Table 2.

Model scenarios and parameters values for rabies transmission processes, characteristics of the environment and dog population, surveillance and response. Reference scenario parameters and model set up are in bold.

	Parameter	Value	Source/Rationale
Transmission	Shape and scale of gamma distribution modelling generation interval	Shape 1.46 days; scale 16.1 days	Hampson et al. [23]
	Mean and dispersion parameter of negative binomial distribution modelling R0	Mean 1.20; k1.33	Hampson et al. [23]; Townsend et al. [31]
Environment and population	Area	500, 5000, 15,000 km2	Ambon Maluku, Indonesia ∼775 km2; Bali, Indonesia ∼5600 km2; Bohol, Philippines ∼4100 km2; Nias, Indonesia ∼5100 km2; 3 districts in Eastern Bhutan ∼7000km2; Flores, Indonesia ∼14,300 km2
	Geometry	Circular, interdigitated	To compare a minimum edge effect versus a large edge effect
	Human-mediated long distance dog movement	0, 2, 5%	Estimated for Bali, Indonesia [31]
	Local movement spatial kernel: mean and dispersion parameter (k) of negative binomial distribution	Mean 0.88 km2; k0.285	Hampson et al. [23]
	Annual dog population turnover	50%	We assume that 50% of dogs vaccinated die one year later, and that the birth and death rates are equivalent
Detection and response	Probability of detection	0.01–0.3, 0.1	See methods: Detection probabilities
	Response mobilization time	1, 6, 12 months	Table 1
	Lag between detected case and laboratory confirmation	0, 14 days	Table 1
	Time period of cases used to determine reactive vaccination	1, 6 months
	Duration of immunity provided by vaccine	2 years	Most commercial vaccines provide 1–3 years of protection
	Vaccination coverage achieved at time and place of vaccination (V)	V = 70%, V∼uniform (35%, 70%),V for 80% of island ∼uniform(35%, 70%) and V for 20% of island ∼uniform(X/2,X) where X∼uniform(0,70%)	70% target vaccination coverage is recommended by WHO and empirically and theoretically supported [40]. Due to the difficulty of estimating coverage and dog population sizes, coverage achieved is expected to be spatially variable and may often fall below the 70% target.
	Vaccination strategy	Proactive, prioritise, react w/o repeat, reactive (see Table S1)	Builds on strategies explored in Townsend et al. [31]
	Cumulative number of cases when start vaccination	500, 5000	5000 is the approx. cumulative number of cases on Bali when mass vaccination started, assuming 10% of cases were confirmed
	Length of monitoring period	2 years	OIE and WHO criteria for rabies-free status requires 2 years without indigenously acquired infection [8,21]
	Months without any detected cases before starting 2-year monitoring period	2, 6 months
	Intervention during monitoring period	No vaccination, proactive vaccination