Limited availability of irrigation water, coupled with irregular rainfall, is driving the search for better water-saving strategies. One strategy is to select more efficient crop varieties that can maintain profitable yields without requiring large amounts of water. In this study, we developed a new approach to determine the soil depletion factor (p) by considering maximum root zone water extraction. The model of Rijtema and Aboukled was modified to account for different initial soil water contents. Three onion varieties with distinct bulb characteristics, ‘Hornet’ (yellow), ‘Mata Hari’ (red) and ‘Amadea’ (white), were evaluated in a field experiment with drip irrigation, in which the plants were subjected to five stress levels measured at 0.20 m: –25, –50, –75, –100 and –125 kPa. The results showed that the estimated p for the three varieties was higher than the reference values proposed by the FAO (p of 0.30 – 0.35 for a crop evapotranspiration of 5.0 mm d-1). Amadea exhibited the highest p values, with an average of 0.58. Mata Hari presented intermediate values and Hornet the lowest, although their average p values (0.48 and 0.39, respectively) were statistically similar. Additionally, Amadea extracted more water than the two other varieties across all analyzed periods, particularly during the final crop stage. It also appeared to be more drought tolerant. Further study focusing on estimating p across different phenological stages is necessary. We conclude that the proposed approach offers an alternative method for estimating the depletion factor and it can be extended to other crops.

Mustard grown as a cover crop retains root zone soil water content (SWC); however, if terminated late, it may lead to excessive water uptake. To test this, we analyzed seven global climate models (GCMs) together with the Decision Support System for Agrotechnology Transfer (DSSAT) to predict the effect of the projected climates on mustard and maize SWC, leaf area index (LAI), aboveground biomass (AGB), and yield in Southwest Texas. GCM drove these predictions in CROPGRO. Our study 1) developed a methodology to predict temporal variations in SWC, 2) calibrated the model and validated it against measured data, and 3) examined the quantity of inputs required for SWC predictions. Results show that mustard retained SWC (0.29-0.38 m3 m-3) only in future scenarios with moderate to high rainfalls. SWC (0.13-0.19) significantly decreased in scenarios where rainfall declined to lower levels. In 2019, simulated SWC for mustard was initially lower than observed but later aligned with it at midseason, indicating effective parameterization of infiltration and evapotranspiration (RMSE = 0.04). In 2020, simulated values of soil moisture closely matched observed ones at the 10 cm depth, although slight underestimations revealed the model’s sensitivity to variations in soil water in root zone. Model simulated LAI well for maize (RMSE = 0.1) and mustard (MAE = 0.1), and AGB for maize (RMSE = 1416.7 kg ha-1) and mustard (MAE = 149.2 kg ha-1; RMSE = 159.2). Under wet conditions, mustard retained soil moisture, supported maize growth, and boosted yield; under dry conditions, its effect was neutral.