Search

Understanding of the shallow shelf system in the Danish Basin during the Early Cretaceous has benefitted significantly from studying the previously overlooked Hauterivian–Aptian section of the Vedsted Formation of the Vinding-1 drill core. The presence of chalks in this section demonstrates that carbonate-rich pelagic sediment accumulated locally in the siliciclastic-dominated Danish Basin and that benthic carbonate production was insignificant. The area was not a carbonate platform in the Early Cretaceous and does not indicate any reworked carbonate supply from platform environments in the vicinity. The scarcity of benthic macrofossils in the cored section is due to the lack of a specialised boreal chalk fauna at that time, and the adjacent nearshore environment apparently did not support any substantial benthic carbonate production. A revised biostratigraphy of the cored section is presented based primarily on calcareous nannofossils, supported by foraminifera, ostracods, and belemnites. Four lithofacies describe the spectrum from marlstone to slightly marly chalk, and the facies succession characterises four depositional units recording two discrete transgressive–regressive cycles. The study provides a depositional record that permits sequence stratigraphic correlation to the Valdemar and Adda Fields in the Central Graben.

A 460 m long profile of the Cretaceous–Paleogene (K–T) boundary strata at Stevns Klint was measured by the late Professor A. Rosenkrantz probably in 1944. The measured profile was inherited by Finn Surlyk around 1974 together with other original boundary data. This material was dug up in a long-forgotten drawer in connection with detailed field work by the co-authors on the boundary succession in the late spring and summer of 2021. The profile illustrates the stratigraphy, geometry and palaeotopography of the boundary strata in unprecedented detail. The part of the cliff illustrated in the profile is today partly covered by beach ridges composed of flint rubble but is situated below the finest section of the lower Danian bryozoan mounds exposed at Stevns Klint. This coastal section is situated immediately adjacent to a large limestone quarry and was planned to be quarried away around 1937, but was saved by A. Rosenkrantz who demonstrated its great scientific and educational value to the authorities.

Outcrops are valuable for analogous subsurface reservoirs in supplying knowledge of fine-scale spatial heterogeneity pattern and stratification types, which are difficult to obtain from subsurface reservoir cores, well logs or seismic data. For petrophysical properties in a domain where the variations are relatively continuous and not dominated by abrupt contrasts, the spatial heterogeneity pattern can be characterized by a semivariogram model. The outcrop information therefore has the potential to constrain the semivariogram for subsurface reservoir modelling, even though it represents different locations and depths, and the petrophysical properties may differ in magnitude or variance. However, the use of outcrop-derived spatial correlation information for petrophysical property modelling in practice has been challenged by the scale difference between the small support volume of the property measurements from outcrops and the typically much larger grid cells used in reservoir models. With an example of modelling the porosity of an outcrop chalk unit in eastern Denmark, this paper illustrates how the fine-scale spatial correlation information obtained from the sampling of outcrops can be transferred to coarser-scale models of analogue rocks. The workflow can be applied to subsurface reservoirs and ultimately improves the representation of geological patterns in reservoir models.

The diagenetic evolution of chalk during burial has previously been established and can be summarized as follows: (1) ooze deposition and dewatering (burial 1000 m). Moreover, chalk particles tend to increase in size as a result of diagenesis, i.e. recrystallization or cementation. As a result of compaction and cementation, the shape, size and connectivity of chalk microporosity evolve during burial. However, if the evolution of bulk porosity during burial is well quantified, actual modifications of pore space topology are less understood. Quantifying pore space properties such as pore and throat sizes, aspect ratio and spatial distribution usually require costly instruments as well as long and complex experimental procedures. Instead, this study documents the application of mathematical morphology to investigate chalk microporosity from two-dimensional (2D) scanning electron microscope (SEM). Chalk samples (n = 105) have been collected from onshore quarries and offshore oil and gas reservoirs. Six chalk lithotypes are defined based on their mineralogical composition, i.e. clay and silica content, as well as the average size of chalk particles, and their apparent diagenetic impact, which appears strongly related to the mean grain size. The latter is used as a proxy for the diagenetic overprint, while the burial-diagenetic model offers a framework for the different lithotypes and their respective pore space properties. During intermediate burial ( 1000 m), pressure solution and cementation become pervasive and produce large clusters of coalescent calcite crystals (3–10 μm; mean = 5.13 μm). Total porosity is strongly reduced as a result (6–11%) while pore body and throat dimensions drop to 1.58 and 0.03 μm, respectively. A high aspect ratio (88.1) and lower proportion of rough porosity relative to total image porosity suggest that throats are proportionally more affected than pore bodies. Cementation leads to less dendritic, more globular and isolated pores. Where processes such as overpressure and oil charge have helped preserve porosity during deep burial (20–30%), pore and throat dimensions were locked to moderately high values (>2.0 and 0.1 μm, respectively), even in silica- and clay-rich chalk. Since the geometry of microporosity exerts a major control on fluid flow, reservoir modelling and recovery predictions could benefit from a better understanding on the geological controls on chalk microporosity. Not only can mathematical morphology quantify pore space properties that are fundamental in pore network and flow modelling, but by being applicable on large datasets, it could provide a new insight into the geological modifications of chalk microporosity during burial.

This paper presents a simulation study on the influence of rock heterogeneity on the distribution and magnitude of trapped CO2 resulting from a drainage/imbibition sequence in a saline aquifer with buoyant CO2. Four scenarios are studied, three simple 1D cases and a 2D case with heterogeneity derived from realistic architecture inspired by tidal sand deposits, having combined layering and crossbedding and sediment property contrasts. The four cases are examined in order to understand which underlying mechanisms are responsible for the results observed and therefore also which processes it is necessary to reflect in the simulation procedure. It is shown that it is important to take into account hysteresis in the capillary pressure. During the imbibition the capillary pressure decreases and the ability to capillary trap the CO2 is reduced and it may in some cases completely vanish. Thus, the capillary pressure hysteresis has a major impact on the amount of trapped CO2. If the imbibition curve has a threshold pressure the sealing power of the barrier is not completely lost, which leads to hyper-trapping, characterised by mobile CO2 being trapped with above end-point saturations. If the imbibition capillary pressure reaches zero, only local barriers constituting effective seals are able to trap free and potentially mobile CO2 and the trapping mechanism is exactly the same as the one acting beneath the top seal. The neglection of capillary pressure hysteresis may result in a large overestimation of the amount of trapped CO2.

In this study, we examined two sandstones bodies of the Camarillas Fm. The porosity and permeability models of each sandstone body, the tsunami deposit at the bottom and the barrier island – tidal inlet deposit at the top, supplied two independent reservoir models reflecting different sedimentary processes. Through study cases from outcrop data, we show that the dissolution of CO2 reached 40% of the total injected where the injection point was in thin zones (1–3 m-thick). The study cases presented here represent nearly one grid block (grid cell) of classic reservoir models, but are very complex in terms of petrophysical distribution. Geological model uncertainty is low (outcrop data), and facies are well correlated to petrophysics. Both deposits stored 50–60% of injected CO2 by residual and dissolution trapping over 7.5 years. The injection rate per day (maximum 200 sm3/day for the tsunami reservoir and 400 sm3/day for the barrier island − tidal inlet (b.i/inlet) reservoir) and the total amount of injected CO2 (108,000–116,000 sm3, ∼200 ton in 1.2–1.7 years) were very low in comparison with an in industrial project; however, results of this study have an important impact on the assessment and estimation of dissolved CO2 in aqueous phase, and could be used in studies on upscaling methods., This research is a contribution to the project: ‘Análisis de Cuencas Sedimentarias Continentales’, of the Gobierno de Aragón, the Análisis de Cuencas Sedimentarias Group of the UCM-CAM, and the projects CGL2011-23717 (Ministerio de Ciencia e Innovación of the Spanish Government and FEDER) and UZ2015-CIE-10 (University of Zaragoza). Additional financial support was provided by the CNPq (Conselho Nacional de Pesquisa e Desenvolvimento Tecnológico, Brasil) Research Grant (200147/2011-0) to F.M.L. Veloso.

The tsunami deposit (up to 3 m thick) of the Cretaceous Camarillas Formation in the Galve sub-basin (eastern Spain) is characterized by a large lateral extent (35 km2) and facies uniformity, consisting in fine to coarse subarkosic–arkosic sandstones. At the scale of outcrop observation, different lithofacies were distinguished and related to sedimentation processes. Sand facies distribution conditioned the facies heterogeneity at both mesoscopic or outcrop scales (10− 1–101 m scale) and sample scale (10− 3–10− 2 m). The sample features were up-scaled to that of a facies model using probability functions and variograms as well as to outcrop-scale data (geometry and size) of facies distribution, and it showed a good correlation with the facies distribution at the outcrop. Porosity is strongly correlated to permeability and the pairs of porosity and permeability values fall into the global hydraulic element (GHEs) 5, so that they can be up-scaled into reservoir models in terms of hydraulic properties. From data analysis, no apparent link between sand sorting and porosity and permeability values was observed. The petrophysical models, which were independent of facies models, were up-scaled taking into account the porosity and permeability values from sample data and the statistical analysis of their distribution along the outcrop. The permeability model was carried out as a function of porosity by applying a linear relation, which simplified the modelling process and discarded permeability uncertainties linked to facies distribution in the deposit. In spite of different source data, the petrophysical models show a distribution of lower and higher values that resembles the facies model. Consequently, our modelling results clearly suggest the link of facies type and their grain size distributions with the petrophysical properties into the deposit. Consistency between facies and petrophysical models and outcrop-scale observations make it possible to extrapolate to other deposits related to similar sedimentary processes., Research was financed by the projects CGL2011-23717 (Ministerio de Ciencia e Innovación of the Spanish Government and FEDER) and UZ2015-CIE-10 (University of Zaragoza). A research grant (200147/2011-0) to F. Veloso of the Conselho Nacional de Pesquisa e Desenvolvimento Tecnologico of Brasil provided additional financial support.

The concept of utilising available pore space in deep saline sandstone aquifers for storage of CO2 was recognised in the late 1980s. In 1996, the first commercial CO2 storage project began with injection into sandstones of the Utsira Formation in Norway. The formation is located above the Sleipner Formation from where the Sleipner field produces natural gas. The project was initiated due to a high CO2 content of the natural gas, which was subjected to a Norwegian offshore carbon tax. The natural gas is produced on the Sleipner platform where the CO2 is separated, captured and reinjected from a neighbouring platform. The potential for using the technology to reduce CO2 emissions from large stationary point sources initiated many research projects aimed at mapping areas with potential CO2 storage capacity around the world.

This study presents a novel laboratory-scale experimental/numerical approach that contributes to our understanding of micro-annular flow paths in cement-caprock interfaces. The approach is based on studying single-phase fluid flow in composite specimens with imperfect bonding between cement and caprock. Imperfect bonding and defect zones in the cement were initially characterized by X-ray micro computed tomography. Subsequently, the bulk permeability of the specimens was determined through single-phase fluid permeability measurements. Permeability was also evaluated numerically by finite-element modelling of steady single-phase flow in the specimens. The numerical modelling revealed that the average permeability of defect zones mapped from μ-CT images was likely to be within 1–10 D (0.1 × 10−11 to 1 × 10−11 m2). If a connected region of elevated permeability is established in situ, it can create a potential well integrity risk by providing a flow path up the annulus of the well. The integrated experimental and numerical framework employed in this study can form the basis for future systematic evaluation of leakage risks related to well integrity issues during underground CO2 storage, as well as in other applications such as drilling and completion of geothermal wells.

Avoiding CO2 leakages from storage reservoirs is crucial to ensure safe and cost-efficient Carbon Capture and Storage (CCS). This can only be done if effort is made to maintain well integrity throughout the entire life-cycle of a well. Cement integrity is especially important, since the interfaces between cement and rock or casing have been identified as weak links in today's well construction. The present paper focuses on the healing of fractures in well cement when the material is exposed to a CO2-brine water-alternating-gas (WAG) flooding scheme. Specimen characterization using computed tomography combined with electron microscopy documents the self-healing procedure in detail for a composite cement-rock specimen subjected to a WAG flooding scheme. The study revealed volumetric data on self-healing of cement cracks and chemical changes in the specimen as well as in aqueous chemistry upon CO2 exposure. The measured aqueous chemistry suggests CO2-cement interaction to be less pronounced with time thereby together with the observed cement self-healing suggesting that the risk of compromising the safety of a storage site by cement-CO2 chemical reactions is minimal.

The paper demonstrates how a potential site for CO2 storage can be evaluated up to a sufficient level of characterization for compiling a storage permit application, even if the site is only sparsely explored. The focus of the paper is on a risk driven characterization procedure. In the initial state of a site characterization process with sparse data coverage, the regional geological and stratigraphic understanding of the area of interest can help strengthen a first model construction for predictive modeling. Static and dynamic modeling in combination with a comprehensive risk assessment can guide the different elements needed to be evaluated for fulfilling a permit application. Several essential parameters must be evaluated; the storage capacity for the site must be acceptable for the project life of the operation, the trap configuration must be efficient to secure long term containment, the injectivity must be sufficient to secure a longstanding stable operation and finally a satisfactory and operational measuring strategy must be designed. The characterization procedure is demonstrated for a deep onshore aquifer in the northern part of Denmark, the Vedsted site. The site is an anticlinal structural closure in an Upper Triassic – Lower Jurassic sandstone formation at 1 800-1 900 m depth.

Long-lasting cement plugging of wells is crucial for successful CO2 storage in underground reservoirs. It requires a profoundly improved understanding of the behavior of fractured cement under realistic subsurface conditions including elevated temperature, high pressure and the presence of CO2 saturated brine. Here we report computed X-ray tomography studies on the effects of CO2 on cement. More specifically, we have exposed cured Portland G cement samples with pre-made microchannels mimicking fractures to CO2 saturated brine at elevated pressure (100 bars) and room temperature. The microchannels were observed to get filled with calcite (CaCO3) during the CO2 exposure. The extent of this self-healing was dependent on the diameter of the leakage path, with narrower channels more readily getting clogged. Chemical simulations taking into account the cement composition, CO2 availability, pH, pressure and temperature gave results consistent with our conceptual understanding of how the differences in dissolution/precipitation profiles in the cement may result from the availability of CO2. In particular, the modelling provides an explanation why calcite precipitates preferentially in the channels rather than on the external cement sample surfaces. We conclude that the localized precipitation can be ascribed to higher pH inside the cavities compared to near the external surfaces, owing to long diffusion distances giving a locally limited CO2 supply within the voids. © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Experiments on shale specimens targeted creep deformation in order to investigate whether this mechanism could be relied upon to close fractures and stop leakage near injection wells. Uniaxial compression tests were conducted with stress ramp up paused near expected failure stress of the tested shale and additional deformation, interpreted as initial creep deformation, was logged. The tests were repeated on shale samples exposed to HCl acid solutions, simulating expected pH conditions near the well upon exposure to injected CO2. The purpose of the tests was to compare the creep propensity of the material after possible dissolution of load-bearing minerals, to the virgin case. The results show slight enhancement of creep deformation. Exposure to super-critical CO2 on the same shale was carried out on hollow cylinder plugs, with radial closure of the borehole measured under radial stress conditions. Results show a doubling in radial deformation when compared to control tests with brine exposure. Finally, analyses of synchrotron beam tomography images were performed to quantify volume change between shale specimens preserved in oil and specimens exposed to HCl, showing that total volume increased by 9%. © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license https://creativecommons.org/licenses/by-nc-nd/4.0/

International audience; We assessed the synergetic benefits of simultaneous formation fluid extraction during CO2 injection for reservoir pressure management by coupled hydro-mechanical simulations at the prospective Vedsted storage site located in northern Denmark. Effectiveness of reservoir pressure management was investigated by simulation of CO2 storage without any fluid extraction as well as with 66% and 100% equivalent volume formation fluid extraction from four wells positioned for geothermal heat recovery. Simulation results demonstrate that a total pressure reduction of up to about 1.1 MPa can be achieved at the injection well. Furthermore, the areal pressure perturbation in the storage reservoir can be significantly decreased compared to the simulation scenario without any formation fluid extraction. Following a stress regime analysis, two stress regimes were considered in the coupled hydro-mechanical simulations indicating that the maximum ground surface uplift is about 0.24 m in the absence of any reservoir pressure management. However, a ground uplift mitigation of up to 37.3% (from 0.24 m to 0.15 m) can be achieved at the injection well by 100% equivalent volume formation fluid extraction. Well-based adaptation of fluid extraction rates can support achieving zero displacements at the proposed formation fluid extraction wells located close to urban infrastructure. Since shear and tensile failure do not occur under both stress regimes for all investigated scenarios, it is concluded that a safe operation of CO2 injection with simultaneous formation fluid extraction for geothermal heat recovery can be implemented at the Vedsted site.

Assessing the pressure buildup in CO 2 storage sites and especially the vertical propagation is vital for evaluation of site behavior and security. Vedsted structure in the Northern part of Jylland in Denmark consists of 290 m thick Gassum Formation at 2100 m depth forming the primary reservoir and is sealed by the 530 m thick Fjerritslev Formation which is mainly shale lithology with very low permeability. Overlying the caprock is a number of formations forming secondary reservoirs and seals including a 420 m thick Chalk Group which is overlain by 20–50 m Quaternary deposits. Seismic profiling of the structure shows the presence of northwest-southeast trending faults of which some originate in the upper layer of the Gassum reservoir and some reach the base Chalk Group layer. Two faults in the upper Gassum reservoir have been interpreted to be connected to the base Chalk Group. In order to evaluate potential risks associated with vertical pressure transmission via the faults through the caprock, a number of simulation cases have been run with various fault permeabilities spanning orders of magnitude to represent both the worst and best case scenarios. Fault rock permeability data were obtained from a literature study and range from 1000 mD (maximum value reported from sedimentary rock environment) for the worst case scenario down to 0.001 mD (sealing faults in sedimentary rock environment) for the best case scenario. The results show that after injecting 60 million tons (Mt) of CO 2 at a rate of 1.5 Mt/year for 40 years, overpressure is developed in the reservoir and about 5 bar is transmitted to the base Chalk Group for the 1000 mD fault permeability (open fault) case, while for the 0.001 mD (sealing fault) case the pressure buildup is confined within the primary caprock. The results also show that, approximately 0.3–5.0 bar overpressure can be transmitted to the base Chalk Group when the fault permeability is above 1.0 mD.

The Fjerritslev Formation in the Norwegian–Danish Basin forms the main seal to Upper Triassic–Lower Jurassic sandstone reservoirs. In order to estimate the sealing potential and rock properties, samples from the deep wells Vedsted-1 in Jylland, and Stenlille-2 and Stenlille-5 on Sjaelland, were studied and compared to samples from Skjold Flank-1in the Central North Sea. Mineralogical analyses based on X-ray diffractometry (XRD) show that onshore shales from the Norwegian–Danish Basin are siltier than offshore shales from the Central Graben. Illite and kaolinite dominate the clay fraction. Porosity measurements obtained using helium porosimetry–mercury immersion (HPMI), mercury injection capillary pressure (MICP) and nuclear magnetic resonance (NMR) techniques on the shale samples show that MICP porosity is 6–10% lower than HPMI or NMR porosity. Compressibility, from uniaxial loading, and elastic wave velocities were measured simultaneously on saturated samples under drained conditions at room temperature. Uniaxial loading tests indicate that shale is significantly stiffer in situ than is normally assumed in geotechnical modelling. Permeability can be predicted from elastic moduli, and from combined MICP and NMR data. The permeability predicted from Brunauer–Emmett–Teller (BET)-specific surface-area measurements using Kozeny’s formulation for these shales, being rich in silt and kaolinite, falls in the same order of magnitude as permeability measured from constant rate of strain (CRS) experiments but is two–three orders of magnitude higher than the permeability predicted from the 1998 model of Yang & Aplin, which is based on clay fraction and average pore radius. When interpreting CRS data, Biot’s coefficient has a significant and systematic influence on the resulting permeability of deeply buried shale.

Large scale CO2 storage has previously been considered for the Vedsted structure located in the Northern part of Jylland in Denmark. Pressure buildup in the Gassum reservoir and transmission to the shallower Chalk Group where the brine–fresh water interface resides need to be investigated as part of site qualification, as overpressure can push brine into the fresh water zone and thereby affecting aquifer performance. Pressure transmission from the reservoir into the surrounding formations, when fractures and faults are ignored, will depend on the properties and thickness of the sealing rock. The most important property to be considered is caprock compressibility and permeability. Laboratory experiments on centimeter-scale plugs and dynamic sonic velocity data from relevant shale formations in Denmark indicate that shale compressibility is lower than often assumed for reservoir simulation studies. The measured compressibility for the Fjerritslev Formation is 0.5 × 10−5 bar−1, which is an order of magnitude lower than the standard compressibility (4.5 × 10−5 bar−1) normally used for reservoir simulation studies. The consequences of this lower compressibility are investigated in a simulation case study and the results indicate that higher overpressure is created in the reservoir and the caprock. Overestimating caprock compressibility can therefore underestimate overpressure within the storage and sealing formations and this can have significant implication in the presence of highly permeable fractures and faults. The caprock permeability is measured on core samples using a geotechnical method of constant rate of strain (CRS) experiments which seem to match the modeled permeability data for the Fjerritslev Formation. We found an average vertical permeability of 0.1 μD for the Fjerritslev Formation from the samples measured. The sensitivity of pressure development for the caprock permeability has been studied by varying from one to three orders of magnitude higher and one to two orders of magnitude lower than the measured permeability of 0.1 μD. Injecting 60 million tons (Mt) of CO2 at a rate of 1.5 Mt/year into the Gassum Formation for 40 years indicates that, with permeability above 1.0 μD, overpressure can be transmitted through the 530 m thick Fjerritslev Formation caprock and further up into the overburden layers.

Potential changes in the recent tectonic stress field as a result of increasing pore pressure due to injection of CO2 may induce mechanical failure of the storage formation and its caprock, reactivation of existing faults or ground uplift if disadvantageous geological conditions are encountered. Coupled hydro-mechanical simulations can be applied to analyse and predict to which extent stress field and mechanical material property changes resulting from pore pressure elevation pose a serious impact on the mechanical stability of faults, reservoir and caprock. Within the frame of the on-going EU-funded SiteChar project, an assessment of CO2 storage performance and associated geomechanical impacts was investigated at a prospective deep saline onshore CO2 storage formation, the Gassum formation at Vedsted, situated in Northern Denmark. Simulations consider the constant injection of 3.15 Mt CO2/year over a time span of 40 years into the marine and fluvial sandstone aquifer of Upper Triassic to Lower Jurassic age, located at a depth between 1,700 m and 2,000 m. For this purpose, coupling of two reservoir and two geomechanical simulators was undertaken by different partners. The widely used commercial Eclipse software and the scientific code TOUGH2-MP were used for the multi-phase flow simulations, whereas the geomechanical modelling was carried out with the commercial codes FLAC3D (coupled to TOUGH2-MP) and VISAGE TM (coupled to ECLIPSE). Pore pressure distribution at selected time steps of the flow simulations conducted was used as input to the geomechanical codes to compute displacement and stress changes. A 3D reservoir-scale model with a lateral extent of 12 km x 16 km was initially applied for the flow simulations and extended to 50 km x 50 km for the geomechanical simulations to minimise the influence of boundary effects. The extended 3D geological model of the Vedsted site also comprises ten different stratigraphic units from the Lower Triassic up to the ground surface to include the overburden. Five northwest-southeast trending major faults implemented into both models were treated as either hydraulically open or closed for cross-flow. Simulation results indicate that changes in the recent stress field after 40 years of CO2 injection are mainly confined to the vicinity of the injection well due to the fact that the greatest pressure build-up occurs in the near-well area of the injector. The magnitudes of stress changes near the faults closest to the injector were insignificant and as such there is as negligible risk for their reactivation. Although shear strength is slightly reduced close to the injection well, the mechanical stability of the reservoir and its caprock is not decreased below critical values at any time during the injection period of 40 years and after. The simulation results further suggest that CO2 injection over a time span of 40 years may induce a vertical displacement of 9.4 cm at the top of the storage formation and a ground uplift of up to 10 cm. The results of the present study show that coupled hydro-geomechanical modelling can be well applied to improve our understanding of the coupled processes during CO2 storage and evaluate associated potential impacts such as ground deformation, fault reactivation and caprock failure.

When injecting CO 2 in to a subsoil aquifer for permanent CO 2 storage the pressure build up in the regional area encircling the site can extend far beyond the site delineation and mitigation procedures must be considered. The pressure build up can be controlled by production of water from the aquifer. In that context the synergy effect by combining CCS with geothermal energy production (GE) is obvious; i.e. the injection site for CCS may be surrounded by several GE plants, where the GE plants are operated so the net production of water can balance the injected CO 2 from the CCS operation sufficiently. Furthermore, the CO 2 plume migration may be controlled by operating the different sites as pressure sinks and sources. The paper illustrates the concept for an area in the northern part of Denmark, where a potential CCS site is characterized together with four prospective locations for GE plants. The GE plants are located in a radius of up to 10 km from the CCS site but still outside the closure of the CCS site. The Eclipse 100 reservoir simulator is used for simulations.

Within the Skagerrak-Kattegat region up to 14 Mt of CO 2 can annually be captured from power and industry sources. To establish a CCS infrastructure in the region it is necessary to identify and characterize potential CO 2 storage sites. Initial screening of the region has revealed large aquifers in the Upper Triassic Gassum Formation. In dynamic simulation studies 250 Mt of CO 2 were injected into the Gassum Formation over a period of 25 years. Identification and analysis of parameters that affect CO 2 storage capacity were performed. Parameters important for the migration speed and the dissolution rate of CO 2 in open dipping aquifers were investigated by a series of simulations on generic tilted reservoir models.

Denmark constitutes a low-enthalpy geothermal area, and currently geothermal production takes places from two sandstone-rich formations: the Bunter Sandstone and the Gassum formations. These formations form major geothermal reservoirs in the Danish area, but exploration is associated with high geological uncertainty and information about reservoir permeability is difficult to obtain. Prediction of porosity and permeability prior to drilling is therefore essential in order to reduce risks. Geologically these two formations represent excellent examples of sandstone diversity, since they were deposited in a variety of environments during arid and humid climatic conditions. The study is based on geological and petrophysical data acquired in deep wells onshore Denmark, including conventional core analysis data and well-logs. A method for assessing and predicting the average porosity and permeability of geothermal prospects within the Danish area is presented. Firstly, a porosity-depth trend is established in order to predict porosity. Subsequently, in order to predict permeability, a porosity–permeability relation is established and then refined in steps. Both one basin-wide and one local permeability model are generated. Two porosity-depth models are established. It is shown that the average permeability of a geothermal prospect can be modelled (predicted) using a local permeability model, i.e. a model valid for a geological province including the prospect. The local permeability model is related to a general permeability model through a constant, and the general model thus acts as a template. The applied averaging technique reduces the scatter that is normally seen in a porosity–permeability plot including all raw core analysis measurements and thus narrows the uncertainty band attached to the average permeability estimate for a reservoir layer. A “best practice” technique for predicting average porosity and permeability of geothermal prospects on the basis of core analysis data and well-logs is suggested. The porosity is primarily related to depth, whereas the permeability also depends on porosity, mineralogy and grain size, which are controlled by the depositional environment. Our results indicate that porosity and permeability assessments should be based on averaged data and not raw conventional core analysis data. The uncertainty range of permeability values is significantly lower, when average values are used.

Simulation studies on CO2 injection operations for geological storage shows that the pressure wave propagates much faster and wider than the CO2 plume. This must be addressed in any site modelling by proper boundary conditions since the pressure rise can reach the model delineation and then cause unrealistic high simulated pressure increase in the site model if the site model boundaries are closed. Any dynamic capacity estimation method depends on the modelled pressure distribution and tends to erroneously results, if the boundary problem is not addressed adequately.This study illustrates how the boundary problem can be analysed by developing a regional scale model surrounding the site model. The use of pore volume multipliers is demonstrated to be a robust option to handle the boundary condition problem.A case study from a Danish onshore locality is presented. Static and dynamic modelling was done in Petrel and Eclipse 100 software respectively.

Investigations of reservoirs, cap-rocks and traps in the Norwegian-Danish Basin have indicated a large potential for geological storage of CO2 onshore Denmark. The possible reservoir rocks include a variety of sandstones with different mineralogical composition, ranging from the Bunter Sandstone Formation containing K-feldspar, clay minerals, calcite, and dolomite as primary reactive minerals to the Gassum Formation containing albite, clay minerals, siderite, and ferroan dolomite as the most reactive minerals. A laboratory and modelling study was carried out to investigate the geochemical response of five potential reservoir rocks to CO2 storage in order to constrain predictions of mineral- CO2 reactions prior to geological storage. The study includes hydrogeochemical experiments, petrographical and mineralogical analyses, and reactive geochemical modelling. During up to 13 months of exposure to CO2 at reservoir conditions (70 °C; 20 MPa), the five different rock samples show little mineral reactivity as compared to similar rock samples exposed to N2 for the same period of time. However, during the period covered by the experiments, dissolution of carbonates present in the host rock is observed both from petrographical analysis and geochemical analysis using speciation calculations. Thus, for the Bunter Sandstone Formation calcite dissolution is apparently taking place in the laboratory experiments while for the Gassum Formation samples ferroan dolomite and siderite dissolution are the dominant mineral dissolution reactions taking place. As a result of the speciation calculations, a procedure for back calculation of chemical analyses to true experimental conditions is suggested in order to reflect the correct saturation state of the pore water with respect to carbonates, primary silicates and aluminosilicates. It is suggested that performing such back calculations is essential to the understanding of the future evolution of the hydrogeochemistry of aquifers aimed at as CO2 storage reservoirs.

Since the end of June 2008, carbon dioxide (CO 2 ) is being injected in the Stuttgart Formation at Ketzin, Germany as part o f the European Union’s CO2SINK project. The injection well (Ktzi 201) is roughly 50 and 100 m away from observation wells Ktzi 200 and Ktzi 202, respectively. CO 2 was detected at the closest observation well on the 15th of July 2008, approximately 20 days after injection commenced. Breakthrough at the Ktzi 202 well was recorded on March 21, more than 8 months after injection was initiated. Dynamic simulations of the injection and flow of CO 2 into the subsurface at Ketzin will be described. The three dimensional (3D) geological model was built based on a 3D seismic survey as well as logging and core analysis data. The formation consists of fluvial sandstone channels within a muddy flood plain at mean depth of 650 m within an anticlinal structure. The geological model uses a cell size of 20×20 m and a layer thickness of 0.5 m, resulting in a total of ∼7 million cells. The model is upscaled (i.e., coarsened) and the grid is refined locally in the zone comprising the injection and observation wells. This upscaled model has corner point geometry with 78×74×91 (525, 252) grid cells in X, Y, and Z directions. X and Y direction grid sizes around the wells are respectively 4 m and 5 m. Average grid size in Z direction is about 0.5 m. Following the upscaling of the geological or static model, a blackoil commercial streamline simulator was used to simulate the flow of injected CO 2 according to the actual injection rate history. The phase behavior of CO 2 and brine were described by blackoil pressure-volume-temperature (PVT) tables. The blackoil PVT was modeled by imposing brine properties to the simulator oil model and CO 2 properties to the simulator gas model. In this manner, solubility of CO 2 into water was taken into account. Salinity of the brine is represented by the appropriate density. Fluid properties (density, viscosity) are pressure dependent and isothermal (at reservoir temperature). The outcome of the simulation is the time for injected CO 2 to arrive at the observation wells and the history of bottom-hole pressures (BHP) as a function o f time that can be compared to measurements taken in the bottom of the injection well. History matching was performed by adjusting the permeability (along and across bedding) until model BHPs agreed with the measured ones. Permeability was chosen as history matching parameter because of its high degree of uncertainty and relies on the porosity/permeability relation derived from core measurements. Good agreement was obtained with a multiplicator of 0.1 applied to the permeability across and along bedding. Breakthrough time for the closest observation well was in good agreement with reality. However the breakthrough time at the most distant well was underestimated by several months. We believe that geological features at distances greater than 50 m from the injection well may be responsible for the mismatch and should be investigated further.

Vattenfall is a major operator on the energy market in Europe and have a broad portfolio of energy production from nuclear power, biomass and coal fired power plants together with a growing amount of wind turbines. Large research programs are looking into optimizing production and reducing environmental impact from the different units. It is Vattenfall vision to be CO 2 neutral in the Nordic countries in 2030 and for the entire group in 2050. To obtain these goals Vattenfall is very active in developing technologies for carbon capture and storage (CCS). In Denmark Vattenfall operates three coal-fired power plants and a screening survey in 2007, matching power plant and storage site location–source sink matching - pointed towards Nordjyllandsvaerket near the city of Aalborg as the most obvious plant with safe onshore storage opportunities only 30 km away. It was decided to develop the project into a full scale demo project with post-combustion capture, pipeline transport and onshore storage in an aquifer. The whole project has been driven by a very firm time schedule in order to meet the 2015 deadline. When the CCS project on the Vattenfall operated power plant Janschwalde in Germany was granted funding through the European Commission the CCS project at Nordjyllandsvaerket was turned into an early commercial project aiming at 2020 for commissioning. The site characterization is still ongoing on the Danish site.

The onshore parts of the Norwegian-Danish Basin and the northern rim of the North German Basin contain extensive saline aquifers of potential interest for CO2 storage. The Triassic Bunter Sandstone and Skagerrak formations comprise arkoses, subarkoses and lithic arkoses, with feldspar being dominated by K-feldspar. The authigenic phases are dominated by carbonates (mainly calcite and dolomite), anhydrite, clay minerals. The Upper Triassic–Lower Jurassic shallow marine and paralic sandstones (Gassum Formation) comprises mainly subarkoses and sublitharenites. The authigenic phases are typically carbonates (typically siderite and ferroan dolomite), clay minerals and pyrite. The Middle Jurassic fluvial, estuarine and shallow marine sandstones (Haldager Sand Formation) is characterised by quartz arenites and subarkoses. The authigenic phases comprise clay minerals, pyrite, carbonate, quartz and feldspar. Experiments in the laboratory were designed and performed in order to investigate the response of these mineralogical different reservoir sandstones to CO2 storage.In the batch experiments, the sandstone plugs were exposed to CO2 saturated synthetic fluid at present day reservoir temperatures and pressures (70 °C at 20 MPa) for a period of 7–14 months. Parallel experiments with N2 gas show minor adjustments of the chemistry of the pore fluids until equilibrium is reached prior to first sampling (∼30 days). After 7 months of experiment the fluids show increased values of HCO3−, Ca2+ and Mg2+ in the Skagerrak and Bunter Sandstone formations. Whereas the Gassum formation (and to a lesser extend the Haldager Sand Formation) show increased values of HCO3−, total Fe and Mn+ in the fluid. These changes are probably related to the dissolution of the typical carbonates in the different formations.

The injection of CO 2 at the Ketzin storage site and the chemical detection of its arrival in the observation well allowed testing of different numerical simulation codes. ECLIPSE 100 (E100, black-oil simulator), ECLIPSE 300 (E300, compositional CO 2 STORE) and MUFTE-UG were used for predictive modelling applying a constant injection rate of 1 kg s −1 CO 2 and for a history match applying the actual variable injection rate which ranged from 0 to 0.7 kg s −1 and averaged 0.23 kg s −1 . The geological model applied, is based on all available geophysical and geological information and has been the same for all programs. The results of the constant injection regime show a good agreement among the programs with a discrepancy of 21–33% for the CO 2 arrival times. However, it is determined from the comparison of the cumulative mass of CO 2 at the time of CO 2 arrival that the injection regime is an important factor for the accurate prediction of CO 2 migration within a saline aquifer. Comparing the actual variable injection regime with the simulations applying a constant injection rate the results are relatively inaccurate. Regarding the actual variable injection regime, which was evaluated using all three simulators, the computational results show a good agreement with the data actually measured at the first observation well. Here, the calculated arrival times exceeded the actual ones by 8.1% (E100), 9.2% (E300) and 17.7% (MUFTE-UG). It can be concluded that irrespective of the deviations of the simulations, due to combinations of different codes and slight differences in input parameters, all three programs are well equipped to give a reliable estimate of the arrival of CO 2 . Deviations in the results mainly occur due to different input data and grid size choices done by the different modelling teams working independently of each other. Deviations of the simulations results compared to the actual CO 2 arrival time result from uncertainties in the implementation of the geological model, which was set up based on well log data and analogue studies.

It is well established that dynamic conditions expressed as tilted fluid contacts characterize most hydrocarbon accumulations in North Sea Chalk reservoirs. Chalk is a low-permeability, high-porosity rock and properties grade smoothly from reservoir over baffle to seal. The natural dynamic conditions prevail because pressure dissipation takes place through the rockmatrix, as fracture-supported flow often is minimal. The dynamic conditions are imposed by processes occurring on a geological time-scale and result mainly in lateral pressure differences in the water zone and even in lateral pressure differences in the oil zone. Re-equilibration of fluid contacts also occurs on a geological time-scale. These processes are of paramount importance for trap definition and impose severe restrictions on migration distances. Reservoir simulation techniques are applied, in combination with back-stripping, to the simulation of geological time-scale secondary migration and trapping. Flow simulation of the filling dynamics of a chalk reservoir shows a complex filling geometry due to the high capillary entry pressures in the low-permeability chalks. Such internal barriers will re-direct hydrocarbons and residual oil can be left on the migration route. The process of hydrocarbon charging is slow and equilibration of hydrocarbons with respect to pressure gradients, therefore, also occurs very slowly. The Kraka Field and the Dan and Halfdan fields are subjected to studies of primary oil charging and re-migration in this paper and the dynamic oil on Dan Field west flank is successfully mimicked. Results show that a time span in the order of 2 Ma is required for the hydrocarbons to reach the summit in an approximately equilibrium state from a flanking position. However, even dynamic equilibrium may not be fully obtained due to re-perturbation by tectonic movements and changing water zone pressure gradients. Results show that saturation profiles in drilled wells may appear in drainage equilibrium while under partial re-imbibition, which impairs saturation modelling.

The VarScale program (1) calculates dispersion variances for different support volumes, (2) performs variogram and histogram scaling, and (3) performs scaling of a linear model of coregionalization (LMC). The input variogram may be at any scale. The required dispersion variances (or average variogram values) are calculated. User input is automatically passed between different components of the program. This paper describes the theoretical background of the scaling algorithms, the program structure of varScale and presents three examples. Variogram scaling from core scale to well-log scale is illustrated with data from a Danish North Sea reservoir. Histogram and LMC scaling are illustrated with data from a West Texas reservoir using porosity and seismic data.

The spatial variability of porosity and permeability was determined for a section in the uppermost Maastrichtian chalk exposed in the Sigerslev quarry at Stevns Klint. The aims were threefold: (1) to quantify the spatial variability pattern and its link to geology by applying systematic variogram analysis procedure, (2) to detect potential cyclicity in the petrophysical properties in this pure and overtly homogeneous chalk, and (3) to compare the section with chalk reservoirs in the North Sea in order to asses to what extent this onshore sequence has a potential as an analogue of the offshore, more deeply buried chalk reservoirs. The interval is of comparable stratigraphic age to the uppermost reservoir zones in the Tor Formation of the North Sea chalk reservoirs. The variability and spatial correlation of porosity and permeability in both horizontal and vertical directions are interpreted in a geological context and show indications of small-scale heterogeneity at 15–25 cm scale, but the clear cyclic layering described from other chalk deposits is not recognised at this locality. The investigated outcrop is not a close analogue to North Sea reservoir chalk, but some aspects are common including basic material properties, porosity/permeability trends and the variability pattern. The outcrop has a potential as analogue for some of the onshore subsurface chalk successions at shallow burial depth that form important aquifers.

ULTimateCO2 will assess the long-term CO2 storage behaviour in terms of efficiency and security. The project is dedicated to studying the main physical processes needed to develop a better, quantitative understanding of the long- term geological storage of CO2, namely: (i) reservoir trapping, (ii) sealing integrity of caprock, and (iii) well leakage. Close collaboration with the NER300 candidates and EEPR demonstration sites will underpin all investigation with relevant supply of data industrial context. ULTimateCO2 will define a set of recommendations that will enable both regulators and operators to demonstrate that site specific long-term site performance will lead to permanent and safe CO2 containment.

At Ketzin, about 25 km west of Berlin (Germany), the saline aquifer of the Triassic Stuttgart Formation is used for a carbon dioxide storage research project. The formation is lithologically very heterogeneous, reflecting a complex fluviatile facies distribution pattern. We focused on the development of a primary geological reservoir model as commonly employed for dynamic modelling during the planning and early injection stages of a storage project. Due to the need to capture the complex geometrical structure of the Stuttgart Formation, despite limited availability of exploration data, stochastic modelling techniques were employed. Firstly, we modelled the facies architecture of the reservoir and, secondly, assigned porosity and permeability values to the facies types included in the model. Petrophysical parameters for each facies type were quantified using site-specific porosity histograms and related permeability functions. The comparison of dynamic flow simulation results and well-test interpretations, and furthermore with the first observed monitoring data, helped to focus the modelling work and to adjust monitoring plans. Modelling is understood as an iterative process, both with respect to data arrival and progressively improving the understanding of the reservoir, but also with respect to the problem which the model is being designed to address.

This paper describes a system for generating variomaps (autocorrelation or variance as a topological map) of binary images. The images are produced from plane sections through porous media. The 2-D autocorrelation function is calculated by Fourier transformation. The resulting map of the autocorrelation illustrates the mean porel size and directionality (anisotropy) in the analyzed section. The method has been tested on porespace combined within simulated grain packs, two types of sandstones, and catalyst material. The software is written in Pascal for PC systems and circumvents the DOS 640K barrier, thus enabling the handling of even large images. Calculation time for a 512 × 512 Fourier transform is less than 2 min on a 33 MHz 486 PC.

Rock samples of the Upper Permian Ca-2 unit of the Løgumkloster-l well have been subjected to chemical analysis. The Ca-2 interval was undoubtedly originally an aragonite which have undergone several episodes of diagenesis including a dolomitization and a heavy anhydrite mineralization. Two original facies types are recognized, an oolitic shoal facies, and a lagoonal carbonate facies. The samples have been characterized by neutron activation, atomic absorption, Xray diffraction, palaeomagnetism, optical and electron microscopical examination. Samples of the drilling mud has been analyzed as well, and it is found that contamination of the samples included in the present study from the drilling mud is unlikely to have happened. Imprints of both the original facies and the diagenetic events are found in the present work. The Sr and Ba abundances reflect the original facies-bound distribution, which is also reflected in the normalized REE distribution patterns. A low-Sr anhydrite component is clearly distinguishable in several parameter cross plots. Iron, Sc, REE and several other elements are found to be situated in the EDTA-insoluble residue, which has been identified primarily as a muscovite or clay mineral phase. Pyrite is present, although of small relative volume it is thought to be responsible for the occurrence of As, Sb, and possibly Se. A small fraction of the Cr, As, and Sb may originate from the heavy part of the crude oil, the introduction of which was the last process of alteration the rock experienced. A geomagnetically reversed period is encountered in the middle of the Ca-2 interval, and is tentatively identified as one of the Tatarian reversed events.

The present paper describes a technique for determination of the free water level (FWL) in low-permeability chalk reservoirs along slanted or horizontal wells from logged saturations and porosities. The calculation is done by utilizing existing empirical correlations for drainage and imbibition capillary pressure curves. For each logged saturation and corresponding porosity the FWL is determined so that the calculated saturation equals the logged value. The method takes into account possible imbibition, which results in a FWL that is shallower than the original paleo FWL. The method has been applied to field data from the North Sea and it is shown that it can capture a tilting water level along a horizontal well.

Geostatistical-simulation techniques are increasingly being used to create heterogeneous realizations for flow modeling and to assess uncertainty in hydrocarbon resources and reserves. These geostatistical-simulation techniques reproduce the input statistics within ergodic fluctuations. The input statistics representing various model parameters must be computed from data that are representative of the entire domain being modeled. Geostatistical simulation does not accommodate a lack of representativeness in the data. Moreover, the extent to which the input statistics are reproduced depends almost exclusively on the size of the modeling domain relative to the range of spatial correlation; fluctuations in realizations of the full reservoir model do not depend entirely on the uncertainty of the input statistics. It is necessary to explicitly incorporate the uncertainty of the input statistics because they have a much larger and more realistic impact on the uncertainty of the full reservoir model than stochastic fluctuations. The best practices for determining representative input values of model parameters and quantification of their uncertainty are presented in this chapter.

Since April 2004, preparatory work prior to CO2 injection has been conducted in the CO2SINK Project, the European Union's first research and development activity on the in-situ testing of geological storage of CO2 near the town of Ketzin, Germany. Carbon dioxide will be injected into a saline aquifer of the Triassic Stuttgart Formation in an anticlinal structure of the northeast German Basin. The drilling of one injection and two observation wells will commence at the end of 2006. The predrilling phase focuses on the baseline geological parameters of the anticline. The Stuttgart Formation is lithologically heterogeneous; it consists of sandy channel-(string)-facies rocks, with good reservoir properties alternating with muddy flood-plain-facies rocks of poor reservoir quality. Playa-type rocks form the immediate cap rock above the CO2SINK reservoir. A geostatistical approach has been applied to describe the reservoir architecture between and beyond well control. This model forms the basis for the generation of reservoir-dynamic models of CO2 injection that assist in the planning of injection operations and in the understanding of CO2 plume evolution. A verification of the geometry of the reservoir and the structural situation of its overburden is expected from a three-dimensional baseline seismic survey that was conducted in the autumn of 2005. Laboratory experiments under simulated in-situ conditions were performed to evaluate the geophysical signature of rocks saturated with CO2. The chemical composition of the groundwater and the CO2 flux in the soil were analyzed across the Ketzin anticline, providing the baseline for a monitoring program during and after injection of CO2, targeted at the detection of potential CO2 leakage from the storage reservoir.