Geoderma 154 (2010) 311–320

Contents lists available at ScienceDirect

Geoderma

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

Transport of Cd, Cu, Pb and Zn in a calcareous soil under wheat and safflowercultivation— A column study

Gholamabbas Sayyad a,⁎, Majid Afyuni b, Sayed-Farhad Mousavi b, Karim C. Abbaspour c,Brian K. Richards d, Rainer Schulin e

a Department of Soil Science, College of Agriculture, Shahid Chamran University of Ahvaz, Ahvaz, Iranb Departments of Soil Science and Water Engineering, College of Agriculture, Isfahan University of Technology, Isfahan 84156-8111, Iranc Swiss Federal Institute for Environmental Science and Technology, EAWAG, Ueberlandstrasse 133 P.O. Box 611, 8600 Dübendorf, Switzerlandd Department of Biological and Environmental Engineering, Riley-Robb Hall, Cornell University, Ithaca, NY 14853-5701, USAe Institute of Terrestrial Ecosystems, ETH Zurich, Universitatstr. 16, CH-8092 Zurich, Switzerland

Abbreviations: AAS, atomic absorption spectrometrypenta-acetic acid; HM, heavy metal; ICP, inductivelapplication; S, planted safflower columns; Sf, fallowdomain reflectometry; W, planted wheat columns; Wf,⁎ Corresponding author. Tel.: +98 611 3364054; fax:

E-mail address: gsayyad@gmail.com (G. Sayyad).

0016-7061/$ – see front matter © 2009 Elsevier B.V. Aldoi:10.1016/j.geoderma.2009.10.019

a b s t r a c t

a r t i c l e i n f o

Article history:Received 23 April 2009Received in revised form 21 October 2009Accepted 28 October 2009Available online 3 December 2009

Keywords:Heavy metal leachingCalcareous soilSafflowerWheatMetal solubilityUndisturbed soil columns

Anthropogenic release of heavy metals (HMs) into the environment has resulted in a continuous buildup ofHMs in agricultural soils. On the one hand, uptake of HMs by crop plants may lead to food chain transfer tohumans, and on the other, leaching of HMs with deep seepage may cause groundwater contamination. Theobjectives of this study were to assess the mobility of cadmium (Cd), copper (Cu), lead (Pb) and zinc (Zn)under two common crop plants with different rooting systems, wheat (Triticum aestivum) with fibrous roots,and safflower (Carthamus tinctorious) with a taproot system in a calcareous silty clay loam soil. The studywas conducted with 50 cm long undisturbed soil columns (Typic Haplocalcid) extracted from a wheat and asafflower field located in the same unit of soil type. The top 10 cm of half of the columns were artificiallycontaminated with Cd, Cu, Pb, and Zn at concentrations of 15, 585, 117 and 1094 mg kg−1, respectively. Halfof the contaminated and uncontaminated columns were planted with wheat and the other half withsafflower according to their previous cultivation history. Discharge was collected continuously. After thecrops were harvested, soil samples were collected at 10-cm intervals and analyzed for HNO3- and DTPA-extractable metals. Plants enhanced the displacement of metals into the subsoil. Safflower resulted in largerconcentrations at deeper depths than wheat. Metal concentrations were significantly (Pb0.05) higher indischarge from contaminated than from uncontaminated columns. Because discharge was reduced in thepresence of plants due to transpiration, total amounts of metals leached were in general larger in fallow thantreatments with plants. The only exception was the leaching of Zn from wheat columns, which was largerunder plants than under fallow. Plants did not only affect metal transport velocity of mobile fraction throughactual root activity, but also through effects on soil properties during previous cultivation history, as therewere clear differences in metal leaching between wheat and safflower soils from fallow columns. These weremetal-dependent, indicating that these effects were not only related to different effects on soil structure bythe root systems, but also to chemical soil properties.

; DTPA, di-ethylene-triamine-y coupled plasma; M, metalsafflower columns; TDR, timefallow wheat columns.+98 611 3330079.

l rights reserved.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Soil pollution by heavy metals (HMs) continues to create seriousenvironmental risks (Hinz and Selim, 1994; Richards et al., 1998;Schwab et al. 2002). Industrial and agricultural activities havereleased large amounts of HMs into soils (Li and Li, 2001). The metalsadded to the soil may eventually end up in the food chain or cause

contamination of groundwater resources. In Isfahan, located in centralIran, for example, a steel factory and other factories release theirwastewater into the surrounding lowland areas where groundwaterlevels are less than 2 m (Shirani, 1996). Also, large-scale applicationsof sewage sludge and fertilizers on agricultural lands can be importantsources of soil contamination by HMs (Amini et al., 2005).

The mobility of HMs in soils has been investigated for severaldecades (Giordano and Mortvedt, 1976; Richards et al., 2000). Manyresearchers have concluded that there is little evidence for HMleaching through the soil into the groundwater (Chang et al., 1984;Higgins, 1984; Williams et al., 1987). However, most of these studieswere conducted using columns of homogenized soil (Giordano andMortvedt, 1976; Emmerich et al., 1982) and often the fact that nosignificant increase in HM concentrations below the contaminated

Table 1Average and standard deviations of soil chemical and physical properties.

Wheat farm Safflower farm

Depth (m) 0–0.25 0.25–0.45 0.45–0.60 0–0.25 0.25–0.45 0.45–0.60pH 7.7±0.1 7.9±0.08 7.9±0.06 7.6±0.05 7.7±0.1 7.8±0.05EC (dS m−1) 10.5±2.45 6.6±0.68 6.5±0.48 9.8±1.64 7.8±1.99 5.9±2.25CaCO3 (%) 38.2±0.85 35.6±0.92 37.6±1.11 37.2±1.68 35.9±0.5 36.2±1.42Cd (mg kg−1) 1.6±0.60 1.7±0.50 1.7±0.34 1.6±0.52 1.7±0.34 1.6±0.50Cu (mg kg−1) 33.4±5.27 28.5±1.71 27.1±2.1 31±1.97 29.5±1.66 28.1±1.56Pb (mg kg−1) 20.3±2.52 16.5±0.70 15.3±1.10 19.3±1.51 15.5±0.90 14.1±1.12Zn (mg kg−1) 64.5±5.10 59.1±4.90 58.1±3.10 64.3±5.51 62.1±5.05 57.3±3.2Cl (%) 2.4±0.51 1.6±0.27 1.6±0.17 2.1±0.51 1.6±0.29 1.1±0.20CEC (cmol kg−1) 14.8±0.83 13.7±0.44 14.6±0.66 13.8±0.71 14.6±0.47 14.3±0.80Organic matter. (%) 1.0±0.08 0.7±0.10 0.7±0.06 0.5±0.03 0.3±0.08 0.4±0.03Sand (g kg−1) 177.0±14.1 172.0±21.0 177.0±14.0 179.0±15.2 171.0±11.2 182.0±5.1Silt (g kg−1) 434.0±23.2 425.0±21.4 431.0±31.1 431.0±21.2 424.0±12.1 433.0±22.Clay (g kg−1) 389.0±21.0 403.0±32.0 392.0±31.0 390.0±23.0 405.0±31.0 385.0±12.0Bulk density (Mg m−3) 1.3±0.13 1.4±0.08 1.4±0.02 1.32±0.15 1.43±0.09 1.45±0.04

Average of three replications±standard deviation.

312 G. Sayyad et al. / Geoderma 154 (2010) 311–320

topsoil was found was considered as evidence that the metals wereimmobile (Chang et al., 1984; Dowdy et al., 1991). In many of thesestudies up to 40–50% of the applied metals were not recovered. Thereasons for poor recoveries were attributed to tillage dispersion(Williams et al., 1987), incomplete analytical recovery and changes inbulk density (Chang et al., 1984). Another initially overlookedpossibility is that metals can be leached by preferential flow andtransport processes through macro-structures such as cracks, worm-holes, plant root channels and other pathways of high hydraulicconductivity. Preferential flow can greatly increase the mobility andvelocity of heavy metals and solute transport through the soil(Camobreco et al., 1996; Richards et al., 1998; Schwab et al., 2002).The effects of preferential flow can be particularly dramatic in claysoils that form cracks upon drying, which increase their hydraulicconductivity by several orders of magnitude while the matrix remainsalmost impermeable (Topp and Davis, 1981). Another factor that canenhance metal mobility is binding with organic ligands or colloids,particularly at near-neutral and high pH values (Camobreco et al.,1996; Richards et al., 1997; McBride, 1998).

Vegetation is another important factor influencing the mobility ofmetals in soil, directly as well as indirectly (Meek et al., 1990; Caronet al., 1996; Madrid et al., 2003; Banks et al., 2006; Caetano et al.,2008; Fernandez et al., 2008). Plants may increase metal mobilitythrough the formation of preferential pathways along root channels orthe complexation of metals with root exudates in the rhizosphere.They may also retard metal leaching through reducing deep seepageby taking up water, adsorption of metals to root surfaces, plant uptakeof metals, and stimulatedmicrobial immobilization in the rhizosphere(Tyler and McBride, 1982; Schwab et al., 2002).

Table 2Freundlich isotherm parameters obtained by batch experiment for different depths.

MetalDepth (cm) kF (mgmetal(1−β)

(cm3)β mgsoil−1)β (−) r2a

Wb S W S W S

Cd0–15 20.4 18.6 0.49 0.58 0.83 0.89Cd15–30 24.6 21.8 0.50 0.49 0.92 0.89Cd30–50 29.5 28.8 0.59 0.51 0.99 0.93Cu0–15 48.9 53.87 0.49 0.96 0.98 0.99Cu15–30 38.0 50.1 0.45 0.37 0.86 0.98Cu30–50 69.2 60.2 0.65 0.83 0.79 0.80Pb0–15 49.8 49.7 1.00 1.00 0.98 0.87Pb15–30 49.8 49.7 0.99 1.00 0.95 0.89Pb30–50 49.8 49.8 1.00 1.00 0.94 0.96Zn0–15 29.1 27.3 1.09 1.11 0.92 0.99Zn15–30 39.8 36.3 0.45 0.27 0.80 0.99Zn30–50 47.8 39.0 0.43 0.33 0.89 0.88

a r 2: coefficient of determination.b W and S stand for “wheat soil” and “safflower soil”, respectively.

The information about HM transport under plants with differentrooting systems is limited, especially in soils with high pH and claycontent, in which cationic metals are generally considered to beimmobile. The objective of this study was to investigate the mobilityof Cd, Cu, Pb and Zn under two crops with different rooting systems,i.e., wheat with a fibrous root system and safflower with a taprootsystem in a typical calcareous soil of central Iran. For this purpose, weperformed a greenhouse experiment with undisturbed soil columns.In order to separate direct plant effects on metal mobility fromindirect effects exerted via the influence of plants on soil structure, thesoils were taken from two fields with the same type of soil, butdifferent previous cultivation histories, i.e. a field that had beencultivated with wheat and another with safflower. Half of the columnswere kept fallow, while the other columns were planted with tworespective crops. This study is an initial assessment in which a widerange of measurements were made in hopes that they show whatfactors were associated with metal transport.

2. Materials and methods

The study was conducted using 24 undisturbed soil columns of22.5 cm in diameter and 50 cm in depth in a greenhouse of IsfahanUniversity of Technology. The soils—fine, mixed, thermic, TypicHaplocalcid—(Soil Survey Staff, 2003) were sampled from the sameunit of soil type on two nearby fields at Kabotarabad Research Stationof Isfahan Agricultural Research Center, 40 km southeast of Isfahan,central Iran (55° 50′ 16″ to 55° 50′ 26″ longitude and 32° 30′ 14″ to32° 30′ 44″ latitude) with an elevation of 1750 m and mean annualprecipitation of 145 mm. One field had been cultivated with safflower

Table 3Water balance for different treatments.

Treatment Soil storage change(mm)

E(mm)

Discharge(mm)

Wa −12.7±5.1bcb 382.0±3.7a 193.0±8.8cWf 7.0±5.6a 260.0±8.1c 335.0±13.6aW+M −7.0±5.7b 336.0±20.2b 245.0±14.3bWf+M 10.1±6.1a 267.0±33.1c 332.0±29.5aS −17.0±8.4c 391.0±11.4a 180.0±19.8cSf 5.0±3.0a 275.0±2.3c 318.0±1.5aS+M −9.0±7.0b 367.0±5.9a 212.0±12.9cSf+M 10.2±1.3a 273.0±3.1c 325.0±5.8aTotal irrigation=588.5 mm

Average of three replications±standard deviation.a W: planted wheat columns; Wf: fallow wheat columns; S: planted safflower

columns; Sf: fallow safflower columns; M: metal application; E: evaporation for fallowand evapotranspiration for planted columns.

b In each column means with similar letters are not significantly different at 0.05level according to the LSD test.

Fig. 1. Cumulative irrigation and discharge for different wheat (a) and safflower(b) treatments for the whole period. W: planted wheat columns; Wf: fallow wheatcolumns, S: planted safflower columns; Sf: fallow safflower columns,M:metal application.

313G. Sayyad et al. / Geoderma 154 (2010) 311–320

and the other with wheat in the previous year. The crop rotation onthe wheat field was wheat–fallow–wheat–wheat and wheat–wheat–fallow–safflower on the safflower field in the 4 years before sampling.For brevity, we simply refer to the first soil as “wheat soil” and to thesecond as “safflower soil”. Continuous macropores of 0.5–3 cm widthwere seen to extend from 25 cm to 60 cm depth in the field profiles.For general characterization of physical and chemical soil properties,soil samples were taken from soil profiles (0–60 cm), air dried,crushed and sieved to b2 mm. Soil pH was measured in saturatedpaste using a Metrohm 828 pH-meter (Herisau, Switzerland).Electrical conductivity (EC) was measured in saturated paste extractusing a Metrohm 644 conductometer (Herisau, Switzerland). Limecontent as the total neutralizing value (TNV) was determined bytitration with NaOH. Cd, Cu, Pb and Zn were extracted with 4 M HNO3

(Sposito et al., 1981) at 70 °C for 12 h. Chloride was determined bytitration with AgNO3, cation exchange capacity (CEC) by the CH3-

COONa method (Sumner and Miller, 1996), organic matter content(OM) by the wet oxidation method (Walkley and Black, 1934), andsoil texture (sand, silt and clay) by the hydrometer method (Gee andBauder, 1986). Bulk density was determined from separately collectedsoil cores (7 cm in diameter and 6 cm in depth) (Blake and Hartge,1986). A batch study was performed on soil samples collected fromeach different depth (0–15, 15–30 and 30–50 cm) of soil profile. Afterhomogenization, 2 g of soil was equilibrated with 20 ml of metalat concentrations of 0.05, 0.1, 0.5, 1, 3, 7, 1, 15, 40, 70, 100, and250 mg l−1 prepared in 0.01 M CaCl2 as background solution.The suspensions were shaken end-over-end for 24 h to ensureequilibrium between soil and solute. The suspension pH wasmeasured after separation of the solute after centrifuging for 15 minat 5000 RPM. Thereafter, the total concentration of metal wasmeasured in the supernatant using an atomic absorption spectrosco-py. Three replicate samples were obtained by repeating the sameprocedure from the start. The amount of sorbed metal was calculatedfrom the difference between the initial and equilibrium concentrationof metal. Sorption of metal in the soils was modelled with Freundlichisotherm expressed as:

s = kFcβ ð1Þ

where kF [ML−3]−β andβ [−] are empirical constants, s [−] is adsorbedmetal concentration and c [ML−3] is concentration in soil solution.

A randomized complete block design with three replications wasconducted. Twelve undisturbed soil columns were taken from eachfield. To prepare the soil columns, two long trenches were dug in eachfield. Polyethylene tubes were driven (by pressure) stepwise into thesoil to a final depth of 50 cm and the surrounding soil was removedafter each step. In the greenhouse (during the experiment averagedaily minimum temperature, maximum temperature and panevaporation were 12.4 °C and 33.2 °C and 5.2 mm, respectively)each column was placed on a support and equipped with a plasticfunnel to collect the discharge. To eliminate any bias that could arisefrom effects of the location of the columns in the greenhouse, thegreenhouse was divided into three areas and in each area one of thethree replicate columns of each treatment was placed at a randomlyassigned location. The upper 10 cm of soil in the half of the columns ofeach field was contaminated with a solution containing CdCl2(19.5 mgCd kg−1dry soil), CuSO4 (750 mgCu kg−1 dry soil), Pb(NO3)2 (150mgPb kg −1 dry soil), and ZnCl2 (1400 mgZn kg −1 drysoil) by spraying and completely mixing it. These concentrationsrepresent half of the maximum permitted cumulative metal loadingsin biosolids applied to soils according to current USEPA Part 503regulations (USEPA, 1993). The topsoils (0–10 cm) of the other half ofthe columns were completely mixed after applying the same amountof distilled water (i.e. without metal).

During 2 weeks after metal application, for normal plant growth,the columns were irrigated until discharge EC decreased around

2dS m−1. Thereafter the columns were fertilized at rates of60 mgN kg−1 and 40 mgK kg−1 by addition of urea and potassiumnitrate at the beginning of the experiment. No P was applied, as theinitial soil analysis indicated sufficient available P for plant growth.After 2 weeks, three of the contaminated wheat soil columns weresown with wheat (W+M), and three of the safflower soil columnswere similarly sown with safflower (S+M), while the remaining sixcolumns were left fallow (Wf+M and Sf+M). Plants were seededmanually on 30March 2003 at a density of 200 seeds perm2 for wheatand 25 seeds perm2 for safflower. Similarly, in the 12 uncontaminatedcolumns, three wheat soil columns were sown with wheat (W), threesafflower soil columns with safflower (S), and the remaining sixcolumnswere left fallow (Wf and Sf). Thus, the crop plantswere chosenin accordance with the previous cultivation history of the reported soil,i.e., columns from the wheat field were planted with wheat, andcolumns from the safflower field were planted with safflower.

After seeding, water content in columns was measured usinghorizontal TDR probes (6050X1 Trase System, Soil moisture Equip-ment Corp., Santa Barbara, California, USA) placed at 15, 30 and 45 cmdepths. Irrigation was applied using a scaled cylinder. The irrigationwas based on 65% depletion of the available soil water capacity. Theapplied rate of irrigation was calculated to saturate available soilwater capacity which had been determined by measuring moisturerelease at 10, 30, 50, 100, 300, 500, 1000 and 1500 kPa (Klute, 1986).

Discharge was collected and sampled 1–2 days after irrigations, bywhich time all percolation had ceased. Water balance for the durationof the experiment was calculated using the following expression:

E = I − D + ΔS ð2Þ

314 G. Sayyad et al. / Geoderma 154 (2010) 311–320

where E is the actual evaporation for unplanted columns and actualevapotranspiration for planted columns (mm), I is irrigation (mm), Dis discharge (mm), and ΔS is the change in column water storage(mm) which calculated using TDR measurements during theexperiment. If I, D and ΔS were known, then E was calculated. Thechange in storage ΔS was determined by the difference in total waterat two time points (the first at the beginning of the experiment, thesecond at the end). For calculating total water content, the measuredvolumetric water content (cm3 cm−3) in unit area (cm water cm−1

soil) in each depth was multiplied by the respective soil depth (cm),resulting in water depth; the summation of water depths in threedepth zones resulted in total water content in column. For example,for the TDR probes at 15 cm depth, the respective effective depth ofwater measurement was assumed to be 0–20 cm. Respectively, depthzones for the TDR probes at 30 and 45 cm were 20–40 and 40–50 cm.

Plants were harvested after 75 days at the end of the pollinationperiod. After the plants had been harvested, the soil columns were cutinto10-cmsections, air dried, crushedand sieved tob2 mm. Sub-sampleswere collected from each section and analyzed for HNO3-extractable(Sposito et al., 1981) anddi-ethylene-triamine-penta-acetic acid (DTPA)-extractable metal concentrations (Soltanpour, 1991). HNO3 extraction

Fig. 2. HNO3-(left) and DTPA-extractable (right)

was carried out by heating a mixture of 9.6 g soil and 60 ml of 4 MHNO3

overnight at 70 °C. Themixture was subsequently centrifuged for 15 minand filtrated using No. 42 Whatman filter paper and analyzed for heavymetals. Total heavymetal accumulation in soil can bemonitored throughacid (e.g., HNO3) soil extractions. Evaluation of total metal levels may beuseful as a global index of contamination, but it provides little indicationof their specific bioavailability, mobility, and reactivity in soils (Cao et al.,1984; Walter et al., 2002). Another monitoring approach consists of theuse of various chemical reagents to estimate the fraction of the soil metalthat is potentially available to plants. The most frequently employedreagents are chelating agents such as DTPA. Metals extracted by DTPAlargely exist in exchangeable, organically complexed and carbonateforms. Such extraction provides more information about metalavailability and tends to correlate with metal uptake by plants (Hoodaand Alloway, 1994). The absolute quantity of metals extracted by DTPAis a usefulmeans of evaluating plant-availablemetals in the soil (Walteret al., 2002).

Metal concentrations in the extracts were analyzed using atomicabsorption spectrometry or inductively coupled plasma spectrometery(ICP-OES, Vista-MPXVarian,Australia). Quality control samples includednumber 962 sandy clay from the Netherlands. Good agreements were

metal concentrations in fallow treatments.

315G. Sayyad et al. / Geoderma 154 (2010) 311–320

achieved between the data obtained and the certified values with thesame extraction protocols, i.e. HNO3 and DTPA.

Statistical analyses were conducted using SAS software (SASInstitute, 1993).Multiple comparisons of variableswere used to performLSD's (least significant difference) separation of means. A probabilitylevel of P≤0.05 was chosen to establish statistical significance.

3. Results and discussion

3.1. Soil properties

Soils had large cation exchange capacity (CEC) and high pH values(Table 1). The organic matter content was generally less than 1%. Thelarge CaCO3 content imparts a large pH buffering capacity. The highpH means that metal solubility was low and thus little metalmovement was to be expected in dissolved ionic form (Plassardet al., 2000). The dominant anion was Cl−, which can form solublecomplexes with Cd (Smolders et al., 1998; Khoshgoftar et al., 2004),and thus increases the mobility of this metal.

Fig. 3. HNO3-(left) and DTPA-extractable (right)

The soils from the two fields did not differ in their basic chemicaland physical properties as listed in Table 1, aside from slightly greaterorganic matter content in the wheat soil. Background metalconcentrations in soil profiles showed that the soil was not polluted(Table 1). Average total Cd, Cu, Pb and Zn concentrations were 1.7,29.6, 16.8 and 60.9 mg kg−1, respectively, which are lower than themaximum acceptable concentrations (MAC) of trace metals inagricultural soils suggested by most countries (Kabata-Pendias andPendias, 1997).

3.2. Metal sorption

Sorption constants and exponents of the Freundlich isothermEq. (1) were obtained using nonlinear least squares method and arelisted in Table 2 for three different soil horizons. The greater claycontent, pH, and CEC in deeper depths may be responsible for thelarger measured kF value. The kF values ranged in “wheat farm soil”from a minimum of 20.4 (mgmetal

(1−β) (cm3)β mgsoil−1) for Cd in 0–15 cmdepth to a maximum of 69.2 (mgmetal

(1−β) (cm3)β mgsoil−1) for Cu in 30–50 cm depth. The minimum and maximum values for “safflower soil”

metal concentrations in planted treatments.

316 G. Sayyad et al. / Geoderma 154 (2010) 311–320

were 18.6 and 60.2 (mgmetal(1−β) (cm3)β mgsoil−1) for Cd in 0–15 cm and Cu

in 30–50 cm depths, respectively. The average kF values for Cd, Cu, Pband Zn in wheat soil were 24.8, 52.0, 49.8, and 38.9 (mgmetal

(1−β) (cm3)β

mgsoil−1) and for safflower soil were 23.1, 54.7, 49.7, and 34.2 (mgmetal(1−β)

(cm3)β mgsoil−1) respectively. As the values show, there were nosignificant differences in metal retention between the two soils. Theorder of metal adsorption to soil particles according to their kF valueswere CdbZnbPbb∼Cu. The isotherms were nonlinear, with theaverage exponent b values of 0.53, 0.53, 1.00, and 0.45 for Cd, Cu, Pband Zn in wheat soil, respectively. The average b values for metalsisotherms in safflower soil were 0.5, 0.7, 1.0, and 0.3 for Cd, Cu, Pb andZn, respectively. These values are comparable with other investiga-tions of similar soils (Harter, 1982; Buchter et al., 1989; Holm et al.,2003; Moradi et al., 2005).

3.3. Effect of contamination on evapotranspiration and shoot drybiomass in different treatments

During the experiment, each soil column received a total of588.5 mm irrigation water. Unsuprisingly, discharge from unplantedcolumns was in general larger than evaporation (Table 3) while theopposite was true for the planted columns. On average, about 35% ofthe applied irrigation water was collected as discharge from the

Fig. 4. Differences between HNO3-(left) and DTPA-extractable (ri

planted columns and 56% from the unplanted columns. The dischargerates varied significantly between different treatments. The differenceevolved and increased with time as the plants developed andconsumed more water (Fig. 1). Although irrigation was generallyincreased during the growing period with increasing demand, therewere short periods in which no discharge occurred from plantedcolumns, in contrast to unplanted columns.

Wheat and safflower cultivars are reported to consume water atsimilar rates (Doorenbos and Pruitt, 1977). As Table 3 shows, that wasalso the case in our study. However, under metal stress, transpirationof wheat was much more reduced than that of safflower. Wheatshowed a 12% decrease in water consumption while the decrease forsafflower was only 6%, indicating that safflower was more tolerant toHM stress thanwheat. This conclusion is confirmed by the effect of themetal treatment on plant biomass. Aboveground wheat biomass was50% (Pb0.05) lower on contaminated columns (mean 12.4 g plantcolumn−1) than on uncontaminated columns (mean 27.6 g plantcolumn−1), while safflower biomass (averaging 29.8 and 27.4 g onuncontaminated and contaminated columns, respectively) was notaffected significantly (Pb0.05).

The decrease inwheat yield (55%)was substantially greater than thedecrease in ET (12%), which suggests that thewater use efficiency of thewheat in the contaminated columns was severely diminished. Indeed,

ght) metal concentrations in planted and fallow treatments.

317G. Sayyad et al. / Geoderma 154 (2010) 311–320

normalizing the water use to the actual dry matter yield for eachtreatment indicates that uncontaminated wheat and all safflowertreatments had similar normalized water uses of circa 13 mm ET per gdry weight; in contrast, the contaminated wheat normalized water usewas much less efficient at over 27 mm per g dry weight (Table 3).

3.4. Metal concentration profiles

As one would expect, there were strong positive relationships(r2N0.90) between HNO3- and DTPA-extractable metal concentra-tions in the top 10 cm for all the treatments. The correlations weremuch weaker for the other depths (r2=0.60).

Metal concentration profiles did not differ significantly (Pb0.05)between unplanted wheat soil and safflower soil, neither in thepresence nor in the absence of the contamination (Fig. 2) except forZn in the subsoil of the uncontaminated columns. The HNO3- as wellas DTPA-extractable Zn concentrations were higher in the safflowersoil than in wheat soil. Lead concentrations were almost identical inthe two soils (mean of 11.00 mg kg−1). Only at 30–40 cm depth wasthe DTPA-extractable Pb concentration higher in contaminated“safflower soil” (Sf+M, mean 0.80 mg kg−1) than in contaminated“wheat soil” (Wf+M, mean 0.55 mg kg−1).

In topsoil of contaminated planted columns (Fig. 3) metal concen-trations (except for Pb) were greater in wheat soil than safflower soil.However, in uncontaminated planted columns there was no significantdifference (Pb0.05) between the two soils. The HNO3- as well as DTPA-extractable metal concentrations in subsoil (except 25 cm) of safflowersoil were greater than wheat soil, whereas at 25 cm depth, metalconcentrations (except for Zn) were greater in wheat soil.

For evaluating the effects of active rooting systemsonmetalmobilityand/or availability, Fig. 4 shows the difference in soil metal concentra-tions between the planted and fallow columns. Differences near zerowould imply that the presence of active rooting systemshadnoeffect onthe respective metal concentrations in the soil. In general, plantedcolumns had higher metal HNO3- and DTPA-extractable values thanfallow columns, as evidenced by the preponderance of differencesgreater than zero throughout the profiles. The fact that the “planted

Fig. 5. Ratio of DTPA-extractable metal concentration to HNO3-extractable metal concentrasafflower (S+M), and uncontaminated safflower (S) columns.

minus fallow” differences were greater for metal-applied columns atdepth suggests some transport of the appliedmetals. While both plantshad no significant effect onmetal transport in uncontaminated columns(thedifferences are aroundzero), thepresence of safflower causedmoremetalmobility in contaminated soil thandidwheat, especially at deeperdepths. Itmay be that this enhancedmobilitywas related to the fact thatsafflower has a tap root system and can produce more preferential flowthanwheatwhichhas afibrous root system. Songet al. (1998, 1999) alsofound similar resultswhen comparing the capability of a dual probeheatpulse technique formeasuring soil water content in the presence of tworooting systems, i.e. a fibrous rooted plant (tall fescue) and a tap-rootedplant (sunflower). Their results showed that fibrous rooted plantcompletely filled the columns, while the tap-rooted plant had areas inthe column where roots did not penetrate. Most of the branch rootsextended out from the crown of the axis near the soil surface and grewhorizontally to the edge of the column and then grew down the sides ofthe column and along the bottom. They concluded that water uptake inthe deeper depths of sunflower planted columns was due to branchroots.

The ratio of DTPA to HNO3-concentration is given in Fig. 5. All HMsshow much larger DTPA-extractable metal concentrations at the sur-face than at depth. Cadmiumhad the largest ratio in safflower, while Pbhad similar ratios in both safflower and wheat columns. As might beexpected, the ratios of DTPA- to HNO3-extractablemetal concentrationsin the contaminated columns (thick lines) were greater than in theuncontaminated columns. In the latter, the metals, representing onlythe background concentrations, had longer time to reach equilibriumbetween the solid phases and solution (Lim et al., 2002).

3.5. Relative metal mobility

For comparingmobility among differentmetals it was necessary tonormalize metal concentrations in the soil profiles because appliedrate of metals were different and metals with larger concentrationswill show larger concentrations (like Zn) in subsoil than metals withsmaller concentrations (like Cd). The differences between the HNO3-extractable concentrations of contaminated and uncontaminated

tion in the contaminated wheat (W+M), uncontaminated wheat (W), contaminated

Table 4Mean of heavy metal concentrations in discharge during the experiment (μg l−1).

Treatment Cd Cu Pb Zn

W 0.15±0.04ea 20.09±5.50cd 5.64±1.01c 103.43±15.35cdWf 0.13±0.04e 28.39±1.81cb 5.13±0.27c 53.72±19.96dW+M 1.08±0.02c 24.56±4.50cd 14.56±2.94b 235.98±22.07bWf+M 1.44±0.29b 31.69±4.46abc 23.95±2.25a 100.63±18.76dS 0.14±0.02e 35.88±1.90ab 5.55±0.35c 289.25±42.74bSf 0.34±0.07e 10.95±1.93d 4.50±0.10c 90.83±10.99dS+M 3.01±0.29a 44.54±0.46a 15.39±0.48b 370.07±92.57aSf+M 0.72±0.11d 21.04±2.69cd 5.77±1.67c 166.81±34.27c

Average of three replications±standard deviation.a In each column means with similar letters are not significantly different at 0.05

level according to the LSD test.

318 G. Sayyad et al. / Geoderma 154 (2010) 311–320

columns (background contamination) at respective depths weredivided by the applied rate of metals (Fig. 6) in order to normalizethem. Values near zero indicate that there was no treatment effect onmetal movement, i.e. metal concentrations in the soil profiles ofcontaminated and uncontaminated treatments were similar. Asexpected, Cd and Zn were much more mobile than Cu and Pb. Thisfinding is in agreement of our batch experiment (Table 2). For bothsoils, average Kd values of Cd (23.95 (mg(1−β)(cm3)β) and Zn (36.55(mg(1−β)(cm3)β) were smaller than Cu (53.35 (mg(1−β)(cm3)β) andPb (49.75 (mg(1−β)(cm3)β)), showingmore Cd and Znmobility in soilthan Cu and Pb. Camobreco et al. (1996) and McBride et al. (1997),among others, concluded the same order of relative mobility for thesemetals. In the presence of wheat plants, movement was generallymore limited to the upper parts of the columns than in the presence ofsafflower. However, except for Cu in the wheat columns, higheramounts of metal were displaced in planted (average values of 0.043,0.026, 0.023 and 0.036 (mg kg−1) for Cd, Cu, Pb and Zn, respectively)than in fallow columns (average values 0.023, 0.018, 0.001 and0.019 mg kg−1 for Cd, Cu, Pb and Zn, respectively), revealing anenhancing effect of the plants on metal movement. Except for Pb, thiseffect was stronger in the presence of safflower than in the presence ofwheat. In the case of Pb, the higher average displacement in the

Fig. 6. Normalized differences between HNO3-extractable metal con

presence of wheat is actually the result of one high concentrationvalue and thus may be due to an outlier value. The higher metalmobility in the presence of safflower could have been due thedifference in root systems. For Zn, there was also a slightly largermovement in safflower than in wheat soil under fallow conditions,which may have been due to residual root channels from the previouscultivation.

centrations in contaminated and uncontaminated treatments.

Table 5Total heavy metal mass exported by discharge (mg column−1).

Treatment Cd Cu Pb Zn

W 0.02±0.01da 4.91±0.85d 0.68±0.03f 14.22±2.69dWf 0.03±0.00d 6.98±0.58bc 1.26±0.11de 12.70±4.22dW+M 0.18±0.04c 8.06±2.85ab 2.53±0.45b 43.67±7.53aWf+M 0.43±0.11a 9.51±1.94a 7.08±0.21a 28.86±7.80bcS 0.01±0.0 d 1.71±0.33f 0.29±0.05g 13.33±0.60dSf 0.08±0.01d 2.63±1.00ef 1.07±0.25e 21.78±5.02cdS+M 0.32±0.03b 4.27±0.05de 1.66±0.01c 36.39±10.15abSf+M 0.18±0.01c 5.17±0.69cd 1.52±0.13cd 44.33±4.83a

Average of three replications±standard deviation.a In each column means with similar letters are not significantly different at 0.05

level according to the LSD test.

319G. Sayyad et al. / Geoderma 154 (2010) 311–320

3.6. Metal concentrations in discharge

Table 4 shows the mean concentrations of HMs in the discharge ofdifferent treatments. Substantial variation between replicate columnswith undisturbed subsoils can be expected due to variable macroporedistributions (Akhtar et al., 2003). Metal contamination of the topsoilsignificantly (Pb0.05) increased metal concentrations in the dis-charge for both planted and fallow treatments. In the contaminatedcolumns planted with wheat, Cd concentrations increased 7.2-fold,while Cu increased 1.2-fold, Pb increased 2.6-fold, and Zn increased2.3-fold in the discharge as the result of contamination thanrespective controls. For safflower these ratios were 21.5, 1.2, 2.8,and 1.3, respectively. The presence of safflower plants increasedmetalconcentrations in the discharge in comparison to fallow safflower soilcolumns for both contaminated and uncontaminated treatments(with the exception of Cd in uncontaminated columns for which theopposite effect was found). In contrast, metal concentrations indischarge from wheat columns were generally greater from soilwithout plants, with the notable exception of Zn. For safflower soil theplant effects on metal concentrations in the discharge, thus, generallyagreed well with the effects observed on metal mobility, whereas forwheat soil this was not the case (see preceding section).

In the presence of plants, higher Zn, Cu and Cd concentrationswere measured in the discharge of safflower than of wheat columns,

Fig. 7. Breakthrough curves of me

while Pb concentrations did not differ. In the absence of plants, onlyZn (90.83 μg l−1) and Cd (0.34 μg l−1) concentrations in uncontam-inated columns were significantly higher in discharge from safflowercolumns, whereas the opposite was true for Cu (28.39 μg l−1) and incontaminated columns also for Pb (23.95 μg l−1). This means that theplants also had a significant effect on metal mobility through theprevious cultivation and not only through their present activity. Asthese effects differed between metals, they could not have beensimply of a physical nature such as effects on soil structure.

Althoughmetal concentrations were in general higher in dischargefromplanted than fallow columns, the total amounts ofmetals leachedwere generally larger under fallow, as the discharge volumes weremuch smaller under plants, due to root water extraction (Table 5). Theonly exceptionswithmore leaching from planted than fallow columnswere Zn leaching from wheat soil and Cd leaching from safflower soil(Table 5). Breakthrough curves ofmetal showed that Cd and Cu are themostmobilemetal in the study since they appeared in the discharge inless pore volumes (i.e. b0.5) than the Pb and Zn (Fig. 7). Alsocomparing planted vs. fallow curves shows the effective plant effect onincreasing metal transport velocity of the mobile fraction especiallywith increasing plant growth inmore pore volumes. This effect ismoreobvious for Cu and Pb (Fig. 7). Maximum and minimum C/C0 ratios inplanted columns were for Zn (85×10−6) and Cu and Cd (35×10−6)while for fallow columns the maximum and minimum ratios were forZn (80×10−6) and Cd (25×10−6). Our results were in accordancewith the finding of Camobreco et al. (1996). Also other researchers(Tyler and McBride, 1982) reported that the relative metal mobilitywas as CdNZnNCuNPb.

4. Conclusions

This study showed that there can be significant leaching of heavymetal contamination even in calcareous soil. Plants increased solubleheavy metal concentrations, but reduced water discharge due to rootwater extraction. As a net effect, the total amounts of metal leachingwere in most cases decreased by the activity of the plants. Safflowergenerally enhanced metal solubility and transport velocity of themobile fraction more than wheat. This may be related partially to thedifferent types of root systems. The fact that metal transport velocity

tals for different treatments.

320 G. Sayyad et al. / Geoderma 154 (2010) 311–320

of mobile fraction also differed between fallow soils of differentcultivation historymeans that plants not only had an immediate effecton metal mobility through the activity of their roots, but also throughlonger lasting changes in soil properties. Because these influencesdiffered between metals, we conclude that these effects did not onlyrelate to soil structure, but also to chemical properties.

Acknowledgments

Wewould like to thank Prof. H. Flühler, Dr. H. Shariatmadari, Dr. M.A. Hajabbasi, and Dr. M. Khajehpour for their advice and also theirhelp in statistical analysis of the experiment results. Also we wouldlike to express appreciation to Rene Saladin and Anna Grunwald fortheir help in soil and plant analyses.

References

Akhtar, M.S., Steenhuis, T.S., Richards, B.K., McBride, M.B., 2003. Chloride and lithiumtransport in large arrays of undisturbed silt loam and sandy loam soil columns.Vadose Zone J. 2, 715–727.

Amini, M., Afyuni, M., Fathianpour, N., Khademi, H., Fluhler, H., 2005. Continuous soilpollution using fuzzy logic and spatial interpolations. Geoderma 124, 223–233.

Banks, M.K., Schwab, A.P., Henderson, C., 2006. Leaching and reduction of chromium insoil as affected by soil organic content and plants. Chemosphere 62, 255–264.

Blake, G.R., Hartge, K.H., 1986. Bulk density. In: Klute, A. (Ed.), Methods of Soil AnalysisPart 1, 2nd ed. Monogr. 9, ASA and SSSA, Madison, Winsconsin, pp. 363–375.

Buchter, B., Davidoff, B., Amacher, M.C., Hinz, C., Iskandar, I.K., Selim, H.M., 1989.Correlation of Freundlich Kd and n retention parameters with soils and elements.Soil Sci. 148, 370–379.

Caetano, M., Vale, C., Cesário, R., Fonseca, N., 2008. Evidence for preferential depths ofmetal retention in roots of salt marsh plants. Sci. Total Environ. 390, 466–474.

Camobreco, V.J., Richards, B.K., Steenhuis, T.S., Peverly, J.H., McBride, M.B., 1996.Movement of heavy metals through undisturbed and homogenized soil columns.Soil Sci. 161, 740–750.

Cao, H., Chang, A.C., Page, A.L., 1984. Heavy metal contents of sludge-treated soils asdetermined by three extraction procedures. J Environ. Qual. 13, 632–634.

Caron, J., Banton, O., Angers, D.A., Villeneuve, J.P., 1996. Preferential bromide transportthrough a clay loam under alfalfa and corn. Geoderma 69, 175–191.

Chang, A.C., Warneke, J.E., Page, A.L., Lund, L.J., 1984. Accumulation of heavy metals insewage sludge-treated soils. J. Environ. Qual. 13, 87–91.

Doorenbos, J., Pruitt, W.O., 1977. Crop Water Requirements, FAO Irrigation andDrainage Paper No. 24. United Nations Food and Agriculture Organization, Rome,Italy.

Dowdy, R.H., Latterell, J.J., Hinesly, T.D., Grossman, R.B., Sullivan, D.L., 1991. Trace metalmovement in an Aeric Ochraqualf following 14 years of annual sludge applications.J. Environ. Qual. 20, 119–123.

Emmerich, W.E., Lund, L.J., Page, A.L., Chang, A.C., 1982. Movement of heavy metals insewage sludge treated soils. J. Environ. Qual. 11, 174–178.

Fernandez, C., Monna, F., Labanowski, J., Loubet, M., van Oort, F., 2008. Anthropogeniclead distribution in soils under arable land and permanent grassland estimated byPb isotopic compositions. Environ. Pollut. 156, 1083–1091.

Gee, G.W., Bauder, J.W., 1986. Particle size analysis. In: Klute, A. (Ed.), Methods of SoilAnalysis, Part 1, 2nd ed., Agron. Monogr. 9., ASA and SSSA, Madison, Wisconsin,pp. 383–411.

Giordano, P.M., Mortvedt, J.J., 1976. Nitrogen effects on mobility and plant uptake ofsewage sludge applied to soil columns. J. Environ. Qual. 5, 165–168.

Harter, R.D., 1982. Competitive sorption of cobalt, copper and nickel ions by a calcium-saturated soil. Soil Sci. Soc. Am. J. 56, 444–449.

Higgins, A.J., 1984. Land application of sewage sludge with regard to cropping systemsand pollution potential. J. Environ. Qual. 13, 441–448.

Hinz, C., Selim, H.M., 1994. Transport of zinc and cadmium in soils: experimentalevidence and modeling approaches. Soil Sci. Soc. Am. J. 58, 1316–1327.

Holm, P.E., Rootzen, H., Borggard, O.K., Moberg, J.P., Christensen, T.H., 2003. Correlationof cadmium distribution coefficients to soil characteristics. J. Environ. Qual. 32,138–145.

Hooda, P.S., Alloway, B.J., 1994. The plant availability and DTPA extractability of tracemetals in sludge-amended soils. Sci. Total Environ. 149, 39–51.

Kabata-Pendias, A., Pendias, H., 1997. Trace Elements in Soils and Plants. CRC Press,Boca Raton, FL.

Khoshgoftar, A.H., Shariatmadari, H., Karimian, N., Kalbasi, M., van der Zee, S.E.A.T.M.,Parker, D.R., 2004. Salinity and zinc application effects on phytoavailability ofcadmium and zinc. Soil Sci. Soc. Am. J. 68, 1885–1889.

Klute, A., 1986. Water retention: laboratory methods. In: Klute, A. (Ed.), Methods ofSoil Analysis, Part 1, 2nd ed., Agron. Monogr. 9, ASA and SSSA, Madison,Wisconsin, pp. 635–662.

Li, L.Y., Li, F., 2001. Heavy metal sorption and hydraulic conductivity studies using threetypes of bentonite admixes. J. Environ. Eng. 127, 420–429.

Lim, T.T., Tay, J.H., Teh, C.I., 2002. Contamination time effect on lead and cadmiumfractionation in a tropical coastal clay. J. Environ. Qual. 31, 806–812.

Madrid, F., Liphadzi, M.S., Kirkham, M.B., 2003. Heavy metal displacement in chelate-irrigated soil during phytoremediation. J. Hydrol. 272, 107–119.

McBride, M.B., Richards, B.K., Steenhuis, T., Russo, J.J., Sauvé, S., 1997. Mobility andsolubility of toxic metals and nutrients in soil fifteen years after sludge application.Soil Sci. 162, 487–500.

McBride, M., 1998. Soluble trace metals in alkaline stabilized sludge products. J. Environ.Qual. 27, 578–584.

Moradi, A., Abbaspour, K.C., Afyuni, M., 2005. Modeling field-scale cadmium transportbelow the root zone of a sewage sludge amended soil in arid region of central Iran.J. Contamin. Hydrol. 79, 187–206.

Meek, B.D., DeTar,W.R., Rolph,D., Rechel, E.R., Carter, L.M., 1990. Infiltration rate as affectedby an alfalfa and no-till cotton cropping system. Soil Sci. Soc. Am. J. 54, 505–508.

Plassard, F., Winiarski, T., Petit-Ramel, M., 2000. Retention and distribution of threeheavy metals in a carbonated soil: comparison between batch and unsaturatedcolumn studies. J. Contamin. Hydrol. 42, 99–111.

Richards, B.K., Steenhuis, T.S., Peverly, J.H., Liebowitz, B.N., 1997. Effect of processingmodeon trace elements in dewatered sludge products. J. Environ. Qual. 26, 782–788.

Richards, B.K., Steenhuis, T.S., Peverly, J.H., McBride, M.B., 1998. Metal mobility at an old,heavily loaded sludge application site. Environ Pollut. 99, 365–377.

Richards, B.K., Steenhuis, T.S., Peverly, J.H., McBride, M.B., 2000. Effect of sludge-processing mode, soil texture and soil pH on metal mobility in undisturbed soilcolumns under accelerated loading. Environ. Pollut. 109, 327–346.

SAS Institute, 1993. SAS User Guide: Statistics. SAS Inst, Cary, NC.Schwab, A.P., Banks, M.K., Erickson, L.E., 2002. Fate and transport of heavy metals and

radionuclides in soil: The impact of vegetation [online], Available at http//www.engg.ksu.edu/HSRC/ fate.html.

Shirani Bidabadi, H., 1996. Investigation of morphological and chemical variability ofthe Zarinshahr soils and evaluation of their properties for sewage waterapplications (In Persian, with English abstract). Master of Science thesis, Collegeof Agriculture, Isfahan University of Technology, Isfahan, Iran, pp. 276.

Smolders, E., Lambergts, R.M., McLaughlin, M.J., Tiller, K.G., 1998. Effect of soil solutionchloride on cadmium availability to Swiss chard. J. Environ. Qual. 27, 426–431.

Soil Survey staff, 2003. Keys to Soil Taxonomy. United States Department of Agriculture,Natural Resources Conservation Service, 9th edition. , pp. 103–127.

Soltanpour, P.N., 1991. Determination of nutrient availability and element toxicity byAB-DTPA soil test and ICPS. Adv. Soil Sci. 16, 165–190.

Song, Y., Ham, J.M., Kirkham, M.B., Kluitenberg, G.J., 1998. Measuring soil water contentunder turfgrass using the dual-probe heat-pulse technique. Am. Soc. Hort. Sci. 123,937–941.

Song, Y., Kirkham, M.B., Ham, J.M., Kluitenberg, G.J., 1999. Dual probe heat pulsetechniqe for measuring soil water content and sunflower water uptake. Soil tillage50, 345–348.

Sposito, G., Lund, L.J., Chang, A.C., 1981. Trace metal chemistry in arid-zone field soilsamended with sewage sludge: I. Fractionation of Ni, Cu, Zn, Cd and Pb in solidphases. Soil Sci. Soc. Am. J. 46, 260–264.

Sumner, M.E., Miller, W.P., 1996. Cation exchange capacity and exchange coefficients.In: Sparks, D.L. (Ed.), Methods of Soil Analysis, Part 1, Agron. Monogr. 9, ASA andSSSA, Madison, Wisconsin, pp. 1201–1229.

Topp, G.C., Davis, J.L., 1981. Detecting infiltration of water through soil cracks by timedomain reflectometery. Geoderma 26, 13–23.

Tyler, L.D., McBride, M.B., 1982. Mobility and extractability of cadmium, copper, nickeland zinc in organic and mineral soil columns. Soil Sci. 134, 198–205.

U.S. Environmental Protection Agency, 1993. Clean Water Act, Section 503, Vol. 58, No.32. USEPA, Washington, DC.

Walkley, A., Black, C.A., 1934. An examination of the Degtareff method for determiningsoil organic matter and a proposed modification of the chromic acid titration.Method. Soil Sci. 37, 29–38.

Walter, I., Martinez, F., Alonso, L., de Gracia, J., Cuevas, G., 2002. Extractable soil heavymetals following the cessation of biosolids application to agricultural soil. Environ.Pollut. 117, 315–321.

Williams, D.E., Vlamis, J., Pukite, A.H., Corney, J.E., 1987. Metal movement in sludgeamended soils: a nine year study. Soil Sci. 143, 124–131.