Global PM2.5 and secondary organic aerosols (SOA) levels with sectorial contribution to anthropogenic and biogenic SOA formation

This study estimates global PM_2.5 and anthropogenic and biogenic Secondary Organic Aerosols (a-SOA and b-SOA) and sources contributing to their formation. The global landscape was divided into eleven domains (North America (NAM); South America (SAM); Europe (EUR); North Africa and Middle East (NAF); Equatorial Africa (EAF); South of Africa (SAF); Russia and Central Asia (RUS); Eastern Asia (EAS); South Asia (SAS); Southeast Asia (SEA) and Australia (AUS)) and 46 cities based on varying populations. Three inventories for global emissions were considered: Community Emissions Data System, Model of Emission of Gases and Aerosol, and Global Fire Emissions Database. WRF-Chem model coupled with atmospheric reactions and the secondary organic aerosol model was employed for estimating PM_2.5, a-SOA, and b-SOA for 2018. No city attained WHO's annual PM_2.5 guideline of 5 μg/m³. Delhi, Dhaka, and Kolkata (63-92 μg/m³) in south Asia were the most polluted, and seven cities (mostly in EUR and NAM) met the WHO target IV (10 μg/m³). The highest SOA levels (2-9 μg/m³) were in the cities of SAS and Africa, but with a low SOA contribution to PM_2.5 (3-22%). However, the low levels of SOA (1-3 μg/m³) in EUR and NAM had a higher contribution of SOA to PM_2.5 (20-33%). b-SOA were consistent with the region's vegetation and forest cover. The SOA contribution was dominated by residential emissions in all domains (except in the NAF and AUS) (maximum in SAS). The non-coal industry was the second highest contributor (except in EAF, NAF, and AUS) and EUR had the maximum contribution from agriculture and transport. Globally, residential and industry (non-coal and coal) sectors showed the maximum contribution to SOA, with a-SOA and b-SOA being nearly equal. Ridding of biomass and residential burning of solid fuel is the single most action benefiting the PM_2.5 and SOA concerns.

Investigating the Relationship between Air Pollutants and Meteorological Parameters Using Satellite Data over Bangladesh

Understanding of the relationship between air pollutants and meteorological parameters on the regional scale is a prerequisite for setting up air pollution prevention and control strategies; however, there is a lack of methodical investigations, particularly in the context of Bangladesh’s deficiency of information on air pollution. This study represents the first attempt to investigate the relationship between air pollutants (NO2, O3, SO2, and CO) and meteorological parameters over Bangladesh using satellite data (OMI and MOPITT) during the period from 2015 to 2020. Geographically weighted regression (GWR) modelling was utilized to assess the relationship between air pollutants and weather variables. The spatial representation and average values of geographically varying coefficients showed that the column densities of air pollutants were affected by the meteorological parameters. For example, NO2 was positively associated with temperature in most of the studied regions, with an average geographically varying coefficient value of 0.12 Dobson units (DU, 1 DU = 2.687 × 1016 molecules/cm2), indicating that NO2 concentrations increase by 0.12 DU/year with every unit increase in temperature. The sources of NO2 and SO2 in Dhaka were identified through emission inventory analysis, and transportation and industry emissions were the most significant influencing factors for NO2 and SO2, respectively. Temperature and pressure showed a higher degree of relationship with all four air pollutants compared with other parameters. The results and discussion presented in this study can be of benefit for policy makers in developing air pollution control strategies in Bangladesh.

Potential of ARIMA-ANN, ARIMA-SVM, DT and CatBoost for Atmospheric PM2.5 Forecasting in Bangladesh

Atmospheric particulate matter (PM) has major threats to global health, especially in urban regions around the world. Dhaka, Narayanganj and Gazipur of Bangladesh are positioned as top ranking polluted metropolitan cities in the world. This study assessed the performance of the application of hybrid models, that is, Autoregressive Integrated Moving Average (ARIMA)-Artificial Neural Network (ANN), ARIMA-Support Vector Machine (SVM) and Principle Component Regression (PCR) along with Decision Tree (DT) and CatBoost deep learning model to predict the ambient PM2.5 concentrations. The data from January 2013 to May 2019 with 2342 observations were utilized in this study. Eighty percent of the data was used as training and the rest of the dataset was employed as testing. The performance of the models was evaluated by R2, RMSE and MAE value. Among the models, CatBoost performed best for predicting PM2.5 for all the stations. The RMSE values during the test period were 12.39 µg m−3, 13.06 µg m−3 and 12.97 µg m−3 for Dhaka, Narayanganj and Gazipur, respectively. Nonetheless, the ARIMA-ANN and DT methods also provided acceptable results. The study suggests adopting deep learning models for predicting atmospheric PM2.5 in Bangladesh.

Assessing the PM2.5 impact of biomass combustion in megacity Dhaka, Bangladesh

In Dhaka, Bangladesh, fine particulate matter (PM_2.5) air pollution shows strong seasonal trends, with significantly higher mean concentrations during winter than during the monsoon (winter = 178.1 μg/m³ vs. monsoon = 30.2 μg/m³). Large-scale open burning of post-harvest agricultural waste across the Indo-Gangetic Plain is a major source of PM_2.5 air pollution in northern India during the non-monsoon period. This study evaluates the extent to which the seasonal differences in PM_2.5 pollution concentrations in Dhaka are accounted for by biomass-burning vs. fossil-fuel combustion sources. To assess this, an index was developed based on elemental potassium (K) as a marker for biomass particulate matter, after adjusting for soil-associated K contributions. Alternatively, particulate sulfur was employed as a tracer index for fossil-fuel combustion PM_2.5. By simultaneously regressing total PM_2.5 on S and adjusted K, the PM_2.5 mass for each day was apportioned into: 1) fossil-fuels combustion associated PM_2.5; 2) biomass-burning associated PM_2.5; and, 3) all other PM_2.5. The results indicated that fossil-fuel combustion contributed 21.6% (19.5 μg/m³), while biomass contributed 40.2% (36.3 μg/m³) of overall average PM_2.5 from September 2013 to December 2017. However, the mean source contributions varied by season: PM_2.5 in Dhaka during the monsoon season was dominated by fossil-fuels sources (44.3%), whereas PM_2.5 mass was dominated by biomass-burning (41.4%) during the remainder of the year. The contribution to PM_2.5 and each of its source components by transport of pollution into Dhaka during non-monsoon time was also evaluated by: 1) Conditional bivariate (CBPF) and pollution rose plots; 2) Concentration weighted trajectories (CWT), and; 3) NASA satellite photos to identify aerosol loading and fire locations on high pollution days. The collective evidence indicates that, while the air pollution in Dhaka is contributed to by both local and transboundary sources, the highest pollution days were dominated by biomass-related PM_2.5, during periods of crop-burning in the Indo-Gangetic Plain.