1 Introduction

Trace metals (TMs) could be toxic to living organisms and even threaten human life and health. The main sources of TMs were weathering of bedrock, volcanic eruptions, and human activities (Mazel et al., 2022). TMs were widely used in the manufacture of automobiles, mining, pesticides, photographic chemicals, etc. Pollution of groundwater and surface water systems caused by human activities was a major environmental problem worldwide (Kumar et al., 2020).

As one of the important land-based inputs to the oceans, the study of trace metal content and sources in rivers was of great significance for offshore marine environmental management (Lee et al., 2007; Zeng et al., 2019a). The studies of trace metal mainly focused on the characteristics of spatial and temporal distribution and their influencing factors, identification of sources, and assessment of pollution and health risks. Concentrations of TMs in most urban rivers were higher than the world average for rivers, and pollution came mainly from fertilizer factories, farms and agricultural run-off from towns, which posed a health risk to adjacent waters (Leventeli and Yalcin, 2021). Seasonal and source analysis of surface water samples in different seasons showed that As, Fe, Se, Sr and V were mainly from the natural environment, while Al, Ba, Cr, Co, Cu, Mn and Ni were mainly from human activities (Giri and Singh, 2014). More than 70 minor and medium rivers along the coast of Jiangsu Province discharged a large number of land-based pollutants into the sea. Doulonggang River, Xinyanggang River, Huangshagang River, Sheyanghe River, Guangaizongqu River, and Linhonghe River served important functions such as drainage, navigation and sewage disposal in Jiangsu Province and they were highly influenced by humans. These areas with high trace metal concentrations were mainly concentrated in estuaries and port areas (Meng et al., 2019). Sheyanghe River, the largest tributary in the north, was contaminated with surface sediments due to pollution from domestic sewage, industrial effluent and agricultural wastewater (Xu et al., 2014). Linhonghe River, as the largest sewage outlet in Haizhou Bay, the TMs (Cu, Zn, Pb, Cd, etc.) in the estuarine waters were higher than those in Bohai and Jiaozhou Bay (Li et al., 2010). The enrichment factors and Analysis of Principal Components-Multiple Linear Regression (APCS-MLR) model and Unmix model indicated four sources of trace metal in Xinyanggang River. Cr, Th, U, Se, Zr and Nb came from industrial and hydrodynamic transport erosion. Ni, Rb, Sc and Ga were from natural sources. Cu, Zn, Mo, Pb and Sn were from mixed sources, including industrial effluents and traffic discharges. As and Sr were mainly related to mixed sources from agriculture and combustion (Cai et al., 2023). TMs had different distribution in different seasons, and TMs in Yancheng coastal wetland had the highest pollution level in spring and summer (Zhang et al., 2020a). Studies on TMs at the northern Jiangsu Province have focused on their distribution (Cao et al., 2020) and the source analysis of river surface sediments and seasonal variation characteristics (Zhang et al., 2020a). But there was alack of systematic and comprehensive studies on the extent and sources of pollution and risk assessment of TMs. In this study, the spatial and temporal distribution characteristics of metals, sources and controlling factors of metals in the major rivers were analyzed.

2 Materials and Methods 2.1 Sampling and Analysis

Sample were collected from Doulonggang River, Xinyanggang River, Huangshayang River, Sheyanghe River, Guangaizongqu River, and Linhonghe River from January to December 2019 (Fig.1). Clean plastic buckets were used to collect surface water samples in 1 L polyethylene bottles. Samples were filtered through a 0.45μm polycarbonate membrane, and the filtrate was treated with a few drops of nitric acid to ensure pH lower than 2 during the determination of dissolved TMs. All membranes and bottles employed during sampling were precleaned with an HCl solution (1:1000) and thereafter rinsed with Milli-Q water.

Salinity (S), pH and dissolved oxygen (DO) were determined in the field using a portable water quality analyzer (Orion3 STAR, Thermo Scientific). TMs V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Mo, Cd, and Pb were determined via inductively coupled plasma mass spectrometry (Thermofisher iCAP Q, ICP-MS, Thermo Fisher Scientific, Germany). The measurement process included the insertion of a standard calibration sample at every 20 th sample. The detection limits for the various TMs were: V (0.02 μg L⁻¹), Cr (0.01μg L⁻¹), Mn (0.04μg L⁻¹), Fe (0.6 μg L⁻¹), Co (0.003 μg L⁻¹), Ni (0.05 μg L⁻¹), Cu (0.02 μg L⁻¹), Zn (0.03 μg L⁻¹), As (0.03 μg L⁻¹), Mo (0.004 μg L⁻¹), Cd (0.01 μg L⁻¹), and Pb (0.01 μg L⁻¹). The recovery rate of the internal standard for each metal element was in the range of 85% to 110%.

2.2 Statistical Analyses

The coefficient of variation (CV) quantifies trends in concentrated and dispersed development of an aggregate and unit over time, or differences in development of different aggregates over time (Reed et al., 2002). Principal component analysis (PCA) was frequently used to establish the interrelationships and potential sources of TMs (Shi et al., 2023). PCA with varimax rotation of standardized component loadings were conducted for maximizing the variation among the variables under each factor, and those PCs with eigen value > 1 were retained. All statistical analyses were carried out by using SPSS 22.

3 Results 3.1 Spatial Distribution of Trace Metal

The annual average concentrations of V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Mo, Cd, and Pb in the six rivers were 2.20, 1.22, 4.53, 21.88, 0.13, 2.79, 1.77, 4.00, 2.97, 3.87, 0.01 and 0.19 μg L⁻¹, respectively (Table 1). Fe had the highest average concentration, which was more than 10 μg L⁻¹. Mn was the second highest element, with an average concentration of 1.15 – 31.60 μg L⁻¹. The highest concentration of TMs was found in Guangaizongqu River and Linhonghe River, whereas the lowest concentration was found in Huangshagang River and Xinyanggang River. The highest concentrations of Cr, Fe, Cu, Zn, Mo, Cd, and Pb were found in Guangaizongqu River, while the highest concentrations of V, Co, Ni, and As were found in the Linhonghe River. The concentrations of Cr, Cu, Cd, Pb, Fe, and Mo varied from north to south; initially increasing and then decreasing. The concentration of trace metal in the Guangaizongqu River was significantly higher than that in others. The concentrations of V and As increased gradually from Doulonggang River in the south to Linhonghe River in the north, with a significant variation. However, the concentration of As was lower in Guangaizongqu River. V was similar in Doulonggang River, Xinyanggang River, and Huangshagang River, with higher concentration in Sheyanghe River. The highest concentrations of V and As were found in the Linhonghe River. Concentrations of Co, Ni, Cu, and Zn varied highly among the rivers. The higher concentrations were in the central and northern regions of the study area.

Compared to other rivers in the world, the concentration of TMs in the rivers of northern Jiangsu Province were typically at a moderate level. The concentrations were higher than those in the Yellow River, Yangtze River, Pearl River and Seine River, which were significantly influenced by human activities. However, the concentrations were lower than those in Mahananda River and Odiel River, which undergo significant industrial, agricultural, and mining activities (Table S1). The concentrations of V, Cr, and Pb was similar to those of the Yangtze River, Pearl River, and Seine River. Fe was higher and was comparable to that of the Subarnarekha River. The concentration was much lower than those of Loutété River, Odiel River, and Huaihe River which were heavily polluted by TMs. The concentrations of Ni, As and Zn were higher than the global average values, but significantly lower than those of Huaihe River and Odiel River. The concentration of Cd was below the average in rivers of the world.

3.2 Seasonal Variation of Trace Metal Concentrations

During rainy season, the concentrations of Cu, Fe, Co, Pb, Cd, V, and As were higher than those in dry season (Fig.2). The average of these elements in different season were showed in Table S2. Except for the Sheyanghe River, the concentrations of Cr and Cd were generally consistent throughout the year in others. The Sheyanghe River showed relatively large fluctuations. During summer, the concentration of Cr was lower compared to other seasons, whereas high Cd concentration occurred in March and April. The other six metals exhibited an upward trend during spring and summer, peaking in summer and decreasing gradually to lower concentrations in autumn and winter. The highest concentrations of Mo, Ni, and Mn were found in spring and autumn. Zn in Linhonghe River and Guangaizongqu River experienced fluctuations due to seasonal changes. The concentrations in Doulonggang River, Huangshagang River, Xinyanggang River, and Sheyanghe River remained fairly stable throughout the year with no significant seasonal pattern.

3.3 Coefficient of Variation

The CV offered a method of comparing trace metal concentrations and the spatial distribution relationships among different rivers located at the northern Jiangsu Province. The CVs of As and V were 11.73% and 14.85%, indicating that the spatial differences in the concentrations of these two elements in different rivers were minor (Fig.3). The CVs of Co, Cu, Mo, and Ni were 41.94%, 46.44%, 49.71%, and 53.44%, which nearly approached 50% and represented a moderate degree of variation. The CVs of Cr, Mn, Fe, Zn, and Pb were 96.95%, 80.24%, 99.40%, 84.69%, and 96.94%, respectively. These coefficients represented high degrees of variation, with large spatial differences. The CV of Cd was higher than 100%, indicating a very high degree of variation. The spatial distribution of these elements exhibited a significant difference.

4 Discussion 4.1 Key Factors Controlled TMs in Different Rivers

On the basis of the coefficient of variation, the spatial variations in the concentrations of the elements can be divided into three categories: The concentrations of the first category increased and then decreased in a north-to-south direction, including Cr, Cu, Cd, Pb, Fe and Mo, with a peak in Guangaizongqu River. Cr, Cd, Cu, Pb, Mo in Xinyanggang River and Huangshagang River were similar and the concentrations were low. Relevant studies have shown that Cr, Cu, Cd, Pb were trace metal components prevalent in urban domestic wastewater, mainly from detergents, lubricants, cosmetics, etc. The trace metal in Guangaizongqu River was greatly affected by human agricultural activities (Lohani et al., 2008). Guangaizongqu River located in the lower reaches of the Huaihe River, was one of the Huaihe River flood discharge into the sea outlet (Yan, 2011). A large amount of industrial and agricultural water containing TMs was concentrated here, resulting in a large accumulation of Cr, Cu, Cd, Pb, Fe, Mo, etc.

The concentrations of the second group, V and As, showed a gradual increase in concentration from Doulonggang River in the south to Linhonghe River in the north. V and As in Guangaizongqu River varied a lot, with a lower As concentration. V was essentially identical in Doulonggang River, Xinyanggang River and Huangshagang River, with higher concentrations in Sheyang River. The highest V and As were found in Linhonghe River, with the higher concentrations in flood season than in dry season. As was widely used in industrial and agricultural raw materials such as antiseptic and pharmaceutical production. There were more plains in the northern Jiangsu Province, where both industry and agriculture thrived. Some of the poor-quality water irrigation could cause spatial variability of arsenic in soil (Hajalilou et al., 2011). The soil matrices material of the coastal plain was marine sedimentary matrices material, and the soil contained high concentration of As. As an important flood channel, Linhonghe River had large discharge volume during the wet season and scouring effect on the soil, which exacerbates the loss of As in the soil and resulted in an increase in As concentration.

The concentration of the third group, including Co, Ni, Mn and Zn, exhibited significant variations. The chemical properties of the third group were similar, leading to nearly uniform spatial distributions, displaying a 95.7% correlation coefficient. Concentrations were higher in the northern part of the study area's central region. The Linhonghe River flows through the Xinyi-Ganyu area, where the soil is relatively rich in Co, Ni, Mn and Zn (Li et al., 2010), resulting in higher concentrations of in the river. Linhonghe River served as a discharge point for industrial effluent and urban domestic sewage, resulting in high concentrations of some TMs. Zn and Mn were among the major trace metal pollutants in urban waste-water, mainly from electroplating, pipelines, batteries, construction materials, etc (Yang et al., 2019). Therefore, the high Zn and Mn concentrations in the Linhonghe River come from the urban wastewater.

The highest concentrations of trace metals were in Guangaizongqu River and Linhonghe River. While the lowest concentrations were observed in Xinyanggang River and Huangshagang River. The watershed contained rocky soils that were rich in metal, which resulted in high background values found in the river water (Huang et al., 2009). There were variations of elemental contents in soils across different subregions within the soil chemistry area of Jiangsu Province. In the northern part, Linhonghe River, Xinyihe River, and Guanhe River flew through the Lianyungang-Gunnan area, where pedogenic rock was related to lagoonal deposition. The soils in this area contained moderate amounts of Fe, As, Cr, Cu, Zn, Ca, and Mg. But Xinyanggang River, Huangshagang River, Doulonggang River and Sheyanghe River, were situated in plain area. The soils generally contained low levels of trace metal (Shen et al., 2005; Liao et al., 2011). Geogenic factors had a significant impact on river trace metal concentrations in northern Jiangsu Province. Even the elevated concentrations in the Ganges Basin were mainly due to geological factors rather than anthropogenic activities (Islam et al., 2023). The concentrations of TMs in Linhonghe River were significantly higher than those in Doulonggang River, Xinyanggang River and Huangshagang River. The land use types in the four river basins of Doulonggang River, Huangshagang River, Xinyanggang River and Sheyanghe River consisted mainly of agricultural and urban land (Chen et al., 2015). The main sources of TMs were municipal wastewater and agricultural irrigation water. There was a significant distinction in the distribution characteristics and influencing factors of TMs among Linhonghe River, Guangaizongqu River, and the others. The banks of the main Guangaizongqu River were mainly used for farmland and residential areas. However, the sampling sites selected for this study were located near estuary so the water salinity was significantly higher than the rest of the river (Table 2). The concentration of base metals had increased, leading to competition for the adsorption sites and resulting in an enhanced resolution of TMs. This effect was especially noticeable for Cd and Cr, which were more prone to forming chlorine complexes. It also reduced their adsorption capacity in the solid phase and led to an increase in the concentrations of Cr, Cu, Cd, Pb, Fe, and Mo (Tam et al., 1999; Zhang et al., 2013). The Guangaizongqu River had a slightly higher average pH than the rest rivers. The elements were more easily desorbed into the water which were sensitively to the acidity fluctuations, such as Cr. There were lots of chemical and high-tech zones in the Linhonghe River and Guangaizongqu River (Yin et al., 2011). Numerous TMs entered the atmosphere, wastewater, and solid waste through various means and eventually contaminated the river through precipitation and surface runoff. Guangaizongqu served various functions, including flood discharge of Huaihe River, agricultural irrigation south of the ruined Yellow River, as well as navigation, power generation, and flood relief. These collective functions resulted in elevated metal concentrations (Lin et al., 2022). Cr, Cu, Cd, Pb, Zn were commonly found in pesticides and fertilizers. And they were released into Guangaizongqu River through farm irrigation wastewater. This resulted in a significant buildup of the respective TMs in Guangaizongqu River (Chen et al., 2017; Liu et al., 2022). The Linhonghe River received a significant amount of trace metal elements such as As, Pb, Zn, Cu, and Cr from atmospheric deposition, and Cd was primarily accumulated through hydrodynamic transport (Zhao et al., 2023). The lower trace metal concentrations at Dulonggang River, Xinyanggang River, and Huangshagang River were due to the dominant land use type in the basin being arable land, including irrigated farmland and less industrial land use; The Huangshagang River watershed was the smallest, covering an area of 865 km² and with a population of just 46000. As a result of negligible human activity, its impact on the concentration of TMs was minor, with most TMs recording the lowest concentrations in Huangshagang. The differences in trace metal concentrations among different rivers were mainly influenced by geogenic factors and human activities in the river basin.

4.2 Source of Trace Metal

The distribution and sources of TMs in anthropogenic environments were more complex compared to natural environments (Liao et al., 2018). The major pollutant source analysis methods currently used in the aquatic environment include Factor Analysis (FA), Principal Component Analysis (PCA), Cluster Analysis (CA), etc. (Vu et al., 2017; Kumar and Singh, 2018). PCA is frequently used to establish the interrelationships and potential sources of TMs (Shi et al., 2023). The northern Jiangsu Province was densely populated with a dense river network and complex industrial and agricultural sectors, so the same pollution source emitted multiple trace metal elements, and the same metal element also came from multiple pollution sources. A single-source analysis mo- del was insufficient to accurately determine the source of pollutants (Mazel et al., 2022). The correlation analysis, coefficient of variation and principal component analysis were used to analyze the source of TMs.

To assess the sources of trace metal, Pearson's correlation coefficients were used in matrices to collapse the dimensionality of the dataset to a few influencing factors while retaining the relationships presented in the original data (Wang et al., 2017b). Significant positive correlation was observed between the concentrations of all metals. A variance closed to 1 for the common factor revealed that extracted information comprehensively and completely retains the original data (Table 3). PCA of TMs extracted two principal components with eigenvalues above 1, with a cumulative contribution rate of 91.2%, which could reasonably explain the source of TMs at the northern Jiangsu Province.

PCA indicated that there was a significant positive correlation among Fe, Mo, and Cd (p < 0.01). There was a certain correlation with Cr, Pb, Cu, and Zn as well (p < 0.05) (Fig.4). Ni was significantly positively correlated with V, Co, and Mn (p < 0.01), suggesting that the geochemical behaviors and the sources of these two groups, were somewhat similar. The factor loadings are categorized as strong, moderate, or weak according to the corresponding loading values, > 0.75, 0.75 – 0.5, < 0.5 (Liu et al., 2003; Jiang and Zhao, 2023). PCA 1 contributed 54.1% and showed high positive loadings on Cr, Fe, Cu, Zn, Mo, Cd and Pb with respective coefficients of 0.988, 0.924, 0.783, 0.909, 0.990 and 0.984. PCA 2 accounted for 37.1% of the variance, with high positive loadings of V, Mn, Co, Ni, and As (Fig.5). The correlation analysis results were consistent with the classification data.

The PCA1 represented the sources of human activities, such as industrial and agricultural activities. The key elements identified in this component were Cr, Fe, Cu, Zn, Mo, Cd, and Pb (Fig.6). The agriculture and industries in northern Jiangsu were complex, with the presence of 15892 industrial enterprises, including fine chemicals, iron, and petroleum industries etc. Cr and its compounds were commonly used as industrial raw materials and have a wide range of applications in the chemical industry, dye production, leather tanning, chromium plating, and alloy production etc. (Giri and Singh, 2014). The main sources of Cd were the electroplating and chemical industries, as well as fertilizer raw materials (Liu et al., 2019b; Dai et al., 2021). Cu was mainly derived from diesel combustion and iron industry. Pb was widely used as an anticorrosive compound in marine coatings and important component of fertilizers, including nitrogen, phosphorus fertilizers, gasoline antifreeze and diesel combustion (Zhang et al., 2015). Vehicle exhaust was one of the important sources of Pb and Cd (Yao et al., 2017). Steel production through the converter and electric furnace processes resulted in trace generation of Fe, Pb, Cr, and Zn wastes. Emissions from the steel and machinery industries were key contributors of Pb, Cr, and Zn in the environment (Simeonov et al., 2003; Chen et al., 2007). Emissions were far above soil background level (Kumar et al., 2017). The industrial wastewater discharge in Jiangsu Province in 2019 was 13.75 t, which contained 11121 kg of Cr, 22 kg of As, 1711 kg of Pb and 4.14 kg of Cd (Annual Environmental Statistics of Jiangsu Province, 2019). The emissions of industrial waste gas were 6.86 trillion cubic meters per year and contained notable amounts of Hg, Cr, and other metal elements (Simeonov et al., 2003; Chen et al., 2007). In summary, trace metal elements such as Cr, Fe, Cu, Zn, Mo, Cd, Pb, etc. at the northern Jiangsu were mainly derived from the anthropogenic impact of agriculture and industry.

The second main component was a mixed source of actions, which included natural sources and anthropogenic sources such as agricultural practices and the chemical industry. This component was represented by V, Mn, Co, Ni, and As (Fig.5). There was advanced agricultural in Jiangsu Province, whose total use of fertilizers and pesticides was 2929500 t in 2019 (National Bureau of Statistics, 2019). The intensity of fertilizer application in Lianyungang and Yancheng ranked second and fourth in Jiangsu Province (Lianyungang Statistical Yearbook, 2019). As and its compounds were widely used in pesticides, fungicides, and herbicides. Agricultural water and fossil fuel emissions containing As entered rivers with groundwater, rainwater discharge and atmospheric deposition (Li et al., 2020), resulting in significant accumulation of As in water. The iron industry, power plants, and paper mills were the primary industrial sources of Mn (Giri and Singh, 2014). Crude oil contained TMs, and catalysts used in oil cracking and fine chemicals contained V, Co, Ni, and Zn (Cao et al., 2021). The aluminum industry in northern Jiangsu Province was another industrial source of V, because V₂O₅ was one of the products of the Bayer method in alumina production (Giri and Singh, 2014). The discharged wastes had a significant impact on TMs in the river (Simpson et al., 2013). The background concentration of TMs in river depended mainly on the rock and soil types. Average concentration of As, V, Mn, Co, Ni in the soil of Jiangsu Province were relatively high (Liao et al., 2011). V, Mn, Co, Ni, As were derived from a mixture of bedrock weathering, agricultural activities and chemical industry.

4.3 Factors Influencing Seasonal Variations

Industrial and agricultural activities manifest distinctive seasonal characteristics and were the primary drivers of seasonal variations in trace metal concentration (Zhang et al., 2014; Ding et al., 2021). Jiangsu Province is situated on the boundary between the northern subtropical and southern temperate zones. This climate was humid and rainy, and the average monthly precipitation in Yancheng and Lianyungang cities in 2019 was shown in Fig.6. The two cities experience uneven precipitation distribution during the year, with more than 65% of the annual precipitation occurring within the wet season, May to September. Yancheng City had the largest annual precipitation in July, reaching 188.1 mm, followed by 178.3 mm in August (Yancheng City Water Resources Bulletin, 2019). Lianyungang City experienced the highest precipitation of the year in August, with a precipitation of 201.6 mm (Fig.7). Compared with the multi-year average monthly precipitation, the precipitation in May and September 2019 was significantly lower, with a decrease of 90% and 82%, respectively (Lianyungang City Water Resources Bulletin, 2019). TMs tend to be physically desorbed due to high temperatures, which led to their release into the water from suspended particulate matter. Additionally, the scouring effect of trace rainfall during the wet season causes the entrance of a large amount of domestic waste into the river via surface runoff, which results in a significant increase of most TMs concentration (Li et al., 2013; Du et al., 2019).

The concentrations of Cu, Fe, Co, Pb, Cd, V, As in all six rivers were higher in the wet season than in the dry season. The elements mainly originated from domestic sewage and industrial wastewater. In summer, the amount of water used for domestic and industrial increased and entered the natural environment. Rice was planted in June, with a significant increase in agricultural water and fertilizer use. Cd, Pb, As originated from the fertilizers, herbicides and pesticides used (Qiu et al., 2020). They were carried into rivers with irrigation water and surface runoff, leading to a sharp increase. Higher levels of suspended particulate matter and high water discharges in summer (Fig.8) caused resuspension of particulate matter and release of TMs into the water. Cr was less variable over the course of a year, but Cr (Ⅲ) was sensitive to changes in acidity and was capable of complexing with acids (Zhou et al., 2005; Sun et al., 2018). High precipitation in the summer reduced water acidity and slightly increased. Cr, Cu, Fe, Co, Pb, Cd, V, and As showed an increasing trend from dry season to wet season mainly caused by agricultural and industry. Studies in the Karasu River and the Fen River had found the same conclusion, the highest TMs (Cd, Fe, Pb etc.) concentration was found in wet season due to the combined effect of weathering processes, increased anthropogenic contributions (such as agricultural activities) (Chai et al., 2021; Varol et al., 2021).

Mo, Ni, and Mn were mainly of anthropogenic origin and show high concentrations in spring and fall. Metal industry products had peak levels in the spring season (Jiangsu Bureau of Statistics, 2019). Cr, Cu, Zn, Ni, Co and Mn were heavily enriched in metal industrial discharges, which entered water through surface water dischargeand atmospheric deposition (Simeonov et al., 2003; Chen et al., 2007). Rivers in northern Jiangsu Province were seasonal and warmer in the spring. TMs mainly originate from weathering and pedogenic processes during the spring season when snowmelt water dominates the river flow (Varol et al., 2021). This could accelerate the rate of contaminant transport and diffusion and widen the area of contamination (Zhang, 2020; Li et al., 2022). The inflow of glacial water from upstream caused the accumulation of pollutants downstream and a gradual increase in trace metal concentrations (Dong et al., 2017a, 2017b). Fall was a dry season with high evaporation from water, so the industrial and agricultural water could not be discharged in a timely manner (Lin et al., 2022), resulting in high levels of TMs in parts of the water. There were results similar to Atuwara River that TMs were slightly higher in the dry seasons compared to those in the wet seasons due to the increase in rainfall in the wet season (Emenike et al., 2020).

Guangaizongqu River showed a specifically seasonal changes, Fe, Zn, Mn concentrations peaked in May related to DO. High-valent Fe-Mn was reduced as an oxidizing agent, and Fe-Mn reduction leads to the release of TMs adsorbed on or co-precipitated with Fe-Mn hydroxides into water (Wang et al., 2009). The lowest DO was found in Guangaizongqu River in May (Fig.9), and the lower DO could promote the release of Zn from the sediments to the overlying water (Chen et al., 2020). There is a correlation between Zn and Fe (Fig.4).

There were similarities in the seasonal distribution of the four rivers, Doulonggang River, Huangshagang River, Xinyanggang River, and Sheyanghe River, and the types of land use in the watersheds of these four rivers were mainly agricultural and urban land. TMs were mainly affected by agricultural irrigation water, the seasonal changes in the four rivers are relatively consistent and stable due to the seasonal farming activities. Linhonghe River, Guangaizongqu River showed differences in the distribution and influencing factors compared with the other four. The Linhonghe River, as the largest sewage inlet of Haizhou Bay, was subject to a wide range of human activities such as industry, agriculture and other human activities. There were more uncertainties that Linhonghe River fluctuated more seasonally in trace metal elements. The sides of Guangaizongqu River were dominated by farmland and residential areas, the sampling site was located near the sea inlet. Guangaizongqu River was affected by domestic sewage, agricultural wastewater and large changes in water salinity caused by seawater intrusion. Frequent anthropogenic activities at the estuaries had also led to significant fluctuations in trace metal concentrations.

5 Conclusions

Concentration, spatial and temporal distribution, pollution status, and sources of 12 TMs in 6 rivers at the northern Jiangsu Province were evaluated. Guangaizongqu River showed high concentrations of Fe, Mo, Cr, Cu, Cd, and Pb. The middle and northern rivers of the study area had high concentrations of Co, Ni, Mn, and Zn. The Linhonghe River had the highest concentration values of V and As. The results of multivariate analysis and PCA models indicated that the source of 12 TMs in rivers at the northern Jiangsu Province could be classified into two categories: 1) Cr, Fe, Cu, Zn, Mo, Cd and Pb; 2) V, Mn, Co, Ni and As. The source and contribution rate (PCA model) represented anthropogenic activities (54.1%) and natural and anthropogenic activities mixing source (37.1%). Higher temperatures, precipitation and emissions led to increased sewage inflow into rivers, resulting in elevated concentrations of Cu, Fe, Co, Pb, Cd, V, and As during the rainy season in comparison to the dry season. Mo, Ni, and Mn concentrations were higher in spring, and fall due to the influx of melted snow from upstream in spring, as well as increased metallurgical production.

Acknowledgements

This study was funded by the National Natural Science Foundation of China (Nos. 42130410), and the Joint Fund between NSFC and Shandong Province (No. U1906210). We thank our laboratory colleagues for their collaboration in sampling and data acquisition.

Author Contributions

Liping Yuan: conceptualization, methodology, software, visualization, investigation, formal analysis, writingoriginal draft. Qian Liu: visualization, investigation, formal analysis. Huimin Jian: methodology, software, investigation. Tiezhu Mi: resources, writing-review & editing. Fuxia Yang: supervision, validation, data curation, investigation. Qingzhen Yao: project administration, resources, writingreview & editing.

Data Availability

Supplementary material is available in the online version of this article at https://doi.org/10.1007/s11802-025-5824-0.

Declarations

Ethics Approval and Consent to Participate

This article does not contain any studies with human participants or animals performed by any of the authors.

Consent for Publication

Informed consent for publication was obtained from all participants.

Conflict of Interests

The authors declare that they have no conflict of interests.