Geological Survey of Denmark and Greenland Bulletin 38, 2017, 37-40 37 The Lower Palaeozoic succession in Scandinavia includes several excellent marine source rocks notably the Alum Shale, the Dicellograptus shale and the Rastrites Shale that have been targets for shale gas exploration since 2008. We here report on samples of these source rocks from cored shallow scientific wells in southern Sweden. The samples contain both free and sorbed hydrocarbon gases with con- centrations significantly above the background gas level. The gases consist of a mixture of thermogenic and bacteri- ally derived gas. The latter likely derives from both carbon- ate reduction and methyl fermentation processes. The pres- ence of both thermogenic and biogenic gas in the Lower Palaeozoic shales is in agreement with results from past and present exploration activities; thermogenic gas is a target in deeply buried, gas-mature shales in southernmost Sweden, Denmark and northern Poland, whereas biogenic gas is a target in shallow, immature-marginally mature shales in south central Sweden. We here document that biogenic gas signatures are present also in gas-mature shallow buried shales in Skåne in southernmost Sweden. In south central Sweden (Västergötland, Östergötland, Närke and Öland, Fig. 1), shallow (present burial <150 m) immature to marginally mature bituminous shale has been known for decades to contain gas and is currently under exploration (see summary in Schultz et al. 2015). Since 2009, many deep (>800 m) exploration drillings in north- ern Poland (including the Lebork S-1 well, Lehr & Keeley 2016), the Vendsyssel-1 well in Denmark (Ferrand et al. 2016) and the A3-1, B2-1, C4-1 wells in Skåne in south- ernmost Sweden (Pool et al. 2012), have demonstrated that the Lower Palaeozoic shale succession contains gas also in these areas (Fig. 1). In Denmark and Skåne, the average gas content is 30 ft3 gas per ton of rock in the organic- rich Alum Shale Formation (Pool et al. 2012; Ferrand et al. 2016). The equivalent shale formation in northern Poland contains up to 268 ft3 gas per ton of rock in the Lebork S-1 well (Lehr & Keeley 2016), which is comparable to the content within the core area of North American shale gas- producing formations (e.g. Jarvie 2012). In this study, we present empirical data on the compo- sition and isotope signatures of gas measured in shallow (<158 m) core samples from five scientific core drillings in southern Sweden, viz. Albjära-1, Lönstorp-1, Gislövs- hammar-2, Hällekis-1 and Djupvik-1 (Fig. 1, Table 1). Sampling and analyses were performed in 1991–1992 as part of the Energy Research Project EFP-1313/88-2 and the Pre-Westphalian Source-Rocks in Northwest Europe (PREW- SOR) project (Schovsbo & Laier 2012). This paper aims at further characterising the gas composition. Samples and analyses The molecular composition and isotope signatures of the occurring gases were measured in 27 core samples (Table 1). The samples were selected during drilling and consist of 4–8 cm long core intervals with a diameter of 5.5 cm. The samples from Albjära-1 and Lönstorp-1 were sealed in metal containers and stored at –18°C until they reached the laboratory for analysis. The free gas was subsequently ana- Generation and origin of natural gas in Lower Palaeozoic shales from southern Sweden Niels Hemmingsen Schovsbo and Arne Thorshøj Nielsen Djupvik-1 Albjära-1 A3-1 C4-1 Gislövshammar-2 Lebork S-1 Hällekis-1 B2-1 Vendsyssel-1 Lönstorp-1 Figur 1. One collumn wide Öland Baltic Sea Bornholm Sweden Denmark Skåne Västergötland Östergötland Närke Poland Lower Palaeozoic strata Caledonian front Exploration well Scientific borehole 200 km Fig. 1: Location of wells mentioned and occurrence of Lower Palaeo- zoic strata in southern Scandinavia. Modified from Nielsen & Schovsbo (2015). © 2017 GEUS. Geological Survey of Denmark and Greenland Bulletin 38, 37–40. Open access: www.geus.dk/publications/bull 3838 lysed by the Geological Survey of Denmark and Greenland (GEUS) by puncturing the containers through a septum before opening. Samples from Gislövshammar-2, Hälle- kis-1 and Djupvik-1 were analysed by the Federal Institute for Geosciences and Natural Resources (BGR), Germany. These samples were stored at –18°C at the drill site and subsequently transferred to a container filled with liquid nitrogen at –196°C. The free gas was measured by allowing the deep-frozen sample to equilibrate to room temperature in a sealed container. For all samples the sorbed gas in the rock matrix was liberated by treating the sample with phos- phoric acid following the procedure outlined by Faber & Stahl (1983). Gas composition The sorbed methane concentrations range from 118 to 23867 ppb (micrograms per kg of rock) and the free gas contents from 3 to 13420 ppm (Fig. 2). The concentrations thus far exceed the level of background gas of 20–50 ppb for methane as defined by Whiticar (1994). The gas content is strongly related to thermal rank of the sampled shales. The gas-mature samples from Gislövshammar-2, Albjära-1 and Lönstorp-1 have c. ten times higher yields than the thermal- ly immature samples from the Hällekis-1 and Djupvik-1 wells (Fig. 2). The gas content appears not to be related to the total organic content in the mature samples. This is ex- emplified by the fact that the highest yields of sorbed gas are found in the Rastrites Shale (TOC average <1%) and the lowest yields in the Alum Shale, which on average contains 9% TOC in Skåne and on Bornholm (Fig. 2). Isotope composition and carbon isotope signatures The gas molecule and isotopic compositions of the analysed samples plot within the bacterial to thermogenic fields in a Bernard diagram (Fig. 3A ). The microbial gas is charac- terised by much higher negative isotope values than seen in the thermogenic-sourced methane. This signature is most clearly expressed in the free gas samples from mature shales that also have 10 to 100 times higher C 1/(C2+C3) molecular ratios than the sorbed gas, which is also typical for biogenic gas (Fig. 3A). One column wide Alme. Ra st rit es Linde. 480 E. C . M a M .C am br ia n500 510 470 460 450 440 430 490 Si lu ri an Fu ro ng ia n Ea rl y O rd ov ic ia n M .O rd ov ic ia n La te O rd ov ic ia n Grey shaleLimestone Siltstone Black shale -80 -60 -40 δ13CMethane ‰ PDBTOC wt% 0 5 1510 Figure 2. BA A lu m S ha le G Dicello. Tø ye n Sh al e K B Gas: Sorbed Free Mature Immature 6 000 ppb 2 000 ppm 13 000 ppm 25 000 ppb Fig. 2. Stratigraphy of the Lower Palaeozoic shales. A: Total organic carbon (TOC) content in shales from the Skåne–Bornholm area (modi- fied from Schovsbo 2003). B: Methane isotope composition of sorbed and free gas from samples of thermally mature shale in the Albjära-1, Gislövshammar-2, and Lönstorp-1 wells and from samples of thermally immature shale in the Hällekis-1 and Djupvik-1 wells. E.C.: Early Cam- brian. G: Gislöv Fm. B: Bjørkåsholmen Fm. K: Komstad Limestone. Alme.: Almelund Shale. Dicello.: Dicellograptus shale. Linde.: Linde- gård Formation. Green area in B outlines the variation field defined by samples from the Djupvik-1 and Hällekis-1 wells. Table 1. Wells and analysed samples§ Well Formation N Depth range (m) Gislövshammar-2 Tøyen S. 1 19.8 Gislövshammar-2 Alum S. 4 31.8–88.3 Djupvik-1 Alum S: 1 2.0 Hällekis-1 Tøyen S. 2 9.9–17.3 Hällekis-1 Alum S, 4 23.4–39.8 Albjära-1 Almelund 1 99.6 Albjära-1 Tøyen S. 3 114.8–134.5 Albjära-1 Alum S. 5 139.6–157.0 §Full analytical results are available on request from the first author. 39 The propane versus ethane isotope gas signature is sug- gested to reflect the maturity of the source and this rela- tionship can be used for a gas-to-source-rock correlation (Whiticar 1994). The analysed gas signatures follow this prediction, as immature Alum Shale samples from the Hällekis-1 well plot with relatively depleted isotope signa- tures of propane and ethane compared to the thermally mature Alum Shale samples (Fig. 3B). Isotope data from Lithuanian and Polish oil and gas reservoirs in Middle Cambrian sandstone, sourced by Alum Shale (Kotarba & Lewan 2013), plot with intermediate signatures (Fig. 3B). The thermally mature samples (Gislövshammar-2 well), however, exhibit a considerable variation in propane iso- tope composition from –22 to –39‰ PDB, suggesting a mixture of differently derived gases. This may be caused by addition of bacterially derived ethane and/or biodegra- dation of a propane component (cf. Whiticar 1994). The thermally immature Alum Shale samples from the Djup- vik-1 well (marked with 0.32 %Gr in Fig. 3B) and sam- ples from Tøyen Shale from the Hällekis-1 well (marked as Tøyen 0.53%Gr in Fig. 3B) plot within an apparent gas maturity of 0.8–1.2% Ro according to the values indicated in Fig. 3B (Whiticar trend line), i.e. with much higher ma- turities than measured, suggesting that gas migration has occurred. In the Hällekis-1 well the migrated gas may have formed in response to intrusion of Permo-Carboniferous dikes that locally matured the shales in south central Swe- den (cf. Schultz et al. 2015). In the vicinity of the Djupvik-1 drill site-mature shale and igneous activity are unknown and the relatively high maturity remains puzzling. Deuterium isotopes Measurements of deuterium and carbon isotopes in meth- ane offer additional information on the source of the nat- ural gases and on the processes that may have modified their composition (Whiticar 1994). Figure 3C shows the deuterium versus carbon isotope composition of methane in the analysed samples. The gas composition exhibits a % Ro 0.1 1 10 100 1000 Thermogenic Bacterial Ke ro ge n ty pe II T yp e I II Microbial oxidation C 1 / ( C 2 + C 3) δ13CMethane (‰ PDB) -80 A -70 -60 -50 -40 -30 3.0 2.0 % Ro 1.5 0.7 0.5 Alum 0.53 % Gr Tøyen 0.53 % Gr 1.7 % Gr Poland–Lithuania 0.6–1.4 % Gr 0.32 % Gr 1.1 -20-30-40-50 -40 -36 -32 -28 -24 -20 δ1 3 C Pr op an e (‰ P D B) δ13CEthane (‰ PDB) B Bacterial carbonate reduction Geothermal Early mature thermogenic Mix and transitionBacterial methyl-type fermentation δ1 3 C M et ha ne (‰ P D B) δDMethane (‰ SMOW) -350 -300 -250 -200 -150 -1 0 0 -8 0 -6 0 -4 0 -2 0 C Whiticar (1994) General trend line Mature Offshore Poland–Lithuania sourced by Alum Shale (Kotorba & Lewan 2013) Immature Fig. 3. Compostion and isotope signatures of organic carbon. A: ‘Ber- nard diagram’ showing the ratio C1/(C2+C3) versus δ 13CMethane for all samples. This type of diagram is used to determine the origin of the gas- es and is modified from Whiticar (1994). For legend, see Fig. 2. B: Re- lationship between δ13CEthane and δ 13CPropane. Tøyen and Alum 0.53% Gr denote stratigraphy and maturity of samples from the Hällekis-1 well. The graptolite ref lectance (% Gr) is from Pedersen et al. (2013); Grönvik locality is used for Djupvik-1; estimate for offshore Poland. C: δD and δ13C for methane. For legend, see Fig. 2. The isotope signatures of the various sources are from Whiticar (1994). 4040 large degree of scatter, but free gas from thermally mature samples plots in the bacterial carbonate reduction field, whereas free gas from immature samples plots towards the bacterial methyl-type fermentation fields, suggesting that different processes generated the depleted methane isotope compositions (Fig. 3C). The sorbed gas compositions in general plot away from bacterial sources, indicating that a thermogenic signature is preserved (Fig. 3C). Implications for shale-gas prospectivity The gas-isotope composition suggests that significant post- generation modification occurred although the timing is un- known. The biogenic isotope signature of the gases in Skåne resembles similar signatures seen in Östergötland (Schultz et al. 2015). Here Schultz et al. inferred that methyl-ferment- ing processes contributed to the methane content. Howev- er, the gas-isotope signature was not as depleted as seen in this study, possibly owing to the mixed shale oil – biogenic nature of the Östergötland Alum Shale play. According to Schultz et al. (2015) the biogenic gas was generated after the Pleistocene glaciation, as modern meteoric water was able to infiltrate the shale and create the right conditions for bacterial activities. We envisage that similar conditions may have affected the shallowly buried shales in Skåne. Krüger et al. (2014), however, show that highly mature kerogen has a much smaller microbial generative gas potential than imma- ture to marginally mature kerogen, since thermal maturity limits the amount of easily biodegradable organic matter that can be transformed to methane. Conclusions Lower Palaeozoic shales in south central Sweden and southernmost Sweden contain natural gas that exceeds the level of background gas. The gas content is strongly related to the thermal rank of the sampled shales, and mature sam- ples have approximately ten times higher yields than the immature samples. The gas is generated by both thermo- genic and bacterial processes. The microbial gas signature is most clearly expressed in the free gas samples from ma- ture shales that also have 10 to 100 times higher molecule C 1/(C2+C3) ratios than the sorbed gas. Migration may have occurred related to gas formation in response to intrusion of Permo-Carboniferous dikes that locally matured the shales in south central Sweden. Acknowledgements We thank Troels Laier (GEUS) for comments and suggestions to an earlier version of the manuscript. The authors wish to thank Mikael Erlström and Maciej Kotarba for constructive comments that improved the paper. References Faber, E. & Stahl, W. 1983: Analytic procedure and results of an isotope geochemical surface survey in an area of the British North Sea. Geo- logical Society Special Publications (London) 12, 51–63. Ferrand, J., Demars, C. & Allache, F. 2016: Denmark – L1/10 Li- cence relinquishment recommendations report. Total E&P, Memo 1-9. Available from: http://www.ft.dk/samling/20151/almdel/efk/bi- lag/353/1651289.pdf. Verified 7.4.2017. Jarvie, D.M. 2012: Shale resource systems for oil and gas: Part 1 – Shale- gas resource systems. A APG Memoir 97, 69–87. Kotarba, M.J. & Lewan, M.D. 2013: Sources of natural gases in Middle Cambrian reservoirs in Polish and Lithuanian Baltic Basin as deter- mined by stable isotopes and hydrous pyrolysis of Lower Palaeozoic source rocks. Chemical Geology 345, 62–76. Krüger, M., van Berk, W., Arning, E.T., Jiménez, N., Schovsbo, N.H., Straaten, N. & Schulz, H.-M. 2014: The biogenic methane potential of European gas shale analogues: Results from incubation experi- ments and thermodynamic modelling. International Journal of Coal Geology 136, 59–74. Lehr, J.H. & Keeley, J. 2016: Alternative energy and shale gas encyclo- pedia. 912 pp. John Wiley & Sons. Nielsen, A.T. & Schovsbo, N.H. 2015: The regressive Early – Mid Cam- brian ‘Hawke Bay Event’ in Baltoscandia: Epeirogenic uplift in con- cert with eustasy. Earth Science Reviews 151, 288–350. Petersen, H.I., Schovsbo, N.H. & Nielsen, A.T. 2013: Ref lectance meas- urements of zooclasts and solid bitumen in Lower Palaeozoic shales, southern Scandinavia: correlation to vitrinite ref lectance. Interna- tional Journal of Coal Petrology 114, 1–18. Pool, W., Geluk, M., Abels, J. & Tiley, G. 2012: Assessment of an unusual European Shale Gas play – The Cambro–Ordovician Alum Shale, southern Sweden: Proceedings of the Society of Petroleum Engineers/European Association of Geoscientists and Engineers Un- conventional Resources Conference, Vienna, Austria, 20–22 March, 2012, 152339. Schovsbo, N.H. 2003: The geochemistry of Lower Palaeozoic sediments deposited on the margins of Baltica. Bulletin of the Geological Soci- ety of Denmark 50, 11–27. Schovsbo, N.H. & Laier, T. 2012: Composition and gas isotope signa- ture of shale samples from 5 scientific wells in Sweden. Geological Survey of Denmark and Greenland Report 2012/17, 1–25. Schulz, H.-M., Biermann, S., van Berk, W., Krüger, M., Straaten, N., Bechtel, A., Wirth, R., Lüders, V., Schovsbo, N.H. & Crabtree, S. 2015: From shale oil to biogenic shale gas: retracing organic-inorgan- ic interactions in the Alum Shale (Furongian-Lower Ordovician) in southern Sweden. A APG Bulletin 99, 927–956. Whiticar, M.J. 1994: Correlation of Natural Gases with their Sources. A APG Memoir 60, 261–283. Authors’ addresses N.H.S., Geological Survey of Denmark and Greenland (GEUS), Øster Voldgade 10, DK-1350 Copenhagen K, Denmark; E-mail: nsc@geus.dk A.T.N., Department of Geosciences and Natural Resource Management, University of Copenhagen. Øster Voldgade 10, DK-1350 Copenhagen K, DK.