Summary Loss of phosphorus (P) from agriculture contributes to the eutrophication of surface waters. We have assessed the magnitude and controls of P leaching and the risk of colloid‐facilitated transport of P from sandy soils in Münster. Concentrations of soluble reactive P in drainage water and groundwater were monitored from 0.9 to 35 m depth. Total P concentrations, P saturation, and P sorption isotherms of soil samples were determined. Concentrations of dispersible soil P and colloidal P in drainage water and groundwater were investigated. The concentrations of soluble reactive P in drainage water and groundwater were close to background concentrations (< 20 µg P l−1). Median concentrations in excess of 100 µg P l−1 were found down to 5.6 m depth at one of four research sites and in the lower part of the aquifer. Experimentally determined equilibrium concentrations and the degree of P saturation were good predictors of P concentrations of drainage water. Large concentrations of dispersible P were released from soil with large concentrations of oxalate‐extractable P and addition of P induced further dispersion. Colloidal P was transported in a P‐rich subsoil when there was a large flow of water and after nitrate had been flushed from the soil profile and total solute concentrations were small. We conclude that the concentration of soluble reactive P in drainage water is controlled by rapid adsorption in the sandy soils. Subsurface transport of dissolved P contributes substantially to the loss of P from the soils we investigated. Accumulation of P in soils increases the risk of colloid‐facilitated leaching of P.

Fertilization exceeding crop requirements causes an accumulation of phosphorus (P) in soils, which might increase concentrations of dissolved and colloidal P in drainage. We sampled soils classified as Typic Haplorthods from four fertilization experiments to test (i) whether increasing degrees of phosphorus saturation (DPS) increase concentrations of dissolved and colloidal P, and (ii) if critical DPS levels can be defined for P release from these soils. Oxalate-extractable concentrations of P, iron (Fe), and aluminum (Al) were quantified to characterize DPS. Turbidity, zeta potential, dissolved P, and colloidal P, Fe, Al, and carbon (C) concentrations were determined in water and KCl extracts. While concentrations of dissolved P decreased with increasing depth, concentrations of water-extractable colloidal P remained constant. In topsoils 28 +/- 17% and in subsoils 94 +/- 8% of water-extractable P was bound to colloids. Concentrations of dissolved P increased sharply for DPS > 0.1. Colloidal P concentrations increased with increasing DPS because of an additional mobilization of colloids and due to an increase of the colloids P contents. In addition to DPS, ionic strength and Ca(2+) affected the release of colloidal P. Hence, using KCl for extraction improved the relationship between DPS and colloidal P compared with water extraction. Accumulation of P in soils increases not only concentrations of dissolved P but also the risk of colloidal P mobilization. Leaching of colloidal P is potentially important for inputs of P into water bodies because colloidal P as the dominant water-extractable P fraction in subsoils was released from soils with relatively low DPS.

In this study, the supply and input-output balances of phosphorus (P) were investigated for a 10-yearperiod at 85 long-term monitoring sites in German forest ecosystems under the European Level II programme. These sites encompass 23 European beech (Fagus sylvatica L.) stands, 9 oak stands comprised of common oak (Quercus robur L.) and/or sessile oak (Quercus petraea Liebl.), 20 Scots pine (Pinus sylvestris L.) and 33 Norway spruce (Picea abies H.Karst.) stands. We quantified P concentrations in needles and leaves, P inputs from the atmosphere, P outputs through leaching and harvesting, and total P in the soil and humus layers. The P concentrations in European beech leaves from two sites ([1 mg P g -1 dry weight), and in Norway spruce needles from four sites ([1.2 mg P g -1 dry weight), were deficient over several years. In contrast, the oak and Scots pine sites were well supplied with P. When P removal through harvesting was disregarded, P balances were positive or stable (median 0.21 kg P ha -1 a -1 ). With harvesting, balances were mostly negative (median -0.35 kg P ha -1 a -1 ), with longterm P removal from the forest ecosystems.

Increasing the phosphorus (P) saturation of sandy soils may cause an increase in the rate of transport of dissolved P to groundwater. We hypothesize that by increasing sorption of P, soil colloids such as iron (Fe) oxides are also mobilized, because the adsorption of P causes the surface charge to become more negative, which increases the repulsive forces between the colloids and the sand grains, and between the colloids in suspension. Goethite particles adsorbed to fine quartz sand and precipitated goethite coatings on coarse quartz sand were used as model systems to test this hypothesis. Soil samples from a Cambisol Bw horizon and a Gleysol Bg horizon were also investigated. We conducted a series of batch experiments with increasing concentrations of ortho-P and inositol hexaphosphate (IHP). The adsorption of P and the dispersion of colloids were determined by measuring P, Fe, aluminium and carbon concentrations in supernatants before and after ultracentrifugation. Dispersed colloids were characterized according to their optical density, zeta potential and particle size. The addition of P caused mobilization of goethite and soil colloids when a critical P saturation, corresponding to a zeta potential of about À20 mV, was exceeded. To induce colloid mobilization in soils, one to two orders of magnitude larger equilibrium concentrations of dissolved P were necessary, compared with those required for the model systems. The adsorption of IHP reduced the zeta potential of colloids more effectively than the adsorption of ortho-P per mol P. Environmentally significant concentrations of colloidal P (> 0.1 mg P litre À1 ) were released from soil samples at equilibrium concentrations of dissolved P < 0.1 mg P litre À1 . We conclude that the sorption and accumulation of P in sandy subsoils that might occur as a result of excessive fertilization might induce the mobilization of colloids and colloidal P.

The leaching of colloidal phosphorus (P(coll)) contributes to P losses from agricultural soils. In an irrigation experiment with undisturbed soil columns, we investigated whether the accumulation of P in soils due to excess P additions enhances the leaching of colloids and P(coll) from sandy soils. Furthermore, we hypothesized that large concentrations of P(coll) occur at the onset of leaching events and that P(coll) mobilized from topsoils is retained in subsoils. Soil columns of different P saturation and depth (0-25 and 0-40 cm) were collected at a former disposal site for liquid manure and at the Thyrow fertilization experiment in northeastern Germany. Concentrations of total dissolved P, P(coll), Fe(coll), Al(coll), optical density, zeta potential, pH, and electrical conductivity of the leachates were determined. Colloidal P concentrations ranged from 0.46 to 10 micromol L(-1) and contributed between 1 and 37% to total P leaching. Large P(coll) concentrations leached from the P-rich soil of the manure disposal site were rather related to a large P-content of colloids than to the mobilization of additional colloids. Concentrations of colloids and P(coll) in leachates from P-poor and P-rich columns from Thyrow did not differ significantly. In contrast, accumulation of P in the Werbellin and the Thyrow soil consistently increased dissolved P concentrations to maximum values as high as 300 micromol L(-1). We observed no first-flush of colloids and P(coll) at the beginning of the leaching event. Concentrations of P(coll) leached from 40-cm soil columns were not smaller than those leached from 25-cm columns. Our results illustrate that an accumulation of P in sandy soils does not necessarily lead to an enhanced leaching of colloids and P(coll), because a multitude of factors independent from the P status of soils control the mobility of colloids. In contrast, P accumulation generally increases dissolved P concentrations in noncalcareous soils due to the saturation of the P sorption capacity. This indicates that leaching of dissolved P might be a more widespread environmental problem in areas with P-saturated sandy soils than leaching of P(coll).