Accurate estimates of soil organic carbon (SOC) are needed to determine SOC changes resulting from agricultural management practices. The objectives of this study were to: (1) determine total SOC, and estimate contributions of light fraction C (LF-C) and resistant C (RC) to total SOC; and (2) simulate SOC dynamics for three cropping systems in the Pacific Northwest using the CQESTR model. Three cropping systems (sweep-tillage winter wheat (Triticum aestivum L.)–tillage fallow rotation, WW–TF; no-till (NT) spring wheat–chemical fallow rotation, SW–CF/NT; and spring barley (Hordeum vulgare L.)–spring wheat rotation, SB–SW/NT) were established in the summer of 1995, in a randomized complete block design with each rotation replicated four times. Soil samples were taken each year from 1995-2002, and total SOC and LF-C were determined every fall and spring of each year, and RC were determined for spring and fall of 1997 and 2001. Simulated data indicated no significant changes in SOC in the top 30 cm of the WW–TF and SW–CF/NT, however, SOC increased in the SB–SW/NT. Resistant C across cropping systems was not significantly different and constituted an average of 48% of the total SOC. The apparent increase in measured SOC with continuous NT spring cropping was due to accumulated undecomposed crop residues that contributed to the labile C pool, and was confirmed by the LF-C analysis. The contributions of the LF-C to total SOC across cropping systems ranged from 13.4 to 18.4% (fall samples) and 14.4 to 18.9% (spring samples) and increased with time/duration of the study.  The LF-C masked the changes in measured SOC. Differences between the measured and simulated SOC were due to artifacts associated with protocols used to determine SOC that did not completely remove accrued crop residue and was explained by LF-C.