Constructed treatment wetlands are a relatively low-cost alternative used for tertiary treatment of wastewater. Phosphorus (P) removal capacity of these wetlands may decline, however, as P is released from the accrued organic soils. Little research has been done on methods to restore the treatment capacity of aging constructed wetlands. One possibility is the seasonal addition of alum during periods of low productivity and nutrient removal. Our 3-mo mesocosm study investigated the effectiveness of alum in immobilizing P during periods of reduced treatment efficiency, as well as the effects on soil biogeochemistry. Eighteen mesocosms were established, triplicate experimental and control units for Typha sp., Schoenoplectus californicus, and submerged aquatic vegetation (SAV) (Najas guadalupensis dominated). Alum was slowly dripped to the water column of the experimental units at a rate of 0.91 g Al m(-2) d(-1) and water quality parameters were monitored. Soil cores were collected at experiment initiation and completion and sectioned into 0- to 5- and 5- to 10-cm intervals for characterization. The alum floc remained in the 0- to 5-cm surface soil, however, soil pH and microbial parameters were impacted throughout the upper 10 cm with the lowest pH found in the Typha treatment. Plant type did not impact most biogeochemical parameters; however, data were more variable in the SAV mesocosms. Amorphous Al was greater in the surface soil of alum-treated mesocosms, inversely correlated with soil pH and microbial biomass P in both soil layers. Microbial activity was also suppressed in the surface soil of alum-treated mesocosms. This research suggests alum may significantly affect the biogeochemistry of treatment wetlands and needs further investigation.

Application of low doses of alum to treatment wetlands to reduce elevated outflow winter phosphorus concentrations were tested in mesocosms vegetated with either Typhadomingensis, Schoenoplectus californicus, or submerged aquatic vegetation (SAV) (Najas guadalupensis-dominated). Alum was pumped to experimental units at a rate of 0.91 g Al m(-2) d(-1) and water quality monitored for 3 months. The alum application significantly improved the outflow water quality and overall the growth of the plants was unaffected by the alum application. Biomass and growth varied between species and through time, but no significant effects of alum application were detected. The concentrations of nutrients and mineral elements in the aboveground tissues differed between species and over time, but only the concentration of Al in plant tissue was increased by alum additions. The concentration of Al was 50-fold higher in alum-treated SAV as compared to the control, and in Typha and Schoenoplectus the concentrations were 4- and 2-fold, higher, respectively. The N/P ratios in the plant tissues were low (<10) suggesting that their growth and biomass was limited by nitrogen. The research suggests that a continuous or seasonal low-dosage alum application to treatment wetlands provides an effective tool to maintain discharge concentrations within permitted values during the inefficient winter treatment times. We suggest that the use of alum should be restricted to treatment wetland areas dominated by emergent vegetation as the effects of the elevated Al concentrations in SAV needs further study.

Little research has been conducted on methods to restore the P removal capacity of older constructed wetlands, as P treatment capacity can decrease with time. We investigated the capacity of alum and three Al‐containing alternatives (alum residual, polyaluminum chloride [PAC], and partially neutralized aluminum sulfate [PNAS]), at three rates (9, 18, and 36 g Al m−2) to reduce P concentrations. Water column pH of the alum treatment was significantly less than all other treatments, averaging 3.65 ± 1.12, while PAC (4.85 ± 0.96) and PNAS (4.21 ± 0.93) treatments had pH values significantly less than the alum residual and controls. Soluble reactive P decreased in the water column of all the treatments (−60.41 to −2.11 mg m−2 d−1). At all dosage rates, alum and PNAS were most effective at binding P, followed by PAC, while the alum residual was least effective. Significantly less P removal occurred in the 9 g Al m−2 alum, PNAS, and PAC treatments than the higher Al dosages. Dosage was inversely related to microbial biomass and activity in the surface soil, suggesting short‐term negative effects of chemical additions. Results suggest that a low‐dosage Al amendment application to wetland soil can, in the short term, prevent release of P from organic soil into the water column as well as remove P from the water column. Long‐term studies are needed, however, to verify P removal efficacy with time and the effects of continued applications on nutrient availability.

Nutrient removal in treatment wetlands declines during winter months due to temperature. A 3-mo (wintertime) mesocosm study was conducted to determine the effectiveness of alum in immobilizing P as well as other nutrients during this period of reduced treatment efficiency. Eighteen mesocosms, triplicate alum, and three controls or no alum were established with either Typha spp., Schoenoplectus californicus, or SAV (Najas guadalupensis-dominated). Alum was delivered by timer at a rate of 0.81 g Al m(-2) d(-1) and parameters measured included: pH, soluble reactive phosphorus (SRP), total phosphorus (TP), dissolved organic carbon (DOC), dissolved inorganic nitrogen (DIN), total kjeldahl nitrogen (TKN), and soluble aluminum (Al). Alum-treated mesocosms had significantly lower pH values (8.1) than controls (8.8), but well within the elevated pH range for aluminum toxicity. Alum significantly reduced all measured water column nutrients with the exception of ammonium N, which remained unaffected, and particulate P, which increased. This study demonstrated that seasonal low-dosage alum application to different vegetation communities in a treatment wetland can significantly improve treatment efficiencies for SRP (87 vs. 58%) and TP (62 vs. 44%) but also increase DOC (19 vs. 0%) and TKN (12 vs. -3%) removal capacity to a lesser degree. Alum applications within close proximity of the treatment wetland effluent points should be implemented with caution due to the production of alum floc-bound P which could potentially affect discharge permit compliance for total suspended solids or total P.