City-specific estimates of PM2.5-attributable premature deaths per capita in 2016 versus other city indicators. (a) Population-weighted annual average PM2.5 concentration (µg/m3) vs. annual CO2 emissions rate (t C per 100,000 people); (b) PM2.5 death rate (deaths per 100,000 people) vs. annual CO2 emissions rate; (c) comparison of population-weighted PM2.5, PM2.5 death rate, CO2 emissions rate, and 2013 carbon footprint rate (kt CO2 per 100,000 people) vs. GDP per capita ($) in 2015 in 82 cities. Colors indicate world regions (see Fig. 1 legend). Linear regression lines are shown where correlations are significant, r is the correlation coefficient, and p is the correlation significance level. (Note: Riyadh was removed from panels a and b to show more detail in the rest of the dataset. Its CO2 emission rate is likely unrealistically high due to very low population estimate in the GPWv4 dataset: CO2 emission rate = 290,000 kt CO2 per 100,000 people, PM2.5 pop-wt = 280 µg/m3, and PM2.5 death rate = 40.) Similar graphs for each region (using “super-regions” from the Global Burden of Disease 2016 Study) and the 50 most populous cities globally are in the Supplemental Information (Figs S5–S12).