Hydrogeochemistry and isotope geochemistry of Velenje Basin groundwater Hidrogeokemija in izotopska geokemija podzemnih vod Velenjskega bazena

The geochemical and isotopic composition of groundwater in the Velenje Basin, Slovenia, was investigated between the years 2014 to 2015 to identify the geochemical processes in the major aquifers (Pliocene and Triassic) and the water–rock interactions. Thirty-eight samples of groundwater were taken from the aquifers, 19 in the mine and 19 from the surface. Groundwater in the Triassic aquifer is dominated by HCO3 –, Ca2+ and Mg2+ with δCDIC values in the range from -19.3 to -2.8 ‰, indicating degradation of soil organic matter and dissolution of carbonate minerals. In contrast, groundwater in the Pliocene aquifers is enriched in Mg2+, Na+, Ca2+, K+, and Si, and has high alkalinity, with δCDIC values in the range of -14.4 to +4.6 ‰. Based on the δ CDIC values in all the aquifers (Pliocene and Triassic), both processes influence the dissolution of carbonate minerals and dissolution of organic matter and in the Pliocene aquifers, methanogenesis as well. Based on Principal Component Analysis (PCA), and on geochemical and isotopic data we conclude that the following types of groundwater in Velenje Basin are present: Triassic aquifers with higher pH and lower conductivity and chloride, Pliocene, Pliocene 1 and Pliocene 2 aquifers with lower pH and higher conductivity and chloride contents, and Pliocene 3 and Pliocene 2, 3 aquifers with the highest pH values and lowest conductivities and chloride contents. 87Sr/86Sr tracer was used for the first time in Slovenia to determine geochemical processes (dissolution of silicate versus carbonate fraction) in Velenje Basin groundwater of different aquifers dewatering Pliocene and Triassic strata. 87Sr/86Sr values range from 0.70820 to 0.71056 in groundwater of Pliocene aquifers and from 0.70808 to 0.70910 in groundwater of the Triassic aquifer. This indicates that dissolution of the carbonate fraction prevails in both aquifers, while in Pliocene aquifers, an additional silicate weathering prevails with higher 87Sr/86Sr isotope ratios.


Introduction
The Velenje Basin in Slovenia is one of the largest actively mined coal basins in central Europe, producing around 4 million tons of lignite per year (MihelaK, 2010). Large amounts of groundwater are extracted from Velenje Basin aquifers to facilitate underground mining of coal, and coal seam gas outbursts are a serious mine safety concern KanDuč et al., 2015a). The main problem in the past production years of Velenje lignite mine has been the groundwater standing directly over the main production lignite strata inside the Pliocene's clastic rocks. In the 1970's many groundwater intrusions occurred in the mine. To reduce the hydrostatic pressure during the last four decades, advanced and massive dewatering works have been performed. As a result, constant water pumping in the production area and its surroundings has produced a typical dewatering groundwater cone (vižintin et al., 2009).

Hydrogeochemistry
and stable isotope compositions of groundwater provide critical information regarding water-rock interaction along flow paths, and mixing of distinct groundwater bodies (cartwriGht et al., 2012). Groundwater geochemistry, especially where aquifers have contrasting mineralogy, is an effective method of determining regional scale inter-aquifer flow and patterns of recharge (DoGraMaci & herczeG, 2002;eDMunDs, 2009;cartwriGht, 2010). In this study we use major ion geochemistry and environmental isotopes to constrain geochemical processes in groundwater of Velenje Basin. In addition, characterization of geochemical properties of groundwater may further aid in understanding of hydrocarbon entrapment and mineral deposition, and help in identifying the origin of fluids (aravena et al., 2003).
Concentrations of dissolved inorganic carbon (DIC) and its isotopic composition (δ 13 C DIC ) are governed by processes occurring in the water system. Changes of dissolved inorganic carbon concentrations result from addition or removal of carbon from the DIC pool (ateKwana & KrishnaMurthy, 1998). The major sources of carbon to the groundwater DIC system are the dissolution of carbonate minerals, and soil CO 2 derived from root respiration and from microbial decomposition of organic matter. Constraining water-rock interaction, evapotranspiration, dissolution and precipitation of minerals, ionexchange and groundwater mixing are all required to understand and manage hydrogeological systems (cartwriGht et al., 2007).
The use of strontium isotopes in hydrogeological studies has many applications. First, minerals that interact with groundwater and surface water exhibit a wide and predictable range of 87 Sr/ 86 Sr ratios. 87 Sr is produced by the decay of 87 Rb with a half-life of 48.8 Ga (Faure, 1991). Strontium derived from K-rich minerals such as biotite and K-feldspar, have high 87 Sr/ 86 Sr ratios, minerals such as plagioclase that have lower K/Ca ratios will contain strontium with moderate 87 Sr/ 86 Sr ratios, while the strontium in Ca-rich minerals such as calcite or gypsum, have low 87 Sr/ 86 Sr ratios that remain essentially unchanged over time (Faure, 1991;Mcnutt, 2000). The geochemical and isotopic composition of carbon and 87 Sr/ 86 Sr isotopes of groundwater in the Velenje Basin, combined with mineralogical information, are applied to investigate the groundwater-rock interactions and geochemical processes that contribute to understanding the generation and source of coal seam gases, namely carbon dioxide and methane (KanDuč et al., 2012;seDlar et al., 2014;KanDuč et al., 2015A).

Study area
The Velenje Basin and associated coal mines are situated in the NE part of Slovenia (Fig. 1). It is situated at the junction of the WNW-ESE trending Šoštanj fault and the E--W trending Periadriatic zone, bounded to the south by the Smrekovec fault segment. The Šoštanj and Smrekovec faults were generated by the collision of continental plates. In general, two hydrogeological systems can be distinguished in the Velenje coal mine ( Fig. 1): (1) Plioquaternary and Pliocene aquifers, each composed of gravel, sand and silt, which are further divided to upper, middle and lower aquifers. Below the latter aquifers, Triassic carbonates are present as a basement of the basin. The upper aquifers are schematically divided into Plioquaternary alluvial sediments (not analyzed in this study) and at least three parts of the Pliocene aquifers (Pliocene 1, Pliocene 2, Pliocene 3) and (2) carbonate aquifers, which are further divided into Lower Triassic (Scythian) limestone and dolomite, Anisian dolomite and limestone, and Miocene Lithotamnium limestone (vižintin et al., 2009 & references therein). Pliocene and Pliocene 2, 3 aquifers were introduced due to the fact that in some cases, it was not possible to attribute the water to a single aquifer, so a combination of all (Pliocene) or two aquifers (Pliocene 2, 3) was used instead. Hydrogeological properties of these dolomites are presented in detail in the works of verBovšeK & veselič (2008) and verBovšeK (2008a, 2008b. Groundwater recharging the Velenje Basin is drained by hanging filters to prevent inrush of water into the mine. The average discharge of water from Pliocene sands is 800 l/min, while that from Triassic limestone aquifers is 3400 l/ min (suPovec et al., 2012a(suPovec et al., , 2012b. In the Velenje coal basin, the shallow aquifers are part of the Quaternary and Pliocene sections. Pliocene aquifers are further classified as aquifers above the coal 0-20 m (Pl-1), aquifers 20-80 m above the coal, Pl-2, and upper Pliocene aquifers (Pl-3) (Fig. 1). Fig. 1. Geological sketch map (BreziGar et al., 1988) of the Velenje Basin with a NNE-SSW cross section. Main aquifers are Pliocene (Pl 1 , Pl 2 and Pl 3 ), and Triassic (T 1 , T 2 and T 2,3 ) dolostones (prevailing) and limestones. Other Triassic and Paleozoic (Pz) lithologies are composed of relatively impermeable strata and therefore do not include significant aquifers. Profiles A-B and C-D show the locations of groundwater sampling in the study of KanDuč et al. (2014).
Hydrogeochemistry and isotope geochemistry of Velenje Basin groundwater

Materials and methods
In a previous study (KanDuč et al., 2014), only 14 groundwater samples were taken seasonally in Velenje Basin for geochemical and isotope characterization. Due to complex groundwater flow paths in the Velenje Basin, a more dense sampling network was established for investigation, in detail of spatial variability in hydrogeochemistry and isotope geochemistry.
Location of groundwater sampling location is presented on Figure 2. In this study, following locations of groundwater dewatering Triassic and Pliocene strata were sampled: groundwater dewatering Triassic strata (15 locations), Pliocene 2 (10 locations), Pliocene (1 location), Pliocene 1 (9 locations), Pliocene 3 (2 locations), Pliocene 2, 3 (1 location) (Fig. 2). Samples were taken from: a) overflow piezometers from the surface, b) groundwater samples sampled with Solinst sampler and c) mine wells. Groundwater samples were taken wherever possible since the complete Velenje Basin is subsiding due to mining activity and piezometers are sheared. Altogether 38 samples were taken, 19 from the mine and 19 from the surface. Temperature, pH and electrical conductivity were measured in the field. Some groundwater samples (sampling b) were sampled with a Discrete Solinst Sampler (KanDuč et al., 2014(KanDuč et al., , 2015a, modified for depth measurements and for sampling from depths to almost 400 m. Sample aliquots collected for chemical analysis (major, trace elements, 87 Sr/ 86 Sr) were immediately passed through 0.45 µ PTFE membranes, which were leached with 1-2 ml of dilute clean HCl, then 2-4 ml of MilliQ water (Millipore) and finally 1-2 ml of the sample itself before collecting the sample filtrate. For cations, 87 Sr/ 86 Sr and trace element analysis, 30 ml LDPE (Nalgene) bottles were used. The bottles were filled with 10 % HCl leaving some headspace and placed in in a zip lock bag in an oven at 70 °C for one day. The bag was then tightened and left for a second day. The bottles were then rinsed repeatedly with MilliQ water, and filled completely with MilliQ water and capped the bottles. They were finally emptied in a dust free place (fume cupboard) and left until dried. Samples for δ 13 C DIC analysis were stored in glass vials filled to the top, with no headspace. Total alkalinity was measured within 24 h of sample collection by Gran titration (GiesKes, 1974) with a precision of ±1 %. Major anions (Cl -, SO 4 2and NO 3 -) were analyzed (with ± 2 % precision) with a Dionex Ion Chromatograph (IC) Model 3000, using an AS23 analytical column in the Hydrology and Water Resources Department at the University of Arizona (UA).
Major and trace elements (Na + , K + , Ca 2+ , Mg 2+ , Sr 2+ ) in water were determined using a quadrupole inductively coupled plasma mass spectrometer (ICP-MS, Agilent 7500ce) with a double-pass spray chamber, PTFE 100 L/min nebulizer, platinum cones and sapphire injector within a quartz platinum-shielded torch. The instrument is located in a filtered air positive pressure lab and sample handling and chemistry was routinely performed on laminar flow benches. External calibration curves for all the elements reported were prepared from 1,000 mg/L single element standard solutions in HNO 3 (Inorganic Ventures). Both the samples, blanks and the calibration solutions were prepared in 2.4 % HNO 3 (BDH Aristair Plus). Standard Reference Material (SRM) 1643e ("Trace Elements in Water" , National Institute of Standards and Technology, USA) was prepared at 1:20 dilution and run together with blanks, samples and calibration curve. Indium at a concentration of 20 ppb was added to the calibration curve, reference material, samples and blanks as internal standard. Samples were diluted in 1:5 and run with inductively coupled plasma mass spectrometer (MC-ICPMS) in blocks of five samples, two blanks and one SRM. This analysis were performed at University of Utah.
The stable isotope composition of dissolved inorganic carbon (δ 13 C DIC ) was determined on Europa scientific TG isotope ratio mass spectrometer coupled with a ANCA-TG preparation module. Phosphoric acid (100 %) was added (100-200 µl) to a septum tube, then purged with pure He. The water sample (1 ml) was then injected into the septum tube and CO 2 measured directly from the headspace. Two standard solutions of Na 2 CO 3 Tjaša KANDUČ, Nataša MORI, Ajda KOCELI & Timotej VERBOVŠEK (Carlo Erba and Scientific Fisher) with known δ 13 C DIC values of -10.8±0.2 ‰ and -4.8±0.2‰ were used to calibrate δ 13 C DIC measurements (sPötl 2005;KanDuč, 2006). This analysis were performed at Jožef Stefan Institute.
Isotopic analysis of strontium was performed using an inductively coupled plasma mass spectrometer (MC-ICPMS) Neptune (ThermoFisher Scientific) at the Department of Geology and Geophysics, University of Utah. Purified fractions, with a strontium concentration around 30 ppb, were introduced in the mass spectrometer through a PTFE nebulizer, quartz double-pass spray chamber, quartz torch and nickel cones.
Principal component analysis (PCA), based on correlation matrix was used to examine the variation of selected environmental variables (temperature, pH, conductivity, Na + , K + , Cl -, Mg 2+ , Ca 2+ , nitrate (NO 3 -), sulphate (SO 4 2-) and HCO 3 ions and δ 13 C DIC ) measured in 38 samples of Velenje Basin groundwater. The CANOCO software package (ter BraaK & šMilauer, 2002) was applied, and the results presented as an ordination diagram, on which points represent samples and arrows the measured parameters.
Chemical speciation of the carbonate systeme.g. partial pressures of CO 2 (pCO 2 ), saturation indices of calcite (SI calcite ) -was evaluated using water composition, pH, alkalinity, and temperature as inputs to the PHREEQC speciation program and employing thermodynamic computations (ParKhurst & aPPelo, 1999).

Groundwater geochemistry of Velenje Basin groundwater
The pH of groundwater in the Triassic aquifers ranged from 6.8 to 8.7 and, in the Pliocene aquifers from 6.6 to 9.7 (Table 1). Dissolved oxygen (DO) was not measured in this study due to risk of explosion (from elevated methane concentration), but it ranges from 10 to 30 % in groundwater dewatering Pliocene strata and from 60 to 80 % in groundwater dewatering Triassic strata (KanDuč et al., 2014). Conductivity ranged from 205.4 to 4410 µS/cm in groundwater dewatering Pliocene strata, and from 232.8 to 4330 µS/cm in groundwater dewatering Triassic strata (Table 1). Only in one sample (j.v. 3051/01, Fig 2), from dewatering Pliocene strata, was very high conductivity found (Table 1). The very high alkalinities can be contributed to by the methanogenesis process, that occurs in the Pliocene groundwaters and are discussed below.
Concentrations of major and minor ions in Velenje Basin groundwater dewatering Pliocene strata and Triassic strata are presented on Figure 3. Groundwater in the Triassic aquifer is dominated by HCO 3 -, Ca 2+ , Mg 2+ (mostly water types Ca-Mg-HCO 3 and Mg-Ca-HCO 3 ), while groundwater in the Pliocene aquifers is enriched in Mg 2+ , Na + , Ca 2+ , K + and Si. Total alkalinity of Pliocene aquifers ranges from 1.95 to 44.1 mM in groundwater dewatering Pliocene strata and from 1.3 to 10.4 mM in groundwater dewatering Triassic strata (Table 1). The highest alkalinity   (Table 1). SO 4 2concentrations in groundwater from Pliocene aquifers are low due to reduction conditions in the aquifer (KanDuč et al., 2014) and range from 0.00078 mM (0.074 mg/l) to 0.50 mM (48 mg/l), NO 3 concentrations range from 0.0002 mM (0.011 mg/l) to 0.11 mM (6.28 mg/l) mM in groundwater from Pliocene aquifers (Table 1). In addition, SO 4 2and NO 3 concentrations were low in the Pliocene aquifers (up to 0.5 mM (48 mg/l) and up to 0.11 mM (6.82 mg/l)) compared to groundwater from Triassic aquifer indicating sulfate and nitrate reduction (KanDuč et al., 2014), and consistent with microbial methanogenesis (coetsiers & walraevens 2009). PCA analysis explained 49.9 % of all variability in data by the first, and an additional 19.4 % by the second axis (Fig. 4). Most of the samples are distributed along the first axis, described mostly as differences in pH and conductivity linked to chloride. Higher conductivity and chloride concentrations and lower pH were observed in the groundwater samples originating from Pliocene strata. The second axis represents mainly a gradient in sulfate and δ 13 C DIC . Generally, the groundwater samples dewatering Triassic strata grouped together in the ordination diagram, while samples of groundwater dewatering Pliocene strata are scattered along both axes, indicating greater variation in measured variables. The outlier sample j.v. 3051/01 is from dewatering Triassic strata, having a high concentration of sulfate, 20.18 mM (1937 mg/l) (  (Ca 2+ +Mg 2+ ) line (Fig. 5A). Dissolution of calcite produces waters with a Mg 2+ : Ca 2+ molar ratio of less than 0.1, a ratio of 0.33 in the case of congruent dissolution of calcite and dolomite, and a ratio equal to 1 if only dolomite is dissolving (szraMeK et al., 2011). Groundwater in the Triassic aquifer falls between lines 0.5 and 1, indicating weathering of dolomite (Fig. 5B), while groundwater in Pliocene aquifers deviates from the dolomite line (having a slope of 1) indicating that some other minerals (e.g. albite, plagioclase) contribute to alkalinity, already confirmed in previous studies by X-ray diffraction (KanDuč et al., 2014 and references therein).
Carbon cycling and strontium geochemistry of Velenje Basin groundwater Partial pressures of CO 2 (pCO 2 ) in groundwater of Pliocene and Triassic aquifers ranged from 15.8 to 575,439.9 ppm and from 169.8 to 107,151.9 ppm, on average 80800 ppm, which is 202 times greater (supersaturated) than that of atmospheric CO 2 concentrations (400 ppm). SI calcite ranged from -0.8 to 1.48 and SI dolomite ranged from -0.32 to 2.93 in Pliocene aquifer. SI calcite ranged from -0.75 to 0.52 in groundwater dewatering Triassic aquifer and SI dolomite ranged from -1.54 to 1.06 in groundwater dewatering Triassic aquifer (Fig. 6). Groundwater samples from the Pliocene and Triassic aquifers therefore have calcite saturation indices (SI calcite ) Hydrogeochemistry and isotope geochemistry of Velenje Basin groundwater generally well above the equilibrium (SI calcite = 0), indicating that calcite is supersaturated and that precipitation is favored thermodynamically due to high alkalinities, especially in the Pliocene aquifer. Only two samples from mining wells and three samples that overflow dewatering Triassic strata are undersaturated with calcite and dolomite (Fig. 6). We also observed precipitates of dolomite/ calcite as visible incrustations in some of the wells. Such incrustations have formed on the well casing, and were scratched from the casing walls when measuring with the Solinst Discrete Interval Probe (photo on Fig. 6).
Values of δ 13 C DIC for groundwater of dewatering Pliocene strata ranges from -14.4 to +4.6 ‰. Groundwater dewatering Triassic strata ranges from -19.3 to -2.8 ‰ (Table 1). Geochemical processes influencing the value of δ 13 C DIC in groundwater are presented on Figure 7. Groundwater samples from the Triassic aquifer have generally lower δ 13 C DIC and are similar to those investigated previously for surface waters (KanDuč et al., 2014). They fall approximately on the line of carbonate dissolution by carbonic acid produced from the soil zone with a δ 13 C CO2 of -26.6 ‰ (KanDuč et al., 2007). Groundwater samples from the Pliocene aquifers have higher values of δ 13 C DIC (up to +4.6 ‰), which are attributed to microbial methanogenesis, causing enrichment in 13 C (KanDuč et al., 2012). Isotopic compositions of carbon in methane (δ 13 C CH4 ) and of deuterium in methane (δD CH4 ), analyzed directly from groundwater ranged from -77.7 to -51.4 ‰ and from -246.6 to -162.0 ‰, indicating bacterial gas (KanDuč et al., 2014;KanDuč et al., 2015b). Higher δ 13 C DIC values are due to methanogenesis within aquifer and are related to microbial methanogenesis associated with coal seam. We also found microbial gas collected from piezometers (as free gas) in previous studies and its δ 13 C CH4 have values in the range of -70.5 to -34.2‰ (KanDuč et al., 2012;KanDuč et al., 2014).
• Line 3. An average δ 13 C value of particulate organic carbon (POC) of -26.6 ‰ was assumed to represent the isotopic composition of POC that was transferred to DIC by in stream respiration. Open system equilibration of DIC with CO 2 enriches DIC in 13 C by about 9 ‰ (MOOK et al., 1974), which corresponds to the value of -17.6 ‰ (Fig. 7). • Line 4. This represents open system equilibration of DIC with soil CO 2 originating from degradation of organic matter with δ 13 C CO2 of -26.6 ‰ (Fig. 7).  figure 6 represents precipitation of carbonates on a probe of "Solinst sampler".
Groundwater in our study is considered as a closed system (clarK & Fritz, 1997), therefore any interactions with atmosphere are not possible.
It was found in previous studies that concentrations of dissolved organic carbon (DOC) in the Pliocene and Triassic aquifers in the Velenje basin are low, ranging from 1.54 to 14.69 mg/L (KanDuč et al., 2010), compared to groundwater associated with coalbeds in the Powder River Basin (oreM et al., 2010). In the latter, high DOC concentrations are related to methanogenesis and contact of groundwater with coal, but in Velenje Basin we can conclude that the methanogenesis (Fig. 7) in groundwater is probably related to the presence of low dissolved oxygen (DO), high alkalinity and high δ 13 C DIC values, occurring within the Pliocene aquifers (KanDuč et al., 2014).
The ratio 87 Sr/ 86 Sr ranges from 0.70820 to 0.71056 in groundwater dewatering Pliocene aquifers and from 0.70808 to 0.70910 in groundwater dewatering Triassic strata (Table 1). Figure 8 represents 87 Sr/ 86 Sr ratios and indicates characteristic values of 87 Sr/ 86 Sr isotopes in different environment with different geological composition e.g. marine limestone 0.711, while silicate fraction has higher 87 Sr/ 86 Sr ratio. The relations between 87 Sr/ 86 Sr ratios and δ 13 C DIC for groundwater in Pliocene and Triassic aquifers for Velenje Basin groundwater are presented on Figure 8. Based on the 87 Sr/ 86 Sr ratio in groundwater from both aquifers (Triassic and Pliocene), the carbonate fraction prevails (Fig. 8). The highest 87 Sr/ 86 Sr ratio was found in Pliocene 2, 3 strata, implying that more silicate fraction is leached in the aquifer. Generally, in Pliocene aquifers of Velenje Basin groundwater, higher 87 Sr/ 86 Sr ratios are observed than in Triassic aquifers (Table 1, Fig. 8). Given that carbonate dissolution rates are higher than those of silicates (lasaGa, 1984) as is the case for the Triassic and Pliocene aquifers of Velenje Basin carbonate dissolution may be important in controlling Sr isotope geochemistry (harrinGton & herzeG, 2003). However in the study of Murray Basin, Australia, 87 Sr/ 86 Sr ratios of groundwater lie between 0.71068 and 0.71905 (median = 0.71595), generally higher than values for potential carbonate sources (cartwriGht et al., 2007). Trends in 87 Sr/ 86 Sr ratios versus δ 13 C DIC values for the mixing of groundwater from silicate and carbonate aquifers of Velenje Basin groundwater are shown schematically in Figure 8. Trends in 87 Sr/ 86 Sr ratios versus δ 13 C DIC values for the mixing of groundwater from silicate and carbonate aquifers or for the progressive dissolution of carbonates are shown in Figure 8. 87 Sr/ 86 Sr ratios of the silicate and carbonate fraction of aquifer Lines indicate processes in aquifer as follows: 1: open system DIC equilibration with atmosphere (Exchange), 2: dissolution of carbonates with average δ 13 C CaCO3 = -2 ‰, 3: nonequilibrium carbonate dissolution by carbonic acid produced from soil zone with a δ 13 C CO2 of -26.6 ‰, 4: open system equilibration of DIC with soil CO 2 originating from degradation of organic matter with δ 13 C CO2 of -26.6 ‰. material in Velenje Basin groundwater, rainfall (ullMan & collerson, 1994), typical 87 Sr/ 86 Sr ratios and δ 13 C DIC values of marine and terrestrial carbonates (clarK & Fritz, 1997;Faure, 1991), and typical δ 13 C values of soil carbon (clarK & Fritz, 1997), are also shown.

Conclusions
Sampling in this study was performed only where possible due to the subsidence of the area, which destroyed some of the piezometers. New wells were sampled in the following three sampling campaigns during the period 2014-2015: a) overflow piezometers, b) groundwater and c) mine wells. Altogether, 38 wells were investigated for geochemical and isotopic analyses.
The major ion geochemistry and isotope geochemistry of the Velenje Basin groundwater are indicated in this study. The major ion geochemistry, isotopic composition of carbon and Sr ratios (δ 13 C DIC , 87 Sr/ 86 Sr) provides a new insight into geochemical processes (water -rock-interactions) in Velenje Basin groundwater.
The concentrations of dissolved ions and cations decreases in the order HCO 3 ->Mg 2+ >Na + >Ca 2+ in the Pliocene aquifers and HCO 3 ->Ca 2+ >Mg 2+ in the Triassic aquifer. Multivariate PCA analysis, based on "in situ" measured parameters (conductivity, pH, temperature), major geochemical parameters (Ca 2+ , Mg 2+ , K + , Na + , NO 3 -, Cl -, SO 4 2-) and δ 13 C DIC revealed that groundwater samples dewatering Triassic strata were grouped together in the ordination diagram, while the samples of groundwater dewatering Pliocene strata were scattered along both axes due mainly to variations in pH, conductivity, chloride, sulphate and δ 13 C DIC .
The values of δ 13 C DIC and 87 Sr/ 86 Sr, together with major ion geochemistry help explain geochemical processes in Velenje Basin groundwater. δ 13 C DIC values of groundwater in Pliocene aquifers range from -14.4 to +4.6 ‰ and in groundwater of Triassic aquifer from -19.3 to -2.8 ‰. Alkalinities range from 1.95 to 44.12 mM in Pliocene aquifers and from 1.32 to 10.43 mM in Triassic aquifer. δ 13 C DIC of groundwater in the Triassic aquifer indicate carbonate dissolu-tion (dolostones and limestones) and organic matter degradation. In Pliocene aquifers, however methanogenesis (higher δ 13 C DIC values up to +4.6 ‰) is present in addition to degradation of organic matter and dissolution of carbonates, contributing to the very high alkalinity in these aquifers. Calculated value of pCO 2 in groundwater of Pliocene aquifers ranged from 15.8 to 575,439.9 ppm and in groundwater of Triassic aquifer from 169.8 to 107,151.9 ppm. Most groundwater in Pliocene and Triassic aquifers was oversaturated with calcite and dolomite. Pliocene groundwaters have much higher oversaturation values than the Triassic ones, also due to methanogenesis and high alkalinity.
The 87 Sr/ 86 Sr ratios in Pliocene aquifers range from 0.70820 to 0.71056 and, in groundwater of Triassic aquifer, from 0.70808 to 0.70912. 87 Sr/ 86 Sr isotopes in combination with δ 13 C DIC , provide added value to the study in deciphering geochemical processes, water-rock interactions and silicate/ carbonate weathering within the aquifers. The carbonate was found to have an 87 Sr/ 86 Sr ratio around 0.71120, the silicate fraction being in the range of 0.7120 to 0.7190. Groundwater from Pliocene aquifers has more positive δ 13 C DIC values and 87 Sr/ 86 Sr ratio, except for two samples located in the Pliocene 3 aquifer ( 87 Sr/ 86 Sr ratio 0.70972, alkalinity 3.23 mM and δ 13 C DIC of -9.13 ‰, PM13/84) and Pliocene 2, 3 aquifer ( 87 Sr/ 86 Sr ratio 0.71056 with low alkalinity of 1.95 mM and δ 13 C DIC of -14.37‰, PE 16/84 sample). Groundwater from Triassic aquifer has more negative δ 13 C DIC values and lower 87 Sr/ 86 Sr.
From the hydrogeological (dewatering system) and geochemical (biogeochemical processes and age of groundwater) points of view it would be useful to investigate further the geochemical and hydrogeological relationships between shallow and deeper aquifers in different strata formations (Pliocene and Triassic). New observation objects (piezometers) will be drilled at northern part of the basin where mining activity is present.
As due to the mining activity, the basin is presently subsiding, a geochemical investigation in combination with hydrogeological studies is also necessary from the safety reasons and groundwater inrush into the mine.
Besides the groundwater investigations it would be useful to continue investigation of gases in groundwater and also geochemical characteristics of coalbed gases in coal matrix at active excavation fields. These investigations would be very useful for future development of CBM technology after the termination of mining activities in Velenje coal basin.