A landscape-scale and small-plot assessment of the nitrogen and non-nitrogen rotation benefits of pea

The inclusion of a pulse crop in a rotation often leads to higher seed yields for the succeeding cereal crop. This effect is known as the rotation benefit. It is not clear how much of the yield advantage is caused by factors related to soil N availability (N benefit) or those factors not related to soil N availability (non-N benefit). The major objective of this study was to quantify the N and non-N benefits of pea to a succeeding wheat crop. Pea-wheat and wheat-wheat rotations were established in two adjacent 1-ha areas in 1993 using a landscape-scale research approach. The same rotations were established in small plots at three sites. Percent N2 fixation by pea was 16 percentage units lower in the high-catchment footslopes as compared to the shoulders and low-catchment footslopes in the landscape-scale study. A poor correlation $(r=0.377)$ was found between estimates of percent N2 fixation estimated with two similar isotope dilution approaches despite corresponding spatial patterns. The poor correlation was thought to be the result of micro-scale controls which ultimately regulated N2 fixation. In the second year of the landscape-study, wheat seed yield was 982 kg ha-1 greater and total N accumulation was 46 kg ha-1 greater in the pea-wheat than in the wheat-wheat rotation. The non-N rotation benefit of pea was credited to reduced severity of leaf disease and grassy weed infestation, and explained 91% of the yield advantage in the pea-wheat rotation. The N rotation benefit was ascribed to a 129 kg N ha-1 increase in the A-value (soil N supplying power) and accounted for 9% of the yield advantage in the pea-wheat rotation. In both rotations, seed yield was about 400 kg ha-1 lower in the high-catchment footslopes as compared to the low-catchment footslopes and shoulders. In the pea-wheat rotation, the landform effect on seed yield was related to greater soil water and inorganic N content in the high-catchment footslopes. Grassy weed infestations were associated with the yield reduction in the high-catchment areas of the wheat-wheat rotation. In the small-plot study sites, wheat seed yield was 43% (698 kg ha$\sp{-1})$ greater when preceded by pea rather than wheat. Six to 14 kg ha-1 of the extra 27 kg ha-1 of N accumulated by wheat in the pea-wheat rotation was derived from the N in incorporated pea residue. The A-value was 114 kg N ha-1 greater in the pea-wheat as compared to wheat-wheat rotation, a difference which explained 8% of the rotation benefit. The remaining 92% of the yield advantage in the pea-wheat rotation was attributed to reduced root diseases and unidentified non-N effects (e.g., P availability). The larger rotation benefit from pea in the landscape-scale study could be due to topographic variation not included in small level areas of a small-plot research site. Footslope and depressional areas can serve as the source of diseases and weeds for the entire field, and accentuate the expression of the non-N rotation benefit of pea to the succeeding wheat crop