Model input values from the low transmission scenario described by Shekhalage (2007) and Mwerinde (2005), and the high transmission scenario described by Paganotti (2004 and 2006) [24]–[27].

Ranges were assigned to all input values. Random samples (n = 1000) were generated from within these ranges and used for model calibration.*for parasite and gametocyte rates Shekalaghe (2007) differentiates between wet and dry season, we used the mean values;**Mwerinde et al. report a mean EIR of 3.4/a for the Moshi region. Since this is a very low value, based on a very limited number of observations (sporozoite rate of 3/5634 sampled mosquitoes), we assume a wide (log normally distributed) possible EIR range around this mean;***Paganotti et. al. investigated two ethnic groups (Mossi and Fulani). We used the data reported for the Fulani group.