Quantitative Aerosol Optical Depth Detection during Dust Outbreaks from Meteosat Imagery Using an Artificial Neural Network Model

Long-term study of desert dust deposition effects on phytoplankton biomass in the Persian Gulf using Google Earth Engine

Persian Gulf is surrounded by the most active dust source regions of the world. Every year, millions of tons of dust are deposited in this area causing significant adverse effects. This research aimed to fill the gap in the field of dust impact quantification regarding different and important regions in the Persian Gulf. We evaluated aerosol optical thickness (AOT) as dust, chlorophyll a (Chl a) as phytoplankton biomass, sea surface temperature (SST), and particulate organic carbon (POC) using Moderate Resolution Imaging Spectroradiometer (MODIS) data to quantify and analyze the impacts of dust. According to dust frequency maps over 21 years, 4 critical regions were selected including the Northern, Central, and Southern regions of the Persian Gulf and Strait of Hormuz. Slightly lower Chl a values were noted in the southern region and Strait of Hormuz compared to the Northern and Central regions. The Mann-Kendall (MK) test and Sen's slope estimator, which can determine trend parameters, were used to analyze 4 selected regions using 8 days, monthly, and yearly averaged data. The results of the MK test showed a significant positive trend for SST and a significant negative trend for POC. AOT and Chl a trends varied based on their locations. The cross-correlation test with a time lag showed maximum correlations between Chl a and dust (AOT) with a delay of 6-7 months. Also, to quantify the impacts of dust on phytoplankton biomass, R2 and analysis of variance (ANOVA) regression were used for data with a 6-7-month time lag. The results showed that the contribution of dust in the amount of Chl a variation was about 10 %-20 %. This study showed that the simultaneous use of remote sensing and statistical methods could provide necessary and timely warnings (regarding management) to prevent algal blooms and aquatic loss in this important and strategic area.

Impacts of cement industry air pollutants on the environment and satellite data applications for air quality monitoring and management

The cement industry plays critical role in any country's development and economic growth. Cement is extensively used in construction sector and infrastructural projects. Abundant raw material availability, infrastructure demands, urbanization, and recent government initiatives-Atal Mission for Rejuvenation and Urban Transformation (AMRUT) project and housing for all under the Pradhan Mantri Awas Yojana (PMAY), India-stood at second place in cement-producing country in the world. Cement plants are emitting 15% of global pollutions into the environment among various industries. Cement industry byproducts include dust/particulate matter (PM_2.5 and PM₁₀), toxic gases (COx, NO_x, SOx, CH₄, and VOCs), noise, and heavy metals (Cr, Ni, Co, Pb, and Hg), which cause climate change, global warming, and health risks as well as negative impact on flora and fauna. The Terra, Aura, Sentinel-5P, GOSAT, and other satellite datasets allow estimation of cement industry major air pollutants such as particulate matter (PM), sulfur dioxide (SO₂), nitrogen dioxide (NO₂), carbon dioxide (CO₂), and volatile organic compounds (VOCs) using regression models, artificial neural network-based models, machine learning models, and the tropospheric NO2 vertical column density (VCD) retrieval algorithm. This review article explores the evolution of the Indian cement industry, air pollutants from the cement industry, social and environmental implications, satellite datasets, models used for assessing air pollutants, and challenges for the long-term sustainability of the cement industry.

A benchmark dataset for binary segmentation and quantification of dust emissions from unsealed roads

The generation of reference data for machine learning models is challenging for dust emissions due to perpetually dynamic environmental conditions. We generated a new vision dataset with the goal of advancing semantic segmentation to identify and quantify vehicle-induced dust clouds from images. We conducted field experiments on 10 unsealed road segments with different types of road surface materials in varying climatic conditions to capture vehicle-induced road dust. A direct single-lens reflex (DSLR) camera was used to capture the dust clouds generated due to a utility vehicle travelling at different speeds. A research-grade dust monitor was used to measure the dust emissions due to traffic. A total of ~210,000 images were photographed and refined to obtain ~7,000 images. These images were manually annotated to generate masks for dust segmentation. The baseline performance of a truncated sample of ~900 images from the dataset is evaluated for U-Net architecture.

Himawari-8/AHI Aerosol Optical Depth Detection Based on Machine Learning Algorithm

Due to the advantage of geostationary satellites, Himawari-8/AHI can provide near-real-time air quality monitoring over China with a high temporal resolution. Satellite-based aerosol optical depth (AOD) retrieval over land is a challenge because of the large surface contribution to the top of atmosphere (TOA) signal and the uncertainty of aerosol modes. Here, by combining satellite TOA reflectance, sun-sensor geometries, meteorological factors and vegetation information, we propose a data-driven AOD detection algorithm based on a deep neural network (DNN) model for Himawari-8/AHI. It is trained by sample data of 2018 and 2019 and is applied to derive hourly AODs over China in 2020. By comparison with ground-based AERONET measurements, R2 for DNN-estimated AOD is up to 0.8702, which is much higher than that for the AHI AOD product with R2 = 0.4869. The hourly AOD results indicate that the DNN model has a good potential in improving the performance of AOD retrieval in the early morning and in the late afternoon, and the spatial distribution is reliable for capturing the variation of aerosol pollution on the regional scale. By analyzing different DNN modeling strategies, it is found that seasonal modeling can hardly increase the accuracy of AOD retrieval to a certain extent, and R2 increases from 0.7394 to 0.8168 when meteorological features, especially air pressure, are involved in the model training.

Himawari-8 Aerosol Optical Depth (AOD) Retrieval Using a Deep Neural Network Trained Using AERONET Observations

Spectral aerosol optical depth (AOD) estimation from satellite-measured top of atmosphere (TOA) reflectances is challenging because of the complicated TOA-AOD relationship and a nexus of land surface and atmospheric state variations. This task is usually undertaken using a physical model to provide a first estimate of the TOA reflectances which are then optimized by comparison with the satellite data. Recently developed deep neural network (DNN) models provide a powerful tool to represent the complicated relationship statistically. This study presents a methodology based on DNN to estimate AOD using Himawari-8 Advanced Himawari Imager (AHI) TOA observations. A year (2017) of AHI TOA observations over the Himawari-8 full disk collocated in space and time with Aerosol Robotic Network (AERONET) AOD data were used to derive a total of 14,154 training and validation samples. The TOA reflectance in all six AHI solar bands, three TOA reflectance ratios derived based on the dark-target assumptions, sun-sensor geometry, and auxiliary data are used as predictors to estimate AOD at 500 nm. The DNN AOD is validated by separating training and validation samples using random k-fold cross-validation and using AERONET site-specific leave-one-station-out validation, and is compared with a random forest regression estimator and Japan Meteorological Agency (JMA) AOD. The DNN AOD shows high accuracy: (1) RMSE = 0.094, R2 = 0.915 for k-fold cross-validation, and (2) RMSE = 0.172, R2 = 0.730 for leave-one-station-out validation. The k-fold cross-validation overestimates the DNN accuracy as the training and validation samples may come from the same AHI pixel location. The leave-one-station-out validation reflects the accuracy for large-area applications where there are no training samples for the pixel location to be estimated. The DNN AOD has better accuracy than the random forest AOD and JMA AOD. In addition, the contribution of the dark-target derived TOA ratio predictors is examined and confirmed, and the sensitivity to the DNN structure is discussed.

Extracting Taklimakan Dust Parameters from AIRS with Artificial Neural Network Method

Two back-propagation artificial neural network retrieval models have been developed for obtaining the dust aerosol optical depth (AOD) and dust-top height (DTH), respectively, from Atmospheric InfraRed Sounder (AIRS) brightness temperature (BT) measurements over Taklimakan Desert area. China Aerosol Remote Sensing Network (CARSNET) measurements at Tazhong station were used for dust AOD validation. Results show that the correlation coefficient of dust AODs between AIRS and CARSNET reaches 0.88 with a deviation of −0.21, which is the same correlation coefficient as the AIRS dust AOD and the Moderate-Resolution Imaging Spectroradiometer (MODIS) aerosol optical depth (AOD) product. In the AIRS DTH retrieval model, there is an option to include the collocated MODIS deep blue (DB) AOD as additional input for daytime retrieval; the independent dust heights from Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) are used for AIRS DTH validation, and results show that the DTHs derived from the combined AIRS BT measurements and MODIS DB AOD product have better accuracy than those from AIRS BT measurements alone. The correlation coefficient of DTHs between AIRS and independent CALIOP dust heights is 0.79 with a standard deviation of 0.41 km when MODIS DB AOD product is included in the retrieval model. A series of case studies from different seasons were examined to demonstrate the feasibility of retrieving dust parameters from AIRS and potential applications. The method and approaches can be applied to process measurements from advanced infrared (IR) sounder and high-resolution imager onboard the same platform.