To compare the spatial-temporal multicomponent model with different spatial weight settings in different terrains, and to explore the optimal weight settings in different terrains. Then to study the distribution and influencing factors of influenza incidence in different terrains of Sichuan Province, so as to provide scientific basis for spatial weight settings and identification of risk influencing factors as well as the formulation of prevention and control measures.

Based on the influenza incidence data in Sichuan Province from 2016 to 2018, we conducted a descriptive analysis of the spatial and temporal influenza situation in different terrain areas（plains, hills and plateau mountainous regions）, explored the optimal weight settings for different terrains using a spatial-temporal multicomponent model, and further analyzed the spatial and temporal influenza propagation dynamics through the optimal model to quantify the effects of meteorological and socio-economic factors.

The optimal weights were inverse distance weights for plains and Power-law weights for hills and plateau and mountainous regions. The results of the factor analysis of the optimal model showed that the average temperature in the plains（RR=0.970, 95% CI: 0.958-0.982, P<0.001）, hills（RR=0.960, 95% CI: 0.942-0.978, P<0.001）and plateau and mountainous regions（RR=0.913, 95% CI: 0.868-0.960, P<0.001）were all inhibitory to the risk of influenza. Air pressure was protective in the plains（RR=0.990, 95% CI: 0.979-1.000, P=0.049）, while showing elevated risk in the hills（RR=1.044, 95% CI: 1.016-1.072, P=0.002）and plateau and mountainous regions（RR=1.005, 95% CI: 1.002-1.009, P=0.004）.

The spatial weights of plains, hills, plateau and mountainous regions should be carefully considered when constructing the model. Inverse distance weights are more suitable for plains, while Power-law weights are more suitable for hills, plateau and mountainous regions. Meanwhile, there are differences in the influence of influenza morbidity risk in different topographic areas, and it is suggested that different early warning and control measures should be developed according to the key risk factors in different areas and key risk characteristics of the areas.