Phosphate fixation and the response of maize to fertilizer phosphate in Kenyan soils

In tropical soils, plant growth is often limited by a low P availability. In addition, these soils often have high P-fixation capacities due to high amounts of iron and aluminum oxyhydroxides. Furthermore, small-scale farming systems in which subsistence crops are produced for local markets are common in the tropics. Such conditions exit in South West Kenya, in the Kish area, where this study was carried out.The overall aim of the study was to improve our understanding of fertilizer P - soil - maize crop interactions and, thereby, to increase the yield response and improve the P recovery and utilization in plants. Initial and residual yield responses to fertilizer P applications; P uptake in plants, as well as P distribution and utilization within the plants; apparent fertilizer P recovery; and dissolution, transport and sorption of fertilizer P in the soils were studied under field conditions during three successive growing seasons. This was done within the concept of the low-input strategy, whereby the high-input strategy was used as a reference.The soils were classified as Ultisols and Mollisols, or as Nitisols and Luvic Phaeozems. They were low in available native P (P-Olsen 1.6 to 3.2 mg/kg) and had medium to high P sorption capacities (445 to 870 mg/kg) measured under laboratory conditions. Total P ranged from 490 to 1,035 mg/kg. Triple superphosphate was used as P fertilizer and hybrid maize as test crop.Good initial and residual yield responses were found in all soils even at low P application rates. The concept of P-fixation, as defined in this study, was not appropriate for the soils in the Kisii area. The second 'slow' phase of the fixation process was shown to proceed at a very slow rate under field conditions. Thus, P fixation is much less of a problem for farmers than expected. When P limited grain yield, the relationship between grain yield (z) and P in plant (y) could be described by the equation: z=620(y-0.50). The efficiency of utilization increased with P application to about 550 kg/kg at P rates of 22 to 131 kg/ha and decreased at higher rates due to a luxurious P uptake. Granular fertilizer residues with P in the form of Fluorapatite, Brushite and an amorphous P compound could be recovered in the soil until the end of the trials, about 600 days after application. Fertilizer P was retained in the soil immediately surrounding the residual fertilizer granules to total P values up to>8,000 mg/kg, which was much higher than predicted from the P-sorption studies in the laboratory. Available P levels (P-Olsen) in the P-enriched soil volume were high, up to>800 mg/kg. The available fertilizer P fraction (ΔP-Olsen/ΔP total) remained fairly constant with time. It was suggested that the number of roots, i.e. root surface area, present in the small P-enriched volume was the limiting factor in fertilizer P uptake. As a result, the apparent P recoveries per crop were low. They were about 10 % each growing season at a P rate of 22 kg/ha.Subseed placement at the start of each growing season is recommended, either placed alone or together with small amounts of cow dung. Then, plants benefit highly from both the startereffect of freshly-applied fertilizer and the residual effect of previously-applied (residual) fertilizer. P application rates of 20 to 40 kg/ha (in P 2 O 5 : 45 to 90 kg/ha) are advised for maize. At these rates, slightly more P is applied than plants require for producing maximum grain yields and the P availability of the soils will gradually increase by enlarging the soil volume enriched with fertilizer P. This repeated subseed placement fits in well with the low-input strategy and is probably easily adaptable in tropical small-scale farming systems.