A comparison of soil test calibrations for the estimation of phosphorus fertilizer requirements of wheat

Abstract

The value of a range of soil analyses as tests for the phosphorus fertilizer requirements of wheat is assessed, for northern New South Wales. The best individual analyses were the standard soil tests, P extracted by ammonium lactate (Egner), 0.5M sodium bicarbonate (Colwell), 0.01N sulphuric acid (Kerr and Von Stieglitz), and Truog's reagent, and the best calibration equation was obtained using both the ammonium lactate and sodium bicarbonate tests. Sorbed phosphate equilibrating with the phosphorus in these extract solutions, estimated by isotopic dilution, made only small contributions to test calibration equations, in the presence of other simpler P analyses. The likely effects of variation in P sorption capacity were represented by analyses for iron and aluminium soluble in Tamm's acid oxalate reagent, but these analyses added little to the calibration equations, suggesting that P sorption capacity is not an important variable affecting wheat production in northern New South Wales. Analyses of variance on the yield data demonstrate the importance of experimental error in field experiments as a factor limiting the precision of soil test calibrations.