Big city, small world: Density, contact rates, and transmission of dengue across Pakistan.

Moritz U. G. Kraemer; T. Alex Perkins; Derek A.T. Cummings; Rubeena Zakar; Simon I. Hay; David L. Smith; Robert C. Reiner

Finally, we performed mathematical analyses of the fitted model to assess the significance of 114 spatial variation in mixing inhomogeneity for how time series data are interpreted and used to 115 Environmental conditions are instrumental in defining the risk of transmission of dengue (Bhatt 131 et al. 2013) . Transmission is limited by the availability of a competent disease vector. Due to a 132 lack of resources and political instability no comprehensive nation wide entomological surveys 133 have been performed in Pakistan. Therefore we use a probabilistic model to infer the 134 probability of occurrence of Ae. aegypti and Ae. albopictus in Pakistan derived from a globally 135 comprehensive dataset containing more than 20,000 records for each species (Figure 1a and 136 b). Such model outputs have proven useful in identifying areas of risk of transmission of 137 dengue as well as malaria (Gething et al. 2011; Sinka et al. 2012; Bhatt et al. 2013) . Other 138 important environmental conditions defining the risk of transmission of dengue are 139 temperature, water availability, and vegetation cover (Messina et al. 2015) . To account for such 140 variation, raster layers of daytime land surface temperature (LST) were processed from the 141 MOD11A2 satellite, gap-filled to remove missing values, and then averaged to a monthly 142 temporal resolution for all four years (Weiss et al. 2015) . The density of vegetation coverage 143 has been shown to be associated with vector abundance (Eisen & Lozano-Fuentes 2009) . 144