2.1. Study area
The Patalganga River, located in the Raigad district of Maharashtra, India, is one of the west-flowing rivers originating from the northern-western ghats near Khopoli village. This region is part of the Western Ghats, a biodiversity hotspot. The river flows into Dharamtar Creek near Kharpada village, creating a unique geomorphological structure. The Patalganga and Amba rivers converge in this creek, which extends towards Uran town. As India is a subtropical country, the riverine system experiences significant seasonal changes. Unlike other tropical regions that receive rainfall throughout the year, the southwest winds bring monsoonal showers to this area for a limited period from June to September. This results in a distinct annual pattern of three seasons: summer (pre-monsoon), rainy (monsoon), and winter (post-monsoon).
In summer, air temperatures in the region can soar above 40°C, impacting the upper stream and intertidal areas by causing high rates of water evaporation, minor changes in salinity, and the drying up of exposed intertidal mudflats. The arrival of the monsoon swiftly alters the weather, counteracting the effects of summer. The monsoon, characterized by a high influx of freshwater, washes large quantities of sediment and organic matter from upstream and adjacent land areas into the estuary, providing essential nutrients for biota. Winter brings relatively mild conditions, with moderate temperatures creating an ideal environment for the growth of various organisms. This season allows a gradual transition from the monsoon to summer, unlike the abrupt change from summer to monsoon.
Additionally, the region is influenced by anthropogenic activities. The Patalganga industrial area, one of the 13 chemical industrial zones developed by the Maharashtra Industrial Development Corporation (MIDC), may discharge treated effluents into the nearby rivers. Consequently, pollutant chemical elements (PCEs) are considered alongside environmental variables in assessing the region's ecological health.
2.2. Sampling locations
In the Patalganga River, three zones were selected based on salinity gradients for this study (Table 1; Fig. 1). The division also considered the distribution of mangrove trees, as their leaf litter could influence total organic carbon levels. Samples were collected during two major seasons: pre-monsoon (April 2015) and post-monsoon (November 2015). Monsoon samples were not collected because the heavy influx of freshwater during this season would dilute the effects of pollutant chemical elements (PCEs), potentially altering the original response of the meiofauna. Three sites were randomly chosen within each zone to capture the maximum salinity variation along the river's length.
Table 1
Details of sampling sites with comment on mangrove features at Patalganga River in three different zone.
Zones | Sub-zones | Latitude (°N) | Longitude (°E) | Depth* (m) | Salinity range** | Mangroves | Remarks |
Zone 1 (Z1) | P1 | 18.83135278 | 72.91396111 | 6.5 | 34.4–35.2 | Sparse | At the Arabian sea, fishing activities |
P2 | 18.83105556 | 72.94255556 | 12 | 34.1–35.1 | Sparse | At the mouth of the creek, few dredging activities |
P3 | 18.80658333 | 72.98408333 | 6.0 | 32.3–34.1 | Sparse | In the creek, fishing activities |
Zone 2 (Z2) | P4 | 18.81494444 | 73.01916111 | 6.0 | 21.0-28.5 | Dense | Near the mouth of the estuary |
P5 | 18.81144444 | 73.05438889 | 4.5 | 14.1–19.1 | Dense | In the middle of the estuary |
P6 | 18.80994444 | 73.07144444 | 4.0 | 10.4–13.7 | Dense | Near splitting zone of the estuary to two arms |
Zone 3 (Z3) | P7 | 18.83866667 | 73.09205556 | 1.5 | 2.3–2.4 | Absent; Terrestrial trees | Surrounded by agricultural land; adjacent to the roadside area |
P8 | 18.84244444 | 73.11405556 | 1 | 0.4–0.8 | Absent; Terrestrial trees | Surrounded by agricultural land; adjacent to road side area |
P9 | 18.84197222 | 73.11955556 | 1 | 0.2–0.8 | Absent; Terrestrial trees | Surrounded by agricultural land; Industrial discharge point; near Apta village |
*maximum depth recorded at full flood (High tide) from P1-P6; **Salinity range recorded during premonsoon and postmonsoon (present study) |
Zone 1 (Z1) is located downstream towards the sea, where mangroves are sparsely distributed. The sites in Z1 are situated in or near the Arabian Sea and are characterized by high salinity. This zone also receives water from the interconnected Amba riverine system and a creek. Zone 2 (Z2) is in the middle of the estuary and features a dense cover of mangroves along both banks of the channel. Zone 3 (Z3) is upstream and characterized by nearly freshwater conditions. It is bordered by terrestrial trees and freshwater weeds along both banks. This zone is also surrounded by agricultural lands and human developments. The study design was structured as 3 × 3 × 2 = 18 (zones × sites × seasons) to ensure comprehensive coverage of the area (spatial), environmental variation associated seasonal changes (temporal).
2.3. Sampling method and processing
2.3.1. Environmental parameters
At each zone, triplicate samples were collected for environmental parameter analysis. Bottom water samples were obtained using a 5L Niskin water sampler to measure various environmental and chemical components, which were then stored in appropriate bottles for further use. Temperature was recorded in situ at the time of sampling with a mercury thermometer accurate to ± 0.1°C. pH was measured using a Thermo Scientific Orion device, calibrated with standard buffer solutions (pH 4, 7, and 10.1). Salinity was analyzed with a Autosal Salinometer (Model: 8400B-50HZ) and measured using the Practical Salinity Scale. Suspended solids were determined gravimetrically by filtering the sample through a Millipore membrane filter with a 0.45 µm pore size. Dissolved oxygen (DO) was analyzed using the methods of Grasshoff et al. (1999).
Undisturbed sediment samples were collected with a van Veen grab, deployed in six replicates at each zone. Three of these replicates were used to collect the upper 0–2 cm of sediment, which was stored in containers for further analysis. Sediment texture (sand, silt, and clay) and organic carbon (Corg) were estimated as environmental parameters. Sediment texture was analyzed according to Buchanan (1984), and organic carbon was measured titrimetrically using the dichromate oxidation method (Walkley and Black 1934).
Pollutant chemical elements (PCEs), including phenol (Phe), heavy metals (Mn, Fe, Co, Ni, Cu, Zn, Hg, Al, and Cr), and petroleum hydrocarbons (PHc), were analyzed to assess the anthropogenic impact on the meiofaunal community. Phenol was estimated using APHA (1998) procedures, and PHCs were analyzed with a Shimadzu RF 5301 Spectrofluorophotometer (IOC-UNESCO 1984). Heavy metals were analyzed following the procedures of Loring and Rantala (1992), using HF, HClO4, HCl, and HNO3 acids (Suprapur grade, Merck). Certified reference standards MESS-3 and PACS-2 were used to validate the method. Sediment was digested with Aqua Regia and oxidized with KMnO4 (US EPA 1983), while mercury (Hg) was analyzed using a Flow Injection Mercury System (FIMS-400, Perkin Elmer, Inc., Shelton, USA).
2.3.2. Meiofaunal samples
Due to the high density of benthic meiofauna in estuarine mud, cores with a diameter of about 1 cm are considered suitable for collection (Somerfield et al. 2005) and have been used in various studies (Nozais et al. 2005; Zhang and Hu 2019). In this study, meiofaunal samples were collected using a handheld Plexiglas corer with a diameter of 3.5 cm and a height of 10 cm. The corer was inserted vertically into the sediment to a depth of 5–6 cm. Meiofaunal assemblages were examined from the upper 5 cm of the sediment layer, as more than 80% of meiofauna is typically found within this depth range (Chen et al. 1999).
Six replicates were collected from each zone and transferred to plastic sediment containers. The samples were then subjected to relaxation, staining, and fixation using 7% magnesium chloride (MgCl₂), 1% rose Bengal stain, and 5% formalin, respectively. In the laboratory, all core samples were sieved through two different mesh sizes: 500 µm and 63 µm. The sediment retained on the 63 µm sieve was examined under a stereomicroscope (Leica S8APO). Meiofauna was sorted to the group level and enumerated. Taxa contributing less than 1% to the mean density were classified as rare taxa.
To describe the meiofaunal community structure, several univariate indices were used, including Margalef's index (d) (Margalef 1968), Pielou's evenness index (J') (Pielou 1969), and the Shannon-Wiener index (H′) (Shannon and Weaver 1949). Additional metrics such as total abundance (N), number of taxa (S), biomass, and the Ne/Co ratio were also analyzed. The Ne/Co ratio, introduced by Raffaelli and Mason (1981) and later validated by Lee et al. (2001), is a useful indicator for assessing heavy metal pollution. In samples lacking copepods, the number of nematodes was used as a substitute for the ratio.
2.3.3. Biomass estimation
For estimation of biomass, less abundant faunal groups were entirely picked out onto slides. Wherever the taxal abundance is high, 30 individuals per group per core were randomly picked out for biomass estimation (Nozais et al. 2005). The length and width of the sorted organisms were then computed using camera-based software IS-Capture, which was pre-calibrated using a standard scale. The length-width measurements were then used to determine the biomass of the organisms following the formulae given by Nozais et al. (2005). Biomass of 10 groups (viz., Nematoda, Ostracoda, Kinorhyncha, Polychaeta, Oligochaeta, Halacaroidea, Nauplii, Tardigrada, Turbellaria) was computed. The remaining groups were excluded from biomass calculations due to less density per core and lack of conversion factors.
2.4. Statistical approach
The environmental data variation was represented overall irrespective of the seasons employing box-plots for each variable and each zone. A principal component analysis (PCA) was performed to show the distribution of environmental parameters. The redundant principal coordinate axes were eliminated by means of less explained variability (less eigenvalue). The first two principal coordinate axes explained the cumulative highest total variance in the scaling act as a proxy for the eliminated ones. PCA was performed separately for different type of datasets to reduce the overlapping of variables on each other due to their difference in values even after transformation, i.e., environmental and pollutant chemical elements (PCEs). Before performing PCA, the datasets of environmental variables and pollutant chemical elements were square-root transformed to increase the symmetry of the data distribution and then normalised. The Pearson correlation was performed between scores of the first two principal components and the environmental variables and PCEs, to understand the significant role of the x- and y-axis in PCA. Univariate and multivariate methods have been applied as benthic assessment tools using the PRIMER v6 software package (Clarke and Gorley 2006) with the PERMANOVA add-on (Anderson et al. 2008). The biotic variables were computed for the three sampling zones Z1, Z2 and Z3, for each replicate and season. The two way-PERMANOVA based on Euclidian distance and Bray-Curtis similarity measures were carried out to test significant differences between the environmental and PCEs, and meiofaunal community structure among zones (Z1, Z2 and Z3), seasons (premonsoon and postmonsoon), and zone×seasons interactions as fixed factors, respectively. The abundance data were log(x + 1) transformed before the analysis. The two way-PERMANOVA based on Euclidean distance was also used to test the significant differences of all the univariate biotic measures (i.e., Total abundance (N), number of taxa (S), Shannon diversity (H'), Pielou's evenness (J'), Margalef's richness (d), Ne/Co ratio and Total biomass). A log (x + 1) transformation of data was computed only for N, Ne/Co ratio and total biomass for significant variation. The significance was computed by 9999 random permutations of sample data among the factor groups. The parallel significance was tested with the Monte Carlo permutation test. For the biological dataset, n-MDS was performed to see the spatiotemporal similarities and difference among the zones. The data were square-root transformed, and then a resemblance matrix was calculated using the Bray-Curtis similarity measure. The clusters formed were then used to identify major meiofaunal taxa's contribution to (dis)similarities among the identified clusters using the similarity percentages (SIMPER) test. BIOENV analysis was performed to test the relationships between environmental variables, PCEs and meiofaunal community (sample×taxa data) and univariate indices of meiofaunal communities (Clarke and Ainsworth 1993). The spearman’s rank correlations were used, and a permutation test (999 permutations) was applied to assess these relationships' significance.