Particulate and Dissolved Metals in the Pearl River Estuary, China—Part 1: Spatial Distributions and Influencing Factors

Abstract

Metals in the Pearl River Estuary are of great importance due to the dense population and rapid industrialization, but they have not been systematically studied. This study investigates the spatial distribution and environmental impacts on dissolved and particulate metals (Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Cd, Tl, Pb) in the Pearl River Estuary by integrating statistical methods and spatial analysis techniques. The areas with high concentrations of particulate metals are mainly located north of the Humen Bridge. Some metals also show higher concentrations near the Hong Kong–Zhuhai–Macao Bridge or around Macau. Overall, the spatial distribution patterns of metals such as Mn, Co, Ni, Cu, Mo, Cd, Tl, and Zn show clustering features. Oxygen content, salinity, and temperature significantly influence most particulate metals. The areas with high concentrations of dissolved metals are mainly located north of the Humen Bridge or in waters closer to the sea. Cr, Cd, Cu, Fe, Ni, Mo, and Tl show a clustered distribution. Dissolved Fe, Ni, and Mo are significantly influenced by environmental factors, except for water depth. This study fills the research gap on dissolved and particulate metals in the Pearl River Estuary, providing essential data to support metal pollution management in the region.

1. Introduction

Metals have long been regarded as one of the main pollutants in aquatic ecosystems. Most metals are present in the environment in large quantities and can both persist and bioaccumulate, making them an important issue. Although some metals are biologically necessary for aquatic organisms, concentrations above certain thresholds are toxic to living organisms [1]. Metals in water exist in different states, e.g., in dissolved form, in the form of suspended solids, and in the form of sediments. When environmental conditions change, physical, chemical, and biological interactions can lead to complex migrations and transformations between different states [2].

Suspended substances are the main carriers of trace metals during transport to the sea and play an important role in the distribution of metals in estuaries. However, variations in the concentration of suspended substances cannot fully explain the variability in the distribution of metals [3]. The environmental behavior of metals in estuaries is diverse and complicated, and their distribution is influenced by many factors such as temperature, salinity, pH, and hydrodynamic conditions [4,5,6,7].

Estuaries are typically densely populated and industrially developed zones, serving as a primary transfer pathway for metals from land to the marine environment and have become an important focus for aquatic metal pollution research [8,9,10,11,12]. Previous research on metals in the Pearl River Estuary mainly focuses on sediments, including analysis of metal content, forms, acid-volatile sulfides, sequential extraction of metals, toxicity assessment, pollution degree, ecological impact assessment [13,14,15,16,17,18], and sources and influencing factors [15,18,19]. Although there is a large number of research on metals in the sediments of the Pearl River Estuary [20,21,22,23,24], studies on metals in suspended particles and water remain limited. In addition, most of the previous studies are limited to the inner estuary areas of the Pearl River, and only a few studies focus on the outer areas of the estuary beyond Lingdingyang Bay and the southern areas. Most existing studies focus on four to six metal types, which makes it difficult to conduct cross-metal comparisons. Finally, differences in sampling locations, timing, and analytical methods between the different studies cannot be eliminated [25,26].

To better understand the distribution and influencing factors of metals in various forms in the Pearl River Estuary and to explore the sources and geochemical behavior of these metals, we surveyed the Pearl River Estuary in January 2022. The specific research objectives of this study are as follows: (1) characterize the spatial distribution of particulate and dissolved metals in the Pearl River Estuary; (2) investigate whether the distribution of metals is influenced by environmental factors; and (3) identify metals that may exhibit similar geochemical behavior.

Water samples were taken, and the concentrations of total suspended matter and metals (Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Cd, Tl, Pb) in suspended particulate matter and filtered water were analyzed in the laboratory. In addition, various water quality parameters were measured at the sampling sites, including water depth, temperature, salinity, oxygen content, and pH value. The effects of these environmental parameters on the distribution of metals in the Pearl River Estuary were then investigated.

2. Materials and Methods

2.1. Study Area

The Pearl River Estuary is located in the southeastern coastal region of China and is the outlet of the Pearl River, the second-largest river in China (measured by annual runoff). In China, the Pearl River is the first river with the highest total deposition of heavy metals, followed by the Yangtze River, the Yellow River, the Haihe River, and the Liaohe River. In terms of the contribution of rivers to coastal metal deposition, the South China Sea also ranks first, followed by the East China Sea, the Bohai Sea, and the Yellow Sea [27]. The Pearl River Estuary is characterized by complex hydrodynamic conditions, large outflows, a low tidal range, a relatively low content of suspended matter, and significant heavy metal pollution. As one of the most economically dynamic and densely populated regions in China, the Pearl River Basin is under enormous environmental pressure due to rapid industrial, agricultural, and urban development. The estuarine waters are influenced not only by upstream pollution sources but also by tides and climate change, which together influence the unique distribution, transformation, and bioavailability of heavy metals. Given the significant changes in the aquatic environment and complex hydrodynamics, the Pearl River Estuary provides an ideal location to study the distribution and drivers of heavy metals.

2.2. Sample Collection and Analysis

We conducted a field survey in the Pearl River Estuary from 5 to 12 January 2021, in which 23 study sites were designed, and approximately 20 L of water samples were collected at each sampling point. The locations and sampling sites are shown in Figure 1. Considering that the hydrodynamic conditions, water depth, and salinity in the Pearl River Estuary vary primarily in the land–sea direction, the sampling sites were designed longitudinally from north to south, i.e., from inland to the sea. Additionally, taking into account the differences in hydrodynamics, the Coriolis force, and the river entry points on the east and west coasts of the estuary, sampling points were set up in an east–west direction. This grid-based sampling design effectively captures the spatial distribution characteristics of metals in the Pearl River Estuary. The water quality parameters of temperature, salinity, oxygen content, and pH value were measured using the Orion FiveStar water quality meter (Thermo Fisher Scientific, MA, USA). The collected water samples were filtered through an acetic acid filter membrane with a pore size of 0.45 μm, and 25 filtrate samples and 25 suspended matter samples were obtained. The filter membranes and other instruments used were rinsed with ultrapure water. Eighteen suspended matter samples and twenty-one filtrate samples were used for the measurement of trace metals. The content of 11 metal elements was measured using the CAP Qc ICP-MS (Thermo Fisher Scientific, Massachusetts, USA), and the entire process of sample digestion was carried out in a dust-free laboratory. The purity of the hydrogen peroxide (H₂O₂, Macklin Incorpration, Shanghai, China) used for the digestion is superior pure, and nitric acid (HNO₃, Macklin Incorpration, Shanghai, China), hydrochloric acid (HCL, Macklin Incorpration, Shanghai, China), and hydrofluoric acid (HF, Macklin Incorpration, Shanghai, China) were further purified based on electronically pure BV-III. The solutions were diluted 4000-fold with 3% HNO₃ (Rh-Re as the internal standard) and determined by ICP-MS.The measurement of the metal content in the filtrate was similar to that in the suspended particles, except for the dilution factor, which was approximately 50-fold. The standard reference materials, AGV-1, BHVO-2, W-2a, SY-4, SARM-4, GSR-1, GSR-2, GSR-3, GSD-9, and GSD-12 (Xindeyuansu Science Compary, Beijing, China) were used to build calibration standard curves for the measurement of metal contents. The total procedural blanks for Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Cd, Tl, and Pb were 1.005 pg, 0.725 pg, 10.36 pg, 0.025 pg, 0.088 pg, 0.078 pg, 1.086 pg, 0.082 pg, 0 pg, 0.001 pg, and 0.368 pg, respectively. The detection limits (3 times the SD of the blank values of 3% HNO₃ solution) were 0.0043 μg/L for Cr, 0.0052 μg/L for Mn, 0.0336 μg/L for Fe,0.0014 μg/L for Co, 0.0027 μg/L for Ni, 0.0025 μg/L for Cu, 0.0007 μg/L for Zn, 0.0007 μg/L for Mo, 0.0010 μg/L for Cd, 0.0002 μg/L for Tl, 0.0014 μg/L for Pb, respectively, reaching the ppt level and meeting the requirements for the sample analysis.

2.3. Saltwater Wedge

The saltwater wedge coefficient (Sw) is a commonly used index to reflect the change and stratification of the salinity of the water and is generally used to distinguish the stratification type of estuaries. The equation for calculating Sw is as follows [28]:

Sw = ds/σs,

(1)

where ds = Sbott − Ssurf, σs = 1/2(Sbott + Ssurf), and Sbott and Ssurf are the salinity at the bottom and the surface of the water column, respectively. When Sw < 0.1, the water column is completely mixed, and when 0.1 < Sw < 1.0, partial mixing occurs. If Sw > 1.0, the stratification is obvious with the presence of salt-wedge is evident.

2.4. Statistical Methods

The statistical analysis methods used in this study primarily included descriptive statistics, correlation analysis, factor analysis, and multiple regression. All statistical analyses were conducted using Microsoft Excel 2010 and SPSS 25.0.

2.4.1. Descriptive Statistics

Descriptive statistics involves the calculation of basic descriptive measures for the sample data, including minimum, maximum, mean, standard deviation (SD), and coefficient of variation (CV). Both SD and CV can be used to measure the volatility of data, and the coefficient of variation can be used to compare the volatility of data of different magnitudes. In this study, the CV was selected to reflect the degree of variation, which was the ratio between the standard deviations of the trace metal concentration at each sampling point and the average values. Weak variability is defined as a CV less than 0.1, while a CV range between 0.1 and 1 indicates moderate variability. A CV greater than 1 indicates strong variability.

2.4.2. Correlation Analysis

Correlation analysis is a statistical method used to evaluate the strength and direction of the linear relationship between two variables. In this study, Pearson’s correlation coefficient (r) was employed to quantify the relationships among different metals, as well as between metals and environmental factors. Pearson’s correlation is appropriate for normally distributed data and measures the degree of association between two continuous variables on a scale ranging from −1 to 1.

An r value close to 1 indicates a strong positive correlation, meaning that as one variable increases, the other also tends to increase. Conversely, an r value close to −1 signifies a strong negative correlation, where an increase in one variable corresponds to a decrease in the other. If r is near 0, it suggests little to no linear relationship between the two variables.

2.4.3. Factor Analysis

Factor analysis (FA) aims to reduce the dimensionality of variables and identify potential influencing factors. Principal component analysis (PCA) is similar to FA, which focuses on transforming the original variables into uncorrelated principal components through linear combinations to maximize variance explanation. It is mainly used for data compression and simplification. In this study, FA was used to explore the commonalities behind the metals and environmental factors.

2.4.4. Multiple Linear Regression Analysis

To quantify the influence of different environmental factors on metals, we employed multiple linear regression to analyze the metals and environmental factors. The standardized regression coefficients (Beta values) were used to compare the relative importance of the independent variables. The larger the absolute value of the Beta coefficient, the greater the influence of the variable on the dependent variable. The change in F must generally be greater than 3, with a significance less than 0.1, which means that the model is meaningful. In this study, the influencing factors often have strong correlations, making it inappropriate to build a fitting model. However, we aim to obtain a clear understanding of the relative importance of these influencing factors.

2.5. Spatial Analysis Methods

The spatial analysis methods used in this study mainly include spatial interpolation and spatial autocorrelation analysis, which were performed in the framework of ArcMap.

2.5.1. Kriging Interpolation

Kriging interpolation is a geostatistical method widely used for the interpolation of spatial data, especially for predicting the value of a variable at a specific geographic location or spatial point. It generates predictions by weighting the known values, with closer points having more influence, and aims to minimize the estimation error. In this study, the kriging interpolation method was used for spatial interpolation and to create a surface map that facilitates the visualization of the spatial distribution patterns of the metals.

2.5.2. Global Moran’s I

Global Moran’s I is a statistical measure used to assess the general spatial autocorrelation of a variable in a study area. It quantifies the extent to which similar values of the variable are spatially clustered or dispersed. A positive value of Moran’s I indicate that similar values are spatially clustered, while a negative value suggests a dispersed pattern. A value close to zero implies a random spatial distribution. Regarding the calculation reference for the Moran’s Index and associated test parameters (Moran PAP, 1948) [29]. This index is commonly used in spatial data analysis to examine patterns of spatial dependence.

Table 1. Global Moran’s I interpretation table.

z-Score	p-Value	Confidence Level
<−1.65 or >+1.65	<0.10	90%
<−1.96 or >+1.96	<0.05	95%
<−2.58 or >+2.58	<0.01	99%

presents the criteria for determining the Global Moran’s I, whereby both the z-score and the p-value must meet the specified conditions.

2.5.3. Local Moran’s I

The Local Moran’s I, based on the Global Moran’s I, analyzes the local autocorrelation of each spatial unit. It helps to reveal whether there are spatial clusters or spatial dispersion in specific regions. The four possible outcomes of Local Moran’s I are as follows:

• High–high cluster: higher observed values form a cluster.
• Low–low cluster: lower observation values form a cluster with low values.
• High–low dispersion: this indicates that the observation value at a given location is high, while the observation values in the neighboring areas are low.
• Low–high dispersion: the observed value at the location is low, while the neighboring values are high.

3. Results and Discussion

3.1. Water Environmental Factors

The basic parameters of the water environment, such as water depth, water temperature, salinity, pH, and total suspended matter (TSM) at the sampling site, are listed in

Table 2. Water environmental factors at sampling points in the Pearl River Estuary.

	Depth of Water (m)	Temperature (°C)	Surface Salinity (psu)	Bottom Salinity (psu)	Oxygen (mg/L)	PH	TSM (mg/L)
Min	5.5	17.8	6.50	9.80	7.93	7.59	0.52
Max	40.0	23.0	32.79	32.95	9.07	8.20	19.99
Mean	18.6	20.3	26.75	28.22	8.75	8.02	6.96
CV	0.6	0.1	0.32	0.25	0.04	0.02	0.70

. The distribution of these parameters and the mixing types of freshwater and saltwater in the estuaries are shown in Figure 2a,b. The water depth ranges from 5.5 to 40.0 m, with an average of 18.6 m. The water depth from Hengmen to Dongao Island and Guishan Island is relatively shallow. The nearshore water depth varies, being deeper in the east and shallower in the west. This pattern results from the long-term deposition of large amounts of sediments transported through the Modao Gate and Jiming Gate on the western coast. The water temperature ranges from 17.8 °C to 23.0 °C, with an average of 20.3 °C. In general, the surface water temperature is lower in the northwest and higher in the southeast. In addition, the water temperature near Hong Kong is significantly higher than in other coastal areas. The salinity of the surface water ranges from 6.50 to 32.79 psu, with an average of 26.75 psu, while the salinity of the bottom water varies from 9.80 to 32.95 psu, with an average of 28.22 psu. Surface salinity generally increases from the inner estuary to the open sea but is slightly lower between Jiming Gate and Modao Gate. The oxygen content of the surface water ranges from 7.93 to 9.07 mg/L, with an average of 8.75 mg/L. In general, the area from Humen to Hengmen serves as a transition zone, with the oxygen content of the surface water being lower in the north and higher in the south. The pH of the surface water ranges from 7.59 to 8.20, with an average of 8.02, and generally increases from the inner estuary towards the Wanshan Islands, being higher in the east and lower in the west. The concentration of total suspended solids ranges from 0.52 to 19.99 mg/L, with an average of 6.96 mg/L. The suspended matter concentration from Hengmen to the Hong Kong–Zhuhai–Macao Bridge(HZM Bridge) is significantly higher than in other areas, which is probably due to the heavy inflow of suspended matter from Hengmen and Hongqimen, which is then partially blocked by the bridge. In addition, the decrease in total suspended solids concentration from the inner estuary to the open sea is not very pronounced, with a slight trend towards higher values in the west and lower values in the east. The water depth at the sampling sites varies considerably, ranging from 5.5 to 40 m. Surface water depth, temperature, salinity, and total suspended solids concentration show moderate variability, while the surface oxygen levels and pH value show weak variability with only slight fluctuations.

3.2. Metals in Suspended Matter

3.2.1. Metals Content in Suspended Particulate Matter

As shown in

Table 3. Basic statistics of metal elements content in suspended particles in Pearl River Estuary (μg/g).

Metals	Min	Max	Mean	CV
Cr	51.77	123.68	71.72	0.27
Mn 1	0.21	5.70	1.09	1.45
Fe 1	15.4	52.2	26.2	0.34
Co	2.64	36.17	12.96	0.67
Ni	18.74	188.21	50.79	0.91
Cu	17.91	91.32	38.12	0.56
Zn	99.82	330.63	169.33	0.39
Mo	0.58	2.46	1.10	0.48
Cd	0.06	1.70	0.39	1.03
Tl	0.09	0.94	0.45	0.48
Pb	34.89	344.73	79.55	0.88

Note: ¹ The unit is mg/g.

, Fe has the highest concentration among the metals examined in suspended particulate matter, with a mean value of 2.62%. Mn ranks second with a mean concentration of 1091.44 µg/g. The mean particulate concentration of Zn is 169.33 µg/g, while the mean concentrations of Cr, Co, Ni, Cu, and Pb are 71.72 µg/g, 12.96 µg/g, 50.79 µg/g, 38.12 µg/g, and 79.55 µg/g, respectively. The mean concentration of Mo in particulate matter is 1.10 µg/g, while Cd and Tl have the lowest concentrations among the analyzed metals, with mean values of 0.39 µg/g and 0.45 µg/g, respectively. The metals in suspended particulate matter, ranked in descending order of their mean concentrations, are as follows: Fe, Mn, Zn, Pb, Cr, Ni, Cu, Co, Mo, Tl, and Cd.

Compared with the concentrations of six heavy metals (Cr, Cu, Zn, Pb, As, Cd) in the suspended particles in the upstream regions of Pearl River Estuary [30], the concentrations of Cr, Cu, Zn, Pb, and Cd in the suspended particles of the Pearl River Estuary are significantly lower. Similarly, the concentrations of Zn, Cu, Ni, Mn, Pb, and Cr in the suspended particles of the upstream Humen are much higher than those in the Pearl River Estuary [31]. This indicates that the concentrations of Cr, Cu, Zn, Pb, Cd, Ni, and Mn in the suspended particulate matter of the Pearl River Estuary are primarily influenced by terrestrial inputs.

Compared to other estuaries (

Table 4. Comparison of metal concentrations (μg/g dry weight) in the suspended matter of Pearl River Estuary and other estuaries.

	Zhanjiang Bay	Major River Estuaries of East Hainan	Yellow River Estuary	Yangtze River Estuary
Metric	Mean	Range	Mean	Mean
Cr	160.76	No data	62.7	78.4
Mn 1	0.89	No data	No data	0.811
Fe 1	22.07	31.81–106.0	No data	38.596
Co	No data	No data	No data	No data
Ni	112.47	18.78–76.31	30.5	No data
Cu	27.14	15.88–56.52	32.5	40.2
Zn	296.9	No data	77.4	182
Mo	No data	No data	No data	No data
Cd	23.55	0.19–0.75	0.278	0.25
Tl	No data	No data	No data	No data
Pb	56.28	22.79–66.30	31	31.5
Date	January 2014	2006–2007	April 2013	2006 dry season
References	[32]	[33]	[34]	[35]

Note: ¹ The unit is mg/g.

), the concentrations of Mn, Cu, and Pb in the suspended particles of the Pearl River Estuary are significantly higher. The concentration of Cd is also higher than in the Yangtze River and Yellow River estuaries. Notably, the coefficient of variation (CV) for these elements is greater than 0.1. However, the CV for all metals except Cr exceeds 0.3, with the CV for Mn and Cd greater than 1 and the CV for Ni and Pb also approaching 1. Overall, the concentrations of metals in suspended particulate matter in the Pearl River Estuary exhibit considerable spatial variability, indicating a strong influence of environmental factors, possibly influenced by human activities, as well as the geochemical behavior of the elements.

The seasonal variations in metal concentrations in aquatic systems are significant [29], primarily due to annual fluctuations in environmental parameters such as temperature and precipitation. For instance, the wet season may increase runoff and sediment influx, which in turn raises the concentrations of specific metals. Long-term monitoring is essential for accurately assessing temporal trends in metal concentrations and the associated environmental risks. Short-term datasets may not adequately capture seasonal fluctuations and long-term trends, whereas continuous monitoring provides a more comprehensive understanding, thus offering a solid empirical foundation for the formulation of effective environmental conservation strategies.

3.2.2. Spatial Distribution of Metals in Suspended Matter

The spatial distribution of particulate metal elements in the Pearl River Estuary is shown in Figure 3. It can be observed that the spatial distribution characteristics of Mn are the simplest and most uniform, with its concentration gradually decreasing from north to south. This straightforward distribution pattern suggests that Mn in the Pearl River Estuary primarily originates from the Humen River estuary, where it is desorbed due to changes in pH or salinity after it enters the estuary.

The concentrations of Co, Ni, Tl, and Fe in suspended matter exhibit a spatial distribution that is higher in the north and lower in the south, with an additional center of elevated values near the Hong Kong–Zhuhai–Macao Bridge. The presence of the bridge alters current velocity and flow patterns, particularly in the intertidal zone, and potentially causes localized water retention or eddies near the bridge. These hydrodynamic effects can result in the deposition and accumulation of suspended particles and metal elements, thereby increasing the concentration of metals in this region. Furthermore, in the intertidal zone, the recurrent changes in water flow direction facilitate the redistribution of suspended particles and metal elements in certain areas. The tidal cycles can transport water with higher metal concentrations toward the bridge area, leading to elevated metal concentrations in this region at certain times. The spatial distribution of Mo and Cu are similar to those of Co, Ni, Tl, and Fe, except that the concentration of Mo increases in the southern part of the Pearl River Estuary (south of Dawan Island), while the concentration of Cu is higher in the southwestern part of the estuary (near the waters of Huangmaohai).

The spatial distribution characteristics of Cd in suspended particulate matter closely resemble those of Cu, with higher concentrations in the north, lower concentrations in the south, and elevated values near Huangmaohai. However, unlike Cu, Cd concentration is not influenced by the Hong Kong–Zhuhai–Macao Bridge, and no areas of high concentration are observed near the bridge. Although the above metal elements generally show a north-to-south distribution pattern, with higher concentrations in the north and lower concentrations in the south, there are subtle differences in the distribution characteristics of some metals in the southern region. The concentrations of Mn, Co, Ni, and Tl gradually decrease towards the sea, while Cr, Fe, Mo, Cu, and Cd exhibit low values before reaching the southernmost waters near the Wanshan Islands.

The distribution patterns of Cr, Zn, and Pb in suspended matter in the Pearl River Estuary are similar. When the estuary is roughly divided into the inner and outer estuaries by the Hong Kong–Zhuhai–Macao Bridge, the concentrations of Cr, Zn, and Pb in the inner estuary decrease progressively downstream. In the outer estuary, south of the bridge, Cr and Pb are primarily influenced by anthropogenic activities, showing a distinct spatial distribution pattern. Based on these distribution patterns, the heavy metals in the suspended particulate matter of the Pearl River Estuary originate primarily from terrestrial sources, with a significant contribution from the Humen River estuary. Both Cd and Cu originate from the Humen Gate and are also influenced by terrestrial inputs from the four western gates. Pb, Zn, and Cr appear to be more strongly influenced by human activities.

In estuaries where freshwater and saltwater mix, the intrusion of saline water promotes the flocculation of fine particles within the sediment [3], resulting in a reduction in the content of fine particles [4]. At the same time, the gravitational circulation formed by the salt front can lead to the accumulation of sediments, creating a zone of turbidity maximum where the content of coarse particles increases. As a result, the concentration of metals may reach a minimum, with various metals exhibiting low-concentration centers near Qi’ao Island or Dong’ao Island. As salinity increases further, there is a gradual sedimentation of larger particles, which is influenced by deposition rates and hydrodynamic conditions [5]. The distribution of heavy metals in the sediments of the Pearl River Estuary is primarily influenced by terrestrial inputs and the hydrodynamic conditions of coastal currents. Under the combined effects of the Coriolis force and coastal currents, runoff from the Pearl River is primarily transported westward. This westward transport results in the accumulation of a larger amount of terrestrial pollutants in the western station areas [7,8]. The distribution of metals in suspended solids closely mirrors this pattern and shows a similar trend.

3.2.3. Spatial Distribution Patterns of Metals in Suspended Matter

The spatial distribution map (Figure 3) provides an intuitive representation of metal concentrations in the Pearl River Estuary. To further investigate the spatial distribution patterns of metals in the estuary, we conducted a spatial autocorrelation analysis. The results of the Moran’s Index analysis are shown in

Table 5. Global Moran’s I Index for particulate metals and its significance test in Pearl River Estuary.

Metals	Moran’s I Index	z-Score	p-Value
Cr	0.043	0.405	0.685
Mn	0.760	3.718	0.000
Fe	0.305	1.525	0.127
Co	0.641	2.863	0.004
Ni	0.708	3.422	0.001
Cu	0.606	2.627	0.009
Zn	0.520	2.280	0.023
Mo	0.721	3.252	0.001
Cd	0.377	1.999	0.046
Tl	0.488	2.131	0.033
Pb	−0.192	−0.997	0.319

. When examining the z-scores and p-values, it was found that the spatial patterns of metals such as Cr, Fe, and Pb did not deviate significantly from a random pattern. In contrast, the spatial distribution patterns of metals such as Mn, Co, Ni, Cu, Mo, Cd, Tl, and Zn exhibit clustering. In particular, the probability that the cluster patterns for the metals Mn, Co, Ni, Cu, and Mo occur randomly is less than 1%, while for the metals Cd, Tl, and Zn, the probability that such cluster patterns occur randomly is less than 5%.

Using Moran’s index analysis, we can assess the overall distribution pattern of metals to determine whether they are clustered or dispersed. To delineate the types and locations of spatial agglomerations of metals, we conducted a local Moran’s I analysis, the results of which are shown in Figure 4. The majority of metals showed local clustering, except for Cr and Pb. Metals such as Co, Cu, and Tl show high-value clustering in the northern region of Shiziyang, while Mn and Mo show high-value clustering near the Nansha Bridge. Ni and Cd show low-value clustering around Wanshan Island, while Zn shows low-value clustering in the waters near Jiminngmen. Fe shows minor clustering in the areas away from the mainland near the Huangmao Sea. Cr is characterized by low-value anomalies north of Wanshan Island, while Pb also shows low-value anomalies north of Wanshan Island and low-value clustering near the Hong Kong–Zhuhai–Macao Bridge, indicating a relatively fragmented spatial distribution.

The spatial accumulation of particulate metals (including Co, Cu, Tl, Mn, and Mo) near Humen may be attributed to three main factors [36]. First, tidal exchanges and strong water currents may transport metals toward the Humen area. Second, changes in salinity and particulate matter concentration could affect the solubility of metals and their adsorption on particles, thereby promoting the accumulation of metals in particulate form. Third, metals may enter the water body through industrial wastewater and urban discharges and combine with particulate matter, leading to the accumulation of particulate metal.

3.2.4. Correlation Analysis of Particulate Metals and Environmental Factors

The results of the correlation analysis between the metal elements in suspended solids and the environmental factors in the Pearl River Estuary are shown in Figure 5. This figure shows a strong correlation between the metal elements in suspended matter, the water environmental factors, and the pairs between the metal elements and the water environmental factors in the Pearl River Estuary. Overall, the metal elements in the suspended matter of the studied area show a high degree of correlation, suggesting a similar environmental behavior. In particular, a significant correlation was found between the pairs of Cr-Fe, Cr-Co, Cr-Cu, Cr-Zn, Cr-Mo, Cr-Tl, Cr-Pb; Fe-Co, Fe-Ni, Fe-Cu, Fe-Mo, Fe-Tl; Co-Ni, Co-Cu, Co-Zn, Co-Mo, Co-Cd, Co-Tl; Ni-Cu, Ni-Zn, Ni-Mo, Ni-Cd, Ni-Tl; Cu-Zn, Cu-Mo, Cu-Cd, Cu-Tl; Zn-Mo, Zn-Cd, Zn-Tl; Mo-Cd, Mo-Tl; and Cd-Tl.

The correlation analysis of environmental factors reveals a significant negative correlation between total suspended solids and water depth, water temperature, salinity, and pH of the water. The physical and chemical properties of the water have a considerable influence on the suspension and deposition processes of suspended solids. A higher salt content in the water can, for example, shorten the suspension time of the solids and thus promote their deposition. Water depth, water temperature, salinity, oxygen content, and pH value of the water are all significantly and positively correlated with each other, except the correlation between oxygen content and water depth. These factors are often interdependent and influence each other.

The correlation analysis also reveals that the metal elements in the suspended matter of the Pearl River Estuary are positively correlated only with the total suspended solids among the water environmental factors, while they are negatively correlated with other factors such as water depth, water temperature, salinity, and oxygen content and pH of the water. This indicates that an increase in suspended matter does not lead to a decrease in metal concentrations in the suspended matter. On the contrary, an increase in other factors (such as water temperature, salinity, pH, oxygen content, and water depth) contributed to a decrease in metal concentrations in the suspended matter. Tl shows a significant correlation with all environmental factors investigated, while Cr and Pb show no significant correlation with any of these factors. Mn, Ni, Cu, Mo, and Cd exhibit a significant negative correlation with temperature, salinity, oxygen content, and water pH. Fe is significantly negatively correlated with water depth and temperature. Co shows a significant negative correlation with water depth, water temperature, salinity, oxygen content, and water pH. Zn is significantly negatively correlated with the salinity, oxygen content, and pH of the water. The results of the correlation analysis indicate that the metals Tl, Co, Mn, Ni, Cu, Mo, Cd, and Zn in the suspended particulate matter of the Pearl River Estuary are highly correlated with environmental factors. In contrast, Cr and Pb show no significant linear correlation with environmental factors, while Fe is only correlated with water depth and temperature.

3.2.5. Factor Analysis of Particulate Metals and Environmental Factors

To simplify the analysis of the influence of the environmental factors on the metal elements, a factor analysis (FA) was conducted on the environmental factors. The results are presented in

Table 6. Loadings of parameters on the first two principal components (91.63% of data variation)—extraction from factor analysis.

Factors	PC1	PC2
Depth of water (m)	0.287	0.899
Surface temperature (°C)	0.635	0.646
Surface salinity (psu)	0.949	0.283
Bottom salinity (psu)	0.908	0.352
Surface oxygen (mg/L)	0.966
Surface PH	0.923	0.351
Surface TSM (mg/L)	−0.109	−0.924
Eigenvalue	4.006	2.407
Cumulative variance (%)	57.234	91.627

Notes: The missing values in the table represent data with absolute values smaller than 0.1. KMO: 0.697; p: 0.000.

. The KMO (Kaiser–Meyer–Olkin) measure was 0.778, exceeding the threshold of 0.6, indicating that the data are suitable for factor analysis. Additionally, Bartlett’s test for sphericity yielded a significance value close to 0, further confirming the appropriateness of conducting factor analysis. After performing the factor component analysis, the seven environmental factors were grouped into two principal components. The eigenvalues of the two principal components were greater than 1 and thus met the criteria for classification as principal components. The total loadings of the two principal components exceed 91%, effectively capturing the information of the original variables. Principal component 1 (PC1) is mainly determined by salinity, oxygen content, and pH and can be regarded as a composite biochemical factor (designated as HJ-1). Principal component 2 (PC2), primarily influenced by water depth and concentration of suspended solids, can be defined as a composite physical factor (designated as HJ-2). The surface temperature of the water body contributes to both the physical and biochemical factors.

By grouping the heavy metals, it is possible to identify which metals have similar sources or are influenced by common underlying factors. To further explore which metals exhibit similar geochemical behavior and are influenced by common factors, a factor analysis (FA) was conducted with the reduced dimensions of the environmental factors and metal elements. The results are presented in

Table 7. Rotated component matrix of suspended particulate metals and environmental factors after dimensionality reduction using FA.

Objects	PC1	PC2	PC3
HJ-1 1	−0.978
HJ-2 2	0.113	−0.824
Cr	0.265	0.381	0.878
Mn	0.972	0.165
Fe	0.345	0.808	0.281
Co	0.847	0.493	0.127
Ni	0.960	0.201	0.103
Cu	0.854	0.303	0.156
Zn	0.801		0.550
Mo	0.897		0.252
Cd	0.871
Tl	0.669	0.699	0.152
Pb			0.939
Eigenvalue	7.127	2.383	2.194
Cumulative variance (%)	54.820	73.152	90.028

Notes: ¹ HJ-1: Principal component obtained from dimensionality reduction analysis of salinity, oxygen content, and pH. ² HJ-2: Principal component obtained from dimensionality reduction analysis of water depth and total suspended solids concentration. The missing values in the table represent data with absolute values smaller than 0.1. KMO: 0.696, p: 0.000.

. The total loadings of the first three principal components exceed 90%, effectively capturing the information of the original variables. PC1 is primarily determined by HJ-1, while PC2 is mainly influenced by HJ-2. PC3 is assumed to be predominantly influenced by anthropogenic factors.

Through factor analysis, we can not only distinguish whether the metal elements are influenced by environmental factors or human activities but also further categorize the environmental factors into biochemical and physical groups. The results indicate that most of the metal elements are primarily influenced by biochemical factors, including Mn, Co, Ni, Cu, Zn, Mo, and Cd. Based on the findings from Section 3.2.2 and Section 3.2.4, it can be deduced that these metal elements have similar geochemical behavior, and Mn, Co, and Ni may share common sources. Fe is primarily influenced by physical factors, while Tl is influenced by both biochemical and physical environmental factors. All environmental factors appear to have an inhibitory effect on the concentrations of these metal elements in suspended matter.

Cr and Pb are not significantly influenced by the environmental factors investigated. Their concentrations and distribution are probably primarily controlled by anthropogenic sources. The reports indicate that various metals in the sediments of the Jiaomen waterway are affected to varying degrees by human activities [9]. Pb in the Pearl River Estuary may be closely related to atmospheric deposition [10]. Mining areas and industrial activities upstream of the Pearl River Estuary serve as significant sources of metals in the Pearl River system. Strengthening environmental regulations is essential to control the discharge of metals from these sources.

The inhibitory effects of environmental factors on concentrations of the metal elements are likely related to the processes of adsorption and desorption, which govern the retention and mobility of the metals in suspended particles. This suggests that environmental conditions may limit the bioavailability or mobility of these metals by promoting their adsorption to suspended matter. However, Cr and Pb appear to remain largely unaffected by the environmental variables considered, suggesting that their concentrations are more influenced by anthropogenic activities rather than natural environmental processes. This highlights the significant impact of human emissions on the distribution and behavior of these metals in the estuarine environment.

3.2.6. Regression Analysis of Particulate Metals and Environmental Factors

To quantify the relative importance of the influencing factors, we performed a multiple regression analysis of the metals and their respective environmental factors. The results presented in

Table 8. Table of multiple regression coefficients for particulate metals and their influencing factors in the Pearl River Estuary.

	Depth of Water	Temperature	Salinity	Oxygen	pH	Suspended Solids
Cr	−0.87 *	−1.49 **	2.56 *	−2.91 ***	1.36	−0.41
Mn	0.08	−0.48 **	0.22	−0.99 ***	0.03	−0.21 **
Fe	−0.58	−1.97 ***	2.23 **	−2.61 ***	1.62 *	−0.19
Co	−0.12	−1.11 ***	0.99 **	−1.51 ***	0.42	−0.16
Ni	−0.01	−0.43 **	0.46	−0.94 ***	−0.27	−0.23 **
Cu	−0.11	−0.94 *	0.88	−1.34 **	0.25	−0.29
Zn	−0.37	−0.51	0.76	−2.04 ***	0.90	−0.41 *
Mo	0.11	−0.65 *	0.08	−1.41 ***	0.83	−0.24
Cd	0.23	0.65	−1.40	0.52	−0.56	0.06
Tl	−0.31	−1.48 ***	1.67 **	−1.99 ***	0.76	−0.12
Pb	−0.69	−0.38	1.45	−1.54	0.58	−0.25

Notes: *: p is less than 0.1. **: p is less than 0.05. ***: p is less than 0.01.

show that particulate metals such as Cd and Pb are only minimally influenced by environmental factors, with their concentration variations being primarily determined by anthropogenic activities or source inputs. According to the factor analysis, Cd is mainly influenced by biochemical factors, and it is speculated that redox potential (Eh) or electrical conductivity (EC) could also serve as potential environmental drivers [12,37]. In contrast, metals such as Fe, Cr, Tl, Co, Cu, Zn, and Mo show a considerable susceptibility to environmental influences. In addition, Mn and Ni show a moderate degree of environmental influence.

The most important factors influencing particulate metals include dissolved oxygen, temperature, and salinity. These parameters have a significant influence on the concentrations of most metals, with dissolved oxygen and temperature being particularly influential. Dissolved oxygen is a key factor for most metals, especially Cr, Fe, Co, Cu, Zn, Mo, and Tl. The absolute values of the regression coefficients for these metals are generally high and have passed significance tests, confirming that dissolved oxygen plays a significant role in the variations in their concentrations. Temperature also has a significant influence on the concentrations of many metals, particularly Cr, Fe, Co, and Tl. Salinity has a strong influence on certain metals, such as Cr, Fe, and Tl. The influence of pH on metal concentrations is more complex. While some metals, such as Fe and Cr, are particularly sensitive to pH, their overall effect is generally less significant than that of dissolved oxygen or temperature. Most metals are not very sensitive to pH. In contrast, suspended solids and water depth have relatively small effects, as indicated by the lower absolute values of their regression coefficients, suggesting that these factors have a weaker influence on the majority of metals.

Oxidative conditions favor the accumulation of heavy metals in sediments, while a high salinity enhances their release [10]. The situation is exactly the opposite for metals in suspended matter. Dissolved oxygen generally shows a negative correlation with most metals, while salinity tends to show a positive correlation. An increase in the concentration of suspended solids in water bodies increases the dilution effect on heavy metals [1], leading to a significant negative correlation between suspended solids and various particulate metals. In addition to the concentration of suspended particles, their particle size also plays a role in influencing the metal content [2].

Redox potential (Eh), electrical conductivity (EC), and soil organic carbon (SOC) are also key factors that influence the spatial distribution patterns of metals [12,37]. Organic carbon promotes the adsorption and precipitation of metals [15]. In addition, sources such as rock and soil weathering debris [6], industrial production and wastewater discharges, transportation activities, and the use of chemicals in agriculture can significantly influence the concentration of particulate metals [8,9,37].

The absolute values of the regression coefficients can be used to determine the relative importance of the influencing factors for each metal. For most metals, oxygen content and salinity are the most influential factors, ranking first in terms of relative importance.

3.3. Metals in Dissolved State

3.3.1. Dissolved Metals Content

Table 9. Basic statistical information of dissolved metal elements content in Pearl River Estuary (μg/L).

Metals	Min	Max	Mean	CV
Cr	1.01	10.25	4.86	0.41
Mn	0.78	1.54	0.98	0.20
Fe	190.90	472.09	381.94	0.25
Co	0.12	0.21	0.18	0.13
Ni	3.49	24.20	18.29	0.33
Cu	1.08	3.55	2.10	0.40
Zn	14.13	48.28	21.78	0.38
Mo	4.61	10.63	8.74	0.24
Cd	0.006	0.062	0.026	0.69
Tl	0.015	0.045	0.025	0.37
Pb	2.43	9.64	4.66	0.39

presents the basic statistical information on the dissolved concentrations of 11 metal elements analyzed in this study, including Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Cd, Tl, and Pb in the Pearl River Estuary. The Fe content was the highest, with an average value of 381.94 μg/L, while the dissolved concentrations of Ni and Zn in the Pearl River Estuary were similar, with an average value of 18.29 μg/L and 21.78 μg/L, respectively. The dissolved concentrations of Cr, Cu, Mo, and Pb were also similar, ranging from 1 to 10, with mean values of 4.86 μg/L, 2.10 μg/L, 8.74 μg/L, and 4.66 μg/L, respectively. The mean concentrations of dissolved Mn and Co were 0.98 μg/L and 0.18 μg/L, respectively, while the concentrations of dissolved Cd and Tl were lowest in the Pearl River Estuary, with mean values of 0.026 μg/L and 0.025 μg/L, respectively. The average dissolved metals showed a general decrease in the following order: Fe > Zn > Ni > Mo > Cr > Pb > Cu > Mn > Co > Cd > Tl.

A larger coefficient of variation (CV) indicates greater variability in the data, suggesting a stronger influence of external factors and typically reflecting a more active geochemical behavior of the element. The CV values of all metal elements range between 0.1 and 1. Mn, Fe, Co, and Mo have CV values between 0.1 and 0.3, while most other metal elements, including Cr, Ni, Cu, Zn, Cd, Tl, and Pb, have CV values between 0.3 and 1. Among these, the variability of dissolved Cd content is highest in the Pearl River Estuary. Overall, the CV reflects the volatility of the data, and the dissolved concentration data of all metals exhibit moderate variability.

The metals have the potential to pose risks to aquatic life, depending on their concentrations, forms of existence, and environmental conditions. Toxic metals (e.g., Cd, Pb, Cu) are known to be highly toxic to aquatic organisms even at low concentrations. They can disrupt physiological processes, impair reproduction, and cause mortality in sensitive species. While essential metals (e.g., Fe, Zn, Cu) are essential for biological functions at trace levels, they can become toxic at elevated concentrations. Metals in dissolved forms are generally more bioavailable and toxic to aquatic organisms. Metals bound to suspended particles are less bioavailable but can still pose risks if released into the water column under changing environmental conditions [2].

3.3.2. Spatial Distribution of Dissolved Metals

The spatial distribution of dissolved metal elements in the Pearl River Estuary is shown in Figure 6. The concentrations of dissolved Tl, Cd, Cu, and Co in the Pearl River Estuary exhibit similar spatial distribution patterns, generally characterized by higher values in the northern region and lower values in the southern region. Tl, Cd, and Cu also show a westward high and eastward low trend, with higher concentrations observed in the western waters around Qi’ao Island, Macau, and Modao Gate, compared to the eastern waters at the same latitude. Additionally, Cd, Cu, and Co display a localized high-value center near Wanshan Island. This distribution pattern is closely linked to terrestrial inputs and human activities.

The spatial distribution trends of Fe and Mo are almost identical, with concentrations gradually increasing from north to south. In the east–west direction, concentrations are higher in the central region than in the areas along the eastern and western coasts. The spatial distribution characteristics of dissolved Fe and Mo can be summarized as a gradual increase from land to sea. This distribution pattern suggests that dissolved Fe and Mo primarily originate from the ocean or are mainly influenced by natural factors such as salinity and oxygen.

The spatial distribution of Ni, Cr, and Pb is generally similar, characterized by high concentrations in the northern and southern regions and lower concentrations in the central region. The areas with high concentrations are located north of the Humen Bridge and south of the Hong Kong–Zhuhai–Macao Bridge, while the areas with low values are primarily found near the Lingdingyang Bridge. In addition, all three elements exhibit a westward high and eastward low trend. However, significant differences were observed south of the Hong Kong–Zhuhai–Macao Bridge: Pb shows higher values in the southeast corner, Cr in the southwest, and Ni is elevated throughout the Pearl River Estuary. The distribution of Ni, Cr, and Pb is influenced by both natural factors and anthropogenic activities. In contrast, Zn and Mn show no clear spatial distribution patterns, with scattered high and low-value centers.

3.3.3. Spatial Distribution Patterns of Dissolved Metals

The results of the Moran’s I analysis for dissolved metals in the Pearl River Estuary are presented in

Table 10. Global Moran’s I Index for dissolved metals and its significance test in Pearl River Estuary.

Metals	Moran’s I Index	z-Score	p-Value
Cr	0.338	1.722	0.085
Mn	−0.351	−1.310	0.190
Fe	0.869	3.797	0.000
Co	0.307	1.470	0.141
Ni	0.716	3.379	0.001
Cu	0.507	2.240	0.025
Zn	−0.213	−0.777	0.437
Mo	0.823	3.580	0.000
Cd	0.420	1.910	0.056
Tl	0.901	3.971	0.000
Pb	0.004	0.231	0.818

. The spatial distributions of dissolved Mn, Co, Zn, and Pb do not significantly deviate from a random pattern. In contrast, Cr, Cd, Cu, Fe, Ni, Mo, and Tl exhibit clustered distributions. Specifically, Cr and Cd show significant clustering at the confidence level of 90%; Cu at the confidence level of 95%; and Fe, Ni, Mo, and Tl at the confidence level of 99%.

To obtain a comprehensive understanding of the distribution patterns of the metals, we conducted a global Moran’s I analysis. To further clarify the clustering patterns of dissolved metals in specific areas of the Pearl River Estuary, we performed a local Moran’s I analysis. The results are shown in Figure 7. The spatial distribution patterns of most dissolved metals in the Pearl River Estuary show a localized clustering. Metals with anomalously high or low values, such as Cd, Mn, and Zn, are relatively rare. The clustering patterns of dissolved metals in the Pearl River Estuary show clear spatial differences, with most clustering zones concentrated north of Hengmen. The distribution patterns can be divided into three classes. For low-value clustering, Cr and Pb are located in the Lingdingyang Bridge–Jiaomen region, Ni near Lingdingyang Bridge, Fe near Nansha Bridge, and Mo near Lingdingyang Bridge and Nansha Bridge. Finally, Zn shows a fragmented pattern with a low-value clustering southeast of Dong’ao Island and the Wanshan Archipelago. For high-value clustering, Ni is located near North Humen, Fe near Wanshan Island, Cu near Nansha Bridge, Mo on Wanshan Island, Cd near Nansha Bridge, and Tl and Co near North Humen. For outliers, Mn shows a high-value anomaly near the northern Dong’ao Island, Cd shows a low-value anomaly near Lingdingyang Bridge, while Zn shows a low-value anomaly in the northern Wanshan Archipelago and a high-value anomaly in the southern offshore waters.

3.3.4. Correlation Analysis of Dissolved Metals and Environmental Factors

The correlation analysis of the metal elements (Figure 8) reveals significant positive correlations between the following pairs: Cr-Ni, Fe-Mo, Cu-Cd, Cu-Tl, and Cd-Tl. Elements with high positive correlations likely have similar sources or are influenced by common factors. Both Fe and Mo show significant negative correlations with Cu, Cd, and Tl, which can be attributed to their different sources.

Many metal elements show strong correlations with environmental factors. For example, dissolved Fe and Mo show positive correlations with all environmental factors except total suspended matter (TSM). Conversely, Tl, Cd, and Cu show negative correlations with all environmental factors except TSM. Co shows a similar trend, although the correlations are not significant. There is no significant linear correlation between dissolved metal element concentrations and TSM, suggesting that their distribution is more influenced by solubility. In general, low pH (acidic conditions) increases the solubility of certain metals, such as iron. However, in the Pearl River Estuary, dissolved Fe exhibits a significant positive correlation with pH. This could be attributed to the relatively neutral pH range (7.59–8.20) in the estuary or to the influence of oceanic sources. The primary factor controlling the concentration of dissolved Fe is likely oceanic input rather than pH.

3.3.5. Factor Analysis of Dissolved Metals and Environmental Factors

Based on the reduction in dimensionality in environmental factors, the factor analysis was applied to the dissolved metal elements. The results are shown in

Table 11. Rotated component matrix of dissolved metals and environmental factors after dimensionality reduction using FA.

Objects	PC1	PC2	PC3	PC4
HJ-1 1	0.973		0.122
HJ-2 2	0.155	0.476	−0.686	0.246
Cr	0.119	0.871
Mn		0.193	0.689
Fe	0.980	0.125
Co	−0.418	0.106	0.434	0.197
Ni		0.914	0.184
Cu	−0.766		0.378
Zn		−0.108		0.971
Mo	0.971	0.187
Cd	−0.809	−0.169	0.297	0.211
Tl	−0.950		0.108
Pb	0.119	0.490	0.581
Eigenvalue	5.229	2.209	1.773	1.109
Cumulative variance (%)	40.221	57.217	70.855	79.383

Notes: ¹ HJ-1: Principal component obtained from dimensionality reduction analysis of salinity, oxygen content, and pH. ² HJ-2: Principal component obtained from dimensionality reduction analysis of water depth and total suspended solids concentration. The missing values in the table represent data with absolute values smaller than 0.1. KMO: 0.592; p: 0.000.

. Together, the first four principal components account for almost 80% of the total variance and thus effectively capture the information of the original variables. The dissolved concentrations of Fe, Cu, Mo, Cd, and Tl are mainly influenced by biochemical factors in the environment. Moreover, biochemical factors promote the concentration of dissolved Fe and Mo while they inhibit that of Cu, Cd, and Tl. Based on the results of the correlation analysis and the spatial distribution of the metal elements, dissolved Fe and Mo not only show a high degree of spatial coherence, with a correlation coefficient of 0.99, but also show very consistent correlations with environmental factors. This suggests that dissolved Fe and Mo are likely to have similar geochemical behaviors and sources. Cu, Cd, and Tl also show similar geochemical behavior. Dissolved Mn, on the other hand, is primarily influenced by physical factors, which could be related to its strong adsorption to particles. The chemical properties of Zn are relatively active, so the dissolved Zn content may be primarily controlled by its intrinsic properties. Therefore, it is assumed that principal component 4 (PC4) reflects the intrinsic properties of the metals, so principal component 2 (PC2) probably indicates the sources of the metal elements. The dissolved concentrations of Cr and Ni may have similar sources, and their spatial distribution further suggests that they are mainly terrestrial in origin. Pb is influenced by both sources and physical factors.

3.3.6. Regression Analysis of Dissolved Metals and Environmental Factors

The results of the multiple regression analysis between dissolved metals in the Pearl River Estuary and environmental factors are presented in

Table 12. Table of multiple regression coefficients for dissolved metals and their influencing factors in the Pearl River Estuary.

	Depth of Water	Temperature	Salinity	Oxygen	PH	Suspended Solids
Cr	−0.03	−0.91	1.57	−1.60	0.77	−0.21
Mn	0.59	−0.27	0.46	−0.14	−0.32	0.64
Fe	0.07	−0.09	0.92 ***	−0.07	0.23	0.18 ***
Co	0.50	−0.70	2.40 *	−0.89	−1.50	0.51
Ni	−0.37	−1.40 **	2.76 **	−2.83 ***	1.17	−0.42
Cu	−0.14	−0.23	−0.58	−0.43	0.49	0.05
Zn	−0.11	0.27	−1.25	0.52	0.53	−0.28
Mo	0.04	−0.12	1.04 ***	−0.25 **	0.29 *	0.08 *
Cd	−0.25	−0.02	−1.35	0.28	0.42	−0.12
Tl	0.10	−0.28	−0.09	−0.43	−0.29	0.03
Pb	0.02	−0.68	1.69	−1.46	0.38	0.15

Note(s): *: p is less than 0.1. **: p is less than 0.05. ***: p is less than 0.01.

. Overall, the metals that are significantly influenced by environmental factors are dissolved Fe, Ni, and Mo, especially Ni. The most important factors that have a remarkable influence on the dissolved metals are salinity and oxygen content, which influence the concentrations of various dissolved metals. Salinity, for example, shows a positive correlation with Fe, Co, Ni, and Mo, suggesting that salinity may be an important cause of concentration variations in these metals. Temperature and oxygen show a negative correlation with most metals, such as Ni. In particular, a low-oxygen environment can lead to an increase in dissolved concentrations of certain metals. The concentration of suspended solids significantly enhances the dissolved concentrations of Fe and Mo. Previous studies have shown that dissolved metals are primarily influenced by salinity and pH [16,17], which is consistent with our results. In addition, we observed that oxygen content also has a significant influence on dissolved metal concentrations.

The absolute values of the regression coefficients can be used to determine the relative importance of the influencing factors. Salinity and pH are consistently among the most important influencing factors for most metals, especially for Fe, Co, Cu, Zn, Mo, Cd, and Pb. Oxygen is a dominant factor for Cr, Ni, Tl, and Pb and also plays an important role for Co, Cu, Zn, Mo, and Cd. Suspended solids are very influential for Mn and Fe.

4. Conclusions

The following conclusions can be drawn from this study based on the analysis of two common forms of metals in water (dissolved and particulate) and water environmental factors:

(1)

The primary spatial distribution characteristics of metals in the suspended particulate matter of the Pearl River Estuary show a decreasing trend from north to south. Several elements, including Co, Ni, Tl, Fe, Mo, and Cu, exhibit a high concentration near the Hong Kong–Zhuhai–Macao Bridge. In contrast, elements such as Cd, Zn, and Pb tend to have higher concentrations near the estuary and in areas influenced by anthropogenic emissions. Overall, the spatial distribution patterns of the vast majority of particulate metals (Mn, Co, Ni, Cu, Mo, Cd, Tl, and Zn) show clustering.

(2)

Most particulate metals are primarily influenced by biochemical factors, in particular oxygen content, temperature, and salinity. However, environmental factors have no significant influence on Cd and Pb.

(3)

The main spatial distribution characteristics of dissolved trace metals in the Pearl River Estuary can be summarized into three types. The most common distribution pattern is characterized by high concentrations in the north and low concentrations in the south, with higher values in the west and lower values in the east, as observed for elements such as Cd, Tl, Co, and Cu. Secondly, concentrations are high in the northern and southern areas and low in the central area is low, such as Cr and Ni. Thirdly, the concentration decreases from the sea towards the land, as in the case of Fe and Mo. The dissolved concentrations of Fe, Cu, Mo, Cd, and Tl are mainly influenced by biochemical factors in the environment. Mn and Pb are influenced to a certain extent by physical factors in the environment. Cr, Cd, Cu, Fe, Ni, Mo, and Tl show a clustered distribution pattern.

(4)

The metals that are significantly affected by environmental factors are dissolved Fe, Ni, and Mo. The factors that have a significant influence on the dissolved metals are salinity, oxygen, and temperature.

(5)

This study fills the spatial gap in previous research on the Pearl River Estuary, which has mainly focused on the large estuaries and their upstream areas. It expands the scope of metal studies in the Pearl River Estuary and provides a more detailed examination of the distribution of metals and their contributing factors. By identifying patterns in elements with similar sources or geochemical behaviors, this study contributes to the development of metal pollution control strategies for the protection of marine and estuarine waters.