Effects of fertiliser on phosphorus pools in soils with contrasting organic matter content: A fractionation and path analysis study

With the intensification of agricultural production in many European countries, more marginal soils with elevated organic matter (OM) content are being brought into cultivation. However, little is known about the transformations in the constituent phosphorus (P) pools of organic soils receiving applications of P fertiliser. Soil P fractions were measured before and after receiving fertiliser in a controlled experiment to determine the change in the soil pools and path analysis was used to evaluate the relationships between P pools. In this study, P deficient soils ranging in OM content from 8 to 76%, were placed in large pots, planted with ryegrass and subjected to P fertiliser applications ranging from 10 to 145 kg ha(-1), and monitored over an eight-month study period. High OM soils had a diminished ability to build-up the labile pool from freshly applied P, with relatively low increases up to 200% of the initial value, compared to mineral soils in which the labile pool increased to > 2500% of the initial concentration. Additionally, organic soils had higher P uptakes in the grass yield than mineral soils, indicating a higher availability of added P in the soil solution than mineral soils due to their limited sorption ability. In general, there was a reduction in the organic P pool over applications from 0 to 55 kg ha(-1), which was indicative of partial mineralization, but was followed by an accumulation of added P over applications from 55 to 145 kg P ha(-1). The residual P pools did not build-up with P additions, but data indicated the occurrence of mineralization in most of the soils with decreases of around 40% of the initial concentrations. Organic and residual pools therefore displayed potential to supply P to more labile P pools across all soils of this study. Path analysis indicated that applied P was the only source of labile P in the soil with the highest OM content, leaving it dependent on continuous P applications to supply P for productivity, whereas in the rest of the soils there were interrelations between the non-labile and labile pools. Low pH strongly immobilised the applied P and should be corrected before the initiation of any fertilisation program, even in soils deficient in plant available P. The results demonstrated that P added as fertiliser to organic soils does not accumulate as in mineral soils, which may leave them susceptible to P losses in surface runoff. Therefore, organic soils under agricultural production located in high status catchments should receive low P applications and only during periods with low probability of precipitation to minimise the possibility of P exports to receiving waters.