On the Effects of Lag-Times in Networks Constructed from Similarities of Monthly Fluctuations of Climate Fields
The complex network framework has been successfully applied to the analysis of climatological data, providing, for example, a better understanding of the mechanisms underlying reduced predictability during El Niño or La Niña years. Despite the large interest that climate networks have attracted, several issues remain to be investigated. Here we focus in the influence of the periodic solar forcing in climate networks constructed via similarities of monthly averaged surface air temperature (SAT) anomalies. We shift the time series in each pair of nodes such as to superpose their seasonal cycles. In this way, when two nodes are located in different hemispheres we are able to quantify the similarity of SAT anomalies during the winters and during the summers. We find that data time-shifting does not significantly modify the network area weighted connectivity (AWC), which is the fraction of the total area of the Earth to which each node is con- nected. This unexpected network property can be understood in terms of how data time-shifting modifies the strength of the links connecting geographical regions in different hemispheres, and how these modifications are washed out by averaging the AWC.