Changes in environmental variables
Changes in water quality variables
Five water quality variables, encompassing water temperature, salinity, pH, dissolved oxygen (DO), and electrical conductivity, have been shown in Table 2. Significant differences in water temperature were observed among the four groups (χ² = 9.12, P < 0.05 in the dry season; χ² = 9.02, P < 0.05 in the inter-monsoon season; χ² = 10.21, P < 0.05 in the wet season). Conductivity differed significantly among groups in the wet season (χ² = 10.65, P < 0.01), while dissolved oxygen showed significant differences among groups in the inter-monsoon season (χ² = 11.10, P < 0.01). In contrast, salinity (P = 0.06 ~ 0.75) and pH (P = 0.07 ~ 0.33) did not differ significantly among groups across all seasons (Fig. 2; Table 2).
Salinity, temperature, pH, conductivity, and dissolved oxygen in the study area among sites in three seasons. For the boxplot figures, the box borders represent the interquartile range (IQR), while the horizontal line within the box indicates the median value. The upper and lower whiskers extend to 1.5 times the IQR beyond the upper and lower quartiles, respectively. Statistical significance is indicated by asterisks (Kruskal-Wallis test, p-values: * < 0.05; ** < 0.01; *** < 0.001).
Changes in sediment variables
During the inter-monsoon season, Pha-a and TOM significantly varied significantly among sites (χ² = 8.93, P < 0.05; χ² = 8.49, P < 0.05). During the dry season, gravel and sand also showed significant variation across sites (χ² = 11.74, P < 0.01; χ² = 11.74, P < 0.01). Chl-a (P = 0.23 ~ 0.79) and silt and clay (P = 0.06 ~ 0.66) did not show significant variation across sites in any season (Fig. 3; Table 3).
Chl-a, Pha-a, TOM, gravel, sand, and silt&clay in the study area among sites in three seasons. For a detailed introduction to boxplot figures, please see the caption of Fig. 2.
Changes in abundance and ecological indices
Macrobenthic abundance and ecological indices
The Kruskal-Wallis’s test showed significant differences in the abundance of macrobenthos across all stations (χ² = 10.16, p < 0.05 in the inter-monsoon season; χ² = 9.22, p < 0.05 in the dry season; χ² = 9.78, p < 0.05 in the wet season). During the inter-monsoon season, the highest abundance was recorded in the FG group (951.54 ± 116.72 ind./m²), while the lowest was observed in the DG group (113.23 ± 59.83 ind./m²). In the dry season, peak abundance was found in both the FG and NG groups (667.54 ± 216.72 and 734.73 ± 17.56 ind./m², respectively), with the lowest in the DG group (205.43 ± 119.75 ind./m²). During the wet season, the highest abundance occurred in the FG and NG groups (817.15 ± 209.69 and 800.54 ± 103.95 ind./m², respectively), and the lowest in the DG group (307.88 ± 65.23 ind./m²) (Fig. 4a).
However, the Margalef, Shannon, and Pielou indices of macrobenthos showed no significant differences among stations across all seasons (P > 0.05) (Fig. 4b, c, d). The highest (0.90 ± 0.08) and lowest (0.45 ± 0.05) values of the Margalef Index were recorded in the wet and dry seasons, respectively. The Shannon Index reached its maximum value (0.97 ± 0.07) during the wet season and its minimum value (0.50 ± 0.03) during the inter-monsoon season. The Pielou Index was highest in the dry season (0.61 ± 0.09) and lowest during the inter-monsoon season (0.29 ± 0.11).
Comparison of macrobenthic abundance and ecological indices among groups in different seasons adjacent to the H-OTEC Pilot Plant. For a detailed introduction to boxplot figures, please see the caption of Fig. 2.
Meiobenthic abundance and ecological indices
The Kruskal-Wallis’s test showed a significant difference in meiobenthic abundance among groups in the inter-monsoon season (χ² = 8.11, p < 0.05), dry season (χ² = 11.27, p < 0.01), and wet season (χ² = 9.68, p < 0.05). The maximum abundance of meiobenthos (1206.62 ± 346.53 ind./10 cm²) was observed in FG group, while the minimum abundance (314.42 ± 84.18 ind./10 cm²) occurred in DG group (Fig. 5a).
The Margalef Index, Shannon Index, and Pielou Index of meiobenthos showed no significant variation among different sites (P > 0.05) (Fig. 5b, c, and d). The highest (1.01 ± 0.11) and lowest (0.45 ± 0.33) values of the Margalef Index occurred in Dry and Inter-monsoon, respectively. Similarly, the Shannon Index recorded its highest value (0.94 ± 0.19) in Dry and its lowest value (0.47 ± 0.11) in Inter-monsoon. However, the Pielou Index reached its highest value (0.62 ± 0.05) in Dry and its lowest value (0.29 ± 0.13) in Inter-monsoon (Fig. 5b, c, and d).
Comparison of meiobenthic abundance and ecological indices among groups in different seasons adjacent to the H-OTEC Pilot Plant. For a detailed introduction to boxplot figures, please refer to the caption of Fig. 2.
Macrobenthic and meiobenthic composition
Macrobenthic composition
The Bray-Curtis index indicated that there was a significant difference in the macro-benthic community structure among different sites during three different seasons (P < 0.01 for the inter-monsoon season; P < 0.01 for the dry season; P < 0.01 for the wet season) (Fig. 6).
Comparison of macrobenthic composition within and among groups in different seasons. For ANOSIM figures, the Y-axis represents the dissimilarity rank distribution. The box border represents the interquartile range (IQR), while the horizontal line within the box indicates the median value. The whiskers extend to 1.5 times the IQR beyond the upper and lower quartiles. Differences among groups are reflected in the between. The test statistic R ranges from − 1 to 1: R > 0 indicates greater similarity within groups, whereas R < 0 suggests greater similarity Between groups than within groups. Significance levels are indicated by P values (*, < 0.05; **, < 0.01; ***, < 0.001).
A total of 2089 macrobenthic individuals were identified, belonging to 22 species, 14 families, 10 orders, and 5 classes (gastropod, bivalvia, pilidiophora, malacostraca, and polychaeta) (Fig. 7; Table 4). The dominant macrobenthic species included Umbonium vestiarium (95.89%), Celebratulus lacteus (1.14%), Nephtys hombergii (0.62%), and Drupella margariticola (0.53%). The remaining species, including Peristernia nassatula, Pagurus minutus, Cerithium traillii, Clithon oualaniensis, and so on, collectively constituted 1.82% of the macrobenthic community.
Comparison of the relative abundance of the 22 macrobenthic species among groups across different seasons.
The results of SIMPER indicated that within four groups, the most dissimilarity of 76.55% was observed between DG and FG group, with species Umbonium vestiarium (Gastropoda: Trochidae), Celebratulus lacteus (Pilidiophora: Lineidae), and Nephtys hombergii (Polychaeta: Nephtyidae) contributing the most to this dissimilarity, accounting for 96.26% of the cumulative difference (Table 5). The least percentage of dissimilarity (31.56%) was observed between the NG and MG groups, with the same three species contributing 92.91% to the cumulative dissimilarity (Table 5). Umbonium vestiarium was the dominant macrobenthos in all sites and seasons, accounting for up to 98.47% of the macrobenthic community in the FG group in the inter-monsoon season (Fig. 7).
Meiobenthic composition
The Bray-Curtis index revealed significant differences in meio-benthic community structure among groups in Dry season (P < 0.05), while no significant differences were observed in Inter-monsoon and Wet season (P > 0.05) (Fig. 8).
Comparison of meiobenthic composition within and among groups during different seasons. For a detailed introduction of ANOSIM figures, please refer to the caption of Fig. 6.
A total of 2,563 individuals representing 9 distinct meiobenthic taxa were recorded. The meiobenthic community was primarily dominated by gastropoda (Table 6). The meiobenthic community exhibited the following distribution in terms of relative abundances: gastropoda (77.02%), nematoda (7.85%), ostracoda (3.97%), bivalvia (3.06%), while foraminifera, polychaeta, copepoda, harpacticoida, and isopoda collectively accounted for 8.10%. Gastropoda was the dominant meiobenthos in all sites and seasons, accounting for up to 86.37% of the meiobenthic community in the FG group in Inter-monsoon season (Fig. 9).
Comparison of the relative abundance of various meiobenthos taxa among groups across different seasons.
The results of the SIMPER analysis indicated that among the four groups, the greatest dissimilarity (26.90%) was observed between the DG and NG groups, with gastropoda, nematodes, and ostracods contributing the most to this difference, accounting for 73.09% of the cumulative dissimilarity (Table 7). The lowest percentage of dissimilarity (18.48%) was observed between the MG and FG groups, with gastropoda, nematodes, and ostracods contributing 66.70% to the cumulative dissimilarity.
Impact of environmental factors on the benthos in the analysis of RDA
A redundancy analysis (RDA) was used to investigate relationships between macrobenthic species and environmental variables (Fig. 10). The first and second axes explained 39.92% and 21.95% of the variation, respectively (Fig. 10). The first axis was mainly contributed by dissolved oxygen, Pha-a, TOM and gravel, whereas the second axis was influenced by water temperature. The most dominant species, Umbonium vestiarium was associated with environments characterized by high water temperature, TOM and gravel concentrations, while it was less prevalent in environments with relatively high silt&clay and sand levels (Fig. 10). Celebratulus lacteus was associated with environments exhibiting elevated Chl-a, conductivity, and salinity.
Redundancy Analysis (RDA) correlation triplots depict the distribution of macrobenthic species (blue line) in relation to sediment environmental variables (red line).
Redundancy analysis (RDA) correlation triplots were used to explore the relationships between meiobenthic taxa and environmental variables (Fig. 11). The first two axes explained 43.97% and 20.03% of the variation, respectively (Fig. 11). The first axis was mainly contributed by Pha-a, gravel and dissolved oxygen, while the second axis was mainly associated with pH, and inversely correlated with conductivity, salinity and Chl-a concentration. Gastropoda were associated with environments characterized by high conductivity, salinity and Chl-a level environment, while nematoda were associated with environments exhibiting relatively high sand content and elevated Chl-a concentrations (Fig. 11).
Redundancy Analysis (RDA) correlation triplots depict the distribution of meiobenthic species (blue line) in relation to environmental variables (red line).
