In arid and semi-arid regions, improving water use irrigation efficiency under limited seasonal water supply is critical for sustainable cotton production. While the effects of seasonal irrigation amount have been widely studied, the independent and interactive roles of irrigation interval under a fixed seasonal irrigation constraint remain insufficiently quantified. This study aimed to evaluate how irrigation amount and interval jointly regulate soil water dynamics, evapotranspiration partitioning, yield formation, and water use efficiency (WUE) in cotton. A two-year, controlled soil-column experiment was conducted using a full-factorial design with two seasonal irrigation amounts (350 and 200 mm) and four irrigation intervals (every 3, 6, 9, or 12 days). The AquaCrop model was locally calibrated with 2024 data and validated with independent 2025 observations. The validated model was then used to conduct scenario simulations across 16 irrigation combinations to analyze process-level responses. The model reproduced canopy cover and soil water storage (SWS) dynamics with good accuracy (R² > 0.89; NRMSE < 16%). The results showed that irrigation amount primarily controlled overall water availability, whereas irrigation interval reshaped the temporal fluctuation pattern of soil water content (SWC) in the shallow root zone. Under moderate irrigation levels, an intermediate interval (every 6 days) improved WUE by stabilizing SWC and maintaining high transpiration proportions. Under severe deficit conditions, prolonged intervals intensified periodic water stress and reduced yield. Simulated transpiration accounted for 95–97% of seasonal evapotranspiration in the controlled system, reflecting limited soil evaporation under column conditions. These findings highlight that irrigation interval, beyond total irrigation amount, is an important management variable for optimizing cotton irrigation scheduling under water-limited conditions. The combined experimental and modeling framework provides practical guidance for irrigation design in arid regions.

Accurate cotton boll identification and boll number estimation from UAV imagery are essential for large-scale yield prediction and precision management, yet severe leaf occlusion and complex canopy backgrounds often hinder robust performance. Here, UAV RGB images were acquired 3 days before defoliant application and at 3, 6, 9, 12, 15, and 18 days after defoliation. Cotton bolls were extracted using Mahalanobis distance, a support vector machine, and a neural network. Boll number was then estimated using an improved random forest model with multi-feature fusion. Across all defoliation stages, the NN produced the most accurate and stable boll extraction, achieving a maximum Kappa of 0.914, an overall accuracy of 95.77%, and an F1 score of 0.96. Extraction accuracy increased rapidly from 3 to 9 days after application and stabilized from 12 to 18 days. For boll number estimation, fusing the boll pixel ratio with color indices and texture features improved accuracy and consistency over time; the best performance was obtained at 18 days after application (R² = 0.7264; rRMSE = 4.9%). Overall, imagery acquired 15–18 days after defoliation provided the most reliable estimation window, supporting operational pre-harvest assessment and harvest-timing decisions.