Using the AquaCrop model for maize in arid water deficient conditions: A new version for improved accuracy when adopting a varied normalized water productivity

Developed by FAO, AquaCrop is one of the most important generic crop growth and yield simulation models. The current research investigated the hypothesis that the AquaCrop model's normalized maize water productivity ( WP* ) remains constant or not under different water stress conditions and it was found that WP* decreases with increasing water stress. To calibrate and evaluate the model, diverse experimental field data sets on maize were used, which included various water stress levels and different irrigation methods (sprinkler, furrow and basin). The performance evaluation of results showed that the original AquaCrop-OS model is accurate in estimating biomass and grain yield for the treatments with mild or no water stress. Conversely, the model showed inaccuracy in estimating biomass and grain yield under severe water stress conditions. The normalized root mean square error ( NRMSE ) values for simulating both maize biomass and grain yield for the entire datasets and all treatments were 20.0 % and 19.5 %, respectively, when validating the original model. When, in the following, the model was modified using its open-source MATLAB version and rewritten with a purposefully modified WP* , i.e. the AquaCrop-WM version, led to NRMSE not exceeding 8.5 % and 11 % for biomass and grain yield, respectively. The new AquaCrop-WM is user-friendly and consists of an independent executable version accessible to all users. Thus, it is proposed its adoption when severe water stress conditions are to be assessed, namely to assess and select appropriate irrigation scheduling practices. • Maize water productivity ( WP* ) drops as water stress intensifies, confirming its high sensitivity to water deficit. • Using adjusted WP* improves maize biomass and yield simulations under severe water stress. • AquaCrop-WM is user-friendly and offered as an independent executable accessible to all users. • The proposed WP* adjustment shows high potential for use with other crops under water-limited conditions.