Effects of phosphate buffer capacity on critical levels and relationships between soil tests and labile phosphate in wheat-growing soils

Thirty-nine soils from northern New South Wales were used to examine the effects of phosphate buffer capacity on (i) the extraction of labile phosphate by four soil tests, (ii) the relationships between the four soil tests, and (iii) the critical level of each soil test required for near-maximum yield of wheat under field conditions. The results confirmed the principle, recently proposed by the author, that the larger the negative effect of buffer capacity on extraction of labile phosphate by a soil test, the higher is the correlation between the soil test and plant yield response to phosphate. The acidic ammonium fluoride extractant of Bray and Kurtz was the most sensitive to buffering in this respect, while the alkaline sodium bicarbonate extractant of Olsen el al. was less sensitive and the modified sodium bicarbonate test of Colwell least sensitive to buffering. Whereas a previous glasshouse study suggested that the ammonium fluoride test was over-sensitive to buffering, and hence underestimated available phosphate in strongly buffered soils, this field study showed that the test is correctly sensitive to buffering. Consequently critical levels for near-maximum wheat yields do not vary for the ammonium fluoride tests, but increase with increasing buffer capacity for the sodium bicarbonate tests. The additional measurement of buffer capacity is therefore required to give precision in the use of the sodium bicarbonate soil test and particularly the Colwell test. The results also suggest that a high correlatiori between two soil tests can only be expected where each test is similarly sensitive to buffering, provided of course that both tests extract phosphate mainly from the labile pool