Long-term cropping effect on phosphorus fractions in an ultisol of the humid forest zone in West Africa

The nature of phosphorus (P) pools in soils after application of P fertilizers should be considered when designing management practices to raise P fertility. Five rates of P: 0(P-0), 45(P-45), 90(P-90), 135(P-135) and 180(P-180) kg ha(-1) in the form of triple superphosphate (TSP) were applied once in 1993 on a Typic Paleudult at Man in Cote d'Iviore and the soil was cropped successively for five years (i.e., up to 1998) with four upland rice cultivators. At the end of the cropping season in 1998, soil samples were collected from the various P treated plots and the following P fractions were analyzed: resin-P, NaHCO3-Pi, NaOH-Pi, HCl-P residual P and organic P. The P fractions were compared with those of stored soil samples collected in the field before the initiation of the experiment in 1993. Five years of cropping the soil reduced labile Pi (resin-P + NaHCO3-Pi) from 10.8 mg P kg(-1) in the uncultivated soil to 7.62 mg P kg(-1) in the control (PO) and to 9.71 mg kg(-1) in the P-45 treatment. Labile Pi in P-90 was the same (10.7 mg P kg(-1)) as in the uncultivated soil but in the P-135 and P-180 soils, labile Pi increased by 1.3 and 2.7 mg P kg(-1) above that of the uncultivated one. The NaOH-Pi and residual P fractions of lower availability increased with increasing rates of P application. The results suggested that on the Ultisol studied, P applied once to a,field, cropped successively over p period of time was not an adequate management option. Frequent P application at low rates at each cropping season is suggested and merits investigation on the Ultisol studied.