Potential for soil legacy phosphorus release from restored riparian wetlands within an agricultural landscape

Wetlands are valuable ecosystems because they are highly productive, support a wide range of wildlife, and serve as hotspots for biogeochemical cycling. Historically, vast areas of wetlands in the United States (US) were drained and converted to agriculture. Efforts are currently underway to restore wetland and floodplain functioning across the US and elsewhere. Re-wetting historically drained and farmed soils can potentially liberate legacy phosphorus (P) to surface waters as soluble reactive P (SRP), offsetting P retained by sedimentation during floods. A better understanding of the controls on SRP release is needed to estimate net P retention in these settings. Soil P saturation ratio (PSR) and soil P storage capacity (SPSC) are two proxies for SRP runoff risk that have shown promise for characterizing restored wetlands but require further testing. In this study, we examined soils at 42 riparian sites ranging from active farms to mature wetlands in the Vermont portion of the Lake Champlain Basin (USA), where phosphorus load reduction is a critical goal to achieve in-lake water quality targets. We additionally quantified potential SRP release to overlying water using intact soil cores from 20 plots spanning 14 sites. Final SRP concentrations in intact cores spanned two orders of magnitude and were predicted well by SPSC and PSR. SRP release was greatest at more recently and frequently farmed sites. Several soil properties, including PSR and SPSC, were correlated with farming frequency and time since farming, indicating that SRP release could be mapped using existing geodata for soils, hydrology and land use. Our findings confirm that soil SRP release during flooding needs to be considered in estimates of net P balance for restored riparian wetlands in agricultural landscapes.