Effect of organic inputs from agroforestry species and urea on crop yield and soil properties at Wondo Genet, Ethiopia

The effect of combined inputs of green manure (GM) and urea on maize (Zea mays L.) productivity and soil characteristics was studied in an on-farm trial at Wondo Genet, southern Ethiopia. The GM species used were the legumes Albizia gummifera G.F. Gmel and Milletia ferruginea (Hochst.) Baker and the non-legumes Cordia africana Lam and Croton macrostachyus Del. The GM and urea were applied separately or combined in different proportions to give equal rates of 100 kg N ha(-1), while varying amounts of P and K were supplied by the GM. The experiment was carried out for two consecutive cropping seasons, but treatments were applied only during the first season. Yield increased by 10-84% in the combined treatments compared to the control, with the increase being significant in those treatments containing GM from Cordia and Croton, which had higher P and K but lower N and total polyphenol contents than those from Albizia and Milletia. Increasing the proportion of GM from the non-legumes tended to increase yield, but the opposite was true when the proportion of GM from the legumes was increased. This suggests that N interacted negatively with polyphenols, and/or the P and K supply from GM was more important than that of N. However, treatment effects on soil properties were not pronounced. It was concluded that a 'modest' input of <= 3.0 Mg ha(-1) GM from Cordia/Croton might effect a reasonable yield but that the additional use of inorganic fertilisers is necessary for the GM to have a substantially increased effect. In addition, when organic materials having both high N and high total polyphenol contents (e.g. Albizia and Milletia) are used, they should constitute the smaller proportion of the GM/fertiliser mixture.