**Abstract**

Climate change is impacting environmental systems including agriculture. In Morocco, declining precipitation and increasing temperatures are negatively impacting crop yields. Consequently, crop yields in Morocco are now dependent on nutrient and water management. Most studies have focused on experimentation through fertilizer application and irrigation without any attention to the intrinsic linear relationships that exist between crop yields, fertilizers, and agricultural water withdrawal. The time series agricultural water withdrawal data were collected from AQUASTAT for the period 1990-2022 while data on nitrogen, phosphorous, and potash fertilizers were collected from FAOSTAT. Yield data for maize, barley, sorghum, and wheat were also collected from FAOSTAT. The data were analyzed using two machine learning models fitted through multiple linear regression. The key results show that for the three fertilizers, phosphates tend to have the strongest impacts and cause changes in crop yield as seen in the context of wheat. When both fertilizers and agricultural water withdrawal are fitted against yield, agricultural water withdrawals tend to have a strong relationship with yields. This work has helped us to identify which crops and management options need to be valorized in terms of increased access to nutrients and water.

**Keywords:** agriculture, water withdrawal, water management nutrient management, trends, multiple linear regression
