Assessment of Potentially Toxic Elements Pollution Pattern and Environmental Risk in Soils from Carpathian Areas Using a GIS-Based Approach and Pollution Indices

Abstract

Understanding the occurrence and spatial variability of potentially toxic elements in soils is essential for tracing pollution origins, assessing ecological risks, and supporting sustainable land use management. This study investigates the soil pollution with Cd, Pb, Ni, Cu, Zn, Cr, As, Mn, Sr, and Fe, their spatial distribution, and environmental risks in two areas in southwestern Romania—Isverna and Tismana—using a combination of pollution indices and Geographic Information System (GIS)-based analyses. Results indicated predominantly low to moderate pollution across both areas, with localized hotspots of high to extreme pollution, particularly with Ni and Pb, near human settlements. In contrast, Tismana showed more uniform, generally lower pollution levels, suggesting the influence of natural (lithogenic) sources. Spatial distribution maps highlighted these differences, showing more heterogeneous localized hotspots in Isverna, likely linked to anthropogenic activities such as agricultural runoff and improper domestic waste disposal. The integrated use of pollution indices and GIS mapping proved effective in identifying contamination patterns and risk zones, providing valuable insights for environmental monitoring and sustainable management of rural land.

1. Introduction

Soil is a vital, non-renewable natural resource situated at the interface of the biosphere, atmosphere, and hydrosphere. It supports food production, maintains or enhances the physico-chemical characteristics of water and air, sustains ecosystems, and promotes human health and habitation [1,2]. In addition, soil functions as both a sink and a source of potentially toxic elements (PTEs), thereby mitigating their environmental impacts [3,4]. Due to its extremely slow rate of formation processes, soil is widely regarded as a non-renewable resource [5].

Among the most common soil contaminants are inorganic compounds, including metals (e.g., Cd, Cr, Cu, Hg, Mn, Ni, Pb, V, and Zn) and metalloids (e.g., As, Be, and Sb) [6]. Some elements function as essential macro- and micronutrients with beneficial effects at trace levels but become toxic when their concentration exceeds certain thresholds. Other elements are toxic even at very low concentrations [6]. Most of these elements are naturally occurring in soils, typically in low concentrations or in forms with low bioavailability [7].

Soil quality may deteriorate due to human-induced pressures, including unsustainable agricultural and forestry practices, industrial activities, tourism, urban expansion, and construction [8,9]. Although on a lower scale, these pressures are also present in rural areas. PTEs enrichment in soil may occur due to both natural processes and anthropogenic activities, and is strongly linked to land use [10]. Elevated concentrations of PTEs, particularly in agricultural soils, represent a significant environmental concern due to their potential accumulation in the food chain and their negative impact on food security and safety [11,12,13,14].

Soil pollution indices are widely recognized as effective tools for assessing the contamination level of PTEs as they integrate complex geochemical datasets into a single, understandable value [4,15]. These indices support the assessment of current PTE contamination status, as well as projections of PTE accumulation and its associated long-term environmental implications [16,17]. Moreover, pollution indices enable the distinction between natural and anthropogenic contamination sources, making them suitable for monitoring human-induced environmental impacts [18,19].

Understanding the spatial variability of soil properties allows data-driven decisions in soil quality evaluation, pollution extent identification, decontamination planning, and land management [20]. Selecting and prioritizing the most suitable areas is one of the most critical steps in land suitability analysis. Geographic Information Systems (GIS) serve as powerful tools for integrating, managing, and analyzing spatial data. As a spatial planning instrument, GIS allows decision-makers to manage, store, analyze, and visualize datasets derived from various sources [21]. GIS also provides a robust geospatial framework for environmental impact assessment, such as soil contamination, by combining spatial and statistical approaches [22]. Standard GIS-based interpolation techniques such as Inverse Distance Weighting (IDW), spline, and various forms of kriging (e.g., Ordinary Kriging, Empirical Bayesian Kriging, Geographical Weighted Regression Kriging, etc.) are widely used to map contamination gradients and identify hotspots. Geostatistical analyses further optimize model parameters, assess prediction uncertainty, and validate performance through cross-validation [23,24,25,26,27]. Spatial autocorrelation tools, such as Moran’s index, detect clustering patterns and highlight significant contamination accumulation zones [28,29]. Several studies have been conducted to assess the spatial distribution of PTEs in soils at different local, regional, and national scales by means of GIS approaches [24,30,31,32,33,34,35].

While numerous studies have assessed pollution levels in industrial areas using contamination indices [36,37,38], information on the occurrence and spatial distribution of PTEs in rural soils remains limited. Although metal contents in Romanian rural soil are generally low [39,40,41,42], maintaining soil quality is essential for ensuring food security. In this context, the current research aims to investigate the extent of PTE contamination in soils exposed to various anthropogenic pressures and agricultural practices in two karstic areas in southwestern Romania. The research combines several pollution and ecological risk indices calculated from soil PTE content, with GIS-based interpolation techniques to map the spatial distribution of PTE contamination and associated environmental risks.

2. Materials and Methods

2.1. Study Area

The study sites are in the SW of the Southern Carpathians: (i) Isverna (IVR) village located in Coșuștea catchment, Mehedinți Mountains, and (ii) Tismana (TSM) town located in the Tismana catchment, Vâlcan Mountains (Figure 1).

The relief is generally mountainous, with some exceptions in the Isverna area, whose southern part is a plateau. In the Vâlcan Mountains (Tismana area), elevations range from 300 to 1000 m, whereas in the Mehedinți Mountains (Isverna area), they range from 430 to 1300 m. The river networks are well-developed in both catchments, with water conduits reorganized underground in karst areas. The sampling points were selected based on the underlying bedrock type and hydraulic transport. In the Tismana basin, seven sampling points are located along the valleys, and 11 points are randomly distributed on the adjacent slopes. On impervious rocks such as granites and metamorphics, mobilization of PTEs by hydraulic transport is very slow (sampling points TSM1-7, IVR6−16), especially on gentle slopes and in ephemeral tributaries. Consequently, PTEs’ content could be higher in these sites. Instead, the residence time is much shorter in porous media such as carbonate rocks (sampling points TSM10−18 and IVR1, 2, 17−20), or at the contact between non-carbonate and carbonate rocks, where percolating waters transport pollutants through the overlying soils directly into the karst aquifers [43]. To the best of our knowledge, no karst polje or border polje has been documented in the study areas. Non-karstic depressions are located at considerable distance from the study area; therefore, both in the Tismana and Isverna karst areas and their overlying soil cover are the main source for aquifers rather than water accumulation zones. The low-altitude sampling points (IVR8−IVR16) in the Isverna area are located within the metamorphic and flysch domains. Transitional zones between carbonate and non-carbonate lithologies, influenced by allogenic water and sediment input, may function as poljes in terms of soil genesis and contaminant transport [44], but are far from the study area.

The climate is classified as humid continental, the warm-summer or hemiboreal subtype (Dfb) in the Koppen–Geiger climate classification system [45]. Vegetation is mainly composed of sessile oak forests (Quercus petraea) transitioning to beech forests (Fagus sylvatica var. moesica) at altitudes higher than 600 m. Under the pronounced south-westerly, Mediterranean climatic influences, the rocky limestone surfaces provide conditions for the local development of thermophilous and calciphilous species in a vegetation type locally called ‘șibleac’, which comprises the common lilac (Syringa vulgaris), the manna ash (Fraxinus ornus), or the Oriental hornbeam (Carpinus orientalis).

The Isverna’s unique Mediterranean scrubs are protected as a nature reserve (IUCN category IV) of 10 ha. Rural communities have transformed the natural landscape for centuries, replacing parts of the forest with pastures, hayfields, and arable land.

Figure 1. (A). Geographic setting of the study areas on the Romanian territory; Geological maps of the (B). Tismana (TSM) and (C). Isverna (IVR) study areas, adapted from [46,47], respectively.

The soils in the Isverna area developed on a complex lithology, favored by the exposure of the main structural units of the Southern Carpathians over a short distance, of only ~5 km: the Danubian units, the Severin Unit, and the Getic Unit. The high-grade metamorphic rocks and granitoids that form the core of the Danubian units are unconformably overlain by lower-grade metamorphosed conglomerates, sandstones, and siltites of Ordovician–Silurian and Devonian ages [48,49]. These are followed by a Mesozoic sedimentary cover, which in the Isverna area consists mainly of Upper Jurassic to Lower Cretaceous reef limestones and Upper Cretaceous deposits: terrigenous and volcaniclastic turbidites, a pre-mélange unit (Nadanova Beds), and a tectonic mélange [46,47]. The Severin Unit comprises Late Jurassic oceanic crust rocks (an ophiolitic mélange with a serpentinitic matrix and siliceous pelagic rocks) and Early Cretaceous turbidites [50].

In the Tismana basin, only the Lower Danubian granitoid rocks (Tismana granite), unconformably overlain by Mesozoic sedimentary rocks (Upper Jurassic reef limestones and Lower Cretaceous siliciclastites), outcrop on two-thirds of the basin’s surface. On the southern edge, these are covered by sub-Carpathian molasse deposits (Upper Sarmatian to Pontian) of the Dacian Basin. Isolated patches of Pleistocene eluvium and fluvial terrace deposits are found (Figure 1).

The distribution of soil types within the two catchments and the sampling sites is shown in Figure 2A. Cambisols are prevalent (brown eu-mesobasic, acidic brown soils, and lithosols). Rendzinas, lithosols, and rare relict patches of terra rosa developed on the limestone bedrock. Protisols (aluvisols, gleysols, and lithosols) developed on the fluvial deposits along the rivers and on the barren grounds, respectively.

Figure 2. (A). Distribution patterns of primary soil types in the Tismana (a) and Isverna (b) areas, compiled from the soil map of Romania, scale 1:200,000 [51]. (B). Land use in Tismana (TSM) and Isverna (IVR) areas (Corine Land Cover 2018).

2.2. Sampling and Analysis

The soil samples from Tismana (TSM, 18 samples) and Isverna (IVR, 20 samples) rural areas were collected from 0.3 to 0.5 m below the surface with a stainless-steel spatula, in April and May 2021. Each sample was prepared as a composite of five sub-samples collected from a 50 m radius. The sampling points were located by a hand-held global positioning system (GPS). Samples were air-dried, crushed, and sieved through a stainless-steel sieve with a 0.5 mm mesh size. Amounts of 0.25 g of each soil sample were digested with a mixture of aqua regia (HNO₃ 65%: HCl 37% (v/v) = 1:3) in a Speedwave XPERT microwave digestion system (Berghof, Eningen, Germany). The obtained solutions were filtered and then diluted with ultrapure water obtained from a PureLab system (Veolia Environnement, Toulouse, France) to 50 mL. The major element contents (Ca, Mg, Na, K, Al, P and Fe) was measured by Optima 5300 DV inductively coupled plasma atomic emission spectrometer (ICP-OES, Perkin Elmer, Waltham, MA, USA) whereas the PTEs contents (Ba, Sr, Mn, Cr, Zn, Pb, Ni, Cu, Cd, Co, and As) by ELAN DRC II inductively coupled plasma mass spectrometer (ICP-MS, Perkin Elmer, Waltham, MA, USA). Multielement ICP Standard 3 containing 1 mgL⁻¹ of Fe, Na, Mg, K, Ca, Ni, Cr, Cu, Zn, Cd, Pb, Mn, Al, and As (Perkin Elmer Pure Plus, Shelton, CT, USA) was used as calibration standards, while Trace Metals—Sandy Loam 8, CRM 025-050 (Merck, Darmstadt, Germany) was used for accuracy check. The calibration was linear, with a correlation coefficient (R²) greater than 0.9997 and a mean recovery between 96% and 104%.

The pH and electrical conductivity (EC) of the soil samples were measured in a mixture of 1:5 soil-to-water ratio (m/v) with a SevenExcellence pH meter S400 (Mettler Toledo, Columbus, OH, USA). The anions (Cl⁻, NO₃⁻, NO₂⁻, SO₄²⁻, F⁻) concentrations were measured using a 761 Compact ion chromatography (Metrohm, Herisau, Switzerland) after filtering the soil: water slurry through a 0.45 µm cellulose acetate membrane filter. Standard solutions containing 1000 mgL⁻¹ Cl⁻, 1000 mgL⁻¹ NO₃⁻, 1000 mgL⁻¹ PO₄³⁻, 1000 mgL⁻¹ F⁻, and 1000 mgL⁻¹ SO₄²⁻ (Certipur, Merck, Darmstadt, Germany) and nitrite standard solution (1000 mgL⁻¹ NO₂⁻, Certipur, Merck, Darmstadt, Germany) were used for the calibration of the ion chromatograph, while the accuracy was checked by analyzing SPS-NUTR WW1 Batch 115 wastewater reference materials (Spectrapure Standards, Oslo, Norway).

2.3. Contamination and Environmental Risk Indices

Contamination indices are divided into two main classes: individual indices that assess each PTE separately, and complex indices that describe contamination across several PTEs [15]. The extent of soil pollution with PTEs and the influence of land use on the pollution extent were assessed using the contamination factor (CF), geo-accumulation index (Igeo), contamination degree (CD), and pollution load index (PLI). In contrast, the environmental impact was assessed using the ecological risk index (ERI) and potential ecological risk index (PERI).

The contamination factor (CF, Equation (1)) is expressed as the ratio of the element content in soil to the preindustrial content of each element [52]. Based on the CF value, the contamination status of individual metals is classified in four classes (Table 1) [53].

Geo-accumulation index (Igeo, Equation (2)) indicates the intensity of the anthropogenic PTEs deposition in surface soil [54,55]. For the current research, the element content in the Earth’s crust was used as a reference background value [56].

$$\mathrm{C}\mathrm{F}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{E}}}{{\mathrm{C}}_{\mathrm{E}\mathrm{G}\mathrm{b}}}}$$

(1)

$$\mathrm{I}\mathrm{g}\mathrm{e}\mathrm{o}=\ln {\displaystyle \raisebox{1ex}{${\mathrm{C}}_{\mathrm{E}}$}\!\left/ \!\raisebox{-1ex}{$1.5\times {\mathrm{C}}_{\mathrm{E}\mathrm{G}\mathrm{b}}$}\right.}$$

(2)

where C_E represents the content of a given element in the analyzed soil, and C_EGb are the geochemical background value 7.2 mgkg⁻¹ for As, 100 mgkg⁻¹ for Cr, 3 mgkg⁻¹ for Cd, 100 mgkg⁻¹ for Cu, 550 mgkg⁻¹ for Mn, 50 mgkg⁻¹ for Ni, 19 mgkg⁻¹ for Pb, 60 mgkg⁻¹ for Pb, 11 mgkg⁻¹ for Sr, 26,000 mgkg⁻¹ for Fe, respectively [54]. Igeo classifies soil contamination into seven categories (Table 1) [55].

Contamination degree (CD, Equation (3)) quantifies the cumulative pollution of all elements [52] and classifies the soils into four classes, as shown in Table 1 [57].

Pollution load index (PLI, Equation (4)) indicates the contribution of each element to environmental pollution [58]. PLI classifies the pollution status of soils in five categories (Table 1) [59].

$$\mathrm{C}\mathrm{D}=\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{C}\mathrm{F}}_{\mathrm{i}},$$

(3)

$$\mathrm{P}\mathrm{L}\mathrm{I}=\sqrt[\mathrm{n}]{{\mathrm{C}\mathrm{F}}_1\times {\mathrm{C}\mathrm{F}}_2\times {\mathrm{C}\mathrm{F}}_3\times \dots .{\times \mathrm{C}\mathrm{F}}_{\mathrm{n}}}$$

(4)

where CF is the contamination factor of the individual element i, and n is the number of elements taken into consideration for complex indices computing.

Table 1. Pollution status as indicated by pollution indices.

Index	Value of Index	Class of Soil Contamination	Reference
CF	CF ≤ 1	Low contamination	[53]
	1 < CF ≤ 3	Moderate contamination
	3 < CF≤ 5	High contamination
	CF > 5	Extremely high contamination
Igeo	Igeo ≤ 0	Unpolluted	[55]
	0 < Igeo ≤ 1	Unpolluted to moderately polluted
	1 < Igeo ≤ 2	Moderately polluted
	2 < Igeo ≤ 3	Moderately to heavily polluted
	3 < Igeo ≤ 4	Heavily polluted
	4 < Igeo ≤ 5	Heavily to extremely polluted
	Igeo > 5	Extremely polluted
CD	CD < 6	Low contamination degree	[57]
	6 ≤ CD < 12	Moderate contamination
	12 ≤ CD < 24	Considerable contamination degree
	CD ≥ 24	Very high contamination degree
PLI	PLI < 0.7	Unpolluted	[60]
	0.7 < PLI < 1	Slightly polluted
	1 < PLI < 2	Moderately polluted
	2 < PLI < 3	Severely polluted
	PLI > 3	Heavily polluted
PERI	PERI < 40	Low risk	[61]
	40 ≤ PERI < 80	Moderate risk
	80 ≤ PERI < 160	Considerable risk
	160 ≤ PERI < 320	High risk
	PERI ≥ 320	Very high risk

Table 1. Pollution status as indicated by pollution indices.

Index	Value of Index	Class of Soil Contamination	Reference
CF	CF ≤ 1	Low contamination	[53]
	1 < CF ≤ 3	Moderate contamination
	3 < CF≤ 5	High contamination
	CF > 5	Extremely high contamination
Igeo	Igeo ≤ 0	Unpolluted	[55]
	0 < Igeo ≤ 1	Unpolluted to moderately polluted
	1 < Igeo ≤ 2	Moderately polluted
	2 < Igeo ≤ 3	Moderately to heavily polluted
	3 < Igeo ≤ 4	Heavily polluted
	4 < Igeo ≤ 5	Heavily to extremely polluted
	Igeo > 5	Extremely polluted
CD	CD < 6	Low contamination degree	[57]
	6 ≤ CD < 12	Moderate contamination
	12 ≤ CD < 24	Considerable contamination degree
	CD ≥ 24	Very high contamination degree
PLI	PLI < 0.7	Unpolluted	[60]
	0.7 < PLI < 1	Slightly polluted
	1 < PLI < 2	Moderately polluted
	2 < PLI < 3	Severely polluted
	PLI > 3	Heavily polluted
PERI	PERI < 40	Low risk	[61]
	40 ≤ PERI < 80	Moderate risk
	80 ≤ PERI < 160	Considerable risk
	160 ≤ PERI < 320	High risk
	PERI ≥ 320	Very high risk

The Potential Ecological Risk Index (PERI, Equation (6)) indicates the environmental response to metal toxicity [52,62].

$${\mathrm{E}}_{\mathrm{r}}={\mathrm{T}}_{\mathrm{n}}^{\mathrm{i}}\times {\displaystyle \frac{{\mathrm{C}}_{\mathrm{P}\mathrm{T}\mathrm{E}}}{{\mathrm{C}}_{\mathrm{P}\mathrm{T}\mathrm{E}\mathrm{b}\mathrm{k}}}},$$

(5)

$$\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{I}={\sum }_{\mathrm{i}}^{\mathrm{n}}{\mathrm{E}}_{\mathrm{r}}^{\mathrm{i}},$$

(6)

where T is the eco-toxicity response coefficient of the ith element, and E_r is the computed value for the potential ecological risk index of the element i. The PERI classifies the potential ecological risk of soils into five categories (Table 1) [62].

2.4. Data Analysis and Distribution Patterns

To evaluate the distribution pattern of PTEs in soils and to identify potential influences, spatial autocorrelation analysis was employed (Moran’s Index). The datasets were further processed using ArcGIS 10.1 Spatial Analyst tools, applying both Inverse Distance Weighting (IDW) and Empirical Bayesian Kriging (EBK) interpolation methods. IDW was used as a reliable method to determine specific areas that were possibly affected by human activities, while EBK was used to identify different trends in the analyzed data and to validate IDW results [24,27,63,64]. IDW interpolation is commonly employed to estimate PTE concentrations with good accuracy, determining weight according to the distance impact, while EBK (kriging interpolation) relies on the spatial distribution of PTE contents by assigning weights to sampling points based on their influence on the predicted values [64].

3. Results and Discussions

3.1. Content of PTEs and Anions in Soil

The content of PTEs and anions in soils from Tismana (TSM) and Isverna (IVR) areas is presented in Table S1 and Figure 3, while the basic statistical parameters are summarized in Table S2. The soils from TSM and IVR were slightly acidic, with pH values of 4.96 and 4.99, respectively. Such acidity is characteristic of soils formed under humid temperate conditions [65]. Slightly acidic conditions can influence nutrient dynamics and the mobility of metals, generally increasing the solubility of cationic elements such as Cd, Zn, and Ni, while reducing the availability of others, such as K and Mo [66,67]. The narrow pH range observed suggests similar pedogenetic conditions across the two sites, though even low pH variations can affect the speciation and bioavailability of both nutrients and PTEs [67]. One of the most important indicators of soil quality is its salinity. Electrical conductivity (EC) is commonly used to determine soil salinity [68]. With values between 8.20 and 202 μScm⁻¹ in the case of TSM and between 12.8 and 167 μScm⁻¹ in the case of IVR, the mean value of EC was 55.1 and 84.8 μScm⁻¹ for TSM and ISV, respectively. Based on the soil salinity classification developed by Richards [69], all soils were classified as non-saline, except for TSM9, a slightly saline soil. Such low conductivity values are typical of well-drained, humid-climate soils, where leaching effectively removes soluble salts from the root zone [70]. The slightly elevated EC observed in sample TSM9 suggests localized salt accumulation, possibly related to restricted drainage or minor anthropogenic inputs. A primary concern with saline soils is the excessive accumulation of soluble salts, primarily chloride (Cl⁻), sulfate (SO₄²⁻), and occasionally nitrate (NO₃⁻) [69]. The presence of a specific anion plays an important role in determining soil type and salinity degree. For example, a predominance of Cl⁻ indicates chloride-dominated salinity, while a high content of sulfate is indicative of sulfate-dominated salinity [71].

Figure 3. The mean content of PTEs in soils in Tismana (TSM-blue) and Isverna (ISV-red) areas.

NO₃⁻ was the most abundant anion detected; its content presented high variation among the studied areas and sampling locations. In the TSM area, NO₃⁻ content ranged from LOD (<0.5 mgkg⁻¹) to 145 mgkg⁻¹, while in the IVR, NO₃⁻ ranged between 2.50 and 275 mgkg⁻¹. The mean NO₃⁻ content was higher in TSM than in IVR, while the standard deviation value (87.9) indicated a considerable variation among sampling sites.

SO₄²⁻ had the lowest contents among the major anions, with mean values of 9.46 and 9.07 mgkg⁻¹ in TSM and IVR, respectively, followed by Cl⁻, with mean values of 6.86 mgkg⁻¹ in TSM and 5.84 mgkg⁻¹ in IVR. The standard deviations for Cl⁻ were 7.72 and 2.20 in TSM and IVR, respectively, reflecting spatial heterogeneity. The anion content also supported the classification of TSM soils as slightly saline. Zhang et al. [72] reported a connection between the low salinity soils and the soil microbiological biomass nitrogen, which was significantly higher than in the case of the saline soils.

Fluoride (F⁻) and nitrite (NO₂⁻) were generally lower than the LOD of the methods (0.5 mgkg⁻¹) or only slightly above it in the case of fluoride, which ranged between 0.60 and 1.20 mgkg⁻¹.

Overall, the mean contents of the analyzed PTEs were comparable between the two study areas but varied widely among sampling sites. In the TSM area, the PTE contents range from 0.506 mg kg⁻¹ for Cd, 4.66 mg kg⁻¹ for As, and 13.5 mg kg⁻¹ for Co, to much higher values of 71.1 mg kg⁻¹ for Zn and 447 mg kg⁻¹ for Mn. Similarly, in the IVR area, PTE contents ranged from 0.440 mg kg⁻¹ for Cd, 5.95 mg kg⁻¹ for As, and 15.2 mg kg⁻¹ for Co, up to 75.1 mg kg⁻¹ for Zn, 75.7 mg kg⁻¹ for Ni, and 480 mg kg⁻¹ for Mn. The slightly higher contents of some PTEs in IVR, compared with TSM, may reflect local lithological characteristics, differences in depositional processes, or anthropogenic influences across the sites [73]. The overall variability highlights the complex interaction of natural and anthropogenic factors governing the PTE distribution in the study area [74]. Median values for most elements were close to or slightly below their respective means, and skewness values below 1 for all the elements except Sr. The kurtosis values varied from −1.36 for Mn and 1.22 for Zn. The low skewness and kurtosis values suggest a near-normal distribution of PTEs content, indicative of minimal anthropogenic influence at most sampling sites. Standard deviations and coefficients of variation (CVs) were relatively low for the majority of PTEs, except for Mn and Ba. However, the absence of locally established geochemical background values of the elements makes it difficult to distinguish between the “natural” and “contaminated” environments.

According to the Romanian legislative standards for sensitive soils [75], none of the samples from the TSM area exceeded the screening or vigilance threshold values. On the other hand, isolated exceedances were observed in the IVR area: Pb and Co exceeded the vigilance threshold in one sample (IVR13 for Pb and IVR15 for Co), while the Ni content from three samples exceeded the alert threshold (IVR4, IVR11, IVR13) and three samples, the intervention threshold (IVR5, IVR14, IVR15).

Co in soils primarily originates from the underlying parent rock, with a global mean concentration in surface soils of approximately 10 mg kg⁻¹ [76]. Elevated Co levels are typically observed in heavy clay soils, such as Cambisols, and occasionally in organic-rich soils. Similarly, Pb is most abundant in Cambisols and Histosols, with a global average content of about 20 mg kg⁻¹ [76]. While both Co and Pb can also derive from anthropogenic sources—including emissions from metallurgical processes, cement production, and power plants—there are no records of such activities in the study area in recent times [76]. Ni is an essential trace element for both human and animal health, playing a vital role in the formation of red blood cells; however, excessive content can be toxic. Plants readily absorb Ni and can therefore accumulate it in crops; thus, they are very susceptible to Ni input into the food chain, with very harmful effects on human health [76].

When compared with the intervention limit established by the Netherlands soil standards, the mean value of the analyzed PTEs does not exceed the ecotoxicological risk content in soils [77,78].

Ca, Mg, K, Al, and Fe were the most abundant elements in this study, based on their mean and maximum values. The median values of macroelements in soils (Ca, Mg, Na, K, Al, P, and Fe) were, generally, close to or lower than their corresponding means, suggesting a relatively symmetric distribution for most elements. In the TSM area, Ca, Mg, Na, and Fe exhibited positive kurtosis values, indicating a more peaked distribution, whereas K, Al, and P exhibited negative kurtosis values. On the other hand, in the IVR area, Na, K, and Fe had positive values of the kurtosis, whereas Ca, Mg, Al, and P displayed negative kurtosis, suggesting a broader spatial spread. Across both areas, macroelements had substantially higher CVs than most microelements, suggesting greater spatial variability in their distribution. The dominance of Ca and Mg reflects the pedogenetic processes typical for carbonate-rich parent material of karstic terrains. Elevated levels of these elements enhance soil buffering capacity and may mitigate the mobility of PTEs through cation-exchange and precipitation reactions [79]. K and Na contents, while generally lower, play essential roles in soil fertility and ionic balance. Their spatial variability, indicated by the higher coefficients of variation and kurtosis values, may be linked to land use patterns—particularly agricultural practices that could influence the distribution of soluble salts and exchangeable cations in soils [80].

Differences in kurtosis between the TSM and IVR areas indicate contrasting lithological backgrounds and soil-forming conditions. The sharper distributions of Ca and Mg in TSM suggest a more uniform parent material, while the broader spread in IVR likely reflects heterogeneous geology and stronger anthropogenic influence [81]. Despite spatial variability, all macroelement contents stayed below international soil quality standards, indicating limited enrichment or contamination from external sources. However, the observed spatial variability implies that even naturally derived elements can show marked heterogeneity in karstic environments [82].

When compared with the soil quality standards, none of the samples surpassed both Dutch soil standards [77,78]. Generally, the median values of the PTEs content were higher in the IVR than in the TSM area.

3.2. PTE Distribution Pattern

The human activities in both areas are primarily focused on grazing, agriculture, and timber harvesting (Figure 4b). The main difference between the two areas lies in land use allocation: in IVR, over 24% of the land use is allocated to grazing, while in TSM, the pastures represent only 3.9% of the land use (Table S3). While the IVR spans an area nearly twice that of TSM (36.5 vs. 17.8 km²), the forest cover is notably lower in IVR, at 68.7%, while in TSM it is over 82%. Another important difference is the degree of forest fragmentation. IVR presents a heterogeneous landscape with both pasture and limestone outcroppings, while in TSM, the forest is mainly continuous and homogeneous. Agricultural activities represent only ~5% (TSM) and 3% (IVR). In TSM, agrarian activity is limited to the southern part of the upper basin due to predominant forest cover, while in the fragmented terrain with limestone outcrops, it is further constrained (Figure 4b).

Figure 4. Spatial distribution pattern of PTEs in soil in (a) Tismana (TSM) and (b) Isverna (IVR) area, generated using Inverse Distance Weighting (IDW) interpolation.

The primary anthropic activity in both studied areas is agriculture; however, small-scale logging activities are also present in IVR. The interpolation methods applied (both IDW in Figure 4 and EBK in Figures S1 and S2) revealed distinct patterns of PTEs occurrence in both studied areas, being linked to a combined influence of natural variation and human activities. The differences in the areas studied are related to geology, climate, and human activities. While the TSM area is characterized by a relatively simple geological framework, a temperate climate, and homogeneous vegetation, the IVR area has a more complex geology (outcropping limestones), a sub-Mediterranean climate (prolonged droughts), and a fragmented vegetation cover (transition from temperate to sub-Mediterranean species). These natural contrasts, along with anthropogenic activities, enhanced environmental pressure on soil formation and increased the area’s susceptibility to contamination. Consequently, IVR appears to be more vulnerable to pollution (soils, groundwater, etc.) than TSM, due to the interplay of complex natural factors and anthropogenic stressors.

To further explore the spatial distribution and clustering of PTEs in the studied soils, spatial autocorrelation analysis was conducted using Moran’s Index. While interpolation methods (IDW and EBK) provided visual insights into patterns of PTE occurrence, spatial statistics allow a quantitative assessment of whether these elements exhibit significant spatial clustering or are randomly distributed [83]. Moran’s I statistic was used to assess spatial autocorrelation in soil pollution data, allowing us to determine whether high or low pollution values exhibit spatial clustering or randomness within the study area. Accordingly, this tool was applied to both investigated sites. Tismana soils exhibit stronger, more significant autocorrelation of PTEs than Isverna soils (Table 2). Most PTEs in Tismana have significant positive Moran’s I values and low p-values, while Isverna shows weak or random patterns. For Tismana, most elements (Cr, Mn, Zn, Cu, Ni, Cd, Co) show significant positive spatial autocorrelation (z-scores > 1.96, p-values < 0.05), indicating clustered distributions. Moran’s index values are positive and relatively high (e.g., Ni = 0.43, Cd = 0.52), suggesting spatial clustering. For Isverna, none of the elements reach statistical significance (p-values > 0.05, Z-scores close to 0), implying a random spatial distribution. Moreover, Moran’s Index values are nearly 0 or slightly negative (e.g., Zn and Pb), suggesting a weak dispersion. Lower variance across most indices indicates greater spatial homogeneity and less local clustering than in Tismana. The contrasting spatial patterns between the two regions likely reflect differences in their natural settings—such as geology, geomorphology, and hydrology—as well as distinct local anthropogenic influences, including grazing, agriculture, and timber harvesting.

Table 2. Distribution of PTEs Moran’s indices in soil from Tismana (TSM) and Isverna (ISV).

PTE	Expected Index	Variance	z-Score	p-Value	Global Moran’s Index
TSM
Cr	−0.058824	0.049662	2.127403	0.033387	0.415265
Mn	−0.058824	0.050280	1.916568	0.055293	0.370931
Zn	−0.058824	0.043878	1.994246	0.046125	0.358914
Pb	−0.058824	0.044973	0.180406	0.856834	−0.020565
Cu	−0.058824	0.047822	2.107200	0.035100	0.401984
Ni	−0.058824	0.049097	2.208871	0.027184	0.430617
Cd	−0.058824	0.049105	2.611585	0.009012	0.519893
Co	−0.058824	0.049675	2.202414	0.027636	0.432049
ISV
Cr	−0.052632	0.013801	0.701186	0.483187	0.029744
Mn	−0.052632	0.015256	−0.215960	0.829019	−0.079306
Zn	−0.052632	0.012976	−0.599969	0.548527	−0.120975
Pb	−0.052632	0.011992	−0.411227	0.680906	−0.097664
Cu	−0.052632	0.015352	0.538192	0.590444	0.014052
Ni	−0.052632	0.013579	0.367126	0.713525	−0.009851
Cd	−0.052632	0.014124	0.520639	0.602618	0.009244
Co	−0.052632	0.014561	0.034596	0.972402	−0.048457

The IDW interpolation identified areas (Figure 4) with variable PTE contents where the high PTEs could be associated with anthropogenic activities (deforestation, agriculture, etc.). In contrast, other PTEs may reflect natural geochemical variability (e.g., different rock types). Soil type also plays a critical role in susceptibility to contamination. Poorly developed soils (e.g., rendzinas and lithosols, mostly found in the IVR area, Table S4) are more susceptible to pollution; whereas more developed soils (e.g., cambisols) demonstrate greater resilience to both natural and anthropogenic impacts. Besides the anthropogenic impact on soil degradation, several natural processes—including surface runoff, landslides, and shifts in vegetation—and climatic patterns (e.g., extended droughts or intense rainfall events) can also alter the soil structure. Both natural and human-induced impacts result in the loss of organic matter and decline in soil biodiversity [84], thereby increasing the soil’s vulnerability to contamination and long-term degradation.

The PTEs distribution pattern differs from element to element and between the study areas. The distribution pattern of the PTEs across the various soil profiles is presented in Figure 4. The spatial distribution of PTEs in IVR shows variations and variability, with contents generally increasing across the area (Figure 4a). In the case of TSM, the highest variation was recorded for Mn, Zn, and Pb, with a coefficient of sample variation of 28,010 for Mn, 227 for Zn and 59 for Pb, while in the case of IVR, Mn, Ni, Cr, and Zn exhibited the highest variability, with coefficients of variation ranging from 42,298 for Mn to 272 for Zn. These patterns likely reflect a combination of heterogeneous parent material, localized anthropogenic inputs, and differential retention within soil horizons.

These findings demonstrate that even in predominantly rural landscapes with limited industrial influence, soil contamination by PTEs can occur, driven by a combination of natural soil properties and localized anthropogenic practices.

The content of Ni, Pb, and Co exceeded the regulatory thresholds established by both national [75] and international regulations [77,78]. Specifically, Ni contents exceeded the alert threshold in samples IVR4, IVR11, and IVR13, while in IVR5, IVR14, and IVR15, Ni contents surpassed the intervention thresholds defined by the Romanian Ministry of Waters, Forests, and Environmental Protection Order No. 756/1997. As observed in Figure 4b (ISV area), Ni contamination was predominantly in the NW and NE sectors of the study area, suggesting a natural dispersion and dilution of the content with increasing distance from the highly contaminated sites. The highest Ni concentrations observed in IVR are comparable to those reported in Shaanxi Province, Northwestern China, a rural area with historical mining activities [85], and to rural soils in Iaşi County, Romania, and Podlasie, Poland, where low-intensity agriculture and natural soil characteristics contributed to metal accumulation [86,87]. This indicates that even moderate anthropic activity can significantly influence soil quality under specific environmental conditions. While organic layers and amendments such as logging residues and wood ash can enhance metal immobilization within the O horizon [88], these natural mitigation processes are spatially heterogeneous and dependent on soil texture, organic matter content, and microbial activity, which may explain the uneven PTE distribution observed in IVR. Additionally, Pb and Co had exceeded the alert threshold in IVR13 for Pb, while Co levels were above the threshold in IVR15. Taken together, these findings indicate that both intrinsic soil properties and localized anthropogenic activities—despite the relatively low intensity of industrial influence—contribute to the observed PTE patterns. This underscores the need for targeted monitoring and risk-based soil management strategies in vulnerable rural areas.

In the absence of established geochemical background values for local soils, the median element contents in soils in TSM and IVR were compared with those reported from other areas in Romania, like Vaslui, Maramureș, Banat, and Iași counties (Table 3). At a national level, Maramureș County exhibited the highest elemental contents, with Pb levels up to 40-fold higher than the element content in TSM or IVR. Similarly, the Banat region showed high mean PTE values: Fe contents were approximately twice the median values recorded in TSM and IVR, while Zn and Mn were 3- and 4–5-fold higher, respectively, and Cu was 4–5-fold higher. In contrast, the contents of Ni and Pb were comparable to those in TSM, but lower than the mean content in the IVR area. On the other hand, in Vaslui County, the contents of Zn, Pb, Cd, and Co were comparable with those in TSM or IVR. However, the contents of Cr and As were higher than those recorded in the areas studied in the present research. Likewise, in Iași county, the contents of Zn, Pb, and Cu were also comparable with those in TSM or IVR. In Vaslui and Iași, anthropogenic activities—particularly vehicle emissions, human activities, and agricultural materials were the main sources of PTE accumulation in soils [39,42].

Table 3. Comparison of the PTEs content (mgkg⁻¹) in soils from various regions in Romania.

Region	Fe	Ba	Sr	Mn	Cr	Zn	Pb	Ni	Cu	Cd	Co	As
TSM (present study)	27,695	88.3	12.7	414	11.8	69.4	25.7	11.5	14.7	0.48	10.2	3.53
IVR (present study)	29,207	82.2	9.05	512	38.3	74.8	21.3	43.6	26.5	0.22	15.3	4.39
Vaslui county [39]	-	-	-	-	62.1	68.2	25.3	47.4	29.8	0.32	10.8	10.2
Maramures county [40]	-	-	-	-	-	377	318	44.5	111	1.10	-	29.5
Banat region [41]	44,853	-	-	2118	-	197	22.0	14.6	47.8	-	-	-
Iași country [42]	-	-	-	-	-	67.2	18.6	-	28.3	-	-	-
Screening value set by Romanian [75] legislation	-	200	-	900	100	100	20	20	20	1	15	5
Alert threshold for sensitive use set by Romanian [75] legislation	-	400	-	1500	300	300	50	75	100	3	30	15
Intervention value established by the Dutch [77,78] standards	-	625	-	-	380	720	530	210	190	12	240	55

Despite the relatively low intensity of anthropogenic activities, their impact on soil contamination is more pronounced in IVR than in TSM. The difference between the two may be linked to specific regional vulnerabilities, in which even small human actions, combined with natural environmental factors, may result in long-lasting soil degradation and loss of soil functions. Similar results have been reported in other rural-dominant landscapes: for instance, Bartkowiak et al. [89] found heavy-metal accumulation in salt-affected soils within a Natura 2000 area in north-central Poland despite limited industrialization. Further, Wiater [87] documented measurable heavy-metal burdens in organic grassland soils in rural Podlasie, Poland. Bojanowski [90] used fingerprinting in an Eastern Polish agricultural catchment to show that even non-industrial landscapes are subject to multiple contamination pathways.

3.3. Soil Pollution

3.3.1. Contamination Factor (CF)

At the TSM sites, the CFs for the analyzed elements ranged from 0.053 (Cd) to 2.77 (Sr), whereas at the IVR sites, CFs spanned between 0.022 (Cd) and 6.00 (Ni) (Table S5). Among the assessed contaminants, Ni, Pb, and Sr contributed the most to overall contamination. The spatial distribution pattern indicated that the highest CFs were recorded for Pb and Sr at TSM, and for Ni and Pb at IVR. Conversely, Cu and Cd consistently showed the lowest CF values at TSM and IVR (Figure 5). Overall, soils from IVR exhibited higher CF values compared to those from TSM. The mean CFs followed the ascending order: Ni < Pb < Zn < Fe < Sr < Mn < As < Cr < Cu < Cd in the IVR area and Pb < Sr < Zn < Fe < Mn < As < Ni < Cr < Cd < Cu in the TSM area. Although Sr and Fe displayed moderate CFs, their limited spatial variation suggests their lithogenic origin, likely derived from rock weathering. This is consistent with previous studies reporting elevated Fe contents in soils [91,92]. Although Fe is naturally abundant in some soils, elevated contents may also result from accelerated weathering under anthropogenic pressures, the application of iron-containing fertilizers, or nearby industrial activities [93]. The enrichment of other PTEs through weathering of rock material has also been reported in the recent literature [94]. In general, the CF values indicated that the soils were low and moderately contaminated with the studied elements, except for Ni and Pb. Notably, at the IVR, two sampling sites (IVR5, IVR14) were categorized as highly contaminated with Ni, and one site (IVR15) as extremely contaminated with Ni. Additionally, IVR13 was found to be highly contaminated with Pb. All these sites are located near human settlements, suggesting potential anthropogenic sources contributing to the elevated contents.

Figure 5. Percentage of soil samples with different contamination factors (CFs).

3.3.2. Geo-Accumulation Index (Igeo)

The Igeo values for all samples were predominantly negative, indicating that the analyzed soils are practically unpolluted (Table S5). In the TSM area, positive Igeo values between 0 and 1 were observed only for Sr, Pb, Fe, and Zn, indicating an unpolluted to moderately polluted status. The highest Igeo values for Zn, Sr, and Fe were observed at sampling site TSM5, whereas Pb reached its peak Igeo value at site TSM11, suggesting localized enrichment, likely influenced by site-specific geochemical conditions or anthropogenic inputs [95]. As with CFs in the IVR area, Igeo indicated moderate pollution at sites IVR13 and IVR15 for Ni and Pb, respectively. These findings highlight the presence of localized pollution hot spots, particularly near human settlements or areas of intensified land use. Compared with regions with a similar history of agricultural activity, the study areas exhibited lower levels of metal accumulation. For instance, in the Thriassio Plain, Greece, Igeo values for Ni, Pb, Cu, and Zn indicated moderate to heavy soil pollution with these elements [96]. Similarly, Barbieri et al. [97] reported positive Igeo values for Ni, Pb, and Cu in surface soils across various locations in the Lazio region, reflecting moderate to heavy pollution levels for those metals. As in the present study, the spatial distribution of PTEs revealed localized enrichment, underscoring the importance of establishing site-specific geochemical background values. This is essential for distinguishing between naturally occurring element contents and those resulting from anthropogenic inputs [97].

3.3.3. Contamination Degree (CD)

The contamination degree (CD), defined as the cumulative sum of contamination factors (CFs) for multiple PTEs, provides a comprehensive measure of the overall contamination level in surface soils. It serves as an indicator of the overall level of contamination in the surface soil layer at a given sampling site or core [98]. In the study areas, the CD indicated three pollution classes within the TSM and ISV areas (Figure 6). Mean CD values of 8.86 for IVR and 7.48 for TSM classify both regions as moderately polluted (Table S5). The highest CD indicated a considerable contamination in soil sample IVR15, mostly due to the high Ni content. Ni was reported to increase its mobility and bioavailability under acidic and neutral conditions [76], a finding confirmed in our research across all locations with high Ni values. CD confirmed the results obtained by using the Igeo and CF indices, reinforcing the reliability of the contamination assessment and highlighting Ni and other PTEs as key contributors to soil pollution in IVR.

Figure 6. The contamination degree (CD) of the analyzed soil samples.

3.3.4. Pollution Load Index (PLI)

The PLI ranged from 0.28 to 1.05 in IVR and from 0.33 to 0.87 in TSM, indicating low to moderate pollution, as shown in Figure 7. The highest PLI values were recorded at site IVR15, primarily driven by elevated Ni content, followed by sites IVR19 and IVR13, where elevated values were attributed to the accumulation effect of multiple elements: Zn, Pb, Mn, and As at IVR19 and Pb, Zn, and Mn at IVR13.

Figure 7. Pollution load index (PLI) in the soil of Isverna (ISV) and Tismana (TSM) area.

The outcomes of the various pollution assessment methods applied in this study revealed no substantial differences between the different metal contamination indicators assessed. All indices—namely the contamination factor (CF), pollution load index (PLI), and geoaccumulation index (Igeo)—consistently pointed to the accumulation of Sr, Pb, Fe, and Zn in the TSM area, as well as slightly elevated levels of Ni and Pb in the IVR area. Among these, the Igeo method is particularly valuable, as it integrates both natural geogenic factors and anthropogenic influences, offering a more intuitive interpretation of heavy metal enrichment in soils [99].

3.4. Potential Ecological Risk Index

PERI is a very effective tool for evaluating the ecological risks associated with PTEs [52,100]. The obtained results highlighted low and moderate risk for the IVR area (Table S6, Figure 8). The moderate ecological risk identified in seven sites (IVR1, IVR5, IVR11, IVR13, IVR15, IVR19, and IVR20) was primarily due to slightly high contents of As, Cd, Pb, and Ni. The ecological risk (Er) ranged from 4.68 to 17.8 for As, from 1.13 to 17.64 for Cd, from 4.15 to 19.8 for Pb, and from 22.9 to 30.0 for Ni. The highest PERI value was determined for IVR15, followed by IVR13 and IVR19. In the case of the TSM area, all samples fell within the low-risk category, with the highest value of PERI being observed for TSM18 (40.3), attributed to moderately increased contents of As, Cd, and Pb, with individual risk factors (Er) of 11.2 for As, 9.25 for Cd, and 10.2 for Pb.

Figure 8. Potential Ecological Risk Index (PERI) value of the soil samples from Isverna (ISV) and Tismana (TSM).

Overall, As, Cd, and Pb were the main contributors to the potential ecological risk in both areas. Several factors, including land use, agricultural practices, and the geological background of the areas, influenced the spatial distribution of PERI values. Several studies have reported moderate ecological risks associated with Cd and Pb in some areas [101,102,103]. Zaakour et al. [101] stated that a possible moderate risk of contamination with Cd and Pb may be related to the application of pesticides (herbicides, insecticides, and fungicides) in agricultural treatment and the utilization of phosphate-based fertilizers with high levels of Cd.

4. Conclusions

The spatial distribution of potentially toxic elements (PTEs) in soil was evaluated in relation to land use, soil type, and environmental vulnerability using pollution indices and distribution maps. Moran’s Index and IDW analyses revealed distinct spatial patterns across the study areas, with Tismana displaying moderate to strong clustering for several PTEs, whereas Isverna exhibited a contrasting random distribution. Overall, PTE contents in both study areas—Tismana and Isverna—were generally low, indicating low anthropogenic impact across most sampling sites. In the Tismana area, all values were below established soil quality thresholds. However, in Isverna, elevated contents of Ni, Pb, and Co are isolated, suggesting localized pollution. The lack of site-specific geochemical background values makes the distinction between natural and anthropogenic sources difficult. Nonetheless, the greater spatial variability of macroelements and slightly higher median PTE contents in Isverna point to increased environmental sensitivity in this area. The spatial distribution patterns of PTEs in the Isverna area reflect both natural and human-induced influences, with its fragmented landscape, complex geology, and weakly developed soils contributing to its increased susceptibility to pollution. Contamination factor and geo-accumulation index indicated low to moderate contamination across both areas. The highest contents of Ni and Pb were found in Isverna, mainly near populated areas, while Zn, Sr, Fe, and Pb were higher in the Tismana area. The localized enrichment of Ni and Pb in Isverna likely originates from anthropogenic inputs, whereas the more uniform spatial distribution of Sr and Fe suggests natural geogenic sources.