Removal, distribution and retention of metals in a constructed wetland over 20 years.

The A-01 wetland treatment system (WTS) was designed to remove metals (primarily copper) from the effluent at the A-01 National Pollution Discharge Elimination System (NPDES) outfall at the Savannah River Site, Aiken, SC. This research investigated metal removal, distribution and retention in the A-01 WTS over a period of 20 years. The findings are important for ensuring continued metal sequestration in the A-01 WTSs over time, providing management guidance for constructed wetlands, and investigating changes in metal remediation effectiveness as a wetland ages. During 20 years of operation, systematic water and sediment sampling validated the wetlands' performance. After passage through the treatment cells, Cu concentrations were well below permit limits during all years of operation, often falling below 10 µg L-1. Cu removal has been consistent over time, averaging about 80% despite large changes in influent Cu concentrations. Most divalent metals were rapidly removed from the water and held in the sediments shortly after the water entered the treatment wetland. Average removal of Pb from water by the wetland system was 67 and 74% in 2004 and 2020, respectively. Comparable values for Zn were 52 and 65%, respectively. Generally, the highest concentrations of Cu, Pb, and Zn were found in the sediment from the first cell in each pair of cells suggesting that most of the Cu, Pb, and Zn in the A-01 effluent was bound to the sediment quickly. Diffusive gradients in thin films (DGT) measurements of Cu and Zn in the sediments were much lower than bulk sediment concentrations. These results suggest that most of the Cu and Zn in the A-01 WTS sediments was not bioavailable, hence not toxic to aquatic organisms, as a likely consequence of adsorption to sediment particles and complexation with organic and inorganic substances.

Effect of bioturbation on contaminated sediment deposited over remediated sediment.

A challenge to all sediment remediation technologies is the continued influx of contaminants from uncontrolled sources following remediation. However, contaminants deposited on sediments remediated with chemically active sequestering agents may be affected by the sequestering agents resulting in reduced impacts. We deposited sediment contaminated with As, Cd, Cu, Ni, Pb, and Zn over clean sediment capped with the sequestering agent, apatite, and clean uncapped sediment in laboratory mesocosms to simulate the recontamination of remediated sediment by influxes of particle-bound contaminants. Cap effectiveness was assessed in the presence and absence of the bioturbating organism Corbicula fluminea based on metal fluxes to sediment pore water and surface water, the distribution of mobile contaminants in sediment and surface water measured by Diffusive Gradients in Thin Films, and contaminant bioaccumulation by Lumbriculus variegatus. The metal sequestration capacity of apatite caps was unaffected or improved by bioturbation for all elements except As. Effects with uncapped sediment were metal specific including reductions in the bioavailable pool for Ni, Cd, and to a lesser extent, Pb, increases in the bioavailable pool for As and Cu, and little effect for Zn. It is likely that the reductions observed for some metals in uncapped, clean sediment were the result of burial and dilution of contaminated sediment combined with chemical processes such as sequestration by minerals and other compounds. These results indicate that apatite caps can control recontamination by metals regardless of bioturbation but point to the complexity of sediment recontamination and the need for further study of this problem.

Assessing effects of dissolved organic carbon and water hardness on metal toxicity to Ceriodaphnia dubia using diffusive gradients in thin films (DGT).

We evaluated the ability of diffusive gradients in thin films (DGT) to assess the effects of water hardness and dissolved organic carbon (DOC consisting of humic acids) on Cu and Zn toxicity (i.e., 48 h LC50s) to Ceriodaphnia dubia. When DOC was high, Cu concentrations measured by DGT (DGT Cu) were significantly lower than Cu concentrations in water (dissolved Cu), and DGT LC50s were significantly lower than LC50s based on dissolved concentrations. When DOC was low, differences between dissolved Cu and DGT Cu were small, as were differences between dissolved LC50s and DGT LC50s. Differences between DGT and dissolved measurements of Zn were small compared with the differences observed for Cu, and DGT Zn LC50s were relatively similar to dissolved Zn LC50s. Humic acids formed strong organic-Cu complexes that were both inaccessible for biological uptake and excluded by DGT, which selected for free or weakly bound Cu. In contrast, Zn did not form strong complexes with DOC that greatly affected either toxicity or the measurement of Zn by DGT. The effects of hardness on DGT measurements of Cu and Zn were smaller and more complex than the effects of DOC. Large, statistically significant differences between DGT measurements of Cu in low and high DOC water accurately reflected the strong effects of DOC on Cu toxicity. However, the effects of DOC were inconsistent for Zn, and DGT provided less information about the toxic fraction of this metal.

Removal of low levels of Cu from ongoing sources in the presence of other elements - Implications for remediated contaminated sediments.

Mesocosms were used to investigate the effects of Cu influx, alone and in the presence of other elements, on sediments remediated by active caps, passive caps, and in situ treatment. Competitive interactions between Cu and other elements were investigated because contaminants often co-occur. Elements in surface water remained at significantly lower concentrations in mesocosms with apatite and mixed amendment caps than in mesocosms with passive sand caps or uncapped sediment. Element concentrations in Lumbriculus variegatus were significantly higher in untreated sediment than in active caps and significantly related to element concentrations in sediment measured by DGT probes. The cumulative toxicity of Cu mixed with other elements was greater than the toxicity of Cu alone in treatments without active caps, but the ability of active caps to control Cu was not affected by the presence of other elements. Active caps can protect remediated sediments by reducing bioavailable elements in ongoing contamination.

Environmental impact of ongoing sources of metal contamination on remediated sediments.

A challenge to all remedial approaches for contaminated sediments is the continued influx of contaminants from uncontrolled sources following remediation. We investigated the effects of ongoing contamination in mesocosms employing sediments remediated by different types of active and passive caps and in-situ treatment. Our hypothesis was that the sequestering agents used in active caps and in situ treatment will bind elements (arsenic, chromium, cadmium, cobalt, copper, nickel, lead, selenium, and zinc) from ongoing sources thereby reducing their bioavailability and protecting underlying remediated sediments from recontamination. Most element concentrations in surface water remained significantly lower in mesocosms with apatite and mixed amendment caps than in mesocosms with passive caps (sand), uncapped sediment, and spike solution throughout the 2520h experiment. Element concentrations were significantly higher in Lumbriculus variegatus from untreated sediment than in Lumbriculus from most active caps. Pearson correlations between element concentrations in Lumbriculus and metal concentrations in the top 2.5cm of sediment or cap measured by diffusive gradient in thin films (DGT) sediment probes were generally strong (as high as 0.98) and significant (p<0.05) for almost all tested elements. Metal concentrations in both Lumbriculus and sediment/cap were lowest in apatite, mixed amendment, and activated carbon treatments. These findings show that some active caps can protect remediated sediments by reducing the bioavailable pool of metals/metalloids in ongoing sources of contamination.

Interactions among phosphate amendments, microbes and uranium mobility in contaminated sediments.

The use of sequestering agents for the transformation of radionuclides in low concentrations in contaminated soils/sediments offers considerable potential for environmental cleanup. This study evaluated the influence of three types of phosphate (rock phosphate, biological phosphate, and calcium phytate) and two microbial amendments (Alcaligenes piechaudii and Pseudomonas putida) on U mobility. All tested phosphate amendments reduced aqueous U concentrations more than 90%, likely due to formation of insoluble phosphate precipitates. The addition of A. piechaudii and P. putida alone were found to reduce U concentrations 63% and 31%, respectively. Uranium removal in phosphate treatments was significantly reduced in the presence of the two microbes. Two sediments were evaluated in experiments on the effects of phosphate amendments on U mobility, one from a stream on the Department of Energy's Savannah River Site near Aiken, SC and the other from the Hanford Site, a Department of Energy facility in Washington state. Increased microbial activity in the treated sediment led to a reduction in phosphate effectiveness. The average U concentration in 1 M MgCl(2) extract from U contaminated sediment was 437 microg/kg, but in the same sediment without microbes (autoclaved), the extractable U concentration was only 103 microg/kg. The U concentration in the 1 M MgCl(2) extract was approximately 0 microg/kg in autoclaved amended sediment treated with autoclaved biological apatite. These results suggest that microbes may reduce phosphate amendment remedial effectiveness.

Metal distribution and stability in constructed wetland sediment.

The A-01 wetland treatment system (WTS) is a surface flow wetland planted with giant bulrush [Schoenoplectus californicus (C.A. Mey.) Palla] that is designed to remove Cu and other metals from the A-01 National Pollution Discharge Elimination System (NPDES) effluent at the Savannah River Site near Aiken, SC. Copper, Zn, and Pb concentrations in water were usually reduced 60 to 80% by passage through the treatment system. The Cu concentrations in the wetland sediments increased from about 4 to 205 and 796 mg kg(-1), respectively, in the organic and floc sediment layers in cell 4A over a 5-yr period. Metal concentrations were higher in the two top layers of sediment (i.e., the floc and organic layers) than in the deeper inorganic layers. Sequential extraction was used to evaluate remobilization and retention of Cu, Pb, Zn, Mn, and Fe in the wetland sediment. Metal remobilization was determined by the potentially mobile fraction (PMF) and metal retention by the recalcitrant factor (RF). The PMF values were high in the floc layer but comparatively low in the organic and inorganic layers. High RF values for Cu, Zn, and Pb in the organic and inorganic layers indicated that these metals were strongly bound in the sediment. The RF values for Mn were lower than for the other elements especially in the floc layer, indicating low retention or binding capacity. Retention of contaminants was also evaluated by distribution coefficient (Kd) values. Distribution coefficient (Kd) values were lower for Cu and Zn than for Pb, indicating a smaller exchangeable fraction for Pb.