PM2.5 Modeling and Historical Reconstruction over the Continental USA Utilizing GOES-16 AOD

Abstract

In this study, we present a nationwide machine learning model for hourly PM2.5 estimation for the continental United States (US) using high temporal resolution Geostationary Operational Environmental Satellites (GOES-16) Aerosol Optical Depth (AOD) data, meteorological variables from the European Center for Medium Range Weather Forecasting (ECMWF) and ancillary data collected between May 2017 and December 2020. A model sensitivity analysis was conducted on predictor variables to determine the optimal model. It turns out that GOES16 AOD, variables from ECMWF, and ancillary data are effective variables in PM2.5 estimation and historical reconstruction, which achieves an average mean absolute error (MAE) of 3.0 μg/m3, and a root mean square error (RMSE) of 5.8 μg/m3. This study also found that the model performance as well as the site measured PM2.5 concentrations demonstrate strong spatial and temporal patterns. Specifically, in the temporal scale, the model performed best between 8:00 p.m. and 11:00 p.m. (UTC TIME) and had the highest coefficient of determination (R2) in Autumn and the lowest MAE and RMSE in Spring. In the spatial scale, the analysis results based on ancillary data show that the R2 scores correlate positively with the mean measured PM2.5 concentration at monitoring sites. Mean measured PM2.5 concentrations are positively correlated with population density and negatively correlated with elevation. Water, forests, and wetlands are associated with low PM2.5 concentrations, whereas developed, cultivated crops, shrubs, and grass are associated with high PM2.5 concentrations. In addition, the reconstructed PM2.5 surfaces serve as an important data source for pollution event tracking and PM2.5 analysis. For this purpose, from May 2017 to December 2020, hourly PM2.5 estimates were made for 10 km by 10 km and the PM2.5 estimates from August through November 2020 during the period of California Santa Clara Unite (SCU) Lightning Complex fires are presented. Based on the quantitative and visualization results, this study reveals that a number of large wildfires in California had a profound impact on the value and spatial-temporal distributions of PM2.5 concentrations.

Providing Fine Temporal and Spatial Resolution Analyses of Airborne Particulate Matter Utilizing Complimentary In-Situ IoT Sensor Network and Remote Sensing Approaches

“…Data collection for this study, encompassing PM 2.5 , meteorological variables, AOD, and solar angles, varied in temporal and spatial resolutions and spanned from January 2020 to June 2023. To analyze these data, tree-based machine learning methods [17,18] were used. These methods were chosen for their effectiveness in handling the highly time-sensitive nature of the data, including the target variable PM 2.5 and other influencing environmental factors.…”

“…This modification affects various atmospheric conditions, including temperature, wind patterns, and precipitation. The presence of particulate matter can lead to the formation of fog and acid rain and contributes to the greenhouse effect, as discussed in [5][6][7][8][9][10][11].…”

“…Given the strong link between various health issues and PM, which exhibits significant variations over time and across different locations, it is crucial to conduct comprehensive studies to better understand the distribution of PM with high temporal and spatial precision [3,11]. Although ground-based monitoring stations are vital, their sparse and uneven distribution across regions makes it difficult to achieve continuous nationwide coverage.…”

Meteorological data source comparison—a case study in geospatial modeling of potential environmental exposure to abandoned uranium mine sites in the Navajo Nation

“…Gridded data products have been extensively applied in geospatial modeling efforts. For example, gridded MET data have been used in Bayesian models to determine the extent of PM 2.5 in urban areas (Nicolis et al, 2019 ), in random-forest models to determine daily concentrations of PM 10 and PM 2.5 (Stafoggia et al, 2019 ), including nationwide prediction of PM 2.5 (Yu et al, 2021 ), and in quantile regressions to determine the spread of fungal spores (Grinn-Gofroń et al, 2019 ). While there have been cases where the uncertainty in the interpolation process has proven negligible (Elaji & Ji, 2020 ), this remains an underdeveloped area for consideration in the model development and evaluation process.…”

“…Previous studies have integrated a wide variety of MET data sources in modeling efforts. As an example, gridded data, reanalysis data, and geostationary satellite data were combined for PM 2.5 modeling (Yu et al, 2021 ).…”