River Mahanadi, together with its three tributaries - Daya, Nuna, and Bhargavi, discharges into the Bay of Bengal in the State of Odisha on the east coast of India (20^°17’08’’ N; 86^°42’24’’ E) ( Figure 1 ). The mixed semi-diurnal tides with a range of 1.45 to 2.20 m (mean, 1.29 m) reach 13 km upstream (depth 5.19 ± 1.11m) and support a rich diversity of the estuarine (mangrove) flora and fauna (Dey et al., 2013; Palei et al., 2014). According to the Indian Water Resources Information System (I-WRIS, 2021), Mahanadi is one of India’s largest and longest (494 km) river systems, with a catchment area of over 65,628 km². The leading seaport, ‘Paradip’ - located 8 km south of the river mouth- is a hub for regional sea-borne trade. With a steadily rising population, the port township is fast developing as a core investment region for petroleum and chemicals (Hazra et al., 2020). Several industries like Indian Farmers and Fertilizers Cooperative Ltd., Paradip Phosphates Limited, Pellet Plant, Indian Oil Corporation Limited, Essar Steel, Ice manufacturing plants, Sea Food Processing units, besides many others, have sprung up in the immediate vicinity, with near denudation of once dense mangrove vegetation fringing the mudflats. Although fishing and cultivation are the traditional livelihood choices of the coastal residents, lucrative shrimp farming has gained precedence in recent years. As a result, the chemicals used for enhancing aquaculture production have been rampant. With the leaching of such substances into the waters, impacts on the estuary are imminent. Besides the untreated sewage from the township, effluents containing oil sludge, sulfur, ash, and gypsum released from industries are proven deterrent to the estuarine ecosystem (Hazra et al., opp. cit. and references therein; SPBO, 2020; Acharyya et al., 2021a; Acharyya et al., 2021b). Overall, the MES has become vulnerable to increased anthropogenic pressures and necessitates further assessment studies.

The climate of Mahanadi Delta is influenced by the premonsoon (March-May), postmonsoon (September-November), and winter (December-February) seasons every year. The southwest monsoon is the major monsoon season during which southwest directional winds blow from June to September. The northwest directional winds influence the winter months (December to February). The weather is typically marked by hot and humid conditions (31.46°C) in April-June and cool and dry (28.66°C) in December-January (https://www.worldweatheronline.com/). The southwest or summer monsoon (mid-June to September) brings heavy precipitation (average, 417.75 mm). The Mahanadi in spate discharges about 45,000 m³ s^-1 into the Bay of Bengal (Mohanti and Swain, 2005), reducing salinity and increasing fluvial loads in the nearshore waters. Furthermore, the southwest monsoon winds generate waves up to ~3 m high or more (Dash et al., 2020). As a result, the coast remains wave-dominated throughout the southwest monsoon and mixed wave and tide-dominated in the non-monsoon periods (Mohanti and Swain, opp. cit.).

Both biotic (macrobenthos) and abiotic (water and sediment) samples were collected trimonthly for nearly five years, covering the premonsoon, postmonsoon, and winter seasons (February 2013– March 2017). However, the pre-and postmonsoon sampling in 2016-2017 could not be completed due to logistic constraints.

A total of 12 sampling stations - from the river mouth to 13 km upstream ( Figure 1 ), were fixed and reached with the help of a handheld GPS (Garmin Etrex 72H, Taiwan). The geographic location of the stations also represents various land-use land-cover types and habitat functioning in the vicinity. For instance, St. 1 (river mouth), St. 2 (adjacent to fishing harbor), and St. 11 (upstream area) act as migratory bird’s foraging zone (BFZ). Similarly, Sts. 3-8 (adjoining mangroves) are the mangrove-associated zone (MAZ). Intensive shrimp farming (at Sts. 3, 6, and 7) and a factory named Indian Farmers Fertiliser Cooperative Limited (IFFCO) (at St. 8) are also present in the MAZ. The remaining Sts. 9, 10, and 12 adjacent to the country boat berthing facility and Essar Steel India Limited, a leading manufacturer and supplier of steel products and iron ore pellets, stand for the mangrove denuded zone (DZ). In addition, the two bifurcates of River Mahanadi at St. 12 bring considerable freshwater influx at the confluence ( Figure 1 ).

Altogether, 264 sediment samples (in replicates) were collected with the help of a van Veen grab (0.04 m²). After separating a small fraction of the sediment (~25g) for textural and organic matter analyses, each grab sample was transferred to a 500 μm sieve and gently washed with the seawater for macrobenthos. All specimens retained on the sieve were fixed in 5% buffered formaldehyde with 1% Rose Bengal for further processing and identification in the laboratory. Species-level identification of the macrobenthos (where possible) was carried out under a stereomicroscope (Leica, E24W, Germany) by following the standard literature (Fauvel, 1953; Fauchald, 1977; Abbott and Dance, 1982; Subba Rao et al., 1991; Blake et al., 2009; Yokoyama and Sukumaran, 2012; Muir and Hossain, 2014; Hutchings and Kupriyanova, 2018). The World Register of Marine Species (http://www.marinespecies.org/index.php) was followed to validate the scientific names. The faunal density at each station was expressed as ind. m⁻². Further, to estimate the wet biomass (gm^-2), preserved specimens were separated into different groups, kept on a mesh, and then moisture blotted out carefully with absorbent paper and weighed (OHAUS PAJ603 electronic balance).

Sand, silt, and clay (%) compositions in the sediment were determined by the pipette method (Krumbein and Pettijohn, 1938) and assigned the textural classes (Shepard, 1954). The organic matter (OM) (%) was estimated by the modified Walkey-Black wet-oxidation method (Gaudette et al., 1974). In the case of hydrographical parameters, salinity (psu), dissolved oxygen (DO) (mg l^-1), nitrite-nitrogen ( NO 2 - ) (µmol l^-1), and orthophosphate ( PO 4 - ) (µmol l^-1) were estimated by following the standard protocols (APHA, 1989; Grasshoff et al., 1999). Water temperature (using a mercury-filled thermometer of 0.5°C sensitivity), pH (using a Hanna HI 98107 with ±0.1 accuracy), depth (m) (using echo sounder), and transparency (m) (using a Secchi disc) were measured in situ. The instruments were calibrated before recording in situ variables in the field.

The faunal diversity indices such as mean abundance, species richness (S), and Shannon-Wiener index (H’) were computed using the PRIMER v.7 software (Plymouth Routines in Multivariate Ecological Research) (Clarke and Gorley, 2015). The (square root-transformed) data were used for Bray-Curtis similarity (clustering through hierarchical group-average linking) and non-metric multidimensional scaling (nMDS) ordinations. The significance of sample groupings was tested through the Analysis of Similarities (ANOSIM), whereas confirmatory evidence of the faunal assemblages was provided through the Similarity Profile Analysis (SIMPROF) (Clarke and Ainsworth, 1993). The species abundance matrix was represented by a shade plot where the gradation of shade from grey to black is linearly proportional to the increase in species abundance (Clarke and Gorley, 2015). The unconstrained binary divisive clustering (UNCTREE) analysis further examined both sample and species associations. The environmental data were normalized and subjected to Principal Component Analysis (PCA) to distinguish the sampling sites in relation to their (local) environmental conditions. The analysis of variance (ANOVA) tests was used to probe the spatial and temporal differences. The PCA was carried out using PRIMER v.7.

AMBI and M-AMBI indices were used to assess the gradient of anthropogenic stress across the MES. The index scores were derived from the AMBI v.6 software (http://www.ambi.azti.es). AMBI index is a univariate measure that uses a ‘differential weighting’ algorithm based on the classification of benthic species into five Ecological Groups (EGs) (i.e., EGI - species very sensitive; EGII - indifferent to enrichment; EGIII - tolerant to excess OME (organic matter enrichment); EGIV - second-order opportunistic species and, EGV -first-order opportunistic species) (Grall and Glémarec, 1997). The macrobenthos of Mahanadi estuary were assigned to different EGs by following the AMBI v.6 December 2020 taxa list. In addition, a few taxa not identified up to species level or not found in the AMBI database were also assigned for their respective EGs availing the WORMS database. However, 18.7% of total taxa were either not assigned to any EG or ignored due to irrelevant species (cf. Borja and Muxika, 2005).

The scores of the AMBI index were used to categorize the ecological quality of the estuary into five classes based on a scale from 0 to 7 (0-1.2: high, 1.2-3.3: good, 3.3-4.3: moderate, 4.3-5.5: poor and, >5.5: bad). On the other hand, M-AMBI is a multimetric index that derives scores from multiple factors (species richness, Shannon-Weiner diversity index, and AMBI scores). Further, it requires setting a reference condition (Muxika et al., 2007) of a high ecological quality ratio (EQR) of environmental and biological parameters specific to the habitat (Borja et al., 2012). Reference conditions are characterized by high biological and environmental quality elements giving the site a high ecological quality ratio (EQR) compared to the impacted site (Bigot et al., 2008). The Water Framework Directive (WFD) offers four criteria to select reference conditions: (1) pristine or minor disturbance, (2) historical data, (3) predictive modelling, and (4) expert judgment (Basset et al., 2013). Since finding the less disturbed or pristine environment is as difficult as getting the historical data in the era of ‘Anthropocene,’ it is always prudent to perform predictive modelling for setting the reference condition. Therefore, an internal control with high diversity, richness, and low AMBI from the dataset was used to set the reference target, and the M-AMBI scores were calculated as suggested by Muxika et al. (2007). The scores of M-AMBI were used to qualify the samples into five grades based on a scale from 0 to 1 (>0.77: high, 0.77–0.53: good, 0.53–0.38: moderate, 0.38–0.20: poor, and <0.20: bad). Procedures adopted to estimate indices scores, algorithm, boundary limits, and community health/ecological quality classifications were based on the WFD scale of indices detailed by Equbal et al. (2017).

The pairwise Mann–Whitney U test was applied to assess differences on a seasonal scale within the faunal groups. Differences were considered significant at p<0.05 for all power test analyses. Finally, Pearsons’ correlation coefficient test was carried out to find the possible influence of environmental parameters on biotic indices (Graph Pad Software, USA).

The water temperature ranged from 21.5 to 36.0°C (mean, 28.34 ± 1.08°C), with the lowest measurements during winter and the highest during premonsoon ( Table 1 ). The upper reaches were relatively warmer than the lower reaches of the estuary ( Figure 2 ). The water pH was slightly alkaline - especially downstream (>7.5), and varied significantly between the seasons (two-way ANOVA F = 19.89, p < 0.01). With increasing salinity from upstream to downstream, brackish water conditions (9.26 ± 1.87 psu) prevailed along the estuary’s entire (13 km long) stretch. Seasonal salinity changed in the order of winter > premonsoon > postmonsoon. The sampling stations in the proximity of industries (e.g., Sts. 8, 9, 10) showed less DO than those adjoining mangroves (Sts. 5, 6, 7). There was a significant difference in DO between pre-and postmonsoon periods (one-way ANOVA F =4.61, p< 0.05).

The distribution of dissolved nutrients revealed higher concentrations of NO 2 - (3.17 ± 0.75 μmol l^-1) and PO 4 - (4.22 ± 1.06 μmol l^-1) at the stations close to shrimp farms and industries (e.g., Sts. 7, 8, 11). Nutrient enrichment was high during the premonsoon, followed by postmonsoon and winter seasons ( Table 1 ). The sampling stations subjected to higher anthropogenic intervention and upstream fluvial loads (Sts. 7-12) also had a lower water column transparency. The waters are more turbid for postmonsoon (0.35 ± 0.03) than premonsoon (0.55 ± 0.05) or winter (0.95 ± 0.09). Station-wise environmental parameters for each year and season are provided as Supplementary Data ( STable 1 ).

The estuarine sediments were comprised mostly of sand (48.03 ± 6.52) and silt (51.81 ± 6.51) ( Table 1 ), with three distinct textural classes (i.e., silty sand, sandy silt, and silt). The OM ranged from 0.07 to 6.63% and was rather high at the creek/canal confluence locations ( Figure 2 ). The correlation between textural grades and OM was clear. For instance, sediments from the river mouth (St. 1) and upstream (St. 12) with a high sand composition contained less OM in contrast to the mangrove sediments (Sts. 2-9) with a high silt composition. Seasonally, silt was distinctive of the postmonsoon. The sediment textural classes also indicated significant spatial and temporal variations (two-way ANOVA, sand F=3.23, p < 0.05; silt F=3.18, p < 0.05; clay F=10.89, p < 0.01).

The PCA revealed significant eigenvalues (>1), and the percentage of variance was 31.9% for PC1, followed by 26.9% for PC2. While most stations in the MAZ with silt and OM were on the positive side of the two axes, the sites of the DZ with a sand abundance were on the negative side ( Figure 3 ). Overall, the first axis had segregation of stations based on their sediment nature, and the second axis was based on their water characteristics. The stations belonging to BFZ were closely associated with the water quality (salinity, DO, pH, NO 2 - ) ( Figure 3 ).

The present study recorded 100,500 individuals from 73 taxa, 64 genera, and 48 families represented by major faunal groups such as Polychaeta, Brachiopoda, Sipuncula, Crustacea, Mollusca, Echinodermata, and Pisces. However, polychaetes (81.41%), and crustaceans (15.42%) were found to be important in terms of their abundance ( Table 2 ). Within Polychaeta, families such as Spionidae (9 species), Capitellidae (4 species), and Nereididae (4 species) were the most diverse and dominant groups, contributing 48% to the total population. The notable species of polychaetes were nereid Perinereis cavifrons, nephtyid Micronephthys oligobranchia, spionids Dipolydora coeca, Malacoceros indicus, and Polydora cornuta. The important species of crustacean and molluscs included Victoriopisa chilkensis, Psammacoma gubernaculum, and Nassarius stolatus, with their juvenile populations ( Table 2 ).

The macrobenthic faunal abundance ranged between 50 and 3,213 ind. m^-2 (760 ± 727) ( Table 2 ) was high at the BFZ in winter (mean, 2,288 ind. m^-2) and postmonsoon (3,213 ind. m^-2). In particular, the prevalence of P. cavifrons for these two seasons (773-970 ind. m^-2) is noteworthy. The faunal abundance of the BFZ varied significantly between premonsoon and postmonsoon (Mann–Whitney test, p = 0.00), whereas it differed between premonsoon and winter for the MAZ (Mann–Whitney, p = 0.03). The wet weight biomass was low in upstream (e.g., 0.03 gm^-2 at St. 9) than in the downstream areas (25 gm^-2 at St. 2). The macrobenthic faunal abundance data (Polychaeta, Mollusca, and Crustacea) tested for seasonal, species and station wise changes together with their interaction effects revealed significant differences, particularly for the molluscs ( Table 3 ). Station-wise data on macrobenthic faunal abundance for each season were provided in the supplementary information ( STable 2 ).

The abundance and distribution of 49 macrobenthic species at one or more sampling stations in the estuary accounted for ≥1% of the total population. Hierarchical clustering and nMDS plots have shown a clear-cut separation of the sampling stations into three major faunal groups (Global R 0.70; p < 0.002) ( Figure 4 ). Group I, representing the BFZ (Sts. 1, 2, and 11), was dominated by Perinereis cavifrons, Cossura coasta, and Dendronereis aestuariana species (π = 1.07, p < 0.07), whereas Group II, for the MAZ (Sts. 3-8), is characterized by Micronephtys-Victoriopisa-Heteromastus assemblage. The sampling sites of DZ (Sts. 9, 10, and 12) as Group III contained Cossura-Dipolydora-Malacoceros assemblage. The UNCTREE analysis further confirmed the presence of these three faunal clusters (BFZ - Global R: 1, B%: 31.1, π: 2.59; MAZ- Global R: 0.61, B%: 48, π:1.06; DZ - Global R: 1, B%: 64.2, π: 1.07 ( Figure 5 ).

Altogether, 61 macrobenthic species (out of 73) were categorized into their corresponding EGs ( Table 2 ). Except for EGV (one species), the other groups were diverse and represented by 12-17 taxa. The highest faunal (mean) density was contributed by EGIV (36.4%), followed by EGII (20.8%), EGIII (20.5%), EGI (19.5%), and EGV (2.7%). The species such as Heteromastus filiformis, C. coasta, D. coeca (of EGIV), P. cavifrons (EGIII), M. oligobranchia (EGII), and V. chilkensis (EGI) were abundant, and collectively formed 60% of the total population. The faunal composition of MAZ (41.5%) and DZ (38.3%) was dominated by EGIV organisms, while BFZ (36.6%) by the EGIII ( Figure 6 ). Seasonally, both premonsoon (41.7%) and winter (35.3.%) periods were characterized by EGIV and the postmonsoon (35.3.%) by EGIII species.

According to the AMBI scores, the estuary’s ecological quality was good throughout except for St. 8, which had moderate conditions ( Table 4 ). The AMBI scores improved from the premonsoon to the winter period. In this context, the samples with poor to moderate status from MAZ and BFZ (Sts. 2 and 7-9) in the premonsoon were represented by moderate to good status during the postmonsoon and winter ( Figure 6B ). The differences between premonsoon and winter were significant (Mann-Whitney, p = 0.01). Also, the sampling stations in the DZ were depicted with good ecological quality. However, in terms of community health, the estuary was unbalanced and polluted ( Table 4 ).

In contrast to AMBI, the M-AMBI scores classified around 60% of the samples as moderate and 33.3% as good quality. M-AMBI followed a trend similar to that of AMBI, improving sample grades from the premonsoon to the winter period ( Figure 6C ). Except for St. 12 and St. 9 (DZ), where samples were downgraded from good (premonsoon) to moderate (postmonsoon and winter) and moderate (premonsoon and postmonsoon) to poor condition (winter). A unidirectional shift in quality grades was observed at St. 2 (BFZ), where samples upgraded from poor (premonsoon) to moderate (postmonsoon) and good condition (winter). In contrast, the samples from Sts. 6, 7, and 8 (MAZ) and St. 10 (DZ) showed quality improvement only during the winter period ( Figure 6C ). Both spatial and temporal differences in the M-AMBI scores of the BFZ (ANOVA F=9.2, P <0.001) and the MAZ were significant (F=.3, P=0.04).

The Pearson’s correlation coefficients indicate a significant relationship between water quality parameters and benthic (AMBI and M-AMBI) indices, whereas sedimentary attributes did not show any marked association ( Table 5 ). Among the water quality parameters, temperature, pH, and salinity were influential in assessing community health. The distribution of macrobenthos in EGI, EGIII, and EGIV seems to have been governed by temperature, pH, salinity, and DO conditions ( Table 5 ).

As physically controlled ecosystems, estuaries are characterized by wide variabilities in their local environmental conditions. The environment, however, becomes imbalanced and stressed when exposed to increased anthropogenic stressors (Kennish, 2005; Brown et al., 2022; Suzzi et al., 2022).

Among the environmental variables observed for the MES, the water temperature was strongly marked by seasonality. In this case, other atmospheric influences, such as precipitation, nearshore sea temperatures, and river run-offs, are likely to influence its variability (Leal Filho et al., 2022). High water temperature in the premonsoon was evident in the present study. The other traceable stressors like municipal wastes, sewage, and litter from Paradip port city, fisher hamlets, and fishing harbour that enter the estuary through Atharbanki, a mangrove channel, are also found to affect the water quality to a greater extent. Nutrient enrichment at the stations close to BFZ, shrimp farms, and the fertilizer industry (Sts. 1-3 and 6-8) can be explained by avifaunal excrements and untreated effluents. The low pH in the proximity of industries (Sts. 8-9) is attributable to the influx of acidic wastes (Sundaray et al., 2009; Acharyya et al., 2021a; Acharyya et al., 2021b). On the other hand, (acidic) humic substances carried by freshwater in colloidal suspension get coagulated upon meeting the seawater and can shift the pH to an alkaline condition (Beer, 1996). The latter was apparent from the downstream measurements at the Mahanadi estuary and was similar to findings from other estuaries elsewhere (Mohanty, 2018; Habib et al., 2021; Jabir et al., 2021).

The east-flowing rivers in India, the Brahmaputra, Ganges, Irrawaddy, Godavari, Mahanadi, Krishna, and Cauvery, contribute significantly to the total freshwater discharge into the Bay of Bengal (BoB) (from 1.5 × 10¹² m³ to 1.83 × 10¹³ m³ per year) (Varkey et al., 1996; Thadathil et al., 2002). These rivers have a maximum climatological discharge between July and September when the southwest monsoon rainfall is at its peak and decreases gradually to a minimum during the winter and premonsoon (Sandeep and Pant, 2019). Hence, salinity with an increasing gradient towards the mouth of the estuary was low for the wet (monsoon) season and high for the dry (winter) season. Higher DO at Sts. 5-7 of the MAZ could be due to sustained autotrophic production in the waters. This process originates through nutrient-rich outwelling from the adjacent mangroves during inundation regimes (Cohen et al., 2004; Prasad and Ramanathan, 2008). However, land-based sulphate leaching from the gypsum dumps (heaped close to the fertilizer industry), and industrial effluents, besides others, are the drivers of near hypoxia conditions at Sts. 8-10 (cf. Nayak et al., 2018; Taillardat et al., 2020; Sharma et al., 2022). Depletion in DO is a strong indicator of poor water quality (Costa et al., 2018). Such low DO can prolong hypoxia and affect major estuarine ecosystem-level processes, including fluxes and rates linked to carbon and nitrogen biogeochemical cycling. Therefore, maintaining the best possible conditions across the river basins, including coastal and offshore waters, is essential (Karydis and Kitsiou, 2013; Barletta et al., 2019) and crucial for ecosystem functioning and maintaining estuarine biodiversity (Ghosh et al., 2022).

Fine sediment deposition is difficult in the regions where strong water currents persist (Mitchell, 2020). Low fine particle accretion, sandy texture, and lesser OM at seaward (St. 1) and near confluence sites (Sts. 10, 12) support this phenomenon in the study area. Similarly, mud (silt and clay) accruals occurred mainly in sheltered seascapes (mangrove-fringed sites, Sts. 3-8) due to weak tidal currents and land-based fine particle sources. Muddy sediments retained more OM (>2%) than sandy sediments because of the better adsorption capacity of the fine-grained particles (Gaonkar et al., 2021; Haddout et al., 2022). Furthermore, litter from mangrove vegetation (leaves, propagules, and twigs) and subsurface root growth significantly deliver organic carbon to mangrove sediments (Alongi, 1998). Therefore, sites adjoining mangroves revealed higher OM arising from autogenic sources and outwelling (Hossain et al., 2014; Mohanty et al., 2019).

Reliable information on aquatic resources is key to improving their management (Karydis and Kitsiou, 2013). Furthermore, the competence in accommodating changes (episodic or permanent) to managerial plans requires an in-depth understanding of the drivers of water quality deviations and the availability of natural resources at different timescales (Costa et al., 2018). Anthropogenic activities have significantly threatened the Mahanadi estuary’s water quality and biota. As a result, continuous monitoring and potential corrective measures are required to mitigate such effects.

Studies on the macrobenthos of the MES and adjacent mangrove waterways/intertidal mudflats are confined mainly to taxonomic accounts, without much inference to the local environmental conditions (Deb, 1998; Pattanayak and Haldar, 1998; Rao, 1998; Surya Rao and Maitra, 1998; Nayak et al., 2018; Tudu et al., 2018; Mohanty et al., 2022). In recent years, industrial expansion has become disproportionately rapid along major waterways due to well-connected transport hubs and habitation sites, as is the case for the MES. The complex network of creeks and canals acting as vectors and sink for the effluents released from the port town, fishing harbour/landing centers, industrial units, etc. (Rodgers et al., 2020) shows significant consequences on the benthic communities.

Macrobenthic diversity and dominance patterns in the Mahanadi estuary reflected varying degrees of natural and man-induced stressors. Spatial and temporal benthic faunal distribution appeared to be primarily structured by sediment texture, OM, salinity, and pH. For instance, the abundance of polychaetes in the estuary could represent their tolerance to a wide array of (shifting/changing) environmental conditions (Sanchis et al., 2021). The deposit-feeding P. cavifrons is an important link between detritus accumulation and higher trophic levels. As an opportunist, P. cavifrons can repopulate in defaunated areas where the perturbations like dredging (St. 2, opposite to fishing harbour) and mangrove denudation (sites in the proximity of industries) occur. Furthermore, the capitellids, including Heteromastus, spionids, and Magelona spp., exhibit a natural proliferation in fine-muddy sediments (Afli et al., 2008; Sivadas and Ingole, 2016), while Paraprionospio and Magelona spp. survive in hypoxic and lesser competitive environments inhospitable to their predators and competitors (Sivadas et al., 2021). The impoverished faunal trends noticed towards the estuary mouth (St. 1) and upstream (St.12) also confirm the observations of Shirodkar and Nayak (2010). Macrobenthos in the proximity of shrimp farms and industries (IFFCO, Essar Steel) could represent the impacted environmental conditions as pollution indicators (Borja et al., 2000; Dauvin et al., 2012; Albano et al., 2013; Shivarudrappa et al., 2019). Seasonally varied diversity indices are perceptible, with maximum abundance in the winter (Kundu et al., 2010). The findings of this study were consistent with other regional (Mulik et al., 2017; Dias et al., 2018; Rehitha et al., 2019; Sivadas et al., 2021; Subramanian et al., 2021; Kumar et al., 2022; Rehitha et al., 2022) and global (Abrogueña et al., 2021; Bravo et al., 2021; Dauvin et al., 2021; Salimi et al., 2021; Sánchez-Ovando et al., 2021; Kanhai, 2022) benthic evaluations.

An estuary is a dynamic environment where the macrobenthos are adapted to live in widely shifting environmental conditions over relatively short distances (Ortega et al., 2018). The benthic assemblages of Mahanadi estuary are distinct in relation to the local environmental settings. For example, the heterogeneity in Micronephtys-Victoriopisa-Heteromastus assemblage of the MAZ could represent anthropic interventions (through industries, ferry traffic, mangrove destruction, and shrimp farms) modifying the habitat and its preference for the sediment composed of silty-sand and high OM. Also, Cossura-Dipolydora-Malacoceros assemblage of the DZ, under similar man-made disturbances (boat berthing, Essar steel), occupied the sediment with silt and high OM. Therefore, each zone showed the macrofaunal responses to diverse stressors in the vicinity. The effect could be revealed through a low diversity, which testifies to the relationship between biodiversity loss and human-induced disturbances (D’Alessandro, 2020).

Apart from the anthropogenic interventions, monsoon-mediated perturbations also lead to a high degree of substrate heterogeneity and changes in the structure and function of biological communities in the estuary (Sivadas et al., 2011). The benthic community attributes such as abundance, richness, and diversity fluctuate mostly through defaunation, migration, and spawning of species with recovery followed by stable conditions, recruitment, and resettlement (Gaonkar et al., 2013; Sivadas et al., 2021). As a result, tolerant or opportunist species often dominate these communities, exhibiting a highly dynamic population under specific situations (Sivadas et al., 2016). The health assessment of the MES based on the biotic indices demonstrated a high agreement with inferences extracted from benthic community structures against natural or anthropogenic disturbances.

The dominance of P. cavifrons, Heteromastus filiformis, and Dipolydora coeca in organically rich and contaminated sediments consequently changed the ecological quality status of MAZ and DZ from poor to moderate. In contrast, good to moderate conditions at sandy sediments (BFZ) suggest that low organic content is also favourable to EGI and EGIII forms (Sigamani et al., 2015). Overall, the present study affirms both MAZ and DZ as polluted, with benthic communities of tolerant (EGIII) and second-order opportunistic (EGIV) species. The Odisha Pollution Control Board (OSPCB) estimates untreated domestic wastewater discharge from urban settlements in the Mahanadi basin at 3,45,000 m³ (m³ = 1000 liters) per day, producing biochemical oxygen demand (BOD) load of about 68.8 tonnes daily. The water quality index (WQI) of the Paradip coastal waters (riverine and estuarine stretch) also indicated a deteriorating trend from “grade C” in 2013-15 to “grade D” in 2017 (SPCB report card, 2020).

The benthic indices obtained through quality-based evaluation in the present study reflect different stages of the estuarine environmental conditions. Large-scale variability in the environmental parameters and ecological quality status of the benthic habitat at each station (or zone) could be attributable to the heterogeneous MES (Elliott and Quintino, 2007). Temporal variation of macrobenthic communities was evident through either increased or decreased benthic indices depending upon life strategy (r-selected and k-selected) and niche performance (sensitive, indifferent and tolerant) of the individual species (Equbal et al., 2017). The improved sample quality between premonsoon and winter is suggestive of satisfactory performance by the benthic indices during stable, less turbulent, and low seasonal impact periods (Kröncke and Reiss, 2010; Karakassis et al., 2013; Sigamani et al., 2015; Bae et al., 2016; Chan et al., 2016; Feebarani et al., 2016; Sivadas et al., 2016; Mulik et al., 2017). In fact, for many estuaries on the east coast of India, premonsoon exhibit the most stressed conditions with excess and continuous accumulation of contaminants (Mitra et al., 2018). However, a good flush-out of pollutants is only possible during the monsoon season (Mulik et al., 2020a), aided further by a breakdown of contaminants owing to changing redox chemistry and consequences on benthic-pelagic coupling (Ghosh et al., 2022).

The applied biotic indices evaluated the benthic quality status by discriminating the highly disturbed from the less disturbed areas of the estuary. While the robustness of the M-AMBI (compared to AMBI) in sample quality assessment was explained by several researchers (Khan et al., 2014; Equbal et al., 2017; Pandey et al., 2021), a few others have observed its reverse performance due to lack of reference conditions and spatial benthic variability, heterogeneity within the estuarine complexes (Sigamani et al., 2015). According to Borja and Tunberg (2011), both AMBI and M-AMBI are sensitive to detecting human-mediated disturbances modified by natural disturbances. However, for regions strongly influenced by season, coastal dynamics, and anthropogenic disturbances, the univariate AMBI could show a high degree of variability than the multimetric M-AMBI (Kröncke and Reiss, 2010). The temporal variability of AMBI was indeed found to be less than M-AMBI (Equbal et al., 2018). Nevertheless, the marine biotic index, especially the M-AMBI, appeared to be more robust and realistic in grading the habitat quality into different classes. Therefore M-AMBI index can be used as far as the assessment of benthic habitat quality of the MES is concerned.

Considering the Mahanadi estuary’s polluted health status, some holistic management initiatives are required immediately. The government can enforce authority and share responsibilities with local stakeholders, academic institutions, and non-governmental organizations as partners (Romañach et al., 2018). Regular monitoring of the environmental conditions and mangrove cover through advanced research (by observing heavy metal concentrations, application of Remote Sensing and GIS, and others) and citizen science approaches are essential (Zauki et al., 2019; Wolswijk et al., 2020; Durango-Cordero et al., 2021; Gopalakrishnan et al., 2021; Wolswijk et al., 2022). Afforestation of degraded mangrove zones, embankment of tidal rivers - without affecting zonation/succession of the wetland species, and biodiversity conservation (as mandated by the Department of Forest and Environment, Government of Odisha) are some crucial endeavours in the right direction. Moreover, the impact of local pollution on human health should be apprised through appropriate awareness campaigns. The approaches of this kind would be able to lessen the present-day waste discharge, habitat fragmentation, and ecological functioning damages.