THE INFLUENCE OF LENTIL, CANOLA, PEA AND WHEAT ON CARBON AND NITROGEN DYNAMICS IN TWO CHERNOZEMIC SOILS

Even though the sustainability of crop rotation with pulses has been demonstrated by ancient civilizations, the effects of legume crops on soil organic carbon (SOC) and soil organic nitrogen (SON) dynamics has not yet been adequately quantified in the Canadian Prairies. The main goal of this thesis was to analyze the impacts of incorporating lentil and pea in rotation with wheat on SON mineralization and the fate of recently fixed C in soil. Two soil types were analyzed in this study, a Brown Chernozem (Cz) from Swift Current SK. and a Dark Brown Cz from Scott SK. To quantify the effect of pulse crops on Nitrogen (N) gross mineralization, stable 15N isotope dilution was used and eight different crop rotations were selected (five on the Dark Brown Cz: canola-wheat, pea-wheat, pea-canola-wheat, continuous wheat without N fertilizer, continuous wheat with N fertilizer; and three on the Brown Cz: lentil-wheat, continuous wheat without N fertilizer, and continuous wheat without N fertilizer). These rotations were in the wheat phase at sampling. Additionally, lentil-wheat in the lentil phase was included. Gross N mineralization was evaluated at seeding (May 2009) and anthesis (July 2009). Mean gross N mineralization was not significantly different between crop rotations on any date. Timing of sampling (seeding versus anthesis) had the strongest effect on mineralization with mineralization rates significantly higher at anthesis than at seeding. No significant effect of the soil type was found and rotations with pulse crops had higher mineralization rates than continuous wheat only when compared to unfertilized continuous wheat. Soil N mineralization and SOC dynamic are interconnected and therefore the transformation of crop residue carbon (C) into SOC was also investigated. Lentil, canola, pea and wheat were grown in intact soil cores extracted from the field plots of the N mineralization experiment (at Swift Current and Scott) and pulse-labeled with 13CO2 in a controlled environment to trace crop residue decomposition. At the end of the labeling season (first growing season) 50% of the soil cores were destroyed to estimate root biomass with a 13C technique. Before the second growing season (without 13C labeling), 13C-enriched shoot residues were ground and incorporated into the soil. At the end of the second growing season the remaining cores were destroyed to assess the amount of remaining derived C from the root and the shoot. For both growing seasons, the soil was fractionated into water extractable organic matter (WEOM) very light fraction (VLF), light fraction (LF) and heavy fraction (HF) and the δ13C was determined for each soil fraction. For all crops, estimated root biomass production was markedly higher than estimates in the literature based on root washing and counting methods. In addition, lentil had root biomass production (0-10 cm) significantly higher than the other crops and the lowest shoot:root ratio at 2.5. Since canola produced three times more straw residue than the other crops, its shoot:root ratio was significantly higher at 13.2. At the end of the second growing season, the amount of root -derived C remaining in the VLF, LF and HF had decreased 91%, 61% and 60% respectively. No significant difference was found among crops. At the end of the second growing season on the Swift Current soil (Brown Cz), lentil had more derived C per ha-1 than wheat and on the Scott soil (Dark Brown Cz), canola and pea had more than wheat. Based on these results, the deduced transformation of plant residue C into SOC was VLF first, then LF, then HF, with all fractions contributing C to the WEOM and the WEOM transferring C back to the HF