High Spatiotemporal Resolution PM2.5 Concentration Estimation with Machine Learning Algorithm: A Case Study for Wildfire in California

As an aggregate of suspended particulate matter in the air, atmospheric aerosols can affect the regional climate. With the help of satellite remote sensing technology to retrieve AOD (aerosol optical depth) on a global or regional scale, accurate estimation of PM2.5 concentration has become an important task to quantify the spatiotemporal distribution of AOD and PM2.5. However, due to the limitations of satellite platforms, sensors, and inversion algorithms, the spatiotemporal resolution of current major AOD products is still relatively low. Meanwhile, for the impact of cloud, the AOD products often have a serious data gap problem, which also objectively limits the spatiotemporal coverage of predicted PM2.5 concentration. Therefore, how to effectively improve the spatiotemporal resolution and coverage of PM2.5 concentration under the requisite accuracy is still a grand challenge. In this study, the fused high spatial-temporal resolution AOD data in our previous study were used to estimate the ground PM2.5 concentration through machine learning algorithms, the deep belief network (DBN). The PM2.5 data had spatiotemporal autocorrelation in geostatistics and followed the Gaussian kernel distribution. Hence, the autocorrelation model modified by Gaussian kernel function integrated with DBN algorithm, named Geoi-DBN, was used to estimate PM2.5 concentration. The cross-validation results showed that the Geoi-DBN (R2 = 0.86, RMSE = 6.84 µg m−3) performed better than the original DBN (R2 = 0.67, RMSE = 10.46 µg m−3). The final high quality PM2.5 concentration data can be applied for urban air quality monitoring and related PM2.5 exposure risk assessment such as wildfire.

Tropospheric aerosols: size-differentiated chemistry and large-scale spatial distributions

Worldwide interest in atmospheric aerosols has emerged since the late 20th century as a part of concerns for air pollution and radiative forcing of the earth's climate. The use of aircraft and balloons for sampling and the use of remote sensing have dramatically expanded knowledge about tropospheric aerosols. Our survey gives an overview of contemporary tropospheric aerosol chemistry based mainly on in situ measurements. It focuses on fine particles less than 1-2.5 microm in diameter. The physical properties of particles by region and altitude are exemplified by particle size distributions, total number and volume concentration, and optical parameters such as extinction coefficient and aerosol optical depth. Particle chemical characterization is size dependent, differentiated by ubiquitous sulfate, and carbon, partially from anthropogenic activity. Large-scale particle distributions extend to intra- and intercontinental proportions involving plumes from population centers to natural disturbances such as dust storms and vegetation fires. In the marine environment, sea salt adds an important component to aerosols. Generally, aerosol components, most of whose sources are at the earth's surface, tend to dilute and decrease in concentration with height, but often show different (layered) profiles depending on meteorological conditions. Key microscopic processes include new particle formation aloft and cloud interactions, both cloud initiation and cloud evaporation. Measurement campaigns aloft are short term, giving snapshots of inherently transient phenomena in the troposphere. Nevertheless, these data, combined with long-term data at the surface and optical depth and transmission observations, yield a unique picture of global tropospheric particle chemistry.