Phosphorus sorption, accumulation and leaching

Arable land is a major diffuse source of P loads to surface waters. The main processes for P losses from agricultural fields are erosion, surface runoff and subsurface leaching. To identify agricultural soils vulnerable to P losses, it is necessary to understand the processes behind these losses. This thesis mainly concerns processes influencing subsurface leaching. Phosphorus sorption and accumulation were studied in the laboratory to evaluate how these processes affect the release of P, which potentially can be leached. Furthermore, P leaching from undisturbed soil cores exposed to natural conditions was studied. All studies were performed on soils from the Swedish long-term fertility experiment, which includes soil types common in southern and central Sweden, each with four soil P levels. P sorption capacity in the soils studied ranged from 6.0 to 12.2 mmol kg-1 soil and was correlated with the amounts of oxalate extractable Fe and Al in the soils. Phosphorus release was correlated with soil test P values (STP) for individual soils, but STP could not explain P release when soils with different P sorption capacities were considered. Instead, estimates of P sorption saturation using a ratio of STP and P sorption capacity proved to be useful for assessing P release. Long-term inorganic P fertilization of 42 to 49 kg P ha-1yr-1 for more than 30 years caused P accumulation down to 40 cm depth in the soil profiles. Below this depth no differences in P properties were found between fertilized and non-fertilized soils. Total P losses from the lysimeters ranged from 0.03 to 1.09 kg ha-1yr-1 and were not correlated with either STP or P sorption saturation. Total P leaching was greatest from a soil in which macropore flow dominated water transport, irrespective of STP. In another soil, with low P sorption capacity, P sorption saturation and P leaching were high from the highest fertilized treatment. In the remaining soils, P sorption and release processes in the subsoils influenced P losses. The subsoils seemed to act either as a source or a sink of P and reduced the differences between treatments.