In this study, we present a nationwide machine learning model for hourly PM${}_{2.5}$ 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 PM${}_{2.5}$ estimation and historical reconstruction, which achieves an average mean absolute error (MAE) of 3.0 $\mathsf{\mu }$g/m${}^3$, and a root mean square error (RMSE) of 5.8 $\mathsf{\mu }$g/m${}^3$. This study also found that the model performance as well as the site measured PM${}_{2.5}$ 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 (R${}^2$) in Autumn and the lowest MAE and RMSE in Spring. In the spatial scale, the analysis results based on ancillary data show that the R${}^2$ scores correlate positively with the mean measured PM${}_{2.5}$ concentration at monitoring sites. Mean measured PM${}_{2.5}$ concentrations are positively correlated with population density and negatively correlated with elevation. Water, forests, and wetlands are associated with low PM${}_{2.5}$ concentrations, whereas developed, cultivated crops, shrubs, and grass are associated with high PM${}_{2.5}$ concentrations. In addition, the reconstructed PM${}_{2.5}$ surfaces serve as an important data source for pollution event tracking and PM${}_{2.5}$ analysis. For this purpose, from May 2017 to December 2020, hourly PM${}_{2.5}$ estimates were made for 10 km by 10 km and the PM${}_{2.5}$ 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 PM${}_{2.5}$ concentrations.

This study aims to provide analyses of the levels of airborne particulate matter (PM) using a two-pronged approach that combines data from in situ Internet of Things (IoT) sensor networks with remotely sensed aerosol optical depth (AOD). Our approach involved setting up a network of custom-designed PM sensors that could be powered by the electrical grid or solar panels. These sensors were strategically placed throughout the densely populated areas of North Texas to collect data on PM levels, weather conditions, and other gases from September 2021 to June 2023. The collected data were then used to create models that predict PM concentrations in different size categories, demonstrating high accuracy with correlation coefficients greater than 0.9. This highlights the importance of collecting hyperlocal data with precise geographic and temporal alignment for PM analysis. Furthermore, we expanded our analysis to a national scale by developing machine learning models that estimate hourly PM${ }_{2.5}$ levels throughout the continental United States. These models used high-resolution data from the Geostationary Operational Environmental Satellites (GOES-16) Aerosol Optical Depth (AOD) dataset, along with meteorological data from the European Center for Medium-Range Weather Forecasting (ECMWF), AOD reanalysis, and air pollutant information from the MERRA-2 database, covering the period from January 2020 to June 2023. Our models were refined using ground truth data from our IoT sensor network, the OpenAQ network, and the National Environmental Protection Agency (EPA) network, enhancing the accuracy of our remote sensing PM estimates. The findings demonstrate that the combination of AOD data with meteorological analyses and additional datasets can effectively model PM${ }_{2.5}$ concentrations, achieving a significant correlation coefficient of 0.849. The reconstructed PM${ }_{2.5}$ surfaces created in this study are invaluable for monitoring pollution events and performing detailed PM${ }_{2.5}$ analyses. These results were further validated through real-world observations from two in situ MINTS sensors located in Joppa (South Dallas) and Austin, confirming the effectiveness of our comprehensive approach to PM analysis. The US Environmental Protection Agency (EPA) recently updated the national standard for PM${ }_{2.5}$ to 9 $\mathsf{\mu }$g/m${ }^3$, a move aimed at significantly reducing air pollution and protecting public health by lowering the allowable concentration of harmful fine particles in the air. Using our analysis approach to reconstruct the fine-time resolution PM${ }_{2.5}$ distribution across the entire United States for our study period, we found that the entire nation encountered PM${ }_{2.5}$ levels that exceeded 9 $\mathsf{\mu }$g/m${ }^3$ for more than 20% of the time of our analysis period, with the eastern United States and California experiencing concentrations exceeding 9 $\mathsf{\mu }$g/m${ }^3$ for over 50% of the time, highlighting the importance of regulatory efforts to maintain annual PM${ }_{2.5}$ concentrations below 9 $\mathsf{\mu }$g/m${ }^3$.

Remote sensing imagery, such as that provided by the United States Geological Survey (USGS) Landsat satellites, has been widely used to study environmental protection, hazard analysis, and urban planning for decades. Clouds are a constant challenge for such imagery and, if not handled correctly, can cause a variety of issues for a wide range of remote sensing analyses. Typically, cloud mask algorithms use the entire image; in this study we present an ensemble of different pixel-based approaches to cloud pixel modeling. Based on four training subsets with a selection of different input features, 12 machine learning models were created. We evaluated these models using the cropped LC8-Biome cloud validation dataset. As a comparison, Fmask was also applied to the cropped scene Biome dataset. One goal of this research is to explore a machine learning modeling approach that uses as small a training data sample as possible but still provides an accurate model. Overall, the model trained on the sample subset (1.3% of the total training samples) that includes unsupervised Self-Organizing Map classification results as an input feature has the best performance. The approach achieves $98.57\%$ overall accuracy, $1.18\%$ cloud omission error, and $0.93\%$ cloud commission error on the 88 cropped test images. By comparison to Fmask 4.0, this model improves the accuracy by $10.12\%$ and reduces the cloud omission error by $6.39\%$. Furthermore, using an additional eight independent validation images that were not sampled in model training, the model trained on the second largest subset with an additional five features has the highest overall accuracy at $86.35\%$, with $12.48\%$ cloud omission error and $7.96\%$ cloud commission error. This model’s overall correctness increased by $3.26\%$, and the cloud omission error decreased by $1.28\%$ compared to Fmask 4.0. The machine learning cloud classification models discussed in this paper could achieve very good performance utilizing only a small portion of the total training pixels available. We showed that a pixel-based cloud classification model, and that as each scene obviously has unique spectral characteristics, and having a small portion of example pixels from each of the sub-regions in a scene can improve the model accuracy significantly.

PM${}_{2.5}$, a type of fine particulate with a diameter equal to or less than 2.5 micrometers, has been identified as a major source of air pollution, and is associated with many health issues. Research on utilizing various data sources, such as remote sensing and in situ sensors, for PM${}_{2.5}$ concentrations modeling remains a hot topic. In this study, the Next Generation Weather Radar (NEXRAD) is used as a supplementary weather data source, along with European Centre for Medium-Range Weather Forecasts (ECMWF), solar angles, and Geostationary Operational Environmental Satellite (GOES16) Aerosol Optical Depth (AOD) to model high spatial-temporal PM${}_{2.5}$ concentrations. PM${}_{2.5}$ concentrations as well as in situ weather condition variables are collected from the 31 sensors that are deployed in the Dallas Metropolitan area. Four machine learning models with different predictor variables are developed based on an ensemble approach. Since in situ weather observations are not widely available, ECMWF is used as an alternative data source for weather conditions in studies. Hence, the four established models are compared in three groups. Both models in this first group use weather variables collected from deployed sensors, but one uses NEXRAD and the other does not. In the second group, the two models use weather variables retrieved from ECMWF, one using NEXRAD and one without. In the third group, one model uses weather variables from ECMWF, and the other uses in situ weather variables, both without NEXRAD. The first two environmental groups investigate how NEXRAD can enhance model performances with weather variables collected from in situ observations and ECMWF, respectively. The third group explores how effective using ECMWF as an alternative source of weather conditions. Based on the results, the incorporation of NEXRAD achieves an R${}^2$ score of 0.86 and 0.83 for groups 1 and 2, respectively, for an improvement of 2.8% and 9.6% over those models without NEXRAD. For group three, the use of ECMWF as an alternative source of in situ weather observations results in a 0.13 R${}^2$ drop. For PM${}_{2.5}$ estimation, weather variables including precipitation, temperature, pressure, and surface pressure from ECMWF and deployed sensors, as well as NEXRAD velocity, are shown to be significant factors.

This study aims to provide analyses of the levels of airborne particulate matter (PM) using a two-pronged approach that combines data from in situ Internet of Things (IoT) sensor networks with remotely sensed aerosol optical depth (AOD). Our approach involved setting up a network of custom-designed PM sensors that could be powered by the electrical grid or solar panels. These sensors were strategically placed throughout densely populated areas of North Texas to collect data on PM levels, weather conditions, and other gases from September 2021 to June 2023. The collected data was then used to create models that predict PM concentrations in different size categories, demonstrating high accuracy with correlation coefficients greater than 0.9. This highlights the importance of collecting hyperlocal data with precise geographic and temporal alignment for PM analysis. Furthermore, we expanded our analysis to a national scale by developing machine learning models that estimate hourly PM2.5 levels throughout the continental United States. These models used high-resolution data from the Geostationary Operational Environmental Satellites (GOES-16) Aerosol Optical Depth (AOD) dataset, along with meteorological data from the European Center for Medium-Range Weather Forecasting (ECMWF), AOD reanalysis, and air pollutant information from the MERRA-2 database, covering the period from January 2020 to June 2023. Our models were refined using ground truth data from our IoT sensor network, the OpenAQ network, and the National Environmental Protection Agency (EPA) network, enhancing the accuracy of our remote sensing PM estimates. The findings demonstrate that the combination of AOD data with meteorological analyses and additional data sets can effectively model PM2.5 concentrations, achieving a significant correlation coefficient of 0.849. The reconstructed PM2.5 surfaces created in this study are invaluable for monitoring pollution events and performing detailed PM2.5 analyzes. These results were further validated through real-world observations from two in situ MINTS sensors located in Joppa (South Dallas) and Austin, confirming the effectiveness of our comprehensive approach to PM analysis. The US Environmental Protection Agency (EPA) recently updated the national standard for PM2.5 to 9 μg/m3, a move aimed at significantly reducing air pollution and protecting public health by lowering the allowable concentration of harmful fine particles in the air. Using our analysis approach to reconstruct the fine-time resolution PM2.5 distribution across the entire United States for our study period, we found that the entire nation encountered PM2.5 levels that exceeded 9 μg/m3 for more than 20% of the time of our analysis period, with the eastern United States and California experiencing concentrations exceeding 9 μg/m3 for over 50% of the time, highlighting the importance of regulatory efforts to maintain annual PM2.5 concentrations below 9 μg/m3.