Improving nitrogen use efficiency in crop plants contributes to increase the sustainability of agriculture. Crop models could be used as a tool to test the impact of climatic conditions on crop growth under several N management practices and to refine N application recommendation and strategy. STICS, a crop growth simulator developed by INRA (France), has the capability to assimilate leaf area index (LAI) from remote sensing to re-initialize input parameters, such as seeding date and seeding density. The coupling with remote sensing derived LAI was successfully tested on corn (Zea mays L.) over small regions for predicting biomass and yield. In this study, we tested the model performance for various nitrogen rate applications. Predictions of corn biomass and N plant uptake were tested against a 16-year experimental database of the Mixedwood Plains ecozone of Eastern Canada (extending from southern Ontario to western Quebec) for N ranging from 0 to 250 kg N ha-1. Data were collected during several growing seasons in the period 1993-2010. Model predictions for LAI, biomass, N in plants were in good agreement with measurements. STICS predicted N in plants with mean errors and root mean square errors below 20% when rainfall was close to the normal and for N application rates close to the recommendations. Results complied with the expected trend that under wet conditions or dry conditions (and low temperatures), yields remained steady whatever the N rate application, especially for higher N rate applications. Under non limited water conditions, increasing N application rates generate a non-linear increase of yield and grain N concentrations until a certain level (140-180N). Scenarios results showed that rainfall has a stronger effect on yield and biomass whereas nitrogen application impacted more the plant N, denitrification and potential N leaching. Between one half and 76% of the inter-annual variance of these variables can be explained by a multiple regression analysis according to climate and N rate applications. The remaining unexplained part of the variance could be explained by interactions occurring between these driving factors.