Impact of COVID-19 on Air Quality in Major Cities of Bangladesh: A Temporal Analysis (2018–2023) Shahanaj Rahman, Mim Mashrur Ahmed, Philip K. Hopke, Emdadul Hoque, Asrafuzzaman, Labib Marwan Hoque, Mansour Almazroui, Talal Suliman Alowaibdi, Arifur Rahman, Firoz Alam, Yingai Jin, Mamdud Hossain, Md. Mahmud Hossain, Mohammad Abdul Motalib, Mizanur Rahman, Kamrul Hasan, Kamrul Hassan Earth Systems and Environment, 2025
A new dynamic approach using data-driven and machine learning models for forecasting particulate matter in Dhaka megacity Kamrul Hasan, Mustafizur Rahman, Momotaj Akhter, Mohammad Mohinuzzaman, Imrul Kayes, Shahanaj Rahman Environmental Pollution and Management, 2024 This study conducts a comprehensive examination of six machine learning models for forecasting PM 2.5 and PM 10 concentrations in Dhaka, Bangladesh, employing average data from three air monitoring stations - Darus Salam, Parliament Area, and BARC established by the Department of Environment (DoE). The analysis utilizes average data from three air monitoring stations spanning January 2016 to December 2022, with meticulous pre-processing to ensure data quality. The employed models for analysis include ARIMA, ANN, ELM, ETS, NAÏVE, and TBATS. Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) are used to validate and rigorously compare model performance. ARIMA shows the best performance for PM 2.5 , while TBATS is slightly better for PM 10 predictions. These insights hold significant value for air quality management in Dhaka, enabling informed and proactive measures to counter particulate pollution and its adverse health implications. Furthermore, this study demonstrates the potential of machine learning models in accounting for local factors influencing air quality, complementing existing research on combining air quality models. This opens doors for further developing even better hybrid models, including weather data and exploring advanced ensemble techniques. • Utilized advanced machine learning model to predict air pollution of Dhaka. • ARIMA and TBATS outperformed other models for PM 2.5 and PM 10 prediction, respectively. • Leveraged real-world air quality data to capture local factors influencing pollution levels. • Demonstrated the potential of machine learning for developing sophisticated air quality forecasting tools. • Findings provide valuable guidance for policymakers to address air pollution challenges.
Particulate matter concentrations around natural gas-fired power plants and their associated health impact assessment Mustafizur Rahman, Kamrul Hasan, Md. Abu Bakar Siddique, Balram Ambade, Md. Alamgir Hussain, Mohammed Abdus Salam, Salman Tariq, Muhammad Ibrahim Journal of King Saud University Science, 2024 The quantification and prediction of particulate matter (PM) concentrations in the air are essential due to their negative impacts on human health and the environment. This study quantified PM concentrations and associated health effects at four natural gas-based power plants in Bangladesh. The measurement of PM2.5 and PM10 using the respirable dust samplers APS-113NL and APS-113BL, respectively from the year of 2015 to 2021 revealed that the concentration of both types of particles fluctuated over the years. The highest recorded levels of particles were in 2019, with PM2.5 at 126 µg/m3 and PM10 at 283 µg/m3 and the lowest recorded levels of particles were in 2017, with PM2.5 at 76.3 µg/m3 and PM10 at 203.3 µg/m3. In 2021, PM2.5 and PM10 concentrations were 88.5 and 225 µg/m3, respectively, lower than in the past two years. Statistical modeling assessed atmospheric contaminants analyzed time series data, and projected air quality. ARIMA, ETS, and ANN modeling methods have been used to predict the monthly average of PM2.5 and PM10 concentrations. RMSE, MAPE, MASE, and MAE have been utilized for model orders, time series analysis, and forecasting validation. There’s a significant variation between the forecasting models and forecasts for average PM2.5 and PM10 concentrations in natural gas-fired power plants from 2022 to 2024. This study also conducted a face-to-face interview with over 100 employees using a structured questionnaire to assess the health effects they are facing due to poor air quality in the power generation complex and found that 2 and 13 % of employees had respiratory and skin issues, respectively. Nonetheless, regular health checks, air filtration, and renewable energy consumption may benefit residents and the environment.
Potential of arima-ann, arima-svm, dt and catboost for atmospheric pm2.5 forecasting in bangladesh Shihab Ahmad Shahriar, Imrul Kayes, Kamrul Hasan, Mahadi Hasan, Rashik Islam, Norrimi Rosaida Awang, Zulhazman Hamzah, Aweng Eh Rak, Mohammed Abdus Salam Atmosphere, 2021 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.
The relationships between meteorological parameters and air pollutants in an urban environment I. Kayes, Shihab Ahmad Shahriar, Kamrul Hasan, M. Akhter, M. Kabir, Salam Global Journal of Environmental Science and Management, 2019 Meteorological parameters play a significant role in affecting ambient air quality of an urban environment. As Dhaka, the capital city of Bangladesh, is one of the air pollution hotspot among the megacities in the world, however the potential meteorological influences on criteria air pollutants for this megacity are remained less studied. The objectives of this research were to examine the relationships between meteorological parameters such as daily mean temperature (o C), relative humidity (%) and rainfall (mm) and, the concentration of criteria air pollutants (SO2, CO, NOx, O3, PM2.5 and PM10) from January, 2013 to December, 2017. This study also focused on the trend analysis of the air pollutants concentration over the period. Spearman correlation was applied to illustrate the relationships between air pollutants concentration and temperature, relative humidity and rainfall. Multiple linear and non-linear regressions were compared to explore potential role of meteorological parameters on air pollutants' concentrations. Trend analysis resulted that concentration of SO2 is increasing in the air of Dhaka while others are decreasing. Most of the pollutants resulted negative correlation with atmospheric temperature and relative humidity, however, they showed variable response to seasonal variation of meteorological parameters. Regression analysis resulted that both the multiple non-linear and linear model performed similar for predicting concentrations of particulate matters but for gaseous pollutants both model performances were poor. This research is expected to contribute in improving the forecast accuracy of air pollution under variable meteorological parameters considering seasonal fluctuations.
RECENT SCHOLAR PUBLICATIONS
Impact of COVID-19 on Air Quality in Major Cities of Bangladesh: A Temporal Analysis (2018–2023) S Rahman, MM Ahmed, PK Hopke, E Hoque, LM Hoque, M Almazroui, ... Earth Systems and Environment, 1-22 , 2024 2024 Citations: 1
A new dynamic approach using data-driven and machine learning models for forecasting particulate matter in Dhaka megacity K Hasan, M Rahman, M Akhter, M Mohinuzzaman, I Kayes, S Rahman Environmental Pollution and Management 1, 235-247 , 2024 2024 Citations: 2
Particulate matter concentrations around natural gas-fired power plants and their associated health impact assessment M Rahman, K Hasan, MAB Siddique, B Ambade, MA Hussain, MA Salam, S Tariq, M ... Journal of King Saud University-Science 36 (7), 103270 , 2024 2024 Citations: 8
Potential of ARIMA-ANN, ARIMA-SVM, DT and CatBoost for Atmospheric PM 2.5 Forecasting in Bangladesh SA Shahriar, I Kayes, K Hasan, M Hasan, R Islam, NR Awang, Z Hamzah, ... Atmosphere 12 (1), 100 , 2021 2021 Citations: 89
Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh SA Shahriar, I Kayes, K Hasan, MA Salam, S Chowdhury Air Quality, Atmosphere & Health 13 (10), 1247-1256 , 2020 2020 Citations: 73
The relationships between meteorological parameters and air pollutants in an urban environment I Kayes, SA Shahriar, K Hasan, M Akhter, MM Kabir, MA Salam Global Journal of Environmental Science and Management 5 (3), 265-278 , 2019 2019 Citations: 206
MOST CITED SCHOLAR PUBLICATIONS
The relationships between meteorological parameters and air pollutants in an urban environment I Kayes, SA Shahriar, K Hasan, M Akhter, MM Kabir, MA Salam Global Journal of Environmental Science and Management 5 (3), 265-278 , 2019 2019 Citations: 206
Potential of ARIMA-ANN, ARIMA-SVM, DT and CatBoost for Atmospheric PM 2.5 Forecasting in Bangladesh SA Shahriar, I Kayes, K Hasan, M Hasan, R Islam, NR Awang, Z Hamzah, ... Atmosphere 12 (1), 100 , 2021 2021 Citations: 89
Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh SA Shahriar, I Kayes, K Hasan, MA Salam, S Chowdhury Air Quality, Atmosphere & Health 13 (10), 1247-1256 , 2020 2020 Citations: 73
Particulate matter concentrations around natural gas-fired power plants and their associated health impact assessment M Rahman, K Hasan, MAB Siddique, B Ambade, MA Hussain, MA Salam, S Tariq, M ... Journal of King Saud University-Science 36 (7), 103270 , 2024 2024 Citations: 8
A new dynamic approach using data-driven and machine learning models for forecasting particulate matter in Dhaka megacity K Hasan, M Rahman, M Akhter, M Mohinuzzaman, I Kayes, S Rahman Environmental Pollution and Management 1, 235-247 , 2024 2024 Citations: 2
Impact of COVID-19 on Air Quality in Major Cities of Bangladesh: A Temporal Analysis (2018–2023) S Rahman, MM Ahmed, PK Hopke, E Hoque, LM Hoque, M Almazroui, ... Earth Systems and Environment, 1-22 , 2024 2024 Citations: 1
