Vol. 48, 01, 05ok.qxd 19 ANNALS OF GEOPHYSICS, VOL. 48, N. 1, February 2005 Key words Eger Rift – French Massif Central – Eifel – carbon dioxide – gas fractionation – isotope composition 1. Introduction In areas with both low CO2 abundance and contents in the gas phase as for example in complexes of crystalline basement or in areas with younger volcanic activity, one frequently encounters CO2 with isotopic compositions of δ13C values lesser than –10 ‰. In many cases, such values are interpreted as biogenic/organic CO2 or mixtures between magmatic and bio- genic CO2 or respectively mantle and crustal end members (e.g., Griesshaber et al., 1992). It appears that CO2 gases with low δ 13C values oc- curring in the margin areas of regions with as- cending magmatic CO2 as for example in min- eral springs of the western Eger Rift, the Eifel or the French Massif Central (Batard et al., 1982) confirm this interpretation. However, due to the high solubility of CO2 in water and the HCO3 formation, fractionations of the CO2-rich gases take place (e.g., Batard et al., 1982; Capasso et al., 1997; D’Alesandro et al., 1997; Chiodini et al., 1999; Weinlich et al., 1999). Commonly, the isotopic data are compared and calculated with a single equilibration. However, in the peripheral areas of regions with magmatic CO2 marked by longer migration Mailing address: Dr. Falk H. Weinlich, Referat Gas- und Isotopengeochemie, Bundesanstalt für Geowissen- schaften und Rohstoffe (BGR), Stilleweg 2, 30655 Han- nover, Germany; e-mail: falkweinlich@gmx.de Isotopically light carbon dioxide in nitrogen rich gases: the gas distribution pattern in the French Massif Central, the Eifel and the western Eger Rift Falk H. Weinlich Bundesanstalt für Geowissenschaften und Rohstoffe (BGR), Hannover, Germany Abstract Based on characteristics of the distribution pattern of the western Eger Rift spring gases, a distribution pattern is presented for the gases of the French Massif Central. The central parts of these areas with ascending magmatic CO2 are characterised by high gas fluxes, high CO2 contents of up to 99.99 vol% and isotopially heavy CO2. In the peripheries, the decrease of δ 13C values of CO2 and CO2 contents in the gas phase is compensated by a rise in N2 contents. It can be demonstrated that gas fractionation in contrary to mixtures with isotopically light biogenic or crustal CO2 controls the distribution pattern of gas composition and isotopic composition of CO2 in these spring gases. Dissolution of CO2 results in formation of HCO3 – causing isotope fractionation of CO2 and an enrichment of N2 in the gas phase. With multiple equilibrations, values of about –17 ‰ or lower are obtained. The scale of gas alteration depends on the gas flux and the gas-water ratios respectively and can result in N2-rich gases. Es- sential for the interpretation are gas flux measurements with mass balances derived for most of the springs. With- out such mass balances it is not possible to discriminate between mixture and fractionation. The processes of iso- topic and chemical solubility fractionations evidently control the gas distribution pattern in other regions as well. 20 Falk H. Weinlich pathways, it can be assumed that gases migrate in different water systems, for example along various faults where dissolution of CO2 and fractionation with concomitantly formed HCO3 – can take place several times. This results in a comprehensive gas fractionation concerning both gas composition and isotopic composition of CO2 and can yield N2-rich gases. 2. Gas distribution pattern in European areas with magmatic CO2 2.1. Western Eger Rift (Czech Republic) In the western Eger Rift the main release of magmatic CO2-rich gases is bound to gas re- lease-centres in the Cheb Basin and Karlovy Vary north and Mariánské Lázně south of the main structures of Eger Rift (Weinlich et al., 1998). These structures of the Krušné hory (Erzgebirge) main fault together with the central fault both dipping to south and the Litoměřice deep fault dipping to north form a Y-structure. This Y-structure splits the gas flux and forms a shielded gas free zone within the Eger Rift. The gases of the mofettes and springs in these areas with the highest gas fluxes are characterised by high CO2 contents of up to 99.99 vol% and δ 13C values of –3.9 up to –1.9 ‰ (fig. 1). In some mofettes and springs the gas flux reach values up to 28 m3/h in Bublák and 35 m3/h in Soos (Cheb Basin) or about 100 m3/h in the Mariiny mofette in Mariánské Lázně. The magmatic nature of these gases is indicated, besides these δ 13C values, by high proportions of mantle-derived helium with R/Ra values up to 6 (Weinlich et al., 1999). With increasing distances from these gas re- lease-centres the gas flux falls and related to it the CO2 becomes isotopically lighter whereas the CO2 contents decrease and are compensated by a rise of the N2 and He contents. To the east and south of Konstantinovy Lázně the CO2 contents drop linked with gas fluxes of lower than a half l/h up to values of 87 and 67 vol%, respectively. To the north of the Eger Rift nitrogen contents of 98 vol% are attained apart from 0.7 vol% CO2 as in the Schönbrunn fluorite mine or in other spring gases in the Erzgebirge (Weinlich, 1989). Linked with the decrease of the CO2 contents in the gas phase is a decrease of the δ13C values of CO2. South of the Eger Rift, the δ13C values fall to –7.4 ‰ in the Konstantinovy Lázně area and to –8.8 ‰ in Ba- varia, respectively. In the same way, the δ13C val- ues of CO2 decrease north of the Eger Rift down to – 6.0 ‰ in Bad Elster and in the N2-rich gases of Schönbrunn down to a value of –17.4 ‰. The gases in the mofettes and mineral springs in the western Eger Rift migrate up- wards along faults in the area with exposed metamorphic rocks of Variscian in the north and of Moldanubian consolidated basement in the south or with exposed Variscan granite in- trusions. The metamorphic rocks in this area have very low potential if any for a CO2 release because all organic carbon is fixed thermody- namically as very stable graphite. In the course of the Variscan, metamorphism mobilised and displaced organic carbon with δ 13C values of – 14 ‰ is present in the form of CO2 gases among others in the fluid inclusions of granitic quartz in Schönbrunn. With a decrepitation temperature of 800°C (Weise et al., 2001), the release of CO2 gases by mineral waters is today hardly possible. 2.2. Eifel (Lower Rhine Graben, Germany) CO2-rich gases linked with Quaternary vol- canic activities occur also in the Eifel. Griess- haber et al. (1992) report isotopic compositions of CO2 ranging from –7.8 ‰ up to –3 ‰. The lighter isotopic values are explained with the aid of lower R/Ra values due to mixing of magmat- ic and biogenic CO2. May (2002) describes the occurrence of CO2-rich gases linked with higher gas fluxes within the central West Eifel and the decrease of CO2 contents in its margin areas (fig. 2). So gases with up to 98.3 vol% CO2 predominate for e.g., in Wallenborn in the central part of the West Eifel and with up to 99 vol% in Laach lake in the East Eifel. The δ 13C values of CO2 in the West Eifel range from –5.7 up to –2.0 ‰ (Hub- berten, 2004, pers. comm.) and in the central parts of the East Eifel from –5.1 up to –3 ‰ (Griesshaber et al., 1992). In contrast, away from these areas with high gas fluxes, gases with higher N2 contents occur 21 Isotopically light carbon dioxide in nitrogen rich gases Fig. 1. Distribution pattern of gas flux (free gas), gas composition (air-free) and δ 13C CO2 values of gases in mineral springs and mofettes in the western Eger Rift, Czech Republic (data taken from Weinlich et al., 1998, 1999, 2003). Legend: ML – Mariánské Lázně; FL – Františkovy Lázně; half-filled squares – uranium mines with gas blowouts. Fig. 2. Gas flux distribution (moving average for circles with 8.6 km diameter), isotope composition of CO2 and schematic profile of CO2 flux, spring density and gas composition across the West Eifel, after May (2002). Carbon iso- tope data from Griesshaber et al. (1992) and Hubberten (2004, pers. comm.). The Brubble spring of Wallenborn (wallen – seethe, born – spring) is a geyser-like spring with gas release of 11.3 m3 per eruption period (May, 2002) the spring with the highest gas flux and the highest δ 13CCO2 value (–2.0 ‰) in the West Eifel. 22 Falk H. Weinlich in the spring gases of Aachen (27 vol% N2, 72 vol% CO2) located to the northwest of Eifel or in the southeast at Bad Bertrich (90 vol% N2, 6.2 vol% CO2) and Bad Wildstein (98.5 vol% N2, 0.13 vol% CO2) (May, 2002). Congruent with the alteration of gas composition Beyer (1995) reports δ13C values of CO2 ranging from –10.2 up to – 6.3 ‰ for the thermal springs of Aachen. In the N2-rich gas of Bad Bertrich, δ 13C value of CO2 amounts to –13.9 ‰ (own analysis). The δ13C values of CO2 in these marginal springs are significantly lower than in the central parts. The gases ascent along faults in an area with exposed Devonian sediments and are occur mainly in the areas of the Maar-type Quater- nary volcanism in the East and West Eifel. Fig. 3. Distribution of gas flux (free gas), gas composition and δ 13C CO2 values of the gases of the mineral springs of the French Massif Central. Compiled after data from Moureu and Lepape (1912), Moureu (1923), Baubron et al. (1979), Schoeller and Schoeller (1979), Batard et al. (1982) and Matthews et al. (1987). Based on the hydrologic map (Risler et al., 1973) and tectonic map (Autran et al., 1980). Fig. 4. Rise of N2 (air-free) in gases with decreasing free gas-water ratio due to increased gas fractiona- tion, i.e. selective CO2 solution in water. The un- avoidable scattering in the data is due to varying air proportions in the gases by different partial pressures influencing the bubble point pressure of the gas-water systems (data from Weinlich et al., 1998, 2003). 23 Isotopically light carbon dioxide in nitrogen rich gases 2.3. French Massif Central In the French Massif Central CO2 occurs with δ13C values ranging from –23 ‰ up to – 4 ‰. The CO2-rich gases of Vichy, Royat or Mont Dore and Cezallier (>99 vol% CO2) are bound to the area of Limagne depression or its direct vicin- ity. Matthews et al. (1987) proved the mantle-de- rived nature of these gases with R/Ra of up to 5.5. Gas flux measurements (Moureu and Lepape, 1912; Batard et al., 1982) carried out on these spring gases in the above region also show that the CO2-rich gases are linked with high gas fluxes. According to the gas composition and isotope data by Moureu (1923), Schoeller and Schoeller (1979), Baubron et al. (1979), Batard et al. (1982) and Matthews et al. (1987) a similar dis- tribution pattern also prevails there. In the central part of the Massif Central isotopically heavy CO2 with δ 13C values of –7 up to –5 ‰ occurs linked exclusively with CO2-rich gases whereas isotopi- cally lighter CO2 with δ 13C values ranging from –23 up to –12 ‰ is linked with N2-rich gases of Evaux (89-93 vol%), Sail-Les-Bains (97.2 vol%) Maizières (88.2 vol%) or Santenay (86-88 vol% N2) in the margin areas (fig. 3). The latter values are in the range of «typical» biogenic CO2. Apart from the Limagne depression filled with Oligocene – Quaternary sediments where the spring gases migrate upward along margin- al faults and the volcanic complex of Mont Dore these spring gases migrate upward along faults in areas with exposed Variscian metamor- phic rocks or granites. 3. CO2 fractionation 3.1. The gases in the western Eger Rift In contrast to other gases, CO2 is very vul- nerable to fractionation processes. Firstly, due to its good solubility in water compared to N2, HC’s and rare gases, the gas composition can be altered solely by solubility fractionations. This results in enrichment of the inert gases as ob- served by an aureole of N2-richer gases in the surroundings of all regions with CO2-rich mag- matic gases in Europe. Figure 4 demonstrates that with an ongoing solution of CO2 resulting in a decrease of the gas/water ratios the gases in the western Eger Rift are enriched in N2 in the gas phase (Weinlich et al., 1998). Secondly, linked with the solution of CO2 are decreasing pH values of these waters. This results in leaching of cations from the adjacent rocks and formation of HCO3 – ions. Between the newly formed HCO3 – and CO2 in the gas phase exists an isotope fractionation of about 10 ‰ (at 10°C) (Wendt, 1968). With ongoing HCO3 – formation, the remain- ing CO2 in the gas phase becomes isotopically lighter. Consequently, the decreasing gas flux correlates with decreasing δ 13C value of CO2 in the gas phase as demonstrated in the distribu- tion pattern for the Eger Rift gases. However, isotopically lighter CO2 can be also a result of mixing of biogenic and magmatic CO2. The key for discrimination between mixing or fractionation is the compilation of complete mass balances of CO2 for each mineral spring with a 24 Falk H. Weinlich scribed by the following equation: m m m m C C wdiss gas HCO gas 13 13 gas total total total HCO3 3 = - + - d d f f _ ` i j free gas phase consisting of gas flux, isotopic and chemical gas composition, contents of HCO3 – and dissolved CO2 contents and water discharge. According to Wendt (1968) the isotope bal- ance for CO2 in the free gas phase can be de- Fig. 5. Dependence of δ 13C CO2 values in free gas phase from the ratio of HCO3 – transport (mHCO3) in water to total CO2 flux in mineral springs and mofettes in the western Eger Rift with complete mass balances, according to the isotope balance formula (see text) (data of HCO3 –, dissolved CO2 and water discharge are taken from Kolářová and Myslil, 1979 and Weinlich et al., 1999, 2003). Lines starting from the scattering range of the dry gas vents (on the y axis) in the mofettes wrap the field of theoretical fractionation according to the fractionation factor ε HCO3-gas at 10°C. Between these lines, the δ 13C values for the free gas are exclusively a result of frac- tionation by the means of formed HCO3 – during a single equilibration, without the necessity to assume an addi- tional biogenic carbon. The δ 13C values below the lines can be explained by twice equilibration. This results in an increase of N2. Multiple equilibrations with solely dissolved CO2 and single equilibration with HCO3 – result in occurrence of N2-richer gases, which fall in between the fractionation lines. Table I. Measured and calculated δ 13C values of CO2 for springs in the vicinity of mofettes in North Bohemia demonstrate that the differences between mofettes and springs are solely caused by HCO3 --fractionation. For the δ 13C primary (total) – value the isotope signature of the neighbouring mofettes, the HCO3 - as well as dissolved CO2 and the water discharge of the respective spring were used according to isotope balance formula (data from Weinlich et al., 1998, 1999). Locality, spring CO2 Gas flux Water CO2 - flux δ 13C CO2 discharge Free gas Dissolved HCO3 − Total Measured Calculated vol% l/h l/h mol/h mol/h mol/h mol/h ‰ ‰ Cheb Basin Soos, mofettes 99.946 21100 0 941.46 0 0 941.46 −− 2.9 Soos, Cisař sý 99.941 7600 2520 339.08 143.18 60.7 542.96 − 3.6 − 3.64 Františkovy Lázně Kostelní spring 99.017 2500 8600 110.51 418.08 84.31 612.90 − 3.6 − 3.34 Mariánské Lázně Smrad’och mofettes 99.923 5200 0 231.96 0 0 231.96 −− 2.27 Farská spring 99.610 149 1166 6.63 67.73 11.55 85.91 − 2.8 − 2.55 Mariiny mofette 99.990 87000 0 3883.54 0 0 3883.54 −− 2.7 Ferdinand spring II 99.969 8240 1440 367.74 70.85 71.86 510.45 − 4.0 − 3.88 Kř ížový spring III 99.625 135.6 72 6.03 4.74 3.27 14.04 − 3.9 − 4.52 Martinov 99.280 4 90 0.18 5.28 1.06 6.52 − 3.5 − 3.22 Chotěnov 85.570 4.1 171.1 0.16 8.54 2.35 11.05 − 4.0 − 3.75 Dolní Kramolín 99.370 164 324 7.28 19.29 1.58 28.15 − 2.6 − 2.35 25 Isotopically light carbon dioxide in nitrogen rich gases where m is the amount of CO2 and ε the fraction- ation factors of –1.3 ‰ for ε diss-gas and 9.6 ‰ for ε HCO3-gas at 10°C (Wendt, 1968; Mook et al., 1974; Zhang et al., 1995). In case of the mofettes the water discharge is 0, i.e. the ratios mdiss /mtotal and mHCO3/mtotal are 0. Consequently, the measured δ13C value of the free gas phase is identical to the total, i.e. the primary isotopic composition of magmatic CO2 in this area. Therefore mHCO3/mtotal ratios near 0 can be used to identify the primary isotopic composition. In the case of mineral springs with a continu- ous transport of leached cations, mainly Ca++ and Mg++, isotopic fractionation occurs with contem- poraneously formed HCO3 – whereas remaining CO2 in the gas phase becomes isotopically lighter and CO2 contents can decrease. Figure 5 exhibits this dependency of the δ13C value of CO2 in the gas phase from the mHCO3/mtotal ratios in the springs. In cases of spring gases in the close vicinity of mofettes it can be demonstrated that the differences in the isotopic composition are solely caused by HCO3 – fractionation (table I). The spring gases, which are further away from the main gas release-centres, can be trans- ported within more than one fault system and thus in different waters. Therefore, the equili- bration between CO2 in the gas phase and HCO3 – can occur several times and the calculat- ed δ 13C CO2-values are consequently higher than expected in a single equilibration. Changes in isotopic composition can also be explained by mixing with lighter biogenic CO2. However, the common change in iso- topic and chemical composition (fig. 6) points to fractionation processes caused by multiple equilibrations as increasing nitrogen contents are not linked with biogenic CO2 admixtures. In the gases of the western Eger Rift a corre- lation between N2 and He contents can be ob- served (fig. 7) which indicates one source and 26 Falk H. Weinlich Fig. 6. Plot of N2 content versus δ 13C values displays the common variations in gas and isotope composition of gases from the Eger Rift area, e.g., from the Cheb Basin/South Vogtland area (CB-SV), Konstantinovy Lázně area (KL) and Bavaria (BY). These variations are caused by fractionation (CO2 solution and HCO3 – formation). In the case of waters with low TDS contents and without HCO3 – formation, the fractionation can only take place with dissolved CO2 and the isotopic heavy CO2 remains therefore in the gas phase. Fig. 7. Correlation between nitrogen and helium contents in gases of the western Eger Rift (data from Wein- lich et al., 1998). a continuous enrichment by solubility frac- tionation (Weinlich et al., 1998). The assump- tion of an additional N2-source in case of the N2-richer gases is therefore not necessary. Compared to the gas release centres the nitro- gen flux decreases in the springs in the mar- ginal areas. An exception is the water inflow about 500 m below the surface with 100 l/h N2 in the Schönbrunn fluorite mine, caused by pressure release in this mine. A special case of these fractionations are springs with very low Ca-Mg-HCO3 contents and with isotopic composition of CO2 being still nearly unchanged due to lack of extensive HCO3 – fractionation and where the CO2 contents are solely decreased by the solution of CO2. This can be explained by fractionation processes. Howev- er, in case of an interpretation of isotopically lighter CO2 in the gas distribution pattern of the western Eger Rift caused by mixing with lighter biogenic CO2 there should be no reason to eluci- date why the mixing should not occur in springs with low Total Dissolved Solids (TDS) contents. Regarding longer migration pathways as mentioned above it is considered that the gas migration occurs within different hydrological systems and therefore these fractionations can take place repeatedly during the migration. This results in a drastic drop of the CO2 contents and the δ13C values in the remaining gas phase. Figure 5 demonstrates the isotopic composition of the N2-rich gases in the Fluorite mine of Schön- brunn (one of the most northern springs shown in fig. 1) with a δ 13C value of –17.4 ‰ which can be explained alone under the assumption that the CO2-HCO3 – system is equilibrated twice. Certainly, an admixture of biogenic CO2 can- not be excluded but in this mine about 3.6 m3 of CO2 gas and 3664 m3 of dissolved CO2 per year were released. Facing the fact that the granite sur- face is located only about 650 m below both the thermal water and gas inflows (Kuschka and Hahn, 1996) this amount is hardly explainable with noticeable proportions of biogenic CO2. It is problematic to derive the nitrogen from crustal sources in terms of the 100 l/h N2 in the free gas phase and about 450 l/h dissolved N2 (air-free over dissolved Ar; procedure in Weinlich et al., 1998) accompanied by 0.45 l/h He and 5.72 l/h dissolved He (in total 54 m3/yr He). It should be considered that due to the intrusion of Variscian granites the metamorphic rocks were exposed to far higher temperatures, as is the case today. Therefore, the N2 is probably also mantle derived, because the N2 gas release including metamor- phic CO2 sourced from these crustal rocks took place during the Variscian intrusions. The nitro- gen isotope composition with δ 15N 0.7 (Weinlich et al., 1999) exhibits a tendency to more positive values of a plume-like mantle (Marty and Dauphas, 2003) which occurs in Central Europe (Wilson and Downes, 1992). Recalculating the CO2 and N2 in the gas phase together with the dissolved N2, and CO2 including the HCO3 – the whole fluid system con- tains about 90 vol% CO2 (possible CaCO3 pre- cipitations would additionally increase this CO2 content) and about 10 vol% N2 and thus this sys- tem is comparable with the Eger Rift gases. 3.2. The gases of the French Massif Central The distribution pattern and the isotopic signature of – 23 up to –12 ‰ of the N2-rich spring gases of the French Massif Central could be explained in the same way. As in the western Eger Rift, the CO2-rich gases with δ13C values of –7 up to – 5 ‰ are linked with high gas flux- es in the concerned mineral springs. Figure 8 demonstrates this correlation between δ13C val- ues and the gas composition. Batard et al. (1982) calculated initial isotope composition for some gases in this area with mass balances according to a single equilibration CO2-HCO3 – . The authors concluded a biogenic or mixed ori- gin for the CO2 because the isotopic composi- tion of the total carbon ranges between –16 and –11 ‰. However, due to multiple equilibra- tions, it cannot be excluded a priori but only the mHCO3/mtotal ratios near 0 should be used to avoid misinterpretations (the calculated δ13 C value of CO2 for Schönbrunn assuming a single equilibration is also –11 ‰). For the gases of the French Massif Central, it can be assumed that the gas of Royat with a δ 13C value of –6.4 ‰ is unfractionated owing to its mHCO3/mtotal ratio of 0.007 and displays the primary composition. Further, geothermometer calculations (Pauwels et al., 1997) indicate that 27 Isotopically light carbon dioxide in nitrogen rich gases 28 Falk H. Weinlich Fig. 8. Correlation of gas composition and isotopic composition of CO2 in gases of the Massif Central (data from Batard et al., 1982 and Matthews et al., 1987). Fig. 9. Mass balance plot displays δ 13C values versus ratio HCO 3 – transport (mHCO3) to total CO2 flux (m0) for gases of mineral springs in the French Massif Central (data from Batard et al., 1982). these waters emerging in the Mont Dore area reach temperatures of about 100-130°C at depth and in Saint-Nectaire (δ 13C value CO2 –7.0 ‰) temperatures of 160-175°C at depth respectively. According to Mook et al. (1974), the fractionation between CO2 and HCO3 – at temperatures of around 120°C is zero. Based on the isotope composition of Royat it can be shown that the low values of –12 and –23 ‰ can be reached (fig. 9) under the assumption Fig. 10. Long-term observation of isotopic composition of the gas of the Eisen spring in Bad Brambach, Eger Rift (Weise et al., 2002) and typical annual changes in the CO2 production in soils (Andrews and Schlesinger, 2001). Growing seasons – light grew. 29 Isotopically light carbon dioxide in nitrogen rich gases that the equilibration between CO2 and HCO3 – takes place only twice and in two cases three times. Therefore, it is not absolutely necessary to assume biogenic contributions in the region as well as in the Eger Rift. Just two isotope val- ues, which represent gases of Santenay and Saint Honoré, situated at the edge of the Mor- van horst and which are associated with Na- SO4 waters are lighter than those lying in the field of gases equilibrated three times. Howev- er, an uncertainty of these mass balances lies within the possibility of influence of non-min- eralised groundwaters and/or mineral waters of different type, which «dilute» the HCO3 – -rich mineral water in the respective springs. In the Cézallier area, Négrel et al. (2000) demonstrate such mixtures of mineral waters with meteoric and different mineralised waters in line with the REE distribution and strontium isotope ratios. Pauwels et al. (1997) state simi- lar processes in the Mont Dore region on the basis of the main element distribution in spring waters. This effect of «dilution» of these miner- al waters can produce lower mHCO3/mtotal ratios, present during equilibration in deeper regions as higher mHCO3/mtotal ratios. Since the CO2 of these N2-rich gases is completely fixed as HCO3 – , it is no longer possible to form new HCO3 – in these waters. The δ 13C values were al- tered by the formed HCO3 – and remained un- changed in less mineralised waters. This results in a shift of these gases in plot of the ratio of HCO3 – versus total CO2 flow (fig. 9) and acts as if a thrice equilibration took place. According to Schoeller and Schoeller (1979) the TDS con- tents and especially the HCO3 – contents decrease with increasing distances from the area of Vichy-Cantal-Devès. 30 Falk H. Weinlich As in the western Eger Rift, the N2-rich gas- es in the Massif Central are enriched in helium. The extreme enrichment of helium, whose con- tents are the highest in Europe, points rather to a complete fractionation than to a simple mix- ing with biogenic components. According to the gas flux measurement of Batard et al. (1982) 30 l/h of N2 are also released in the CO2- rich gases of Royat. On the other hand, N2 re- lease in the case of N2-rich gases are ca. 2.8 l/h in Bourbon-Lancy, 29.6 l/h in Evaux-les-Bains, 0.5 l/h in Sail-les-Bains and 1.7 l/h N2 in the Lithium spring in Santenay. Thus only the gas composition is fractionated and it is not neces- sary to assume additional N2 sources. An additional argument contradicting the in- fluence of mixing processes is that outside these areas with magmatic CO2 there are no springs with biogenic/organic CO2 in the gas phase. The pro- duction rates of biogenic CO2 in soils (Andrews and Schlesinger, 2001) are too small to nourish a free gas phase. A long-term measurement of iso- tope composition of CO2 in the gas phase of the Wettin spring in Bad Brambach (Weise et al., 2001) compared with the biogenic CO2 production rates (Andrews and Schlesinger, 2001) demon- strate that there is no influence (fig. 10). It is also problematic to derive biogenic/or- ganic CO2 from sedimentary rocks, especially in areas of metamorphic rocks, since these wa- ters and gases circulate within fault systems. There, either a far-reaching CO2 exchange be- tween the gases migrating along fault pathways and the surrounding country rocks is impeded or the ascending magmatic CO2 saturates the groundwater with CO2 gas, as it is the case in the Cheb Basin. The CO2 concentration gradi- ent in the close vicinity of the faults prevents the admixture of CO2 from other sources like for example the biogenic/organic CO2. 4. Conclusions As demonstrated, it is possible to elucidate low δ 13C values with gas fractionation, i.e. by isotope fractionation with formed HCO3 – and not necessarily and exclusively by mixing with biogenic or organic CO2. However, without complete mass balances it is not possible to dis- criminate between either or give reasons to pre- fer one of the interpretations. In some cases, it will not be possible to educe the «last proof» for the interpretation. Therefore, it should al- ways be considered that even enhanced CO2- contents in the soil air encountered in the vicin- ity of fractured rocks can also represent com- pletely fractionated magmatic CO2. However, if we have to assume that the iso- topic composition and contents of CO2 in the gas phase can be alternated by fractionation processes, an influence on the C/3He ratios should also be assumed. Marty et al. (1989) de- scribed abating C/3He linked with reduced CO2 contents. Acknowledgements The author would like to express his gratitude to Fausto Grassa, Istituto Nazionale di Geofisica e Vulcanologia - Sezione di Palermo, for provid- ing constructive comments to the review. The support of H.W. Hubberten, A. Wegner Inst. f. Polar Marin Res., Potsdam in his permission to use his unpublished data is greatly appreciated. K. Pütz, Staatsbad Bad Bertrich is thanked for supplying gas samples. For constructive discus- sions I would like to thank Jolanta Kus and Franz May, BGR Hannover. REFERENCES ANDREWS, J.A. and W.H. SCHLESINGER (2001): Soil CO2 dynamics, acidification, and chemical weathering in a temperate forest with experimental CO2 enrichment, Global Biochem. Cycles, 15, 149-162. AUTRAN, A., J-P. BRETON and J.C. CHIRON (coordinator) (1980): Carta tectonquie de la France 1:1 000 000, Mem. BRGM 110, pp. 52. BATARD, F., J.C. BAUBRON, B. BOSCH, A. MARCÉ and J.J. RISLER (1982): Isotopic identification of gases of a deep origin in French thermomineral waters, J. Hy- drol., 56, 1-21. BAUBRON, J-P., B. BOSCH, P. DESGRANGES, M. LELEU, J.J. RISLER and C. SARCIA (1979): Recherches géocimiques sur les eaux thermals de la bordure oust de la Limagne, Bull. Minéral, 102, 676-683. BEYER, F. (1995): Isotopenhydrogeologie der Aachener Thermalquellen, Geol. Inst. Univ. Köln, Sonderveröf- fentlichungen 105, p. 148. CAPASSO, G., R. FAVARA and S. INGUAGGIATO (1997): Chem- 31 Isotopically light carbon dioxide in nitrogen rich gases ical features and isotopic composition of gaseous man- ifestations on Vulcan Island, Aeolian Islands, Italy: an interpretative model of fluid circulation, Geochim. Cosmochim. Acta, 61, 3425-3440. CHIODINI, G., F. FRONDINI, D.M. KERRICK, J. ROGIE, F. PAR- ELLO, L. PERUZZI and A.R. ZANZARI (1999): Quantifi- cation of deep CO2 fluxes from Central Italy. Examples of carbon balance for regional aquifers and of soil dif- fuse degassing, Chem. Geol., 159, 205-222. D’ALESANDRO, W., S. DE GREGORIO, G. DONGARRÀ, S. GURRIERI, F. PARELLO and B. PARISI (1997): Chemical and isotopic characterization of the gases of Mount Et- na (Italy), J. Volcanol. Geotherm. Res., 78, 65-76. GRIESSHABER, E., R.K. O’NIONS and E.R. OXBURGH (1992): Helium and carbon isotope systematics in crustal fluids from the Eifel, the Rhine Graben and Black Forest (FRG), Chem. Geol., 99, 213-135. KOLÁŘOVÁ, M. and V. MYSLIL (1979): Minerální vody Zá- padočeského kraje, UUG (Geol. Surv.) Praha, p. 286. KUSCHKA, E. and W. HAHN (1996): Flußspatlagerstätten des Südwestvogtlandes: Schönbrunn, Bösenbrunn, Wieders- berg, Bergbau in Sachsen, Bd. 2 (Landesamt Umwelt u. Geol, Sächs. Oberbergamt), p. 281. MARTY, B. and N. DAUPHAS (2003): The nitrogen record of crust-mantle interaction and mantle convection from Archean to present, Earth Planet. Sci. Lett., 204, 397-410. MARTY, B., A. JAMBON and Y. SANO (1989): Helium iso- topes and CO2 in volcanic gases of Japan, Chem. Geol., 76, 25-40. MATTHEWS A., C. FOUILLAC, R. HILL, R.K. O’NIONS and E.R. OXBURGH (1987): Mantle-derived volatiles in con- tinental crust: the Massif Central of France, Earth Planet. Sci. Lett., 85, 117-128. MAY, F. (2002): Quantifizierung des CO2-Flusses zur Ab- bildung magmatischer Prozesse im Untergrund der Westeifel (Shaker Verlag), p. 170. MOOK, W.G., J.C. BOMMERSON and W.H. STAVERMAN (1974): Carbon isotope fractionation between dissolved bicarbonate and gaseous Carbon dioxide, Earth Planet. Sci. Lett., 22, 169-176. MOUREU, CH. (1923): Les gaz rares des gaz naturels, J. Chem. Soc., 123, 1905-1947. MOUREU, CH. and A. LEPAPE (1912): Sur quelques mélanges gazeux naturels particulièrement riches en hélium. Gise- ments d’hélium, C.R. Acad. Sci. Paris, 155, 197-200. NÉGREL, PH., C. GUERROT, A. COCHERIE, M. AZAROUAL, M. BRACH and CH. FOUILLAC (2000): Rare earth elements, neodymium and strotium isotopic systemtics in miner- al waters: evidence from the Massif Central, France, Appl. Geochem., 15, 1345-1367. PAUWELS, H. and C. FOULLIAC (1997): The isotopic and chemical composition of CO2-rich thermal waters in the Mont-Dore region (Massif-Central, France), Ap- plied Geochmistry, 12, 411-427. RISLER, J.J., G. CASTANY, J. MARGAT, L. MONITION and D.B. NINARD (1973): Carte des eaux minérales et thermales de la France 1:1 000 000, Mem. BRGM. SCHOELLER, H. and M. SCHOELLER (1979): Une étude des eaux thermominérales du Massif Central Français, Bull. BRGM, Sec. III, 2, 121-156. WEINLICH, F.H. (1989): Geochemie und Genese des Stick- stoffs in den vogtländisch-erzgebirgischen Quellgasen, Z. Angew. Geol., 35, 129-135. WEINLICH, F.H., J. TESAŘ, S.M. WEISE, K. BRÄUER and H. KÄMPF (1998): Gas flux distribution in mineral springs and tectonical structure in the western Eger Rift, J. Czech Geol. Soc., 43, 91-110. WEINLICH, F.H., K. BRÄUER, H. KÄMPF, G. STRAUCH, J. TESAŘ and S.M. WEISE (1999): An active subcontinen- tal mantle volatile system in the western Eger Rift, Central Europe: gas flux, isotopic (He, C and N) and compositional fingerprints, Geochim. Cosmochim. Ac- ta, 63, 3653-3671. WEINLICH, F.H., K. BRÄUER, H. KÄMPF, G. STRAUCH, J. TESAŘ and S.M. WEISE (2003): Gas flux and tectonic structure in the western Eger Rift, Karlovy Vary - Oberpfalz and Oberfranken, Bavaria, GeoLines, Prague, 15, 171-177. WEISE, S.M., K. BRÄUER, H. KÄMPF, G. STRAUCH and U. KOCH (2001): Transport of mantle volatiles through the crust traced by seismically released fluids: a natu- ral experiment in the earthquake swarm area Vogt- land/NW Bohemia, Central Europe, Tectonophysics, 336, 137-150. WENDT, I. (1968): Fractionation of carbon isotopes and its temperature dependence in the system CO2-Gas-CO2 in solution and CO3-CO2 in solution, Earth Planet. Sci. Lett., 4, 64-68. WILSON, M. and H. DOWNES (1992): Mafic alkaline mag- matism associated with the European Cenozoic rift system, Tectonophysics, 208, 172-182. ZHANG, J., P.D. QUAY and D.O. WILBUR (1995): Carbon iso- tope fractionation during gas-water exchange and disso- lution of CO2, Geochim. Cosmochim. Acta, 59, 107-114.