over large distances. The different parts of the depositional system, however, display a different response to variations in climate and ground-water level, and it remains yet unconfirmed how to link the different parts of the basin chronostratigraphically. In order to adequately explain and predict depositional facies, more sophisticated models are required based on a reliable correlation in relation to an understanding of the vertical and lateral facies relationships and the recognition of major significant surfaces.
(3) Base Permian Unconformity: Within the SPB, the boundary between the Permian Upper Rotliegend II and the underlying Carboniferous and older Rotliegend deposits is called the Base Permian Unconformity (Glennie et al., 2003; Geluk 2007). The unconformity consists of a series of amalgamated and superimposed erosive events as a result of prolonged uplift and erosion during the end of the Variscan Orogeny and before the start of the thermal subsidence of the SPB which initiated the deposition of the Upper Rotliegend II (Appendices B1c–B1i).
The contact with the underlying Carboniferous Limburg Group is generally characterised by a sudden downward transition from red-bed sandstones and claystones to grey fine-grained sediments commonly intercalated with coal seams, often combined with a downwards increase in compaction (due to pre-Permian burial of the Carboniferous). The younger part of the Limburg group, however, was deposited in more semiarid conditions and thick red-bed sand bodies may occur, in which case the boundary may be more difficult or impossible to discern based on wire-line logs only.
Core material recovered from several wells allows the characterisation of the sediments overlying and underlying an erosional surface associated with the BPU. The contact has a variable nature caused by variations in lithology, mineralogy, weathering, colour, etc. To date, a wide range of stratigraphic and sedimentological techniques can be applied to identify the boundary more reliably, but it is considered essential that these techniques not be used as stand-alone tools only. Examples of these techniques are:
Palynology: The oxidation of palynomorphs below the BPU due to leaching is a commonly recognised feature. This leaching, which is characterised by a secondary reddening of sediments, can penetrate as far as hundreds of metres below the BPU (Besly et al., 1993). As a result, the application of palynology is often not very useful in this type of setting.
Magnetostratigraphy: During the Permo-Carboniferous a rapid northward drift of the Southern North Sea area has been reconstructed. Although the current study area remained at low latitudes, the tangential relationship between latitude and inclination of the magnetic field does allow measurement of an angle exceeding the error margin associated with demagnetification techniques to be applied. If there is a significant time gap at the BPU, then declinations and in particular inclinations of the characteristic remnant magnetization can be significantly different. This, however, requires recovery of core material straddling the BPU, which is not very often available.
Petrography: Claystone lithologies of the Rotliegend usually reflect the dominance of illite and possibly some mica, whereas Carboniferous claystones reveal higher proportions of kaolinite and mixed-layer clays (smectite) reflecting deposition in more damp conditions. Further to this, observations from petrographic evaluations indicate that the Rotliegend and Carboniferous sandstones have different textural characteristics. A key feature of Rotliegend sandstones is the presence of reddening grain-coating haematite. There is no evidence for grain-coating haematite pre-dating quartz overgrowths in Carboniferous sediments, indicating a lack of primary haematite precipitation.
Compaction: shale compaction involves a series of progressive textural and compositional changes. Variations in the burial history between Carboniferous and Permian deposits may result in variations and stepwise changes in density, sonic velocity, and hydrogen index which can be identified by various wireline logs available.
Chemostratigraphy: The application of chemostratigraphic techniques has shown that variations in the inorganic geochemistry are used to differentiate Permian claystones from Carboniferous claystones. Results from the following techniques have been incorporated: XRD (X-ray diffraction), ICP-MS (inductively coupled plasma–mass spectrometry) and ICP–AES (inductively coupled plasma–atomic emission spectrometry).
Sediments of the Upper Rotliegend II belong to the most studied intervals in northwestern Europe, but despite receiving such a large amount of attention some important problems identified more than fifty years ago still remain, mostly due to the poor biostratigraphic control in these deposits. Nevertheless, significant progress has been achieved in understanding the key drivers and boundary conditions of the depositional system in the Upper Rotliegend II and its incorporation into a more genetic stratigraphic framework. It is evident that a single “silver bullet” stratigraphic or sedimentological solution does not exist for providing a more profound understanding of the depositional model. None of the methods described above should be used in isolation, but rather be integrated into the framework of sequence stratigraphy, providing the petroleum geologist an analytical and predictive geological model.
This paper is published by permission of the Nederlandse Aardolie Maatschappij BV, Shell International Petroleum Maatschappij BV and ExxonMobil. We gratefully acknowledge many of our colleagues for all the fruitful discussions over core, outcrop, wireline and seismic data in the North Sea area helping to formulate the ideas in this paper. The two anonymous reviewers are thanked for their valuable improvements on an earlier version. Reina Van Dijk (NAM) is thanked for her valuable and prompt drawing contributions.”
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.