Climate change is reshaping yield formation and water use in cold-region rice production through rising air temperatures, altered precipitation patterns, and increasing atmospheric CO₂ concentrations. However, the responses of yield, crop evapotranspiration (ET_c), and water use efficiency (WUE) to climate forcing and elevated CO₂ remain insufficiently quantified for cold-region rice systems in Northeast China. This study simulated changes in rice yield, ET_c and WUE during the 2030s–2090s relative to the 2000–2020 baseline period under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios at 10 agro-meteorological stations in Heilongjiang Province. Simulations were conducted using the AquaCrop model driven by CMIP6 multi-model climate data, and the contribution of elevated CO₂ was quantified by comparing the rising-CO₂ and fixed-CO₂ treatments. The results showed that under SSP5-8.5, the maximum air temperature in the 2090s is projected to increase by 5~6 °C relative to the baseline period, while precipitation is projected to range from −10% to 20%. Compared with the fixed-CO₂ treatment, rice yield under the rising-CO₂ treatment is projected to increase by 18.70%. Although ET_c showed an overall increasing trend, rising CO₂ attenuated its increase. Under SSP5-8.5 in the 2090s, ET_c increased by only 2.70% under rising-CO₂ treatment, compared with 11.61% under fixed-CO₂ treatment. As a result of increased yield and ET_c, the WUE improved by 15.42% and 14.28% under SSP2-4.5 and SSP5-8.5, respectively, in the 2090s, whereas it remained below the baseline level under the scenarios without CO₂ effects. These findings indicate that rising CO₂ may enhance yield and moderate ET_c increases, thereby providing useful information for regional grain-yield assessment, agricultural water-resource evaluation, and climate-change adaptation planning.

Accurate precipitation prediction is critical for water security and disaster mitigation, yet remains challenging due to atmospheric complexity and class imbalance in rainfall data. This study introduces an integrated "architecture-feature-augmentation" framework to address these limitations. Through systematic comparison of CNN-LSTM and Trans-former architectures, we identify a fundamental trade-off: CNN-LSTM demonstrates higher enhanceability, achieving 80% recall for heavy rainfall when combined with phys-ics-informed augmentation, while Transformer shows superior inherent sensitivity (75% recall) but greater vulnerability to data distribution shifts. Feature engineering benefits are model-specific, significantly improving CNN-LSTM but often introducing redundancy for Transformer. Notably, oversampling techniques like SMOTE achieve peak F1 scores but with substantial generalization gap (ΔF1 > 0.47), indicating overfitting risks, whereas physics-informed augmentation proves more reliable. We establish a principled decision framework: for robust predictions, use CNN-LSTM with physics-informed augmentation; for peak performance where risks are tolerable, employ CNN-LSTM with SMOTE. These findings provide scientific guidance for extreme weather preparedness and water resource management.

Accurate precipitation forecasting is paramount for water security and disaster mitigation, yet it remains formidable due to atmospheric stochasticity and the inherent class imbalance in rainfall datasets. This study proposes an integrated “architecture-feature-augmentation” framework to circumvent these limitations. Through a systematic evaluation of CNN-LSTM and Transformer architectures, we delineate distinct performance profiles: The Transformer model, when coupled with feature engineering and physics-informed augmentation, yields a peak F1-score of 0.1429, marking the optimal configuration for harmonizing precision and recall. Conversely, CNN-LSTM demonstrates superior robustness in extreme event detection, consistently maintaining high recall rates (up to 0.90) across diverse scenarios. We identify feature engineering as a critical performance modulator, substantially bolstering CNN-LSTM’s baseline metrics while enabling the Transformer to realize its maximum predictive capacity. Although synthetic oversampling techniques—such as SMOTE and GAN—effectively extend the detection range for heavy precipitation, physics-informed augmentation provides the most consistent performance gains, particularly in multi-class contexts. We conclude that the Transformer, augmented by physical constraints, is the optimal candidate for high-precision requirements, whereas CNN-LSTM, integrated with synthetic augmentation, offers a more sensitive alternative for early warning systems prioritizing recall. These findings provide empirical guidance for advancing extreme weather preparedness and strategic water resource management.