Pesticide mobility and leachate toxicity in two abandoned mine soils. Effect of organic amendments

Science of the Total Environment 497–498 (2014) 561–569

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Pesticide mobility and leachate toxicity in two abandoned mine soils. Effect of organic amendments José Antonio Rodríguez-Liébana, M. Dolores Mingorance, Aránzazu Peña ⁎ Instituto Andaluz de Ciencias de la Tierra (IACT), Consejo Superior de Investigaciones Científicas-Universidad de Granada (CSIC-UGR), Avda. de las Palmeras, 4, 18100 Armilla, Granada, Spain

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Mine soils are degraded landscapes which need to be restored through revegetation. • Soil columns to evaluate pesticide and element mobility from two Spanish mine soils • Thiacloprid and fenarimol mobility reduced by application of organic amendments • Only As in leachates at hazardous levels, despite the high soil element content • The toxicity of the eluates diminished after amendment addition.

a r t i c l e

i n f o

Article history: Received 13 June 2014 Received in revised form 29 July 2014 Accepted 3 August 2014 Available online xxxx Editor: D. Barcelo Keywords: Pesticide hazard Mine soil Sewage sludge Compost Leachability

a b s t r a c t Abandoned mine areas, used in the past for the extraction of minerals, constitute a degraded landscape which needs to be reintegrated to productive or leisure activities. However these soils, mainly composed by silt or sand and with low organic matter content, are vulnerable to organic and inorganic pollutants posing a risk to the surrounding ecosystems and groundwater. Soils from two mining areas from Andalusia were evaluated: one from Nerva (NCL) in the Iberian Pyrite Belt (SW Andalusia) and another one from the iron Alquife mine (ALQ) (SE Andalusia). To improve soil properties and fertility two amendments, stabilised sewage sludge (SSL) and composted sewage sludge (CSL), were selected. The effect of amendment addition on the mobility of two model pesticides, thiacloprid and fenarimol, was assessed using soil columns under non-equilibrium conditions. Fenarimol, more hydrophobic than thiacloprid, only leached from native ALQ, a soil with lower organic carbon (OC) content than NCL (0.21 and 1.4%, respectively). Addition of amendments affected differently pesticide mobility: thiacloprid in the leachates was reduced by 14% in NCL-SSL and by 4% in ALQ-CSL. Soil OC and dissolved OC were the parameters which explained pesticide residues in soil. Chemical analysis revealed that leachates from the different soil columns did not contain toxic element levels, except As in NCL soil. Finally ecotoxicological data showed moderate toxicity in the initial leachates, with an increase coinciding with pesticide maximum concentration. The addition of SSL slightly reduced the toxicity towards Vibrio fischeri, likely due to enhanced retention of pesticides by amended soils. © 2014 Elsevier B.V. All rights reserved.

Abbreviations: A254, Absorbance at 254 nm; ALQ, Alquife soil; BTC, Breakthrough curves; CDE, Convection–dispersion equation; CSL, Composted sewage sludge; DOC, Dissolved organic carbon; EC, Electrical conductivity; FEN, Fenarimol; HIX, Humification index; HPLC, High performance liquid chromatography; NC, Nerva soil; NCL, Limed Nerva soil; OC, Organic carbon; R, Retardation factor; SSL, Stabilised sewage sludge; THC, Thiacloprid; V/V0, Relative pore volume. ⁎ Corresponding author. Tel.: +34 958 230000x190119; fax: +34 958 552620. E-mail address: [email protected] (A. Peña).

http://dx.doi.org/10.1016/j.scitotenv.2014.08.010 0048-9697/© 2014 Elsevier B.V. All rights reserved.

562

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569

1. Introduction Abandoned mining areas have a negative influence on environmental and economic development. They were used in the past for the extraction of minerals and exploitation of the deposits but are no more in use, implying a major source of pollution of the surrounding environment, through a combination of eolian dispersion and water erosion processes (Mendez and Maier, 2008; García-Gómez et al., 2014). They constitute large areas of derelict land whose reintegration as productive land requires taking severe actions for the purpose of recovering mine soils for economic activities. Mine tailings lack nutrients supportive of biological growth (N, P, K), have a pH ranging from highly acidic (pH 2) to alkaline (pH 9) depending on the carbonate content and acid-generating potential of the tailings (Mendez and Maier, 2008), are mainly composed by silt or sand and contain almost no organic matter, lost as a result of initial stripping from the site to be mined. These two last features, highly related with soil ability to hold and release chemical compounds in solution (Fernandes et al., 2003), show that these soils are prone to pollutant leaching and groundwater contamination. Revegetation of these mine tailings is thus a good strategy to cope with this situation, because plant canopy serves to reduce eolian dispersion while plant roots help prevent water erosion and leaching (Asensio et al., 2013). Full restoration of these areas will require the establishment of a functional ecosystem through revegetation which demands organic fertilization. Improvement of soil properties by addition of organic wastes is a usual strategy to restore these areas for revegetation purposes (Alvarenga et al., 2008; Mingorance et al., 2014), thus avoiding waste disposal to the environment. Organic wastes, with high organic matter content, affect soil fertility and water holding capacity and improve the overall soil structure and nutrient content, favouring soil reclamation. However, pollutants either present in the soil by past mining activities (metals), or added with the organic wastes or through the use of pesticides to control pests in the plants used for revegetation, are susceptible to reach the soil environment (Marín-Benito et al., 2013; García-Gómez et al., 2014). Understanding the physico-chemical processes involved in pollutant dynamics is an important issue to minimize water source contamination and properly assess the requirements of these soils, intended to be implemented for recreational or leisure activities. The environmental fate of metals, abundant in these soils, has been the subject of numerous researches in recent years (Gilchrist et al., 2011; Deng et al., 2011; Asensio et al., 2013). However, the potential hazard of pesticide leaching from treatments to plants grown in these soils has not been yet evaluated. To do this, two contrasting mine soils from Andalusia (south of Spain) were selected: one located in one of the largest sulphide deposits in the world (Nerva, Huelva, western Andalusia), with extremely acid pH and heavily contaminated with toxic elements and the other one from an abandoned iron ore mine (Alquife, Granada, eastern Andalusia), with basic pH and slightly polluted. The ability of both soils to retain two model pesticides, the polar insecticide thiacloprid and the fungicide fenarimol of intermediate polarity, was evaluated using soil columns. The effect on pesticide mobility of addition of organic wastes (sewage sludge and composted sewage sludge), previously selected for revegetation strategies (Mingorance et al., 2014), was also established. Finally, the chemical composition and toxicity to Vibrio fischeri of the leachates were assessed to establish the potential hazard of the different treatments. 2. Materials and methods 2.1. Soils and amendments Two soils from Andalusia (south of Spain) devoted in the past to mining activities were selected: one situated near Riotinto in the

province of Huelva (SW Spain), used for copper extraction and the other one placed in Alquife, in the province of Granada (SE Spain), aimed at iron extraction. The locations of both mining sites are presented in Fig. 1S in the supplementary material. The soils were collected from the upper 20 cm, air dried and sieved (b 2 mm). The fine earth fraction corresponded in both cases to sandy loam soils (55% sand, 14% clay for the former and 64% sand, 8% clay for the latter). The soil from Riotinto site, very acid (pH 2.4), was limed with Carbocal (Azucarera Ebro), a residue rich in calcium carbonate (83.4%) to correct soil acidity and named as NCL (Table 1). It displays low organic carbon (OC) content and higher electrical conductivity (EC), while the soil from Alquife (named ALQ) presents alkaline pH with low EC and OC content (Table 1). Major and toxic elements for both soils are also shown in Table 1. More information about both sites can be found elsewhere (Rodríguez-Liébana et al., 2013). Two organic amendments from wastewater treatment plants were employed: stabilised sewage sludge (SSL) from the plant of Granada (SE Spain), and composted sewage sludge (CSL) from the plant of Sevilla (SW Spain). Sewage sludges were subjected in the urban water treatment plant to an aerobic digestion and mechanical dehydration. Then they were air-dried in the laboratory for several weeks and sieved b2 mm before use. Main properties and element contents are also presented in Table 1. Amended soils consisted in NCL soil mixed with 2% SSL and ALQ soil mixed with 5% CSL. This study takes part of a project aiming at the restoration of both mining sites through revegetation. Therefore, the selected amendment doses took into account an effective increase in soil OC (Table 1) and previous results reflecting the ability of establishment of plant species in both mining areas (Mingorance et al., 2014 and unpublished results). 2.2. Pesticides Thiacloprid (THC), a neonicotinoid insecticide and fenarimol (FEN), a pyrimidine systemic fungicide, were employed. Standards (purity ≥ 98%) were used without further purification (Dr. Ehrenstorfer, Germany). Their octanol/water partition coefficients (log Kow) are 1.26 and 3.69, and their solubility in water is 185 and 13.7 mg L−1, respectively (Tomlin, 2003). Their environmental DT50 values range between 7 and 21 d for THC and N365 d for FEN. The Groundwater Ubiquity Score (GUS) index for THC (1.44) indicates a low leaching potential, while that of FEN (2.72) can be considered as moderate (Footprint Table 1 Some properties of the mine soils (NCL and ALQ) and the organic amendments, stabilised sewage sludge (SSL) and composted sludge (CSL).

pH EC (dS m−1)a OC (%) HIX Fe2O3 (%) CaO (%) MgO (%) MnO (%) K2O (%) Na2O (%) P2O5 (%) S (mg kg−1) Sr (mg kg−1) As (mg kg−1) Cu (mg kg−1) Cr (mg kg−1) Pb (mg kg−1) Zn (mg kg−1) Ni (mg kg−1)

NCL

ALQ

SSL

CSL

NCL-SSL2

ALQ-CSL5

6.8 1.9 1.4 5.74 23 0.13 0.53 0.05 1.95 0.41 0.51 8320 55 3951 694 97 3976 120 b.d.

8.2 0.07 0.21 1.41 28 15.9 1.6 0.97 2.11 0.49 0.10 127 171 49 49 b.d. b.d. 70 b.d.

7.2 2.8 36 0.43 3.3 13.0 0.99 0.05 1.11 0.06 7.8 b.d. 2327 b.d. 676 169 1163 2470 140

6.8 3.0 10 2.21 4.8 15.5 1.5 0.11 1.46 0.26 4.1 14,012 312 b.d. 314 154 b.d. 934 71

7.0 2.3 1.8 1.91

8.0 0.66 0.37 4.19

Element content by X-ray fluorescence (XRF). b.d., below detection. a EC measurements at 1:2.5 ratio (w:v) for non-amended (NCL and ALQ) and amended soils (NCL-SSL2 and ALQ-CSL5) and at 1:10 ratio for pure amendments (SSL and CSL).

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569

database, http://sitem.herts.ac.uk/aeru/iupac/). The molecular structure of both pesticides is presented in Fig. 2S in the supplementary material. Stock pesticide solutions were prepared at 1000 mg L−1 in HPLC grade acetone (Panreac). 2.3. Transport experiments The solute-transport assays were carried out, per triplicate, in polypropylene columns (2.6 cm i.d.; 10 cm long), which were homogeneously packed by adding successive layers of non-amended or amended soil to establish uniform bulk density up to a height of 5 cm (Peña et al., 2011). Soils and amendments, passed several times through 2 mm sieve, were shaken end-over-end overnight to assure homogeneous mixing before column packing. The columns were saturated by capillarity with 0.01 M CaCl2 (Panreac) for 16 h, allowing them afterwards to drain excess water, so the final soil moisture was close to field capacity. They were wrapped in aluminium foil to avoid pesticide photodegradation. A 100-μL aliquot of a solution containing a mixture of THC and FEN at 1 mg mL−1 in acetone (HPLC grade, Panreac) was carefully added on top of the column and left until the solvent evaporated. Then MgBr2 (0.4 mL, 0.2 M, Aldrich), used as a non-reactive tracer, was added on the top soil for testing column performance and immediately deionised water was applied on top of the soil column at a constant pressure head under steady state flow conditions, allowing the influents to flow through the soil under positive pressure head. Leachates were collected using a fraction collector (Model frac-920, General Electrics), programmed so that collection was more frequent at the beginning of the experiment (typical leachate volume 1.5 mL), slowing down as the experiment progressed (from 3 to 10 mL). Leachates were weighed, filtered and stored at −20 °C if the chemical analysis could not be immediately undertaken, except for pesticides and dissolved organic carbon (DOC) which were always immediately analysed. Bromide concentration in the leachates was determined until a constant background level was reached. At the end of each experiment, carried out at ambient temperature, columns were allowed to drain and then the soil was sectioned into two equal layers (upper and lower) to determine the amount of remaining pesticide. An aliquot of each soil section was oven dried at 105 °C to determine the water content. All the results were referred to dry weight. Pesticide sorption on the column material or on any plastic (i.e., tubing between the column output and the collecting tubes) was previously discarded. 2.4. Analytical determinations The concentrations of THC and FEN in the leachates were determined by reversed-phase HPLC with DAD, using a Zorbax Eclipse XDBC8 column (5 μm, 2.1 × 150 mm), as previously reported (RodríguezLiébana et al., 2013). Briefly, pesticides were eluted with a 50:50 (v:v) mixture of acetonitrile/water and detection wavelength was set at 245 nm for THC and 210 nm for FEN. Linear responses for both pesticides were established between 0.1 and 10 mg L−1 (R2 = 0.998). LODs were 0.06 and 0.05 mg L−1 and LOQs 0.19 and 0.16 mg L−1, for THC and FEN, respectively, calculated by application of a linear regression model to the calibration curve (Cuadros Rodríguez et al., 1993; Sánchez et al., 2004). Extraction of THC and FEN from soil was carried out by sonicating 2.5 g soil (dry weight) (Ultrasons, Selecta) with 15 mL methanol (HPLC grade, Panreac) for 15 min, and centrifuging subsequently the suspensions (Hettich GmbH) at 2500 rpm for 15 min. This process was repeated twice. The combined supernatants were concentrated to dryness (Laborota 4000, Heidolph), and the dry residue was dissolved in 1.5 mL of a 50:50 acetonitrile:water mixture, filtered (PVDF 0.45 μm filters) and analysed by HPLC as described above. Samples from each soil section were analysed per duplicate. Recovery experiments were run with fortified soil samples at two

563

concentration levels (1 and 5 μg g− 1) to validate the proposed analytical method. Soil recoveries (%) for the 1 ppm level ranged from 87.9 to 97.6 for THC and from 95.3 to 105.1 for FEN with standard deviations from 1.7 to 3.7. At the 5 ppm level recoveries ranged from 103.2 to 107.5 and from 105.9 to 109.0 for THC and FEN, respectively, with standard deviations from 1.0 to 5.2. The particle size distribution was determined by sieving and sedimentation, applying the Robinson's pipette method after organic matter had been removed with H2O2, using sodium hexametaphosphate as dispersing agent. Soil pH and electrical conductivity (EC) determinations were carried out in soil/deionised water suspension 1/2.5 (w/v) for soils and 1/10 (w/v) for amendments. Soil OC was determined by a modified Walkey and Black method (Mingorance et al., 2007). The mineralogical composition of soils and amendments was determined by X-ray fluorescence analysis. The soil HIX was determined in sample/deionised water suspensions 1/4 (w/v), according to Zsolnay (2003). Absorbance at 254 nm of the filtered and diluted leachates was used as a rapid estimation of DOC amount (Artiola and Walworth, 2009). For element analysis, leachates were combined in 10 fractions, acidified to pH 1 with HNO3, and analysed by ICP-OES (6500 Duo, Thermo iCAP). Bromide analysis was performed by ion chromatography (761 Compact IC, Metrohm) in diluted leachates. The inhibitory effects of the combined leachates used for chemical element analysis were assessed with the Microtox Acute Toxicity test, a 15-minute luminescence inhibition test which uses freeze-dried luminescent bacteria (V. fischeri NRRL B-11177), according to ISO 11348-3 (2007). Grouped column leachates, and their dilutions with a nontoxic control (2% NaCl solution), 6.25%, 12.5%, 25% and 50% (v/v), were tested and compared with the control. The decrease of luminescence was measured after 15 min contact using Optocomp I (MGM Instruments, Gomensoro, Spain). All measurements were carried out in duplicate. The EC50 values (leachate concentration, % v/v, at which a toxic effect on 50% of the population of organisms can be observed) were calculated, as well as the derived toxic units (TU) as TU ¼ EC150  100 (Persoone et al., 2003; Matejczyk et al., 2011). 2.5. Pesticide transport model According to the miscible displacement theory, movement of solute in porous media occurs as a result of the combined effects of diffusion and convection. The convection–dispersion equation (CDE) is the basic equation used in this mechanistic model. The one-dimensional non reactive solute transport, assuming steady-state flow through homogeneous soils, is given by the following equation δC δ2 C δC ¼ D 2 −υ δt δx δx

ð1Þ

where x is the distance in cm, D (cm2 min−1) is the apparent dispersion coefficient, which refers to the combined influence of diffusion and hydrodynamic dispersion for dissolved chemicals in porous media, υ (cm min− 1) is the pore-water velocity and C (mg L− 1) is the solute concentration. For reactive solutes the CDE two-region non-equilibrium model was assumed. This approach, which considers that solutes can be partitioned into mobile ( m ) and immobile ( im ) liquid phases and that linear adsorption controls solute retention, is written as βR

δC1 1 δ2 C1 δC1 ¼ −ωðC1 −C2 Þ−μC1 − δt δx P δx2

ð2Þ

where R is the retardation factor, which represents the effect of sorption process on solute transport; P is the Péclet number which accounts for the dispersivity of the soil; β is the mobile water partitioning coefficient; ω is a dimensionless mass transfer

564

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569

coefficient; C1 and C2 are the relative concentrations in the mobile (C1) and immobile (C2) liquid phases; and μ is the solute first order degradation coefficient; the rest as described above. In particular, for THC and FEN transport, the model was used under the following conditions: Dirac pulse input water content; flow velocity and dispersion coefficient set constant; linear pesticide sorption; and pesticide degradation in the liquid phase set to zero, because of the short duration of the assay.

2.6. Data evaluation The breakthrough curves (BTCs) for bromide, THC and FEN in non-amended and amended soils were drawn by plotting the corresponding solute concentration in the leachate versus the relative pore volume (V/V0). The data were pooled if replicates showed close agreement, an indication of good control over experimental conditions and sound experimental procedure. Some parameters in the transport model can be independently calculated from experimental measurements. The volumetric water content (V, cm3 cm− 3) was calculated as the resident water content of the soil after the experiment by soil volume unit. The elution fluxes at the column outlet (Q, cm3 min− 1) were measured by weighing the amount of solution eluted at regular time intervals. The Darcy's velocities (q, cm min− 1) were evaluated by dividing Q by the cross sectional area of the soil column. The actual pore-water velocity (υ, cm min− 1) was calculated as υ = q / . Other parameters are obtained by fitting leaching data to the CDE model with STANDMOD software that includes CXTFIT algorithms (Toride et al., 1999). For the bromide tracer, the D value was estimated by fitting the analytical solution of the CDE model under equilibrium, setting R to 1. These parameters are presented in Table 2 and in Fig. 3S in the supplementary material. For THC and FEN, the two-region non-equilibrium CDE model was fitted by introducing D from bromide BTCs and υ as constants to estimate R and non-equilibrium dimensionless parameters, β and ω, for the transport of both pesticides. Pesticide degradation was assumed to be zero, as stated above. An experimental pesticide retar dation factor was approached by equation Rexp ¼ 1 þ ρKd θ , where ρ is the bulk density (g cm− 3) and Kd a linear distribution coefficient (L kg− 1 ) derived from batch sorption assays (Rodríguez-Liébana et al., 2013). ANOVA followed by the Fisher's least significant difference (LSD) test was used to compare several mean groups. The significance level was set at α = 0.05. The data were analysed with SPSS 17.0 software (SPSS, Chicago, IL).

3. Results and discussion 3.1. Leachate properties The determination of physicochemical properties and the analysis of chemical elements in the leachates showed in general good agreement among repetitions, so data for replicated columns were pooled. Initial leachate pH (Table 3) was lower for ALQ than for NCL and, in general, pH values increased during elution, reached a plateau and then levelled or slightly decreased until the final measured pH (Fig. 3S in the supplementary material). In the case of NCL-SSL2 the pH value remained more or less stable during elution but with marked oscillations between consecutive leachates. Leachate conductivity (Table 3) corresponds to the contributions of the amendments and the soils, lower for ALQ than for NCL, in accordance with their properties, reflecting the differences in EC of the mining soils (Table 1). Salts were mainly eluted in the first leachates and then gradually fell as a consequence of the washing effect (data not shown). The values of absorbance at 254 nm (Table 3 and Fig. 3S in the supplementary material) (Artiola and Walworth, 2009) suggested that both non-amended soils leached similar quantities of DOC, in spite of their different OC contents and their different doses. Addition of amendments led to an increase of the amount of leached DOC (Table 3), more for SSL, indicating that its proportion of soluble and polar DOC was greater. This fact is supported by the lower HIX value of this fresh amendment (Table 1), in comparison with the HIX value of CSL, a composted sewage sludge, with more humified OM. With respect to element content in the leachates, macronutrients, in particular exchange cations, were found in all soils (amended and nonamended) due to their greater solubility. In all cases elements peaked in the first fraction and their concentration diminished gradually (Fig. 1). Among the potentially toxic elements, only relatively low concentrations of As and Cu were quantified in NCL leachates (Fig. 1) despite the fact that this soil is highly polluted (Table 1), according to the European Directive establishing the limit values of heavy metals in sludge for agricultural land disposal (ED 86/278/EEC and European Commission, 1986) (Table 1). However the concentration of As in the leachates (N 10 μg L−1) would be above the threshold in water for human consumption, but not that of Cu (b 2 mg L−1) (RD 140/2003, 2003). On the other hand, total element content of ALQ (Table 1) was below the threshold to allow its classification as a contaminated soil (ED 86/ 278/EEC and European Commission, 1986). As a consequence hazardous elements were not detected in the leachates (Fig. 1). It is important to stress that P was not detected in any fraction from ALQ soils (Fig. 1), because it is a P-deficient soil (Sevilla-Perea et al., 2014).

Table 2 Calculated soil column parameters (m, ρ, ϑ, V/V0, V, q, υ) (±standard deviation) and estimated transport parameter (D) (±standard error) for bromide tracer in non-amended (NCL and ALQ) and amended (NCL-SSL2 and ALQ-CSL5) soils. The values of D and R2 were derived from BTCs generated by the CXTFIT modelling using the equilibrium convection–dispersion approach. Soil

ma (g)

NCL NCL-SSL2 ALQ ALQ-CSL5

37.0 36.8 42.9 41.7

a b c d e f g h

± ± ± ±

0.5 0.9 1.3 0.1

Soil mass. Bulk density. Volumetric water content. Pore volume. Total volume leached. Flux density or Darcy velocity. Pore-water velocity. Dispersion coefficient.

ρb (g cm−3)

ϑc (cm3 cm−3)

V/V0d (cm3)

Ve (mL)

qf (cm min−1)

υg (cm min−1)

Dh (cm2 min−1)

R2

1.35 1.31 1.60 1.53

0.435 0.430 0.406 0.410

13.53 14.17 10.73 11.47

168 311 172 133

0.010 0.031 0.021 0.015

0.021 0.061 0.052 0.034

0.012 0.19 0.018 0.013

0.960 0.894 0.921 0.935

± ± ± ±

0.02 0.03 0.05 0.03

± ± ± ±

0.007 0.014 0.022 0.000

± ± ± ±

0.17 0.33 1.13 0.43

± ± ± ±

7 29 21 6

± ± ± ±

0.000 0.003 0.004 0.002

± ± ± ±

0.001 0.003 0.009 0.004

± ± ± ±

0.001 0.06 0.002 0.001

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569 Table 3 Some properties of the leachates (± standard deviation) from non-amended (NCL, ALQ) and amended (NCL-SSL2, ALQ-CSL5) soils. ALQ pHin pHfin pHmax ECin (dS m−1) ECfin (dS m−1) A254max

7.18 7.70 8.01 2.64 0.13 0.216

± ± ± ± ± ±

0.16 0.08 0.07 0.21 0.00 0.035

ALQ-CSL5

NCL

7.29 7.85 7.93 4.55 0.20 0.380

7.59 7.42 7.95 5.53 1.61 0.170

± ± ± ± ± ±

0.01 0.03 0.04 0.20 0.00 0.015

NCL-SSL2 ± ± ± ± ± ±

0.03 0.07 0.01 0.10 0.13 0.008

7.45 7.39 7.78 5.33 0.68 0.411

± ± ± ± ± ±

0.09 0.06 0.05 0.06 0.11 0.082

Subscripts in and fin refer to the measured property in the first and last collected leachates. Subscript max indicates the maximum value reached for the measured property.

As a conclusion, in leachates from ALQ columns, Ca and S were mainly provided by CSL, and Mg and Mn were mobilised from the soil when amendment was added. In NCL columns Ca was mainly provided by the liming agent, Mn and P by SSL, while K, Na and S were mobilised after amendment addition. 3.2. Leaching of the tracer Bromide was selected as the non-retained tracer because, according to X-ray fluorescence analysis, the soils and amendments contained no detectable or negligible amount of Br, in contrast with their content in chloride, another tracer ion widely used in mobility studies (Peña et al., 2011; Marín-Benito et al., 2013). The BTCs of the tracer for amended and non-amended mine soils are presented in Fig. 2. They were in general bell shaped with slight asymmetry and with little tailing, except for NCL-SSL2. Both NCL soils, with higher OC content (Table 1), reached lower peak concentration (maximum leached values) (Fig. 2) in accordance with previous reports using different organic amendments (Marín-Benito et al., 2013). Recovered Br− mass oscillated between 82.7 and 95.7%, a range which can be considered satisfactory (Lennartz, 1999; Vincent et al., 2007). Bromide leaching data

Element concentration (mg L-1)

1200 1000 800 600 400 200 0

As


565

showed a good fitting to the CDE model with R2 ranging from 0.921 to 0.960 for all columns except for NCL-SSL2, which was close to 0.9 (Table 2). The values of D were similar for both non-amended soils and for ALQ-CSL5, but increased for NCL-SSL2, indicating more diffusion–dispersion in this treatment (Table 2) (Pang and Close, 1999; Dal Bosco et al., 2013). It can be possible that, although care was taken for soil homogenisation, the greater heterogeneity of SSL could have favoured a less homogeneous packing (Marín-Benito et al., 2013). The addition of SSL to NCL soil significantly increased the elution flux (Fig. 3S in the supplementary material) possibly by reducing soil compactness and increasing porosity, favouring the rapid percolation of the influent solution. 3.3. Pesticide leaching 3.3.1. Thiacloprid In non-amended ALQ soil THC presented symmetrical BTCs (Fig. 3) and the peak corresponding to the maximum concentration was only slightly delayed relative to the tracer (Table 4). This means that in this soil this polar pesticide is scarcely retained and practically moves with the eluting front. For the rest of the soils (ALQ-CSL5, NCL and NCL-SSL2) the BTCs showed extended tails (Fig. 3), pointing to non-equilibrium sorption due to time-dependent pesticide/soil interactions (Rodriguez-Cruz et al., 2011). According to the mobile water partitioning coefficient (β) THC leached in ALQ soil under equilibrium conditions. For all the other cases (β b 1) non-equilibrium conditions were assumed. The low mass transfer coefficient values (ω b 10) reinforced the non-equilibrium conditions. Therefore, pesticide mass transfer is likely governed by convective transport (Peña et al., 2011). The calculated THC retardation factors (Rmod) indicate that ALQ retained THC less than NC L , and that retention was higher for amended soils than for the corresponding non-amended substrates (Table 4), because the addition of amendments reduces the presence

Ca

Cu

K

Mg

Mn

Na

P

S

864 389 29.2 16.3


20.0 18.8 3.07
86.4 16.4 4.13 1.92

0.52 0.13
47.1 13.5 1.26 0.54


346 5.25 3.89 0.25

1200 1000 800 600 400 200 0

As

Ca

Cu

K

Mg

Mn

Na

P

S

1st NCL-SSL2

0.49

928

1.00

15.9

49.3

0.63

22.0

2.79

421

1st NCL

0.15

1038

0.10

4.83

13.0

0.32

7.15


389

last NCL-SSL2

0.54

107

0.12

4.80

0.15

172

0.07


1.00

last NCL

0.33
0.83
110

0.82

0.75

76

Fig. 1. Element concentration (mg L−1) in the first and last fractions of grouped leachates from columns of non-amended (ALQ and NCL) and amended (ALQ-CSL5 and NCL-SSL2) mine soils. bb.d., below detection.

566

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569

1400

1000

ALQ

1200

ALQ-CSL5

800

Bromide concentration (mg L-1)

1000 800

600

600

400

400 200

200

0

0 0

2

4

6

8

10 12 14 16 18 20

0

700

2

4

6

8

10 12 14 16 18 20

700

NCL

600

NCL-SSL2

600

500

500

400

400

300

300

200

200

100

100

0

0 0

2

4

6

8

10 12 14 16 18 20

0

2

4

6

8

10 12 14 16 18 20

V/V0 Fig. 2. Measured (circles) and simulated (continuous lines) breakthrough curves for the non-retained bromide tracer from columns of non-amended (ALQ and NCL) and amended (ALQCSL5 and NCL-SSL2) mine soils.

of pesticide in the liquid phase. The values of Rmod for THC were significantly (p = 0.05) related with soil OC content (Rmod = 1.15 + 3.36 × OCsoil, R2 = 0.854), confirming the relevant role of this soil parameter on the retention of non-ionic organic compounds. An experimental retardation factor for all soils and pesticides was calculated from previous sorption experiments (Rodríguez-Liébana

5

THC

ALQ-CSL5

ALQ

8

4

6

3

4

2

THC FEN

2

1

0 0

2

4

6

8

10 12 14 16 18 20 0

2

4

6

8

0 10 12 14 16 18 20 5

10

NCL-SSL2

NCL 8

4

6

3

4

2

2

THC

1

THC

0 0

2

4

6

8

10 12 14 16 18 20 0

2

4

Toxic units (TU)

Pesticide concentration (mg L-1)

10

et al., 2013) using non-amended and amended mine soils after an incubation period of 1.5 months (Table 4). Therefore, the calculated Rexp may differ slightly because some soil parameters, such as OC content, which play an essential role in pesticide sorption, change during incubation (Mingorance et al., 2014). Both retardation factors for THC, Rexp and Rmod, could not be statistically related (p N 0.05), although in general

6

8

0 10 12 14 16 18 20

V/V0 Fig. 3. Measured (circles) and simulated (continuous lines) breakthrough curves for thiacloprid and fenarimol from columns of non-amended (ALQ and NCL) and amended (ALQ-CSL5 and NCL-SSL2) mine soils. The diamonds represent the toxic units of the leachates (Persoone et al., 2003).

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569

567

Table 4 Pesticide sorption parameters (Kd, Rexp) and estimated transport parameters (β, ω, Rmod) in non-amended (NCL, ALQ) and amended soils (NCL-SSL2, ALQ-CSL5) for thiacloprid and fenarimol from BTCs generated by the CXTFIT modelling using the non equilibrium convection–dispersion approach. Pesticide/soil

Kd ± SD a

Thiacloprid NCL NCL-SSL2 ALQ ALQ-CSL5

1.35 1.44 0.17 0.54

± ± ± ±

0.05 0.16 0.02 0.09

5.19 5.39 1.67 3.02

Fenarimol NCL NCL-SSL2 ALQ ALQ-CSL5

6.0 12.8 0.83 6.56

± ± ± ±

0.1 1.7 0.04 0.70

19.62 40.00 4.27 25.48

a b c d

Rexpb

β ± SEc

ω ± SE

Rmod ± SE

R2

MSEd

0.69 ± 0.02 0.52 ± 0.07 ≈1 0.67 ± 0.02

0.85 ± 0.08 10.02 ± 3.81 0.75 ± 0.05

4.74 7.98 1.64 2.95

0.17 0.61 0.02 0.09

0.970 0.792 0.966 0.986

0.005 0.010 0.190 0.014

0.64 ± 0.09

4.73 ± 2.72

3.48 ± 0.06

0.954

0.012

± ± ± ±

Kd, linear distribution coefficients ± standard deviation (Rodríguez-Liébana et al., 2013). Rexp calculated as R ¼ 1 þ Kdϑρ. Standard error. Mean squared error.

the Rmod values were lower than those calculated from sorption parameters. Previous results also found divergences between experimental and calculated retardation values and attributed them to less contact time between the pesticide and the soil in column experiments, removal by leaching of the pesticides from the column in contrast with the batch method, possible increase of soil specific surface area in batch methods due to abrasion caused by constant agitation, differences in soil/solution ratios and differences in sorption extent due to sorption non-linearity (Kookana et al., 1992; Chang and Wang, 2002; Dal Bosco et al., 2013; Marín-Benito et al., 2013; Larsbo et al., 2013). Residues of THC in soil were only found in NCL, with a uniform distribution in both soil layers (Table 5). The amount of THC leached from the different soil columns was similar for ALQ, ALQ-CSL and NCL but significantly lower (p b 0.05) for NCL-SSL2, whose OC content is more than 4 times higher than that of amended ALQ soil (Table 1). Finally for NCLSSL2 no pesticide could be extracted from the soil, contrary to what was expected from the amount of pesticide leached and from the Rmod and Rexp values (Table 4). This result was attributed to the strong retention of the pesticide by this amended soil or to higher irreversibility of THC sorption in this organic-amended soil (Fernandes et al., 2003), so that the remaining pesticide was probably present as a bound residue, retained on the amended soil matrix. Besides, and due to the relatively high content of S in NCL soil (Table 1), pesticide degradation could have occurred in NCL and NCL-SSL2 columns mediated by reduced sulphur species as reported for chloroacetanilide pesticides in sediment porewaters (Zeng et al., 2011). 3.3.2. Fenarimol As can be deduced from Table 4, the transport parameters for this fungicide could only be estimated by the CXTFIT model for ALQ soil, the only soil from which this pesticide leached. However, this is not

Table 5 Relative pesticide amount (%) leached and extracted from non-amended (NCL, ALQ) and amended (NCL-SSL2, ALQ-CSL5) soil columns. Pesticide/soil

Leached ± SD (%)

Upper soil ± SD (%)

Lower soil ± SD (%)

Total ± SD (%)

Thiacloprid NCL NCL-SSL2 ALQ ALQ-CSL5

70.48 56.43 88.19 83.90

4.78 ± 0.55 b.d. b.d. b.d.

5.61 ± 0.58 b.d. b.d. b.d.

80.87 56.43 88.19 83.90

± ± ± ±

0.53 8.23 11.50 3.97

Fenarimol NCL NCL-SSL2 ALQ ALQ-CSL5

b.d. b.d. 63.63 ± 7.10 b.d.

50.87 ± 7.04 60.21 ± 10.66 b.d. 34.56 ± 5.49

22.10 ± 3.87 10.78 ± 2.32 b.d. 31.05 ± 1.40

72.97 70.99 63.63 65.61

± ± ± ±

4.15 12.01 7.10 5.89

± ± ± ±

0.23 8.23 11.50 3.97

SD, standard deviation; b.d., below detection.

surprising because Rexp values were ≥20 for all soils but ALQ (Table 4) and because approximately 20 V/V0 were collected from each experimental column (Fig. 3). As in the case of THC, β b 1 and ω b 10 suggested that transport of FEN occurred under non-equilibrium conditions and was probably governed by convective transport. The Rmod, calculated with high R2 values and low MSE, indicated that this pesticide is much more retained than the more polar THC in this OC-poor soil. In both NCL and NCL-SSL2, pesticide amount in soil was significantly greater in the upper than in the lower soil layer (pNCL = 0.0004, pNCL-SSL2 = 0.0001) (Table 5), a fact that reflects the greater capacity of both NCL soils to retain this pesticide of intermediate polarity. On the contrary, in the amended ALQ-CSL5 soil, FEN was similarly distributed in both soil layers (p = 0.0824) suggesting that FEN was further transported than in NCL soils. Unlike for THC, the presence of FEN in the upper soil layer was not significantly related with soil OC (FENupper = 27.70 + 17.52 × OCsoil, R2 = 0.983), though it almost reached statistical significance (p = 0.0595). It may be possible that the mobility through the soil columns of this pesticide, more hydrophobic than THC, was also affected by DOC. It is important to stress that, although OC content is a good predictor of pesticide retention by soil as shown for THC, DOC aids in pesticide solubilisation as previously reported for tebuconazole and isoproturon (Herrero-Hernández et al., 2011; Ding et al., 2011). The organic amendments used in this study differ greatly in their composition: SSL is a fresh residue as shown by its much lower HIX value (Table 1), and therefore dominated by a fraction of polar DOC (Cox et al., 2000). The leachates of the amended NCL-SSL2 column consequently exhibit higher A254 values (Table 3) and more extended in time (Fig. 3S in the supplementary material). On the contrary, the composted CSL, amendment of ALQ soil, presents a higher HIX value, which indicates a more humified material (Mingorance et al., 2014), less water soluble, and accordingly the leachates of the amended column displayed lower A254 values. While DOC seems to affect more the leachability of FEN, of intermediate polarity, the mobility of the more polar THC, already very soluble in water, is not modified by this parameter, in accordance with a recent report of pesticide desorption studies from agricultural soils (Rodríguez-Liébana et al., 2014). Enhanced pesticide transport mediated by DOC molecules has been explained either by competition for sorption sites between the pesticide and DOC molecules or by interaction between the pesticide and DOC in solution (Cox et al., 2000; Fenoll et al., 2011). 3.4. Ecotoxicity test Toxicity assessment of the leachates can provide valuable and complementary information to the chemical analysis, because one of its major advantages is a direct assessment of the hazard to the ecosystem. The first effluent showed lower toxicity indexes for ALQ than for NCL (Fig. 3), suggesting that the composition of the latter may affect more

568

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569

the bioluminescence of V. fischeri. According to the hazard toxic units developed by Persoone et al. (2003), the leachates from all the columns, amended or not, with TU between 1 and 10, would correspond to Class III and considered as acutely toxic (Fig. 3). The toxicity of lixiviates from both soil columns without organic amendments rapidly increased after elution of the first pore volumes, reaching a plateau afterwards. The plateau in EC50 values for NC L (24.9 ± 2.4%), was attained at about 1 V/V0, while for ALQ it was displaced to 2 V/V0 (29.7 ± 4.5%), the leachates from ALQ showing slightly lower but not significantly different (p N 0.05) toxicity values. The general trend of bioluminescence inhibition does not resemble the time-dependent decrease in total metal concentration. On the contrary the toxicity increase during the first pore volumes could be related with pesticide leaching behaviour, in agreement with previous reports concerning organic pollutants in leachates, such as PAHs (Cottin and Merlin, 2008). THC has been classified as moderately hazardous (Class II), while FEN as slightly hazardous (Class III), according to the WHO pesticide classification based on acute oral LD50 values (http://www. who.int/ipcs/publications/pesticides_hazard_2009.pdf). However, the small changes in inhibition for the rest of the experimental period, when pesticide and element concentrations have drastically diminished, suggest that response in ecotoxicity tests cannot be always directly correlated with total or water-soluble pollutant concentrations, but may also reflect overall solution characteristics, including pH or EC (Płaza et al., 2010). Both amendments were able to slightly alleviate soil toxicity towards V. fischeri, in agreement with a general decrease in pesticide and elements leached from the amended columns (Figs. 1 and 3). The decrease in toxicity after amendment was significant for SSL added to NCL soil (p b 0.05). In this column, the leached THC was reduced by 14%, while in ALQ the reduction of THC leached after CSL addition was only 4% (Table 5). Although sewage sludge has been shown to be toxic towards some organisms, effluent toxicity has been only reported for high applications rates (≥5%) (Alvarenga et al., 2008). 4. Conclusions The one dimensional model CXTFIT 2.1. was able to adequately fit the experimental pesticide data. Amendment of organic C-poor mine soils with sewage sludges could be a good strategy to decrease the risk of groundwater contamination by non-ionic pesticides, especially those weakly retained such as thiacloprid. Fenarimol always leached less than thiacloprid; both pesticides were more retained in NCL than in ALQ soils, and more in amended than in non-amended soils. Nature and content in soil OC and/or DOC were determinant in the leachability of both pesticides from soils. However the effect of DOC on thiacloprid mobility seems to be negligible due to its high polarity and water solubility. In spite of the relatively high content in potentially toxic elements of one of the mine soils, leachates from the different soil columns only contained As at hazardous levels. These results were reinforced by ecotoxicological data which showed low/moderate toxicity in the first fractions, increasing afterwards, coinciding with the elution of the maximum pesticide concentration. The addition of organic amendments reduced the toxicity towards V. fischeri, probably due to the higher retention ability of the amended soil, reducing the potential risk of groundwater pollution. The results point to the adequacy of using organic amendments in reducing the vulnerability to pesticides and to toxic elements of these impoverished mine soils, with a view to revegetating them with appropriate plant species. Acknowledgements This work was supported by Proyecto de Excelencia-Junta de Andalucía (P10-RNM5814), cofinanced by European FEDER-ESF funds. JARL thanks CSIC for a predoctoral fellowship (JAE-Pre), also cofinanced

by ESF. EMASAGRA and EMASESA are acknowledged for the kind supply of stabilised (SSL) and composted (CSL) sewage sludge, respectively. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.08.010. References Alvarenga P, Palma P, Gonçalves AP, Baiao N, Fernandes RM, de Varennes A, et al. Assessment of chemical, biochemical and ecotoxicological aspects in a mine soil amended with sludge of either urban or industrial origin. Chemosphere 2008;72:1774–81. Artiola JF, Walworth JL. Irrigation water quality effects on soil carbon fractionation and organic carbon dissolution and leaching in a semiarid calcareous soil. Soil Sci 2009;174: 365–71. Asensio V, Vega FA, Singh BR, Covelo EF. Effects of tree vegetation and waste amendments on the fractionation of Cu, Cr, Ni, Pb and Zn in polluted mine soils. Sci Total Environ 2013;443:446–53. Chang TW, Wang MK. Assessment of sorbent/water ratio effect on adsorption using dimensional analysis and batch experiments. Chemosphere 2002;48:419–26. Cottin N, Merlin G. Removal of PAHs from laboratory columns simulating the humus upper layer of vertical flow constructed wetlands. Chemosphere 2008;73:711–6. Cox L, Celis R, Hermosín MC, Cornejo J, Zsolnay A, Zeller K. Effect of organic amendments on herbicide sorption as related to the nature of the dissolved organic matter. Environ Sci Technol 2000;34:4600–5. Cuadros Rodríguez L, García Campaña AM, Jiménez Linares C, Román Ceba M. Estimation of performance characteristics of an analytical method using the data set of the calibration experiment. Anal Lett 1993;26:1243–58. Dal Bosco TC, Sampaio SC, Coelho SRM, Corrêa MM, Netto AM, Cosmann NJ. The influence of organic matter from swine wastewater on the interaction and transport of alachlor in soil. Acta Sci Agron 2013;35:277–86. Deng C, Zhang C, Li L, Li Z, Li N. Mercury contamination and its potential health effects in a lead–zinc mining area in the karst region of Guangxi, China. Appl Geochem 2011;26: 154–9. Ding Q, Wu HL, Xu Y, Guo LJ, Liu K, Gao HM, et al. Impact of low molecular weight organic acids and dissolved organic matter on sorption and mobility of isoproturon in two soils. J Hazard Mater 2011;190:823–32. ED 86/278/EEC, European Commission. Council Directive 86/278/EEC on the protection of the environment, and in particular of the soil, when sewage sludge is used in agriculture. Off J Eur Communities 1986;L181:6–12. Fenoll J, Ruiz E, Flores P, Vela N, Hellín P, Navarro S. Use of farming and agroindustrial wastes as versatile barriers in reducing pesticide leaching through soil columns. J Hazard Mater 2011;187:206–12. Fernandes MC, Cox L, Hermosín MC, Cornejo J. Adsorption–desorption of metalaxyl as affecting dissipation and leaching in soils: role of mineral and organic components. Pest Manag Sci 2003;59:545–52. García-Gómez C, Sánchez-Pardo B, Esteban E, Peñalosa JM, Fernández MD. Risk assessment of an abandoned pyrite mine soil in Spain based on direct toxicity assays. Sci Total Environ 2014;470–471:390–9. Gilchrist S, Gates A, Elzinga E, Matthew Gorring M, Szabo Z. Source and fate of inorganic soil contamination around the abandoned Phillips sulphide mine, Hudson Highlands, New York. Soil Sediment Contam 2011;20:54–74. Herrero-Hernández E, Andrades MS, Marín-Benito JM, Sánchez-Martín MJ, RodríguezCruz MS. Field-scale dissipation of tebuconazole in a vineyard soil amended with spent mushroom substrate and its potential environmental impact. Ecotoxicol Environ Saf 2011;74:1480–8. ISO 11348-3. Water quality — determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (luminescent bacteria test). Part 3: method using freeze-dried bacteria. Geneva, Switzerland: International Organisation for Standardisation; 2007. Kookana RS, Aylmore LAG, Gerritse RG. Time-dependent sorption of pesticides during transport in soils. Soil Sci 1992;154:214–25. Larsbo M, Löfstrand E, van Alphen de Veer D, Ulén B. Pesticide leaching from two Swedish topsoils of contrasting texture amended with biochar. J Contam Hydrol 2013;147: 73–81. Lennartz B. Variation of herbicide transport parameters within a single field and its relation to water flux and soil properties. Geoderma 1999;91:327–45. Marín-Benito JM, Brown CB, Herrero-Hernández E, Arienzo M, Sánchez-Martín MJ, Rodríguez-Cruz MS. Use of raw or incubated organic wastes as amendments in reducing pesticide leaching through soil columns. Sci Total Environ 2013;463–464:589–99. Matejczyk M, Płaza GA, Nałęcz-Jawecki G, Ulfig K, Marlowska-Szczupak A. Estimation of the environmental risk posed by landfills using chemical, microbiological and ecotoxicological testing of leachates. Chemosphere 2011;82:1017–23. Mendez MO, Maier RM. Phytoremediation of mine tailings in temperate and arid environments. Rev Environ Sci Biotechnol 2008;7:47–59. Mingorance MD, Barahona E, Fernández-Gálvez J. Guidelines for improving organic carbon recovery by the wet oxidation method. Chemosphere 2007;68:409–13. Mingorance MD, Rossini-Oliva S, Valdés B, Pina Gata FJ, Leidi EO, Guzmán I, et al. Stabilized sewage sludge addition to improve properties of an acid mine soil for plant growth. J Soils Sediments 2014;14:703–12. Pang L, Close ME. Non-equilibrium transport of Cd in alluvial gravels. J Contam Hydrol 1999;36:185–206.

J.A. Rodríguez-Liébana et al. / Science of the Total Environment 497–498 (2014) 561–569 Peña A, Palma R, Mingorance MD. Transport of dimethoate through a Mediterranean soil under flowing surfactant solutions and treated wastewater. Colloids Surf A 2011;384: 507–12. Persoone G, Marsalek B, Blinova I, Törökne A, Zarina D, Manusadzianas L, et al. A practical and user-friendly toxicity classification system with microbiotests for natural waters and wastewaters. Environ Toxicol 2003;18:395–402. Płaza GA, Nałęcz-Jawecki G, Pinyakong O, Illmer P, Margesin R. Ecotoxicological and microbiological characterization of soils from heavy-metal and hydrocarboncontaminated sites. Environ Monit Assess 2010;163:477–88. RD 140/2003. Real Decreto por el que se establecen los criterios sanitarios de la calidad del agua de consumo humano. BOE 2003;45:7228–45. Rodríguez-Cruz MS, Ordax JM, Arienzo M, Sánchez-Martín MJ. Enhanced retention of linuron, alachlor and metalaxyl in sandy soil columns intercalated with wood barriers. Chemosphere 2011;82:1415–21. Rodríguez-Liébana JA, Mingorance MD, Peña A. Pesticide sorption on two contrasting mining soils by addition of organic wastes: effect of organic matter composition and soil solution properties. Colloids Surf A 2013;435:71–7. Rodríguez-Liébana JA, Mingorance MD, Peña A. Role of irrigation with raw and artificial wastewaters on pesticide desorption from two Mediterranean calcareous soils. Water Air Soil Pollut 2014;225:2049. http://dx.doi.org/10.1007/s11270-014-2049-z.

569

Sánchez L, Mingorance MD, Peña A. Chemical and physical factors affecting the extractability of methidathion from soil samples. Anal Bioanal Chem 2004;378:764–9. Sevilla-Perea A, Almendros G, Mingorance MD. Quadratic response models for N and P mineralization in domestic sewage sludge for mining dump reclamation. Appl Soil Ecol 2014;75:106–15. Tomlin CDS. The pesticide manual. 13th ed. Bracknell, Berks, UK: British Crop Protection Council; 2003 [1344 pp.]. Toride N, Leij FJ, van Genuchten MTh. The CXTFIT code for estimating transport parameters from laboratory or field tracer experiments, version 2.1. Research report no. 137. Riverside, CA: U.S. Salinity Laboratory, USDA, ARS; 1999. Vincent A, Benoit B, Pot V, Madrigal I, Delgado-Moreno L, Labat C. Impact of different land uses on the migration of two herbicides in a silt loam soil: unsaturated soil column displacement studies. Eur J Soil Sci 2007;58:320–8. Zeng T, Ziegelgruber KL, Chin Y-P, Arnold WA. Pesticide processing potential in prairie pothole porewaters. Environ Sci Technol 2011;45:6814–22. Zsolnay A. Dissolved organic matter: artefacts, definitions and functions. Geoderma 2003; 113:187–209.