Dynamics of Phosphorus Release from Wetlands Restored on Agricultural Land.

摘要

Wetland restoration is done, in part, to improve water quality. However, P release to surface waters has been observed in many wetlands restored from agricultural land. Phosphorus is a limiting nutrient in many freshwater systems, so it is important to understand how wetland restoration on agricultural land may affect surface water quality. Numerous studies have examined such water quality impacts of wetland restorations, but the role that plants play in governing the rate of P loss in isolated wetlands is not well understood. The general goal of this research was to examine the role of wetland vegetation in P cycling within restored wetlands. Special attention was paid to the effects of the rhizosphere of bald cypress (Taxodium distichum L.) roots on P dissolution in greenhouse and field settings in soils from a 291 ha Carolina bay wetland restored from agricultural land named Juniper Bay. The role of plants in P cycling in restored wetlands was assessed by developing a P budget for that same wetland.;The first study used root-box rhizotrons to examine rhizosphere effects on P dissolution. Rhizotrons were planted with bald cypress saplings or left unplanted to simulate rhizosphere and matrix conditions, respectively. Two soil treatments were imposed to simulate the two dominant soil types -- mineral (Aeric Alaquods) and organic (Terric Haplosaprists). The rhizosphere treatment did not cause higher P concentrations in solution than matrix values for either soil type. This was because labile C was not limiting to reduction processes in the matrix of these two soils. Redistribution of roots was observed with root death in deep, reduced soil layers and root growth in oxidizing surface soil layers.;A second study was conducted at Juniper Bay to "field-truth" the root-box rhizotron results. Bald cypress trees were instrumented with minirhizotron tubes, porewater samplers, and groundwater monitoring wells at sites located on mineral or organic soils. The trees exhibited vigorous root growth during drought conditions in 2011, then root death during the wet seasons of 2012 and 2013. However, redistribution of roots from the deeper subsoil to the surface, as was seen in the root-box rhizotron study, was not observed in this minirhizotron study. This is attributed to a difference in plant age and prior exposure to reducing conditions between the two studies. Soil solution chemistry measurements corresponded closely with results from the root-box rhizotron study and suggested that P dissolution was dependent on Fe reduction under saturated conditions.;A third study examined if Juniper Bay is contributing P to surface waters using a P balance. The change in soil P was evaluated between archived samples taken at restoration (2005), and eight years after restoration (2013). The P pool at the time of restoration was 800 kg P ha-1. After eight years of restoration that P pool declined to 740 kg P ha-1, but that difference was not significant at the alpha=0.05 level. Atmospheric deposition contributed 7 kg P ha-1, plants extracted 27 kg P ha-1 and incorporated it into woody biomass, and 0.5 Mg P was lost to surface waters draining the site. Because P loss to surface waters was small, and that P concentrations were not high enough to cause eutrophication (< 0.1 mg/L), we concluded that Juniper Bay is not contributing to the degradation of surface water quality of nearby streams following restoration. This is due to little groundwater flowing either into or out of the site as a result of the small hydrologic gradient that exists in this flat wetland system. Further, "isolated" wetlands such as this Carolina bay are ideal sites for future wetland mitigation projects due to limited impacts on surface water quality.