Based on five CMIP6 climate models and using the SSP126 and SSP245 scenarios, this study employed the high−resolution meteorological grid data for China (CN05.1) from 1970 to 2014 (historical period) as the baseline dataset. Delta bias correction, Taylor diagram and Bayesian model averaging (BMA) were used to assess the spatiotemporal changes in agricultural precipitation and heat resources in China from 2015 to 2050. The aim was to provide scientific evidence for optimizing planting systems, adjusting agricultural layout and adapting to climate change. The results showed that: (1) the five CMIP6 climate models and BMA ensemble demonstrate good performance in simulating temperature and precipitation, effectively capturing regional climate characteristics, with better accuracy in temperature simulations. BMA could effectively balance the performance of multiple models in simulating temperature and precipitation. (2) Under the SSP126 and SSP245 scenarios, the average temperature increase rate from 1970 to 2050 was 0.37°C·10y−1 and 0.40°C·10y−1, respectively. With the most significant temperature increasing in the Tibetan plateau, northwest, north and northeast regions, generally exceeding 0.4°C·10y−1. Annual precipitation showed a slight increasing trend, with rates of 5.6mm·10y−1 and 4.8mm·10y−1. Significant increases (≥10mm·10y−1) were observed in south, east and northeast regions, while the southwest region showed a varying degrees of decrease. (3) Compared with the historical period (1970−2014), the regions with an average temperature ≤0°C during 2015−2050 showed significant warming, with the area gradually shrinking. The isotherms for 5°C, 10°C, 15°C and 20°C moved northward by 2.1°, 2.9°, 4.2° and 2.2°, respectively. The region in south China with annual precipitation ≥1500mm slightly expands. (4) Under the SSP245 scenario during From 1970 to 2050, the trend rates of effective accumulated temperature for thresholds ≥0°C, ≥5°C, ≥10°C, and ≥15°C were 10.4°C·d·y−1, 9.3°C·d·y−1, 7.6°C·d·y−1 and 5.9°C·d·y−1, respectively, showing a pattern in which higher temperature thresholds correspond to smaller increases in effective accumulated temperatures. (5) From the distribution of effective accumulated temperature at different thresholds, the low value areas in high−altitude and mid−latitude regions shrinked in the future, while the high value areas in south China expanded to varying degrees. From 1970 to 2050, the area with an increase in effective accumulated temperature ≥10℃·d·y−1 for thresholds of ≥0℃, ≥5℃, ≥10℃ and ≥15℃ gradually decreased as the temperature threshold rises, while the area with an increase in the range of 0−10℃·d·y−1 continued to expand. Climate change has led to an extended growing season, a northward expansion of planting boundaries and an increase in the cropping index. However, it also presented new challenges for the growth of cool−season crops such as winter wheat, pest control and disease prevention. Strengthening agricultural adaptation strategies to cope with the uncertainties of future climate change is crucial.
