Bunching of extreme events on complex network.

Area of Science:

• Complex systems
• Network science
• Statistical physics

Background:

• Extreme events like earthquakes and floods often occur in bursts.
• The underlying mechanisms driving these burst phenomena are not fully understood.
• Network structure's role in event correlations remains an active area of research.

Purpose of the Study:

• To investigate the influence of network topology on the bunching of extreme events.
• To model and understand how network structure can induce correlations among independent events.
• To identify network characteristics that promote or suppress extreme event clustering.

Main Methods:

• Utilizing a computational model of independent random walkers on complex networks.
• Analyzing network structures with varying cluster sizes and inter-cluster connectivity.
• Employing statistical techniques including recurrence time distribution, autocorrelation function, and burstiness parameter.

Main Results:

• Sparsely connected network clusters exhibit oscillatory behavior and event bunching.
• A small cluster sparsely connected to a larger network shows emergent correlations and bunching.
• These correlations and bunching phenomena are driven by network structure, not walker interdependence.
• No significant bunching or correlations were observed in the large cluster.

Conclusions:

• Network topology is a critical determinant of extreme event bunching.
• Sparse connectivity in specific network modules can naturally induce correlations and burstiness.
• Understanding network structure is key to predicting and potentially mitigating clustered extreme events.