RESUMO
Constructed surface flow (SF) wetlands are commonly used for phosphorus (P) removal. Geochemistry of wetlands provides explicit mechanisms for permanent P-sequestration in sediments. This study had two goals: (1) Find P removal performance and rate at the highest alum doses that do not produce floc in an SF wetland; and (2) Determine potential improvements to P removal performance with low alum doses in a 140-ha land application system downstream from the wetland. The study started with a small fraction of a conventional, flocculation/sedimentation alum dose, then progressively increased the dose to observe initiation of floc formation and removal of P. For flows near 10â¯megaliters per day in an 0.8â¯ha SF wetland, doses started 189â¯Lâ¯d-1 for two weeks, then increasing by 189â¯Lâ¯d-1 every two weeks until the final two weeks at 946â¯Lâ¯d-1. At an alum dosing rate of 189â¯Lâ¯d-1 (alum concentration of 9.5â¯mgâ¯L-1), there was an order of magnitude improvement in P removal rates over literature values. Floc formation in the wetland was observed at 567â¯Lâ¯d-1, but no significant improvement in P removal rates were observed until a conventional alum dose of 946â¯Lâ¯d-1 was applied. Alum addition improved P removal performance in the land application system. In 2014, during which there was no alum dosing, the median effluent total P (TP) during the July-September dry season (groundwater dominated outflow) was 0.43â¯mgâ¯L-1. In 2015, (alum dosing August-October) median dry weather TP of 0.18â¯mgâ¯L-1 was significantly lower (pâ¯<â¯0.0001). Alum dosing in 2016 at 189â¯Lâ¯d-1 produced a dry weather median of 0.28â¯mgâ¯L-1, which was significantly lower (pâ¯=â¯0.015) than in the 2014 median. Mean daily dry weather TP loads to the land application system were 44â¯kgâ¯d-1 in 2014, 45â¯kgâ¯d-1 in 2015, and 41â¯kgâ¯d-1 in 2016.