Figure 1.

Fluctuation power spectra of infective time series for the SIR model on dynamic networks of $N={10}^6$ nodes for small-world parameter p equal to (a) 1, (b) 0.4 and (c) 0.2 and β_eff equal to (a) 1.32, (b) 1.01 and (c) 0.79; the networks are built over regular 1000×1000 lattices with 12 first neighbours and periodic boundary conditions. The solid lines are the averaged numerical power spectra of 400 stationary time series, each 32 768 time steps long. The dashed lines are the analytic power spectra given by (2.2), where for p smaller than 1 the value of the parameter β has been taken as β_eff, defined as the actual number of new infections per time step and infective individual divided by the density of susceptibles. Parameter values: μ=6×10⁻⁴, β=1.32, τ=1/8.