Overview of isotopic investigations of groundwaters in a fractured aquifer system near Rogaška Slatina, Slovenia Pregled izotopskih raziskav podzemne vode v razpoklinskem vodonosnem sistemu na območju Rogaške Slatine Branka TRČEK1 & Albrecht LEIS2 1University of Maribor, Faculty of Civil Engineering, Transportation Engineering and Architecture, Smetanova ulica 17, SI-2000 Maribor, Slovenia; e-mail: branka.trcek@um.si 2JR-AquaConSol GmbH, Elisabethstraße 18/II, 8010 Graz, Austria; e-mail: albrecht.leis@jr-aquaconsol.at Prejeto / Received 15. 11. 2016; Sprejeto / Accepted 9.5.2017; Objavljeno na spletu / Published online 9.6.2017 Dedicated to Professor Jože Pezdič on the occasion of his 70th birthday Key words: fractured aquifer, mineral waters, environmental isotopes, Rogaška Slatina, Slovenia Ključne besede: razpoklinski vodonosnik, mineralna voda, naravni izotopi, Rogaška Slatina, Slovenija Abstract The isotopic investigations of groundwaters stored in the Rogaška Slatina fractured aquifer system were performed in the periods 1978–1985 and 2007–2011 aiming at answering open questions on the groundwater recharge and dynamics, on connections between different types of aquifers and on solute transport. Environmental isotopes 2H, 18O, 3H, 13C of dissolved inorganic carbon and 14C were analysed in mineral, thermo-mineral and spring waters. Results indicated the source and mechanism of groundwater recharge, its renewability, a transit time distribution, hydraulic interrelationships, the groundwater origin and its evolution due to effects of water- rock interaction. They proved the hypothesis that the Boča massif should be a catchment area of the Rogaška mineral waters. The estimates of the mean residence time of mineral waters in the aquifer system are between 7200 and 3400 years, depending on the location and depth. Thermo-mineral water is the oldest water in the study area with the mean residence time of 14000 years. Izvleček V obdobjih 1978–1985 in 2007–2011 so bile izvedene izotopske raziskave podzemne vode, uskladiščene v razpoklinskem vodonosnem sistemu na območju Rogaške Slatine, z namenom, da se odgovori na odprta vprašanja o napajanju podzemne vode in njeni dinamiki, o povezavi različnih vrst vodonosnikov in o prenosu snovi. V mineralnih, termomineralnih in izvirskih vodah smo analizirali 2H, 18O, 3H, 13C v raztopljenem anorganskem ogljiku in 14C. V rezultatih smo opisali vire in mehanizme napajanja podzemne vode, njeno obnovljivost, zadrževalni čas v vodonosniku, izvor in geokemijski razvoj zaradi reakcij s kamninami. Potrdili smo hipotezo, da je masiv Boč napajalno območje Rogaških mineralnih vod, ki se zadržujejo v vodonosniku s povprečnim zadrževalnim časom od 7200 do 3400 let. Le ta je odvisen od lokacije in globine. Termomineralna voda je starejša – njen povprečen zadrževalni čas je 14000 let. GEOLOGIJA 60/1, 49-60, Ljubljana 2017 https://doi.org/10.5474/geologija.2017.004 © Author(s) 2017. CC Atribution 4.0 License Introduction Rogaška Slatina is famous by mineral water, which was discovered in this place in the time of ancient Rome. Numerous investigations of the Rogaška groundwaters were subject to balne- ology and to the larger exploitation quantities (nosan, 1975), whereas information essential for the definition of the Rogaška aquifer system and for its protection has been still missing. Questions on the groundwater recharge area and dynamics, on connections between aquifers and on solute transport have remained open, which depends on the field geology and structure. The latter is very complicated – three regional faults intersect in this area, which is folded to anticlinal and syn- clinal folds. The nature of geological structures, their mutual relations and extent have not been 50 Branka TRČEK & Albrecht LEIS explained at a satisfactory level in many parts of the system. With regard to results of previ- ous hydrogeochemical investigations (Nosan, 1973, 1975; Pezdič, 1986, 1997) it was presumed that the Boč massif near Rogaška Slatina is the catchment area of the Rogaška mineral waters, although geological data did not support this hy- pothesis (Aničić & Juriša, 1984). Aiming at answering the discussed open ques- tions also the isotopic investigations of ground- waters stored in the Rogaška Slatina fractured aquifer system were performed. The first studies took place in the period 1978–1985 (Pezdič, 1997), while the last studies were performed during a period 2007–2011 (TrčeK et al., 2010; TrčeK & Leis, 2011). The isotopic investigations based on the environmental isotopes of H, O and C, which were used as tracers of geological and hydrogeo- logical processes. The applications of stable isotope ratios of hydrogen and oxygen in groundwater hydrol- ogy are based primarily on isotopic variations in precipitation as the predominant groundwa- ter source. After the infiltration of precipitation into the aquifer, only physical processes, such as diffusion, dispersion, mixing and evapora- tion, alter the groundwater isotopic composition (ClarK & Fritz, 1997). The stable isotope content of water may be considered conservative under low-temperature and low-circulation ground- water systems, as long as the relative amount of water involved in chemical reactions remains limited (ClarK & Fritz, 1997; HoeFs, 1997). The exchange with oxygen (possibly also hydrogen) bearing minerals of the host rocks is particularly important in geothermal environment. With low reaction rates in low-temperature environments a long time is needed for a significant exchange to take place and equilibrium will generally not be reached (IAEA, 1983). Another process, which may modify the initial stable isotope content of groundwater, is the isotopic exchange with a gas phase which is not initially in equilibrium with the environmental water (e.g. CO2 or H2S). The stable carbon isotope composition of dis- solved inorganic carbon (δ13C-DIC) is not a con- servative tracer. Nevertheless, the δ13C can trace the carbon sources and reactions for numerous interacting organic and inorganic species. The isotope ratio 13C/12C is an important tool for quantifying the water-rock interactions, identi- fying the proportion of different CO2 sources in water, and determining the initial geological set- tings of the groundwater recharge (HoeFs, 1997; KenDall & McDonnell, 1998). Groundwater dating is the main field of ap- plication of the 14C and 3H radioactive isotopes. Their input source functions, which describe the time-varying global fluxes of isotopes deduced from atmospheric, cosmogenic and anthropo- genic production, are well known (MooK, 1980; HoeFs, 1997). The measured activity concentra- tions are compared with the input functions to get fairly informative age determinations over the past several decades. 3H has a half-life of 12.3 years and is a very applicable tracer for deter- mining spring residence times when recharge processes took place within a timescale of less than 50 years. Groundwaters seldom have more than 50 TU today and are typically in the 5–10 TU range (criss et al., 2007; rose, 2007). The re- cent 3H study (KOvačič, 2015) indicated that Slo- vene groundwaters could be divided into four categories – groundwaters that are older than 100 years (3H concentration is below the detec- tion limit), groundwaters with the prevailing old component (3H concentration is between 0 and 2.5 TU), groundwaters with the age between 30 and 60 years (3H concentration is on average 8 TU) and recent groundwaters with age up to 15 years (3H concentration is about 6 TU). On the other hand 14C with a half-life of 5,730 years is applicable for dating groundwaters re- charged prior to 1,000–2,000 years (Bajjali et al., 1997) and as old as 30,000 years (clarK & Fritz, 1997). Study area The study area with its broader surroundings is geologically one of the most complex parts of Slovenia (TrčeK et al., 2010). It is the juncture of three major regional fault systems, separat- ing three tectonic units (Figs. 1 and 2). The Boč massif belongs to the Southern Karavanke unit. It borders to the south with the Donat line to the narrow tectonic unit between the Donat line and Šoštanj fault close to which the town of Rogaš- ka Slatina is situated. To the north, the Dravinja fault as part of the Periadriatic line separates the Southern Karavanke unit from the Upper Aus- tro–Alpine unit. The ongoing dextral strike-slip zone along the Lavantal fault and the Šoštanj fault (Fig. 2) associates eastward extrusion of the Eastern Alps as a result of the northward shift 51Overview of isotopic investigations of groundwaters in a fractured aquifer system near Rogaška Slatina, Slovenia which are relatively displaced and folded (Figs. 1 and 2). Most of the area is composed of mas- sive limestones and dolomites of the Carnian age. They are capping the Boč mountain crest. The Oligocene and Miocene beds, covering the northernmost part of the territory and the lower slopes north of Rogaška Slatina, are composed of alternating sandstones, sands, shaly claystones and marlstones and conglomerates in the lower parts. A wide belt of volcanic rocks (andesite, its tuffs and volcanic breccias) is also present in within these units. The upper part is almost entirely composed of hard and bituminous marl- stones (aničić & Juriša, 1984). Mineral water is stored in fractured layers of the Oligocene tuff covered by the Upper Oligo- cene and Lower Miocene beds (Figs. 1 and 2). The discussed water belongs to the magnesium-sodi- um-hydrogen carbonate-sulphate facies. It was exploited from five boreholes that are 24 to 600 m deep (Fig. 1). Donat Mg is the most famous among Rogaška mineral waters. It has the highest min- eral content (Tab. 1) and contains more than 1,000 mg/l of Mg. The content of gaseous CO2 in the water is normally in the range of 2–30 g/l. Gas is practically pure CO2 with minor share of nitrogen and negligible concentration of oxygen and methane (Pezdič, 1997). Fig. 1. Study area with locations of boreholes with mineral, thermo–mineral and spring water (TrčeK et al., 2010). Sl. 1. Raziskovalno območje z lokacijami vrtin z mineralno, termomineralno in izvirsko vodo (TrčeK et al., 2010) and the counter clockwise rotation of the Adria microplate. The complexity of the study area is reflected in a lithological heterogeneity as a re- sult of faulting into many small tectonic blocks, Fig. 2. Geological cross section A-A’ (see Fig. 1) of the Rogaška Slatina area (TrčeK et al., 2010). Sl. 2. Geološki prerez A-A’ (glej sl. 1) območja Rogaške Slatine (TrčeK et al., 2010). 52 Branka TRČEK & Albrecht LEIS Borehole name Borehole depth (m) Q (l/s) M (g/l) SEC (μS/cm) T (ºC) pH Mineral water RSL-2 274 0.5 12.75 10999 14.6 6.8 RSL-3 24 0.2 5.8 5165 11.9 6.4 RSL-6 606 1 11.8 10539 28.4 6.9 RSL-7 603 0.4 8.65 8627 15.1 6.5 RSL-11 170 0.1 8.08 7102 12.0 6.5 Thermo- mineral water RSL-1 1700 6 6 6418 55.4 7.1 Spring water from limestone RSL-4 215 2.2 459 12.6 7.1 RSL-10 130 2.1 379 11.0 7.5 Spring water from dolomite RSL-8 170 1.4 0.5 570 11.7 7.3 RSL-14 38 40 481 12.9 7.5 Spring water from sandstone RSL-12 100 2 382 9.9 7.7 Spring water and thermo–mineral water are also exploited at the study area (Fig. 1, Tab. 1). The former is stored in fractured Triassic car- bonate rocks and Miocene sandstones of the Boč massif and the latter in the dolomitized kerato- phyre at depths between 1,500 and 1,700 m. The Rogaška mineral waters discharge with a help of a gas lift, thermo–mineral water dis- charges with a help of a thermo lift, while spring waters of boreholes RSL-4, RSL-8, RSL- 10 and RSL-14 are artesian. Methods Isotopic investigations were fundamental for studies of groundwater dynamics and solute transport in aquifers with mineral and spring water and among them. The recent investigations included the monitoring of 5 boreholes with mi- neral water (RSL-2, RSL-3, RSL-6, RSL-7 and RSL-11), 1 borehole with thermo–mineral wa- ter (RSL-1), 2 boreholes with spring water from a limestone (RSL-4, RSL-10), 2 boreholes with spring water from a dolomite (RSL-8, RSL-14) and 1 borehole with spring water from a san- dstone (RSL-12; Fig. 1, Tab. 1). Besides the grou- ndwater CO2 was sampled and also precipitation at altitudes of 340, 530 and 710 m asl. The previ- ous investigations involved only the sampling of mineral waters and spring water from carbonate rocks (Pezdič, 1997). Recent isotopic analyses of stable isotopes of H and O were made in Laboratory Centre for Iso- tope Hydrology and Environmental Analytics, Joanneum Research, Graz, Austria. The oxygen isotopic composition (δ18O) of the water was mea- sured by the classic CO2 – H2O equilibrium tech- nique (EPstein & MayeDa, 1953) with a fully au- tomated device adapted from Horita et al. (1989) coupled to a Finnigan DELTAplus Mass Spectro- meter. Deuterium (δ2H) was measured in a conti- nuous flow mode by chromium reduction using a ceramic reactor slightly modified from Morrison et al. (2001). Stable oxygen and hydrogen isotopic ratios are reported relative to the VSMOW (Vi- enna-SMOW) standard with an overall precision of 0.1 and 1 ‰, respectively. The data of H and O stable isotopic composi- tion in precipitation were applied to define the local meteoric waterline (LMWL) of the Rogaš- ka Slatina area. Due to small number of samples (57) the ordinary linear regression analysis was conducted that based on the method of seasonally weighted mean values (rozansKi et al., 1993). Table 1. Average values of sampled water discharges (Q), electroconductivity (SEC), mineralisation (M), temperature (T) and pH in the period 2007-2011 Tabela 1. Povprečne vrednosti pretoka vzorčenih vod ter njene elektroprevodnosti (SEC), mineralizacije (M), temperature (T) in pH v obdobju 2007-2011. 53Overview of isotopic investigations of groundwaters in a fractured aquifer system near Rogaška Slatina, Slovenia The 13C content was also measured in the Sta- ble Isotope Laboratory of the Institute of Water Resources Management, Joanneum Research, Graz, mass spectrometrically by Thermo Fin- nigan DELTAplusXP (CF-IRMS). The values are reported as ‰ vs. V-PDB standard with an over- all precision of 0.1 ‰. Measurements of the radioactive isotopes 3H and 14C were performed in the Isotope laboratory HYDROSYS – Water and Environmental Protec- tion Developing Ltd, Budapest, Hungary. 3H was measured by counting β-decay events in a liquid scintillation counter (LSC). Values are report- ed in absolute concentrations as tritium units (TU), where one TU corresponds to one 3H atom per 1018 atoms of hydrogen 1H or to an activity of 0.118 Bq/kg in water. 14C was measured by the super low level liq- uid scintillation analyser (PerkinElmer Tri-Carb 3170TR/SL) based on ASTM D6866-06 standard. The values are expressed in pmC (Percent Mod- ern Carbon). Groundwater age (t) is determined using the radioactive decay equation (Eq. 1), em- ploying the pre-industrial baseline 14C activity of the soil CO2 equal to 100 pmC as an initial con- centration and the 14C activity of the DIC (HCO3 ¯) measured in the groundwater sample. However, the dissolved C mass balance of groundwaters is altered during the groundwater flow path due to geochemical reactions that take place within the aquifer. In such cases, a factor q is incorporated into the radioactive decay equation to take into account the 14C dilution effect (other than radio- active decay) during the groundwater flow path caused by a dissolution of 14C–free marine car- bonates and incorporation of a geogenic 14C–free CO2 of the magmatic or metamorphic source. The groundwater age is thus determined: t = −8267ln at 14C qa0 14C         (Eq. 1) where t is groundwater age, a0 14C is a modern 14C activity in the soil zone (100 pmC), at 14C is a 14C activity of DIC in the groundwater sample and q is a dilution factor. The chemical mass-balance correction (CMB model) has been used to quantify the 14C dilution due to incorporation of 14C–free geogenic CO2, assuming that the carbonate dissolution in the recharge area evolves under the closed system condition. In such cases the 14C dilution factor of carbonate groundwaters is about 0.5. The subse- quent 14C dilution caused by the incorporation of 14C–free geogenic CO2 and resulting geochemi- cal reactions has been determined as a ratio of the HCO3 - concentration in the recharge area to the HCO3 - concentration in samples of mineral and thermos-mineral waters. The HCO3 - concen- tration in the recharge area is given as an aver- age value of sampled spring waters that equals to 304 mg/l. The total 14C dilution factor is thus the product of the 14C dilution ensuing in the re- charge area (0.5) and the subsequent 14C dilution induced by the incorporation of 14C-free geogenic CO2. The first analyses of O, H and C environmen- tal isotopes were made at Institute Jožef Stefan, Ljubljana (Pezdič, 1997). For δ18O determination from water the classic CO2 – H2O equilibrium technique was also applied (EPstein & MayeDa, 1953). Hydrogen gas was prepared by reducing water vapor on hot zinc wool at 400 °C or on zinc granules at 490 °C (coleMan et al., 1982). Precip- itated DIC and carbonates reacted with 100 % orthophosphoric acid at 55 ºC. Dissolved car- bonate species are treated with phosphoric acid to measure just δ13C. The isotopic composition of prepared gaseous compounds (H2 and CO2) were measured on the Varian MAT 250 mass spec- trometer. Measuring accuracy exceeds ± 0.05 ‰ for oxygen and carbon and ± 0.5 ‰ for hydrogen. Results were given relative to the SMOW stan- dard for oxygen and hydrogen and PDB for car- bon. The tritium content was measured with a liq- uid scintillation counter at Institute Jožef Stefan, Ljubljana (Pezdič, 1997). Concentrations were given in TU and Bq. The activity of radiogenic carbon (14C) was measured at Rudjer Boikovic In- stitute in Zagreb using a gas proportional count- er. 14C data were presented as % Modern Carbon (pmC) vs. the activity of standard oxalic acid (Ox) formed before the nuclear period with a correc- tion factor of 0.95 (Pezdič, 1997). Results and discussion Average values of sampled water discharges, electroconductivity, mineralisation, temperature and pH during a monitoring period 2008–2010 are presented in Table 1. The level of gaseous CO2 in water is normally in the range 2-30 g/l, but in some areas is as high as 40 g/1 (nosan, 1973). Gas analysis revealed that the gas is practically pure 54 Branka TRČEK & Albrecht LEIS CO2 with a minor share of nitrogen (0.3 %) and negligible concentrations of oxygen and methane (below 0.01 %; Pezdič, 1997). The relationship between O and H stable iso- topic compositions of sampled water is illustrat- ed in Figure 3. The average groundwater O and H isotopic composition approximately equals the weighted average of the precipitation O and H composition as a rule, while surface water is may be enriched more with heavier (more positive) isotopes than precipitation, due to the evapora- tion process (clarK & Fritz, 1997; rozansKi et al., 1992, 1993; TrčeK & zOJer, 2010). Mineral waters have 18O in the range between -12.2 and -10.3 ‰ and δ2H in the range between -80.5 and -63.9 ‰. The lowest δ values refer to wa- ters with the highest mineralization, RSL-2 and RSL-6, and the highest to water of RSL-3 with the lowest mineralisation (Tab. 1), which reflects mixing processes of older and younger waters (Fig. 3). The δ18O and δ2H values range in spring waters from carbonate rocks is -10.7 to -9.9 ‰ and -78.9 to -65.0 ‰, respectively. It is slightly different and more narrow in spring water from sandstone, -10.9 to -10.6 ‰ and -76.9 to -67.2 ‰, respectively (Fig. 3). The δ18O and δ2H of thermo–mineral wa- ter is similar to spring water from a dolomite. Similar δ18O and δ2H results were obtained in previous investigations (Pezdič, 1997). Miner- al waters had a δ18O range -11.2 to -9.55 ‰ and a δ2H range -84.0 to -76.9 ‰, while the parameters ranged in spring waters from carbonate rocks from -10.25 to - 8.19 ‰ and from -78.6 to -63.6 ‰, respectively. CraiG (1961) firstly described the relationship of discussed isotopes in the precipitation by the global meteoric waterline (GMWL, Fig. 3). The local meteoric waterline (LMWL) of the Rogaška Slatina area slightly differs from the GMWL in both the slope and the intercept (Fig. 3). It is a fun- ction of temperature during the secondary evapo- ration as rain falls from a cloud, which results in effects of the isotopic fractionation with respect to the latitude, altitude, and climate. The isoto- pic composition of precipitation is affected by the season, latitude, altitude, precipitation amount, and distance from the coast (rozansKi et al., 1992, 1993). All sampled waters from recent and pre- vious investigation (Pezdič, 1997) are distributed in the LMWL vicinity. The δ18O and δ2H values Fig. 3. The relationship between stable oxygen and hydrogen isotopic composition of sampled water in the period 2007-2011. Sl. 3. Razmerje med sestavo stabilnih izotopov kisika in vodika v vzorčenih vodah v obdobju 2007-2011. 55Overview of isotopic investigations of groundwaters in a fractured aquifer system near Rogaška Slatina, Slovenia and their distribution indicate that sampled wa- ters are of a meteoric origin, hence the Rogaška fractured aquifer system is recharged with local precipitation. The δ values of mineral waters fall on the same LMWL, but on its lower part with the depleted heavy isotopes (Fig. 3, Pezdič, 1997), which reflects colder climate conditions during the infiltration processes (clarK & Fritz, 1997; HoeFs, 1997; iaea, 1983). The exception is the shal- low RSL-3 mineral water (Tab. 1) that is mixed with surface waters (Figs. 3 and 4). The δ values of mineral waters sampled from RSL-2, RSL-6 and RSL-3 slightly deviate from LMWL (Fig. 3). The level of gaseous CO2 in these waters is the highest (up to 40 g/1; nosan, 1973), therefore it is assu- med that 18O is depleted due to the exchange of water with the geogenic CO2 (clarK & Fritz, 1997; HoeFs, 1997; iaea, 1983). During the period of previous isotopic inves- tigations (1978–1985) the 3H contents of 1.9 to 176 TU were measured in mineral waters and of 37 to 150 TU in spring waters from carbonate rocks (Pezdič, 1997). The 3H data of investigations in July 2008, when almost three half-lives of 3H has passed, are illustrated in Figure 4. The precipita- tion had the 3H content of 10.7 TU, which coincides with the summer values measured in Ljubljana (KOvačič, 2015). 3H was not detected in mineral waters from boreholes RSL-6 and RSL-7, hence these waters are not in contact with recent pre- cipitation infiltration. According to KOvačič (2015) mineral waters from boreholes RSL-6 and RSL- 7 are older than 100 years. The samples of ther- mo–mineral and mineral waters from boreholes RSL-1, RSL-2 and RSL-11 contain 0.6 TU or less 3H, which indicates the prevailing old water com- ponent (KOvačič, 2015). Evidently the recharge of these groundwaters occurred prior to the 1950s, therefore they are relatively unblemished by hu- man activities. Similar findings were published in previous investigations (Pezdič, 1997). Groundwater samples from RSL-3, RSL-4, RSL-8, RSL-10, RSL-12 and RSL-14 with tri- tium contents 4.6 to 9.4 TU (Fig. 4) should con- tain modern water that infiltrated predominantly after the 1960s, suggesting the vulnerability of these groundwater systems to man–made im- pacts. According to KOvačič (2015) the mean resi- dence time of spring water of shallower boreholes RSL-10, RSL-12 and RSL-14 (Tab. 1) should be around 15 years, around 30 years for spring water of the deeper borehole RSL-4 and up to 60 years for spring water of the deeper borehole RSL-8. Among listed waters only RSL-3 is mineralized. The water is captured from a depth of 20 m, where it should be mixed with young fresh water. Pezdič (1997) reported that spring waters from carbonate rocks are no more than 15 years old. During the period 2007–2011 the δ13C values of the dissolved inorganic carbon varied between -13 and +2 ‰ in sampled groundwater (Figs. 5 and 6). The parameter values between -2 and +2 ‰ are characteristic for groundwaters that are influ- enced by the volcanic CO2: RSL-2, RSL-3, RSL-6, RSL-7 and RSL-11 (clarK & Fritz, 1997). These waters are highly mineralized, as it is evidenced in Table 1. The RSL-1 water has lower mineral- ization, which is reflected in δ13C-DIC values. On the other hand low δ13C values (-12 to -13 ‰) are typical for spring waters, collected from RSL-4, RSL-8, RSL-10 and RSL-14. According to clarK & Fritz (1997) these values closely resemble the δ3C content of carbonate groundwater that evolves under a closed system condition. During the period 1978–1985 similar δ13C val- ues were detected in mineral waters, between -3.1 and +3.3 ‰, and between -16.9 and 13.1 ‰ in spring waters of carbonate rocks (Pezdič, 1997). The main sources of the carbon dissolved in groundwater are soil CO2, CO2 of a geogenic or- igin or a magmatic CO2 (from deep crustal or mantle sources), carbonate minerals, an organic matter in soils and rocks, fluid inclusions, and methane. Each of these sources has a different carbon isotopic composition and contribute to the totally dissolved carbon in various proportions. In the studied groundwater systems the total dis- solved inorganic carbon (DIC) exists practically Fig. 4. Tritium concentrations in groundwater and in preci- pitation of the Rogaška Slatina area in the period 2007–2011. Sl. 4. Koncentracije tricija v podzemnih vodah in padavinah na območju Rogaške Slatine v obdobju 2007–2011. 56 Branka TRČEK & Albrecht LEIS all in a HCO3 - form. The HCO3 - concentration var- ies between 224–382 mg/l in groundwater sam- ples from RSL-4, RSL-8, RSL-10, RSL-12 and RSL-14, where the carbonate rocks were disso- lute in the reaction with biogenic CO2 in soil. The HCO3 - concentration of mineral and thermo–min- eral waters from RSL-1, RSL-2, RSL-3, RSL-6, RSL-7 and RSL-11 is increased to about 1950– 8280 mg/l (Tab. 2). The hydrochemical composi- tion was most probably altered to a great extent by the incorporation of the geogenic CO2 influx that originates from a volcanic source (Fig. 5). The average δ13C of CO2 in mineral waters is -4.1 ‰ and -9.9 ‰ in thermo–mineral water, re- spectively. It induced the intensive rock-water in- teractions and consequently the dissolved solute contents increased significantly. It is well known that the δ13C is about -25 ‰ in the soil CO2 (simi- lar to plants), -7 ‰ in the atmospheric CO2, -8 ‰ to -3 ‰ in the CO2 originating from the geothermal and volcanic systems and about 0 ‰ in marine carbonate rocks (KenDall & McDonnell, 1998). δ values from -2.0 ‰ to +4.1 ‰, with an average of +2.2 ‰ were analysed in Slovene carbonate rocks (Koceli et al., 2013; oGorelec et al., 2000). The re- lationship between δ13C and HCO3 - concentration in sampled water is illustrated in Figure 6. The 14C investigation was done on five ground- water samples of mineral and thermo–mineral waters collected from the boreholes RSL-1, RSL- 2, RSL-3, RSL-6 and RSL-7 in 2009. Very low 14C contents, ranging between 0.9 to 1.4 pmC, have been determined (Tab. 2), which reflects very long mean residence times of groundwaters. With the aim to determine the resident time of ground- water according to Equation 1 the geogenic CO2 was characterised first based on the δ13C value of the CO2 gas phase of water samples. The δ 13C values show a very wide spread, between -10 and -3 ‰ (Fig. 5). They are within the range of δ13C reported in the literature for magmatic CO2, -12 to -1 ‰ (truesDell & huston, 1980; eXley et al., 1986; rOSe & daviSSOn, 1996). Further the chem- ical mass-balance correction (CMB model) has been used to quantify the 14C dilution due to in- corporation of the 14C–free volcanic CO2, assum- ing that the carbonate dissolution in the recharge area evolves under a closed system condition. From stoichiometry in such cases the initial 14C concentration (~ 100 pmC) is expected to become diluted for about 50 %. It follows that the 14C dilution factor is about 0.5. The subsequent 14C dilution caused by the incorporation of the 14C– free volcanic CO2 and the resulting geochemical reactions has been determined as a ratio of the HCO3 - concentration acquired in the recharge zone to the HCO3 - concentration in the ground- water samples from RSL-1, RSL-2, RSL-3, RSL- 6 and RSL-7. The HCO3 - concentration acquired in the recharge zone is given as an average value from RSL-4, RSL-8, RSL-10 and RSL-14, which is about 304 mg/l. The ages of sampled mineral Fig. 5. Stable C isotopic composition of total dissolved inorganic carbon (DIC) in sampled water and its CO2 in the period 2007–2011. Sl. 5. Izotopska sestava 13C raztopljenega anorganskega ogljika v vzorčenih vodah in v njihovem CO2 v obdobju 2007–2011. Fig. 6. The relationship between stable C isotopic compositi- on and HCO3 - concentration in sampled waters in the period 2007–2011. Sl. 6. Razmerje med izotopsko sestavo 13C in koncentracijo raztopljenega anorganskega ogljik v vzorčenih vodah v ob- dobju 2007–2011. 57Overview of isotopic investigations of groundwaters in a fractured aquifer system near Rogaška Slatina, Slovenia waters were calculated employing the total 14C dilution factor, which is the product of the 14C di- lution ensuing in the recharge zone (= 0.5) and the subsequent 14C dilution induced by the incorpo- ration of the 14C–free volcanic CO2. Table 2. HCO3 – concentration, stable C isotopic compositions, 14C content, dilution factor qtotal and age estimates of the min- eralized waters and thermos mineral water (RSL-1) sampled in 2009. Tabela 2. Koncentracije HCO3 –, izotopska sestava 13C, vseb- nost 14C, faktor razredčenja qtotal in ocena starosti mineralne in termomineralne vode, vzorčene leta 2009. Sampling point HCO3 – (mg/l) δ13C (‰) 14C (pmC) qtotal Age (year) RSL-1 1949.6 -4.37 1.43 0.078 14000 RSL-2 8176.7 2.05 1.23 0.019 3400 RSL-3 3771.0 0.53 0.95 0.040 ---- RSL-6 7279.7 1,99 0.88 0.021 7100 RSL-7 4875.5 1,75 1.31 0.031 7200 The estimated ages of mineral waters are list- ed Table 2. The thermo–mineral water, captured from RSL-1 at depths 1500–1700 m exhibits a considerably high mean residence time – 14000 years, corresponding to the Pleistocene epoch, the Bølling-Allerød interglacial period that lasted from 14700 to 12900 years ago. During this peri- od glaciers almost disappeared in many Europe- an mountain ranges, also in the Alps (Palacios et al., 2016; DielForDer & hetzel, 2014; hiPPe et al., 2014; ivy-ochs, 2015). The wormer climate during the infiltration process of thermo–mineral water is reflected also in the stable isotopic composition of O and H (Fig. 3). The mean residence time of mineral waters RSL-6, RSL-7 and RSL-2 is 7200, 7100 and 3400 years, respectively. The first two correspond to the Holocene epoch, the end of the colder Bore- al period that lasted from 8600 to 7200 years ago (roBerts, 2014). According to the 18O and 2H iso- topic data (Fig. 3) waters from RSL-6 and RSL-2 should be recharged and discharged under simi- lar conditions, but they are captured at different depths (606 and 274 m respectively, Tab. 1) and hence fallow different flow paths, which should result in age differences. It is also worth to note that despite the low 14C concentration in the groundwater sample from RSL-3, no groundwater age was calculated due to the presence of a higher tritium concentration (4.6 TU) in the groundwater sample. Most likely it represents mixing of recent and old groundwater, thus groundwater age dat- ing using 14C is not appropriate. The 14C was investigated in the RSL-7 miner- al water also in 1980s (Pezdič, 1997). A content of 2 pmC was recorded. Two corrections were made to estimate the water mean residence time based on 14C data: a) according to a dilution factor deter- mined as the ratio between the predicted active and measured total DIC (equals to 0.037), which indicated that the mixture had about 3.7 pmC of initial 14C activity (ICA) and measured 2 pmC and b) according to estimation that primary miner- al water had become mixed with about 20 % of young meteoric water. The recalculation of val- ues indicated the relative 14C activity of prima- ry RSL-7 mineral water to be approximately 35 pmC. This pointed to an estimated age of about 8000 years, which is compatible with the result of recent investigations. Conclusions The results of isotopic investigations of groundwaters in the Rogaška Slatina area per- formed in the period 1978–1985 (Pezdič, 1997) and 2007–2011 coupled with available information on a physical hydrogeology and a water chemistry help in understanding the source and mechanism of groundwater recharge, groundwater circu- lation and its renewability, groundwater transit time distribution, hydraulic interrelationships, the groundwater origin and its evolution due to effects of water-rock interaction. The δ18O and δ2H values indicated the meteor- ic origin of mineral and thermos-mineral waters and proved the hypothesis that the Boč massif near Rogaška Slatina should be a catchment area of the Rogaška mineral waters. They reflected the mixing processes between younger and older waters or mineral and spring waters and hence contributed to the water body vulnerability as- sessment together with 3H data. The latter point- ed out waters that infiltrated predominantly after the 1960s, suggesting the vulnerability of these groundwater systems to man-made impacts. The estimations of the mean residence time of mineral, thermos-mineral and spring waters based on 3H, δ13C and 14C data provided additional information on groundwater recharge processes, 58 Branka TRČEK & Albrecht LEIS its renewability and vulnerability. Thermo-min- eral water, captured from RSL-1 at depths 1500– 1700 m is the oldest water in the study area with the mean residence time of 14000 years. The mean residence time of mineral waters captured at depth around 600 m from RSL-6 and RSL-7 is estimated to 7100 and 7200 years respectively. However, the residence time of mineral water captured at depth around 270 m from RSL-2 is shorter – 3400 years. Nevertheless, the 18O and 2H data indicated that mineral waters from RSL-6 and RSL-2 should be recharged and discharged under similar condi- tions. Based on 3H data it was estimated that shal- low spring waters are around 15 years old, while the age of deeper ones is 30 to 60 years. The presented findings were integrated with the structural analysis of the study area (TrčeK et al., 2010; ŽibreT, 2016). The results improved a conceptual hydraulic model of the Rogaška Sla- tina fractured aquifer system and are essential to determine the optimal balance between envi- ronmental protection and economic use of mine- ral and spring water resources in the study area. Acknowledgement The author would like to acknowledge the Droga Kolinska, Živilska industrija d.d., Slovenian Research Agency and Knet Water for the financial support of the researches. Pregled izotopskih raziskav podzemne vode v razpoklinskem vodonosnem sistemu na območju Rogaške Slatine (daljši povzetek) V obdobjih 1978–1985 in 2007–2011 so bile izvedene izotopske raziskave podzemne vode, uskladiščene v razpoklinskem vodonosnem sis- temu na območju Rogaške Slatine z namenom, da se odgovori na odprta vprašanja o napajanju podzemne vode in njeni dinamiki, o povezavi različnih vrst vodonosnikov in o prenosu snovi. V mineralnih, termomineralni in izvirskih vo- dah so se analizirali 2H, 18O, 3H, 13C-DIC in 14C (sl. 1). Mineralna voda je uskladiščena v razpo- kanih plasteh oligocenskega tufa, ki jih prekri- vajo zgornje oligocenske in spodnje miocenske kamnine, izvirska voda pa v razpokanih zgornje triasnih karbonatnih Boča in miocenskih pešče- njakih (sl. 1 in 2, tab. 1). Termomineralna voda je zajeta v vrtini RSL-1 na globinah 1500 do 1700 m, v dolomitiziranem keratofirju. Razmerje med izotopsko sestavo 18O in 2H v vzorčenih vodah na območju Rogaške Slatine je predstavljeno na sliki 3. Lokalna meteorna premica (LMWL) ne odstopa dosti od globalne (GMWL). Vrednosti parametrov, izmerjenih v površinskih in podzemnih vodah, so razporejene v območju, ki ga pokrivajo vrednosti, izmerjene v padavinah. Na podlagi tega in sorodnih rezul- tatov predhodnih raziskav (Pezdič, 1997) lahko zaključimo, da so vse vzorčene vode meteornega izvora. Na sliki 3 odstopajo nekoliko vrednosti mineralnih vod, vzorčenih na vrtinah RSL-2, RSL-6 in RSL-3. Možnih je več vzrokov: drugač- ne klimatske razmere v času infiltracije padavin, drugačni pogoji napajanja vodonosnika in izo- topska izmenjava s CO2 v mineralni vodi. Razbe- remo, katere vode imajo podoben režim napajanja ali podobno napajalno območje. Med mineralni- mi vodami je treba izpostaviti podobnost med vodami vzorčnih mest RSL-2 in RSL-6 oziroma RSL-11 in RSL-7. Iz tega sledi, da pripada par najverjetneje istemu vodonosniku. Termomine- ralna voda vrtine RSL-1 ima podobno izotopsko sestavo 18O in 2H kot dolomitne vode vrtin RSL-8 in RSL-14, kar kaže na podobno nadmorsko viši- no napajalnih zaledij vodonosnikov. Rezultati analiz 3H v vzorčenih vodah so pri- kazani na sliki 4. Mineralne in termomineralne vode vrtin RSL-1, RSL-2, RSL-6, RSL-7 in RSL- 11 vsebujejo 0,6 TU ali manj, kar pomeni, da so se napajale pred letom 1960. Po drugi strani lahko sklepamo, da so omenjene vode zavarovane pred antropogenimi vplivi iz površja. Vsebnost 3H v mineralni vodi RSL-3, več kot 4 TU, opozar- ja da se meša mineralna voda v vrtini z mlajši- mi vodami, ki dotekajo s površja. Izvirske vode vrtin RSL-3, RSL-4, RSL-8, RSL-10, RSL-12 in RSL-14 vsebujejo 4,6 do 9,4 TU, kar pomeni, da so mlajše od 60 let. Ocenjujemo, da je starost pli- tvejših izvirskih vod okoli 15 let, medtem ko so globlje stare med 30 in 60 let. Na podlagi podatkov δ13C, HCO3 - in 14C se je določila starost mineralnih in termo mineral- nih vod (sl. 5 in 6, tab. 2). Uporabila se je enačba radioaktivnega razpada (enačba 1), ki se ji doda faktor q, s pomočjo katerega se upošteva učinek razredčenja 14C zaradi raztapljanja karbonatnih kamenin in geogenega CO2 (vulkanskega izvo- ra). Izotopska sestava 13C-DIC v vzorčenih vo- dah je predstavljena na sliki 5. Le ta izpostavlja mineralne vode, ki so pod vplivom vulkanskega CO2: RSL-2, RSL-3, RSL-6, RSL-7 in RSL-11. Omenjene vode so visoko mineralizirane, kar je 59Overview of isotopic investigations of groundwaters in a fractured aquifer system near Rogaška Slatina, Slovenia evidentirano v tabeli 1. Termomineralna voda RSL-1 ima nižjo mineralizacijo, kar se kaže tudi v vrednosti δ13C-DIC. V izvirskih vodah RSL-4, RSL-8, RSL-10 in RSL-14 so bile izmerjene vred- nosti δ13C-DIC med -13 in -14 ‰. Te vrednosti so značilne za zaprt sistem raztapljanja karbonat- nih kamnin v območju napajanja vodonosnika. Za določitev vpliva vulkanskega CO2 na mer- jene vrednosti 14C se je uporabil model korekcije kemijske masne bilance (CMB model) ob predpo- stavki, da poteka proces raztaplanja karbonatnih kamenin v napajanem območju vodonosnika pod zaprtimi pogoji. V takih primerih se upošteva, da se zmanjša začetna koncentracija 14C (~ 100 pmC) za okoli 50 %. Iz tega sledi, da je faktor razred- čenja 14C v karbonatnih podzemnih vodah raz- iskovalnega območja okoli 0,5. Dodatno razred- čenje 14C povzroči še vulkanski CO2. Le-to se je ocenilo s pomočjo razmerja med koncentracijo HCO3 - v vodah z območja napajanja vodonosnega sistema (karbonatne izvirske vode) in koncentra- cijo HCO3 - v mineralnih in termo mineralnih vo- dah RSL-1, RSL-2, RSL-3, RSL-6 in RSL-7. Za koncentracijo HCO3 - na območju napajanja vodo- nosnika z mineralno vodo se je upoštevala pov- prečna vrednost vzorčenih vod RSL-4, RSL- 8, RSL-10 in RSL-14, 304 mg/l. Opisani podatki in informacije so se upošteva- li pri določitvi starosti mineralnih vod, podanih v tabeli 2. Najdaljši zadrževalni čas, 14000 let, je značilen za termomineralno vodo RSL-1, ki je zajeta v globinah 1500-1700 m. Povprečni zadrže- valni časi mineralnih vod RSL-6, RSL-7 in RSL- 2 so 7200, 7100 oziroma 3400 let. Glede na podat- ke δ18O in δ2H (sl. 3) imata vodi RSL-6 in RSL-2 isto napajalno območje in pripadata istemu vo- donosnemu sistemu, kjer potekajo geokemijske reakcije pod podobnimi pogoji. Ker pa se zajema- ta obravnavani vodi na različnih globinah, 606 oziroma 274 m, je različen njun zadrževalni čas v vodonosniku. V predhodnih raziskavah je bilo ocenjeno, da je zadrževani čas mineralne vode RSL-7 okoli 8000 let (Pezdič, 1997). Rezultati izotopskih raziskav so opisali vire in mehanizme napajanja podzemne vode, njeno ob- novljivost, zadrževalni čas v vodonosniku, izvor in geokemijski razvoj zaradi reakcij s kamninami. Potrdili so hipotezo, da je masiv Boča napajalno območje Rogaških mineralnih vod, ki se zadržuje- jo v vodonosniku s povprečnim zadrževalnim ča- som od 7200 do 3400 let, ki je odvisen od lokacije in globine. References aničić, B. &Juriša, M. 1984: Basic geological map SFRJ, 1:100.000, Sheet Rogatec. Zvezni geolo- ški zavod, Beograd. Bajjali, W., ClarK, I. & Fritz, P. 1997: The ar- tesian thermal groundwaters of northern Jordan: insights into their recharge history and age. J. Hydrol., 192: 355–382, doi:10.1016/ S0022-1694(96)03082-X. ClarK, I. & Fritz, P. 1997: Environmental Isotopes in Hydrogeology, Lewis Publishers. Criss, R. E., Davisson, M. L., SurBecK, H., & Winston, W. E. 2007: Isotopic Techniques. In: GolDscheiDer, N. & Drew, D. (eds.): Methods in karst hydrogeology, International con- tribution to hydrogeology 26. Taylor and Francis, London: 123–145. coleMan, M.L., shePherD, t.j., DurhaM, j.j., rouse, j.e. & Moore, G.R. 1982: Reduction of wa- ter with zinc for hydrogen isotope analysis. Anal. Chem., 54/6: 993–995, doi:10.1021/ac00243a035. DielForDer, a. & hetzel, R. 2014: The deglaciation history of the Simplon region (southern Swiss Alps) constrained by 10Be exposure dating of ice-molded bedrock surfaces. Quat. Sci. Rev, 84: 26–38, doi:10.1016/j.quascirev.2013.11.008. EPstein, S. & MayeDa,T. 1953: Variation of O18 content of waters from natural sources. Geochimica et Cosmochimica Acta, 4/5: 213– 224, doi:10.1016/0016-7037(53)90051-9. EXley, R.A., Mattey, D.P., ClaGue, D.A. & PillinGer, C.T. 1986: Carbon isotope syste- matics of a mantle “hotspot”: a comparison of Loihi Seamount and MORB glasses. Earth and Planetary Science Letters, 78/2-3: 189– 199, doi:10.1016/0012-821X(86)90060-9. HoeFs, J. 1997: Stable isotope geochemistry, fo- urth ed. Springer-Verlag, Berlin-Heidelberg. Horita, J., UeDa, A., MizuKaMi, K. & TaKatori, I. 1989: Automatic δD and δ18O analyses of multi-water samples using H2 and CO2-water equilibration methods with a common equil- ibration set-up. Appl. Radiat. Isot., 40/9: 801– 805, doi:10.1016/0883-2889(89)90100-7. hiPPe, K., ivy-ochs, s., KoBer, F., zasaDni, j., wieler, r., wacKer, l., KuBiK, P.w. & schlüchter, c. 2014: Chronology of Lateglacial ice flow reor- ganization and deglaciation in the Gotthard Pass area, Central Swiss Alps, based on cos- mogenic 10Be and in situ 14C. Quat. Geochronol, 19: 14–26, doi:10.1016/j.quageo.2013.03.003. international atoMic enerGy aGency 1983: Isotope techniques in the hydrogeological assessment of potencial sites for the disposal of high-level radioactive wastes. Technical 60 Branka TRČEK & Albrecht LEIS reports series 228, International Atomic Energy Agency, Vienna: 151 p. ivy-ochs, S. 2015: Glacier variations in the European Alps at the end of the last glacia- tion. Cuadernos de Investigación Geográfica, 41/2: 295–315, doi:10.18172/cig.2750. KenDall, C. & McDonnell, J.J. 1998: Isotope tracers in catchment hydrology. Elsevier, Amsterdam. Koceli, A., Kanduč, T. & verBovšeK, T. 2013: Anorganski ogljikov cikel v sistemu tla-kamnina-podzemna voda v kraško-raz- poklinskih vodonosnikih. Geologija, 56/2: 219–228, doi:10.5474/geologija.2013.014. KOvačič, K. 2015: Radioactive isotopes in ground waters of Slovenia. Ph.D. Thesis. University of Nova Gorica Graduate School, Nova Gorica: 127 p. Márton, E., FoDor, L., Jelen, B., Márton, P., RiFelj, H. & Kevrić, R. 2002: Miocene to Quaternary deformation in NE Slovenia: complex pa- leomagnetic and structural study. Journal of Geodynamics, 34: 627–651, doi:10.1016/ S0264-3707(02)00036-4. MooK, W.G. 1980: Carbon-14 in hydrogeo- logical studies. In: Fritz, P. & Fontes, J. CH. (eds.): Handbook of Environmental Isotope Geochemistry 1A, The Terrestrial Environment, 75–134. Morrison, J., BrocKwell, T., Merren, T., Fourel, F. & PhiliPs, A.M. 2001. On-line high-precision stable hydrogen isotopic analyses on nanoli- ter water samples. Anal. Chem., 73/15: 3570– 3575, doi:10.1021/ac001447t. Nosan, A. 1973: Thermal and mineral wells in the Slovenia = Termalni in mineralni vrelci v Sloveniji). Geologija, 16: 6–81. Nosan, A. 1975: Hydrogeological investigations of mineral waters in Rogaška Slatina from 1952 to 1972. Geological Survey of Slovenia, Ljubljana: 44 p. OGOrelec, b., dOlenec, T. & Pezdič, J. 2000: Izotopska sestava O in C v mezozojskih karbo- natnih kamninah Slovenije - vpliv faciesa in diageneze. Geologija, 42: 171–205, doi:10.5474/ geologija.1999.012. Palacios, D., GoMez-ortiz, a., anDres, n., SalvadOr, F. & Oliva, m. 2016: Timing and new geomorphologic evidence of the last deglacia- tion stages in Sierra Nevada (southern Spain). Quat. Sci. Rev., 150: 110–129, doi:10.1016/j. quascirev.2016.08.012. Pezdič, J. 1997: Recharge and retention time study of a partly karstified area of Boč (Eastern Slovenia) using hydrogen, oxygen and carbon isotope composition as natural tracers. Isotopes Environ. Health Stud., 33: 293–306, doi:10.1080/10256019708234040. roBerts, N. 2014: The Holocene, the Environmental History, third edition. John Wiley & sons: 358 pp. Rose, S. 2007: Utilization of decadal tritium variation for assessing the residence time of base flow. Ground Water, 45/3: 309–317, doi:10.1111/j.1745-6584.2006.00295.x. Rose, T.P. & Davisson, M.L. 1996: Radiocarbon in Hydrologic Systems Containing Dissolved Magmatic Carbon Dioxide. Science, 273: 1367–1370, doi:10.1126/science.273.5280.1367. RozansKi, K., araGuDs–araGuDs, L. & GonFiantini, R. 1992: Relation between long–term trends of 18O isotope composition of precipitation and climate. Science, 258: 981–985, doi:10.1126/ science.258.5084.981. RozansKi, K., araGuDs-araGuDs, L. & GonFiantini, R. 1993: Isotopic patterns in modern global precipitation. In: Swart, P.K., LohMan, K.C., McKenzie, J. & Savin, S. (eds.): Climate change in continental isotopic records–Geophysical monograph 78. American Geophysical Union, Washington, DC: 1–36. TrčeK, B., NovaK, M., Celarc, B. & Leis, A. 2010: Origin of mineral water from Rogaška Slatina (Slovenia). In: ZuBer, A., Kania, J. & KMieciK, E. (eds.): XXXVIII IAH Congres, Groundwater Quality Sustainability. University of Silesia, Krakow: 1769–1775. TrčeK, B. & Zojer, H. 2010: Recharge of sprin- gs. In: Krešić, N. & STevanOvić, Z. (eds.): Groundwater hydrology of springs: enginee- ring, theory, management, and sustainability. Butterworth-Heinemann: 87–127. TrčeK, B & Leis, A. 2011: Isotopic investigati- ons of mineral waters in Rogaška Slatina. In: Groundwater: our source of security in an uncertain future. Papers presented at the international conference incorporating the Biennial Conference of the Ground Water Division (GWD) of the Geological Society of South Africa (GSSA) and Meeting of the International Association of Hydrogeologists (IAH). Beta Products, Pretoria: 8 p. truesDell, A.H. & Huston, J.R. 1980: Isotopic evi- dence on environments of geothermal systems. In: Fritz, P. & Fontes, J. CH. (eds.): Handbook of Environmental Isotope Geochemistry, The Terrestrial Environment, A, 1: 179–226. ŽibreT, L. 2016: A contribution to better un- derstanding of structural characteristics and tectonic phases of the Boč region, Periadriatic Fault Zone. Geologija, 59/2: 243–257, doi: 10.5474/geologija.2016.015.