Prediction of fertilizer requirements for spring wheat grown on stubble in southwestern Saskatchewan

Abstract

Results of fertilizer trials conducted throughout southwestern Saskatchewan were used to develop a production function relating grain yield increases of stubble-seeded spring wheat (Triticum aestivum L.) to available water, soil NO3-N content to 60 cm depth, soil PO4-P content to 15 cm depth, and fertilizer N and P applications. The trials were carried out from 1967 to 1971 and from 1978 to 1983 on mid-textured Brown Chernozemic soils. The model accounted for 67% of the variability in yield increases, and identified water availability as the single most limiting production factor, which by itself accounted for 15% of the variability in yield increases. Further, the effects of soil and fertilizer N and P were only significant in their interaction with water. Response to N fertilization was adequately described by a second-degree polynomial. The rates of fertilizer N required for maximum economic yield increases were related to the availability of soil NO3-N and PO4-P and to the rate of fertilizer P application, and inversely proportional to available water and the ratio of N cost to wheat price. The model indicated that soil NO3-N was equally effective as fertilizer N in raising yields, but considered soil PO4-P only one-fifth as effective as seed-placed fertilizer P in modifying the rates of fertilizer N required for maximum return. The effect of fertilizer P in the model was only significant in interactions with fertilizer N and soil PO4-P. Water use efficiency was proportional to the availability of soil and fertilizer nutrients. Key words: Fertilizer N, fertilizer P, soil testing, moisture use, soil N, soil P