Abstract

Many soils in western Kenya are acidic and potassium (K) deficient. Subsequently, maize (Zea mays) productivity is far below crop potential for the region. While the use of nitrogen (N) and phosphorus (P) fertilizers has been widely accepted among local farmers, K application is still ignored.

The present study aims to sustainably increase maize yields through the application of K under balanced fertilization. The specific objectives were to study maize response to K application rate, with and without lime; to test possible benefits of applying slow-release N fertilizer compared with top-dressed N; and to evaluate the effects of three types of K fertilizers on maize yield. Experiments took place in 2018 in five locations: Ndengelwa and Mabanga (Bungoma County), and Kamidi, Wepukhulu and Githanga (Trans Nzoia County). Two experiments were conducted at each location. The first experiment evaluated eight fertilizer treatments, with lime (2 Mg ha^-1) and without lime. The treatments included six pre-planting K rates of 0, 40, 80, 120, 160, and 200 kg K₂O ha^-1, slow-release N applied with 80 kg K₂O ha^-1, and an unfertilized control (UFC).

All treatments except UFC received pre-planting N and P through the application of di-ammonium phosphate and side dressing with urea to bring the levels up to 150 kg N and 100 kg P₂O₅ per ha. In the second experiment, three types of K fertilizer - muriate of potash (MOP), sulphate of potash (SOP), and NPK-17-17-17 (SSS) were compared at 80 kg K₂O ha^-1, with and without liming.

Application of N and P at different K levels increased maize yields from 2 and 3 Mg ha^-1 under UFC, to 6 and 12 Mg ha^-1, in Bungoma and Trans Nzoia County, respectively.

Water availability significantly restricted the duration of the cropping season and, consequently, reduced yields. Beyond this obstacle, liming displayed insignificant and inconsistent impacts on maize yields. K application displayed a significant potential to increase yields, although adverse effects were evident at rates higher than 40 kg K₂O ha^-1. In conclusion, K application should be divided into separate doses and delivered throughout the season or through slow-release fertilizers. Moreover, the amount and type of K fertilizer, as well as liming requirements, should be evaluated in each location according to the local soil properties and in consideration of the economic benefits.

Keywords: Liming; relative yield; slow release; soil acidity; Zea mays.