Abstract: 【Objective】 To investigate the effects of key climatic factors on Glycine max yield in different climatic regions of Shanxi Province,providing a theoretical basis for ensuring G. max production and optimizing G. max planting regionalization. 【Methods】 Meteorological data—including mean temperature,sunshine duration,daily precipitation,and mean relative humidity—during different growth stages of G. max from May to September at 28 meteorological stations in Shanxi Province from 2005 to 2022,along with G. max yield per unit area,were collected. Cluster analysis was used to classify the climatic regions of Shanxi. Correlation analysis,principal component analysis,and multiple linear regression analysis were applied to clarify the effects of climatic factors on G. max yield per unit area in different climatic regions. 【Results】 Shanxi Province was classified into three climatic regions （A,B,and C） by cluster analysis,containing 13,14,and 1 meteorological stations,respectively. Correlation analysis showed that all growth stages in regions A and B,and mean relative humidity from the flowering stage to the pod-setting stage in region C,as well as mean temperature from the sowing stage to the flowering stage in region A,were significantly correlated with G. max yield per unit area （P<0.05）. In region C,daily precipitation from the sowing stage to the flowering stage and from the flowering stage to the pod-setting stage was positively correlated with G. max yield per unit area,with relatively high correlation coefficients,while mean temperature from the flowering stage to the pod-setting stage was negatively correlated with G. max yield per unit area. Principal component analysis indicated that the cumulative contribution rates of the first five principal components were 78.353%,80.592%,and 83.431% in regions A,B,and C,respectively. The dominant climatic factors in region A were mean temperature and mean relative humidity during all growth stages,among which mean temperature had a positive effect;in region B,dominant climatic factors were mean temperature during all growth stages and mean relative humidity from the flowering stage to the pod-setting stage and from the pod-setting stage to the maturity stage,among which mean temperature had a negative effect;in region C,dominant climatic factors were daily precipitation,sunshine duration,relative humidity,and mean temperature. Multiple linear regression analysis showed that the explanatory power of the model for region C（R²=0.786） was significantly higher than that for region A（R²=0.281） and region B（R²=0.309）,and the effects of different climatic regions on G. max yield followed the order region C>region B>region A. 【Conclusion】 Among the three climatic regions of Shanxi Province,in region A,climatic factors had the least effect on G. max yield per unit area,with the main influencing factors being mean temperature from the sowing stage to the flowering stage and mean relative humidity from the flowering stage to the pod-setting stage;in region B,climatic factors had a moderate effect on G. max yield per unit area,with the main influencing factors being mean relative humidity from the sowing stage to the flowering stage and from the pod-setting stage to the maturity stage,and sunshine duration from the flowering stage to the pod-setting stage;in region C,climatic factors had the greatest effect on G. max yield per unit area,with the main influencing factors being mean relative humidity from the flowering stage to the pod-setting stage and daily precipitation from the sowing stage to the flowering stage.

Key words: Shanxi Province, Climatic region, Climatic factor, Glycine max, Yield, Principal component analysis, Multiple linear regression analysis