Effective resistance against pandemics: Mobility network sparsification for high-fidelity epidemic simulations
Fig 6
Map of U.S. Census Tracts by RUCA category.
(A) map of the U.S. where each census tract is color-coded according to its RUCA designation: 1, 2, 3 are metropolitan core, high, and low commuting, respectively, 4, 5, 6 are micropolitan core, high, and low commuting, respectively, 7, 8, 9 are small town core, high, and low commuting, respectively, and 10 is rural. On the right, we show the arrival time error for sparsifiers that sample s = 0.1m edges across the 10 RUCA categories for (B) the localized initial condition and (C) the dispersed initial condition. Across all RUCA categories, effective resistance performs better than uniform or weight-based sampling. This effect is especially pronounced for low-commuting areas, which have fewer or lower-weighted edges connecting them to the rest of the network. Base map provided by U.S. Census Bureau (https://www2.census.gov/geo/tiger/TIGER2016/TRACT/) and is covered under public domain.