Impact of global warming assessed from evolution of temperature relevance networks

This study assesses the impact of global warming in evolution process of temperature relevance networks. Temperature relevance networks are constructed from temperature curves with permutation tests to select statistically significant edges. After the networks were constructed, we employ sub-graph ration profile to study structure of the networks, come to the conclusion that mean temperature and relative sub-graph ratio for four-node complete graph changes in opposite directions and temperature relevance network will lose some features and become similar to corresponding randomized graph if the temperature keeps rising. Community structures are found in the networks, with motifs of the networks, we find that highly connected sub-graphs are more likely to appear in temperature relevance network than in randomized networks.