The aim of this work is to assess the impact of no-till based cropping system on soil erosion using two indicators. The experiment plot was under no-tillage (NT) for four years, adjacent to a plot under conventional tillage (CT) with same other management practices. The two used indicators are Le Bissonnais soil aggregates’ stability test and the activities of the fallout radionuclides (FRN) 7Be and 137Cs. For each radionuclide, the reference sites were identified and sampled using grid sampling approach and the study sites (the two plots) were sampled by applying a one-dimensional point transect sampling. Five samples were collected from each study site with 10m incre-ment. The results showed that the mean weighted diameter (MWD) was of 2.2 for the NT plot and 2.0 for the CT plot, this indicates a lower soil detachability under the no-till system. For the FRN results, the 7Be activity showed that the NT plot retained 79% of the reference site activity and the CT plot retained 54%. The 137Cs activity tests showed also that the NT plot retained more of reference site activity. The mass balance conversion models application to the FRN results showed that the no-till system generated 10% less soil erosion rate than the con-ventional tillage. The results showed that in spite of needing more than 4 years implementation for statistical significance, no-till helps reducing water erosion in the hilly agricultural lands of Northern Morocco.