Heavy Metal Content and Pollution Assessment in Typical Check Dam Sediment in a Watershed of Loess Plateau, China

Abstract

To understand historical trends and assess the ecological risk associated with heavy metal pollution, the concentration of eight species of heavy metals (vanadium (V), chromium (Cr), manganese (Mn), cobalt (Co), nickel (Ni), zinc (Zn), copper (Cu), and arsenic (As)) in typical silt dam sediments on the Loess Plateau were analyzed. The radionuclide ¹³⁷Cs was used to quantify rates of erosion, deposition, and heavy metal contamination in the soils of a watershed that supplies a check dam. The sediment record revealed three time periods distinguished by trends in erosion and pollutant accumulation (1960–1967, 1968–1981, and 1985–1991). Heavy metal concentrations were highest but exhibited significant fluctuation in the first two periods (1960–1967 and 1968–1981). From 1985 to 1991, heavy metal pollution showed a downward trend and tended to be stable. The potential risks of heavy metals in silt dam sediments were explored by applying the geo-accumulation index and the potential ecological risk index. The results indicated medium risk associated with Cu and As accumulation, especially in 1963, 1971, and 1986 when the assessed values increased significantly from previous levels. Agricultural practices and high rates of slope erosion may be responsible for the enrichment of As and Cu in soil and the accompanying increase in risk. Land use optimization and the careful use of fertilizers could be used to control or intercept heavy metal pollutants in dammed lands. The results provide the basis for evaluating the current status and ecological risk of heavy metal contamination in dam sediments and for predicting possible heavy metal pollution in the future.

1. Introduction

Rapid industrial and agricultural expansion has resulted in widespread heavy metal contamination in soils. Because heavy metals are persistent and toxic, and in some cases, irreversibly bound to soil minerals or soil organic matter, this type of pollution presents serious risks to human health and economic development. [1,2]. Heavy metals can originate from natural sources (e.g., rock weathering, soil parent material, etc.) or from artificial sources (e.g., industrial and agricultural production, traffic pollution, etc.) and can accumulate over time [3]. When agricultural production takes place on soils enriched in heavy metals, these toxic substances infiltrate the food chain, creating substantial risks to human health [4,5,6,7].

On the Loess Plateau in Northwest China, heavy rainfall coupled with intense human activity has resulted in heavy erosion [8,9,10,11]. Starting in the 1950s, the government began implementing engineering and forestry measures intended to conserve water and reduce rates of erosion [12]. As part of this strategy, an extensive network of warping dams was constructed across the Plateau. These dams block sediment export by retaining and depositing soil materials dislodged by erosive rainfall. Not only do warping dams reduce soil loss from erosion, but the resulting alluvial deposits enhance the Plateau’s fertility [13,14].

The sediment retained by warping dams integrates information about natural and anthropogenic processes occurring in the area (e.g., pollution of As and Cu in cultivated soil). Because soils are a major sink for heavy metals, the sediment contains a record of pollution in the area. Moreover, when environmental conditions (e.g., temperature, pH, and organic matter content) change, heavy metals may be redistributed in the sediment, causing secondary contamination [15,16]. Thus, establishing the sequence of sedimentary ages can reveal the historical inputs and associated trends in heavy metal pollution.

Soil pollution is typically evaluated using the geo-accumulation index (I_geo) in conjunction with the potential ecological risk coefficient ($E_r^i$) [17,18]. This method has been used by researchers to study, for example, Cadmium contamination in near-shore sediments of the Yellow River [17] and heavy metal contamination in surface sediments of the Bay of Bengal estuary [18]. Recently, scientists have begun augmenting these methods with GIS geostatistical analysis and artificial neural network models to produce maps depicting the spatial distribution of pollutants [19].

Heavy metal pollution in river sediment has been widely studied [20,21,22,23]. However, there are few studies describing how the distribution and chemistry of heavy metals changes over time in check dam sediments. In this study, the age and deposition rate of heavy metals were determined using the radionuclide cesium-137 (¹³⁷Cs). The heavy metal content of stratified samples was also assessed to derive age-specific heavy metal accumulation characteristics. Together, this information provides the basis for evaluating the current status of dam site sediments and predicting possible future heavy metal contamination. The objectives of this study were: (1) to identify the age of check dam sediments; (2) to identify how the heavy metal content in check dam sediment varies over time; and (3) to evaluate the extent of heavy metal pollution in dam land.

2. Materials and Methods

2.1. Study Sites

The 5.97 km² Wangmaogou watershed is situated in the Loess Plateau (110°20′49.5″ E–110°21′00.3″ E, 37°35′36.3″–37°35′11.2″ N) at an altitude of 940–1188 m, with an annual average erosion modulus of 18,000 t/km²·a [24]. The Nianyangou watershed is a small branch of the Wangmaogou watershed, with an elevation between 1027 and 1188 m. The Nianyangou watershed has three silt dams, of which two have drainage structures while the third has no water release structure. This study focused on the basin of the third Nianyangou dam located within the first branch of the Wangmaogou watershed. Built in 1956 and silted in 1990, this dam has controlled an area of 0.46 km² for 34 years. As rainfall scours the area, the dam intercepts sediments formed when topsoil on the slope of the basin is dislodged from areas of different land use types (e.g., slope farmland, shrubland, and grassland).

2.2. Sampling and Testing

Samples were collected from Nianyan Ditch in Suide County on 25 July 2020 (Figure 1). Three gravity samplers (10 cm in diameter) were used to collect columnar sediment samples along the dam at a depth of 6.5 m. After collection, complete sediment samples were divided into cyclic layers if boundaries were easily distinguishable based on color and particle composition. Each of the 77 columnar samples collected was divided into 29 layers and placed in a clean plastic bag for immediate shipment to the laboratory for cryogenic (4 °C) storage. After samples were air-dried, impurities were removed and they were passed through a 0.155 mm (100 mesh) sieve in preparation for testing.

Each sample was evenly mixed and 0.10 g (accurate to 0.0001 g) subsamples were placed in an X-press digestion tube. Aqua regia (nitrohydrochloric acid) was prepared by mixing concentrated hydrochloric acid (HCl) and concentrated nitric acid (HNO₃) in a 3:1 ratio by volume. Microwave digestion was performed using the following procedure: A total of 6 mL aqua regia was added to each X-press tube containing a sample, after which the tube was placed in a microwave digestion instrument. The sample was heated to 120 °C over an interval of 5 min and held at the target temperature for 2 min. The temperature was increased again over a period of 8 min to 185 °C, where it was held for 40 min. Samples were allowed to cool to room temperature, after which they were placed in a fume hood. Acid was driven off by heating the digested samples to 120 °C and holding for 2 min. The solution was passed through filter paper into a 50 mL volumetric flask, after which the digestion tube and filter paper were cleaned 3 times with 5% dilute nitric acid. The final volume was fixed using deionized water.

An inductively coupled plasma mass spectrometer (ICP-MS, Agilent Technologies, Santa Clara, CA, USA) was used to determine the concentration of the eight heavy metals of interest (V, Cr, Mn, Co, Ni, Cu, Zn, and As). To ensure analytical precision, relative standard deviation (RSD) values were calculated for instrument sensitivity, oxides, double charge parameters, and the signal intensity of elements in the tuning solution. Experimental results with RSD values >5% for any factor were rejected. Before running a blank measurement, the system was flushed with 5% nitric acid until the analytical signal was stabilized. The recoveries of soil samples ranged from 87.8% to 111.0%.

Soil ¹³⁷Cs was measured by a high purity germanium γ spectrometer (ORTEC, Oak Ridge, TN, USA) analyzer. The determination time for each sample was 28,800 s. The mass activity of ¹³⁷Cs was calculated from the net area of 661.62 KeV rays, and the relative error of samples was less than 10%. Soil bulk density was determined for each divided cyclic layer by collecting the undisturbed 200 cm³ soil core using a ring knife. The sample was weighed, dried in an oven at 105 °C for 16 h, then reweighed after cooling to 25 °C.

2.3. Data Analysis

2.3.1. Calculation of Soil Bulk Density

Soil bulk density was calculated using [25]:

$${\rho }_b=G\times 100/\left(v\times \left(100+w\right)\right)$$

(1)

where ${\rho }_b$ is the soil bulk density (g/cm³), $G$ is weight of the field moist soil sample (g), $v$ is the internal volume of the ring knife (cm³), and $w$ is the soil moisture content.

2.3.2. Calculation of Cyclic Layer Sediment Weight and Total Sediment Weight

Cyclic layer sediment and total sediment volume were calculated using the formulas from [26]:

$$M_i=V_i\times {\rho }_{i}i=1,2,\dots \mathrm{n}$$

(2)

$$M={{\displaystyle \sum }}_1^n{M}_i$$

(3)

where $V_i$ is the volume of the i-th sedimentary cycle (m³); ${\rho }_i$ is the unit weight of the i-th sedimentary cycle (g·cm⁻³); $M_i$ is the sediment deposition volume of the i-th sediment cycle (t); M is the gross amount of sediment retained (t).

2.3.3. Calculation of Heavy Metal Deposition Rate

Heavy metal deposition rate was calculated using [26]:

$$W=\Delta M/\left(T_2-T_1\right)$$

(4)

$$\Delta M=\sum B_i\times h_i$$

(5)

$$\sum h_i=\Delta H$$

(6)

where T₁ and T₂ are the start and end years of the period estimated using ¹³⁷Cs, respectively; ∆H is the thickness corresponding to the interval between T₁ and T₂; $\Delta M$ is the mass of sediment per unit area (g/cm²) corresponding to ∆H; $B_i$ and $h_i$ are the deposit bulk density (g/cm³) and thickness (cm), respectively; and W is the deposition rate of heavy metals (g/cm²·a).

2.3.4. Risk Analysis

The geo-accumulation index (I_geo), also known as the Muller index, is an important parameter used to distinguish the impact of human activities [27]. It is calculated using the equation:

$$I_{geo}=lo{g}_2\left[C_n/\left(1.5B_n\right)\right]$$

(7)

where $I_{geo}$ is the geo-accumulation index, $C_n$ is the measured concentration of heavy metals in soil, $B_n$ is the geochemical background concentration of soil heavy metals, and 1.5 is the background matrix correction factor.

The potential risk index ($E_r^i$) considers the types, contents, and toxicity levels of pollutants in sediments to quantify ecological risk [28]. It was calculated using the equation:

$$C_i^f=\frac{C_s^i}{C_n^i}$$

(8)

$$E_r^i=T_r^i\times C_f^i$$

(9)

$$RI={{\displaystyle \sum }}_{i=1}^n{E}_r^i$$

(10)

where $C_i^f$ is the pollution coefficient of a metal, $C_s^i$ the measured value of heavy metals, $C_n^i$ is the parameter ratio, $E_r^i$ is the potential ecological risk coefficient of heavy metals $, T_r^i$ is the response coefficient of metal toxicity, and $RI$ is the comprehensive potential ecological damage index of various heavy metals.

Evaluations of risk using this metric vary depending on the background knowledge of the person using it. Because of its subjective nature, the potential risk index is notoriously difficult to use. The types of heavy metal elements measured in this paper are different from the eight elements evaluated by Hakanson (1980). Therefore, the Hakanson criteria cannot be used to evaluate heavy metal pollution in sediments collected in the Nianyan Ditch [29]. $E_r^i$ for these samples was calculated using $T_r^i$ values determined by Xu et al. (2008) (

Table 1. Response coefficient values of heavy metals in dam sediments.

Element	V	Cr	Mn	Co	Ni	Cu	Zn	As
Toxicity Coefficient	2	2	1	5	5	5	1	10

) [30].

To fit the area assessed in Nianyan Ditch, the ecological hazard index for heavy metals in the study area was adjusted using the risk intensity limit adjustment method developed by Lin et al. (

Table 2. Ecological risk index and risk intensity grade associated with Nianyan Ditch sediments.

${\mathit{E}}_{\mathit{r}}^{\mathit{i}}$	Degree	$\mathit{R}\mathit{I}$	Degree
<9	Slight	<35	Slight
9 ≤ r < 18	Medium	35 ≤ RI < 70	Medium
18 ≤ r < 37	Considerable	70 ≤ RI < 140	Considerable
37 ≤ r < 75	Very strong	RI ≥ 140	Very strong
75 ≤ r	Extremely strong

Note: $E_r^i$ is the potential ecological risk coefficient of heavy metals; RI is the comprehensive potential ecological damage index.

) [31].

3. Results

3.1. Determination of Sediment Age

During a typical rainstorm, eroded materials are transported along the runoff path and deposited in front of the dam, forming a layer of sediment. Each of these layers is formed by a single depositional event, resulting in the development of a stratified sedimentary bed [32]. The date at which each cyclic layer formed can be determined using weather records, sediment volume, and ¹³⁷Cs activity.

The mass specific activity of ¹³⁷Cs in the samples collected for this study varies with depth (Figure 2). Two storage peaks are obvious: the main storage peak (4.86 Bq/kg) at a depth of 530 cm and the second storage peak (3.99 Bq/kg) at a depth of 154.8 cm. Globally, the abundance of ¹³⁷Cs peaked in 1963 as the result of nuclear weapons testing [33] and crested again in 1986 following the Chernobyl nuclear leakage event [34]. Based on the depth profile of ¹³⁷Cs activity, it was determined that the deep storage peak (layer 24) formed in 1963 and the shallow storage peak (layer 7) formed in 1986. Cyclic layers containing greater sedimentary volume correspond to heavy, erosive rainfall events, which allowed for the development of the more highly resolved timeline presented in Figure 2.

3.2. Patterns of Heavy Metal Concentration in the Dam Sediment Site over Time

The soil profile distribution of heavy metals in the sediment reflects the depositional record of each element, and the change in each stage reflects different geochemical behaviors. Concentrations of heavy metals in the soil profile at Nianyangou are greatly varied. The contents of V, Cr, Mn, Co, Ni, As, Zn, and Cu are 32.83–75.04, 30.36–122.54, 440.06–774.69, 12.60–28.35, 19.26–44.56, 17.99–51.34, 56.23–130.44, and 10.97–22.08, respectively. The sampled sediments are relatively enriched in Zn and Mn, mostly due to their high abundance in the crust. The variation of each heavy metal with depth is shown in Figure 3, and the direction of change in concentration (i.e., increasing or decreasing) is roughly mirrored across all eight metals at a given depth.

Based on ¹³⁷Cs activity, thickness of the silt layer, and the vertical distribution of heavy metals, the sediment core was divided into three distinct time periods. The deepest core segment corresponds to the period between 1960 and 1967. This section is distinctive because of its thick silt layer, reflecting the strong erosion of the drainage basin and high rates of sediment deposition during this period. This segment is also notable because a decrease in Cr concentration in 1963 was coupled with large increases in the concentration of each of the seven other heavy metals. The middle segment corresponds to a period spanning from 1968 to 1981. Though there were a few thick layers in this section, cyclic layers are generally thin, indicating reduced rates of erosion. The concentration of heavy metals in this segment greatly fluctuated: the concentration of Cr increased and then decreased, while the concentrations of V, Mn, Co, Ni, As, Zn, and Cu demonstrate an opposite trend, decreasing first before increasing. The shallowest segment corresponds to a period from 1985 to 1991. This section is notable because the thickness of the silt layer increased, indicating higher rates of erosion. However, the content of heavy metals in this stage tended to be stable, with generally lower rates of pollution compared to the previous two periods.

3.3. Assessment of Heavy Metal Pollution

Overall, values of $I_{geo}$ in sampled sediment were between −1.62 to 0.83 (Figure 4), indicating slight to no pollution. Values of $I_{geo}$ for Cr, Co, Cu, Zn, and As indicate slight contamination, with pollution proportions of 37.93%, 64.29%, 75%, 32.14%, and 32.14%, respectively.

In the first time period, heavy metal contamination reached its peak in 1963, with $I_{geo}$ values in the slight pollution range: 0.52, 0.24, and 0.62 for Co, Cu, and As, respectively. The second time period was also marked by slight pollution, which peaked in 1971 with $I_{geo}$ values of 0.83, 0.86, 0.51, and 0.28 for Co, Cu, Zn, and As, respectively. In the third period, heavy metal pollution was at its lowest, with only two heavy metals having $I_{geo}$ values indicating light pollution: 0.38 and 0.32 for Cu and Cr, respectively.

The potential risk index values ($E_r^i$) were ranked from from highest to lowest: As > Cu > Co > Ni > Cr > V > Zn > Mn. Among them, the risk degree of V, Cr, Mn, Ni, and Zn was low, and their $E_r^i$ values were less than nine. The $E_r^i$ values of Co and Cu ranged from 5.94 to 15.09 and 4.76 to 13.58, respectively. For Co and Cu, 65.52% and 62.10% of the sampling points showed low risk, respectively, and 34.40% and 37.90% of the sampling points showed medium risk. The values of As ranged from 9.86 to 18.69, and 89.60% of the sampling points were at medium risk (

Table 3. Contribution of each heavy metal to ecological risk assessment values.

${\mathit{E}}_{\mathit{r}}^{\mathit{i}}$	Proportion(%)								RI	Proportion (%)
	V	Cr	Mn	Co	Ni	Cu	Zn	As
Slight	100	100	100	65.52	100	62.1	100	10	Slight	24.14
medium	0	0	0	34.48	0	37.93	0	89.65	medium	75.86
Considerable	0	0	0	0	0	0	0	0	Considerable	0
Very strong	0	0	0	0	0	0	0	0	Very strong	0
Extremely strong	0	0	0	0	0	0	0	0

). The $E_r^i$ values of the three heavy metals Co, Cu, and As increased significantly in 1963, 1971, and 1986 to values indicating moderate risk, which was consistent $I_{geo}$ values for those years. The $RI$ of the eight heavy metals ranged from 31.78 to 59.69, and 78.56% of the sampling points were at medium risk. The average contributions of As and Cu to the $RI$ were 32.9% and 20.9%, respectively, while the average contribution rate of the remaining eight heavy metals to the $RI$ was 46.20% (Figure 5). Although concentrations of As and Cu in the dam sediments were low, their potential ecological risks were high.

The degree of risk indicated by values of I_geo and $E_r^i$ were generally similar. However, land accumulation index values indicated slight pollution, while $E_r^i$ indicated high potential ecological risk. This difference results from the two methods having different emphases: I_geo provides an intuitive level of heavy metal pollution but doesn’t discuss the harm of heavy metal pollution to organisms, while the $E_r^i$ reflects the biological toxicity of heavy metals. The two metrics can be used in tandem to evaluate the risks associated with heavy metal pollution more comprehensively.

4. Discussion

4.1. Effects of Watershed Erosion and Sediment Yield on the Distribution of Heavy Metals

Check dam sediment can be used to restore soil in areas that have been degraded by heavy erosion [34]. The dam traps eroded materials from across the watershed, serving as a sink for sediment exported from the surrounding slopes [35]. Studies have shown that soil erosion is usually accompanied by the movement of pollutants. Due to the strong adsorption capacity of soil colloids, heavy metals are tightly bound to soil materials and migrate under the action of erosion [36]. Thus, eroded sediments may be enriched in any heavy metals contained within the watershed [36].

Human activities have significantly altered the properties of soil. Land use change (e.g., cultivation for use in agricultural production) results in compositional changes to local vegetation. The resulting difference in plant cover and root characteristics alters soil fertility, water retention, and other edaphic properties that influence microbial community structure. Collectively, these changes affect the behavior of heavy metals by altering patterns of pollutant migration and diffusion in the soil [37,38,39]. Before the 1990s, the Wangmaogou watershed was mainly sloping farmland. After the 1990s, conservationists successfully advocated for converting cropland to forest and grassland. Similar land use changes have been reported in the basin [38,40].

In this study (Figure 6), the total amount and deposition rates of heavy metals varied over time, with rates of heavy metal deposition peaking in 1963. Heavy rainfall coupled with severe erosion and sediment generation on slopes in the watershed resulted in the transport of large amounts of silt. As a result, heavy metals exhibited high rates of vertical migration, producing a spike in total heavy metal concentrations in dam sediments, with peaks in 1963, 1971, and 1986. These peaks correspond to historical heavy rainfall events (rainfall events of 1963/7 (55 mm), 1971/7 (87 mm), and 1986/8 (70 mm), in Figure 2), particularly in 1971 and 1986. Compared to trends in the subsequent periods (1968–1981, 1982, and 1985–1991), total heavy metal concentrations decreased gradually. This finding is consistent with other studies, linking the dam’s location in a gully to the increased accumulation of easily transported materials (i.e., nutrients) from surrounding hillslopes. Moreover, higher rates of erosion are associated with an increase in phosphorus accumulation in dam water and sediment [40,41]. At the same time, the conversion from cultivated land to woodland or grassland promotes water infiltration into the soil, reduces sediment transport and runoff, and enhances soil organic carbon (SOC) storage [42,43]. Land use reversion to woodland or grassland results in SOC distribution [44]. In general, soil erosion and land use transitions are significant factors affecting soil nutrient and heavy metal contents.

4.2. Causes and Prevention Measures of Heavy Metal Pollution

Both the sources and effects of heavy metal pollution vary with land use [45,46]. In the surface soil of western Iran, for example, the content of heavy metals (Co and Ni) is related to the composition of the soil’s parent material (a natural source) [47]. In contrast, Arsenic contamination in cultivated soils results from the extensive use of chemical fertilizers in agricultural production (an artificial source) [48]. In the area studied here, soil contamination assessment methods indicated a medium risk for Cu, Co, and As (Figure 7). The accumulation of these pollutants may be related to agricultural practices on land in the watershed’s hillslope. For example, some fertilizers contain heavy metal elements, such as, Cr, Hg, and Ni, and some pesticide formulations (e.g., dicopper chloride trihydroxide, Bordeaux mixture) contain Cu and As [49]. In addition, manure applied as fertilizer has become an important source of heavy metals; these environmental contaminants are often added to livestock feed to prevent diseases and increase weight gain. Because heavy metals are poorly utilized by livestock, most of it is excreted in manure and urine. For example, cattle manure contains more than 90% of Cu consumed by the animal [50]. Thus, indiscriminate use of livestock manure by local farmers contributes to soil contamination. The history of land use in the basin and the results presented here suggest that the source of heavy metal pollution is largely caused by soil erosion and human activity. To minimize heavy metal contamination in soils, the selection and dosage rates of nitrogen, phosphorus, and potassium fertilizers should be carefully managed. In addition, the migration of heavy metals can be attenuated by ensuring reasonable vegetation cover on hillslopes (i.e., reversion to grassland, shrubland, or forest). This practice can reduce soil erosion, intercept rainfall, and change surface runoff and hydrological processes. Moreover, beyond intercepting sediment and preventing river clogging, the check dam can control surrounding slope grades to mitigate soil erosion.

The deployment of multiple control measures can greatly reduce rates of erosion. Landscape degradation can be mitigated by adjusting the proportion of land used for agriculture, forestry, and animal husbandry. When combined with judicious fertilization strategies, such as the application of passivators, soil conditioners, etc., the activity of pollutants in the soil could be reduced or eliminated. It is important to note that absorption of substances into soil materials converts toxic and harmful heavy metal ions from effective states into chemically inactive forms, thereby reducing their migration in the soil environment, their availability to plants, and their biological toxicity. By reducing rates of erosion and lowering contaminant inputs, risks to human health can be attenuated.

5. Conclusions

This investigation found that variations in the soil profile concentrations of eight heavy metals, considering depth and time, follow similar patterns. In the first (1960–1967) and second (1968–1981) periods investigated, the concentration of heavy metals greatly fluctuated. In the third period (1985–1991), the concentration of heavy metals tended to be stable. Erosion caused the longitudinal migration of heavy metals along the slope, resulting in the peak concentration of heavy metals in 1963, 1971, and 1986. Two soil pollution metrics indicated that the ecological risk of Cu pollution significantly increased in 1963, 1971, and 1986, which was closely related to the impact of land use decisions and slope water and soil loss. Agricultural production, particularly the indiscriminate use of chemical and organic fertilizers, is likely the main source of As and Cu contamination in dam sediments. Cu, Co, Mn, Ni, V, and Zn concentrations in the sediment reflect the composition of soil parent materials (i.e., natural sources). The potential ecological risks of As and Cu in the dam sediments are relatively high, indicating a need for additional mitigation efforts. Measures such as land use optimization and careful fertilizer use should be considered to reduce heavy metal pollution beyond the point sources.