Protocol for Defining the Agricultural Calendar and Yield Gap Using the AquaCrop Model: A Case Study for Sorghum Under High‐Salinity Water Irrigation in Current and Future Climate Conditions in Pernambuco, Brazil

In the face of climate change, agricultural modelling combined with experimental research is essential for predicting yields in diverse climatic scenarios. This study calibrated, validated and applied the AquaCrop model for forage sorghum under future climate scenarios in a semi‐arid region between 2020 and 2022. Meteorological data, soil moisture, canopy cover and crop productivity were collected. Model performance was evaluated using: Pearson's correlation coefficient, Willmott's index of agreement, confidence index, coefficient of determination, root mean square error, normalised RMSE, mean bias error and the Nash–Sutcliffe model efficiency coefficient. After calibration, simulations were performed for 11 semi‐arid locations to estimate yield gaps, optimal sowing periods and future projections (2030–2100). Future climate data were obtained from the CanESM2 model under the RCP 4.5 scenario and adjusted for bias correction based on historical observations (1961–2020). AquaCrop demonstrated strong predictive accuracy for productivity and canopy cover. Under the RCP 4.5 scenario, increased rainfall during the growing season resulted in an approximately 35% increase in achievable productivity. Yield gaps were more affected by management practices than by water deficit, averaging 42%. Overall, AquaCrop has proven to be an effective tool for defining agricultural calendars and supporting water management strategies in diverse environmental conditions.