Response of three cereal crops in continuous arable or ley-arable rotations to fertilizer nitrogen and soil nitrogen at Rothamsted's Woburn Ley-arable experiment

Rothamsted's Woburn Ley-arable experiment, started in 1938 on a sandy loam soil, provides valuable real-world data on the effects of all-arable and ley-arable rotations. In this study, six rotations were compared from 1973 to 2001. Two had 3-year arable "treatment " crops, two had 3-year leys, and two had 8-year leys; the leys being all-grass given fertilizer nitrogen (Ln3 and Ln8), or grass/clover (Lc3 and Lc8). Here, we present the yields of two test crops, winter wheat (1981-2000) followed by spring barley (1982-1991) or winter rye (1997-2001) in each of the six rotations, and their response to four rates of fertilizer N and soil N. From fitted yield/N response curves, we show that maximum wheat yields were least (7.10 t ha(-1)) in the AB rotation, slightly higher, but not significantly so (7.65 t ha(-1)) following Ln leys but significantly higher (8.12 t ha(-1)) following Lc leys. Significantly less fertilizer N (30 kg ha(-1)) was needed to achieve the higher yields following Lc leys. Yields of the second cereal following the leys were 0.3-0.8 t ha(-1) higher than those in the AB rotation; these increases were not statistically significant. However, significantly less fertilizer N, 26-38 kg N ha(-1), was required to achieve those yields. There was no difference found between the type of ley. The initial benefit of the Lc leys was short-lived. If leys are to be introduced into mainly arable farming systems, they may need to be subsidized to make them financially viable.