Assessment of the long-term impact of fertilizers and other management factors on crop production and environmental sustainability of cropping systems in the semi-arid Canadian prairies is needed. This paper discusses the long-term influence of N and P fertilizers on crop production, N uptake and water use of hard red spring wheat (Triticum aestivum L.), and the effect of the preceding crop type [flax (Linum usitatissimum L.) and fall rye (Secale cereale L.)] on wheat grown on a medium-textured, Orthic Brown Chernozem at Swift Current, Saskatchewan. We analysed 36 yr of results (1967–2002) from eight crop rotation-fertility treatments: viz., fallow-wheat receiving N and P (F-W, N + P), three F-W-W treatments fertilized with (i) N + P, (ii) P only, and (iii) N only; two other 3-yr mixed rotations with N + P (i) F-flax-W (F-Flx-W) and (ii) F-fall rye-W (F-Rye-W); and two continuous wheat rotations (Cont W), one receiving N + P and the other only P. Growing season weather conditions during the 36-yr period were near the long-term mean, but the first 22 yr were generally drier than normal while the last 14 yr (1989–2002) had average to above-average growing conditions. This was partly responsible for grain and N yield being greater in the latter period than in the first 22 yr. The 36-yr average response of wheat grown on fallow to P fertilizer was 339 kg ha-1, while the response to N fertilizer over this period was only 123 kg ha-1. The 36-yr average response of wheat grown on stubble to N was 344 kg ha-1 for F-W-(W) and 393 kg ha-1 for Cont W. Neither flax nor fall rye influenced the yield response of the following wheat crops. Annualized grain production for F-W (N + P), F-W-W (+ N) and F-W-W (+ P) rotations were similar (1130 kg ha-1 yr-1); this was about 15% lower than for F-W-W (N + P), 40% lower than for Cont W (N + P), and 5% lower than for Cont W (+ P). Annualized aboveground N yield for Cont W (N + P) was 57% higher than for Cont W (+ P). Regressions were developed relating straw to grain yields for wheat, flax and fall rye. The amount of NO3-N left in the soil was directly related to amount of N applied and inversely to N removed in the crop. Thus, F-(W)-W (+ N) left about 28% more NO3-N in the rooting zone than F-(W)-W (N + P), while F-W-(W) (N + P) left 20% more than F-W-(W) (+ P), and Cont W (N + P) left 39% more than Cont W (+ P). F-Rye-W (N + P) left much less NO3-N in the soil than any other fallow-containing system and similar amounts to Cont W (N + P). Key words: Yields, grain protein, N and P fertilizer, straw/grain regressions, water use, soil nitrate
