Urban forest biodiversity and cardiovascular disease: Potential health benefits from California’s street trees

Though the plots in Fig 3 have a linear shape, the density of observations of the Shannon Index is weighted toward higher values. In columns two and five, we estimate a non-linear specification where the omitted category is the lowest quantile of the Shannon index (Shannon I), zip codes with Index values below 2.495. The interpretation of the coefficient for each quan- tile is the difference in the mortality rate for zip codes within the selected quantile as compared to those in the omitted quantile. These coefficients are precisely estimated for heart disease, with a potential threshold effect for the zip codes above the median (Shannon III and IV). We see a similar potential threshold effect for stroke, however the coefficients are estimated more precisely as the Shannon index quantile increases. Results hold when we interact biodiversity and trees per 100,000 individuals (columns three and six), suggesting biodiversity has an inde- pendent effect on mortality. The coefficient on the number of trees per hundred thousand peo- ple increases in magnitude but remains imprecise. Sensitivity analysis In this section we present a sensitivity analysis of our linear estimate. In Table 3 we repeat our main estimate from the linear model in column one and consider seven alternative specifica- tions to assess whether our estimates are sensitive to empirical design choices, as well as two additional tables in Supporting Information. Fig 3. Urban forest diversity and mortality. Notes: Data for 857 California zip codes for Heart Disease and 241 zip codes for Stroke for 2010–2018 from

California Department of Public Health. Shannon index calculated from data from consortium of private tree maintenance companies. https://doi.org/10.1371/journal.pone.0254973.g003 PLOS ONE Biodiversity and cardiovascular disease PLOS ONE | https://doi.org/10.1371/journal.pone.0254973 November 3, 2021 10 / 23