Distribution of Mercury in Drained Peatlands as the Effect of Secondary Transformation of Soil Organic Matter

Abstract

Peat organic soils play a major role in the accumulation of soil organic matter (SOM) and the mercury (Hg) cycle. Large mercury resources in peatlands can be a source of methylmercury for many decades and centuries, even if deposition is significantly reduced. The organic matter of peatland soils drained for agricultural use is subject to secondary transformation, which may affect the accumulation and resources of mercury. The aim of our work is to assess the secondary transformation of organic matter in the soils of drained peatlands of the temperate climate zone and to examine whether it affects mercury resources and profile distribution in organic soils. Field research was conducted in peatlands located in eastern Poland. In the present study, evaluation of secondary transformations occurring after drainage was based on observations of soil morphological characteristics, physical and chemical properties as well as fractional composition of organic matter of the identified soil horizons (to depth 70 cm). Standard cold vapor atomic absorption spectrometry (CV-AAS) was used to measure the total mercury content. In our research, we found a significant effect of the secondary transformation of organic matter occurring in drained peatlands of the temperate climate zone on the total mercury content and stock in soils. The highest content and differentiation of mercury occurred in murshic horizons (up to a maximum depth of 43 cm). The average mercury content of the distinguished soil horizons is grouped in the following series (in μg kg⁻¹): M1 (212.0) > M2 (182.8) > M3 (126.3) > Pt (84.9). The mercury stock, up to a depth of 70 cm in the tested soils, ranged from 17.5 to 39.6 mg m⁻². As much as 82.2% of soil mercury was found in the upper murshic horizons. We found strong correlations between soil properties characteristically variable in the secondary transformation process and total mercury content. The increased content of humic substances in murshic horizons caused a significant increase in the total mercury content. Our research is of great importance for soil monitoring, as the amount of determined mercury was greatly influenced by the depth of sampling (up to 25 cm). The results of the research should be taken into account when planning the restoration of peatlands of the temperate climate zone. There is a potential risk that elevated mercury concentrations in the upper murshic horizons may be a source of methylmercury for a long period of time. In peat soils with a high concentration of mercury, the risk of contamination with this toxic metal should be determined before re-irrigation.

1. Introduction

Soil is a basic component of the natural environment and it plays a key role in the geochemistry of elements and food production security. Therefore, it has to be regularly checked for potential contamination with trace elements. In this context, special importance is attributed to mercury, which is one of the most geochemically active and toxic metals [1,2,3,4,5]. In the natural environment, mercury is converted to methylmercury which is easily bioaccumulated [6,7,8,9]. Elevated mercury contents in the food chain markedly increase the risk of health issues in animals and people [10,11,12]. As regards mercury, the European Food Safety Authority (EFSA) adopted on 24 February 2004 an opinion related to mercury and methylmercury in food and endorsed the provisional tolerable weekly intake of 1.6 μg/kg body weight [13]. Methylmercury can make up more than 90% of the total mercury in fish and seafood. Forms of mercury present in other foods than fish and seafood are mainly not methylmercury and they are therefore considered to be of lower risk. Maximum levels of mercury in fish and seafood range between from 0.30 to 1.0 mg/kg wet weight.

Global mercury emissions from anthropogenic sources have increased 3–4 times compared with the pre-industrial period [14]. Major anthropogenic mercury sources since the beginning of the industrial era have included metal production, fossil fuel (mainly coal) combustion, waste incineration and sewage dumping [15,16]. The time when the elemental form of mercury is present in the atmosphere, spanning 0.8 to 1.7 years (from the beginning of emission to deposition on the soil surface), means the element is likely to be transported by way of the atmosphere over long distances, and indicates a global source of soil contamination with this metal [17,18,19,20]. Despite the reduction in global mercury emissions, it still poses a serious threat to the environment and human health, especially where there are favourable conditions for the production of methylmercury [21,22,23].

After being released into the atmosphere, a mercury atom can remain in circulation for >1000 years [2].

Estimated mercury pools in the European Union soils amount to around 44.8 Gg, the average concentration being 103 g ha⁻¹ [4,5]. Agricultural soils account for 85% of mercury released into the environment [5]. The concentration of mercury in soil varies greatly and depends on many factors, including its origin. In regions free of anthropopressure, it has been assumed that deposition from the atmosphere is the major mercury source in the soil [2,24,25]. The share of mercury inputs from the atmosphere reaches 50% [26,27], the highest mercury content being recorded in the topsoil layer of soils [20,28]. The availability of mercury to plants is low because there is a tendency for mercury bioaccumulation in their roots which form a specific barrier against translocation to the above-ground parts [29]. The concentration of mercury in soils is associated with organic matter content and is affected by temperature and vegetation type [4]. Mercury sequestration in soil is positively correlated with organic matter content but the amount of mercury in terrestrial ecosystems is generally determined by atmospheric mercury deposition [30,31]. Most soil mercury received from the atmosphere is associated with organic carbon and does not form a permanent mercury sink because it is released as a result of decomposition of soil organic matter.

Soils of raised bogs (ombrotrophic bogs) are mainly fed by rainwater and spread above the groundwater level. These are the reasons why they are commonly chosen for research into contamination with airborne mercury [20,32,33]. Much less scientific attention has been paid to the issues of mercury accumulation in fen peatlands although quite a number of these peatlands are used as grasslands to produce biomass for fodder purposes. Organic soils of peatland areas used for farming purposes, in the context of mercury accumulation and circulation, have the following characteristics:

(i)

They are a component linking terrestrial habitats with bodies of surface water. Organic soils are a significant source of methylmercury in ecosystems adjacent to bodies of water in which it can be accumulated in fish [34,35,36]. In general, peatlands are potentially highly productive providers of methylmercury and are its main source for the aquatic food chain [37,38,39]. Anaerobic conditions in soil are conductive to the production of this toxic compound [39,40,41]. In the studies of Busaan et al. [42] the amounts of methylmercury in wetland sediments ranged from 0.36 to 1.43 ng g⁻¹. In addition, the authors in pioneering in situ studies proved that the indicators mercury methylation and demethylation rates vary seasonally (highest in summer and lowest in winter). Moreover, the risk exists of mercury accumulation in plants followed by its accumulation in milk, meat and other foodstuffs [43,44,45];

(ii)

They are fed not only directly by rainwater but to a great extent by river water and surface water within the catchment area, which creates another possibility for mercury transport and accumulation in the soils of river valleys [17,46,47]. An inflow of mercury through natural watercourses is estimated to account for 25% of its current content [35,48]. Mason et al. [26] estimated that as much as 90% of mercury derived from the atmosphere is retained in soils in the areas of river catchments. Due to water erosion of European and UK soils, an approximate annual amount of 43 Mg mercury is released, of which around 6 Mg mercury reaches rivers [5];

(iii)

The solid phase of organic soils mainly consists of soil organic matter (SOM). Strong binding of mercury in soils is due to its high chemical affinity for organic matter [4,6,18,49,50,51]. Peatland drainage (very frequent in case of utilisation of organic soils) stimulates aerobic organic matter decomposition, production of dissolved organic matter (DOM) and humic substances (HS) [52,53,54,55]. Dissolved organic matter affects mercury speciation, mobility and toxicity in the environment [8,9,19,50,56,57,58,59]. DOM increases mercury mobility in the environment by, e.g., reducing precipitation of mercury sulphide [9,60]. HS are an example of the soil parameters which enhance mercury methylation [61];

(iv)

Soils of peatland areas (particularly drained ones) are characterised by great dynamics of oxidation-reduction conditions. Fluctuating groundwater levels result in, e.g., thermal changes in the soil and in pH values; they also affect vegetation cover [62,63,64]. These factors may affect mercury accumulation in soil [1,4,7,38,47,49,57,65,66];

(v)

Peatland drainage and agricultural utilisation result in increased soil trophicity which is followed by the enhanced microbiological activity of the soil [54,67,68]. In the study by Wang et al. [66], net production of methylmercury was positively correlated with the trophic requirements of vegetation and an increased availability of electron acceptors and donors for mercury-methylating microorganisms. Moreover, the authors believe that ecosystem characteristics which intensify microbial processes involved in mercury methylation also contribute to mercury reduction, which may lead to mercury re-emission into the atmosphere. Busaan et al. [42] showed that in wetland sediments with increasing temperature the rate of mercury methylation increases.

Peatland drainage for farming purposes is a major cause of fen peatland degradation in the temperate climate [69]. Peat degradation is followed by the formation of a murshic horizons which has an aggregate structure and a greater quantity of humic substances [52,54,68,70]. The process of peatland drainage is progressing due to climatic change which contributes to a negative water balance that is unfavourable for peatland development [71,72]. Disturbing forecasts of climate change in Central Europe point to declining annual precipitation and increased air temperature [73,74].

The aim of this research is to assess the secondary transformation of organic matter in soils of drained peatlands of the temperate climate zone and to investigate whether this affects the total mercury content and distribution in the soil profile. We pose the hypothesis that the process of secondary transformation of organic matter, intensified by organic soil drainage, significantly affects the distribution of total mercury in the soil profile.

2. Materials and Methods

2.1. Location and Soil Sampling

Field research was conducted in peatlands located in eastern Poland (the river Bug catchment):

−

The valley of the river Liwiec on the Siedlecka Upland (the number of physio-geographical region is 318.94) [75]. The region is a poorly peaty old glacial moraine upland with a 3.1% share of peatlands (region I) [76];

−

Peatlands in the area of the Wieprz-Krzna Canal (the number of physio-geographical region is 845.11) [75]. The region is a swampy sandy proglacial stream valley in Western Polesie (Pripet Marshes) with a high share (13.4%) of peatlands (region II) [76].

The study of peatlands in region I were drained in the 1960s whereas drainage of the soils in region II was performed somewhat later, that is in the 1970s. At present, they are intensively managed for agricultural purposes to produce cattle feed. According to the Köppen–Geiger’s climate classification [77], the peatlands are located in the fully humid warm temperate climatic zone with warm summers.

Nine soil profiles of grassland soils were chosen for laboratory analyses (4 from region I, 5 from region II). According to the FAO WBR system of soil classification [78], the examined soils are classified as Drainic Histosols. The study material comprised soil samples collected from 36 soil horizons that were determined according to the Polish Soil Classification System [79]. The soil samples were taken to the depth of 70 cm. The overall description of the examined soil horizons is presented in

Table 1. Soil morphological characteristics.

Horizon (Soil Material)	No.	Depth (cm)	Thickness (cm) (Mean ± SD)	Morphological Features
M1	9	Min. 0–9 Max. 0–10	10.0 ± 0.8	Murshic horizons. The top level is strongly overgrown with plant roots, with a fine grain structure.
M2	9	Min. 9–21 Max. 10–25	13.7 ± 1.3	Murshic horizons. Lying below the M1 horizons, with a medium grain structure.
M3	9	Min. 21–40 Max. 25–43	18.9 ± 1.2	Murshic horizons in the initial phase of formation. Intensely cracked organic mass with preserved peat properties.
Pt Peat (parent material)	9	Min. 40–70 Max. 43–70	28.0 ± 1.0	Rush peat, fibrous, degree of decomposition on the von Post scale: H4–H8

SD—standard deviation.

.

2.2. Laboratory Analysis

The following parameters were determined in the samples of fresh soil material:

−

Bulk density (ρ_a) by the gravimetric method. Soil samples, taken so as to preserve the original layout, were placed in cylinders (v = 100 cm³) and dried at 105 °C;

−

pH value by the potentiometric method, using a pH meter, in a suspension of H₂O_dist. (soil/water = 1/5).

The remaining analyses were performed in soil samples dried at room temperature and ground in a porcelain mortar (ø < 0.25 mm). The following analyses were carried out:

−

Ash content (ash) by the weighing method, after decrepitation in a muff furnace (T = 600 °C);

−

Total carbon (TC) and nitrogen (TN) contents using the PerkinElmer^® 2400 Series II elemental analyzer with thermal conductivity detection (TCD) and acetanilide (C = 71.09%; N = 10.34%) as reference calibration standard;

−

Sequential fractioning of organic matter [80] (

Table 2. Fractionation methods for soil organic matter.

Fraction of Organic Carbon (SOM Fractions)	Methods
Labile fractionLab-C	Extraction 0.05 M H2SO4 (decalcitation of soil sample); extraction time 24 h; m/v = 1/50; centrifugation (g = 4000 rpm) and filtration through a cellulose filter. The carbon content of the solution was determined.
Fraction of humic substancesHS-C	Extraction 0.1 M NaOH; extraction time = 24 h; m/v = 1/50; centrifugation (g = 4000 rpm) and filtration through a cellulose filter. The carbon content of the solution was determined.
Fraction of fulvic acidsFAs-C	Acidification (2.5 M H2SO4, pH = 1.80) of a measured part of the extract with 0.1 M NaOH; separation of humic substances into the fraction of fulvic and humic acids as a result of sedimentation of humic acids (24 h). Carbon content was determined in the solution of fulvic acids.

). Carbon content in extraction solutions was determined by the oxidation-titration method. The fractional composition of organic matter was expressed as the share of fraction carbon in TC (% TC);

−

Standard cold vapour atomic absorption spectrometry (CV-AAS) was used to measure the total mercury content. Based on in situ dry ashing followed by gold amalgamation cold vapour, AAS was evaluated. Using the mercury analyser AMA 254 (Altec, Prague, Czech Republic), the total content of mercury was determined in 3 replications. The detection limit was 0.01 ng Hg. Measurement conditions: wavelength 253.65 nm, gas–oxygen (purity ≥ 99.5%), with pressure 200–250 kPa. Times of the individual stages of the analysis were: 120, 150 (decomposition at 550 °C) and 60 s.

The analytical procedure of mercury determination was controlled with the use of the TILL-3 certified reference material (

Table 3. Analysis of certified reference material (Geological Survey of Canada, Ottawa, Ontario).

Metal	Certified Value	Determined Mean Value	Recovery Rates	Replications
Hg [µg·kg−1]	107	111 ± 2.43	104%	5

). Mercury content was expressed in units of mass per soil mass (Hg_m/m, µg∙kg⁻¹) and units of mass per soil volume (Hg_m/V, µg∙dm⁻³).

Chemical analyses were replicated 3 times. The obtained results were related to absolute dry mass of the soil sample. The content of absolute dry mass was determined after the sample was dried at 105 °C.

2.3. Calculations and Formulas

−

Humin acid fraction content was calculated following the formula:

HAs-C = HS-C − FAs-C,

where:

(see Table 2).

−

Residual fraction content (post-extraction residue) was computed following the formula:

Res-C = TC − (Lab-C + HS-C).

where:

(see Table 2).

−

The values of the index of mercury distribution in the profile (Hg(DI)) were calculated following the formula:

(a)

Hg(DI_m/m) = Hg_m/m(M)/Hg_m/m(Pt);

(b)

Hg(DI_m/V) = Hg_m/V(M)/Hg_m/V(Pt);

where:

Hg_m/m—mercury content (in µg·kg⁻¹) in murshic (M) and peat (Pt) horizons;

Hg_m/V—mercury content (in µg·dm⁻³) in murshic (M) and peat (Pt) horizons.

−

Mercury pool in the examined soils (to the depth of 70 cm) was calculated following the formula:

$${\mathrm{Hg}}_{\mathrm{stock}}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{V}}_1·{\mathrm{Hg}}_{\mathrm{m}1/\mathrm{V}1})+({\mathrm{V}}_2·{\mathrm{Hg}}_{\mathrm{m}2/\mathrm{V}2})+\dots ({\mathrm{V}}_{\mathrm{n}}·{\mathrm{Hg}}_{\mathrm{m}\mathrm{n}/\mathrm{V}\mathrm{n}}),$$

where:

Hg_stock—soil mercury pool (w mg·m⁻²);

i—number of soil horizon;

V = h·m²—soil horizon volume (in m³) whose thickness is h (in m) and area is 1 m²;

Hg_m/V—mercury content (in μg·dm⁻³) in the soil horizon.

2.4. Statistical Analysis

Statistical calculations were performed using the statistical software STATISTICA 13 PL (TIBCO Software Inc., Palo Alto, CA, USA). Mean values of the examined parameters were compared using a one-way variance analysis and the post-hoc Tukey test. Statistically significant differences were determined at the significance level of p < 0.05. Relationships between the examined parameters were checked using a linear correlation coefficient (r). Selected relationships were presented as linear regression equations.

3. Results and Discussion

3.1. Evaluation of Secondary Transformations

In the present study, evaluation of secondary transformations occurring after drainage was based on observations of soil morphological characteristics, physical and chemical properties as well as fractional composition of organic matter of the identified soil horizons.

3.1.1. Soil Properties

During field studies, there were determined murshic horizons (M1, M2) with poorly diversified thickness, regardless of soil profile location. They contained stable soil aggregates with granular structure. Parental material layers of the study soils were made up of peat with a fibrous structure (Pt). An M3 horizon was determined between the M2 murshic horizon and peat horizon (Pt), which has morphological properties of both these horizons. A clear morphological distinction of top murshic horizons, compared to peat horizons, were confirmed by results of laboratory analyses as well as ANOVA (

Table 4. Soil parameters.

Parameters	Statistical Measure	Murshic Horizons			Peat (Pt)
		M1	M2	M3
pH	range	5.42–6.90	5.76–6.45	5.85–6.37	5.69–6.05
Ash (%)	Mean	26.9 b	28.7 b	17.1 a	12.1 a
	SD	4.31	6.88	10.1	4.41
	CV (%)	16.0	23.9	58.9	36.6
Bulk density (ρa) (Mg·m−3)	Mean	0.290 bc	0.323 c	0.236 b	0.169 a
	SD	0.051	0.055	0.063	0.026
	CV (%)	17.5	17.1	26.5	15.2
TC (g·kg−1)	Mean	385.8 a	386.2 a	468.1 b	506.8 b
	SD	25.3	45.7	68.8	35.5
	CV (%)	6.56	11.8	14.7	7.00
TC/TN	Mean	11.7 a	12.1 a	14.3 b	15.6 c
	SD	0.383	0.909	1.06	0.898
	CV (%)	3.29	7.52	7.41	5.75

TC—Total carbon; TN—Total nitrogen; SD—standard deviation; CV—coefficient of variation; ^a, ^b, ^c—homogeneous groups.

). The M1 and M2 horizons were found to form a uniform group compared to the layers below them; the group was characterised by a significantly higher ash content, greater density (higher ρ_a values), lower TC content and lower values of the TC/TN ratio. The obtained variation in the properties of the examined profiles is typical of drained organic soils of fern peatlands [52,54,55,67,68,70]. Processes of rapid organic matter decomposition contributed to a significant decline in organic matter content and nitrogen accumulation and, as a consequence, to lowering the values of the TC/TN ratio. The properties of the examined soils, in particular the carbon-to-nitrogen ratio, relatively low acidification and potentially good aeration of the M1 and M2 horizons are indicative of the fact that they are biologically active soils. This may increase the intensity of processes of organic matter decomposition and more rapid eutrophication of the soil environment [53,55,81,82,83].

3.1.2. Soil Organic Matter Fractions

Analysis of the carbon content in the identified organic matter fractions confirmed distinctiveness of the M1 and M2 horizons compared with the M3 and Pt horizons of the study soils (

Table 5. Share of organic carbon fraction (% TC).

Parameters	Statistical Measure	Murshic Horizons			Peat (Pt)
		M1	M2	M3
Lab-C	Mean	0.739 c	0.462 b	0.277 a	0.208 a
	SD	0.221	0.103	0.065	0.058
	CV (%)	29.8	22.4	23.4	28.1
HS-C	Mean	35.0 b	33.8 b	25.2 ab	17.6 a
	SD	8.32	10.1	10.1	6.98
	CV (%)	23.8	29.8	40.2	39.7
FAs-C	Mean	8.00 b	7.32 b	4.71 a	3.63 a
	SD	0.785	1.64	2.16	1.83
	CV (%)	9.82	22.4	45.8	50.4
HAs-C	Mean	27.0 b	26.5 b	20.5 ab	14.0 a
	SD	8.10	9.6	8.94	6.57
	CV (%)	30.1	36.5	43.6	47.1
Res-C	Mean	64.3 a	65.8 a	74.5 ab	82.2 b
	SD	8.18	10.1	10.2	6.98
	CV (%)	12.7	15.4	13.6	8.49
HAs-C/FAs-C	Mean	3.26 a	3.65 a	4.66 a	4.18 a
	SD	0.947	1.26	1.88	1.77
	CV (%)	29.1	34.6	40.3	42.4

TC—Total carbon; Lab-C—labile carbon; HS-C—carbon of humus substances; FAs-C—carbon of fulvic acids; HAs-C—carbon of humic acids; Res-C—carbon in residual fraction; SD—standard deviation; CV—coefficient of variation; ^a, ^b, ^c—homogeneous groups.

). The M1 and M2 horizons had a significantly higher share of labile organic matter forms (Lab-C) and humic substances extracted with 0.1 M NaOH (HS-C), both the fulvic acid fraction (FAs-C) and humic acid fraction (HAs-C). Simultaneously, no significant differences were found between values of the HAs-C/FAs-C ratio. Moreover, the M1 surface horizon had significantly the highest share of Lab-C fraction compared with the remaining horizons. It may be the effect of considerable biological activity as well as the inflow of fresh organic matter from the existing vegetation. Hence, we confirmed the association reported in the literature that peatland soil drainage and intensification of the process of secondary transformations enhance organic matter humification and increase the share of humic substances in the soil matter [52,54,68,84].

An increase in the ash content of the study soils, and a simultaneous decline in TC content and increase in bulk density, was accompanied by significantly lower values of the TC/TN ratio (

Table 6. Significant correlation coefficient p < 0.05.

Parameter	Ash	ρa	pH	TC	TC/TN	Lab-C	HS-C	FAs-C	HAs-C
ρa	0.898	-
TC	−0.984	−0.892		-
TC/TN	−0.730	−0.655	−0.457	0.765	-
Lab-C	0.587	0.495	0.539	−0.625	−0.757	-
HS-C	0.817	0.871		−0.806	−0.545	0.453	-
FAs-C	0.765	0.758		−0.804	−0.660	0.596	0.687	-
HAs-C	0.752	0.816		−0.729	−0.464		0.983	0.542	-
Res-C	−0.822	−0.873		0.811	0.556	−0.470	−0.999	−0.693	−0.981
HAs-C/FAs-C			−0.465		0.329			−0.437	0.452

ρ_a—bulk dencity, TC—Total carbon; TN—Total nitrogen; Lab-C—labile carbon; HS-C—carbon of humus substances; FAs-C—carbon of fulvic acids; HAs-C—carbon of humic acids; Res-C—carbon in residual fraction.

). Enhanced decomposition of organic matter is followed by lowering of the TC/TN values in the soil matter, as indicated by a significant negative correlation of the TC/TN ratio with Lab-C and HS-C, and a significant positive association with Res-C. Values of the Pearson coefficient of linear correlation undoubtedly indicate there were favourable conditions in the examined soils for organic matter transformations due to drainage and oxygenation of surface soil layers. Organic matter transformation under such conditions results in an increase in the amount of labile forms of organic matter and humic substances.

3.2. Mercury Content and Distribution in the Profile

Variation was noted in the total mercury content in the soil profiles of both the study regions (

Table 7. Total mercury content and values of enrichment factors in murshic horizons.

Region	Horizon	Depth (cm)	Hgm/m (μg·kg−1)	Hg(DIm/m)	Hgm/v (μg·dm−3)	Hg(DIm/V)
I	M1	0–10	262.0	3.21	93.8	6.81
	M2	10–22	225.7	2.77	73.8	5.35
	M3	22–40	161.2	1.98	53.2	3.86
	Pt	40–70	81.5	-	13.8	-
	M1	0–10	194.7	2.10	53.0	3.08
	M2	10–23	187.0	2.02	59.5	3.45
	M3	23–43	97.9	1.06	21.2	1.23
	Pt	43–70	92.6	-	17.2	-
	M1	0–10	211.1	2.60	69.9	5.45
	M2	10–25	182.8	2.25	71.8	5.60
	M3	25–42	100.8	1.24	18.7	1.46
	Pt	42–70	81.1	-	12.8	-
	M1	0–9	215.6	2.09	81.9	3.55
	M2	9–24	201.1	1.95	86.1	3.73
	M3	24–43	195.9	1.90	69.0	2.98
	Pt	43–70	103.1	-	23.1	-
II	M1	0–10	251.9	3.79	57.4	3.91
	M2	10–22	203.8	3.06	55.2	3.76
	M3	22–42	71.7	1.08	16.3	1.11
	Pt	42–70	66.5	-	14.7
	M1	0–9	222.6	3.29	49.2	5.35
	M2	9–21	145.0	2.14	42.8	4.65
	M3	21–41	118.0	1.75	26.6	2.89
	Pt	41–70	67.6	-	9.20	-
	M1	0–10	201.0	1.96	59.7	3.55
	M2	10–24	141.2	1.38	37.4	2.22
	M3	24–42	114.4	1.11	24.1	1.43
	Pt	42–70	102.7	-	16.8	-
	M1	0–12	153.1	1.71	39.6	2.89
	M2	12–25	163.3	1.82	52.4	3.82
	M3	25–43	154.8	1.72	29.7	2.16
	Pt	43–70	89.7	-	13.7	-
	M1	0–10	195.9	2.48	52.1	3.71
	M2	10–22	195.6	2.48	55.7	3.97
	M3	22–42	122.0	1.55	28.9	2.06
	Pt	42–70	78.9	-	14.0	-

Hg_m/m—mercury content in µg·kg⁻¹; Hg(DI_m/m)—values of distribution index of mercury (µg·kg⁻¹) in profile; Hg_m/v—mercury content in μg·dm⁻³; Hg(DI_m/V)—values of distribution index of mercury (μg·dm⁻³) in profile.

). Mercury concentration declined with an increasing depth in the profiles.

Peat in the soil parent material of both the regions had a very similar total mercury content (p = 0.999). The M1, M2 and M3 horizons in region I soils had a higher mercury content compared with surface horizons of soils in region II. No significant variation was found between them. The length of the period following drainage had no effect on the mercury concentration in the surface layer of soils in both the regions.

As soils of both the regions had a similar total mercury content (

Table 8. Results of statistical analysis (Anova, the Tukey test).

Horizon	Mean Content of Hgm/m (μg·kg−1)		Significant Level	Mean Content of Hgm/v (μg·dm−3)		Significant Level
	Region I (N = 4)	Region II (N = 5)		Region I (N = 4)	Region II (N = 5)
M1	220.9	204.9	p = 0.993	74.6	51.6	p = 0.138
M2	199.1	169.8	p = 0.839	72.8	48.7	p = 0.107
M3	138.9	116.2	p = 0.950	40.5	25.1	p = 0.582
Pt	89.6	81.1	p = 0.999	16.7	13.7	p = 0.999

), the need was recognised to present the mechanism of mercury accumulation as a universal phenomenon in the soils. Taking into account the average total mercury content in the examined profiles, their horizons can be ranked in the following order: M1 > M2 > M3 > Pt. The highest values of enrichment factor indicate a substantial mercury accumulation in the M1 and M2 horizons, that is in the rhizosphere zone of the study soils. In particular, this applies to values of the factor calculated using soil volume (μg Hg·dm⁻³). Mercury accumulation in the M1 and M2 horizons was over four times as high as the values for M3 and Pt. The M1 and M2 murshic horizons formed uniform groups with a significantly higher total mercury contents compared with M3 and Pt (

Table 9. Values of descriptive statistics for mercury content and values of distribution index of mercury.

Parameters	Statistical Measure	MURSHIC Horizons			Peat
		M1	M2	M3
Hgm/m (μg·kg−1)	Mean	212.0 c	182.8 c	126.3 b	84.9 a
	SD	29.6	28.1	38.0	13.4
	CV (%)	14.0	15.4	30.1	15.7
Hg(DIm/m)	Mean	2.58	2.21	1.67	-
	SD	0.633	0.509	0.310	-
	CV (%)	24.5	23.0	18.6	-
Hgm/v (μg·dm−3)	Mean	61.9 b	59.4 b	32.0 a	15.0 a
	SD	15.8	15.5	17.5	3.82
	CV (%)	25.5	26.0	54.8	25.4
Hg(DIm/V)	Mean	4.25	4.06	2.56	-
	SD	1.20	1.03	0.855	-
	CV (%)	28.1	25.2	33.3	-

Hg_m/m—mercury content in µg·kg⁻¹; Hg(DI_m/m)—values of distribution index of mercury (µg·kg⁻¹) in profile; Hg_m/v—mercury content in μg·dm⁻³; Hg(DI_m/V) values of distribution index of mercury (μg·dm⁻³) in profile; SD—standard deviation; CV—coefficient of variation; ^a, ^b, ^c—homogeneous groups.

).

Mercury pools calculated to the depth of 0.7 m of the study soils were within the range of 17.5 to 39.6 mg·m⁻² (

Table 10. Mercury stock in soils.

Statistical Measure	Hgstock (mg·m−2)	% Hgstock for Soil Horizons
		Murshic Horizons				Peat
		M1	M2	M3	M1 + M2 + M3
Mean	24.2	25.7	32.5	24.0	82.2	17.8
Range	17.5–39.6	18.6–29.5	25.9–43.9	13.0–33.1	76.7–87.1	12.9–23.3
SD	7.09	3.37	5.23	6.84	3.45	3.45
CV (%)	29.3	13.1	16.1	28.5	4.20	19.4

SD—standard deviation; CV—coefficient of variation.

). A varied mercury accumulation in profiles resulted in distinctively higher pools of this element in the M1, M2 and M3 horizons (on average, 82.2% of total mercury pool), the M2 horizon having 32.5% of the total mercury pool. The mercury pool in the soils of region I was significantly higher compared with the pool in region II soils (

Table 11. Result of statistical analysis of mercury stocks in the regions (Anova, Tukey test).

Hgstock (mg·m−2) Mean ± SD		Significant Level
Region I (N = 4)	Region II (N = 5)
29.5 b ± 7.97	20.0 a ± 1.52	p = 0.0395

a,b – homogeneous groups.

).

Estimation of mercury pools in soils made it possible to evaluate interactions between the total mercury content and other soil parameters and was used to determine the degree of its accumulation. It should be stressed that mercury contents in the examined soils were multiple times lower than the permissible content in Poland which poses a risk of contamination [85], and which is 5.5 Hg kg⁻¹ dry soil mass for the 0–25 cm layer of organic soils. The regulatory depth of 0.25 cm corresponds approximately to the combined thickness of the M1 and M2 horizons which form the rhizosphere zone of the examined soils. The mercury contents determined in the horizons of the study soils are far higher than those reported for European Union soils by Ballabio et al. [4] who estimated the content’s median to be around 38.3 μg·kg⁻¹. Data reported for various soils on a worldwide basis show that mean concentrations of mercury in surface soils do not exceed 400 µg/kg [24]. The organic soils and paddy soils are likely to retain more than any other soils. The mean content of total mercury in agricultural surface soils and forest soils of Poland is estimated at 61 µg/kg, within the range 3.4 to 284.4 µg/kg.

The total mercury content in the analysed soils may be associated with both natural and anthropogenic sources. It is difficult to distinguish between these two sources. Many researchers point to the atmosphere as the main source of peatland contamination with mercury, it being either dry or wet deposition [14,20,25,86,87]. The phenomenon is global in character [1,19]. According to research by Benoit et al. [88], pre-industrial deposition in Minnesota (the USA) was approximately 4 μg·m⁻²·year⁻¹ and, in the 1990s, it exceeded 190 μg·m⁻²·year⁻¹. In the work by Coggins et al. [89], the index of mercury accumulation from 1950 to 1970 in various Irish peatlands was from 6 to 24 μg·m⁻²·year⁻¹. The large amount of mercury in peatlands compared to deposition and export in surface water runoff indicates that they have accumulated mercury from past atmospheric deposition [6].

Mercury accumulation in surface horizons of the study soils was the result of direct atmospheric deposition of this metal from local/global anthropogenic sources of its emission. Moreover, the murshic horizon (M1) is rich in humus colloids and it can bind more mercury whereas the existing vegetation may exacerbate the interception of this metal from the atmosphere. Osterwalder et al. [17] point out that mercury pools in peatlands can be a source of this metal for decades even if deposition decreases. The store of mercury in peat and organic soils is very large compared to runoff fluxes [17]. Changes in metal concentration in peats are associated with bulk density and ash content in the peat [89]. Numerous studies point to the trend of higher heavy metal concentrations in upper peat layers compared with lower ones [65,90,91,92].

In our study, values of correlation coefficients indicate that mercury accumulation in drained organic soils to a large extent may have resulted from organic matter transformation (

Table 12. Significant correlation coefficients between soil properties and mercury content, p < 0.05.

Parameter	Hgm/m	Hgm/v
Ash	0.762	0.873
ρa	0.766	0.933
TC	−0.802	−0.889
TC/TN	−0.770	−0.716
Lab-C (% TC)	0.626	0.567
HS-C (% TC)	0.796	0.886
FAs-C (% TC)	0.753	0.785
HAs-C (% TC)	0.730	0.826
Res-C (% TC)	−0.801	−0.889

ρ_a—bulk dencity, TC—Total organic carbon; TN—Total nitrogen; Lab-C—labile carbon; HS-C—carbon of humus substances; FAs-C—carbon of fulvic acids; HAs-C—carbon of humic acids; Res-C—carbon in residual fraction.

). Quantitative changes in physical and chemical parameters of the study soils during the process of secondary transformation are significantly correlated with mercury content. Organic matter mineralisation and increased ash content had a significant positive effect on mercury accumulation in the examined surface layer of the soils (Figure 1). Many works have confirmed that organic matter mineralisation in the process of secondary transformation results in an increase in the concentration of ash components in surface soil horizons [54,67,93]. In addition, research by Golovatskaya and Lyapina [49] has demonstrated a higher mercury content in surface and oxygenated layers of peat. Plants can act as pathways for the transfer of mercury from the geosphere to the atmosphere or as a source of mercury in the air–plant–soil system after the decomposition of organic matter [94,95].

A significant relationship between mercury accumulation and share of humic substances in the study soils is presented in Figure 2. Many researchers believe the extent of SOM decomposition and an increased share of humic substances in the soil mass account for the higher mercury content in drained organic soils compared with natural peatlands [8,49]. It is these substances that play an important role in the process of mercury binding in this type of soils [49]. Concentration of mercury in soil is a function of deposition rate and carbon turnover time. Mercury accumulation in soil is strongly influenced by the organic matter decomposition degree [96].

4. Conclusions

In our research, we found a significant effect of the secondary transformation of organic matter occurring in drained peatlands of the temperate climate zone on the total mercury content and stock in soils. The highest content and differentiation of mercury occurred in murshic horizons (up to a maximum depth of 43 cm). The average mercury content of the distinguished soil horizons is grouped in the following series (in μg kg⁻¹): M1 (212.0) > M2 (182.8) > M3 (126.3) > Pt (84.9). The mercury stock, up to a depth of 70 cm in the tested soils, ranged from 17.5 to 39.6 mg m⁻². As much as 82.2% of soil mercury was found in the upper murshic horizons. We found strong correlations between soil properties characteristically variable in the secondary transformation process and total mercury content. The increased content of humic substances in murshic horizons caused a significant increase in the total mercury content.

Our research is of great importance for soil monitoring, as the amount of determined mercury was greatly influenced by the depth of sampling (up to 25 cm). The results of the research should be taken into account when planning the restoration of peatlands of the temperate climate zone. There is a potential risk that elevated mercury concentrations in the upper murshic horizons may be a source of methylmercury for a long period of time. In peat soils with a high concentration of mercury, the risk of contamination with this toxic metal should be determined before re-irrigation.

Conscious actions should be taken to slow down the murshic process and thus reduce the emission of significant amounts of methylmercury and CO₂ into the atmosphere. Organic carbon contained in peatlands and natural meadows is a significant share in the circulation of this element in the environment. Mercury reserves in peat soils can be a reservoir of this metal for many decades and centuries. The assessment of this risk requires clarification and comprehensive scientific research in the future.