Arch Environ Contam Toxicol (2014) 66:277–286 DOI 10.1007/s00244-013-9982-5

Input and Leaching Potential of Copper, Zinc, and Selenium in Agricultural Soil from Swine Slurry Jordi Comas • Carmen Domı´nguez • Dora I. Salas-Va´zquez • Juan Parera Sergi Dı´ez • Josep M. Bayona

•

Received: 10 September 2013 / Accepted: 6 December 2013 / Published online: 29 December 2013 Ó Springer Science+Business Media New York 2013

Abstract Trace elements, such as copper, zinc, and selenium, used as feed additives were determined in samples of both fresh (N = 14) and anaerobically digested (N = 6) swine slurry collected on medium- to large-size farms in northeast Spain. Considering both fresh and anaerobically digested samples, mean concentrations of zinc (1,500 mg kg-1 dry mass [dm]) were greater than those of copper (mean 239 mg kg-1 dm), and the selenium concentrations detected were even lower (mean 139 lg kg-1 dm). Zinc concentrations were significantly greater in anaerobically digested samples, whereas no significant differences were found for copper or selenium. In addition, the leaching potential of zinc, copper, and selenium in cropped (lettuce heart) and uncropped experimental units subject to drip irrigation was assessed in a greenhouse experiment. Generally, the addition of swine slurry to soil (1.7 g kg-1 dm) significantly increased zinc, copper, and selenium concentrations in leachates, which decreased in accordance with the volume of leachate eluted. Under the experimental conditions, the leaching potential of zinc and selenium was more strongly correlated with bulk parameters directly associated with the composition of the pig slurry (dissolved organic carbon, electrical conductivity, and ammonium), whereas copper mobility was more strongly associated with the crop

J. Comas  D. I. Salas-Va´zquez DEAB-UPC, Esteve Terrades 8, Building D4, 08860 Castelldefels, Spain C. Domı´nguez (&)  S. Dı´ez  J. M. Bayona IDAEA-CSIC, Jordi Girona 18, 08034 Barcelona, Spain e-mail: [email protected] J. Parera DARPAMN-GENCAT, Ronda Sant Martı´, 2-6, 25002 Lleida, Spain

root exudates. Although selenium has been shown to be mobile in soil, the selenium content found in the leachates did not pose any appreciable risk according to current drinking water standards.

Intensive farming gives rise to a large volume of manure. The most cost-effective and sustainable way to dispose of this manure is to apply it to agricultural soil, preferably near the manure source. However, in countries with large amounts of livestock, farms are usually concentrated in relatively small areas, and manure production significantly exceeds the local agricultural demand for manure as fertilizer. In the European Union (EU), manure is considered an organic byproduct of farming and as such can be applied to agricultural soil with no other requirement than compliance with the European Directive 91/676/EEC (1991) concerning the protection of waters against pollution caused by nitrates from agricultural sources. On large hog and poultry farms, the IPPC Directive (European Directive 2008/1/EC of the European Parliament and of the Council 2008), concerning integrated pollution prevention and control, also applies. Transporting livestock slurry far from its source to comply with disposal regulations is not cost-effective from an agronomical point of view because it is highly diluted (liquid fraction *90 %) (Nolan et al. 2012). In practice, this usually leads to excessive application in fields located near intensive livestock farms. One way to decrease the transport cost is to use dehydrated products, although dehydration processes are also energy intensive. Accordingly, the most widely used practices and processes to treat livestock slurry include reception pits, lagoons, aerobic and/or anaerobic digestion, and nitrification–denitrification

123

278

Arch Environ Contam Toxicol (2014) 66:277–286

Table 1 Mean TE concentration average (maximum–minimum) in swine manures (dry matter) and threshold values in the EU, United States, and Spanish legislations Manure treatment

Area/herd size

Cu (mg kg-1)

Zn (mg kg-1)

Se (mg kg-1)

References

Untreated (N = 8)

Sta. Catarina St (Brazil)

(523–110)

(1,445–427)

NR

Steinmetz et al. (2009)

Untreated

Mantua (N Italy)

(800–250)

(1,000–600)

NR

Cortellini and Piccinini (1993)

Untreated (N = 7)

NE China (200–800)

94 (164–11)

136 (207–37)

NR

Zhang et al. (2012)

Untreated (N = 21)

NE China ([800) Re´union Island (France)

118 (258–2.3)

135 (207–37)

NR

Zhang et al. (2012)

Untreated

399 (575–271)

635 (792–348)

NR

Legros et al. (2013)

Untreated

Hangzhou (China)

321

689

NR

Miamiao et al. (2009)

Untreated Untreated (N = 14)a

Austria Catalonia (NE Spain)

282–84 233 ± 134b

1,156–399 1335 ± 783b

3.37–1.35 0.144 ± 0.072b

Sager (2007) This study

Anaerobically digested (N = 16)

Jiangsu (China)

204 ± 30

477 ± 40

NR

Jin and Chang (2011)

Combustion and thermal gasification

Denmark

589–72

4,493–809

NR

Kuligowski et al. (2008)

Anaerobically digested (N = 6)

Catalonia (NE Spain)

245 ± 85b

1664 ± 722b

0.133 ± 0.044b

This study

Threshold values mg kg-1

Cu

Zn

Legal text

1,000–1,750

2,500–4,000

European Directive 86/278/EEC (1986)

70–400

200–1,000

Spanish RD 05/824

1,500

2,800

USEPA 40 CFR Part 503 (1993)

NR not reported a

Herd size comprises between 600 and 2,000 animals

b

Relative SD

(NDN) (Kunz et al. 2009; Flotats et al. 2009; Bernet and Be´line 2009; Nasir et al. 2012; Viancelli et al. 2013). In the EU, treated manure is no longer considered a farming byproduct. However, there is no European directive establishing the criteria for the use of such manure as fertilizer or an organic amendment. Nevertheless, European Directive 86/278/EEC (1986), concerning the protection of the environment, in particular of the soil, when sewage sludge is used in agriculture is often applied by default. Therefore, the actual criteria for applying such products are established by each Member State. In Spain, the relevant law is Royal Decree R.D. 824/2005 (B.O.E. 171 BOE 2005). Both the EU Directive and the Spanish Royal Decree establish thresholds for trace element (TE) contents used as feed additives to control some diseases. However, the standards under Spanish law are much more restrictive (Table 1). Similarly, the United States Environmental Protection Agency’s (USEPA) standards for the use or disposal of sewage sludge (Code of Federal Regulations, Title 40 [dealing with the protection of human health and the environment], part 503) establish a single reference limit for organic fertilizers and composts derived from manure or sewage sludge (Table 1). Nevertheless, these regulations do not address the potential contamination of soil and water with metals or antibiotics. In fact, the heavy application of slurries to agricultural soil has a variety of environmental impacts, such as odor, excess nutrients, TEs (e.g., copper and zinc),

123

steroid hormones (Combalbert et al. 2012), and the presence of certain antibiotics (e.g., chlortetracyclines, sulfonamides, fluoroquinolones) (Stone et al. 2011), C [ which could contaminate surface runoff or groundwater in vulnerable aquifers (Krapac et al. 2002; Aldrich et al. 2002; Hseu 2004; Ku¨mmerer 2009). Moreover, antimicrobial resistance genes have been associated with the co-occurrence of copper, zinc, and certain antibiotics (Ho¨lzel et al. 2012). Phytotoxicity has also been reported in some sensitive cultivars (e.g., amaranth [Amaranthus tricolor]) in sandy loam soil amended with swine manure (Hao et al. 2008). Animal manure is also one of the largest sources of TEs in soil, as shown, e.g., in England and Wales (Nicholson et al. 2003) and a nation-wide assessment performed in France that monitored the input of 10 TEs and the most significant sources thereof (Belon et al. 2012). Significant evidence exists regarding the mobility of TEs from livestock waste in agricultural soil. After rainfall events, copper, zinc, and cadmium have been detected in drainage water and river catchments from areas with extensive agricultural activity in which the soil is amended with manure (Xue et al. 2000; Aldrich et al. 2002). In these studies, a large percentage of the copper (99.9 %) and a smaller percentage of the zinc (50 %) were complexed with abundant organic ligands, such as dispersed colloids. Accordingly, metals complexed with dispersed colloids associated with dissolved organic matter (DOM) can

Arch Environ Contam Toxicol (2014) 66:277–286

behave as mobile elements through soil macropores (Li and Shuman 1997; Karathanasis 1999; Bao et al. 2011). Manure composting can also affect metal content and mobility in soil (Hsu and Lo 2001; Hseu 2004). Cultivation practices (e.g., fertilization, mulching) can also affect the migration of TEs from soil to plants. Their incorporation into crops is thus variable and not directly correlated with soil contamination (Li et al. 2010). Moreover, plant root exudates consisting of a complex mixture of organic acid anions, phytosiderophores, sugars, vitamins, amino acids, purines, nucleosides, inorganic ions, and enzymes have major direct and indirect effects on TEs mobility and thus both TE plant bioavailability and TE movement through the soil profile (Bais et al. 2006). In contrast, the effect of cropping on the mobility of TEs in soil has scarcely been addressed (Banks et al. 2006; Kassaye et al. 2012), and the mechanisms involved are poorly understood (Degryse et al. Degryse et al. 2012). Moreover, the uptaken amount of TEs does not seem to pose any significant risk according to current European legislation (Mantovi et al. 2003). The main objectives of this study were to assess the occurrence of major (copper and zinc) and minor (selenium) TEs in swine slurry of different origins in northeast (NE) Spain subjected to different treatment processes (NDN, anaerobic digestion, and thermal treatment). The mobility of copper, zinc, and selenium in agricultural soil amended with digested swine slurry and the effect of cropping (i.e., Lactuca sativa L.) on their vertical transport were evaluated in experimental units in controlled conditions (i.e., greenhouse drip irrigation).

279

(N = 3) were obtained. Only plastic tools and containers were used for sampling and sample storage. Samples were transported to the laboratory and kept frozen at -20 °C until analysis. Experimental Layout

Materials and Methods

The experiment was performed at the mesoscale level in a greenhouse (Agro`polis, UPC) located in Viladecans (Barcelona, Spain) using 2.5 L cylindrical amber glass containers (/ = 15 cm) fitted with a bottom outlet (/ = 3 cm). This outlet was threaded with a plastic plug attached by polytetrafluoroethylene tubing to a 1 L amber glass bottle located beneath. Each glass container was filled with 2.4 kg of topsoil (0–20 cm). The irrigation water was taken from harvested rainwater. A drip irrigation system was used, and the soil’s water content was monitored by gravity. The irrigation water flow was regulated using online drippers (2.2 L h-1) placed on polyethylene drip tubes (16/13.6 optical density (OD)/ID mm) and driven by microtubes (0.250/0.170 OD/ID mm). The duration (1 min) and frequency of irrigation were regulated by an automatic programmer. The watering frequency was increased as the crops grew from once every 2 days to twice daily by the end of the crop cycle. Thus, after each irrigation, the soil was brought to near field capacity, except before each leachate-sampling campaign, when the soil was brought to saturation and the irrigation was continued until enough leachates were obtained (*400 mL). Accordingly, the relation between soil and drainage water was 0.17 L of water/kg of soil, a value not far from field conditions (0.5–12 L kg-1 of topsoil). This procedure resulted in the collection of a total of three leachates every 2 weeks for a 6-week study period.

Sampling and Sample Handling

Soil

Swine Slurry

The soil used was collected from the surface horizon of a typical Xerorthent soil (Soil Survey Staff 1998) from the Llobregat River Delta’s agricultural area (longitude 2°030 E, latitude 41°170 N). It was characterized by conventional analytical methods (Burt 2004). Its main features were as follows: loamy sand texture (90 % sand, 8 % silt, and 2 % clay); pH 8.12 as measured in water suspension (soil-to-water ratio 1:2.5); (3) electrical conductivity 3.8 dS m-1 as measured in a saturated slurry; (4) total organic carbon content 5 g of kg-1; (5) total N content 0.7 g of kg-1; (6) Olsen P 29 mg kg-1; cation-exchange capacity (CEC) 3.8 meq 100 g-1; and exchangeable Ca2?, Mg2?, Na? and K? of 2.82, 0.64, 0.25, and 0.15 meq 100 g-1 soil, respectively. Native copper and zinc concentrations in this soil were 74 and 176 mg kg-1 (dm), respectively, and selenium was \10 mg kg-1 (dm).

Samples were collected from medium- to large-size farms (680–2,900 heads) in Catalonia (NE Spain), the largest pig producer in Spain (total herd size in 2011 = 25.7 9 106). Fresh slurry grab samples were collected at pig farms, mostly from the fattening and postfattening stages. A representative sample of fresh (N = 14), anaerobically digested (N = 6), and anaerobically digested plus thermally dehydrated (N = 2) swine (Sus scrofa ssp domestica) slurry were studied. Subsampling was used for the sampling of large farms ([2,000 animals) to increase representativeness. The subsamples were then analyzed as integrated samples. In addition, dehydrated swine slurry (pellets from an anaerobic digestion and thermal dehydration plant [N = 3]) and anaerobically digested slurry

123

280

Soil–Swine Slurry Preparation Anaerobically digested and thermally dehydrated swine slurry (278 g) was mechanically mixed with sieved soil (36 kg) and 165 g of water-soluble nitrogen–phosphorus– potassium (NPK) compound fertilizer with magnesium, sulfur, and micronutrients (Hakaphos Orange) using an industrial roller table homogenizer at the University of Barcelona’s MAT Control Laboratory facilities until the mixture was completely homogenized (*7 days). The amount of dehydrated swine slurry added to the soil (7.7 g of kg-1 dm) was equivalent to the legal limit for nitrogen in nitrate-vulnerable areas (170 kg N ha-1; European Directive 91/676/EEC 1991). Treatments To assess the impact of both the addition of swine slurry at an agronomic rate and cropping on the mobility of TEs, the following treatments were applied: (1) uncultivated topsoil (uncropped control), (2) cultivation of the topsoil without swine slurry (cropped control), (3) uncultivated topsoil with swine slurry, and (4) cultivation of the topsoil with swine slurry. As model, crop lettuce (Lactuca sativa L. cv. Cogollo de Tudela) was cultivated. Uncultivated experimental units were replicated four times, and those cultivated with lettuce were replicated eight times. Bulk Parameters in Leachates and Elemental Analysis in Swine Slurry Conventional wastewater parameters, including pH, conductivity and ammonia, were determined using a Hach HQ40d portable digital meter (Dusseldorf, Germany) equipped with interchangeable IntelliCAL electrodes (Hach). DOC was analyzed by high-temperature catalytic oxidation on a Shimadzu TOC-VCSH ? TNM-1 ? ASI-V analyzer (Kyoto, Japan) according to a previously published methodology ´ lvarez-Salgado and Miller 1998). (A Determination of TE in Pig Slurry and Leachates Samples were transported to the laboratory and frozen at -20 °C. Collected samples were freeze-dried and milled (150 lm diameter) using a ball mill (P100; Retsch GmbH, Haan, Germany). Total nitrogen and carbon of the samples (whole sample and nonhydrolysable fraction) were obtained using a Carlo Erba Model EA 1108 elemental analyzer (Carlo Erba, Milan, Italy). Copper, zinc, and selenium concentrations were determined by atomic absorption spectrometry (SpectrAA 110; Varian, Palo Alto, CA, USA) after dissolution of ashes (ignition at 470 °C) in 3 mol L-1 HNO3 (Commission Directive 93/1/CEE).

123

Arch Environ Contam Toxicol (2014) 66:277–286

TE determination in leachates was performed an inductively coupled plasma mass spectrometer from Thermo Scientific (XSeries II model; Bremen, Germany) equipped with a collision-induced dissociation cell. Helium was used as collision gas for greater sensitivity and to decrease possible isobaric interferences. Leachates were filtered through a 0.45-lm nylon filter and acidified to 2 % with ultra high-purity HNO3. Appropriate dilutions were made from original leachates for the final inductively coupled plasma-mass spectrometry (ICP-MS) determination. Single-element ICP standards from JT Baker (in a nitric acid 1 % medium) were used to quantify TEs. Lithium, scandium, gallium, yttrium, rhodium, and indium were used as internal standards. Quality control of the analytical determinations was performed by evaluating fortified blank samples and control standards to check the robustness, reproducibility, and accuracy of the method as well as possible memory effects. Data Analysis Content of TE in treated and untreated pig slurry and the effect of treatments on bulk parameters and TEs in each campaign were tested using one-way analysis of variance (ANOVA) (considered significant when the p value was \0.05) and Tukey least significant difference multiplecomparison test (setting a to 0.05). To investigate associations between measured bulk parameters and TEs and the possible clustering of treatments, multivariate principal component analysis (PCA) was performed and corresponding biplots created. All statistical analyses were performed using the R statistical package (R Development Core Team, Vienna, Austria).

Results and Discussion Occurrence of Copper, Zinc, and Selenium in Swine Slurry Table 1 lists the mean concentrations of the total TEs, namely, copper, zinc, and selenium, in the fresh and digested swine slurry analyzed in this study and compared with previously published data. TE content of the different samples was quite variable (coefficients of variation between 33 and 58 %) reflecting relatively moderate differences in management at the farm level. A greater concentration was found of zinc than copper, which is consistent with previously reported findings in pig feed (Sager 2007; Zhang et al. 2012). Selenium concentrations were 3–4 orders of magnitude lower than copper and zinc concentrations. The zinc content found in this study (mean 1,335 mg kg-1) was among the highest ever reported in

Arch Environ Contam Toxicol (2014) 66:277–286

b Fig. 1 Temporal evolution of a electrical conductivity (EC), b NH4?

30

(a)

281

and c Dissolved organic matter (DOC) in leachates over three samplings dates. At each sampling date, different letters indicate significant differences between treatments (a = 0.05)

Pig slurry & crop

a

Pig slurry

b 20

Uncropped control

-1

EC (mS cm )

Cropped control

c

a

10

d

b

a

c

0

b c d

10

15

20

25

30

35

40

45

Days from planting Pig slurry & crop Pig slurry Cropped control

b

600

Uncropped control

400

-1

NH4 (mg L )

(b)

800

a

a b

200

c

a b c

0

c 10

15

20

25

30

35

40

45

(c)

800

Days from planting

a 600

Pig slurry & crop Pig slurry Cropped control

-1

DOC (mg L )

Uncropped control

swine slurry. Selenium (mean 0.139 mg kg-1) was 1 order of magnitude lower than the values reported in Austria (Sager 2007). Moreover, significantly lower concentrations of zinc were obtained in the fresh samples than in the digested ones (mean values 1,335 and 1,664 mg kg-1, respectively), whereas no significant differences were found in this regard for selenium or copper. The high copper and zinc content in swine slurry found in this study is consistent with previous data in which zinc was usually the predominant element (Table 1). As expected, swine slurry treatment processes, such as NDN, anaerobic digestion, and thermal treatment, do not significantly affect copper and zinc concentrations. The slight enrichment, especially regarding zinc, may be attributable to the elimination of nitrogen during the NDN process and of carbon during the anaerobic digestion processes, which would lead to TE enrichment. This occurs because, with the exception of selenium, which can lead to the formation of volatile species under strongly reduced conditions, TEs behave almost conservatively in those treatment processes. It should be noted that the zinc content in pig slurry (both untreated and digested) found in this study is compliant with the threshold value set by European Directive 86/278/ EEC (1986) and USEPA 40 CFR Part 503 (1993). However, it does not meet the limits set by Spanish R.D. 824/2005. Therefore, it could not be used directly as fertilizer or an organic amendment in Spain. Conversely, copper concentrations are lower than the maximum legal limits. Selenium is also supplied in pig feed but at lower concentrations than copper and zinc (Sager 2007). Accordingly, selenium concentrations in manure are 3–4 orders of magnitude lower than those of copper and zinc. To the extent that selenium content decreases during the digestion of the manure, it behaves differently than copper and zinc, which behave conservatively.

400

Bulk Parameters in Leachates

200

a b

b

a ab b c

0

c

10

15

20

25

30

35

Days from planting

40

45

Figure 1 shows the temporal variability of bulk parameters in collected leachates according to the different treatments. In most cases, a decrease was observed in all measured parameters, but some differences were detected among treatments. Moreover, the values were quite similar at the last sampling regardless of treatment. A detailed discussion regarding electrical conductivity, ammonium, and DOC is offered in the following text.

123

282

Electrical Conductivity In the first sampling campaign, the electrical conductivity of leachates (ECL) from samples treated with pig slurry was almost twice that of controls (25.9 and 12.7 mS cm-1, respectively), and cropping increased significantly ECL in leachates (5.9 and 14.7 %, respectively) in samples treated with pig slurry and uncropped control, respectively (Fig. 1a). In the second sampling campaign, ECL decreased in all treatments, but it remained greater in treatments receiving swine slurry than in the controls (13.5 vs. 7.4 mS cm-1), and now cropping decreased ECL (31 %) but only in treatments that did not receive pig sludge. Finally, in the third sampling campaign, ECL decreased less than in the second sampling; the ECL of experimental units receiving pig slurry remained greater than the ECL of the controls; and cropping again caused a significant decrease in ECL in both experimental units where slurry was added and the controls (*40 %). These results are consistent with the swine’s slurry high content of watersoluble ions, which are mobile and transported by the irrigation water with almost no interaction with soil. Ammonium The NH4? content in leachates (NH4?L) followed a similar pattern to that observed for the ECL, i.e.: (1) NH4?L decreased over time; (2) NH4?L was greater in treatments where slurry was added (with and without crop) than in the controls; and (3) cropping caused NH4?L to decrease in the third sampling campaign, although in this case the decrease was only significant for samples treated with pig slurry (Fig. 1b). The relatively high NH4?L content in treatments receiving pig slurry in the first sampling campaign indicates a rapid mineralization of an organic material rich in nitrogen, which is consistent with the ratio of carbon to nitrogen in pig slurry (*6.5). Moreover, part of this ammonium had not yet been oxidized to nitrate. The presence of ammonium in leachates also indicates the presence of competing cations in the soil’s CAC. In the second sampling, the effect of the application of pig slurry was much less pronounced, and it was imperceptible in the third. The greater decrease of NH4?L in the cropped experimental units in the last sampling is consistent with plant uptake. DOC In the first two samplings, DOCL behaved similarly to ECL and NH4?L. However, in the last sampling DOCL behaved differently insofar as cropping increased significantly DOCL in both the control (by 230 %) and treatments where pig slurry was added (by 37 %) (Fig. 1c). This could

123

Arch Environ Contam Toxicol (2014) 66:277–286

indicate the existence of two sources of DOC: (1) pig slurry, which would account for the majority; and (2) root exudates, which were most evident in the last sampling data. This is consistent with the fact that in soil with lower nutrient content, root exudate is greater (Dakora and Phillips 2002). Mobility of TE in Agricultural Soil Figure 2 shows the copper, zinc, and selenium concentrations in the leachates in the three samplings. Like the bulk parameters, TE concentrations decreased in accordance with the volume of eluted leachate. However, certain differences were observed between the different TEs analyzed. They are discussed in the following text. Copper In the first sampling, the mean copper content in leachates (CuL) was 15 % greater in treatments receiving swine slurry than in controls. Cropping increased CuL by 9 % in samples treated with swine slurry and by 20 % in untreated samples. In the second sampling, CuL decreased significantly in samples treated with swine slurry (with and without crop) and in the cropped control, which showed CuL values 50 % lower than those in the uncropped control. In the third sampling campaign, CuL continued to decrease in samples treated with swine slurry, and cropping increased the Cu content in leachates in both treated and untreated samples with swine slurry. These results are consistent with the poor mobility of copper from pig slurry compared with native copper in soil, which appears to be more mobile. It could be attributable to the fact that copper can occur as different chemical species with different mobilities in soil depending on the source. Because the pig slurry used in this experiment was thermally dehydrated after an anaerobic digestion process, the highly insoluble CuO form would predominate, whereas native copper in soil might occur as oxidized forms (Cu?/Cu2?), which can form water-soluble complexes with ammonia. Moreover, copper can form insoluble compounds (copper sulphide) with the sulphur species present in manure or soil (Legros et al. 2013). However, these results might suggest that the greater the production of root exudates (Fig. 1c), the greater the presence of copper in leachates (Fig 1a), which is consistent with findings for other dicotyledons (Degryse et al. 2012). Zinc In the first sampling, the mean zinc content in leachates (ZnL) was much greater (130 %) in treatments where pig slurry was added (with and without crop) than in controls.

(a)

283

1.5

Arch Environ Contam Toxicol (2014) 66:277–286

The effect of cultivation was only perceived in the uncropped control, in which cropping led to a 23 % increase in ZnL. In the second sampling, ZnL decreased in all treatments. Finally, in the third sampling, the zinc content in leachates continued to decrease but at a more moderate rate. These results evidence a significant ZnL input from swine slurry with a greater mobility than copper, which is consistent with findings in the literature (e.g., Legros et al. 2013). Here it could be explained by the fact that zinc forms complexes with organic molecules from pig slurry more easily than does copper. This behavior could be attributable to the occurrence of oxidized water-soluble species that can interact with DOC to form chelates, which are much more mobile than positively charged TE species.

Pig slurry & crop Pig slurry Cropped control

-1

Cu (mg L )

1.0

Uncropped control

a ab b c 0.5

a

a b b

0.0

c

10

15

20

25

30

35

40

45

(b)

4.0

Days from planting

Pig slurry & crop

a

Pig slurry

Uncropped control

2.0

-1

Zn (mg L )

3.0

Cropped control

b

1.0

c a b 0.0

a b 10

15

20

25

30

35

40

45

(c)

0.4

Days from planting

Pig slurry & crop

0.3

Pig slurry

a

The selenium content in leachates (SeL) was much lower than that observed for zinc and copper. In the first sampling, SeL was much greater (36 %) in treatments receiving pig slurry (with and without crop) than in controls. No cropping effect was observed. In the second sampling, SeL decreased, especially in samples treated with swine slurry, such that SeL values were similar in all treatments. The same trend was observed in the third sampling; however, in that sampling, a slightly greater SeL content was found in the uncropped samples treated with swine slurry than in the other treatments. These results show that swine slurry contributes moderately to SeL content and, in the second sampling, that selenium content was convergent regardless of treatment. Probably the occurrence of selenium as the negatively charged selenite or selenate and its organic forms would facilitate its transport due to the repulsion with the negatively charged soil silicates. However, the low concentrations of selenium found in swine slurry and its relatively low toxicity do not pose any risk to groundwater in soil that is not rich in natural selenium.

Cropped control

Multivariate Statistical Analysis

Uncropped control

-1

Se (mg L )

Selenium

0.2

b

a

0.0

0.1

b

10

15

20

25

30

35

40

45

Days from planting

Fig. 2 Temporal evolution of a Cu, b Zn and c Se content in leachates over three sampling dates. At each sampling date, different letters indicate significant differences between treatments (a = 0.05)

PCA of data from the first sampling showed that the first two PCs (PC1 and PC2) accounted for 97 % of the total variability. Specifically, PC1 accounted for 87 % and PC2 for 10 %. Indeed, the biplot shown in Fig. 3 indicates that the bulk parameters, zinc and selenium, were strongly correlated with each other as well as with PC1. In contrast, copper behaved somewhat differently. Its variability was also limited to PC2 (which had a much lower weight). The biplot also shows that experimental units receiving swine slurry and controls were clearly separated in the direction of PC1, i.e., the greater bulk parameter values in leachates were associated with samples treated with swine slurry. Although the DOC from root exudates in the first sampling

123

284

Arch Environ Contam Toxicol (2014) 66:277–286 Cu Pig slurry & crop

4

1

(a)

Pig slurry Cropped control

2

0.5

Uncropped control

0

0

Se EC + NH4 , DOC

-2

-0.5

Zn

seen between ECL and NH4?; however, they were less strongly associated with DOC. The association between zinc and selenium and the bulk parameters decreased, and, as before, copper behaved somewhat differently. As in the first sampling, the biplot shows a clear separation between the data from the control and the treatments receiving pig slurry. Finally, the PCA results for the third sampling were similar to those for the second sampling campaign (not shown).

-4

-1

Conclusion

Treatment scores -4 -1

-2

0

2

4

-0.5

0

0.5

1

Pig slurry & crop

Cu

4

(b)

1

Parameter loadings (arrows)

Pig slurry

Se

Cropped control

2 0 -2 -4

0.5 -0.5 -1

0

Uncropped control

Zn

EC, DOC + NH4

Treatment scores -4 -1

-2

0

2

4

-0.5

0

0.5

1

Parameter loadings (arrows)

Fig. 3 Principal components biplot showing bulk parameters and trace elements (as arrows) and scores of treatments over the first two principal components in the a first and b second sampling campaign. The first component is shown in the x-axis and the second in the y-axis

was much lower than that from the slurry, this is not reflected in the figure. Moreover, the fact that the copper content in leachates is not associated with the DOC shown in the figure may corroborate the hypothesis that the mobility of copper is more strongly associated with complexing molecules from roots than from the pig slurry (Fig. 3a). The statistical quality of the biplot for the second sampling is also high (accounting for 95 % of the data variability), but this time 63 % is found in PC1 and 32 % in PC2. As in the first sampling, a strong correlation can be

123

TEs used in pig feed, such as copper, zinc, and selenium, were detected in both treated and untreated swine slurry. The studied treatment processes (NDN, anaerobic digestion, anaerobic digestion–thermal treatment) did not significantly affect the slurry’s content of such metals, and only a moderate increase in concentration was detected, particularly in the case of zinc. Of the three elements analyzed, zinc exhibited the highest concentration in slurry and the highest mobility in soil. Given the high zinc concentrations in the slurry, it is thus a matter of concern in terms of groundwater contamination. Selenium was also mobile, but due to the low concentrations found in slurry and its lower toxicity, it does not seem to pose a significant environmental risk. However, in the EU a threshold value of 10 lg L-1 for selenium in the drinking water exists (European Directive 98/83/EC 1998). The effects of cropping on TE mobility in soil were case-specific but generally low; however, the presence of root exudates can increase TE mobility. Finally, it should be noted that in Europe, the rules that determine the agricultural application of livestock manures and slurries is exclusively geared to the risk of contamination of groundwater with nitrates (European Directive 91/676/EEC 1991); thus, the regulations affecting sewage sludge focus on the risk of soil contamination with TE (European Directive 86/278/EEC 1986). The results listed here and those of other investigators suggest that it would be advisable to consider the potential hazards of TE in livestock manures and slurries, specially taking into account that soil mobility of TE is enhanced by DOC. Zinc is of special concern due to its wide use as animal growth promoter and its mobility in the soil. As previously reported, decreasing the dietary supply of these elements to pigs would be the most effective way to control heavymetal content in swine manures and slurries and thus decrease the environmental hazard (Jin and Chang 2011). Moreover, the co-occurrence of TEs with antibiotics in manure posses an additional risk regarding to the release of antimicrobial-resistant genes (Ho¨lzel et al. 2012). Further research is needed to address environmental and health

Arch Environ Contam Toxicol (2014) 66:277–286

concerns associated with groundwater contamination associated with livestock waste disposal in agricultural soil because groundwater is used as drinking water in many countries. A particularly important research issue is the role of organic compounds and plant exudates, particularly in the mobility and transport of heavy metals, because the mechanisms involved are not yet clearly understood. Acknowledgments Funding was obtained from the Spanish Ministry of Science and Innovation (VALPUR Project, TRA_0279) and the Catalan Ministry of Farming, Livestock, Fisheries, Food, and the Natural Environment. Technical assistance was provided by Miquel Massip of Agro`polis (UPC-Viladecans, Catalonia, Spain) and Yolanda Rodrı´guez. A. Sahuquillo from the University of Barcelona is kindly acknowledged for sample homogenization in the MAT Control laboratory. Conflict of interest Authors state that we do not have any financial relationship with the Spanish Ministryof Science and Innovation that funded this research.

References Aldrich AP, Kistler D, Sigg L (2002) Speciation of Cu and Zn in drainage water from agricultural soil. Environ Sci Technol 36:4824–4830 ´ lvarez-Salgado XA, Miller AEJ (1998) Simultaneous determination A of dissolved organic carbon and total dissolved nitrogen in seawater by high temperature catalytic oxidation: conditions for precise shipboard measurements. Mar Chem 62:325–333 Bais HP, Weir TL, Perry LG, Gilroy S, Vivanco JM (2006) The role of root exudates in rhizosphere interactions with plants and other organisms. Annu Rev Plant Biol 57:233–266 Banks MK, Scwab AP, Henderson C (2006) Leaching and reduction of chromium in soil as affected by soil organic content and plants. Chemosphere 62:255–264 Bao Q, Lin Q, Tian G, Wang G, Yu J, Peng G (2011) Copper distribution in water-dispersible colloids of swine manure and its transport through quartz. J Hazard Mat 186:1660–1666 Belon E, Boisson M, Deportes IZ, Eglin TK, Feix I, Bispo AO et al (2012) An inventory of trace elements inputs to French agricultural soils. Sci Total Environ 439:87–95 Bernet N, Be´line F (2009) Challenges and innovations on biological treatment of livestock effluents. Bioresour Technol 100: 5431–5436 Boletı´n Oficial del Estado (BOE) 171 (2005) Royal Decree R.D. 824/2005 on fertilizers [in Spanish], pp 25592–25669 Burt R (ed.) (2004) Soil Survey Laboratory methods manual. Soil Survey Investigations Report No. 42, Version 4.0. United States Department of Agriculture Natural Resources Conservation Service Combalbert S, Bellet V, Dabert P, Bernet N, Balaguer P, HernadezRaquet G (2012) Fate of steroid hormones and endocrine activities in swine manure disposal and treatment facilities. Water Res 46:895–906 Cortellini L, Piccinini S (1993) Determinazione delle caratteristiche chimiche dei reflui zootecnici. In: Manuale per la Gestione e l’Utilizzazione Agronomica dei Reflui Zootecnici. Centro Ricerche Produzioni Animali, Regione Emilia Romagna, Assessorato all’Agricoltura, pp 27–42 Dakora FD, Phillips DP (2002) Root exudates as mediators of mineral acquisition in low-nutrient environments. Plant Soil 245:35–47

285 Degryse F, Verma VKE, Smolders E (2012) Mobilization of Cu and Zn by root exudates of dicotyledonous plants in resin-buffered solutions and in soil. Plant Soil 306:69–84 European Directive 2008/1/EC of the European Parliament and of the Council (2008) Integrated pollution prevention and control, OJ L 24 European Directive 86/278/EEC (1986) Protection of the environment, and in particular of the soil, when sewage sludge is used in agriculture, OJ L 181, pp 6–12 European Directive 91/676/EEC (1991) Protection of waters against pollution caused by nitrates from agricultural sources, OJ L 375, pp 1–8 European Directive 98/83/EC (1998) Quality of water intended for human consumption, OJ L 330, pp 32–54 Flotats X, Bonmatı´ A, Ferna´ndez B, Magrı´ A (2009) Manure treatment technologies: on-farm versus centralized strategies. NE Spain as case study. Bioresour Technol 100:5519–5526 Hao X, Zhou D, Sun L, Li L, Zhang H (2008) Dynamic chemical characteristics of soil solution after pig manure application: a column study. J Environ Sci Health B 43:429–436 Ho¨lzel CSA, Mu¨ller CB, Harms KSA, Mikolajewski SC, Scha¨fer SA, Schwaiger KA et al (2012) Heavy metals in liquid pig manure in light of bacterial antimicrobial resistance. Environ Res 113:21–27 Hseu ZY (2004) Evaluating heavy metal contents in nine composts using four digestion methods. Bioresour Technol 95:53–59 Hsu JH, Lo SL (2001) Effect of composting on characterization and leaching of copper, manganese, and zinc from swine manure. Environ Pollut 114:119–127 Jin H, Chang Z (2011) Distribution of heavy metal contents and chemical fractions in anaerobically digested manure slurry. Appl Biochem Biotechnol 164:268–282 Karathanasis AD (1999) Subsurface migration of copper and zinc mediated by soil colloids. Soil Sci Soc Am J 63:830–838 Kassaye YA, Skipperud L, Meland S, Dadebo E, Einset J, Salbu B (2012) Trace element mobility and transfer to vegetation within the Ethiopian Rift Valley lake areas. J Environ Monit 14:2698–2709 Krapac IG, Dey WS, Roy WR, Smyth CA, Storment E, Sargent SL et al (2002) Impacts of swine manure pits on groundwater quality. Environ Pollut 120:475–492 Kuligowski K, Poulsen TG, Stoholm P, Pind N, Laursen J (2008) Nutrients and heavy metals distribution in thermally treated pig manure. Waste Manag Res 26:347–354 Ku¨mmerer K (2009) Antibiotics in the aquatic environment—A review: Part I. Chemosphere 75:417–434 Kunz A, Miele M, Steinmetz RLR (2009) Advanced swine manure treatment and utilization in Brazil. Bioresour Technol 100: 5485–5489 Legros S, Doelsch E, Feder F, Moussard G, Sansoulet J, Gaudet J-P et al (2013) Fate and behavior of Cu and Zn from pig slurry spreading in a tropical water-soil-plant system. Agric Ecosyst Environ 164:70–79 Li Z, Shuman LM (1997) Mobility of Zn, Cd and Pb in soils as affected by poultry litter extract—I leaching in soil columns. Environ Pollut 95:219–226 Li F, Ni L, Yuan J, Sheng GD (2010) Cultivation practices affect heavy metal migration between soil and Vicia faba (broad bean). Chemosphere 80:1393–1398 Mantovi P, Bonazzi G, Maestri E, Marmiroli N (2003) Accumulation of copper and zinc from liquid manure in agricultural soils and crop plants. Plant Soil 250:249–257 Miamiao H, Wenhong L, Xinqiang L, Donglei W, Guangming T (2009) Effect of composting process on phytotoxicity and speciation of copper, zinc and lead in sewage sludge and swine manure. Waste Manag 29:590–597 Nasir IM, Mohd Ghazi TI, Omar R (2012) Anaerobic digestion technology in livestock manure treatment for biogas production: a review. Eng Life Sci 12:258–269

123

286 Nicholson FA, Smith SR, Alloway BJ, Carlton-Smith C, Chambers BJ (2003) An inventory of heavy metals inputs to agricultural soils in England and Wales. Sci Total Environ 311:205–219 Nolan T, Troy SM, Gilkinson S, Frost P, Xie S, Zhan X et al (2012) Economic analysis of pig manure treatment options in Ireland. Bioresour Technol 105:15–23 Sager M (2007) Trace and nutrient elements in manure, dung and compost samples in Austria. Soil Biol Biochem 39:1383–1390 Soil Survey Staff (1998) Keys to soil taxonomy, 8th edn. United States Department of Agriculture Natural Resource Conservation Service, United States Government Printing Office, Washington DC Steinmetz RLR, Kunz A, Drensler VL, Flores EMD, Martins AF (2009) Study of metal distribution in raw and screened swine manure. Clean 37:239–244 Stone JJ, Oswald AS, Lupo CHD, Clay AA, Mott HV (2011) Impact of chlortetracycline on sequencing batch reactor performance for swine manure treatment. Bioresour Technol 102:7807–7814

123

Arch Environ Contam Toxicol (2014) 66:277–286 United States Environmental Protection Agency (USEPA) 40 CFR Part 503—Standards for use and disposal of sewage sludge: Final rules (1993) Federal Register 58:9248–9415 Viancelli A, Kunz A, Steinmetz RLR, Kich JD, Souza CK, Canal CW et al (2013) Performance of two swine manure treatment systems on chemical composition and on the reduction of pathogens. Chemosphere 90:1539–1544 Xue H, Sigg L, Ga¨chter R (2000) Transport of Cu, Zn, and Cd in a small agricultural catchment. Water Res 34:2558–2568 Zhang F, Li Y, Yang M, Li W (2012) Content of heavy metals in animal feeds and manures from farms of different scales in Northeast China. Int J Environ Res Public Health 9:2658–2668