Hydrologic and Vegetation Effects on Water Column Phosphorus in Wetland Mesocosms

Historic phosphorus (P) loading from agricultural areas has been identified as one of the major causes for ecological changes occurring in the Florida Everglades. The restoration plan for the Everglades includes construction of large stormwater treatment areas (STAs) to intercept and treat this relatively high nutrient water down to very low total P (TP) concentrations. One such STA has been in operation for approximately 10 yr and contains both emergent aquatic vegetation (EAV) and submerged aquatic vegetation (SAV) communities. The surface water TP concentrations in areas near the outflow range from 0.02 to 0.05 mg TP L. To simulate these areas, we investigated the interaction of vegetation type; EAV or SAV; and hydrology; continuously flooded or periodic drawdown; on the P removal capacity in mesocosms packed with peat soil obtained from STA-1W. The surface water had lowTP concentrations with an annualmean5 0.023mgL. For SRP and TP, hydrologic fluctuations alone had no discernable impactonPtreatmentwhilevegetationtypeshowedasignificant impact. Influent soluble reactive P (SRP) decreased by 49% for the SAV treatments compared with 41% for the EAV treatments, irrespective of hydrology treatment. The reduction of dissolved organic P (DOP) was also higher for the SAV treatment averaging 33% while showing a reduction of 11% for the EAV treatments. There was no significant difference in the treatment efficiency of particulate P (PP) across the treatments. For TP, SAV treatments removed 45% of TP while EAV removed significantly less at 34%. By mass calculations, the EAV required 85%more P for plant growth than was removed from the water column in 1 yr compared with only 47% for the SAV. Therefore, the EAV ‘‘mined’’ substantially more P from the relatively stable peat soil, translocating it into the detrital pool. PHOSPHORUS RETENTION by constructed wetlands may include the following processes: surface adsorption on soil minerals, precipitation reactions, microbial immobilization, and plant uptake (Reddy et al., 1995). These processes may be combined into two distinct P retention pathways for wetlands: sorption and burial (Reddy et al., 1999). Phosphorus sorption includes both adsorption and precipitation reactions as mechanisms for the removal of phosphate from the soil solution to the solid phase. As plants senesce, some of the P contained in detrital tissue can be recycled within the wetland, and released into thewater column.Remaining refractory detrital tissue may eventually become incorporated as organic matter in the wetland soil profile as organic matter accretes. Accretion of organic matter has been reported as a major sink for P in wetlands (Craft and Richardson, 1993; Reddy et al., 1993). Wetlands tend to accumulate organic matter due to the production of detrital material from biota and experience relatively low rates of decomposition under flooded conditions (DeBusk and Reddy, 1998). Soil accretion rates for constructed wetlands are on the order of millimeters per year, although accretion rates in productive natural systems such as the Everglades have been reported as high as 1 cm or more per year (Craft and Richardson, 1993; Reddy et al., 1993). Over time, productive constructed wetland systems will accumulate organic matter that has different physical and biological characteristics than the original preconstruction soil. Eventually, this new material settles and compacts to form new soil, which may exhibit a different P removal capacity than the original soil. Phosphorus accretion increases with P loading to the wetland (Reddy et al., 1993). However, an increase in accretion does not assure low surface water outflow P concentrations, especially for intermittently flooded wetland systems where decomposition of organic detritus releases available P back into the water column. Decomposition of detrital material was found to increase under high P conditions (DeBusk and Reddy, 2003, Wright and Reddy, 2001) lower water levels (White and Reddy, 2000), and higher redox conditions (White and Reddy, 2001). Scientific investigations of P reductions in constructed wetlands generally focus on wetlands receiving much higher inflow concentrations (.0.100 mg L) than this study. The goal of this study was to investigate the P treatment capacity of EAVand SAV communities under both continuously flooded and under periodic drawdown with a mean inflow TP concentration of 0.023 mg L.