Paul Murphy, Forestry and Environmental Management, Univ of New Brunswick, 28 Dineen Dr, Fredericton, NB E3B 6C2, Cape Verde, Andrew Bell, Queens University Belfast (formerly), Belfast, Ireland, and Benjamin Turner, Smithsonian Tropical Research Inst., Unit 0948, Apo, AA 34002-0948, Panama.
To assess the potential contribution of soil organic phosphorus to pasture nutrition and phosphorus loss to surface waters, the chemical nature of soil phosphorus in 25 Irish grassland soils was determined by NaOH–EDTA extraction and solution 31P NMR spectroscopy. All soils were under perennial ryegrass for silage or grazing and included 20 basaltic soils that do not typically show a response to P fertilization, despite low concentrations of bicarbonate-extractable P. Bicarbonate- and water-extractable P were ineffective indicators of potentially plant available P in basaltic soils due to their high Fe and Al content and resultant strong sorption of P. At least some of this undetected plant available P is likely to be in organic form, probably largely as phosphate monoesters. Degree of P saturation (DPS) is a better indicator of P status for such soils and nutrient management should account for this. Soil organic P was dominated by orthophosphate monoesters, with smaller concentrations of DNA and phosphonates. Inorganic orthophosphate was a large proportion of the extracted P, with smaller concentrations of pyrophosphate. The distribution of P species was strongly influenced by their relative stabilities; DNA and phosphonate were associated with higher OM content and lower pH, indicating poor drainage conditions and tendency to become saturated. P species distribution was also strongly influenced by management history; P accumulated largely as orthophosphate and phosphate monoesters with increasing P loading, but to a greater degree as orthophosphate. While phosphate monoesters dominate lower P soils, orthophosphate dominates high P soils, which pose the greatest risk of P loss to surface waters.