Journal of Geochemical Exploration xxx (2014) xxx–xxx

GEXPLO-05420; No of Pages 8

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

j ourna l homepage: www.e lsev ie r .com/ locate / jgeoexp

Distribution of water-soluble inorganic ions in the soils of Cyprus

Andreas M. Zissimos a,⁎, Irene C. Christoforou a, Eleni Morisseau a, David R. Cohen b, Neil F. Rutherford b

a Geological Survey Department of Cyprus, Ministry of Agriculture, Natural Resources and Environment, Lefkonos 1, Lefkosia 1415, Cyprusb School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, NSW 2052, Australia

⁎ Corresponding author. Tel.: +357 22409200.E-mail address: azissimos@gsd.moa.gov.cy (A.M. Zissi

http://dx.doi.org/10.1016/j.gexplo.2014.07.0040375-6742/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Zissimos, A.M., ethttp://dx.doi.org/10.1016/j.gexplo.2014.07.0

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 January 2014Accepted 5 July 2014Available online xxxx

Keywords:ChromatographySoilGeochemistryFractals

The water-soluble anion content of soils reflects variable input of geogenic and anthropogenic influences, and isan important indicator of soil fertility, the existence or risk of salinization and effects of other pollutants. A rapidion chromatography method has been used to determine the concentration of water-soluble F−, Cl−, NO3

− andSO4

2− on a suite of 3771 top soil samples collected as part of the geochemical atlas of Cyprus project. Excludingareas with salt flats and beach deposits, the average soil concentrations in Cyprus are 193 mg/kg for Cl−,17 mg/kg for F−, 67 mg/kg for NO3

− and 866 mg/kg for SO42−. Parent lithology (especially sedimentary environ-

ment) and the effects of seawater are the dominant controls on F− and Cl−, whereas SO42− andNO3

− display amixof geogenic and anthropogenic influences (including the effects of agriculture and mining of the basalt-hostedsulphide deposits). They display distinct fractal population characteristics that can be related to different sourcesof anions.

mos).

al., Distribution of water-soluble inorganic ion04

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Analysis of the water-soluble inorganic ion concentrations in soils isimportant in both environmental management and the assessment ofthe suitability of areas for various agricultural activities (Jackson,2000). Some of the more critical threats to soils, such as salinisationand desertification, are closely linked to increased concentrations of in-organic ions and related salts (Soil Atlas of Europe, 2005). Althoughsalinisation primarily originates from the weathering of minerals orinteraction of groundwater with palaeo-salt deposits, there can also beanthropogenic sources such as irrigation, municipal runoff and fertiliserapplication (Bloem et al., 2008).

Geochemical atlases, such as the GEMAS mapping of agriculturalsoils in Europe (De Vivo et al., 2013), provide a reference to measurespatial changes in the distribution of chemical species in the naturalenvironment, assist in determining factors that control regional geo-chemical variations, and provide a basis for setting environmental tar-gets and policies in different areas (Barth et al., 2009; Darnley et al.,1995; Reimann and Garrett, 2005; Salminen et al., 2005). The highsampling density soil geochemical atlas of Cyprus (Cohen et al., 2011)provides an opportunity to examine the effects of variation in geology,land use and other environmental factors on inorganic ions at theregional scale.

The island of Cyprus lies in the north-easternMediterranean Sea andcovers an area of 9251 km2. The geology of Cyprus (Fig. 1)may be divid-ed into a number of distinct terranes, including the mafic–ultramaficTroodos ophiolite complex (TOC) in the centre of the island and the

carbonate-dominated circum-Troodos sedimentary sequence (CTSS).There are a series of Cyprus-type sulphide deposits in the basalts withextensive historical mining activity. The soil characteristics of Cyprusreflect the terrane. On the ridges and peaks of Troodos, up to 1 m ofFe-rich, organic-rich A and residual B horizon soils overlie a C horizon.The carbonate-rich CTSS units that flank Troodos producemassive chalkysoils (calcisols) on hill slopes and terra rossa (chromic luvisols) over ex-posed limestones in areas with flat topography (Hadjiparaskevas,2005). In coastal areas there are thick colluvium and alluvium depositsand saline lake deposits on the Akrotiri Peninsula and near Larnaca.Cyprus has a typical Mediterranean climate, with hot and dry sum-mers from mid-May to mid-September and rainy winters from mid-November to mid-March.

Land use in Cyprus can be subdivided into the following majorclassifications:

(i) Undeveloped forests that cover much of central and north-western Troodos Massif,

(ii) Scrubland (garique–maquise) and degraded land that occupiesmuch of eastern Troodos and the southern Troodos flanks as faras the coast (excluding parts of the Polis Valley and coastal fringearound Pafos and Lemesos) with some valleys heavily terraced,

(iii) Cropping areas that aremainly found in the eastern end of the is-land between Kokkinochoria and the edge of the Troodos Moun-tains. The principal crop is wheat, followed by olives, vineyardsand citrus fruits. There is intense farming with irrigation anduse of fertilisers in areas such as the Kokkinochoria, west ofLemesos on the Akrotiri Peninsula east of Paphos and in thevalley south from Polis. Historically, most of the irrigation waterwas obtained from ground water resources. Groundwater

s in the soils of Cyprus, J. Geochem. Explor. (2014),

200015001200800500200100

0

m asl

Mamonia Terrane

Quaternary deposits

KeryneiaTerraneCircum-TroodosSedimentarySequence

Troodos Ophiolite Complex &Arakapas Transform Sequence

ultramafic unitmafic unitspillow lavas

0 20km

Pafos

Lefkosia

Larnaca

Lemesos

Polis

Akrotiri

Kokkinochoria

Fig. 1. Simplified geology and topography of Cyprus.

2 A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

aquifers have been highly overexploited over the years withpiezometric levels decreasing rapidly with concurrent aquifers'deterioration – quantity and quality – seawater intrusion (espe-cially in the east of the island). Today irrigationwater in croppingareas is supplied mainly from dams, to a lesser extent fromground water and also from tertiary treatment of sewage efflu-ent. Irrigation water is distributed through an open channel andpipe network. Modern irrigation systems such as drippers, minisprinklers and low capacity sprinklers have been used in Cyprusagriculture for the last 30 years. Most fertilisers applied inCyprus soils are mineral or chemical fertilisers.

(iv) Urban areas that incorporate industrial, commercial and residen-tial areas.

(v) Mining areas, for which the greatest environmental impact isderived from the (historical) copper mines and processing oper-ations that ring Troodos in the pillow basalts. Cyprus has a verylong history of mining which spans over 4000 years. Over200 Mt of mine wastes are estimated to have been producedsince the early 1900s (Demetriades et al., 2006).

2. Sampling and analysis

Sampling for the atlas involved collection of top soils (0–25 cmdepth) and sub-soil (50–75 cm) on a 1 km × 1 km grid (Cohen et al.,2011). The samples were sieved to b2 mm mesh, air-dried and milled(95% passing b 75 μm sieve) using a Cr-steel mill. A typical sample sizewas 1 kg of soil depending of the topography of the sampling site. Allsamples were analysed for metals via both aqua regia extraction in-duced coupled plasma mass spectrometry (ICPMS) and total metals byinstrumental neutron activation analysis (INAA). Soil pH and electricalconductivity (EC) were measured on 1:5 soil:water slurries.

Ion chromatography (IC) has proved to be a useful technique in thepast for measuring inorganic ions in water samples such as under-ground, surface and seawater (Carrozzino and Righini, 1995; Gros andGorenc, 1997; Jackson, 2000; Morales et al., 2000; Rowland et al.,1995). The method has also been applied for the measurement of inor-ganic ions in soils but to a lesser extent (Park et al., 2002). Ideally, solu-tions injected for ion chromatography analysis should be low in organicmaterials such as organic solvents, pollutants, strong acids, and soluble

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

salts; hence, many of the traditional methods used for the extractionor digestion of soils are not compatible with ion chromatography(Tabatabai and Basta, 1982). When it comes to regional geochemicalmapping of soils the methods of analyses have to be rapid and practicalbecause of the large number of samples processed. We have developeda method designed for this kind of application where within less than 4h of laboratory work (including sample preparation) fit for purpose re-sults could be obtained.

As part of this study, the effects of changing the sample massto (18.2 MΩ/cm) deionized water ratio (solid–solution ratio) andleach time on extraction levels were tested using a bulk soil sample(CYP-A), derived from a calcareous marl in Cyprus milled to b75 μmand homogenised. The various combinations of parameters and extrac-tion levels are summarised in Fig. 2. The results indicated that higherconcentrations of inorganic ions were extracted as the dilution factorand the shaker time was increased.

Prior to analysis, all solutions were filtered through a 0.45 μmmem-brane to remove larger particulate matter. Conductivity of the filtratewasmeasured using a Mettler Toledo MC226meter. Samples of electri-cal conductivity greater than 600 μS/cm or subsequently found to haveanion concentrations exceeding the calibration range were diluted.

Chromatographic analysiswas carried out using a Shimadzu ion chro-matograph (model SCL-10Asp)with 50 μL sample loop. The system com-ponents consisted of one anion guard columnShim-pack IC-SA2(G) and aseparator column Shim-pack IC-SA2 (250mmL × 4.0mm) for the simul-taneous separation of anions, an anion self-generating suppressor pack-age HK-10Asuper, a pulsed electrochemical detector CDD-10Asp and anadvanced gradient pump LC-10ADsp. The eluents were delivered tothe columns by a Shimadzu eluent degasser module DGU-14A andthe above setup was used in combination with the SLC-10Asp systemcontroller. Calibration curves for each anion were obtained usingcertified single element calibration solutions with concentrations1.000 ± 0.002 g/L (Table 1). All the calibration curves had squaredcorrelation coefficient R2 greater than 0.99.

For every batch of 25 samples an internal QC (CYP-A) was run. Forevery run of the QC sample the whole extraction process and analysiswere applied and therefore reproducibility estimates incorporate ex-traction and analytical variance. The coefficient of variance (CVR) wasestimated from the QC results as a percentage giving the total variance

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),

Conc. (mg/L)

0

40

80

120

160

200F-

Cl-

NO3-

SO42-

5100

60

10100

60

20100

60

5100

30

10100

30

20100

30

5100

120

10100

120

20100

120

5200

120

5

300

120

volume (ml)time (min)

mass (g)

Fig. 2. Variation in inorganic ion extraction concentrations as a function of sample mass, leachate water volume and digestion time.

3A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

for each measured parameter (Table 2). Lower reproducibility was ob-served in the case of fluorides and chlorides which can be attributedto the sample pretreatment conditions since anion adsorption on thewalls of the container used for the water extractions was observed.

Of the main suite of 5515 top soil samples collected as part of theatlas project, 3771 were analysed for water-soluble anions. Based onthe results of the initial optimizing of the method, 5 g of milled soilwas leached in 200 mL deionised water for 120 min (with shaking)prior to the IC analysis.

3. Results and discussion

3.1. General patterns

The results for the anion analysis are summarised in Table 3, withmeans and standard deviations calculated separately for the main suitof samples and those collected from the saltflats and beachdeposits. Ex-cluding the latter, the average F− value is low and Cl− is high comparedwith averages across other parts of Europe.

The results are mapped in Fig. 3. The maps were generated byconverting the original point data to a 300 × 300 m grid using a 2 kmsearch radius and an inverse distance weighting of values relative to thedistance from the grid centre. The colours were allocated using ten divi-sions based on data percentiles. During mapping below detection limitvalues were replaced with half detection limit (LD) values. For reportingsummary statistics the conventional approach of reporting values as halfDL was adopted, though there were few samples below DL.

3.2. Water-soluble ions

3.2.1. FluorideThe highest F− value occurs at the periphery of the Skouriotissa Cu

mine (Cu mines are shown in Fig. 3). Elevated F− (N100 mg/kg) is

Table 1Single column ion chromatographic parameters for the ion chromatographic system andsingle element concentration ranges.

Variable Setting

Eluent 1.8 mM Na2CO3 + 1.7 mM NaHCO3

Flow rate 1 mL/minSeparator column 250 mmL × 4.0 mm Shim-pack IC-SA2Guard column Shim-pack IC-SA2(G)Injection volume 50 μLDetector CDD-10Asp suppressed conductivityLow concentration rangefor F−, Cl−, NO3

−, SO42−

0.05–10 mg/L

High concentration range for Cl− 20–75 mg/LHigh concentration range for SO4

2− 10–50 mg/L

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

observed in the vicinity of salt lakes with values typically N100 mg/kg,but elsewhere is consistently low (b20 mg/kg).

The main natural source of inorganic fluorides in soil is the parentrock (WHO, 1984). Whereas some fluoride minerals weather rapidly,the more common F-bearing minerals of fluorapatite and fluoriteweather slowly (Fuge and Andrews, 1988; Kabata-Pendias and Pendias,1992). Fertilisers derived from apatite are the main non-geogenicsource of F− contamination of soils (Neumüller, 1981; Thomson et al.,1979), with such phosphate fertilisers generally containing ~3.5% F−

(McLaughlin et al., 1996). Generally, F− applied through fertilisertends to have high residence times within the soil matrix, particularlyin soils of high clay content or high organic carbon content, and a num-ber of studies have categorized fluoride as extremely immobile in soil(Bower and Hatcher, 1967; Polomski et al., 1982). Weakly elevated F−

to the east of Larnaca, near some villages in the western part of the is-land and at various locations near the coast is not attributed tofertilisers, due to the low correlation between F− and NO3

− suggestingthat the major sources of F- are geogenic.

3.2.2. ChlorideChloride concentrations are high in the soils from the salt lakes at

Larnaca and Akrotiri (some samples containing N20% Cl−) and coastalareas affected by sea-salt spray where concentrations are typicallyN2000 mg/kg. In some agricultural areas both irrigation and land clear-ing have probably contributed to the rise of groundwater tables (partlydue to deforestation) and rapid evaporation in the dry climate typical ofregions similar to Cyprus (Goldsmith and Hildyard, 1984). Values werevery low (typically b100 mg/kg) in the Troodos Massif and between250 and 400 mg/kg in the carbonate-dominated Troodos flanks.

Most of the chloride present in soil comes frommarine aerosols, vol-canic emissions, irrigation, and use of fertilisers (Xu et al., 2002). Chlo-ride is a highly mobile element and is only weakly absorbed by claysand soil organics under acidic conditions (Mortvedt, 2000). Apart fromthe areas affected by acid derived from the weathering and oxidationof sulphide mineralisation and from related mining operations, thesoils of Cyprus have an alkaline pH (Cohen et al., 2012b).

There is poor spatial and statistical correlation between Cl− and Na+

in the soils, apart from the saline lakes where most samples plot along

Table 2Estimates of reproducibility (CVR) for F−, Cl−, NO3

− and SO42− on in-house soil reference

material CYP-A.

Cl− (mg/kg) F− (mg/kg) NO3− (mg/kg) SO4

2− (mg/kg)

N 151 151 151 151Mean 3.31 0.25 0.56 3.29SD 1.24 0.08 0.10 0.41RSD 0.38 0.33 0.19 0.12%CVR 37.6 32.7 18.6 12.4

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),

Table 3Summary statistics for the 3771 samples (including 29 from salt flats and beach deposits). Population splitting was based on Q–Q plots.

Cl− (mg/kg) F− (mg/kg) NO3− (mg/kg) SO4

2− (mg/kg)

All samples excluding salt flats and beachdeposits (n = 3741)

Range Population 1 b200 b50 b1000 b100Geometric mean 110 7 8 41Range Population 2 200–15,000 100–35,000Geometric mean 370 – – 355

Typical global average valuesa ~100 200–300Salt flats and beach deposits (n = 30) Geometric mean 6310 71 710 13,740All samples (n = 3771) Minimum 2 1 b1 b1

25th percentile 94 9 14 3350th percentile 134 13 26 5075th percentile 188 19 64 8690th percentile 278 26 163 20695th percentile 519 33 259 553Maximum 72,070 3536 3000 231,700

a Data from Bowen (1979), Mortvedt (2000), and De Vos et al. (2006).

4 A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

the Cl:Na molar equivalence line (Na+ data obtained from Cohen et al.,2012a). As Na was analysed using aqua regia extraction which attackscarbonates, exchangeable Na on clays and organics, and most Na salts,the results would indicate that most of the Na is related to Na-substituted carbonate in the CTSS and Na-altered (spilitised) pillowbasalts. Moderate to high values of chlorides were measured aroundthe Cu mines but may relate to the effects of hydrothermal alteration

3.00%0.330.250.200.160.04

3,000 mg/kg500290200180

90

Top soil(0 –25 cm)

3,771 samples

Pafos

Lefkosia

Larnaca

500 mg/kg221916128Lemesos

Polis

Akrotiri

Kokkinochoria

Na+

Cl-

F

A-

Fig. 3. a. Map of distribution of water-extractable F−, Cl−, on 1:5 soil:water and aqua-regia extrtribution ofwater-extractable SO4

2−, NO3− and ECon 1:5 soil: water in Cyprus soils. Coppermine

zones (NVZ) are shown on the NO3− distribution map.

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

by fluids sourced from seawater that generated the Cu deposits(Constantinou and Govett, 1972).

3.2.3. SulphatePatterns in the SO4

2− distribution appear to be derived from threemain sources. The first is from the weathering and oxidation ofsulphur-bearing minerals (pyrite and chalcopyrite) that characterise

B

4.00 mS/cm0.340.210.170.150.08

220 mg/kg8050302215

Rivers4th to 7th order

5000 mg/kg280135754525

SkouriotissaLimni

Kokkinopesula

3,771 samples

EC

NO3-

SO42-

Cu Mines Basins

Kalavassos

Top soil(0 –25 cm)

actable Na+ in Cyprus soils. Major urban centres are shown on the F− map. b. Map of dis-s andmajor basins are shownon the SO4

2− distributionmap. Rivers and nitrogen vulnerable

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),

EC (mS/cm)

10

100

1k

10k

100k

0.01 0.1 10 100

10

100

1k

10k

SO42-

Cl-

Saline lakesamples

1

EC (mS/cm)

10

100

1k

NO3-

F-

10

100

1k

0.01 0.1 10 1001

Mine areas

Fig. 4. Correlation between water-soluble F−, Cl−, NO3− and SO4

2− and electrical conductivity on 1:5 soil:water slurry in Cyprus soils.

F-

1

10

100

Cl-

100

10

100

1k

10k

5A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

the pillow basalt-hosted Cyprus-type Cu deposits, and areas whereopen cut mining operations and mineral processing plants have beendeveloped to exploit the mineralisation. Acid mine drainage in groundand surface waters and possibly dispersion of dust and sedimentscontaining sulphides, Fe-sulphates such as jarosite and gypsum, hasgenerated elevated SO4

2− down-slope of these deposits, especiallyaround the Limni, Kalavassos, Skouriotisa and Kokkinopesula miningareas (Fig. 3) where there is corresponding elevations in the soil con-centrations of Cu, Zn, In and Fe (Cohen et al., 2012a). Some areas havebeen subjected to mining operations over various periods stretchingback to the Late BronzeAge (Koucky and Steinberg, 1982)with substan-tial historical mine wastes. Contamination relating to such deposits is acritical environmental management issue in Cyprus (Geological SurveyDepartment of Cyprus, 1998) and was a trigger for the GeochemicalAtlas of Cyprus project. The second is the elevated SO4

2− values(N1000 mg/kg) contained within the salt lake deposits and otherareas along the coastal strip. The third zone is in the southern PolisValley and SE of Kalavassos which correspond to the exposuresof deep marine carbonates of the Lefkara Formation that contains anumber of outcropping lenses of gypsum deposited during periods oflow sea levels and aridity that have affected the Mediterranean in theMessinian–Pliocene boundary, including sedimentation in the Polemi,Pissouri, Psemantismenos and Lemesos Basins (Rouchy et al., 2001).

1.0

0.8

0.6

0.4

0.2

0.0

0.00.0 0.2 0.4 0.6 0.8 1.0

0.2

0.4

0.6

0.8

1.0 Cl -

SO42-

Total soluble anions

1,000 mg/kg

500

25Salt flats

Mines andsulfidedeposits

Main cluster ofsamples fromagricultural areas

≥

Fig. 5. Ternary diagram of relative proportions of water-soluble Cl−, NO3− and SO4

2− inCyprus soils with total water-soluble anions indicated.

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

3.2.4. NitrateWhereas the salt lake sediments contain elevated NO3

−, most of thevalues above 50 mg/kg are located in the main agricultural areas andare thus likely to be related to both the use of application of synthetic

SO42-

10

100

1k

Landuse Geological groups

NO3-

1

10

Fig. 6. Boxplot comparison of water-soluble F−, Cl−, NO3− and SO4

2− in Cyprus soils sepa-rated by land use and main parent geology groups.

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),

10 100 1,000 10,000 100,000101

102

103

104

105

106

107

SO42-F- Cl-NO3

-

ΣΣN>r

Concentration (ppm) = r

Fig. 7. Number-concentration fractal plot for water-extractable fluoride, nitrate, chloride and sulphate ions.

6 A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

fertilisers or manure to fields, or residues derived from the crops them-selves. Another potential anthropogenic source of nitrates is the leakagefrom domestic septic fields, municipal sewage systems and livestockfacilities.Most of the high values are observedwithin the designated ni-trate vulnerable zones (NVZ), mainly within the Kokkinochoria regionand Mesaoria Plain NE of Troodos. Moderate to high nitrate values arealso found within artificial surfaces and along the coastal areas. Thereare also typically higher NO3

− values along the ridges than the adjacentriver valleys.

Due to the eutrophication (increasing algae growth, degrading hab-itat for aquatic organisms) and adverse effects on human health causedby excess nitrates in soil and ground or surface waters the EuropeanUnion has taken steps within the last few decades to monitor and re-strict the use of fertilisers in agricultural soils. NVZ have been designatedby the Cyprus government through specific studies (Geological SurveyDepartment, 2000) in an effort to comply with the EU Nitrate Directive(91/676/EEC). Whereas the NO3

− value in the NVZs is ~120 mg/kg(twice themean value observed for thewhole dataset), the low correla-tion of nitrates with phosphorus within the NVZ zones (Cohen et al.,2011) suggests that there may be either, a limited use of nitrogen-phosphorus rich fertilisers, or input balance crop demand.

3.3. Other patterns

There is a weak correlation between the SO42− and Cl− contents and

soil EC (mainly at high EC values and related to salt-affected soils andthe sulphide deposits), but little correlation with NO3

− or F− (Fig. 4).These patterns indicate some contribution to EC from the carbonates,which dominate much of the geology away from Troodos.

A ternary plot of Cl−–NO3−–SO4

2− indicates two main trends in thedata (Fig. 5). The first are the series of low-NO3

− samples extendingfrom the SO4

2− apex (mainly samples near the mines) to the Cl− apex(mainly the salt flats and some coastal areas) and contains themajorityof samples with high total soluble anions. The second trend extendsfrom around 0.3 on the SO4

2− axis towards the NO3− apex, with most

samples containing b200 mg/kg total anions and covers the agriculturalareas and CTSS.

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

3.4. Population splitting

The effect of different land use and parent geology type is indicatedby way of comparative boxplots in Fig. 6. There is little apparent influ-ence of differing land use on Cl−, F− and SO4

2− apart from the beachdeposits. Nitrate values are strongly affected by land use and are signif-icantly higher in areas used for growing grain and other crops comparedwith forests and grasslands. In terms of the effect of different parentrock type, Cl− shows little variation, but the other anions show distinctdifferences. SO4

2− is highest in the basalts and is related to both the pres-ence of sulphide in deposits and related alteration halos and the occur-rence of disseminated sulphides in the basalts. F− is highest in soilsderived from the carbonates. Nitrate is highest in the areas underlainby basalts and alluvium/colluvium but these are also the areas contain-ingmost of the cropping and urban areas (geology affecting topographyand soil types which, in turn, influences land use).

Further subdivision of the populations has been undertaken usingthe number-size fractal methods previously used in the analysis of soiland sediment geochemical data by Shen and Cohen (2005) andGeranian et al. (2013). Individual populations displaying fractalbehaviour display the relationship:

N ≥rð Þ ¼ Cr−D

where r represents concentration, N (≥r) is the cumulative concentra-tion of observations greater than or equal to r, C is a constant and D isthe power scaling or fractal dimension of the distribution. If a plot of N(≥r) versus r (in log-log coordinates) is linear then all data belong toa single population and the distribution is a simple fractal. If, however,the plots can be fitted with several straight-line segments, then thedistribution is multifractal and the break-points between straight-linesegments are the thresholds which separate the populations.

All the ions displayed distinctmulti-fractal behaviour (linear plots inlog–log coordinates) for the populations other than the first (lowest)concentration population (Fig. 7). The breaks in slope on the fractalplots were determined visually. Segments that deviate from linearityindicate that some data (sub-populations) do not exhibit ideal fractalbehaviour within the C–N method used. The upper two or three fractal

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),

SO42-

F-

Cl-

NO3-

0 20km

Fractalpopulation

2

3

4

Cu Mines

Fig. 8. Distribution of fractal populations 2, 3 and 4 for water-extractable F−, Cl−, NO3− and SO4

2−, with underlying geological groups (key in Fig. 1).

7A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

populations are plotted in Fig. 8. The upper two Cl− populations (3 and4) relate to the saline flats and other locations along the coastal fringe.Cl− population 2 is largely restricted to the agricultural areas thatoccur on the areas of low relief that flank Troodos and correlates spatial-ly with population 2 for nitrate— both suggesting the effects of fertiliserand/or irrigation. The samples in the second and third F− and SO4

2− pop-ulations are largely restricted to the deep basins containing LefkharaFormation carbonates, as well as the areas around and down-slope ofthe abandoned sulphide mines.

4. Conclusions

The distribution of the concentrations of water-soluble anions in thesoils of Cyprus is largely controlled by geogenic factors (parent rock, in-cluding the presence of sulphide mineralisation) and natural environ-mental effect (proximity to the coast and the areas of salt flats). TheNO3

− distribution is affected by landuse (mainly the effects of fertilisers)and SO4

2− values are elevated in the vicinity of the now-abandonedsulphide mining operations (although such areas also have naturallyelevated SO4

2− in the soils). These results mirror those of the metalanalysis (Cohen et al., 2012a, 2012b) where the dominant control onsoil geochemistry is parent lithology but with some detectable anthro-pogenic influences for some metals in the urban areas and miningoperations.

Acknowledgements

This project was funded by the Cyprus Ministry of Agriculture Natu-ral Resources and Environment (contract AR 2005/12). The authorsthank Actlabs and the various field and laboratory crews for theirvaluable assistance.

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

References

Barth, J.A.C., Grathwohl, P., Fowler, H.J., Bellin, A., Gerzabek,M.H., Lair, G.J., Barcelo, D., et al.,2009. Mobility, turnover and storage of pollutants in soils, sediments and waters:achievements and results of the EU project AquaTerra. Agron. Sustain. Dev. 29,161–173.

Bloem, E., van der Zee, S.E.A.T.M., Toth, T., Hagyó, A., 2008. Risk assessment methods ofsalinity (RAMSO42-IL). Sixth Framework Program, Scientific Support to Policies.

Bowen, H.J.M., 1979. Environmental Chemistry of Elements. Academic Press, New York.Bower, C.A., Hatcher, J.T., 1967. Adsorption of fluoride by soils and minerals. Soil Sci. 103,

151–154.Carrozzino, S., Righini, F., 1995. Ion chromatographic determination of nutrients in sea

water. J. Chromatogr. A 706, 277–280.Cohen, D.R., Rutherford, N.F., Morisseau, E., Zissimos, A.M., 2011. Geochemical Atlas of

Cyprus. UNSW Press, Sydney.Cohen, D.R., Rutherford, N.F., Morisseau, E., Zissimos, A.M., 2012a. Geochemical patterns

in the soils of Cyprus. Sci. Total Environ. 420, 250–262.Cohen, D.R., Rutherford, N.F., Morisseau, E., Christoforou, E., Zissimos, A.M., 2012b.

Anthropogenic versus lithological influences on soil geochemical patterns inCyprus. GEEA 12, 349–360.

Constantinou, G., Govett, G.J.S., 1972. Genesis of sulfide deposits, ochre and umber ofCyprus. Inst. Min. Metall. 81, B34–B46.

Darnley, A.G., Björklund, A., Bølviken, B., Gustavsson, N., Koval, P.V., Plant, J.A., et al., 1995.A global geochemical database for environmental and resource management: recom-mendations for international geochemical mapping. IGCP Project 259UNESCO Publ.

De Vivo, B., Albanese, S., Lima, A., Cicchella, D., Dinelli, E., Valera, P., Reimann, C., Birke,M., Demetriades, A., et al., 2013. GEMAS: the geochemical mapping of agriculturaland grazing land soils of Europe. E3S Web of Conferences 1, #38004.

De Vos, W. and Tarvainen, T. (Chief-editors), 2006. Geochemical Atlas of Europe, Part 2:Interpretation of Geochemical Maps, Additional Tables, Figures, Maps, and RelatedPublications. Geological Survey of Finland, Espoo, Finland, 690 pp.

Demetriades, A., Androulakakis, N., Charalambides, A., 2006. National inventory of poten-tial sources of soil contamination in Cyprus. Report to Cyprus Ministry of Agriculture,Lefkosia.

Fuge, R., Andrews, M.J., 1988. Fluorine in the UK environment. Environ. Geochem. Health10, 96–104.

Geological Survey Department of Cyprus, 1998. Mining waste management of Cyprus:assessment, strategy, development and implementation — final rpt. LIFE ProgrammeContract No. 94/CY/B21/CY/0977/MED.

Geological Survey Department of Cyprus, 2000. Study of the Protection of Waters Causedby Nitrates from Agricultural Sources According to Directive 91/676/EEC. GeologicalSurvey Department of Cyprus.

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),

8 A.M. Zissimos et al. / Journal of Geochemical Exploration xxx (2014) xxx–xxx

Geranian, H., Mokhtari, A.R., Cohen, D.R., 2013. A comparison of fractal methods andprobability plots in identifying and mapping soil metal contamination near an activemining area, Iran. Sci. Total Environ. 463, 845–854.

Goldsmith, Ε., Hildyard,Ν., 1984. The Social and Environmental Effects of Large Dams. vol. 1.Wadebridge Ecological Centre, UK (Ch 11).

Gros, N., Gorenc, B., 1997. Performance of ion chromatography in the determination of an-ions and cations in various naturalwaterswith elevatedmineralization. J. Chromatogr.A 770, 119–124.

Hadjiparaskevas, C., 2005. European Soil Bureau — research report no. 9. Soil Survey andMonitoring in Cyprus.

Jackson, P.E., 2000. Ion chromatography in environmental analysis. In: Meyers, R.A. (Ed.),Encyclopedia of Analytical Chemistry. John Wiley & Sons Ltd, Chichester, pp.2779–2801.

Kabata-Pendias, A., Pendias, H., 1992. Trace Metals in Soils and Plants, 2nd edition. CRCPress, Boca Raton, Fla, USA.

Koucky, F.L., Steinberg, A., 1982. The ancient slags of Cyprus. In: Muhly, J.D., Maddin, R.,Karageorghis, V. (Eds.), EarlyMetallurgy in Cyprus 4000–500 BC. Pierides Foundation,Nicosia, pp. 117–137.

McLaughlin, M.J., Tiller, K.G., Naidu, R., Stevens, D.P., 1996. Review: the behaviour andenvironmental impact of contaminants in fertilisers. Aust. J. Soil Res. 34, 1–54.

Morales, J.A., de Graterol, L.S., Mesa, J., 2000. Determination of chloride, sulphate andnitrate in groundwater samples by ion chromatography. J. Chromatogr. A 884,185–190.

Mortvedt, J., 2000. Handbook of Soil Science. CRC Press, New York, pp. D71–D88.Neumüller, O., 1981. 8th ed. Römpps Chemie Lexikon, vol. 2. Franck'sch Verlagshandlung,

Stuttgart.Park, H.-M., Kim, Y.-M., Lee, D.W., Lee, S.-w, Lee, K.-B., 2002. Ion chromatographic determi-

nation of inorganic anions in environmental samples of Korea. Anal. Sci. 18, 343–346.

Please cite this article as: Zissimos, A.M., et al., Distribution of water-soluhttp://dx.doi.org/10.1016/j.gexplo.2014.07.004

Polomski, J., Fluhler, H., Blaser, P., 1982. Accumulation of airborne fluoride in soils. J. Environ.Qual. 11, 457–461.

Reimann, C., Garrett, R.G., 2005. Geochemical background: concept and reality. Sci. TotalEnviron. 350, 12–27.

Rouchy, J.M., Orszag-Sperber, F., Blanc-Valleron, M.M., Pierre, C., Rivière, M., Combourieu-Nebout, N., Panayides, I., 2001. Paleoenvironmental changes at theMessinian–Plioceneboundary in the eastern Mediterranean (southern Cyprus basins): significance of theMessinian Lago-Mare. Sed. Geol. 145, 93–117.

Rowland, A.P., Woods, C., Kennedy, V.H., 1995. Control of errors in anion chromatographyapplied to environmental research. J. Chromatogr. A 706, 229–239.

Salminen, R., Batista, M.J., Bidovec, M., et al., 2005. Geochemical Atlas of Europe. Part 1:background information methodology and maps. Geol. Surv. Finland ISBN 951-690-913-2 (electronic version), (http://www.weppi.gtk.fi/publ/foregsatlas/index.php).

Shen, W., Cohen, D.R., 2005. Fractally invariant distributions and an application ingeochemical exploration. Math. Geol. 37, 895–909.

Soil Atlas of Europe, 2005. European Soil Bureau Network, European Commission, JointResearch Center.

Tabatabai, M.A., Basta, N.T., 1982. Ion Chromatography in Soil Analysis: Modern Instru-mental Techniques, In: Smith, K.A. (Ed.), 2nd editionMarcel Dekker, New York.

Thomson, L.K., Sidhu, S.S., Roberts, B.A., 1979. Fluoride accumulations in soils and vegetationin the vicinity of a phosphorus plant. Environ. Pollut. 18, 221–234.

WHO, 1984. Fluorine and Fluorides. Environmental Health Criteria, Geneva, p. 36.Xu, G., Magen, H., Tarchitzky, J., Kafkaf, U., 2002. Advances in chlorine nutrition of plants.

Adv. Agron. 28, 97–150.

ble inorganic ions in the soils of Cyprus, J. Geochem. Explor. (2014),