SUMMARY TECHNICAL REPORT of CO2SINK

CO2SINK Summary Technical Report

SES6-CT-2004-502599

CO2SINK

In-situ R&D Laboratory for Geological Storage of CO2 Integrated Project Thematic Priority: 6.1.ii

Start date of project: 01.04.04 Duration: 72 months Project co-funded by the European Commission within the Sixth Framework Programme (2002-2006) Dissemination Level PU Public PP Restricted to other programme participants (including the Commission Services) RE Restricted to a group specified by the consortium (including the Commission Services) CO Confidential, only for members of the consortium (including the Commission Services)

CO2SINK Summary Technical Report

CONTENT

SUMMARY ...... 3

INTRODUCTION ...... 88

GEOLOGICAL AND HYDROGEOLOGICAL SITE CHARACTERIZATION ...... 1212

LABEXPERIMENTS: ROCK FLUID INTERACTION ...... 1818

DYNAMIC FLOW MODELING ...... 2222

RISK ASSESSMENT ...... 2727

SITE PREPARATION AND INJECTION FACILITY ...... 3131

EXPERIENCES OF STORAGE OPERATION ...... 3838

MONITORING CO2 MIGRATION ...... 4141

DISSEMINATION AND PUBLIC OUTREACH ...... 5454

OUTLOOK ...... 5959

CO2SINK Summary Technical Report Chapter 1

SUMMARY

Hilke Würdemann1, Günter Borm 1,2,Frank R. Schilling1,2

1Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ), Telegrafenberg, 14473 Potsdam, Germany 2 Universität Karlsruhe, KIT, Engler-Bunte-Ring 14, 76131 Karlsruhe, Germany

CO2SINK Summary Technical Report

The CO2SINK (CO2 Storage by Injection into a Natural Saline Aquifer at Ketzin) pilot project aimed for a better understanding of geological CO2 storage in a saline aquifer. The main focus of the project was the development and testing of monitoring techniques. Furthermore, CO2SINK served as a test site for CO2 injection and safety measures. Through the European project CO2SINK the most advanced monitoring project of the subsurface worldwide was established. It triggered a broad variety of national and international activities, which contributed greatly to the success of the storage operation. The EUROGIA Projects COSMOS (GEOTECHNOLOGIEN Germany), COSMOS I (France) and COSMOS II (France) on monitoring strategies, the EU FP6 projects GRASP and CO2ReMoVe allowed for a European wide capacity building. The German Projects CORTIS provided the CO2 supply whereas the project CORDRILL enabled the project to drill three wells; both were funded by industrial partners and the German Ministry of Economy and Technology (BMWi – COORETEC Initiative). The Project CHEMKIN (development of new sensors – financed by GEOTECHNOLOGIEN programme of the German Ministry of Education and Research (BMBF) supported the development of a new gas membrane sensor. CO2SINK triggered national and international activities, allowed for a better connection and interaction of European Scientists and Companies and supported the European legislation process. A broad outreach project was a necessity to gain public acceptance for the project and is used as a lighthouse project in Europe, for Science, R&D activities, public acceptance and networking.

Food-grade CO2 was injected into Upper Triassic sandstones (Stuttgart Formation) of a double-anticline structure. The Stuttgart Formation represents a fluvial environment composed of sandstone channels and silty to muddy deposits. Undisturbed, initial reservoir conditions are ~ 35 °C and 62 bar at depth of injection at 650 m below surface. The initial reservoir fluid was highly saline with about 235 g/l total dissolved solids primarily composed of sodium chloride with notable amounts of calcium chloride. The initial pH value was 6.6. Hydraulic tests revealed a permeability between 50 and 100 mDarcy for the sand channels of the storage formation.

The anticline forms a classical multibarrier system: The first caprock is a playa type mudstone of the Weser and Arnstadt formations directly overlying the Stuttgart formation. Laboratory tests revealed permeabilities in a µDarcy-range. The second main caprock is a tertiary clay, the so-called Rupelton. Maximum injection pressure reservoir integrity tests resulted in values of about 120 bar for the first cap rock. Due to safety standards, the pressure threshold was set to 82 bar until more experience on the reservoir behaviour was available. The sealing property of the secondary cap rock was well known from natural gas storage operations at the testing site and was the basis for the permission by the mining authority to operate the CO2 storage.

th CO2 injection started on 30 June 2008 and by end March 2010 - after 21 months a total of 33,000 tons of CO2 has been injected. This is slightly more than half of the original target of up to 60,000 tons. The injection was operated successfully and safely. The mean injection rates were around 67 tons of CO2 per day when in service. The monitored wellhead pressure was almost constant at 63 to 64 bar and the calculated flow pressure at reservoir depth was 77 to 78 bar.

CO2SINK Summary Technical Report

The effects of CO2 injection were observed with surface and downhole measurements. A long-term field monitoring of natural diffusive CO2 soil gas emissions, gas fluxes along faults and fissures as well as CO2 dissolved in regional ground waters at shallow levels with different commercially available geochemical sensors has been conducted over the whole life time of the project. To date no evidence of any gas seepage or leakage from the CO2 storage reservoir has been detected.

Spreading of the CO2 plume in the reservoir was monitored by a broad range of geophysical and geochemical techniques. Three wells (one injection and two observation wells) were equipped with “smart casings” containing electrodes for Electrical Resistivity Tomography (ERT) behind casing, facing the rocks and a Distributed Temperature Sensing (DTS). Changes in borehole temperatures were continuously monitored since the start of CO2 injection with permanently installed downhole fiber-optic sensor cables using DTS. In-situ saturation changes were monitored by repeated pulsed neutron-gamma wire line logging, as well as a heat pulse technique using an electrical heater cable installed parallel to the DTS sensor cables.

A newly developed Gas Membrane Sensor (GMS), enabling a continuous monitoring of the gas composition in the observation wells (OWs), observed the arrival of CO2 and gaseous tracers at the observation wells. CO2 arrival was detected at the first OW (50 m away from injection) after about 500 tons of injected CO2 on July 15, 2008. Arrival at the second OW (at 112 m distance) was recorded after about 11,000 tons of injected CO2 on March 21, 2009. Pressure and temperature logs revealed a supercritical state of the CO2 in all three wells at depth of the storage formation after arrival of CO2.

The propagation of the carbon dioxide plume in the underground was observed with seismic and geoelectrical measurements as well. Repeated seismic measurements indicated changes in the elastic properties of rocks due to the influence of carbon dioxide. Repeated geoelectrical observations imaged changes in the electrical conductivity caused by the injection of carbon dioxide. Both methods allowed the performance of long-term observations of the carbon dioxide during injection phase.

The seismic monitoring included crosshole seismic experiments, Vertical Seismic Profiling (VSP) and Moving Source Profiling (MSP), star seismic experiments and 4-D seismics. Seismic measurements for structural interpretation and baseline characterization were carried out in 2005 and 2007 using different methods in order to cover the near-injection to the regional scale. Time-lapse crosshole seismics showed no considerable changes in seismic velocity between the two observation wells within the first two repeats after injection of 660 tons and 1,700 tons of CO2, respectively. However, after injection of 18,000 tons CO2 all time-lapse surveys showed a clearly observable signature of the CO2 propagating in the Stuttgart formation. In summer and fall 2009 surface-downhole observations (MSP, 2D and 3D surface surveys) were repeated providing a multi-scale view on the time-lapse effect of more than 20,000 tons of injected CO2. The crosshole tomography revealed a significant reduction of propagation velocity within the injection horizon, while MSP and surface reflection surveys revealed both an increased reflectivity from the depth range of the top of the

CO2SINK Summary Technical Report Stuttgart Formation around the injection location which can be attributed to a signature of the CO2 migration in the reservoir.

Electrical Resistivity Tomography (ERT) proved to be sensitive to saturation changes caused by propagation of supercritical CO2 within the reservoir. Forward modeling indicated an increase of electrical resistivity of about 200 % caused by CO2 injection, which is verifiable in the field data from geoelectrical cross-hole and surface-downhole measurements, respectively. Time-lapse results from the permanently installed Vertical Electrical Resistivity Array (VERA) show significant resistivity changes in the plane between injection and first observation well. ERT measurements indicated an anisotopic flow of CO2 coinciding with the “on-time” arrival of CO2 at OW 1 and the late arrival at the second observation well.

In order to enhance the understanding of geological CO2 storage, monitoring data are used to verify the numerical models. Data used for history matching were injection pressure and CO2 arrival times at both OWs. CO2 arrival at the first OW was in good agreement with the predictions made by different modeling approaches. However, the arrival of CO2 at the second OW was notably later than predicted. The reason for this discrepancy between predicted and observed arrival times is under further investigation. The impact of spatial heterogeneity of the permeability within a sandstone channel of the Stuttgart Formation was studied with 2D vertical and horizontal models in a stochastic Monte-Carlo framework. The study showed that this heterogeneity cannot be the only reason for the late CO2 arrival. The heterogeneity of the geological units (e.g. channel geometry) is possibly the decisive factor and the small scale heterogeneity emphasizes this effect.

Monitoring of biological and geochemical processes in the reservoir aimed at assessing their impact on the long-term safety of geological CO2 storage, e.g. on inducing mineral dissolution and/or precipitation and corrosion. Downhole sampling revealed quantitative and qualitative changes after CO2 arrival in the first OW. An increase of the iron content and shifts in the microbial community from chemoorganotrophic to chemolithotrophic populations were observed, as evidenced by the temporarily outcompetition of sulphate- reducing bacteria by methanogenic archaea. Additionally, an enhanced activity of the microbial population after five months CO2 storage indicated that the microbial community was able to adapt to the changes in environmental conditions (e.g. pH decrease, pressure increase).

Long-term CO2-fluid exchange experiments under simulated reservoir conditions (p,T) were performed to study petrophysical, mineralogical, geochemical and microbiological 2+ 2+ changes induced by CO2 exposure in detail. Ca and Mg concentrations in the sampled fluid exceeded Ketzin reservoir fluid concentrations by 15 % and 8 %, respectively, which may reflect mineral dissolution in response to CO2 exposure. This is consistent with the dissolution of Ca-rich feldspar observed. The majority of the microbes were able to adapt to the changed conditions, since only minor changes in the composition of the microbial community were observed. Extraction experiments with CO2 at supercritical conditions showed a preferred extraction of formate, acetate, chloride and nitrate.

CO2SINK Summary Technical Report Considerable efforts were made to inform the wider community about the project and CCS in general. There have been many visits to the site where a small visitor’s centre and a tour of the site have been available. Press coverage following several open events has been quite extensive, ranging from local, national, European and international journalists. More than 1000 new articles were printed in newspapers and weekly journals, all big TV stations of Germany and from several European countries (e.g. France, Denmark) and beyond (e.g. Japan) put CO2SINK into the media. The visit of Andris Piebalgs, the European Commissioner for Energy, was one of the highlights of public communication. The communication activity where focused on the local acceptance by educating, informing and involving the local public. Little if any adverse reaction has been received from the local inhabitants.

CO2SINK Summary Technical Report Chapter 2

INTRODUCTION

Hilke Würdemann1, Günter Borm 1,2,Frank R. Schilling1,2

1Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ), Telegrafenberg, 14473 Potsdam, Germany 2 Universität Karlsruhe, KIT, Engler-Bunte-Ring 14, 76131 Karlsruhe, Germany

CO2SINK Summary Technical Report The development of strategies for sustainable and secure technologies to reduce substantially emission of greenhouse gases to the atmosphere is one of the major challenges of the next decades. Geological CO2 storage is a promising technology to effectively reduce anthropogenic greenhouse gas emissions to the atmosphere (IPCC, 2005). Different storage technologies are under development and are being tested in large scale experiments like the Norwegian Sleipner and the Canadian Weyburn projects. Improving monitoring and verification, and the efficiency of storage operations in saline aquifers is the focus of the on-shore pilot projects in USA (Frio), Australia (Otway), Japan (Nagaoka), Algeria (In Salah) and Germany (CO2SINK).

The CO2SINK project is the first pilot project for on-shore geological CO2 storage into a saline aquifer in Europe. Within the frame of the project, an “in-situ laboratory” for CO2 storage was developed to fill the gap between numerous conceptual engineering and scientific studies on geological storage and a full-fledged on-shore storagedemonstration. The fate of the injected CO2 as well as the performance of the reservoir has to be monitored to ensure safe and responsible long-term storage. This requires the combination of robust monitoring technologies from several disciplines (geophysics, geochemistry, geomicrobiology) applied at the surface and downhole.

Geophysical monitoring tools were used to verify the detection limit, to observe CO2 migration in the underground and to provide a tomography of the CO2 evolution in the reservoir (Borm and Förster, 2005, Schilling et al. 2009, Würdemann et al., 2010). Surface and downhole measurements were applied to test and optimise the resolution of different methods and to follow the developing CO2-plume (Giese et al. 2009, Kießling et al., 2010 and Lueth et al. 2010). Time-lapse seismic monitoring of migration in the deep underground proved to be a sensitive tool using crosshole seismic data as demonstrated for the Frio and the Nagaoka project (Spetzler et al., 2008, Onishi et al., 2009). 4D seismic surveys were successful to monitor CO2 evolution in large scale projects like Sleipner and Weyburn (Arts et al., 2004, Li, 2003). Ramirez et al. (2003) and Christensen et al. (2006) have evaluated the feasibility of monitoring CO2 migration due to electrical resistivity changes in field applications. Joint evaluation of geophysical and geochemical monitoring results will help reducing non-unique images of tomographic inversions.

Within the frame of the CO2SINK project the following issues were of particular interest:

 migration of CO2 within the reservoir  rate at which CO2 dissolves in brine or reacts with minerals  quality of the seals, including leakage to the shallow subsurface  upward migration of CO2 along artificial pathways (casing and cementation of injection/observation boreholes)

The project aimed to enhance the understanding of the underlying processes involved in geological CO2 storage in saline aquifers, to advance and get practical experience with monitoring and verification of CO2 storage, to prove the predictability of different coupled models, and to test technologies required for a safe storage operation for CO2. The main focus of the project was the development, testing and benchmarking of monitoring techniques. The best practice in performing of wells within geological storage projects is 9

CO2SINK Summary Technical Report enhanced by new technologies for CO2 monitoring via these wellbores making extensive use of a smart casing concept for the first time. An intensive well bore logging program was applied to control well integrity. The geophysical monitoring of CO2 migration is complemented by geochemical and microbial investigations of rock cores and water samples taken during drilling, testing and storage operation to study the influence of microorganisms on precipitation, mineral alteration and corrosion processes. Gaseous and water soluble tracers were applied to better follow the migration of CO2 within the reservoir. A gas membrane sensor (GMS) was used to monitor changes in the reservoir gas composition in the observation wells in near real time, as a low cost high resolution alternative to U-tube sampling performed in Frio and Otway.

A critical issue of the project was the need of sufficient amounts of CO2 for realistic testing of the geological storage. Our Expression of Interest to the EU dated 02 May 2002 for submission of the FP6 Integrated Project CO2SINK proposed that the project would demonstrate fuel production from bio-mass integrated with underground storage of CO2 into deep saline aquifers. The CO2 which the crop releases as it is burnt or gasified is balanced by an equal amount removed from the atmosphere as it grew. Therefore, if the CO2 produced during biomass conversion were captured in subsurface permeable strata, a negative CO2 balance were achieved, so that this bioenergy production site would act as an effective sink for CO2. Unfortunately, the planning and establishment of the biogas facility adjacent to the CO2SINK storage site took longer than expected, and – even worse - the capture of high-grade CO2 from biogas turned out to be unreasonably complex and costly. Therefore, the CO2 had to be purchased and transported from a remote hydrogen refinery plant. However, since the CO2 storage together with the biogas plant are now existing on the same site, the perspective is remaining that the original vision of a real SINK of CO2 at Ketzin eventually may become true in the future.

REFERENCES

Arts, R., Eiken, O., Chadwick, A., Zweigel, P., van der Meer, B., Kirby, G., (2004). Seismic monitoring at the Sleipner underground CO2 storage site (North Sea). Geological Society, London, Special Publications 233, 1, 181-191.

Borm, G. and Förster, A., 2005. Tiefe salzwasserführende Aquifere - eine Möglichkeit zur geologischen Speicherung von CO2. Energiewirtschaftliche Tagesfragen - Zeitschrift für Energiewirtschaft, Recht, Technik und Umwelt. 8, 15-20.

Christensen, N. B., Sherlock, D. and Dodds, K., (2006). Monitoring CO2 injection with cross-hole electrical resistivity tomography. Exploration Geophysics. 37, 44-49.

Giese, R., Henninges, J., Lüth, S., Mozorova, D., Schmidt-Hattenberger, C., Würdemann, H., Zimmer, M., Cosma, C., Juhlin, C., CO2SINK Group, 2009. Monitoring at the CO2SINK site: A concept integrating geophysics, geochemistry

CO2SINK Summary Technical Report and microbiology. Greenhouse Gas Technology Conference 9 (Washington, USA). Energy Procedia. 1, 2251−2259.

Kiessling, D., Schmidt-Hattenberger, C., Schuett, H., Schilling, F., Krueger, K., Schoebel, B., Danckwardt, E., Kummerow, J., and the CO2SINK Group, (2010). Geoelectrical methods for monitoring geological CO2 storage: First results from crosshole and surface-downhole measurements from the CO2SINK test site at Ketzin (Germany), Int. J. Greenhouse Gas Control, 4, pp. 816-826.

Li, G., (2003). 4D seismic monitoring of CO2 flood in a thin fractured carbonate reservoir (in Carbonate geophysics; an introduction). Leading Edge. 22, 7, 690-695.

Lüth, S. Bergmann, P., Giese, R., Götz, J., Ivanova, A., Juhlin, C., Cosma, C., 2010. Time-Lapse Seismic Surface and Down-Hole Measurements for Monitoring CO2 Storage in the CO2SINK Project (Ketzin, Germany). Submitted abstract to GHGT- 10, 10th International Conference on Greenhouse Gas Control Technologies.

Onishi, K., Ueyama, T., Matsuoka, T., Nobuoka, D., Saito, H., Azuma, H., Xue, Z., (2009). Application of crosswell seismic tomography using difference analysis with data normalization to monitor CO2 flooding in an aquifer. International Journal of Greenhouse Gas Control. 3, 3, 311-321.

Schilling, F., Borm, G., Würdemann, H., Möller, F., Kühn, M., 2009. Status Report on the First European on-shore CO2 Storage Site at Ketzin (Germany). Energy Procedia. 1,

Spetzler, J., Xue, Z., Saito, H., Nishizawa, O., (2008). Case story: time-lapse seismic crosswell monitoring of CO2 injected in an onshore sandstone aquifer. Geophysical Journal International. 172, 1, 214-225.

Ramirez, A. L., Newmark, R. L., Daily, W. D., (2003). Monitoring Carbon Dioxide Floods Using Electrical Resistance Tomography (ERT): Sensitivity Studies. Journal of Environmental and Engineering Geophysics, Volume 8, Issue 3, pp.187–208.

Würdemann, H., Moeller, F., Kühn, M., Heidug, W., Christensen, N.P., Borm, G., Schilling, F.R. and the CO2SINK Group (2010). "CO2SINK – From Site Characterisation and Risk Assessment to Monitoring and Verification: One Year of Operational Experience with the Field Laboratory for CO2 Storage at Ketzin, Germany." International Journal of Greenhouse Gas Control. doi: 10.1016/j.ijggc.2010.08.010.

CO2SINK Summary Technical Report Chapter 2

GEOLOGICAL AND HYDROGEOLOGICAL SITE CHARACTERIZATION

Andrea Förster1, Ben Norden1, Peter Frykman2, Niels Springer2, Bernd Wiese1

1GFZ German Research Centre for Geosciences 2 GEUS The Geological Survey of Denmark and Greenland

CO2SINK Summary Technical Report Work on baseline geology was required for baseline seismic surveys, for the planning of the drilling, and for the design of CO2 injection and monitoring. In the pre-drilling phase of the project (Förster et al., 2006) scientific activities exceeded a reconnaissance available from former exploration campaigns in the area by considering the geology of the entire Ketzin anticline down to the Stuttgart Formation (the CO2 injection formation), and its underlying seal. Especially the data from CO2SINK baseline seismic survey (Juhlin et al., 2006), formed an integrative part in the construction of a first geological site model, including the major geological structure of the Ketzin anticline and major fault zones intersecting large parts of the geological sequence. The aim of the first period of CO2SINK also included a study of the geology of the shallow groundwater system in the larger Ketzin area as a basis for risk assessment (Norden, 2007). Groundwater modeling was used to characterize the subsurface flow conditions and served for the detection of possible leakage pathways to optimize the surface CO2 monitoring program. In addition, the detailed near-surface hydrogeological model built the basis for the interpretation of a baseline seismic tomography study of the shallow underground, which can be repeated for the detection of changes caused by CO2 accumulations (Yordkayhun et al., 2009). During and after the drilling phase in 2007 (Prevedel et al., 2008) a wealth of new data became available from well-logs and core analyses. These data formed the basis for in-depth studies on the geology of the anticline and, in particular, of the caprock and reservoir. In this phase of the project, the aim of study focused on the lithology and petrophysical properties of the reservoir and parts of the caprock as well as the hydraulic properties of the reservoir verified by hydraulic borehole testing. Data from the repeats of seismic surface surveys in the post-drilling phase of the project formed another pool of information that needed to be integrated in the geological site model, as they provided valuable information for the verification of a stochastic reservoir model.

Lithology and mineralogy of the reservoir formation

Various types of geophysical well logs obtained in the three CO2SINK boreholes, together with a macroscopic core inspection, were used to characterize the lithology of the Stuttgart Formation (Fig. 1). The CO2 reservoir is lithologically heterogeneous and made up by fluvial sandstones and siltstones interbedded with mudstones showing remarkable differences in porosity. The thickest sandstone units are associated with channel sandstone, whose thickness varies over short lateral distances.

The in-depth petrographic, mineralogical, mineral-chemical, and whole-rock geochemical analysis were performed focusing on the sandstone intervals, which display the best reservoir properties for CO2 injection. The dominantly fine-grained and well to moderately-well sorted, immature sandstones classify as feldspathic litharenites and lithic arkoses (Förster et al., 2010). Quartz, plagioclase, and K-feldspar predominate mineralogically. Muscovite plus illite and mixed-layer minerals are omnipresent. Quartz, feldspar, as well as meta-sedimentary and volcanic rock fragments comprise the most abundant detrital components, which often are rimmed by thin, early diagenetic coatings of ferric oxides, and locally of clay minerals. Feldspar grains may be unaltered and optically clear, partially to completely dissolved, partially altered to sheet silicates (mainly illite), or albitized. Analcime and anhydrite constitute the most widespread, often spatially associated pore-filling cement minerals, Authigenic dolomite, barite, and 13

CO2SINK Summary Technical Report coelestine is minor. The percentage of cements ranges in total from about 5 vol.% to 32 vol.%. Except of samples intensely cemented by anhydrite and analcime, total porosities of the sandstones range from 13% to 26%. The fraction of intergranular porosity varies between 12% and 21%. About 1–5% porosity has been generated by dissolution of detrital plagioclase, K-feldspar, and volcanic rock fragments.

Fig. 1: Generalized lithology of the Stuttgart Formation from core and well-log interpretation. Porosity adapted from Norden et al. (2010). CS is channel sandstone section; OB is overbank section (from Förster et al., 2010).

The total porosity from textural analysis correlates with the results of a hot-shot routine, laboratory core analysis performed to evaluate in particular the sandstone/siltstone reservoir units. In turn these values are supported by well-log analysis. The high porosities of the sandstones correspond with an average brine permeability of 500 mD 14

CO2SINK Summary Technical Report (upper channel sandstone) and 1300 mD (lower channel sandstone) (Norden et al., 2010). In addition to porosity and permeability, the suitability of a consolidated sedimentary reservoir for CO2 storage is critically dependent on the modal amount of minerals that may dissolve during interaction with the CO2 and the formation fluid.

The immature (shaly) sandstones of the Stuttgart Formation at Ketzin are generally rich in potentially reactive minerals, including plagioclase, K-feldspar, biotite, muscovite, chlorite, analcime, and anhydrite, and, thus, may be a reservoir prone to extensive chemical reactions (Förster et al., 2010). Grain coatings, although usually minor in the sandstones from this study, are likely to be reactive due to their mineralogy and accessibility to the pore fluid. The presence of variable amounts of hematite and clay- mineral grain coatings constitutes a limiting factor for the overall accessibility of the reactive minerals to pore fluids. Illite and mixed-layer minerals accumulated in pore spaces and deposited on grain surfaces effectively reduce the connections between pore fluids and reactive minerals.

Petrophysical and hydraulic properties of reservoir

The good core control of the first two CO2SINK wells was used to guide the petrophysical well-log interpretation and allow a porosity and permeability prediction by analogy for the third well (Norden et al., 2010).

Pumping tests were performed in three wells, with simultaneous pressure monitoring in each well to determine the hydraulic properties on reservoir scale. In general, core permeabilities are much higher than the observed pumping test values. This observation is unusual and appears to be a unique feature of the field site. The difference is most likely related to a low-permeability feature present in the system.

Petrophysical properties of caprock

Porosity and grain density was measured by the Helium injection technique, and gas permeability was measured relative to dry N2 gas at a confining stress of 2.8 MPa (400 psi) on the samples. The samples have porosity values close to 12 % and slightly elevated grain densities due to the cementation with dolomite and anhydrite. Gas permeability for un-fractured vertical plugs was determined between 0.002 and 0.018 mD. The kv/kh ratio was calculated in the range of 0.33 (CO2 Ktzi 200/2007) to 0.63 (CO2 Ktzi 201/2007) and for all plugs with 0.56. The determined kv/kh ratio is very uncertain due to the scarce number of unbiased measurements. There is no significant correlation between gas permeability and porosity; the distribution pattern seems to be controlled mainly by fractures and the dolomite / anhydrite cementation (Springer, 2008). Five samples were selected for a detailed characterization of the pore system and determination of capillary properties to estimate the caprock seal capacity.

Geological reservoir model (FLUVSIM/PETREL)

The Ketzin area imposes complexity in the geometry and spatial distribution of reservoir properties to be considered in the prediction of CO2 migration. This complexity was shown in a first 3-D geological reservoir model (Förster et al., 2006). After drilling of all 3 15

CO2SINK Summary Technical Report Ketzin wells, the detailed interpretations and information from well logs and core material have been incorporated for an updated model (version 3.0; Frykman et al., 2010), which has been used as a base-case for estimating break-through timing at the observation wells, and the model has also been used for guiding the operations and released for other studies. Due to problems in matching the arrival time for the second observation well (CO2 Ktzi 202/2007), the base-case model was locally modified in a version called iso4. This model proved able to delay the arrival, but was abandoned since the model procedure was not efficient for history matching purposes.

REFERENCES

Förster, A., Norden, B., Zinck-Jørgensen, K., Frykman, P., Kulenkampff, J., Spangenberg, E., Erzinger, J., Zimmer, M., Kopp, J., Borm, G., Juhlin, C., Cosma, C.-G., Hurter, S. (2006) Baseline characterization of the CO2SINK geological storage site at Ketzin, Germany, Environmental Geosciences, 13, pp145-161

Förster, A., Schöner, R., Förster, H.-J., Norden, B., Blaschke, A.-W., Luckert, J., Beutler, G., Gaupp, R., Rhede, D. (2010). Reservoir characterization of a CO2 storgae aquifer: The Upper Triassic Stuttgart Formation in the Northeast German Basin, Marine and Petroleum Geology, doi:10.1016/j.marpetgeo.2010.07.010.

Frykman, P., Zinck-Jørgensen, K., Norden, B., Johannessen, P., Bech, N., Nielsen, C.M. (2010) Geological Modelling of the Ketzin Stuttgart Formation Storage Aquifer, Post-Drilling Version 3.0, Internal CO2SINK report D6.1-11, 25pp

Juhlin, C., Giese, R., Zinck-Jørgensen, K., Cosma, C., Kazemeini, H., Juhojuntti, N., Lüth, S., Norden, B., Förster, A. (2007) 3D baseline seismics at Ketzin, Germany: the CO2SINK project, Geophysics, 72, 5, ppB121-B132

Norden, B. (2007) Report on Shallow Groundwater Model for Ketzin, Internal CO2SINK report D2.1-7.

Norden, B., Förster, A., Vu-Hoang, D., Marcelis, F., Springer, N., Le Nir, I. (2010). Lithological and petrophysical core-log interpretation in CO2SINK, the European CO2 onshore research storage and verification project, SPE Reservoir Evaluation & Engineering, 13, 2, pp179-192.

Prevedel, B., Wohlgemuth, L., Henninges, J., Krüger, K., Norden, B., Förster, A., CO2SINK Drilling Group (2008) The CO2SINK boreholes for geological storage testing, Scientific Drilling, 6, pp32-37.

Springer, N. (2008) Preliminary Caprock Analysis Data from New Wells, Internal CO2SINK report D3.1—11, 21pp

CO2SINK Summary Technical Report Wigand, M., Carey, J.W., Schütt, H., Spangenberg, E., Erzinger, J. (2008) Geochemical effects of CO2 sequestration in sandstones under simulated in-situ conditions of deep saline aquifers, Appl. Geochemistry, 33, pp1-40

Wiese, B., Böhner, J., Enachescu, C., Würdemann, H., Zimmermann, G. (2010) Hydraulic characterization of the Stuttgart Formation at the Ketzin test site, International Journal of Greenhouse Gas Control, in press

Yordkayhun, S., Juhlin, C., Norden, B. (2009) 3D seismic reflection surveying at the CO2SINK project site, Ketzin, Germany: A study fro extracting shallow subsurface information, Near Surface Geophysics, 7, pp75-91

CO2SINK Summary Technical Report Chapter 3

LABEXPERIMENTS: ROCK FLUID INTERACTION

A. Liebscher1, J. Kulenkampff1, J. Kummerow1, C. Otto2, K. Rempel1, A.-K. Scherf1, E. 1 1 1 1 1 3 Spangenberg , A. Vieth , M. Wandrey , H. Würdemann , K. Zemke , M. Zettlitzer

1German Research Centre for Geosciences, Telegrafenberg, D-14473 Potsdam 2Shell International, Kesslerpark 1, NL-2288GS Rijswijk 3 RWE Dea AG, Wietze Laboratory, Industriestraße 2, D-29323 Wietze, Germany

CO2SINK Summary Technical Report An extensive experimental program including a wide variety of different experimental techniques has been performed within the frame of the CO2SINK project. The aim of these experiments was to study the CO2 induced inorganic and organic geochemical interactions between CO2, formation fluid, reservoir rock, cap rock and microbial community as well as the CO2 induced changes in petrophysical properties of the reservoir system. Initial physico-chemical equilibrium between saline formation fluid, reservoir rock, cap rock and microbial community can be assumed for these deep saline aquifer systems. Injection of CO2 into these saline aquifer systems will drive the systems out of equilibrium and trigger inorganic and organic chemical interactions between injected CO2, saline formation fluid, reservoir rock, cap rock and microbial community. These interactions may lead to dissolution and/or precipitation of specific minerals and to mobilization of certain inorganic and/or organic compounds. In addition, these interactions may change the physical properties of the reservoir system like permeability, porosity, electrical conductivity and sound wave velocity. Precise knowledge of the CO2 induced interactions between injected CO2, saline formation fluid, reservoir rock, cap rock and microbial community and of the resulting changes in chemical and physical properties of the reservoir system is therefore a prerequisite for any secure operation of a storage site. However, core samples are typically only taken before start of injection and any information on CO2 induced interactions at real storage sites are based on gas or fluid samples recovered from observation wells, which may be influenced by the drilling procedure. To overcome this problem, an extensive experimental program at simulated reservoir pressure and temperature conditions has been performed within the frame of the CO2SINK project. This program includes i) petrophysical laboratory experiments to determine fundamental petrophysical properties of the Ketzin reservoir rocks and to monitor CO2 induced changes of these properties by structural sensitive petrophysical methods, ii) long-term exposure experiments to study CO2 induced mineralogical, petrophysical, inorganic and organic geochemical, and microbiological interactions, iii) extraction experiments with CO2 at supercritical conditions to study the mobilization of organic compounds, and iv) experiments on fluid- fluid interactions to study the role of CO2 as a separate solvent phase.

Flow experiments showed a notable decrease in P-wave velocities, almost constant S- wave velocities and a significant increase in electrical resistivity due to CO2 injection. Residual water saturation in sandstone samples from the Ketzin reservoir range from 36 to 55 %. Long term CO2 exposure experiments on freshly drilled, pristine Ketzin reservoir core samples were accomplished for 24 months using synthetic brine(Wandrey et al., 2010a). Slight changes in the composition of the indigenous microbial community dominated by chemoorganotrophic bacteria and hydrogen oxidizing bacteria as well as changes in porosities were observed with time (Wandrey et al., 2010c).. During the experiments porosity first increased due to mineral dissolution but then tended to decrease due to mineral precipitation. Mineralogical investigations indicate preferred dissolution of anhydrite, Ca- and K-bearing feldspar and precipitation of albite. These mineralogical changes are consistent with changes in fluid composition during the + 2+ 2+ 2- course of the experiments that indicate notably increased K , Ca , Mg , and SO4 concentrations. K+, Ca2+, Mg2+ concentrations exceeded the reservoir brine composition significantly and can be attributed to the CO2 exposure (Wandrey et al., 2010b). Extraction experiments with CO2 at supercritical conditions reveal mainly the extraction of formate, acetate, chloride and nitrate. 19

CO2SINK Summary Technical Report

Extraction experiments at supercritical conditions using CO2 as the solvent were done under reservoir conditions on rock samples (clay, silt and sandstones) from the injection well and two observation wells to assess the potential to release organic compounds as well as inorganic anions. The analyses of fluid extracts reveal mainly the extraction of formate, acetate, chloride and nitrate. The compound specific concentrations of e.g. formate and acetate are lower than 0.2 mg/kg rock for all samples, but differ in orders of magnitudes between certain samples. Summed concentrations of the organic solvent soluble compounds from the CO2 extraction experiments amount to up to 8 % of the total organic carbon (TOC) from the respective rock sample.

Within the CO2SINK project the extensive experimental research program has created a comprehensive and rather unique database on the CO2 induced inorganic and organic geochemical interactions between CO2, formation fluid, reservoir rock, and microbial community as well as the CO2 induced changes in petrophysical properties of the reservoir system. The interdisciplinary long-term experiments feature run-durations generally not applicable to normal research experiments and thus more reliably allow assess questions of the mid- to long-term integrity of the Ketzin reservoir. But although specifically addressing the conditions at the Ketzin site the results have also implications for CO2 induced processes within CO2 storage reservoirs in general. Within this context, long-term experimental studies on the role of microbiological processes on mineral alteration, e.g., carbonate precipitation, are needed. Re-location of organic and inorganic matter may create micro-habitats, in which diverse metabolic processes could locally reduce the permeability by formation of extracellular polymeric substances and mineral precipitates. Within faults, these processes would support and enhance self- healing effects. Additionally, microbes may convert and fix CO2 and other gases transported within the CO2 into minerals. For assessing questions of the mid- to long- term integrity of CO2 storage systems, data on the inorganic and organic geochemical interactions between CO2 and cap rock and their effects on petrophysical properties of the cap rocks are urgently needed.

Wandrey, M., Morozova, D., Zettlitzer, M., Würdemann, H. (2010a). "Contamination assessment of rock core and brine samples at the Ketzin storage site using fluorescent tracers." International Journal of Greenhouse Gas Control. doi:10.1016/j.ijggc.2010.05.012.

Wandrey, M., Fischer, S., Zemke, K., Liebscher, A., Scherf, A., Vieth, A., Zettlitzer, M. and Würdemann, H. (2010b). Monitoring petrophysical, mineralogical, geochemical and microbiological effects of CO2 exposure - Results of long-term experiments under in situ conditions. Energy Procedia 10, in press.

Wandrey, M., Pellizari, L., Zettlitzer, M. and Würdemann, H. (2010c). Microbial community and inorganic fluid analysis during CO2 storage within the frame of CO2SINK – Long-term experiments under in situ conditions. Energy Procedia 10, in press.

CO2SINK Summary Technical Report

Zemke, K., Liebscher, A., Wandrey, M. and the CO2SINK Group (2010). Petrophysical analysis to investigate the effects of carbon dioxide storage in a subsurface saline aquifer at Ketzin, Germany (CO2SINK). International Journal of Greenhouse Gas Control, 4, pp990-999.

CO2SINK Summary Technical Report Chapter 4

DYNAMIC FLOW MODELING

Peter Frykman1, Holger Class2, Thomas Kempka3

1Geological Survey of Denmark and Greenland – GEUS, Øster Voldgade 10, 1350 Copenhagen, Denmark, [email protected] 2University of Stuttgart, Hydromechanics and Modeling of Hydrosystems, Pfaffenwaldring 61, 70569 Stuttgart, Germany, [email protected] 3Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ), Telegrafenberg, 14473 Potsdam, Germany, [email protected]

CO2SINK Summary Technical Report

Dynamic flow modeling of the CO2 injection at the Ketzin storage site was conducted in different phases depending on increasing data availability with time. First modeling activities relevant for issues of permitting and operational site design were conducted based on pre-existing information extracted from data of wells drilled nearby the planned injection site and natural analogues. For that purpose, simplified radially symmetric models were applied until a first version of the static geological 3D model was available. Using this preliminary 3D model the first predictions of CO2 arrival time at the observation wells were made with different simulators resulting in a reasonable agreement with the later real arrival time depending on the choice of injection rate and boundary conditions. Subsequent to the drilling of the three wells at the Ketzin site, well data was available for the further development of the static geological model followed by simulations to predict the CO2 arrival times at both observation wells. The numerical simulation codes ECLIPSE 100 (black-oil simulator), ECLIPSE 300 (compositional CO2STORE) and MUFTE_UG (2p2ni_CO2) were used for a history match applying the real injection rates. Regarding the real injection regime the computational results show a relatively good agreement with the data effectively measured at the first observation well. The calculated arrival times exceeded the real ones by 8% (ECLIPSE100), 9% (ECLIPSE300) and 18% (MUFTE-UG). However, the real arrival time at the second observation well exceeded the calculated ones by a factor of 3. Uncertainties related to the geological model and parameter choices made by the modelers as well as in the conceptual approaches applied in the different simulators have to be discussed in detail to reflect these deviations. Further studies, such as the implementation of heterogeneities on a structural geological scale (low permeability barrier between the injection and second observation well) revealed one reasonable explanation for the late arrival. However, further research on the sensitivity of input parameters is required involving integration of recent geophysical monitoring data to improve the understanding of geological heterogeneities at the Ketzin site and their impact on the CO2 plume distribution.

Eventually, the modeling work described above requires a discussion that summarizes the achievements and findings but provides also an overview of issues and open questions that, due to limited resources, could not be addressed in the desired detail during the CO2SINK project.

Considering the fundamental question “Why do we do all this modeling?” might help to judge the relevance of the achievements and insights gained in the project. Basically, we consider modeling to be a decisive tool for predicting the long-term fate of a storage reservoir and the injected CO2, provided that data availability is not limiting (which unfortunately, to some degree, is always the case). Besides supplementary contributions to decision making during the operation of a storage site or for the layout of a monitoring concept, mainly the evaluation of risks associated with a storage project are of particular interest. The selection of risk-related scenarios is, of course, site- specific. However, leakage through faults or wells and overpressures in the reservoir or in particular in the caprock region are probably the most obvious risks. The actual assessment of risks with a focus on the site in Ketzin was, due to a lack of resources, not a major topic of this work and was therefore restricted to a limited study of fault transmission rates.

CO2SINK Summary Technical Report However, Kopp et al. (2010) developed and proposed a general methodology that allows the quantification of risks related to storage scenarios. This includes the definition of certain failure scenarios, the determination of the likelihood of a failure event to occur from a multitude of numerical simulations, and the quantification of the average damage that the suite of leakage events produces. While, for the sake of an intuitively comprehensive demonstration, Kopp et al. (2010) applied this methodology to a simplified homogeneous radially-symmetric reservoir, it can be easily adapted to any specific storage site, thus also for Ketzin.

Another fundamental question is “How confident are we in the results of our modeling?” The experience during CO2SINK showed clearly that the degree of confidence into predictive simulations will crucially depend on the quality of history matches with reliable monitoring data. As was pointed out in this report, the geological model is the key to this. However, for a final judgment on confidence and reliability, one should always see the spatial and temporal scales of interest in relation to the required accuracy of the geological model. The larger the spatial scale the more decreasing is the influence of small-scale geological features that are not covered by the geological model, and for larger time scales also other processes in addition to the pure two-phase flow displacement of brine by the injected CO2 become relevant. For example, convective mixing was not investigated in detail within the CO2SINK project, although the applied simulators are capable of describing it in principal. However, the considered time scales were far below the expected onset time of fingering (Riaz et al. 2006) and, in addition to that, the small bed thickness of the storage layers in Ketzin will prevent convection ever happening on a significant scale.

In the following, we list what we consider the most important achievements and new insights gained during the CO2SINK project:

 A geological model of the Ketzin site has been continuously developed, adapted, and updated with available field data collected by other groups during the project.

 Modeling has provided a good insight into the dominant processes and improved the understanding of their interaction on different spatial and temporal scales.

 Modeling has been able to support operational decisions. Exchange of data with practical engineers (model parameter from field data in, simulation results out) has proven to be a mutual benefit both for modelers and for the engineers who are in charge of operational decisions.

 The history match of the arrival time at CO2 Ktzi 200/2007 was successful and all modelers and models were in satisfactory agreement on that.

 For Ketzin, one could conclude that a history match of pressure data is associated with quite some uncertainty since the boundary conditions influence the simulated pressure results. However, the extent and shape of the CO2 plume is expected to be rather independent of the boundary conditions because the injection itself is modeled with a flux (Neumann) condition in the injection well. Thus, extent and shape of the plume are mainly determined by the injected mass, 24

CO2SINK Summary Technical Report and the required pressure gradient will establish accordingly. The absolute values of simulated pressures, which are relevant for matching the monitoring data are still influenced by the boundary. This can have some limited effect on the development of the CO2 plume if the fluid properties (mainly density and viscosity) are very sensitive to pressure in the given range.

 The experience from this project (and supported by the benchmark study by Class et al. (2009) showed clearly that both the commercial and the open source simulators applied here are capable of reproducing the fundamental processes in excellent agreement. Consequently, the reliability of predictions is expected to depend not on the choice of the simulator (as long as it is capable of reproducing the dominant processes!), but rather on the reliability of the geological model. Monitoring data are important that would help improving the geological model by identifying distinct geological features like the hypothetical barrier mentioned in this report.

There are also a couple of questions that arose or have been left open since available resources were limited. Addressing those issues leads to some perspectives for future work.

A discussion point is the fact that the geological model did not allow a reliable history match at CO2 Ktzi 202/2007. The question is how influential this mismatch is on the general reliability of predictions to be made with this model. One must be aware that the underestimated arrival time at CO2 Ktzi 202/2007 is a mismatch of a very detailed monitoring data (point observation: now the CO2 is there) in a very heterogeneous reservoir. This mismatch has not a dramatic influence on our view of the overall behavior on a larger time-scale. For example, the barrier study shows that once the plume has passed the barrier, the overall shape of the further developing plume looks rather similar while the barrier makes a decisive difference for the arrival time at CO2 Ktzi 202/2007. The investigations here were based on one selected realization of the geological model out of a suite of equally probable scenarios. Therefore, we have not explored the range of uncertainty and cannot quantify the range of variation of the CO2 plume propagation. Better knowledge of probable geological scenarios and identification of geological features would require snap-shots of the plume evolution at different times. However, simplified modeling showed that the late arrival at CO2 Ktzi 202/2007 with high probability cannot be explained by small-scale (REV-scale) heterogeneity but rather requires distinct geological features.

A further issue would also deserve some in-depth investigations. Different modelers can make (and indeed did make) different choices for the size of the domains, the gridding, the boundary conditions, or for various input parameters, for example, concerning the endpoints of the applied relative permeability relationships. However, the influence of these partly different choices was not investigated in a systematic way within the project. This concerns also the simulation of the dissolution process of CO2 into brine. Instantaneous dissolution, which is assumed in all the simulators, is very dependent on the gridding scheme. Predicting the amount of dissolution requires also a very good knowledge of the pattern of the geological heterogeneity since it influences the surface- to-volume ratio of the CO2 plume and thus the rate of dissolution. 25

CO2SINK Summary Technical Report

ACKNOWLEDGEMENTS

The success of the simulation studies has fully depended on input, feedback, and discussions with the groups dealing with the geological data and geological interpretations as well as operational data. (Although: without data this whole study would have been much easier.) Thereto, we would like to thank the whole CO2SINK team and industrial partners for cooperation and provision of data. Our special thanks are dedicated to Andrea Förster and Ben Norden (both GFZ Potsdam) for their work on the geological interpretation which was required for implementation of the geological model used within this study.

REFERENCES

Class H., Ebigbo A., Helmig R., Dahle H.K., Nordbotten J.M., Celia M.A., Audigane P., Darcis M., Ennis-King J., Fan Y., Flemisch B., Gasda S.E., Jin M., Krug S., Labregere D., Naderi Beni A., Pawar R.J., Sbai A., Thomas S.G., Trenty L., Wei L. (2009) A benchmark study on problems related to CO2 storage in geologic formations: Summary and discussion of the results. Computational Geosciences 13(4):409-434.

Kopp A., Binning P., Johannsen K., Helmig R., Class H. (2010). A contribution to risk analysis for leakage through abandoned wells in geological CO2 storage. Advances in Water Resources, in Review.

Riaz, A., Hesse, M., Tcheepi, H. A., Orr, F.M.Jr. (2006). Onset of convection in a gravitationally unstable, diffusive boundary layer in porous media, Journal of Fluid Mechanics, 548:87-111.

CO2SINK Summary Technical Report Chapter 5

RISK ASSESSMENT

Todd Flach1, Semere Solomon1

1 Det Norske Veritas (DNV) Research, Energy and Resources, POBox 300, 1322 Hovik, Norway Assisting Partners: Shell, GFZ, StatoilHydro, RWE

CO2SINK Summary Technical Report Risk assessment for gas storage is usually based on expert knowledge. Due to the limited number of CO2 storage sites and the lack of long-term experiences with geological CO2 storage, expert knowledge of the oil and gas industry will remain an important pillar for risk assessment. Different risk assessment approaches were applied to evaluate the risk of storage operations in Ketzin involving two teams of experts that worked independently: Det Norske Veritas (DNV) and Untergrundspeicher- und Geotechnologie-Systeme GmbH (UGS) performed independent risk assessments. The operator (Verbundnetz Gas AG, VNG) and the project leader (Deutsches GeoForschungsZentrum GFZ) were not directly involved in the risk assessments of DNV and UGS but did contribute actively in data collection required by risk assessment. The multi-independent risk assessment was mainly driven by Shell, Statoil, RWE, GFZ and the LBGR, supported by essential input from GEUS, IWS, IAEG, GFZ and others.

The work performed by UGS on site risk assessment was defined by the compliance regime for natural gas storage projects in the Brandenburg region, as per standard procedure applied by the regulator of underground storage activities, Landgesamt für Bergbau, Geologie und Rohstoffe Brandenburg, which requires that a document that presents the case for injection and storage suitability (“Eignungsnachweis”), be submitted, reviewed and approved. Although the required coverage of this report is not specific for the case of CO2 geological storage for the purpose of avoiding climate change, there is considerable overlap with the issues of wellbore integrity and cap rock integrity. This document is traditionally used in permitting and managing natural gas seasonal storage sites. The regulator has indicated however that any CO2-specific issues should also be included in this document, without specifying or prescribing what these might be, since the Ketzin storage site is the very first of its kind (aquifer storage on land).

Part of the risk assessment and risk management plan involved traditional worker safety issues related to hazardous operation identification, repeated incident reporting, post-job analysis, and risk management of these. These issues were handled by independent service suppliers, authorities and certified specialists who were not partners (they performed services on contract to the CO2SINK project) in the project, to avoid any potential conflict of interest. The particular focus of this part of risk management was on the CO2 storage, handling and injection facilities on site, and the robustness of the safety systems associated with these.

The scope of risk assessment tasks performed by CO2SINK partners were identifying, evaluating and proposing mitigative actions for all potential risks within the CO2SINK project, including risks of events or processes that could reduce the research productivity of the project. The most important research topic related to risk management was ensuring that the injected CO2 does not return to the shallow underground and eventually the atmosphere The majority of resources and efforts were put on this. The reasoning is that even if a storage project has excellent site operational safety risk management, it is successful only if it provides storage permanence, which in the long-term perspective (centuries and millenia), is not a site safety issue, i.e. non- permanence does not imply any risk of injury to people or property on site. If storage permanence is not achieved, then the storage project does not contribute to reducing emissions that contribute to climate change. 28

CO2SINK Summary Technical Report

The following text describes the main learning during the CO2SINK project from work flows, formats and platforms applied in the period of 2004-2010 that are especially relevant for decision making and risk assessment in the context of CO2 geologic storage (CGS) projects.

The very first risk assessment and decision-making process began in 2004 with a single, focused workshop in which all sub-project leaders, project management and key industry representatives discussed the most influential risk issues in a structured facilitation conducted by DNV. The goal of this was to identify the very most critical and highest priority risk sources to begin mitigating at the very start of the project.

Following this, but still in the initial, pre-drilling phase of the project, additional risks across all parts of the project were identified through a series of facilitated workshops. One of the key risks identified was long-term leakage out of the storage complex (decades and centuries after closure of the storage site). The use of Features, Events and Processes (FEPs) databases was essential in the workshop sessions dedicated to this.

Workshop results were captured in a fit-for-purpose risk database (EasyRisk ManagerTM), and it is highly recommended that this process and similar software platform be used on all CO2 geologic storage (CGS) projects. Facilitated workshops with representatives from several related technical specialist areas provided improved risk identification and evaluation, as well as mitigative actions. The risk database provided superior transparency, access and traceability in updating and following-up the identified risks as mitigative actions were implemented. The combination of broad participation from all specialist areas in facilitated workshops and capture of results in a fit-for- purpose risk database is recommended as an effective work flow and format for the initial phase of risk assessment and management. An additional benefit from using the fit-for-purpose risk database is that established CGS project learning results can be transferred more efficiently to new projects and work groups.

The primary research focus for risk assessment of CGS projects is on the long-term storage integrity, i.e. making sure that the stored CO2 is contained over a sufficiently long period of time. The format chosen for achieving this was the Safety Case, which has been applied in other types of subsurface storage projects. The distinguishing feature of the Safety Case is that it is a holistic approach with a set of top-level subject areas, but within each subject area, the project organisation is responsible for finding the appropriate level of detail, analyses, estimation methodologies and expert input according to the specific needs of the project. In other words, the Safety Case is not a set, standardized work flow with a prescribed set of standardized calculation procedures with standardized inputs and outputs. It is expected that every storage site has its own set of unique challenges and uncertainties related to mapping the underground, as well as a fit-for-purpose injection and storage solution. Therefore the Safety Case is formed to allow flexibility and scalability to address common thematic challenges which are otherwise dominated by site-specific and project-specific characteristics.

CO2SINK Summary Technical Report

The CO2SINK project has established an in-situ field laboratory for testing novel and innovative monitoring technologies. Therefore there has been no attempt to design and choose an optimal monitoring programme, and no decision making or risk analysis has focused on alternative monitoring solutions and strategies.

By the end of the CO2SINK project, more than 30 thousand tons CO2 were injected. The first time-lapse seismic data survey had been collected, but results were not ready for detailed analysis. It was also too early to judge the effectiveness of electrical tomography in mapping the CO2 plume compared to what can be achieved using various acoustic seismic techniques. These will be the starting point of any follow-up activities at the Ketzin site. It is hoped that the Safety Case for Long-term Storage will function as a reference document for the follow-up projects and that it will be updated as new and more focused field monitoring data and simulation modelling might indicate relevant deviation from predicted performance. At the end of the CO2SINK project, it would appear that sufficient injectivity is achievable despite the high expectation of halite precipitation near the wellbore. This positive result is very likely due to the specific pre- flush performed before CO2 injection began. Some reduced injectivity over time may nonetheless occur and continuation of monitoring of injection parameters is highly warranted.

REFERENCES

EasyRisk Manager™ User Guide ©DNV 2008.

CO2SINK Summary Technical Report Chapter 6

SITE PREPARATION AND INJECTION FACILITY

CO2SINK Summary Technical Report Chapter 6-1 DRILLING AND COMPLETION Bernhard Prevedel 1,Claus Otto 2, Thor Harald Hanssen 3

1 Deutsches GeoForschungsZentrum Potsdam, Telegrafenberg, 14473 Potsdam, Germany 2 Shell International, The Hague,The Netherlands 3 StatoilHydro, Stavanger, Norway

In the framework of the CO2Sink research project three wells have been drilled on the former gas storage facility of VNG. The layout of three wells (Ktzi 200, 201 and 202) in close proximity was planned to obtain high resolution in both, ERT and crosshole seismic experiments. Furthermore, the CO2 should arrive at the observation wells within the lifetime of the project. The activity started on 13, March, 2007 with the drilling of the first well Ktzi-200 and was completed in September of the same year with running the injection string completion in well Ktzi 201.

The objectives for the drilling of the 3 wells were:  Construction of 3 independent suitable drilling pads

 Construction of one injection well with following functionalities: o Drilling of an 8 ½” pilot hole with diverter to a gas flare to a depth of 170m for the exploration for the potential presence of shallow high pressurized gas originating from the previous gas storage operation; o Hole opening of the 8 ½” pilot hole to 23“ and deepening of the well in sections 23“ to 170m, 12 ¼“ to 600m and 8 ½“ to 810m TD; o 6 ¼” coring of the Stuttgart reservoir section; o Running casing / cementing the well in sections 18 5/8“ to 170 m, 9 5/8“ to 600 m; and 5 ½“ to 810 m; o Open hole well logging in sections to 170 m, 600 m, 810 m; o Production testing the potential storage reservoir (Stuttgart); o Installation of the 3 ½“ packer completion; and o Assembly of the wellheads.

 Construction of two observation wells with following functionalities: o Drilling of a 23” hole to a depth of 170m; o Drilling of the well in sections 12 ¼“ to 600m and 8 ½“ to 810m TD; o 6 ¼” coring of the Stuttgart reservoir section; o Running casing / cementing the well in sections 9 5/8“ to 600 m; 5 ½“ to 810 m; o Open hole well logging in sections to 170 m, 600 m, 810 m; o Production tests in the Stuttgart formation; and o Replacement of the drilling mud with brine and assembly of the wellheads.

In the 3 Ketzin wells, with a surface separation of 100 m and 50 m from each other, following geologic profile was expected to be drilled:

CO2SINK Summary Technical Report

CO2SINK Summary Technical Report On 29/30.01.2007 a Drilling the Well on Paper /DWOP) pre-spud workshop was held in Potsdam with participation of representatives from the project partners, the drilling contractor and all service providers. Aim of the 2-day workshop was to work jointly through the drilling plan, develop improvements and identify risks and hazards and ways to mitigate them. As main deliverables for the wells were identified: 1) Seamless acquisition of project data during the drilling phase; 2) Zonal isolation in the Stuttgart formation to the cap-rock; 3) Establishment of good communication with the telemetry; and 4) Determination of the injection regime and intervals 5) at the highest quality, health, safety and environmental standards possible

Well Ktzi-200 reached the geologic target horizon of the Stuttgart formation. Drill cores were retrieved from the target horizon between depths of 610 m and 709,6 m in good quality and 97,7% recovery. The well was fully logged and the casings were installed as per drilling plan and their integrity and pressure tightness confirmed. Due to losses in the Sinemur und Hexeter Formation the 9 5/8“ stinger cementation could not pump the cement to surface. The temperature log detected the cement head at 63 m. The 5 ½“ cementation (above the top sand screen) was irregular due to a failure of a dowh-hole tool. Because of a not opening of the cement ports of this stage cementing tool (POST) the cement was pumped across the sand filters into the annulus. The CBL logs confirmed good cement bond between 455 m – 630 m, meaning still a sufficient isolation of the 9 5/8“ x 5 ½“ annulus. Despite cement flow through the filters the 2 production tests produced formation water from the Stuttgart formation and hence confirmed connectivity to the reservoir. Due to the irregularities during the 5 ½“ production string cementation it was decided to discount well Ktzi-200 as the injector and convert it to an observation well. The planned running of the 3 ½“ packer completion was therefore not realised.

Well Ktzi-201 reached the geologic target horizon of the Stuttgart formation. Drill cores were retrieved from the target horizon between depths of 620 m to 703,5 m in good quality and 100% recovery. The well was fully logged, the casings were installed as per drilling plan and their integrity and pressure tightness confirmed. Due to losses in the Sinemur und Hexeter Formation the 9 5/8“ stinger cementation could not pump the cement to surface. The CBL log detected the cement head at 69 m. CBL logging conducted on 14.09.2007 confirmed good bonding from casing shoe (588,8 m) to 465 m. The cementation of the 5 ½“ production string was performed as planned. Well Ktzi-201 was completed in replacement for Ktzi-200 with the 3 ½” injection string, permanent production packer, subsurface safety valve and fibre-optical pressure and temperature sensors.

Well Ktzi-202 reached the geologic target horizon of the Stuttgart formation. Drill cores were retrieved from the target horizon between depths of 626 m to 644,5 m in good quality and 100% recovery. 34

CO2SINK Summary Technical Report The well was fully logged and the casings were installed as per drilling plan and their integrity and pressure tightness confirmed. The cement head of the 9 5/8” cementation was detected by means of CBL logging at 269 m. CBL logging conducted on 12.09.2007 confirmed good bonding from casing shoe at 579,5 m to 460 m. The cementation of the 5 ½“ production string was performed as planned.

The entire operation in Ketzin lasted for 201 rig days, resulting in three „light“ accidents, two theft incidents, resulting in a total LTI=7. 46 incidents were reported on RIR form, and one major contractor tool failure (stage cementation) occurred.

The operation had 460 hours of rig repair time (10% of Total), plus 1068 hours down- time for tool failures, mainly swellable downhole casing packer equipment (21% of Total).

0 50 100 150 200 0

-100

-200

-300

-400

-500

-600

-700

-800

-900 Time (days)

Fig. 6.1-1: Time-depth curve Ketzin 200+201+202

CO2SINK was the first project that extensively uses behind-casing installations for a study of the CO2 injection and storage process in a geological medium. In this regard, CO2SINK differs from other scientific projects of CO2 test storage, such as the Frio experiment in the Texas, the Nagaoka experiment in Japan, the field test in the West Pearl Queen Reservoir in New Mexico, and the Otway Basin pilot project in Australia. It is envisaged that the extensive set of data generated by cross-correlation of seismic surface monitoring, well-logging, monitoring and simulations will allow for verification of a priori scenarios of storage/migration of fluids. Thus, the observations in progress in Ketzin will contribute to a sound understanding of the thermodynamic processes of CO2 injection at well-scale as well as in the short and longer term the processes during CO2 storage at larger scale.

Based on the results from the drilling and completion of the Ketzin wells a continuation project was proposed to the German government with the drilling of two more wells on the current Ketzin storage site.

CO2SINK Summary Technical Report Chapter 6-2 INJECTION FACILITY Fabian Möller1, Andreas Bannach2

1Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ), Telegrafenberg, 14473 Potsdam, Germany 2ESK-Erdgasspeicher Kalle GmbH (RWE Group), Halsbruecker Str. 34, 09599 Freiberg

The CO2 for injection at the Ketzin site has been delivered from a petroleum refinery by road tankers in a liquid state. Thus, it had to be conditioned prior to injection. Since the CO2 needs a certain overpressure against the reservoir to enable inflow, pressurising was needed. For best monitoring practice the injection facility was requested to be operated under flow-control. Appropriate safety devices have been installed to mitigate any damage of equipment and/or reservoir (formation breakdown) by unintended overpressure. To avoid multiphase-flow at the given pressure conditions, it was decided to heat up the CO2 so that it stays in a single „dense“ phase. Fig. 6-2.1 Phase diagram of CO2 The main goal was to be every time either above the critical pressure and/or critical temperature (see figure 6-2.1).

For the injection of CO2 at the Ketzin site, technical installations have been made to pressurise and heat up the CO2. The injection facility has been tendered for on the European market. It has been erected by Linde AG and has been operated constantly since the start of injection operation (June, 30th 2008). The injection facility is connected to a SCADA (supervisory control and data acquisition) system. The SCADA system includes an audited ESD (emergency shut-down) loop to meet all HSE (health, safety & environment) requirements. The whole facility has been technically and electrically audited by the German „TÜV Nord“ organisation. -1 The CO2 injection facility consists of five main plunger pumps (0 - 1,000 kg h ), a heating device (300 kWel) and two intermediate storage tanks (50 t, each). The facility -1 -1 -1 was designed to handle a CO2 stream of 300 kg h to 3,250 kg h (200 kg h stepwise) at 50 °C at the heater outlet, resulting in a maximum injection of 78 t of CO2 per day. During commissioning the control parameters were optimised, and an additional air- heater was installed between the CO2 pumps and the electrical heater to pre-heat the cooled and liquid CO2 to ambient conditions in order to reduce the required amount of electrical energy for gas conditioning and to ensure a smooth injection regime. This additional air heaters significantly reduced energy consumption and variation in CO2 flux during injection.

CO2SINK Summary Technical Report An overall control and automation system is used for controlling the injection process and for monitoring the relevant injection parameters: CO2 flow, temperature along the injection string, pressure data from the formation, and the pressures at the wellheads. All emergency shut-down (ESD) functionality is software independent and has been certified by TÜV and the technical control board of the project. Furthermore, the mining authorities acknowledged the emergency and operation plans of the plant. The CO2 flux is measured using a Coriolis gas flow meter ten meters away from the well head. The well head and the annuli pressures of all three wells are recorded continuously using pressure transducers connected to the overall control and automation system. In the injection well, temperature and pressure are additionally monitored with a bottom hole sensor. Following the same standards used in the gas storage industry, surface and subsurface safety valves are operated independently from the control and automation system with a hydraulic single-well control panel close to the well. Additional manual valves act as further safety equipment.

CO2SINK Summary Technical Report Chapter 7

EXPERIENCES OF STORAGE OPERATION

Fabian Möller1, Andreas Bannach2, Sebastian Köhler3, Axel Liebscher1, Thomas Thielemann3

1Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ), Telegrafenberg, 14473 Potsdam, Germany 2ESK-Erdgasspeicher Kalle GmbH (RWE Group), Halsbruecker Str. 34, 09599 Freiberg 3UGS GmbH Mittenwalde, Berliner Chaussee 2, Mittenwalde 4RWE Power AG, Stüttgenweg 2, 50935 Köln

CO2SINK Summary Technical Report

A safe operation of injection at the Ketzin site had to be ensured. The injection operation had to be in line with the legal and scientific requirements. To ensure a smooth and safe operation at the Ketzin site, the following decisions have been made:

 Personnel should be at the site – no remote control of the storage activities  A partner from the CO2SINK consortium with experience in the field of underground storage should run the injection

In consequence, the GFZ has assigned the injection operation to VNG AG. The VNG personnel looks after the injection facility and all three wellheads at least once an hour. They ensure that the injection facility runs as desired, check wellhead and annuli pressures and CO2 sensors. In addition to this on-site staff, additional engineering companies were awarded to accompany the injection operation. These are UGS for periodic reporting to the mining authority and additional legal underground monitoring issues and ESK for on-call duty. Figure 7-1 gives an overview on the organisation of the injection operation.

Fig. 7-1 Organisation of the injection operation

To integrate the scientific needs and ensure the smooth implementation of an integrated scientific-technical injection regime, weekly meetings were scheduled to discuss injection rates and shut-in phases for scientific experiments. A very conservative pressure limit has been set by the CO2SINK consortium to not exceed 82 bar at the injection point (sand-face) – this translates to a ~ 80 bar pressure reading at the P/T gauge.

CO2SINK Summary Technical Report The injection rate has been ramped up carefully in the beginnig of the injection operation. The first time, only daily operation has been carried out under engineer supervision at the site to carefully test all functionality of the injection facility. Injection started officially on June 30th 2008. All possible injection rates have been tested from a few 100 kg h-1 to the maximum design rate of 3.250 kg h-1.

Regular operations started on September 24th, 2008 with a constant injection rate setpoint of 2.000 kg h-1. From December, 12th 2008 on, the setpoint was 2.600 kg h-1. In a last step, the setpoint was risen to maximum (3.250 kg h-1) on March, 3rd 2008. This stepwise ramping up has been carried out to perform as reservoir test.

Before the CO2 injection in June 2008 hydraulic tests were conducted to get detailed information about the hydrodynamic characteristics of the reservoir. During the injection phase, tests were designed and realized by the operational management (e.g. flow after flow test, modified isochronal test). Based on operational conditions (e.g. high oscillation of the injection rate), it was not possible to get results with workable accuracy. However, with the results of the tests, it was possible to confirm the hydraulic parameter of the pre-injection tests and to rule out a degradation of the reservoir characteristics.

The reservoir behaviour has been monitored by recording the CO2 massflow in the injection piping and a pressure/temperature (P/T) gauge. Before the start of injection, there was much uncertainty about possible phase changes of CO2 inside the injection tubing. A P/T gauge was installed at 550 m depth to constantly monitor the pressure/temperature downhole. Technically, it was impossible to place the sensor directly at formation depth (630 m). Thus a correction for the remaining 80 m of CO2 column underneath the P/T gauge is necessary and has been calculated via numerical flow simulations. A fairly good correction factor has been „P/T-gauge + ~2 bar“ to get the formation pressure. Until end of January 2010, approx. 30.000 tons of CO2 have been injected. The overall injection rate since start of injection (June 30th, 2008) is 1.500 tons per month (50 tons per day). The maximum amount was 2.188 tons a month (71 tons per day). The overall performance of the injection operation was very well and safe. There have been no HSE issues to report. The operation at Ketzin could prove for the first time in Europe that safe injection and storage of CO2 into an onshore saline aquifer can be realized. This operation of injection benefits from best available technology which has been developed within the oil and gas producing industry the last decades. Having a smoothly running research & storage site at Ketzin has given a major contribution to facilitate this development. CO2 injection at Ketzin will carry on with the help of the nationally funded CORTIS project. It is intended to further inject CO2 within the follow-up project CO2MAN with the same companies and persons responsible.

CO2SINK Summary Technical Report Chapter 8

MONITORING CO2 MIGRATION

CO2SINK Summary Technical Report Chapter 8-1

BASELINE SURFACE GEOCHEMISTRY OF CO2 AND GEOMICROBIOLOGY

PAS MEERI - Polish Academy of Sciences Mineral and Energy Economy Research Institute, Wybickiego 7, 31-261 Cracow, Poland

Methods which use soil microorganisms to monitor concentration changes of carbon dioxide in soil gas have not been well recognized, although sufficient knowledge already exists to do so. To discover indicatory microorganisms different samples, from the Ketzin site and South Poland, were analyzed. For the investigation of the microbiological parameter variability depending on CO2 supply laboratory experiments were performed. The work focused on the definition of the impact of different CO2 concentrations on indicative microorganism cell numbers. Ground samples were investigated using standard methods. A special method adapted to project needs was developed to read out bacterial cell numbers. For the estimation of a possible seasonal variability of number and occurrence of indicative microorganisms several lab experiments with samples taken during several campaigns were carried out using standard methods. Sampling was conducted according to a grid placed along a previously defined profile through the top zone of the Ketzin anticline, nearby the CO2 injection well. Based on the results of all the analyses and experiments, which were performed monitoring guidelines were formulated.

Clostridium kluyveri and the nitrification process can be applied as indicators to estimate the CO2 supply within the soil environment. The existence of a correlation between CO2 supply and an increase of the bacterial cell numbers of Clostridium kluyveri species and bacterial consortia could be confirmed. Calibration curves drawn-up according to the results, displaying dependence between indicative bacterial cell number and carbon dioxide concentration, can be applied to monitor carbon dioxide underground storage sites. Results obtained from microbiological examinations of soil samples taken over the whole year from the Ketzin site show the sensitivity of the bacteria to CO2 supply, the dependency of the cell number on the sampling location as well as the applicability of Clostridium kluyveri bacteria and the nitrification process as tools to read CO2 supply variability in soil air. Carbon dioxide storage site monitoring using microbiological methods was sucessfully developed within the frame of the CO2SINK Project.

CO2SINK Summary Technical Report Chapter 8-2 GAS GEOCHEMICAL INVESTIGATIONS, METHODS AND TOOLS APPLIED FOR SURFACE AND SUBSURFACE GAS CONCENTRATION MONITORING AT THE KETZIN CO2- STORAGE TEST SITE Martin Zimmer1, Jörg Erzinger1, Peter Pilz1 & Christian Kujawa1

1Helmholtz Centre Potsdam, German Research Centre for Geosciences (GFZ), Telegrafenberg, 14473 Potsdam, Germany

Subsurface storage of carbon dioxide in saline aquifers demands for sophisticated monitoring tools for long term safety and environmental impact issues. Despite extensive research, many factors governing the fate of injected carbon dioxide remained unclear. Further comprehensive research is essential to better understand the behaviour of CO2 during geological storage. Therefore, a program for permanent and direct monitoring of gases in the reservoir and for determining the CO2 flux at the ground surface was started at the CO2SINK site, targeted to measure the behaviour of injected CO2 in the storage horizon and to identify possible risks by leakage of the CO2 storage reservoir at depth. Different standard techniques have been applied for the surface monitoring, whereas for the borehole measurements a newly developed innovative geochemical Gas- Membrane-Sensor (GMS) was deployed. A meteorological station was set up on the test site, to monitor the local weather conditions. These data are necessary to correlate with the geochemical data. To gain adequate long term baseline data on the natural local and seasonal background CO2-flux variations, surface measurements have been made once a month over a five years period. Shortly before the injection started, the two GMS monitoring tools were installed in the deep observations wells to continuously measure the gas composition. It was possible to successful detect the arrival of the injected gases in the two observation wells.

The special concern of this work was to establish a gas monitor concept which allows for long term geochemical monitoring at depth, to trace the composition of the formation gas and the concentration and spatial migration of CO2 in the subsurface reservoir horizon as well as to monitor the CO2-flux and the soil gas composition at the surface for safety and environmental issues. Based on the work carried out in the project, it was successfully proven, that the gas chemical monitoring concept used in Ketzin is an effective method for the permanent surveillance of the CO2 storage. It allows for the measurement of surface CO2 fluxes as well as for the on-line determination of dissolved gases in formation fluids in deep bore holes. The observed trends of the CO2 flux at the surface mainly depend on rising ambient temperatures and, thus, growing bio-activity in the root zone. No unusual CO2 concentrations below the base plate of the bore hole cellars have been detected.

CO2SINK Summary Technical Report Furthermore, no methane or hydrocarbon gas concentrations above the natural atmospheric background were detected in the soil gas. From the gas concentration depth profiles the highest gas concentrations were encountered in porous permeable sandstone layers, whereas the relative impermeable cap rock yield only small amounts of gas. The gas composition of the formation fluids changed drastically from a N2 to a CO2 dominated gas phase after the start of injection, respectively the arrival of the gases in the observation wells. The gas water ratio increased significantly and the CO2 concentration rose successively to 99,4 %. The arrival of the Kr gas tracer, technical N2 and the breakthrough of the CO2 were detected and the relative velocities of the injected gases at the observation wells were evaluated. The GMS measurements, therefore, provide important benchmark data for the calibration and improvement of geological models.

CO2SINK Summary Technical Report Chapter 8-3 WELL MONITORING Jan Henninges1, Barry Freifeld2, Matteo Loizzo3, Karl Schwab4, Dat Vu-Hoang3

1 GFZ German Research Centre for Geosciences, 14473 Potsdam, Germany 2 Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA 3 Schlumberger Carbon Services, 92926 La Defense Cedex, France 4 Schlumberger Oilfield Services, 49377 Vechta, Germany

In the framework of the CO2SINK project geophysical borehole measurements are carried out in order to monitor migration of CO2 during injection into a saline aquifer. Both the migration inside the reservoir and along the boreholes is considered. Fluid saturation changes and the vertical CO2 distribution in the injection zone are determined using a combination of pulsed neutron-gamma (PNG) wireline logging, distributed temperature sensing (DTS), and a newly developed method using thermal perturbation measurements (DTPS, Freifeld et al., 2008). The integrity of the well cementation, which is crucial to ensure zonal isolation and avoid vertical migration along the boreholes, was monitored using a combination of classical cement-bond logging and ultrasonic imaging of the cement sheaths, together with DTS monitoring during the cementation process. The measurements described herein enable to derive detailed information about in-situ conditions at borehole-scale and provide reference points for the other geophysical methods applied at Ketzin working at larger scales.

Owing to the high formation water salinity, the high formation porosity along with a high contrast in capture cross section between saline formation water and CO2, distinct changes indicative of the displacement of brine by CO2 in the near-wellbore environment could be measured at the Ketzin wells using PNG wireline logging. The precise determination of formation saturation changes is nevertheless partially hampered because of the change of the borehole fluid between the baseline and repeat logging runs, as well as the invasion of CO2 into the uncemented sections of the well annuli in the reservoir section.

As a novelty compared to other CO2 storage pilot projects, both the injection well Ktzi 201 and the two observation wells Ktzi 200 and Ktzi 202 are equipped with “smart casings” containing permanent downhole sensors. For DTS monitoring, fiber-optic sensor cables have been installed behind the borehole casing and were cemented in place (Prevedel et al., 2008). The temperature data enables to detect behind-casing flow and to delineate the reservoir sections which take up the CO2 in the injection well from the temperature changes recorded during the early time after start of injection. The intervals determined correlate well with the positions of the sand- and siltstone intervals with higher porosity and permeability, and aid the interpretation of the PNG wireline logging for determining reservoir sections with increased CO2 saturations. The results of the first repeat DTPS measurement revealed decreased thermal conductivities at the level of the main reservoir sands, which also indicates elevated CO2 saturations in these strata. An integrated analysis of the monitoring results shows that at the injection well 45

CO2SINK Summary Technical Report

CO2 is predominantly taken up by a 17m-thick sandstone interval where the highest CO2 saturations up to 82 % are reached, and a thinner more silt-dominated interval with significantly lower saturations underneath. At the observation wells increasing CO2 saturations with values up to about 66 % at Ktzi 200, and 45 % at Ktzi 202 have been observed.

The temperature anomalies detected around the time of arrival of CO2 at the Ktzi 200 observation well (Giese et al., 2009) are predominantly related to flow processes within the wells. The development of a two-phase region in both observation wells, with P/T conditions fixed to the vapor pressure curve, resulted in significant temperature anomalies after the arrival of CO2 (Henninges et al., 2010). Both pressure and temperature log data enable the localization of liquid and gaseous CO2 phase zones inside the wells. This could aid in the determination of the downhole pressure conditions which are very difficult to predict under these circumstances using standard procedures used in reservoir engineering.

When comparing completion designs among the three wells, the solution adopted for Ktzi 200 seems superior to the others with respect to zonal isolation. In particular, having a section of cement isolating the open hole just below the 9⅝” casing shoe (and above the filter screens) outside the coated 5½“ production casing section would serve a double essential purpose: Firstly it provides an additional barrier to possible leaks before exposing the 9⅝”-12¼” annulus, secondly it allows repeat logging of cement in contact with CO2. The presence of large uncemented intervals of the wells in the reservoir sections has resulted in unfavorable conditions for the interpretation of PNG wireline logs because of uncontrolled fluid movement in the annular space between the casing and the borehole wall during the injection process.

REFERENCES: Freifeld, B. M., S. Finsterle, T. C. Onstott, P. Toole, and L. M. Pratt (2008), Ground surface temperature reconstructions: Using in situ estimates for thermal conductivity acquired with a fiber-optic distributed thermal perturbation sensor, Geophysical Research Letters, 35, L14309.

Giese, R., J. Henninges, S. Lüth, D. Morozova, C. Schmidt-Hattenberger, H. Würdemann, M. Zimmer, C.-G. Cosma, C. Juhlin, and CO2SINK Group (2009), Monitoring at the CO2SINK Site: A Concept Integrating Geophysics, Geochemistry and Microbiology, Energy Procedia, 1(1), 2251-2259.

Henninges, J., A. Liebscher, A. Bannach, W. Brandt, S. Hurter, S. Köhler, F. Möller, and CO2SINK Group (2010), P-T-ρ and two-phase fluid conditions with inverted density profile in observation wells at the CO2 storage site at Ketzin (Germany), 10th International Conference on Greenhouse Gas Control Technologies, Amsterdam, September 20-23, 2010.

Prevedel, B., L. Wohlgemuth, J. Henninges, K. Krüger, B. Norden, A. Förster, and the CO2SINK Drilling Group (2008), The CO2SINK boreholes for geological storage testing, Scientific Drilling, 6, 32-37.

CO2SINK Summary Technical Report Chapter 8-4

SURFACE AND BOREHOLE SEISMICS

Peter Bergmann1, Calin Cosma2, Nicoleta Enescu2, Rüdiger Giese1, Julia Götz1, Alexandra Ivanova1, Chris Juhlin3, Niklas Juhojuntti3, Artem Kashubin3, Stefan Lüth1, C. Yang3 and F. Zhang3

1GFZ German Research Centre for Geosciences 14473 Potsdam, Germany 2 Vibrometric Oy Cosma, Taipaleentie 127, 01860 Perttula, Finland 3 Uppsala University P.O. Box 256, SE-751 05 Uppsala, SWEDEN

An important component of the project, and for geological storage of CO2, is monitoring the movement of injected CO2 using seismic methods. The seismic characterization strategy adopted within the CO2SINK project aims to cover the site at kilometer scale, while resolving down to meter scale. Hence, time-lapse investigations at all scales are being done, together with modeling and data integration. The time-lapse multi-line 2D survey is meant to tie-in the down-well surveys with the 3D, while the time-lapse VSP and MSP surveys are meant to cover, with higher resolution a smaller region (~300m radius) from more diverse view angles and identify possible steeply inclined migration paths of the CO2. As a first step in the seismic program a 3D seismic survey with about 12 square km of sub-surface coverage was acquired in 2005 with the objectives of providing (1) if possible, an understanding of the structural geometry for flow pathways within the reservoir, (2) a baseline for later evaluation of the time evolution of rock properties as CO2 is injected into the reservoir, and (3) detailed sub-surface images near the injection borehole for planning of the drilling operations. The uppermost 1000 m were well imaged on the 3D survey, showing an anticlinal structure with an east-west striking central graben on its top. No faults were imaged near the drill sites. Remnant gas, cushion and residual gas, from a previous natural gas storage facility, is present near the top of the anticline in the depth interval of about 250–400 m and has a clear seismic signature on the 3D survey. In addition to the 3D survey, seven 2D lines in a “star” configuration were acquired during the 3D campaign. During the autumn of 2007, baseline VSP and MSP data were acquired at the injection site. The MSP and “star” acquisition campaigns were repeated in autumn 2009, along with the acquisition of a smaller repeat 3D centered on the injection site. Crosshole seismics were acquired shortly before injection started and have since then been repeated three times, twice in the early stages of the injection process. The most important monitoring result is from the repeat 3D survey acquired in 2009. Comparisons with the baseline data show that the injected CO2 has its highest concentrations near the injection borehole. CO2 can be observed about 150-200 m away from the injection borehole, but the plume is not radially symmetric indicating the strong lateral heterogeneity of the reservoir. More CO2 has migrated downdip than previously expected and there is a tendency for a NW-SE trend in the observed plume. CO2 may 47

CO2SINK Summary Technical Report have spread further than the seismic data indicate since the resolution of the seismic data is limited. Interpretations of the 2D “star“ repeat data, as well as the repeat MSP data are consistent with the interpretation of the 3D repeat data.

ACKNOWLEDGEMENTS

The extension of the seismic monitoring programme (the first seismic 3D repeat survey) is partially funded by the German Ministry of Education and Research (BMBF), within the GEOTECHNOLOGIEN programme, which we gratefully acknowledge.

REFERENCES

Juhlin C., Giese R., Zinck-Jørgensen K., Cosma C., Kazemeini H. and Juhojuntti N. (2007). 3D baseline seismics at Ketzin, Germany: The CO2SINK project. Geophysics 72, 121–132. doi: 10.1190/1.2754667.

Kazemeini H., Juhlin C., Zinck-Jørgensen K. and Norden B. (2009). Application of the continuous wavelet transform on seismic data for mapping of channel deposits and gas detection at the CO2SINK site, Ketzin, Germany. Geophysical Prospecting 57, 111–123. doi: 10.1111/j.1365-2478.2008.00723.x.

Yordkayhun S., Ivanova A., Giese R., Juhlin C., and Cosma C. (2009). Comparison of surface seismic sources at the CO2SINK site, Ketzin, Germany. Geophysical Prospecting 57, 125-139.

CO2SINK Summary Technical Report Chapter 8-5 GEOELECTRICAL CROSS-HOLE AND LARGE-SCALE SURFACE / SURFACE-DOWNHOLE MONITORING AT THE KETZIN CO2-STORAGE TEST SITE

Conny Schmidt-Hattenberger1, Dana Kießling1, Peter Bergmann1, Kay Krüger1, Carsten Rücker2, Hartmut Schütt3

1 Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences, Centre for CO2 Storage, Telegrafenberg, 14473 Potsdam, Germany, [email protected] 2University of Leipzig, Institute of Geophysics and Geology, Talstr.35, 04103 Leipzig, Germany, [email protected] 3Statoil ASA, Grenseveien 21, 4035 Stavanger, Norway, [email protected]

If geologic formations are intended to be used for carbon dioxide (CO2) storage over long time periods, thoroughly monitoring of the underground reservoir is imperative to confirm the containment of CO2, assess leakage paths, and gain understanding into interactions between CO2, the formation and corresponding fluids. To support these challenging objectives the related national-funded project COSMOS had been closely linked to CO2SINK to strengthen the development and application of an innovative geophysical monitoring technique based on geoelectrical methods. An array of electrodes is permanently located in boreholes across the storage formation and resolves the distribution of fluid saturation around the boreholes. This sophisticated approach of electrical logging is very similar to the one used in surface electrical surveys, but electrodes and cables will be vertically deployed in the borehole and cemented outside of an electrically insulated metallic casing. The prototype-design was called Vertical Electrical Resistivity Array (VERA)-system, and has been deployed for the first time at a CO2-injection test site since summer 2007. Geoelectrical methods are particularly suited for monitoring CO2 storage in deep saline aquifers due to the high conductivity contrast between CO2 and brine (Ramirez et al. 2003, Christensen et al., 2006). They provide independent information on the electric resistivity of the fluid-bearing rock that can help to constrain and refine seismic and gravity models of the subsurface. The electrical resistivity can be interpreted in terms of the relative CO2 and brine saturation. The field method and the interpretation scheme as well are easily transferable to tackle other fluid-related problems, such as oil and gas production, enhanced oil recovery operations, or environmental monitoring. In the report, the feasibility of monitoring CO2 migration in the Ketzin saline aquifer at a depth of about 650 m with cross-hole and surface-downhole electrical resistivity tomography (ERT) is described. The deepest European permanent VERA-system consists of 45 electrodes (15 in the injection well Ktzi201 and 15 in each of the two observation wells Ktzi200 and Ktzi202), successfully placed on the electrically insulated casings, in the depth range of about 590 m to 740 m with a spacing of about 10 m. First synthetic modelling studies indicate an increase of the electrical resistivity of about 200 % caused by CO2 injection, corresponding to a bulk CO2 saturation of 50 %, which is in good agreement with laboratory studies. Finite difference inversion of field data delivers

CO2SINK Summary Technical Report three-dimensional resistivity distributions between the wells which are consistent with first multiphase fluid-flow reservoir modelling studies (Lengler et al., 2010). To increase the limited observation area provided by the cross-hole measurements, additional surface-downhole measurements were deployed. A considerable CO2 migration in SE-NW direction is deduced from surface to downhole resistivity experiments. The first cross-hole time-lapse results show that the resolution and the coverage of the electrode array in the Ketzin setting are sufficient to resolve the expected resistivity changes on the characteristic length scale of the electrode array. Significant resistivity changes could be measured, however detailed information on CO2 plume could not be resolved yet by VERA under the existing geological circumstances (Kiessling et al., 2010). This will be the objective of further data integration and joint interpretation of all deployed monitoring methods, planned in future research projects related to Ketzin.

At Ketzin, ERT has been successfully deployed for tracking the CO2 plume between the monitoring wells and around the near-wellbore area. Repeat surveys should be carried out to verify the time-lapse effect of CO2 migration. Joint data interpretation of seismic and geoelectric results, constrained by laboratory and logging results, can help to develop an approach for a more quantitative analysis of the stored CO2 volume. Embedded in the frame of the geological structural model this integrative evaluation can contribute to performance and risk assessment work for the Ketzin test site.

ACKNOWLEDGMENTS This ERT research work was supported in part by COSMOS (CO2 Storage, Monitoring and Safety Technology), financed by the German Federal Ministry of Education and Research and its R&D program “GEOTECHNOLOGIEN”, and additional EU-funded by the CO2SINK project. Special thanks are extended to Dr. William D. Daily from the Lawrence Livermore National Laboratory for his helpful comments and fruitful discussions in the phase of planning the VERA system.

REFERENCES: Kiessling, D., Schmidt-Hattenberger, C., Schuett, H., Schilling, F., Krueger, K., Schoebel, B., Danckwardt, E., Kummerow, J., and the CO2SINK Group, (2010). Geoelectrical methods for monitoring geological CO2 storage: First results from crosshole and surface-downhole measurements from the CO2SINK test site at Ketzin (Germany), Int. J. Greenhouse Gas Control, 4, pp. 816-826.

Lengler, U., De Lucia M., Kühn, M., (2009). The impact of heterogeneity on the distribution of CO2: Numerical simulation of CO2 storage at Ketzin, Int. J. Greenhouse Gas Control, doi:10.1016/j.ijggc.2010.07.004

CO2SINK Summary Technical Report Chapter 8-6 MONITORING HYDROGEOCHEMICAL AND MICROBIOLOGICAL EFFECTS

Hilke Würdemann1, Andrea Vieth1, Daria Morozova1 , Ann-Kathrin Scherf1

1 Helmholtz Centre GeoForschungsZentrum Potsdam, Centre for CO2 storage, Telegrafenberg, 14473 Potsdam, Germany 2 Centre GeoForschungsZentrum Potsdam, Organic Geochemistry Section, Telegrafenberg, 14473 Potsdam, Germany

Since microorganisms represent very effective geochemical catalysts, they may influence the process of CO2 storage significantly. The interactions between the supercritical CO2, the microorganisms, the reservoir fluids as well as the minerals and the natural organic matter of both the reservoir and the cap rock will cause changes to the microbial activity and the structure and chemical composition of the rock formations. Due to the rearrangement of minerals and/or precipitation of organic matter the reservoir permeability can be influenced locally. In addition, precipitation and corrosion may be induced around the well affecting the casing and the casing cement. Therefore, analyses of the composition of microbial communities and its changes in combination with monitoring of hydrogeochemical effects will contribute to an evaluation of the effectiveness and reliability of the long-term CO2 storage technique.

This study presents the monitoring of the geochemical and microbiological changes in fluids taken from the injection and the observation wells of the saline CO2 storage aquifer. The injection of supercritical CO2 caused changes in the concentration of inorganic and organic substances in the reservoir fluid. This could be shown by short term extraction experiments and long-term exposure experiments (see Chapter 3) and partly by downhole monitoring. The representativeness of downhole samples for processes within the reservoir was limited because the hydraulic connection of wells with the reservoir decreased with increasing well head pressure (Würdemann et al. 2010). Therefore in situ monitoring focuses on characterising the composition and activity of the autochthonous microbial community in brine samples and their changes in response to CO2 exposure (Morozova et al. 2010). A prerequisite of these studies is the acquisition of samples free of or negligibly affected by drill mud and technical fluids releasing organic and inorganic compounds as well as microorganisms. A contamination control for the fluid samples was done by the application of the fluorescent dye tracer Fluorescein (Wandrey et al. 2010). Prior to the CO2 injection, a temporal injectivity loss in the injection well Ktzi 201 due to black solids was observed. Chemical and XRD analyses proved that the black solids consisted mainly of iron sulphide (Zettlitzer et al. 2010). Sulphate reducing bacteria (SRB) were detected in fluid samples indicating that the formation of iron sulphide was caused by bacterial activity. The N2 lift of Ktzi 201 succeeded in the full restoration of the well productivity and injectivity as well as a reduction of SRB and organic compounds in the well. The acetate concentration was reduced by two orders of magnitude. 51

CO2SINK Summary Technical Report Although saline aquifers could be characterised as an extreme habitat for microorganisms due to reduced conditions, high pressure and salinity, a high number of diverse groups of microorganisms were detected with downhole sampling in the injection and observation wells. By using Fluorescence in situ Hybridisation (FISH) and molecular fingerprinting we have shown that the microbial community was strongly 6 -1 influenced by the CO2 injection. Before CO2 arrival, up to 6x10 cells ml were detected by DAPI-staining at a depth of about 650 m below the surface. The microbial community was dominated by the domain Bacteria that represented approximately 60 to 90 % of the total cell number, with Proteobacteria and Firmicutes as the most abundant phyla comprising up to 47 % and 45 % of the entire population, respectively. Both the FISH and fingerprinting analyses revealed quantitative and qualitative changes after CO2 arrival. Our study revealed temporal outcompetition of sulphate reducing bacteria by methanogenic archaea. In addition, an enhanced activity of the microbial population after five months CO2 storage indicated that the bacterial community was able to adapt to the extreme conditions of the deep biosphere and to the extreme changes of these conditions due to CO2 exposure (Morozova et al. 2010a,b).

Further in situ studies of microbiological and geochemical effects of CO2 storage are necessary to improve the geoscientific understanding of biological processes and their contribution to the mineral creation and dissolution. The injection of supercritical CO2 caused changes in the temperature, the pressure and the concentration of inorganic and organic compounds in the reservoir fluid. Flow of CO2 and other reservoir fluids resulted in mobilization of the organic and inorganic substances since supercritical CO2 is known to be an excellent solvent. Mineral and organic precipitations can lead to a local reduction in permeability.

Morozova, D., Wandrey, M., Alawi, M., Zimmer, M., Vieth, A., Zettlitzer, M., Würdemann, H., and the CO2SINK Group (2010a online first). "Monitoring of the microbial community composition in saline aquifers during CO2 storage by fluorescence in situ hybridisation." International Journal of Greenhouse Gas Control. doi: 10.1016/j.ijggc.2009.11.014.

Morozova, D., Zettlitzer, M., Let, D., Würdemann, H. and CO2SINK Group (2010b). Monitoring of the microbial community composition in deep subsurface saline aquifers during CO2 storage in Ketzin, Germany. Energy Procedia 10, in press.

Wandrey, M., Morozova, D., Zettlitzer, M., Würdemann, H. (2010). "Contamination assessment of rock core and brine samples at the Ketzin storage site using fluorescent tracers." International Journal of Greenhouse Gas Control. doi:10.1016/j.ijggc.2010.05.012.

Würdemann, H., Moeller, F., Kühn, M., Heidug, W., Christensen, N.P., Borm, G., Schilling, F.R., and the CO2SINKGroup (2010). " CO2SINK – From Site Characterisation and Risk Assessment to Monitoring and Verification: One Year of Operational Experience with the Field Laboratory for CO2 Storage at Ketzin,

CO2SINK Summary Technical Report Germany." International Journal of Greenhouse Gas Control. doi:10.1016/j.ijggc.2010.08.010.

Zettlitzer, M., Moeller, F., Morozova, D., Lokay, P., Würdemann, H., and the CO2SINK Group (2010 online first). "Re-Establishment of proper injectivity of the CO2- injection well Ketzin-201." International Journal of Greenhouse Gas Control. doi: 10.1016/j.ijggc.2010.05.006.

CO2SINK Summary Technical Report Chapter 9

DISSEMINATION AND PUBLIC OUTREACH

Michael Haines1, Gunther Borm2,3, Michael Kühn2, Hilke Würdemann2, Frank Schilling2,3

1 IEAGHG, Orchard Business Centre, Stoke Orchard, Cheltenham, UK 2 GFZ, Telegrafenberg, Potsdam, Germany 3 Universität Karlsruhe, KIT, Engler-Bunte-Ring 14, 76131 Karlsruhe, Germany

CO2SINK Summary Technical Report

Public acceptance is one of the fundamental prerequisites for geological CO2 storage. After technology development and demonstration, good communication is essential to gaining public acceptance. In highly populated areas like central Europe, especially in the vicinity of metropolitan areas, the effect of underground operations are very much the focus of attention for the people living next to the site, the media, and politicians. To gain acceptance, all these groups – the people in the neighbourhood, journalists, and authorities – need to be confident of the security of the planned storage operation as well as the long term security of storage and the technical effectiveness of carbon capture and storage (CCS) .

Dissemination of information and public outreach, particularly to those living in the immediate vicinity of the CO2SINK injection site were recognized from the beginning as important activities for the success of the project. Dissemination was considered to range from provision of general information to a non-scientific audience through to the sharing of detailed scientific and technical information with the scientific community and amongst the many project partners. The main activities carried out were:- Writing and printing brochures and flyers about the project and its progress Setting up and maintaining a project website Holding scientific meetings and forums Holding stakeholder forums and events

A communications strategy was drawn up and formed the basis for planning and tailoring the communication activities. Primary objectives of the project were defined as follows:- 1) To advance understanding of the science and practical processes involved in underground storage of CO2 to reduce emissions of greenhouse gases to atmosphere. 2) To build confidence towards future European Carbon Dioxide capture and geological storage. 3) To provide real case experience for use in development of future regulatory frameworks for CO2 geological storage.

Communication objectives were formulated as follows:- 1) To inform all interested parties of the principles of geological storage and plans and progress with the CO2 storage experiments at Ketzin. 2) To provide a platform for interested parties to give input to the experimental programme. 3) To enable exchange of ideas amongst researchers. 4) To assist in demonstrating to the local community the safety, environmental impact and security of the experimental programme. 5) To identify issues likely to arise as a result of the project and prepare response.

A list of key audiences and their specific interests was drawn up and finally a set of key messages to be delivered were formulated as follows:- 1) Green house gases contribute to climate change. 2) There is a strong desire in the world community to reduce CO2 emissions because of this. 3) One option is to put CO2 back into the deep underground. 55

CO2SINK Summary Technical Report 4) This option has very large potential and is one of the few options which can give deep cuts in emissions. 5) Much work has to be done to develop it as a safe option even though the principle is very simple. It will be a major engineering undertaking. 6) This is one of a series of key experiments.

Consortium members were issued with an overview of the project which could be used to answer questions from third parties and also a detailed guidance note on how to handle such approaches. Where possible such requests were always referred to the Project Office so that consistent information would be provided.The following principles were adopted with regard to release of information and data about the project.  All written publications, papers, presentations, data etc should be approved by the Project Co-ordination Committee before release.  All consortium members should abide by the clauses governing handling of proprietary information contained in the Consortium Agreement.

All researchers were encouraged to publish scientific material. A log of all publications was kept by the project office and made available on the project website. The project hosted an international technical meeting of the CSLF (Carbon Sequestration Leadership Forum) early in the project at which the plans for CO2SINK were presented alongside plans for other CCS demonstration projects around the world. In April 2009 a number of papers on the results of research under the CO2SINK project were presented at the European Geophysical Union meeting. This material was subsequently published in a special issue of the new Elsevier’s scientific publication the International Journal of Greenhouse Gas Control Technology (IJGGC). Later in the project two of the regular progress meetings were combined with a forum for young researchers under the GRASP (Greenhouse-gas Removal Apprenticeship and Student Programme) initiative.

More general engagement was carried out by:- Preparing a brochure and yearly project flyers for distribution at conferences and events. Having a stand at the Communicating European Research event in Brussels in 2005 Organising several “stakeholder engagements” A press event in Berlin for which a special video on CCS was prepared A local information exchange meeting at the nearby town of Ketzin Start of drilling and start of injection ceremonies at site to which press, public and officials were invited Setting up and maintaining a website with both a public and a consortium member section. The members section was used as a repository for much drilling and geological information which needed to be shared. Setting up a public information room at the injection site to which visitors were taken on request. A small pressurized viewing chamber was constructed for display at the site so that visitors could see first hand the strange behaviour of supercritical CO2.

A tally of the various dissemination and outreach activities was kept as the project progressed and gives a good illustration of the overall extent of dissemination activities for a project of this type.

CO2SINK Summary Technical Report 1,238 releases by press, radio and TV 73 conference contributions 51 conference attendances 8 reports 7 colloquia 6 reviews 43 workshops 8 flyers 18 hearings 3 popular scientific journal articles 47 scientific publications 17 films/videos

The public outreach and dissemination elements of a scientific project of this nature can easily be underestimated and a key lesson is that they need to be just as carefully planned and budgeted from the beginning as are the scientific and engineering activities. The exact response of interested parties cannot be predicted in advance. What can be anticipated is that persons from a wide range of backgrounds and locations may become interested and, when they do, will expect immediate responses. To satisfy this interest, a range of information has to be readily to hand in forms that can be easily understood and assimilated. The Project Co-ordination Committee (PCC) ensured that all publications, scientific or otherwise were always reviewed for consistency before release. This is considered to have been most valuable and recommended for other projects of this type. Outside events seemingly unconnected with a project of this nature can have a significant influence on interest and opinions. One example is the occurrence of minor earth tremors often thought to be associated with oil and gas production operations. For future projects early consideration of the type of external events which could have an impact and the appropriate response is recommended.

The existence of the information centre at the site and the possibilities to see the equipment and activity first hand proved invaluable since they provide a real and immediate point of contact. Otherwise projects of this nature can appear rather faceless to the rest of the community. CCS was not an easy subject to make interesting. Presentation material which can help are items such as rock cores, the CO2 phase visualiser, items of drilling and casing hardware, timelapse representation of reservoir behaviour. CO2SINK took place in an area where underground operations were already established. The generally favourable response to the project so far may in part be due familiarity and may not be indicative of responses if CO2 geological storage is carried out at other locations.

A very important point to increase public acceptance and confidence was to show that the technical risks of CO2 storage can be managed with the help of a proper short and long term monitoring concept as well as appropriate mitigation technologies, e.g. adequate abandonment procedures for leaking wells. Examples for possible leakage scenarios helped the public to assess and accept the technical risks of CO2 storage at Ketzin.

Outlook

Leakage experiments will be essential to demonstrate that even worst case releases will affect a limited area of the environment and that the resulting damage will be quite low. Furthermore, leakage experiments can be used to verify detection limits of monitoring methods and to improve the emergency plans. To this end a new FP7 project “RISCS” is 57

CO2SINK Summary Technical Report in progress and a recent report for the UK Energy Institute summarises the state of knowledge for onshore CCS systems following a programme of experimental releases undertaken at UK Spadeadam test site in 2007. Development of leakage scenarios and simulation of leakage is of vital importance to increase credibility. Undefined fear seems to be the major risk in public acceptance of geological CO2 storage. Misinformation and missing communication further enhance resistance to geological storage of CO2. Another important issue is to show how benefits of CCS for the local people could be generated and how it fits in with concepts for sustainable use of energy. Up to now CCS technology is considered as obstructive to the development and supply of renewable energy by Non Governmental Organisations (NGOs), even though CCS can only be an interim technology for the fast reduction of CO2 emissions, due to finite fossil resources and limited geological storage sites. The public acceptance could be improved considerably, if CCS could be combined with development of renewable energy sources, like geothermal energy and biomass. In particular if a sufficient amount CO2 arising out of biomass utilisation could be stored geologically, this would offer a viable method for removing CO2 from the atmosphere which is becoming an increasingly likely need later this century.

CO2SINK Summary Technical Report Chapter 10

OUTLOOK

Michael Kühn1, Sonja Martens1

1GFZ German Research Centre for Geosciences, Telegrafenberg, 14473 Potsdam, Germany

CO2SINK Summary Technical Report

The CO2SINK project as well as national funded projects provided an excellent base for the set-up of a reliable infrastructure for CO2 injection and comprehensive research activities at the Ketzin site. Up to now, the Ketzin site is the only active CO2 storage site in Germany with a high local, national and international interest. Based on the outcome of the various Ketzin projects the aim is to continue and complement the CO2 injection as well as the monitoring activities at the Ketzin test site for a second period within national and international research projects.

Following the CO2SINK project and further national funded projects for Ketzin the aim is to continue and complement the research activities at the CO2 storage site for a second period. For this reason, the GFZ German Research Centre for Geosciences coordinates the two proposals CO2MAN and CO2CARE.

It is planned that the two projects CO2MAN (CO2 Reservoir Management) and CO2CARE (CO2 Site Closure Assessment Research) will succeed the CO2SINK project which ended in March 2010 and further national projects which funded R&D activities for Ketzin. The following provides an overview of the scientific content of both proposals.

The CO2MAN project

The CO2MAN project is fundet by the Federal Ministry of Education and Research (BMBF) and is organized in the following four parts:

Scientific Infrastructure All operational aspects of the CO2 storage are handled within this topic. It will provide the base for all other scientific work. Hence, the most important precondition for conducting further research at the Ketzin site is the continuation of the CO2 injection. It is the integral part e.g. of the planned monitoring concept and well abandonment under the presence of CO2.

Two new boreholes are planned for the Ketzin site (Fig. 11-1). The first drilling represents a shallow groundwater monitoring well into the Exter formation above the cap-rock of the storage reservoir. The second borehole (Ktzi 203) represents an appraisal drilling into the pressurized and already CO2 loaded storage reservoir.

CO2SINK Summary Technical Report

Figure 11-1. Planned (white) and existing (yellow) boreholes at the Ketzin site

Both boreholes will be implemented as slim holes. The well Ktzi 203 will be constructed in the reservoir section for the first time with fully cemented and corrosive resistant composite casing pipes and carrying on the outside several monitoring cables and sensors, like DTS (Distributed Temperature Sensing), seismometers as well as fibre- optical pressure / temperature point sensors. At the end of the CO2MAN project, Ktzi 203 shall be fully abandoned and back constructed under preservation of the down-hole sensor installations in order to be in a position to continue the monitoring until breakdown of the sensors and/or their monitoring transmission cables to the surface.

Geophysical Monitoring From the beginning of the CO2 injection, the monitoring of its propagation has been a critical part of the research activities in Ketzin. Thus this work package aims at continuation of the monitoring activities carried out so far.

The borehole monitoring will be continued in the three existing wells as well as in the planned new well (Ktzi 203). The measurements will comprise different methods in order to observe the CO2 propagation close to the wells. One focus will be on temperature measurements providing important base data to assess the phase of the CO2. Active heat pulse experiments will be performed and Pulsed Neutron Gamma and Induction Log runs will be interpreted providing evidence for CO2 saturation and flow processes in the reservoir horizon.

Active seismic measurements have been performed before the injection started (baseline) and were later repeated after the injection of a certain amount of CO2. The

CO2SINK Summary Technical Report evaluation of these repeat surveys has shown that seismic measurements, performed at different scales, provide a detailed image of the CO2 propagation in the reservoir. With the continued injection the CO2 plume will propagate further in the reservoir and repeat surveys, performed from the surface and from the monitoring wells, are necessary to update the image of the CO2 behaviour within the reservoir.

The geoelectric monitoring will continue the borehole measurements (Electrical Resistivity Tomography - ERT) and surface-borehole measurements in order to image the spatial distribution of electrical resistivity around the injection and monitoring wells. The surface-borehole measurements extend the observed volume to the region not covered by the wells. The combination of tomographic evaluation of geoelectric measurements and interpretation of the seismic measurements will contribute to an enhanced qualitative and quantitative understanding of the CO2 propagation in the reservoir.

Reservoir Processes Within this part the chemical and physical interactions between injected CO2, reservoir fluid, reservoir and cap rock, and the deep biosphere will be further studied. Aim is here twofold: first, it characterizes and quantifies the principal processes that are triggered by the injection of CO2; second, it monitors and surveillance of these processes. The main focus is on the study of natural fluid, gas, and rock samples from the Ketzin storage system. These studies on natural samples are complemented by laboratory and experimental studies in synthetic model systems.

Central to this work are core samples from the planned new wells Ktzi 203 and the monitoring well into the Exter Formation. Especially the core samples from Ktzi 203 provide the rather unique opportunity to study reservoir and cap rock samples from a storage site that have been exposed to CO2 under real storage conditions.

Modelling & Simulations The topic modelling and simulation aims at the investigation of coupled processes in the Stuttgart formation and its caprock by numerical modelling taking into account a coupling of hydrodynamics (e.g. multi-phase and multi-component flow), thermodynamics, hydro(geo)chemics and geomechanics.

The geological models will integrate all data from the available Ketzin wells and the different monitoring techniques to provide a holistic view of the Ketzin site.

The reservoir simulation and visualization part will compile and integrate all available data. This is the basis for an extensive parameterization of the numerical simulators. Here, inverse modeling and history matching methods will be used to determine the relevant reservoir parameters. Subsequently, the adapted models will be applied for prediction of CO2 distribution considering the scheduled injection. Visualization of all modeling results is planned during the entire project runtime for analysis and improvement of knowledge about reservoir processes in the Stuttgart formation.

The CO2CARE project

CO2SINK Summary Technical Report

The CO2CARE project is complementary to CO2MAN and is fundet by the EU (Work programme topics addressed: ENERGY.2010.5.2-3: CCS - site abandonment). The consortium of CO2CARE gathers 22 European and non-European partners from research and industry, all major and long experienced actors in research on CO2.

In contrast to CO2MAN which is focused solely on the Ketzin site, it is aimed within CO2CARE to study nine international storage sites which all differ in location (offshore vs. onshore) and geology. As shown in Figure 11-2, CO2CARE will incorporate up-to- date results and monitoring data from the main European and international CO2 injection sites including Ketzin.

Fig. 11-2 Field Site Portfolio with CO2CARE

CO2CARE is dedicated to abandonment, that is, to the evolution of the storage site behaviour after end of injection. Prediction of long-term behaviour is based on the observations from monitoring, experiments and numerical simulations performed during the storage life.

The research activities in CO2CARE are embedded in the regulatory frame of the EU Directive on CO2 Geological Storage (EU Directive). The fundamental regulatory principles surrounding site closure, transfer of responsibility (liability) to the competent authority, and post-closure obligations are set out in the EU Directive. Pursuant to Article 18, the overall philosophy of the Directive regarding site closure is enshrined in the three minimum criteria for transfer of liability:

 Observed behaviour of the injected CO2 conforms with the modelled behaviour;  No detectable leakage; 63

CO2SINK Summary Technical Report  Site is evolving towards a situation of long-term stability.

These conditions define satisfactory long-term site performance at a high-level. Significant gaps however remain in understanding how these high-level principles will be implemented at real sites, not only at the closure stage, but also at storage commencement, when future closure arrangements will need to be incorporated into a site storage license acceptable to both operator and regulator. The Directive does not define the specific technical acceptance criteria, based on real site performance data, which can demonstrate that a site meets the three requirements.

The identification of such criteria, and the development of site abandonment procedures and technologies which guarantee the fulfilment of these criteria are the main objectives of CO2CARE. Research in CO2CARE is organized in six work packages: (1) current Practices and regulatory framework, (2) well abandonment, (3) post-closure reservoir management, (4) risk management, (5) best practice and regulatory compliance in abandonment, and (6) dissemination.

For some of the sites, including Ketzin, a dry-run of abandonment will be carried out, that is, a realistic simulation of abandonment, including the fulfilment of all obligations of the mining authorities. Therefore, CO2CARE will be the first European experience with abandonment and will contribute to the development of the national legislations about CCS.