PhotosVideosAllPeople
Impacts of aerosols on aspen growth under different sky conditions. (a–c) Beeswarm plots showing the relative influence of predictors on stem relative growth rate (RGR) in the random forest model. Predictors are ordered by importance based on the mean absolute Shapley value. Each point represents the mean Shapley value from 1000 iterations, corresponding to daily tree-level observations under different sky conditions. Points indicate the values of the predictor variable as shown in the color bar. The position along the x-axis reflects the magnitude and direction of each variable’s effect on stem RGR in the model. Overlapping points are jittered to show density. (d–f) Estimated radiative and climatic effects of aerosol on stem RGR. These estimates were derived from a sensitivity analysis of the random forest model, using changes in radiation and climatic variables resulting from a 50 μg m−3 increase in PM2.5 concentration. The bar plot shows the mean effect from 1000 simulation cycles, with error bars representing the standard deviation. Variable abbreviations: TotalR, total radiation; DiffuseF, diffuse radiation fraction; Tair, air temperature; VPD, vapor pressure deficit; SWC, soil water content; DBH, diameter at breast height; RDC, radial daily contraction.