Dependence of the TER on the value of <i>M</i> used, for the stochastic SIR model with seasonal transmission.

A. The TER (obtained by solving system of equations <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1012364#pcbi.1012364.e132" target="_blank">(11)</a> numerically) when sustained transmission is possible throughout the year (<i>β</i>₀ = 10, and <i>γ</i> = 4 . 9 month^-1), for different values of <i>M</i> (100, 200, 300, 400 and 500). For <i>M</i> = 500, the TER as approximated using model simulations is also plotted (blue dots). B. Analogous results to panel A, but in a scenario in which sustained transmission can only occur for some of the year (<i>β</i>₀ = 4, <i>β</i>₁ = 5 and <i>γ</i> = 4 . 9 month^-1). In both panels, the overall population size was assumed to be <i>N</i> = 1 , 000 individuals. When we computed the TER numerically, we used the time step <i>Δt</i> = 0.00033 months. When we approximated the TER using model simulations, 10 , 000 simulations were run for each time of introduction considered. (PDF)