Abstract The Kilmar Field, part of the Tors complex (Kilmar and Garrow fields), was discovered in 1992 and is located on the northern margins of the Southern North Sea Basin. Gas is produced from Namurian sandstones, at a depth of 11 000 ft, from a 25 km 2 closure at the Base Permian level. The reservoir was deposited as a series of stacked channel sandstones in a fluvio-deltaic setting. Seismic imaging of intra-Carboniferous strata is limited, so mapping of individual bodies of sandstone is not achievable. The development philosophy has been to maximize the drilled lengths of specific reservoir units and to contact multiple sandstone bodies by drilling long, high-angle, multi-bore production wells. The sandstones are of low to medium porosity and permeability, supplemented by connection through a fracture network. At project sanction in 2005, the combined gas-in-place resource in Kilmar was estimated to be 311 bcf and a total of 75 bcf gas recovery from three wells was forecast. Cumulative gas production to date is 69 bcf. Whilst the gas-in-place has changed little, the distribution has changed between segments. The recovery factor for the field is 24%. Infill drilling opportunities have been identified but are gas price dependent.

Abstract Machar is one of several diapir fields located in the Eastern Trough of the UK Central North Sea. It contains light oil in fractured Cretaceous–Danian chalk and Paleocene sandstones draped over and around a tall, steeply-dipping salt diapir that had expressed seafloor relief during chalk deposition. The reservoir geology represents a complex interplay of sedimentology and evolving structure, with slope-related redeposition of both the chalk and sandstone reservoirs affecting distribution and reservoir quality. The best reservoir quality occurs in resedimented chalk (debris flows) and high-density turbidite sandstones. Mapping and characterizing the different facies present has been key to reservoir understanding. The field has been developed by water injection, with conventional sweep in the sandstones and imbibition drive in the chalk. Although the chalk has high matrix microporosity, permeability is typically less than 2 mD, and fractures are essential for achieving high flow rates; test permeabilities can be up to 1500 mD. The next phase of development is blowdown, where water injection is stopped and the field allowed to depressurize. This commenced in February 2018.

Abstract The Stella Field is located in UKCS Block 30/6a, c. 230 km SE of Aberdeen. Stella Field is a four-way, dip-closed and salt-cored dome with a 600 ft gas-condensate column underlain by a 250 ft oil column within the 5–30 ft thick Paleocene Andrew Sandstone Member. Oil is also present in the underlying chalk reservoirs of the Danian Ekofisk and Maastrichtian Tor formations. Following the discovery of the field in 1979 there have been three phases of appraisal followed by the recent development, with first oil in February 2017. The field development has been challenging as the 38-year gap between discovery and first oil illustrates. Principal challenges have included structural undulations and radial faulting combined with a thin primary reservoir, variation in hydrocarbon, compartmentalization and depletion relating to producing fields. These issues have been reviewed following the latest drilling results and ideas on the petroleum geology updated. Block 30/6a, containing the Stella Field, was originally awarded to Shell/Esso in the first Licensing Round in 1964 as part of multi-block licence P011. The current Stella P.011 licence holders are Ithaca Energy (UK) Limited (operator) with 54.66%, Dyas UK Limited with 25.34% and Petrofac GSA Limited with 20%.

Abstract The application of production geochemistry techniques has been shown to provide abundant and often low-cost high-value fluid information that helps to maximize and safeguard production. Critical aspects to providing successful data relate to the appropriate sampling strategy and sampling selection which are generally project-aim-specific. In addition, the continuous direct integration of the production geochemistry data with subsurface and surface understanding is pivotal. Examples from two specific areas have been presented including: (a) the effective use of IsoTubes in the production realm; and (b) the application of geochemical fingerprinting primarily based on multidimensional gas chromatography. Mud gas stable carbon isotopes from low-cost IsoTubes have been shown to be very effective in recognizing within-well fluid compartments, as well as recognizing specific hydrocarbon seals in overburden section, including the selective partial seal for only C 2+ gas species. With respect to geochemical fingerprinting, examples have been presented related to reservoir surveillance including compartmentalization, lateral and vertical connectivity, as well as fluid movements and fault/baffle breakthrough. The production-related examples focus on fluid allocation within a single well, as well as on its application for pipeline residence times, fluid identification and well testing.

ABSTRACT Anticlines along the western margin of the South Viking Graben in the Brae area of the U.K. North Sea form the structural components of large structural and stratigraphic traps within Upper Jurassic Brae Formation submarine fan deposits. Various interpretations of the origin of these anticlines and their attendant inboard synclines at the graben boundary have been previously published, with both gravity-driven processes and inversion being invoked. Based on regional interpretation of 3-D seismic data sets and analysis of thickness variations in uppermost Jurassic and Cretaceous sequences in numerous wells, it is concluded that gravity-driven processes were more important than inversion. Differential compaction of mudstone-rich slope deposits laterally adjacent to coarse clastic submarine fan reservoirs has resulted in the field reservoirs now being at slightly higher elevations than the finer grained deposits along the length of the anticlines. Compaction of the very thick sandstone- and mudstone-dominated successions in the basin center has also been greater than that of the more conglomeratic successions adjacent to the basin margin, where sequences are underpinned by the slope of the footwall, resulting in over-steepened slopes toward the basin on the outboard side of the anticlines. Movement of Permian salt that underlies the Jurassic (and Triassic) in the basin has also had significant broad effects on the Upper Jurassic structures, creating depressions and underpinning some anticlines. Continued slow subsidence of the basin-fill down the main graben boundary fault system in the under-filled rift during the latest Jurassic and Early Cretaceous, above changes in footwall slope (from eroded slope to graben-boundary fault, or, in the case of East Brae, across a plunging basement nose) is considered to be the primary cause of the anticlines and their inboard synclines. Reversal of movement along the main boundary fault, causing inversion of the graben-margin sequences, is considered unlikely as the primary mechanism for anticline formation. Additional movement down the graben-boundary fault system in the early Maastrichtian may have slightly tightened the anticlines. Final minor fault movement along the graben margin occurred in the mid Eocene, but this is unlikely to have significantly affected the Brae structures. Some of the anticlines provide evidence of the presence of the underlying thick reservoir sequences (due to differential compaction over conglomeratic sections), but not all positive structural features contain coarse clastic sediments.

Abstract The Tyra Field in the central North Sea is located in Palaeogene and Upper Cretaceous chalk. It contains a natural gas zone underlain by an oil leg. Based on analysis of logs and core data from ten wells drilled prior to the field being put into production, normalized water saturation depth-trends from logs were compared with normalized water saturation depth-trends predicted from capillary pressure core data. The ten wells lie close to a SW–NE cross section of the field. For the gas–oil contact, a free contact measured in one well corresponds to a practically horizontal contact interpreted from logging data in the remaining wells. A westerly dipping oil–water contact was found from logging data. Comparison of the depth-wise trends in normalized water saturation among the different wells indicates a regional pattern: in the western side of the field, the trends correspond to a situation of imbibition, where the free water level overlies an interval of residual oil, whereas in the eastern part of the field, the depth-wise trends in normalized water saturation correspond to a situation of drainage. The free water level apparently dips to the east due either to hydrodynamic action or to pressure inequilibrium in the aquifer following tectonic tilting.

Abstract A biostratigraphic review, conducted on 34 wells from the chalk of the Eldfisk Field, Norwegian Central Graben, has been integrated with petrophysical, geophysical and sedimentological information resulting in a revised lithostratigraphic framework for the chalk on this structure. Chalk of Danian to Turonian age is divided into five formations: the established Ekofisk Formation of Danian age and Tor Formation of Maastrichtian age, together with a new three-fold division of the Hod Formation, namely the Magne Formation of Campanian age, the Thud Formation of Santonian age and Narve Formation of Coniacian to Turonian age. This work demonstrates the application of this three-fold division of the Hod Formation. Internal field specific subdivisions of all formations are also presented for the Eldfisk Field. This lithostratigraphic framework is applied across the Eldfisk Field, together with the recognition of erosional features, unconformities, areas of non-deposition, reworking and lateral changes in biofacies. The results have also allowed recognition of the following regionally synchronous tectonic phases for the first time on a Norwegian chalk structure: Stille's Ilsede phase (Late Turonian–Coniacian) and Wernigerode phase (Late Santonian–‘earliest’ Campanian), Mittel–Santon phase (Middle Santonian) of Niebuhr et al. and Reidel's Peine phase (‘latest’ Early Campanian), together with un-named phases of ‘latest’ Campanian, intra Mid Maastrichtian and (previously unrecognized?) intra Danian age. Evidence for these tectonic phases is compared with work from Denmark, Germany and the Anglo-Paris Basin. An innovative approach to mapping lateral biofacies (principally water depth) variations has been applied using the microfaunal database. This enhances understanding of the timing of structural phases when integrated with time lines generated by nanoplankton data. Biofacies proxies for silica content in the sediment may also correlate with changes in reservoir quality. Biofacies interpretations have also facilitated the identification and mapping of allochthonous bioclastic rich debris flow deposits. The fully calibrated biostratigraphic, lithostratigraphic and tectonostratigraphic frameworks presented can be applied to chalk structures regionally.

Abstract A high magnitude of overpressure is a characteristic of the deep, sub-Chalk reservoirs of the Central North Sea. The Upper Cretaceous chalk there comprises both reservoir and non-reservoir intervals, the former volumetrically minor but most commonly identified near the top of the Tor Formation. The majority of non-reservoir chalk has been extensively cemented with average fractional gross porosity of 0.08, and permeability in the nano- to microDarcy range (10 −18 –10 −21 m 2 ), and sealing properties comparable to shale. Hence deeply buried chalk is comparable to shale in preventing dewatering and allowing overpressure to develop. Direct pressure measurements in the Chalk are restricted to the reservoir intervals, plus in rare fractured chalk, but reveal that Chalk pressures lie on a pressure gradient which links to the Lower Cenozoic reservoir above the Chalk and the Jurassic/Triassic reservoir pressures below. Hence a pore pressure profile of constantly increasing overpressure with increasing depth is indicated. Mud weight profiles through the Chalk, by contrast, show many borehole pressures lower than those indicated by these direct measurements, implying wells are routinely drilled underbalanced. The Chalk is therefore considered the main pressure transition zone to high pressures in sub-Chalk reservoirs. In addition to its role as a regional seal for overpressure, the Base Chalk can be shown to be highly significant to trap integrity. Analysis of dry holes and hydrocarbon discoveries relative to their aquifer seal capacity (the difference between water pressure and minimum stress) shows that the best empirical relationship exists at Base Chalk, rather than Base Seal/Top Reservoir, where the relationship is traditionally examined. Using a database of 65 wells from the HP/HT area of the Central North Sea, and extending the known aquifer gradients from the Fulmar reservoirs via Base Cretaceous to Base Chalk, leads to a risking threshold at 5.2 MPa (750 psi) aquifer seal capacity. Discoveries constitute 88% of the wells above the threshold and 36% below, with 100% dry holes where the aquifer seal capacity is zero (i.e. predicted breached trap). This relationship at Base Chalk can be used to identify leak points which control vertical hydrocarbon migration as well as assessing the risk associated with drilling high-pressure prospects in the Central North Sea.

Abstract At South Arne a highly repeatable time-lapse seismic survey (normalized root-mean-square error or NRMS of less than 0.1) allowed us to reliably monitor reservoir production processes during five years of reservoir depletion. Time-lapse AVO (amplitude v. offset) inversion and rock-physics analysis enables accurate monitoring of fluid pathways. On the crest of the field, water injection results in a heterogeneous sweep of the reservoir, whereby the majority of the injected water intrudes into a highly porous body. This is in contrast to a pre-existing reservoir simulation model predicting a homogeneous sweep. On the SW flank, time-lapse AVO inversion to changes in water saturation Δ S w reveals that the drainage pattern is fault controlled. Time-lapse seismic data furthermore explain the lack of production from the far end of a horizontal producer (as observed by production logging), by showing that the injected water does not result in the expected pressure support. On the highly porous crest of the reservoir compaction occurs. Time-lapse time shifts in the overburden are used as a measure for compaction and are compared with predictions of reservoir compaction from reservoir geomechanical modelling. In areas where compaction observations and predictions disagree, time-lapse seismic data give the necessary insight to validate, calibrate and update the reservoir geomechanical model. The information contained in time-lapse seismic data can only be fully extracted and used when the reservoir simulation model, the reservoir geomechanical model and the time-lapse seismic inversion models are co-visualized and available in the same software application with one set of coordinates. This allows for easy and reliable investigation of reservoir depletion and gives deeper insight than using reservoir simulation or time-lapse seismic individually.