Catena 150 (2017) 206–222

Contents lists available at ScienceDirect

Catena

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

Assessment of arsenic and associated metals in the soil-plant-watersystem in Neogene basins of Attica, Greece

Evdokia E. Kampouroglou ⁎, Maria Economou-EliopoulosDept. of Geology & Geoenvironment, Section of Economic Geology & Geochemistry, National University of Athens, Panepistimiopolis, 15784 Athens, Greece

⁎ Corresponding author.E-mail addresses: ekamp@geol.uoa.gr (E.E. Kampouro

(M. Economou-Eliopoulos).

http://dx.doi.org/10.1016/j.catena.2016.11.0180341-8162/© 2016 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 17 February 2016Received in revised form 10 November 2016Accepted 14 November 2016Available online xxxx

This paper presents a study on the soil-plant-water system in the Neogene basins of Attica (Greece), where anuppermost yellow-brown travertine limestone has been deposited. The principal goal of our study is to identifythe percentage of arsenic (As) and metal transferred into plants and crops (bio-accumulation) in order to de-scribe the incorporation of As in plants from rocks and soils and to assess potential groundwater contamination.The bio-accumulation factor for plants exhibits a wide range, including relatively low (1.5–7.6%) for As, Pb, Ni,Mn, Cr, Ba, Sb, Fe, much higher (29–67%) for Cu, Zn, Co, Ca andMg, and exceptionally high (265%) for P. Althoughacceptable limits for As and heavy metals for plants have not been defined, As contents (dry weight) in plantsfrom the Neogene basins of Attica are often higher than normal or limited values. The bio-accumulation factorfor plants (Asplant / Assoil * 100) in Neogene basins of Attica exhibited a positive correlation between As and Fe,Cr, Mn, Pb and Sb.The estimated risk assessment maps for As, Na, Cl and Se in water are produced according to the parametricvalues of Directive 98/83/EC, although these elements showed higher risk values in the southwest and centralpart of the Mesogeia basin than in the Kalamos-Varnavas basin. The elevated Na, Cl, As, Se, Li and B concentra-tions,measurement of salinity and factor analysis in groundwater in theMesogeia basinwere attributed to a con-tribution by seawater in this aquifer.The estimated risk assessment maps of As in soils and ground waters in Neogene basins of Attica may indicate apotential human health risk and environmental significance of an integratedwater-soil-plant investigation of Ascontamination in similar Neogene lacustrine formations.

© 2016 Elsevier B.V. All rights reserved.

Keywords:ArsenicContaminationRiskBio-accumulationGround waterNeogene basinsGreece

1. Introduction

Arsenic (As), which is associated with igneous and sedimentaryrocks, and sulphide ores, is the 20th most common element in theearth's crust (0.5–2.5 mg/kg) (Kabata-Pendias and Mukherjee, 2007).Arsenic compounds are observed in rock, soil, water, air and plant andanimal tissues. The risk of As entering the food chain through water,soil and plant/crop contaminations has been a topic of global interest(WHO, 2004). The recommended limit of As in drinking water is10 μgL−1 As (USEPA, 2001), although at present, there is no knownsafe limit for As. Recent research suggests that tuberous vegetables ac-cumulate higher amount of arsenic than leafy vegetables, althoughleafy vegetables accumulate higher amounts of arsenic than fruityvegetables (Alam et al., 2002; Bhattacharya et al., 2010a, 2010b;Roychowdhury et al., 2002; Samal, 2005). The rate of uptake and thebio-accumulation factor of As in crops usually depends on its availability

glou), econom@geol.uoa.gr

in soils, the soil pH, organicmatter, redox potential andmineral compo-sition (Mandal and Suzuki, 2002).

In aquatic systems, inorganic arsenic occurs primarily in two oxida-tion states, As(V) and As(III). Both forms generally co-exist, althoughAs(V) predominates under oxidizing conditions and As(III) predomi-nates under reducing conditions, depending on the Eh, pH, salinity,metal concentrations, temperature, and distribution and compositionof the biota (USEPA, 1984). Natural levels of arsenic in soil usuallyrange from 1 to 40 mg/kg, with a mean of 5 mg/kg. Arsenic is observedin many foods at contents that usually range from 20 to140 μg/kg.

Elevated As contents were recently recorded for the first time in alimestone quarry in Varnavas basin (NE Attica, Greece) that is exploitedfor a popular multicolour building material and in the associated soil(Kampouroglou and Economou-Eliopoulos, 2013). The results of a geo-chemical investigation at Neogene basins,which cover a significant por-tion of Attica, were presented using the geographical informationsystem (GIS), geostatistical techniques and mapping software to showthe extent and intensity of As contamination and other elements (Fe,Mn, Ni, Cr and Ba) in travertine limestone and associated soils(Kampouroglou and Economou-Eliopoulos, 2016). These researchers

207E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

concluded that the major contamination sources are most likely sul-phide and Fe-Mn mineralization in Attica, basement rocks that involvemylonitic ophiolitic blocks and, to a lesser extent, human activities.

Given that the uppermost yellow-brown travertine limestone,which is distributed across Attica, has high contents of As and heavymetals, our present study focused on the investigation of plants andwater samples (springs, wells and boreholes) from the major Neogenebasins of Attica. Our study aimed to assess the percentage of As andmetal transferred into plants/crops (bio-accumulation) in order to de-scribe the incorporation of arsenic in plants from soils and potentialgroundwater contamination.

1.1. Description of the site-sampling

The study area of Attica is composed of alpine basement, bothmetamorphic and non-metamorphic rocks, and post-alpine forma-tions. The metamorphic section of the Attica-Cyclades zone occursin the eastern and southern parts of Attica, where it is transformedby high pressure and low temperature conditions and consists of Tri-assic schists, metabasic, and quartz-feldspathic rocks (Papanikolaouand Papanikolaou, 2007).

The geomorphology of the area is dominated by the Parnitha andAegaleo Mountains in the west, and the Penteli and Hymettus Moun-tains in the east that, border the Athens basin. During the Upper Plio-cene-Pleistocene, activation of the Penteli detachment fault lifted theHymettus Mountain and caused the separation of the Athens basinfrom the Mesogeia basin (Mposkos, 2008). Maps showing contaminat-ed and potentially contaminated sites coupled with mineralogical and

Fig. 1. Topographical map showing the sampling lo

geochemical data confirm the geotectonic literature data suggest theseparation of an initially single basin in Attica into smaller basins(Kampouroglou and Economou-Eliopoulos, 2016). Three major drain-age basins can be distinguished from the coastal zone of Oropos-Kalamos in the southern Evoikos gulf to the Saronic gulf: 1) TheKalamos-Varnavas basin toward the northwest, 2) The Athens basin inthe south with a NNE-SSW flow direction, and 3) The Mesogeia basinwith a major W-E flow direction between the Hymettus Mountain inthe west and the Penteli Mountain in the north (Fig. 1).

The alpine bedrock is covered by post-alpineNeogene to Quaternaryformations. These formations include 1) Upper Pliocenic deposits(marine-coastal) of Southern Athens and Eastern Mesogeia basins,2) Upper Miocenic deposits (lacustrine formations-travertine lime-stone,fluvial-lacustrine,fluvial-continental and lacustrine-continental),and 3) Quaternary alluvial deposits (I.G.M.E., 2000, 2002, 2003). TheNeogene lacustrine deposits of Kalamos-Varnavas basin consist ofmarls, marl limestone with lignite intercalations of the Malakasa-Oropos area and travertine, while upward to these, clays, sandstonesand conglomerates are developed (Ioakim et al., 2005; Mettos, 1992).In the central part of the Mesogeia basin, alternation marls and marllimestonewith lignite intercalations of Rafina are found (Mettos, 1992).

The most important aquifers in the Kalamos-Varnavas basin areconstituted by Triassic-Jurassic limestones, Upper Cretaceous lime-stones, marly limestones and travertine limestones, which are char-acterized as permeable rocks. The Triassic-Jurassic limestones of theSub-Pelagonian zone in the Kalamos-Varnavas basin are character-ized as karstic aquifer and with outlets through submarine andcoastal springs located in the north part of the Kalamos area. The

cations in Neogene basins of Attica (Greece).

208 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

terrestrial springs are the spring of contact between the travertinelimestone and basement schists. Upper Miocene conglomerates andmarly formations with lignite intercalations (fluvial-lacustrine de-posits) form a poor aquifer. The alluvial deposits form a phreaticaquifer that possess low hydraulic characteristics and feed a largenumber of wells and boreholes, with depths ranging from 10 to20 m (Stamatis et al., 2011).

The main part of Mesogeia basin is occupied by Quaternary forma-tions (gravels, sand, and clays) that characterize a low productive phre-atic aquifer. The basis of this aquifer consists of schists, carbonate rocksand Neogene deposits. In the Neogene formations, alternate permeableand impermeable layers or lenses result in the development of uncon-fined and confined aquifers (Champidi et al., 2011). The impermeablebottom of the Neogene basin is composed of basement schists, whichcontrols the regional groundwater flow and causes high surface runoff. Carbonate rocks, limestone and marble are found in the schists,and these constitute themost important aquifer of the basin by showinghigh permeability due to their intense karstification and fracture poros-ity (Stamatis et al., 2006).

2. Materials and methods

2.1. Sampling and preparation

Soil, plant and water samples have been collected from the basinsassociated with the lacustrine formations of travertine limestone(Kampouroglou and Economou-Eliopoulos, 2013, 2016). Soil samples(n=37)were collected from the rhizosphere of plants during thewin-ter of 2012 and spring of 2013 from amaximumdepth of approximately20 cm inorder to investigate the relationship between element contentsin plants and soil. Soils were dried at 50 °C, crumbled mechanically andpassed through a sievewith a 2mmmesh. Soil thatwasfiner than 2mmwas pulverized and analysed by Inductively Coupled PlasmaMass Spec-troscopy (ICP/MS) after Aqua Regia Digestion at the ACME AnalyticalLaboratories in Canada. These soil data have been previously published

Table 1Descriptive statistics of soil samples (n= 37) from the Neogene basins of Attica. The metal andground values in soils (n = 6) established in this study.

Soils associated with travertine limestones

Units Average Min. Max. Standard deviation Curvature

As mg/kg 210 16 1200 33 16Cu 30 5.3 77 2.8 1.6Pb 70 14 240 8.4 4.6Zn 167 39 500 17 2.5Ni 144 27 660 22 5.2Co 23 9 64 1.6 7.4Mn 776 290 1400 52 −0.7Cr 172 17 1400 40 19Ba 158 6 410 18 −0.7Th 4.9 0.05 11 0.5 −0.3Sr 55 15 140 4.7 1.4Cd 0.5 0.2 1.5 0.1 3.2Sb 5.7 0.4 21 0.7 2.8V 58 13 160 6.2 0.1La 19 0.5 41 1.7 0.0Sc 7.3 0.3 19 0.8 −0.8Fe wt% 3.0 0.83 5.6 0.2 0.1Ca 13 0.6 36 1.5 1.5Mg 0.6 0.1 2.6 0.1 6.9Ti 0.1 0.0005 0.5 0.0 −1.0Al 3.2 0.03 9.3 0.4 −0.8Na 0.2 0.0005 0.7 0.0 0.1K 0.7 0.005 2.1 0.1 −1.1P 0.1 0.02 1.3 0.0 36organic matter 4.0 0.6 19.9 0.6 8.5pH 8.0 7.4 8.3 0.04 1.2Eh mV −87 −111 −57 2.4 0.1

(Kampouroglou and Economou-Eliopoulos, 2016). Six (6) referencesoils un-associated with travertine limestone and developed in theclose vicinity of exposed rock outcrops in the Neogene basins of Attica(area of Holargos and the western part of the Kalamos-Varnavasbasin)were collected and considered to beunaffected soils by travertinelimestone (Tian et al., 2017).

The vegetation of Attica is composed of forests of Pinus halepensis,Querqus coccifera, Cupressus sempervirensvar, and Olea. Phryganagrow in the forest gaps and in newly afforested areas. Grassesfound in the region are classified as species of shrubby pseudo-steppe derived from degradation of the natural forest vegetation(e.g., Brachypodium ramosum, Poa bulbosa, Avena sp., and Bromussp.). There are also important species of geophytes such as, Cyclamengraecum. The following plant species were collected, at the sametime with the soils, and analysed in the present study: Stachysgermanica (KAL-PL1, KAL-PL4, KAL-PL6, KAL-PL13, KAL-PL14, VAR-PL21), Thymus sibthorpii (KAL-PL2, KAL-PL3, KAL-PL5, KAL-PL7,KAL-PL15, VAR-PL20, VAR-PL24, DR-PL1, DR-PL5, AR-PL5), Sonchusoleraceus (KAL-PL8, KAL-PL11, VAR-PL22, DR-PL6, AR-PL1),Cichorium intybus (KAL-PL10, KAL-PL12, VAR-PL23), Bromus arvensis(DR-PL3, AR-PL2), Foeniculum vulgare (DR-PL4),Malva sylvestri (AR-PL7), Sinapis arvensis (AR-PL8), Phlomis fruticosa (DR-PL2), Inulaverbascifolia ssp. methanea (AR-PL4, AR-PL6) and Lactuca sativa(VAR-PL26, VAR-PL30, VAR-PL31). The investigated plant speciesare edible by humans and animals except for the species Phlomisfruticosa and Inula verbascifolia ssp. methanea.

The plant sample numberswere determined by the plant's availabil-ity and presence in the areas of interest. In certain cases, more than oneplant species growing on a single site in the same soil were collected toestimate the role of plant physiology andmetabolism tometal bioavail-ability. Plantswere cut to remove roots from stems andwashedwith co-pious amounts of distilled water to remove the adhered soil and dustparticles. Washed plants were oven dried at 50 °C overnight in the lab-oratory. A total of 32 wild and three cultivated (Lactuca sativa) plantswere powdered in an agatemortar and analysed by Inductively Coupled

OM contents are detailed in Kampouroglou and Economou-Eliopoulos (2016). The back-

Reference values Background values

Asymmetry Gazette 641B/1991 Dutch limits (2009) Average Range

3.4 – 76 16 7–251.4 50–140 190 38 28–502.1 50–300 530 35 15–761.5 150–300 720 81 58–1302.1 30–75 100 255 130–4702.1 – 190 28 19–390.3 – – 888 680–11804.0 – 180 126 67–2000.6 – 8 87 54–1400.2 – –1.1 – – 75 18–1041.8 – 13 0.4 0,2–0,81.6 – 22 0.6 0,2–1,40.9 – – 40 23–530.5 – – 11 7–160.6 – –0.3 – – 2.9 2,2–3,31.0 – – 6.8 2,2–112.2 – – 1.4 0,5–20.6 – – 0.015 0,01–0,020.6 – – 1.6 1–2,31.0 – – 0.02 0,01–0,060.6 – – 0.36 0,17–0,676.0 – – 0.11 0,06–0,222.4 – – 1 0,7–1,4−1.2 – –0.4 – –

Table 2Major and trace element contents in plants.

Location Samplemg/kg t%

Shoots As Mo Cu Pb Zn Ni Co Mn Th Sr Cd Sb La Cr Ba B e Ca P Mg Al Na K S

Kalamos KAL_PL1*1 0.05 0.7 11 0.4 27 0.9 40 37 0.05 5.2 0.005 0.05 0.2 2.2 2.4 4 .03 0.4 0.1 0.1 0.02 0.3 1.2 0.2KAL_PL2*2 5.4 0.4 7.5 1.5 36 7.6 18 54 0.1 9.3 0.05 0.1 0.8 8.1 28 12 .1 1.2 0.09 0.1 0.1 0.04 0.6 0.1KAL_PL3*2 30 0.4 8.7 1.1 29 18 16 58 0.2 10 0.02 0.6 0.5 37 28 10 .2 1.2 0.08 0.1 0.1 0.04 0.7 0.09KAL_PL4*1 3.9 2.6 3.6 0.7 19 2.8 22 39 0.03 5.1 0.005 0.1 0.1 6.9 2.8 3 .04 0.4 0.07 0.1 0.02 0.4 1.0 0.1KAL_PL5*2 0.9 0.2 8.4 1.1 25 2.7 12 26 0.08 9.0 0.04 0.05 0.4 3.9 8.8 11 .06 1.2 0.05 0.1 0.05 0.03 0.8 0.1KAL_PL6*1 0.7 0.7 3.1 1.3 44 2.3 34 53 0.07 5.1 0.03 0.06 0.3 3 9.1 3 .05 0.4 0.04 0.06 0.04 0.01 0.6 0.1KAL_PL7*2 1.2 0.3 6.7 2.2 27 4.1 6.2 33 0.2 12 0.07 0.09 0.7 4.9 34 10 .09 1.3 0.05 0.1 0.09 0.06 0.3 0.1KAL_PL8*3 0.4 0.9 11 0.5 60 1.5 8.1 47 0.02 11 0.1 0.1 0.1 3 5.8 17 .03 1.7 0.08 0.2 0.02 0.5 2.6 0.2KAL_PL10*4 1.0 2.7 5.4 0.4 57 3.5 4.6 50 0.03 13 0.005 0.08 0.1 4.8 6.1 26 .04 1.7 0.2 0.2 0.02 0.5 2.0 0.3KAL_PL11*3 1.3 0.6 4.2 0.2 26 1.4 16 23 0.01 16 0.05 0.1 0.06 2.9 6.4 17 .02 1.2 0.09 0.2 0.01 0.9 2.4 0.2KAL_PL12*4 0.6 0.3 5.8 0.6 18 2.8 17 27 0.06 12 0.03 0.05 0.2 3.9 26 16 .05 1.9 0.2 0.2 0.04 0.1 1.4 0.1KAL_PL13*1 0.2 0.6 3.5 0.5 20 1.4 33 27 0.02 2.0 0.01 0.07 0.1 3.8 1.9 3 .02 0.2 0.2 0.07 0.02 0.2 0.8 0.08KAL_PL14*1 0.2 0.6 3.6 0.2 26 2.1 11 52 0.005 2.1 0.01 0.03 0.03 2.4 2.2 3 .01 0.2 0.07 0.04 0.005 0.01 1.0 0.1KAL_PL15*2 0.4 0.2 7.1 0.7 30 1.9 24 15 0.03 8.0 0.02 0.05 0.2 3.4 9.7 11 .03 1.2 0.07 0.1 0.02 0.08 0.9 0.07

Average 3.3 0.8 6.4 0.8 32 3.8 19 39 0.1 8.6 0.03 0.1 0.3 6.4 12 10 .1 1.0 0.1 0.1 0.04 0.2 1.2 0.1Varnavas VAR_PL20*2 1.9 0.5 8.2 3.0 29 2.7 35 27 0.1 7.5 0.04 0.2 0.5 4.9 13 11 .07 1.1 0.04 0.1 0.05 0.1 0.7 0.08

VAR_PL21*1 0.3 0.8 2.9 0.5 12 1.4 11 17 0.02 2.6 0.01 0.05 0.09 2.3 3.5 2 .02 0.2 0.04 0.03 0.01 0.04 0.7 0.05VAR_PL22*3 0.3 1.9 12 0.5 55 0.8 21 36 0.02 9.0 0.05 0.05 0.09 4.2 3.9 27 .03 1.5 0.1 0.2 0.01 0.4 2.6 0.2VAR_PL23*4 1.8 0.4 7.7 1.7 48 1.6 19 26 0.07 16 0.1 0.1 0.3 3.4 12 23 .07 2.7 0.04 0.2 0.04 1.9 2.1 0.1VAR_PL24*2 0.2 0.4 5.2 0.7 37 0.8 10 14 0.02 7.0 0.01 0.05 0.1 2.1 8.4 11 .02 1.0 0.05 0.1 0.02 0.02 0.9 0.1VAR-PL26*11 4.3 1.4 11 0.4 60 2.5 0.2 250 b0.01 16 0.6 0.2 0.05 2.0 16 33 .05 1.5 0.5 0.6 b0.01 0.4 8.7 0.3VAR-PL30*11 6.5 1.7 12 2.6 70 1.6 0.2 130 0.02 13 0.8 0.4 0.4 3.1 11 38 .05 1.1 0.5 0.4 0.01 0.4 7.7 0.3VAR-PL31*11 1.4 1.2 61 3.8 170 16 5.6 240 b0.01 18 1.7 4.4 0.1 3.2 43 74 .2 2.0 0.1 0.5 0.03 0.8 3.9 0.4

Average 2.1 1.0 15.1 1.7 60.0 3.4 12.7 92.5 0.0 11.1 0.4 0.7 0.2 3.2 13.8 27.4 .1 1.4 0.2 0.3 0.0 0.5 3.4 0.2Drafi DR_PL1*2 1.0 0.8 5.8 1.5 27 13 15 31 0.09 35 0.02 0.06 0.3 6.3 11 14 .09 1.5 0.04 0.2 0.04 0.04 1.1 0.2

DR_PL2*5 2.5 0.2 3.2 1.3 44 7.7 8.9 20 0.08 19 0.06 0.07 0.3 3.8 6.4 15 .06 1.1 0.06 0.2 0.04 0.03 0.6 0.08

(continued on next page) 209E.E.Kam

pouroglou,M.Econom

ou-Eliopoulos/Catena

150(2017)

206–222

w

F

00000000000000000000000000

Table 2 (continued)

Location Samplemg/kg wt%

Shoots As Mo Cu Pb Zn Ni Co Mn Th Sr Cd Sb La Cr Ba B Fe Ca P Mg Al Na K S

DR_PL3*6 1.5 1.8 3.6 1.0 25 5.0 13 31 0.06 6.5 0.02 0.06 0.2 3.1 2.4 7 0.04 0.4 0.05 0.06 0.03 0.04 1.0 0.1DR_PL4*7 0.1 0.5 8.4 0.2 41 1.4 12 39 0.005 45 0.005 0.02 0.02 2.1 5.1 21 0.02 1.8 0.2 0.3 0.005 2.0 0.9 0.2DR_PL5*2 0.9 0.4 6.8 0.6 26 2.5 30 17 0.04 32 0.01 0.05 0.2 3 8.9 8 0.04 1.0 0.05 0.1 0.03 0.1 0.7 0.1DR_PL6*3 1.3 0.6 8.4 0.4 29 2.5 15 17 0.02 14 0.03 0.05 0.1 3 3.6 13 0.03 0.9 0.08 0.1 0.02 1.2 2.0 0.3

Average 1.2 0.7 6.0 0.8 32 5.4 16 26 0.05 25 0.02 0.1 0.2 3.6 6.2 13 0.05 1.1 0.1 0.2 0.03 0.6 1.1 0.2Artemida AR_PL1*3 0.8 0.6 8.5 1.5 30 1.9 8.0 33 0.07 28 0.03 0.07 0.3 3.6 7.1 26 0.07 2.3 0.06 0.2 0.04 0.3 2.4 0.2

AR_PL2*6 0.3 0.4 3.8 0.7 17 1.3 31 20 0.06 3.7 0.005 0.04 0.2 3.1 1.8 2 0.02 0.2 0.03 0.05 0.02 0.04 0.7 0.1AR_PL4*8 0.4 0.1 2.8 0.8 18 0.9 5.6 26 0.03 16 0.04 0.05 0.2 2.6 5.5 13 0.03 0.5 0.04 0.1 0.02 0.01 0.6 0.1AR_PL5*2 0.05 0.1 4.4 0.7 9.6 1.1 7.3 14 0.03 36 0.03 0.04 0.2 3.3 7 13 0.03 1.0 0.03 0.1 0.02 0.1 0.6 0.1AR_PL6*8 6.4 0.5 11 4.1 32 5.0 62 63 0.2 11 0.2 0.1 0.6 5.3 7.3 19 0.1 1.0 0.05 0.1 0.09 0.3 1.1 0.1AR_PL7*9 0.8 0.7 16 1.2 38 20 46 49 0.01 33 0.09 0.05 0.5 27 9.5 22 0.2 2.2 0.3 0.4 0.1 0.06 2.9 0.3AR_PL8*10 0.05 0.8 8.8 0.2 11 1.3 12 15 0.005 18 0.4 0.02 0.03 3.1 4 11 0.02 1.2 0.3 0.1 0.005 0.03 1.9 0.4

Average 1.3 0.5 7.9 1.3 22 4.5 25 31 0.06 21 0.1 0.05 0.3 6.9 6.0 15 0.06 1.2 0.1 0.2 0.05 0.1 1.5 0.2Kalamos KAL_PL2R*2 2.8 0.2 12 1.2 32 3.8 20 45 0.09 9.4 0.07 0.1 0.5 5.4 16 5 0.07 0.8 0.06 0.1 0.07 0.01 0.3 0.07

KAL_PL5R*2 2.3 0.2 13 1.2 12 3.4 26 42 0.1 4.4 0.04 0.09 0.6 4.5 6.4 4 0.07 0.4 0.03 0.07 0.07 0.04 0.5 0.06KAL_PL15R*2 1 0.1 8.4 1.0 13 1.9 14 22 0.06 3.7 0.03 0.07 0.3 3.6 5.2 4 0.04 0.4 0.02 0.05 0.04 0.02 0.3 0.04

Average 2.0 0.2 11 1.1 19 3.0 20 36 0.09 5.8 0.05 0.09 0.4 4.5 9.3 4.3 0.06 0.5 0.04 0.07 0.06 0.02 0.3 0.06Varnavas VAR_PL24R*2 1.5 0.2 12 2.3 31 1.2 14 17 0.04 4.8 0.03 0.2 0.3 2.7 7.7 5 0.04 0.7 0.02 0.06 0.03 0.02 0.3 0.07Drafi DR_PL1R*2 1.6 0.7 12 1.3 16 15 68 38 0.08 17 0.02 0.08 0.3 7.9 5.4 6 0.08 0.6 0.02 0.1 0.05 0.02 0.6 0.05

DR_PL2R*5 24 0.08 51 5.3 19 4.7 0.3 23 0.01 15 0.07 0.7 0.09 3.1 6 5 0.03 0.6 0.02 0.06 0.02 0.008 0.2 0.02DR_PL3R*6 34 0.7 8.6 17 145 72 22 169 0.7 15 0.1 0.5 2.4 27 18 5 0.5 2.0 0.04 0.2 0.4 0.007 0.6 0.1DR_PL4R*7 2.5 0.1 11 0.8 61 4.8 7.6 37 0.1 36 0.02 0.1 0.2 3.9 8.8 11 0.06 1.0 0.06 0.3 0.04 0.9 0.2 0.09

Average 15 0.4 21 6.2 60 24 24 67 0.2 21 0.06 0.3 0.8 11 9.6 6.8 0.2 1.0 0.03 0.2 0.1 0.2 0.4 0.07Artemida AR_PL1R*3 2.8 0.5 25 2.9 73 3.4 10 69 0.08 23 0.04 0.08 0.4 4.5 9.6 15 0.08 1.4 0.06 0.1 0.06 0.2 0.7 0.1

AR_PL4R*8 0.9 0.08 4.8 0.9 9.5 1 8.6 13 0.04 11 0.03 0.1 0.2 2.9 3.8 5 0.03 0.3 0.02 0.05 0.03 0.001 0.4 0.04Average 1.9 0.3 15 1.9 41 2.2 9.5 41 0.1 17 0.04 0.1 0.3 3.7 6.7 10 0.1 0.9 0.04 0.1 0.05 0.1 0.5 0.1Detection limit 0.1 0.01 0.01 0.01 0.1 0.1 0.01 1 0.01 0.5 0.01 0.02 0.01 0.1 0.1 1 0.001 0.01 0.001 0.001 0.01 0.001 0.01 0.01Reference materialsSTD V14 11 0.1 4.8 0.9 14 1.4 0.8 2103 0.01 6.6 0.2 0.1 0.03 1.1 1.5 11 0.02 0.6 0.1 0.1 0.1 0.001 0.5 0.1STD V16 1.5 1.3 6.1 2.8 37 6.4 1.0 704 0.01 11 0.1 0.1 0.04 276 2.1 5.3 0.4 0.3 0.05 0.1 0.05 0.00 0.2 0.03STD CDV-1 1.1 0.2 7.5 0.8 21 5.8 1.8 340 0.6 103 0.03 0.03 2.1 11 7.8 12 0.2 1.7 0.03 0.1 0.1 0.005 0.2 0.10Excessive or toxic in shootsa 5–20 10–50 20–100 30–300 100–400 10–100 15–50 400–100 – – 5–30 150 – 5–30 – 50–200 – – – – – – – –

Symbols: *1: Stachys germanica; *2: Thymus sibthorpii; *3: Sonchus oleraceus; *4: Cichorium intybus; *5: Phlomis fruticosa; *6: Bromus arvensis; *7: Foeniculum vulgare; *8: Inula verbascifolia ssp. Methanea; *9:Malva sylvestri; *10: Sinapis arvensis; *11: Lactucasativa (cultivated plants).

a After Kabata-Pendias and Pendias, 2001.

210E.E.Kam

pouroglou,M.Econom

ou-Eliopoulos/Catena

150(2017)

206–222

211E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

Plasma Mass Spectroscopy (ICP/MS), after Aqua Regia Digestion, at theACME Analytical Laboratories in Canada. Roots were analysed separate-ly from the shoots.

Water samples (n=20) were collected from the surface dischargesof springs, agricultural and domestic wells, and boreholes. The springswere characterized as the contact between the travertine limestoneand basement schists. The water of wells and boreholes are associatedwith unconfined and confined aquifers, which develop in Neogene for-mations. A single round of groundwater samples was collected from 20locations within the study area in October 2012 (wet period). Measure-ments of T, pH, Eh, electric conductivity (EC), TDS and salinity were per-formed in the field at the time of groundwater sampling with ConsortC561 portable multiparameter analyzer. The measurement of T wasconducted in the wet period (October 2012) and ranged from 16 to18 °C for boreholes, 13 to 14 °C for springs and 14 to 16 °C for wells.The samples were collected and divided in two portions, and one por-tionwas stored in polyethylene containers at 4 °C in a portable refriger-ator. The second portion of each sample was acidified by addition ofconcentrated HNO3 acid and also stored at 4 °C. The groundwater sam-ples were analysed by inductively coupled plasma mass spectroscopy(ICP-MS) at the ACME Analytical Laboratories Ltd., Vancouver, Canada.

The detection limit of the analytical method and results of standard(STD) and black analysis (BLK) for all samples are provided at the end ofeach table.

2.2. Statistical analysis and geochemical mapping

Statistical software codes (Microsoft Excel 2007, MINITAB v. 15.0and SPSS v. 17.0) were used to represent the spatial distribution ofground water parameters and bio-accumulation in the Neogene basinsof Attica. For the correlation analysis, selected elements with relation-ships N0.50 were plotted in diagrams using Microsoft Excel software.

The R-mode factor analysiswas conducted by applying the Varimax-raw rotational technique with Kaiser Normalization processing to thegeochemical data set. The correct number of factors was selectedusing a combination of common criteria as to be expressed higher per-centage variation of geochemical data (% of variance) after using theScree plot method. For determination of inter-element relationships,factor loadings N0.50 were selected.

A geographical information system (GIS) of ArcGIS v. 10.2 is used todigitize available and generated information of geomorphology andchemical analysis. All maps required geo-referencing to the area coordi-nates according to the HGRS 87.

Hot Spot Analysis (Getis-Ord Gi*) is a tool of Spatial Statistics forArcMap that identifies statistically significant spatial clusters of highvalues (hot spots) and low values (cold spots). This analysis creates anew Output Feature Class with a z-score and p-value for each featurein the Input Feature Class.

Kriging (Geostatistical Analyst for ArcMap) is an interpolator thatcan be exact or smoothed depending on the measurement errormodel. Kriging is very flexible and allows a user to investigate graphsfor spatial auto- and cross-correlation. Kriging uses statistical modelsthat allow a variety of output surfaces including predictionmap by ordi-nary Cokriging and probability maps by indicator Kriging. In ordinaryCokriging, statistical weighting is based on the distance between mea-sured points and the overall spatial relationships between locations.This method also enable for assessment of multiple variables (z-scoreof hot spots analysis) that have an effect on the variable of interest (con-centration of the plant or water samples). Indicator Kriging assumes anunknown constant mean. The specified threshold (value from a guide-line) is an important parameter that determines which predictionswill receive a 0 or 1.

Saturation indices for selected minerals were calculated using thesoftware code Aquachem 2012.1 by Schlumberger Water Services tobetter understand the hydrochemical processes that occur in theaquifers.

3. Results

3.1. Trace element contents in plants and soils

Kampouroglou and Economou-Eliopoulos (2016) presentedthat the average contents of As, Ni, Cr and Ba in soils is higherthan the parametric values of Dutch limits (2009). The soil pHranges from 7.4 to 8.3 and Eh from −111 to −57 mV (Table 1).There is significant variation in the contents of As (16–1200 mg/kg),Cu (5.3–77 mg/kg), Pb (14–240 mg/kg), Ni (58–660 mg/kg), Co(15–64 mg/kg), Mn (290–1400 mg/kg), Cr (82–1400 mg/kg), Ba(6–410 mg/kg), Sb (1.6–21 mg/kg), V (15–160 mg/kg), Sc (0.3–19mg/kg), Fe (0.83–5.6 wt% ), and P (0.02–1.3 wt%) in soils, whichare commonly higher than in rock samples. Comparedwith the back-ground values, the average contents of As, Pb, Zn, Cr, Ba, Sb, V, La, Ca,Ti, Al, Na, K and organic matter are the highest, whereas the averagecontents of Cd and Fe are only slightly higher than or similar to thebackground values (Table 1).

The arsenic content in roots (Table 2) ranges from 0.8 to 34 mg/kgand from 0.05 to 30 mg/kg in shoots. The highest values for As, Mo,Pb, Zn, Ni, Mn, Th, Cd, La, Cr, Ba, Ti, Fe, Ca, Al and S in roots belong tothe sample DR_PL3R (Bromus arvensis). This plant showed low valuesof elements in shoots. The cultivated plants of lettuce had higher valuesof As, Mo, Cu, Pb, Zn,Mn, Cd, Sb, Ba, B, Fe, Ca, P, Mg, Na, K and S in shootsthan the average values of shoot samples in Neogene basins of Attica.

The highest contents of As were found in samples coming from theKalamos-Varnavas basin, particularly Thymus sibthorpii of the Kalamosarea and cultivated Lactuca sativa of the Varnavas area. The percentageof soil metals in plants [(metal in plant × 100) / metal in soil] or bio-accumulation factor for As, Pb, Ni, Mn, Cr, Ba, Sb and Fe ranging from1.5% to 7.6% is lower than that for Cu, Zn, Co, Ca and Mg, ranging from29% to 67%, and the exceptionally high average of 265% for P (Table3). Of these elements, Mn, Fe, Cu, Zn, Ca, P and Mg are characterizedas essential for plant growth. There is a positive correlation concerningthe bio-accumulation factor for As versus Νi, Fe and Cr for plants in allNeogene basis of Attica (Fig. 2; Table 4). Any relationship between thebio-accumulation factor for As and P is not obvious (r = 0.2), butthere is a strong positive correlation (r N 0.8) betweenAs and P contentsin plant roots from the Kalamos and Artemida basins, and a negativecorrelation (r = − 0.8) in plant shoots from the Drafi basin (Fig. 2d).There is also a positive correlation between As and Pb, Mn, Ba and Sband a negative correlation between As and Ca and Co (Table 4).

3.2. Elemental spatial distribution patterns of bioaccumulation

The predictable spatial distribution of the bio-accumulation factorfor As and Fe in the Neogene basins of Attica are presented (Fig. 3a–b). Cr, Ni and Mn bio-accumulation factor showed similar spatial distri-bution, as expected by the correlation analysis (Table 4). After Hot Spotanalysis, these maps are in a good agreement with the hypothesis thathot spots are present in the Varnavas area. Arsenic, Fe, Cr, Ni showhigh contents in the Varnavas area up to the northern part of theMesogeia basin (Drafi area). Manganese shows higher values in theVarnavas area up to the N site of Penteli Mount. The Kalamos basinpresents lower to intermediate values of As, Fe, Cr, Ni and Mn. The low-est values were found in the Mesogeia basin.

3.3. Health risk assessment in soils

The spatial maps of Fig. 4a-c indicate the probability that As, Ni andCr in soils to exceed the correspondingDutch Guideline (2009). There isa high probability of health risks from arsenic contamination for theKalamos-Varnavas basin. There is a medium probability of arsenic andnickel contamination in the eastern part of Mesogeia basin (Fig. 4a-b).In Fig. 4b, thewestern part of theAttica detachment has a high probabil-ity for health risks associated with nickel. Chromium shows high risk

Table 3Percentage of soil metals in plants (mplant/msoil * 100) from Neogene basins of Attica. Data from Table 1; Kampouroglou and Economou-Eliopoulos (2016), soils *: Kampouroglou andEconomou-Eliopoulos (2013).

mp/ms * 100

Location Description As Cu Pb Zn Ni Co Mn Cr Ba Sb Fe Ca P Mg

Kalamos PL1-S1 0.02 19 1.1 7.2 0.3 130 3.4 0.7 0.8 2 0.5 19 186 14PL2-S1 1.7 13 4.3 9.6 2.7 59 4.9 2.6 9.4 4 2.7 53 126 12PL3-S2 2.5 11 3.4 13 2.8 25 4.8 2.6 15 2.9 3.2 7.7 159 13PL4-S2 0.3 4.7 2.2 8.0 0.4 34 3.3 0.5 1.5 0.5 0.6 2.7 129 9.4PL5-S3 0.8 28 1.1 15 1.7 42 1.9 2.1 2.5 2 1.4 97 112 12PL6-S3 0.6 10 1.3 26 1.4 118 3.8 1.6 2.6 2.4 1.1 31 90 7.0PL7-S4 1.9 33 5.2 30 4.6 41 5.8 3.8 14 3.6 3.0 7.9 91 18PL8-S5 0.2 27 0.8 50 0.7 27 4.3 0.9 1.4 1 0.6 177 231 18PL10-S6 0.7 16 2.8 119 1.8 18 11 2.1 3.8 0.7 2 8.2 678 21PL11-S6 0.9 12 1.5 53 0.7 62 5.1 1.3 4 0.8 1.1 5.8 404 18PL12-S7 1.1 8.6 1.3 4.2 1.8 61 2.1 1.6 8.1 2 1.4 45 15 24PL13-S7 0.4 5.1 1.1 4.5 0.9 124 2.1 1.6 0.6 2.8 0.5 5.8 13 9.2PL14-S8 0.2 5.3 0.4 15 0.8 33 4.7 0.7 0.7 1.2 0.2 29 149 3.1PL15-S8 0.4 10 1.5 16 0.7 76 1.4 1.0 3.2 2 0.8 182 147 9.1

Varnavas PL20-S10 1.6 34 2.2 17 4.7 204 5.2 6.0 8.7 1.5 2.8 8.5 108 28PL21-S10 0.3 12 0.4 6.8 2.4 63 3.3 2.8 2.3 0.4 0.6 1.9 110 8.5PL22-S11 0.4 37 0.7 42 0.7 75 5 2.6 1.4 0.8 0.6 238 430 27PL23-S12 0.6 30 0.8 9.6 1.2 64 2.2 2 3.4 0.8 1.6 104 102 30PL24-S12 0.1 20 0.3 7.5 0.6 36 1.2 1.2 2.5 0.3 0.5 37 124 18

Drafi PL1-S1 0.3 29 3.4 37 3.4 43 4.4 2.2 6.5 2.4 2.7 12 186 33PL2-S1 0.8 16 3.0 62 2.0 25 2.9 1.3 3.8 2.8 1.9 9.0 255 23PL3-S1 0.5 18 2.3 34 1.3 37 4.4 1.1 1.4 2.4 1.2 2.9 223 8.8PL4-S2 0.2 38 0.5 65 1.1 54 5.2 1.6 2.1 0.8 0.7 16 540 43PL5-S2 1.4 31 1.4 42 1.9 135 2.3 2.3 3.7 2 1.4 8.8 173 19PL6-S2 2.0 38 1.0 46 1.9 70 2.3 2.3 1.5 2 1.0 8.0 280 17

Artemida PL1-S1 0.6 33 1.6 40 1.9 42 3 2.4 3.2 2.8 2.7 19 76 30PL2-S1 0.2 15 0.7 23 1.3 166 1.8 2.1 0.8 1.6 0.9 1.8 36 6.5PL4-S2 0.4 13 1.1 20 0.9 31 3.9 1.9 2.1 2 1.0 4.5 88 15PL5-S2 0.1 20 1.0 11 1.1 41 2.1 2.4 2.7 1.6 0.9 8.9 72 15PL6-S3 2.4 35 1.7 24 4.2 309 6.8 3.3 3.2 1.6 3.7 23 91 14PL7-S4 2.2 25 4.4 61 4.8 140 7.9 4.3 7.3 2 5.7 20 381 16PL8-S4 0.1 14 0.7 18 0.3 36 2.4 0.5 3.1 0.8 0.6 11 422 5.2

Drazaiza* EV1 -ES1 3.3 42 17 74 8.8 21 8.9 5.3 19 20 3.9 18 164 23Pourithi* EV2 -ES2 2.9 72 2.6 89 3.5 17 10 5.3 7.7 6 2.5 13 428 82

EV4 - ES4 6.1 161 6.8 83 6.3 20 11 12 13 12 6.3 10 371 59Varnavas quarry* EV5 - ES5 2.9 40 2.3 33 13 41 17 6.8 11 5.1 2.7 5 417 31

EV7 - ES7 9 95 17 62 14 40 45 15 19 19 11 11 568 76EV8 - ES8 1.7 30 2 22 3.8 28 11 3.8 25 4.6 2.4 26 1329 81PL31 - ES5 0.5 320 6 95 30 39 34 7.8 59 61 7.9 14 648 195

Symbols: mp = metal content in plant; ms = metal content in soil.

212 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

values in the northwestern to northern part of Kalamos-Varnavas basinand in the north side of Athens basin. In the central part of theMesogeiabasin, there is amedium risk for chromium. Arsenic accumulates in veg-etables, fruits, and other plants that grow in contaminated soils, sug-gesting that this is an important pathway for the transfer of arsenic tothe food chain (Meharg and Hartley-Whitaker, 2002).

3.4. Trace element concentrations in water

Concentrations of As, B, Cr, Cu, Mn, Ni, Sb, Se, Na and SO4 in watersampleswere lower than the parametric values established by theEuro-pean Community (Council Directive 98/83/EC). In contrast, the ele-ments Ag, Al, Au, Bi, Cd, Ce, Co, Er, Eu, Fe, Ga, Hf, Hg, Ho, In, La, Lu, Nd,Pd, Pr, Pt, Sm, Ta, Tb, Te, Th, Ti and Tm were lower than the detectionlimit, so were not presented in Table 5. In water, pH ranges from 6.6to 7.8 and Eh ranges from −83 to 23 mV (Table 5). The water samplesof Artemida (Mesogeia basin) are characterized as brackish water be-cause of salinity values that are 0.8 to 0.9‰. The salinity values of thewater samples are between 0.2 and 0.4‰ in the Kalamos-Varnavasbasin and described as fresh water. Arsenic and sulphur showed highervalues in the southern and southwestern part of theMesogeia basin. In-termediate to high concentrations of Na and Cr were also measured inthe Mesogeia basin. There is a significant positive correlation betweenAs and Li, Se, K, Mg, Na and S (Fig. 5). In Fig. 5, groundwater valuesfrom the Mesogeia basin of the present study were compared to

published data (wet period) for the same basin in the area of Koropi(n = 31) (Chrisanthaki, 2010; Pavlopoulos et al., 2011).

The water samples from the areas of Kalamos, Varnavas andKaisariani are classified as a Ca\\HCO3 water type, while samples fromArtemida are a Ca\\Cl water type (Fig. 6).

The mineralogical composition of water-bearing formations incombination with the calculation for thermodynamic equilibriumconditions of ground waters (Table 6) showed that the main min-erals present in the aquifers are: anhydrite (CaSO4), aragonite(CaCO3), barite (BaSO4), calcite (CaCO3), chalcedony (SiO2), dolo-mite (CaMg(HCO3)2), gypsum (CaSO42H2O), halite (NaCl), quartz(SiO2), talc (Mg3Si4O10(OH)2) and witherite (BaCO3). The groundwater of the study area is saturated in quartz and unsaturated in an-hydrite, aragonite, chalcedony, gypsum, halite, talc and witherite.Aragonite is saturated only in agricultural wells of the Varnavasarea. Barite is saturated in agricultural wells of the Artemida area.Calcite is saturated in springs of Varnavas and in agricultural wellsof Varnavas and Artemida. Dolomite is unsaturated in all samplesites except from the area of Artemida.

3.5. Statistics-factor analysis in waters

The results of multivariate analysis (factor) in the ground water ofKalamos-Varnavas and Mesogeia basin are presented in Table 7 andFig. 7a-c. Three factors were identified that account for 71.7% of thetotal variance. The first factor explained 52.5% of the total variability

R² = 0,77

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10

Ni m

p/m

s*10

0

As mp/ms*100

R² = 0,90

0

2

4

6

8

10

12

14

16

18

0 2 4 6 8 10

Cr

mp/

ms*

100

As mp/ms*100

R² = 0,87

0

2

4

6

8

10

12

0 2 4 6 8 10

Fe

mp/

ms*

100

As mp/ms*100

Kalamos Varnavas Drafi Artemida

c

0,01

0,10

1,00

0,01 0,10 1,00 10,00 100,00 1000,00

P (

wt%

)As (mg/kg)

Roots Artemida

Shoots Artemida

d

a b

Roots Drafi

Shoots Drafi

Roots Kalamos-Varnavas

Shoots Kalamos-Varnavas

Fig. 2. Correlation analysis diagrams for selected elements in the percentage of bioaccumulation (a–c) and As versus P content in shoots and roots (d). Data from Tables 1 and 3.

213E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

that includes As, B, Cr, Li, Se, Cl, K, Mg, Na and S. This factor (Fig. 7a),which could be considered the salinity factor, expresses the total saltconcentration of the groundwater samples and is determined by theseions. Factor 2 explains 10.2% of the total variance and has positive load-ings for Li and Si. This factor explains the dissolution of silicates from themetamorphic basement (Fig. 7b). Factor 3, which explains 9.0% of thevariance, has high positive loading (N0.75) on Zn, and may beinterpreted as leaching from the lignite intercalations found in the Neo-gene lacustrine deposits (Fig. 7c).

3.6. Health risk assessment in waters

The estimated risk assessment maps for As and Na in water showthat these elements exceed the corresponding Directive 98/83/EC(Fig. 8a–b). Similar risk values were presented for Cl and Se. Thesefour elements showed high risk values in the southwestern and cen-tral part of the Mesogeia basin. In contrast, our results showed thatthe Kalamos-Varnavas basin is safe.

Table 4Correlation matrix of selected major and trace elements in the percentage of soil metals in pla

As Cu Pb Zn Ni Comp/ms * 100

As mp/ms * 100 1Cu 0.41 1Pb 0.71 0.47 1Zn 0.42 0.64 0.47 1Ni 0.74 0.68 0.73 0.44 1Co −0.12 −0.16 −0.26 −0.31 −0.11 1Mn 0.59 0.61 0.69 0.56 0.77 −0.27Cr 0.74 0.74 0.65 0.42 0.91 −0.06Ba 0.66 0.61 0.75 0.36 0.82 −0.31Sb 0.58 0.63 0.79 0.41 0.79 −0.21Fe 0.77 0.67 0.86 0.48 0.87 −0.11Ca −0.09 0.08 −0.15 −0.05 −0.18 0.03P 0.24 0.53 0.31 0.67 0.3 −0.41Mg 0.44 0.82 0.5 0.5 0.65 −0.28

4. Discussion

4.1. Bioaccumulation of As

Based on the presented data, the level of accumulated elementsdiffers between and within species, most likely due to genetic dif-ferences (Meharg and Hartley-Whitaker, 2002). As contents inroots, are often higher than normal or limited values (Kabata-Pendias and Pendias, 2001). Additionally, Co and Cr in shoots andNi, Co and Cr in roots are often higher than normal or limited values(Kabata-Pendias and Pendias, 2001). Acceptable limits for As andheavy metals for plants, however, have not been established. Arse-nic content appears to be elevated in green vegetables and plantsgrown in the most contaminated soils. The bio-accumulation factorfor As (Asplant / Assoil * 100), ranges from 0.02 to 9.0% and it has apositive correlation with that for Fe, Cr, Ni, Mn, Pb and Sb, suggest-ing that even low As/metal contents in the soil can be significant inplants.

nts from the Neogene basins of Attica. Data from Table 3.

Mn Cr Ba Sb Fe Ca P Mg

10.73 10.67 0.73 10.69 0.71 0.7 10.71 0.83 0.85 0.74 1−0.17 −0.13 −0.01 −0.06 −0.13 10.62 0.29 0.41 0.24 0.33 0.04 10.65 0.71 0.7 0.6 0.68 0.07 0.51 1

a

b

Fig. 3. Prediction maps showing the spatial distribution pattern of the bio-accumulation factor for (a): As and (b): Fe in Neogene basins of Attica. Data from Table 3.

214 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

a b

c

Fig. 4. Estimated risk assessment maps for (a): As; (b): Ni; (c): Cr in soils corresponding Dutch Guideline (2009). Data from Table 1; Kampouroglou and Economou-Eliopoulos (2016).

Table 5The concentrations of major and trace elements in waters from the Neogene basins of Attica and sea-water from the Evoikos Gulf.

μg/l mg/lmV μS/cm ‰

Location Samples As B Cr Zn Li Se U Si Ca Cl K Mg Na S HCO3 SO4 TDS pH Eh EC sal

Kalamos KAL W1*1 6.4 31 19 1.8 9.8 1.4 0.4 12 120 77 1.5 14 29 6 366 18 350 7.2 −1 654 0.3KAL G01*2 6.2 16 6.5 1300 17 1.1 0.4 8.3 93 50 1.4 18 24 3 283 9 312 6.6 23 582 0.3KAL G02*2 1.6 12 8.6 2.7 1.8 0.0 0.4 5.6 96 17 0.7 18 10 2 293 6 251 7.1 5 469 0.2KAL_W10*1 6.3 31 12 b0.5 9.2 1.4 0.4 7.4 73 60 1.3 15 39 16 223 48 343 7.8 −83 640 0.3

Varnavas VAR R100*1 1.3 105 6.0 1.4 2.4 1.3 0.3 10 110 70 3.2 11 37 16 336 48 350 7.3 −13 648 0.3VAR R101*1 1.2 89 6.2 0.7 3.0 0.7 0.3 11 130 78 2.5 12 39 17 397 51 375 7.4 −16 704 0.3VAR R102*1 1.5 29 5.1 b0.5 6.1 1.3 0.2 11 130 82 0.8 15 40 24 397 72 413 7.5 −19 776 0.4KAPA F1*3 2.6 12 7.0 500 2.5 1.3 0.5 7.5 140 46 0.8 8 24 12 427 36 368 7.2 −4 686 0.3KAPA F2*3 3.2 15 6.0 400 1.6 0.5 0.6 6.5 140 42 0.8 7 23 15 427 45 344 7.2 −7 644 0.3KAPA F3*3 2.3 17 6.9 80 2.3 0.9 0.4 7.8 140 41 1.3 8 25 17 427 51 370 7.1 2 692 0.3KPW 100*1 3.6 25 7.3 0.7 3.5 1.0 0.4 8.3 120 48 0.9 10 28 9 366 27 327 7 6 605 0.3KPW 101*1 2.2 18 6.1 0.5 3.3 1.2 0.4 9.4 130 59 0.8 10 33 14 397 42 261 6.9 11 489 0.2KAPR2000*1 3.7 23 6.1 b0.5 4.5 0.9 0.3 9.7 100 48 0.8 11 31 8 305 24 319 7.3 −11 594 0.3KAP G1*2 0.3 15 5.6 b0.5 3.2 0.8 0.5 7.8 120 45 0.7 6 25 8 366 24 340 7.2 −3 638 0.3MET_W1*4 2.0 24 15 b0.5 14 1.5 1.0 13 71 46 1.4 30 42 12 216 36 480 7.6 −72 900 0.4MET_W2*4 2.2 15 13 380 6.7 0.7 0.2 7.4 97 26 1.6 5.7 16 11 295 33 268 7.7 −78 502 0.2

Artemida AR_W1*3 9.2 130 17 b0.5 12 5.0 3.3 12 140 310 5.6 64 120 34 427 102 900 7.4 −64 1670 0.8AR_W2*3 5.3 96 17 b0.5 14 6.6 1.8 12 160 370 2.7 47 130 26 488 78 950 7.2 −53 1770 0.9AR_W3*3 4.7 120 14 b0.5 12 5.5 3.2 11 120 300 5.4 62 120 34 366 102 880 7.5 −71 1650 0.8

Kaisariani W·UN·CAM*2 2.5 62 8.3 b0.5 7.8 1.8 2.6 8.5 85 71 1.9 45 42 9 259 27Evoikos Gulf sea water 80 4200 60 b50 160 360 b2 2 390 490 1200 6400 1200 1190 3600Detection Limit 0.5 5 0.5 0.5 0.1 0.5 0.02 0.04 0.05 1 0.05 0.05 0.05 1Reference MaterialsSTD TMDA-70 42 17 404 497 23 26 59 0.4 22 13 1.0 5.8 8.8 7STD TMDA-70 43 18 414 504 21 26 62 0.5 24 13 1.0 6.1 8.7 b1Reference values a 10 1000 50 – – 10 – – – 250 – – 200 – – 250 2500

Symbols: *1: spring; *2: borehole; *3: agricultural well; *4: domestic well.a Reference values according to Directive 98/83/EC.

215E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

R² = 0,94

0,01

0,10

1,00

10,00

100,00

1000,00

0 20 40 60 80 100

Se (

µg/

l)

As (µg/l)

R² = 0,84

1

10

100

1000

10000

0 20 40 60 80 100

S (m

g/l)

As (µg/l)

R² = 0,90

0,1

1,0

10,0

100,0

1000,0

0 20 40 60 80 100

K (

mg/

l)

As (µg/l)

R² = 0,86

1

10

100

1000

10000

0 20 40 60 80 100

Mg

(mg/

l)

As (µg/l)

R² = 0,35

1

10

100

1000

10000

0 20 40 60 80 100

Na

(mg/

l)

As (µg/l)

R² = 0,66

1

10

100

1000

0 20 40 60 80 100

Li

(µg/

l)

As (µg/l)

Koropi Kalamos Varnavas Artemida Kaisariani Sea water

Fig. 5. Correlation analysis diagrams for selected elements in water samples. Data from Table 5; Chrisanthaki (2010), for the Koropi area.

216 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

Factors that can influence the arsenic content in plants may be thespecies type, the ability of arsenic to enter the plant (actively or passive-ly), and the presence of species adhered to the outside surface of theplant roots (Vamerali et al., 2010). Baker and Brooks (1989) classifiedthe plants growing in metal-contaminated soils into three categories:i) excluders that maintain the concentration in shoots at a low levelup to the critical soil value, which is above relatively unrestricted root-to-shoot transport results, ii) accumulators that concentrate metals inhigh levels, and iii) indicators that uptake and transport the metals tothe shoot until toxicity occurs.

The mobility, transport and availability of As in soil and its mobiliza-tion from soil to shoots of the plants (bio-accumulation factor) ismostlydependent on the soil pH, the As content in soil and its speciation(Mench et al., 2009). Soil pH that ranges from7.4 to 8.3 suggests that ar-senic must be in the form of arsenate. More specifically, inorganic arse-nic in the form of arsenite [As(III)] is more mobile and considered to bemore toxic than arsenate [As(V)]. Nevertheless, both types of com-pounds are characterized as harmful to living organisms (Caruso et al.,2001; Kumaresan and Riyazuddin, 2001; Zhou et al., 2003). Negatively

charged arsenate is strongly adsorbed onto the surface of several com-mon minerals. Arsenite adsorbs less strongly, which is a property thatenables it to be more mobile (Mandal and Suzuki, 2002).

The mineralogical and geochemical data, particularly theFe\\Mn-hydroxides andminor base metal sulphides, such as pyrrho-tite, pyrite, chalcopyrite, sphalerite, barite, galenite, enargite andtennantite, as well as F-apatite, quartz, Ti-oxides, zircon, sphene, rutileandREE-phosphateminerals in the travertine limestones studied inNeo-gene basins of Attica (Kampouroglou and Economou-Eliopoulos, 2016),may provide evidence for their contribution of detrital and chemicalcomponents from the erosion of Grammatiko Fe-Mn and Lavrionminer-alization and host rocks (Marinos and Petrascheck, 1956; Skarpelis,2007; Voudouris et al., 2008). Moreover, the presence of Fe and manga-nese oxides also increases As mobility and its availability in soil (Zavalaand Duxbury, 2008).

Apart from presenting a health risk, the presence of As in irrigationwater or soil at an elevated level could hamper normal growth of plantswith toxicity symptoms such as biomass reduction (Carbonell-Barrachina et al., 1997) and yield loss (Jiang and Singh, 1994). Although

Fig. 6. A plot of water samples on a Piper diagram. Symbols: inverted triangles: Kalamos; triangles: Varnavas; orange circles: Artemida; green stars: Koropi; square: Kaisariani. Data fromTable 5; Chrisanthaki (2010), for the Koropi area. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

217E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

the mechanism of As bio-accumulation is not yet well understood todate, relatively high As, Fe, Cr, Mn, Ni and Co contents in the sample ofThymus sibthorpii are consistent with the relatively high contents ofthese elements in the soil of the area. This may confirm suggestionsthat As can be loaded into the xylem system (Verbruggen et al., 2009).In addition, the positive correlation between As and Fe and Cr (Fig. 2)may suggest that they have a common driving force for their bio-accumulation as they are affected by the same uptake mechanisms.With respect to Cr bio-accumulation, it may enter in plants by reductionand/or complexation with root exudates, such as organic acids, whichincrease the solubility and mobility of Cr through the root xylem(Bluskov et al., 2005). At neutral pH, Cr(VI) compounds are tetrahedraland transported across cell membranes through similar tetrahedral ion

Table 6Saturation indices for selected minerals in waters from the Neogene basins of Attica.

Location Samples Anhydrite Aragonite Barite Calcite Cha

Kalamos KAL W1*1 −2.41 −0.083 −0.624 0.065 −0KAL G01*2 −2.779 −0.279 −0.664 −0.131 −0KAL G02*2 −2.935 −0.244 −1.632 −0.097 −0KAL_W10*1 −2.138 −0.488 −0.24 −0.341 −0

Varnavas VAR R100*1 −2.013 −0.16 −0.278 −0.012 −0VAR R101*1 −1.947 −0.032 −0.295 0.116 −0VAR R102*1 −1.809 −0.041 −0.105 0.106 −0KAPA F1*3 −2.059 0.038 −0.501 0.186 −0KAPA F2*3 −1.963 0.037 −0.408 0.184 −0KAPA F3*3 −1.912 0.034 −0.361 0.181 −0KPW 100*1 −2.224 −0.08 −0.585 0.068 −0KPW 101*1 −2.02 −0.025 −0.375 0.123 −0KAPR2000*1 −2.325 −0.222 −0.529 −0.075 −0KAP G1*2 −2.265 −0.075 −0.589 0.073 −0MET_W1*4 −2.298 −0.518 −0.915 −0.371 −0MET_W2*4 −2.178 −0.241 −0.592 −0.094 −0

Artemida AR_W1*3 −1.753 −0.042 0.177 0.105AR_W2*3 −1.811 0.072 0.038 0.219 −0AR_W3*3 −1.798 −0.165 0.165 −0.017 −0

Kaisariani W·UN·CAM*2 −2.397 −0.383 −0.622 −0.235 −0

Symbols: *1: spring; *2: borehole; *3: agricultural well; *4: domestic well.

channels, while Cr(III) is octahedral and transported through diffusionacross membranes (Cohen et al., 2006).

Arsenic is toxic whereas phosphorus is essential for plants, althoughthey are both Group VA elements in the periodic system and havesimilar electron configurations and chemical properties. Due totheir similarity, arsenate competes for the same adsorption sites in thesoil, which results in a reduction of their sorption by soil and an increasein solution concentrations (Livesey and Huang, 1981; Manning andGoldberg, 1996; Smith et al., 2002). This assumes that arsenate canreact with the PO4 analogue present in soils and form stable insolublecompounds (Meharg, 1994).

With respect to the plants studied from Attica, any relationshipbetween the bio-accumulation factor for As and P is not obvious

lcedony Dolomite Gypsum Halite Quartz Talc Witherite

.088 −0.518 −2.173 −7.229 0.357 −5.108 −3.737

.249 −0.69 −2.542 −7.491 0.196 −5.36 −3.604

.42 −0.635 −2.698 −8.337 0.025 −6.029 −4.381

.299 −1.085 −1.901 −7.198 0.146 −5.814 −4.029

.167 −0.74 −1.776 −7.164 0.278 −5.755 −3.865

.126 −0.518 −1.71 −7.1 0.319 −5.512 −3.82

.126 −0.44 −1.572 −7.069 0.319 −5.247 −3.777

.292 −0.586 −1.822 −7.538 0.153 −6.688 −3.844

.354 −0.648 −1.726 −7.596 0.091 −7.118 −3.849

.275 −0.596 −1.675 −7.571 0.17 −6.635 −3.855

.248 −0.658 −1.987 −7.448 0.197 −6.184 −3.88

.194 −0.583 −1.783 −7.291 0.251 −6.005 −3.82

.181 −0.823 −2.088 −7.398 0.264 −5.754 −3.866

.275 −0.87 −2.028 −7.523 0.17 −6.947 −3.839

.054 −0.831 −2.061 −7.284 0.391 −3.932 −4.574

.299 −1.134 −1.941 −7.947 0.146 −7.066 −4.0950.158 −1.516 −6.04 −3.553

.086 0.194 −1.574 −5.93 0.359 −3.742 −3.521

.125 −0.035 −1.561 −6.05 0.32 −3.508 −3.636

.238 −0.461 −2.16 −7.105 0.207 −4.182 −4.047

Table 7Varimax rotated component loadings of 3 factors and variance explained for 13 elementsin waters of Neogene basins of Attica. Data from Table 5; Chrisanthaki (2010)—Koropiarea.

Variable

Factor

Communality1 2 3

As 0.84 0.24 0.22 0.81B 0.61 0.24 −0.41 0.6Cr 0.67 0.2 0.25 0.55Zn −0.06 −0.1 0.85 0.74Li 0.67 0.55 0.12 0.76Se 0.93 −0.09 −0.03 0.87Si −0.13 0.88 −0.19 0.82Ca 0.4 0.13 −0.14 0.2Cl 0.77 −0.19 −0.12 0.65K 0.91 0 −0.02 0.84Mg 0.91 0.09 −0.15 0.86Na 0.92 −0.03 −0.15 0.87S 0.84 −0.13 −0.18 0.75Total 6.82 1.33 1.17 9.32% of Variance 52.48 10.2 9.01 71.68

218 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

(r=0.2). However, a strong positive correlation (r N 0.8) between Asand P contents in plant roots from the Kalamos and Artemida basinsmay suggest that arsenate and phosphate are taken up into plantroots by the same mechanism (Meharg and Hartley-Whitaker,2002), whereas a negative correlation (r = −0.8) in plant shootsfrom the Drafi basin (Fig. 2d) may suggest that arsenite is more

a

c

Fig. 7. Graduated symbol plots of (8a): Log Factor 1 (As, B, Cr, Li, Se, Cl, K, Mg, Na, S); (8b): Locomparison with geology. Data from Table 5; Chrisanthaki (2010), for the Koropi area.

mobile and taken up by the plant through a P-independent mecha-nism (Wang et al., 2002).

4.2. Variation of the trace element concentrations in groundwater

The concentrations of As andother elements (B, Cr, Cu,Mn,Ni, Sb, Se,Cl, Na, Ba, Co,Mo, Zn, Li, Ge, Cs, Be, Nb, Rb, Ru, Sc, Sr, U, Si, V,W, Ca, K,Mg,S and REE) in water samples from the Neogene basis of Attica werelower than the parametric values of European Community (CouncilDirective 93/83/EC) (Table 5). However, certain ground water samplesfrom an area extending into the broader area of Koropi-Markopoulo-Paiania-Spata, south of Hymettus (Chrisanthaki, 2010), that have devel-oped within Quaternary and Neogene deposits have up to 68 μg/l of As(dry period) and a positive correlation between As and Na, Cl, Se, Li andB. The variability of As, B, Cr, Li, Se, Cl, K, Mg, Na and S in Factor 1 can becharacterized by two processes: a)mixing of the groundwaterwith sea-water mainly in Mesogeia basin and b) dissolution and leaching of ter-restrial salts present in the Neogene and Quaternary formations in theKalamos-Varnavas basin (Fig. 7a).

In addition, based on data for ground water from the area of Koropi(Chrisanthaki, 2010), the water can be divided into five water types:Ca\\HCO3 (39%), Na\\Cl (29%), Ca\\Cl (19%), Na\\HCO3 (10%) andMg\\SO4 (3%) (Fig. 6). In contrast, water samples from the areas ofKalamos, Varnavas and Kaisariani can be classified into Ca\\HCO3

water type and the samples from Artemida into Ca\\Cl water type(Fig. 6).

b

g Factor 2 (Li, Si); (8c): Log Factor 3 (Zn) scores for waters in Neogene basins of Attica in

a

b

Fig. 8. Estimated risk assessment maps for (8a): As; (8b): Na in waters corresponding Directive 98/83/EC. Data from Table 5; Chrisanthaki (2010), for the Koropi area.

219E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

220 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

Arsenic concentration in ground water may result from the natu-ral weathering of soil and rocks and/or anthropogenic sources, in-cluding mining, smelting, waste water, dumping of sewage sludge,coal mining, coal burning power plants, manufacturing processes,urban runoff and atmospheric deposition (Garbarino et al., 2003;Francis and White, 1987; Nriagu and Pacyna, 1988; Pacyna et al.,1995; Wadge and Hutton, 1987). Also because As and Na, Cl, Se, Liand B are elevated in the sea-water (Table 5; Rose et al., 1979) therelatively high concentrations of these elements in the area of Koropi(SW part of Mesogeia basin) are attributed to a major contribution ofsea-water to aquifers, which ismost likely due to their over exploitation(Chrisanthaki, 2010; Pavlopoulos et al., 2011). Among the As-bearingminerals in travertine limestone samples are bacterio-morphic aggre-gates of goethite containing up to 3.4 wt% As2O3, Fe-(hydro)oxidesand Mn\\Ba-(hydro)oxides (hollandite) containing up to 1.7 wt%As2O3 and siderite with up to 3.0 wt% As2O3 (Kampouroglou andEconomou-Eliopoulos, 2016). Proposed mechanisms of transport andreaction pathways of arsenic into groundwater include the oxidationof pyrite containing arsenic (Das et al., 1996; Mandal et al., 1998), re-duction of adsorbed arsenate to arsenite (BGS and DPHE, 2001; Boseand Sharma, 2002), competitive anion exchange of adsorbed arsenic(Acharyya et al., 1999; Appelo et al., 2002) and reductive dissolutionof iron oxides containing arsenic (Nickson et al., 2000; McArthur et al.,2001; Swartz et al., 2004).

Arsenic is mobilized with pH values typically found in groundwater(pH 6.5–8.5) and under both oxidizing and reducing conditions. For ourwater samples, the pH ranges from6.6 to 7.8 and Eh from−83 to 23mV(Table 5). The investigation of As(III) and As(V) sorption onto Fe-richsoil concretions demonstrated that As(V) sorption onto iron oxideshas a strong dependence on pH. In addition, the negative Gibbs free en-ergy (ΔG°) and positive entropy (ΔS°) values for arsenite and arsenatesorption on Fe-rich soils are consistent with spontaneous reaction be-tween the species and medium (Gupta, 1998; Partey, 2008). The meta-stable oxyhydroxide ferrihydrite with respect to goethite is exclusivelynano particulate under typical surface conditions, and has high specificsurface area and significant reactivity toward the sorption of aqueouscontaminants AsO4

3 − and SO42− (Gasparatos, 2013; Navrotsky et al.,

2008; Pinney and Morgan, 2013). Due to chemical similarity of phos-phorus and arsenic, the influence of phosphorus [P(V)] on As(III) andAs(V) adsorption to Fe oxides is greatly influenced by pH and P(V) con-centration (Jain and Loeppert, 2000).

Arsenic sorption on calcite has also been studied (Memon et al.,2009). Arsenate sorption on calcite increases from pH 6 to 10, peaks be-tween pH 10 and 12, and decreases above pH 12 (Goldberg andGlaubig,1988). In natural systems, arsenic may be incorporated into the latticestructure of calcite as arsenite under alkaline pH (Di Benedetto et al.,2006), whilst the arsenite retention mechanisms on calcite changedfrom adsorption to co-precipitation with increasing As(III) concentra-tion (Roman-Ross et al., 2006). Arsenate anions exhibit a great affinityfor calcite surface sites at pH 8.3 and form inner-sphere complexes atthe calcite surface (Alexandratos et al., 2007). Romero et al. (2004) sug-gested that arsenic retention in carbonate-rich aquifermaterial could bepartly due to adsorption onto calcite.

Elevated arsenic contents are also common in coal basins. Investiga-tion in coal basins of Czech Republic (Slejkovec and Kanduc, 2005) indi-cated that there are unexpected organic arsenic compounds in low-rankcoals. Organic acids such as humic acid and fulvic acid may competestrongly with As(III) and As(V) for active adsorption sites on mineralsurfaces that influence As mobility. The competition for active bindingsites on mineral surfaces between organic acids and As compoundsmay result in lowering As retention levels, especially under acidic con-ditions (Wang andMulligan, 2006). The presence of high As concentra-tions (from11 to 247 μg/l) in groundwater has been reported in an areabetween Kalamos, Markopoulo and Mavrosouvala due to lignite inter-calations in marly formations, marly limestone, travertines and/or towaste piles of lignite mining (Stamatis et al., 2011).

Arsenic and heavy metals in the majority of soil samples exceededthe parametric values of Dutch limits (2009), and the backgroundvalues from the area of Holargos andwestern part of Kalamos-Varnavasbasin. In contrast, parametric values of the Greek Directive are not yetavailable. There is a small portion of water samples (14%) at the south-western and central part of theMesogeia basin that present higher con-centrations of As than the parametric values of the EuropeanCommunity (Council Directive 93/83/EC). Given that groundwater con-tamination by As is already considered a serious global environmentalproblem and crops/vegetables grown on As-contaminated soils can bethe main source of arsenic for humans (Williams et al., 2005; Mehargand Hartley-Whitaker, 2002; Pigna et al., 2010), the As level in ground-water in Neogene basins of Attica combined with the As content in cer-tain soil and plant samples may point to possible health risks to humanhealth.

5. Conclusions

The compilation of trace element data corresponding to soils, plantsand groundwater from the Neogene basins of Attica led to the followingconclusions:

• The groundwater samples of Kalamos, Varnavas and Kaisariani areclassified as a Ca\\HCO3 water type, while those from Artemida andKoropi are a Ca\\Cl type.

• On the basis of the calculated saturation index, the groundwater is sat-urated in quartz and aragonite only in agricultural wells of Varnavasarea, and in calcite in springs and agricultural wells of Varnavas.

• The elevated Na, Cl, As, Se, Li and B concentrations in groundwater inthe Mesogeia basin suggest a contribution of sea water to this aquifer.

• Although acceptable limits for As and heavy metals for plants are lack-ing, the presentedAs contents (dryweight) in plants fromNeogene ba-sins of Attica are often higher than normal or sufficient values.

• Thebio-accumulation factor for As (Asplant / Assoil * 100) inNeogeneba-sins of Attica showed a positive correlation with that for Fe, Cr, Mn, Pband Sb.

• The bio-accumulation factor for plants is relatively low (1.5–7.6%) forAs, Pb, Cr, Ba and Sb, which is characterized as toxic for their growth.

• Uptakeof arsenic increaseswithhigher arsenic concentration in the soil.• The integrated water-soil-plant investigation of the arsenic contamina-tion in the Neogene basins of Attica may indicate a potential humanhealth risk in similar Neogene lacustrine formations.

Acknowledgments

The University of Athens is greatly acknowledged for the financialsupport of this work (Grant No. KE_11078). The Editor Dr. MarkusEgli, the Associated Editor Dr. Christina Siebe and reviewers are greatlyacknowledged for their constructive criticism and suggestions for im-provement of our manuscript. The Elsevier Language Editing Servicesare greatly acknowledged for editing the manuscript in British Englishlanguage.

Appendix A. Supplementary data

Supplementary data associated with this article can be found in theonline version, at doi: 10.1016/j.catena.2016.11.018. These data includethe Google map of the most important areas described in this article.

References

Acharyya, S.K., Chakraborty, P., Lahiri, S., Raymahashay, B.C., Guba, S., Bhowmik, A., 1999.Arsenic poisoning in the Ganges delta. Nature 401, 545–547.

Alam, M.G.M., Allinson, G., Stagnitti, F., Tanaka, A., Westbrooke, M., 2002. Metal concen-trations in rice and pulses of Samta village, Bangladesh. Bull. Environ. Contam.Toxicol. 69, 323–329.

221E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

Alexandratos, V.G., Elzinga, E.J., Reeder, R.J., 2007. Arsenate uptake by calcite: macroscopicand spectroscopic characterization of adsorption and incorporation mechanisms.Geochim. Cosmochim. Acta 71, 4172–4187.

Appelo, C.A.J., van der Weiden, M.J.J., Tournassat, C., Charlet, L., 2002. Surface complexa-tion of ferrous iron and carbonate on ferrihydrite and the mobilization of arsenic. En-viron. Sci. Technol. 36, 3096–3103.

Baker, A.J.M., Brooks, R.R., 1989. Terrestrial higher plants which hyperaccumulate metallicelements—a review of their distribution, ecology and phytochemistry. Biorecovery 1,81–126.

BGS, DPHE, 2001. Arsenic contamination of groundwater in Bangladesh. In: Kinniburgh,D.G., Smedley, P.L. (Eds.), Summary. British Geological Survey Report WC/00/19Volumen 1. British Geological Survey, Keyworth.

Bhattacharya, P., Samal, A.C., Majumdar, J., Santra, S.C., 2010a. Arsenic contamination inrice, wheat, pulses, and vegetables: a study in an arsenic affected area ofWest Bengal,India. Water Air Soil Pollut. 213 (1–4), 3–13.

Bhattacharya, P., Samal, A.C., Majumdar, J., Santra, S.C., 2010b. Accumulation of arsenicand its distribution in rice plant (Oryza sativa L.) in Gangetic West Bengal, India.Paddy Water Environ. 8 (1), 63–70.

Bluskov, S., Arocena, J., Omotoso, O., Young, J., 2005. Uptake, distribution, and speciationof chromium in Brassica juncea. Int. J. Phytoremediation 7:153–165. http://dx.doi.org/10.1080/16226510590950441.

Bose, P., Sharma, A., 2002. Role of iron on controlling speciation and mobilization of arse-nic in subsurface environment. Water Res. 36, 4916–4926.

Carbonell-Barrachina, A.A., Burlo, F., Burgos-Hernandez, A., Lopez, E., Mataix, J., 1997. Theinfluence of arsenite concentration on arsenic accumulation in tomato and beanplants. Sci. Hortic. 71, 167–176.

Caruso, F., Shi, X., Caruso, R.A., Susha, A., 2001. Hollow titania spheres from layeredprecursor deposition on sacrificial colloidal core particles. Adv. Mater. 13,740–744.

Champidi, P., Stamatis, G., Zagana, E., 2011. Groundwater quality assessment andgeogenic and anthropogenic effect estimation in Erasinos basin (E. Attica). Eur.Water 33, 11–27.

Chrisanthaki, I., 2010. Hydrogeological and Hydrochemical Study of the Toxic Metals Be-havior in the Wider Area of Koropi. (MSc thesis). Harokopio University.

Cohen, M.D., Prophete, C., Sisco, M., Chen, L., Zelikoff, J.T., 2006. Pulmonary immunotoxicpotentials of metals are governed by select physicochemical properties: chromiumagents. J. Immunotoxicol. 3, 69–81.

Council Directive 98/83/EC. On the quality of water intended for human consumption.Das, D., Samata, G., Mandal, B.K., Chowdhury, T.R., Chanda, C.R., Chowdhury, P.P., Basu,

G.K., Chakraborti, D., 1996. Arsenic in groundwater in six districts of West Bengal,India. Environ. Geochem. Health 18, 5–15.

Di Benedetto, F., Costagliola, P., Benvenuti, M., Lattanzi, P., Romanelli, M., Tanelli, G., 2006.Arsenic incorporation in natural calcite lattice evidence from electron spin echo spec-troscopy. Earth Planet. Sci. Lett. 246, 458–465.

Francis, C.W., White, G.H., 1987. Leaching of toxic metals from incinerator ashes. J. WaterPollut. Control Fed. 59 (11), 979–986.

Garbarino, J.R., Bednar, A.J., Rutherford, D.W., Beyer, R.S., Wershaw, R.L., 2003. Environ-mental fate of roxarsone in poultry litter. I. Degradation of roxarsone duringcomposting. Environ. Sci. Technol. 37, 1509–1514.

Gasparatos, D., 2013. Sequestration of heavymetals from soil with Fe-Mn concretions andnodules. Environ. Chem. Lett. 11 (1), 1–9.

Goldberg, S., Glaubig, R., 1988. Anion sorption on a calcareous, montmorillonitic soil–ar-senic. Soil Sci. Soc. Am. J. 52, 1297–1300.

Guideline, Dutch, 2009. Soil Remediation Circular 2009. Ministry of Housing, Spatial Plan-ning and Environment Directorate-General for Environmental Protection.

Gupta, A., 1998. Postcolonial Developments: Agriculture in the Making of Modern India.Duke University Press, Durham, London.

I.G.M.E, 2000. Geological map of Greece 1:50.000. Eretria sheet. Institute of Geology andMineral Exploration.

I.G.M.E, 2002. Geological map of Greece 1:50.000. Kifissia sheet. Institute of Geology andMineral Exploration.

I.G.M.E, 2003. Geological map of Greece 1:50.000. Koropi–Plaka sheet. Institute of Geologyand Mineral Exploration.

Ioakim, C., Rondoyanni, T., Mettos, A., 2005. The Miocene basins of Greece (EasternMediterranean) from a paleoclimatic perspective. Rev. Paléobiol. 24 (2),735–748.

Jain, A., Loeppert, R.H., 2000. Effect of competing anions on the adsorption of arsenate andarsenite by ferrihydrite. J. Environ. Qual. 29, 1422–1430.

Jiang, Q.Q., Singh, B.R., 1994. Effect of different forms and sources of arsenic on crop yieldand arsenic concentration. Water Air Soil Pollut. 74, 321–343.

Kabata-Pendias, A., Mukherjee, A., 2007. Trace Elements from Soil to Human. Springer-Verlag, Berlin Heidelberg, New York, p. 550.

Kabata-Pendias, A., Pendias, H., 2001. Trace Elements in Soils and Plants. CRC Press, USA,p. 534.

Kampouroglou, E., Economou-Eliopoulos, M., 2013. Natural contamination by As andheavy metals in soil, their bio-accumulation and potential sources: the case of a trav-ertine limestone quarry, Greece. Cent. Eur. J. Geosci. 5 (1), 174–188.

Kampouroglou, E., Economou-Eliopoulos, M., 2016. Assessment of the environmen-tal impact by As and heavy metals in lacustrine travertine limestone and soil inAttica, Greece: mapping of potentially contaminated sites. Catena 139,137–166.

Kumaresan, M., Riyazuddin, P., 2001. Overview of speciation chemistry of arsenic. Curr.Sci. 80, 837–846.

Livesey, N.T., Huang, P.M., 1981. Adsorption of arsenate by soils and its relation to selectedchemical properties and anions. Soil Sci. 131, 88–94.

Mandal, B.K., Suzuki, K.T., 2002. Arsenic round the world: a review. Talanta 58, 201–235.

Mandal, B.K., Chowdhury, T.R., Samata, G., Mukherjee, D.P., Chanda, C.R., Saha, K.C.,Chankraborti, D., 1998. Impact of safe water for drinking and cooking on five ar-senic-affected families for 2 years in West Bengal, India. Sci. Total Environ. 218,185–201.

Manning, B.A., Goldberg, S., 1996. Modeling competitive adsorption of arsenate withphosphate and molybdate on oxide minerals. Soil Sci. Soc. Am. J. 60, 121–131.

Marinos, G.P., Petrascheck, W.E., 1956. Laurium. Institute for Geology and Subsurface Re-search, Athens, Greece.

McArthur, J.M., Ravenscroft, P., Safiulla, S., Thirlwall, M.F., 2001. Arsenic in groundwater:testing pollution mechanisms for sedimentary aquifers in Bangladesh. Water Resour.Res. 37, 109–117.

Meharg, A.A., 1994. Integrated tolerance mechanisms constitutive and adaptive plant re-sponses to elevated metal concentrations in the environment. Plant Cell Environ. 17,989–993.

Meharg, A.A., Hartley-Whitaker, J., 2002. Arsenic uptake and metabolism in arsenic resis-tant and nonresistant plant species. New Phytol. 154, 29–43.

Memon, J.R., Memon, S.Q., Bhanger, M.I., Ei-Turki, A., Hallam, K.R., Allen, G.C., 2009. Ba-nana peel: a green and economical sorbent for the selective removal of Cr(VI) fromindustrial wastewater. Colloids Surf. B. Biointerfaces 70 (2), 232–237.

Mench, M., Schwitzguébel, J.P., Schroeder, P., Bert, V., Gawronski, S., Gupta, S., 2009. As-sessment of successful experiments and limitations of phytotechnologies: contami-nant uptake, detoxification, and sequestration, and consequences to food safety.Environ. Sci. Pollut. Res. 16, 876–900.

Mettos, A., 1992. Geological and Paleogeographical Study of the Continental Neogene andQuaternary Formations of NE Attica and SE Boeotia. (PhD thesis). University of Ath-ens, Greece.

Mposkos, E., 2008. Neogene and Quaternary Sediments of the Northern Part of AthensBasin and the Broad Area of Kapandriti. Contribution to the Palaeogeographic Evolu-tion of Athens Basin from Late Miocene to Present 149. Mineral Wealth.

Navrotsky, A., Mazeina, L., Majzslan, J., 2008. Size-driven structural and thermodynamiccomplexity in iron oxides. Science 319, 1635.

Nickson, R.T., McArthur, J.M., Ravenscroft, P., Burgess, W.G., Ahmed, K.M., 2000. Mecha-nism of arsenic release to groundwater, Bangladesh andWest Bengal. Appl. Geochem.15, 403–413.

Nriagu, J.O., Pacyna, J.M., 1988. Quantitative assessment of worldwide contamination ofair, water and soils by trace metals. Nature 333, 134–139.

Pacyna, J.M., Scholtz, M.T., Li, Y.E., 1995. Global budget for trace metal sources. Environ.Rev. 3, 145–159.

Papanikolaou, D., Papanikolaou, I., 2007. Geological, geomorphological and tectonic struc-ture of the NE Attica and seismic hazard implications for the northern edge of theAthens plain. Bull. Geol. Soc. Greece 40, 425–438.

Partey, F.P., 2008. Mechanism of Arsenic Sorption onto Laterite Concretions. Ph.D.,NewMexico Institute of Mining and Technology. Department of Earth and Environ-mental Science (177 pp.).

Pavlopoulos, K., Chrisanthaki, I., Economou–Eliopoulos, M., Lekkas, S., 2011.Hydrochemical study of metals in the groundwater of the wider area of Koropi. Ad-vances in the Research of Aquatic Environment, pp. 169–176.

Pigna, M., Cozzolino, V., Caporale, A.G., Mora, M.L., Di Meo, V., Jara, A.A., Violante, A., 2010.Effect of phosphorus fertilization on arsenic uptake by wheat grown in polluted soils.J. Soil Sci. Plant Nutr. 10 (4), 428–442.

Pinney, N., Morgan, D., 2013. Thermodynamics of Al-substitution in Fe-oxyhydroxides.Geochim. Cosmochim. Acta 120, 513–530.

Roman-Ross, G., Cuello, G.J., Turrillas, X., Fernandez-Martinez, A., Charlet, L., 2006. Arse-nite sorption and co-precipitation with calcite. Chem. Geol. 233, 328–336.

Romero, F.M., Armienta, M.A., Carrillo, C.A., 2004. Arsenic sorption by carbonate-rich aqui-fer material, a control on arsenicmobility at Zimapan.Mexico. Arch. Environ. Contam.Toxicol. 47, 1–13.

Rose,W., Hawkes, E., Webb, S., 1979. Geochemistry inMineral Exploration. second ed. Ac-ademic Press.

Roychowdhury, T., Uchino, T., Tokunaga, H., Ando, M., 2002. Survey of arsenic in foodcomposites from arsenic affected area of West Bengal, India. Food Chem. Toxicol.40, 1611–1621.

Samal, A.C., 2005. An Investigation on Accumulation of Arsenic in Ecosystem of GangeticWest Bengal and Assessment of Potential Health Risk. (Ph. D. Thesis). University ofKalyani, West Bengal, India.

Skarpelis, N., 2007. The Lavrion deposit (SE Attica, Greece): geology, mineralogy andminor element chemistry. Neues Jb. Mineral. Abh. 183 (3), 227–249.

Slejkovec, Z., Kanduc, T., 2005. Unexpected arsenic compounds in low-rank coals. Environ.Sci. Technol. 39, 3450–3454.

Smith, A.H., Lopipero, P.A., Bates, M.N., Steinmaus, C.M., 2002. Arsenic epidemiology anddrinking water standards. Science 296, 2145–2146.

Stamatis, G., Lambrakis, N., Alexakis, D., Zagana, E., 2006. Groundwater quality inMesogeabasin in eastern Attica (Greece). Hydrol. Process. 20, 2803–2818.

Stamatis, G., Alexakis, D., Gamvroula, D., Migiros, G., 2011. Groundwater quality as-sessment in Oropos-Kalamos basin, Attica, Greece. Environ. Earth Sci. 64 (4),973–988.

Swartz, C.H., Blute, N.K., Badruzzman, B., Ali, A., Brabander, D., Jay, J., Besancon, J., Islam, S.,Hemond, H.F., Harvey, C., 2004. Mobility of arsenic I a Bangladesh aquifer: inferencesfrom geochemical profiles, leaching data, and mineralogical characterization.Geochim. Cosmochim. Acta 68, 4539–4557.

Tian, K., Huang, B., Xing, Z., Hu,W., 2017. Geochemical baseline establishment and ecolog-ical risk evaluation of heavy metals in greenhouse soils from Dongtai, China. Ecol.Indic. 72, 510–520.

U.S. EPA, 1984. Health Assessment Document for Inorganic Arsenic. EPA-600/8-83-021F. Office of Health and Environmental Assessment, Office of Research andDevelopment.

222 E.E. Kampouroglou, M. Economou-Eliopoulos / Catena 150 (2017) 206–222

U.S. EPA, 2001. National Primary DrinkingWater Regulations; arsenic and clarifications tocompliance and new source contaminants monitoring; final rule. EPA-815-Z-01-001.Fed. Regist. 66, 6976–7066.

Vamerali, T., Bandiera, M., Mosca, G., 2010. Field crops for phytoremediation of metal-contaminated land. A review. Environ. Chem. Lett. 8, 1–17.

Verbruggen, N., Hermans, C., Schat, H., 2009. Molecular mechanisms of metalhyperaccumulation in plants. New Phytol. 181, 759–776.

Voudouris, P., Melfos, V., Spry, P.G., Bonsall, T., Tarkian, M., Economou-Eliopoulos, M.,2008. Mineralogy and fluid inclusion constraints on the evolution of the Plaka intru-sion-related ore system, Lavrion, Greece. Mineral. Petrol. 93, 79–110.

W.H.O, 2004. Some drinking-water desinfectants and contaminants, including arsenic.IARC Monogr. Eval. Carcinog. Risks Hum. 84.

Wadge, A., Hutton, M., 1987. The leachability and chemical speciation of selected trace el-ements in fly ash from coal combustion and refuse incineration. Environ. Pollut. 48,85–99.

Wang, S., Mulligan, C.N., 2006. Effect of natural organic matter on arsenic releasefrom soils and sediments to groundwater. Environ. Geochem. Health 28,197–214.

Wang, J.R., Zhao, F.J., Meharg, A.A., Raab, A., Feldmann, J., McGrath, S.P., 2002. Mechanismsof arsenic hyperaccumulation in Pteris vittata. Uptake kinetics, interactions withphosphate, and arsenic speciation. Plant Physiol. 130, 1552–1561.

Williams, P.N., Price, A.H., Raab, A., Hossain, S.A., Feldmann, J., Meharg, A.A., 2005. Varia-tion in arsenic speciation and concentration in paddy rice related to dietary exposure.Environ. Sci. Technol. 39, 5531–5540.

Zavala, Y.J., Duxbury, J.M., 2008. Arsenic in rice: estimating normal levels of total arsenicin rice grain. Environ. Sci. Technol. 42, 3856–3860.

Zhou, X., Yoshida, K., Kuroda, K., Endo, Y., Endo, G., 2003. Effects of cyst on the cytotoxicityof arsenic compounds. Arch. Environ. Contam. Toxicol. 45, 324–330.