Phosphorus retention and soil organic carbon in restored and natural freshwater wetlands

Nutrient (e.g., phosphorus) retention is an important function of wetlands that can improve water quality. We examined soil physical and chemical characteristics and phosphorus (P) sorption capacities in three recently restored herbaceous wetlands (RWs) on previously cultivated soils and three adjacent natural forested wetlands (NWs) on Kent Island, Maryland, USA. Our objective was to compare P retention in these two wetland types. As hypothesized, NW soils differed fundamentally in soil chemistry and had significantly higher total organic carbon (TOC) contents than RW soils (5.7 +/- 1.7% vs. 1.2 +/- 0.1%, respectively, p < 0.05). A number of soil properties (bulk density, pH, labile organic and microbial P, total N, and total N: total P ratios) differed between natural and restored wetlands, as expected from the differences in TOC. Concentrations of pyrophosphate-extractable (organically-bound) Al (Al-p) were an order of magnitude larger in NW than in RW soils (2099.1 +/- 365.5 vs. 767.0 +/- 194.7 kg/ha, respectively). Although past studies have suggested that higher concentrations of organically-bound Al can enhance P sorption, P-sorption capacities were significantly greater in the RW soils, likely due to differences in soil chemistry. In the RWs, 15 soil chemical parameters were significantly correlated with P sorption (based on single factor regression), including residual Al, oxalate-extractable Al and Fe, clay, HCl-extractable Fe and pyrophosphate-extractable Fe (r(2) = 0.90, 0.89, 0.87, 0.85, 0.83 and 0.82, respectively). In contrast, P sorption in the NWs was correlated only with Al-p (r(2) = 0.68). As restored wetland soils are likely in transition from a non-hydric to a hydric state, they should be reevaluated periodically to determine the ultimate effects of this transition on their capacity to retain P.