A transition from river networks to scale-free networks

TL;DR

This paper investigates how spatially embedded networks transition from river-like structures to scale-free networks by tuning a parameter that influences attachment probability based on degree and distance.

Contribution

It introduces a tunable model that interpolates between river networks and scale-free networks, identifying a critical transition point at alpha = -2.

Findings

Transition point at alpha = -2 identified

Model unifies river networks and scale-free networks

Critical behavior changes at the transition

Abstract

A spatial network is constructed on a two dimensional space where the nodes are geometrical points located at randomly distributed positions which are labeled sequentially in increasing order of one of their co-ordinates. Starting with $N$ such points the network is grown by including them one by one according to the serial number into the growing network. The $t$-th point is attached to the $i$-th node of the network using the probability: $\pi_i(t) \sim k_i(t)\ell_{ti}^{\alpha}$ where $k_i(t)$ is the degree of the $i$-th node and $\ell_{ti}$ is the Euclidean distance between the points $t$ and $i$. Here $\alpha$ is a continuously tunable parameter and while for $\alpha=0$ one gets the simple Barab\'asi-Albert network, the case for $\alpha \to -\infty$ corresponds to the spatially continuous version of the well known Scheidegger's river network problem. The modulating parameter $\alpha$ is tuned to study the transition between the two different critical behaviors at a specific value $\alpha_c$ which we numerically estimate to be -2.