Phosphorus storage and release in response to flooding: implications for Everglades stormwater treatment areas

As part of the Everglades restoration program, 16 000 ha of constructed wet lands will be reestablished on land presently in agricultural production. T hese wetlands will be used to remove Phosphorus (P) from agricultural runof f before it enters the Everglades. Histosols, organic soils, are the predom inant soil type in the Everglades Agricultural Area (EAA), and the conversi on of these soils from drained to flooded conditions has important implicat ions for P storage. Phosphorus storage in organic soils has been shown to b e both positively and negatively affected by anaerobic conditions, In this study, P storage and release was followed in a 146 ha area during its conve rsion from farmland to wetland. The development of a productive biological community, as evidenced by strong diel dissolved O-2 and pH cycles, occurre d within 3 weeks of flooding at one site and 2 months at a second site. Thi s biological community was considered influential in maintaining the low co ncentrations of both N and P in the water column relative to soil porewater concentrations, Maximum total P (TP) and total Kjeldahl N (TKN) concentrat ions of 0.3 and 5 mg 1(-1), respectively, were recorded in the water column following flooding. These concentrations declined to background levels wit hin 2-3 months. Soil porewater TP and total dissolved Kjeldahl N (TDKN) con centrations increased to maxima of 4 and 24 mg 1(-1), respectively, 2 month s following flooding. Nutrient profiles across the soil-water interface wer e used to estimate flux rates. Calculated NH4-N flux rates ranged between 0 .18 and 0.74 mug cm(-2) d(-1) and P fluxes ranged between 0.03 and 0.15 mug cm(-1) d(-1). Phosphorus fluxes from the soil to the overlying water are a function of the mobility of different P fractions. Phosphorus fractions wi thin soil cores, collected immediately upon flooding and again I year after flooding, were identified using bicarbonate (labile), sodium hydroxide (Fe - and Al-bound) and hydrochloric acid (Ca- and Mg-bound) extractions. Labil e inorganic P increased, while labile organic P decreased in response to fl ooding. Phosphorus associated with Ca and Mg increased in the surface 0-45 cm soil profile in response to flooding. These data suggest that immediatel y following flooding the reestablished wetlands will act as a source rather than a sink for P, and P concentrations in the water column will not meet discharge requirements. Although this only occurs for a short time period, steps need to be taken to contain or recycle the water during this initial start-up. Soil fractionation data indicate that while organic P is the prim ary means of P retention within these soils, Ca-phosphates may play a signi ficant role in P storage. Therefore, the reestablished wetlands should be o perated to enhance Ca phosphate formation in addition to biological P uptak e. (C) 2001 Elsevier Science B.V. All rights reserved.