PS 35-30
Effects of urbanization on stream physicochemical and bacterial communities after flooding events

Wednesday, August 7, 2013
Exhibit Hall B, Minneapolis Convention Center
Alescia Roberto, Department of Biological Sciences, Kent State University, Kent, OH
Laura G. Leff, Department of Biological Sciences, Kent State University, Kent, OH
Background/Question/Methods

Urbanization directly affects the quantity of pollutants entering urban streams from both point and non-point sources (e.g. waste water treatment plants and runoff attributed to impervious surfaces) during floods. The negative impacts of pollutants on fish and macro-invertebrates are well known; however, less is known about the effects of these pollutants on communities of stream bacteria. In this study, we used DNA fingerprinting techniques, terminal restriction length polymorphisms (T-RFLP) and quantitative PCR (qPCR), to evaluate the impact of stormwater runoff on stream bacterial communities. Water column, and benthic samples were collected before and after large flood (caused by Superstorm/ Hurricane Sandy) from multiple sites along a highly impacted stream (Tinker’s Creek, Cuyahoga watershed, Northeast Ohio) fed by multiple tributaries impacted by urbanization and two reference streams. Physicochemical properties, including: dissolved organic carbon (DOC), benthic organic matter (BOM), total dissolved nitrogen (TN), nitrate and nitrite, soluble reactive phosphorus, conductivity, pH, water temperature, redox potential, dissolved oxygen (DO), dissolved metals, turbidity, and flow. Biological variable measured were total bacterial abundance (T-Bac) and community composition based on T-RFLP of 16S rRNA genes.

Results/Conclusions

There were significant (p ≤0.05) interactions between sampling site and date (pre-versus post-flood dates) for multiple physicochemical properties, namely pH, temperature, DO, redox potential, turbidity, flow, and TN. Benthic organic matter and T-Bac differed between dates and among sites, but the interaction was not significant.  Dissolved organic carbon only differed significantly among sampling sites (p ≤0.05). Reference sites tended to have different bacteriological and physicochemical attributes; for example, T-Bac and nutrient concentrations were higher for each sampling site in Tinker’s Creek compared to the reference sites. Urbanized areas along Tinker’s Creek saw an increase in all measured parameters for both pre- and post-flood dates. Flooding impacted stream physicochemical properties and T-Bac concentrations, resulting in an increase (+5%) in water pH and temperature, while redox potential, turbidity, and flow were ~100% lower pre-flood. Bacterial abundances were lower pre-flood (-17%), whereas DO concentrations were greater post-flood (+ 6%) at all sites except for two Tinker’s Creek sites. The opposite trend was observed in TN; lower concentrations (-17%) occurred in all sites except at the reference sites post-flood. These results suggest that differences in urbanization along the impacted stream influence stormwater input and consequently alters bacterial communities and physicochemical conditions.