Int. J. Aquat. Biol. (2017) 5(2): 95-107; DOI: ISSN: 2322-5270; P-ISSN: 2383-0956 Journal homepage: www.ij-aquaticbiology.com © 2017 Iranian Society of Ichthyology Original Article Assessment of suitability of macrobenthic mollusc diversity to monitor water quality and shallow sediment quality in a tropical rehabilitated and non-rehabilitated wetland system W.M. Dimuthu Nilmini Wijeyaratne*,1Bellanthudawage Kushan Aravinda Bellanthudawa Department of Zoology and Environmental Management, Faculty of Science, University of Kelaniya, Sri Lanka. Article history: Received 12 January 2017 Accepted 9 April 2017 Available online 2 5 April 2017 Keywords: Biomonitoring Principal Component Analysis Diversity indices Sri Lanka Abstract: Six sampling sites were selected to represent different land use types in the rehabilitated and non-rehabilitated areas of a recreational wetland in Sri Lanka to study the suitability of macrobenthic mollusc diversity to monitor spatial and temporal variation in physico-chemical parameters of water and shallow sediments. Individuals belonging to six families and eight species were recorded during the study. The significantly highest mean abundance (individuals) of Bithynia tentaculata and Pila globosa were recorded in sites from the rehabilitated area and there was no significant temporal variation of mollusc abundance during the study. The abundance and diversity of mollusc community showed significant spatial variations and this study identified that B. tentaculata and P. globosa can be used as possible bioindicators to detect changes in water and shallow sediment quality in tropical wetland ecosystems Introduction Molluscs are a very diverse and species rich phylum that is widely distributed in many types of ecosystems throughout the world. Different species of molluscs perform key functions in ecosystems maintaining the ecosystem balance and material cycling. Also, they serve as important components in the food webs of ecosystems as they have representatives in almost all the trophic levels. In addition, molluscs have been identified as excellent bioindicators due to their relatively immobility or sedentary adult stages, close contact with both bottom sediments and the water column, and ability to respond to many types of anthropogenic impacts, such as nutrient enrichment, oxygen availability, and changes in habitat structure (Bonada et al., 2006; Feio et al., 2007; Mereta et al., 2013). Adult stages of mollusc life cycle mostly occupy the bottom habitats in a wetland and therefore they are always in contact with shallow soft sediments and overlying water column. This allows them to respond quickly to any physical or chemical change in these environments. Therefore, the temporal and * Corresponding author: W.M. Dimuthu Nilmini Wijeyaratne DOI: https://doi.org/10.22034/ijab.v5i2.288 E-mail address: dimuthu.wijeyaratne@kln.ac.lk spatial variation of abundance, composition, and distribution of molluscs provide an index of the ecosystem (Barbour et al., 1999). The wide distribution, ecological and ecotoxico- logical importance of the molluscs in ecosystem monitoring have attracted a lot of attention. Several biomonitoring studies have indicated that river bed compaction due to agricultural practices, eutrophic- cation and different types of pollutants as causal factors for decline of freshwater molluscs species (Hartmut and Gerstmann, 2007). The adult and juvenile stages of freshwater mussel populations (Villosa iris and Anodonta grandis) were recorded to be declining due to activities such as dredging, agricultural runoff, industrial pollution and sedimentation (Jacobson et al., 1993). The accumulation of chromium (Cr) and cadmium (Cd) in soft parts of bivalve Caelatura companyoi and the gastropod snail Cleopatra bulimoides have showed the suitability of these species as suitable bioindicators of heavy metal pollution (Moloukhia and Sleem, 2011). In addition, juvenile stages of Lampsilis cardium has 96 Wijeyaratne and Bellanthudawa/ Use of macrobenthic diversity as bioindicator in a tropical wetland system been used for assessment of pore water ammonia in several regions of USA (Bartsch et al., 2003). Heavy metal contamination of wetland ecosystems has also been assessed using bivalves such as mussels and oysters in temperate waters (Paez-Osuna et al., 1995; Riget et al., 1997). Also some recent ecological studies have used the presence-absence and abundance data of the gastropod Patella caerulea to assess heavy metal contamination in the Iskenderun Bay (Turkmen et al., 2005; Yuzereroglu et al., 2010). Gastropod and pelecypod species were identified as ideal bio-indicators for trophic stages classification of lentic and lotic environments (Clarke, 1979; Choubisa, 1992). The gastropod Corbicula fluminea has been used in Taiwan to study Arsenic (As) pollution via a biological early warning system in wetland health assessment. In Malaysia, Melonides tuberculata a freshwater mollusc has been utilized to assess the acute toxicity of Cu, Cd, Zn, Pb, Ni, Fe, Al, and Mn (Shuhaimi-Othman et al., 2012). Although the molluscs are considered as an important bioindicator in most different ecosystems in the world, in Sri Lanka, mollusc species are less popular as bioindicator organisms compared to fish and other bioindicator species. There are very few published research on use of mollusks to trace ecotoxicity in the aquatic systems in Sri Lanka. In a study in Bolgoda canal and Waga stream, Paludomus loricatus, P. zeylanicus, Planorbella trivolvis, Promacea bridgesi, and Cerithiidae sp. have shown significant relationships with the variation of physicochemical parameters (Idroos and Manage, 2012). A comprehensive study done in the stretch in the Dutch canal, Sri Lanka, highlighted the importance of Melanoides tuberculata, Thiara acanthica, Faunas ater, and Neritina perottetiana as bioindicators (Gamlath and Wijeyaratne, 1997). Also another study conducted in the Negombo estuary to evaluate the diversity and distribution of microbenthic community with special reference to changing environmental variables indicated that the molluscs species such as Cerithidea cingulate, Dentalium sp., Hydrobiid sp., Terebralia palustris, Faunus ater, Thiarid sp., and several Venerid sp., Tellind sp., Mytilid sp., are applicable as a good bioindicators in wetland health assessments (Dahanayaka and Wijeyaratne, 2010). The present study was conducted in the Diyawwannawa wetland system which is a very popular recreational wetland located in the commercial capital of Sri Lanka. This wetland system is located in a very urbanized environment and part of the wetland is maintained under pristine conditions, while other part is a rehabilitated area surrounding the Parliament of Sri Lanka. The rehabilitated part of the wetland is dedicated for recreational purposes and effects of human disturbances in the rehabilitated area is higher compared to the natural wetland area. Due to the human interferences, there can be changes in the water and sediment quality of the rehabilitated area of the wetland system compared to the natural area and these changes may have significant effects on the associated sediment community structure. However, no previous research have been conducted to compare the changes of water and sediment quality and benthic community structure of these two areas of the Dyawannawa wetland system. Therefore, the present study was conducted to examine differences of water and shallow sediment quality parameters in the rehabilitated and natural areas of the Diyawannawa wetland system. In addition, as diversity of molluscs is closely related to the sediment and water quality of their associated habitats, hence, this study focused on effects of differences in water and sediment quality parameters on the diversity and abundance of macrobenthic molluscs in the Diyawannawa wetland system. Materials and Methods Study area: Diyawannawa wetland is one of the most popular urban wetlands in Colombo district Sri Lanka. This wetland is composed of rehabilitated and non- rehabilitated areas. In the rehabilitated areas of the wetland, some parts are dredged, banks are restored and the aquatic vegetation is removed using mechanical removal methods to improve the water retention capacity of the wetland. In the non– rehabilitated area of the wetland, the human 97 Int. J. Aquat. Biol. (2017) 5(2): 95-107 interventions on the wetland ecosystem are minimum and this area is rich in biodiversity and provides habitats for several threatened and endemic species in Sri Lanka. The present study was carried out from April to December 2016. Six sampling sites were selected from the study area. The location of the sampling sites in the Diyawannawa wetland is shown in Figure 1. Site A (06°54'.585''N, 079°54'.722''E), site B (06°54'- .664''N, 079°54'.633''E) and site C (06°54'.609''N, 079°54'.604''E) were located in the non-rehabilitated area of the wetland. Site D (06°54'.68''N, 079° 54'.610''E), site E (06°54'.751''N, 079°54'.735''E) and site E (06°54'.741''N, 079°54'.525''E) were located in the rehabilitated area. Sites A is characterized by thick vegetation cover in the wetland banks as well as presence of high density of floating aquatic macrophytes. Sites B and C are characterized by having a thick vegetation cover in the wetland banks and site D was characterized by absence of vegetation cover in the banks and presence of thick growth of aquatic macrophytes. Site E was receiving runoff from small scale poultry farms located within 100 meters of the wetland. Site F was located in a highly urbanized area. Water and sediment quality parameters: From each site, water samples and shallow sediments samples (0- 0.5 m depth) were collected. Sampling was carried out once in 6 weeks for a period of 7 months from April to December in 2016. At each sampling sites, water pH, temperature, conductivity, total dissolved solids (TDS) and salinity were measured in-situ using a calibrated digital multi parameter (YSI Environmental Model-556 MPS). Dissolved oxygen concentration (DO) was measured using DO meter (HQ 40b model- Hach). Visibility was recorded using a secchi disk. The biological oxygen demand 5 days after incubation (BOD5), chemical oxygen demand (COD), nitrate concentration, chlorophyll 𝛼 concentration and total and dissolved phosphorus concentrations were measured based on APHA (1992). At each sampling site, sediment pH was measured in-situ using the calibrated digital multiparameter (YSI Environmental Model-556 MPS). Sediment organic matter content was measured in the laboratory using the loss on ignition method and the percentage sand, silt and clay content of the sediments were measured using the sedimentation jar. Water and sediment quality parameters were measured in triplicate within a 1 m2 area and averaged for each sampling site. Benthic mollusc diversity: Benthic mollusc sampling was carried out using the Peterson grab sampling method (Dahanayake and Wijeyaratne, 2006). The collected samples were preserved in 5% Rose bengal solution on site and were transported to the laboratory. In the laboratory at the University of Kelaniya, Sri Lanka, the samples were subjected to wet sieving through 4 mm, 2 mm and 1 mm mesh sieves to separate benthic molluscs. The organisms retained in each sieve were collected and preserved in 10% formalin and after a week they were transferred into 10% ethyl alcohol to prevent dehydration as described Figure 1. The study sites in the Diyawannawa wetland, Sri Lanka. Sites A, B and C are located in the non-rehabilitated area and sites D, E and F are in the rehabilitated area. 98 Wijeyaratne and Bellanthudawa/ Use of macrobenthic diversity as bioindicator in a tropical wetland system by Dahanayaka and Wijeyaratne (2006). Molluscs were identified to family level using keys given by Fernando and Weerawardhena (2002). The number of individuals of each taxon and the number of families at each sampling site were counted. Using the species abundance data, species richness, taxa richness (TR), species density, Simpson’s diversity index (D), Shannon-Wiener diversity index (H’), Pielou's evenness index (J), Modified Hilsenhoff Biotic index and percent contribution of dominant family were calculated for each sampling site at each sampling occasion and data were averaged over time for six sampling sites. Statistical analysis: After confirming the normality using Anderson darling test, the data were analyzed using one way ANOVA followed by Tukey’s pairwise comparison to determine the significance of the spatial and temporal variation of selected water quality and sediment quality parameters in the rehabilitated and non-rehabilitated sites of the Diyawannawa wetland. A Principal Component Analysis (PCA) was used to determine water and sediment quality parameters and diversity and biotic indices that describes the distribution of benthic mollusks in the Diyawannawa wetland. MINITAB 14 statistical software package was used in the statistical analysis. Results Spatial variation of water quality parameters in sampling sites is given in Table 1. Temporal variation of shallow bottom water quality parameters is given in Table 2. There was no spatial variation in temperature, salinity and total phosphorous (P>0.05). Site A of the non-rehabilitated area showed significantly higher water pH, conductivity and TDS. However, this site showed significantly lower COD, DO and BOD5 (P<0.05). The site B of the non-rehabilitated area and all the sites in the rehabilitated area showed significantly higher chlorophyll 𝛼 concentration throughout the study period (P<0.05). There was no significant temporal variation in temperature, DO and salinity (P>0.05). However, the other physico- chemical parameters showed significant temporal fluctuations throughout the study period (Table 2). The spatial variation of shallow sediment quality parameters in each sampling site is given in Table 3. The temporal variation of shallow sediment quality parameters is given in Table 4. The pH, percentage total organic carbon (TOC) and the percentage silt content of the shallow sediments did not show significant spatial variations (P>0.05), but the percentage sand content at sites E and F of the rehabilitated area were significantly lower and Table 1. Spatial variation of water quality parameters at the sampling sites of the Diyawannawa wetland during the study period. The results are presented as mean±Standard Error of Mean. Different superscripts in each row indicate statistically significant differences (P<0.05). Parameter Non-rehabilitated area Rehabilitated area Site A Site B Site C Site D Site E Site F pH 5.98±0.3b 7.54±0.2a 7.63±0.3a 7.78±0.2a 7.38±0.2a 8.05±0.3a Visibility (Cm) 42.7±4.0a 40.2±4.4a 37.7±1.5a 43.3±2.6b 30.8±1.7bc 23.76±1.2c Temperature (oC) 30.74±0.4a 31.34±0.3a 31.44±0.3a 31.54±0.4a 31.75±0.5a 32.14±0.3a Conductivity (µs/cm) 345.5±10.5a 253.7±9.7b 271.6±14.6b 252.4±8.3b 248.69±4.8b 245.23±7.4b Depth (cm) 85.5±3.2bc 118.6±9.5a 98.9±5.9ab 119.4±2.5a 85.7±6.2bc 62.7±6.8c TDS (mg.L-1) 166.25±5.0a 121.68±4.7b 129.98±7.1b 120.47±4.0b 109.90±5.3b 116.84±3.6b DO (mg.L-1) 2.82±0.09a 6.84±0.4b 7.68±0.4b 7.81±0.4b 10.61±0.2c 10.28±1.0c Salinity (o/oo) 0.16±0.004 a 0.12±0.004a 0.12±0.006a 0.12±0.004a 0.11±0.006a 0.12±0.003a BOD5 (mg.L -1) 1.20±0.5c 2.13±0.5c 3.85±0.3b 3.79±0.3b 6.56±0.3a 5.13±0.3ab Total phosphate (mg.L-1) 0.02±0.004a 0.02±0.003a 0.02±0.004a 0.03±0.005a 0.03±0.006a 0.04±0.008a Nitrate (mg.L-1) 0.01±0.001a 0.04±0.001b 0.02±0.002a 0.02±0.003a 0.04±0.004b 0.04±0.002b Chlorophyll-a (mg/dm3) 2.05±0.2a 11.45±1.7b 2.40±0.6a 10.61±0.5b 12.13±2.6b 12.42±1.9b COD (mg.L-1) 173.1±36.4a 305.3±44.7b 285.5±46.9b 384.9±38.8b 387.5±38.2b 454.8±30.1b 99 Int. J. Aquat. Biol. (2017) 5(2): 95-107 percentage clay content of these sites were significantly higher compared to the other sites (Table 3). The sites A and B of the non-rehabilitated area showed significantly lower sediment conductivity compared to other sites (P<0.05). There was no significant temporal variation of the percentage TOC, conductivity and pH of the shallow sediments of the study sites (P>0.05). However, the percentage sand and silt contents showed significant decrease and percentage clay content of the sediments showed a significant increase during September and October compared to other months (P<0.05). The spatial variation of mean abundance of macrobenthic mollusk species in each sampling site is given in Table 5. The temporal variation of mean abundance of macrobenthic mollusc species during the study period is given in Table 6. In the present study, a total of 8 freshwater macrobenthic molluscs, Table 2. Temporal variation of water quality parameters at the sampling sites of the Diyawannawa wetland during the study period. The results are presented as mean±Standard Error of Mean. Different superscripts in each row indicate statistically significant differences (P<0.05). Parameter April May June September October pH 7.48±0.2a 6.13±0.2b 6.37±0.18b 8.73±0.12c 8.24±0.14c Visibility (cm) 31.7±3.2a 48.0±3.4b 34.3±3.1a 36.0±1.3a 32.0±1.9a Temperature (oC) 33.82±0.16a 30.78±0.19a 29.91±0.18a 30.59±0.11a 32.37±0.24a Conductivity (µs/cm) 278.11±4.8ab 221.28±8.6c 254.3±12.7bc 289.4±11.8ab 304.50±9.8c Depth (cm) 116.2±5.8a 114.4±6.5b 79.9±5.9b 81.6±6.1b 83.5±6.7b TDS (mg.L-1) 133.57±2.52a 98.20±4.75b 121.49±6.13a 138.88±6.05a 145.45±4.72a Dissolved Oxygen (mg.L-1) 8.05±0.8a 6.90±0.7a 8.14±0.8a 7.26±0.5a 9.00±0.7a Salinity (o/oo) 0.13±0.003 a 0.09±0.005a 0.12±0.006a 0.13±0.006a 0.14±0.004a BOD5 (mg.L -1) 2.29±0.4a 3.57±0.5ab 3.94±0.4ab 3.72±0.5ab 5.37±0.4b total phosphate (mg.L-1) 0.009±0.0005a 0.011±0.001a 0.009±0.0009a 0.055±0.004b 0.044±0.004b Nitrate (mg.L-1) 0.035±0.003a 0.031±0.004ab 0.027±0.003ab 0.024±0.003ab 0.021±0.002a Chlorophyll-a (mg/dm3) 8.29±1.0a 5.961±0.8a 4.578±0.7a 14.83±2.8b 4.571±0.6a COD (mg.L-1) 504.6±16.9a 445.6±42.3a 288±20.1b 253.2±37.6b 167.7±19.5c Table 3. Spatial variation of shallow sediment quality parameters at the sampling sites of the Diyawannawa wetland during the study period. The results are presented as mean±Standard Error of Mean. Different superscripts in each row indicate statistically significant differences (P<0.05). Parameter Non-rehabilitated area Rehabilitated area Site A Site B Site C Site D Site E Site F Percentage sand content 44.6±9.7a 54.3±11.9a 48.7±10.6a 42.7± 9.4a 7.3±1.6b 5.5±1.2b Percentage silt content 9.9±2.1a 3.9±1.2a 8.1±1.7a 13.6± 3.1a 11.9±2.6a 13.1±2.8a Percentage clay content 45.5±11.9a 41.8±12.7a 43.3±12.4a 43.8± 12.3a 80.76±4.20a 81.50±4.04a Percentage TOC 12.42±0.01ab 12.34±0.02b 12.36±0.03b 12.35± 0.02b 12.48±0.03a 12.48±0.03a Conductivity (µs/cm) 47.98±0.6c 43.80±0.7c 70.87±2.5b 74.49± 2.4b 77.05±1.5b 91.44±1.2a pH 6.19±0.08a 5.76±0.24a 6.17±0.11a 6.12± 0.12a 6.23±0.13a 6.31±0.10a Table 4. Temporal variation of shallow sediment quality parameters at the sampling sites of the Diyawannawa wetland during the study period. The results are presented as mean±Standard Error of Mean. Different superscripts in each row indicate statistically significant differences (P<0.05). Parameter April May June September October Percentage sand content 55.0±7.6a 57.2±8.2a 56.8±8.2a 3.41±0.01b 3.59±0.02b Percentage silt content 18.4±1.3a 16.2±1.5a 15.9±1.5a 6.32±0.01b 5.86±0.02b Sediment clay % 26.74±7.25b 26.49±7.15b 27.24±7.14b 90.3±0.01a 90.6±0.02a Shallow sediment TOC% 12.35±0.02a 12.43±0.03a 12.455±0.03a 12.42±0.03a 12.37±0.03a Shallow sediment conductivity (µs/cm) 68.24±4.4a 68.39±4.4a 68.22±4.2a 68.01±4.04a 65.17±4.5a Shallow sediment pH 6.13±0.23bc 5.75±0.02c 5.83±0.05c 6.32±0.06ab 6.60±0.05a 100 Wijeyaratne and Bellanthudawa/ Use of macrobenthic diversity as bioindicator in a tropical wetland system species belonging to 6 families were recorded. The mean abundance of Bithynia tentaculata was significantly higher in site F in the rehabilitated area (P<0.05, Table 5). Only B. tentaculata and Melanoides turbeculata were recorded from site A of the non-rehabilitated area and the recorded numbers of these two species in this site were significantly lower compared to other study sites (Table 5). Pila globosa was recorded only at sites E and F of the rehabilitated area (Table 5). There was no significant temporal variation of the mean abundance of freshwater macrobenthic molluscs during the study period (Table 6). The Shannon wiener diversity index, Simpsons index, modified biotic index and Pielou’s evenness index for each sampling site are given in Table 7. The lowest values for Simpsons diversity index, Shannon wiener diversity index and Pielou’s evenness index were recorded from site F in the rehabilitated area (Table 7). The highest values for Simpsons diversity index, Shannon wiener diversity index and modified biotic index were recorded from the Site D in the Table 5. Spatial variation of the mean abundance of freshwater macrobenthic molluscs species in each sampling sites of the Diyawannawa wetland during the study period. Different superscripts in each row indicate statistically significant differences (P<0.05). Species Name Non-rehabilitated Area Rehabilitated Area Site 1 Site 2 Site 3 Site 4 Site 5 Site 6 Bithynia tentaculata 3d 3d 8cd 15bc 17b 35a Melanoides turbeculata 0b 0b 1b 4a 1b 5a Melanoides turriculus 1b 2ab 4ab 5a 2ab 3ab Thiara scabra 0b 1b 2b 10a 1b 3ab Lamellidens marginalis 0c 1c 1c 2b 0c 3a Pila globosa 0b 0b 0b 0b 1b 2a Gyraulus saigonensis 1b 1b 1b 1b 3a 2ab Lymnaea stagnalis 1b 0b 0b 1b 3a 1b Table 6. Temporal variation of the mean abundance of freshwater macrobenthic molluscs species in each sampling sites in the Diyawannawa wetland during the study period. Different superscripts in each row indicate statistically significant differences (P<0.05). Species Name April May June September October Bithynia tentaculata 8a 12a 18a 15a 15a Melanoides turbeculata 3a 2a 3a 2a 1a Melanoides turriculus 4a 3a 3a 3a 2a Thiara scabra 7a 2a 2a 2a 1a Lamellidens marginalis 1a 1a 1a 1a 1a Pila globosa 1a 1a 1a 1a 1a Gyraulus saigonensis 2a 1a 1a 2a 2a Lymnaea stagnalis 1a 1a 1a 1a 1a Table 7. Spatial variation of diversity and biotic indices of macrobenthic molluscs in the Diyawannawa wetland during the study period. Site Simpson’s Diversity Index (D) Shannon-Wiener Diversity Index (H’) Modified Biotic Index Pielou's Evenness Index A 0.613 1.1376 7.371 1.890 B 0.635 1.2033 7.160 1.721 C 0.660 1.3116 7.093 1.686 D 0.708 1.4177 6.873 1.678 E 0.506 1.1167 7.619 1.321 F 0.472 1.0978 7.548 1.216 Table 8. The percentage dominance of dominant macrobenthic mollusc family in each study site during the study period. Site. Dominant family % Dominance A Bithyniidae 55.71 B Bithyniidae 50.62 C Bithyniidae 50.93 D Thiaridae 53.50 E Bithyniidae 68.66 F Bithyniidae 71.56 101 Int. J. Aquat. Biol. (2017) 5(2): 95-107 rehabilitated area (Table 7). The diversity and biotic index values of the sites in the non-rehabilitated area showed similar values to each other (Table 7). The percentage dominance of dominant macrobenthic mollusc family in each study site during the study period is given in Table 8. All the study sites except Site D were dominated by the members of the family Bithyniidae and Site D was dominated by the family Thiaridae (Table 8). PCA showed the categorization six sampling sites Figure 2. Ordination of the study sites based on PC1 and PC2 scores of PCA of the physico-chemical parameters of over lying water and sediments of the study sites in the Diyawannawa wetland (Sites A, B and C=non-rehabilitated area and sites D, E and F=rehabilitated area). Figure 3. Ordination of the study sites based on PC1 and PC2 scores of PCA of the abundance of macrobenthic molluscs of the study sites in the Diyawannawa wetland. (Sites A, B and C=non-rehabilitated area and sites D, E and F=rehabilitated area). 102 Wijeyaratne and Bellanthudawa/ Use of macrobenthic diversity as bioindicator in a tropical wetland system according to the water and sediment quality parameters. Two principal components displaying a cumulative variance of 85.8% were obtained after applying PCA on water and sediment quality parameters. PCA score plot for variation of water quality and sediment quality parameters among the study sites in the Diyawannawa wetland is given in Figure 2. The eigenvalues of the first two principal components, eigenvectors of the water and sediment quality variables and the principal component scores for the study sites are given in Table 9. According to the results of the PCA on water quality and sediment quality parameters, the sites C and B of the non- rehabilitated area and site D of the rehabilitated area grouped together. Sites E and F of the rehabilitated area were grouped together and site A of the non- rehabilitated area was separated from the other groups (Fig. 2). Two principal components displaying a cumulative variance of 99.3% were obtained after applying PCA on total abundance of macrobenthic molluscs. PCA score plot for variation of total abundance of macrobenthic molluscs among the study sites in the Diyawannawa wetland is given in Figure 3. The eigenvalues of the first two principal components, eigenvectors of the total abundance of macrobenthic molluscs and the principal component scores for the study sites are given in Table 10. The results of the PCA on total abundance of macrobenthic molluscs also categorized six sampling sites into three groups. The Site F of the rehabilitated area was separated from other study sites. The three sampling sites of the non- rehabilitated area were grouped together and sites E and F of the rehabilitated area formed a separate group Table 9. Summary of the PCA of physico-chemical parameters of water and shallow sediments of the study sites at the Diyawannawa wetland. Cumulative % variation of only the PC1 and PC2 are shown. A high cumulative percentage as high as 85.8% of the total variation among physico- chemical parameters are explained by PC1 and PC2 axis. Eigenvalues PC Eigenvalues %Variation Cum.%Variation 1 11.93 62.8 62.8 2 4.38 23.1 85.8 Eigenvectors (Coefficients in the linear combinations of variables making up PC's) Variable PC1 PC2 PC3 PC4 PC5 Water pH -0.214 0.263 -0.204 0.145 -0.388 Temperature -0.281 0.068 0.055 0.069 -0.252 EC 0.229 -0.290 -0.057 -0.026 0.021 TDS 0.240 -0.258 -0.018 -0.044 -0.187 DO -0.277 0.104 0.026 0.186 0.144 BOD5 -0.264 -0.023 0.080 0.241 0.470 COD -0.274 0.100 0.123 -0.197 -0.143 Nitrate -0.213 0.184 -0.459 -0.046 -0.029 Chlorophyll a -0.216 0.189 -0.180 -0.575 0.075 Total Phosphorus -0.255 -0.117 0.116 -0.384 -0.289 Visibility 0.249 0.137 0.253 -0.225 0.303 Depth 0.135 0.380 0.225 -0.187 0.322 %TOC -0.177 -0.337 -0.274 -0.046 0.191 Sediment pH -0.128 -0.386 -0.290 -0.192 -0.058 %sand 0.249 0.218 0.141 0.088 -0.177 % Silt -0.157 -0.257 0.453 -0.373 0.179 %clay -0.248 0.194 -0.241 -0.028 0.160 Sediment conductivity -0.251 -0.115 0.334 0.147 -0.134 Principal Component Scores Sample SCORE1 SCORE2 SCORE3 SCORE4 SCORE5 Site A(Non Rehabilitated Site) 4.871 -2.879 -0.373 -0.363 0.033 Site B(Non Rehabilitated Site) 1.797 3.130 -1.451 -0.261 -0.250 Site C (Non Rehabilitated Site) 1.207 0.448 0.839 1.573 -0.170 Site D (Rehabilitated site) -0.071 1.198 1.957 -0.915 0.169 Site E (Rehabilitated site) -3.286 -0.513 -0.683 0.156 1.208 Site F (Rehabilitated site) -4.519 -1.384 -0.289 -0.181 -0.990 103 Int. J. Aquat. Biol. (2017) 5(2): 95-107 on the PCA plot (Fig. 3). The results of the Pearsons’ correlation analysis on the total abundance of macrobenthic molluscs with selected water quality parameters is given in Table 11. The abundance of B. tentaculata, Lamellidens marginalis, M. turbeculata and P. globosa showed significant positive correlations with sediment conductivity. In addition, the abundance B. tentaculata showed significant positive correlations with percentage clay content and significant negative Table 10. Summary of the PCA of total abundance of macrobenthic molluscs of the study sites at the Diyawannawa wetland. Cumulative % variation of only the PC1 and PC2 are shown. A high cumulative percentage as high as 99.3% of the total variation among physico-chemical parameters are explained by PC1 and PC2 axis. Eigenvalues PC Eigenvalues %Variation Cum.%Variation 1 7.58 94.7 94.7 2 0.36 4.6 99.3 Eigenvectors (Coefficients in the linear combinations of variables making up PC's) Variable PC1 PC2 PC3 PC4 PC5 Bithynia tentaculata -0.326 0.731 -0.037 -0.153 0.531 Melanoides turbeculata -0.360 -0.199 0.176 0.282 0.288 Melanoides turriculus -0.351 0.372 0.545 0.138 -0.650 Thiara scabra -0.362 0.002 -0.316 -0.666 -0.209 Lamellidens marginalis -0.361 -0.096 -0.374 0.267 -0.116 Pila globosa -0.357 -0.298 -0.185 -0.209 -0.178 Gyraulus saigonensis -0.362 -0.016 -0.320 0.526 0.030 Lymnaea stagnalis -0.348 -0.435 0.543 -0.206 -0.350 Principal Component Scores Sample SCORE1 SCORE2 SCORE3 SCORE4 SCORE5 Site A(Non Rehabilitated Site) 2.590 -0.059 -0.051 0.040 0.039 Site B(Non Rehabilitated Site) 2.550 -0.016 0.012 0.056 -0.036 Site C (Non Rehabilitated Site) 2.269 0.373 0.065 -0.095 -0.003 Site D (Rehabilitated Site) -2.098 -0.751 0.327 -0.005 0.002 Site E (Rehabilitated Site) -1.935 -0.486 -0.387 -0.022 -0.005 Site F (Rehabilitated Site) -3.375 -0.939 0.033 0.026 0.002 Table 11. Pearsons’ Correlation coefficients between the abundance of fresh water molluscs species and shallow sediment quality parameters (* indicates significant correlations at 95% level of significance). Species Shallow sediment TOC Shallow sediment pH Shallow sediment conductivity Sand % Clay % Silt % Bithynia tentaculata .327 .134 .645* -.445* .365* .133 Gyraulus saigonensis .120 .282 .310 -0.341 .314 -.050 Lamellidens marginalis .216* .126 .497* -.267 .231 .023 Lymnaea stagnalis .665* .154 .109 -.226 .189 .048 Melanoides turbeculata 0.195 0.063 0.511* .005 -.092 .392 Melanoides turriculus -.062 .057 .148 .156 -.187 .221 Pila globosa .465* .144 .468* -.332 .285 .042 Thiara scabra -.028 .140 .166 .122 -.178 .310 Table 12. Criteria for evaluation of water quality using the family-level biotic index (Hilsenhoff, 1988). Biotic index Value Water quality Degree of organic pollution 0.00-3.50 Excellent No apparent organic pollution 3.51-4.50 Very good Possible slight organic pollution 4.51-5.50 Good Some organic pollution 5.51-6.50 Fair Fairly significant organic pollution 6.51-7.50 Fairly poor Significant organic pollution 7.51-8.50 Poor Very significant organic pollution 8.50-10.0 Very poor Severe organic pollution 104 Wijeyaratne and Bellanthudawa/ Use of macrobenthic diversity as bioindicator in a tropical wetland system correlation with percentage sand content of the sediments. Pila globosa and Lymnaea stagnalis showed significant correlations with percentage total organic carbon content of the shallow sediments (Table 11). Discussion The distribution of moluscs in aquatic ecosystems are influenced by various ecological features of the sediments an overlying water column. Sediment characteristics are an important aspect of physical habitat in aquatic ecosystems. Sediments are the primary component of the substrate upon which macrobenthic molluscs move, rest, shelter, and feed. Therefore changes in the sediment characteristics can directly influence the mollusk community structure of the ecosystem. Molluscs are commonly used for biological monitoring of freshwater ecosystems in many parts of the world as they have limited mobility and are quite easy to collect using standard sampling techniques (Hellawell, 1986; Abel, 1989). Begon et al. (2006) stated that in a healthy aquatic community, the evenness index ranges between 0.0 and 2.0, and the Shannon-Wiener’s diversity index ranges between 1.5 and 3.5. During the present study in each site, Shonnon-Wiener Diversity index values were ranged in between 1.09- 1.42 and Pioulos’ eveness indices values ranged between 1.2-1.9 indicating less variation of diversity and a healthy aquatic community in both non rehabilitated and rehabilitated areas of the Diyawanawa wetland system. Further, the significantly high value for Pielou’s Evenness Index in sampling sites in non- rehabilitated area compared to the rehabilitated area reflect that there is an evenly distributed, less diverse freshwater macrobenthic molluscs community in rehabilitated area compared to the non-rehabilitated area. However, among the species of molluscs recorded during the study, the number of individuals of B. tentaculata collected from the rehabilitated sites were significantly higher compared to that of the non- rehabilitated sites during the study period. The reason may be that the environmental parameters of the rehabilitated area may be significantly favorable for the survival and distribution of B. tentaculata. According to the Pearsons’ correlation analysis the abundance of B. tentaculata showed significant positive correlations with sediment clay content and sediment conductivity. The PCA analysis based on water and shallow sediment quality parameters of the present study categorized sites E and F of the rehabilitated area together and these sites were characterized by high sediment clay content and high sediment conductivity. According to the PCA on abundance of macrobenthic molluscs, site F of the rehabilitated area was separated from the other study sites and B. tentaculata was the responsible species for the separation of site F. Therefore, the results of the present study agrees with studies conducted by Kołodziejczyk (1984) and Savage and Gazey (1987) indicating that there is a statistical significant relationship with bottom sediments clay percentage and organic matter with the abundance of B. tentaculata in lentic ecosystems. In the present study, sites D and E of the rehabilitated area were grouped together and was characterized by P. globosa and L. stagnalis. Further, P. globosa and L. stagnalis showed significant correlations with percentage total organic carbon content of the shallow sediments. This finding agrees with a study conducted in a tropical aquatic system in Bangladesh indicating tolerance of these species to organic enrichment and poor water quality (Badruzzaman et al., 2007). In addition, according to the PCA on water and sediment quality parameters the sites A and B of the non-rehabilitated area and site D of the rehabilitated area were grouped together and were characterized by high visibility and high percentage sand content in the sediments. The sites A, B and D were located in close proximity to each other and this may have caused these sites to share common physical parameters. In the PCA of abundance of macrobenthic molluscs, the sites (A, B and C) in the non-rehabilitated area were grouped together and were characterized by abundance of M. turriculus. This shows that M. turriculus is a pollution sensitive species and therefore restricted to non-rehabilitated area. Also 105 Int. J. Aquat. Biol. (2017) 5(2): 95-107 previous studies in tropical wetland systems has not identified the mussel species M. turriculus as a biomonitoring indicator that can be used in assessing the wetland health. Plafkin et al. (1989) and Bode et al. (1991, 1996) stated that Modified Biotic Index is highly applicable and modified to incorporate non-arthropod species including molluscs to characterize the level of organic pollution input in the aquatic ecosystems. The input of organic pollutants and the water quality of the receiving aquatic system can be categorized based on the values of MBI and the categorization is given in Table 12 (Hilsenhoff, 1988). Significantly, high Modified Biotic Index (MBI) was recorded in sites E and F indicating a very significant level of input of organic pollutants and poor water quality in those sites. In addition, according to the variation of MBI sites, A, B and C in the non-rehabilitated area and site D of the rehabilitated area also indicated a significant organic pollution and fairly poor water quality. In a study conducted to assess the water quality in the Diyawannawa wetland system for a five year period have recorded that the water nitrate concentration varied within a range 0.01-0.8 mg.L-1 (Dharmasoma and Piyadasa, 2010). In the present study, a similar variation of nitrate concentration from 0.02 to 0.09 mg.L-1 was recorded and the values recorded in the study are below the maximum standard ambient nitrate concentration (5 mg.L-1) to maintain a healthy aquatic life established by the Central Environmental Authority of Sri Lanka (CEA, 2001). Based on the results, there was no statistically significant spatial variation in the total phosphate concentration. But compared to the study sites in the non-rehabilitated area, the sites in the rehabilitated area showed higher total phosphate concentrations. The total phosphate concentration of the study sites ranged from 0.0009 to 0.066 mg.L-1and it was also less than the maximum standard ambient nitrate concentration (0.4 mg.L-1) to maintain a healthy aquatic life (CEA, 2001). DO is identified to be negatively correlated with the organic pollution status of a wetland (Kazi et al., 2009). As recorded in a study conducted by Dharmasoma and Piyadasa (2014), the mean dissolved oxygen concentration of the selected sites of the Diyawannawa wetland ranged from 4.69-5.86 mg.L-1. In the present study, mean dissolved oxygen concentration showed a wider range of variation with a range from 2.40 to 14.91 mg.L-1. The dissolved oxygen concentration showed a significant spatial variation indicating a significantly lower dissolved oxygen concentration at site A of the non-rehabilitated area compared to other sites. During the field work, it was noted that this site had lot of floating macrophytes compared to other sites, thus disturbing the contact of aquatic and atmospheric interphase. This may have reduced the atmospheric input of oxygen of this site. The mean oxygen concentration recorded in site A of the non-rehabilitated area was less than the standard minimum ambient dissolved oxygen concentration (3 mg.L-1) established by the CEA, but the other sites showed higher values than the ambient minimum standard. In the present study, there is a significant increment in BOD5 and COD in the rehabilitated area compared to those of non-rehabilitated area. COD is a measurement used to detect the input of domestic sewage, agricultural and industrial waste into aquatic systems and high BOD and COD can be recorded in aquatic systems where extensive domestic waste dumping is pracised (Kazi et al., 2009; Triest et al., 2001). In the present study, improper solid waste dumping and discharging of waste from nearby residents was observed in sites E and F of the rehabilitated area, which may have contributed to significantly high levels of BOD and COD. However, the BOD values recorded during the study were less than the maximum ambient BOD (4 mg.L-1) and the COD values were higher than the maximum ambient COD (15 mg.L-1) values established by the CEA of Sri Lanka. The results of the present study showed a significant spatial and temporal variation of physicochemical parameters of water and shallow sediment quality in the rehabilitated and non- rehabilitated areas of the Diyawannawa wetland system affecting the abundance and diversity of 106 Wijeyaratne and Bellanthudawa/ Use of macrobenthic diversity as bioindicator in a tropical wetland system macrobenthic molluscs. Therefore, it highlights the importance of continuous monitoring of water and sediment quality, and fresh water macrobenthic mollusc diversity as reflect the health of this wetland system. The results also showed that abundance of B. tentaculata and P. globosa can be used as a biomonitoring tool to assess sediment quality in relation to their positive interaction with sediment quality and their abundance. References Abel P.D. (1989). Water Pollution Biology. 2nd ed. Ellis Horwood, Chichester, UK. Badruzzaman A.B.M., Bari M.F., Alam M.S., Hoque M.M., Habib M.E., Saha M. (2007). Effects of thermal effluent discharge on the macroinvertebrate abundance in the Sitalakhya River in Bangaladesh. In: Proceedings of the Scientific Conference; Rivers of the Hindu-Kush Himalaya-Ecology and Environmental Assessment, pp: 113. Barbour M.T., Gerritsen J., Snyder B.D., Stribling J.B. (1999). Rapid bioassessment protocols for use in streams and wadeable rivers: Periphyton, benthic macroinvertebrates and fish, 2nd ed. U.S. Environmental Protection Agency; Office of Water; Washington, D.C. Bartsch M.R., Newton T.J., Allran J.W., O'Donnell J.A., Richardson W.B. (2003). Effects of pore‐water ammonia on in situ survival and growth of juvenile mussels (Lampsilis cardium); in the St. Croix Riverway, Wisconsin, USA. Environmental Toxicology and Chemistry, 22(11): 2561-2568. Begon M., Townsend C.R., Harper J.L. (2006). Ecology: from individuals to ecosystems. 4th ed. Wiley- Blackwell, USA. Bode R.W., Novak M.A., Abele L.E. (1996). Quality assurance work plan for biological stream monitoring in New York State. NYS Department of Environmental Conservation, Albany, NY. 89 p. Bode R.W., Novak M.W., Abele L.E. (1991). Methods for rapid biological assessment of streams. Stream Biomonitoring Unit, Bureau of Monitoring and Assessment, Division of Water, NYS Dept. of Environmental Conservation. 57 p. Bonada N., Prat N., Resh V.H., Statzner B. (2006). Developments in aquatic insect biomonitoring: a comparative analysis of recent approaches. Annual Review of Entomology, 51: 495- 523. CEA (2001). Proposed Ambient Water quality Standards for Inland Waters in Sri Lanka. Colombo, Sri Lanka: Environment Action Project (Funded by ADB). Choubisa S.L. (1992). Molluscs as bio-indicators for the trophic stages of lakes and lotic environments. Bulletin of Pure and Applied Science, 11: 35-40. Clarke A.H. (1979). Gastropods as indicators of trophic lake stages. Nautilus, 94: 138-142. Dahanayaka D.D.G.L., Wijeyaratne M.J.S. (2006). Diversity of macrobenthic community in the Negombo estuary, Sri Lanka with special reference to environmental conditions. Sri Lanka Journal of Aquatic Science, 11: 43-61. Dharmasoma U.Y.I.L., Piyadasa R.U. (2014). An Assessment of seasonal variation water quality in the inlets of Diyawannawa Lake, Sri Lanka. In: Proceedings of the SAITM Research Symposium on Engineering Advancements. pp: 110-112. Feio M.J., Almeida S.F.P., Craveiro S.C., Calado A.J. (2007). Diatoms and macroinvertebrates provide consistent and complementary information on environmental quality. Fundamental and Applied Limnology, 169: 247-258. Fernando C.H., Weerawardhena S.R. (2002). A guide to the fresh water fauna of Ceylon (Sri Lanka). A reprint of A.S. Mendis and C.H. Fernando (1962) with a forward, a check list of Amphibians of Sri Lanka and selected references on freshwater fauna. 2nd ed. Department of Fisheries, Sri Lanka. 173 p. Gamlath G.A.R.K., Wijeyaratne M.J.S. (1997). Indicator organism of environmental conditions in a lotic water body in Sri Lanka. Sri Lanka Journal of Aquatic Science, 2: 121-129. Hartmut F., Gerstmann S. (2007). Declining populations of freshwater pearl mussels (Margaritifera margaritifera) are burdened with heavy metals and DDT/DDE. Ambio, 36(7): 571-574. Hellawell J.M. (1986). Biological indicators of freshwater pollution and environmental management. 1st ed. Elsevier, London. 518 p. Hilsenhoff W.L. (1988). Using a biotic index to evaluate water quality in streams. Tech. Bull. No 132 Wisconsin Department of Natural Resources. Idroos F.S., Manage P.M. (2012). Aquatic life health quality assessment of the Bolgoda Canal and Waga Stream with respect to selected physico- chemical parameters and bioindicators. 107 Int. J. Aquat. Biol. (2017) 5(2): 95-107 Journal of Tropical Forestry and Environment, 2(2): 13- 26. Jacobson P.J., Farris J.L., Cherry D.S., Neves R.J. (1993). Juvenile Frshwater mussel (Bivalvia, Uninoidae) responses to acute toxicity testing with copper. Environmental Toxicology and Chemistry, 12: 879- 883. Kazi T.G., Arain M.B., Jamali, M.K., Jalbani N., Afridi H.I., Sarfraz R.A., Baig J.A., Shah A.Q. (2009). Assessment of water quality of polluted lake using multivariate statistical techniques: A case study. Ecotoxicology and Environmental Safety, 72(2): 301-309. Kolodziejczyk A. (1984). Occurrence of Gastropoda in the lake littoral and their role in the production and transformation of Detritus 1. Snails in the Littoral of Mikolajskie Lake- General Characteristics of Occurrence. Polish Journal of Ecology, 32(3): 441-468. Mereta S.T., Boets, P., De Meester L., Goethals P.L.M. (2013). Development of a multimetric index based on benthic macroinvertebrates for the assessment of natural wetlands in Southwest Ethiopia. Ecological Indicators, 29: 510-521. Moloukhia H., Sleem S. (2011). Bioaccumulation, fate and toxicity of two heavy metals common in industrial wastes in two aquatic Molluscs. Journal of American Science, 7(8): 459-464. Páez-osuna P., Frias-espericueta M.G., Osunalópez J.I. (1995). Trace metal concentrations in relation to season and gonadal maturation in the oyster Crassostrea iridescens. Marine Environmental Research, 40(1): 19- 31. Plafkin J.L., Barbour M.T., Porter K.D., Gross S.K., Hughes R.M. (1989). Rapid bioassessment protocols for use in streams and rivers: benthic macroinvertebrates and fish. EPA. Riget F., Johansen P.., Asmund G. (1997). Uptake and release of lead and zinc by blue mussels. Experience from transplantation experiments in Greenland. Marine Pollution Bulletin, 34(10): 805-815. Savage A.A., Gazey G.M. (1987). Relationships of physical and chemical conditions to species diversity and density of gastropods in English lakes. Biological conservation, 42(2): 95-113. Shuhaimi-Othman M., Nur-Amalina R., Nadzifah Y. (2012). Toxicity of metals to a freshwater snail, Melanoides tuberculata. The Scientific World Journal, 1-10. Triest L., Kaur P., Heylen S., De Pauw N. (2001). Comparative monitoring of diatoms, macroinvertebrates and macrophytes in the Woluwe River (Brussels, Belgium). Aquatic Ecology, 35(2): 183-194. Turkmen M., Turkmen A., Akyurt I., Tepe Y. (2005). Limpet, Patella caerulea Linnaeus, 1758 and barnacle, Balanus sp., as biomonitors of trace metal availabilities in Iskenderun Bay, northern east Mediterranean sea. Bulletin of Environmental Contamination and Toxicology, 74(2): 301-307. Yuzereroglu T.A., Gok G., Cogun H.Y., Firat O., Aslanyavrusu S., Maruldali O., Kargin F. (2010). Heavy metals in Patella caerulea (mollusca, gastropoda) in polluted and non-polluted areas from the Iskenderun Gulf (Mediterranean Turkey). Environmental Monitoring and Assessment, 167: 257-264.