Climate change exerts significant impacts on regional agricultural production. This study assesses the implications of climate change on winter wheat yields in the Huang-Huai-Hai Plain (3H Plain), utilizing bias-corrected climate projections from the Coupled Model Intercomparison Project Phase 6 (CMIP6) for mid-21st century (2041–2060) and late 21st century (2081–2100) periods under two shared socioeconomic pathways (SSP2–4.5 and SSP5–8.5). These projections were incorporated into the decision support system for agrotechnology transfer (DSSAT) CERES-Wheat model to forecast potential alterations in winter wheat production. Initial findings reveal that uncorrected CMIP6 projections underestimated temperature and precipitation while overestimating solar radiation across the southern 3H Plain. Following bias correction through the equidistant cumulative distribution function (EDCDF) method, the regional average biases for temperature, precipitation, and solar radiation were reduced by 18.3%, 5.6%, and 30.7%, respectively. Under the SSP2–4.5 and SSP5–8.5 scenarios, mid-21st century simulations predicted a 13% increase in winter wheat yields. Late 21st century projections indicated yield increases of 11.3% and 3.6% under SSP2-4.5 and SSP5-8.5 scenarios, respectively, with a notable 8.4% decrease in yields south of 36° N under the SSP5-8.5 scenario. The analysis of climate change factors and winter wheat yields in the 3H Plain under both scenarios identified precipitation as the key contributing factor to yield increases in the northern 3H Plain, while temperature limitations were the primary constraint on yields in the southern region. Consequently, adaptive strategies are essential to mitigate climate change impacts, with a particular focus on addressing the challenges posed by elevated temperature in the southern 3H Plain.

Understanding future changes in crop phenology and climate suitability is essential for sustaining winter wheat production in the Huang-Huai-Hai (3H) Plain under climate change. This study integrates bias-corrected CMIP6 climate projections, the DSSAT CERES-Wheat crop model, and Random Forest analysis to assess spatiotemporal shifts in winter wheat phenology and climate suitability. The assessment focuses on the mid- (2041–2060) and late 21st century (2081–2100) under the SSP2-4.5 and SSP5-8.5 scenarios. The results indicate that the vegetative and whole growing periods (VGP and WGP) will be extended in the mid-century but shorten by the late century. In contrast, the reproductive growing period (RGP) will be slightly reduced in the mid-century and extended under high emissions in the late century. Temperature suitability is projected to increase during the VGP and WGP but decline during the RGP. Precipitation suitability generally improves, except for a decrease during the reproductive period south of 32° N. Solar radiation suitability is expected to decline across all stages. Temperature is identified as the primary driver of phenological changes, with solar radiation and precipitation playing increasingly important roles in the mid- and late 21st century, respectively. Adaptive strategies, including the adoption of heat-tolerant varieties, longer reproductive periods, and earlier sowing, are recommended to enhance yield stability under future climate conditions.

Hourly precipitation is a standard variable for hydrological applications, flood warning, and quantitative assessment of precipitation-related hazards. Weather radar provides high-frequency observations, but these sub-hourly measurements must be linked to hourly rain-gauge accumulations for radar-based quantitative precipitation estimation (QPE). This study develops a machine-learning-based hourly radar QPE framework using intra-hour temporal features from dual-polarization observations. Each hourly gauge accumulation was matched with ten consecutive 6 min radar observations. Six temporal features were derived from dual-polarization observations, including the mean, sum, standard deviation, linear trend, maximum, and skewness. The results show clear rainfall regime dependence in the contribution of polarimetric variables. Evaluation of machine-learning models confirms the value of intra-hour temporal features, with the best-performing model achieving an RMSE of 3.96 mm h⁻¹, compared with 6.42 mm h⁻¹ for the linear baseline. The experiments further show that the complete temporal feature set improves QPE performance. Diagnostic evaluation also indicates that model bias is sensitive to the representation of heavy rainfall samples. These findings suggest that ML-based radar QPE should emphasize physically meaningful intra-hour polarimetric features and rainfall-imbalance-aware sampling.