Abstract: To solve the destructive impact of climate change in Yunan Province on human production, life, and the ecological environment and to understand the main influencing factors of climate change in this area, this study uses the complex network theory to abstract meteorological stations into network nodes based on the climate data of 32 meteorological stations in Yunnan Province from
1984 to 2016. The Pearson correlation coefficient is used to determine the connection weight of edges. A single-layer and hierarchical
climate complex network model in Yunnan Province are constructed, and the network characteristic parameters and stability are
analyzed. The research finds that in single-layer networks, the correlation between temperature, pressure, relative humidity, and wind
speed networks are strong, while the correlation between precipitation and sunlight networks is weak. According to the results of
principal component analysis and single-layer network experiments, temperature, pressure, and relative humidity are the main
influencing factors of climate change in Yunnan Province. In the hierarchical network, temperature has a strong negative correlation
with pressure, relative humidity, and precipitation, temperature mainly shows a strong positive correlation with wind speed and
sunlight. There is an inherent correlation between various meteorological elements. Deletion of the three meteorological factors of
temperature, relative humidity and pressure would significantly damage the stability of the entire network. Therefore, it is recommended to protect the climate and environment of Yunnan Province from these three main influencing factors.

Key words:  complex network, climate network, hierarchical network, network modeling, stability