Objective To explore the effects of different irrigation limits on physiological indicators and yield of winter wheat in northern Xinjiang, and to optimize its irrigation scheduling.

Methods Taking the field experiment data of dryland winter wheat in Changji City, Xinjiang from 2023 to 2025 as a reference, an improved APSIM model based on Bayesian Monte Carlo Markov chain method was used to simulate the changes in winter wheat growth indicators and yield under six different irrigation scheduling. Twelve irrigation scenarios were constructed to select the appropriate irrigation scheduling for local winter wheat.

Results The growth indicators and yield of wheat showed a trend of first increasing and then decreasing with the increase of irrigation quota, among which W3 treatment had better growth indicators and yield compared to other treatments. The W6 treatment, which is relatively poor, increased plant height by 18.11% to 19.29%, LAI by 18.73% to 30.22%, and yield by 38.77% to 44.82%; Comparing the simulation results of the APSIM model before and after improvement, the coefficient of determination (R²) before improvement was 0.768 and 0.666, the consistency index (D) was 0.885 and 0.918, the relative root mean square error (NRMSE) was 0.09 and 0.10, and the root mean square error (RMSE) was 821.07 and 889.55 kg/hm². After improvement, the R² was 0.955 and 0.974, D was 0.961 and 0.995, NRMSE was 0.05 and 0.04, and RMSE was 454.83 and 319.45 kg/hm². The simulation results of the improved model were good and can be applied to winter wheat simulation research in this region; Analyzing the 12 irrigation scheduling simulated by the APSIM model, the results showed that the P6 treatment (irrigation quota of 440.14 mm) had better crop yield.

Conclusion The optimal irrigation scheduling for winter wheat in Changji City, Xinjiang is as follows: 70 mm irrigation during the emergence to overwintering period, 32.76 mm irrigation during the greening period, 65.52 mm irrigation during the jointing period, 102.32 mm irrigation during the heading period, 74.24 mm irrigation during the flowering period, 46.16 mm irrigation during the grain filling period, and 49.14 mm irrigation during the maturity period. The crop yield is 10707.42 and 10404.76 kg/hm², respectively. This study provides a theoretical basis for optimizing the irrigation scheduling of winter wheat in Northern Xinjiang using the APSIM model.