A monitoring study was conducted on the Sandusky River, in north-central Ohio, to characterize changes in water quality and quantity before and after the removal of a low-head dam (St. John). Short-term time series of flow and turbidity were taken during the dewatering and the removal of the dam in order to determine sediment loads. High resolution spatial data, based on vertical profiles and longitudinal surveys, were collected before and after the removal to determine changes in temperature, dissolved oxygen (DO), pH, turbidity, specific conductivity, and oxidation-reduction potential (ORP) in the 18 km reservoir. Surface water samples were collected upstream and downstream of the dam before and after removal and analyzed for nitrate, ammonia, and phosphate. After dam removal, denitrification potential was studied within the former reservoir to compare the potential in newly exposed mudflats with the riverbed and the floodplain.
During the breach of the dam, change in depth and turbidity was measured at a sampling station 200 m downstream. A small increase in volumetric flow was measured, however, time series data taken at the same sampling station during the removal showed a larger increase in flow. The discharge from the breach was not discernable at a gaging station 53 km downstream, however, the larger discharge from the removal was evident, though clearly attenuated. The removal also caused an increase in turbidity less than that of a bankfull storm event that occurred less than one week later, however, the duration of increased turbidity during removal was shorter than the storm event.
The vertical profile of temperature and DO measured behind the dam before removal showed a decrease in temperature and DO with increasing depth from the surface. This was similar to lake stratification, not river conditions. The longitudinal surveys of temperature and DO were determined to be more variable before dam removal and more variable in the 13.7 km reservoir than in the control reach upstream of the reservoir. Longitudinal surveys of specific conductivity indicated no spatial trends. Values mimicked historical trends (1970-1973) at the St. John Dam by increasing with low discharge and decreasing with high discharge. Before and after dam removal, longitudinal surveys of turbidity varied throughout the reservoir and were affected mostly by discharge and storm events. ORP and pH were similar upstream and downstream, before and after removal, suggesting no significant differences.
Phosphate and ammonia were not detected above 0.5 mg L-1 upstream or downstream of the dam before or after removal. Before dam removal, nitrate concentrations upstream from the reservoir were statistically different than concentrations in the reservoir. However, this was also true of the former reservoir after the dam was removed, concluding dam removal had no affect on nitrate distribution in the Sandusky River system.
Denitrification potential was measured in the reservoir after removal. Data showed there was no significant difference in removal potential longitudinally or laterally in the river. There was, however, a significant increase in potential when samples were treated with nitrate (increased from 5 to 100 mg N20 g-1 soil h-1). This indicated that the river system was nitrate limited. Furthermore, studies also determined that the loss of sediment-water (denitrification potential) contact from dewatering the reservoir was insignificant. Even though sediment-water contact was lost, there was still enough sediment-water contact left to adequately remove nitrate from the system.
The dam breach did not significantly affect discharge or turbidity, however, dam removal increased turbidity and discharge similar to a small storm event. Dam removal did not significantly affect longitudinal values of nutrients, pH, ORP, turbidity, or specific conductivity, however, it did decrease the variability of temperature and DO in the former reservoir. Also, dam removal did not significantly affect the river system's ability to remove nitrate.