Redox control on trace element geochemistry and provenance of groundwater in fractured basement of Blantyre, Malawi

Journal of African Earth Sciences 100 (2014) 335–345

Contents lists available at ScienceDirect

Journal of African Earth Sciences journal homepage: www.elsevier.com/locate/jafrearsci

Redox control on trace element geochemistry and provenance of groundwater in fractured basement of Blantyre, Malawi Harold Wilson Tumwitike Mapoma a,b, Xianjun Xie a,⇑, Liping Zhang a a b

School of Environmental Studies, China University of Geosciences (Wuhan), Lumo Road 388, Wuhan 430074, Hubei, China Department of Physics and Biochemical Sciences, University of Malawi, The Polytechnic, Private Bag 303, Blantyre 3, Malawi

a r t i c l e

i n f o

Article history: Received 12 March 2014 Received in revised form 14 July 2014 Accepted 15 July 2014 Available online 26 July 2014 Keywords: Isotope Irrigation return flow Groundwater Trace elements PHREEQC d-Excess

a b s t r a c t Assessment of redox state, pH, environmental isotope ratios (d18O, d2H) coupled with PHREEQC speciation modeling investigations were conducted to understand trace element geochemical controls in basement complex aquifer in Blantyre, Malawi. Groundwater in the area is typical Ca–Mg–Na–HCO3 type suggesting more of carbonate weathering and significance of carbon dioxide with dissolution of evaporites, silicate weathering and cation exchange being part of the processes contributing to groundwater mineralization. The significance of pH and redox status of groundwater was observed. The groundwater redox state was mostly O2-controlled with few exceptions where mixed (oxic–anoxic) O2–Mn(IV) and O2–Fe(III)/SO4 controlled redox states were modeled. More so, some of the main trace element species modeled with PHREEQC varied with respect to pH. For instance vanadium(III) and vanadium(IV) decreased with increase in field pH contrasting the trend observed for vanadium(V). The isotopic composition of the sampled groundwater varied between 5.89‰ and 3.32‰ for d18O and 36.98‰ and 20.42‰ for d2H. The d2H/d18O and d18O/Cl ratios revealed that groundwater is of meteoric origin through vertical recharge and mixing processes. The d-excess value approximated the y-intercept of GMWL of 10 (d-excess = 9.269, SD = 1.240) implying that influence of secondary evaporative processes on isotopic signature of the study area is minimal. Thus, there is evidence to suggest that groundwater chemistry in the studied aquifer is influenced by inherent processes with contribution from human activities and furthermore, the water originates from rainwater recharge. With such results, more studies are recommended to further constrain the processes involved in mineralization through isotopic fractionation investigations. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Groundwater geochemistry is controlled by various factors. These factors include pH, availability of mineral sources, weathering processes, temperature, reduction/oxidation (redox) reactions and anthropogenic input. Each factor plays an inherent role in groundwater mineralization or may be induced by human activities. Groundwater is a primary resource for domestic and agricultural purpose in rural communities. Groundwater exploitation may increase the levels of natural contaminants or inject additional parameters into the aquifer. Besides, evapotranspiration mostly in arid and semi-arid regions increases the potential for evaporate increase in groundwater through leaching processes. Processes controlling groundwater mineralization in these areas

⇑ Corresponding author. Tel.: +86 27 67883170; fax: +86 27 87436235. E-mail addresses: [email protected], [email protected] (X. Xie). http://dx.doi.org/10.1016/j.jafrearsci.2014.07.010 1464-343X/Ó 2014 Elsevier Ltd. All rights reserved.

requires proper evaluation in order to effectively manage and conserve groundwater (Xie et al., 2013). Groundwater chemical composition is generally influenced by mineral dissolution within the aquifer matrix, evaporation and anthropogenic activities or combination of these processes (Ghabayen et al., 2006; Fass et al., 2007). Whenever redox processes are involved in groundwater, they can mobilize or immobilize potentially toxic elements from geogenic materials (Smedley and Kinniburgh, 2002). Consequently, redox processes enhance degradation or preservation of anthropogenic contaminants (Bradley, 2003). Mobilization and degradation of chemical elements in groundwater may generate undesirable by-products such as dissolved ferrous iron (Fe2+), hydrogen sulfide (H2S) and methane (CH4) (Chapelle and Lovley, 1992). Determining the kind of redox processes that occur in an aquifer, documenting spatial distribution and understanding how they affect concentrations of natural or anthropogenic contaminants, is central to assessing and describing the chemical quality of groundwater (Christensen et al., 2000; McMahon and Chapelle, 2008).

336

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

Then again, isotopic investigation has proven useful in ascertaining the origin of groundwater. In recent years various studies have been conducted on stable isotopes elsewhere (Dindane et al., 2003; Ako et al., 2012; Amer et al., 2012; Hamed and Dhahri, 2013; Xie et al., 2013) and results have been very significant to hydrological assessments of groundwater. Various studies done in basement complex of Malawi have documented the effect of hydrolysis of aluminosilicate, evaporite dissolution, cation exchange and carbonate weathering (Msonda et al., 2007; Wanda et al., 2013) and anthropogenic contamination (Mkandawire, 2008; Kanyerere et al., 2012; Msilimba and Wanda, 2013). However, not much has been done on redox processes and isotopic investigation in the complex (Mapoma and Xie, 2014). Therefore, the foremost objective of the present study was to investigate how redox processes affect trace element chemistry in the fractured basement complex of rural Blantyre. Subsequently, we analyzed and assessed the provenance of groundwater in the fractured basement aquifer through stable oxygen and hydrogen isotope concentrations. The study results will contribute to better groundwater management and conservation in rural Blantyre district. 2. Geological and hydrological setting of the study area The location of the study area is illustrated in Fig. 1 which is within Blantyre district: where Blantyre district is located at 15°4200 S and 35°E coordinates. The site elevation and coordinates of each sampling point are indicated in Table 1 under Section 4. The lithology of the area is characterized by crystalline basement complex mostly of metamorphic and igneous rocks of Pre-cambrian to lower Palaeozoic age (Carter and Bennett, 1973; Chilton and Smith-Carington, 1984). The study area is on the eastern edge of the southern branch of the East African Rift Valley where prominent faults occur. The basement is fractured enough for potential

aquifers to exist for groundwater exploitation (Mapoma and Xie, 2014). The major aquifer lithological units are syenitic granites, charnockitic and ultra-basic gneisses, schistis, granulalite and quartzites (Mkandawire, 2004; Monjerezi and Ngongondo, 2012). The lithofacies of the study area are mainly dark grey gneisses strongly enriched in amphibole and pyroxene. Charnockitic gneisses (calcium–iron–magnesium) and syenitic gneisses (abundance of feldspar) have been identified in the area (Dill et al., 2005). Common mineral bearing rocks present in this area include quartz, feldspar, amphibole, hematite, hydrobiotite, kaolinite, calcite, gypsum, vermiculite, goethite, halloysite and smectite (Dill et al., 2005). The presence of such mineral phases determines the abundance (or trace existence) of elements in the study area. Various mechanisms controlling mineralization of elements in basement complex have been studied (e.g. Monjerezi et al., 2011; Wanda et al., 2013). Of significant are cation exchange, rock–water interaction, carbonate and silicate weathering and hydrolysis. Oxic conditions in fractured basement promote redox reactions that may mobilize redox sensitive elements such as Fe, Mn and V. Some of the mineral end members identified to be of significance in the entire basement complex that are involved in trace element geochemistry include aluminosilicates, evaporites (anglesite, barite and celestite), carbonates (cerussite, siderite, strontianite and witherite), oxides (cassiterite, cuprite, hematite and karelianite), sulfides (chalcopyrite and galena), mineral alloys e.g. electrum (Ag–Au alloy) and hydroxides (garnierite and nickeliferous limonite) (Monjerezi et al., 2011; Wanda et al., 2013; Mapoma et al., submitted for publication-b). Aquifer recharges from higher grounds and groundwater discharge in surface depressions (Fig. 1). Although there are variations in groundwater flow direction on a small scale dependent on landscape geodynamics, it can be generalized that the flow is from the east in the uplands (shire highlands) towards the west into Shire Valley (part of Zambezi Valley). Fig. 2 is a conceptualized cross

Fig. 1. Location of the study area in the shire highlands of Blantyre showing sampling points and geological features.

Table 1 Oxidation states of main species in lmol/L realized from PHREEQC speciation modeling aligned with field measurements. Elevation (m. amsl)

X

Y

pH

Temp (°C)

EC (lS/cm)

TDS (mg/L)

Au(+1)

B(+3)

Ba(+2)

Co(+2)

Co(+3)

Cr(+2)

Cr(+3)

Cr(+5)

Cr(+6)

BH01 BH02 BH03 BH04 BH05 BH06 BH07 BH08 BH09 BH10 BH11 BH12 BH13 BH14 BH15 BH16 BH17 BH18 BH19 BH20 BH21

1118 1111 1088 1070 1058 1101 1093 1102 1106 1037 1059 1050 1021 1031 1071 960 985 944 938 890 886

722,199 722,563 722,294 723,536 722,598 723,324 723,852 724,377 723,805 725,495 726,904 726,691 725,411 725,033 724,662 724,034 724,523 723,783 723,565 722,709 722,932

8,255,012 8,256,578 8,261,044 8,260,543 8,261,792 8,261,994 8,262,139 8,262,527 8,263,720 8,265,334 8,265,307 8,265,100 8,266,031 8,266,539 8,266,205 8,265,358 8,265,508 8,266,868 8,266,807 8,266,815 8,267,200

6.44 6.54 6.44 6.35 6.32 6.1 6.35 6.33 6.45 6.43 6.61 6.57 6.4 6.34 6.64 7.1 7.13 6.97 6.81 6.91 7.21

23.3 22.5 23.2 23.2 23.4 23.1 23.3 23.3 23.5 23.8 23.4 23.3 23.8 23.8 24 24.4 23.8 24.4 25.2 25.1 25.3

825 639 279 307 265 159.7 422 248 301 307 290 304 403 280 266 873 379 414 371 554 873

413 319 139 155 134 79.9 211 125 150 153 146 148 204 142 132 438 191 206 187 277 417

0.49 0.49 0.56 0.57 0.54 0.47 0.57 0.58 0.41 0.65 0.59 0.47 0.71 0.67 0.67 0.55 0.59 0.57 0.60 0.67 0.53

3.60 3.22 2.39 2.32 1.70 2.24 2.19 2.22 2.35 1.91 1.92 1.70 2.13 1.93 1.87 1.41 1.86 1.39 1.47 2.48 1.53

4.49 4.24 0.58 0.54 0.45 0.25 0.79 0.48 0.53 0.34 0.54 0.71 3.48 0.87 1.24 0.18 1.62 0.32 0.36 2.37 0.20

0.09 0.02 0.01 0.08 0.06 0.02 0.04 0.04 0.03 0.02 0.06 0.06 0.07 0.04 0.04 0.07 0.04 0.08 0.08 0.05 0.06

3.55  1034 n.d n.d 4.41  1034 2.50  1034 1.12  1034 2.54  1034 1.63  1034 n.d n.d 1.85  1034 1.66  1034 3.42  1034 1.30  1034 1.79  1034 1.12  1034 n.d 1.87  1034 3.28  1034 1.21  1034 1.11  1034

8.90  1015 3.23  1015 9.80  1015 4.79  1015 2.83  1015 1.43  1014 5.34  1015 1.39  1014 5.28  1015 6.16  1015 8.95  1015 6.77  1015 7.17  1015 8.43  1014 n.d 1.16  1016 6.17  1016 1.27  1015 n.d 2.06  1014 7.24  1016

0.01 0.01 0.02 0.01 0.00 0.02 0.01 0.03 0.01 0.02 0.04 0.02 0.01 0.11 0.01 0.00 0.01 0.01 n.d 0.12 0.03

1.77  1024 4.71  1025 7.81  1025 7.77  1025 1.10  1025 5.09  1026 8.19  1025 9.47  1025 8.51  1025 3.25  1024 3.53  1023 7.56  1024 1.29  1024 2.91  1024 3.17  1023 2.31  1021 3.75  1021 1.95  1021 n.d 6.18  1021 3.75  1019

6.91  1026 8.76  1027 3.60  1026 8.37  1026 1.01  1026 1.26  1026 7.62  1026 9.79  1026 4.60  1026 3.29  1025 1.73  1024 3.35  1025 9.51  1026 2.18  1025 2.10  1024 3.65  1023 5.26  1023 5.80  1023 n.d 1.55  1022 1.49  1020

Sample

Cu(+1)

Cu(+2)

Fe(+2)

Fe(+3)

Li(+1)

Mn(+2)

Mn(+3)

Ni(+2)

Pb(+2)

Se(2)

Se(+4)

Se(+6)

Sn (+2)

Sn (+4)

Sr

V(+3)

V(+4)

V(+5)

BH01 BH02 BH03 BH04 BH05 BH06 BH07 BH08 BH09 BH10 BH11 BH12 BH13 BH14 BH15 BH16 BH17 BH18 BH19 BH20 BH21

0.39 0.03 0.08 0.04 0.09 0.07 0.05 0.11 0.07 0.40 0.02 0.01 0.07 0.01 0.07 0.02 0.08 0.06 0.16 0.02 0.01

3.4  103 2.6  104 1.1  103 1.1  103 1.8  103 1.6  103 1.0  103 2.1  103 1.3  103 5.9  103 4.4  104 1.1  104 1.4  103 1.8  104 1.7  103 6.1  104 1.9  103 1.4  103 4.1  103 4.9  104 1.6  104

0.08 0.69 0.15 0.56 0.48 0.28 0.29 0.13 0.14 0.14 0.10 0.17 0.12 0.33 0.14 19.37 0.26 0.19 0.12 0.43 19.62

3.6  107 3.0  106 5.5  107 2.7  106 1.5  106 6.2  107 1.6  106 5.2  107 5.6  107 1.4  106 1.4  106 1.8  106 5.3  107 1.1  106 2.9  106 3.1  103 3.6  105 1.7  105 6.1  106 1.6  105 1.1  102

1.14 0.81 0.32 0.48 0.20 0.26 0.39 0.26 0.23 0.46 0.42 0.55 0.52 0.29 0.35 0.92 0.22 0.24 0.42 0.53 1.17

14.52 0.02 0.17 0.01 0.03 0.08 0.01 0.01 0.19 0.01 0.18 2.13 0.01 0.01 n.d 1.27 0.07 2.01 0.02 0.01 1.64

1.8  1024 1.1  1027 1.3  1026 1.9  1027 3.2  1027 1.6  1026 6.5  1028 5.1  1028 1.8  1026 1.9  1027 1.8  1026 1.8  1025 1.2  1027 5.8  1028 n.d 1.4  1025 3.6  1027 1.6  1025 1.7  1027 7.5  1028 1.9  1025

0.03 0.25 0.11 n.d n.d n.d 0.04 0.13 0.04 0.18 0.22 0.21 n.d 0.04 0.22 0.15 0.01 0.27 n.d 0.28 n.d

0.049 0.056 0.080 0.079 0.019 0.018 0.033 0.031 n.d 0.043 0.031 0.029 0.058 0.053 0.049 0.042 0.053 0.056 0.038 0.058 0.047

0.08 n.d 0.26 0.05 0.11 n.d 0.02 0.04 0.09 n.d n.d n.d 0.16 n.d 0.02 0.01 n.d 0.14 0.12 0.18 n.d

2.1  1005 n.d 1.9  1005 1.0  1004 2.8  1005 n.d 2.9  1005 2.5  1005 1.9  1005 n.d n.d n.d 1.1  1004 n.d 3.1  1004 4.7  1003 n.d 1.6  1002 1.5  1002 2.2  1003 n.d

1.4  1021 n.d 7.5  1022 1.1  1020 1.4  1021 n.d 2.9  1021 1.7  1021 1.1  1021 n.d n.d n.d 9.1  1021 n.d 1.2  1019 1.1  1017 n.d 1.9  1017 1.7  1017 1.3  1018 n.d

6.2  1020 2.3  1020 n.d 1.6  1020 1.7  1019 3.6  1020 9.9  1020 1.1  1019 1.3  1019 8.7  1020 2.5  1020 9.7  1020 1.1  1019 3.2  1019 n.d 1.6  1020 2.6  1020 6.7  1021 3.5  1020 3.1  1020 2.8  1021

1.3  102 2.5  103 n.d 7.6  103 3.9  102 1.0  102 4.1  102 3.5  102 2.6  102 4.3  102 1.3  102 3.0  102 3.2  102 5.1  102 n.d 3.0  102 2.6  102 8.4  103 4.1  102 1.7  102 2.0  102

8.24 7.07 2.34 2.22 2.07 1.56 3.09 2.33 2.80 3.00 3.34 3.59 7.17 3.92 3.78 12.70 2.71 2.71 2.49 3.66 12.93

1.5  103 2.0  103 1.1  103 5.6  104 9.7  104 1.1  103 7.1  104 6.7  104 6.6  104 3.1  104 1.4  104 2.5  104 8.5  104 1.3  103 7.2  105 3.8  106 6.3  106 6.5  106 1.2  105 3.6  105 2.9  107

3.9  102 3.3  102 2.4  102 2.2  102 2.6  102 3.5  102 2.6  102 2.1  102 1.7  102 1.3  102 5.2  103 7.2  103 2.7  102 3.3  102 3.5  103 2.8  104 3.1  104 3.9  104 8.0  104 1.5  103 4.5  105

1.20 1.06 0.61 0.60 0.45 0.19 0.70 0.44 0.55 0.62 0.62 0.54 0.84 0.54 0.71 1.43 1.14 0.68 0.64 1.46 0.93

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

Sample

amsl = Above mean sea level. Brackets values and sign stand for oxidation state.

337

338

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

Fig. 2. Hydrogeological cross-section along the line A–B in Fig. 1.

section of the study area along the localized line A–B (Fig. 1) illustrating the terrain from upland towards lower areas (discharge point). The groundwater depth was not measured since the boreholes are tight sealed except during maintenance. Based on the geormophology of the study area (Fig. 1) and the conceptual model provided (Fig. 2), the groundwater recharge zones is the Chiradzulu Mountains (and Ndirande Mountains), even though direct rainfall injection may contribute significant recharge into the fractured aquifer.

Geosciences in Wuhan. d18O values were determined through equilibration with H2O–CO2 at 25 °C for 24 h followed by continuous introduction to the mass spectrometer using a Thermo Finnigan GasBench on-line gas preparation system. d2H was measured by reaction with Cr at 850 °C, using an automated Finnigan MAT H/Device (Thermo Scientific). d18O and d2H values were measured relative to internal standards calibrated using Vienna Standard Mean Ocean Water (V-SMOW). Isotopic composition (d18O and d2H) were reported in standard d notation representing per mil deviations from the V-SMOW standard (Eq. (1)), where

3. Materials and methods

d2 H ðor d18 OÞ ¼

Rsample  Rstandard  1000 Rstandard

ð1Þ

3.1. Sampling and analytical work 2

Groundwater samples were collected from 21 sampling points within Machinjiri area in Blantyre. Field work included immediate in situ measurements of pH (Wagtech pH meter, S/N 297046), dissolved oxygen (HANNA, Model HI9743), temperature, electric conductivity (EC) and total dissolved solids (TDS) (Wagtech EC/TDS/°C meter, S/N 303178) whereas bicarbonate (HCO3) was analyzed in the Department of Physics and Biochemical Sciences’ chemistry laboratory via ALPHA titration technique. Borehole site elevation and coordinates were done using a portable hand held global positioning system (GPS) meter. Subsequently, water samples for major and trace element analysis were filtered through a 0.45 lm millipore filter on site into 100 mL washed high density polyethylene (HDPE) bottles. 21 samples for cation (major and trace elements) analysis were acidified to pH < 2 using 6 M HNO3 and preserved immediately in a refrigerator. On the other hand 21 samples for anion analysis were not acidified but were immediately stored in a refrigerator as well prior to transportation to the laboratory in Wuhan, China. The laboratory analysis of all the samples was carried out at State Key Laboratory of Biogeology and Environmental Geology of China University of Geosciences (Wuhan). Major anions fluorides (F), chlorides (Cl), nitrates (NO3), nitrites (NO2), sulfates (SO4) and bromine (Br) from unacidified aliquots were analyzed using Ion Chromatography (DX-1100 -Dionex). Acidified samples were analyzed for trace elements by inductively coupled plasma optical emission spectrometry (ICP-OES) (ICAP6300). Trace elements analyzed were silver (Ag) arsenic (As), aluminum (Al), gold (Au), boron (B), barium (Ba), beryllium (Be), cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), lithium (Li), manganese (Mn), nickel (Ni), iron (Pb), selenium (Se), tin (Sn), strontium (Sr), vanadium (V) and zinc (Zn). Likewise, we analyzed all major water quality cations namely calcium (Ca), magnesium (Mg), potassium (K), silicate (Si) and sodium (Na). The average analytical error for major and trace chemical constituents using ICP-OES is less than ±5%. d18O and d2H values were measured using a Finnigan MAT 253 stable isotope ratio mass spectrometer at the State Key Laboratory of Biogeology and Environmental Geology, China University of

R¼1

H or H

18

O

16 O

Precisions for d2H and d18O were ±1.0‰ and ±0.1‰, respectively. Deuterium excess values (d-excess) were calculated by d = d2H  8  d18O in order to evaluate the changes in moisture sources. 3.2. Data analysis and intepretation In order to assess the significance of redox processes in the investigated samples, we used the USGS redox scheme for interpreting redox states of groundwater designed by McMahon and Chapelle (2008). Methods for inferring redox state from commonly measured chemical constituents, including concentrations of DO, provide a useful context for evaluating trace element occurrence in groundwater. Redox is a measure of the oxidation–reduction state of groundwater, which in natural systems can be measured by hierarchical progression of terminal electron acceptors for the reduction of compounds in groundwater (Chapelle et al., 2002; Paschke, 2007; McMahon and Chapelle, 2008; McMahon et al., 2009). The geochemical modeling program PHREEQC v2.8 (Parkhurst and Appelo, 1999), implemented with LLNL (Lawrence Livermore National Laboratory) database V8.R6230 was used in trace element speciation analysis and computation of carbon dioxide partial pressure. 4. Results and discussion 4.1. General physico-chemical characteristics In terms of physical properties, a discernible pH variation was observed amongst sampling points. The pH values straddled around pH 6.5 varying from 6.1 to 7.1. This indicates that the groundwater is mainly of slightly acidic nature consistent with studies in basement complex elsewhere (Wanda et al., 2011; Mkandawire, 2004). The partial pressure of carbon dioxide

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

(pCO2) calculated with PHREEQC program (Parkhurst and Appelo, 1999) executed using LLNL database, ranged from 0.005 to 0.05 atm. The modeled values are higher than the atmospheric pCO2 (103.5 atm) suggesting that groundwaters have gained most of its CO2 from root respiration and decay of organic matter (Hamed and Dhahri, 2013). Subsequently, the increase in pCO2 causes a drop in pH as indicated in Fig. 3. EC and TDS varied from borehole to borehole indicating evidence of considerable differences in mineralization influenced by various factors. The groundwater major hydrochemical element loads are illustrated in Fig. 4. Most of the samples were typical Mg (Ca)–Na–HCO3 type hinting that carbonate weathering (e.g. aragonite, dolomite, calcite and siderite) is one of the main contributors to the observed water types (Mapoma et al., submitted for publication-a). Evidence of considerable amounts of NO3 and Cl anions is apparent while SO4 dominated water is found in two boreholes (BH16 and 21). High TDS water exhibited high concentrations of SO4, Fe and NO3. Evaporite dissolution such as anhydrite (CaSO4), halite (NaCl) and gypsum (CaSO4:2H2O) is suggested as one of the main sources of Cl, SO4 and Fe (Mapoma et al., submitted for publication-a). This is evident by the strong Pearson’s correlations (IBMÒ SPSSÒ version 21) between TDS and Ca (r = 0.95; p < 0.001), Mg (r = 0.63; p = 0.003), K (r = 0.95; p < 0.001), Na (r = 0.77; p < 0.001) and SO4 (r = 0.87; p < 0.001). Moreover, the apparent relationship between TDS and major elements is evidence of Ca, Mg, Na, SO4, NO3, HCO3 being the primary contributors to salinity (Fig. 4). The mild correlation between TDS and Mg suggests otherwise a cation exchange as another plausible source of Mg (Eq. (2)).

MgðCaÞ-X þ 2Naþ ! Na-Clay þ Mg2þ ðCa2þ Þ

ð2Þ

As alluded to above carbonate weathering is another source of Mg concentration in groundwater. Moreover, there was no significant correlation between TDS and F (r = 0.38; p = 0.092) and TDS against NO3 (r = 0.36; p = 0.106) suggesting other sources of F and NO3 apart from aquifer material. The predominance of NO3 over other anions in some boreholes (Fig. 4) is an indication of input of NO3 bearing fertilizers. Such observations are also true for borehole samples that exhibited high concentrations of SO4 and Cl. Over reliance on fertilizers for domestic crop production in rural areas has increased the probabilities of enhanced contributions of SO4 and NO3 in groundwater through the effect of evapotranspiration and leaching processes. Such anthropogenic

Fig. 3. Evolution of carbon dioxide activity modeled using PHREEQC with respect to field pH.

339

Fig. 4. Variations of major chemical constituents of groundwater investigated in Machinjiri area, Balntyre district.

activities would keep the groundwater quality with respect to SO4 and NO3 higher than the expected values though not exceeding benchmark values, except in few cases. The relatively shallow boreholes are expected to have oxic conditions that would promote oxidation of NO2 to NO3 and S to SO4. Thus, the absence of NO2 in the borehole samples is related to interaction of aquifer with atmospheric oxygen consequently depleting NO2 concentration through oxidation reactions. However, oxidation of pyrite maybe involved in controlling SO4 concentration besides anthropogenic input and evaporite dissolution as illustrated in the reaction mechanism (Fig. 5). The poor relationship between Ca and F (r = 0.409; p = 0.066), and the oversaturation of fluorite (CaF2) is evidence of low contribution of fluorite dissolution with the index suggesting a more precipitation control mechanism. As such, the very little fluoride dissolution into the groundwater is being controlled by various factors within the aquifer system. Mostly, fluorite is the main control of F mineralization in groundwater. However, there are few cases of samples that had evidence of F concentrations above WHO benchmark but still being below the Malawian Standards. With respect to Cl, the poor relationship between Na and Cl (r = 0.49, p = 0.024) and oversaturation with respect to halite is an explanation of possible cation exchange and supporting the possibility of other sources of Na other than halite such as albite and mirabilite. Therefore, dissolution of evaporites, carbonate and silicate weathering simultaneously contribute to SO4, Cl and HCO3 concentrations while anthropogenic controls inject additional NO3 and F into the system with simultaneous addition of SO4

Fig. 5. Pyrite reaction mechanism presumed to be one of the processes involved in sulfate concentration.

340

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

Fig. 6. Summary of trace elements analyzed in the 21 boreholes in machinjiri area, Blantyre district.

and possible Cl from fertilizer residues. Chemical transport from surface through rainwater into the subsurface is likely in the area with respect to time. Fig. 6 summarizes the trace element constituents of groundwater sampled in Machinjiri area (data from Mapoma et al., submitted for publication-b). From Fig. 6, it can be observed that groundwater in the area has a consistent concentration of Au and B yet with high variations in Fe and Zn concentrations from borehole to borehole. Furthermore, PHREEQC speciation identified various main species for each trace element analyzed based on element constrains analyzed in our investigation of the groundwater samples. These results are summarized in Table 1. Presented in the table are the possible oxidation states and accompanied molar concentrations for the trace elements analyzed. 4.2. Factors controlling trace element geochemistry Time elapsed since recharge significantly influences chemical evolution of trace elements in groundwater. Age of water and its flow dynamics within the aquifer ensures mineralization of groundwater through contact with underlying aquifer materials. However, ‘‘measurement of the time water is in an aquifer is complicated by the potential mixing of waters of different ages’’ (Ayotte et al., 2011). Other factors that assist scientists in understanding the geochemical controls on trace element concentrations despite the degree of chemical evolution of the water are redox status and pH. Redox conditions for the water in each well in this analysis were determined from a scheme that evaluates concentrations of DO, NO3, Mn, Fe and SO4 (Paschke, 2007; McMahon and Chapelle, 2008). The results from the scheme identified two groups of water samples (1) groundwater that was O2 or nitrate-reducing was considered oxic and (2) groundwater that met more than one redox process criterion was considered mixed. Mixed redox status in this case implied a mixture of both oxic and anoxic conditions depending on process control for which one or the other may predominate. Evidence of anoxic conditions was found in five out of twentyone boreholes investigated using the USGS scheme. Of these five boreholes three (BH01, 12 and 18) had mixed (oxic–anoxic) O2– Mn(IV) redox processes whose electron acceptor (reduction) half reactions are (Eqs. (3) and (4)):

O2 þ 4Hþ þ 4e ! 2H2 O

ð3Þ

MnO2 ðsÞ þ 4Hþ þ 2e ! Mn2þ þ 2H2 O

ð4Þ

The other two samples (B16 and 21) with mixed (oxic–anoxic) conditions are controlled by O2–Fe(III)/SO4 as electron acceptors (Eqs. (5)–(7)):

FeðOHÞ3 ðsÞ þ Hþ þ e ! Fe2þ þ H2 O

ð5Þ

FeOOHðsÞ þ 3Hþ þ e ! Fe2þ þ 2H2 O

ð6Þ

 þ  SO2 4 þ 9H þ 8e ! HS þ 4H2 O

ð7Þ

In general, the concentration of oxygen in these boreholes ensured oxic conditions to prevail, an indication of close interaction with the atmospheric direct oxygen dissolution. The other factor, pH, can affect the solubility and mobility of trace elements, often through sorption/desorption processes (Welch and Stollenwerk, 2003; Paschke, 2007). Metals such as Cu, Fe and Mn solubility in water often decreases with increasing pH while As solubility can increase linearly with pH over the normal pH range of most natural waters. As mentioned earlier, the pH range for the studied groundwater samples was 6.1–7.1. 4.3. Relationship of trace elements to pH and redox state The occurrence of trace elements in groundwater was variable. The variations in concentrations may reflect several processes, including differences in precipitation, the effects of land use, ion competition, complexation, redox and evaporative concentrations. The effect of redox state and pH accounts for substantial additional variability in concentrations of trace elements in groundwater within an aquifer. Other factors such as water type, TDS, organic matter and biological activity also may influence trace element concentrations and may account for unexplained variability in the expected redox- and pH-related behaviors of some trace elements. Most trace elements are sensitive to redox conditions in groundwater, either because that elements can occupy multiple redox states in natural waters, or because the redox state controls the concentrations of soluble complexes. Variation in pH has an effect on the adsorption of many trace elements, as well as, on the solubility of some concentration-limited hydroxide solid phases (such as Fe and Mn). Some redox sensitive trace elements are more mobile under low pH and oxic conditions. For example Cd, Cu, Pb and Zn can increase from 0% to 100% adsorption on oxide–hydroxides over a pH range of 2 standard units (Salbu and Steinnes, 1994). Other elements such as Fe and Mn are mobile under low pH and (or) anoxic conditions. Some oxyanion-forming trace elements, by contrast are generally more mobile under high pH conditions and include As, Cr, Se and V. Arsenic sorption for example, decreases as pH increases (Smedley and Kinniburgh, 2002; Stollenwerk, 2003). Most samples were oxic (Table 2) according to the scheme that uses commonly measured ions and properties of water samples to assess the redox state (Chapelle et al., 2002; Paschke, 2007; McMahon and Chapelle, 2008; McMahon et al., 2009). 4.3.1. Iron (Fe) and manganese (Mn) Based on the redox classification scheme used in this study, samples with high Fe or Mn (concentration > 100 lg/L or 50 lg/L, respectively) were considered to be anoxic (Table 2). Therefore, high Fe and Mn occurred most often in mixed (oxic–anoxic) samples and relatively low concentrations observed in oxic samples consistent with studies elsewhere (Ayotte et al., 2011). However, BH1 was an exception to the aforementioned observation in that the concentration of Mn was anomalously high (797.2 lg/L) compared to other samples despite a higher dissolved oxygen content (4.82 mg/L). However, there was no remarkable linear variation of Mn (Mn2+ and Mn3+) with respect to the field pH range in this study. Nevertheless, higher concentrations of Fe2+ were also observed above neutral pH (Fig. 7a). It is known that for many metals including

341

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345 Table 2 Results of USGS scheme used to assess redox state of natural water (McMahon and Chapelle (2008)). Sample ID

Dissolved O2 (mg/L)

NO 3 N (mg/L)

Mn2+ (lg/L)

Fe2+ (lg/L)

SO2 4 (mg/L)

Redox category

BH01 BH02 BH03 BH04 BH05 BH06 BH07 BH08 BH09 BH10 BH11 BH12 BH13 BH14 BH15 BH16 BH17 BH18 BH19 BH20 BH21

4.82 3.42 0.87 1.10 1.32 1.29 1.82 1.47 2.64 1.17 1.24 0.77 2.25 1.92 1.96 0.90 1.22 1.18 2.47 2.12 1.22

166.09 116.85 15.70 31.08 6.78 7.96 83.76 15.75 18.08 7.74 22.10 16.11 49.33 33.49 10.85 0.00 18.51 50.37 33.42 39.17 0.00

797.20 1.00 9.50 0.60 1.40 4.30 0.20 0.20 10.80 0.70 10.00 117.20 0.50 0.30 0.00 69.90 3.90 110.30 0.80 0.60 89.90

4.60 38.90 8.20 31.00 26.70 15.50 16.60 7.00 7.60 8.00 5.80 9.70 6.70 18.60 7.80 1080.00 14.50 10.30 6.60 24.20 1096.00

21.08 15.18 7.04 7.99 7.95 4.80 9.39 5.85 8.50 9.57 8.28 9.27 8.02 5.47 5.53 210.75 9.50 25.91 19.12 9.30 207.65

Mixed(oxic–anoxic) Oxic Oxic Oxic Oxic Oxic Oxic Oxic Oxic Oxic Oxic Mixed(oxic–anoxic) Oxic Oxic Oxic Mixed(oxic–anoxic) Oxic Mixed(oxic–anoxic) Oxic Oxic Mixed(oxic–anoxic)

Fe, solubility decreases with increasing pH (Ayotte et al., 2011). As such, our study results can be explained by other factors such as differences in mineral content of the aquifer and redox conditions. 4.3.2. Copper (Cu+ and Cu2+) and lead (Pb2+) Copper and lead are common elements and products used in drinking water well construction and plumbing. It is therefore difficult to determine the extent to which the concentrations of these elements in water samples from drinking water wells are related to these materials or to other sources. Studies on redox in relation to the two elements show that Cu was on average relatively high under oxic conditions. However, our study could not prove that Cu was sensitive to field pH (Table 2). This observation is similar to studies elsewhere (Ayotte et al., 2011). Mostly, Pb2+ was higher in oxic conditions than mixed conditions. Since the scheme did not exclusively identify completely anoxic conditions, it implies that whenever dominant anoxic conditions exist, Pb2+ would definitely be lower than the observed values in oxic conditions (Ayotte et al., 2011). But, based on solubility considerations, Pb is more mobile under low-pH conditions (Hem, 1985). Fig. 7b illustrates the sensitivity of Pb2+ to increase in pH from our investigated groundwater samples. Within the field pH of our investigated groundwater samples, Pb2+ shows a regressive behavior with decrease in pH. According to some studies, Pb is more mobile under low pH (as mentioned in this paragraph). It is therefore assumed that the regressive behavior of Pb mobility in this aquifer is mixed factor controlled. 4.3.3. Boron (B), chromium (Cr) and vanadium (V) On average, relatively high concentrations of B3+ were observed in oxic conditions compared to mixed conditions. This shows that boron is redox sensitive (Ayotte et al., 2011). But, no tangible observation can be said for Cr with respect to redox conditions. However, species of vanadium showed sensitivity to redox conditions (Table 2). These three elements commonly form oxyanions in groundwater that are affected by redox and pH (Ayotte et al., 2011). The relationship between B3+ and pH (Fig. 7c) indicate that the element decreased with increase in field pH suggesting low mobility of boron under high pH conditions in the studied aquifer. Nonetheless, sensitivity of Cr species to the field pH was not compellingly momentous compared to that of V species (Fig. 7d and e) where

V3+ and V4+ decreased sharply beyond pH 6.5 while V5+ increased with increasing pH beyond 6.5. 4.3.4. Barium (Ba), lithium (Li) and strontium (Sr) Generally, low concentrations of Ba were observed under mixed conditions. However, the results were not very convincing. Equally, Li did not show significant variations with respect to field pH. Nonetheless, Sr exhibited high concentrations in groundwater with mixed redox conditions with very high Sr concentration in the groundwater group controlled by O2–Fe(III)/SO4 redox process. With respect to pH, there was no discernible control on Ba, Li and Sr at field pH range measured in our study, except for high Sr observed for two boreholes at pH greater than neutral (Fig. 7f). 4.3.5. The rest of the trace elements With the exception of few cases (Table 2), higher concentrations of Au+ were observed in completely oxic conditions compared to mixed conditions. The opposite was observed for Co2+. However, BH04, 13 and 19 showed higher concentrations of Co within the oxic conditions compared to mixed conditions, making it difficult to conclude on sensitivity of Co to redox in the studied samples. It should also be noted that nickel was not detected in afore indicated boreholes where Co was relatively higher (Mapoma et al., submitted for publication-b) and no observed sensitivity to redox and pH either. Sn concentrations did not vary with respect to redox conditions and pH. Similar non-sensitive observations where noted for Se and Zn however, the narrow range in pH may bias the conclusion on pH-sensitivity of the mentioned trace elements. Despite that, the field pH of the studied aquifer contributes to variations in trace element species and mineralization of the groundwater just as redox processes do. 4.4. Stable isotope investigation The environmental isotopic ratio of oxygen (d18O) and hydrogen (d2H) are excellent tracers for determining the origin of groundwater. They are widely used for studying aquifer recharge and mixing of waters from different sources (Subyani, 2004). The hydrogen and oxygen isotopic composition of the investigated boreholes are presented in Table 3. Our study is the first to report on the isotopic composition of fractured basement complex in Blantyre district.

342

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

Fig. 7. Variations of PHREEQC modeled main trace element species (a) iron, (b) lead, (c) boron, (d, e) vanadium and (f) strontium that showed some observed trends with respect to field pH. The units for the species are mmol/L.

The isotopic composition of the sampled groundwater varied between 5.89‰ and 3.32‰ for d18O and 36.98‰ and 20.42‰ for d2H (Table 3). This variation can be the result of seasonal rainfall differences caused by local climatic factors and altitude effects. Oxygen and hydrogen isotopic analysis for the investigated groundwater are plotted in Fig. 8. The so called global meteoric water line (GMWL) is also included in the diagram. The GMWL is defined by the average relationship between hydrogen and oxygen isotope ratios in natural terrestrial waters based on

precipitation data from locations around the globe with an R2 > 0.95 (Craig, 1961). It has a linear relationship expressed by d2H = 8 d18O + 10. The d2H and d18O diagrams usually provide interpretive information on the origin of groundwater (Gat and Gonfiantini, 1981). Moreover, the local meteoric water line (LMWL) is referenced in the diagram (Fig. 8), which is the line derived from precipitation collected from a single site or set of ‘‘local’’ sites. The LMWL can be significantly different from the GMWL displaying generally slopes in the range of 5 and 9 (USGS

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345 Table 3 Isotopic data of groundwater from Machinjiri area. Sample ID

d18O

d2H

d-Excess

BH01 BH02 BH03 BH04 BH05 BH06 BH07 BH08 BH09 BH10 BH11 BH12 BH13 BH14 BH15 BH16 BH17 BH18 BH19 BH20 BH21 Average STDEV

5.43 3.68 3.32 4.74 5.25 5.32 5.12 5.49 5.14 5.55 5.89 5.66 5.68 5.81 5.56 5.75 4.88 4.94 4.92 4.94 5.87 5.19 0.66

33.2 20.4 20.6 29.8 31.6 33.1 32.0 34.8 32.2 35.1 36.6 35.2 35.8 37.0 34.8 36.5 31.6 31.0 29.5 31.3 34.8 32.2 4.5

10.28 9.02 5.99 8.09 10.40 9.46 8.94 9.14 8.94 9.34 10.54 10.06 9.66 9.50 9.65 9.47 7.47 8.57 9.83 8.18 12.12 9.27 1.24

the effect of processes such as evaporation within the unsaturated zone during recharge or mixing with an evaporated source. However, a significant rapid infiltration before evaporation takes place is suggested since the study area is underlain by fractured basement of basaltic gneiss lithology. In fractured aquifer system, direct infiltration through preferential channels is likely to preserve the isotopic composition of original rain which minimizes isotopic fractionation processes. Evaporation and exchange with rock minerals are common fractionation processes that affect the relationship between d2H and d18O (Fontes et al., 1980). As mentioned earlier in this discussion, the groundwater appears to be largely meteoric derived. It therefore means that, rock water interaction have not been sufficiently strong enough to result in significant shift from the LMWL and the GMWL. The samples plotting to the right of the GMWL suggest that they have undergone some evaporation before or during their transit into the aquifer below. The shifting apart indicate the action of combined local processes such as selective infiltration, direct percolation through preferential channels that do not change the isotopic composition of the original rainwater, some degree of fractionation in the vadose zone and most probably surface water contribution and mixing mechanisms by anthropogenic activities especially groundwater drawing. Nevertheless, evaporation increases the concentration of dissolved salts in water. A relationship between evaporation and dissolution of evaporates can be illustrated by a d18O vs TDS plot (Fig. 9a). In our diagram of d18O vs TDS, there is no discernible relationship between dissolved solids and variation in isotopic composition. This implies that there is very limited impact of

Fig. 8. d2H vs d18O plot of groundwater samples from Machinjiri area in Blantyre compared to GMWL (Craig, 1961) and LMWL (Monjerezi et al., 2011).

(2004) http://www.rcamnl.wr.usgs.gov/isoig/period/h_iig.html). Since data values of d2H and d18O for local precipitation are not documented, a generalized local meteoric water line is used in this case (Monjerezi et al., 2011), defined by d2H = 7.75 d18O + 9.68. This line runs slightly apart from the GMWL as can be observed through the slope (7.75) and the intercept (9.68) compared to that of the GMWL, respectively. Most of the groundwater samples plotted on or close to the GMWL, producing a regression line defined by d2H = 6.65 d18O + 2.24 suggesting a meteoric origin with a mean isotopic composition of borehole water being d2H = 32.23 and d18O = 5.19‰. This approximates the observations made elsewhere (Boutaleb et al., 2000; Monjerezi et al., 2011). The straddling of isotopic composition close to the GMWL and LMWL demonstrates a precipitation origin of groundwater in the studied area. Within the dataset we observed that two water samples (BH02 and 03) were relatively enriched in d2H and d18O compared to the rest of the samples. This may be attributed to evaporation (Monjerezi et al., 2011). Some samples had their isotopic composition shifted apart from the GMWL and LMWL and plotted to the right indicating

343

Fig. 9. Relationship between d18O with TDS (a) and NO3 (b).

344

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345

evaporation on the salinization of groundwater in the study area. We may reasonably assume that the observed deviation from the LMWL is not mostly explained by evaporative enrichment of the stable isotopes prior to groundwater recharge. Mostly, the results suggest that the isotopic variation may arise from mixing of groundwater along the flow paths. Irrigation return flow may modify d18O as water infiltrates and percolates. Nonetheless, NO3 increase in groundwater does not show significant relationship with changes in d18O (Fig. 9b). As such return flow in the study area is not significant. Moreover, irrigation though encouraged is not significantly practiced in the area as farmers practice seasonal farming. The d-excess is a useful parameter for measuring non-equilibrium conditions during source water evaporation. It is dependent on relative humidity and sea surface temperature in the moisture source region (Breitenbach et al., 2010). In tropical regions temperature has a rather weak influence on stable isotopes in precipitation (Aggarwal et al., 2004). But stable isotopes are controlled by amount of precipitation and the degree of air mass rainout during moisture transport. Moreover, the d-excess of the continental atmospheric vapor mass may be altered by contribution from secondary evaporative processes (Marfia et al., 2004). Our study was a one-time sampling of groundwater and analysis of isotopes which explains the absence of d18O vs precipitation analysis. This evaluation has to wait for other time when resources are available. The d-excess values shown in Table 3 approximate the y-intercept of the GMWL of 10. As such the influence of secondary evaporative processes is minimal (Marfia et al., 2004). 5. Conclusions Our study, which is the first of its kind in the area, indicates that redox processes and pH are significant factors in trace element distribution. In some instances, the significance of these two processes was observed especially for iron, manganese, lead and vanadium. Depending on species present, pH increase may mobilize or restrain the element. According to literature some elements such as iron, manganese and vanadium are redox sensitive, which has been highlighted in our study. It therefore shows that besides presence and activity of mineral phases such as evaporites, carbonates and silicates, mineralization in the area is pH and redox dependent for some trace elements. Furthermore, d18O and d2H isotopic investigation revealed that groundwater in this area is meteoric in origin and that vertical recharge and mixing are the main process as observed from the d18O vs TDS plot. Moreover, the relationship between NO3 and d18O shows that impact of irrigation return flow on groundwater stable isotope modification is not significant. Also, influence of secondary evaporative processes on isotopic signature of the study area is minimal as shown by the values of d-excess (M = 9.269, SD = 1.240). Thus, results suggest that groundwater chemistry in the studied aquifer is influenced by inherent processes with contribution from human activities and mostly the water originates from rainwater influx. With such results, further studies are recommended to further delineate the processes involved in mineralization incorporating strontium, carbon and sulfur isotopes. Acknowledgements This study had the full support from China University of Geosciences (CUG) (Wuhan) and University of Malawi, The Polytechnic. The full support from the technical staff of Physics and Biochemical Sciences and Department of Water (Blantyre) disserves credit. We are beholden to expert reviewers for their input.

References Aggarwal, P.K., Fröhlich, K., Kulkarni, K.M., Gourcy, L.L., 2004. Stable isotope evidence for moisture sources in the Asian summer monsoon under present and past climate regimes. Geophys. Res. Lett. 31, L08203. Amer, R., Ripperdan, R., Wang, T., Encarnación, J., 2012. Groundwater quality and management in arid and semi-arid regions: Case study, Central Eastern Desert of Egypt. J. Afr. Earth Sci. 69, 13–25. Ako, A.A., Shimada, J., Hosono, T., Ichiyanagi, K., Nkeng, G.E., Eyong, G.E.T., Roger, N.N., 2012. Hydrogeochemical and isotopic characteristics of groundwater in Mbanga, Njombe and Penja (Banana Plain)–Cameroon. J. Afr. Earth Sci. 75, 25– 36. Ayotte, J.D., Gronberg, J.A.M., Apodaca, L.E., 2011. Trace Elements and Radon in Groundwater Across the United States, 1992–2003. USGS Scientific Investigation Report 2011-5059, pp 115. Boutaleb, S., Bouchaou, L., Mudry, J., Hsissou, Y., Mania, J., Chauve, P., 2000. Hydrogeologic effect on the quality of water in the Oued Issen watershed, western Upper Atlas Mountains, Morocco. Hydrogeol. J. 8, 230–336. Bradley, P.M., 2003. History and ecology of chloroethene biodegradation: a review. Bioremediat. J. 7, 81–109. Breitenbach, S.F.M., Adkins, J.F., Meyer, J.F., Marwan, N., Krishna Kumar, K., Haug, G.H., 2010. Strong influence of water vapour source dynamics on stable isotopes in precipitation observed in southern Meghalaya, NE India. Earth Planet. Sci. Lett. 292, 212–220. Carter, G.S., Bennett, J.D., 1973. The geology and mineral resources of Malawi. Bull. Geol. Surv. Malawi. 6. Chapelle, F.H., Lovley, D.R., 1992. Competitive exclusion of sulfate reduction by Fe(III)-reducing bacteria: a mechanism for producing discrete zones of highiron groundwater. Ground Water 30, 12–23. Chapelle, F.H., Bradley, P.M., Lovley, D.R., O’Neill, K., Landmeyer, J.E., 2002. Rapid evolution of redox processes in a petroleum hydrocarbon-contaminated aquifer. Ground Water 40, 353–360. Chilton, P.J., Smith-Carington, A.K., 1984. Hydrogeological characterization and water-supply potential of basement aquifers in tropical Africa. Hydrogeol. J. 3, 36–49. Christensen, T.H., Bjerg, P.L., Banwart, S.A., Jakobsen, R., Heron, G., Albrechtsen, H.-J., 2000. Characterization of redox conditions in groundwater contaminant plumes. J. Contam. Hydrol. 5, 165–241. Craig, H., 1961. Isotope variation in meteoric waters. Science 133, 1702–1703. Dill, H.G., Ludwig, R.R., Kathewera, A., Mwenelupembe, J., 2005. A lithofacies terrain model for the Blantyre region: implications for the interpretation of palaeosavanna depositional systems and for environmental geology and economic geology in southern Malawi. J. Afr. Earth Sci. 41, 341–393. Dindane, K., Bouchaou, L., Hsissou, Y., Krimissa, M., 2003. Hydrochemical and isotopic characteristics of groundwater in the Souss Upstream Basin, southwestern Morocco. J. Afr. Earth Sc. 36, 315–327. Fass, T., Cook, P.G., Stieglitz, T., Herczeg, A.L., 2007. Development of saline ground water through transpiration of sea water. Ground Water 45, 703–710. Fontes, J.C., Pouchon, P., Saliege, J.F., Zuppi, G.M., 1980. Environmental Isotope study of Groundwater Systems in the Republic of Djibouti. Arid-zone Hydrology: Investigations with Isotope Techniques, 237–262. Gat, J.R., Gonfiantini, R., 1981. Stable Isotope Hydrology. Deuterium and Oxygen-18 in the Water Cycle. International Atomic Energy Agency. Vienna, Technical Reports Series, (210). Ghabayen, S., McKee, M., Kemblowski, M., 2006. Ionic and isotopic ratios for identification of salinity sources and missing data in the Gaza aquifer. J. Hydrol. 318, 360–373. Hem, J.D., 1985. Study and Interpretation of the Chemical Characteristics of Natural Water, vol. 2254. Department of the Interior, US Geological Survey. Hamed, Y., Dhahri, F., 2013. Hydro-geochemical and isotopic composition of groundwater, with emphasis on sources of salinity, in the aquifer system in Northwestern Tunisia. J. Afr. Earth Sci. 83, 10–24. Kanyerere, T., Levy, J., Xu, Y., Saka, J., 2012. Assessment of microbial contamination of groundwater in upper Limphasa River catchment, located in a rural area of northern Malawi. Water SA 38, 581–596. Mapoma, H.W.T., Xie, X., 2014. Basement and alluvial aquifers of Malawi: an overview of groundwater quality and policies. Afr. J. Environ. Sci. Technol. 8, 190–202. Mapoma, H.W.T., Xie, X., Zhang, L., Nyirenda, M.T., submitted for publication-a. Hydrochemical Characteristics of Rural Community Groundwater Supply in Blantyre, Malawi. Mapoma, H.W.T., Xie, X., Nyirenda, M.T., Zhang, L., submitted for publication-b. Trace Element Geochemistry of Fractured Basement Aquifer of Southern Malawi: A Case of Machinjiri Area in Blantyre. Marfia, A.M., Krishnamurthy, R.V., Atekwana, E.A., Panton, W.F., 2004. Isotopic and geochemical evolution of ground and surface waters in a karst dominated geological setting: a case study from Belize, Central America. Appl. Geochem. 19 (6), 937–946. McMahon, P.B., Chapelle, F.H., 2008. Redox processes and water quality of selected principal aquifer systems. Ground Water 46, 259–271. McMahon, P.B., Cowdery, T.K., Chapelle, F.H., Jurgens, B.C., 2009. Redox Conditions in Selected Principal Aquifers of the United States: U.S. Geological Survey Fact Sheet 2009–3041, pp 6.

H.W.T. Mapoma et al. / Journal of African Earth Sciences 100 (2014) 335–345 Mkandawire, P.P., 2004. Groundwater resources of Malawi. In: Appelgren, B., (Ed.), Managing Shared Aquifer Resources in Africa. IHP-VI, Series on Groundwater No. 8. UNESCO, France. Mkandawire, T., 2008. Quality of groundwater from shallow wells of selected villages in Blantyre District, Malawi. Phys. Chem. Earth 33, 807–811. Monjerezi, M., Ngongondo, C., 2012. Quality of groundwater resources in Chikhwawa, lower Shire valley. Malawi. Water Qual. Exp. Health 4, 39–53. Monjerezi, M., Vogt, R.D., Aagaard, P., Gebru, A.G., Saka, J.D., 2011. Using 87Sr/86Sr, d18O and d2H isotopes along with major chemical composition to assess groundwater salinization in lower Shire valley, Malawi. Appl. Geochem. 26, 2201–2214. Msilimba, G., Wanda, E.M.M., 2013. Microbial and geochemical quality of shallow well water in high-density areas in Mzuzu City in Malawi. Phys. Chem. Earth, Parts A/B/C 66, 173–180. Msonda, K.W.M., Masamba, W.R.L., Fabiano, E., 2007. A study of fluoride groundwater occurrence in Nathenje Lilongwe Malawi. Phys. Chem. Earth 32, 1178–1184. Parkhurst, D.L., Appelo, C.A.J., 1999. User’s Guide to PHREEQC (Version 2): A Computer Program for Speciation, Batch-reaction, One-dimensional Transport, and Inverse Geochemical Calculations. Paschke, S.S., 2007. Hydrogeologic Settings and Ground-water Flow Simulations for Regional Studies of the Transport of Anthropogenic and Natural Contaminants to Public-supply Wells-Studies begun in 2001: U.S. Geological Survey Professional Papers 1737-A, pp 244.

345

Salbu, B., Steinnes, E., 1994. Trace Elements in Natural Waters. CRC Press, pp 302. Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source, behaviour and distribution of arsenic in natural waters. Appl. Geochem. 17, 517–568. Stollenwerk, K.G., 2003. Geochemical processes controlling transport of arsenic in groundwater – a review of adsorption. In: Welch, A.H., Stollenwerk, K.G., (Eds.), Arsenic in Ground Water: Geochemistry and Occurrence. Kluwer Academic Publishers, Boston, pp. 67–100. Subyani, A.M., 2004. Use of chloride-mass balance and environmental isotopes for evaluation of groundwater recharge in the alluvial aquifer, Wadi Tharad, western Saudi Arabia. Environ. Geol. 46, 741–749. USGS, 2004. (accessed 11.12.14). Wanda, E.M.M., Gulula, L.C., Phiri, A., 2013. Hydrochemical assessment of groundwater used for irrigation in Rumphi and Karonga districts, Northern Malawi. Phys. Chem. Earth, Parts A/B/C 6, 51–59. Wanda, E., Monjerezi, M., Mwatseteza, J.F., Kazembe, L.N., 2011. Hydro-geochemical appraisal of groundwater quality from weathered basement aquifers in Northern Malawi. Phys. Chem. Earth. 36, 1197–1207. Welch, A.H., Stollenwerk, K.G. (Eds.), 2003. Arsenic in Ground Water: Geochemistry and Occurrence. Springer. Xie, X., Wang, Y., Li, J., Su, C., Duan, M., 2013. Hydrogeochemical and Isotopic Investigations on Groundwater Salinization in the Datong Basin, Northern China1.