The black-land in Northeast China is one of three black-lands in the world and has become the largest grain producing area and commercial grain base of China. Therefore, black-land soil erosion caused by natural factors and human activities has attracted much attention. The Revised Universal Soil Loss Equation (RUSLE) developed by the Agriculture Research Service of the United States Department of Agriculture provides a comprehensive framework for assessing soil erosion and has been widely applied all over the world. Compared with factors such as rainfall erosivity, soil erodibility, topography, and conservation practice, the vegetation coverage and management factor (C-factor in RUSLE) of cropland encompasses the most easily optimized measures. However, in some studies, the quantification of C-factor only considers vegetation cover and ignores other farmland management measures due to the limitation of field management information at regional scale, which implies a huge room for improvement in quantification of C-factor. In this study, the quantification of C-factor considered not only vegetation cover but also crop residue cover, agricultural machinery total power, and fertilizer application rate. Among them, the former is related to weakening soil erosivity, and the latter two are related to enhancing soil erosivity. Monitored sediment transport modulus (STM) at the sub-watershed scale was used to compare changes in the estimation accuracy of the soil erosion modulus (SEM) before and after improving the C-factor. The results showed that the improved cropland C-factor in RUSLE produced a better linear fit accuracy between SEM and STM, with an average increase of 0.115 for R2. Moreover, the order of SEM from high to low in different years was: 12.01 t·ha−1·y−1 (2019), 11.43 t·ha−1·yr−1 (2005), 11.17 t·ha−1·yr−1 (2010), 11.01 t·ha−1·yr−1 (2015), and 10.30 t·ha−1·yr−1 (2000), which was positively correlated with interannual variation of precipitation in Northeast China. Spatially, the order of multi-year average SEM in agricultural zones from high to low was as follows: 22.53 t·ha−1·yr−1 (Liaoning Plain and Hilly Zone; LPH), 20.44 t·ha−1·yr−1 (Baekdu Mountain Zone; BM), 15.11 t·ha−1·yr−1 (Western Liao River Zone; WLR), 11.36 t·ha−1·yr−1 (Lesser Khingan Mountain Zone; LKM), 7.56 t·ha−1·yr−1 (Greater Khingan Mountain Zone; GKM), 3.41 t·ha−1·yr−1 (Sanjiang Plain Zone; SJP), 3.14 t·ha−1·yr−1 (Songnen Plain Zone; SNP), and 2.06 t·ha−1·yr−1 (Hulunbuir Grassland Zone; HG), whereas the soil loss amount from high to low in various agricultural zones was in the order of BM > LPH > WLR > GKM > LKM > SNP > SJP > HG. Our study verified the feasibility of improving the C-factor of croplands in areas where cropland is the dominant land-use type. Moreover, our method will contribute to the use of RUSLE with higher precision in other regional-scale soil erosion assessments worldwide.