MEASUREMENTS OF MICROBIAL AND ALGAL COMMUNITIES ASSOCIATED WITH HARMFUL ALGAL BLOOMS IN OLD WOMEN CREEK NATIONAL ESTUARY OF LAKE ERIE USING REFLECTANCE SPECTROSCOPY AND QUANTITATIVE PCR

Old Women Creek National Estuary (OWC), a State Nature and Research Reserve, is a freshwater estuary connected to Lake Erie. During the week of July 16-26, 2018, 55 water samples were collected from five sites within OWC. Three different algal bloom-monitoring methods; reflectance spectra, chlorophyll concentrations, and quantitative PCR (qPCR), were compared with each other and the environmental factors measured at these sites using Pearson correlations with Benjamini Hochberg adjusted p values. For qPCR analysis, 10 mL of each water sample was filtered through a 0.2µm, 13mm polycarbonate filter, and genomic DNA was extracted from the filters using the Qiagen DNeasy kit. Visible color reflectance was measured using a Konica Minolta 2600D spectrophotometer on oven dried 0.4µm GF/F filter paper after filtering 250 ml of the sample through them. Four principal components (VPCs) were extracted by varimax-rotated principal component analysis of reflectance data. Comparison of VPCs to known algal and/or pigment spectra using a stepwise linear regression correspond to a mixture of cyanobacteria and phycoerythrin (VPC1), chlorite+amphibole and cyanobacteria (VPC2), fucoxanthin+chlorophyll-a (VPC3), and calcite+ chlorophyll-a (VPC4). Across all sampling sites, VPC 2 and 3 are negatively correlated with total suspended solids (TSS), but are positively correlated to Secchi depth suggesting that lower VPC 2 and 3 scores correspond to higher turbidity and vice versa. VPC1 is positively correlated with temperature and maximum daily wind speed suggesting increasing cyanobacteria productivity with increasing water temperature. Resuspension of sediment associated with wind makes more nutrients available for cyanobacteria growth and dispersal within the estuary. Chlorophyll concentration had no correlation with qPCR data, but negatively correlated with VPC1. Bacterial and archaeal qPCR gene counts did not significantly correlate with VPC loadings, which may result from the VPCs’ associations with several algal and sediment spectra. Neither VPC loadings nor qPCR gene counts correlated with pH, temperature, salinity, nitrates, precipitation, or oxidation-reduction potential. However, the Bacterial and archaeal 16S gene counts were positively correlated to each other.