The long-term use of perennial forages in crop rotations can increase soil carbon (C) and lower nitrous oxide (N2O) emissions compared with continuous annual cropping. However, less is known of the short-term (within 2 yr) benefit of inclusion of perennial forages in an annual crop rotation on net carbon dioxide (CO2) and N2O fluxes. Perennial forage, primarily composed of alfalfa (Medicago sativa L.) and a minor component of timothy grass (Phleum pretense L.) was sown in 2008 on two 4-ha plots previously in annual cropping in the Red River Valley, Manitoba. Spring wheat (Triticum aestivum L.) and industrial rapeseed (Brassica napus L.) were grown on two adjacent plots in 2008 and 2009, respectively. Carbon dioxide and N2O fluxes were measured continuously using the flux-gradient micrometeorological method from 2008 May 01 to 2010 Apr. 30. During the 2-yr study, the newly established perennial forage was nearly twice the sink for atmospheric CO2 (mean and standard deviation of 4480 +/- 1840 kg C ha(-1)) as the annual crops (2470 700 kg C ha(-1)). The annual crop emitted more than four times the N2O (7.8 +/- 0.7 kg N ha(-1)) as the perennial forage stand (1.8 +/- 0.7 kg N ha(-1)). When accounting for harvest C removals (grain, straw, hay) and considering the greenhouse gas (GHG) emissions in CO2-equivalents (eq.), the newly established perennial forage was a net sink of 8470 5640 kg CO2-eq. ha(-1) and the annual crop was a source of 3760 +/- 2450 kg CO2-eq. ha(-1) during the study. The results indicate an immediate reduction in soil GHG emissions with the inclusion of perennial forage in the rotation, primarily from reduced N2O emissions, the lack of crop removal in the forage establishment year and the longer growing season period of net CO2 uptake of the perennial crop.