Population growth rate
We applied the non-parametric skygrowth method (Volz & Didelot 2018) to examine if the FIVPco population growth rate fluctuated across time and if this was related to changes in male or female population size in the treatment region. We did not do the same for the stable region as similar estimates were not available. We fitted these models using MCMC (100,000 iterations) assuming that FIVPco population size fluctuated every 6 months over a 14-year period (the estimated time to most recent common ancestor of this clade, Fig. S7). Otherwise, the default settings were used. We then performed a Pearson correlation test to assess if the trend in FIVPco population growth was related to male and female population size estimates (Logan & Runge 2020a). Measuring the correlation between population size estimates and patterns of population growth using generalized linear models (Gill et al. 2016; Volz & Didelot 2018) was not feasible due to the relatively small size of this dataset.