A nitrogen (N) x phosphorus (P) fertiliser experiment commenced in 1985 on a uniform area of a deep Black Vertosol in the Darling Downs, Southern Queensland, Australia. Bicarbonate extractable P (Colwell, 1963) in the surface soil of the nil P treatment has remained unchanged despite continuous cereal cropping and the removal of approximately 200 kg P/ha since the start of the experiment.

As the labile bicarbonate extractable pool was not declining, it was hypothesized that plants were accessing soil P either from subsoil layers (10-60 cm) or from surface soil P pools not extracted by the Colwell method. To investigate this, surface and subsoil soil samples were collected in 2003 for detailed P fractionation. Treatments sampled had had 80 kg N/ha applied with every crop, and either 0 or 20 kg P/ha.crop. Care was needed in sampling, because zero tillage practices result in bands of residual P in the +P treatments. Surface 0-10 cm samples were collected from multiple points from an area 60 cm across x 10 cm along the row.. Samples at depths 10-30 cm and 30-60 cm were taken using a 32 mm soil core at 5 points across the bed. Variability from samples within a plot was negligible using this sampling method. A reference sample was taken from 20 m outside the planted area from soil that had been uncropped since the start of the trial . Phosphorus was measured in soil extracts obtained by using, in sequence: anion exchange resin, 0.1 M sodium bicarbonate, 0.1 M NaOH/1 M NaCl, 1 M HCl, and perchloric acid/sulfuric acid digestion for the residual pool (Guppy et al., 2000). For the NaOH and HCl extracts, both inorganic and organic P were measured. Archived soil samples from 1994 were available for some treatment combinations, and these were used to assess changes in P fractions over time.

Compared to the nil treatment, the 20 kg P/ha.crop treatment, showed a significant increase in the resin , bicarbonate, inorganic hydroxide and inorganic acid pools for the 0-10 cm sample depth. Below 10 cm depth there was no change in the size of any of the P fractions between 1994 and 2003. No change in organic P fractions was observed in the last 9 years of intensive cropping. By comparison with the reference site, an estimated 180 kg P/ha has been removed in grain from the top 30 cm of the profile since 1985, 95% of which appears to have come from the surface 10 cm. This roughly equates to that estimated to have been exported from the site over 18 years. However of the total P removed from the surface soil, only 20% was removed from the labile P fractions (resin/bicarbonate/hydroxide); with 40% coming from the acid extractable fraction and 40% from the residual P fractions. Where banded P fertiliser was regularly applied, the surface soil should have accumulated approximately 110 kg P/ha. Although increases in labile and acid inorganic P fractions account for 90 kg P/ha, concomitant decreases in the organic and residual fractions resulted in no increase in total P levels over 18 years.

Uncertainty remains as to whether plants are driving the transfer of P from less available P fractions through to labile P fractions and buffering the soil test P level over time, or whether plant roots are directly accessing P fractions previously considered to be unavailable. It is clear that plants in this Black Vertosol a) do not access P from below 30 cm; b) do not significantly rely on organic P cycling for plant P requirements and c) may be reaching the limit of their capacity to draw P from residual and acid P fractions without significant inputs of P to labile P fractions. Further research is needed to identify whether particular soil P fractions can be used to indicate how long P can be supplied to plants without replacement.

Colwell, J. D. (1963). The estimation of the phosphorus fertiliser requirements of wheat in southern New South Wales by soil analysis. Australian Journal of Experimental Agriculture and Animal Husbandry 3,190-197

Guppy, C.N., Menzies, N.W., Moody, P.W., Compton, B.L., and Blamey, F.P.C. (2000). A simplified sequential phosphorus fractionation method. Communications in Soil Science and Plant Analysis 31, 1981-1991