Coastal aquifers of southeastern Brazil have a fundamental role in economic and social terms, as it is a densely populated region with intense economic activity. The present research was carried out in the coastal region of the State of Rio de Janeiro and enclosed the Brazilian Nuclear Power Plants (NPP) site. The study aimed to assess the hydrochemistry of the shallow aquifer at the NPP's site and surroundings. The hydrochemical study was carried out based on data from twelve water samples, collected in October 2018, in which the following parameters, among others, were measured: electrical conductivity, total dissolved solids, temperature, and pH were measured in situ. The analytical determinations carried out were as follows: alkalinity (titration); Ca2+ and Mg2+, (atomic absorption/spectrometry); Na+ and K+ (liquid chromatography); Cl- and SO42+ (liquid chromatography), and; silicon (inductively coupled plasma mass spectrometry). Using Piper and Chadha diagrams, the research identified four water types: Group 1 (Ca2+-Na+-HCO3-), Group 2 (Ca2+-HCO3-), Group 3 (Ca2+ (Na+ rich)-HCO3-), and Group 4 (Na+-Cl-). From the hillsides towards the beach line, the research results suggest a hydrochemical evolution of local groundwater from Group 1 (high rainwater influence), passing through Group 3 (intermediate between Group 1 and 2) and Group 2 (rock influence), to Group 4 (seawater like). The waters of groups 1 to 3 showed potential for base cation exchange (softening or freshening process), while, in Group 4, the potential is for reverse cation exchange (hardening or salinizing process). In a general hydrogeology context, the methods and applied techniques proved to be particularly efficient in identifying potential hydrochemical processes. Noteworthy is the freshening cation exchange identified in the shallow aquifer at NPP's site. Therefore, theoretically, in case of an accidental release of radionuclides from the NPP to the aqueous phase, the local clay lenses can capture the divalent cation Ra2+ in a natural contamination attenuation process.
Research Article
Use of Major Ions to Evaluate Groundwater Chemistry and Identify Hydrochemical Processes in a Shallow Coastal Aquifer in Southeast Brazil
https://doi.org/10.21203/rs.3.rs-946502/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
hydrochemistry
coastal aquifer
hydrochemical processes
seawater interference
All industrial activity can affect the environment, with water resources being one of the most vulnerable environmental compartments for human actions. In this context, the International Atomic Energy Agency (IAEA) conducted a Coordinated Research Project (CRP) with the objective of better evaluating the hydrogeological conditions in the geological configuration of Nuclear Power Plants (NPP's) sites by using specific isotope techniques coupled with classic hydrochemical ones (IAEA 2021). The results of the CRP aimed at providing basic information that could be used for predicting chemical behavior and the fate of radioactive and conventional contaminants that may be released into groundwater during normal operations conditions or due to an accident.
A detailed understanding of groundwater hydrochemistry is relevant for the protection of the water resources as a whole in order to establish reference values for water quality and quantity for sustainable groundwater management (Sefie et al. 2018; Tizro and Voudouris 2008), mainly in regions where water is demanded by different uses with risk of conflicts.
The Brazilian nuclear power plant complex known as "Centrais Nucleares Almirante Álvaro Alberto-CNAAA" comprises two nuclear power plants in full operation, Angra I and Angra II, and another under construction (Angra III). The complex is situated in the municipality of Angra dos Reis (RJ), southeastern Brazilian coastline, near the limits of the states of São Paulo (SP) and Rio de Janeiro (RJ), the first and third largest Brazilian states in population, respectively.
Coastal aquifers of southeastern Brazil have a fundamental role in economic and social terms, as it is a densely populated region with intense economic activity. These aquifers belong to the Atlantic Coastal Hydrogeological Province (Bocanegra et al. 2010), extending intermittently along the Atlantic coast, from Uruguay (in the south) to the north, passing through Brazil, Guianas, Venezuela, and Colombia.
The objective of this study was to assess the hydrochemistry of a shallow aquifer at the location of Brazilian nuclear power plants and the surrounding area. Classical hydrochemical diagrams, scatter graphics, and hydrochemical indexes were used to identify the types and sources of water, the water mixtures, and the potentially feasible hydrochemical processes related to seawater influence, such as ion exchange and salinization.
Specifically, identifying the potential for cation exchange reactions is essential where nuclear or radioactive facilities are involved. It is not new that radium, being a divalent cation, similar to calcium and magnesium, can also be removed from water by the cation exchange process (EPA 1983; Nirdosh et al. 1984; Bi et al. 2016; Robin et al. 2017). Thus, natural cation exchange can attenuate aquifer contamination by radium.
The study was carried out in a highly touristic coastal region of the State of Rio de Janeiro (Fig. 1), about 20 km from the city of Angra dos Reis (204 thousand inhabitants), 120 km from the city of Rio de Janeiro (7 million inhabitants) and 220 km from the city of São Paulo (12 million inhabitants). The nuclear power plants occupy a coastal area called "Praia de Itaorna" (Itaorna Beach). The Highway BR-101 traverses the region at an average of 200 meters north of the NPP site limits.
The study area presents a typical microclimate regime of a tropical coastal region, influenced by factors like latitude, longitude, the proximity of the ocean, relief, nature of the vegetation cover, and, in particular, the action of atmospheric circulation systems in large and mid-scales, such as cold fronts and sea/land wind regimes. The site has a humid tropical climate, with an average annual rainfall of 1817 mm, with maximum rainfall occurring mainly from October to March, and an average annual temperature of 21.4ºC (data from 1982 to 2001, according to Eletronuclear 2005).
In general terms, the geomorphology of the place is characterized to the north by elevated terrains, constituted by the hills and escarpments of the Serra do Mar mountain range. To the south, occurs the fluvio-marine coastal plain, in the coves and beaches. The main hydrological pattern consists of rivers and streams that have their springs at the hillsides of the Serra do Mar and flow downhill towards the coastal plains and beaches. The Mambucaba River (Fig. 1) is the primary watercourse in the study area.
Geological and hydrogeological conceptual model
The study area has been the subject of hydrogeological and geological studies in the last 20 years, mainly due to surveys for licensing of nuclear power plants, as UFRJ (2003) and Eletronuclear (2005), among others. Figure 1 shows a geological map of the study area. A geological and hydrogeological schematic cross-section is shown in Fig. 2. The crystalline basement comprises granite-gneissic proterozoic rocks belonging to the Paraíba do Sul and Rio Novo Complexes (Eletronuclear 2005; CPRM 2007). These rocks are characterized by the predominance of garnet-biotite gneisses, hornblende-biotite gneisses, migmatites, and kinzigites. Porphyritic granitoid bodies containing microcline, oligoclase, quartz, and biotite occur embedded in the orthogneisses of the Rio Novo Complex. This Neoproterozoic unit has outcrops at NPP's site at Itaorna Beach. The Cambro-Ordoviciano Mambucaba Granite, mainly composed of oligoclase, biotite, and quartz, is located in the region of Mambucaba (east of the homonymous river plain) with outcrops at Itaorna Beach. It consists of small stocks and clusters of intrusive dykes in the rocks of the Rio Novo Complex. Expressive systems of basic (tholeiitic) rock dykes of Mesozoic age (Guedes et al. 2005) are observed in the study area, including Itaorna Beach (Eletronuclear 2005; CPRM 2007). These dykes have hydrogeological importance as they are preferably lodged along with pre-existing fractures and have internal orthogonal joint systems, which can store and conduct water.
The quaternary units are represented by alluvial deposits associated with the present-day and ancient drainage network, consisting of gravel, fine and medium sand, silt, and clay, including the deposits of terraces and floodplains. Fluvial-marine and marine deposits are also present, constituted by present-day and older sediments, representing deposits of the continental-marine transition environment to essentially marine. Fluvial-marine and marine sediments generally consist of layers of fine, medium, and coarse sand, alternated by levels of clay and silt, formed due to successive oscillations in sea level (Eletronuclear 2005).
The hillsides close to the NPP's site present colluvial and talus deposits, consisting of soils with a brownish, sandy, micaceous texture, with several blocks and boulders rolled in the soil matrix and intense vegetation. The intense cracking of the rocks causes numerous water springs at the foot of the hillsides next to the bed of the BR-101 highway (Eletronuclear 2005). There are mangrove deposits at the mouth of some rivers, characterized by silt-clayey muds and sludge rich in organic matter, dark in color, characteristically formed in a reducing environment derived from the decomposition of plant remains. Generally, the gneisses show local and regional schistosity and around N 50º to 70º E, with high dip angles, 60º to 80º NW. Two fracture directions are dominant in the area: NE and NNW.
Two main types of aquifers can be found in the region (Eletronuclear 2005). The first one is a free, porous, and shallow aquifer type (upper aquifer). The other is a fractured semi-confined type, characterized by fractures of crystalline rocks (Fig. 2). The unconfined and porous aquifer (0 to 40 meters deep) occurs in the lowlands, river, and coastal plains; it is represented by the river and marine river sediments, composed in its upper portion by sandy sediments (locally silt-rich), containing fragments of seashells. These fluvial-marine sediments occur over a sandy-clayey saprolite, containing fragments of the fresh rock and pebbles (eventually boulders) resulting from the weathering of the granitic-gneiss basement, forming the basis of the porous aquifer. In the NPP's facilities area, this aquifer is commonly covered by a layer of landscaping material. On the hillsides, this aquifer is represented by the saturated portions of the talus and colluvium deposits, residual soil and saprolite.
Locally, the fluvial-marine sediments are interspersed with layers of clay, one of which, of greater thickness (1 to 10 m thickness) and lateral continuity, is constituted by organic, dark, and plastic clay, presenting fragments of shells, and that mainly occurs over the saprolite. This clay layer must result from sea-level fluctuations, frequent in the quaternary (Martin et al. 1996). Due to the absence of more boreholes and piezometers, the lateral continuity of this clay layer can only be confirmed in the vicinity of the NPP's facilities. Undoubtedly, it is a clay aquifuge or aquitard that, when present, could be acting as a confining layer for the groundwater below. According to Eletronuclear (2005), this porous aquifer is in hydraulic connection with the superficial portions of the fractured aquifer.
The fractured and semi-confined aquifer (0 to undefined m deep) is represented by the fracture network of the crystalline basement rocks. It occurs both on the surface (outcrops) and in-depth under the porous aquifer. According to Eletronuclear (2005), the gradients and the hydraulic load are high, as the recharge is frequently at the top of the hillsides (up to 500 m high) located north of the NPP's site. This aquifer is unconfined in the elevated portions of the hillsides, occupied by extensive outcrops of the crystalline basement. It becomes confined under the layers of saprolite and colluvium/ talus deposits and the fluvial-marine deposits in the coastal region. Sometimes this aquifer has its piezometric level at or above the land surface (Eletronuclear, 2005). This fractured aquifer in granitic-gneiss crystalline rocks is very common in the coastal region of southeastern Brazil, with the recharge areas located in the highest parts of the hills (Silva-Filho et al. 2009)
Previous hydrogeochemistry studies at the NPP's site (UFRJ 2003; Eletronuclear 2005) identified three distinct groups of waters: Group I: Na-Cl water type, with a sub-group of these waters in which the concentration of calcium was a little greater: Ca-Na-Cl- water type; Group II, waters from the deep tubular wells: Ca-Na-HCO3-Cl water type and Group III: Ca-HCO3 water type. Regarding the groundwater flow, its direction is mainly from NNE to SSW and discharging into the ocean at the Itaorna Bay, which agrees with the ground surface topography. The water table is about 2 m below the ground surface, with a slope of 0.5% towards the ocean (Eletronuclear 2005).
Along with the duration of the CRP (2016–2020), about 70 water samples were collected, but this article will focus on hydrochemical data from twelve water samples collected in October-2018 (Fig. 1 and Table 1). Eight samples (ANG-16 to ANG-23) were collected from piezometers used for monitoring purposes (8–10 m deep) and located at the NPP's site, on Itaorna beach. The water samples from the piezometers were obtained with a low flow pumping system. The ANG-24 and ANG-25 samples were collected on the hillside close to the NPP's site: the first comes from a small dam, close to a water treatment station (EPTA); the second comes from a groundwater spring. For the Mambucaba River Plain, sampling (ANG-26 and ANG-27) was carried out from two shallow wells (4–8 meters deep), used for domestic purposes. In these wells, water samples were collected directly from the pumps installed in them, after repeated purges of the wells until the stability of electrical conductivity, to ensure that the samples will represent the groundwater.
Electrical conductivity (EC), total dissolved solids (TDS), redox potential (Eh), temperature (T), and pH were measured in situ using an Ultrameter II multiparameter handheld meter (Myron L Company). The water samples were collected in properly clean polyethylene bottles (250–500 mL). They were filtered through 0.45 µm membranes, and the samples for cation analyses were acidified with HNO3 (pH < 2) to stop bacterial growth and metal oxidation (Appelo and Postma 2005). The bottles were placed in thermally insulated boxes (± 4°C) and then taken to the laboratory.
The analytical determinations carried out were as follows: alkalinity (titration); cations Ca2+ and Mg2+, (atomic absorption/spectrometry); cations Na+ and K+ (liquid chromatography); anions, Cl− and SO42+ (liquid chromatography); Si (Inductively coupled plasma mass spectrometry-ICP-MS). All these analyses were performed at the Nuclear Technology Development Center (CDTN/Brazil). The hydrochemical data were evaluated through diagrams and scatter plots. The reliability of the chemical analyses was evaluated according to the acceptance criteria proposed by APWA (1999), using the following formula:
(1)The global mean composition for seawater estimated by Millero (2013) was used as a reference value. It was also used as reference data the ionic composition of rainwater (volume weighted average - VWA) determined by Souza et al. (2006) in rain samples collected at Ilha Grande, which is an island about 20 km offshore from Itaorna beach (Fig. 1). As the direct analysis of alkalinity was not done by Souza et al. (2006), the concentration of bicarbonate (HCO3-) ion was estimated from pH and atmospheric CO2 partial pressure, assuming gas–liquid equilibrium (Brimblecombe 1986; Tiwari et al. 2012), using the equation
[HCO3−] = KH K1 pCO2 / [H+] (2)
where,
KH Henry's law constant for CO2 solubility = 3.4x104 µeq/L atm− 1
K1 First dissociation constant for H2CO3 = 4.5x10− 1 µeq/L
pCO2 Partial pressure of CO2 = 3.87x10-4 atm
General Hydrogeochemistry
Descriptive statistics (Table 1), considering all data, show a significant variation in parameter values, resulting in high standard deviation (SD) values. Certainly, this is related to considering both continental water and waters with greater influence from the sea. The total dissolved solids (TDS) varied from 13.3-21130.0 mg/L with a mean value of 1900.3 mg/L. The pH ranged from acidic (5.4) to alkaline (7.9) with a neutral mean value (7.0). Considering the mean value, the abundance of the major ions is as follows: Na+ > Mg2+ > Ca2+ > K+ and Cl−> HCO3− > SO42+. However, this analysis considering all samples must be interpreted with care, as the extreme values measured in the Group 4 sample (ANG-22), with strong interference from the seawater, as we will see later, caused a large standard deviation in all parameters. The groups and their statistical data will be considered later when the water types are discussed. As regarding the seawater (Millero 2013), the TDS is 35046 mg/L (very high), the pH is a little alkaline 8.1, and the major concentration rank is: Na+ > Mg2+ > Ca2+ > K+ and Cl− > SO42− > HCO3−. As expected in relation to seawater, the rainwater (Souza et al. 2006) was more acidic, pH = 5.22, less mineralized, TDS = 13.174 mg/L and a similar ionic abundance rank: Na+ > Mg2+ > K+ > Ca2+ and Cl− > SO42− > HCO3−.
Water types
Based upon the Piper diagram (Piper 1944), the waters under investigation can be classified into four major groups (Fig. 3):
Group 1 – Ca-Na-HCO3 waters type. It is represented by a groundwater sample (ANG-23), a sample from a small reservoir (ANG-24) and a sample from a groundwater spring (ANG-25). These samples were collected in the hillside environment, where predominate granitic-gneissic basement outcrops, residual soil, talus deposits, and dense vegetation, representing an aquifer recharge zone, but with occurrences of discharge points from the fractured aquifer. Compared to the other groups (Table 1), the waters in this group are the least mineralized, with an average TDS of 28.3 mg/L and ranging from 13.3–52.0 mg/L. The lowest TDS value is similar to the value measured by Souza et al. (2006) in regional rainwater, 13.2. As well as the minimum pH value, 5.4, is similar to that found by this author in rainwater, 5.22. The average pH is 6.0, and the maximum is 6.8. Considering the mean value, the abundance of the major ions is as follows: Na+ > Ca2+ > K+ > Mg2+and HCO3− > Cl−> SO42+. Especially noteworthy is the high silica content of this group compared to the others.
Group 2 – Ca-HCO3 waters type. It is represented by samples ANG-16, ANG-17, ANG-18, ANG-19, ANG-20, and ANG-21. These are all groundwater samples collected at Itaorna Beach, close to the current and future NPP facilities. These samples come from the porous aquifer and seem to correspond to group 3, individualized in previous hydrogeological surveys carried out in the study area by UFRJ (2003) and Eletronuclear (2005). As shown in Table 1, the pH range is 6.8–7.9 (neutral to alkaline) with an average of 7.5. The TDS range is 119.0-417.2 mg/L with an average of 235.7 mg/L. These values indicate more mineralized and alkaline waters than those in Group 1. The major ions abundance is as follows: Na+ ≈ Ca2+ > Mg2+ > K+ and HCO3− > Cl−> SO42+
Group 3 – Ca-(Na)-HCO3 waters type. According to Piper Diagram, this group could be classified as a subgroup of Group 2. However, considering the Chadha Diagram (Chadha 1999, Fig. 3) and other geochemical plots shown later, it is better to consider it a separate group. This group seems to represent a mixture or an intermediate stage (flow and chemical composition) between the waters of groups 1 and 2. It is represented by two samples (ANG-26 and ANG-27) of groundwater collected in the Mambucaba River plain (Fig. 1). These samples were collected from the sandy porous aquifer on a coastal plain where ancient and current fluvial and marine sediments occur. The TDS values vary in an intermediate-range between Groups 1 and 3, 74–101 mg/L (87.1 average), and pH more similar to Group 1. The ionic abundance is as follows: Ca2+ > Na+ > K+ > Mg2+ and HCO3− > Cl−> SO42+
Group 4 – Na-Cl waters type. One sample (ANG-22) is represented, bearing an ionic composition typical of great seawater influence: It has the highest values for TDS (21130 mg/L) and ions. Presents the following rank in terms of ionic abundance (like the seawater): Na+ > Mg2+ > Ca2+ > K+ and Cl− > SO42+> HCO3−.This group corresponds to group 1 defined by the previous researches (UFRJ 2003; Eletronuclear 2005).
The sample data was plotted in the Chadha diagram (Chadha 1999) to identify the different types of water. In this diagram (Fig. 4), the X-axis represents the difference between (Ca2+ + Mg2+) and (Na+ + K+) in meq/L % and the Y-axis represents the difference between (HCO3−) and (Cl− + SO42−) in meq/L %. The four water types suggested by Chadha (1999) are shown in the quadrants of Fig. 4. It is found that the Group 1 samples fell in the Na-HCO3 type quadrant and the Group 2 samples fell in the Ca-HCO3 type quadrant, as noted in Piper's diagram. The Group 3 samples fell into the Ca-HCO3 type, but they can be considered an intermediate group: with less Ca than Group1 and less Na than Group 2, as proposed in the analysis of Piper's diagram (Fig. 3). The Group 4 sample fell in the Na-Cl type quadrant, very close to the global mean value for seawater (Millero 2013), clearly indicating seawater interference. This piezometer was perforated very close to the coastline and is subject to the fluctuation of the ocean tides.
Considering the elevation of the samples (Fig. 1 and Fig. 2) and the Chadha diagram (Fig. 4), it can be interpreted that starting from the recharge zone (hills and flat hillsides) towards the coastal zone, the sodium bicarbonate waters (Group 1) will enrich themselves in calcium (Groups 3 and 2) and at a certain point, they start to be influenced by seawater (Group 4).
The pH versus TDS (mg/l) scatter plot (Fig. 5) brings some information that confirms the observations made so far about groups, types of water, and direction of flow. It is observed in this plot that Group 1 has a pH ranging from 5.6 to 6.8 and the TDS < 70 mg/L. In relative terms, it is the group with the lowest pH and TDS values; thus, they are the most acidic and least mineralized waters. This behavior is undoubtedly related to the fact that they are waters from the recharge zone or have a short residence time in contact with the aquifers lithologies.
Showing pH and TDS (Fig. 5) values of 6.0-6.5 and 70–100 mg/L, respectively, are the samples from Group 3, which are clearly in an intermediate position between Group 1 and Group 2, in terms of H+ concentration and dissolved ions. Group 2 fell in a range with pH 6.8-8.0 and TDS between 100–1000 mg/L, more mineralized and alkaline than the samples of Group 1 and Group 3. Both aspects suggest the influence of carbonate weathering in the waters of Group 2, following the porous marine sedimentary aquifer characteristics. Group 4 presents a very high TDS value (21130 mg/L): one more indication of seawater interference, while the alkaline pH value indicates weathering of the carbonate terms of the porous coastal aquifer.
Hydrochemical processes
The Gibbs diagram (Fig. 6, Gibbs 1970) has been widely used as a first approach for understanding the relative importance of the significant natural processes that control and determine the composition of groundwater (Tiwari et al. 2019; Sefie et al. 2020; Nandakumaran et al. 2020). It depicts log TDS against the ratios of cations Na+/(Na++Ca2+) and the ratio of anions Cl−/(Cl−+HCO3−) and defines three main zones: evaporation, rainfall, and rock dominance.
The ANG-22 sample (Group 4) fell in the evaporation domain, which is one more indication of the seawater influence in this sample. Most of the samples (Group 2) fell into the rock dominance zone, which indicates that the rock-water interactions are the main factor that controls the chemical composition of these samples. Group 1 samples (ANG-23, ANG-24, and ANG-25) are in the zone of the rainfall domain. This suggests that although these samples must suffer some influence from the weathering of the rocks (mainly ANG-23), they have contact with the atmosphere, mainly concerning ANG-24 (spring area) and ANG-25 (surface water). The fact that these three samples are influenced by precipitation is consistent with how they occur on the slopes in places with crystalline rocks outcrops, residual soils, talus deposits, and denser vegetation (Fig. 1) and this also explains the low mineralization of these waters and the more acidic pH (Fig. 5). This acidification was likely due to one of the following factors, or their combination: organic matter decomposition (reaction 3), respiration of plants (reaction 4), and the natural pH of uncontaminated rainwater in the atmosphere, 5.5-6.0 (reaction 5) (Hounslow 1995; Appelo and Postma 2005).
CH2 + O2 = > CO2 + H2O = > H2CO3 (3)
C6H12O6 + 6O2 = > 6CO2 + 6H2O = > H2CO3 (4)
CO2(g) = > CO2(aq) ; CO2(aq) + H2O = > H2CO3 (5)
It is observed (Fig. 6) that the samples ANG-26 and ANG-27 (Group 3) are in an intermediate position between rock dominance (Group 2) and rainfall dominance (Group 1) zones. This suggests that they are water resulting from the mixture of these two groups and/or that depicts an intermediate step in the chemical evolution of the waters of the study area, departing from Group 1, less mineralized waters, to Group 2, with more significant rock-water interaction.
Scatter plot HCO3 /SIO2 ratio versus Mg/(Ca + Mg) ratio. The scatter plot in Fig. 7 was prepared based on Hounslow (1995) and aimed to indicate the type of rock with which the water interacted when passing through pores and/or fractures. It can be seen in this figure that the samples ANG-16 to ANG-21 (Group 2) fell in area 1A, indicating that their composition is predominantly related to the "Not silicate weathering" and, more specifically, to the weathering of carbonate rocks (calcite-dolomite minerals). The ANG-22 sample (Group 4) is also related to the weathering of non-silicate rocks but fell in area 1B, indicating seawater interference. These observations made for Groups 2 and 4 are in line with the characteristics defined previously for the porous fluvial-marine aquifer of the coastal region, locally with interference from seawater.
The samples ANG-23 to ANG-25 (Group 1) are in area 2A, suggesting waters with chemical composition influenced by the weathering of silicates, and more specifically, granite composition rocks. This behavior is consistent since these samples were collected on the hills where the fractured aquifer of granitic-gneisses composition outcrops, and therefore its chemical weathering products are present. The ANG-26 and ANG-27 (Group 3) samples are positioned on or slightly above line y = 5 (HCO3/SiO2 ratio) in area 3, depicting an ambiguous character and reinforcing that they have an ionic composition between Groups 1 and 2.
The chloro-alkaline index-CAI (Schoeller 1965). This index is calculated to assess the potential for ion exchange reactions between groundwater and mineral constituents in aquifers. CAI also indicates whether an aquifer is salinizing or freshening or has gone through these processes in the past. (Tiwari et al 2019; Stuyfzand 2008).
The ion exchange can be of the type called "base ion exchange process", in which Mg2+ and/or Ca2+ ions in the water are exchanged for Na+ and/or K+ from the mineral aquifer matrix. This is a softening process, since Ca2+ and Mg2+ are removed from the aqueous phase, reducing the water hardness. On the other hand, in the mechanism called "reverse ion exchange process", the opposite occurs (or hardening), that is, sequestration of Na+ and K+ ions by the mineral matrix of the aquifer and releasing of Ca2+ and/or Mg2+ ions to the aqueous phase. Reaction 6 presents an example of an ion-exchange reaction involving Ca2+ and Na+ (Hounslow 1995, Appelo and Postma 2005).
½ Ca2+ + Na-X => ½ Ca-X + Na+ (6)
→ Base cation exchange, softening, or freshening process →
← Reverse cation exchange, hardening, or salinizing process ←
Where X = solid exchanger (mainly clay and hydroxides minerals)
The CAI was calculated using the Schoeller (1965) formula:
CAI= [Cl-(Na + K)]/Cl (meq/L) (7)
CAI values will be positive if there is a hardening process, negative if a softening process is present, and there is no ion exchange process if CAI values are close to zero (Abu-alnaee et al. 2018; Tiwari et al. 2019). The scatter plot CAI values versus Cl−concentrations (Fig. 8a) shows that ANG-22 (Group 4) has CAI slightly higher than zero, indicating no ion exchange or a hardening (salinization) process, confirming the effect of seawater intrusion and saltwater upcoming for this water sample. The other samples (Groups 1 to 3) presented CAI < 0, indicating the potential for softening or freshening process, where Ca2+ decreases and Na+ increases in the aqueous phase.
This indication of a possible ion exchange softening process in the shallow porous aquifer at Itaorna Beach, the NPP site, should be pointed out. In radium-contaminated water, an artificial cation exchange resin can be used for removing Ra2+ (EPA 1983; Nirdosh et al. 1984; Bi et al. 2016; Robin et al. 2017). Thus, theoretically, in case of an accidental release of radionuclides from NPP to the aqueous phase, the fluvial-marines clays can capture Ra2+, releasing Na+ for instance, and mitigating groundwater radium contamination.
Scatter plots (Ca2+ +Mg2+ ) versus HCO3− (Fig. 8b) and Ca2+ versus HCO3− (Fig. 8c). These plots were done to identify the potential sources of Ca2+ and Mg2+ (Shansi et al. 2018). Figure 8b shows that the sample ANG-22 (Group 4) has an excess of (Ca + Mg), as indicative of seawater, enriched mainly in Mg2+ (Millero 2013) and the reverse ion exchange process (hardening). Regarding the other samples, it is observed that most of them (Group 2) fall on the 1:1 stoichiometric equilibrium line (Ca2++ Mg2+) versus HCO3−, suggesting dissolution of dolomite or Mg-rich calcite according to reaction 8 (Hounslow 1995; Appelo and Postma 2005). However, some (Groups 1 and 3 samples) fell slightly below line 1:1, suggesting a base ion exchange process (softening).
CaMg (CO3)2 + 2CO2 + 2H2O -> Ca2+ + Mg2+ + 4HCO3− (8)
In Fig. 8c, except for the ANG-22 sample (Group 4), all samples fall close or below 1:1 stoichiometric equilibrium line Ca2+ versus HCO3−. The samples on the line suggest waters composition with the main contribution of calcite weathering, according to reaction 9 (Hounslow 1995; Appelo and Postma 2005). The samples below the 1:1 line suggest a Ca2+ deficit, which may be due to calcium capture by the aquifer's solid phase (mainly clays) in an ion exchange process (softening) involving Na+ (mostly Group 1). The indications provided by these plots (Fig. 8b and 8c) are consistent with the previous observations
CaCO3 + H2O + CO2 -> Ca2+ + 2HCO3− (9)
Scatter plot Na+ versus Cl−. This plot is used to identify potential sources of Na+ and Cl− (Shansi et al. 2018). Figure 8d clearly shows that, for the ANG-22 sample (Group 4), the source of Na+ and Cl− is the seawater, which confirms the strong influence of seawater on the ionic composition of this sample. Most of the data are slightly above the halite dissolution line. The Na+/Cl− ratio greater than 1 indicates either the presence of silicate weathering or base ion exchange (softening) between Na+ and Ca2+ (Shansi et al. 2018). For Group 2 samples, it is suggested that ion exchange (Ca-Na) is the leading cause for Na + excess. For samples from Groups 1 and 3, both processes must occur, whereas for samples ANG-23 to ANG-25 (Group 1), it is suggested that the excess of Na+ is predominantly due to the weathering of sodium-rich silicates (Na-rich plagioclase). These observations confirm the interpretation performed in Fig. 8a, 8b, and 8c. In other words, these data show a Na+ enrichment and concomitant Ca2+ depletion in most of the water samples, except for the ANG-22 sample, in which the saline influence is significant.
The study area is characterized by a porous, free and shallow aquifer, consisting of talus deposits, residual soil, and marine to marine-fluvial sediments, occurring predominantly in the coastal plain. A fractured aquifer, with a granite-gneissic composition, generally occurring under the porous aquifer but outcropping in the hills and hillsides. Based on the hydrochemical assessment, four types of water were identified in the shallow aquifer. The waters of Group 1 occur in a hillside environment, where it predominates granitic-gneissic basement outcrops, residual soil, talus deposits, and dense vegetation, characterizing an aquifer recharge zone, but with occurrences of discharge points from the fractured aquifer. Where the fluvial-marine sediments prevail, the waters of Groups 2 and 3 are present, in addition to Group 4, which is the closest point to the sea. The waters in Group 1 are Ca-Na+-HCO3−, not very mineralized (TDS < 70 mg/L), with ionic composition indicative of the weathering of granite-gneissic rocks. Based on the Gibbs diagram, Group 1 is in the rainfall dominance zone, which indicates that the rainwater infiltration is the main factor governing its chemical composition. Group 2 consists of Ca2+-HCO3− waters, more mineralized (100–500 mg/L TDS) than those of group 1, with ionic composition indicative of carbonate weathering, with rock dominance (Gibbs diagram), which indicates that the dissolution of rock-forming mineral is the main factor controlling the water chemistry of this group. Group 3 has intermediate hydrochemical characteristics between groups 1 and 2. For example, it has 70–100 mg/L TDS. The sample that characterized Group 4 clearly shows some interference from seawater, such as 20,000 mg/L of TDS and indicative of halite dissolution. In the Gibbs diagram, the sample felt in the typical evaporation dominance zone, indicating that the evaporation of seawater is the predominant factor in its chemical composition. The waters of groups 1 to 3 showed potential for Base cation exchange (softening or freshening process), while in Group 4, the potential is for reverse cation exchange (hardening or salinizing process). From the hillsides towards the coastline, the research results suggest a hydrochemical evolution of local groundwater from Group 1 (high rainwater influence), passing through Group 3 (intermediate between Group 1 and 2) and Group 2 (rock influence), to Group 4 ( seawater like).
In a general hydrogeology context, the methods and techniques applied proved to be particularly efficient in identifying water sources, water mixtures, and the salinization and freshening processes involved. The freshening cation exchange identified in the shallow aquifer in Itaorna Beach is noteworthy. In case of an accidental release of radionuclides from the Nuclear Power Plants to the aqueous phase, the local clay lenses can capture Ra2+ in a natural contamination attenuation process.
Acknowledgments: The authors would like to thank the International Atomic Energy Agency (IAEA) for funding this research through CRP F33022 and also wish to thank Eletronuclear for its technical and operational support.
Funding International Atomic Energy Agency (IAEA)
Conflicts of interest/Competing interests Not applicable
Availability of data and material Not applicable
Code availability Not applicable
Authors' contributions Not applicable
Additional declarations Not applicable
Ethics approval Not applicable
Consent to participate Not applicable
- Abu-alnaeem MF, Yusoff I, Ng TF, Alias Y, Raksmey M (2018) Assessment of groundwater salinity and quality in Gaza coastal aquifer, Gaza Strip, Palestine: an integrated statistical, geostatistical and hydrogeochemical approaches study. Sci Total Environ 615:972e989
- APHA American Public Health Association (1999) Standard methods for the Examination of water and wastewater. American Public Health Association, Washington, DC
- Appelo CAJ, Postma D (2005) Geochemistry, Groundwater and Pollution. Balkema, Rotterdam, 2ND edition, p. 634
- Bi Y, Zhang H, Ellis BR, Hayes KF (2016) Removal of Radium from Synthetic Shale Gas Brines. ENVIRONMENTAL ENGINEERING SCIENCE, Volume 33, Number 10, 2016. DOI: 10.1089/ees.2016.0002
- Bocanegra E, Da Silva Jr GC, Custodio E, Manzano M, Montenegro S (2010) State of knowledge of coastal aquifer management in South. America Hydrogeology Journal 18:261–267
- Brimblecombe P (1986) Air Composition and Chemistry. Cambridge University Press, VIII, 224 pp
- Chadha DK (1999) A proposed new diagram for geochemical classification of natural waters and interpretation of chemical data. Hydrogeology Journal; 7:431–439
- CPRM – Serviço Geológico do Brasil (2007) Geologia da Folha Angra dos Reis SF.23-Z-CII (escala 1:100.000): Ministério das Minas e Energia, Brasília (2007)
- Eletronuclear (2005) Estudo de Impacto Ambiental Unidade 3 da Central Nuclear Almirante Álvaro Alberto – ANGRA 3 – volume 2, maio de 2005
- EPA U.S. Environmental Protection Agency (1983) Radionuclide Removal for Small Public Water System. Washington, DC
- Gibbs RJ (1970) Mechanisms controlling world water chemistry. Science 170:1088–1090. https://doi.org/10.1126/science.170.3962.1088
- Guedes E, Heilbron M, Vasconcelos PM, Valeriano CM, Almeida JCH, Teixeira W, Thomaz Filho A (2005) K-Ar and 40Ar/39Ar ages of dikes emplaced in the onshore basement of Santos Basin, Resende area, SE Brazil: implications for the South Atlantic opening and Tertiary reactivation. J S Am Earth Sci 18:371–382
- Hounslow AW (1995) Water Quality Data: Analysis and Interpretation. CRC Press LLC, Lewis Publishers, Boca Raton
- IAEA International Atomic Energy Agency (2021) Use of Isotope Hydrology to Characterize Groundwater Systems in the Vicinity of Nuclear Power Plants. MONOGRAPH, Results of a Coordinated Research Project (CRP) F33022, 2016–2020. Vienna
- Martin L, Suguio K, Flexor J, Dominguez JML, Bittencourt ACSP (1996) Quaternary sea-level history and variation in dynamics along the Central Brazilian Coast: consequences on coastal plain construction. An Acad Bras Ciências v 68(3):303–354
- Millero FJ (2013) Chemical Oceanography. CRC Press, 4ª edição, 591 p
- Nandakumaran P, Balakrishnan K (2020) Groundwater quality variations in Precambrian hard rock aquifers: a case study from Kerala, India. Applied Water Science (2020) 10:2. https://doi.org/10.1007/s13201-019-1084-8
- Nirdosh I (1984) Radium in uranium mill tailings -- some observations on retention and removal. Hydrometallurgy, 12 (1984) 151–176 151
- Piper AM (1944) A graphic procedure in the geochemical interpretation of water analyses. Trans Am Geophys Union 25:914–923
- Robin V, Tertre E, Beaucaire C, Regnault O, Descostes M (2017) Experimental data and assessment of predictive modeling for radium ion-exchange on beidellite, a swelling clay mineral with a tetrahedral charge. Appl Geochem 85:1–9. https://doi.org/10.1016/j.apgeochem.2017.07.009
- Schoeller H (1965) Qualitative evaluation of groundwater resources. Methods Tech. Groundw. Investig. Dev. UNESCO, pp 54e83
- Sefie A, Aris AZ, Ramli MF, Narany TS, Shamsuddin MKN, Saadudin SB, Zali MA (2018) Hydrogeochemistry and groundwater quality assessment of the multilayered aquifer in Lower Kelantan Basin, Kelantan, Malaysia. Environmental Earth Sciences (2018) 77:397.https://doi.org/10,10007/s12665-018-7561-9
- Shamsi A, Karami GH, Hunkeler D, Taheri A (2019) Isotopic and hydrogeochemical evaluation of springs discharging from high-elevation karst aquifers in Lar National Park, northern Iran. Hydrogeol J 27(2):655–667. https://doi.org/10.1007/s10040-018-1873-4
- Silva-Filho EV, Sobral Barcellos RG, Emblanch C, Blavoux B, Sella SM, Daniel M, Simler R, Wasserman JC (2009) Groundwater chemical characterization of a Rio de Janeiro coastal aquifer, SE e Brazil. J South Am Earth Sci 27:100e108
- Souza PA, Mello WZ, Maldonado J (2006) Composição química da chuva e aporte atmosférico na Ilha Grande, RJ. Quim. Nova. Vol. 29. No. 3, 471–476
- Stuyfzand PJ (2008) Base exchange indices as indicators of salinization or freshening of (coastal) aquifers. 20th Salt. Water Intrusion Meeting. Naples, Florida
- Tiwari S, Chate DM, Bisht DS, Srivastava MK, Padmanabhamurty B (2012) Rainwater chemistry in the North Western Himalayan Region, India. Atmospheric Research 104–105 (2012) 128–138
- Tiwari AK, Pisciotta A, De Maio M (2019) Evaluation of groundwater salinization and pollution level on Favignana Island, Italy. Environmental Pollution 249 (2019) 969 – 98. https://doi.org/10.1016/j.envpol.2019.03.016
- Tizro AT, Voudouris KS (2008) Groundwater quality in the semiarid region of the Chahardouly basin, West Iran. Hydrol Process 22(6):3066–3078
- UFRJ Universidade Federal do Rio de Janeiro – Instituto de Geociências (2003) Levantamento e Diagnóstico Ambiental (Meio Físico) da Área de Influência da Central Nuclear (CNAAA) – Volume II, Eixo 2 – Geologia e Recursos Hídricos. UFRJ, Rio de Janeiro
|
Group
|
Stats
|
EC µS/cm
|
TDS mg/L
|
pH
|
HCO3−mg/L
|
Cl− mg/L
|
SO42− mg/L
|
K+ mg/L
|
Na+ mg/L
|
Mg2+ mg/L
|
Ca2+ mg/L
|
SiO2 mg/L
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
ALL N = 12
|
Mini
|
23
|
13.3
|
5.4
|
11
|
2.33
|
0.94
|
0.84
|
2.61
|
0.3
|
0.6
|
6.37
|
|
Maxi
|
23190
|
21130.0
|
7.9
|
357
|
8167.86
|
1200.36
|
164.36
|
4388.15
|
490.0
|
211.0
|
16.48
|
|
|
Mean
|
2161
|
1900.3
|
7.0
|
134
|
688.24
|
105.28
|
18.56
|
377.241
|
46.0
|
40.3
|
9.51
|
|
|
SD
|
6625
|
6056.9
|
0.9
|
108
|
2355.48
|
344.89
|
46.16
|
1263.19
|
140.0
|
57.3
|
3.11
|
|
|
1 N = 3
|
Mini
|
23
|
13.3
|
5.4
|
11
|
2.33
|
0.94
|
0.84
|
2.61
|
0.3
|
0.6
|
11.84
|
|
Maxi
|
88
|
52.0
|
6.8
|
32
|
6.87
|
1.28
|
1.32
|
5.37
|
0.8
|
1.9
|
16.48
|
|
|
Mean
|
48
|
28.3
|
6.0
|
18
|
4.30
|
1.06
|
1.11
|
3.72
|
0.6
|
1.4
|
14.08
|
|
|
SD
|
35
|
20.8
|
0.7
|
12
|
2.33
|
0.19
|
0.24
|
1.46
|
0.2
|
0.7
|
2.32
|
|
|
2 N = 6
|
Mini
|
200
|
119.9
|
6.8
|
99
|
3.94
|
3.22
|
3.32
|
4.65
|
1.2
|
25.0
|
6.37
|
|
Maxi
|
666
|
417.2
|
7.9
|
357
|
28.07
|
9.34
|
15.77
|
53.27
|
23.0
|
52.0
|
9.49
|
|
|
Mean
|
384
|
235.7
|
7.5
|
207
|
11.17
|
5.84
|
8.24
|
18.67
|
9.6
|
41.7
|
7.80
|
|
|
SD
|
155
|
99.2
|
0.4
|
86
|
9.06
|
2.28
|
5.01
|
18.33
|
7.7
|
9.6
|
1.29
|
|
|
3 N = 2
|
Mini
|
125
|
74.0
|
6.0
|
42
|
4.75
|
10.24
|
2.58
|
6.76
|
1.1
|
7.9
|
7.72
|
|
Maxi
|
169
|
100.1
|
6.6
|
58
|
6.35
|
14.53
|
2.99
|
8.79
|
1.8
|
11.0
|
9.74
|
|
|
Mean
|
147
|
87.1
|
6.3
|
50
|
5.55
|
12.39
|
2.79
|
7.78
|
1.5
|
9.5
|
8.73
|
|
|
SD
|
31
|
18.5
|
0.4
|
11
|
1.13
|
3.03
|
0.29
|
1.44
|
0.5
|
2.2
|
1.43
|
|
|
4 N = 1
|
value
|
23190
|
21130
|
7.6
|
210
|
8167.86
|
1200
|
164.36
|
4388
|
490.0
|
211.0
|
7.62
|
|
Seawater*
|
-
|
35046
|
8.1
|
105
|
19353
|
2712
|
399
|
10781
|
1284
|
412
|
-
|
|
|
Rainwater**
|
-
|
13.174
|
5.2
|
0.06
|
6.32
|
1.67
|
0.28
|
3.27
|
0.49
|
0.18
|
-
|
|
*Global mean composition for seawater estimated by Millero (2013)
**Ionic composition of rainwater determined by Souza et al. (2006) at Ilha Grande