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 PM_2.5 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 PM_2.5 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 PM_2.5 in Bangladesh.

Ground-level ozone (O₃) is mainly produced during daytime in the presence of ultraviolet (UV) light and later destroyed by nitrogen oxides during nighttime. However, light pollution caused by the excessive use of artificial lights may disrupt the chemistry of night-time ground-level O₃ by providing enough energy to initiate nighttime ground-level O₃ production. In this study, nighttime (7 p.m. to 7 a.m.) ground-level O₃, nitrogen oxide (NO), and nitrogen dioxides (NO₂) concentrations were observed for three years (2013, 2014, and 2015). The existence of O₃ was found during nighttime, especially in urban areas with a concentration range of 8–20 ppb. The results suggested that nighttime variations of ground-level O₃ concentrations were higher in urban areas than in suburban areas. The mean nighttime O₃ concentration at urban sites varied, possibly because the distribution of anthropogenic lights around the urban sites is brighter than in suburban locations, as indicated by the data from the light-pollution map. This anthropogenic light has not caused the suspected nighttime photolysis processes, which directly slowed nighttime oxidation. The photochemistry rate of JNO₂/k₃ was supposed to be near zero because of the absence of photochemical reactions at night. However, the minimum concentration in all urban and suburban sites ranged from 2–3 ppb, indicating that O₃ might also form at night, albeit not due to light pollution.

Ground-level ozone (O3) is a significant air pollutant formed when sunlight reacts with nitrogen oxides (NOx) and volatile organic compounds (VOCs). It poses risks to human health, vegetation, crops, and materials. This study focuses on ground-level ozone in Klang Valley, Malaysia, using hourly average data from 2010 to 2020 acquired from the Department of Environment in Klang, Shah Alam, and Petaling Jaya. Descriptive statistics and GIS tools are employed to analyse the distribution and variations of pollutants during the study periods. The study revealed that the maximum concentration in all locations in most of the year exceeded 100 ppb, with the maximum recorded in Shah Alam at 161 ppb, while mean concentrations were 13.69 – 21.18 ppb in Klang, 17.99 – 23.88 ppb in Shah Alam, and 9.40 – 17.93 ppb in Petaling Jaya, with the daily maximum concentration still exceeding the hourly average recommended value of 90 ppb. Spatial analysis using GIS reveals the distribution of O3 concentrations is highly affected by seasonal wind direction, as high O3 concentration is observed from January to March. The study underscores the pressing issue of ozone pollution in Klang Valley and the need for mitigation strategies to improve air quality.

Ozone is a gas that plays important role in balancing the Earth’s atmosphere thermal and chemical. Ozone is capable of absorbing the harmful sun’s ultraviolet (UV) radiation. The stratospheric ozone has protecting humans and the ecosystems on Earth from solar radiation. However, ground level ozone (GLO) has been related to secondary pollutants and greenhouse gases that give adverse impacts on human health and crop yield. Hence, determining ozone concentration measurement is important for air quality status for the environmental and public health warnings. Variations of ground level ozone (GLO) and total column ozone (TCO) concentrations over Klang Valley, Malaysia in 2014 and 2015 were determined in this study. The GLO measurement data was acquired from Department of Environment (DOE) Malaysia and is measured using standard method instrument, UV Absorption Ozone Analyzer Model 400A while, TCO measurement using Ozone Monitoring Instrument (OMI). The analysis suggests that GLO concentrations in all four stations in Klang Valley exceeded the Malaysia Ambient Air Quality Standard (NMAAQS) 2020 prescribed limit of 90 ppb per hour. In both 2014 and 2015, TCO concentration observed higher from May to August with the highest concentration of 290 DU and lowest from November to February with the lowest concentration of 230 DU, suggesting that there is seasonal fluctuation caused by Northeast Monsoon (NEM) and Southwest Monsoon (SWM). This is due to environmental factors during SWM that favour a rise in TCO concentration, such as low precipitation, low relative humidity, high temperature, and prolonged sunshine hours. On the other hand, NEM results in lower TCO concentration due to high precipitation, high relative humidity, cold temperature, and short sunshine hours. Meanwhile, GLO peaks are usually seen towards the end of March through April and in October, according to GLO trend research (40 to 50 ppb). This is during the transitional monsoon season, where the west part of Peninsular Malaysia received precipitation lower than usual.