Optimizing phosphorus availability for bean production on a high altitude andosol

Abstract

In Sub-Saharan Africa, common bean (Phaseolus vulgaris L.) is a food security and income sourcing crop as well as a contributor to soil fertility improvement through biological nitrogen fixation. In Uganda, bean production is widespread, with the highlands in the southwestern part contributing up to 78% of the overall national production. Although bean is endowed with capacity to fix nitrogen, its performance is heavily constrained by low soil fertility, particularly phosphorus deficiency. This is especially so in southwestern Uganda where soil fertility management is rarely practiced. The first step in developing effective P management strategies is to understand P chemistry in terms of sorption and release by soil. Some studies have been done on soils elsewhere in the country, but not on the Andosols that predominate the southwestern part of the country. This study was conducted with the aim of making P available in order to increase bean production in the highlands of southwestern Uganda. Specifically, the study involved determination P sorption capacity as well as desorption/release under laboratory conditions. Greenhouse and field study experiments were also conducted. For the sorption/desorption study, 3 g subsamples of the Andosol were administered with P in the order of 0, 0.015, 0.03, 0.045 and 0.06 mg P equivalent to 0, 5, 10, 15 and 20 mg P kg-1. Then lime (Ca(OH)2) was also factorially administered to the soil and P mixture at rates of 0, 1.5, 2.25 and 3 mg lime per 3 g of soil. These rates corresponded to 0, 10, 20, 30 and 40 kg P ha-1 for phosphorus treatments and 0, 1, 1.5 and 2 t ha-1 for lime treatments. After 56 days of incubation, samples were then equilibrated with 30 ml of 0.01M CaCl2 containing seven varying concentrations of P (0, 200, 300, 400, 500, 1000, 3000, and 5000 ppm P) as KH2PO4 . Based on the Two-surface Langmuir model, Kisoro soil had P sorption maxima range of 1667 to 10 000 mg kg-1, which represented about 95% of the P applied to the soil. In order to satisfy the P sorption sites and, thus initiate P availability for adequate crop growth, the external P requirement was at least 1491.5mg P kg-1. There was no significant effect of applied lime and phosphorus rates on phosphorus sorption capacity of the Andosols. Phosphorus desorption displayed a response curve similar to that of P sorption, with the highest cumulative desorbed P rate of 167 to 204 mg P kg-1 per hour occurring in the first five hours. As for the greenhouse experiment, soil used was collected from Kisoro district, and 4 kg was used in pots of 5-L capacity. Treatments included lime (0, 2, 3 and 4 mg) and P (0, 0.20, 0.30, 0.40 and 0.60 g) applied factorially. These rates corresponded respectively to 0, 1, 1.5 and 2 t ha-1 of lime and 0, 10, 20, 30 and 40 kg P ha-1. The experiment was laid out in a completely randomised design with, three replications. The best lime and P combination, which yielded 12.82 g pot-1 of bean grains, was 30 kg P ha-1 + 1 t lime ha-1. The field experiment was done on-farm in Kisoro district on ridges following the farmers’ practice. Treatments included lime at 0, 1, 1.5 and 2 t ha-1 and P at 0, 10, 20, 30 and 40 kg ha-1. The study was laid out in a randomized complete block design with three replicates. Bush bean variety NABE 14 was the test crop. The best lime and P combination which yielded 2.93 t ha-1 of bean grains, was 30 kg P ha-1 + 1.5 t lime ha-1.