Assessing Effect of Targeting Reduction of PM2.5 Concentration on Human Exposure and Health Burden in Hong Kong Using Satellite Observation

Abstract

Targeting reduction of PM2.5 concentration lessens population exposure level and health burden more effectively than uniform reduction does. Quantitative assessment of effect of the targeting reduction is limited because of the lack of spatially explicit PM2.5 data. This study aimed to investigate extent of exposure and health benefits resulting from the targeting reduction of PM2.5 concentration. We took advantage of satellite observations to characterize spatial distribution of PM2.5 concentration at a resolution of 1 km. Using Hong Kong of China as the study region (804 satellite's pixels covering its residential areas), human exposure level (c) and premature mortality attributable to PM2.5 (Mort) for 2015 were estimated to be 25.9 g/m(3) and 4112 people per year, respectively. We then performed 804 diagnostic tests that reduced PM2.5 concentrations by -1 g/m(3) in different areas and a reference test that uniformly spread the -1 g/m(3). We used a benefit rate from targeting reduction (BRT), which represented a ratio of declines in c (or Mort) with and without the targeting reduction, to quantify the extent of benefits. The diagnostic tests estimated the BRT levels for both human exposure and premature mortality to be 4.3 over Hong Kong. It indicates that the declines in human exposure and premature mortality quadrupled with a targeting reduction of PM2.5 concentration over Hong Kong. The BRT values for districts of Hong Kong could be as high as 5.6 and they were positively correlated to their spatial variabilities in population density. Our results underscore the substantial exposure and health benefits from the targeting reduction of PM2.5 concentration. To better protect public health in Hong Kong, super-regional and regional cooperation are essential. Meanwhile, local environmental policy is suggested to aim at reducing anthropogenic emissions from mobile and area (e.g., residential) sources in central and northwestern areas.