Geological Survey of Denmark and Greenland Bulletin 41, 2018, 13-16 13 This study is based on a feasibility study for the Danish En- erginet.dk to identify potential formations for brine storage near the gas storage facility at Lille Torup, northern Jylland, Denmark (Fig. 1; Hjuler et al. 2017). Located on top of a salt structure, the gas storage facility com- prises seven caverns, which have been washed out by circulating water in the salt dome. One cavern contains c. 520.000 m3 of intrusive brine that must be disposed of in order to increase the storage volume for gas. One option is to inject the brine into the subsurface if a target with appropriate storage properties can be identified, but it is a prerequisite that the stored brine does not compromise freshwater reservoirs. Due to cost considerations, the brine storage should be situated within a radius of 50 km of the gas storage facility and at a depth not exceeding 2000 m. Based on the national geothermal research conducted during the last decade, a number of sandy formations are considered potential storage reservoirs (Fig. 2; e.g. Mathiesen et al. 2009; Vosgerau et al. 2016). Around Lille Torup, these include the Bunter Sandstone/Skagerrak, Gassum, Haldager Sand and Frederikshavn formations where the two former formations are discarded due to present-day burial depths ex- ceeding 2000 m. In addition, the Chalk Group is considered a potential storage formation due to its importance as a hy- drocarbon reservoir in the North Sea, however, due to risk of leakage to the younger sediments and risk of environmental issues, the chalk was discarded as potential storage zone. Geological background The Lille Torup area is located centrally in the Danish Ba- sin, where the Upper Permian–Mesozoic succession is 5–5.5 km thick. The basin was formed in the Late Carbonifer- ous–Early Permian with basal Rotliegendes coarse-grained clastic sediments and thick Zechstein salts overlain by Trias- sic sandstone, mudstone, carbonate rocks and salt (Nielsen 2003). These are followed by Lower Jurassic mudstone, Mid- dle Jurassic sandstone, Upper Jurassic–Lower Cretaceous mudstone and siltstone with few sandstone layers. The Mes- ozoic succession terminates with c. 1200 m thick carbonate deposits. The salt structure at Lille Torup consists of mobilised Zechstein salt penetrating the Mesozoic succession. Its top point is c. 250 m below the present-day surface. The Haldager Sand Formation in the northern part of the basin is 2–150 m thick, but may exceed 200 m in rim syn- clines of salt structures, where sandstone commonly domi- nates the lithology. The Frederikshavn Formation is primar- ily present in the northern part of the basin and frequently includes sandstone layers. Its thickness decreases southwards from 150–300 m to a few metres. The more than 1000 m thick chalk- and limestone-dominated Chalk Group consti- tutes the topmost pre-Quaternary formation in large parts of the Danish Basin. Potential for brine storage near the gas storage facility at Lille Torup, northern Jylland, Denmark Morten Leth Hjuler, Morten Sparre Andersen, Carsten Møller Nielsen, Anders Mathiesen, Lars Kristensen, Nina Skaarup and Lars Henrik Nielsen R i n g k ø b i n g - F y n H i g hN o r t h G e r m a n B a s i n N o r w e g i a n - D a n i s h B a s i n 50 km Lille Torup gas storage facility Salt diapir Fault Selected storage area 2D seismic line Aarhus Silkeborg RandersViborg Hobro Skive Aalborg Holstebro 1 2 3 4 5 Erslev-1 Erslev-2 Hyllebjerg-1 Aars-1 Farsø-1 Kvols-1 Hobro-1 Gassum-1 Skive-1 Skive-2 Mors-1 Rødding-1 15 km Fig. 1. The study area and selected storage areas within a radius of 50 km of the Lille Torup gas storage facility. © 2018 GEUS. Geological Survey of Denmark and Greenland Bulletin 41, 13–16. Open access: www.geus.dk/bulletin http://www.geus.dk/bulletin 1414 Methods The subsurface within a radius of 50 km of the Lille Torup storage facility was screened for potential sandstone reser- voirs suitable for storage. The local database comprises 11 vertical deep wells and an open grid of regional 2D seismic profiles of variable quality and resolution (Fig. 1). Some wells were excluded from the database due to location on top of salt structures (Erslev-1–2 and Skive-1–2), uncertain data quality (Aars-1) or separation from the Lille Torup storage facility by fjord water (Mors-1). The Danish geothermal WebGIS application (Vosgerau et al. 2016) provided maps of formation depth, formation thickness and potential reservoir sandstone thickness as well as reservoir parameters of relevant wells. The potential reser- voir sandstone thickness map was developed for assessment of the geothermal potential and is used in this study as an indicator for injection capacity. Reservoir properties derived from well logs include the depths of formation top and base, formation thickness, gross sand thickness (i.e. cumulated thickness of all sandstone lay- ers), potential reservoir sandstone thickness (i.e. cumulated thickness of sandstone layers with a shale content <30% and a porosity >15%), as well as averaged values of porosity, permeability and transmissivity of the potential reservoir sandstone. For uncertainty considerations, see http://dybge- otermi.geus.dk/. Potential formations for brine storage Five potential storage areas were defined based on reservoir quality assessments obtained by integration of well-log data and WebGIS data improved with locally refined seismic in- terpretations. Injectivity assessments were performed using ECLIPSE 100 reservoir simulation software and Petrel soft- ware. See Hjuler et al. (2017) for details. The Haldager Sand Formation (Figs 3A–C) is presently buried more than 2000 m in large parts of the study area, but more shallow occurrences exist. The generally 50–150 m thick formation is dominated by sandstone known to be quartz-rich, which points to good reservoir properties. In ar- eas of relatively shallow burial (<2000 m), the Haldager Sand Formation may constitute a storage formation. The Frederikshavn Formation (Figs 3D–F) is buried less than 2000 m and generally more than 100 m thick; it is thickest east of Lille Torup. Several potential storage res- ervoirs with sufficient lateral extent can be identified. The Frederikshavn Formation constitutes a storage option. The Chalk Group (Figs 3G, H) is buried at 100–700 m depth and is more than 1 km thick. On top of the Lille Torup salt structure, the salt movements may have fractured the c. 250 m of chalk, and increased permeability and thus reser- voir quality. However, the overlaying Quaternary deposits are not expected to possess sealing qualities and brine storage in the chalk could lead to environmental issues. The Chalk Group is therefore discarded as a potential storage formation. Potential seals The Chalk Group outside the top of the Lille Torup salt structure is expected to effectively seal off pore water from the sandy formations beneath it due to its low permeability and great thickness. In addition, the clayey Lower Creta- ceous unit is assumed to be of sufficiently low permeability to prevent pressure and pore-water propagation from below. Reservoir parameters of the formations The Haldager Sand Formation mainly comprises sandstone layers with porosities in the 18–22% range and permeabili- ties in the 140–360 mD range (Table 1). Disregarding burial depth, the Haldager Sand Formation is assumed to provide suitable storage properties. The sandstone layers of the Frederikshavn Formation have porosities in the 17–30% range, permeabilities in the 110–1500 mD range and the thickness of potential reservoir sandstone in the 6–66 m range (Table 1). In the Kvols-1 well, however, the formation seems to have little or no storage po- tential, assumedly because clay minerals reduce both pore Quaternary System Lithostratigraphic unit Post Chalk Group Chalk Group Lower Cretaceous unit Reservoir Reservoir Reservoir Reservoir Assumed seal Assumed seal Assumed seal Frederikshavn Fm Børglum Fm Flyvbjerg Fm Haldager Sand Fm Fjerritslev Fm Gassum Fm Vinding Fm Oddesund Fm Tønder Fm Falster Fm Ørslev Fm Bunter Sandstone Fm Zechstein Group Cretaceous Jurassic Triassic Permian Fig. 2. Lithostratigraphic chart showing potential reservoirs (yellow) and assumed seals (brown). 15 space and permeability. The thickness of potential reservoir sandstone decreases from east to west (Fig. 3F). Suggested areas for brine storage The WebGIS-based maps (Fig. 3) are suitable for assessing reservoir quality trends on a regional scale of tens of kilo- metres, but not on a local scale of up to a few kilometres as uncertainties will be significant. Thus, the areas suggested for brine storage cover several square kilometres (Fig. 1). Area 1 includes the Hyllebjerg-1 and Farsø-1 wells (Figs 1, 3), in which the Haldager Sand Formation is evaluated to provide the better storage reservoir, with higher porosities and permeabilities than the Frederikshavn Formation (Ta- ble 1). The two formations offer two storage options within a narrow depth interval. Area 2 is situated above the rim syncline next to the salt structure beneath Lille Torup (Figs 1, 3), where the Frederiks- havn and Haldager Sand formations may be up to 300–400 m thick (Figs 3A, D) and include potential reservoir sand- stone units more than 15 m thick at depths shallower than 1400 m. However, existing seismic data are insufficient to confirm these thickness estimates and the formation depths may be closer to 2000 m due to deposition in the rim syn- m below mean sea level Thickness Thickness H al da ge r Sa nd F m Fr ed er ik sh av n Fm C ha lk G ro up Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 Hy-1 A-1F-1 K-1 Ho-1 G-1 M-1 R-1 S-1 S-2R-1 S-1 S-2 R-1 S-1 S-2R-1 S-1 S-2 R-1 S-1 S-2R-1 S-1 S-2 R-1 S-1 S-2R-1 S-1 S-2 R-1 S-1 S-2R-1 S-1 S-2 Above sea level ≥ 15 m 0–100 100–200 200–300 300–400 400–500 500–600 600–700 700–800 1200–1300 1300–1400 1400–1500 1500–1600 800–900 900–1000 1000–1100 1100–1200 Not present 0–50 m 50–100 100–150 150–200 200–250 250–300 300–350 350–400 400–450 450–500 500–550 550–600 600–650 > 700 650–700 Not present at 800–3000 m below mean sea level 50 km Lille Torup gas storage facility Salt diapir Fault Wells: Study area E-1 Erslev-1 E-2: Erslev-2 F-1: Farsø-1 G-1: Gassum-1 Ho-1: Hobro-1 Hy-1: Hyldebjerg-1 K-1: Kvols-1 M-1: Mors-1 R-1: Rødding-1 S-1: Skive-1 S-2: Skive-2 A-1: Aars-1 Depth to formation Formation thickness Potential reservoir sand thickness A B C D E F G H Fig. 3. WebGIS-generated maps of potential brine storage formations in the study area. Reservoir quality is assessed based on depth to formation top, formation thickness and potential reservoir sand thickness. 1616 cline. Thus, the estimated reservoir parameters are based on average values of the nearby wells, Hyllebjerg-1, Farsø-1, Ho- bro-1 and Kvols-1. Area 3 includes the Hobro-1 well (Figs 1, 3), in which the Haldager Sand and Frederikshavn formations are estimated to offer two storage options within a narrow depth interval, the Haldager Sand Formation providing the better reservoir (Table 1). Area 4 includes the Gassum-1 well (Figs 1, 3). The Fred- erikshavn Formation is more shallowly buried in this area (<1000 m; Fig. 3D) and offers the most excellent storage properties in the study area (Table 1). The Haldager Sand Formation is not present in Area 4 (Fig. 3A). Area 5 includes the Kvols-1 well (Figs 1, 3). This well in- dicates good reservoir properties of the Haldager Sand For- mation at a burial depth of 1940–1955 m, but also that the Frederikshavn Formation is a poor reservoir (Table 1). Assessment of injection rate and pore pressure Indicative injection rates (shown as a well injection index) and pore pressures for the Frederikshavn and Haldager Sand formations in the entire study area are presented in Table 1. Reservoir parameters are calculated as averages of all wells. The well injection index is assumed to correspond to the well production index; the pressure of the subsurface pore f luid in the formations is assumed to be hydrostatic. Conclusions The Frederikshavn and Haldager Sand formations consti- tute potential brine storage formations in the larger Lille Torup area and are sealed off by the Chalk Group and Lower Cretaceous unit. The Haldager Sand Formation offers the best reservoir properties but at greater depth than the Fred- erikshavn Formation. The well injectivity index and pore pressure indicate favourable conditions for brine storage. Five potential storage areas are suggested, compromising be- tween burial depth and distance from the gas storage facility. References Hjuler, M.L., Andersen, M.S., Nielsen, C.M., Mathiesen, A., Kristensen, L., Skaarup, N. & Nielsen, L.H. 2017: Potential for establishing an in- jection well for brine storage near the gas storage facility at Lille Torup. Danmarks og Grønlands Geologiske Undersøgelse Rapport 2017/20, 34 pp. Mathiesen, A., Kristensen, L., Bidstrup, T. & Nielsen, L.H. 2009: Vurde- ring af det geotermiske potentiale i Danmark. Danmarks og Grønlands Geologiske Undersøgelse Rapport 2009/59, 85 pp. Nielsen, L.H. 2003: Late Triassic – Jurassic development of the Danish Basin and the Fennoscandian Border Zone, southern Scandinavia. In: Ineson, J.R. & Surlyk, F. (eds): The Jurassic of Denmark and Greenland. Geological Survey of Denmark and Greenland Bulletin 1, 459–526. Vosgerau H., Mathiesen A., Andersen M.S., Boldreel L.O., Hjuler M.L., Kamla E., Kristensen L., Pedersen C.B., Pjetursson B. & Nielsen L.H. 2016: A WebGIS portal for exploration of deep geothermal energ y based on geological and geophysical data. Geological Survey of Den- mark and Greenland Bulletin 35, 23–26. Table 1. Reservoir parameters of the Haldager Sand and Frederikshavn formations in five areas suggested for brine storage Reservoir parameters based on the vertical wells in the suggested areas (Fig. 1). Reservoir parameters of area 2 based on average values of the Hyllebjerg-1, Farsø-1, Hobro-1 and Kvols-1 wells. For detailed descriptions of parameters and uncertainties, see http://dybgeotermi.geus.dk/. Area 1 Farsø-1 Haldager Sand 1934 1952 18 16 11 18 138 2 Hyllebjerg-1 Haldager Sand 1885 1894 9 9 9 22 344 3 Average Haldager Sand 1909 1923 14 12 10 20 241 3 Farsø-1 Frederikshavn 1689 1839 150 25 9 17 106 1 Hyllebjerg-1 Frederikshavn 1664 1810 146 37 15 19 188 3 Average Frederikshavn 1676 1824 148 31 12 18 147 2 Area 2 Average Haldager Sand 1900 1920 20 16 12 20 255 3 Average Frederikshavn 1700 1756 56 33 8 15 133 1 Area 3 Hobro-1 Haldager Sand 1852 1891 39 24 18 19 175 3 Hobro-1 Frederikshavn 1741 1806 65 12 6 19 238 1 Area 4 Gassum-1 Haldager Sand 1176 1178 2 2 2 19 188 0.5 Gassum-1 Frederikshavn 1053 1154 101 66 65 30 1500 97 Area 5 Kvols-1 Haldager Sand 1940 1955 15 15 9 21 363 3 Kvols-1 Frederikshavn 1856 1912 56 56 0 5 0 0 Well average Haldager Sand 1757 1774 17 13 10 20 242 2 9 Well average Frederikshavn 1601 1704 104 39 19 28 406 8 27 Well Formation Thickness Property To p be lo w m ea n se a le ve l ( m ) Ba se b el ow m ea n se a le ve l ( m ) Fo rm at io n th ic kn es s (m ) G ro ss s an d (m ) Po te nt ia l r es er vo ir s an ds to ne (m ) A ve ra ge p or os ity (m D ) A ve ra ge r es er vo ir pe rm ea bi lit y (m D ) Re se rv oi r tr an sm iss iv ity (m D ) W el l i nj ec tio n in de x (m 3 / ba r/ da y) Authors’ address Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen, Denmark. E-mail: MLH@geus.dk. mailto:MLH@geus.dk