Reactive Fe controls the relative amount of PO4 extracted from acidic soils by NaHCO3 and by acidic fluoride I. Soils without recent P additions

Abstract

The relations of reactive Fe and Al fractions (oxalate-extractable) with 2 measures of extractable orthophosphate (Colwell and Bray-1) were examined for 47 acidic, mineral, surface soils from eastern New South Wales. The soils had developed from 3 different parent materials and had a wide range of chemical compositions; most would have been P-deficient in their virgin state. Some of the soils had received P fertiliser to promote pasture growth but none had been fertilised with P for at least 2 years before sampling. Variation among soils in the log of the ratio of values for the P tests (Colwell/Bray-1) was related to the log of the concentration of either oxalate Fe (R2 = 0 · 87) or oxalate Al (r2 = 0 · 24), or P sorption (r2 = 0 · 26). Soil parent material had no significant effect on these relations. Two independent data sets support the strong relation of log (Colwell/Bray-1) with log(oxalate Fe); however, not all of the relations are identical and possible reasons for the differences are discussed. The complete set of relations indicates that under the described conditions, much of the orthophosphate may be bound to reactive Fe. It is inferred that across a set of acidic, mineral soils that have not had recent additions of P and which vary widely in Feox, the critical levels of the Colwell test should vary over a much wider range than the corresponding Bray-1 values. Consequently, for the Colwell test to provide as reliable a guide to P responses as the Bray-1 test, the Colwell test should be used in conjunction with a measure of reactive Fe, such as oxalate-extractable Fe.