Risk assessment and source analysis of heavy metal contamination in the soil of the Juanshui River Mouth

ABSTRACT

Heavy metals with a high concentration value in the soil are detrimental to the soil ecosystem. In this study, 96 samples were taken from two soil sections (A-A, B-B) in the mouth of the Juanshui River to analyze heavy metal pollution. Soil pollution status and the ecological risk were investigated using Nemerow and the potential ecological risk index. The sources affecting pollution and pollution distribution characteristics were exhibited based on Inverse Distance Weighted (IDW), principal component analysis (PCA), and Statistics. Pollution in the soil of the mouth of the Juanshui River had reached a moderate level. Hg and Cd were the main potential ecological risk contributions. Coal burning and agricultural chemicals were the predominant pollution source in the study area. In order to avoid the soil of the river mouth becoming a secondary pollution source for water bodies and resulting in a vicious circle of heavy metal pollution, heavy metals in the soil of the river mouth of China should be mainly managed and controlled.

KEYWORDS:

• Heavy metal
• Juanshui river
• soil pollution
• risk assessment
• source analysis

1. Introduction

Heavy metal contamination in soils mainly comes from the parent soil material and human activities [1,2]. Naturally, heavy metals are released by geological processes such as the weathering and erosion of mineralized deposits as well as bedrock. Anthropogenically, heavy metals interfere with soil properties through the release of industrial wastewater, traffic discharge, agricultural production, mining, electroplating, and smelting [3–5]. Among heavy metals, most of them are essential for maintaining human health, such as Fe, Cu, and Zn. But essential elements are equally dangerous in excess [6]. Moreover, a few heavy metals (i.e. Pb, Cr, Cd, As, and Ni) are harmful in trace amounts [7]. The continuous degradation of soil quality and potential effects on human health are caused by toxicity, poor degradation ability, and high mobility of heavy metals in the soil [8–10]. Due to the consumption of crops rich in harmful heavy metals and long-term exposure to heavy metals in soil particles, the eight systems of the human body are difficult to coordinate, leading to pathological changes in the organs [11]. Finally, humans are at risk of disease and even death [12].

Due to rapid economization and urbanization [13–16], the scope and frequency of human activities are expanding. Heavy metals are exacerbating soil contamination at an alarming rate through multiple pathways. According to statistics from the latest national survey of soil contamination in 2014, more than 20 million hectares of land were contaminated to varying degrees in China. Compared with northern China, soil pollution in southern China was even less optimistic [17,18]. In addition, China Ecological Environment Status Bulletin 2021 showed heavy metals were the first pollutants that cause serious damage to soil environmental quality [19]. Soil serves as a sink of pollutants and can also serve as a source of pollution for other mediums in the biosphere. For example, when soil is subjected to long-term erosion by the river, certain biochemical reactions occur, and then heavy metals migrate and transform between the river and the soil, resulting in secondary pollution. In this situation, heavy metals caused great damage to both terrestrial and aquatic ecosystems simultaneously. Therefore, it is of great importance to determine and monitor heavy metals in soil to protect the ecosystem.

Determining the pollution degree, potential ecological risk and pollution sources of heavy metals in soil are primary prerequisites for effective pollution prevention and control. Currently, numerous indices have been applied for the research related to heavy metal pollution in soil Geoaccumulation index (Igeo), Nemerow index, contamination factor (CF), pollution load index (PLI), and enrichment factor (EF) have been widely used to evaluate the pollution degree of heavy metals [20–22]. Potential ecological risk index (RI) and ecological risk factor (Ei r) have been developed to evaluate associated ecological risks [23]. The positive matrix factorization (PMF) model and some multivariate statistical methods such as PCA, factor analysis, correlation analysis, and cluster analysis are also commonly used to identify pollution sources of heavy metals [24–26].

Juanshui River provides convenience for the coastal basin in industry and agriculture, supporting the economic development of Xiangtan County, and is the main drinking water source of Xiangtan County. The mouth of the Juanshui River is the intersection of the Xiangjiang River and the Juanshui River. The soil in the mouth of the Juanshui River has been eroded by the Xiangjiang River and the Juanshui River for a long time. Heavy metals are released from the soil into the surrounding rivers. The ecological security of the Juanshui River and the health of coastal residents will be difficult to be ensured if the non-point source pollution caused by the soil is ignored. At present, the research about soil heavy metal pollution only focuses on agricultural and industrial areas, but the pollution of heavy metals in the soil of the river mouth has never been investigated. In order to fill this gap, this study investigates the levels of heavy metals, PH, and OM in the soil of the mouth of the Juanshui River. The objectives of the present study were to (1) analyze the content characteristics of heavy metals, PH, and OM in two soil sections of the mouth of the Juanshui River; (2) evaluate the contamination status using the Nemerow index; (3) assess the ecological risk level in the soil of the mouth of the Juanshui River using potential ecological risk index method; (4) define significant contribution source as well as secondary contribution sources of soil pollution heavy metals using PCA. The research findings play a fundamental role in further controlling and repairing contaminated soils and the environment.

2. Samples and methods

2.1 Study area

Juanshui River is a first-class tributary of the Xiangjiang River located in the north of Mount Heng. It originates from Shuangfeng County, Loudi City, flows through Hengshan County, and finally merges into the Xiangjiang River at Xiangtan County, Xiangtan City, with a total length of 118.5 km and a drainage area of 1764 km². The study area is located in the mouth of the Juanshui River (Figure 1), which is the intersection of the Juanshui River and the Xiangjiang River in Xiangtan County. It belongs to a humid subtropical monsoon climate and has the characteristics of flat terrain, plentiful rainfall, and abundant resources [27]. In addition, the soil types include red loam, downy loam, and clay. Industrial production activities in the study area are frequent due to four large-scale industries of new construction material, PVC sections, agricultural products processing, and electromechanical manufacturing. In terms of the ecological environment in Xiangtan County, research showed that Cd pollution was serious in part of the soil, and different degrees of enrichment for Zn, Cd, and As were detected in agricultural products [28].

Figure 1. Map of the study area.

2.2 Sample collection and preparation

Under certain chemical conditions, heavy metals will migrate and transform between the river and the soil. Taking this into consideration, it is possible that the Xiangjiang River is one of the pollution sources in the study area. This study selected the mouth of the Juanshui River at the intersection of the Xiangjiang River and the Juanshui River as the research area. In order to compare the contents of heavy metals, we set two sections according to the standard of the distance from the Xiangjiang River. Section A-A closer to the Xiangjiang River and section B-B further away from the Xiangjiang River were set. Each section was composed of 6 sampling points. A total of 12 sampling sites were in the study area and distributed in the floodplain and cultivated land. The sampling points were symmetrically distributed on both sides of the Juanshui River, and within a range of two kilometres in a straight line from the Juanshui River (Figure 2, Table 1). Soil samples were collected vertically using auger boring at a depth of 80 cm. Three groups of samples were taken from each sampling point according to the random principle. Each 10 cm depth of soil was considered a complete sample. A total of 96 soil samples were collected from the mouth of the Juanshui River, then stored in cloth bags and air-dried in the laboratory. After the plant roots, stones, plastic products, and other litter were discarded from samples, a 20-mesh sieve was used for testing pH, and a 200-mesh sieve for testing heavy metal elements and organic matter (OM), respectively. Samples of this study were sent to the Hunan Institute Of Geological Survey testing centre for testing.

Figure 2. Map of the sampling sites.

Table 1. Location of sampling sites.

		1	2	3	4	5	6
Section A-A	Longitude	112.92	112.92	112.92	112.92	112.92	112.92
	Latitude	27.78	27.78	27.77	27.77	27.78	27.78
Section B-B	Longitude	112.90	112.90	112.90	112.90	112.90	112.90
	Latitude	27.76	27.76	27.76	27.76	27.76	27.76

2.3 Analysis methods

2.3.1 Nemerow index method

Nemerow is a weighted multi-factor environmental quality index that considers the extreme values or highlights the maximum values. The comprehensive pollution index obtained by the Nemerow method is used to evaluate the pollution situation in the study area. The method can arrive at the pollution index of each pollutant, which reflects the integrated pollution level of different pollutants in the soil. It is suitable to evaluate comprehensive pollution, such as soil heavy metal pollution [29]. The mathematical expression is as follows:

(1) $p_i=\frac{C_i}{C_n}$(1)

Where P_i is the pollution index of element i; C_i is the measured content of element i; C_n is the environmental quality standard value of element i. In this study, the environmental quality standard of this study was taken from the background value of heavy metals in the Xiangjiang River Basin [30]. The classification criteria are shown in Table 2.

(2) $P_N=\sqrt{\frac{{\left(P_{i\mathrm{v}\mathrm{a}\mathrm{e}}\right)}^2+{\left(P_{imax}\right)}^2}{2}}$(2)

Table 2. Classification criteria of single-factor index pollution.

Pi	Pi≤ 1		1< Pi ≤ 2	2< Pi ≤ 3		Pi > 3
Pollution level	No pollution		Low pollution	Medium Pollution		Heavy Pollution

where P_N is the comprehensive pollution index of heavy metals in soils; P_imax is the maximum of the individual P_i of heavy metals in soils; $P_{\mathrm{i}\mathrm{v}\mathrm{a}\mathrm{e}}=\frac{1}{n}\sum _{i=1}^n{P}_i$ is the average of the individual P_i of heavy metals in soils, and classification criteria of comprehensive soil pollution are as in Table 3.

Table 3. Classification criteria of comprehensive soil pollution degree.

NIPI	Pollution Grade	Quality Value
PN≤ 0.7	1	Practically uncontaminated
0.7< PN≤ 1.0	2	Relatively uncontaminated
1.0< PN≤ 2.0	3	Low contamination
2.0< PN≤ 3.0	4	Moderate contamination
PN> 3.0	5	High contamination

2.3.2 Potential ecological risk index method

Swedish scientist Hakanson [31] proposed the potential ecological risk index method, according to the properties and environmental behaviour characteristics of heavy metals from the sedimentological perspective. The potential ecological risk index is a method for evaluating the ecological risk of heavy metals in sediment or soil [32]. It not only considers the content of heavy metals comprehensively but also considers some factors, such as the multi-element synergistic, toxicity levels, contamination degree, and the sensitivity of the environment for heavy metals pollution [33]. In addition, this evaluation methodology is widely used at present that can reflect the environmental impact of a single heavy metal element and the comprehensive effects of multiple heavy metal pollutants [34–36]. Therefore, the potential ecological risk index method was selected to evaluate the ecological risk of the soil at the mouth of the Juanshui River. The mathematical expression is as follows:

(3) $C_f^i=\frac{C_s^i}{C_n^i}$(3)

(4) $E_r^i=T_r^i\times C_f^i$(4)

(5) $RI=\sum _{i=1}^n{E}_r^i=\sum _{i=1}^n{T}_r^i\times C_f^i=\sum _{i=1}^n{T}_r^i\times \frac{C_s^i}{C_n^i}$(5)

where RI is the comprehensive index of multi-element environmental risk; Ei r is the ecological risk index of heavy metal element i; Ci f is the pollution coefficient of heavy metal element i against reference ratio; Ci s is the measured concentration of heavy metal element i; Ci n is the assessment reference ratio of heavy metal element i (The background values of heavy metals in Xiangjiang River Basin was used in this study) [30]; Ti r is the toxicity response coefficient of heavy metal element i that reflects the toxic level of heavy metals and the sensitivity of environment for heavy metals pollution. The toxicity response coefficient of heavy metals is shown in Table 4. Classification criteria for potential ecological risk levels are shown in Table 5.

Table 4. Toxicity coefficient of heavy metals.

Heavy metals	Cr	Cu	Cd	Pb	Hg	As
${\mathrm{T}}_r^i$	2	5	30	5	40	10

Table 5. Classification criteria of potential ecological risk levels.

Ei r	Ecological Grade	RI
$E_r^i$< 40	Low risk	$RI$<150
40≤ $E_r^i$< 40	Moderate risk	150≤$RI$<300
80≤ $E_r^i$< 160	Considerable risk	300≤$RI$<600
160≤ $E_r^i$< 320	High risk	$RI$≥600
$E_r^i$≥ 320	Very high risk

2.2.3 PCA

PCA is a multivariate statistical method in accordance with dimensionality reduction thinking [37], which reduces many indicators to a smaller set of comprehensive indicators on the premise of losing little information [38]. Hence, relationships between a smaller set of comprehensive indicators can effectively be examined [39]. The set of small comprehensive indicators is used to explain the pollution status of heavy metals in soil. Combined with the characteristics of heavy metals released by various industries and rock weathering, the predominant sources of heavy metals were effectively identified in the study area [40,41]. Kaiser-Meyer-Olkin (KMO) and Bartlett’s sphericity tests were used to confirm the convenience of the data for PCA [42].

3. Results and discussion

3.1 Heavy metal contents and spatial distribution in soil sections

The results of descriptive statistics for Cr, Cu, Cd, Pb, Hg, As, PH, and OM are shown below (Table 6). Among 48 samples in section A-A, the mean concentrations of Cr, Cu, Cd, Pb, Hg, and As were 85.98, 38.59, 1.37, 75.30, 0.17, and 29.10 mg kg⁻¹, respectively. The mean concentration of Cd was 2.3 times the background value of heavy metals in Xiangjiang River Basin, and other heavy metals were 1.6 to 1.9 times the background value of heavy metals in the Xiangjiang River Basin. In contrast, there was a significantly high enrichment of Cd. Among 48 samples in section B-B, the mean concentrations of Cr, Cu, Cd, Pb, Hg, and As were 90.44, 35.25, 0.98, 55.03, 0.13, and 17.20 mg kg⁻¹, respectively. More specifically, the mean concentrations of heavy metals in section B-B were 1.3 to 1.7 times the background value of heavy metals in the Xiangjiang River Basin. The aforementioned results implied heavy metals enrichment to a certain degree. In addition to Cr, the mean concentrations of all heavy metals were lower in section B-B compared to section A-A. The mean concentrations of heavy metals in the soil of the mouth of the Juanshui River followed a sequence of Cr > Pb > Cu > As > Cd > Hg. Among them, the mean concentrations of Cr and Pb were higher. PH ranges of soil section A-A and section B-B were 4.26 to 7.21 and 5.28 to 7.81, with a mean PH value of 6.42 and 6.43, respectively. It indicated that the soil is slightly acidic in the mouth of the Juanshui River. The mean contents of soil OM for section A-A and section B-B were 13.85 and 13.67 mg kg⁻¹, respectively. A high abundance of soil OM indicated a strong migration ability of heavy metals in the soil of the mouth of the Juanshui River. Therefore, the increased intensity exchange of materials and energy leads to decreased PH value in the soil of the mouth of the Juanshui River, showing slight acidity.

Table 6. Soil heavy metal contents in the mouth of the Juanshui River (mg/kg).

		Cr	Cu	Cd	Pb	Hg	As	pH	OM
Section A-A	Min	72.23	33.97	0.503	49.68	0.106	17.78	4.26	6.23
	Max	102.98	47.86	5.526	117.78	0.304	39.03	7.21	34.67
	Mean	85.98	38.59	1.37	75.30	0.17	29.10	6.42	13.85
	SD^a	5.36	1.76	1.03	10.56	0.03	4.92	0.62	12.84
	CV	0.06	0.05	0.78	0.14	0.15	0.17	0.10	0.93
Section B-B	Min	75.23	25.80	0.222	38.08	0.090	11.42	5.28	6.23
	Max	105.64	118.12	2.561	74.22	0.399	24.99	7.81	34.67
	Mean	90.44	35.25	0.98	55.03	0.13	17.20	6.43	13.67
	SD^a	1.76	6.74	0.43	5.59	0.02	1.56	0.20	14.49
	CV	0.02	0.19	0.45	0.10	0.12	0.09	0.50	1.06
Background		54	22	0.60	40	0.10	15	-	-

^aSD: standard deviation.

The Xiangtan Iron and Steel Plant and a large quantity of farmland were distributed in the vicinity of the study area. The iron and steel plant engaged in frequent smelting activities to maintain working order. In order to make a profit, farmers applied fertilizers and pesticides to the farmland around the study area. In the study of soil heavy metals contamination in industrial and agricultural regions of China, Zhou et al. [43] showed that heavy metals pollution and soil acidification had become increasing problems with the development of industry and agriculture. The pollution of heavy metals Cd, Pb, and As were more serious, among which Cd has the most obvious accumulation. Liu et al. [44] showed that Cr, Pb, Cu, As, Cd, and Hg exceeded to varying degrees in the study of heavy metal pollution in the sediments of the Xiangjiang River. The Juanshui River was a tributary of the Xiangjiang River. The type and accumulation degree of heavy metals pollution in the study area were quite similar to the Xiangjiang River. Due to the influence of human activities, heavy metals had accumulated in the soil of the study area.

The CV is applied to reflect the average variation degree of heavy metals in soil samples and the fluctuation degree of pollutants on a spatial scale. Cd has the maximum variability with a CV value of 0.78 and 0.45 in section A-A and section B-B, respectively. 0.1 < CV < 1 belongs to moderate variability. Higher CV denotes the more inhomogeneous distribution for heavy metals on the spatial scale. Therefore, in all probability, point source pollution occurs in the soil of the mouth of the Juanshui River. Liu et al. [45] found that Cd has moderate variability, which is affected by anthropogenic factors to a certain degree.

The spatial distribution of heavy metals in the soil of the mouth of the Juanshui River was mapped using the IDW interpolation method (Figure 3). Spatial variation of heavy metal content indirectly reflects pollution sources. The concentrations of Cu, Cd, Pb, Hg, and As were almost identical in spatial distribution. Meanwhile, Cu, Cd, Pb, Hg, and As with higher concentrations in section A-A. Contrary to the aforementioned heavy metals, the concentrations of Cr were higher in section B-B. Regarding geographical position, a large quantity of farmland is distributed around sections B-B. In a study of chromium accumulation in Chinese agricultural soils, Li et al. [46] found that the application of Cr-containing fertilizers was the primary pathway for Cr pollution in agricultural soils. Therefore, it is speculated that chemical fertilizer was the main reason for the high Cr content.

Figure 3. Spatial distribution of heavy metal contents (mg kg⁻¹⁾ in soil sections.

3.2 Pollution assessment in soil sections

The background values of heavy metals in the Xiangjiang River Basin were used as the environmental quality standard for assessing the pollution degree of heavy metals in this study. The soil pollution status was displayed in Table 7, which was calculated based on the Nemerow index method. The mean P_i of heavy metals in section A-A with a descending order that Cd (3.00) > Pb (1.96) > As (1.94) > Cu (1.76) > Hg (1.71) > Cr (1.58). Cd was in moderate contamination, while other heavy metals were in low contamination in section A-A. The mean P_i of heavy metals in section B-B with a descending order that Cr (1.68) > Cd (1.61) = Cu (1.61) > Pb (1.37) > Hg (1.30) > As (1.15). All heavy metals belong to low contamination in section B-B. Cr has the highest mean P_i followed by Cd and Cu. It was observed from the spatial distribution map of heavy metal contents that Cr shows a high concentration in section B-B. It was highly consistent with the results of the Nemerow method. Section A-A had a moderate degree of pollution with a P_N value of 2.84, whereas the P_N value of section B-B was 1.76, which was within the low pollution range. Sampling sites No. 5 and No. 6 located in the floodplain of the Juanshui River, were obviously polluted, especially sampling sites No. 5 and No. 6 in section A-A. The soil pollution status of sampling sites (Figure 4) generally showed a symmetrical pattern (No. 5-No. 6, No. 1-No. 4, and No. 2-No. 3). Sampling sites were located within two kilometres of the Juanshui River. It may be that heavy metals are prove to migrate from the water body to soil due to the water body rising in flood season, causing serious soil pollution in the floodplain. Zhao et al. [47] reported that close to the floodplain, soil heavy metal pollution is more serious.

Figure 4. Pollution index (P) on A-A section (left) and B-B section (right).

Table 7. Results of pollution index of heavy metals in soil sections.

Number		Pi						PN
		Cr	Cu	Cd	Pb	Hg	As
Section A-A	1	1.64	1.91	1.48	2.35	1.91	2.34	2.13
	2	1.54	1.77	1.29	1.87	1.66	2.23	1.99
	3	1.42	1.65	2.07	1.89	1.49	1.99	1.87
	4	1.56	1.74	1.18	1.85	2.10	2.11	1.94
	5	1.65	1.75	5.98	1.90	1.56	1.47	4.55
	6	1.65	1.75	5.98	1.90	1.56	1.47	4.55
	Mean	1.58	1.76	3.00	1.96	1.71	1.94	2.84
Section B-B	1	1.69	2.24	0.88	1.30	1.27	1.19	1.87
	2	1.66	1.60	1.76	1.48	1.38	1.22	1.64
	3	1.74	1.61	1.83	1.48	1.34	1.27	1.69
	4	1.65	1.25	0.64	1.09	1.14	0.94	1.41
	5	1.64	1.52	1.67	1.44	1.57	1.14	1.59
	6	1.66	1.43	2.86	1.46	1.09	1.12	2.32
	Mean	1.68	1.61	1.61	1.37	1.30	1.15	1.76

In summary, Cd should be given priority attention in the soil of the mouth of the Juanshui River. Moreover, the P_N value of section B-B was obviously lower than section A-A, which was closer to the Xiangjiang River. Many studies indicated that Cd, Pb, As, Cu, Hg, and Cr were the major pollution element in the Xiangjiang River basin [48]. Among them, Cd showed the strongest pollution [49]. These findings were consistent with the results of this study. Therefore, except for the surrounding residents and a large quantity of farmland, the enrichment of heavy metals might be related to the rising water in the Juanshui River and the Xiangjiang River.

3.3 Ecological risk assessment in soil sections

Based on the potential ecological risk index method, the ecological hazards of heavy metals in the soil of the mouth of the Juanshui River were evaluated in Table 8. The mean values of Ei r followed a descending order that Cd> Hg> As> Pb> Cu> Cr in section A-A. As to section B-B, the sequence was Hg> Cd> As> Pb> Cu> Cr. Moreover, all the mean values of Ei r for As, Pb, Cu, and Cr were below 40. There was a close similarity between the two soil sections for the risk level of heavy metal. Hg and Cd had the highest risk in the study area owing to their high toxicity. It was also found that Hg and Cd were major risk factors for soil pollution in the urban renewal area of Beijing [50]. Shi et al. [9] reported that Hg and Cd were major risk factors for soil pollution in China, and the mean values of Ei r for Pb, Cr, As, and Cu in soil were below 40. The above findings were consistent with our research results. Therefore, Cd and Hg deserved more attention considering the high potential ecological risks. The mean value of RI was 199.57 in section A-A, meaning moderate risk. By contrast, the mean value of RI was 129.86 in section B-B, which was relatively far from Xiangjiang River, meaning low risk. Meanwhile, Guo et al. [51] also found that the farther away from the estuary, the lower the potential ecological risk. RI value of soil samples showed a symmetrical pattern of No. 1-No. 4, No. 2-No. 3, and No. 5-No. 6. Namely, heavy metal pollution was symmetrically distributed based on the mouth of the Juanshui River. The mean values of RI of sampling sites No. 5 and No. 6 were significantly higher than other sampling sites in two soil sections (Figure 5). It was consistent with the result of the Nemerow index method, which has been successfully verified.

Figure 5. Potential ecological risk index (RI) on A-A section (left) and B-B section (right).

Table 8. Results of potential ecological risk index of heavy metals in soil sections.

Number	Ei r							RI
		Cr	Cu	Cd	Pb	Hg	As
Section A-A	Min	2.84	8.27	38.63	9.25	59.42	14.71	148.60
	Max	3.31	9.56	179.37	11.73	76.57	23.44	278.23
	Mean	3.16	8.81	89.87	9.80	68.58	19.36	199.57
	RI	18.93	52.88	539.19	58.80	411.46	116.14	-
Section B-B	Min	3.29	6.24	19.23	5.47	43.58	9.43	158.35
	Max	3.48	11.18	85.83	7.39	62.70	12.68	158.35
	Mean	3.35	8.03	48.25	6.87	51.89	11.47	129.86
	RI	20.10	48.17	289.48	41.22	311.35	68.82	-

3.4 Source analysis

The correlation between heavy metals was strong evidence for inferring common origin or similar geochemical processes. The correlation coefficients between heavy metals, PH, and OM were displayed in Table 9, which were calculated by SPSS. OM was positively and significantly correlated with Cu (r = 0.634, p < 0.01), Pb (r = 0.350, p < 0.05) as well as Hg (r = 0.314, p < 0.05), respectively, revealing that homology presented in each group above. There was a significant negative correlation (r= −0.501, p < 0.01) between Pb and pH, indicating that Pb was prove to be released from acidic soil conditions. In a word, heavy metals in soils were easily affected by the physicochemical properties of soil. Additionally, Pb was positively and significantly correlated with Hg (r= −0.346, p < 0.05) and As (r= −0.381, p < 0.01), respectively, and the correlation coefficient between As and Cd (r= −0.433, p < 0.01) significantly showed a negative correlation. So the source of Pb, Hg, and As was common. Jiang et al. [52] reported that Pb, Hg, and As had a common origin in the sediments of the Xiangjiang River. Overall, it was further evidenced that the Xiangjiang River was a significant source of heavy metals in the soil of the mouth of the Juanshui River.

Table 9. Correlation coefficients of heavy metals.

	Cr	Cu	Cd	Pb	Hg	As	OM	pH
Cr	1.000**
Cu	0.095	1.000**
Cd	0.223	0.011	1.000**
Pb	−0.094	0.079	0.211	1.000**
Hg	−0.156	−0.003	−0.103	0.346*	1.000**
As	−0.199	−0.185	−0.433**	0.381**	0.281*	1.000**
OM	0.123	0.634**	0.158	0.350*	0.314*	−0.274	1.000**
pH	0.191	0.095	−0.164	−0.501**	−0.247	−0.231	−0.385**	1.000**

*Correlation is significant at the 0.05 level.

**Correlation is significant at the 0.01 level.

KMO and Bartlett’s sphericity tests were performed to show effectiveness for PCA with a KMO index of 0.591 (>0.5) and a p-value of 0.002 (<0.005) [53]. In Table 10, the results of PCA showed that the eigenvalues of three principal components were all>1, which explained 88.08% of the total variables. The variance contribution rate on PC1 was 53.082%, with a higher load factor for Pb (0.93), Hg (0.88), and As (0.957). From previous studies, Pb and Hg were considered the marker elements for vehicle exhaust and coal burning, respectively [54–56]. According to field surveys, a large-scale steel mill was found in the north of the study area. Coal is one of the fundamental fuels for smelting. For a normal working order, smelting required transporting and burning large amounts of coal. Due to the consumption of leaded gasoline and coal, Pb and Hg entered the soil through atmospheric deposition. The CV for Pb and Hg ranged from 0.14 to 0.19, which was a medium variance degree, indicating the planar pollution rather than point source pollution of Pb and Hg was attributable to the waste gas dispersed by wind. Zhao et al. [57] indicated that the high As concentration in the soil might be related to industrial discharges. Meanwhile, Huang et al. [58] also reported that the illegal discharge of industrial wastewater during smelting caused As enrichment. Undoubtedly, the steel mill was the most predominant source of As pollution. Therefore, PC1 was inferred as a compound source of industry, traffic, and coal burning.

Table 10. Results of PCA for sources of heavy metals.

	PC1	PC2	PC3
Eigenvalue	3.185	3.185	1.004
Of variance(%)	53.082	18.256	16.739
Cumulative(%)	53.082	71.338	88.077
Cr	−0.625	0.003	0.634
Cu	0.482	−0.243	0.752
Cd	0.082	0.975	0.165
Pb	0.930	0.249	0.057
Hg	0.880	−0.075	0.031
As	0.957	−0.131	−0.064

The variance contribution rate on PC2 was 18.256%, with a higher load factor for Cd (0.975). Among all heavy metal elements, Cd had the highest CV, indicating Cd was under the influence of human disturbances. Sun et al. [59] considered Cd in the soil as a marker for using chemical fertilizers in agronomic activities. The application of phosphate fertilizer led to Cd enrichment [60]. A review of soil heavy metal pollution in China industrial zone showed that Cd in the atmosphere around industrial areas entered into the soil by rainfall and sedimentation, leading to a high Cd concentration in the soil of the industrial zone [43]. Considering the fact that farmland and the steel mill were distributed around the study area, PC2 was defined as the agricultural and industrial source.

The variance contribution rate on PC3 was 16.739%, with a higher load factor for Cr (0.634) and Cu (0.752). Many studies indicated that Cr was mainly influenced by soil parent material [61,62]. However, a large quantity of farmland was distributed around section B-B. As explained above, the concentrations of Cr were higher in section B-B. There was a strong suspicion that the high Cr content in the soil was associated with farmland. In addition, Liang et al [63]. found that agrichemicals (e.g. chemical fertilizer and insecticide) were leading factors for the high concentration of Cu. Consequently, there was a compound source for nature and agriculture on PC3.

4. Conclusion

This study investigated the pollution status, ecological risk, and pollution source in the soil of the mouth of the Juanshui River. Research indicated that the mean concentrations of all heavy metals exceeded the Xiangjiang River Basin background values and showed moderate variability. Among all heavy metals, Cd has the maximum variability. A high abundance of soil OM indicated a strong migration ability of heavy metals in the soil of the mouth of the Juanshui River.

According to the Nemerow method, soil pollution in the mouth of the Juanshui River has reached a moderate level. The soil in the study area was moderately polluted by Cd and slightly polluted by Cr, Cu, Pb, Hg, and As. In terms of sampling sites, No. 5 and No. 6 located in the floodplain of the Juanshui River were obviously polluted, especially sampling sites No. 5 and No. 6 in section A-A, which was closer to the Xiangjiang River. The potential ecological risk index indicated that Cd and Hg were the main risk contributions.

PCA speculated that As and Hg were associated with industrial activities and coal burning. Pb was delivered from traffic fumes, which indirectly resulted from industrial production. Cu and Cd were delivered from agricultural chemicals. In view of the spatial distribution pattern, Cr pollution may be affected by both soil parent material and agricultural chemicals.

The government should strictly control the usage of industrial wastewater, traffic exhaust, coal burning, and agricultural chemicals to ensure sustainable development. From the perspective of the location and pollution regularity of the study area, The Juanshui River and the Xiangjiang River are secondary sources of heavy metals pollution in the study area. Except for industrial and agricultural activities, secondary pollution sources also have a great impact on the ecological environment. The methods in this study were insufficient to support the aim that the mutual pollution between soil and water is fundamentally solved. Therefore, combining a new method to study the forms, migration, and transformation of heavy metals is necessary. Finally, this study has practical significance for maintaining soil equilibrium in similar areas of China and restoring global ecological resources.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Key R&D project of Science and Technology Department of Hunan Province (2022SK2073).