INORGANIC AND ORGANIC PHOSPHORUS DYNAMICS DURING A BUILDUP AND DECLINE OF AVAILABLE PHOSPHORUS IN AN ULTISOL

Development of efficient, cost effective P management strategies for h ighly weathered tropical soils is limited by our understanding of the fate of fertilizer P and the availability of organic P. A 4-year field study an an Ultisol was conducted to examine P fertilizer management options. Four levels of fertilizer P were applied to each of four crop s during the first 2 years. The cumulative P fertilizer input for each treatment after 2 years was 0 (control), 155 (low), 310 (medium), and 930 (high) kg P ha(-1),The residual benefits of these inputs were mea sured throughout the next 2 years, during which no additional P was ap plied. A sequential P fractionation procedure (extraction with Fe oxid e impregnated filter paper (FeO), 0.5 M NaHCO3, 0.1 M NaOH, 1.0 M HCl, concentrated HCl (Con-HCl), and H2SO4 digestion) was used to measure progressively more recalcitrant inorganic (Pi) and organic (Po) P frac tions,Total soil P (sum of all P pools) in the control, which received no P fertilizer, was 1780 mg P kg(-1). Despite this high amount of P, soybean dry matter yield and P uptake increased with P inputs. Dry ma tter yield and P uptake declined during the first year of the residual phase for all P treatments. In the high P treatment, approximately 9, 54, and 37% of the added fertilizer P was recovered from labile (FeO and NaHCO, Pi), moderately labile (NaOH Pi), and recalcitrant (Cone. H Cl Pi and H2SO4 P) pools, respectively, 3.5 months after the first P a pplication. The distribution of fertilizer P among the pools remained relatively constant throughout the 4-year experiment, suggesting that the observed decline in plant available P, after P additions ceased, w as not caused by conversion of P to more recalcitrant forms. Across al l P treatments, FeO and NaHCO3 were correlated highly (r(2) > 0.95) wi th Mehlich-1 extractable P but not with soybean yield or P uptake. Tot al Po (excluding Po in H2SO4 P) was 18% of total P. The NaHCO3 Po pool declined at the same rate as soil organic C and total N, suggesting t hat NaHCO3 Po and C mineralization may be coupled. In the control, NaH CO3 Po was correlated with labile Pi and soybean dry matter yield and P uptake, suggesting that mineralized Po is an important source of pla nt P in unfertilized systems.