The hot water extractable organic carbon (HWEOC) and K2Cr2O7 oxidizable organic carbon (OOC) have been suggested as indicators to assess soil management effects on soil organic matter; however, traditional methods for measuring these C fractions are costly and tedious. The potential of using near-infrared reflectance spectroscopy (NIRS) with partial least squares (PLS) regression to predict HWEOC and OOC concentrations in cultivated and uncultivated Mollisols in China were explored in this paper. The soil organic carbon (SOC), OOC, and HWEOC in 030 cm layer were 37.6, 41.2, and 58.8% lower in cultivated than in uncultivated soils. The HWEOC is more sensitive to soil management relative to SOC or OOC. HWEOC concentrations were accurately predicted using NIRS-PLS model, with high coefficient of determination (R 2=0.89), residual prediction deviation (RPD=3.69) for model calibration, and high R 2 (0.85), RPD (3.03), and correlation coefficient (r=0.92) of predicted and measured values in the validation set. Excellent prediction for OOC was acquired with R 2 and RPD at 0.97 and 6.11 for model calibration, respectively, and R 2 and RPD and r at 0.92, 5.75, and 0.97 for model validation, respectively. This study indicated that the HWEOC could be used to illustrate the impacts of agronomic management on soil quality. Both of HWEOC and OOC can be accurately quantified using NIRS-PLS approach.