Considering that the treatments were replicated at four different sites and the same treatment as applied to the same plot over the years, a mixed model was used to statistically analyze the effects of treatments and time (year), as well as their interactions, on the outputs (e.g., annual yield and NUE).