The polycyclic aromatic hydrocarbon concentrations in soils in the Region of Valasske Mezirici, the Czech Republic

The polycyclic aromatic hydrocarbon (PAH) contamination of urban, agricultural and forest soil samples was investigated from samples obtained in the surroundings of Valasske Mezirici. Valasske Mezirici is a town located in the north-east mountainous part of the Czech Republic, where a coal tar refinery is situated. 16 PAHs listed in the US EPA were investigated. Organic oxidizable carbon was also observed in the forest soils. The PAH concentrations ranged from 0.86-10.84 (with one anomalous value of 35.14) and 7.66-79.39 mg/kg dm in the urban/agricultural and forest soils, respectively. While the PAH levels in the urban/agricultural soils are within the range typically found in industrialized areas, the forest soils showed elevated PAH concentrations compared to other forest soils in Western and Northern Europe. The PAH concentrations and their molecular distribution ratios were studied as functions of the sample location and the meteorological history. The soils from localities at higher altitudes above sea level have the highest PAH concentrations, and the PAH concentrations decrease with increasing distance from the town.


Background
Many sites in the Czech Republic have been negatively affected by industrial pollution. The town of Valasske Mezirici and its surroundings, in the northeast of the republic, is one of these sites. In the 1960s, a coal tar refinery was established there. Coal tar refineries are known sources of pollution; among the pollutants they produce are various organic compounds, including polycyclic aromatic hydrocarbons (PAHs).
The PAHs form a group of chemical compounds that are ubiquitous in the environment [1]. These compounds are well known for their characteristic properties, such as toxicity and carcinogenity, environmental persistence and tendency for bioaccumulation [1][2][3][4][5]. They form through the incomplete combustion or pyrolysis of organic matter, and their release into the atmosphere is connected with energy and heat production, local heating facilities, vehicle exhausts, refuse burning, coke ovens, and so on. They also enter the environment via natural processes such as volcanic activity or forest and prairie fires [6,7]. After being emitted into the atmosphere, they are redistributed between the gas and particle phases, and subsequently deposited to the terrestrial environment through dry or The town is an important industrial center and a significant railway and road junction, with a relatively high traffic density. The main industrial source of PAH pollution is the coal tar refinery (DEZA Corporation) and the carbon black production plant (CABOT CS). Both of these sources are located in the north of the town (Figure 1). DEZA a.s. is the only Czech producer of aromatic hydrocarbons that are obtained by processing coal tar and benzol. The annual processing capacity of DEZA is 160,000 Mt of crude benzol and 450,000 Mt of crude coal tar. The company's business is processing coal tar, mixed tar oils, benzol and raw materials for the production of phenols. Coal tars are composed of hundreds of organic compounds: aromatic compounds, including PAHs, heterocyclic PAHs, phenols, benzene, toluene, and xylenes, as well as aliphatic and polar hydrocarbons, but the chemical composition of coal tar may vary as a function of many factors, including feed stock, plant operating conditions and weathering once released to the environment [13,14].
The products of DEZA are mainly aromatic hydrocarbons -benzene, toluene, mixtures of xylenes, phenols, cresols, and xylenols -followed by polycyclic aromatic hydrocarbons -anthracene, carbazol and naphthalene, phthalic anhydride, dioctyl phthalate. Coal tar pitch is also among their important products. The company is the supplier of raw materials for the production of carbon black in the affiliated company CS-CABOT [15]. According to the Czech Integrated pollution register, the company reported annual naphthalene emissions of 50,000 kg. The emitted quantities of the other PAH compounds are not reported [16].

Sampling of soils and analytical methods
Forty samples were collected for the determination of the PAH distribution in soils from the area of Valasske Mezirici. Twenty samples were collected from urban grass-covered and agricultural soils in the town and in its nearest surroundings. Agricultural soil refers to soil that is used for agricultural uses, for example, soil for food production, pasture land and grassland. Urban soil refers to soil not used for agriculture, occurring in the territory of the urban settlement. In this context, it represents soil in town gardens, soil along the roads and road junctions, parking places, and so on. These type of soils are affected by human activities [17,18]. The sampling sites were selected in order to reflect the diverse exposure of soils to the pollution sources and dominant wind directions. They represent samples directly affected by industrial emissions, and directly subjected to transport and/or local heating emissions.
Twenty samples were collected from the humus layer of forest soils on the hill slopes in the surroundings of the town; 15 in coniferous forests and 5 in mixed forests. The sampling sites were also selected to represent a diversity of dominant wind directions.
The sampling sites are shown in Figure 3 and described in Tables 1 and 2. The sampling procedure, the physical sample preparations and the representative sample division were accomplished by using simple random shallow sampling according to Tan (2005) and ISO 10381-4:2003 and ISO 10381-5:2005 [19][20][21]. The samples were collected from the surface layer (from A-horizon of 0-10 cm depth), after removing grass in urban/agricultural soil and needles and litter in forest soil. Approximately 2 kg of material were collected from a square area of 10 × 10 m in each sampling site. All samples were put into glass containers with a minimum headspace of air, and immediately transported to the laboratory. In each series of samples, one The local landscape around the town of Valasske Mezirici town and the coal tar refinery Figure 1 The local landscape around the town of Valasske Mezirici town and the coal tar refinery. Some altitudes are marked (in orange). The limits of detection and measurement uncertainties for PAH compounds are shown in Table 3. The values of the limits of detection were derived 1) from the smallest integrable areas for each compound peaks in signal and 2) from the ratio between the analyte signal and the baseline noise (limit of detection = peak height/noise ratio of 3).
The limits of detection were determined for contaminated soil matrices [22]. The measurement uncertainties are expanded uncertainties calculated by multiplying the combined standard uncertainties by a coverage factor k = 2 for an approximate level of confidence of 95% [23].
A sufficient part of each sample was taken for the PAH extraction. The samples were not dried before extraction. The PAH compounds were extracted from the soil samples  New analysis of urban/agricultural soils New analysis of forest soils using a 1:1 (v:v) mixture of acetone and hexane in a Soxwav 3.6 microwave extractor (5 min of 20% microwave performance, following 55 min of 60% performance). The final extracts were filtered, dried with Na 2 SO 4 and evaporated to produce near-dry residue. The residue was redissolved in 1 ml of hexane and divided into two aliquots; one half of the volume was used for gas chromatog-raphy -mass spectrometry (GC/MS) analysis and the second half was analyzed by high pressure liquid chromatography with fluorescence detector (HPLC/FD) after solvent exchange (hexane for methanol). The chromatographic analyses were performed using a GCQ/ Polaris Q ThermoFinnigan gas chromatograph, equipped with a DB-5MS 30 m × 0.25 mm × 0.25 μm column, and   The dry matter content of each soil sample was determined after drying in an oven at 105 ± 3°C to constant weight. The organic oxidizable carbon (C ox ) content (in percentage) was determined in each forest soil sample by the modified Walkley-Black (VB) dichromate methods [24]. The values were determined to be in the range 3.45-19.20% (Table 2). ISO 10390:2005 was used for the routine determination of pH using a glass electrode in a 1:5 (volume fraction) suspension of soil in water (pH in H 2 O) and 1 mol.l -1 potassium chloride solution (pH in KCl) ( Table 1 and 2).

PAH content in urban/agricultural soil
The results of the PAH analysis of urban/agricultural soil samples are shown in Tables 4, 5 in mg/kg dm, and the statistical evaluation appears in  Table 7. The observed values of the PAH concentrations in the B-horizon were similar to the concentrations in the A-horizon. The values were in the range that was observed for typical concentrations in urban soils (0.600-3.000 mg/kg), and they are higher than typical concentrations in forest and rural soils (up to 1.000 mg/kg) [25]. We do not know the full history of the site, but the relatively high PAH concentration values show an influence due to human activity. The concentrations of the carcinogenic fraction -benzo [a]pyrene were in the range of 0.06-2.25 mg/kg dm, with an arithmetic mean of 0.36 mg/kg dm and a median of 0.22 mg/kg dm.
The highest concentrations of the 16 selected PAHs were identified in the urban soils in the Valasske Mezirici town territory (ID 137, 109, 17,18,41,11), and at the Jurinka (ID 6) and Policna (ID 108) sites. All of these sampling sites are situated south of the coal tar refinery and are in the proximity of the town center and the coal tar refinery. Influences from all PAH sources (industrial activity, local heating and transport) are evident. The remaining samples were collected north of the study area, and the identified PAH concentrations were lower in these sites. The sites (ID 13, 15, 20, 54, 78) are situated in the direction of the prevailing wind flow, and the landscape of this region decreases in altitude in this direction. The PAH atmos- pheric dispersion is greater in this direction, and a lower level of PAH deposition occurs at these sites.
Anomalous concentrations of the PAHs (35.16 mg/kg dm) and benzo [a]pyrene (2.25 mg/kg dm) were identified in the sample ID 116, in Stepanov. The values of the PAHs were significantly higher than in the other soil samples. This sample was collected in agricultural soil on the slope above the town in the southeast direction. It suggests that PAH deposition at higher altitudes in the proximity of PAH sources can occur as a consequence of unfavorable meteorological conditions associated with the low wind directions or calm conditions that are frequent in this region (see 2.1). These conditions are favorable for the occurrence of near ground high pollutant concentrations and are not appropriate for good atmospheric dispersion of pollutants [26]. Graphical presentations of the results are shown in Figures 4 and 5, illustrating the concentrations of the studied PAHs and benzo [a]pyrene in urban/agricultural as well as forest soils.
The average PAH16 distribution pattern is shown in Table  8 and in Figure 6. Fluoranthene, pyrene, chrysene and indeno [1,2,3-cd]pyrene are the most prominent compounds in the samples (calculated from average values). The predominance of 4-6 ring compounds in the soil samples suggests pyrogenic sources [27]. However, higher concentrations of naphthalene were found in 12 samples ( IDs 7,12,13,15,16,17,18,19,20,54,78,138). Six of these samples were collected in the coal tar refinery surroundings (IDs 7,13,15,16,54,78), four were collected in the town territory (IDs 12, 17,18,19) in the proximity of the main routes, and one (ID 20) was from the background site. The proportion of naphthalene in the PAH samples was generally in the range of 6.2-22.3%, and in one case even 35.3% (ID 138).
The presence of 2-and 3-ring PAHs can indicate more recent PAH deposition [27]. It is well known that light PAH compounds occur in soils in lower proportions than heavier ones, due to their physico-chemical properties, such as higher water solubility, volatility and biodegradability and lower sorption ability to soil organic matter or particles [28,29]. Removal of the PAHs from the atmosphere is accomplished by dry or wet deposition of particles and vapors [30,31]. Generally, it is known that for high level sources, the level of deposition near the source is small and increases with downwind distance from the source until it reaches a maximum, after which it decreases. For ground-level sources, the highest dry deposition is directly next to the source, and decreases downwind. Thus, it is possible to explain higher naphthalene contents in the soils in the territory of Valasske Mezirici due to naphthalene in PAH emissions from transport, local heating and the coal tar refinery.
Diagnostic ratios of selected PAH compounds are generally considered to be a good indicator of the pollution sources and of the mechanism of PAH transport into the soil. The ratios Ind/(Ind+BghiP) and Flt/(Flt+Pyr) are often used to distinguish between pyrogenic and petrogenic sources. The value of Ind/(Ind+BghiP) > 0.  Figure 7 and in Table 9.
The percentage of the sum of 8 PAHs compounds (with 4 or more aromatic rings, excluding fluoranthene and pyrene) and the sum of naphthalene and phenanthrene concentrations from the full set of PAHs are alternative    The Czech Republic requires the regular monitoring of agricultural soils [35]. The PAHs level ranged from 0.056 to 9.489 mg/kg dm for PAHs (16 except acenaphthylene) and < 0.004-0.624 mg/kg dm for benzo [a]pyrene, with annual arithmetic means ranging from 0.797-1.297 mg/ kg dm for PAHs (16 except for acenaphthylene) and from The PAH concentrations in urban/agricultural and forest soils in the Valasske Mezirici Region Figure 4 The PAH concentrations in urban/agricultural and forest soils in the Valasske Mezirici Region.

PAHs in forest soils
Two processes affect the fate of PAHs in forest soils: deposition and decomposition. In addition to the PAHs input to the soil through litter, dry and wet deposition of aerosol particles is an important pathway of path contamination of acid forest soils. The highest PAH concentrations in the air were in aerosol particles of grain size of 0.1-3 μm [36]. PAHs may be degraded in forest soils/forest humus, although microbial degradation activity decreases at pH levels below 3.6 [37,38]. The scavenging effect is a key source of forest soil contamination, and the high amount of natural organic matter in forest soil also pro-  vides better conditions for the binding and accumulation of PAHs in soils [8,29,39]. Coniferous trees with needles are more efficient at collecting particles and cloud droplets than leafy trees [30].

Comparison of PAH percentage distributions in urban/agricultural and forest soils in the Valasske Mezirici Region
The observed PAH concentrations in forest soils are summarized in Tables 10, 11 and the statistical evaluation is in  Table 7. The PAH content of the Bhorizon are higher than those corresponding to typical forest soils (0.005-0.100 mg/kg), which showed a higher pollution load in the forest soils in the region [25]. All the analyzed PAHs were identified in each soil sample, except acenaphthylene, which was below the detection limit in most samples.
Generally, high levels of the 16 investigated PAHs were identified in the transect zone of the north-south direction. The highest concentrations of the PAHs, as well as of The comparison of molecular diagnostic ratios Figure 7 The comparison of molecular diagnostic ratios.
Diagnostic r atios of in ur ban and for est soils: ur ban/agr icultur al soils for est soils  (IDs 13,14,15). The lowest concentrations were identified to the west of the town (IDs 7,12). These findings correspond to the meteorological conditions in the region, with the frequent south and north wind directions, inversion situations and the east wind direction being the least frequent.
The distribution of pollutants in the environment is influenced by landscape relief. This aspect is apparent in this region, due to the meteorological conditions low wind velocities and calm are the two most frequent wind flow situations. The altitude increases to the south and east from the town of Valasske Mezirici. In the north, the altitude decreases, and atmospheric pollutants can be better dispersed by the south wind ( Figure 1). However, the PAH dispersion during frequent calm and inversion situations is very slow and is associated with pollutant concentrations increasing near the ground. These meteorological conditions are mainly correlated with low temperatures, and the PAHs are deposited on the slopes of the hills surrounding the town of Valasske Mezirici.
The PAH concentration trend shows that the highest concentrations are obtained at higher altitudes in the proximity of the town (Figure 4 and 8). In the case of the altitude of 400 m above sea level, the PAH concentrations decrease with increased distance from the town centre; this is in agreement with other authors [39]. The three last values in Figure 8 show the PAHs and benzo [a]pyrene concentrations at the highest altitudes of 450, 625 and 670 m, to the southeast of the town (IDs 13,14, respectively 15). The site with the altitude of 450 m is the nearest to the town of these three points, which is the reason for the highest PAH value there.
For two sites at higher altitudes (500 and 550 m above sea level), the PAH and benzo [a]pyrene concentrations were not as high as in the preceding three sites. Policna (ID 7, 500 m above sea level) is located to the west of the town, outside the frequent wind flow, and the site Ruzdka (ID 16, 550 m above sea level) is the furthest from the town. It is situated to the south and is sheltered from pollutant deposition by higher hill barriers (481-691 m above sea level) between the site and the town PAH sources. The levels of PAHs and benzo [a]pyrene at the altitudes of 450, 625 and 670 m markedly exceeded other values. Comparing the distributions of single PAHs compounds with altitude, we found that all the compounds had the same trend, except fluorene, acenaphthene and phenanthrene, where some deviations appeared. Fluorene, acenaphthene and anthracene are the least abundant in forest soil. The behavior of lower molecular PAHs is affected by the collecting effect of the bark. Due to their physico-chemical characteristics, lower molecular PAHs are more efficiently adsorbed by the bark, they are leached from the organic layers, they can be re-volatilized or biodegraded, and so on [40]. The sum of the 7 carcinogenic PAHs (according to IARC, see 3.1) in the selected forest soil ranged from 4.260-44.400 mg/kg dm, and they formed 47-61% of the sum of the 16 PAHs. The average PAH distribution pattern for forest soils is shown in Table  8 and in Figure 6. Fluoranthene, pyrene, chrysene, indeno  Figure 7). The dominant pollution sources according to these values are pyrogenic sources, coal/ wood/grass combustion, and the ratio of BaP/BghiP confirms a contribution from traffic. We can observe some differences in the Baa/(Baa+Chry) ratio, whose values are lower than 0.35, corresponding to a mixture of pyrogenic and petrogenic sources in the 10 samples of forest soils. Only pyrogenic sources are found in urban/agricultural soils using this molecular ratio. Most of the forest soil samples had a higher value of Ind/(Ind+BghiP) than the urban/agricultural soils (Figure 7).
The percentage of the sum of 8 PAH compounds (4)(5)(6) and of the sum of naphthalene and phenanthrene concentrations in the complete PAH set were in the ranges 48-63% and 5-10%, respectively. In case of the naphthalene and phenanthrene sum, we see a lower proportion in the sum of the 16 PAHs than in the case of urban/agricultural soil, whereas the proportion of the 8-PAH sum is slightly higher.

Comparison of the PAHs in soils from other geographic locations
The PAH concentrations in the forest soils in the Valasske Mezirici Region are much higher than the concentrations presented in [10,27,29,41]. PAH concentrations were significantly higher at upper than at lower slope sites, indicating long-distance transport [40]. The comparison of PAH concentrations in forest soils of different regions in Europe with area of Valasske Mezirici is summarized in Table 13. The PAHs concentrations presented in this paper correspond to those identified in forest soils in regions with elevated atmospheric PAH deposition: in Northern Bavaria (0.644-19.919 mg/kg dm in A-horizon) and near a blast furnace plant located in the region of Hoogovens, in the Netherlands [39,42]. The PAH levels in urban/agricultural soils are in accordance with those in industrialized areas [11,32,[43][44][45]. In total, the PAH concentrations in forest soils were significantly greater than those identified in urban/agricultural soils.
In their studies, Farrar et al. (2005) and Tao et al. (2007) observed clear decreasing trends in PAH concentrations in the vertical distribution in the urban boundary atmospheric layer. There are three possible scenarios of the vertical concentration profiles due to differences in emission sources, advection and vertical mixing conditions: 1) even vertical mixing, with weak fresh emissions, dominant advection and well-mixing conditions, 2) decrease with height, with ground source domination and stable atmospheric boundary layer conditions, 3) increase with height with upper boundary layer emission sources [46,47]. We can suppose, especially in the winter and autumn periods that Scenario 2 predominates in the studied area, when pyrogenic sources and traffic are the dominant pollution sources according to the determined values of molecular diagnostic ratios (see 3.1 and 3.2). Local heating and traffic are sources that emit pollutants at ground level. However, an important source of PAH pollution is located in this region the coal tar refinery. During coal tar treatment, some operations are performed (for example coal tar launching) that lead to organic compound escape to the ambient air. The escape level depends on the air temperature. The possible emissions sources in coal tar refinery are 15 m of height and they can contribute to the pollution sources at ground level not only in the winter days, but also in periods with higher temperatures.
A higher level of deposition of PAH compounds on the hill slopes in the vicinity of the town occurs in this region as a consequence of the prevailing meteorological conditions (on average, 66% of meteorological situations are calm with wind velocities 0-2.5 m/s, 27.5% have wind LD -limit of detection velocity 2.6-7.5 m/s and 6.5% have wind velocities >7.6 m/s, according to the Czech Hydrometeorological Institute) and topographical terrain. The meteorological conditions are not favorable for atmospheric pollutant dispersion, and the pollutants are accumulated in the boundary layer, the altitude of the inversion layer generally reaches a number of tens or hundreds meters) [26]. The unfavorable meteorological conditions usually persist for several days and are repeated (see 3.1).
The molecular diagnostic ratios indicate pyrogenic sources and combustion of coal, wood and grass as main pollution sources. However, the coal tar composition is formed especially from naphthalene (10% by weight), phenanthrene (4.5% by weight), fluoranthene (3.0% by weight), acenaphthylene (2.5% by weight), pyrene (2.0% by weight), fluorene (1.8% by weight), anthracene (1.3% by weight), chrysene (1.0% by weight) and acenaphthene (0.2% by weight) [48]. This fact can explain the higher concentrations of lighter PAHs, especially naphthalene, in urban/agricultural soils in the vicinity of the refinery and their higher contributions to the sum of the 16 PAHs in the urban/agricultural soils compared to the forest soils (Table 8, Figure 6). The diagnostic ratios determined for coal tar are 0.18 for Ant/(Ant+Phe), 0.58 for Flt/(Flt+Pyr), 0.54 for Baa/(Baa+Chry) and 0.53 for Ind/(Ind+BghiP) [49]. Thus, these values correspond to those in our samples in the case of Ant/(Ant+Phe) and Flt/(Flt+Pyr). The values of Baa/(Baa+Chry) were lower, and those of Ind/ (Ind+BghiP) were higher in our samples. All the values correspond to coal/wood/grass combustion and pyrogenic sources. Brown et al. (2006) compared eleven samples of coal tars from 10 former manufactured gas plant sites in the Eastern United States [16]. The diagnostic ratios determined in these coal tar samples were within the ranges (Table 9): 0,22-0,27 for Ant/(Ant+Phe), 0,36-0,58 for Flt/(Flt+Pyr), 0,48-0,58 for Baa/(Baa+Chry), 0,38-0,59 for Ind/(Ind+BghiP) and 1,67-3,25 for BaP/ BghiP. Thus, these values correspond to those in our samples in case of Ant/(Ant+Phe) and Flt/(Flt+Pyr). It is not possible to distinguish between pyrogenic sources and the coal tar refinery in this way, because the ratios presented are determined for cruel coal tar, but the coal tar refinery may contribute to soil by emitted PAH deposition. The diagnostic ratios in depositions can be completely different.
We observed in our samples higher concentrations of indeno [ The total PAH concentration as a function of altitude Figure 8 The total PAH concentration as a function of altitude.
for petroleum, single source combustion and environmental samples; the Ind/(Ind+BghiP) values were in the range of 0.09-0.70, while we determined these values to be above 0.85 in our samples, and higher in forest soils than in urban soils [49]. Indeno [1,2,3-cd]pyrene and benzo [ghi]perylene are degraded photochemically at comparable rates, and the original composition is preserved during atmospheric transport. The concentrations of benzo [a]pyrene were also significantly higher than those of benzo [ghi]perylene (4-6 times higher), and the ratio values were higher in forest soils than urban/agricultural ones. An inverse trend was observed in the case of Baa/(Baa+Chry), where mixed sources (pyrogenic and petrogenic) were determined in most of forest soils, and pyrogenic sources in the urban soils. Because we did not suppose petroleum sources in forest soils (the samples were collected in remote areas) this fact may correspond to faster benzo [a]anthracene degradation in forest soils or in the air during the atmospheric transport.

Conclusions
The urban, agricultural and forest soils in the region of Valasske Mezirici are contaminated with polycyclic aromatic hydrocarbons, which are emitted through industrial processes, transport and local heating sources. The highest PAH concentrations were found in forest soils from the forest-covered slopes of hills. The observed values were much higher than the PAH concentrations reported for forest soils in Western and Northern Europe. The PAH concentrations in the urban soils were higher than in the agricultural soils, but comparable with other soils from urban areas in the world. The highest PAH concentrations were observed in forest soils collected at higher altitudes above sea level. The PAH distribution in the mountainous region is related to altitude and influenced by meteorolog-ical conditions. The frequent calm conditions that are connected with inversion conditions contribute to PAH deposition at higher altitudes in the surroundings of the town. Elevated PAH concentrations were identified at sites further from the town, but the concentrations decreased in comparison with sites in the proximity of the town. Compared to the PAH load in other world areas, it is clearly confirmed that the soils in the region of Valasske Mezirici, especially the forest soils, are highly polluted. There is a primary source of PAH pollution in the region a coal tar refinery. There are also contributions from other sources local heating, vehicle transport and long range pollutant transport. It is not possible to distinguish the contributions of each source on the basis of diagnostic ratios.
designing of the study, in sample collecting and data evaluation. MHR participated in design of the manuscript. All authors have read and approved the final manuscript.