Spatial correlation of wheat yield and yield pattern changes under two weather scenarios had been analyzed using Moran’s I and LISA. The results indicated there were most significant spatial autocorrelations among different yield patterns. Statistically, wheat yields were highly spatially correlated in cold years. With global and local spatial statistics, LISA map of yield differences in two weather patterns, stable and unstable yield areas were identified. Most significant spatial autocorrelations had been found in the northeast, west and south region. The northeast and west region of the field tended to be unstable, while the south region tended to be stable between different weather patterns. 