Dynamics of Current and Residual Phosphorus in Tropical Acid Soil

The use of phosphate rocks (PRs) has been proposed as an agroeconomically sound alternative to the more expensive superphoshates particularly, for P 'recapitalization' based on their purportedly high residual effects in acid tropical soils. To understand the current and residual dynamics of P from different P sources, one laboratory incubation trial and two field experiments were conducted. Experimental treatments comprised factorial combinations (3x4~2o) f three P sources [Triple superphosphate (TSP), Gafsa PR (GPR) and Christmas Island PR (CIPR)] at four P rates with and without manure in three replications. Setaria grass (Setaria Anceps Stapf Cv. Kazungula) was used as the test crop and was sequentially harvested bimonthly for 14 months in the field. Results indicated that the amounts of P extracted from treated soils using three soil tests: Mehlich-3 (M3P), Bray-I (B1 P) and Pi-strip (Pi-P) were in the order: M3P>Bl P (85% M3P) >Pi-P (53% M3P) and were significantly related ( ~ ~ = 0 . 4to2 0 .83, n=294). Phosphorus extracted from the different P sources was in the order: TSP>GPR>CIPR and increased with increasing P rates. When P rates were raised from 0 to 300 kg decreased substantially at the highest P rate. The degree of phosphorus saturation (DPS) (defined as a ratio of P already adsorbed to P adsorption capacity of a soil) also varied with P sources (5.1 to 15.8%) and extractants (2.1 to 44%) following a similar order to extractable P. The variously estimated DPS values were all significantly correlated (r =0.91 to 0.98) and therefore were equally suitable in estimating DPS. Sequential strip P indicated that P released was described by power (~~=0 . 7to9 0.95) (TSP) and exponential (PR) functions ( ~ ~ = 0 . 7to7 0.99), while from the field trial, a power function described RV of P in both PRs and TSP (~*=0.64to 0.96). Total extracted Pi-P was related to NaHCOsinorganic and organic P (Bic-Pi and Po), and HCI-P. However, though the Pi-strip could estimate P release, it could not estimate residual value (RV). Residual value is the ratio of amount of freshly applied TSP required to produce yield X to the amount of previously applied fertilizer required to produce the same yield X. The dry matter yield (DMY) exhibited a quadratic relationship with P rates. The maximum DMY (6- 11 t ha-') was attained at 150-200 kg P ha-' and over time, DMY increased to a maximum(11 t ha") and then declined to a constant yield (24 t ha.') after one year. Manure-CIPR integration increased DMY while, manure-GPR and manure-TSP integration depressed yields except in the initial harvest. Setaria DMY was found to be related to NaOH-organic and -inorganic P (Hyd-Po and Hyd-Pi, respectively), and BioPo fractions, which are not accounted for in M3P or B1P thus explaining the low R~ between DMY and M3P (~~=0.0o8r) B1P (R~=o1.0 ). The current RVs of the three fertilizers determined in the field were 100, 100 and 140% for CIPR, GPR and TSP, respectively at 100 kg P ha-'. At the same rate, fertilizer-manure integration increased initial RV to 180, 160 and 110% in TSP, CIPR and GPR treatments. Wth an exception of CIPR-manure, RV was depressed by manure in the other fertilizers in subsequent harvests after the initial one. The RV also declined with increasing P rates as evidenced by current RV decline to 20, 60 and 50% in CIPR, GPR and TSP, respectively when P rate was raised from 100 to 300 kg P ha". The RV declined to 3040% for the first six months and then gradually to 10 - 20% after one year (power function). These results indicated that GPR and CIPR-manure combination were as good as TSP and the optimal P rate was 100 - 150 kg P ha". The results did not support P 'recapitalization' in this soil but annual applications. The RV could be estimated from the P fractions: Bic-Po, Hyd-Pi and Hyd-Po, and DPS computed from Mehlich extracted Al, Fe and P. The RV had no relationship with M3P or B1 P.

Text FP_2005_2.pdf Download (1MB)

Actions (login required)

View Item