Source apportionment and ecological risk assessment of heavy metals in sediments of Dongping lake based on PCA-PMF model

Abstract

Accurately characterizing the spatial distribution of heavy metals in lake surface sediments, analyzing their sources, and identifying potential ecological risks are important scientific supports for lake ecosystem management. This study selected Dongping Lake, a typical inland shallow lake in the eastern region, as the research object. It systematically analyzed the distribution characteristics, source identification, and ecological risks of eight heavy metals (As, Zn, Cu, Ni, Cd, Hg, Pb, and Cr) in the sediments of Dongping Lake by employing a combination of statistical analysis, cluster analysis, principal component analysis (PCA), and positive matrix factorization (PMF) models, along with the enrichment factor method, geoaccumulation index method, and potential ecological risk assessment method. The results show that the concentrations of the eight heavy metals (As, Zn, Cu, Ni, Cd, Hg, Pb, and Cr) in the study area exhibit an overall improving trend, with significant spatial heterogeneity, and relatively higher concentrations in the central and southern regions. Through multivariate statistical analysis, three major sources of metal accumulation were identified: industrial and agricultural activities (As, Cu, Hg, Cd), traffic emissions (Zn, Ni, Pb), and natural geological background (Cr). Enrichment characteristics showed that Cd, As, and Hg were moderately to heavily enriched, and the potential ecological risk index (RI) indicated that Hg and Cd are the main risk factors, with contribution rates of 41.83% and 37.77%, respectively. The study revealed the dominant role of human activities in the accumulation of heavy metals in the sediments of Dongping Lake, providing a scientific basis for precise pollution control and ecological restoration in the watershed.

Introduction

Heavy metals have a wide range of sources within ecological environmental systems and can migrate between different biological communities for extended periods. They are characterized by non-degradability, bioaccumulation, carcinogenicity, and other harmful properties, making their remediation extremely challenging and posing significant difficulties for environmental protection agencies. The pollution of various heavy metals such as As, Zn, Cu, Ni, Cd, Hg, Pb, and Cr presents serious threats to both the ecological environment and human health¹. Lakes, particularly urban lakes, serve as unique natural and cultural landscapes, while also playing a vital role in ecological regulation². Furthermore, they provide a variety of human activities, including fisheries, tourism, and recreational activities³. With the rapid development of the economy, society, and agriculture, lakes, as discharge zones for rivers and groundwater, have received industrial, agricultural, and domestic wastewater from the surrounding watersheds. A significant amount of pollutants is discharged into the lake basin in various forms, leading to their accumulation within the lake water.

Sediments, as a crucial component of lake water systems⁴, play a key role in the accumulation and enrichment of various pollutants, with the concentrations of heavy metals significantly exceeding local background levels⁵. Heavy metal pollution in lake sediments is persistent, severely affecting the health and diversity of aquatic organisms. The release of accumulated heavy metals from sediments into the water body causes secondary pollution of the lake’s water resources⁶. Additionally, processes such as food chain magnification and biological accumulation can impact human health, thus significantly restricting the healthy functioning of lake ecosystems⁷. As a primary accumulation zone for heavy metal pollution, lake sediments have become a focal point for water resource researchers and scholars. Common research methods include the single pollution index, comprehensive pollution index, Nemero index method, enrichment factor method, and geoaccumulation index method⁸. Bayram Yüksel³ and others have evaluated pollution levels in sediments using enrichment factors, pollution indices, and pollution load indices, while also assessing the ecotoxicological risks of Simenit Lake based on the potential ecological risk index. Proshad et al.⁹ conducted a source analysis of surface sediment distribution in the Rupsa River in Bangladesh using the enrichment factor method. Wijesiri et al.¹⁰ studied the migration and accumulation processes of heavy metals between sediments and water bodies in urban rivers. Sun Bowen¹¹, Zhang Chuang¹², and Wang Henian¹³ respectively evaluated the distribution characteristics of heavy metal concentrations in lake sediments, including Baiyangdian, Hengshui Lake, a city lake in Jiangxi, and Erhai Lake, using the single pollution index method, Nemero pollution index method, and pollution load index. Additionally, scholars have applied methods such as the geoaccumulation index, enrichment factor method, and potential ecological risk index to analyze heavy metal pollution risks and sources in sediment and wetland ecosystems, including those in urban lakes of Ezhou, Hubei¹⁴, the Daqaidam Salt Lake¹⁵, urban rivers in Jiangsu¹⁶, Miliç Wetland¹⁷, and the Değirmendere River Basin and Estuary¹⁸. Fikret Ustaoglu et al.¹⁹conducted a systematic study on the potential toxicological elements in the surface sediments of the Abdal River system. They assessed the environmental risk levels by calculating ecological risk and pollution indices and conducted a systematic source analysis using multivariate statistical methods such as Pearson correlation analysis (PCC), principal component analysis (PCA), and hierarchical cluster analysis (HCA). However, most of the studies are limited to the statistical characteristics of heavy metals and risk assessment or are only capable of qualitatively identifying major pollutants and their sources, without providing quantitative evaluations of the specific contributions of each pollutant²⁰.

Dongping Lake, located at the junction of Henan and Shandong provinces, is a typical shallow inland lake in the eastern plains of China. It also serves as an important flood retention area within the Yellow River Basin and as a regulating lake for the Eastern Route of the South-to-North Water Diversion Project. The heavy metal characteristics of its sediments have been a subject of significant attention among researchers. Scholars such as Ge Huimin²¹, Zhang Ju²², and Ai Liuhuan²³ have evaluated the distribution characteristics and ecological risks of heavy metals in the lake sediments. However, most studies have focused on the distribution characteristics of heavy metals in the sediments and the ecological health risks, with relatively few sampling points. This study integrates the enrichment factor, geoaccumulation index, and potential ecological risk assessment methods to analyze the spatial distribution characteristics of eight typical heavy metals in the sediments of Dongping Lake. At the same time, cluster analysis, correlation analysis, principal component analysis (PCA), and Positive Matrix Factorization (PMF) were employed for a qualitative analysis of the enrichment mechanisms of these metals. Additionally, a quantitative evaluation of the pollution contribution values of the typical heavy metals was conducted. The results of this study provide data support for the prevention and control of heavy metal pollution in Dongping Lake, the formulation of monitoring measures, and the management of ecological risks in the watershed.

Materials and methods

Study area

Dongping Lake is located in the western part of Dongping County, Tai’an City, Shandong Province, at the mouth of the Da Wen River where it enters the Yellow River. The coordinates are between 116°00′–116°30′E and 35°30′–36°20′N, as shown in Fig. 1. To the west, it is close to the Beijing-Hangzhou Grand Canal, to the east it connects to the Daqing River, and to the north it flows into the Yellow River. The lake has a year-round water storage area of approximately 209 km², making it the second-largest freshwater lake in Shandong Province and a typical shallow lake in eastern China²⁴. Dongping Lake serves as one of the important water transmission and regulation reservoirs for the Eastern Route of the South-to-North Water Diversion Project. The lake spans across Dongping, Liangshan, and Wenshang counties, forming a large, multi-year regulated, gourd-shaped shallow water lake. The northern section consists of the narrow Xiao Qing River, while the southern section is a wide body of water. The water system is well-developed, and human activities in the region have a history of 4,500 years. The lake plays a comprehensive role in flood retention, drought resistance, water storage, water diversion, as well as navigation and tourism. Its water quality is of direct significance to the implementation of the South-to-North Water Diversion Project, as well as to the ecological environment and economic development along its route. Therefore, it is a crucial node for ecological security in Shandong Province.

Fig. 1

Geological conditions and sampling site distribution of the study area. Note: The map of China uses the review number GS(2024)0650, with no modifications to the base map. The imagery is derived from ArcGIS 10.8 online imagery. The DEM (Digital Elevation Model) elevation data was downloaded from the Geospatial Data Cloud (http://www.gscloud.cn/), with a resolution of 30 m × 30 m. The data for urban areas, residential districts, and industrial and mining land uses is sourced from the Chinese Academy of Sciences Resource and Environmental Science and Data Center, specifically the 30-meter resolution land use grid data (2020). This dataset employs a two-level classification system: the first level categorizes land into six types, primarily based on land resources and utilization attributes, including cultivated land, forest land, grassland, water bodies, construction land, and unused land. The second level further classifies these into 23 types based on the natural attributes of the land resources, with urban areas, residential districts, and industrial and mining land assigned the codes 51, 52, and 53, respectively.

Sample collection and analysis

Sample collection

According to the “Lake Sediment Survey Standards” and the “Technical Guidelines for the Evaluation of Heavy Metal Pollution in Sediments” (DB37/T4471-2021), sampling points were selected to avoid areas where hydrodynamic changes are intense (such as those subject to severe sediment scouring caused by sluice gate discharge) or areas recently disturbed by significant human activities (such as external soil input). The sampling points were distributed comprehensively based on the morphology of the Dongping Lake watershed and the boundaries of the protected areas, following a grid-based sampling approach.

To ensure the accuracy and reliability of trace element analysis data in environmental samples, this study established a stringent pollution control system. The SR6 Pro RTK GPS system was used for centimeter-level precision in geographic coordinate positioning and recording. A stainless steel grab sampler was employed to collect surface (0–20 cm) sediment samples. To minimize sampling errors and cross-contamination risks, several quality control measures were implemented, including the collection of three independent parallel samples at each sampling point followed by onsite homogenization. Care was taken to avoid sampling potential contamination zones at the interface of the sampler, and the samplers were rinsed with ultra-pure water prior to use. Additionally, representative samples weighing 500 ± 10 g were placed in polyethylene sample bags and immediately transferred to a 4 °C constant temperature storage box to ensure that the samples were transported under cold chain conditions (≤ 6 h) to the laboratory. Based on a systematic grid sampling approach, 20 sediment sampling points were set within the target area (spatial distribution shown in Fig. 1). All samples were collected by June 14, 2024, and detailed records of the sampling process, including GPS coordinates, sampling depth, onsite temperature, and other metadata, were kept in sampling logs to ensure the traceability of experimental data.

Analytical testing

In this sample analysis, the determination of Zn, Cu, Ni, Cd, and Pb was carried out using closed acid digestion coupled with Inductively Coupled Plasma Mass Spectrometry (ICP-MS). A Thermo Scientific Inductively Coupled Plasma Mass Spectrometer was used for the analysis. The detection limits for each element are as follows: Zn 0.5 mg/kg, Cu 1.0 mg/kg, Ni 2.0 mg/kg, Cd 0.01 mg/kg, and Pb 1.0 mg/kg. The analytical methods were conducted in accordance with the standards GB/T 14506.30–2010 “Chemical Analysis Methods for Silicate Rocks” and DZ/T 0279.1–2016 “Geochemical Sample Analysis Methods for Regional Geochemistry.”

The determination of Cr was performed using powder pellet X-ray fluorescence spectroscopy (XRF). Analysis was conducted using a PANalytical Axios X-ray fluorescence spectrometer, with a detection limit of 0.01 mg/kg. The analytical procedure strictly followed the requirements outlined in the standard DZ/T 0279.13–2016 “Geochemical Sample Analysis Methods for Regional Geochemistry.”

The determination of As was performed using hydride generation atomic fluorescence spectroscopy (HG-AFS). Analysis was carried out using a Haiguang AFS-830 atomic fluorescence spectrophotometer, with a method detection limit of 0.2 mg/kg. The analytical procedure was conducted in accordance with the standard DZ/T 0279.17–2016 “Geochemical Sample Analysis Methods for Regional Geochemistry.”

The determination of Hg was performed using cold vapor atomic fluorescence spectroscopy (CV-AFS) with a Haiguang AFS-830 atomic fluorescence spectrophotometer. The method detection limit was 0.05 mg/kg. The analytical procedure strictly adhered to the requirements outlined in the standard DZ/T 0279.17–2016 “Geochemical Sample Analysis Methods for Regional Geochemistry.”

During the laboratory analysis process, to ensure the accuracy of the analytical results and minimize analytical errors, 20% of the samples were randomly selected for parallel analysis. Additionally, certified reference materials for watercourse sediment composition were used for quality control. The results of the heavy metal element determinations showed that the relative standard deviations for all elements were less than 3%, meeting the precision requirements specified in the relevant standards.

Source apportionment methods for sediment heavy metals

In source research, methods such as cluster analysis, correlation analysis, principal component analysis (PCA), and positive matrix factorization (PMF) were employed. Cluster analysis and correlation analysis were used to reveal the relationships between heavy metal elements, while PCA was applied for data dimensionality reduction and the extraction of main pollution sources. PMF, through non-negative matrix factorization, was used to analyze pollution sources and their contributions, addressing the interpretational limitations of PCA^25,26,27。.

Cluster analysis and correlation analysis methods

Cluster analysis is a method that groups a set of objects based on selected features or attributes, ensuring that objects within the same cluster (Cluster) are as similar as possible, while objects in different clusters are as dissimilar as possible. Correlation analysis, on the other hand, is used to measure the strength of the linear relationship between two or more variables, typically expressed in the form of correlation coefficients. In this study, the Pearson Correlation Coefficient was applied for statistical analysis of the correlations between heavy metal elements in the sediment of Dongping Lake.

Principal component analysis

The PCA model plays a crucial role in the evaluation and analysis of heavy metals in lake sediment. It performs exceptionally well in data dimensionality reduction by extracting the main variances in the data, eliminating redundant information, and transforming multiple correlated variables into a few independent composite variables. This simplification of the data structure enhances the intuitiveness and comprehensibility of the analysis. PCA enables the identification and extraction of these correlations, revealing the potential sources and distribution patterns of heavy metal pollution. By retaining the most significant data features, it improves the efficiency and accuracy of the analysis.

The calculation of the correlation coefficient R involves the eigenvalues λ₁ ≥ λ₂ ≥ … ≥ λ_m, with a₁, a₂, …, a_n being the corresponding eigenvectors, where aj = (a_1j, a_2j, …, a_nj)T. The indicator variables consist of m sample numbers composed of the eigenvectors.

$$\:\left\{\begin{array}{c}{\text{F}}_{1}={a}_{11}{\text{X}}_{1}+\:{a}_{21}{\text{X}}_{2}+\dots\:+{a}_{\text{n}1}{\text{X}}_{\text{n}}\\\:{\text{F}}_{2}={a}_{12}{\text{X}}_{1}+\:{a}_{22}{\text{X}}_{2}+\dots\:+{a}_{\text{n}2}{\text{X}}_{\text{n}}\\\:\dots\:\dots\:.\\\:{\:\:\:\:\text{F}}_{\text{m}}={a}_{1\text{m}}{\text{X}}_{1}+\:{a}_{2\text{m}}{\text{X}}_{2}+\dots\:+{a}_{\text{n}\text{m}}{\text{X}}_{\text{n}}\end{array}\right.$$

(1)

In the equation, F₁ represents the first principal component, F₂ represents the second principal component, and F_m represents the m-th principal component.

Positive matrix factorization

The PMF model is based on iterative operations that directly compare the original data, enabling source apportionment of key components. It is widely applied in the fields of environmental science and data analysis, and has also been recognized by experts and scholars in ecological environmental assessment, including the analysis of soil, sediment, and air in ecological studies. Its advantage lies in the non-negative constraints imposed on the factor loadings and scores of heavy metals during matrix decomposition, which corrects data bias^25,26, providing interpretability and clear physical meaning to the decomposition results. It simultaneously identifies the source profiles and their contributions, addressing the limitations of PCA in terms of interpretability and its inability to retain all significant information²⁷. Therefore, in this study on heavy metals in Dongping Lake sediments, the PMF model was employed to quantify the sources of eight heavy metal elements. The basic principle involves using the least squares method to perform iterative operations on the original data²⁸, while simultaneously determining the pollution source contribution matrix A_ik, the source profile matrix B_kj, and a residual matrix e_ij, with the calculation formula as follows:

$$\:{X}_{ij}={\sum\:}_{k=1}^{p}\:{A}_{ij}{B}_{ij}+{e}_{ij}$$

(2)

In the equation, X_ij represents the concentration (mg/kg) of heavy metal element j in the i-th sediment sample; A_ij is the concentration of heavy metal element j in pollution source k; B_ij represents the contribution of pollution source k to the i-th sediment sample; e_ij is the residual matrix²⁹. This formula is computed using the least squares method in iterative steps to minimize the objective function Q, with the calculation process as follows:

$$\:Q={\sum\:}_{i=1}^{n}\:{\sum\:}_{j=1}^{m}\:{\left(\frac{{X}_{ij}-{\sum\:}_{k=1}^{p}\:{A}_{ik}{B}_{kj}}{{u}_{ij}}\right)}^{2}$$

(3)

In the equation, $\:{u}_{ij}$ represents the uncertainty of heavy metal j in sediment sample i.

The main advantage of the PMF model lies in the application of non-negative constraints, which ensure that the solution has clear physical meaning, allowing the extracted factors to directly correspond to actual pollution sources or natural processes. Additionally, the model is flexible in handling measurement uncertainties in the data and is capable of addressing nonlinear relationships and complex mixing issues. The calculation of uncertainty primarily depends on whether the detected concentrations of heavy metal elements at sampling points are above or below the laboratory detection limit for each species. The calculation formula is as follows:

$$\:\begin{array}{c}c\leqslant\text{M}\text{D}\text{L},{u}_{ij}=5/6\times\:\text{M}\text{D}\text{L}\\\:c>\text{M}\text{D}\text{L},{u}_{ij}=\sqrt{\left((\text{E}\text{F}\times\:c{)}^{2}+(0.5\times\:\text{M}\text{D}\text{L}{)}^{2}\right.}\end{array}$$

(4)

In the equation, c represents the concentration of the element in the sample; MDL is the method detection limit; and EF is the relative standard deviation.

Heavy metal source apportionment and ecological risk assessment

Three methods were employed: the enrichment factor method, the geoaccumulation index, and the potential ecological risk assessment method. The enrichment factor method determines the degree of heavy metal enrichment by comparing the ratio to environmental background values³⁰, typically using aluminum as the reference element. The geoaccumulation index quantifies the pollution level by comparing the measured values with the background values, with a correction factor introduced for adjustments^28,31. The potential ecological risk index method combines the concentration of heavy metal elements, ecological environment, and toxicity to categorize the degree of sediment heavy metal pollution and its potential ecological risks¹³. The classification of $\:EF$、$\:{I}_{geo}, RI$ is shown in Table 1.

Enrichment factor method

The Enrichment Factor (EF) method is a technique used to determine the degree of heavy metal enrichment in sediments by comparing the ratio of total heavy metal concentration to the regional environmental background values³⁰. To minimize the impact of sediment particle size on heavy metal concentrations during the calculation process, a reference element is selected to normalize the data before calculating the EF value. Aluminum (Al), primarily derived from natural soil-forming processes and being immobile in the soil, is commonly used as the reference element. The calculation formula for EF is as follows:

$$\:EF=\frac{{({C}_{i}/Al)}_{sediment}}{{({C}_{i}/Al)}_{background}}$$

(5)

In Eq. (1), EF represents the enrichment factor of a single heavy metal, $\:{({C}_{i}/Al)}_{sediment}$ is the ratio of the mass fraction of the heavy metal element to the mass fraction of Al in the sediment, and $\:{({C}_{i}/Al)}_{background}$ is the ratio of the mass fraction of the heavy metal element to the mass fraction of Al in the soil background.

Geoaccumulation index

The Geoaccumulation Index (I_geo) was proposed in 1979 by the renowned German scientist Muller as a quantitative method to assess the potential heavy metal pollution levels in soil and sediments. It has been widely used to evaluate heavy metal pollution in aquatic environments and sediments^28,31. The calculation formula for the Geoaccumulation Index is as follows:

$$\:{I}_{geo}=\text{log}\frac{{C}_{i}}{k\times\:{B}_{i}}$$

(6)

In Eq. (2), $\:{I}_{geo}$represents the geoaccumulation index; $\:{C}_{i}$ is the measured concentration of the heavy metal element in the sediment; $\:{B}_{i}$ is the background value of the heavy metal element in the soil; and k is the correction factor, typically set to 1.5. The classification of $\:{I}_{geo}$ is shown in Table 1.

Potential ecological risk assessment

The Potential Ecological Risk Index method combines the concentration of heavy metal elements, ecological environment, and toxicity to categorize the degree of sediment heavy metal pollution and its potential ecological risks¹³. The calculation formula for the Potential Ecological Risk Index is as follows:

$$\:RI=\sum\:_{i=1}^{n}{E}_{r}^{i}=\sum\:_{i=1}^{n}({T}_{i}\times\:\frac{{C}_{i}}{{C}_{n}^{i}})$$

(7)

In Eq. (3), RI represents the potential ecological risk index; $\:{E}_{r}^{i}$ is the potential ecological risk index for heavy metal element i; $\:{T}_{i}$ is the toxicity weighting factor for heavy metal i. The toxicity coefficients for eight heavy metal elements are Zn = 1, Cr = 2, Cu = Ni = Pb = 5, As = 10, Cd = 30, and Hg = 40; $\:{C}_{i}$ is the measured concentration of the i-th heavy metal in the sediment; and $\:{C}_{n}^{i}$ is the background concentration of the i-th heavy metal in the soil of the study area³². The classification of $\:EF$、$\:{I}_{geo}\:and\:RI$ is shown in Table 1.

Table 1 The classification criteria for EF, I_geo, and RI.

Data processing and graphical representation

The data processing and map preparation in this study primarily utilized the ArcGIS 10.7 geostatistical analysis module to generate a regional location map of the study area. The Kriging spatial interpolation method was employed to create spatial distribution maps of sediment heavy metal elements and potential ecological risk assessment maps. IBM SPSS Statistics 26 was used for statistical analysis of sediment heavy metal element concentrations, cluster analysis, principal component analysis (PCA), and potential ecological risk assessment statistical results. The Pearson correlation coefficient was applied to investigate the relationships between various heavy metal element indicators. Additionally, Origin 2022 was used to generate box plots for the Geoaccumulation Index (Igeo), Enrichment Factor (EF), and sediment heavy metal source contribution maps.

Results and discussion

Statistical characteristics of heavy metals in sediments

Concentration characteristics of heavy metals

In the 0–50 cm depth sediment samples from 20 sampling points, the mass fractions of the eight heavy metal elements As, Zn, Cu, Ni, Cd, Hg, Pb, and Cr were found to be in the ranges of 9.17 ~ 27.8 mg/kg, 72.5–204.0 mg/kg, 18.9 ~ 53.9 mg/kg, 23.3 ~ 38.0 mg/kg, 0.081 ~ 0.3 mg/kg, 0.018 ~ 0.055 mg/kg, 19.0 ~ 34.8 mg/kg, and 40.0 ~ 135.0 mg/kg, respectively (Table 2).

Table 2 Contents of 8 kinds of heavy metal elements in various sampling sites in Dongping lake.

The statistical characteristics of the eight heavy metal elements in the sediment samples from Dongping Lake were analyzed using IBM SPSS Statistics 26 (see Table 3). The results indicated that the concentration values of the heavy metal elements, or their transformed values, approximately followed a normal distribution. Therefore, the mean concentrations meet the characterization requirements for statistical analysis.

The “Soil Element Background Values of Shandong Province,” published by the China Geological Environmental Monitoring Station in 1990, was determined based on systematic sampling across the province and laboratory analysis. This dataset represents the natural elemental content of the major soil types in Shandong Province. Although the data is relatively old, it remains an important reference for determining regional environmental baseline conditions, as it represents the first systematic study of soil background values in Shandong Province. The current “Soil Environmental Quality-Soil Pollution Risk Control Standard for Agricultural Land” (GB 15618 − 2018) also references such background data for risk screening values. Therefore, in this study, the heavy metal content from these background values is used as the regional environmental baseline.

The analysis results (Fig. 2) indicate that the average concentrations of the eight heavy metals in the sediment of Dongping Lake were 19.21, 107.32, 32.87, 29.83, 0.20, 0.04, 27.47, and 70.32 mg/kg, significantly higher than the soil background values of Shandong Province. For example, the concentrations of elements such as As (19.21 vs. 8.7), Zn (107.32 vs. 40.0), and Cd (0.20 vs. 0.078) were notably above the regional background levels. Further analysis revealed that the average concentrations of As, Zn, Cu, Ni, Cd, Hg, Pb, and Cr were 2.21, 2.68, 1.51, 1.49, 2.56, 2.50, 1.13, and 1.08 times their respective background values, with the maximum concentrations of Cd, Hg, and As even exceeding three times the background values. Moreover, except for Cu, Pb, and Cr, which had exceedance rates of 95%, 80%, and 50%, respectively, the exceedance rates for the other five heavy metals were all 100%. This result not only corroborates the typical phenomenon of sedimentary deposition and the long-term accumulation of heavy metals in soil erosion but also highlights the ongoing pressure on the ecological environment of Dongping Lake and the potential risks it faces.

Fig. 2

Created by Origin Pro 2025 (Software accessible at https://www.originlab.com/).

Statistical Box Plots of Heavy Metals in Dongping Lake Sediments.

This study constructs the trend and variation graphs (Fig. 3; Table 3) of heavy metal concentrations in the sediment of Dongping Lake from 1990 to 2024. The analysis results show that the concentrations of arsenic (As) and zinc (Zn) have been increasing year by year, closely associated with the intensification of industrial activities and rapid urbanization. The concentrations of chromium (Cr) and lead (Pb) decreased between 1990 and 2009, but showed a significant rebound in 2014 and 2024, indicating the potential presence of new pollution sources or further effects of heavy metal enrichment under natural background conditions in the Dongping Lake and its surrounding river areas. The concentrations of nickel (Ni) and mercury (Hg) displayed a pattern of increase followed by decrease, reflecting the significant environmental remediation efforts in Dongping Lake in recent years. The copper (Cu) concentration exhibited a more complex trend, first rising, then declining, and slightly increasing again, which may be closely related to the lake’s dynamic changes and fluctuations in human activities. These trends suggest that, although the overall concentration of heavy metals in the sediment of Dongping Lake is stabilizing, the lake, as a sedimentary discharge area, continues to be affected by the long-term accumulation of heavy metals. The ecological environment is still facing severe challenges. Therefore, further analysis of the sources of heavy metal pollution and their potential ecological risks in Dongping Lake is urgently needed to provide scientific evidence for comprehensive water management and ecological environmental protection.

Fig. 3

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Heavy metal concentrations and temporal trends in Dongping lake sediments.

Table 3 Statistical summary of heavy metals in Dongping lake sediments.

Spatial distribution characteristics of heavy metals in sediments

The coefficient of variation (CV) is a statistical measure used to assess the degree of dispersion in the heavy metal concentration data from the sediment of Dongping Lake, indirectly revealing the spatial distribution characteristics of different heavy metals. Some studies have suggested that when the CV exceeds 20%, human activities become the primary driver of spatial variability in sediment heavy metal distribution^36,37,38,39. As shown in Table 2, the CV of the eight heavy metals in Dongping Lake sediment indicates that only Ni and Pb have values below 20%, while the CVs for the remaining metals exceed 20%, with Zn and Cr having CVs greater than 30%, representing moderate variation². This suggests that the spatial distribution of As, Zn, Cu, Cd, Hg, and Cr concentrations is heavily influenced by anthropogenic activities or hydrodynamic disturbances, which are the main driving factors.

The spatial distribution map of heavy metals (Fig. 4) reveals significant spatial variability. The spatial distribution characteristics of arsenic (As), mercury (Hg), copper (Cu), and lead (Pb) show remarkable consistency, with high-concentration areas mainly concentrated in the central, central-western, and southern regions of Dongping Lake. This is primarily related to the following factors: firstly, the inflow of Liuchang River water contributes to the accumulation of heavy metals in the sediment, leading to the deposition of heavy metals in the central-western and southern regions of the lake. Secondly, frequent human activities in the southern region, including agriculture, livestock farming, and industrial and mining activities, serve as one of the significant sources of heavy metal pollution in Dongping Lake. Additionally, the high concentrations in the central region may be associated with the former residential area on the island in the lake and the gentle flow of water, resulting in long-term accumulation of heavy metal pollutants. On the other hand, the lower concentrations in the Dawen River inlet and the western region indicate that the water entering the lake from the Dawen River contains relatively low levels of As, Hg, Cu, and Pb. The high-concentration areas of zinc (Zn), nickel (Ni), and lead (Pb) are mainly concentrated in the central and southwestern regions of Dongping Lake, suggesting that traffic routes and agricultural non-point source pollution in the southern area are the primary contributors to the heavy metal pollution load in this region^40,41. Meanwhile, a distinct low-concentration zone is observed near the inlet of the South-to-North Water Diversion project, indicating that the water diversion process and its disturbance effects have a certain flushing and desorption effect on the heavy metals in the sediment. Additionally, the northern and eastern regions mainly consist of low-concentration zones for heavy metals. This is closely related to the relatively thin Quaternary sedimentary layers in the northeast region, as well as the strong scouring effect of lake water, leading to faster sediment turnover. The spatial distribution characteristics of cadmium (Cd) and chromium (Cr) concentrations exhibit significant heterogeneity. The high-concentration zone of Cd is primarily concentrated in the northern and western regions, while the high-concentration zone of Cr is mainly concentrated at the Dawen River inlet, suggesting the heterogeneity of pollution sources for these two heavy metals. The Cd pollution load is mainly attributed to agricultural pollution, while Cr may primarily be influenced by factors such as soil leaching, industrial wastewater, and domestic sewage discharge along the Dawen River. Overall, the concentrations of heavy metals in the central and southern regions of Dongping Lake are generally higher, mainly due to human activities, industrial emissions, or the influence of Liuchang River. The lower concentrations in the northern and eastern regions can be attributed to relatively less human interference and the dilution effect of Dawen River water. The higher concentrations in the central region are associated with factors such as the former residential area on the island in the lake, slow water flow, and severe algal accumulation.

Fig. 4

Generated by ArcGIS 10.8 (Software available at https://www.arcgis.net.cn/).

Spatial distribution of eight heavy metals in Dongping lake sediments.

Source apportionment of heavy metals in sediments

Cluster analysis

Cluster analysis groups heavy metal elements with similar sources into the same category, distinguishing elements with different pollution load sources. The closer the clustering distance, the more similar the sources of the different heavy metals⁴⁰. To explore the correlations among the heavy metal elements in the sediment of Dongping Lake, a cluster analysis was performed on the raw heavy metal data (Fig. 5). The results of this cluster analysis indicate that the heavy metal elements in the sediment of the study area can be divided into three categories:

Class I: This can be further divided into two subcategories. The first subcategory consists of arsenic (As), copper (Cu), and mercury (Hg). These three heavy metals have similar clustering distances, corresponding with the spatial distribution characteristics, thereby indicating their homogeneity. Mercury (Hg) is primarily considered to have increased due to coal combustion for heating, which releases Hg into the environment. This Hg then settles onto the surrounding soil via atmospheric deposition⁴², with rainwater washing and agricultural irrigation contributing to its accumulation in Dongping Lake. Arsenic (As) is typically regarded as primarily derived from sources such as metal smelting, pesticides, fertilizers, and livestock wastewater³⁷, indicating that the pollution sources in this category are mainly industrial emissions, domestic sewage, and agricultural non-point sources from nearby villages; The other subcategory consists of zinc (Zn), nickel (Ni), and lead (Pb). These three heavy metals have similar clustering distances, which also correspond with the previously assessed results, indicating their homogeneity. Zinc (Zn) and lead (Pb) are considered to be traffic-related pollutants^40,41, with the combustion of leaded gasoline being a significant source of Pb in automobile exhaust emissions⁴³. The accumulation of these elements is closely associated with traffic volume. The exhaust gases from motor vehicles, as well as the wear and tear from brakes and tires, all contribute to the accumulation of Zn in the ecological environment⁴⁴. Nickel (Ni) is primarily influenced by industries such as metal processing and chemicals⁴⁵. This suggests that this category is primarily influenced by a combination of traffic pollution, industrial pollution, and agricultural activities.

Class II: Cadmium (Cd). Cd is generally considered to originate from agricultural activities, such as the use of phosphate fertilizers, manure, and pesticides, making it a typical element that enters the ecological environment due to human activities. According to the spatial distribution map of Cd, the high-concentration zones are primarily located in areas with high agricultural and livestock farming activities^45,46. This indicates that this category is mainly influenced by agricultural activities.

Class III: Chromium (Cr). The correlation of Cr with the other seven heavy metal ions is relatively weak, which is consistent with the spatial distribution characteristics observed earlier. Additionally, the exceedance rate of Cr is low, suggesting that the presence of Cr is primarily related to the soil background in the areas through which the Dawen River flows.

Fig. 5

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Cluster analysis of heavy metals in Dongping lake sediments.

Correlation analysis

The correlation analysis of heavy metals in the sediment of Dongping Lake is shown in Fig. 6. Arsenic (As) is highly significantly (P ≤ 0.001) positively correlated with copper (Cu) and mercury (Hg), and Cu is also highly significantly positively correlated with Hg, indicating that As, Cu, and Hg in the sediment share the same source. Zinc (Zn) is highly significantly positively correlated with nickel (Ni) and lead (Pb), and Ni is similarly highly significantly positively correlated with Pb, suggesting that Zn, Ni, and Pb in the sediment have the same source. Mercury (Hg) is highly significantly positively correlated with Pb, indicating that Hg and Pb in the sediment have the same source. Chromium (Cr) is only highly significantly negatively correlated with cadmium (Cd), suggesting an antagonistic relationship between Cr and Cd. The correlation between Cr and other heavy metals is weak, and the average enrichment factor of Cr is 1.01, further suggesting that the source of Cr is primarily related to the soil parent material and pedogenesis in the areas through which the Dawen River flows.

Fig. 6

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Correlations among heavy metals in Dongping lake sediments.

Principal component analysis

Principal Component Analysis (PCA) is a dimensionality reduction method used to identify a few composite variables that can replace the original numerous variables, ensuring that these composite variables represent the information of the original variables as effectively as possible. It is an effective method for analyzing the sources of soil heavy metals⁴⁷. The PCA results of heavy metals in the sediment of Dongping Lake are shown in Table 4, exploring the potential pollution sources of heavy metals. Prior to this, to ensure the reliability and representativeness of the data, Kaiser-Meyer-Olkin (KMO) and Bartlett’s test of sphericity were conducted⁴⁸. The KMO statistic was 0.688, which is generally considered acceptable for factor analysis when the KMO value exceeds 0.5. The significance level of Bartlett’s test of sphericity was 0.000, indicating that the sediment data were suitable for factor analysis²⁰. The maximum variance method was used to rotate the component matrix, selecting factors with initial eigenvalues greater than 1 as the main influencing factors. In this study, three principal components (PC1, PC2, and PC3) were extracted, with the eigenvalue of PC3 equaling 1. Although the eigenvalue of PC3 was at the critical value, it significantly improved the cumulative variance contribution rate (from 76.76 to 89.22%), reaching the 80% threshold required in the field, and thus it was retained. As shown in Table 5, the cumulative contribution rate of the principal component analysis reached 89.22%, which can adequately explain the information of the original variables.

Table 4 Principal component analysis (PCA) of heavy metals in Dongping lake sediments.

The first principal component (PC1) contributes 54.62%, with the major loadings from arsenic (As) (0.88), copper (Cu) (0.85), and mercury (Hg) (0.87). The spatial distribution of these three heavy metals in relatively high concentration areas is similar, showing a strong correlation, which is consistent with the results of the previous correlation analysis. Therefore, it is inferred that As, Cu, and Hg share the same source. According to the earlier statistical analysis, their spatial distribution is similar, with significant correlation and a high coefficient of variation, indicating a prominent influence of human activities. The high-concentration areas of As, Cu, and Hg are concentrated in the central and southern parts of Dongping Lake. As and Cu primarily originate from mineral smelting and arsenic-containing herbicides or pesticides⁴⁹, while Hg mainly originates from coal combustion, aligning with the spatial distribution of local iron ore resources and agricultural and livestock activities. Combining the geochemical characteristics of the elements, it is concluded that PC1 represents a mixed industrial and agricultural pollution source.

The second principal component (PC2) contributes 22.14%, with the major loadings from zinc (Zn) (0.95), nickel (Ni) (0.83), and lead (Pb) (0.71). The spatial distribution of these three metals is similar, all exhibiting high concentrations in the southwestern region. Zn primarily originates from automobile exhaust, tire wear, and industrial emissions⁵⁰, while Ni is associated with traffic and metal processing activities. Pb, according to the soil background values in Shandong Province, has a relatively high natural background level, with its current concentration being close to the background value, indicating that Pb is a natural source. The study area is located near national and provincial highways and is a fishery zone, where traffic and fishing boat emissions of metal dust enter the lake through atmospheric deposition and runoff, leading to the accumulation of heavy metals in the sediment. Therefore, PC2 represents both traffic and natural sources.

The third principal component (PC3) contributes 12.46% and is primarily associated with cadmium (Cd), which is a typical indicator of agricultural pollution⁵⁰. Studies have shown that its sources include high cadmium content in agricultural fertilizers (such as phosphate fertilizers), improper application of livestock manure, and feed additives used in aquaculture^51,52. Therefore, PC3 can be identified as an agricultural source of pollution.

The results of the principal component analysis indicate that heavy metal pollution in the study area mainly originates from three types of activities: industrial and mining activities, traffic emissions, and agricultural activities. The industrial and agricultural mixed source is characterized by arsenic (As), copper (Cu), and mercury (Hg), with its spatial distribution highly corresponding to regional mineral resource development and the distribution of agricultural and pastoral activities. The traffic source is represented by zinc (Zn), nickel (Ni), and lead (Pb), with high-concentration areas aligning with the locations of major transportation routes. The agricultural source is primarily characterized by cadmium (Cd) pollution, which is closely linked to fertilizer use and livestock farming activities. The spatial distribution characteristics of these three pollution sources show a significant correspondence with the regional industrial layout, and the model analysis results effectively explain the composition of heavy metal pollution sources in the study area.

PMF source apportionment

Quantitative source analysis of various heavy metal elements in the study area was conducted using PMF, following the guidance of the EPA PMF 5.0 software manual. The content data and uncertainty data were input as required, and the metals were classified based on the signal-to-noise ratio (S/N) into the “Strong” category. The number of factors was set to range from 3 to 5, and 20 iterations were performed. By comparing the QRobust/Qtrue values under different numbers of factors, it was found that when the number of factors was set to 3, the QRobust/Qtrue value was consistently 17.4, with all element residuals falling within the [−3, 3] range and showing a normal distribution, indicating the best fit and stable calculation results. The source analysis conducted using this model showed that the R² values for arsenic (As), zinc (Zn), copper (Cu), nickel (Ni), cadmium (Cd), mercury (Hg), lead (Pb), and chromium (Cr) were 0.75, 0.69, 0.92, 0.94, 0.92, 0.86, 0.96, and 0.63, respectively. This suggests that the model’s source analysis is effective and can adequately explain the information contained in the original data.

Principal component analysis indicates that the eight heavy metal ions in the sediment of the study area primarily originate from three pollution sources. The sources of arsenic (As), copper (Cu), and mercury (Hg) are mainly associated with industrial and agricultural activities, and the factor loadings show that Factor 1 contributes the most to As, Cu, and Hg. Therefore, Factor 1 is identified as the industrial and agricultural source. Factor 2 contributes the most to chromium (Cr), which, as analyzed earlier, primarily originates from natural background sources, so Factor 2 is classified as the natural source. Cadmium (Cd) is mainly sourced from agricultural activities, and since Factor 3 contributes the most to Cd, Factor 3 is designated as the agricultural source.

According to the heavy metal pollution source composition spectrum of the sediment (Fig. 7), Factor 1 (industrial and agricultural source) contributes the highest percentage to As (45.78%), Cu (45.44%), and Hg (46.11%) (45%−46%), followed by lead (Pb) (36.72%) and chromium (Cr) (37.19%) (36%−37%). The contributions to zinc (Zn) (32.86%), nickel (Ni) (32.11%), and cadmium (Cd) (30.17%) are in the range of 30%−33%, with an overall average contribution rate of 38.30%. Factor 2 (natural source) shows a dominant contribution to Cr (62.81%) (> 60%), while the contributions to Zn (39.55%), Ni (37.28%), and Pb (34.01%) are about 34%−40%. The contributions to Cu (29.91%) are close to 30%, and those to As (23.59%) and Hg (20.44%) are lower (< 25%). The contribution to Cd (9.59%) is the lowest (< 10%), with an average contribution rate of 32.15%. Factor 3 (agricultural source) has a significantly higher contribution to Cd (60.24%) (> 60%), followed by As (30.63%), Ni (30.61%), and Pb (29.28%) at approximately 30%. The contribution to Hg (33.45%) is around 33%, while the contributions to Zn (27.59%) and Cu (24.66%) range from 24% to 28%. There is no contribution to Cr. The average contribution rate of Factor 3 is 29.56%.

Based on the contribution rates of the three pollution sources to heavy metal pollution in the sediment of Dongping Lake (Fig. 8), the contribution of natural parent material sources accounts for 32.1%, while human activities account for 67.9%. Among human activities, the industrial and agricultural source contributes 38.3%, and the agricultural source contributes 29.6%. Overall, the contribution of human activities to heavy metal pollution in the study area is much higher than that of the natural source. This study concludes that the sources of heavy metal elements in the study area are complex and diverse, indicating a mixed pollution scenario. As shown in Fig. 8, most heavy metal elements are affected to varying degrees by natural parent material and industrial activities. Except for chromium (Cr), other heavy metals are also influenced by agricultural sources to varying extents. It is inferred that the parent material of the soil determines the initial distribution and content of heavy metals in the surface soil, while long-term industrial and agricultural activities significantly alter the accumulation of heavy metals in the surface soil. Therefore, industrial and agricultural activities have become the primary sources of heavy metal pollution in the surface soil of the study area.

Fig. 7

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Heavy metal pollution source composition spectrum of Dongping lake sediment.

Fig. 8

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Contribution rates of different pollution sources to heavy metal contamination in Dongping lake sediment.

This study conducted an in-depth investigation into the sources of heavy metal pollution in the sediment of Dongping Lake through Principal Component Analysis (PCA) and Positive Matrix Factorization (PMF) source apportionment. First, PCA identified three primary pollution sources: industrial and agricultural mixed pollution, traffic pollution, and agricultural pollution. PC1 (industrial and agricultural source) is characterized by arsenic (As), copper (Cu), and mercury (Hg), reflecting the contributions of mineral smelting, agricultural activities, and coal combustion to heavy metal pollution. PC2 (traffic and natural sources) is primarily associated with the pollution of Zn, Ni, and Pb, with high-concentration areas of these elements being closely related to major transportation routes and high background values. PC3 (agricultural source) is closely related to cadmium (Cd) pollution, indicating that agricultural activities, especially the use of fertilizers and feed, have a significant impact on heavy metal accumulation. The cumulative contribution rate of PCA reached 89.22%, which adequately explains the primary sources of heavy metal pollution in the sediment.

Further analysis through PMF source apportionment confirmed the presence of the three pollution sources and provided a quantitative assessment of each source’s contribution. The results showed that industrial and agricultural activities contributed most significantly to As, Cu, and Hg, while agricultural activities had a notable impact on the accumulation of Cd. Chromium (Cr) was primarily attributed to natural background sources. PMF analysis also revealed the dominant role of human activity sources, especially industrial and agricultural sources, in heavy metal contamination, with the overall contribution rate of human activities being 67.9%, significantly higher than the natural source contribution of 32.1%. These findings suggest that the heavy metal pollution in Dongping Lake sediment primarily results from mixed pollution, with industrial and agricultural activities as the main sources.

In conclusion, the heavy metal pollution problem in Dongping Lake is complex, involving multiple sources, with human activities contributing far more to the pollution than natural factors. The study recommends targeted remediation measures for different pollution sources, particularly controlling pollution from industrial and agricultural activities, to reduce the negative impact of heavy metals on the ecological environment.

Enrichment characteristics of heavy metals in sediments

Enrichment characteristics of heavy metals

Based on the enrichment factor (EF) box plot of eight heavy metals in the Dongping Lake sediment (Fig. 9), the average enrichment factor values are as follows: Cd (2.35) > As (1.97) > Hg (1.92) > Zn (1.61) > Cu (1.30) > Ni (1.11) > Cr (1.01) > Pb (0.41). Among these, Pb does not exhibit significant enrichment, while other elements show varying degrees of enrichment influenced by human activities. Cd is moderately enriched, with its enrichment factor significantly higher than that of other heavy metals. Among the sampling points, 65% exhibit moderate enrichment, 30% show slight enrichment, and only 5% show no enrichment. As, Hg, Zn, Cu, Ni, and Cr are slightly enriched, with As, Hg, and Zn at some sampling points reaching moderate enrichment levels, accounting for 45%, 40%, and 20%, respectively. Moreover, Cr’s enrichment factor (EF) is close to 1, with only 30% of the points showing slight enrichment, suggesting that chromium contamination may primarily be influenced by the parent material of the soil during pedogenesis. Overall, the enrichment factor analysis indicates that, except for Pb, the heavy metals in Dongping Lake sediment are significantly affected by human activities. Cd exhibits moderate enrichment and is the most pronounced, while Cr contamination is mainly influenced by the natural pedogenic parent material with a lower contribution from human activities. The intertwined effects of human and natural backgrounds have shaped the distribution characteristics of heavy metals in the sediment of Dongping Lake.

Fig. 9

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Boxplots of enrichment factors (EF) for eight heavy metals in Dongping lake sediments.

Geoaccumulation index I _geo of heavy metals in sediments

The sediment heavy metal accumulation index (I_geo) was calculated, with the results shown in Fig. 10. The average I_geo values of the heavy metal elements, ranked from highest to lowest, are as follows: As > Zn > Cu > Ni = Hg > Cr > Cd > Pb. As and Zn have the highest I_geo values, averaging 1.11 and 1.02, respectively, indicating moderate pollution levels. Among all the sampling points, 95% of the points show moderate pollution for As, while 50% show moderate pollution for Zn. The average I_geo values of Cu, Ni, Hg, and Cr are close to 1, suggesting that they are between unpolluted and moderately polluted states. The proportion of unpolluted points for these four elements is 75%, 95%, 90%, and 85%, respectively, indicating that these metals are mainly influenced by geological background, though they are also subject to some degree of anthropogenic influence. Cd and Pb have relatively low I_geo values, at 0.79 and 0.72, respectively, placing them in the transitional zone between unpolluted and moderately polluted. However, 10% of the sampling points show signs of moderate pollution. In contrast, Pb does not show any signs of pollution in the sediment throughout Dongping Lake, which may result from the influence of the parent material in the sediment under natural conditions.In conclusion, As is the primary contaminant heavy metal in the sediment of Dongping Lake, followed by Zn. Cu, Ni, Hg, and Cr also show some degree of pollution, with a proportion of sampling points reaching moderate pollution levels, further confirming the contribution of human activities to the contamination of these heavy metals. The absence of pollution by Pb is primarily attributed to the natural geological background.

Fig. 10

Created by OriginPro 2025 (Software accessible at https://www.originlab.com/).

Boxplots of Geoaccumulation index for eight heavy metals in Dongping lake sediments..

Potential ecological risk assessment

The results of the potential ecological risk index ($\:{E}_{r}^{i}$) and potential ecological hazard index (RI) analysis for the Dongping Lake sediment are shown in Table 3. The average $\:{E}_{r}^{i}$ values for each heavy metal element are ranked as follows: Hg (80.53) > Cd (72.70) > As (20.65) > Cu (6.85) > Ni (5.78) > Pb (2.16) > Cr (2.13) > Zn (1.69). This ranking is consistent with the results of the 2019 study by Ai Liuhuan et al. on the distribution and risk assessment of heavy metals in Dongping Lake water and sediment. The average $\:{E}_{r}^{i}$ values for each heavy metal have decreased, indicating that the overall ecological environment of Dongping Lake sediment has improved under the influence of different policies and ecological protection measures. According to the standard for potential ecological risk of heavy metals in sediments (Table 2), the analysis of heavy metal pollution in Dongping Lake sediment shows that As, Zn, Cu, Ni, Pb, and Cr all have $\:{E}_{r}^{i}$ values less than 40, indicating light pollution. However, the variation coefficients for As, Zn, Cu, and Cr exceed 20%, suggesting that these four heavy metals are subject to a certain degree of anthropogenic influence. In contrast, Hg and Cd have significantly higher average values (80.53 and 72.70, respectively) compared to other elements, and all sampling points show moderate to strong pollution levels. The maximum values of Hg and Cd reach 107.14 and 115.79, respectively, with strong risk points accounting for 50% and 30%. The variation coefficients for Hg and Cd are relatively high (26.40 and 29.17, respectively), indicating significant fluctuations in their concentrations across different sampling points, with certain local areas potentially facing higher ecological risks. Furthermore, the contributions of Hg and Cd to the total ecological risk are 41.83% and 37.77%, respectively, far exceeding those of other elements, making them the primary contributors to the potential ecological risk in Dongping Lake sediment.

According to the RI results for the sediment in Dongping Lake (Table 6), only the sampling point DN06 exhibits a potential ecological hazard at a mild pollution level, while all other points show moderate pollution levels. Notably, the potential ecological hazard index (RI) at sampling points DN07 and DN17 is the highest, indicating that the heavy metal pollution in these areas is particularly severe. An in-depth spatial analysis of mercury (Hg), cadmium (Cd), and the overall RI is presented in Fig. 11. The high RI values for Hg are mainly concentrated in the central and southeastern parts of the lake, with these areas classified as strong risk zones. Only a small low-risk area is found at the mouth of the Dawen River where it enters the lake. The RI values for Cd show a decreasing trend from east to west, with strong risk zones primarily located in the central and northern parts of Dongping Lake, while the low-risk areas are predominantly in the western region. Based on the distribution of the potential ecological hazard index (RI), it is evident that the central and southern regions face higher ecological risks. Heavy metal pollution in these areas is relatively severe, with significant impacts on the ecological environment, necessitating priority action for pollution control and ecological restoration. Although the pollution status in the Dawen River inlet in the central and eastern parts of the lake is relatively better, continued monitoring of the pollution conditions in moderate risk areas is required to prevent further deterioration of the pollution problem.

Table 5 Statistical summary of $\:{E}_{r}^{i}$and $\:RI$ for heavy metals in Dongping lake sediments.

Table 6 Statistical summary of RI for sampling sites in Dongping lake sediments.

Fig. 11

Generated by ArcGIS 10.8 (Software available at https://www.arcgis.net.cn/).

Spatial Distribution of the RI in Dongping Lake Sediments.

Conclusions

(1) Analysis of the heavy metal content and spatial distribution characteristics in the sediment of Dongping Lake revealed that the average concentrations of most heavy metals are significantly higher than the soil background values of Shandong Province, indicating a widespread enrichment of heavy metals in the lake sediment. Coefficient of variation analysis suggests that, with the exception of Ni and Pb, the spatial distribution of other heavy metals is significantly influenced by human activities. Time series analysis indicates that the overall concentration of heavy metals in the Dongping Lake sediment is stable and shows a tendency for improvement. However, the concentrations of As and Zn have been increasing annually, which is associated with the intensification of industrialization and urbanization. The concentrations of Cr and Pb, after a period of decline, have rebounded, primarily due to the accumulation effect under natural background conditions. Meanwhile, the concentrations of Cd, Ni, and Hg exhibit a trend of first increasing and then decreasing, reflecting the phased effectiveness of environmental management measures. Spatial distribution characteristics show that the high-concentration areas of heavy metals are mainly concentrated in the central, central-western, and southern parts of Dongping Lake, closely related to the inflow of Liuchang River water, as well as agricultural and industrial activities. The northern and eastern regions have relatively low concentrations, benefiting from minimal human disturbance and natural purification processes.

(2) Source apportionment analysis of the heavy metals in Dongping Lake sediment reveals three major sources: industrial sources, agricultural sources, traffic sources, and natural sources. Among these, industrial and agricultural activities contribute significantly to the heavy metal content in the lake sediment. Cluster analysis and principal component analysis (PCA) show that PC1 (As, Cu, Hg) is primarily derived from industrial and agricultural activities; PC2 (Zn, Ni, Pb) is mainly influenced by traffic emissions and high background values; and PC3 (Cd) is predominantly associated with agricultural activities. Positive Matrix Factorization (PMF) source apportionment further confirms these findings, with the contribution rates of industrial and agricultural sources being 38.3% and 29.6%, respectively, and the natural source contributing 32.1%.

(3) The enrichment characteristics and geoaccumulation index analysis of heavy metals in Dongping Lake sediment indicate that cadmium (Cd) exhibits the most significant enrichment, with a moderate level of enrichment primarily influenced by human activities. Arsenic (As), mercury (Hg), zinc (Zn), copper (Cu), nickel (Ni), and chromium (Cr) show slight enrichment, with chromium exhibiting the weakest enrichment. Lead (Pb) is not enriched. The geoaccumulation index results reveal that arsenic (As) and zinc (Zn) are at moderate pollution levels, with arsenic showing a high proportion (95%) in the moderate pollution category, and zinc at 50%. Copper (Cu), nickel (Ni), mercury (Hg), chromium (Cr), cadmium (Cd), and lead (Pb) fall within the transition zone from no pollution to moderate pollution, largely influenced by geological background, though human activities also contribute to their enrichment.

(4) The potential ecological risks of heavy metals in Dongping Lake sediment exhibit significant spatial variation. The potential ecological risk assessment indicates that mercury (Hg) and cadmium (Cd) are the primary contributors to ecological risks, with their average values significantly higher than those of other heavy metals. In certain regions, the risk levels reach the strong pollution category, posing a considerable threat to the ecological environment. From a spatial distribution perspective, the central and southeastern regions show higher ecological risks, particularly at sampling points DN07 and DN17, where the potential ecological hazard index (RI) is the highest. This suggests that heavy metal pollution in these areas is particularly severe, necessitating the prioritization of pollution control and ecological restoration measures.

The study indicates that the ecological environment quality of heavy metals in Dongping Lake sediment is showing a gradual improvement trend. Through comprehensive analysis of the heavy metal elements, including statistical analysis, source apportionment, enrichment characteristics, and ecological risk assessment, it was found that the primary sources of heavy metals are industrial activities (38.3%), agricultural activities (29.6%), and natural background (32.1%). Among these, arsenic (As), copper (Cu), and mercury (Hg) are closely related to industrial and agricultural activities; zinc (Zn) and nickel (Ni) are influenced by traffic and industrial activities; cadmium (Cd) mainly originates from agricultural activities, while chromium (Cr) and lead (Pb) are primarily derived from natural background sources. Notably, mercury (Hg) and cadmium (Cd) are the main contributors to ecological risk, with higher ecological risks observed in the central and southeastern regions. Sampling points DN07 and DN17 show the highest potential ecological hazard index, indicating that heavy metals transported by the Dawen River into the lake are transferred through surface water migration and, as the water flow velocity decreases in the central areas, they gradually settle into the sediment. Overall, the heavy metal pollution in Dongping Lake sediment is the result of both human activities and natural background factors, and urgent action is required through scientific prevention, control, and comprehensive management to achieve sustainable development.