ORIGINAL ARTICLE Trace elements, REEs and stable isotopes (B, Sr) in GAS groundwater, São Paulo State, Brazil Daniel Marcos Bonotto1 • Trevor Elliot2 Received: 16 January 2017 / Accepted: 24 March 2017 / Published online: 30 March 2017 � Springer-Verlag Berlin Heidelberg 2017 Abstract This investigation was carried out in the Guarani Aquifer System (GAS) following a transect in São Paulo State, Brazil, and involved the analysis of trace elements, REEs and stable isotopes (B, Sr) in both rainwater and groundwater samples (the latter sampled from tube wells drilled in 10 cities). The Brazilian Code for Mineral Waters (BCMW) has been adopted for classifying the groundwa- ters according to their temperature and was useful for identifying the major trends of the hydrochemical data. Three water categories are identified: (\25 �C), hypother- mal (values ranging from 25 to 33 �C) and hyperthermal ([38 �C). The hyperthermal waters exhibited geostatic pressures [250 bar, whereas the cold/hypothermal waters values \100 bar. REEs concentrations were higher at the monitoring point BCS (Bernardino de Campos). Dissolved strontium in the groundwater behaves like other alkaline earth metals (calcium and barium) in samples collected along the studied transect. The hyperthermal waters tended to exhibit similar 87Sr/86Sr ratios (between 0.7088 and 0.7099), approximately corresponding to the value of ca. 0.709 for seawater Sr isotopic ratio at the end of the Proterozoic. The cold and hypothermal waters exhibited lower B contents than the hyperthermal waters. The d11B ranged from -8.1 to ?12.0%, where the d11B-values in cold/hypothermal waters were characteristically positive in clear distinction to the negative d11B signatures found in hyperthermal waters. Keywords Stable isotopes � Rare earth elements � Groundwater � Guarani Aquifer System � Paraná sedimentary basin Introduction The huge, transboundary Guarani Aquifer System (GAS) in South America has been investigated using many different approaches since the 1970s, many of the studies arising out of the IAEA-sponsored Global Environment Fund (GEF)— Guarani Aquifer Program for Groundwater Resource Sus- tainability and Environmental Protection (Foster et al. 2006). The GAS represents an important hydrological resource for about 90 million inhabitants living in the Mercosur nations, and its waters are used extensively for potable supply with many water supply systems using its waters at least as part of their networks. As a consequence, several water resource investigations have been undertaken detailing hydrochemistry, stable iso- topes (H, O, C, S) and radionuclides (14C, 36Cl, 238U, 234U, 226Ra, 222Rn, 228Ra, 210Pb, 210Po) to investigate groundwater flow patterns and ages, paleoclimatic conditions, mineral dissolution and precipitation processes, water qualities and resource availability (e.g., Sracek and Hirata 2002; Bonotto and Caprioglio 2002; Bonotto 2004, 2006, 2011, 2012, 2013; Bonotto and Mello 2006; Bonotto and Bueno 2008; Bonotto and Armada 2008; Cresswell and Bonotto 2008; Bonotto et al. 2009a, b; Gastmans et al. 2010a, b; Hirata et al. 2011; i Gil and Bonotto 2015). Multiple studies have been realized already in the Paraná sedimentary basin focusing on the rare earth & Daniel Marcos Bonotto danielbonotto@yahoo.com.br 1 Instituto de Geociências e Ciências Exatas-IGCE, Universidade Estadual Paulista-UNESP, Av. 24-A No. 1515, P.O. Box 178, Rio Claro, São Paulo CEP 13506-900, Brazil 2 School of Natural and Built Environment (SNBE), Queeńs University Belfast, Stranmillis Road, Belfast BT9 5AG, Northern Ireland, UK 123 Environ Earth Sci (2017) 76:265 DOI 10.1007/s12665-017-6590-0 http://crossmark.crossref.org/dialog/?doi=10.1007/s12665-017-6590-0&domain=pdf http://crossmark.crossref.org/dialog/?doi=10.1007/s12665-017-6590-0&domain=pdf elements (REEs) distributions in the igneous basaltic rocks (e.g., Marques et al. 1999). However, there is a lack of investigation focusing on their presence in groundwater, mainly due to the difficulties for their quantification in the liquid phase as a consequence of the very low dissolved contents. Unlike major ions chemistry, the REEs abun- dance exhibits patterns reflecting those of the host aquifers, thus constituting useful tracers of groundwater flow in aquifers where the mineralogy may vary (cf. Tweed et al. 2006). This paper focuses a single transect previously studied in São Paulo State, Brazil, also including additional monitoring points sampled there. The REEs abundance has been deter- mined in rainwater and groundwater samples, as well as other dissolved trace elements not previously discussed for the GAS and also the stable isotopes of strontium (87Sr/86Sr ratios) and boron (d11B-values) in some selected wells. Such novel database has allowed to perform new insights on the water/rock–soil interactions taking place in this GAS portion at São Paulo State, with possible implications to other areas of its occurrence in the Paraná sedimentary basin. Study area and experimental The GAS consists of Triassic–Jurassic age aeolian–fluvio– lacustrine sandstones confined by thick Cretaceous basalt flows of the Serra Geral Formation and extends over some 1.2 million km2 within the Paraná sedimentary basin comprising southern Brazil, eastern Paraguay, NW Uru- guay and the NE extreme corner of Argentina. Table 1 shows that the basalts and diabases package of the Serra Geral Formation overlying the GAS occur in all wells sampled in this study, except in São Pedro (SPO) city. The lithological description of the bores (Table 1) indicates that the main GAS sediments (Botucatu Formation and Pir- ambóia Formation) in the study area overlie greatly vari- able thick layers of the Passa Dois Group, Tubarão Group and Paraná Group sediments at Paraguaçu Paulista (PPA), Presidente Prudente (PPE) and Presidente Epitácio (PEO) municipalities. Rainwater sample was collected at a station located in the GAS recharge beds close to Rio Claro city area (Fig. 1) and sampled during the middle of the wet season Table 1 Location and description of the tubular wells drilled at the GAS—Guarani Aquifer System, at São Paulo State, Brazil, whose waters were analyzed in this paper Sample code Site State Latitude Longitude Altitude (m) Depth (m) GP (bar) Stratigraphy ITI Itirapina SP 22�150130 0S 47�490050 0W 880 129 0.9 BP (0–69); DI (69–115); BP (115–129) SPO São Pedro SP 22�320070 0S 47�540330 0W 590 150 0.9 BP (0–150) AVR Avaré SP 23�060460 0S 48�540330 0W 640 150 10.2 M (0–8); BA (8–12); SG (12–35); BP (35–150) SUT Sarutaiá SP 23�160030 0S 49�290050 0W 750 152 7.8 M (0–4); SG (4–26); BO (26–145); DI (145–152) ASB Águas de Santa Bárbara SP 22�520240 0S 49�130380 0W 560 120 3.0 SG (0–8); BO (8–56); SG (56–104); BO (104–120) BCS Bernardino de Campos SP 23�010410 0S 49�290050 0W 660 509 87.4 M (0–21); SG (21–327); BO (327–402); PI (402–509) SCP Santa Cruz do Rio Pardo SP 22�540560 0S 49�390050 0W 440 124 31.1 SG (0–114); BO (114–124) PPA Paraguaçu Paulista SP 22�250210 0S 50�330380 0W 474 3663 258.7 BA (0–64); SG(64–974); BO (974–1250); PD (1250–2050); TU (2050–3554); PA (3554–3663) PPE Presidente Prudente SP 22�060450 0S 51�220430 0W 407 1800 382.0 BA (0–218); SG (218–1440); BO (1440–1570); PI (1570–1730); PD (1730–1800) PEO Presidente Epitácio SP 21�460290 0S 52�050270 0W 258 3953 430.4 BA (0–90); SG (90–1623); BO (1623–1976); PI– PD–TU–PA (?–?) RCL Rio Claro SP 22�240410 0S 47�330410 0W 625 – – – RCL, rainwater; GP, geostatic pressure (Castany 1982), M, weathered mantle; DI, diabase sill; BA, Bauru Group (sandstones, siltstones, mudstones, carbonatic nodules); SG, Serra Geral Formation (basalts, diabases); BO, Botucatu Formation (eolic sandstones); PI, Pirambóia Formation (fluvial sandstones); BP, Undifferentiated Botucatu–Pirambóia Formations; PD, Passa Dois Group (siltstones, mudstones, shales, limestones); TU, Tubarão Group (sandstones, conglomerates, diamictites, tillites, siltstones, shales, rhythmites, silex); PA, Paraná Group (Devonian) (sandstones, conglomerates) In parentheses = depth range (in meters). Major lithologies according to Almeida and Melo (1981) 265 Page 2 of 15 Environ Earth Sci (2017) 76:265 123 (December 2009) to ensure collection of pristine (less affected by evaporation) signature. Samples were collected with bulk (dry and wet deposition) collectors consisting of large rectangular funnels coupled to polyethylene flasks (25 L) that allowed rapidly sample the same rainfall event, without the need of specific protocol for collections per- formed over long periods of time (Pelicho et al. 2006). Groundwater samples in this study provided from 10 municipalities in São Paulo State (Fig. 1, bottom). They were collected along the transect AA0 in São Paulo State Fig. 1 (top) Simplified map modified from Silva (1983) showing the outcrop and groundwater flow direction in the GAS, São Paulo State, Brazil, as well the transect AA’ from Avaré up to Presidente Epitácio (SE–NW direction); the star symbol indicates the location of the rainwater monitoring point (RCL). (bottom) Location at São Paulo State of all sampling sites, whose codes are: AVR, Avaré; SUT, Sarutaiá; ASB, Águas de Santa Bárbara; BCS, Bernardino de Campos; PPA, Paraguaçu Paulista; PPE, Presidente Prudente; PEO, Presidente Epitácio; SCP, Santa Cruz do Rio Pardo; SPO, São Pedro; ITI, Itirapina; RCL, Rio Claro Environ Earth Sci (2017) 76:265 Page 3 of 15 265 123 already focused in previous hydrogeochemical investiga- tions (Fig. 1, top): AVR—Avaré, SUT—Sarutaiá, BCS— Bernardino de Campos, PPA—Paraguaçu Paulista, PPE— Presidente Prudente and PEO—Presidente Epitácio (Sracek and Hirata 2002; Bonotto 2006; Cresswel and Bonotto 2008; i Gil and Bonotto 2015). Four additional samples were collected at São Paulo State from 120- to 150-m-deep tube wells drilled into the GAS recharge beds at Itirapina (ITI), São Pedro (SPO), Águas de Santa Bárbara (ASB) and Santa Cruz do Rio Pardo (SCP) municipalities (Fig. 1, bottom). Both rainwater and groundwater samples were stored in polyethylene bottles, with the field temperature, dissolved oxygen, pH, redox potential (Eh), electrical conductivity (EC) and alkalinity readings being performed according to the protocols described by Bonotto (2006). The filled bot- tles were transported up to LABIDRO-Isotopes and Hydrochemistry Laboratory of UNESP at Rio Claro city, where aliquots were divided for evaluating the major/trace elements, REEs and stable (Sr and B) isotopes. Suspended solids were separated on filtering the samples through a 47-mm-diameter Millipore membrane of 0.45-lm porosity. The dry residue (DR) (which equates to *TDS, total dis- solved solids) content and free dissolved CO2 data have been given in Bonotto (2006). The filtered aliquots for dissolved cations were pre- served with HNO3 or HCl. Na and K were measured by atomic absorption spectrometry (AAS); Ca, Mg and Si contents by inductively coupled plasma atomic emission spectrometry (ICP-AES). Major anions measurements on unacidified aliquots for chloride, fluoride, nitrate and sul- fate were taken using potentiometry and colorimetry as reported by Bonotto (2006). Trace elements Al, B, Br, Li, Ba, Mo, Cr, Zn, As, Rb and Sr were all quantified by inductively coupled plasma mass spectrometry (ICP-MS) analysis performed at Nicholas School of the Environment, Duke University, USA. The aliquots (10–50 L) for REEs analysis were acidified to pH\ 2 on using HCl, and about 1 g FeCl3 added. The REEs were co-precipitated on Fe(OH)3 by increasing the pH to 7–8 through addition of concentrated NH4OH solution. The precipitate was recovered, dissolved in 8 M HCl, and Fe3? was extracted into an equal volume of isopropyl ether. The acid solution containing REEs then was evaporated to dryness, and the dry residue dissolved with 1.75 M HCl to a volume of 20 mL. The acidic REEs-bearing solution was purified by cation-exchange chromatography on a Cl- col- umn of 100–200 mesh Dowex resin. The REEs then were eluted from the Cl- column with 4 M HCl and after evap- oration to dryness was dissolved in 20 mL of 1.75 M HCl. The REEs concentration in the acid solution then was mea- sured by ICP-AES Model 3410 Spectrometer, Fisons Instruments at LABOGEO—Geochemistry Laboratory, UNESP, Rio Claro city. The detection limit (DL) for the lanthanides was: La = 9.18 lg/L; Ce = 21.87 lg/L; Nd = 8.95 lg/L; Sm = 4.14 lg/L; Eu = 1.09 lg/L; Gd = 15.77 lg/L; Dy = 8.74 lg/L; Er = 0.66 lg/L; Yb = 0.79 lg/L; Lu = 0.56 lg/L. Analyses of strontium and boron isotopes in ground- water along transect AA0 (Fig. 2) were made at Nicholas School of the Environment, Duke University, USA, using a fully automated Thermo Scientific Triton Thermal Ioniza- tion Mass Spectrometer (TIMS) with Virtual Amplifiers, Dynamic Zoom and all-carbon plug-in Faraday cups, whose abundance sensitivity is *1 ppm. The technique included the use of a low-blank matrix solution that enhances BO2 - ionization and provides a stable ion beam with minimum isotopic fractionation for boron isotopes (Vengosh et al. 1989); the boron blank levels tested by isotope dilution methodology were less than 15 pg. The Sr isotopes data were expressed as 87Sr/86Sr ratios, whereas Fig. 2 Simplified geological cross section along transect AA’ according to Silva (1983) showing the depth and stratigraphy of the bores, as well the groundwater flow direction in the GAS. The main lithologies are (Almeida and Melo 1981): Bauru, sandstones, siltstones, mudstones, carbonatic nodules; Serra Geral, basalts, diabases; Botucatu/ Pirambóia, sandstones; Passa Dois/Tubarão, sandstones, conglomerates, siltstones, mudstones, shales, limestones, diamictites, tillites, rhythmites 265 Page 4 of 15 Environ Earth Sci (2017) 76:265 123 the B isotopes as d11B (in %), which was determined by the equation: d11B ¼ f½ð11 B=10 BÞsample=ð 11 B=10 BÞstd� � 1g � 1000 ð1Þ where (11B/10B)std is the 11B/10B = 4.04367 of boric acid (SRM 951). Results All hydrochemical data for the rainwater and groundwater samples are reported in Tables 2 and 3. Discussion Water quality, hydrochemical trends and temperature WHO (2011) has established guideline values for the following parameters in Table 2 that are of health sig- nificance in drinking water: nitrate (50 mg/L), fluoride (1.5 mg/L), boron (2400 lg/L), barium (700 lg/L), chro- mium (50 lg/L), arsenic (10 lg/L), uranium (30 lg/L) and 226Ra (1 Bq/L). Concentrations exceed the maximum admissible concentration for fluoride in the hyperthermal Table 2 Results of the analysis of the rainwater and groundwater samples considered in this paper Parameter Unit ITI SPO AVR SUT ASB BCS SCP PPA PPE PEO RCL Temp. �C 25 32 23 23 28 28 24 43 63 70 25 pH – 4.03 5.89 5.94 6.39 7.58 6.60 8.26 9.64 8.80 8.70 5.90 Eh mV 208 196 144 164 112 200 457 -66 -55 -72 – Diss. O2 mg/L 8.0 9.5 6.0 8.0 8.5 7.0 9.0 3.2 2.4 2.8 – Diss. CO2 mg/L 300 16 180 30 3 45 0.3 0.1 0.5 0.6 – EC lS/cm 70 13 70 70 160 160 110 520 910 760 110 DR1 mg/L 200 100 300 200 200 300 150 400 200 600 – Si mg/L 3.7 4.5 24.4 20.8 18.3 18.1 11.2 21.4 13.3 16.8 – HCO3 -? CO3 2- mg/L 2 6 70 41 73 102 36 72 216 222 7 Cl- mg/L \0.15 \0.15 \0.15 \0.15 \0.15 \0.15 6.6 9.6 110.0 56.0 1.1 NO3 - mg/L 0.7 1.2 0.5 0.6 0.6 0.7 0.6 0.7 0.7 0.6 3.5 SO4 2- mg/L \0.3 \0.3 \0.3 \0.3 0.5 0.5 \0.3 9.3 69.8 81.7 1.2 F- mg/L \0.02 \0.02 \0.02 0.05 \0.02 \0.02 0.84 1.60 8.80 6.60 – Br- mg/L – – 0.02 0.01 – 0.01 – 0.04 0.17 0.06 0.01 Na mg/L 0.5 0.8 3.3 4.9 7.3 14.5 22.6 117.0 214.0 178.0 0.9 K mg/L 0.18 0.69 1.55 2.73 3.07 1.09 0.56 0.55 2.12 1.39 0.30 Ca mg/L 0.21 0.21 8.58 7.75 25.50 20.10 4.43 0.52 4.83 1.99 0.83 Mg mg/L 0.18 0.41 4.18 1.21 0.90 2.14 \0.10 \0.10 0.71 \0.10 0.25 Li lg/L – – 0.9 2.0 – 3.0 – 19.8 57.4 29.5 0.1 Al lg/L – – 3.4 3.0 – 5.9 – 78.1 58.4 71.7 17.8 B lg/L – – 0.8 4.0 – 4.6 8 318.6 2063.0 897.8 5.0 Rb lg/L – – 3.1 5.0 – 1.0 – 0.8 5.3 1.3 1.2 Sr lg/L – – 98.7 83.1 – 146.8 – 5.5 144.0 69.1 4.1 Ba lg/L 13.0 35.0 16.6 11.5 9.0 10.9 1.0 0.2 29.3 12.9 3.1 Zn lg/L 30.0 14.0 2.2 6.6 10.0 0.6 7.0 22.8 18.0 4.9 0.1 Cr lg/L – – 0.98 1.39 – 2.05 – 3.53 15.58 2.27 0.40 As lg/L – – 0.03 0.44 – 0.10 – 19.62 4.73 6.95 0.24 Mo lg/L – – \0.01 \0.01 – 0.02 – 0.65 4.60 3.59 0.90 Ua lg/L 0.01 0.09 0.09 0.10 4.82 0.71 0.11 0.31 0.06 1.01 – 234U/238Ua A.R. 1.81 1.29 6.32 3.87 2.19 5.31 2.24 1.77 1.61 4.29 – 222Rnb Bq/L 38.4 57.2 40.8 4.1 12.5 86.5 45.9 44.5 2.5 37.3 – 226Rab Bq/L 0.12 0.14 0.40 0.31 0.18 0.42 0.32 0.44 0.19 0.24 – d11B % – – – ?12.0 – ?10.9 – -6.6 -7.3 -8.1 – 87Sr/86Sr – – – 0.7072 0.7117 – 0.7130 – 0.7094 0.7099 0.7088 – DR, dry residue a Data reported by Bonotto (2006); bData reported by Bonotto (2004) Environ Earth Sci (2017) 76:265 Page 5 of 15 265 123 ([38 �C) waters PPA, PPE and PEO, as well for arsenic in PPA. However, none of these waters is used for human consumption, only for recreational (thermal swimming pools) purposes. For dissolved radionuclides, the adoption of a dose con- version factor (DCF: IAEA 1996; WHO 2011) is required to estimate effective doses from ingestion of radionuclides in waters. There is, however, no consensus in the literature on the DCF value for 222Rn. A value of 10-8 Sv/Bq resulting from the application of a modified ICRP model for the 222Rn ingestion in water sometimes has been adopted (Kendall et al. 1988; Oliveira et al. 2001; Bonotto 2004). Bonotto (2011) suggested a value of 1.4910-6 mSv/Bq that is uti- lized in this paper. Assuming an annual dietary water con- sumption of 2 L (WHO 2011) and applying this DCF to the 222Rn activity concentration data (Table 2), it is possible to estimate the dose range due to ingested 222Rn as 0.002–0.09 mSv/year. Such annual doses are lower than the WHO (2011) guideline reference value of 0.1 mSv/year for the ingestion of all radionuclides dissolved in drinking water. The parameters pH and Eh reflect, respectively, the proton (pH) and electron (pe) activities in the environment. In natural systems, reactions in which both protons and electrons are transferred are common, effecting pH and Eh according to the following general trends: lower Eh values tend to occur under higher pH conditions, and higher Eh values are contrarily obtained under lower pH conditions (Baas Becking et al. 1960). There is little significant cor- relation (r = -0.46; n = 10) between the pH and Eh data in the study groundwaters. The Eh–pH diagram (Fig. 3) shows that the circulation environment for the waters is variable: reducing transitional acidic (ITI, SPO, SUT, AVR, BCS), oxidizing basic (SCP), transitional basic (ASB) and reducing basic (PPA, PPE, PEO). 238U and 234U are considered useful isotopes for the hydrogeochemical prospection for concealed, subsurface U Table 3 REE concentration of groundwater and rainwater samples analyzed in this paper Sample codea Unit Volume (L) La Ce Nd Sm Eu Gd Dy Er Yb Lu ITI ng/L 43.5 45.1 36.6 8.1 3.7 9.9 45.1 7.2 7.2 7.2 1.4 ITI ng/L 43.2 47.0 43.7 9.3 4.9 11.0 47.0 7.5 6.2 6.2 1.1 SPO ng/L 43.9 23.6 25.9 15.2 2.8 1.0 5.4 4.8 3.3 2.8 0.8 AVR ng/L 43.9 38.4 44.5 28.8 5.3 1.8 5.9 4.4 2.6 2.3 0.5 SUT ng/L 10.0 15.0 30.0 9.0 3.0 25.0 3.0 \DL \DL 1.0 1.0 SUT ng/L 45.1 17.4 22.1 10.1 1.1 \DL 3.6 \DL 2.1 2.0 0.7 ASB ng/L 45.1 20.0 19.1 12.2 2.8 1.0 3.2 \DL 2.0 1.8 0.3 BCS ng/L 45.7 281.2 494.4 276.3 45.7 19.0 42.6 28.6 13.0 7.1 1.0 SCP ng/L 46.3 24.7 28.4 13.0 2.3 \DL 3.4 \DL 2.2 2.4 0.6 PPA ng/L 10.0 \DL \DL \DL 1.0 2.0 \DL \DL \DL 1.0 \DL PPE ng/L 45.3 19.9 23.3 10.9 1.8 \DL 2.8 3.8 2.9 3.5 0.7 PEO ng/L 44.1 16.0 17.5 7.0 0.6 \DL \DL \DL \DL 1.5 0.3 RCLb ng/L 50.0 29.2 29.1 12.8 2.2 \DL 2.7 \DL \DL 0.4 0.2 NASC lg/g – 34.0 66.7 30.1 5.8 1.16 5.12 4.67 2.73 2.67 0.41 NASC-normalized REE abundance ITI 910-7 13.3 5.5 2.7 6.4 85.8 88.2 15.4 26.3 26.9 33.9 ITI 910-7 13.8 6.6 3.1 8.4 95.2 91.8 16.1 22.6 23.2 26.8 SPO 910-7 7.0 3.9 5.1 4.8 8.9 10.5 10.4 12.2 10.6 18.3 AVR 910-7 11.3 6.7 9.6 9.2 15.1 11.5 9.3 9.4 8.5 12.7 SUT 910-7 4.4 4.5 3.0 5.2 215.5 5.9 \DL \DL 3.8 24.4 SUT 910-7 5.1 3.3 3.3 1.9 \DL 7.0 \DL 7.6 7.3 16.3 ASB 910-7 5.9 2.9 4.0 4.8 9.0 6.3 \DL 7.3 6.7 8.3 BCS 910-7 82.7 74.1 91.8 78.8 164.2 83.3 61.1 47.7 26.7 23.7 SCP 910-7 7.3 4.3 4.3 4.0 \DL 6.6 \DL 8.0 8.9 15.4 PPA 910-7 \DL \DL \DL 1.7 17.2 \DL \DL \DL 3.8 \DL PPE 910-7 5.9 3.5 3.6 3.1 \DL 5.5 8.2 10.5 13.0 16.1 PEO 910-7 4.7 2.6 2.3 1.0 \DL \DL \DL \DL 5.6 7.1 RCLb 910-7 8.6 4.4 4.3 3.8 \DL 5.2 \DL \DL 1.6 5.1 NASC, North American Shale Composite (Goldstein and Jacobsen 1988); DL, detection limit a Site location in Fig. 1; bRainwater 265 Page 6 of 15 Environ Earth Sci (2017) 76:265 123 deposits. The data for dissolved U content and 234U/238U activity ratio (AR) in the GAS groundwaters (Table 2) are plotted on a two-dimensional U content vs. AR diagram containing several areas of associative significance (Cowart and Osmond 1980; Osmond and Cowart 1981; Chatam et al. 1981). Most of the samples are categorized as reducing as defined also by the Eh–pH diagram. The groundwater sample SCP is oxidizing (Fig. 3) rather than reducing (U content vs. AR diagram; Bonotto 2006), perhaps due to a mixed signa- ture or dominance of another couple). The data for pH, EC, major cations (Na?, K?, Ca2?, Mg2?) and major anions (HCO3 -? CO3 2-, Cl-, NO3 -, SO4 2-) in each groundwater sample have been autocorrelated with those of the rainwater. The following Pearson correlation coefficient values were found: ITI: r = 0.99; SPO: r = 0.88; AVR: r = 0.70; SUT: r = 0.88; ASB: r = 0.92; BCS: r = 0.85; SCP: r = 0.95; PPA: r = 0.97; PPE: r = 0.96; PEO: r = 0.95. These coefficients suggest a strong influence of the rainwater signature on the groundwater chemistry. Such behavior should be expected in the boreholes drilled close to the recharge beds (ITI, SPO, AVR, SUT, ASB, BCS and SCP) but not necessarily in the deeper (PPA, PPE and PEO) tube wells where higher water–rock interaction impacts might be expected. Thus, such major element criteria then are not particularly useful for discriminating unconfined and confined groundwater in the GAS. Discrimination appears when the trace constituents (Br-, Li, Al, B, Rb, Sr, Ba, Zn, Cr, As, Mo) and REEs in ground- water and rainwater are considered. The autocorrelations here are not significant for the boreholes close to the recharge beds (AVR: r = 0.17; SUT: r = 0.17; BCS: r = 0.18) nor par- ticularly for those exploiting the deeper wells (PPA: r = 0.36; PPE: r = 0.22; PEO: r = 0.26). The lack of autocorrelation with rainwater emphasizes the sensitivity of minor/trace ele- ments to discriminate groundwater processes (water–rock interactions) here; however, unlike for the case that strong autocorrelation with rainwater signatures implies lack of discrimination between unconfined/confined groundwater we cannot simply preclude the same here (as the deeper wells show a higher r than the recharge bed waters). Temperature categories from the Brazilian Code for Mineral Waters (BCMW; DFPM 1966) also prove useful for discriminating the groundwater samples in this study (Table 2): cold (\25 �C: AVR, SUT, SCP); hypothermal (25–33 �C: ITI, SPO, ASB, BCS); and hyperthermal ([38 �C: PPA, PPE, PEO). Two major water groupings are identified: cold/hypothermal waters (Group I) and hyper- thermal waters (Group II). Figure 4 shows that the Group I waters exhibit geostatic pressure \100 bar, dissolved O2 [6 mg/L, positive Eh values, EC \160 lS/cm, sodium \23 mg/L, chloride\7 mg/L and sulfate\0.5 mg/L. The trace constituents Al, Li, F, Br, As and Mo then are seen to be enhanced in the Group II waters relative to the Group I waters (Fig. 5). Nevertheless, despite its usefulness, this temperature criterion when looking at only the major hydrochemical facies fails to show coherent discrimination of waters; for instance, the hyperthermal waters PPA, PPE and PEO are dominated by sodium and (bi)carbonate, the same occurring with the cold groundwater SCP. REEs dissolved in groundwater and rainwater REEs generally are lithophile in character and strongly enriched in the continental crust relative to mantle and oceanic crust. Their crustal enrichment factors decrease with increasing atomic number (Faure 1991) from 26.8 for La (Z = 57) to 4.9 for Lu (Z = 71). Systematic variation of the abundances of the REEs in the continental crust cor- relates with the decrease in their ionic radii (the so-called lanthanide contraction), which results from the progressive filling of 4f orbitals and the consequent contraction of their electron clouds (Faure 1991). Thus, effective ionic radii of the REEs decrease from 1.13 Å for La3? to 0.94 Å for Fig. 3 Plotting of the pH and Eh data in the study area in an Eh–pH diagram (Krauskopf and Bird 1995). Codes: AVR, Avaré; SUT, Sarutaiá; ASB, Águas de Santa Bárbara; BCS, Bernardino de Campos; PPA, Paraguaçu Paulista; PPE, Presidente Prudente; PEO, Presidente Epitácio; SCP, Santa Cruz do Rio Pardo; SPO, São Pedro; ITI, Itirapina Environ Earth Sci (2017) 76:265 Page 7 of 15 265 123 Lu3? in sixfold coordination and the ionic potential of the REEs increases from 2.65 (La3?) to 3.19 (Lu3?), which suggest, for example, that Lu3? forms stronger ionic bonds in crystals than Lu3? (Faure 1991). Thus, the lanthanide contraction often reflects in a systematic variation of the crystal/liquid distribution coefficients of the REEs (Faure 1991). REE abundance patterns have been extensively utilized in rock geochemical studies where, in general, the light REEs are concentrated in the late-stage felsic differentiates of magma, whereas the heavier REEs are concentrated in the early-formed mafic products (Faure 1991). Table 3 shows the REEs analyzed in groundwaters of the GAS. One rainwater and twelve groundwater samples were chemically analyzed for the REEs abundance patterns. The following REE ranges above the detection limit were found in groundwaters (Table 3): La = 15–281 ng/L; Ce = 18–494 ng/L; Nd = 7-276 ng/L; Sm = 0.6-46 ng/L; Eu = 1-25 ng/L; Gd = 3-47 ng/L; Dy = 4-29 ng/L; Er = 2-13 ng/L; Yb = 1-7 ng/L; Lu = 0.3-1 ng/L. Figure 6 shows that, in general, the REEs abundance pat- terns indeed decrease with increasing atomic number from La to Lu, as often seen in the continental crust (Faure 1991). Table 4 shows a matrix of the Pearson correlation coefficients (r) for all REEs analyzed. The two-tailed P value was estimated by GraphPad software (Arsham 1988) from each Pearson correlation coefficient. Statisti- cally significant correlations (practically 50% of the cases showing the coherence of the REE variations) are high- lighted in bold in Table 4. The REEs concentration values are found highest at the monitoring point BCS (Table 4, Fig. 6). In this area, the Fig. 4 a Geostatic pressure, b dissolved oxygen, c redox potential Eh, d electrical conductivity (EC), e sodium, f chloride and g sulfate plotted against the groundwater temperature in the GAS 265 Page 8 of 15 Environ Earth Sci (2017) 76:265 123 occurrence of extensive faulting has affected the relative positioning of the layers of basalts (Serra Geral Formation) and sandstones (Botucatu and Pirambóia Formations) (Guedes et al. 2015) (Fig. 2). Such processes have exposed rock surfaces containing some minerals assembly enriched in REEs whose transfer to the liquid phase would be favored by preferential paths of leaching that could allow the REEs to be more accessible to water. Much of the knowledge of magmatic processes and nat- ural aqueous systems based on the relative abundance of individual lanthanide elements has been made using a log- arithmic plot of lanthanide abundances normalized to abundances in chondritic (stony) meteorites (Hedrick and Templeton 1991), primordial mantle (Sun and McDonough 1989), UCC—Upper Continental Crust (Taylor and McLennan 1985), PAAS—Post Archean American Shale (Taylor and McLennan 1985), NASC—North American Shale Composite (Goldstein and Jacobsen 1988) and 3SA— 3-Shale Average (Sholkovitz 1988). Recent studies focusing on the REEs distribution in groundwater have adopted the normalization to shales (Smedley 1991; Johannesson et al. 1996, 1997; Tang and Johannesson 2006). Squisato et al. (2009) report the REEs concentration in flood basalts of Serra Geral Formation occurring in four dif- ferent regions of São Paulo State (Jaú, Ribeirão Preto, Franca and Fernandópolis), which represent almost the total area of outcrops of basalts in the São Paulo State. The mean values were: La = 30.0 lg/g; Ce = 66.2 lg/g; Nd = 37.4 lg/g; Sm = 8.0 lg/g; Eu = 2.5 lg/g; Gd = 7.5 lg/g; Dy = 6.8 lg/g; Er = 3.5 lg/g; Yb = 2.8 lg/g; Lu = 0.4 lg/g. Figure 7 shows that these average values are very well cor- related (r = 0.99) with the NASC data (Table 3), thus justi- fying the NASC normalization of the lanthanides abundance in groundwater. The NASC-normalized REEs abundance is also given in Table 3. Fig. 5 Dissolved a aluminum, b fluoride, c bromide, d lithium, e arsenic and f molybdenum plotted against the groundwater temperature in the GAS Fig. 6 REEs abundance in groundwater from the GAS. Codes: BCS, Bernardino de Campos; ASB, Águas de Santa Bárbara; SUT, Sarutaiá; SPO, São Pedro; AVR, Avaré; PPE, Presidente Prudente; SCP, Santa Cruz do Rio Pardo; PEO, Presidente Epitácio Environ Earth Sci (2017) 76:265 Page 9 of 15 265 123 Eu-anomalies (Eu/Eu*) from the available data in this paper are defined as the ratio between measured Eu in the groundwater and the value expected for Eu on a smooth NASC-normalized plot. The Eu-anomalies (Eu/Eu*) have been calculated in four sites where all REEs analyzed were above the DL. This was done using the simple arithmetic [(SmN ? GdN)/2] mean to estimate Eu*, where SmN and GdN are the NASC-normalized values for Sm and Gd. Positive Eu-anomalies (mean Eu/Eu* = 1.6) were found: ITI = 1.8; SPO = 1.2; AVR = 1.4; BCS = 2.0 in line with Sant́Anna et al. (2006) where Eu/Eu* ratio of 1.5 is shown in some highly crystalline illite-type clays occurring in sedimentary rocks of the Rio Bonito Formation (Per- mian), Paraná basin, which are characterized by high water/rock ratios. These findings suggest that water–rock processes taking place in the GAS strata could release REEs into the liquid phase, which retain equivalent sig- natures to those seen in the mineral assembly of the rock matrices. Table 4 Correlation matrix involving the REEs in GAS groundwater, São Paulo State, Brazil La 1.00 Ce 0.99 1.00 Nd 0.99 0.99 1.00 Sm 0.99 0.99 0.99 1.00 Eu 0.41 0.44 0.41 0.47 1.00 Gd 0.68 0.64 0.60 0.64 0.33 1.00 Dy 0.99 0.99 0.98 0.99 0.91 0.66 1.00 Er 0.95 0.92 0.91 0.92 0.41 0.87 0.97 1.00 Yb 0.73 0.68 0.65 0.69 0.30 0.94 0.74 0.89 1.00 Lu 0.36 0.34 0.29 0.34 0.67 0.73 0.39 0.54 0.66 1.00 La Ce Nd Sm Eu Gd Dy Er Yb Lu Statistically significant correlations are highlighted in bold Fig. 7 Average REEs concentration in flood basalts of Serra Geral Formation occurring in São Paulo State (SP—Squisato et al. (2009) compared to the NASC, North American Shale Composite (Goldstein and Jacobsen 1988) Fig. 8 Dissolved (top) calcium and (middle) barium content plotted against the strontium concentration in GAS groundwater. The 87Sr/86Sr ratio is also plotted in the bottom against the reciprocal of the dissolved strontium content 265 Page 10 of 15 Environ Earth Sci (2017) 76:265 123 Strontium and boron isotopes in groundwater Dissolved strontium in the groundwater behaves like other alkaline earth metals such as calcium and barium as evi- denced by the significant regression correlations between them (Fig. 8). There is a decrease in both Sr and Ba con- tents in PPA hyperthermal waters, possibly following the same trend for higher dissolved Ca in cold/hypothermal waters SUT, SCP, ASB and BCS (Table 2). This could be attributed to the decrease in the carbonates reactivity in aqueous solution, according to increased temperature (Pokrovsky et al. 2009). The hyperthermal waters tend to exhibit restricted range 87Sr/86Sr ratios (between 0.7088 and 0.7099) (Table 2, Fig. 8) similar to the value of ca. 0.709 for seawater Sr isotopic ratio at the end of Protero- zoic (between 610 and 550 Ma; Asmerom et al. 1991) although there is no evidence of a seawater component from other chemistries. The 87Sr/86Sr ratios in the cold/hypothermal waters, however, differ from the values found in the hyperthermal waters (Table 2, Fig. 8). The carbonates reactivity increa- ses in the lower temperatures, and leaching processes are expected to be more pronounced. The ratio in AVR cold groundwater (0.7072) approaches the mean value of 0.7065 ± 0.0006 reported by Wildner et al. (2006) for 18 samples of basaltic lavas belonging to the Serra Geral Formation and sampled at southwestern of Paraná State. On the other hand, the higher 87Sr/86Sr ratios (0.7117 and 0.7130) in the cold (SUT) and hypothermal (BCS) waters correspond to more enriched 87Sr radiogenic values pos- sibly generated from leaching of secondary calcite phase in the Botucatu/Pirambóia sandstones. This is supported by the 87Sr/86Sr ratios (0.7161–0.7171) given by Vieira (1980) and Gilg et al. (2003) in sandstones of these formations. Additionally, in studies made in the Great Artesian Basin (Australia) this was also found as de Caritat et al. (2005) suggested that 87Sr/86Sr ratios from 0.712 to 0.715 indicate the influence of carbonate dissolution, which partly includes our samples. However, Moya et al. (2016) low- ered the 0.712 limit from de Caritat et al. (2005), corre- lating with all our ‘‘more radiogenic samples.’’ Boron stable isotopes (atomic masses of 10 and 11) possess natural abundances of about 19.82 and 80.18%, respectively (Mather and Porteous 2001). Boron isotope fractionations are often controlled by the partitioning between the undissociated boric acid, B(OH)3 (planar trigonal) and the anion B(OH)4 - (tetrahedral), through the equilibrium reaction: B(OH)3 ? OH- = B(OH)4 - (Pennisi et al. 2006a). Both dissolved species are pH dependent and, at 25 �C and B concentration of 10-3 mol.L-1, B(OH)3 is dominant for pH\ 9, whereas B(OH)4 - predominates for pH[ 9 (Tonarini et al. 2004). In general, the combination of the isotopic fractionation in reaction involving B(OH)3 and B(OH)4 - and removal of H4BO4 - by the preferential adsorption on clay minerals (Palmer et al. 1987) constitutes the main mechanism for the large isotopic variations of B seen in natural waters (Pennisi et al. 2006a, b). In conti- nental waters, the upper limit of the 11B/10B ratios is ?59%, as observed in Australian crater lakes (Vengosh et al. 1991), and the lower limit is -27%, as shown in groundwater (Pennisi et al. 2006a). Dissolved boron and d11B in the groundwater samples are plotted in Fig. 9 against temperature. Cold/hypothermal waters (Group I) show lower B contents than hyperthermal waters (Group II), following trends for the trace con- stituents Al, Li, F, Br, As and Mo which are enhanced in the Group II waters relative to the Group I waters (Fig. 5). The d11B ranged from -8.1 to ?12.0%, where the d11B- values in the cold/hypothermal waters (Group I) are posi- tive, in a clear distinction of the negative d11B-values found in the hyperthermal waters (Group II). Because there is no evidence of anthropogenic inputs in these groundwater samples (Hirata et al. 2011), geogenic factors are considered responsible for the large variability in the d11B-values. Boron is a trace element that has been considered a good indicator of paleosalinity in sedimenta- tion sites since the pioneering studies focusing its presence in illite (Frederickson and Reynolds 1960). High B has been already identified in deep waters and Paleozoic sed- iments of the Paraná sedimentary basin (Ramos and For- moso 1975; Rodrigues and Quadros 1976; Szikszay and Teissedre 1981). In the current transect, the boron content increases from Rio Bonito Formation up to the Irati For- mation (Rodrigues and Quadros 1976) and illite is the dominant clay mineral in sediments of the Passa Dois, Fig. 9 (top) Dissolved boron and (bottom) d11B plotted against the groundwater temperature in the GAS Environ Earth Sci (2017) 76:265 Page 11 of 15 265 123 Tubarão and Paraná Groups in bore PPA (cf. Ramos and Formoso 1975). Boric acid, B(OH)3, undergoes adsorption–desorption processes on the ubiquitous clay minerals (Pennisi et al. 2006b). 11B separates preferentially into the B(OH)3 spe- cies in solution, prevailing at lower pHs (Spivack and Edmond 1987; Palmer et al. 1987; Tonarini et al. 2004). This could explain the positive d11B-values found in the cold/hypothermal waters (Table 2; Fig. 9). Adsorption depends on higher pHs, prevailing B(OH)4 - that is more easily adsorbed than its equilibrium competitor B(OH)3 (Pennisi et al. 2006b). 10B is dominantly incor- porated into B(OH)4 - in the solid phase, which substitutes for Al in silicate minerals (Spivack and Edmond 1987; Palmer et al. 1987). Water desorption from clays is a function of burial depth (geostatic/lithostatic pressure) in sediments, i.e., the number of adsorbed water layers decreases as the temperature and pressure of the sediments increase (Velde 1992). Thus, desorption processes affect- ing clay minerals and occurring at higher temperatures could support the negative d11B-values found in the hyperthermal waters (Table 2; Fig. 9). Such processes would also explain the enhanced concentrations of some major/trace constituents in the hyperthermal waters (Group II) relative to the cold/hypothermal ones (Group I) (Figs. 4, 5) as confirmed by significant regression correlations with Fig. 10 Dissolved a sodium, b chloride, c sulfate, d lithium, e fluoride, f bromide and g molybdenum plotted against the boron content in GAS groundwater 265 Page 12 of 15 Environ Earth Sci (2017) 76:265 123 boron (Fig. 10): Na (r = 0.91), Cl- (0.99), SO4 2- (r = 0.85), Li (r = 0.98), F (r = 0.96), Br (r = 0.99) and Mo (r = 0.95). Conclusions Groundwaters of the Guarani Aquifer System (GAS) are an important resource in South America and are extensively used for drinking water. This investigation was realized in a single transect in São Paulo State, Brazil, and involved the sampling of several boreholes drilled for exploiting the GAS. Two major water groups were identified according to temperature: cold/hypothermal waters (Group I) and hyperthermal waters (Group II). The Group I waters show geostatic pressure \100 bar, dissolved O2 [6 mg/L, posi- tive Eh values, EC\160 lS/cm, sodium\23 mg/L, chlo- ride \7 mg/L and sulfate \0.5 mg/L. The trace constituents Al, Li, F, Br, As and Mo are enhanced in the Group II waters relative to the Group I waters. When comparing the hydrochemical data in groundwater to WHO guideline values for chemicals that are of health signifi- cance in drinking water (nitrate, fluoride, boron, barium, chromium, arsenic, uranium 226Ra and 222Rn), it was possible to identify concentrations exceeding the maximum allowable fluoride in some hyperthermal waters, as well for arsenic in one groundwater sample (PPA). However, none of these waters is used for human consumption, only for recreation (thermal swimming pools) purposes. The REEs concentration values are higher at the monitoring point BCS characterized by extensive faulting that has affected the relative positioning of the layers of basalts/sandstones from the Serra Geral Formation and Botucatu–Pirambóia Formations. Positive Eu-anomalies were found at ITI, SPO, AVR and BCS that are compatible with values reported in some highly crystalline illite-type clays occurring in sedi- mentary rocks of the Rio Bonito Formation (Permian), Paraná basin characterized by high water/rock ratios. 87Sr/86Sr ratios in the cold/hypothermal waters are different of the values found in the hyperthermal waters. Desorption processes affecting clay minerals at higher temperatures as shown along the studied transect could explain the negative d11B-values found in the hyperthermal waters. Such pro- cesses also would explain the enhanced concentrations of some major/trace constituents in the hyperthermal waters relative to the cold/hypothermal ones. Acknowledgements FAPESP (Foundation Supporting Research in São Paulo State) and CNPq (National Council for Scientific and Technologic Development) in Brazil are greatly thanked for financial support of this investigation. Reinhard Jung from Hochschule Man- nheim, Germany, is thanked by the REE analysis made at LABO- GEO/IGCE-UNESP-Rio Claro. Avner Vengosh from Duke University, USA, is thanked by ICP-MS and TIMS analyses. One anonymous reviewer is greatly thanked by helpful comments that improved the readability of the manuscript. References Almeida FFM, Melo MS (1981) The Paraná basin and Mesozoic volcanism. In: IPT (Technological Research Institute of São Paulo State) (ed) Geological map of São Paulo State, vol 1. Promocet, São Paulo, pp 46–81 Arsham H (1988) Kuiper’s P-value as a measuring tool and decision procedure for the goodness-of-fit test. J Appl Stat 15(3):131–135 Asmerom Y, Jacobsen SB, Knoll AH, Butterfield NJ, Swet K (1991) Strontium isotopic variations of Neoproterozoic seawater: implications for crustal evolution. Geochim Cosmochim Acta 55:2883–2894 Baas Becking LGM, Kaplan IR, Moore D (1960) Limits of the natural environment in terms of pH and oxidation-reduction potential. J Geol 68:243–284 Bonotto DM (2004) Doses from 222Rn, 226Ra, and 228Ra in groundwater from Guarani aquifer, South America. J Environ Radioact 76:319–335 Bonotto DM (2006) Hydro(radio)chemical relationships in the giant Guarani aquifer, Brazil. J Hydrol 323:353–386 Bonotto DM (2011) Natural radionuclides in major aquifer systems of the Paraná sedimentary basin, Brazil. Appl Radiat Isot 69:1572–1584 Bonotto DM (2012) A comparative study of aquifer systems occurring at the Paraná sedimentary basin, Brazil: major hydrochemical trends. Environ Earth Sci 67:2285–2300 Bonotto DM (2013) A comparative study of aquifer systems occurring at the Paraná sedimentary basin, Brazil: U-isotopes contribution. Environ Earth Sci 68:1405–1418 Bonotto DM, Armada PCP (2008) Radon and progeny (214Pb and 214Bi) in urban water-supply systems at São Paulo State, Brazil. Appl Geochem 23:2829–2844 Bonotto DM, Bueno TO (2008) The natural radioactivity in Guarani aquifer groundwater, Brazil. Appl Radiat Isot 66:1507–1522 Bonotto DM, Caprioglio L (2002) Radon in groundwaters from Guarany aquifer, South America: environmental and exploration implications. Appl Radiat Isot 57:931–940 Bonotto DM, Mello CB (2006) A combined method for evaluating radon and progeny in waters and its use at Guarani aquifer, São Paulo State, Brazil. J Environ Radioact 86:337–353 Bonotto DM, Bueno TO, Tessari BW, Silva A (2009a) The natural radioactivity in water by gross alpha and beta measurements. Radiat Meas 44:92–101 Bonotto DM, Caprioglio L, Bueno TO, Lazarindo JR (2009b) Dissolved 210Po and 210Pb in Guarani aquifer groundwater, Brazil. Radiat Meas 44:311–324 Castany G (1982) Principles and methods in hydrogeology. Dunod, Paris Chatam JR, Wanty RB, Langmuir D (1981) Groundwater prospecting for sandstone-type uranium deposits: the merits of mineral- solution equilibria versus single element tracer methods. US DOE, Rep. GJBX-129 (81), p 216 Cowart JB, Osmond JK (1980) Uranium isotopes in groundwater: their use in prospecting for sandstone-type uranium deposits. US DOE, Rep. GJBX-119 (80), p 112 Cresswell RG, Bonotto DM (2008) Some possible evolutionary scenarios suggested by Cl-36 measurements in Guarani aquifer groundwaters. Appl Radiat Isot 66:1160–1174 de Caritat P, Kirste D, Carr G, McCulloch M (2005) Groundwater in the Broken Hill region, Australia: recognising interaction with Environ Earth Sci (2017) 76:265 Page 13 of 15 265 123 bedrock and mineralisation using S, Sr and Pb isotopes. Appl Geochem 20:767–787 DFPM (Division for Supporting the Mineral Production) (1966) The mining code, the mineral waters code and how applying research in a mineral deposit, 8th edn, Rep. 91, Rio de Janeiro, DFPM, p 109 Faure G (1991) Principles and applications of inorganic geochem- istry. MacMillan Publishing Co, New York, p 626 Foster S, Kemper K, Garduño H, Hirata R, Nanni M (2006) The Guarani aquifer initiative for transboundary groundwater man- agement. Case profile collection, 9, sustainable groundwater management—lessons from practice. The World Bank, Wash- ington, DC Frederickson AF, Reynolds RC (1960) Geochemical method of determining paleosalinity. Clay Miner 8:203–213 Gastmans D, Chang HK, Hutcheon I (2010a) Stable isotopes (2H, 18O and 13C) in groundwaters from the northwestern portion of the Guarani Aquifer System (Brazil). Hydrogeol J 18:1497–1513 Gastmans D, Chang HK, Hutcheon I (2010b) Groundwater geochem- ical evolution in the northern portion of the Guarani Aquifer System (Brazil) and its relationship to diagenetic features. Appl Geochem 25:16–33 Gilg HA, Morteani G, Kostitsyn Y, Preinfalk C, Gatter I, Strieder AJ (2003) Genesis of amethyst geodes in basaltic rocks of the Serra Geral Formation (Ametista do Sul, Rio Grande do Sul, Brazil): a fluid inclusion, REE, oxygen, carbon, and Sr isotope study on basalt, quartz, and calcite. Miner Deposita 38:1009–1025 Goldstein SJ, Jacobsen SB (1988) Rare earth elements in river waters. Earth Planet Sci Lett 89:35–47 Guedes C, Morales N, Etchebehere MLC, Saad AR (2015) Indications of neotectonic deformations at Rio Pardo-SP basin by the analysis of fluvial morphometric parameters and SRTM images. Geociências 34(3):364–380 Hedrick JB, Templeton DA (1991) Rare earth minerals and metals in 1989. USBM, Washington, DC Hirata R, Gesicki A, Sracek O, Bertolo R, Giannini PC, Aravena R (2011) Relation between sedimentary framework and hydroge- ology in the Guarani Aquifer System in São Paulo state, Brazil. J S Am Earth Sci 31:444–456 i Gil AS, Bonotto DM (2015) Hydrochemical and stable isotopes (H, O, S) signatures in deep groundwaters of Paraná basin, Brazil. Environ Earth Sci 73:95–113 IAEA (International Atomic Energy Agency) (1996) International basic safety standards for protection against ionizing radiation and for the safety of radiation sources. Safety series, vol 15. IAEA, Vienna, p 353 Johannesson KH, Stetzenbach KJ, Hodge VF, Lyons WB (1996) Rare earth element complexation behavior in circumneutral pH groundwaters: assessing the role of carbonate and phosphate ions. Earth Planet Sci Lett 139:305–319 Johannesson KH, Stetzenbach KJ, Hodge VF (1997) Rare earth elements as geochemical tracers of regional groundwater mixing. Geochim Cosmochim Acta 61(17):3605–3618 Kendall GM, Fell TP, Phipps AW (1988) A model to evaluate doses from radon in drinking water. Radiological Protection Bulletin 97:7–8 Krauskopf KB, Bird DK (1995) Introduction to geochemistry. McGraw-Hill Inc, New York Marques LS, Duprè B, Piccirillo EM (1999) Mantle source compo- sitions of the Paraná Magmatic Province (southern Brazil): evidence from trace element and Sr-Nd-Pb isotope geochem- istry. J Geodyn 28:439–458 Mather JD, Porteous NC (2001) The geochemistry of boron and its isotopes in groundwaters from marine and non-marine sandstone aquifers. Appl Geochem 16:821–834 Moya CE, Raiber M, Taulis M, Cox ME (2016) Using environmental isotopes and dissolved methane concentrations to constrain hydrochemical processes and inter-aquifer mixing in the Galilee and Eromanga Basins, Great Artesian Basin, Australia. J Hydrol 539:304–318 Oliveira J, Mazzilli BP, Sampa MHO, Bambalas E (2001) Natural radionuclides in drinking water supplies of São Paulo State Brazil and consequent population doses. J Environ Radioact 53:99–109 Osmond JK, Cowart JB (1981) Uranium-series disequilibrium in groundwater and core composite samples from the San Juan Basin and Copper Mountain research sites. US DOE, Rep. GJBX-364 (81), p 126 Palmer MR, Spivack AJ, Edmond JM (1987) Temperature and pH controls over isotopic fractionation during adsorption of boron on marine clays. Geochim Cosmochim Acta 51:2319–2323 Pelicho AF, Martins LD, Nomi SN, Solci MC (2006) Integrated and sequential bulk and wet-only samplings of atmospheric precip- itation in Londrina, South Brazil (1998-2002). Atmos Environ 40:6827–6835 Pennisi M, Gonfiantini R, Grassi S, Squarci P (2006a) The utilization of boron and strontium isotopes for the assessment of boron contamination of the Cecina River alluvial aquifer (central- western Tuscany, Italy). Appl Geochem 21:643–655 Pennisi M, Bianchini G, Muti A, Kloppmann W, Gonfiantini R (2006b) Behaviour of boron and strontium isotopes in ground- water–aquifer interactions in the Cornia Plain (Tuscany, Italy). Appl Geochem 21:1169–1183 Pokrovsky OS, Golubev SV, Schott J, Castillo A (2009) Calcite, dolomite and magnesite dissolution kinetics in aqueous solutions at acid to circumneutral pH, 25 to 150�C and 1 to 55 atm pCO2: new constraints on CO2 sequestration in sedimentary basins. Chem Geol 265(1–2):20–32 Ramos AN, Formoso MLL (1975) Clay minerals from sedimentary rocks of the Paraná basin. PETROBRÁS-CENPES-DINTEP, Rio de Janeiro, p 72 Rodrigues R, Quadros LP (1976) Mineralogy of clays and boron content of Paleozoic formations from Paraná basin. In: SBG (Brazilian Society of Geology) (ed) Proceedings of XXIX Brazilian geolog- ical congress, vol 1. SBG, Ouro Preto, pp 351–379 Sant́Anna LG, Clauer N, Cordani UG, Riccomini C, Velázquez VF, Liewig N (2006) Origin and migration timing of hydrothermal fluids in sedimentary rocks of the Paraná basin, South America. Chem Geol 230:1–21 Sholkovitz ER (1988) Rare-earth elements in marine sediments and geochemical standards. Chem Geol 88:333–347 Silva RBG (1983) Hydrochemical and isotopic study of groundwaters from Botucatu aquifer in São Paulo State. Ph.D. Thesis, USP- São Paulo University, São Paulo Smedley PL (1991) The geochemistry of rare earth elements in groundwater from the Carnmenellis area, southwest England. Geochim Cosmochim Acta 55:2767–2779 Spivack AJ, Edmond JM (1987) Boron isotope exchange between seawater and the oceanic crust. Geochim Cosmochim Acta 51:1033–1043 Squisato E, Nardy AJR, Machado FB, Marques LS, Rocha ERV Jr, Oliveira MAF (2009) Lithogeochemistry and petrogenetic aspects of basalts from Paraná Magmatic Province at the north-central portion of São Paulo State. Geociências 28(1):27–41 Sracek O, Hirata R (2002) Geochemical and stable isotopic evolution of the Guarani Aquifer System in the state of São Paulo, Brazil. Hydrogeol J 10:643–655 Sun SS, McDonough WF (1989) Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and 265 Page 14 of 15 Environ Earth Sci (2017) 76:265 123 processes. In: Sauders MJ (ed) Magmatism in the ocean basins, vol 42. Geological Society Special Publ, London, pp 313–345 Szikszay M, Teissedre JM (1981) Springs of the Paraná sedimentary basin, São Paulo State. Revista Águas Subterrâneas 3:85–102 Tang J, Johannesson KH (2006) Controls on the geochemistry of rare earth elements along a groundwater flow path in the Carrizo Sand aquifer, Texas, USA. Chem Geol 225:156–171 Taylor SR, McLennan SM (1985) The continental crust: its compo- sition and evolution. Blackwell, Melbourne, p 312 Tonarini S, Pennisi M, Gonfiantini R (2004) Boron isotope determi- nations in waters and other geological materials: analytical techniques and inter-calibration of measurements. In: IAEA (International Atomic Energy Agency) (ed) Book of extended synopses of the international symposium on quality assurance for analytical methods in isotope hydrology, Vienna, IAEA, pp 55–56 Tweed SO, Weaver TR, Cartwright I, Schaefer B (2006) Behavior of rare earth elements in groundwater during flow and mixing in fractured rock aquifers: an example from the Dandenong Ranges, southeast Australia. Chem Geol 234:291–307 Velde B (1992) Introduction to clay minerals: chemistry, origins, uses and environmental significance. Chapman & Hall, London Vengosh A, Chivas AR, McCulloch MT (1989) Direct determination of boron and chlorine isotopes in geological materials by negative thermal-ionization mass spectrometry. Chem Geol 79:333–343 Vengosh A, Chivas AR, McCulloch MT, Starinski A, Kolodny Y (1991) Boron isotope geochemistry of Australian salt lakes. Geochim Cosmochim Acta 55:2591–2606 Vieira PC (1980) Preliminary geochronological interpretation of the Paraná basin. Revista IG 1(2):25–32 WHO (World Health Organization) (2011) Guidelines for drinking- water quality, 4th edn. WHO Press, Geneva Wildner W, Brito RSC, Licht OAB, Arioli EE (2006) Geology and Mineral Resources from Paraná State, scale 1:200,000. Brası́lia, CPRM/Mineropar, p 95 Environ Earth Sci (2017) 76:265 Page 15 of 15 265 123 Trace elements, REEs and stable isotopes (B, Sr) in GAS groundwater, São Paulo State, Brazil Abstract Introduction Study area and experimental Results Discussion Water quality, hydrochemical trends and temperature REEs dissolved in groundwater and rainwater Strontium and boron isotopes in groundwater Conclusions Acknowledgements References