Evaluation of Phosphorus Fertilizer Materials on Two Benchmark Soils of the Tropics

Abstract

One possible strategy for improving the economic attractiveness of P fertilization on acid P-deficient soils in the tropics was investigated agronomically in two field experiments on sites of the Benchmark Soils Project in the Philippines and Indonesia. The specific strategy is the use of phosphate rock for direct application or partially acidulated or thermally altered phosphate rock in place of acidulated phosphates, e.g., superphosphate. The experiments were conducted on Hydric Dystr'andept soil on a site near Naga City, Camarines Sur, Philippines and on a Typic Paleudult on a site located near Kotabumi, Lampung on the island of Sumatra, Indonesia. The experiments compared crop response (primarily maize) to one highly reactive phosphate rock (North Carolina, NC) and one moderately reactive phosphate rock (Central Florida, CF), which were used because they are well characterized with respect Co reactivity. These two P-sources were used in finely-ground form and also each was used in a different minimally altered form. For the North Carolina rock, a minigranular form was used form, Co determine if the minigranulation process can be used Co overcome the handling and transportation problems of the dusty, finely-ground form without reducing agronomic effectiveness. For the Central Florida rock a minigranular form which was 20% partially acidulated with phosphoric acid was used as the minimally altered form to determine if partial acidualcion of moderately reactive phosphate rocks could adequately improve agronomic effectiveness on acid soils. Comparing the results for the two soils sites in the first crop, a large difference between sources is seem on the Hydric Dystrandept, while no significant differences are seen between sources on the Typic Paleudult. In the second crop, fresh applications of superphosphate were made on former control plots on both sites and the other treatments were left as residuals. On the Hydric Dystrandept significant differences between sources are seen in the residual effects. On the Typic Paleudult, soybeans were planted, and no significant differences are seen between residual effects of the sources. The higher P requirement evidenced by the Hydric Dystrandept as compared to the Typic Paleudult, made reapplication necessary at all rates for the third crop on the Hydric Dystrandept, whereas for the Typic Paleudult, reapplication of all sources was made only on the LO kg/ha plots, on which 70 kg P/ha were applied to permit comparison with the residual 80 kg P/ha treatments. The third season yield results again show no significant differences between sources on the Typic Paleudult, even for the reapplications of 70 kg P/ha, whereas some significant differences are seem for the reapplication on the Hydric Dystrandept. With respect to rock modifications on the Hydric Dystrandept, where differences between sources are detected, minigranulation of the NC rock did not reduce its effectivesness, thus confirming the agronomic viability of the process. Also on the Hydric Dystrandept, partial acidulation did improve the performance of the OF rock, but not sufficiently to make it equal Co superphosphate or the NC forms. In summary, these results indicate that the Typic Paleudult is well suited for direct application of phosphate rocks of moderate and possibly lower reactivity, whereas the Hydric Dystrandept requires phosphate rocks of high reactivity to obtain yields comparable to those of superphosphate. This difference in performance of phoshpate rock between the two soils can be related to the slightly lower pH of the Typic Paleudult and also, its lower P-sorption capacity, in comparison to the Hydric Dystrandept. The results from the post-harvest P soil test analyses indicate that neither the modified Truog nor the Bray P I procedure can be satisfactorily used to predict residual P in the Hydric Dystrandept when P-sources varying in solubility are used. In the Typic Paleudult there is some indication that the Bray I procedure underestimates residual P from the CF sources; however, it appears that minor modification of the procedure by increasing the solution to soil ratio could overcome this problem. This is considered to be a worthwhile area for research since direct application of phosphate rock over the complete range of solubility appears to be an agronomically advisable practice for the Typic Paleudult and similar soils.