An investigation of the limits of phosphorus removal in wetlandsUSaa mesocosm study of a shallow periphyton-dominated treatment system

摘要

The performance of a mesocosm-scale experimental wetland system (9-cm deep raceways with a limerock substrate followed by a limerock filter) that received effluent from a treatment wetland in south Florida was monitored for 19 months.The raceways were operated at a constant hydraulic load of 11 cm per day,resulting in a nominal hydraulic retention time (HRT) of 0.8 days.The vegetation community that developed in these systems varied along the longitudinal phosphorus (P) gradient and became dominated by eutrophic periphyton at the inflow (Cladophora and Microspora) and oligotrophic taxa along the raceway's mid-regions and outflow (Schizothrix and Scytonema).The inflow total P (TP) concentration (mean = 18 mugPL~(-1)) was reduced to a mean of 10 (mug P L~(-1) at the raceway outflow and 8 mug P L~(-1) exiting the limerock filters.Total P removal for the entire system was 0.32 g m~2 per year,or 46% of the inflow TP load,and uptake rates (k values) were 24 and 32 m per year for the raceway and the raceway/limerock filter system,respectively.Most TP removal occurred within the first 3 m of the raceways at an HRT of 1.3 h.Both soluble reactive and particulate P were reduced to their analytical detection limits,while dissolved organic P was removed less efficiently.There was little evidence for a diel trend in P removal.Following a 1-week dry-out,the raceways exported P for only 24 h after rehydration,after which treatment performance quickly returned to pre-desiccation levels.Nutrient ratios suggest that the raceways were never N-limited and that P limitation increased with distance from the inflow ports.Accrued sediment had a high percentage of non-labile P (>70%).A number of operational and construction issues need to be resolved before this technology could be scaled up for use in conjunction with Everglades restoration.