Metal fluxes across the sediment water interface in a drinking water reservoir

Elevated concentrations of iron (Fe) and manganese (Mn) in drinking water degrade water quality by affecting taste, odor, and color. Under oxic conditions (dissolved oxygen (DO) >2 mg/L), Fe and Mn are rarely present in soluble form in natural waters, as they occur as insoluble, oxidized minerals in sediments. However, the development of low DO concentrations in the bottom waters of some lakes and drinking water reservoirs during thermal stratification can lead to the reduction of oxidized, insoluble Fe and Mn in sediments to soluble forms, which are then released into the water column. In response, many water utilities have installed oxygenation systems to control metal concentrations in situ in drinking water reservoirs. However, previous research has found anoxic (DO < 0.5 mg/L) conditions still develop within sediments, even with operational oxygenation systems, allowing for the reduction and release of soluble Fe and Mn into the water column.

To examine the drivers of metal release from sediments into the water column, I conducted sediment flux chamber experiments to directly quantify Fe and Mn fluxes at the sediment-water interface of a small, eutrophic drinking water reservoir (Falling Creek Reservoir, Vinton, VA). The experiments were conducted twice during the 2018 summer stratification period (April 24 – October 21). Using the flux chambers, I measured total and soluble Fe and Mn concentrations under changing oxygen conditions over 10-day periods to calculate fluxes. Throughout the experiments, I monitored DO, oxidation-reduction potential (ORP), temperature, and pH. In addition to the direct measurements, I also estimated metal fluxes using a mass balance method, which relies on measurements of metal inputs and outputs into the bottom waters of the reservoir.

Overall, our results showed that fluxes are highly variable during the stratification period, with some periods having positive fluxes (release of metals from sediment to the water column) and some with negative fluxes (return of metals from the water column to sediment). The metal fluxes are highly sensitive to redox conditions in the water column, sediment-water interface and sediments.

Metal fluxes measured using the chambers are 91-105% higher than those estimated using the mass balance method. This difference supports result of previous work that the flux chamber method likely provides maximum values of metal fluxes as the isolated chamber water does not allow for mixing with the bottom waters. In contrast, because the mass balance method relies on water column data, results are affected by mixing and biogeochemical reactions that can remove metals from the water column; thus, flux estimates using this method likely reflect minimum values. However, when used together, these two methods provide a useful tool for constraining metal fluxes under different redox conditions and highlight the importance of measuring ORP in addition to DO. The results of this study can be used by water utilities to improve the effectiveness of engineered oxygenation systems and water quality management practices related to iron and manganese.