Limited data on greenhouse gas emissions (GHG) emission from cropland suggest the need to understand the interaction of tillage, manure and conventional fertilization on carbon (C) sequestration and GHG emission from cropland. The objective of this work is to quantify GHG emission affected by tillage and liquid manure application using eddy covariance techniques in a corn-soybean rotation. Two adjacent fields (32.4 ha each), one with manure and the other with fertilizer, on similar soil types were instrumented with eddy covariance towers. The first dry cattle manure application was 17 October, 2003 on corn stalks and moldboard plowed on 20 October. Both corn (2005) and soybean (2006) were planted at the normal times for this area. The net cumulative C flux in 2005 was higher on the manure treatment than the fertilizer treatment. The net accumulated C flux for corn in 2005 was considerably larger than that for soybean in 2006, related to crop biomass. As of 4 October, 2006 (harvest date), the net cumulative carbon flux was 24.4 g C m^-2 higher on the manure treatment compared to the fertilizer treatment. The extreme drought (74 mm rain in 64 days) starting after 25 June caused both fields to show stress symptoms and the data indicated the manure treatment evapotranspiration (ET) was 43 mm higher than the non-manure treatment at harvest. After soybean harvest, the manure system received an additional 149,600 l ha^-1 of dairy manure incorporated on 14 October, 2006. For the first 10 days after the liquid manure application, the manure treatment exhibited 21 g C m^-2 more loss than the non-disturbed conventional fertilizer treatment. Short-term tillage effects showed flux increases that were subtle, but realistic based on soil temperature and water conditions. The annual carbon budget for conservation tillage systems demonstrates soil carbon(C) loss from year round emissions.