The spatial variation of physicochemical parameters observed during the present study is shown in Figure 2 . The SST ranged from 26.54 to 30.01°C, with the highest and lowest SST recorded at RAM (offshore) and THA (offshore) stations, respectively. The pH decreased from north to south, with a maximum (8.47) at PUD and a minimum (8.05) at the TIR region. Salinity ranged from 28.35 to 34.38 PSU, with higher and lower salinities observed offshore of RAM (RAM7) and PUD (PUD3), respectively. The mean salinity was higher (31.96 ± 1.42 PSU) at RAM. Compared to the nearshore, the offshore recorded optimal DO and TSS, which might be attributed to less disturbance in the water column leading to good transparency and productivity. The THA stations recorded lower DO and higher BOD, which could be attributed to organic influx from sewage discharge into the marine environment causing oxidative degradation by bacteria, resulting in high BOD and reduced DO (Sahu et al., 2013; Pandey and Ganesh, 2019). TSS showed a significant spatial variation (F=11.785, p < 0.01), which ranged from 15.8 to 67.8 mg/L, with a maximum recorded at the nearshore stations. Freshwater input from many perennial (Krishna, Godavari, and Mahanadi) and seasonal (Palar, Vellar, and Coleroon) rivers in the nearshore region could be attributed to the higher TSS (Shanthi et al., 2013). In addition, three minor rivers, Tamiraparani, Vembar, and Vaipar, drain into the GoM (Rao et al., 2008). Variations in dissolved nutrient levels between the nearshore and offshore of each district are shown in Figure 2 . Most of the nutrients were recorded as comparatively high at the northern stations and low at the southern stations, which could be attributed to anthropogenic activities in the surrounding vicinity. Nutrients such as NO₃ ⁻, PO₄ ³⁻, and SiO₄ ²⁻ were low in the southern stations and increased toward the north, which could be due to anthropogenic activities such as untreated domestic sewage, industrial wastewater, and agricultural runoff (Xu et al., 2010; Kumar et al., 2018). SiO₄ ²⁻ (18.95 ± 11.1 µM) and NH₄ ⁺ (4.85 ± 1.9 µM) were found to be high at PUD and RAM nearshore stations, respectively, while NO₃ ⁻ (3.49 ± 0.4 µM) was found to be high at THA stations.

In addition to the hydrographic parameters, 11 important heavy metals were also examined in the present study and are summarized in Table 1 , with Cd and Hg being below the detection limit (BDL). The mean concentration of heavy metals were in the order of Al >Fe >Zn >Cu >Mn >Cr >Pb >Ni >Co. Al ranged from 14.4 to 420.3 ppb, with a maximum recorded at the THA station, which was above the permissible level (Conama, 2003). The concentrations of Fe (225.6 ppb) and Zn (67.2 ppb) were high at RAM. The concentrations of Cu (26.01 ppb) and Mn (12.5 ppb), which were found to be high in the PUD region, could be due to the usage of biocides attributed to anthropogenic activities (Anbuselvan et al., 2018). Ni (0.51 ppb) was high at THA stations, while Cr (1.52 ppb) was high at THO and PUD stations. Higher Ni concentrations at THA stations could increase phytoplankton production as Ni is essential for algal growth (Price and Morel, 1991). However, the recorded Fe and Ni concentrations were comparatively lower than in nearby coastal regions (Jha et al., 2019). During the present study, most of the metal concentrations were comparatively higher at northern stations. However, the lower concentration of metals like Fe at southern stations could be one of the reason for the lower phytoplankton abundance. Continuous monitoring could help regulate coastal water quality and thus support the fishing industry. Additionally, metal analysis can help identify the sources of contamination and prevent further pollution of the marine ecosystem.

Cluster analysis was used to assess the relationship of physicochemical parameters among the five coastal districts, and the study revealed two statistically significant clusters ( Figure 3A ). In the present study, lower activities such as industrial and pesticides use might have contributed fewer contaminants and nutrients in PUD, RAM, and TIR districts, thus forming cluster 1. Wherein RAM and TIR districts formed clusters with less distance could be attributed to similar sea water quality with lower nutrient and anthropogenic activity, which resulted these two districts as undisturbed (Pandey et al., 2022). Higher concentrations of NO₃ ⁻and PO₄ ³⁻ could be the reason for the formation of cluster 2 between THO and THA. This could be attributed to higher domestic sewage, solid waste disposal, industries, agriculture, and aquaculture activities around the coastal districts. The results are similar to earlier observations that indicated a higher concentration of NO₃ ⁻and PO₄ ³⁻ in THO due to domestic and industrial waste through anthropogenic runoff (Muthukumar et al., 2022).

KMO values were used to assess the sampling efficiency, which revealed variance significance. For the nearshore ( Figure 3B ) data, the KMO was 0.52, indicating that FA can significantly reduce the dimensionality of the original dataset (Wu et al., 2010). Four factors (eigenvalue >1) identified by factor analysis explained 62.5% of the overall variation in the dataset, but the first two factors were selected for detailed discussion ( Table S1 ; Supplementary Material ). Factor 1 showed strong positive loadings of BOD (0.85) and pH (0.84) with a 21.2% variance. In factor 2, moderate positive loadings of PO₄ ^3¯ (0.64), NO₂ ^¯ (0.59), and moderate negative loadings of salinity (-0.71) and NH₄ ⁺ (-0.53) were observed, accounting for 16.2% of the total variation. This could be attributed to higher nutrient loading in the nearshore area due to river inflow (Zhou et al., 2007; Sahu et al., 2013). The offshore area ( Figure 3C ) indicated a significant reduction of the original dataset with 0.565 KMO, and FA explained 65.2% of the total variance with four factors (>1 eigenvalue). Factors 1 and 2 explained 20.9% and 18.2% of the total variance, respectively. Factor 1 showed strong positive loadings of NO₃ ^¯(0.78), PO₄ ^3¯(0.75), and moderate negative loading of SiO² ₄ ^¯ (-0.73). On the other hand, factor 2 showed strong and moderate positive loadings of pH (0.81) and BOD (0.68), respectively. Nevertheless, a similar hydrographic condition was observed in the nearshore and offshore areas. However, TSS and nutrients were comparatively high at the nearshore, which could be due to the anthropogenic input from land runoff (Iscen et al., 2008). The FA analysis revealed a significant correlation matrix between different environmental parameters and their influence on phytoplankton density and diversity.

Phytoplankton biomass (Chl-a) ranged from 0.03 to 0.5 mgm^-3 ( Table S2 ; Supplementary Material ). The mean Chl-a concentration was higher in TIR (0.30 ± 0.03 mgm^-3) and lower in RAM (0.15 ± 0.02 mgm^-3). A similar trend was observed for pheophytin concentrations, which ranged from 0.05 to 0.60 mgm^-3. However, nearshore stations showed comparatively higher biomass than offshore stations, which could be attributed to higher nutrient loading from land runoff. During the present study, 85 phytoplankton species were recorded, such as diatoms (64), dinoflagellates (18), silicoflagellates (1), and cyanophyceae (2). A detailed list of phytoplankton recorded in the present study is shown in Table S3 ( Supplementary Material ). As with Chl-a, phytoplankton abundance was comparatively higher in the nearshore than the offshore stations due to the nutrient input from the land. Phytoplankton densities ranged from 0.7 to 210×10³cells L^-1, with lower and higher densities observed in the THA at offshore (Stn-8) and nearshore (Stn-2) stations, respectively. The mean phytoplankton density was observed as being maximum at THA (59.8 ± 16.2×10³cells L^-1) and decreasing abruptly toward the southern stations, with the lowest density at THO (6.3 ± 1.4×10³cells L^-1). Although THO receives more nutrients from the land runoff, the phytoplankton biomass and abundance were low. This could be due to changes in the hydrographic features, such as a decrease in salinity, pH and an increase in turbidity due to the inflow of freshwater (Bharathi et al., 2018). THA alone contributed 60% of the total phytoplankton abundance, which was ~6 times higher than the other coastal stations in TN. This could be attributed to higher nutrient concentrations at THA, resulting in higher phytoplankton abundance. The phytoplankton community was dominated by diatoms (75.2%), followed by cyanophyceae (17.4%), and dinoflagellates (7.3%). Among the diatoms, pennate diatoms contributed 66% and centric diatoms contributed 34%. The dominance of pennate diatoms could lead to higher fouling rates (Mitbavkar and Anil, 2008). During the present study, seven species dominated the phytoplankton community, i.e., Pseudo-nitzschia sp., Cylindrotheca closterium, Chaetoceros diversus, Cyanobacteria sp. 1 in nearshore stations and Rhizosolenia styliformis, Pleurosigma sp. 1, Rhizosolenia sp. 2 in offshore stations. Most of the species belonged to the pennate diatoms at many stations.

The Shannon-Wiener diversity index (H’) ranged from 0.16 to 2.50 ( Table S2 ). The mean high diversity was observed in TIR (1.99 ± 0.13), followed by RAM (1.80 ± 0.08), and lower in THA (1.43 ± 0.17). Phytoplankton taxa ranged from 4 to 18 at nearshore stations and from 2 to 26 at offshore stations. Optimal hydrographic conditions may have supported the higher phytoplankton taxa at RAM and TIR, while anthropogenic disturbance may have reduced the taxa at THO stations. Maximum phytoplankton abundance and minimum species diversity & evenness were observed at THA, which could be attributed to the dominance of one phytoplankton species (Pseudo-nitzschia sp.) as reported previously (Sahu et al., 2012; Kumar et al., 2018). Species richness and evenness ranged from 0.11 to 2.49 and 0.07 to 1.0, respectively, in the present study. Higher diversity, richness, and evenness were recorded at the TIR stations, where the maximum number of phytoplankton taxa (26) was also recorded. In contrast, lower phytoplankton species richness and diversity were recorded in the THO region, both nearshore and offshore, due to intense anthropogenic activities such as power plants, ports, and various other industrial activities (Bharathi et al., 2018; Jha et al., 2022a).

Pearson correlation was used to identify the significant relationships between phytoplankton and hydrographic factors. In offshore stations, phytoplankton density was positively correlated with NO₃ ⁻ (r = 0.67, P< 0.05), followed by PO₄ ³⁻(r = 0.51, P< 0.05), pH (r = 0.32, P< 0.05) and negatively correlated with SiO₄ ²⁻(r = - 0.27, P< 0.05). A similar result was also observed through multivariate regression analysis, which revealed a positive correlation of NO₃ ⁻ with phytoplankton (r^2 = 0.18, P< 0.05) and pennate diatoms (r^2 = 0.33, P< 0.05). The Chl-a biomass was positively and negatively correlated with NO₂ ⁻(r = 0.22, P < 0.05), and salinity (r = - 0.31, P< 0.05), respectively. In the nearshore, phytoplankton showed the positive and negative correlation with NO₃ ⁻(r = 0.36, P< 0.05) and SiO₄ ²⁻(r = - 0.24, P< 0.05). Compared to the offshore, NO₃ ⁻and SiO₄ ²⁻ were comparatively high in the nearshore, which might be due to the nutrient input from land runoff (Zhou et al., 2008). However, the low nutrient concentration leads to low phytoplankton abundance under natural oceanic conditions, which was evident in the present study except for THA stations. At the same time, increased TSS concentrations in the southern stations may also be one of the reasons for the lower phytoplankton productivity. Previous studies also indicated that increasing TSS as a primary limiting factor by reducing light penetration resulted in lower phytoplankton abundance (Yuvaraj et al., 2018; Kumar et al., 2022). It is understood that NO₃− was the main factor that influenced the phytoplankton especially the pennate diatoms in the present study. The ability of diatoms to quickly absorb and store NO₃ ⁻ resulted in rapid growth (Cermeno et al., 2011), which could support fisheries by implementing artificial reefs. One earlier study also revealed the importance of fisheries production by maintaining good water quality, productivity and through the deployment of artificial reefs (Jha et al., 2022b).