1 Introduction

Li et al. (2016a) analyzed all of the publications on sewage sludge from 1991 to 2012 (using the Science Citation Index Expanded) and found that among the most popular disposal and treatment methods mentioned were composting and anaerobic digestion. However, the soil application was suggested as the most important approach to sewage sludge disposal. Although heavy metals were the contaminants of greatest concern for several decades, PPCPs (pharmaceuticals and personal care products) and PBDEs (polybrominated diphenyl ethers) showed comprehensive attention in recent years. McClellan and Halden (2010) showed that antibiotics such as triclocarban, triclosan, ciprofloxacin, and ofloxacin, followed by a number of tetracycline antibiotics, were the prevalent contaminants in biosolids. A comparison of different sludge processing methods showed galaxolide (representing musk compounds) and polybrominated diphenyl ether BDE-209 (representing PBDEs) among the substances most frequently detected under different treatment processes (Guerra et al. 2015). However, treatments such as composting or anaerobic fermentation should be at least partially effective for the elimination of PPCPs in biosolids and farmyard manure (Amarakoon et al. 2016; Ray et al. 2017; Spielmeyer et al. 2017).

Based on present knowledge, Roccaro and Vagliasindi (2014) assumed that the health risk related to the presence of PPCPs in biosolids used for land application is very low. However, the information concerning their toxicity, ecotoxicity, increased resistance of bacteria, possible synergistic effects due to the presence of emerging organic contaminant mixtures, etc., is limited and needs further research. In particular, the possible interrelationships between pollutants should be taken into account. For instance, risk elements such as Pb were observed by Li et al. (2011) to have an effect on the sorption of pharmaceuticals onto sediments. The potential transport of PPCPs and veterinary drugs into runoff after land application of biosolids from municipal treatment plants and manure from livestock or poultry production has already been investigated, as well (Topp et al. 2010). The suppressive effect of veterinary antibiotics on soil microbiota has already been documented. Yi et al. (2016) observed decreasing soil microbial cellulose decomposition activity and microbial respiration as well as ammonification with the addition of tetracycline, chlortetracycline, norfloxacin, and enrofloxacin, i.e., both human and veterinary antibiotics. The soil amended with a single application of antibiotics resulted in changed soil microbial diversity; however, communities appeared to converge over time, and no long-term significant effect was detected with the single application (Nelson et al. 2011). However, as a long-term effect of the antibiotics in the soil, an increased level of antibiotic resistance genes and shifts in the microbial community structure has been documented (Cleary et al. 2016).

For the most common persistent organic pollutants (POPs), such as polychlorinated dibenzo-p-dioxins/furans (PCDD/Fs), polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), organochlorine pesticides (OCPs), hexachlorocyclohexanes (HCHs) isomers, and the dichlorodiphenyl trichloroethane (DDT) family, the soil can represent an important reservoir of these compounds (Rychen et al. 2014; Clarke and Cummins 2015). In this context, sewage sludge is considered as the material that is able to increase the input of POPs into the soil (Elskens et al. 2013). The half-lives of PCBs and PCDD/Fs in sewage sludge vary between 9 and 12 years (Zennegg et al. 2013). Thus, the loads of these compounds in the soil can increase with repeated application of sewage sludge to the soil. In addition to these compounds, recently investigated pollutants such as PBDEs should be taken into account. The distribution of PBDE congeners in sewage sludge was published by Lee et al. (2014) who showed the dominant content of BDE-209 in all sludge samples. After BDE-209, relatively high levels were found for BDE-28 and -47 from industrial sludge, BDE-47 and -99 from domestic wastewater sludge, and BDE-206, -207, and -208 from livestock sludge. The contents of PBDEs in soils were measured at various industrial and electronic waste recycling sites. Total PBDE concentrations were dominated by BDE-209, while BDE-17, -85, -138, -191, -204, and -205 were the least abundant compounds (Li et al. 2016b).

Long-term (for several decades) fertilizer application and crop rotation experiments were established in many countries for the assessment of the sustainability of crop production on a particular soil type, in most cases to study the effects of different soil-climate conditions, the effects of various fertilizers, and crop rotation on soil properties and the nutrient content of crops (Körschens 2006; Madaras et al. 2013). However, a systematic assessment of the potential impact of fertilizers on the contents and accumulation of anthropogenically-derived organic pollutants is lacking. The fate and behavior of sewage sludge–derived contaminants in the soil, i.e., their persistence, mobility, leaching, plant uptake, and biodegradation depend on various factors such as soil properties, sewage sludge composition, contaminant properties, and interrelationships among the compounds (Oleszczuk 2007).

This experiment is based on the samples collected from four long-term established experimental sites in the Czech Republic with different soil and climate characteristics treated with various mineral and organic fertilizers, including sewage sludge. The main goal of our study was to investigate the direct as well as subsequent effects of different treatments on the accumulation of a large group of contaminants belonging to POPs and PPCPs and to assess the abundance of the individual compounds in the treated soils as a result of the long-term regular pollutant load via sewage sludge and farmyard manure applications.

2 Material and methods

2.1 Experimental sites, soil sampling, and characterization

The long-term field experiment (since 1996) established at four sites in the Czech Republic (Hněvčeves, Humpolec, Lukavec, Praha-Suchdol) with different soil and climate characteristics was used for this study. The whole design of the experimental tool and procedure as well as the main characteristics of the experimental sites and soils are described in detail elsewhere (Kulhánek et al. 2014; Vašák et al. 2015; Száková et al. 2018), and in the Supplementary material (Table S1, Electronic Supplementary Material—ESM). Composite soil samples consisting of ten subsamples collected throughout the plot to a depth of 30 cm were collected during April 2016; therefore, three composite soil samples were prepared for each treatment. Soil samples were air-dried and sieved through a 2-mm mesh. More detailed characteristics of the soils are summarized in the Supplementary material (Table S2—ESM). The individual treatments were as follows: (i) NPK mineral fertilization; (ii) sewage sludge with Lowe application, labeled SS1; (iii) sewage sludge with higher application rate, labeled SS2; (iv) farmyard manure, labeled FYM; and (v) untreated control, labeled Control.

2.2 Analytical methods

2.2.1 Chemicals

Certified standards with purity higher than 98% of PCBs (Nos. 28, 52, 101, 118, 138, 153, 170, and 180) dissolved in isooctane were provided by Dr. Ehrenstorfer, GmbH (Germany) as well as selected OCPs in the solid phase: isomers of hexachlorocyclohexane (α-, β-, γ-, δ-HCH), hexachlorobenzene (HCB), dichlorodiphenyltrichloroethane (o,p′-, p,p ′-DDT), dichlorodiphenyldichloroethan (o,p′-, p,p′-DDD), and dichlorodiphenyldichloroethylene (o,p′-, p,p′-DDE). Certified standards of selected synthetic musk compounds (SMs) were purchased from Neochema GmbH (Germany): galaxolide, tonalide, and cashmeran. Certified standards of 13 polybrominated diphenyl ethers congeners (# 47, 49, 99, 100, 153, 154, 183, 196, 197, 203, 206, 207, and 209) were provided by Wellington Laboratories, Inc. (Canada).

The following certified standards were purchased from Wellington Laboratories, Inc. (Canada): perfluoroheptanoic acid (PFHpA), perfluorooctanoic acid (PFOA), perfluorodecanoic acid (PFDA), and sodium salt of perfluorooctanesulfonic acid (linear and branched isomers of PFOS). Certified standards of 12 priority PAHs were supplied by the National Institute of Standards and Technology (NIST, Gaithersburg, USA): anthracene (AN), benz[a]anthracene (BaA), benzo[a]pyrene (BaP), benzo[k]fluoranthene (BkFA), fluoranthene (FLT), benzo[b]fluoranthene (BbFA), chrysene (CHR), benzo[ghi]perylene (BghiP), dibenz[a,h]anthracene (DBahA), indeno[1,2,3-cd]pyrene (IP), phenanthrene (PHE), and pyrene (PY).

Methanol, acetonitrile, and ethyl-acetate were provided by Honeywell (USA). Hexane, dichloromethane, and isooctane were supplied by Sigma-Aldrich (Merck, Germany). Sodium chloride and magnesium sulfate were provided by Lach-Ner s.r.o. (Czech Republic).

The performance characteristics of all the following analytical methods are tabulated in Table S3 (ESM).

2.2.2 Determination of PCBs, OCPs, PAHs, SMs

A mixture of 10 g of soil sample and anhydrous sodium sulfate (Na2SO4) was prepared and transferred into a cellulose extraction thimble. Then, the sample was extracted for 7 h by dichloromethane. The extract was evaporated and dissolved in a mixture of cyclohexane:ethyl-acetate (1:1, v/v) and purified using gel permeation chromatography in a Bio-Beads S-X3 column (styrene-divinylbenzene copolymer) with a mixture of cyclohexane:ethyl-acetate (1:1, v/v) employed as the mobile phase. A purified extract was also evaporated to dryness, re-dissolved in isooctane, and transferred into a vial for analysis of PCBs, OCPS, BFRs, and SMs using gas chromatography coupled to (tandem) mass spectrometry (GC-MS(/MS)), as described by Száková et al. (2018). In the case of PAHs, a crude extract was purified separately also using a Bio Beads S-X3 column, and after evaporation, the collected fraction was re-dissolved in acetonitrile and analyzed using high-performance liquid chromatography with a fluorescence detector (HPLC-FLD), for details see Száková et al. (2018).

2.2.3 Determination of PFASs

A mixture was prepared by transferring 5 g of soil sample into a 50-mL polypropylene tube, adding 10 mL of acetonitrile and 10 mL of deionized water, and shaking the mixture for 1 min. Subsequently, 1.5 g of NaCl and 6 g of MgSO4 were added into the tube with the sample; the mixture was shaken once again for 1 min and then centrifuged for 5 min at 10,000 RPM. Into a second 50-mL PP tube, 5 mL of the upper organic layer was transferred, and the solvent was evaporated to dryness. The residue was re-dissolved in methanol and transferred into a vial for subsequent analysis by ultra-high performance chromatography coupled to tandem mass spectrometry (UHPLC-MS/MS), see Száková et al. (2018).

2.2.4 Determination of antibiotics

Soil samples were prepared by dilution of a dry sample (5 g) with 2.5 mL of 0.05 M EDTA, 2.5 mL of McIlvaine buffer (pH 4), and 5 mL of methanol. Diluted samples were properly shaken for 1 min and then filtered using microfilters to ensure that all solid particles were removed prior to UHPLC-MS/MS analysis (Acquity UPLC with mass detector Xevo TQ-S (Waters, USA)).

3 Results and discussion

3.1 Polychlorinated biphenyls (PCBs)

The results showed detectable levels of all of the measured congeners in all of the analyzed samples, except for PCB 52 (detection frequency 95%), and the total PCB contents in the soils varied between 0.44 and 12.1 μg/kg dw (Table 1). In the Czech Republic, the preventive values of PCBs agricultural soils are given by the public notice characterizing the conditions for the protection of agricultural soil quality (Public Notice No. 153/2016 2016). The maximum values of the sum of seven congeners (PCB 28, 52, 101, 118, 138, 153, and 180) did not exceed the preventive value in soil, i.e., 20 μg/kg dw. Among the analyzed congeners, the highly chlorinated compounds presented the most important concentrations; No. 180 varied between 0.05 and 2.93 μg/kg dw, No. 153 between 0.08 and 3.42 μg/kg dw, and No. 138 between 0.11 and 3.63 μg/kg dw. The results also documented a difference between elevated PCB contents in the soil at the Suchdol location (in all of the samples including the untreated control) compared to the other locations with low levels of the sum of the investigated congeners in the untreated control samples. Profiles of PCBs Nos. 138, 153, and 180 and their levels were comparable between the Humpolec, Lukavec, and Hněvčeves locations and between the methods of fertilization and were 2–3 times lower compared to the Suchdol location. Moreover, the concentrations of all PCBs were very similar between the untreated control, fertilization with NPK, and fertilization with FYM, which shows that NPK and FYM are not significant sources of PCBs and their contribution to the overall levels of PCBs is negligible. The elevated overall levels of PCBs at the Suchdol location could be connected with the proximity of the most industrialized city in the Czech Republic, Prague. This statement could be supported by findings for urban soils in London, United Kingdom, where the sum of three congeners (Nos. 138, 153, and 180) reached up to 12.4 μg/kg dw (Vane et al. 2014). In this study, the maximum value of the sum of these three congeners in soil at the Suchdol location was 8.9 μg/kg dw, indicating similar levels of PCBs in both cities. Results originating from an industrial area in China (from 1.46 to 19.2 μg/kg dw) also showed the impact of industrial activities on the PCB level in agricultural soils (Zhu et al. 2014).

Table 1 Median concentrations of detected PCBs and OCPs in soils; n = 3; concentrations of HCBD, β-HCH, γ-HCH, and δ-HCH were below the limit of quantification or were detected in less than 50% of samples
Full size table

Regardless of the location, the results showed increased PCB contents in the soils treated with SS compared to all of the remaining variants, and the increase was related to the sludge application rate in the soil. The concentrations of PCBs were 1.8 times higher in the soil fertilized with SS2 compared to soil fertilized with SS1. The PCB content in the sludge exceeded the soil PCB levels ca. 150-fold, confirming that SS could be an important source of PCBs in agricultural soil. These statements are only speculative because the regularly applied SSs were not monitored from the beginning of the long-term experiment. However, the potential impact of SS application on the PCB levels in soils could be estimated according to the previously published results. Previous findings pointed out that the contents of dioxin-like PCBs (DL-PCBs) in various fertilizers in Belgium showed that the contents of these compounds in SS were the highest among both organic and inorganic fertilizers (Elskens et al. 2013). However, these authors assumed that their contribution to the overall soil contamination was low. Low PCB contents were reported by Abad et al. (2005) in various SSs for agricultural purpose in Spain, as well. Sludge from anaerobic wastewater stabilization ponds was even determined by Santos et al. (2009) to have no detectable PCB contents. Moreover, many authors have documented the degradability of SS-derived PCBs. For instance, a decrease in the soil PCB content was observed in composted SS-treated soil compared to the original soil before the treatment (Antolin-Rodriguez et al. 2016) indicating the positive stabilization effect of the SS-derived organic matter. Tillage could also be a factor in decreasing the accumulation of PCBs in the upper layer of soils (During and Gath 2002). The biodegradation rate of the SS-derived PCBs in soil can substantially decrease within 128 days of sludge amendment (Wilson et al. 1997). However, our results indicate substantial persistence of sludge-derived PCBs in the long-term trials if the sludge was regularly and repeatedly applied. Thus, the long-term experiment can allow us to make a more reasonable assessment of the potential impact of SS on the soil PCB levels. The long-term effect (39 years) of SS application on DL-PCBs in the soil was investigated by Umlauf et al. (2011). They found that the soils treated with SS showed higher values compared to those treated with a mineral fertilizer, but both were still below the guideline value of these compounds for arable land. Thus, they stated that the common practice of SS application in Germany does not pose a current threat to the agro-environment with regard to DL-PCBs. However, they highlighted the necessity of investigation into the time-dependent changes in pollutant contents in long-term experiments. Our results could be concluded by the same statement: the PCB levels in the SS-treated soils do not represent a direct environmental risk, but regular assessment of the potential changes in the PCB contents in the sludge-soil system should be carried out.

3.2 Organochlorinated pesticides (OCPs)

Except for HBCD, all of the targeted OCPs were found in the soil samples at detectable levels (Table 1). Among them, α-HCH, HCB, DDT isomers (o,p′-DDT, p,p′-DDT), DDE isomers (o,p′-DDE, p,p′-DDE), and one DDD isomer (p,p′-DDD) were identified in all of the analyzed samples. Minor amounts of o,p′-DDD and pentachlorobenzene were identified in 97% and 60% of the soil samples, respectively. The total OCP content in the soils varied between 3.31 and 51.3 μg/kg dw, and the most abundant compounds were DDE isomers representing from 23 to 60% of the total OCP content. Thus, the OCP contents did not exceed the preventive value of these compounds in soil given by Singh and Agrawal (2008), i.e., 75 μg/kg dw. In the Czech Republic (Public Notice No. 153/2016 2016), however, the preventive value of ∑DDTs is 7.5 μg/kg and for HCB and α-HCH, it is 20 μg/kg and 10 μg/kg, respectively. Although the HCB and α-HCH contents were far below these limits, the sum of DDTs exceeded these limits in all of the samples at the Humpolec location and in numerous samples at the Lukavec location. However, the data did not reach the value which can directly threaten human and animal health, i.e., 8 mg/kg (Public Notice No. 153/2016 2016).

Comparing the individual locations, the highest OCP contents varying between 21.5 and 51.3 μg/kg dw were found at the Humpolec location, regardless of the soil treatment. OCP contents at Hněvčeves and Lukavec were quite similar, varying within the range 7.02–13.9 and 9.91–16.0 μg/kg dw, respectively; however, these values were 2–4 times lower compared to Humpolec. The lowest concentrations of OCPs were measured in soil samples originating from the Suchdol location: 3.34–7.30 μg/kg dw, which is 3–16 times lower compared to the most contaminated area. The amounts of chlorinated pesticides in the SS-treated soils from long-term experiments in the United Kingdom typically ranged from 0.1 to 10 μg/kg dw, and the highest concentrations were found for γ-HCH and p,p′-DDE (Meijer et al. 2001). In this experiment, the proportions of HCH isomers in the sludge-treated soils did not exceed 2% of the total OCP content. Thus, the long-term impact of SS application on the soil OCP content seems to be negligible, and the concentrations of these compounds could be connected with their intensive application as pesticides in the past.

DDD and DDE are considered to be products of the partial degradation of DDT, and a ratio of the DDT content to the sum of DDE and DDD that is less than one indicates past application of DDT as an insecticide. This ratio was recorded for all of the locations except Hněvčeves, where the ratio of DDT/(DDD + DDE) was > 1 indicating either more recent application of DDT (although the application of DDT has been banned for several decades), or a slow degradation rate of DDT compared to other locations. The last possibility should be supported by the presence of low content of total organic carbon (TOC) in this soil (Table S2—ESM), indicating possible lower activity of soil microbiome in these conditions. This speculation should be supported by the detailed microbiological analysis in further research. However, the results measured in the samples from this study did not reach the soil levels of DDD and DDE isomers observed in an area affected by the illegal application of these compounds (79 μg/kg dw) in Uganda (Sseburege et al. 2010). Odukkathil and Vasudevan (2013) summarized and reviewed the behavior of chlorinated pesticides in soils. According to the previous investigations, chlorinated pesticides like DDT and HCH are persistent in soil, and their half-lives range from 100 to 200 days. In this experiment, however, the detectable OCP contents were found many years after the application ban. Unfortunately, the application rates and/or duration of the application of OCPs at the individual locations during the period of permitted application of these chemicals before the establishment of the long-term experiment was not documented. Thus, we can only speculate that the locations differed in their OCP contents due to their different DDT rates.

3.3 Per- and polyfluoroalkylated substances (PFASs)

Among the 19 targeted PFASs, only branched isomers of PFOS (br-PFOS), linear PFOS (L-PFOS), PFOA, and PFHpA were detected in 90%, 83%, 43%, and 3% of the samples, respectively. Concentrations of the remaining PFASs were below the quantification limit (LOQ). The total PFOS content varied between < 0.1–2.46 μg/kg dw and that of PFOA between < 0.3–0.97 μg/kg dw (Table 2). The results showed no significant difference in the PFAS content among all of the locations, but fertilization of the soil with SS was the highest contributor to soil pollution with PFASs. These results indicate that SS is a significant source of these compounds and therefore has to be carefully checked before possible application to soil. No differences were observed in the PFAS content in the control soil and soil fertilized with the FYM or NPK. Another interesting fact is that the ratio between br- and L-PFOS was 1:4 in all locations and all types of fertilization, which corresponds to the ratio of PFOS isomers in the technical mixture. This result indicates that PFOS probably does not undergo transformation between its forms in soil.

Table 2 Median concentrations of detected PFASs and BFRs in soils; n = 3; concentrations of PFBA, PFPeA, PFHxA, PFNA, PFUDA, PFDoDA, PFTrDA, PFTeDA, PFBS, PFHxS, PFDS, PFOSA, N-MeFOSA, N-EtFOSA, BDE 28, BDE 66, BDE 85, BTBPE, DBDPE, HBB, α-HBCD, β-HBCD, γ-HBCD, OBIND, PBEB, and PBT were below the limit of quantification or were detected in less than 50% of samples
Full size table

Higher contents of PFOA and PFOS in soil were observed in a study from the USA (Xiao et al. 2015). PFOA and PFOS were detected in all samples, and concentrations ranged between 0.2–28.2 μg/kg dw for PFOA and 5.5–125.7 μg/kg dw for PFOS. Similar to our study, concentrations of PFOS were higher compared to PFOA. Another trend was identified in a study from Sweden (Plassmann and Berger 2013) in which the concentrations of PFOA (< 1.67–3.26 μg/kg dw) were higher compared to PFOS (< 0.147–0.272 μg/kg dw) but lower than in our study. In another survey conducted in Korea (Naile et al. 2013), higher concentrations of PFOA (< 0.2–6.4 μg/kg dw) compared to PFOS (< 0.2–1.7 μg/kg dw) were also measured. The results from Korea are comparable with the data from our study. It should be noted that accumulation of PFASs strongly depends on the soil organic carbon content and the available cations (Higgins and Luthy 2006; Hong et al. 2013), and this is probably the reason why the concentrations of PFASs among several studies are so different. In this experiment, however, the TOC levels of the untreated soils (Table S2—ESM) varied in relatively narrow range, between 0.93 (Hněvčeves) and 1.76% (Suchdol), thus, differences similar to the cited studies were not expected.

3.4 Brominated flame retardants (BFRs)

Among the group of BFRs, 15 compounds representing brominated diphenyl ethers (PBDEs) and other flame retardants (PBT and PBEB) and isomers of HBCD were detected in the examined soil samples, where the total contents of these compounds varied between 0.69 and 16.9 μg/kg dw (Table 2). The most abundant PBDEs were BDE-47 and -99 (both identified in 78% of the samples), γ-HBCD (68% of the samples), BDE-100 (62% of the samples), and BDE-183 (57% of the samples); the remaining congeners occurred at detectable levels in less than 50% of the samples. The individual congeners with highest contents were BDE-209 (from 0.5 to 5.88 μg/kg dw) and BDE-206 (from 0.05 to 2.78 μg/kg dw), but they were only detected in 30% of the samples. On the other hand, the concentrations of BDE-28, HBB, BTBPE, OBIND, and DBDPE were below the LOQ. Concentrations of BDE 47 and BDE 99 were somewhat higher in soil from the Suchdol location compared to other locations, whereas the concentrations in soil from Lukavec, Humpolec, and Hněvčeves were similar. The results also indicated the impact of SS application on the PBDE levels in the soil: increasing PBDE contents were observed with increasing sludge application rate. On the contrary, the PBDE levels in the soils treated by FYM and NPK did not differ from the control variants. Among the locations, the levels at Hněvčeves tended toward higher values compared to all of the remaining sites.

PBDE content is one of the soil quality parameters that is often investigated, and the levels determined by investigators are highly variable. Low contents of BDE-47 and -99 (not exceeding 1 μg/kg dw) were found in long-term sludge-treated soils in Sweden (Österas et al. 2015), whereas higher values (up to 17 μg/kg dw) were determined in rice fields in China (Wang et al. 2011). Similar to PCBs and OCPs, BFRs showed long persistence in the soil. Venkatesan and Halden (2014) found that the initial BFR amount in sludge/soil mixtures (1:2) persisted over the monitoring duration of 3 years. Thus, the repeated application of SS can be a source of potential accumulation of these compounds in soil. The possible enrichment of soils with BFRs should be regularly monitored also because of the potential intake of these compounds into the food chain. Uptake and transformation of BDE-99 by Oryza sativa, Triticum aestivum, and Glycine max have been observed via recordings of the corresponding homologous methoxylated metabolites (Chow et al. 2015; Pan et al. 2016).

3.5 PAHs

All 12 targeted compounds from the PAHs group were detected in all samples. PAHs are one of the main contributors to soil pollution, and their concentrations varied between 54.1 and 1443 μg/kg dw (Table 3). The public notice No. 153/2016 (2016) gives the preventive value of ∑12 PAHs in the soil at the level 1000 μg/kg. Fluorathene and pyrene were the most abundant PAHs, representing 12–22% and 11–17% of the total PAH content, respectively. The highest concentrations of PAHs, occasionally exceeding the preventive value, were found in soil samples from the Humpolec location, and concentrations in this location were 2–5 times higher compared to the others. These results imply a probable local source of PAHs pollution in Humpolec. No difference in the PAH contents was observed between the different types of fertilization, indicating that the pollution of soil with PAHs is more likely caused through atmospheric deposition than by fertilization, and SS represents only a negligible risk. The ratio between all PAHs in all samples is quite similar, and the ratio between B[a]A, CHR, B[b]F, and B[a]P was approximately 1:1:1:1 in all samples and all locations with no significant differences.

Table 3 Median concentrations of detected PAHs and SMs in soils; n = 3; concentrations of Celestolide, Musk Ambrette, Musk Ketone, Musk Moskene, Musk Tibetene, Musk Xylene, Phantolide, and Traseolide were below the limit of quantification
Full size table

3.6 Musk compounds

Among the targeted SMs, only galaxolide (most abundant, identified in 88% of the soil samples), tonalide (70% of the soil samples), and cashmeran (28% of the soil samples) were detected (Table 3). The total content of the determined compounds varied between 6.22 and 80.9 μg/kg dw. The predominant compound in all of the samples was galaxolide, varying between 1.6 and 58.9 μg/kg dw. Although the differences among the individual locations were negligible, the content of musk compounds increased with increasing SS application rate at all of the locations. Monitoring of PPCP contents in SSs from wastewater treatment plants in the United Kingdom showed a few high contents of total SMs ranging from 2.1 to 86 mg/kg dw. Thus, SS could be an important source of musk compounds in soil, where assessment of the potential environmental risk is aggravated by the high variability of these pollutants in sludge. The variability of musk compound levels in SS could be documented by the musk compound levels in the SS-treated soils as presented by other authors. Chen et al. (2014a) determined galaxolide and tonalide levels in Chinese soils reaching up to 82.4 μg/kg, whereas Yang and Metcalfe (2006) found levels of these compounds up to only 4.6 μg/kg in SS-treated soils in Canada.

The potential risk of the intake of SMs into the terrestrial food chain needs to be taken into account as well. The potential uptake of PPCPs by plant roots has already been proved. The introduction of persistent semi- and nonpolar pollutants such as musk compounds (galaxolide, tonalide) into the food chain via edible plants like carrots could be of particular relevance when sludge is applied as fertilizer (Macherius et al. 2012). The potential relationships of musk compounds (galaxolide) and risk elements (Cd) have also been observed in regard to the uptake of these substances by plants (Chen et al. 2014b). Thus, the uptake and translocation of risk elements by plants and their interactions with sludge-derived organic contaminants should also be included in comprehensive assessments of the potential impact of SS on agricultural soils.

3.7 Antibiotics

Within this contaminant group, ten chemicals, i.e., sulfadiazine, sulfathiazole, sulfamerazine, sulfamethoxazole, sulfadimidine, norfloxacin, enrofloxacin, tetracycline, oxytetracycline, and chlortetracycline, were targeted but not detected at concentrations above the LOQ. Hou et al. (2016) monitored the contents of human antibiotics including tetracyclines within the processing stages in pharmaceutical wastewater treatment plants, and although the antibiotics were partially degraded during the procedure, large amounts of tetracyclines were still discharged through the effluent (up to 32.0 ± 6.0 mg/L) and dewatered SS (up to 5481 ± 123 mg/kg). Limited biodegradation of human antibiotics in wastewater treatment plants was observed also by Plosz et al. (2010). On the contrary, Gao et al. (2012) showed the high efficiency of removal of antibiotics via biotransformation during the wastewater cleaning procedure and observed only an insignificant role of the sorption of these compounds onto the SS.

The contents of tetracyclines and sulfonamides in manures were monitored by An et al. (2015) in northeast China, where the highest concentration of antibiotics in manure was 143.9 mg/kg for chlortetracycline. In the manure-treated soils, the concentrations of antibiotics varied from 2.56 μg/kg dw of sulfadimidine to 1590 μg/kg dw of chlortetracycline. Ostermann et al. (2014) identified the hotspots of antibiotic-contaminated soils in China, where the concentrations reached 110 μg/kg of sulfamethazine, 111 μg/kg dw of chlortetracycline, and 62 μg/kg dw of enrofloxacin in the topsoil of agricultural fields.

The screening provided by Xiang et al. (2016) showed that a wide range of tetracycline compounds (oxytetracycline, tetracycline, chlortetracycline, and doxycycline) was present in the soil, ranging from 0.04 to 184.8 μg/kg dw. Ding et al. (2016) investigated the potential biodegradation of antibiotics and found that the removal of these substances from the soil solution showed complementary effects on the laccase-mediated oxidation and soil adsorption with the adsorption rate depending on the chemical structure of the antibiotics. Moreover, Allaire et al. (2006) observed that sorption of chlortetracycline in sandy loam and clay occurs very fast: more than 95% of the chlortetracycline adsorption was completed within 10 min. However, the sorption/desorption behavior of antibiotics (oxytetracycline) was substantially affected by the soil type and physicochemical properties (Feng et al. 2016). In contrast to the previous investigations, this study showed that neither manure nor SS application represents a risk for long-term loads of these contaminants in the treated soils. The lower use of antibiotics in the Czech Republic compared to other countries, and/or the degradation of these compounds within the wastewater cleaning procedure or in soil after SS and manure application can be speculated as the reasons for the low antibiotic contents in the long-term treated soils.

4 Conclusions

Regardless of the location and fertilizer application, the results showed two major patterns of the fate of contaminants in the soils (Fig. 1): (i) For PAHs and OCPs (especially the DDT family), the results indicated long-term soil contamination, most probably originating from the past before the establishment of the long-term experiment, as also stated by Száková et al. (2018). The DDT levels especially documented high persistence and low biodegradability even a few decades after the application ban. We can also speculate that the differences in the DDT contents among the individual locations are determined by the different loads of the contaminants at the individual locations rather than the potential differences in the biodegradability of the pollutants due to the different soil physicochemical and biological properties. (ii) For PFASs, BFRs, and SMs, the results indicated that the soil could become a sink for these compounds due to the application of sewage sludge, confirming the original assumption. Evidently, none of the levels of the targeted contaminants represent a direct environmental risk. However, the potential sorption/desorption and/or potential degradation pattern should be more intensively investigated to elucidate the potential risk of sewage sludge-derived pollution from the long-term perspective. Moreover, contaminants such as PFASs, BFRs, and SMs should be regularly monitored in the soils and sewage sludge together with the “classical” contaminants (PAHs, OCPs, PCBs), i.e., compounds already monitored following the Public notice No. 153/2016 (2016).

Fig. 1
figure 1figure 1

A comparison of the sums of the individual groups of pollutants at the individual locations as affected by the individual treatments

Full size image