Methods

Diatom Sample Analysis

Benthic diatoms were collected from 30 sampling sites across selected wetlands during both prior to restoration (October 2009) and post-restoration (November 2011) from all available habitats (Epiphytic, Epilithic and Episammic). Samples were carried to laboratory and observed immediately in order to record live and dead valves. Samples with dead valves were not considered for further analysis. Cleaning and enumeration of samples was carried out following laboratory procedures as per Taylor et al., (2005) and Karthick et al., (2010). Samples were cleaned using Hot HCl and KMnO4 and slides were prepared using Naphrax® as the mounting solution. 400 valves were counted using light microscope model Olympus BX51 equipped with JENOPTIC mircophotographic system from each sample to determine percentage relative abundance of each taxon. Diatoms were identified to species level using Taylor et al., 2007; Krammer and Lange-Bertalot, 1986-1991 and Gandhi, 1998.

Water sampling

Water samples were collected from 10 to 30 cm underneath the water surface and stored in disinfected plastic bottles for laboratory analyses. Samples were immediately transported to the laboratory and refrigerated for subsequent water quality analysis. Parameters such as water temperature, pH, electrical conductivity and dissolved oxygen were assessed onsite using portable electrode probe. Laboratory analyses included total alkalinity, biological oxygen demand (BOD), chemical oxygen demand (COD), total hardness, inorganic phosphates (PO43-), nitrates (NO3–) and chlorides (Cl-) following standard protocol of American Public Health Association (APHA, 2005).

Data Analysis

Principal component analysis (PCA) was carried out to prioritize the environmental factors that reflect variation in species across wetland types. Diatom taxa occurring in at least one sample with a relative abundance >10% were considered for diversity estimations and statistical analyses. Diversity indices like Shannon, species richness, evenness and dominance were estimated. % Eutrophic taxa were calculated following van Dam et al. 1994 classification. Non- metric multidimensional scaling (NMDS) was performed using PAST version 2.19 (Hammer et al., 2001), to describe patterns in diatom composition with respect to five groups of wetlands. The resulting pattern represents similar wetlands cluster in ordination space while no/dissimilar wetlands were spaced apart. The stress value assigned reflects how well the ordination summarizes the observed distances among the samples.