Soil phosphorus fractions as influenced by phosphorus and nitrogen sources on two sites in Western Kenya.

Abstract

The effects of inorganic phosphorus fertilizers on soil P fractions in the highly weathered soils of the sub-humid tropics are poorly understood. A study to evaluate effects of triple super phosphate (TSP, 20% P) and Minjingu phosphate rock (MPR, 13% P, 3% citrate-soluble P), on soil phosphorus fractions and their relationship to maize yields was conducted on a clay loam Oxisol and a sandy clay loam Alfisol in western Kenya. The two sources of P, applied at a basal rate of 0, 50 and 250 kg ha-1 each were combined with 60 kg N ha- applied as either urea or tithonia [Tithonia diversifolia (Hemsley) A. Gray] green biomass. Sequential and non-sequential extractable inorganic (Pi) and organic (Po) fractions were determined at the end of each cropping season. Microbial biomass P (MBP) and C (MBC) were determined on samples taken two weeks after fertilizer application. All the fractions were correlated against maize yield from the next harvest. Phosphorus applied at the rate of 250 kg P ha-1, as either TSP or MPR had a significant increase on the labile and moderately labile inorganic Pi fractions. Enrichment of Pi fractions above the control indicates that use of inorganic P fertilizers (either TSP or MPR) can substantially increase P in both the liquid capital (approximated by Resin + NaHC03 extractable inorganic P) and reserve capital (approximated by NaOH + HCI extractable inorganic P) fractions. Extractable inorganic P varied between the P sources and sites. For example the TSP-treated soil, NaHC03-Pi and NaOH-Pi were the major P sinks in both soils, while for MPR treated soils the Resin and HCI extractable inorganic P fractions were highest in the Alfisol. MPR treatments showed moderate increases between the 2 seasons in both the labile (RP-Pi and NaHC03-Pi) and moderately labile (NaOH-Pi) fractions at both sites. Extractable Pi fractions from tithonia treatments were generally slightly higher than those from urea treatments. The organic P (Po) fractions were not influenced by the Prate, P source or N source (tithonia vs. urea). Coefficients of determination (r2) of P pools against maize yield were generally higher for TSP as compared with MPR on both sites. Non-sequential RP-Pi, NaHC03-Pi and NaHC03 + EDTA-Pi had stronger correlations with maize yield than Pi fractions from the sequential extractions in TSP-treated soils. Resin-Pi (both sequential and non-sequential) had the strongest correlations with maize yield in MPR-treated soils. The organic P fractions and microbial biomass fractions (P and C) were not significantly correlated with maize yields. The relative agronomic effectiveness (RAE), determined as increase of maize yield due to MPR divided by increase due to TSP, expressed as a percentage, varied with both site and season. The RAE of MPR (averaged over N sources) increased from 72% in season 1 to 92% in season 2 on the clay loam and decreased from 114% to 93% on the sandy clay loam. The RAE for MPR with tithonia as N source, averaged over both seasons, was 77% on the clay loam and 94% on the sandy clay loam, while with urea, RAE averaged 87% on the clay loam and 113% on the sandy clay loam.