Fig. 1

Spatial distribution of sampling sites in Beibu Gulf. Sixteen black dots represent the sixteen sampling sites"

Fig. 2

Relative abundance of invertebrates in Sanniang Bay and the Dafeng River estuary in March (a), July (b), September (c), and December (d)"

Fig. 3

Seasonal variation in alpha diversity indices: Simpson index (a) and species richness (b). Note: for one-way ANOVA, * denotes P<0.05, ** denotes P <0.01, *** denotes P<0.001"

Tab. 1

Linear regression analysis of alpha diversity indices: Simpson index, Shannon index, and species richness with environmental factors in March, July, September, and December"

Fig. S1

Bray-Curtis dissimilarity between two sampling sites in different seasons"

Fig. S2

Relationships between Bray-Curtis dissimilarity and geographical distance, pH, TN, $\text{NO}_{2}^{-}$, and Chl-a in March. The black line represents the linear regression, r indicates the degree of correlation, and P-values were calculated to indicate significant differences"

Fig. S3

Relationships between Bray-Curtis dissimilarity and geographical distance, pH, salinity, and TN in July. The black line represents the linear regression, r indicates the degree of correlation, and P-values were calculated to indicate significant differences"

Fig. S4

Relationships between Bray-Curtis dissimilarity and geographical distance, pH, turbidity, $\text{PO}_{4}^{3-}$, $\text{NO}_{3}^{-}$, and TN in September. The black line represents the linear regression, r indicates the degree of correlation, and P-values were calculated to indicate significant differences"

Fig. S5

Relationships between Bray-Curtis dissimilarity and temperature, pH, $\text{PO}_{4}^{3-}$, $\text{NO}_{3}^{-}$, $\text{NO}_{2}^{-}$, and TN in December. The black line represents the linear regression, r indicates the degree of correlation, and P-values were calculated to indicate significant differences"

Fig. 4

Seasonal CCA biplots of invertebrate community composition and environmental variables in March (a), July (b), September (c), and December (d)"

Fig. 5

PLS-PM analysis of invertebrate communities responses to environmental variables in March (a), July (b), September (c), and December (d). Invertebrate community formation correlates well with seawater properties and nutritional factors, as indicated by strong model fit indices (R2 values between 0.876 and 0.956). Positive and negative interactions are represented by red and blue lines respectively. Solid lines denote significant interactions, while dashed lines indicate non-significant ones"

Fig. S6

Heatmaps showing the relative abundance of top 20 invertebrate communities in March (a), July (b), September (c), and December (d) and Spearman's rank correlation of these communities with environmental factors (*P<0.05, ** P<0.01, *** P<0.001)"

Fig. 6

The co-occurrence networks in invertebrate communities based on OTU analysis in March (a), July (b), September (c), and December (d). Node colors differentiate phyla in the network. Connections indicate strong (Spearman’s |r|>0.80) and statistically significant (FDR-adjusted P<0.05) correlations within the invertebrate community. Node size reflects the eigenvector centrality, representing each node’s influence in the network"

Fig. 7

Proportional influence of deterministic and stochastic processes on invertebrate community assembly. In Fig.a-Fig.d, blue dots: taxa above the neutral model prediction curve (stochastic process-dominated); black dots: taxa fitting the neutral model (neutral process-explained); red dots: taxa deviating from the neutral model (deterministic process-dominated). Nm is the product of metacommunity size (N) and migration rate (m), representing dispersal flux (alternatively, the effective number of migrants) to quantify the overall species dispersal capacity among communities"