Landscape-scale variability of nitrogen fixation by pea and the availability of its residue-N for the succeeding crop

Non-Peer ReviewedThe main objectives of this study were to estimate the variability in N2-fixation across a large field and to determine whether the availability of legume residue-N to the succeeding crop was controlled by position of incorporation. In 1991, a field was selected near Blaine Lake, Saskatchewan and a 130 x 130 m grid composed of 169 sample sites was laid out. Six landform elements (upper and lower level, divergent and convergent shoulders and footslopes) were identified and the variability of N2-fixation (15N-isotope dilution) by pea was determined for 60 sites (six landforms x 10 replicates). At each site, the 15N-labeled pea residue was incorporated in a nearby unlabeled area in the spring of 1992. The percent N derived from N2-fixation (% Ndfa) by pea had a median of 57%. A difference in% Ndfa between landforms was observed with the highest% Ndfa at the divergent footslopes (69) and the lowest on the convergent shoulders (28). The total N2 fixed (seed+residue) did not show a landform effect and had a median of 57 kg N ha-1. The total N in pea residue (21-30 kg N ha-1) translated into C:N ratios ranging from 37-56. In 1992, landform differences for grain yield of spring wheat were present. Grain yield ranged from 1160 kg ha-1 on convergent footslopes to 1880 kg ha-1 on divergent shoulders. Due to the early frost, the median harvest index was low (0.24). The% Ndfr (N derived from residue) and% RUE (residue use efficiency) in the wheat grain and residue suggested that almost none of the pea residue-N had become available for wheat. The main reasons for the low N availability of the residue were: 1) incorporation of the pea residue at time of seeding (potential net N-mineralization of the residue in the fall and spring was excluded), and 2) the below average temperatures and precipitation in 1992 which would have reduced soil microbial activity and therefore net N -mineralization