Edaphic factors affecting the vertical distribution of radionuclides in the different soil types of Belgrade, Serbia

Abstract

The specific activities of natural radionuclides (⁴⁰K, ²²⁶Ra and ²³²Th) and Chernobyl-derived ¹³⁷Cs were measured in soil profiles representing typical soil types of Belgrade (Serbia): chernozems, fluvisols, humic gleysols, eutric cambisols, vertisols and gleyic fluvisols. The influence of soil properties and content of stable elements on radionuclide distribution down the soil profiles (at 5 cm intervals up to 50 cm depth) was analysed. Correlation analysis identified associations of ⁴⁰K, ²²⁶Ra and ¹³⁷Cs with fine-grained soil fractions. Significant positive correlations were found between ¹³⁷Cs specific activity and both organic matter content and cation exchange capacity. Saturated hydraulic conductivity and specific electrical conductivity were also positively correlated with the specific activity of ¹³⁷Cs. The strong positive correlations between ²²⁶Ra and ²³²Th specific activities and Fe and Mn indicate an association with oxides of these elements in soil. The correlations observed between ⁴⁰K and Cr, Ni, Pb and Zn and also between ¹³⁷Cs and Cd, Cr, Pb and Zn could be attributed to their common affinity for clay minerals. These results provide insight into the main factors that affect radionuclide migration in the soil, which contributes to knowledge about radionuclide behaviour in the environment and factors governing their mobility within terrestrial ecosystems.

---

Environmental impact

The influence of edaphic factors on radionuclide distribution down soil profiles representing soil types typical for Belgrade, Serbia, and also for the region was examined in this study. Statistically significant correlations between radionuclides and some soil properties and stable element contents were identified. The knowledge gained contributes significantly to understanding the biogeochemical behaviour of radionuclides in the environment and factors governing their mobility within terrestrial ecosystems.

1. Introduction

Natural radioactivity in the soil arises mainly from primordial radionuclides formed in nucleosynthesis processes in stars before the early stage of the formation of the solar system, i.e.⁴⁰K and nuclides from the ²³²Th and ²³⁸U series and their decay products.¹ Natural radioactivity is of special environmental concern because the majority of the total radiation dose to the world population is from natural sources. Natural environmental radioactivity and the associated external exposure due to gamma radiation depend primarily on the geological and geographical conditions. Distribution of natural radionuclides is governed by weathering, sedimentation, leaching/sorption and precipitation from percolating groundwater or dilution with other materials with different composition, resulting in high variability in their specific activities.²

Anthropogenic radionuclides are derived from the radioactive fallout of nuclear fission products during nuclear weapon tests and from the Chernobyl accident in 1986. This accident resulted in significant deposition of ¹³⁷Cs on surface soils throughout Europe. Due to its relatively long half-life of 30.2 years and chemical behaviour similar to that of potassium, ¹³⁷Cs is one of the most significant radionuclides in the environment. Measurements of Chernobyl fallout clearly indicate that soil is the main reservoir of ¹³⁷Cs (ref. 3–5) but its migration behaviour and associated profile distribution are site-specific and depend on soil characteristics and environmental conditions.

Soil acts as a medium for transfer of radionuclides to biological systems and hence it is the basic indicator of radiological contamination of the environment. As part of different soil compounds, radionuclides are subjected to various biogeochemical processes that eventually determine their mobilization and availability for ecological processes. An understanding of the pathways by which radioactivity reaches biota requires an assessment of the soil's physical and chemical properties that affect the abundance and distribution of radionuclides. In recent years much knowledge on the effects of soil properties on biogeochemical mobilization of these radionuclides has been gained. Thus, Navas et al. (2002) showed that depth distribution of radionuclides is affected by soil properties, including pH, carbonates, organic matter and particle size, and soil processes, such as leaching and adsorption.⁶ An association between ²²⁶Ra and ²³²Th with oxides of iron and manganese was reported by Navas et al. (2005).⁷ Blanco Rodríguez et al. (2008) found a strong correlation between the finer particles of soil and radionuclide specific activities.⁸ Buccianti et al. (2009) and Belivermis et al. (2010) confirmed the influence of particle size on radionuclide specific activities in soil.^9,10 Nevertheless, there is a scarcity of information on the behaviour of radionuclides in the soils of Serbia.^11,12

In this study the depth distribution of the natural radionuclides, ⁴⁰K, ²²⁶Ra and ²³²Th, and Chernobyl-derived ¹³⁷Cs, was studied in profiles representing typical soil types of the Belgrade area.¹³ Although a substantial body of knowledge already exists in this area, in the majority of studies correlations between natural or anthropogenic radionuclides and a limited number of soil parameters, mainly particle size,^{6–10,14–22} pH, organic matter content,^{6,10,16–18,20–22}carbonate content^6,7,16 or selected stable elements,^7,20,21 have been investigated. To assess the pedogenic effect on the distribution of radionuclides, the relationships between radionuclide specific activities and a number of soil properties (particle size distribution, pH, organic matter content, carbonate content, cation exchange capacity, bulk density, particle density, saturated hydraulic conductivity and specific electrical conductivity) were analysed. The relationships between radionuclides and major and trace element concentrations in soil (Al, Ca, Cd, Co, Cr, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Sr, Pb, Zn and Ti) were also assessed to gain insight into the biogeochemical behaviour of the analysed radionuclides. An additional contribution of the study is that the analysed soil types are also major ones in the region.¹³

2. Materials and methods

2.1. Study area and soil sampling

Belgrade is the capital and the largest city of Serbia with a population of about 1.8 million. It lies 117 m above sea level in the northern part of central Serbia (N-44°49′14′′, E-20°27′44′′) at the confluence of the Danube and Sava rivers. The climate is moderate continental with four seasons. The high plasticity of the Belgrade relief, south of both rivers, has allowed the city to spread over many hills. North of the rivers Sava and Danube there are alluvial plains and loessial plateaus, which are divided by a steep section. There are two structural floors in the Belgrade area tectonics. The first is composed of jurassic and cretaceous sediments and magmatic products. Parts of the mesozoic joist on the west consist of flysh sediments of the upper cretaceous series, covered by neogen sediments with horizontal stratification important for the development of pedological cover. The second structural floor is composed of neogen sediments. More than 70% of the territory of Belgrade is covered by quaternary deposits of alluvial lake, alluvial marsh, alluvial, delluvial, delluvial–prolluvial, prolluvial and eolic sediments.²³

Soil profiles were sampled in May 2008 at six locations where the soil types are representative for the territory of Belgrade (Fig. 1). According to the FAO (2006) the soils are classified as: chernozems (A), fluvisols (B), humic gleysols (C), eutric cambisols (D), vertisols (E) and gleyic fluvisols (F).²⁴ To ensure representativeness, three subsamples were taken from each site. They were mixed to form a composite sample for analysis. Soil samples were collected every 5 cm from the uppermost layer down to 50 cm depth at each site. Samples were collected using a stainless steel spade, cylinders and a plastic scoop. To avoid cross-contamination disposable gloves were used and the equipment was first brushed to eliminate residues from the previous sample and then flushed with soil from the new sampling site. The surface vegetation, visible roots, stones and other debris were removed from the samples. The samples were then dried at room temperature to constant weight, homogenized and passed through a 2 mm sieve for radioactivity measurements and particle size analysis and ground in a mortar for other analyses.

Fig. 1 Simplified pedologic map of Belgrade showing the location of sampling sites.

2.2. Analytical methods

Radioactivity measurements were performed using an HPGe gamma-ray spectrometer (ORTEC-AMETEK, 34% relative efficiency and 1.65 keV FWHM for ⁶⁰Co at 1.33 MeV). The detector was surrounded by a shield consisting of 100 mm low-level lead (less than 50 Bq kg⁻¹ of ²¹⁰Pb), including an interior lining consisting of 2 mm Cu, 1 mm Cd and 4 mm Perspex. The samples were then kept in hermetically sealed Marinelli beakers of 500 mL volume for about 4 weeks to allow equilibrium between ²²⁶Ra and its daughters. The energy calibration and relative efficiency calibration of the gamma spectrometer were carried out using a 500 mL Marinelli calibration source MBSS2 supplied by Czech Metrological Institute and the standard gamma-ray spectrometry reference material IAEA-RGU-1 supplied by International Atomic Energy Agency (IAEA). For the purpose of quality assurance, the calibration was checked using standard reference materials (IAEA-375, EML 596 QAP 9803). Prior to the sample measurement, the environmental background at the laboratory site was determined with an empty Marinelli beaker under identical measurement conditions. Background spectral intensities were later subtracted from corresponding sample intensities. The activity of each sample was measured for 60 ks.

The specific activities of ⁴⁰K and ¹³⁷Cs were determined from their gamma-ray lines at 1460.8 keV and 661.6 keV, respectively. The specific activity of ²²⁶Ra was assessed from the gamma-ray lines of ²¹⁴Bi (609.3 keV) and ²¹⁴Pb (295.2 and 352.0 keV). The specific activity of ²³²Th was evaluated from the gamma-ray lines of ²²⁸Ac at 338.4, 911.1 and 968.9 keV, assuming that a state of secular equilibrium exists between ²³²Th, ²²⁸Ra and ²²⁸Ac. Gamma Vision 32 MCA emulation software was used to analyse gamma-ray spectra.²⁵ Self-absorption corrections were made for a given geometric setup based on coefficients obtained experimentally through measurements and comparison of the efficiency for samples with different densities.

The ¹³⁷Cs inventory was calculated using the following equation:²⁶

where Cs_inv is the ¹³⁷Cs inventory (Bq m⁻²), i is the sample index, n is the number of the deepest sample, C_i is the specific activity of ¹³⁷Cs in i^th soil sample (Bq kg⁻¹), BD_i is the bulk density of the soil (kg m⁻³) and DI_i is the thickness of i^th sample (m).

The traditional pipette method was used for particle size analysis.²⁷ Once the organic matter was removed by H₂O₂, the remaining mineral sample was weighed and subjected to particle size analysis to determine the following fractions: sand (0.05–2 mm), silt (0.002–0.05 mm) and clay (<0.002 mm). Soil pH was measured in 1 : 5 soil–water suspensions.²⁸ The organic matter content was determined by dichromate digestion based on the Walkley–Black method.²⁹Carbonates were measured volumetrically using Scheibler's calcimeter for CaCO₃ content.³⁰ The total cation exchange capacity of the sorptive complex was calculated as the sum of the hydrolytic acidity and total exchange bases, both measured according to Kappen (1929).³¹ Dry bulk density was determined using undisturbed soil cylinders (100 cm³).³² Particle density was measured with a pycnometer.³² Saturated hydraulic conductivity was measured by the falling-head method according to Klute and Dirksen (1986).³³ The specific electrical conductivity was measured in a 1 : 5 water suspension using a WTW inoLab ph/Cond 720 instrument.³⁴

For metal concentrations, samples were digested with HNO₃ and H₂O₂ in a microwave oven and then concentrations were determined using a Perkin Elmer PE3100/MHS-1 Atomic Absorption Spectrometer.³⁵ By the applied method all elements that could become environmentally available were dissolved. The standard reference material (SRM 2711) from the National Institute of Standards and Technology was used for quality assurance and quality control. The concentrations obtained were within 10% range of the certified values. Reagent blanks were also included in each batch of analyses to check for any contamination of the soil sample extracts. Average values for two replicates were taken for each determination.

The influence of soil depth and soil type on the distribution of analysed parameters was analysed by one-way ANOVA using the software package SPSS 16.0 for Windows.³⁶ The same software package was also used for evaluation of significant relationships between specific activities of radionuclides and soil physicochemical characteristics and stable element contents by Pearson correlation.

3. Results and discussion

3.1. Distribution of natural radionuclides and ¹³⁷Cs in soils

The mean values of natural radionuclide specific activities, i.e. 600 Bq kg⁻¹ for ⁴⁰K, 38.9 Bq kg⁻¹ for ²²⁶Ra and 46.6 Bq kg⁻¹ for ²³²Th (Table 1), fell within the range for Serbian soils and those from the Belgrade area.^37,38 The specific activities of ⁴⁰K, ²²⁶Ra and ²³²Th were found to be influenced by soil type (Table 2, Sig. ≤0.05). The mean specific activity of ¹³⁷Cs was found to be 18.5 Bq kg⁻¹. The specific activities of radionuclides (Table 1) varied by a factor up to 1.7 for ⁴⁰K, 2.8 for ²²⁶Ra and ²³²Th and 540 for ¹³⁷Cs.

Table 1 Basic descriptive statistics for radionuclide specific activities (Bq kg⁻¹)

Parameter	^40K	^226Ra	^232Th	^137Cs
Mean	600	38.9	46.6	18.5
Median	600	38.1	46.7	4.35
Mode	600	26.1	28.4	0.50
Standard deviation	74.4	11.8	11.5	33.2
Range	310	42.7	48.6	160
Minimum	450	23.0	27.6	0.30
Maximum	760	65.7	76.2	160

---

Table 2 The analysis of variance for radionuclides, major and trace elements and soil properties with depth and type of soil (F—the ratio of variation between groups and variation within groups, Sig.—significance)

Variable	Source of variation	F	Sig.
^40K	Depth	0.16	0.997
	Soil type	37.8	0.000
^226Ra	Depth	0.04	1.000
	Soil type	192	0.000
^232Th	Depth	0.19	0.994
	Soil type	54.6	0.000
^137Cs	Depth	9.32	0.000
	Soil type	1.033	0.408
Al	Depth	1.30	0.265
	Soil type	3.24	0.012
Ca	Depth	0.06	1.000
	Soil type	57.3	0.000
Cd	Depth	0.40	0.936
	Soil type	3.20	0.013
Co	Depth	0.45	0.899
	Soil type	8.57	0.000
Cr	Depth	0.26	0.982
	Soil type	52.3	0.000
Cu	Depth	0.48	0.883
	Soil type	1.80	0.127
Fe	Depth	0.55	0.832
	Soil type	8.27	0.000
K	Depth	1.08	0.397
	Soil type	10.1	0.000
Li	Depth	1.78	0.094
	Soil type	3.40	0.010
Mg	Depth	0.25	0.985
	Soil type	19.3	0.000
Mn	Depth	0.76	0.655
	Soil type	3.95	0.004
Na	Depth	0.80	0.624
	Soil type	1.70	0.152
Ni	Depth	0.77	0.647
	Soil type	3.33	0.011
Sr	Depth	0.88	0.545
	Soil type	3.70	0.006
Pb	Depth	1.45	0.190
	Soil type	7.84	0.000
Zn	Depth	0.57	0.818
	Soil type	8.80	0.000
Ti	Depth	1.17	0.335
	Soil type	2.05	0.086
Sand	Depth	0.10	1.000
	Soil type	50.6	0.000
Silt	Depth	1.71	0.112
	Soil type	8.31	0.000
Clay	Depth	1.20	0.311
	Soil type	9.81	0.000
pH	Depth	0.26	0.983
	Soil type	135	0.000
Organic matter	Depth	7.53	0.000
	Soil type	1.90	0.111
Cation exchange capacity	Depth	0.02	1.000
	Soil type	855	0.000
Carbonates	Depth	0.08	1.000
	Soil type	79.2	0.000
Bulk density	Depth	8.74	0.000
	Soil type	3.15	0.014
Particle density	Depth	5.98	0.000
	Soil type	4.48	0.002
Sat. hydraulic conductivity	Depth	61.5	0.000
	Soil type	0.32	0.897
Spec. electrical conductivity	Depth	3.78	0.001
	Soil type	8.30	0.000

---

The distribution of radionuclide specific activities down the depth profiles is presented in Fig. 2. In all soil profiles specific activities of natural radionuclides showed homogeneous distribution (Table 2). Homogeneous distribution of natural radionuclides was also reported by Navas et al. (2005) for soils in the Central Spanish Pyrenees and those from Burullus Lake, Egypt,^7,39 while increasing ⁴⁰K specific activities with depth were reported by Fujiyoshi and Sawamura (2004).⁴⁰ The differences in ⁴⁰K depth distribution have generally been attributed to the variability of organic matter and mineral composition of the soil and to soil biological activity including root distribution.⁴⁰ The specific activities of ¹³⁷Cs were maximal at 10–15 cm depth, which is in accordance with numerous studies showing that ¹³⁷Cs deposition is mostly contained within the surface soil layer.^41–43 The ¹³⁷Cs specific activities in the surface soil layers showed a wide range of values (from 8.3 to 160 Bq kg⁻¹ in the 0–5 cm layer and from 8.4 to 120 Bq kg⁻¹ in the 5–10 cm layer), which could be attributed to non-homogeneous surface contamination after the Chernobyl accident. The ¹³⁷Cs inventories calculated according to eqn (1) were found to vary between 60 Bq m⁻² for profile C and 320 Bq m⁻² for profile E, which is in accordance with values reported for Belgrade area after the Chernobyl accident.⁴⁴ The migration of ¹³⁷Cs in soil is slow because it is adsorbed onto clay minerals, silt and humic substances.^3,45,46 The extent of ¹³⁷Cs binding to soil components, i.e. the amount and distribution of this radionuclide in plant rooting zone, greatly influences its bioavailability.⁴⁷ The vertical distribution of ¹³⁷Cs is influenced not only by physical factors, such as the quantity and intensity of precipitation following deposition and the soil physicochemical properties, but also by biological factors, such as absorption and re-deposition by plants and biological characteristics of the soil.³ It has been found that microflora strongly contribute to the immobilization of ¹³⁷Cs in soils.⁴⁸ In profile B the specific activities of ¹³⁷Cs decreased with depth up to 25 cm and then slightly increased down the profile. As seen in Tables 3 and 4 some other physical properties of the soil (pH, organic matter, carbonates, K) showed discontinuity at the 25 cm depth. This soil profile (fluvisols) is collected in the alluvial plain of the Sava River which is flooded periodically. The physical and chemical characteristics of fluvisols depend on the nature and sequence of layers and length of periods of soil formation after or between flood events. The observed trend in the ¹³⁷Cs specific activities and other soil properties could be attributable to the different mechanical, mineralogical and chemical composition of layers in the studied fluvisols profile.⁴⁹ The ¹³⁷Cs depth distribution could be also influenced by bioturbation, which is considered by many authors to be an important cause of vertical movement of ¹³⁷Cs.⁴⁰ Following the Chernobyl accident there were a number of studies on migration of ¹³⁷Cs in soils. An exponential decrease of ¹³⁷Cs specific activities with soil depth and a strong influence of soil density on the ¹³⁷Cs profile were reported by Miller et al. (1990).⁵⁰ The application of a model based on an advection–dispersion equation showed significant regional differences in the apparent soil water velocity and in the dispersion coefficients, which were attributed to real physical differences in the soils.⁵¹

Fig. 2 Depth distribution of radionuclides in the 5 cm interval samples of the studied soil profiles (A—chernozems, B—fluvisols, C—humic gleysols, D—eutric cambisols, E—vertisols, and F—gleyic fluvisols). The uncertainties of the radioactivity measurements were 9–10% at the 95% level of confidence.

Table 3 Physicochemical characteristics of analysed soil profiles (the measurement uncertainties varied from 3 to 25%)

Site	Depth/cm	pH	Org. matter (%)	Carbonates (%)	Cation exch. capacity/cmol kg^−1	Bulk density/g cm^−3	Particle density/g cm^−3	Sat. hidr. cond./mm h^−1	Spec. el. cond./μS cm^−1
A	0–5	6.11	5.81	0.19	30.7	1.10	2.49	2370	170
	5–10	6.27	4.17	0.18	29.9	1.32	2.49	1300	85.8
	10–15	6.75	3.64	0.14	28.4	1.37	2.50	640	76.8
	15–20	6.51	3.26	0.00	29.6	1.28	2.52	410	85.5
	20–25	7.20	4.33	0.21	27.1	1.28	2.54	290	82.1
	25–30	6.77	5.03	0.19	29.8	1.45	2.58	210	81.3
	30–35	6.99	3.47	0.12	29.8	1.54	2.61	100	79.7
	35–40	7.61	3.38	0.17	32.6	1.48	2.64	110	76.1
	40–45	7.47	3.64	0.20	29.2	1.42	2.66	82.9	79.4
	45–50	7.87	4.12	0.18	30.6	1.38	2.67	59.8	84.1
B	0–5	8.50	4.81	10.1	100	1.03	2.59	3070	234
	5–10	8.75	2.81	9.48	100	1.19	2.59	2130	192
	10–15	8.82	2.16	9.77	100	1.14	2.58	1330	180
	15–20	8.83	2.26	9.10	100	1.14	2.62	690	177
	20–25	8.69	1.83	0.10	110	1.17	2.61	310	174
	25–30	8.87	2.09	9.65	100	1.18	2.64	260	176
	30–35	8.89	2.10	10.5	110	1.28	2.57	150	177
	35–40	8.94	2.17	10.1	100	1.31	2.61	100	178
	40–45	8.83	2.34	9.91	100	1.30	2.64	56.0	181
	45–50	8.97	3.05	9.43	100	1.30	2.67	21.1	184
C	0–5	8.02	8.38	4.29	97.3	1.10	2.34	1710	304
	5–10	8.16	7.05	4.84	110	1.30	2.39	1300	254
	10–15	8.39	5.90	5.75	98.1	1.35	2.39	960	220
	15–20	8.49	6.14	5.81	100	1.36	2.39	310	203
	20–25	8.60	3.91	6.11	99.2	1.38	2.48	120	189
	25–30	8.72	4.15	5.53	98.4	1.42	2.53	64.1	174
	30–35	8.82	4.00	5.43	97.5	1.38	2.59	34.0	167
	35–40	8.66	6.12	4.53	97.6	1.41	2.59	24.3	168
	40–45	8.83	3.76	4.09	96.4	1.49	2.61	10.5	167
	45–50	8.96	3.19	3.83	97.6	1.51	2.61	10.7	167
D	0–5	5.43	12.93	0.17	20.7	1.00	2.51	2110	242
	5–10	5.06	7.69	0.21	16.6	1.18	2.57	1640	113
	10–15	5.19	5.76	0.19	18.7	1.31	2.60	1060	92.4
	15–20	5.20	4.78	0.19	18	1.37	2.62	880	91.1
	20–25	5.43	3.79	0.16	14.5	1.36	2.64	600	79.9
	25–30	5.74	3.03	0.17	10.4	1.36	2.66	390	73.5
	30–35	5.82	1.53	0.24	11.1	1.52	2.65	150	54.3
	35–40	5.70	1.74	0.19	12.7	1.55	2.66	99.0	53.5
	40–45	5.53	1.41	0.21	13.6	1.60	2.67	40.0	52.1
	45–50	5.77	1.36	0.19	13.6	1.65	2.68	9.8	52.6
E	0–5	7.08	10.55	0.27	37.9	1.24	2.31	2520	180
	5–10	7.13	5.64	0.35	29.1	1.36	2.33	1240	109
	10–15	7.28	5.91	0.29	24	1.51	2.34	210	87.2
	15–20	7.22	5.26	0.26	26.4	1.42	2.39	160	85.5
	20–25	7.35	4.72	0.27	26	1.46	2.52	100	85.8
	25–30	7.60	5.09	0.29	24.4	1.46	2.59	54.1	81.8
	30–35	7.59	5.07	0.26	29	1.45	2.61	23.2	76.2
	35–40	7.74	2.86	0.30	27.6	1.47	2.64	20.5	72.4
	40–45	7.84	2.02	0.32	26.5	1.44	2.65	18.4	72.1
	45–50	7.85	2.07	0.30	30.4	1.47	2.68	12.4	78.9
F	0–5	6.94	10.46	0.32	47.8	0.94	2.53	2470	405
	5–10	7.22	7.50	0.27	36.5	1.28	2.53	1580	203
	10–15	7.20	5.31	0.05	28.6	1.35	2.56	610	145
	15–20	7.30	4.03	0.16	30.1	1.33	2.59	320	112
	20–25	7.33	3.14	0.16	31.8	1.38	2.62	220	93.6
	25–30	7.20	4.02	0.17	28.2	1.43	2.64	210	87.1
	30–35	7.06	2.72	0.10	23.6	1.48	2.66	98.5	81.1
	35–40	7.00	4.21	0.10	23.8	1.50	2.66	61.2	76.3
	40–45	7.55	3.97	0.21	22.4	1.53	2.67	19.9	74.7
	45–50	7.39	3.07	0.09	24.6	1.56	2.69	9.0	69.4

---

Table 4 Concentrations of major and trace elements (mg kg⁻¹) in analysed soil profiles (the measurement uncertainties varied from 5 to 20%)

Site	Depth/cm	Al	Ca	Cd	Co	Cr	Cu	Fe	K	Li	Mg	Mn	Na	Ni	Sr	Pb	Zn	Ti
A	0–5	10 900	5550	1.88	19.3	111	21.6	67 660	1080	14.1	5260	1040	128	55.6	26.0	72.5	59.0	71.3
	5–10	10 560	7780	2.25	20.8	122	22.4	79 340	1050	19.8	5330	860	120	56.9	28.2	62.1	71.3	62.9
	10–15	10 000	4950	2.20	18.7	118	19.9	68 800	1070	19.8	5400	980	122	46.4	29.5	81.4	64.3	63.5
	15–20	9690	5140	1.79	19.9	128	32.0	61 190	1040	25.3	5420	880	129	46.3	29.1	91.0	65.4	60.5
	20–25	9800	6460	2.07	20.7	124	25.4	80 010	1100	23.8	5640	1030	104	43.6	26.7	49.9	70.4	48.4
	25–30	10 200	5950	2.14	18.6	128	21.5	73 970	1080	18.6	5750	1170	117	63.1	28.7	66.7	70.9	48.7
	30–35	10 680	6590	1.80	20.3	108	20.3	82 000	1140	15.9	6040	1420	120	64.7	26.8	80.8	66.9	59.1
	35–40	10 630	9410	2.17	21.2	126	22.9	61 320	1240	21.7	6450	1370	122	57.5	30.3	72.0	58.4	68.3
	40–45	11 110	8560	2.50	20.7	132	28.3	74 550	1260	26.0	5980	1100	114	66.5	26.1	70.4	64.9	66.8
	45–50	10 230	7850	2.44	19.6	114	25.4	69 850	1180	20.1	5550	960	115	64.5	25.4	90.8	58.6	66.5
B	0–5	9140	19 090	1.48	14.0	113	18.3	71 670	1280	17.2	4740	850	94	43.9	25.0	59.2	69.8	49.9
	5–10	9590	23 520	1.75	14.3	136	25.5	61 890	1080	20.4	5240	930	103	53.1	27.9	63.2	67.7	40.9
	10–15	9920	31 060	1.67	16.9	153	25.5	94 180	1160	22.0	5510	1040	97.1	54.5	30.9	61.5	88.2	47.4
	15–20	9020	24 400	1.85	13.7	131	22.5	69 160	1240	20.4	5040	910	113	46.9	27.0	54.8	71.2	43.4
	20–25	10 140	28 160	1.74	14.4	129	28.7	85 590	1210	24.6	4920	1200	113	48.2	33.1	47.8	87.7	50.7
	25–30	9430	29 370	1.33	11.6	139	27.0	83 030	920	24.7	5330	1380	121	41.5	29.7	45.9	77.5	62.9
	30–35	10 430	33 330	2.13	15.6	151	25.7	99 750	1410	16.6	4980	1420	148	57.3	29.4	49.9	87.7	71.3
	35–40	10 540	32 710	2.14	21.8	126	27.0	86 560	1510	15.4	5680	1370	124	63.4	28.6	64.8	93.3	57.3
	40–45	11 080	30 040	2.05	19.9	124	26.9	80 970	1340	18.9	5390	1460	124	65.6	31.5	76.7	79.0	47.1
	45–50	10 140	23 610	2.20	19.2	127	22.0	67 180	1140	20.9	4250	1390	103	53.6	27.1	79.3	68.7	63.6
C	0–5	10 850	33 800	1.54	14.9	129	29.8	75 550	900	25.2	5020	950	111	45.4	28.3	47.2	45.8	47.3
	5–10	9520	18 270	1.82	18.6	141	30.0	73 840	910	17.3	4830	850	114	53.9	26.4	60.1	80.9	59.2
	10–15	10 290	20 780	2.06	17.9	157	31.9	93 400	970	18.5	5160	920	118	53.8	30.1	56.3	90.0	68.9
	15–20	9680	19 080	2.02	15.0	134	33.0	87 500	1460	24.0	5000	930	128	55.7	29.8	55.2	89.2	68.5
	20–25	10 260	22 690	2.10	18.2	137	27.0	85 780	1660	26.8	5400	1020	110	49.4	33.3	64.0	85.2	64.0
	25–30	8870	17 450	1.75	14.7	127	25.0	78 510	1160	23.5	4460	670	102	37.8	27.3	48.9	75.2	58.3
	30–35	9250	13 480	1.91	16.2	120	25.4	77 850	1380	22.9	5110	870	109	61.1	26.4	59.4	71.0	62.5
	35–40	9780	11 400	1.38	10.6	122	28.2	73 740	1300	14.9	5300	910	127	50.7	21.9	59.6	64.3	63.3
	40–45	9820	11 360	1.37	13.8	128	21.6	80 140	1430	21.9	4880	800	141	41.7	20.2	53.4	67.3	68.0
	45–50	9700	11 500	1.25	15.1	128	19.4	71 320	1433	23.2	5090	780	107	43.6	22.6	61.3	62.7	50.4
D	0–5	10 960	3410	2.17	22.7	46.4	24.4	84 510	1350	22.5	3500	1380	125	56.8	33.9	73.5	88.0	66.6
	5–10	10 720	3270	1.90	18.9	52.0	21.7	43 240	1360	21.6	3050	1370	127	44.0	31.6	56.9	36.1	57.4
	10–15	9560	4260	2.04	17.6	69.2	23.5	40 300	1320	25.0	2790	940	127	43.0	30.0	81.0	33.1	59.2
	15–20	9680	1700	1.77	18.4	68.7	23.6	40 060	1340	21.4	2890	770	116	47.3	27.5	76.1	36.9	60.8
	20–25	9670	1680	1.98	17.6	73.8	27.3	38 620	1350	20.4	2640	790	96.7	58.2	29.8	53.2	36.5	63.6
	25–30	10 520	3920	2.27	18.3	61.8	27.4	44 960	1350	25.5	2760	810	119	46.0	32.5	50.2	45.9	57.1
	30–35	10 410	4220	2.08	19.1	70.2	28.0	49 270	1490	20.8	2720	860	140	69.0	30.5	80.7	41.5	69.1
	35–40	9480	4270	1.90	17.1	76.3	25.2	33 650	1430	15.7	2900	760	112	47.5	28.8	73.2	36.7	65.6
	40–45	10 510	7370	2.39	19.1	70.3	26.7	42 010	1580	21.7	5120	860	106	58.2	28.8	71.4	44.2	57.9
	45–50	11 040	7000	2.10	19.4	78.6	29.0	51 480	1160	26.0	5640	990	121	67.9	34.1	72.2	39.2	72.2
E	0–5	10 860	11 010	1.72	18.2	96.7	21.9	55 580	1070	8.06	5590	670	123	42.1	31.7	34.3	61.7	51.2
	5–10	8610	5790	1.39	15.4	94.0	21.1	45 600	840	6.32	4810	730	96.1	32.7	19.6	46.0	45.5	54.7
	10–15	10 650	4980	1.90	20.1	110	24.4	61 580	980	7.78	5770	820	141	52.5	27.8	52.7	59.9	88.4
	15–20	9380	4970	2.26	14.5	96.4	24.4	58 750	960	16.6	4980	830	108	40.9	24.8	58.5	45.1	41.8
	20–25	9580	5600	2.18	15.9	115	22.4	48 470	970	17.5	5270	930	110	46.1	28.6	53.2	47.2	47.2
	25–30	9120	4780	1.68	15.8	84.1	22.3	57 970	840	21.1	4850	1020	121	47.2	24.0	45.5	47.4	62.2
	30–35	9040	5190	1.80	17.6	104	18.7	47 550	890	20.7	4530	1200	97.2	43.6	28.3	57.0	34.3	57.3
	35–40	9520	5110	2.38	17.1	106	26.4	96 570	890	16.8	4800	1040	105	40.5	27.2	62.0	44.0	52.1
	40–45	9060	3740	1.92	17.6	91.3	23.7	58 520	930	15.5	4680	1650	107	42.8	24.8	46.3	43.5	43.7
	45–50	9450	12 390	1.60	18.9	100	28.0	52 950	940	19.6	4900	1130	117	47.9	28.7	47.6	41.9	59.4
F	0–5	10 130	11 440	2.51	17.4	114	26.7	38 830	990	20.3	5570	1030	93.8	56.6	21.4	73.5	39.5	49.5
	5–10	10 020	6280	2.02	18.1	101	27.6	56 620	1040	18.1	5050	1140	96.3	48.3	22.5	54.7	63.1	44.2
	10–15	9660	6210	2.07	18.5	113	25.1	46 920	930	25.0	4770	980	96.9	47.7	25.0	75.4	54.5	46.3
	15–20	10 290	8180	1.94	20.0	128	24.4	59 840	1070	27.5	4490	1260	98.8	44.0	24.6	79.9	65.8	66.4
	20–25	10 290	8370	1.90	21.7	138	21.9	70 140	1470	26.8	5780	1450	97.5	64.1	30.1	68.1	71.5	43.4
	25–30	10 290	7610	1.61	19.0	134	22.1	80 650	1100	25.3	5320	1490	109	57.9	28.8	72.6	71.6	50.6
	30–35	10 880	3810	1.94	19.0	146	24.7	122 670	940	16.1	4820	1220	114	44.5	29.6	62.7	63.0	67.2
	35–40	10 130	7820	2.25	20.9	133	27.0	83 780	1010	12.4	5040	1180	127	41.0	28.0	69.8	80.9	64.7
	40–45	12 040	7500	1.96	19.8	112	23.7	81 250	1070	13.0	5240	1120	104	48.3	27.8	53.1	67.8	46.2
	45–50	10 100	7290	1.96	20.7	111	21.4	63 410	1070	12.4	5250	1100	119	48.7	24.6	83.2	58.7	62.1

---

3.2. The edaphic characteristics of analysed soils

The main physicochemical characteristics of the analysed soils are presented in Table 3. Most soils were found to be neutral to strongly alkaline and only those from sampling site D were moderately to strongly acidic. The average organic matter content varied from 2.56% for fluvisols to 5.26% for humic gleysols. For all soils, significant variations (Sig. ≤0.05) of organic matter content (decreasing trend) with soil depth were observed (Table 2). The carbonate content differed among profiles, with the average varying from 0.16% for gleyic fluvisols and chernozems to 8.81% for fluvisols. The cation exchange capacity varied from 15.0 cmol kg⁻¹ for gleyic fluvisols to 103 cmol kg⁻¹ for fluvisols. Soils were found to be similar with respect to density and moisture content. Saturated hydraulic conductivity of the analysed soils varied from 436 mm h⁻¹ for vertisols to 812 mm h⁻¹ for fluvisols. The electric conductivity ranged from 90.1 μS cm⁻¹ for chernozems to 201 μS cm⁻¹ for humic gleysols.

The particle size distribution with soil profile depth is shown in Fig. 3. The textural characteristics varied significantly (Sig. ≤0.05) among the soil types (Table 2). According to the USDA (1999), the analysed soil falls into the silty loam, silty clay loam or silt clay textural classes.⁵² No significant variance was observed for particle size fractions of analysed soils with depth (Table 2).

Fig. 3 Grain size distribution in the 5 cm interval samples of the studied soil profiles.

The concentrations of all major and trace elements in the soil profiles are shown in Table 4. No significant variation of these elements with soil depth was found (Table 2), while significant variation (Sig. ≤0.05) of all elements, except Cu, Na and Ti, with the soil type was observed.

3.3. Relationships between edaphic factors and radionuclides

The effects of soil physicochemical characteristics and major and trace element contents on radionuclide specific activities were analysed through correlations and the results are summarized in Table 5. Concerning the soil particle size, it is well documented that radionuclides are adsorbed onto clay surfaces or fixed within the lattice structure.^14,53,54 In our soils, clay and silt contents were positively correlated with ⁴⁰K, ²²⁶Ra (p < 0.01) and ¹³⁷Cs (p < 0.05) specific activities. The sand content was negatively correlated (p < 0.01) with ²²⁶Ra specific activity. This correlation analysis between radionuclide specific activities and particle size distribution confirmed results obtained world-wide showing that the fine-grained soil fraction has a higher tendency for radionuclide adsorption than coarse-grained soils since the soil particle surface area is larger.^15,16Minerals such as smectite, illite, vermiculite, chlorite, allophone and imogolite, as well oxides and hydroxides of silica, aluminium, iron and manganese are the most important for adsorption of radionuclides. This is due to the surface charge of these soil constituents and their three-dimensional structure. Clay minerals carry different kinds of charge, a variable charge, which can be either positive or negative, and a permanent charge which is merely negative.⁵⁵ The results of the previous study in the Belgrade area have shown that the mineral composition of the clay size mineral fraction, containing illite, smectite, vermiculite, chlorite, kaolinite, quartz and feldspar mix layer silicates, is the main factor of distribution and immobilization of radionuclides in soil.⁵⁶

Table 5 Pearson correlation coefficients between radionuclides and both soil properties and stable element contents of the studied soils

	^40K	^226Ra	^232Th	^137Cs
a Correlation is significant at the 0.01 level. b Correlation is significant at the 0.05 level.
Sand	0.05	−0.49^a	−0.24	0.01
Silt	0.33^a	0.49^a	−0.11	0.28^b
Clay	0.34^a	0.28^a	0.20	0.29^b
pH	0.10	−0.30^b	−0.59^a	−0.13
Organic matter	−0.18	−0.19	0.09	0.73^a
Cation exch. capacity	−0.27	−0.27	−0.25	0.41^a
Carbonates	0.08	−0.67^a	−0.73^a	−0.13
Bulk density	−0.05	0.05	0.50	−0.52^a
Particle density	0.15	0.11	−0.07	−0.61^a
Sat. hydraulic conductivity	−0.04	−0.03	−0.19	0.68^a
Spec. electrical conductivity	−0.09	−0.44^a	−0.57^a	0.61^a
Al	0.19	0.13	−0.15	0.05
Ca	0.12	−0.57^a	−0.75^a	0.01
Cd	0.15	0.30^b	0.11	0.41^a
Co	0.18	0.32^b	0.12	0.02
Cr	0.45^a	−0.09	0.60^a	0.68^a
Cu	−0.06	−0.24	−0.16	0.06
Fe	0.31	0.48^a	0.51^a	−0.22
K	0.81^a	−0.06	−0.16	−0.33^b
Li	0.17	−0.04	−0.18	−0.26^b
Mg	0.14	0.21	−0.28^b	0.06
Mn	0.12	0.54^a	0.37^a	−0.16
Na	−0.01	0.17	−0.05	−0.21
Ni	0.35^a	0.26^b	0.54^a	−0.14
Sr	−0.06	0.08	−0.13	−0.17
Pb	0.44^a	0.36^a	0.67^a	0.56^a
Zn	0.35^a	0.60^a	0.74^a	0.48^a
Ti	0.05	0.14	0.06	−0.16

---

The soil pH was found to be unrelated to ⁴⁰K and ¹³⁷Cs specific activities, but was negatively correlated with ²²⁶Ra (p < 0.05) and ²³²Th (p < 0.01) specific activities. Negative correlations between pH and radionuclides of the uranium and thorium series were recorded by Tsai et al. (2011).¹⁷ Belivermis et al. (2010) reported a negative correlation between pH and ²³²Th specific activity in soil.¹⁰ The lack of correlation between pH and natural radionuclides specific activity reported by Navas et al. (2002) was attributed to a variety of pedogenic processes, such as eluviation and sorption to soil components, which also affect the distribution of radionuclides in soils.⁶ However, Baeza et al. (1995) did find negative correlations between ¹³⁷Cs specific activity and pH.¹⁸

Positive correlations between ¹³⁷Cs specific activity and both organic matter content and cation exchange capacity (p < 0.01) were found. Elejalde et al. (1996) also observed that ¹³⁷Cs specific activities were related to organic matter and cation exchange phenomena.⁵⁷ Organic matter is a component of great importance because it tends to form soluble or insoluble complexes with radionuclides, which can migrate throughout the profile or be retained in the soil.^58,59 No significant correlation was found between the organic matter content and ²²⁶Ra and ²³²Th specific activities which is in accordance with earlier results.^16,19,57 The absence of significant correlation between ²²⁶Ra and ²³²Th and organic matter content in analysed soils can be explained by the strong tendency of organic matter content to be inversely correlated with depth while these radionuclides showed homogeneous depth distribution. However, Vandenhove and Van Hees (2007), working with spiked soils, showed that the radium concentrations in soil solution are related to the organic matter or cation exchange capacity of soils.⁶⁰ Taboada et al. (2006) observed that thorium mobility in soils developed on granitic rocks is related to the presence of organic matter, particularly of simple organic acids and fulvic and humic acids of low molecular weight, which form complexes with thorium in the soil profile.²⁰ Since organic matter is extremely heterogeneous and consists of organic acids, lipids, lignin, fulvic and humic acids, there are a large number of possible reactions and interactions of radionuclides with organic matter. Organic matter contains functional groups that can form complexes with radionuclides. This complexation affects radionuclide mobility, adsorption to soils and bioavailability.⁵⁵

Negative correlations (p < 0.01) were observed between the carbonate content and ²²⁶Ra and ²³²Th specific activities, which confirms earlier findings.^6,7,57 The negative correlation of these radionuclides with carbonates suggests their binding in soils with minerals other than calcite, probably in silicates derived during weathering processes from parent rocks.

Soil density was negatively correlated (p < 0.01) with ¹³⁷Cs, which is in accordance with the findings of Ligero et al. (2001).¹⁹ No correlation between soil density and natural radionuclides was found, which is consistent with the results of Tsai et al. (2011).¹⁷ However, in the soils of German forests, the specific activities of ⁴⁰K were positively correlated with soil density, which indicated that most of the potassium was contained within the mineral components of the soil.⁴⁰

Positive correlations (p < 0.01) between ¹³⁷Cs specific activity and both saturated hydraulic conductivity and specific electrical conductivity were observed. Electrical conductivity was negatively correlated with ²²⁶Ra and ²³²Th specific activities. Tsai et al. (2011) found a weak correlation between ⁴⁰K and electrical conductivity.¹⁷

The correlations between radionuclides and major and trace elements (Table 5) revealed that ⁴⁰K was positively correlated (p < 0.01) with K (as it is present in natural potassium with a constant abundance), Cr, Ni, Pb and Zn, which is in accordance with their common correlations to clay. Strong positive correlations of ⁴⁰K with Cr and Ni were also reported by Van der Graaf et al. (2007) and those with Pb and Zn by Al-Trabulsy et al. (2010).^21,61

Close positive correlations (p < 0.01) were found between ²²⁶Ra and ²³²Th and Fe and Mn, which indicate association of these radionuclides with Fe and Mn oxides (assuming the secular equilibrium between ²³²Th, ²²⁸Ra and ²²⁸Ac) or deposition of Fe and Mn oxides on the surfaces of ²²⁶Ra and ²³²Th minerals. This association is supported by strong positive correlations with Ni, Pb and Zn which also appear to be related to the above-mentioned oxides. Similar relationships have been found in other studies.^7,62 A positive correlation (p < 0.05) between ²²⁶Ra and Co was also observed, which agrees with the findings of Chao and Chuang (2011).⁶² A negative correlation (p < 0.01) between ²²⁶Ra and ²³²Th and Ca was found, which is in accordance with the negative correlation of these radionuclides with carbonates. A similar relationship was also found by Navas et al. (2005).⁷ The strong positive correlation (p < 0.01) observed between ²³²Th and Cr confirms the results of Navas et al. (2005) and Van der Graaf et al. (2007).^7,61

The anthropogenic ¹³⁷Cs was found to be positively correlated (p < 0.01) with Cd, Cr, Pb and Zn, which is in accordance with earlier results for soils from the Zlatibor area, Serbia.²² These correlations could be attributed to the common affinity of these elements for clay minerals.⁶¹ The strength of these correlations may be influenced by variations in pH and/or organic substances.

The behaviour of radionuclides depends on elemental properties of the radionuclides, on the mineral and organic inventory of the soil and the chemical reaction milieu.⁵⁵ Radionuclides and metal ions can be retained in soil by (ad)sorption, precipitation and complexation reactions.⁶³ Their interaction with the soil environment depends on both soil properties and environmental factors. The concentration of competitive elements present in the soil is of particular importance for determining radionuclide distribution between soil and soil solution. Given that the number of sites at which ions may be adsorbed are limited, the adsorption of any particular species decreases as the concentration of competing ions increases. Thus, the adsorption of radium has been shown to be strongly dependent on ionic strength and concentrations of other competing ions.⁶⁴ The results of Nathwani and Phillips (1979) showed that the addition of competing alkaline earth cations to the soil system can greatly affect radium sorption on the clay minerals.⁶⁵ Sturchio et al. (2001) observed the decrease in radium sorption with increasing concentrations of Ca²⁺, Ba²⁺ and Mg²⁺in soil which occupy sorption sites available for radium.⁶⁶ These elements are likely to undergo cation exchange on clay minerals. The concentration of an ion in solution in most soils is determined by cation exchange reactions with the soil matrix which by their nature are competitive but other processes, e.g.co-precipitation, also depend on the concentrations of competing substances in solution.⁶⁷ For radiocesium, these competitive effects formed the basis of countermeasures at the soil-to-plant level after a nuclear accident.⁶⁸

Conclusion

In this study the influence of a number of edaphic factors on radionuclide distribution down the depth profile was analysed. The significant variations in radionuclide specific activities and major and trace element concentrations among different soil types were observed. Among radionuclides, different behaviour was distinguished with respect to the soil depth. The distribution of ⁴⁰K, ²²⁶Ra and ²³²Th was found to be constant down the soil profile depth, while the large variability was observed for ¹³⁷Cs depth distribution. The correlations between soil properties and radionuclide specific activities found in this study could serve as a basis for determination of radionuclide bioavailability in different ecosystems. Clay and silt contents were positively correlated with ⁴⁰K, ²²⁶Ra and ¹³⁷Cs. Close positive correlations were found between ²²⁶Ra and ²³²Th and Fe and Mn, which indicates the association of these radionuclides with Fe and Mn oxides, supported also by strong positive correlations with Ni, Pb and Zn. A negative correlation between ²²⁶Ra and ²³²Th and Ca was found, which is in accordance with the negative correlation of these radionuclides with carbonates. The strong positive correlation was observed between ²³²Th and Cr. The anthropogenic ¹³⁷Cs was found to be positively correlated with Cd, Cr, Pb and Zn, which confirmed their common affinity for clay minerals. As the large number of parameters which could influence radionuclide migration in soil were analysed, the findings of the study could be useful in the modeling of radionuclide migration in soils. A number of models are already developed to quantify radionuclide mobility of radionuclides in soils but they are not generally applicable because they require many input parameters that are specific for each site. The behaviour of radionuclides in soil is confirmed as a complex phenomenon that depends on a number of physical, chemical and biological properties of the soil. In addition to the soil parameters examined in this study, in further investigations bioturbation should be analysed, by including data on populations and activities of (micro)organisms and root distributions.

Acknowledgements

This work was supported by the Ministry of Education and Science of the Republic of Serbia (projects III43009 and 173030).

References

1. UNSCEAR (United Nations Scientific Committee on the Effect of Atomic Radiation), Sources and Effects of Ionizing Radiation, UNSCEAR 2008 Report, United Nations, New York, 2010.
2. M. Dowdal and J. O'Dea, J. Environ. Radioact., 2002, 59, 91–104.
3. P. L. Nimis, Studia Geobot., 1996, 15, 3–49.
4. B. Rafferty, M. Brennan, D. Dawson and D. Dowding, J. Environ. Radioact., 2000, 48, 131–143.
5. L. E. C. Ciuffo, M. Belli, A. Pasquale, S. Menegon and H. R. Velasco, Sci. Total Environ., 2002, 295, 69–80.
6. A. Navas, J. Soto and J. Machín, Eur. J. Soil Sci., 2002, 53, 629–638.
7. A. Navas, J. Machín and J. Soto, Soil Sci., 2005, 170, 743–757.
8. P. Blanco Rodríguez, F. Vera Tomé, J. C. Lozano and M. A. Péréz-Fernández, J. Environ. Radioact., 2008, 99, 1247–1254.
9. A. Buccianti, C. Apollaro, A. Bloise, R. De Rosa, G. Falcone, F. Scarciglia, A. Tallarico and G. Vecchio, Geoderma, 2009, 152, 145–156.
10. M. Belivermis, Ö. Kiliç, Y. Çotuk and S. Topcuoğlu, Environ. Monit. Assess., 2009, 163, 15–26.
11. D. Krstić, D. Nikezić, N. Stevanović and M. Jelić, Appl. Radiat. Isot., 2004, 61, 1487–1492.
12. S. Nenadović, M. Nenadović, I. Vukanac, A. Djordjević, S. Dragićević and M. Lješević, Nucl. Technol. Radiat. Prot., 2010, 25, 30–36.
13. European Soil Bureau Network, Soil Atlas of Europe, European Commission, Office for Official Publications of the European CommunitiesLuxembourg 2005.
14. A. J. VandenBygaart and R. Protz, Can. J. Soil Sci., 1995, 75, 73–84.
15. A. Baeza, M. Del Río, A. Jiménez, C. Miró and J. Paniagua, J. Radioanal. Nucl. Chem., 1995, 189, 289–299.
16. A. Navas, J. Soto and J. Machín, Appl. Radiat. Isot., 2002, 57, 579–589.
17. T. Tsai, C. Liu, C. Chuang, H. Wei and L. Men, J. Radioanal. Nucl. Chem., 2011, 288, 927–936.
18. A. Baeza, M. Del Río, A. Jiménez and C. Miró, Radiochim. Acta, 1995, 68, 135–140.
19. R. A. Ligero, I. Ramos-Lerate, M. Barrera and M. Casas-Ruiz, J. Environ. Radioact., 2001, 57, 7–19.
20. T. Taboada, A. Martínez Cortizas, C. García and E. García-Rodeja, Sci. Total Environ., 2006, 356, 192–206.
21. H. A. M. Al-Trabulsy, A. E. M. Khater and F. I. Habbani, Arabian J. Chem., 2010 DOI:10.1016/j.arabjc.2010.10.001/.
22. S. Dragović, N. Mihailović and B. Gajić, Chemosphere, 2008, 72, 491–495.
23. M. Komatina and S. M. Komatina, in Groundwater in the Urban Environment. Selected City Profiles, ed. J. Chilton, Balkema, Rotterdam, 1999, pp. 317–322.
24. FAO (Food Agricultural Organization), World Reference Base for Soil Resources. A Framework for International Classification and Communication, World Soil Resources Reports No. 103, Food and Agriculture Organization of the United Nations, Rome, 2006.
25. ORTEC, Gamma Vision 32, Gamma-Ray Spectrum Analysis and MCA Emulation, ORTEC, Oak Ridge, Version 5.3, 2001.
26. R. A. Sutherland, Hydrol. Processes, 1992, 6, 215–225.
27. D. L. Rowell, Bodenkunde. Untersuchungsmethoden und ihre Anwendungen, Springer, Berlin, 1997.
28. ISO 10390:2005, Soil Quality—Determination of pH, International Standard Organization, Geneva.
29. L. P. Van Reeuwijk, Procedures for Soil Analysis. Technical Paper 9, International Soil Reference and Information Centre (ISRIC), Wageningen, 1986, p. 106.
30. ISO 10693:1995, Soil Quality—Determination of carbonate content–Volumetric Method, International Standard Organization, Geneva.
31. H. Kappen, Die Bodenaziditaät, Springer Verlag, Berlin, 1929.
32. G. R. Blake and K. H. Hartge, in Methods of Soil Analysis. Part 1-Physical and Mineralogical Methods, ed. A. Klute, Soil Science Society of America, Madison, WI, 2nd edn, 1986, pp. 363–382.
33. A. Klute and C. Dirksen, in Methods of Soil Analysis. Part 1-Physical and Mineralogical Methods, ed. A. Klute, Soil Science Society of America, Madison, WI, 2nd edn, 1986, pp. 687–734.
34. ISO 11265:1994, Soil Quality—Determination of the Specific Electrical Conductivity, International Standard Organization, Geneva.
35. USEPA (United States Environmental Protection Agency), Method 3050B: Acid Digestion of Sediments, Sludges, and Soils, Revision 2, Washington DC, 1996.
36. SPSS (Statistical Package for the Social Sciences) 16.0, Chicago, Illinois, 2007.
37. S. Dragović, Lj. Janković, A. Onjia and G. Bačić, Radiat. Meas., 2006, 41, 611–616.
38. Lj. Janković-Mandić and S. Dragović, Radiat. Prot. Dosim., 2010, 140, 369–377.
39. H. I. El-Reefy, T. Sharshar, R. Zaghloul and H. M. Badran, J. Environ. Radioact., 2006, 87, 148–169.
40. R. Fujiyoshi and S. Sawamura, Sci. Total Environ., 2004, 320, 177–188.
41. Z. Pietrzak-Flis, I. Radwan, L. Rosiak and E. Wirth, Sci. Total Environ., 1996, 186, 243–250.
42. S. Yoshida, Y. Muramatsu, A. M. Dvornik, T. A. Zhuchenko and I. Linkov, J. Environ. Radioact., 2004, 75, 301–313.
43. Ö. Karadeniz and G. Yaprak, Radiat. Prot. Dosim., 2008, 131, 346–355.
44. Federal Secretariat for Labour, Health, Veterans Affairs and Social Policy, Radioactivity of the Environment of Yugoslavia, Belgrade, 1990.
45. J. P. Absalom, S. D. Young, N. M. J. Crout, A. Sanchez, S. M. Wright, E. Smolders, A. F. Nisbet and A. G. Gillett, J. Environ. Radioact., 2001, 52, 31–43.
46. N. Kruyts and B. Delvaux, J. Environ. Radioact., 2002, 58, 175–190.
47. J. T. Smith and D. G. Elder, Eur. J. Soil Sci., 1999, 50, 295–307.
48. A. Bruckmann and V. Wolters, Sci. Total Environ., 1994, 157, 249–256.
49. G. Tóth, L. Montanarella, V. Stolbovoy, F. Máté, K. Bódis, A. Jones, P. Panagos and M. Van Liedekerke, Soils of the European Union, JRC Scientific and Technical Reports, Office for Official Publications of the European Communities, Luxembourg, 2008.
50. K. M. Miller, J. L. Kuiper and I. K. Helfer, J. Environ. Radioact., 1990, 12, 23–47.
51. P. Bosew and G. Kirchner, J. Environ. Radioact., 2004, 73, 127–150.
52. USDA (United States Department of Agriculture), Soil Taxonomy. A Basic System of Soil Classification for Making and Interpreting Soil Surveys, Handbook No. 436. Soil Survey Staff, Washington DC, 1999.
53. M. Jasinka, T. Niewiadomski and J. Schwbenthan, in Natural Radiation Environment, ed. K. Vohra, U. C. Mishra, K. C. Pillai and S. Sadasivan, Willey, New York, 1982, pp. 206–211.
54. A. Cremers, A. Elsen, P. De Preter and A. Maes, Nature, 1988, 335, 247–249.
55. H. Koch-Steindl and G. Pröhl, Radiat. Environ. Biophys., 2001, 40, 93–104.
56. Z. P. Tomić, A. R. Đorđević, M. B. Rajković, I. Vukašinović, N. S. Nikolić, V. Pavlović and Č.M. Lačnjevac, Sens. Transducers J., 2011, 125, 115–130.
57. C. Elejalde, M. Herranz, F. Romero and F. Legarda, Water, Air, Soil Pollut., 1996, 89, 23–31.
58. C. A. Campbell, in Soil Organic Matter (Developments in Soil Science), ed. M. Schnitzer and S. V. Khan, Elsevier, Amsterdam, 1978, vol. 8, ch. 5, pp. 173–265.
59. F. A. Vega, E. F. Covelo, M. L. Andrade and P. Marcet, Anal. Chim. Acta, 2004, 524, 141–150.
60. H. Vandenhove and M. Van Hees, Chemosphere, 2007, 69, 664–674.
61. E. R. Van der Graaf, R. L. Koomans, J. Limburg and K. De Vries, Appl. Radiat. Isot., 2007, 65, 619–633.
62. J. H. Chao and C. Y. Chuang, Appl. Radiat. Isot., 2011, 69, 261–267.
63. E. D. van Hullebusch, P. N. L. Lens and H. H. Tabak, Rev. Environ. Sci. Bio/Technol., 2005, 4, 185–212.
64. USEPA (United States Environmental Protection Agency), Understanding Variation in Partition Coefficient, K_d, values, EPA 402-R-04-002C, Washington DC, 2004, vol. 3.
65. J. S. Nathwani and C. R. Phillips, Chemosphere, 1979, 8, 293–299.
66. N. C. Sturchio, J. L. Banner, C. M. Binz, L. B. Heraty and M. Musgrove, Appl. Geochem., 2001, 16, 109–122.
67. S. Ehlken and G. Kirchner, J. Environ. Radioact., 2002, 58, 97–112.
68. B. J. Howard and G. Desmet, Sci. Total Environ., 1993, 137, 1–315.

---