Global estimates of daily ambient fine particulate matter concentrations and unequal spatiotemporal distribution of population exposure: a machine learning modelling study

Background Short-term exposure to ambient PM2 center dot 5 is a leading contributor to the global burden of diseases and mortality. However, few studies have provided the global spatiotemporal variations of daily PM2 center dot 5 concentrations over recent decades. Methods In this modelling study, we implemented deep ensemble machine learning (DEML) to estimate global daily ambient PM2 center dot 5 concentrations at 0.1 degrees x 0.1 degrees spatial resolution between Jan 1, 2000, and Dec 31, 2019. In the DEML framework, ground-based PM2 center dot 5 measurements from 5446 monitoring stations in 65 countries worldwide were combined with GEOS-Chem chemical transport model simulations of PM2 center dot 5 concentration, meteorological data, and geographical features. At the global and regional levels, we investigated annual population-weighted PM2 center dot 5 concentrations and annual population-weighted exposed days to PM2 center dot 5 concentrations higher than 15 mu g/m(3) (2021 WHO daily limit) to assess spatiotemporal exposure in 2000, 2010, and 2019. Land area and population exposures to PM2 center dot 5 above 5 mu g/m(3) (2021 WHO annual limit) were also assessed for the year 2019. PM2 center dot 5 concentrations for each calendar month were averaged across the 20-year period to investigate global seasonal patterns. Findings Our DEML model showed good performance in capturing the global variability in ground-measured daily PM2 center dot 5, with a cross-validation R-2 of 0.91 and root mean square error of 7.86 mu g/m(3). Globally, across 175 countries, the mean annual population-weighted PM2 center dot 5 concentration for the period 2000-19 was estimated at 32.8 mu g/m(3) (SD 0.6). During the two decades, population-weighted PM2 center dot 5 concentration and annual population-weighted exposed days (PM2.(5) >15 mu g/m(3)) decreased in Europe and northern America, whereas exposures increased in southern Asia, Australia and New Zealand, and Latin America and the Caribbean. In 2019, only 0.18% of the global land area and 0.001% of the global population had an annual exposure to PM2 center dot 5 at concentrations lower than 5 mu g/m(3), with more than 70% of days having daily PM2.(5) concentrations higher than 15 mu g/m(3). Distinct seasonal patterns were indicated in many regions of the world. Interpretation The high-resolution estimates of daily PM2.(5) provide the first global view of the unequal spatiotemporal distribution of PM2 center dot 5 exposure for a recent 20-year period, which is of value for assessing short-term and long-term health effects of PM2 center dot 5, especially for areas where monitoring station data are not available.