Conventional soil test phosphorus failed to accurately predict dissolved phosphorus release in agricultural hydromorphic soils in Brittany, Western France

Abstract

Researches have proved that agricultural phosphorus (P) loss contributes significantly to surface water eutrophication. Various soil test P (STP) methods have been developed to assess the P loss risk from agricultural soils. In the intensively-cultivated Brittany region of Western France, hydromorphic soils in wetland domains exhibit high risks of leaching and transferring dissolved P - the most bio-available form of P - to surface waters. It remains unclear whether STP conventionally developed for well-drained soils can accurately predict the risk of dissolved P release from these hydromorphic soils. In this study, we measured the dissolved reactive P (DRP) concentrations in soil solutions sampled in situ from 26 hydromorphic soils in the Brittany region and examined their relationship with several STPs available on the corresponding soils, such as the degree of soil P saturation, the equilibrium soil P concentration, or the soil Olsen P, Dyer P, and water extractable P contents. DRP concentrations ranged from 0.01 to 0.310 mg P l−1 (mean = 0.075 mg P l−1), highlighting the potential of hydromorphic soils as hotspots for DRP release in agricultural landscapes. Correlations between DRP concentrations and STPs were relatively weak (0.09 < r2 < 0.64), indicating that conventional STPs are generally unable to accurately predict the DRP release risks in hydromorphic soils. Tentatively, Olsen P showed promises as a useful risk indicator, with a relatively high r2 value of 0.6 and wide inclusion in the current STP database, especially in the Brittany region. Nevertheless, this hypothesis requires further evaluation with additional data. This study confirms the high risk of dissolved P release from hydromorphic soils in agricultural wetland domains and emphasizes the need for developing specific risk assessment tools to these hydromorphic soils.