Toward understanding atmospheric physics impacting the relationship between columnar aerosol optical depth and near-surface PM2.5 mass concentrations in Nevada and California, U.S.A., during 2013

Determining the relationship between columnar aerosol optical depth (τext) and surface particulate matter concentrations (PM2.5) is desired to estimate surface aerosol concentrations over broad spatial and temporal scales using satellite remote sensing. However, remote sensing studies incur challenges when surface aerosol pollution (i.e. PM2.5) is not correlated with columnar conditions (i.e., τext). PM2.5 data fusion models that rely on satellite data and statistical relationships of τext and PM2.5 may not be able to capture the physical conditions impacting the relationships that cause columnar and surface aerosols to not be correlated in the western U.S. Therefore, an extensive examination of the atmospheric conditions is required to improve surface estimates of PM2.5 that rely on columnar aerosol measurements. This investigation uses datasets from both routine monitoring networks and models of meteorological variables and aerosol physical parameters to understand the atmospheric conditions under which surface aerosol pollution can be explained by column measurements in California and Nevada during 2013. A novel quadrant method, that utilizes statistical analysis, was developed to investigate the relationship between τext and PM2.5. The results from this investigation show that τext and PM2.5 had a positive association (τext and PM2.5 increase together) when local sources of pollution or wildfires dominated aerosol pollution in the presence of a deep and well-mixed planetary boundary layer (PBL). Moreover, τext and PM2.5 had no association (where the variables are not related) when stable conditions, long-range transport, or entrainment of air from above the PBL were observed. It was found that seasonal categorization of the relationship between τext and PM2.5, an approach commonly used in statistical models to estimate surface concentrations with satellite remote sensing, may not be enough to account for the atmospheric conditions that drive the relationships between τext and PM2.5. For all stations, winter showed the maximum average PM2.5 concentrations (14.1 μg m−3, σ = 11.6 μg m−3) meanwhile, τext reached minimum values (0.06 μg m−3, σ = 0.04) during the same season. Conversely, spring presented the minimum average PM2.5 concentrations (9.4 μg m−3, σ = 6.9 μg m−3) and the average values of τext during spring had the second highest values (0.11, σ = 0.06) averaged for all stations.