Comprehensive health risk assessment of urban ambient air pollution (PM2.5, NO2 and O3) in Ghana Boansi Adu Ababio a,e,*, Gerheart Winfred Ashong a, Thomas Peprah Agyekum b, Blessed Adjei Yeboah c , Marian Asantewah Nkansah a, Jonathan Nartey Hogarh d, Michael Kweku Commeh e, Edward Ebow Kwaansa-Ansah a, Kwabena Dabie f, Felix Adulley f, Eldad Boansi e, Lorenda Sarbeng g, Birago Adu Ababio h, Maame Serwaa Boapea i, Nana Kwabena Oduro Darko j, Meshach Kojo Appiah k a Department of Chemistry, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana b Department of Occupational & Environmental Health & Safety, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana c Department of Computer Science, Kumasi Technical University, Kumasi, Ghana d Department of Environmental Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana e Technology Consultancy Centre International Centre for Innovation, Manufacturing, Technology Transfer and Entrepreneurship, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana f Department of Biochemistry, University of Cape Coast, Cape Coast, Ghana g Department of Geography and Regional Planning, University of Cape Coast, Cape Coast, Ghana h Department of Biomedical Sciences, University of Health and Allied Sciences, Ho, Ghana i Department of Virology, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon, Ghana j Department of Biochemistry and Biotechnology, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana k Department of Basic Education, University of Education, Winneba, Ghana A R T I C L E I N F O Edited by Professor Bing Yan Keywords: Air pollution Health risk Air quality Ghana Particulate matter Nitrogen dioxide Ozone A B S T R A C T Urbanization and industrialization have drastically increased ambient air pollution in urban areas globally from vehicle emissions, solid fuel combustion and industrial activities leading to some of the worst air quality con- ditions. Air pollution in Ghana causes approximately 28,000 premature deaths and disabilities annually, ranking as a leading cause of mortality and disability-adjusted life years. This study evaluated the annual concentrations of PM2.5, NO2 and O3 in the ambient air of 57 cities in Ghana for two decades using historical and forecasted data from satellite measurements. The study assessed urban air quality and evaluated both carcinogenic and non- carcinogenic health risks associated with human exposure to ambient air pollutants. Alarmingly, our findings revealed the yearly median PM2.5 concentrations (50.79–67.97 µg m− 3) to be significantly higher than the WHO recommendation of 5 µg m− 3. Tropospheric ozone concentrations (72.21–92.58 µg m− 3 ) also exceeded the WHO annual standard of 60 µg m− 3. Furthermore, NO2 concentrations (3.65–12.15 µg m− 3 ) surpassed the WHO threshold of 10 µg/m³ in multiple cities. Hazard indices indicated that PM2.5 and O3 pose significant non- carcinogenic health risks for younger age groups for a daily exposure duration of three hours and beyond. Ac- cording to the Air Quality Life Index (AQLI) in our study, exposure to PM2.5 shortens life expectancy by 4.5–6.2 years. The ambient air of the majority (98 %) of the cities was unhealthy for sensitive groups. This study reveals the urgent need for comprehensive air quality policies in Ghanaian cities. It emphasizes the significance of robust real-time monitoring of air pollutants and the investigation of seasonal dust storm effects, to fill data gaps in Ghana and West Africa, facilitating evidence-based interventions that improve urban air quality and public health outcomes. * Corresponding author at: Department of Chemistry, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. E-mail address: boansiaduababio@gmail.com (B.A. Ababio). Contents lists available at ScienceDirect Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv https://doi.org/10.1016/j.ecoenv.2024.117591 Received 2 August 2024; Received in revised form 16 December 2024; Accepted 19 December 2024 Ecotoxicology and Environmental Safety 289 (2025) 117591 Available online 7 January 2025 0147-6513/© 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by- nc-nd/4.0/ ). https://orcid.org/0009-0005-3133-9271 https://orcid.org/0009-0005-3133-9271 mailto:boansiaduababio@gmail.com www.sciencedirect.com/science/journal/01476513 https://www.elsevier.com/locate/ecoenv https://doi.org/10.1016/j.ecoenv.2024.117591 https://doi.org/10.1016/j.ecoenv.2024.117591 http://crossmark.crossref.org/dialog/?doi=10.1016/j.ecoenv.2024.117591&domain=pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ http://creativecommons.org/licenses/by-nc-nd/4.0/ 1. Introduction Air is a mixture of gases and tiny dust particles. It is a basic element of life, as a source of oxygen needed for respiration. It is a microbial habitat and a superhighway for the water cycle (Ababio, 2023). Air pollution is caused by human activities or natural processes that intro- duce harmful substances into the air, posing risks to the health and well-being of humans and endangering the ecosystem at large. Ambient air pollution is a major health risk in the 21st century. It is responsible for millions of annual premature mortalities across the globe in urban and rural environments (Li et al., 2023). Half of the world’s population resides in urban cities where they are exposed to gaseous and particulate pollutants (Fan et al., 2020). It is projected that by 2050 approximately 70 % of the global population will live in urban areas (Sicard et al., 2023). Nitrogen dioxide (NO2), particulates with aerodynamic sizes of 2.5 µm (PM2.5) and tropospheric ozone (O3) are among the most harmful air pollutants to public health in urban settings (Sicard et al., 2020; Cakaj et al., 2023). Millions of people are exposed to the concentrations of these pollutants at levels beyond the WHO-recommended thresholds (Anenberg et al., 2022; Sicard et al., 2021; Southerland et al., 2022). Chronic exposure to these pollutants results in increased risks of ischaemic heart disease, stroke, acute lower respiratory infections, chronic obstructive pulmonary diseases, acute kidney injury, dementia, cognitive decline, liver and lung cancer (Kumar et al., 2023; Wang et al., 2023; Ma et al., 2023; 2024). Urbanization and industrialization have led to a significant increase in ambient air pollution levels in urban cities across the globe making them have some of the worst air quality on the planet. Anthropogenic sources of air pollution comprise vehicle emissions, the combustion of solid fuels and domestic waste as well as industrial, agricultural, con- struction and mining activities (Nyarku et al., 2021; Okedere et al., 2021). The Global Burden of Disease estimates ambient air pollution to be a primary contributor to global mortality and the loss of disability-adjusted life years, with 4 million premature deaths annually linked to it (Johnson et al., 2024). Air pollution is the second-highest risk factor in Ghana, accounting for about 28,000 premature deaths and disabilities per annum (Ababio et al., 2023). The World Bank esti- mates that the annual cost of air pollution in Ghana is 4.2 % of GDP, which is equivalent to US $2.5 billion. Also, it costs approximately US $264 million per annum alone for the largest urban cities, in Ghana, such as Accra and Kumasi (World Bank Group, 2020). Various kinds of studies including monitoring and measurement studies, epidemiological investigations, source apportionment analyses, air quality modeling, policy and regulatory assessments, control and mitigation studies have been conducted on air pollution in the ambient environment across the globe. Despite these numerous studies, the state of ambient air pollution in urban cities in Sub-Saharan Africa specifically Ghanaian cities remains understudied. This is due to factors such as the lack of air quality monitoring instruments, unenforced national emis- sions standards, limited funding for research initiatives, insufficient infrastructure for data collection and analysis, and disparities in the prioritization of environmental health. Notable among recent studies on ambient air pollution in Ghana include a survey on the knowledge, perception and awareness of air pollution by Odonkor and Mahami (2020). The study reported that most of its respondents who were residents of the capital city, Accra, knew about air pollution and its recurring impacts on their health. An air quality study of 130 locations in the Greater Accra Metropolis by Wang et al. (2024) predicted annual, harmattan and non-harmattan mean NO2 levels for the metropolis to be 37, 50, and 28 µg m− 3 respectively. The study identified road traffic as the primary source of NO2 emissions with the annual levels in the city exceeding the World Health Organization (WHO) guideline of 10 µg m− 3. A PM2.5 monitoring study by Amegah et al. (2022) in ten different traffic hotspots in Accra during the dry and wet seasons reported median levels ranging from 27.08 – 51.37 and 20.04 – 48.92 µg m− 3 respectively. An increase of 1 µg m− 3 in PM2.5 exposure was associated with slight increases in respiratory and car- diovascular, and symptoms among street traders. Nyadanu et al. (2022) recorded an annual PM2.5 average of 59.97 µg m− 3 with an increase of 10 µg m− 3 in the annual average linked with a 3 % risk of stillbirth. Existing studies on ambient air pollution in Ghana have focused on the capital city, Accra, making the air quality status of other cities un- known (Arku et al., 2008, 2015; Armah et al., 2010; Ofosu et al., 2012; Rooney et al., 2012; Dionisio et al., 2010; Odonkor and Mahami 2020; Kanhai et al., 2021; Amegah et al., 2022; Alli et al., 2023; Wang et al., 2024). Expanding the evaluation of ambient air pollution beyond Accra is crucial for a comprehensive assessment of air quality across Ghana. Without up-to-date information on the trend of ambient air pollutants for different urban cities in Ghana over the past two decades, policy- makers will face challenges in identifying priority areas for intervention and evaluating the efficacy of existing measures. This study conducts a comprehensive health risk assessment of ambient air pollutants in urban cities across Ghana from the year 2000–2023. It evaluates urban air quality and health risks of long-term exposure to PM2.5, O3 and NO2 in the urban population of Ghana. Furthermore, this study is crucial for addressing health inequalities as it sheds light on how vulnerable populations such as the elderly and children are affected by urban air quality. It provides essential infor- mation to aid the tailoring of nation-specific public health interventions. To the best of our knowledge, this study is the first to comprehensively assess the ambient air quality of urban cities across different regions of Ghana. This study seeks to inform policymakers, the scientific commu- nity and public health officials about the status of urban air pollution in Ghana. This study not only fills a critical gap on urban air quality in Ghana but serves as a benchmark for facilitating informed decision- making processes, equipping stakeholders with the knowledge needed to develop effective evidence-based strategies to improve air quality and safeguard the public health of urban populations in the country 2. Methodology 2.1. Study area Ghana is situated in Sub-Saharan Africa, along the West African coast, bordered by Côte d′Ivoire to the west, Burkina Faso to the north, Togo to the east, and the Gulf of Guinea to the south. Its diverse land- scape includes coastal plains, tropical rainforests, savannahs, and the Volta River Basin. It covers a land area of approximately 240,000 km2 with a population of about 30 million people (Agodzo et al., 2023). A majority of this population (57 %) reside in urban areas with almost half of this urban population living in the Accra and Kumasi Metropolis (Ghana Statistical Service, 2021; Iddrisu et al., 2023). The urban cities involved in this study are illustrated in Fig. 1. 2.2. Data This study used air quality observation data for 57 urban cities in different regions of Ghana from 2000 to 2019. The data were obtained from https://urbanairquality.online/ which provides a downloadable dataset of air pollutant concentrations in all cities worldwide, as described below. The obtained historical data from 2000 to 2019 for the 57 urban cities were used to estimate forecast values for 2020–2023 by employing time series forecasting in Excel with a 95 % confidence in- terval using the Exponential Smoothing (ETS) method. The retrieved data included annual concentrations and population attributable frac- tion (%) (PAF) for PM2.5, O3 and NO2 in 57 Ghanaian urban cities. The obtained PM2.5 urban data were from a higher spatial resolution dataset (1 km x 1 km) which integrated information from satellite- retrieved aerosol optical depth, chemical transport modeling, and ground monitor data incorporating geographically weighted regression. Southerland et al. (2022) provide additional details on the method used to generate the PM2.5 dataset. The obtained O3 data were from combined B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 2 https://urbanairquality.online/ ground measurement data with chemical transport model estimates, downscaled to create finer resolution ozone concentration estimates from 1990 to 2017 which were further extrapolated to 2019 based on log-linear trends from 2008 to 2017 and re-gridded to a spatial resolu- tion of 1 km to align with population estimates. Further information regarding the method utilized to generate the O3 dataset can be found in Malashock et al. (2022). The obtained NO2 data were from a resolution of approximately 1 km2, calculated in 5-year increments between 1990 and 2010, and annually from 2010 to 2019. Additional details about the methodology used in generating the NO2 dataset are provided by Anenberg et al., (2022). The concentrations in parts per billion (ppb) were converted to μg/ m3 using 1 ppb NO2 = 1.88 μg/m3 and 1 ppb O3 = 1.96 μg/m3 (Shen et al., 2023). In addition, the population attributable fraction (%) (PAF) for PM2.5, O3 and NO2 per annum also obtained from https://urbanai rquality.online were used to estimate the mortality, disability-adjusted life years (DALY), years of healthy life lost due to disability (YLD), years of life lost from mortality (YLL) associated with ambient urban air pollution in Ghana. These estimates were made for Chronic Respiratory Diseases (CRD) and also specifically for Chronic obstructive pulmonary disease (COPD), pneumoconiosis, asthma and Interstitial Lung disease and pulmonary sarcoidosis (ILDS). The national burden of chronic res- piratory diseases used for the estimates was retrieved from the Global Burden of Disease Study in 2019 (Momtazmanesh et al., 2023). Fig. 1. Map of Ghana showing the 57 urban cities in the study. B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 3 https://urbanairquality.online https://urbanairquality.online 2.3. Ambient air quality Ambient air quality is the extent to which the ambient air is free of pollution and healthy for breathing. This can be assessed using a stan- dard numerical rating known as the air quality index (AQI) (Table S1) (Ababio et al., 2024). AQI is estimated from the concentrations of air pollutants that are considered criteria for air quality. The sub-air quality ratings of the ambient air pollutants (PM2.5, O3 and NO2) were used to compute the air quality index. The worst sub-air quality indices were used to adjudge the overall air quality index for each city and year in this study. The sub-air quality indices for each air pollutant were calculated using Eq. (1) (Abdul Raheem et al., 2022). IP = (CP − BPLo)x I Hi − I Lo BP Hi − BP Lo + ILo (1) where IP: Index value for the air pollutant CP: Concentration of air pollutant BP Lo: Lower Breakpoint value of CP BP Hi: Higher Breakpoint value of CP I Hi: Index Breakpoint value of BP Hi I Lo: Index Breakpoint value of BP Lo The breakpoint values for the concentrations of the air pollutants are provided in Table S2. The six classifications of air quality index have with their respective colour codes and ranges which are provided in Table S1. 2.4. Dominant pollution types of PM2.5 and O3 The respective annual concentration thresholds [PM2.5 = 5 µg m− 3, O3 = 60 µg m− 3] were used to categorize the dominant pollution types of PM2.5 and O3 into four which are; Compound Pollution of PM2.5 and O3 (P-O), PM2.5 Dominant Pollution, O3 Dominant Pollution, and Clean. The categorization criteria which were modified from He et al. (2024) are given in Table S3. 2.5. Analysis of synergistic changes in compound pollution The relative rate of change (ROC) of PM2.5 and O3 concentrations in 2000 and 2023 was used to quantify the degree of synergistic changes in pollution levels for the urban cities using Eq. 2 and Eq 3. (He et al., 2024). ROCi = Ci, 2023 Ci, 2000 (2) If = ⎧ ⎪⎪⎨ ⎪⎪⎩ ROCi,PM2.5 ≥1 and ROCi,O3 ≥1 Synergistic Increase ROCi,PM2.5 <1 and ROCi,O3 <1 Synergistic Decrease ROCi,PM2.5 ≥1 and ROCi,O3 <1 PM2.5 Increase and O3 Decrease ROCi,PM2.5 <1 and ROCi,O3 ≤1 PM2.5 Decrease and O3 Increase (3) 2.6. Health risk assessment 2.6.1. Daily dose Daily Dose (DD) is the chronic inhalation of an ambient air pollutant daily. DD was calculated using Eq. (4) which was modified from (Ababio et al., 2023). The parameter values for different age groups used for evaluating the corresponding health risks for 1 h, 3 h, 6 h, 9 h and 12 h of exposure to the air pollutants have been provided in Table 1. DD = C x R x T x F x D BWxAT (4) where C: Concentration of PM2.5 or NO2 or O3 R: rate of inhalation (m3 hr− 1) T: time of exposure F: frequency of exposure (365 d yr− 1) D: duration of exposure BW: average body weight AT: average time (duration of exposure x frequency of exposure) The parameter values used for estimating the daily dose in the study were adopted from the exposure factor handbook of the United States Environmental Protection Agency (2011). The employed values are provided in Table 1. 2.6.2. Non-carcinogenic risk As a ratio between the daily dose and reference dose, the Hazard Quotient (HQ) reveals how the ambient air pollutant compares to its reference dose. The hazard quotient and Hazard index (HI) are in- dicators of non-carcinogenic health risks calculated using Eq. 5 and Eq. 6 (Ababio et al., 2023). HI predicts the sum of the individual HQ for the criteria pollutants through an exposure route to the reference dose (RfD) for each pollutant (Hogarh et al., 2018). An HI/HQ < 1 and HI/HQ > 1 imply negligible non-cancer health risk and a higher chronic non-cancer risk respectively. HQ = DD RfD (5) HI = HQPM2.5 + HQNO2 + HQO3 (6) The reference dose (RfD) determined using Eq. (7) is the concen- tration at which daily exposure to a pollutant such as PM2.5, O3, and NO2 does not cause adverse health effects over a lifetime (Ababio et al., 2023). RfD = RfC x IR BW (7) Where RfC: Reference concentration (WHO annual guideline values PM2.5 = 5 µg m− 3, O3 = 60 µg m− 3 and NO2 = 10 µg m− 3 (World Health Organization, 2021) IR: Inhalation rate (m3 d− 1) BW: average body weight 2.6.3. Carcinogenic risk for PM2.5 exposure The carcinogenic risk which is the incremental chance of developing cancer from exposure to ambient PM2.5 in urban cities was calculated using Eq. (8). The accepted range for carcinogenic risk is 10− 4 to 10− 6, as risks < 10− 6 are considered negligible and > 10− 4 are considered highly unacceptable and a cause for alarm. CR = DD x SF (8) The slope factor (SF) was used to estimate the likelihood of devel- oping cancer from exposure to a carcinogenic pollutant such as ambient PM2.5. SF was calculated using Eq. (9) for adults, adolescents, children, toddlers and infants (Mbazima, 2022). SF = IUR BWxIR (9) Where SF: Slope Factor IUR: Inhalation Unit Risk (0.008) IR: Inhalation rate BW: average body weight Table 1 Parameter values for estimating the daily dose in different age groups. Age Group Inhalation rate (m3 d¡1) Exposure duration (yrs) Body weight (kg) Infant 5 1 7 Toddler 9 4 12 Children 12 12 23 Adolescents 16 19 62 Adults 20 70 70 B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 4 2.6.4. Relative and excess risk 2.6.4.1. Relative risk. Relative risk (RR) is a measure of the degree of correlation between exposure to a particular pollutant and the likeli- hood of developing health outcomes such as a disease or mortality. RR was obtained using Eq. (10) (Sharma et al., 2023). RR = exp [β (C − Co)] (10) Where β: the exposure-response coefficient is the corresponding additional health risk per unit increase in an air pollutant beyond its threshold concentration. β = 0.038 %, 0.13 % and 0.048 % per 1 µg m− 3 increase of PM2.5, NO2 and O3 respectively. (Shen et al., 2020) C: Concentration of PM2.5 or NO2 or O3 Co: the threshold concentration (WHO annual guideline values) 2.6.4.2. Excess risk. Excess Risk (ER) is the difference in risk between exposed and unexposed individuals. It was used to estimate the addi- tional risk of health outcomes or mortalities due to exposure to the pollutant. ER was computed using Eq. (11) (Sharma et al., 2023) ER = RR – 1 (11) Where RR: Relative Risk 2.6.5. Air quality life index The Air Quality Life Index (AQLI) estimates the impact of air pollu- tion on life expectancy (Greenstone and Fan, 2018). It measures the potential gains in life expectancy per person (Life Years Saved) if PM2.5 concentrations meet WHO or national air quality standards. It is based on findings that sustained exposure to an additional 10 µg m− 3 of PM2.5 above the recommended threshold reduces life expectancy by 0.98 years (Chen et al., 2013; Ebenstein et al., 2017; Greenstone and Fan, 2018; Rana, 2022). It was calculated using Eq. (12) Life Years Saved = ( C − S 10 ) x 0.98 (12) Where C = the annual concentration of PM2.5 in µg m− 3 S = PM2.5 WHO annual guideline value = 5 µg m− 3 2.6.6. Mortality, Disability-Adjusted Life Years (DALY), Years of Healthy Life Lost due to Disability (YLD) and Years of Life Lost from Mortality (YLL) estimates for all age groups The estimates for the deaths, DALY, YLD and YLL associated with ambient concentrations of PM2.5, NO2 and O3 were made to assess the impact on respiratory health burden in Ghana. Using the annual median population attributable fraction (%) (PAF) data spanning from 2000 to 2023, the adjusted deaths, DALYs, YLLs, and YLDs for CRD and specific respiratory diseases (COPD, pneumoconiosis, asthma, and ILSD) were calculated accordingly using Eqs. (13), (14), (15), (16), (17) and (18). The mortality, DALY, YLL and YLD values for Ghana from the Global Burden of Disease Study in 2019 (Momtazmanesh et al., 2023) used for the estimation are provided in Table S4. IncidencePM2.5 or NO2 or O3 = Mortalitydisease x PAFPM2.5 or NO2 or O3 (13) PrevalencePM2.5 or NO2 or O3 = Mortalitydisease x PAFPM2.5 or NO2 or O3 (14) MortalityPM2.5 or NO2 or O3 = Mortalitydisease x PAFPM2.5 or NO2 or O3 (15) DALYPM2.5 or NO2 or O3 = Mortalitydisease x PAFPM2.5 or NO2 or O3 (16) YLLPM2.5 or NO2 or O3 = Mortalitydisease x PAFPM2.5 or NO2 or O3 (17) YLDPM2.5 or NO2 or O3 = Mortalitydisease x PAFPM2.5 or NO2 or O3 (18) 2.7. Statistical analysis The retrieved data underwent statistical analysis using the Stata and Microsoft Excel. Quantile regression was employed in this study to analyze the relationship between ambient air pollution in urban cities. It was used in this study due to the skewed distribution exhibited by the obtained data. It was used to estimate the median and other quan- tiles of ambient air pollutant concentrations. It allowed for the com- parison of the quantile estimates across urban cities. It made it possible to assess the differences in pollutant concentrations for the different years. Additionally, regression coefficients (β) were also estimated from quantile regressions. The p-values associated with the regression coefficients were also generated using quantile regression which indicated the statistical sig- nificance of the differences in pollutant levels between the cities. Moreover, the Kendall Tau (KT) test was also used to assess trends in the time series data of the air pollutants. The Kendall’s tau coefficients (τₐ and τᵦ) and the associated p-value were estimated. Kendall’s tau coefficients were used to measure the association between the variables "year" and "pollutant concentration", with positive values indicating an increasing trend, negative values indicating a decreasing trend, and values close to zero indicating no significant trend. The associated p-value provided a measure of the statistical significance of the observed trend. In addition, Sen’s slope was also employed to analyze trends in pollutant concentrations over time using the estimated slope coefficient, p-value which determined the significance level, and the confidence interval which assessed the pre- cision and reliability of the estimated trend. Moreover, Principal Component Analysis (PCA) was used to analyze air pollutant patterns in the various urban cities compressing the dataset into components to reveal the relationship between pollutants and their spatial distribution. 3. Results and discussion 3.1. Ambient air pollutant concentrations, population attributable fraction (%), epidemiological impacts and air quality indices of the urban cities The annual concentrations of PM2.5, NO2, and O3 for different urban cities in Ghana from 2000 to 2023 are reported as median and quartiles in Table 2. The corresponding p-values, regression coefficients, and confidence intervals for the annual concentrations of PM2.5, NO2, and O3 using the capital city, Accra as the reference are provided in Tables S1 and S2. The annual PM2.5, and O3 concentrations for each city in the study are provided in Fig. S1a – Fig. S1e and that of NO2 in Fig. S2a – Fig. S2e. The annual median PM2.5 concentrations of the urban cities spanned from 50.79 to 67.97 µg m− 3. These were five to six-fold higher than the WHO annual PM2.5 air quality guideline value (5 µg m− 3). Spatial dis- tribution revealed the urban populace in Akatsi and Jirapa situated in Southeastern and Northwestern Ghana to be exposed to the highest and lowest median annual PM2.5 concentrations respectively (Fig. 2a). It was observed that cities such as Accra, Koforidua, and Kumasi in the Southern regions of Ghana consistently exhibited higher annual median PM2.5 concentrations compared to lower levels in cities in the Northern belt such as Wa, Tamale, and Nalerigu. The PM2.5 findings of this study were similar to the findings of Moro et al. (2022) as Savannah belt cities had the lowest PM2.5 concentrations, followed by Forest belt and Coastal belt cities. The inter-urban comparison of PM2.5 concen- trations using the capital city, Accra, as the reference indicated 20 of the urban cities in the study to have positive regression coefficients (β) (0.31 – 7.80) (Table S5). These revealed residents of these cities to be exposed to higher levels of PM2.5 as compared to urban dwellers in the capital. These might be attributed to anthropogenic factors such as higher reli- ance on traditional biomass cooking sources, lack of green spaces, untarred roads, and combustion of waste coupled with meteorological B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 5 conditions such as higher temperatures and low wind speeds that could have exacerbated PM2.5 pollution in these cities. Negative regression coefficients (β) (-0.40 to − 10.40) revealed the majority (36) of residents in the urban cities in this study to be exposed to lower levels of PM2.5 as compared to the capital. These might have been due to lower population densities, limited industrial activities and vehicular traffic. Furthermore, fine particulate matter (PM2.5) concentrations in the urban cities except Jirapa and Wa were not significantly different from the capital with p > 0.05 (Table S5). The AQLI estimates for gains in life expectancy if ambient PM2.5 concentrations per annum in the urban cities met WHO guidelines are given in Fig. 3 and Table S6. The AQLI revealed that ambient PM2.5 pollution in urban cities cut life expectancy short by 4.5–6.2 years relative to if concentrations in the cities met the WHO threshold. This revealed the detrimental impact of PM2.5 pollution on the life expec- tancy of habitants in Ghanaian urban cities. The relative rate of changes for P-O compound pollution and the dominant pollution classifications from the ambient concentrations of PM2.5 and O3 in the urban cities are provided in Table S7. The ambient environment of all the urban cities in the study except Cape Coast and Sekondi Takoradi exhibited a PM2.5-O3 dominant compound pollution as concentrations of PM2.5 > 5 µg m− 3 and O3 > 31 ppb. In the absence of corresponding ozone concentrations, the ambient environs of Cape Coast and Sekondi-Takoradi were classified as PM2.5 dominated pollu- tion as PM2.5 > 5 µg m− 3. The relative rate of changes revealed PM2.5-O3 compound pollution in Sekondi-Takoradi and Cape Coast to have syn- ergistic changes of PM2.5 decrease. The observed level of synergistic changes for ambient PM2.5-O3 pollution in Akim Oda and Mampong was synergistic decreases in P-O compound pollution. In contrast, the syn- ergistic increase in P-O compound pollution was observed for Nkawkaw for the study’s duration. The remaining 54 urban cities exhibited syn- ergistic changes of PM2.5 decrease and O3 increase for P-O compound pollution from 2000 to 2023. The findings for urban ambient environ- ments in this study was similar to a comparative study by He et al. (2024) which observed P-O compound pollution in majority of global cities, predominantly in urban Asian cities. The median tropospheric ozone concentrations in the urban cities varied from 72.21 – 92.58 µg m− 3 which exceeded the WHO annual guideline equivalent recommendation of 60 µg m− 3. The highest and lowest median concentrations of ozone were found in Tatale and the capital, Accra, situated in the Northeastern and Southern parts of the country respectively (Fig. 2b). Inter-urban comparison indicated cities in Southern Ghana had lower median concentrations of ground-level ozone as compared to higher concentrations in Northern Ghana cities. Furthermore, a com- parison of the ozone concentrations in reference to the capital city only gave positive regression coefficients (β) (0.36 – 22.26) (Table S5). This suggested that the urban population of the other cities in the study were exposed to higher concentrations of ground-level ozone than in the capital. This could have been due to factors like cleaner energy initia- tives and favourable meteorological conditions for dispersion and degradation that led to the lesser emission of ozone precursors such as nitrogen oxides (NOx) and volatile organic compounds (VOCs) in Accra. Tropospheric ozone concentrations in 33 urban cities were significantly different from that of the capital with p < 0.05 while those of 21 cities were not with p > 0.05 (Table S5). Median NO2 concentrations of the urban cities ranged from 3.65 – 12.15 µg m− 3. The highest and lowest were observed in Garu and Cape Coast accordingly. Few of the urban cities had nitrogen dioxide con- centrations exceeding the WHO guideline value of 10 µg m− 3. Urban cities situated within the Upper East and Northern Regions of Ghana recorded the highest NO2 concentrations whiles cities in the Central Region of Ghana recorded the lowest (Fig. 2.c). Positive regression coefficients (β) (0.30 – 0.94) revealed the resi- dents of Bawku, Garu. and Nalerigu to be exposed to concentrations of NO2 above that of the capital. Furthermore p > 0.05 indicated the NO2 concentrations in Bolgatanga, Bawku, and Wa to be significantly different from that of the capital. The median PM2.5 concentrations per annum varied from 46.69 – 82.52 µg m− 3 (Fig. 4A). These were 5 – 8 times above the WHO rec- ommended threshold. The highest was observed in 2000 while the lowest was observed in 2014. The PM2.5 concentrations exhibited fluc- tuations across the 23-year duration. The years 2000 – 2001, 2002 – 2003, 2004 – 2006, 2008 – 2009, and 2012 – 2013 exhibited significant decreases in PM2.5 concentrations while significant increases were observed for the years, 2001 – 2002, 2003 – 2004, 2006 – 2007, 2014 – 2015, and 2019 – 2020. In comparison with the year 2000, negative Table 2 Concentrations of PM2.5, NO2 and O3 in Ghanaian Urban Cities. City PM2.5 (µg m¡3) O3 (µg m¡3) NO2 (µg m¡3) Median IQR Median IQR Median IQR Accra 61.43 13.53 72.21 24.09 11.06 0.95 Agona Swedru 62.98 15.93 73.04 24.41 6.25 0.48 Akatsi 67.97 11.98 78.48 24.99 6.96 0.53 Akim Oda 62.33 19.08 80.09 24.34 6.65 0.34 Asamankese 63.42 16.18 80.59 24.33 4.94 0.35 Assin Fosu 64.80 20.33 74.81 23.30 6.14 0.67 Atebubu 61.93 11.77 89.37 22.98 5.97 1.18 Bawku 58.22 12.47 85.94 16.93 11.43 0.92 Berekum 55.72 13.67 83.99 22.94 7.87 0.76 Bimbila 61.17 9.11 92.01 19.06 8.41 0.93 Bole 56.17 8.90 88.79 15.13 7.91 2.79 Bolgatanga 56.34 9.31 86.43 17.19 11.08 0.95 Cape Coast 60.27 16.95 nd Nd 3.65 0.28 Dambai 64.56 10.96 90.15 21.42 7.34 1.26 Dormaa Ahenkro 57.64 17.35 80.81 22.96 6.82 0.95 Dunkwa Offin 60.55 17.83 78.23 22.45 5.50 0.57 Ejura 60.29 11.11 87.62 23.87 7.27 0.87 Enchi 59.43 16.26 72.99 21.65 4.69 0.48 Garu 58.07 13.52 87.47 17.55 12.15 0.84 Gushiegu 56.28 8.86 90.88 18.06 10.01 0.70 Ho 63.58 8.00 83.50 26.12 5.88 2.02 Hohoe 64.24 11.20 87.52 24.84 5.31 0.50 Jirapa 50.79 6.87 85.42 16.42 9.08 1.42 Karaga 54.49 10.73 90.59 18.06 9.76 1.84 Kete Krachi 61.91 10.80 90.08 23.62 6.54 1.02 Kintampo 58.24 12.90 88.10 19.60 5.94 1.05 Koforidua 61.14 13.31 81.10 26.05 7.05 0.83 Konongo 59.93 15.82 82.33 24.56 7.00 1.08 Kumasi 63.15 18.65 83.42 24.24 10.27 0.32 Kwame Danso 62.22 12.73 88.77 23.09 6.16 2.61 Mampong 58.81 11.31 85.81 23.88 6.27 0.61 Nalerigu 54.80 7.66 87.49 17.55 11.58 0.40 Navrongo 56.48 9.86 85.65 16.83 9.82 0.67 New Tafo Akim 60.03 13.66 82.55 26.03 6.46 0.45 Nkawkaw 60.43 12.79 83.48 25.47 6.49 0.71 Nkoranza 57.76 12.81 86.33 21.96 5.84 0.86 Nsawam 62.34 15.32 82.35 24.12 8.05 0.47 Obuasi 61.43 18.83 77.51 23.89 7.48 0.56 Paga 55.52 9.27 85.09 16.72 8.86 0.71 Salaga 62.03 9.01 89.40 19.41 7.15 1.78 Sandema 55.83 11.15 85.89 16.83 8.89 1.83 Savelugu 55.60 9.74 90.33 17.81 9.08 1.51 Sawla 55.93 9.48 89.16 15.32 7.52 2.64 Somanya 60.51 11.03 81.45 25.73 6.07 0.60 Suhum 62.09 16.81 80.42 26.06 5.61 0.52 Sunyani 57.26 13.10 83.95 23.36 9.11 1.47 Sekondi-Takoradi 59.27 15.04 nd Nd 4.66 0.25 Tamale 57.21 11.88 91.43 18.28 10.45 1.42 Tatale 57.71 9.92 92.58 17.45 8.21 1.51 Techiman 57.18 12.55 86.31 21.96 7.35 0.42 Wa 51.57 8.70 86.78 15.87 10.62 2.19 Walewale 55.34 9.80 88.27 17.58 10.17 1.14 Winneba 63.73 16.18 73.58 22.28 5.83 0.41 Wulensi 63.58 10.28 91.23 19.21 7.97 1.67 Yeji 61.65 9.41 89.86 21.19 7.29 1.82 Yendi 55.77 11.38 92.21 18.00 9.72 1.46 Zebilla 56.50 10.92 86.51 17.50 9.80 0.68 IQR means Interquartile range; nd means no data B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 6 quantile regression coefficients (β) (-4.90 to − 35.83) showed significant reductions in PM2.5 concentrations in the urban cities per annum (Table S6). The highest and lowest reductions in annual PM2.5 concen- trations were observed in the years 2000 and 2014 respectively. The median PM2.5 concentrations in the urban cities per annum were significantly different as p < 0.001 (Table S8). Negative Kendall tau coefficients (τ) (τₐ= − 0.1354 and τᵦ = − 0.1383) and a p-value < 0.0001 were obtained from the KT test which revealed a weak significant negative correlation between years and PM2.5 concentrations (Table S9). Also, negative slope coefficient β (-0.14), a p-value < 0.001 and 95 % CI [-0.17, − 0.11] were obtained from the Sen’s slope analysis (Table S10). These indicated a significant decreasing trend in PM2.5 concentrations over the observed years. This could have been due to the transition from biomass combustion sources to cleaner energies such as the increased adoption of liquified petroleum gas for cooking in urban households. Increased greening initiatives such as tree planting programs and the tarring of unpaved roads and streets in urban municipalities reducing the resuspension of dust. Moreover, stringent air quality policies Fig. 2. (a). Spatial Distribution of PM2.5 Concentrations in Urban Cities in Ghana (b) Spatial Distribution of O3 Concentrations in Urban Cities in Ghana (c) Spatial Distribution of NO2 Concentrations in Urban Cities in Ghana. Fig. 3. Life Years Saved per Annum if PM2.5 Concentrations met Recommended Threshold. B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 7 curtailing particulate emissions from industrial activities, together with improved urban waste management reducing the open burning of waste might have accounted for the observed PM2.5 decline. Median ground- level ozone concentrations per year ranged from 64.36 – 105.37 µg m− 3 which were in exceedance of the WHO guideline limit (Fig. 2A). The highest and lowest were recorded for the years 2021 and 2003 respectively. A trend of fluctuations and plateaus was observed for ozone concentrations across the 23-year duration. Noticeable plateaus in ozone concentrations were observed for the years, 2000 – 2001, 2006 – 2007, 2016 – 2017 and 2020 – 2021. Relatively rapid increases in ozone occurred from 2004 – 2006, 2009 – 2015. Notable decreases were observed from 2001 – 2004, 2007 – 2009 and 2015 – 2016. Further- more, in the last few years (2021 – 2023), there has been a significant decrease in ozone levels following the peak in 2021. The variability in ground-level ozone levels may have been due to the complex interplay of factors such as fluctuating weather conditions, including temperature, humidity, and wind speed, which might have influenced the formation and dispersion of ozone. Also, anthropogenic activities such as industrial processes and transportation emissions might have contributed to ozone formation through the emission of precursor pollutants. The KT test indicated a strong positive correlation between years and O3 concen- tration, with Kendall’s tau coefficients (τ) of 0.6097 (τₐ) and 0.6229 (τᵦ). In addition, a p-value < 0.0001 indicated a significant increasing trend in tropospheric O3 concentration over the 23 years (Table S9). In addition, a positive slope coefficient β (0.61), a p-value < 0.001 and 95 % CI [0.59, 0.63] were obtained from the Sen’s slope analysis (Table S10). These confirmed a consistently increasing trend in O3 concentrations over the years in urban cities. This could have been due to the continuous decline in Nitrogen Oxides (NOx) accounting for an increase in the ratio of Volatile Organic Compounds (VOCs) to NOx. This might have favoured ozone formation in a NOx-limited regime through the suppression of ozone titration, increased radical activities and peroxide formation over nitric acid formation. In addition, higher tem- peratures increasing ozone production by rising emissions of precursors such as VOCs alongside higher solar radiation intensities in the study area might have accounted for the observed increase in tropospheric ozone. Median NO2 concentrations per year in the urban cities spanned from 6.41 – 8.18 µg m− 3 below the WHO threshold (Fig. 4B). The lowest and highest concentrations were recorded in the years 2023 and 2005 Fig. 4. (a) Annual Median Concentrations of PM2.5 and O3 in Ghanaian Urban Cities (b) Annual Median Concentrations of NO2 in Ghanaian Urban Cities. B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 8 respectively. A general trend of decreasing NO2 concentrations was observed with variabilities in NO2 concentrations from year to year as some years showed significant increases (2000 – 2005, 2010 – 2011, 2014 – 2015) or decreases (2005 – 2010, 2017 – 2019) compared to the preceding years, indicating fluctuations in NO2 levels. The general declining trend in urban NO2 concentrations appears to have stabilized in the most recent years (2021–2023), indicating that the trend may have slowed down or plateaued. Kendall’s tau coefficients (τ) [-0.1192 (τₐ) and − 0.1218 (τᵦ)] and a p-value < 0.0001 indicated a weak nega- tive correlation between years and NO2 concentrations (Table S9). Furthermore, a negative slope coefficient β (-0.12), a p-value < 0.001 and 95 % CI [-0.15, − 0.08] were obtained from the Sen’s slope analysis (Table S10). These suggested a significant decreasing trend in NO2 concentration over the years. This could be attributed to improved regulations of vehicular emissions and industrial processes through the adoption of cleaner energy alternatives, usage of catalytic converters in vehicles, improved public transit systems coupled with meteorological conditions such as dispersing wind patterns might have led to the decreasing trend in ambient urban NO2 concentrations. Comparing the annual concentrations of PM2.5 in Ghanaian urban cities in this study to those in other global regions revealed notable differences (Table 3). Research by Tariq and colleagues (2023), and Mandal et al. (2020) reported higher concentrations in Niger and India respectively. The findings of Liu et al. (2023), Xu and Zhang (2020) and Rahman et al. (2021) reported a broader range of annual concentrations in urban Asian cities. In contrast, Park et al. (2020) reported a smaller range of annual concentrations in Seoul. Furthermore, Li et al. (2021) observed lower concentrations in Iraq and Kuwait. Azhari et al. (2021) also reported lower concentration ranges in Malaysia. Moreover, a local study by Bahino et al. (2024) reported notably lower concentrations in Accra, Ghana than this study. The comparison of tropospheric ozone concentrations in the study with other studies in different geographical locations also revealed sig- nificant variability. The research conducted by He et al. (2023) reported wider concentration ranges compared to this study. The reported annual ranges in this study were above the reported levels of McHugh et al. (2023) and Xu et al. (2019). Higher and ground O3 concentration ranges than this study’s findings were observed by Huang et al. (2018) and Lee et al. (2023) in China. The annual nitrogen dioxide findings of this study were lower than the reported concentrations from related studies in France and China (Jurado et al., 2020; Sanyal et al., 2018; Kuerban et al., 2020). The differences in concentrations observed in this study could be due to its longer duration compared to related studies such as Wang et al. (2024) which had a shorter timeframe. Principal Component Analysis (PCA) was employed in this study to determine spatial variability of air pollutants and the influence of each pollutant on principal components using three-factor loading categories; strong factors (> 0.75), moderate (0.50 – 0.75), and weak factors (< 0.50) (Ashong et al., 2024). The PCA revealed three principal compo- nents (total variance); PC1 (62.02 %), PC2 (22.72 %) and PC3 (15.26 %) in the study (Fig. 5). The first principal component (PC1) represented the most dominant air pollution pattern in the urban cities revealing an inverse relationship between PM2.5 (- 0.61) and the gaseous pollutants; NO2 (0.60) and O3 (0.52). This could have been due to the prevalence of particulate emission sources (such as the combustion of solid fuels and dust from construction and mining activities) over gaseous pollutant sources. Also, finer PM2.5 providing surfaces for heterogeneous reactions might have favoured the scavenging of gaseous pollutants in the atmo- sphere. The second principal component (PC2) characterized by a strong positive loading of O3 (0.85) and a negative loading of NO2 (- 0.45) indicated an inverse relationship between the two gaseous pollutants. This could be due to a non-linear photochemical ozone formation with NO2 as a precursor in urban environs where higher solar radiation en- hances ozone formation with lower NO2 reducing ozone titration. This suggests PC2 represents photochemical processes governing ozone for- mation and depletion in the troposphere. The third principal component (PC3) showed strong positive loadings for both PM2.5 (0.74) and NO2 (0.67) with minimal influence from O3 (0.1). This component likely represents localized pollution patterns and specific pollution sources affecting both PM2.5 and NO2 concentrations simultaneously. The co-variation of PM2.5 and NO2 in PC3 could be due to common pollution sources such as high traffic areas and industrial zones in urban cities. The dominant role of PC1 suggests that air pollution mitigation strategies targeting PM2.5 reduction may not proportionally affect NO2 and O3. Necessitating multi-pollutant approaches to addressing urban air pollution in Ghana. Furthermore, the distinct behaviour of O3 in PC2 highlights the need for interventions tailored to target ozone formation particularly in urban areas prone to photochemical smog. In addition, PC3 suggests the possibility of some targeted interventions mitigating both PM2.5 and NO2 simultaneously in certain urban areas. The PCA biplot revealed southern urban cities exhibited elevated concentrations of PM2.which could be attributed to higher vehicular traffic, industrial processes and combustion activities in southern Ghana. In contrast, northern urban cities had higher O3 concentrations likely due to photochemical reactions involving NO2 and VOC precursors under favourable conditions such as higher solar radiations in northern Ghana. The population attributable fraction (%) (PAF) for air pollution is an epidemiologic measure which estimates the proportion of disease cases in a population that can be attributed to exposure to a specific air Table 3 Comparison of concentrations of ambient urban air pollutants with comparative studies. Reference Location Pollutant Study Duration Annual Concentrations This Study Ghana PM2.5 2000 – 2023 46.69 – 82.52 µg m− 3 Tariq et al. (2023) Niger 1998 – 2019 68.85 – 70.47 µg m− 3 Liu et al. (2023) China 2013 – 2021 6.5 – 94.2 µg m− 3 Xu and Zhang (2020) Beijing, China 2013 – 2018 5 – 470 µg m− 3 Park et al. (2020) Seoul, South Korea 2014 – 2015 41 – 46.7 µg m− 3 Mandal et al. (2020) Delhi, India 2010 – 2016 87 – 138 µg m− 3 Li et al. (2021) Iraq 2001 – 2018 33 – 44 µg m− 3 Kuwait 36 – 49 µg m− 3 Azhari et al. (2021) Kuala Lumpur, Malaysia 2019 30.4 – 43.7 µg m− 3 Rahman et al. (2021) Dhaka, Bangladesh 2016 5.27 – 105 µg m− 3 Bahino et al. (2024) Accra, Ghana 2020 – 2021 21.1 – 24.8 µg m− 3 This Study Ghana O3 2000 – 2023 64.36 – 105.37 µg m− 3 He et al. (2023) China 2013 – 2018 1.84 – 160 µg m− 3 McHugh et al. (2023) Ireland 2010 – 2019 39 – 43 µg m− 3 Lee et al. (2023) California, USA 2005 – 2017 94.08 – 103.88 µg m− 3 Huang et al. (2018) China 2013 – 2017 141.0 – 163.5 µg m− 3 Xu et al. (2019) Windsor, Canada 1996 – 2015 39.79 – 52.92 µg m− 3 This Study Ghana NO2 2000 – 2023 6.41 – 8.18 µg m− 3 Jurado et al. (2020) France 2013 – 2017 5 – 95 µg m− 3 Kuerban et al. (2020) China 2015 – 2018 5.9–64.4 µg m− 3 Sanyal et al. (2018) France 1999 – 2000 4.55 – 46.96 µg m− 3 Wang et al. (2022) Accra, Ghana 2019 – 2020 70 µg m− 3 Wang et al. (2024) Accra, Ghana 2019 – 2020 37 µg m− 3 B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 9 pollutant. Thus, it represents the percentage of disease burden in the population that could be avoided if exposure to the air pollutant were mitigated. The median PAF for annual air pollutant exposure in the urban cities is given in Table S11 and S12.The median PAF associated with ambient exposure to PM2.5, O3, and NO2 in the urban cities varied from 14.25 – 34.61 %, 1.94 – 8.28 %, and 1.14 – 11.72 % respectively. The highest annual PAF for PM2.5, O3, and NO2 exposure were observed for the years 2000, 2020 and 2005 respectively (Table S12). The Mann- Kendall test was used to examine the relationship between years and PAFs for PM2.5, O3, and NO2. The obtained Kendall’s tau coefficients (τ) [PM2.5PAF = -0.1326 (τₐ) and − 0.1366 (τᵦ); NO2PAF = -0.0804 (τₐ) and − 0.0821 (τᵦ)] revealed weak negative correlations for PM2.5PAF and NO2PAF with years (Table S12). Additionally, p-values < 0.0001 showed a significant decreasing trend in PM2.5PAF and NO2PAF over the years (Table S13). This suggested existing interventions in Ghana to tackle PM2.5PAF and NO2PAF were effective. In contrast, positive Ken- dall’s tau coefficients (τ) [-0.1326 (τₐ) and − 0.1366 (τᵦ)] a p-value < 0.0001 were obtained for O3PAF which indicated a significant increasing trend over the years. This indicated the need for targeted strategies to address O3PAF in Ghana. The epidemiological contribution of PM2.5, NO2, and O3 to in- cidences, prevalences, mortalities, DALYs, YLLs and YLDs in Ghana were estimated for chronic respiratory diseases in general and specifically for COPD, pneumoconiosis, asthma, and ILSD. PM2.5 had the highest inci- dence of respiratory diseases. CRD had the most cases at 63190, fol- lowed by asthma with 43296 cases. COPD and ILDS also recorded high incidences of 7019 and 12847 cases, respectively. PCN had the lowest incidence among the diseases, with only 28 cases attributed to PM2.5 exposure. Exposure to ground-level ozone also resulted in the incidence of respiratory diseases albeit with a reduced impact compared to PM2.5. CRD had the highest incidence (12667) followed by asthma (8679). COPD and ILDS had lower incidences of 1407 and 2575 cases, respec- tively. PCN registered the lowest incidence, with only 6 cases linked to ambient O3 exposure. Nitrogen dioxide had the lowest incidence among the three air pollutants as CRD had the highest incidence at 9239 cases followed by asthma with 6331 cases. COPD and ILDS had lower in- cidences of 1026 and 1878 cases, respectively. PCN has the lowest incidence, with only 4 cases attributed to NO2 exposure (Table S14). The estimated prevalence of CRD for annual PM2.5 exposure was found to be 329880, which is significantly higher than the prevalence of COPD (115965), PCN (59), asthma (245644), and ILDS (1736). The prevalence of CRD due to O3 exposure was estimated to be 66127, while that for NO2 is 48233. The prevalence from ambient urban O3 had estimates of 23246 for COPD, 12 for PCN, 43950 for asthma, and 348 for ILDS. NO2 had the lowest prevalence among the three pollutants with prevalence of 16956 for COPD, 9 for PCN, 32057 for asthma, and 254 for ILDS (Table S15). The estimated annual CRD mortalities varied from 1734 – 2185, 45 – 704, and 233 – 344 for ambient exposure to PM2.5, O3, and NO2 respectively. The average annual estimated deaths for COPD, pneumo- coniosis, asthma, and ILSD from ambient PM2.5 were 1079, 1, 696 and 42 respectively. Tropospheric ozone was estimated to account for 391, 216, 140 and 8 annual average mortalities in Ghana from CRD, COPD, asthma, and ILSD respectively. Furthermore, the ambient NO2 in the study was found to result in 285, 158, 102 and 6 annual average mor- talities in Ghana from CRD, COPD, asthma, and ILSD respectively. Es- timates from the study indicated no mortalities for pneumoconiosis from O3 and NO2 ambient urban exposure in Ghana (Table S16). The annual DALY estimates revealed substantial impacts of the urban air pollutants on CRD, COPD, PCN, Asthma, and ILDS. It was observed that PM2.5 consistently had the highest annual DALY average, generally for CRD (78653) and particularly for COPD (37933). These showed it Fig. 5. Principal component analysis biplot of air pollutants in Ghanaian urban cities. B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 10 significantly exacerbated CRDs in comparison with the other pollutants in Ghana. Annual average DALYs of 11500 (NO2) and 15767 (O3) were observed for CRD. In addition, DALYs linked to COPs were 7604 and 5546 for O3 and NO2 respectively (Table S17). The obtained annual average YLL from CRD, COPD, PCN asthma, and ILSD ranged from 34 – 53938, 7 – 10812, 5 – 7887, for PM2.5, O3, and NO2 ambient exposure respectively. The highest and lowest YLLs were observed for CRD and PCN respectively. The highest YLLs for all the diseases were consistently associated with ambient PM2.5 in the urban cities (Table S18). The highest and lowest annual YLD averages were observed for CRD (24715) and particularly for COPD (13456) respec- tively. The YLD estimates from the study indicated PM2.5 had the most substantial impact on YLD, with notably high YLDs for CPD (24715), COPD (13456), asthma (8664) and ILDS (197). In comparison, O3 recorded lower YLDs for CRD (4954), COPD (2697), asthma (1737), and ILDS (40). On the other hand, NO2 registered the lowest YLDs for CRD (3614), COPD (1967), asthma (1267) and ILDS (29) among the three pollutants. The YLD estimates for pneumoconiosis associated with exposure to the three pollutants were < 10 (Table S19). The epidemiological contribution of PM2.5, O3, and NO2 to CRD in Ghana revealed a significant public health burden with PM2.5 consis- tently showing the highest impact across all health outcomes. The clear hierarchy of impact for incidence, prevalence, mortality, DALY, YLD and YLL in Ghana was PM2.5 > O3 > NO2. Notably, CRD bore the highest brunt of air pollution effects while PCN had minimal impact. These suggest the need for stricter PM2.5 targeted reduction policies and the adequate resourcing of the Ghanaian health care system to improve the early detection, management and treatment of CRDs to reduce mortalities. The ambient air pollutant concentrations in the urban cities were indexed using the corresponding estimated sub-indices (Q) for each air pollutant in the cities. The air quality index of the towns was obtained from the worst sub-indices. The estimated air quality indices and sub-air quality indices are provided in Table 4. The PM2.5 sub-air quality indices (121.25 – 152.43) rated the ambient concentrations in 56 urban cities as moderately good for human exposure while that of Akatsi was unhealthy. The sub-indices of NO2 (2.34 – 6.09) and tropospheric O3 (34.11 – 43.74) revealed the con- centrations in all the urban cities were good for breathing. Overall, the air quality indices of 56 urban cities were found to be unhealthy for sensitive groups and that of Akatsi was unhealthy for all groups of people. The PM2.5 sub-air quality indices per annum rated its ambient pollution in urban cities as moderately good for breathing during the years 2001–2003, 2005–2006, 2009–2014, and 2018–2023. However, PM2.5 sub-air quality indices per annum rated its pollution of the ambient air in the urban cities during the years 2000, 2004, 2007–2008, and 2015–2017 as unhealthy for breathing. Furthermore, NO2 and O3 sub-air quality indices per year indicated their pollution in the urban cities during the years 2000–2023 to be good for breathing. The years 2000, 2004, 2007 – 2008, and 2015 – 2017 had air quality indices indicating the ambient air of Ghanaian urban cities was unhealthy for breathing while that of the other years in this study’s duration was unhealthy for sensitive groups (Table S20). 3.2. Carcinogenic, non-carcinogenic, relative and excess health risk assessments of ambient air pollutant concentrations in urban cities The Hazard Quotients (HQ) for one-hour exposure in the urban cities were < 1 for all demographics, ranging from 0.49 – 0.50 for PM2.5, 0.058 – 0.059 for O3 and 0.032 – 0.033 for NO2. The hazard indices (HI) [0.58 – 0.59] for the different age groups were below the threshold of 1, denoting no significant non-carcinogenic risks from one hour of expo- sure to the air pollutants. HQ values ranging from 1.47 – 1.49 for PM2.5, 0.096 – 0.099 for NO2 and 0.175 – 0.177 for O3 were obtained for three hours of exposure across all age groups. Non-carcinogenic HI of 1.73–1.75 indicated significant risks for three-hour exposure to ambient air pollution in the cities. For six hours of exposure to PM2.5 HQ values of 2.94–2.97 were observed whereas HQ values of 0.19 – 0.20 and 0.350 – 0.353 were obtained for NO2 and O3 respectively. Non-carcinogenic HI ranging from 1.18 – 3.51 revealed significant non-cancer risks for six- hour exposure. Nine-hour exposure had a non-carcinogenic HQ of 4.41 – 4.46 for PM2.5, 0.288 – 0.295 NO2 and 0.52 – 0.53 for O3 expo- sure. Non-carcinogenic HI of 5.20–5.26 was obtained for nine-hour exposure to the air pollutants. Worst-case scenario assessments gave HQ values of 5.88 – 5.94 for half-day and 11.76 – 11.88 for full-day adult ambient exposure to PM2.5. The HQ values for O3 and NO2 half-day exposure were < 1. HQ values > 1 (1.39 – 1.41) were obtained for full-day ambient exposure to O3 while the HQ values of 0.77 – 0.79 were obtained for NO2. Significant non-carcinogenic deleterious risks with HI ranging from 6.94 – 7.01 for half-day and 13.96 – 14.03 for full-day exposure were observed in this study for all the different age groups (Table S21). Slight differences were observed between the different age groups regarding potential non-carcinogenic risks to ambient air pollutants. The non-carcinogenic risks from both PM2.5 and NO2 exposure in the urban cities could be ranked in descending order as Toddlers > Children > Infants > Adults > Adolescents while that for O3 was Children >Toddlers > Infants > Adults > Adolescents. This suggested that younger age groups such as children and toddlers are more susceptible to adverse non-cancer effects from urban air pollution in Ghana. The individual contribution of the air pollutants to non-cancer risks in this study could be ranked in descending order as PM2.5 > O3 > NO2. The findings of the study suggest that traders, auto-artisans, farmers, ma- sons, traffic wardens, street hawkers and other outdoor workers who quotidianly spend three to nine hours in ambient environs performing their occupational activities in urban cities in the country are more vulnerable to non-cancer health risks. Furthermore, the exposure of sensitive groups (such as people with pre-existing lung and cardiovas- cular diseases) for three hours and beyond could lead to the exacerba- tion of pre-existing health conditions. The carcinogenic risks (CR) for human exposure to PM2.5 concen- trations in urban cities ranged from 2.82E-06–6.76E-05 for adults, 3.95E-06–9.48E-05 for adolescents, 1.44E-05–3.44E-04 for children, 1.13E-04–2.72E-03 for infants and 3.67E-05–8.80E-04 for toddlers. These indicated acceptable carcinogenic risks for adults, adolescents, children and toddlers as the CR were within the tolerable carcinogenic risk limit of 1E-04–1E-06. Contrarily, the CR for infants’ exposure to ambient PM2.5 concentrations in urban cities indicated significant carcinogenic risks CR for an exposure time of nine hours and beyond (Table S22). The estimated relative and excess risks obtained for the concentra- tions of ambient air pollutants in the urban cities are provided in Table S23. The relative risks for PM2.5 and O3 in the 57 urban cities in the study were > 1 and the respective excess risks were positive. These both suggested concentrations of PM2.5 and O3 in the ambient air of the cities posed significant health risks to the dwellers thereof. Except for Wa, Walewale, Tamale, Navrongo, Garu, Gushiegu, Bolgatanga, Bawku, Accra, Kumasi and Nalerigu the relative risks for NO2 < 1 and the cor- responding excess risks were negative. These both indicated ambient concentrations of NO2 in most of the urban cities posed no significant health risks. The excess risks for PM2.5, O3 and NO2 in the urban cities varied from 0.0176 – 0.0242, 0.0059–0.0158 and (-) 0.0082 to (+) 0.0028 respectively. Furthermore, the relative risks for PM2.5 and O3 for annual concen- trations per annum were > 1 which indicated they posed significant risks to the health of urban habitants in Ghana from 2000 – 2023. In contrast, the relative risks for concentrations of NO2 per year were < 1 suggesting no significant health risks from its exposure. The excess risks for PM2.5, O3 and NO2 per annum varied from 0.0160 to 0.0299, 0.0030–0.0222 and (-) 0.0024 to (-) 0.0046 respectively (Table S24). B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 11 Table 4 Classification of air quality index of ghanaian urban cities. City QPM2.5 QO3 QNO2 AQI Accra 142.19 34.11 5.56 Unhealthy for Sensitive Groups Agona Swedru 145.24 34.50 3.14 Unhealthy for Sensitive Groups Akatsi 152.43 37.07 3.49 Unhealthy Akim Oda 143.96 37.83 3.34 Unhealthy for Sensitive Groups Asamankese 146.10 38.07 2.48 Unhealthy for Sensitive Groups Assin Fosu 148.82 35.34 3.08 Unhealthy for Sensitive Groups Atebubu 143.17 42.21 3.00 Unhealthy for Sensitive Groups Bawku 135.87 40.59 5.74 Unhealthy for Sensitive Groups Berekum 130.95 39.68 3.95 Unhealthy for Sensitive Groups Bimbila 141.68 43.46 4.23 Unhealthy for Sensitive Groups Bole 131.84 41.94 3.97 Unhealthy for Sensitive Groups Bolgatanga 132.17 40.83 5.57 Unhealthy for Sensitive Groups Cape Coast 139.90 - 1.83 Unhealthy for Sensitive Groups Dambai 148.35 42.59 3.69 Unhealthy for Sensitive Groups Dormaa Ahenkro 134.73 38.18 3.42 Unhealthy for Sensitive Groups Dunkwa Offin 140.46 36.95 2.76 Unhealthy for Sensitive Groups Ejura 139.94 41.39 3.65 Unhealthy for Sensitive Groups Enchi 138.25 34.48 2.36 Unhealthy for Sensitive Groups Garu 135.58 41.32 6.09 Unhealthy for Sensitive Groups Gushiegu 132.05 42.94 5.03 Unhealthy for Sensitive Groups Ho 146.42 39.44 2.95 Unhealthy for Sensitive Groups Hohoe 147.72 41.34 2.66 Unhealthy for Sensitive Groups Jirapa 121.25 40.35 4.56 Unhealthy for Sensitive Groups Karaga 128.53 42.79 4.90 Unhealthy for Sensitive Groups Kete Krachi 143.13 42.56 3.28 Unhealthy for Sensitive Groups Kintampo 135.91 41.62 2.98 Unhealthy for Sensitive Groups Koforidua 141.62 38.31 3.54 Unhealthy for Sensitive Groups Konongo 139.24 38.89 3.51 Unhealthy for Sensitive Groups Kumasi 145.57 39.41 5.16 Unhealthy for Sensitive Groups Kwame Danso 143.74 41.94 3.09 Unhealthy for Sensitive Groups Mampong 137.03 40.54 3.15 Unhealthy for Sensitive Groups Nalerigu 129.14 41.33 5.81 Unhealthy for Sensitive Groups Navrongo 132.45 40.47 4.92 Unhealthy for Sensitive Groups New Tafo Akim 139.43 39.00 3.24 Unhealthy for Sensitive Groups Nkawkaw 140.22 39.44 3.25 Unhealthy for Sensitive Groups Nkoranza 134.97 40.78 2.93 Unhealthy for Sensitive Groups Nsawam 143.98 38.91 4.04 Unhealthy for Sensitive Groups Obuasi 142.19 36.61 3.75 Unhealthy for Sensitive Groups Paga 130.56 40.19 4.44 Unhealthy for Sensitive Groups Salaga 143.37 42.23 3.59 Unhealthy for Sensitive Groups Sandema 131.17 40.57 4.46 Unhealthy for Sensitive Groups Savelugu 130.71 42.68 4.56 Unhealthy for Sensitive Groups Sawla 131.36 42.12 3.77 Unhealthy for Sensitive Groups Somanya 140.38 38.47 3.05 Unhealthy for Sensitive Groups Suhum 143.49 37.99 2.82 Unhealthy for Sensitive Groups Sunyani 133.98 39.66 4.58 Unhealthy for Sensitive Groups Sekondi-Takoradi 137.94 - 2.34 Unhealthy for Sensitive Groups Tamale 133.88 43.19 5.25 Unhealthy for Sensitive Groups Tatale 134.87 43.74 4.12 Unhealthy for Sensitive Groups Techiman 133.82 40.77 3.69 Unhealthy for Sensitive Groups Wa 122.78 40.99 5.33 Unhealthy for Sensitive Groups Walewale 130.20 41.69 5.10 Unhealthy for Sensitive Groups Winneba 146.71 34.76 2.92 Unhealthy for Sensitive Groups Wulensi 146.42 43.10 4.00 Unhealthy for Sensitive Groups Yeji 142.62 42.45 3.66 Unhealthy for Sensitive Groups Yendi 131.05 43.56 4.88 Unhealthy for Sensitive Groups Zebilla 132.49 40.86 4.92 Unhealthy for Sensitive Groups Where Q = sub-air quality index, AQI = Air Quality Index, Green Colour = Good, Orange Colour = Unhealthy for Sensitive Groups and Red Colour = Unhealthy, B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 12 3.3. Limitations A notable limitation of this study is that the impact of seasonal regional dust storms on the concentrations of pollutants for the study’s duration was not assessed. This is due to the lack of ground and satellite harmattan and non-harmattan pollutant measurements in the urban cities. Furthermore, the lack of data on VOCs concentrations limited the study from assessing the ratio of VOC to NOx to shed further light on potential sources of ground-level ozone in the cities for the two decades. Moreover, the concentrations of pollutants in the study are based on satellite measurements, and the dearth of ground measurements of the pollutants for the study’s duration prevented the study from comparing these satellite measurements with annual ground-monitored concen- trations. In addition, the forecasting of air pollution in the study did not account for the impact of the COVID-19 lockdowns on air pollution trends in the country. 3.4. Policy suggestions Efficient ground monitoring and reporting of real-time urban air quality along with strict enforcement of air quality standards as outlined by the WHO and Ghana’s EPA are needed to safeguard the public’s health. Rapid transition to renewable energy, promotion of electric ve- hicles, improved public transportation and stricter emissions controls and incentives are essential globally, including in Ghana. Effective air quality management plans should be implemented in Ghana’s urban cities. These should include reducing emissions from transportation, industry, and waste burning. They should also promote clean cooking technologies, enhance transportation infrastructure and encourage ecologically conscious development. Ghana must strengthen its regulatory framework by imposing stringent emission limits on ve- hicles, industry and other significant polluters. Enhancing the capacity of regulatory bodies to oversee, document and enforce adherence to air quality guidelines is crucial. Local governments and communities should receive training and resources to create effective air quality management plans, alongside international collaboration for sharing best practices and technical support. Ghana should foster cooperation between educational in- stitutions, public and private organizations, and businesses to develop innovative solutions for regional air pollution focusing on health im- pacts. National budgets, foreign donations and private sector in- vestments should fund initiatives to improve air quality, along with financial incentives for adopting greener practices and technologies. Public health campaigns are needed nationwide to increase the aware- ness of residents of urban cities on air pollution and its significant impact on their health. 4. Conclusion The development in urbanization, population, and industrialization has significantly increased ambient air pollution in urban cities both locally and globally causing some of the most serious air quality chal- lenges. The comprehensive health risk evaluation of ambient air pollu- tion in urban areas in Ghanaian cities from 2000 to 2023 revealed significant findings. The findings showed that with significant regional variations, the yearly median PM2.5 concentrations being the most detrimental overall for chronic respiratory diseases such as asthma, were five to six times higher than WHO recommendations. Compared to northern cities, Accra, Koforidua, and Kumasi consistently showed higher PM2.5 levels. Ozone and NO2 concentrations also varied and periodically surpassed recommended thresholds. Most urban cities had a dominant PM2.5-O3 pollution profile, which suggests significant health risks. Severe health risks are highlighted by high median population attributable fractions (PAF) for air pollution exposure especially for PM2.5. Hazard indices from PM2.5 exposure through inhalation indicated significant non-cancer health risks for an exposure duration of three hours and beyond per day. Carcinogenic risks for human exposure to urban PM2.5 concentrations were mostly within the acceptable limit of 1E-04–1E-06. The air quality indices of 56 urban cities were found to be unhealthy for sensitive groups and that of Akatsi was unhealthy for all groups of people. Urban air quality indices show moderately good ambient pollution from 2001 to 2003, 2005–2006, 2009–2014, and 2018–2023, with unhealthy pollution from 2000 to 2017, and good sub-air quality from 2000 to 2023. Relative risks > 1 for PM2.5 and O3 confirmed substantial health risks for urban residents with PM2.5 pollution contributing to a reduction in life expectancy by 4.5–6.2 years. Regarding the severe adverse health effects of PM2.5 and O3 pollu- tion, this study highlights the urgent need for focused actions to enhance air quality in Ghanaian metropolitan areas. For an in-depth awareness of air pollution dynamics in Ghana and West Africa, future studies should prioritize bridging data gaps, including the effects of periodic dust storms and volatile organic compounds levels. CRediT authorship contribution statement Marian Asantewah Nkansah: Writing – review & editing, Valida- tion, Supervision, Project administration. Nana Kwabena Oduro Darko: Writing – review & editing, Validation. Blessed Adjei Yeboah: Writing – review & editing, Software, Investigation, Data curation. Maame Serwaa Boapea: Writing – review & editing, Validation, Investigation. Thomas Peprah Agyekum: Writing – review & editing, Supervision, Project administration, Formal analysis, Data curation, Conceptualization. Meshach Kojo Appiah: Writing – review & editing, Visualization, Formal analysis. Birago Adu Ababio: Writing – review & editing, Visualization, Formal analysis. Gerheart Winfred Ashong: Writing – review & editing, Validation, Methodology, Formal analysis. Lorenda Sarbeng: Writing – review & editing, Visualization, Software. Boansi Adu Ababio: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Eldad Boansi: Writing – review & editing, Validation, Conceptualization. Felix Adulley: Writing – review & editing, Validation, Investigation. Kwabena Dabie: Writing – review & editing, Validation, Methodology. Edward Ebow Kwaansa-Ansah: Writing – review & editing, Validation, Supervision, Project adminis- tration. Michael Kweku Commeh: Writing – review & editing, Super- vision, Project administration. Jonathan Nartey Hogarh: Writing – review & editing, Supervision, Project administration. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements We thank the Milken Institute of Public Health for making their data available. We also thank Susan Anenberg for her assistance and valuable feedback during this work. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ecoenv.2024.117591. Data availability Data will be made available on request. B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 13 https://doi.org/10.1016/j.ecoenv.2024.117591 References Ababio, B.A., 2023. Evaluation of indoor air quality of senior high school kitchens in the Kumasi metropolis (Masters Dissertation). Ababio, B.A., Nkansah, M.A., Hogarh, J.N., Agyekum, T.P., Commeh, M.K., 2023. In- kitchen particulate matter emissions in high schools in the Kumasi metropolis, Ghana: levels and the health risk assessment. J. Hazard. Mater. Adv., 100358 Ababio, B.A., Nkansah, M.A., Hogarh, J.N., Agyekum, T.P., Commeh, M.K., 2024. Gaseous air quality and health risk assessment of high school kitchens in the Kumasi Metropolis. Environ. Adv. 17, 100576. Abdul Raheem, M., Jimoh, G., Abdulrahim, H., 2022. Assessment of kitchen air pollution: Health implications for the residents of ilorin south, nigeria. J. Environ. Public Health 2022 (1), 7689141. Agodzo, S.K., Bessah, E., Nyatuame, M., 2023. A review of the water resources of Ghana in a changing climate and anthropogenic stresses. Front. Water 4, 973825. Alli, A.S., Clark, S.N., Wang, J., Bennett, J., Hughes, A.F., Ezzati, M., Brauer, M., Nimo, J., Bedford-Moses, J., Baah, S., Cavanaugh, A., 2023. High-resolution patterns and inequalities in ambient fine particle mass (PM2. 5) and black carbon (BC) in the Greater Accra Metropolis, Ghana. Sci. Total Environ. 875, 162582. Amegah, A.K., Dakuu, G., Mudu, P., Jaakkola, J.J., 2022. Particulate matter pollution at traffic hotspots of Accra, Ghana: levels, exposure experiences of street traders, and associated respiratory and cardiovascular symptoms. J. Expo. Sci. Environ. Epidemiol. 32 (2), 333–342. Anenberg, S.C., Mohegh, A., Goldberg, D.L., Kerr, G.H., Brauer, M., Burkart, K., Hystad, P., Larkin, A., Wozniak, S., Lamsal, L., 2022b. Long-term trends in urban NO2 concentrations and associated paediatric asthma incidence: estimates from global datasets. Lancet Planet. Health 6 (1), e49–e58. Arku, R.E., Dionisio, K.L., Hughes, A.F., Vallarino, J., Spengler, J.D., Castro, M.C., Agyei- Mensah, S., Ezzati, M., 2015. Personal particulate matter exposures and locations of students in four neighborhoods in Accra, Ghana. J. Expo. Sci. Environ. Epidemiol. 25 (6), 557–566. Arku, R.E., Vallarino, J., Dionisio, K.L., Willis, R., Choi, H., Wilson, J.G., Hemphill, C., Agyei-Mensah, S., Spengler, J.D., Ezzati, M., 2008. Characterizing air pollution in two low-income neighborhoods in Accra, Ghana. Sci. Total Environ. 402 (2-3), 217–231. Armah, F.A., Yawson, D.O., Pappoe, A.A., 2010. A systems dynamics approach to explore traffic congestion and air pollution link in the city of Accra, Ghana. Sustainability 2 (1), 252–265. Ashong, G.W., Ababio, B.A., Kwaansa-Ansah, E.E., Gyabeng, E., Nti, S.O., 2024. Human and ecotoxicological risk assessment of heavy metals in polymer post treatment sludge from Barekese Drinking Water Treatment Plant, Kumasi. Toxicol. Rep. 12, 404–413. Azhari, A., Halim, N.D.A., Mohtar, A.A.A., Aiyub, K., Latif, M.T., Ketzel, M., 2021. Evaluation and prediction of PM10 and PM2. 5 from road source emissions in Kuala Lumpur City Centre. Sustainability 13 (10), 5402. Bahino, J., Giordano, M., Beekmann, M., Yoboue, V., Ochou, A., Galy-Lacaux, C., Liousse, C., Hughes, A., Nimo, J., Lemmouchi, F., Cuesta, J., 2024. Temporal variability and regional influences of PM 2.5 in the West African cities of Abidjan (Côte d′Ivoire) and Accra (Ghana). Environ. Sci.: Atmos. 4 (4), 468–487. Cakaj, A., Lisiak-Zielińska, M., Khaniabadi, Y.O., Sicard, P., 2023. Premature deaths related to urban air pollution in Poland. Atmos. Environ. 301, 119723. Chen, Y., Ebenstein, A., Greenstone, M., Li, H., 2013. Evidence on the impact of sustained exposure to air pollution on life expectancy from China’s Huai River policy. Proc. Natl. Acad. Sci. 110 (32), 12936–12941. Dionisio, K.L., Arku, R.E., Hughes, A.F., Vallarino, J., Carmichael, H., Spengler, J.D., Agyei-Mensah, S. and Ezzati, M., 2010. Air pollution in Accra neighborhoods: spatial, socioeconomic, and temporal patterns. Ebenstein, A., Fan, M., Greenstone, M., He, G., Zhou, M., 2017. New evidence on the impact of sustained exposure to air pollution on life expectancy from China’s Huai River Policy. Proc. Natl. Acad. Sci. 114 (39), 10384–10389. Fan, H., Zhao, C., Yang, Y., 2020. A comprehensive analysis of the spatio-temporal variation of urban air pollution in China during 2014–2018. Atmos. Environ. 220, 117066. Ghana, G.S.S., 2021. Population and Housing Census: Population of Regions and Districts. Ghana Statistical Service. Greenstone, M. and Fan, C.Q., 2018. Introducing the air quality life index. AQLI Annual Report. He, C., Liu, J., Zhou, Y., Zhou, J., Zhang, L., Wang, Y., Liu, L., Peng, S., 2024. Synergistic PM2. 5 and O3 control to address the emerging global PM2. 5-O3 compound pollution challenges. Eco-Environ. Health. https://doi.org/10.1016/j. eehl.2024.04.004. He, C., Wu, Q., Li, B., Liu, J., Gong, X., Zhang, L., 2023. Surface ozone pollution in China: trends, exposure risks, and drivers. Front. Public Health 11, 1131753. Hogarh, J.N., Agyekum, T.P., Bempah, C.K., Owusu-Ansah, E.D.J., Avicor, S.W., Awandare, G.A., Fobil, J.N., Obiri-Danso, K., 2018. Environmental health risks and benefits of the use of mosquito coils as malaria prevention and control strategy. Malar. J. 17 (1), 1–12. https://doi.org/10.1186/s12936-018-2412-4. Huang, J., Pan, X., Guo, X., Li, G., 2018. Health impact of China’s air pollution prevention and control action plan: an analysis of national air quality monitoring and mortality data. Lancet Planet. Health 2 (7), e313–e323. Iddrisu, S., Siiba, A., Alhassan, J., Abass, K., 2023. Land-use and land cover change dynamics in urban Ghana: implications for peri-urban livelihoods. Int. J. Urban Sustain. Dev. 15 (1), 80–96. Johnson, M., Mazur, L., Fisher, M., Fraser, W.D., Sun, L., Hystad, P., Gandhi, C.K., 2024. Prenatal exposure to air pollution and respiratory distress in term newborns: results from the MIREC prospective pregnancy cohort. Environ. Health Perspect. 132 (1), 017007. Jurado, X., Reiminger, N., Vazquez, J., Wemmert, C., Dufresne, M., Blond, N., Wertel, J., 2020. Assessment of mean annual NO2 concentration based on a partial dataset. Atmos. Environ. 221, 117087. Kanhai, G., Fobil, J.N., Nartey, B.A., Spadaro, J.V., Mudu, P., 2021. Urban Municipal Solid Waste management: modeling air pollution scenarios and health impacts in the case of Accra, Ghana. Waste Manag. 123, 15–22. Kuerban, M., Waili, Y., Fan, F., Liu, Y., Qin, W., Dore, A.J., Peng, J., Xu, W., Zhang, F., 2020. Spatio-temporal patterns of air pollution in China from 2015 to 2018 and implications for health risks. Environ. Pollut. 258, 113659. Kumar, P., Singh, A.B., Arora, T., Singh, S., Singh, R., 2023. Critical review on emerging health effects associated with the indoor air quality and its sustainable management. Sci. Total Environ. 872, 162163. Lee, H.J., Kuwayama, T., FitzGibbon, M., 2023. Trends of ambient O3 levels associated with O3 precursor gases and meteorology in California: synergies from ground and satellite observations. Remote Sens. Environ. 284, 113358. Li, C., van Donkelaar, A., Hammer, M.S., McDuffie, E.E., Burnett, R.T., Spadaro, J.V., Chatterjee, D., Cohen, A.J., Apte, J.S., Southerland, V.A., Anenberg, S.C., 2023. Reversal of trends in global fine particulate matter air pollution. Nat. Commun. 14 (1), 5349. Li, J., Garshick, E., Hart, J.E., Li, L., Shi, L., Al-Hemoud, A., Huang, S., Koutrakis, P., 2021. Estimation of ambient PM2. 5 in Iraq and Kuwait from 2001 to 2018 using machine learning and remote sensing. Environ. Int. 151, 106445. Liu, X., Yi, G., Zhou, X., Zhang, T., Bie, X., Li, J., Tan, H., 2023. Spatio-temporal variations of PM2. 5 and O3 in China during 2013–2021: impact factor analysis. Environ. Pollut. 334, 122189. Ma, Y.H., Chen, H.S., Liu, C., Feng, Q.S., Feng, L., Zhang, Y.R., Hu, H., Dong, Q., Tan, L., Kan, H.D., Zhang, C., 2023. Association of long-term exposure to ambient air pollution with cognitive decline and Alzheimer’s disease–related amyloidosis. Biol. Psychiatry 93 (9), 780–789. Malashock, D.A., Delang, M.N., Becker, J.S., Serre, M.L., West, J.J., Chang, K.L., Cooper, O.R., Anenberg, S.C., 2022. Global trends in ozone concentration and attributable mortality for urban, peri-urban, and rural areas between 2000 and 2019: a modelling study. Lancet Planet. Health 6 (12), e958–e967. Mandal, S., Madhipatla, K.K., Guttikunda, S., Kloog, I., Prabhakaran, D., Schwartz, J.D., Team, G.H.I., 2020. Ensemble averaging based assessment of spatiotemporal variations in ambient PM2. 5 concentrations over Delhi, India, during 2010–2016. Atmos. Environ. 224, 117309. Mbazima, S.J., 2022. Health risk assessment of particulate matter 2.5 in an academic metallurgy workshop. Indoor Air 32 (9), e13111. McHugh, K., Cummins, T., Aherne, J., 2023. Distribution and long-term trends of tropospheric ozone concentrations in Ireland. Atmosphere 14 (3), 569. Min, J., Kang, D.H., Kang, C., Bell, M.L., Kim, H., Yang, J., Gasparrini, A., Lavigne, E., Hashizume, M., Kim, Y., Ng, C.F.S., 2024. Fluctuating risk of acute kidney injury- related mortality for four weeks after exposure to air pollution: a multi-country time- series study in 6 countries. Environ. Int. 183, 108367. Momtazmanesh, S., Moghaddam, S.S., Ghamari, S.H., Rad, E.M., Rezaei, N., Shobeiri, P., Aali, A., Abbasi-Kangevari, M., Abbasi-Kangevari, Z., Abdelmasseh, M., Abdoun, M., 2023. Global burden of chronic respiratory diseases and risk factors, 1990–2019: an update from the Global Burden of Disease Study 2019. EClinicalMedicine 59. Moro, A., Aborigo, R., Nonterah, E., Seyram, K., Nartey, K. and Soyiri, I., 2022, September. Variation in PM2. 5, PM10 and atmospheric conditions in the three (3) ecological zones in Ghana. In ISEE Conference Abstracts (Vol. 2022, No. 1). Nyadanu, S.D., Tessema, G.A., Mullins, B., Kumi-Boateng, B., Ofosu, A.A., Pereira, G., 2022. Ambient particulate matter air pollution and stillbirth in Ghana: a difference- in-differences approach. Atmos. Pollut. Res. 13 (7), 101471. Nyarku, M., Buonanno, G., Ofosu, F., Jayaratne, R., Mazaheri, M., Morawska, L., 2021. Ultrafine particles in key microenvironments in rural and urban areas of Ghana. Atmos. Pollut. Res. 12 (11), 101212. Odonkor, S.T., Mahami, T., 2020. Knowledge, attitudes, and perceptions of air pollution in Accra, Ghana: a critical survey. J. Environ. Public Health. Ofosu, F.G., Hopke, P.K., Aboh, I.J., Bamford, S.A., 2012. Characterization of fine particulate sources at Ashaiman in Greater Accra, Ghana. Atmos. Pollut. Res. 3 (3), 301–310. Okedere, O.B., Elehinafe, F.B., Oyelami, S., Ayeni, A.O., 2021. Drivers of anthropogenic air emissions in Nigeria-a review. Heliyon 7 (3). Park, E.H., Heo, J., Kim, H., Yi, S.M., 2020. Long term trends of chemical constituents and source contributions of PM2. 5 in Seoul. Chemosphere 251, 126371. Rahman, M.S., Bhuiyan, S.S., Ahmed, Z., Saha, N., Begum, B.A., 2021. Characterization and source apportionment of elemental species in PM2. 5 with especial emphasis on seasonal variation in the capital city “Dhaka”, Bangladesh. Urban Clim. 36, 100804. Rana, S., 2022. Determination of Air Quality Life Index (AQLI) In Medinipur City of West Bengal (India) during 2019 to 2020: a contextual study. Curr. World Environ. 12 (1). Rooney, M.S., Arku, R.E., Dionisio, K.L., Paciorek, C., Friedman, A.B., Carmichael, H., Zhou, Z., Hughes, A.F., Vallarino, J., Agyei-Mensah, S., Spengler, J.D., 2012. Spatial and temporal patterns of particulate matter sources and pollution in four communities in Accra, Ghana. Sci. Total Environ. 435, 107–114. Sanyal, S., Rochereau, T., Maesano, C.N., Com-Ruelle, L., Annesi-Maesano, I., 2018. Long-term effect of outdoor air pollution on mortality and morbidity: a 12-year follow-up study for metropolitan France. Int. J. Environ. Res. Public Health 15 (11), 2487. Sharma, K., Garg, A., Joshi, V., Kumar, A., 2023. Assessment of health risks for criteria air pollutants present in 11 non-attainment cities of Uttar Pradesh, India. Hum. Ecol. Risk Assess.: Int. J. 29 (1), 103–122. B.A. Ababio et al. Ecotoxicology and Environmental Safety 289 (2025) 117591 14 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref1 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref1 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref1 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref2 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref2 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref2 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref3 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref3 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref3 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref4 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref4 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref5 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref5 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref5 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref5 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref6 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref6 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref6 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref6 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref7 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref7 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref7 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref7 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref8 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref8 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref8 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref8 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref9 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref9 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref9 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref9 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref10 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref10 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref10 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref11 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref11 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref11 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref11 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref12 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref12 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref12 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref13 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref13 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref13 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref13 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref14 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref14 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref15 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref15 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref15 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref16 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref16 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref16 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref17 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref17 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref17 https://doi.org/10.1016/j.eehl.2024.04.004 https://doi.org/10.1016/j.eehl.2024.04.004 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref19 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref19 https://doi.org/10.1186/s12936-018-2412-4 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref21 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref21 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref21 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref22 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref22 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref22 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref23 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref23 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref23 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref23 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref24 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref24 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref24 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref25 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref25 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref25 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref26 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref26 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref26 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref27 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref27 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref27 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref28 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref28 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref28 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref29 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref29 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref29 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref29 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref30 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref30 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref30 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref31 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref31 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref31 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref32 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref32 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref32 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref32 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref33 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref33 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref33 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref33 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref34 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref34 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref34 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref34 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref35 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref35 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref36 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref36 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref37 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref37 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref37 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref37 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref38 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref38 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref38 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref38 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref39 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref39 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref39 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref40 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref40 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref40 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref41 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref41 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref42 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref42 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref42 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref43 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref43 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref44 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref44 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref45 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref45 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref45 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref46 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref46 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref47 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref47 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref47 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref47 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref48 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref48 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref48 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref48 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref49 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref49 http://refhub.elsevier.com/S0147-6513(24)01667-1/sbref49 Shen, L.T., Ge, M.W., Hu, F.H., Jia, Y.J., Tang, W., Zhang, W.Q., Zhao, D.Y., Shen, W.Q., Chen, H.L., 2023. The connection between six common air pollution particles and adult brain tumors: a meta-analysis of 26,217,930 individuals. Environ. Sci. Pollut. Res. 30 (50), 108525–108537. Shen, F., Zhang, L., Jiang, L., Tang, M., Gai, X., Chen, M., Ge, X., 2020. Temporal variations of six ambient criteria air pollutants from 2015 to 2018, their spatial distributions, health risks and relationships with socioeconomic factors during 2018 in China. Environ. Int. 137, 105556. Sicard, P., Agathokleous, E., Anenberg, S.C., De Marco, A., Paoletti, E., Calatayud, V., 2023. Trends in urban air pollution over the last two decades: a global perspective. Sci. Total Environ. 858, 160064. Sicard, P., Agathokleous, E., De Marco, A., Paoletti, E., Calatayud, V., 2021. Urban population exposure to air pollution in Europe over the last decades. Environ. Sci. Eur. 33, 28. Sicard, P., Paoletti, E., Agathokleous, E., Araminienė, V., Proietti, C., Coulibaly, F., De Marco, A., 2020. Ozone weekend effect in cities: deep insights for urban air pollution control. Environ. Res. 191, 110193. Southerland, V.A., Brauer, M., Mohegh, A., Hammer, M.S., Van Donkelaar, A., Martin, R. V., Apte, J.S., Anenberg, S.C., 2022. Global urban temporal trends in fine particulate matter (PM2⋅ 5) and attributable health burdens: estimates from global datasets. Lancet Planet. Health 6 (2), e139–e146. Tariq, S., Mariam, A., Mehmood, U., 2023. Assessment of variability in PM2. 5 and its impact on human health in a West African country. Chemosphere 344, 140357. United States Environmental Protection Agency, 2011. Exposure factors handbook: 2011 Edition. Office of Research and Development, Washington, DC. VoPham, T., Jones, R.R., 2023. State of the science on outdoor air pollution exposure and liver cancer risk. Environ. Adv., 100354 Wang, J., Alli, A.S., Clark, S., Hughes, A., Ezzati, M., Beddows, A., Vallarino, J., Nimo, J., Bedford-Moses, J., Baah, S., Owusu, G., 2022. Nitrogen oxides (NO and NO2) pollution in the Accra metropolis: spatiotemporal patterns and the role of meteorology. Sci. Total Environ. 803, 149931. Wang, J., Alli, A.S., Clark, S.N., Ezzati, M., Brauer, M., Hughes, A.F., Nimo, J., Bedford- Moses, J., Baah, S., Nathvani, R. and Agyei-Mensah, S., 2024. Inequalities in urban air pollution in sub-Saharan Africa: An empirical modelling of ambient NO and NO2 concentrations in Accra, Ghana. Environmental Research Letters. Wang, R., Liu, J., Qin, Y., Chen, Z., Li, J., Guo, P., Shan, L., Li, Y., Hao, Y., Jiao, M., Qi, X., 2023. Global attributed burden of death for air pollution: demographic decomposition and birth cohort effect. Sci. Total Environ. 860, 160444. World Bank Group. 2020. Ghana Country Environmental Analysis. Washington, D.C. 〈htt p://documents.worldbank.org/curated/en/419871588578973802/Ghana-Countr y-Environmental-Analysis〉 p. xix. World Health Organization, 2021. . WHO global air quality guidelines: particulate matter (PM2. 5 and PM10), ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide. World Health Organization. Xu, X., Zhang, T., Su, Y., 2019. Temporal variations and trend of ground-level ozone based on long-term m