Integrated Assessment of Ecological Quality Combining Biological and Environmental Data in the Yellow River Estuary

Abstract

The integrated assessment of ecological quality in estuarine ecosystems holds significant importance for environmental management. Previous monitoring programs predominantly focused on environmental data, lacking a comprehensive quality assessment approach. To address this gap, this study aimed to integrate environmental factors with macrofaunal community information to evaluate the ecological quality status of the Yellow River Estuary. A total of 13 stations were routinely monitored in August for four consecutive years to collect environmental and biological data. Candidate indicators were screened based on variation coefficients, distribution ranges, and redundancy analysis, identifying 16 indicators belonging to three categories (i.e., seawater, sediment, and biology). The model fit and the interrelationship of the components were determined using structural equation modelling (SEM). The main results were as follows. (1) A total of 144 macrofaunal taxa, belonging to eight animal phyla and 98 families, were identified, with a dominance of Annelida (37.8%) and Mollusca (33.3%). The environmental variables most strongly correlated with the macrofaunal community were TOC, DO, Cd, and Md. (2) NO₂ and heavy metals represented the two most direct factors of environmental pollution, while the factor load of biodiversity indices (H’, J, and D) was large in the biology category. (3) The evaluation results indicated that 78.85% of the total samples were between the average and upper levels of ecological quality, but only 7.69% of samples were at the “high” level. The framework system for the evaluation of ecological quality constructed in this study provides a theoretical and practical basis for the evaluation of the effectiveness of conservation management of the Yellow River Estuary.

1. Introduction

Estuaries and coastal ecosystems are specific habitats situated at the interface of land and sea, characterized by highly dynamic natural disturbances [1,2]. They serve a critical function in delivering a variety of ecosystem services, including regulatory (e.g., as carbon sinks and feeding areas), provisioning (e.g., as fishery and aquaculture resources), and cultural (e.g., as areas for recreational activities and sources of local ecological knowledge) services [3,4]. However, in recent years, estuarine ecosystems have been increasingly threatened by a variety of anthropogenic activities, leading to habitat deterioration and biodiversity degradation [5]. Consequently, routine assessments of ecological quality are essential for effective management aimed at enhancing deteriorating estuarine ecosystems worldwide [6,7,8,9,10]. These assessments provide important information for the evaluation of the effectiveness of nature-based solutions (NBSs) [11]. As both natural and anthropogenic stressors can affect ecosystems simultaneously, it is essential to distinguish their interrelation with observed effects and properly account for the inherent complexity of ecological data [12,13,14].

Traditional evaluations of habitat quality in estuarine waters have predominantly depended on chemical parameters, such as nitrogen, phosphorus, heavy metals, and organic pollutants [15,16,17]. This approach entails the comparison of the content of these substances with environmental quality standards to determine the quality level [18]. However, conducting a comprehensive evaluation solely based on chemical data presents significant challenges, because such data fail to capture the biological impacts resulting from the mixing of multiple pollutants. Fortunately, the European Water Framework Directive (WFD) provides a roadmap for the establishment of integrative standards, combining hydro-morphological elements, physicochemical elements, and biological quality elements [19,20]. It requires that member states establish specific ecological quality standards and targets, thus ensuring the attainment of a “good status” for water bodies [21]. Accordingly, numerous tools have been developed to assess the coastal ecological quality status (EcoQs). Among these, macrobenthos-based indices have been widely accepted by the scientific community [22,23,24,25].

Macrofauna can be categorized into different ecological groups based on their sensitivity and tolerance to disturbance and pollution. Several benthic biotic indices (BBIs), based on the relative abundance of these ecological groups, have been developed, including AZTI Marine Biotic Index (AMBI) [26], Multivariate AZTI Marine Biological Index (M-AMBI) [27], and BENTIX [28]. Although they were initially designed to assess the ecological quality of coastal areas in Europe, they have been widely applied, either individually or in combination, to coastal areas in other countries. In China, AMBI and M-AMBI have been utilized in the estuaries of the Yellow River, Yangtze River, and Xiaoqing River [29,30,31]; similarly, AMBI and BENTIX have found application in the intertidal zone of subtropical islands [32].

The monitoring programs described in most ecological quality assessment guidelines for coastal waters primarily concentrate on environmental parameters. When biological data are included, a comprehensive quality assessment method based on multiple parameters has yet to be developed. To fill this gap, this study aimed to explore the importance of both abiotic and biotic information, for assessing the ecological quality of the Yellow River Estuary in an integrative way. We aim to (1) characterize the temporal patterns of macrofauna for four consecutive years; (2) identify the set of environmental variables that better explains biological community patterns; and (3) develop an integrative method to assess ecological quality status using structural equation modeling (SEM). We hypothesized that the biological data contribute equally to the ecological quality evaluation compared with the seawater and sediment variables using SEM.

2. Materials and Methods

2.1. Study Area

The Yellow River, which flows westward into the Bohai Sea, is China’s second longest river and holds the position of the world’s second largest river by sediment load [33]. Over the past several thousand years, it has accounted for 6% of the worldwide sediment flux into the ocean [34]. Like many large rivers in other countries, the Yellow River has a series of dams and reservoirs, which were constructed to support local industrial, urbanized, and economic developments [35]. Consequently, the Yellow River has experienced a rapid decline in water discharge and sediment load. Since 2002, the water-sediment regulation scheme (WSRS) has been employed to tackle the issue of inadequate water supply for sediment transport in the Yellow River [36]. This scheme releases a substantial volume of water over a short period, representing over 60% of the yearly sediment load and 50% of the annual nutrient flux [37].

2.2. Sampling and Laboratory Analyses

Thirteen stations were routinely sampled for macrofauna investigation in August from 2019 to 2022 in the Yellow River Estuary (Figure 1). Environmental and biological samples were concurrently collected. Bottom seawater samples were collected at each station by using Niskin bottles. Seawater environmental factors including salinity (Sal), dissolved oxygen (DO, mg/L), and pH were measured in situ using a YSI 6600 system (YSI ProPlus Incorporated, USA). Seawater samples were taken to the laboratory to analyze the nitrite nitrogen (NO₂-N), nitrate nitrogen (NO₃-N), ammonia nitrogen (NH₄-N), and phosphate (PO₄-P) [15]. In compliance with the China National Standard [38], two replicate samples were obtained using a Van Veen grab sampler (0.1 m²). Upon sieving with a 0.5 mm mesh size, the biological samples were immediately preserved in a 5% buffered formaldehyde solution. Within the laboratory, all macrofaunal organisms were sorted using a stereomicroscope. Individuals were identified to the most specific taxonomic level achievable. The species names were validated against the database of the World Register of Marine Species (http://www.marinespecies.org/) accessed on 5 August 2023. Taxon density was recorded for each station by counting all individuals collected. Meanwhile, 100 g of sediment were taken to measure the total organic carbon (TOC), sulfide (Sul), and heavy metal contents (Cu, Pb, Zn, Cd, Hg, and Cr). Sediment grain size was analyzed using a laser particle size analyzer (Mastersizer 2000; Malvern, UK) [18].

2.3. Data Analyses

Univariate analysis was performed using the PRIMER v7 package [39]. The main macrofaunal community parameters were the species richness (S: number of taxa), abundance (A: number of individuals/m²), dominance, Shannon–Wiener (H’, log₂), Simpson (D), and Pielou (J) indices. The AMBI and M-AMBI indices were determined using AZTI’s AMBI software (ver. 5.0) (http://ambi.azti.es/) and the species list from December 2022, following the guidelines. Additionally, the BENTIX index was calculated based on the relative percentages of the ecological groups to which the species belonged. Regarding the temporal variation, differences among different years were evaluated using Student’s t-test.

For multivariate analysis, macrofaunal abundances underwent square-root transformation to diminish the influence of dominant taxa. Significant differences in macrofaunal community composition in different years were evaluated via principal coordinate analysis (PCoA) based on the Bray–Curtis distance using the “Vegan” packages. The influence of environmental factors on macrofaunal community structures was analyzed through redundancy analysis (RDA) with the “Vegetation” packages. Concurrently, distance-based linear modeling (DistLM) was employed to evaluate the relative contributions of abiotic parameters to the observed variability in macrofaunal community structures [34]. The model construction employed the best combination of predictors according to the “Best” procedure, utilizing AIC (Akaike information criteria).

2.4. Development of the Integrated Ecological Quality Index

Initially, three important components of the coastal ecological integrity framework system, i.e., water, sediment, and biology, were selected from the perspective of the basic characteristics of estuarine ecosystems. Then, the ecological quality indicators were screened and determined. Next, the standardized method was used to unify the indicator dimensions, and the factors loading in the structural equation model were used to calculate the indicator weights. Finally, an integrated ecological quality index was calculated via the weighted summation method, and the results were graded according to the equidistant method.

(1)

Screening and identification of evaluation indicators

To ensure the uniqueness of the information provided by the selected indicators, this study employed variability analyses, normality tests, collinearity diagnostics, and redundancy tests. These analyses validated the representativeness and non-redundancy of each selected indicator. The analysis of candidate indicators’ variability and distribution range revealed that most water quality indicators exhibited high variability (CV > 50%), signifying significant fluctuations. Conversely, sediment indicators, with the exception of Sul and Hg, displayed low variability, indicative of minor fluctuations. Given the non-normal distribution of most indicators (p < 0.05), spearman correlation analysis and collinearity diagnostics were conducted for redundancy testing in the candidate indicators. The results indicated that the M-AMBI index demonstrated high correlations with several indicators (p > 0.7), and significant collinearity was observed among the H’, J, D, and M-AMBI indices (VIF > 10). Consequently, the M-AMBI indicator was excluded from the candidate list. Consequently, the integrated ecological quality index in this study consisted of 16 indicators.

(2)

Standardization of evaluation indicators

To comprehensively compare how each indicator represents ecological quality, we categorized the indicators into two types: positive indicators and negative indicators. Positive indicators imply that the higher the original value, the better the estuarine ecological quality, while negative indicators indicate that the higher the original value, the worse the estuarine ecological quality. Each indicator is standardized according to the following formulae.

Positive indicator:

$$\mathrm{z}={\mathrm{x}}_{\mathrm{i}}\in \left[{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}+\frac{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{5}\times \left(\mathrm{z}-1\right),{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}+\frac{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{5}\times \mathrm{z}\right]$$

(1)

Negative indicators:

$$\mathrm{z}={\mathrm{x}}_{\mathrm{i}}\in \left[{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-\frac{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{5}\times \mathrm{z},{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-\frac{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{5}\times \left(\mathrm{z}-1\right)\right]$$

(2)

x_i represents any given data point for each indicator, x_max is the maximum value for each indicator, x_min is the minimum value for each indicator, and z is 1, 2, 3, 4, 5.

(3)

Weight calculation for evaluation indicators

According to the factor load in the structural equation model, the weight w_a of each indicator was calculated as follows:

$${\mathrm{w}}_{\mathrm{i}}=\frac{{\mathrm{q}}_{\mathrm{i}}}{{\mathrm{q}}_{\mathrm{t}}}$$

(3)

$$\begin{array}{c}{\mathrm{w}}_{\mathrm{j}}=\frac{{\mathrm{q}}_{\mathrm{j}}}{{\mathrm{q}}_{\mathrm{t}}} \end{array}$$

(4)

$$\begin{array}{c}{\mathrm{w}}_{\mathrm{a}}={\mathrm{w}}_{\mathrm{i}}\times {\mathrm{w}}_{\mathrm{j}} \end{array}$$

(5)

w_i and w_j correspond to the first-order and second-order weights, respectively; q_i is the loading of the observed variable on the first-order latent variable, q_j is the loading on the second-order latent variable, and q_t is the sum of the absolute values of all loadings on that latent variable; w_a represents the total weight of each indicator (

Table 1. The weight distribution of each indicator at all levels.

Goal	Component	First-Order Weight (wi)	Indicator	Second-Order Weight (wj)	Total Weight (wa)
Ecological quality	Water	0.314	NO2	0.353	0.111
			NO3	0.278	0.087
			NH4	0.204	0.064
			PO4	0.165	0.052
	Sediment	0.386	TOC	0.106	0.041
			Sul	0.099	0.038
			Cu	0.217	0.084
			Zn	0.186	0.072
			Cr	0.113	0.044
			Cd	0.152	0.059
			Pb	0.128	0.049
	Biology	0.300	H’	0.241	0.072
			J	0.262	0.079
			D	0.298	0.089
			AMBI	0.083	0.025
			Bentix	0.116	0.035

).

(4)

Determination of ecological quality status

We used the weighted sum method to synthesize multiple indicators to obtain a comprehensive value that could reflect the ecological quality status. The equidistant scoring method was used to divide the scores obtained from small to large into five grades, namely: high, good, moderate, poor, and bad (

Table 2. Classification of the grades of the ecological quality of the Yellow River Estuary.

Grade	High	Good	Moderate	Poor	Bad
Ecological quality index	x ≥ 4.0	3.4 ≤ x < 4.0	2.8 ≤ x < 3.4	2.2 ≤ x < 2.8	x < 2.2
Number of samples	4	12	25	6	5
Proportion (%)	7.69	23.08	48.08	11.54	9.61

).

3. Results

3.1. Environmental Characteristics

In this study, the percentage of silt and clay was high in the majority of stations in the Yellow River Estuary. There was no obvious temporal variation in Sal, NH₄-N, NO₃-N, PO₄-P, Sul, Cu, or Zn, while DO, NO₂-N, Md, TOC, Cd, Cr, Hg, and Pb varied significantly (Figure 2, Table S1). The sediment properties (e.g., TOC, slit, and sand) exhibited higher spatial variations than the water variables (e.g., DO and Sal). In general, the farther offshore, the higher the silt and clay content, while the TOC exhibited similar patterns.

3.2. Macrofaunal Community

A total of 144 macrofaunal species were collected from the Yellow River Estuary during the four years (Table S2). Annelida had the highest number of species, with 54 species, accounting for 37.8% of the total species number, followed by Mollusca, with 48 species, accounting for 33.3% of the total species number. Others included twenty-nine Arthropoda, five Echinodermata, four Chordata, two Nemertea, one Platyhelminthes, and one Brachiopoda. The average species number was 78, and the average density ranged from 120 to 172 ind./m² (Figure 3a). The community was dominated by Mollusca in 2020, whereas it was dominated by Annelida in the other three years (Figure 3b). The composition of dominant species showed great temporal variation (Figure 3b,

Table 3. The characteristics of dominant species in the macrofauna in the Yellow River Estuary.

Dominant Species	Ecological Group	Dominance
		2019	2020	2021	2022
Aricidea fragilis Webster	EGI	0.039	+	0.020	+
Ringicula doliaris A. Gould	EGI	+	0.035	+	+
Goniada maculata Örsted	EGII	0.057	+	−	−
Glycinde bonhourei gurjanovae	EGII	−	−	0.096	0.167
Arcuatula senhousia	EGIII	−	0.116	+	+
Sternaspis scutata Ranzani	EGIII	+	−	−	0.025
Heteromastus filiformis Claparède	EGIV	0.076	0.025	0.077	0.089
Sigambra bassi Hartman	EGIV	0.031	+	0.033	+

Note: the species is defined as a dominant species (Y ≥ 0.02). + and − indicate the dominant species present or absent during the different years.

). A Venn plot showed that there were 39 species in common between 2019 and 2020, 45 species in common between 2020 and 2021, and 65 species in common between 2021 and 2022 (Figure 3c). To further clarify the variability of community species composition across years, PCoA analysis was used to measure the similarity of the species composition over the four years, and the results showed that communities from 2020 distinctly clustered compared to other years, and they were separated by the PC2 axis (Figure 3d).

3.3. Influence of Environmental Data on Macrofaunal Assemblages

The relationship between the macrofaunal communities and the environmental factors is demonstrated in the RDA ordination diagram (Figure 4). Monte Carlo tests for the all ordination axes were highly significant (p < 0.05). The first two axes of the RDA explained 37.23% of the total variation. The first axis was positively correlated with Md, TOC, Hg, and Pb but negatively correlated with DO and NO₂-N. It was obvious that the macrofaunal communities structure of the Yellow River Estuary in different years was affected by different environmental factors. According to DistLM, the environmental variables which significantly correlated with the whole macrofaunal community (identified at the species level) were mainly TOC (5.2% of variation explained), DO (4.8%), and Cd (4.6%) (

Table 4. Results of DistLM (distance-based linear modelling) analysis. The variables explaining macrofaunal community (species abundances) are included. Table shows significance levels for each predictor variable and the proportion of variation explained. Only variables with significant results (p < 0.05) in marginal tests are shown. Overall best models are also included. AIC: Akaike information criterion; SS (trace): sum of squares of sequential test, Prop.: proportion of variability explained by each factor (in the marginal tests without coaction of factors), Sal: salinity, DO: dissolved oxygen, TOC: total organic carbon.

Variables	SS (Trace)	Pseudo-F	p	Proportion (%)
TOC	8307	2.72	0.001	5.2
DO	7668.2	2.50	0.001	4.8
Cd	7491.5	2.44	0.001	4.6
Md	6432.7	2.08	0.007	4.0
NO2-N	5833.8	1.88	0.014	3.6
NO3-N	5881.8	1.90	0.012	3.6
NH4-N	5583.3	1.80	0.014	3.5
Hg	5202.2	1.67	0.035	3.2
Sal	5031.6	1.61	0.045	3.1
AIC	R2	Variables
418.5	13.5	DO, NH4-N, Cd
418.8	16.2	DO, NO2-N, NH4-N, Cd
418.9	16.1	Sal, DO, NH4-N, Cd

), while the Md, NO₂-N, NO₃-N, NH₄-N, Hg, and Sal were also significant. The best model (AIC = 418.5, R² = 13.5) was obtained for the combination of DO, NH₄-N, and Cd (Table 3). When the three main phyla (Annelida, Arthropoda, and Mollusca) were analyzed separately, DistLM showed that variables explained a higher proportion of variability in Arthropoda, followed by Annelida and, finally, Mollusca (Table S3); therefore, the Annelida phylum can be concluded to best reflect the environmental conditions in the Yellow River Estuary.

3.4. Integrated Ecological Quality Evaluation

In this study, we utilized the SEM to construct an estuarine ecological quality assessment model, validating its fit and obtaining the intrinsic connections among various secondary indicators. The results of the data verification indicated that the primary indicators of water, sediment, and biology all had Cronbach’s alphas above 0.6 (Table S4), confirming data reliability and fulfilling the modeling requirements for constructing the integrated ecological quality model. The model fit results (

Table 5. Model fitting criteria and results.

Classify Fit Measures	Absolute Fit Measures		Incremental Fit Measures
	CMIN/DF	RMSEA	CFI	TLI	IFI
Reference standard	<2	<0.08	>0.9	>0.9	>0.9
Initial fit index	1.665	0.114	0.800	0.762	0.810
Modified fit index	1.126	0.050	0.963	0.955	0.966

Notes: CMIN/DF is the ratio of likelihood ratio to degrees of freedom; RMSEA is the root-mean-square error of approximation; CFI is the comparative fit index; TLI is the Tucker–Lewis index; IFI is the incremental fit index.

) indicated optimal model fitting according to absolute fit indices like chi-square/degrees of freedom (CMIN/DF = 1.126) and root-mean-square error of approximation (RMSEA = 0.05). Incremental fit indices like the comparative fit index (CFI = 0.968) also reached optimal standards, suggesting that the constructed ecological quality evaluation model for the estuary closely resembles the real model, with a high overall fit. These validation results indicated that the model outputs could effectively explain the interrelations between ecological quality and secondary indicators in the Yellow River Estuary. The path analysis results for these secondary indicators and primary indicators indicated significant relationships, demonstrating that the selected secondary indicators could be effectively applied to ecological quality assessment (Figure 5). Using the weighted sum method for assessing ecological quality based on SEM, our results showed that the ecological quality values for the study area ranged from 1.70 to 4.59. Although 78.85% of the total samples were between the average and upper levels of ecological quality, only 7.69% of samples were at the “high” level (Table 2, Table S5).

4. Discussion

In this research, we conducted a routine four-year investigation of macrobenthic communities in the Yellow River Estuary. In total, 144 species were collected (62 species in 2019, 75 species in 2020, 81 species in 2021, and 97 species in 2022). These species belong to 125 genera, distributed across 98 families, 39 orders, 14 classes, and 8 phyla. These results are consistent with those of Li et al. [40], who identified 88 species of macrofauna at 26 stations in this area. Our findings also indicated an increasing trend year on year in the species number of macrofauna, which likely reflects the successful impact of ecological protection and restoration initiatives within the Yellow River Estuary. In most years, the macrofaunal communities were dominated by polychaetes in this study, consistent with findings from other studies in the Yellow River Estuary and adjacent sea areas such as Laizhou Bay [30,41]. Compared with historical data, the macrofaunal communities have shown a trend toward miniaturization in the Bohai Sea. Large individual echinoderms, crustaceans, and mollusks have gradually been replaced by smaller individual polychaetes and mollusks [42,43]. This trend may be associated with marine eutrophication, overfishing, and continuous anthropogenic disturbances [44]. However, in 2020, mollusks were dominant in relative abundance (71%), primarily due to the influence of Arcuatula senhousia collected at S13, which exhibited a high abundance of up to 6520 ind./m². Overall, the sediment in this area is characterized by a higher clay component and a smaller median particle size, conducive to the habitation of molluscan species. Additionally, the explosive reproduction and gregarious habituation of Arcuatula senhousia made the large quantity collected at S13 an expected phenomenon. Similar findings were reported by Joel et al. [45], in whose study a significant number of Arcuatula senhousia were also collected at a certain station.

The patterns of macrofaunal communities in this study were primarily explained by TOC, DO, Cd, and Md. These variables are also supported by the findings of previous research [46,47]. Sediment granularity is considered one of the main factors regulating soft-bottom macrofaunal communities. Blanchet et al. (2008) [48] also reported granularity, TOC, and heavy metals as significant variables in explaining the distribution of mollusks, crustaceans, and polychaetes in estuarine environments. The importance of sediment grain size distribution to the structure of macrobenthic animal communities underscores the need to consider the impact of human activities on sediment transport systems, as well as terrestrial organic matter and nutrient loads, in the management of estuarine ecosystems.

Estuarine ecological quality evaluation is a complex system project, and although many scholars have made significant progress in this area, considerable controversy remains over the objectivity and accuracy of their evaluation results [8]. Our study aimed to optimize the monitoring and evaluation of estuarine ecological quality by establishing an evaluation framework, selecting evaluation indicators, and adopting objective quantification methods. The construction of an index system represented the initial stage in the assessment of ecological quality, with the selection of indices that accurately mirrored the critical elements of the estuarine ecosystem being imperative. Commonly, macrobenthic communities are recognized as robust indicators of ecosystem health due to their significant associations with sediment processes, which are concurrently connected to water column dynamics [7,10,49,50]. Hence, in this study, five indicators (H’, J, D, AMBI, and Bentix) were combined to describe the characteristics of the benthic macroinvertebrate community rather than a universal index of other indicators. Given the scope of national routine monitoring programs implemented across China, the index system devised for ecological quality assessment is appropriate for application across the general coastal waters of China. Future validations of the ecological quality evaluation model should incorporate both spatial and temporal dimensions to enhance its robustness and applicability.

Structural equation modeling, which is used in this research, has been the fastest growing branch of statistics over the past 30 years. Its major advantage lies in the analysis of latent variables that cannot be directly measured [51], allowing for the use of multi-scale indicators, including both biological and abiotic variables, to better represent the hierarchy and heterogeneity of complex ecosystems [52,53,54]. Indicator weights form a critical basis for evaluating the ecological quality of estuaries. SEM derives path coefficients from the intrinsic logical relationships in original data to establish weights, providing a new perspective for determining weights for various assessment indicators of estuarine ecological quality [55,56]. In this study, model fit indices such as χ²/df and RMSEA meet excellent standards, indicating that the estuarine ecological quality index model is well adapted to the 16 selected indicators. Among environmental factors, the sediment weights are higher than that of the water. Specifically, the heavy metal factor load of the sediment was high, and this represents the main pollution of estuaries. Coastal and estuarine ecosystems in China are now facing increasing metal pollution pressure [57,58], which affects the growth and reproduction of benthic organisms. The weights of the four water quality indicators (NO₂-N > NO₃-N > NH₄-N > PO₄-P) suggest that nitrogenous input might be more serious than phosphorous input in the study area. Urban industrial and agricultural nitrogen and phosphorus sewage inputs, along with sediment nitrogen and phosphorus releases, jointly affect the estuarine water quality [59,60]. Biological indices have been proven to be effective indicators of ecological quality. In this study, the high weights of the biodiversity indices D, H’, and J indicate that species richness and evenness can greatly influence the biological quality of the study area. This may be due to the uneven distribution characteristics of the macrofaunal community in the Yellow River Estuary. Furthermore, high sedimentation rates in the Yellow River Estuary lead to deteriorating habitat conditions for benthic organisms, reduced food availability, thus affecting biodiversity and ecosystem stability. Overall, the results obtained in this study indicate that effective measures should be taken to control nutrient and heavy metal pollution, to restore biodiversity in the Yellow River Estuary.

5. Conclusions

In conclusion, this research examines the macrofaunal communities and ecological quality in the Yellow River Estuary. Generally, Polychaeta was identified as the dominant taxonomic group within the macrofaunal community, and TOC, DO, Cd, and Md were significant environmental factors that influence the community. The structural equation modeling reveals that environmental factors are highly weighted, particularly emphasizing the substantial disruption caused by nutrients and heavy metals on estuarine ecological quality. The factor load of biodiversity indices (H’, J, and D) was high, indicating the important role of macrofaunal biodiversity in the studied area. The evaluation results indicate that 78.85% of the total samples were between the average and upper levels of ecological quality, but only 7.69% of samples were at the “high” level. Future efforts should focus on the continuous monitoring of environmental factors and macrobenthic community dynamics.