Air, Volume 2, Issue 4 (December 2024) – 6 articles

Mineral dust from desert areas accounts for a large portion of aerosols globally, estimated at 3–4 billion tons per year. Aerosols emitted from arid and semi-arid areas, e.g., from parched lakes or rivers, are transported over long distances and have effects on a global scale, affecting the planet’s radiative balance, atmospheric chemistry, cloud formation and precipitation, marine biological processes, air quality, and human health. Desert dust transport takes place in the atmosphere as the result of a dynamical sequence beginning with dust uplift from desert areas, then followed by the long-range transport and terminating with the surface deposition of mineral dust in areas even very far from dust sources. The Mediterranean basin is characterized by frequent dust intrusion events, particularly affecting Spain, France, Italy, and Greece. Such events contribute to the increase in PM₁₀ and PM_2.5 concentration values, causing legal threshold values to be exceeded. In recent years, these events have shown a non-negligible increase in frequency and intensity. The present work reports the results of an analysis of the dust events that in recent years (2018–2023) affected the Mediterranean area and in particular central Italy, focusing on the more recurrent meteorological configurations leading to long-range transport and on the consequent increase in aerosol concentration values. A method for desert intrusion episodes identification has been developed using both numerical forecast model data and PM₁₀ observed data. A multi-year dataset has been analyzed by applying such an identification method and the resulting set of dust events episodes, affecting central Italy, has been studied in order to highlight their frequency on a seasonal basis and their interannual variability. In addition, a first attempt at a meteorological classification of desert intrusions has been carried out to identify the most recurrent circulation patterns related to dust intrusions. Understanding their annual and seasonal variations in frequency and intensity is a key topic, whose relevance is steeply growing in the context of ongoing climate change. Full article

Research into the adverse health effects of air pollution exposure has repeatedly considered smaller particles, to the point where particle number concentration might be a more relevant metric than mass concentration. Here, we highlight some historical research which developed metrics for air pollution severity based on particle number concentration. Because this work was published in a national journal and prior to the internet and open access, this historical research is not easy to find, and it was more through the history of the aerosol research community in Ireland that this work is now being presented. Multiple online searches for published research papers on “particle number concentrations” and “air pollution severity” were undertaken. Even when specific searches were undertaken using the author names and publication year, these featured papers were not found on any internet search. O’Dea and O’Connor proposed that air pollution severity could be classified based on particle number concentration of condensation nuclei, with ‘little’ air pollution <50 × 10³ particles per cm³, ‘mean’ 50–70 × 10³ particles per cm³, ‘strong’ 70–100 × 10³ particles per cm³, and ‘very strong’ >100 × 10³ particles per cm³. Applying their assumptions on density and mean particle size, equated to mass concentrations for a mean of 6 µgm⁻³, strong at 8.5 µgm⁻³, and very strong >10 µgm⁻³. These are consistent with the current WHO guideline values for PM_2.5. Additionally, we highlight the 1955 work by Burke and Nolan on the retention of inhaled particles, where ~40% of the inhaled number concentration is retained in the respiratory system. This is also consistent with the more recently published work on particle retention. In summary, the proposed categories of pollution severity, based on number concentrations, could form a basis for the development of future guidelines. This paper highlights that sometimes research has already been published, but it is difficult to find. We challenge researchers to find publications from their own countries which pre-date the WWW to inform current and future research. Additionally, there is scope for a repository for such information on historical publications. We have presented historical research on aerosol number concentrations, classifications of air pollution severity, and particle retention, which present lessons for current researchers. Full article

This study assesses the performance of the Copernicus Atmosphere Monitoring Service (CAMS) reanalysis dataset in estimating ground-level concentrations (GLCs) of aerosols and criteria pollutants across the Himawari-8 observational area, covering China, Indonesia, and Australia, from 2016 to 2023. Ground-based monitoring networks in these regions are limited in scope, making it necessary to rely on satellite-derived aerosol optical depth (AOD) as a proxy for GLCs. While AOD offers broad coverage, it presents challenges, particularly in capturing surface-level pollution accurately during episodic events. CAMS, which integrates satellite data with atmospheric models, is evaluated here to determine its effectiveness in addressing these issues. The study employs square root transformation to normalize pollutant concentration data and calculates monthly–hourly long-term averages to isolate pollution anomalies. Geographically weighted regression (GWR) and Jacobian matrix (dY/dX) methods are applied to assess the spatial variability of pollutant concentrations and their relationship with meteorological factors. Results show that while CAMS captures large-scale pollution episodes, such as the 2019/2020 Australian wildfires, discrepancies in representing GLCs are apparent, especially when vertical aerosol stratification occurs during short-term pollution events. The study emphasizes the need for integrating CAMS data with higher-resolution satellite observations, like Himawari-8, to improve the accuracy of real-time air quality monitoring. The findings highlight important implications for public health interventions and environmental policy-making, particularly in regions with insufficient ground-based data. Full article

Overtraffic is one of the main keys to air pollution in urban areas. The aim of the present work is to review the approaches and explore the potentiality of AI in reducing traffic pollution in urban areas, ranging over three main areas: the optimization of traffic lights timing to reduce delays, the use of AI-powered drones to monitor pollution levels in real-time, and the use of fixed AI-based sensors to detect the levels of pollutants in the air with the use of AI models to identify patterns in the collected data and predict air quality in near-real time. Some attention was also dedicated to possible problems arising from privacy protection and data security, and the case study of the Piemonte area and of the city of Turin in the north–west of Italy is presented: the current situation is depicted, and possible local future applications of AI are explored. The use of AI has proven to be very promising in all three areas, particularly in the field of optimization of traffic lights’ timing and coordination in increasingly larger traffic networks. Full article

This paper presents a novel machine learning methodology able to extend the results of detailed local emission inventories to larger domains where emission estimates are not available. The first part of this work consists in the development of an emission inventory of elemental carbon (EC), black carbon (BC), organic carbon (OC), and levoglucosan (LG) obtained from the detailed emission estimates available from the Project LIFE PREPAIR for the Po Basin in north Italy. The emissions of these chemical species in combination with particulate primary emissions and gaseous precursors are very important information in source apportionment and in the impact assessment of the different emission sources in air quality. To gain a better understanding of the origins of atmospheric pollution, it is possible to combine measurements with emission estimates for the particulate matter fractions known as EC, BC, OC, and LG. To identify the sources of emissions, it is usual practice to use the ratio of the measured EC, OC, TC (Total Carbon), and LG. The PREPAIR emission estimates and these new calculations are then used to train the Random Forest (RF) algorithm, considering a large array of local variables, such as taxes, the characteristics of urbanization and dwellings, the number of employees detailed for economic activities, occupation levels and land cover. The outcome of the comparison of the predictions of the machine learning implemented model (ML) with the estimates obtained for the same areas by two independent methods, local disaggregation of the national emission inventory and Copernicus Air Modelling Service (CAMS) emissions estimates, is extremely encouraging and confirms it also as a promising approach in terms of effort saving. The implemented modelling approach identifies the most important variables affecting the spatialization of different pollutants in agreement with the main emission source characteristics and is suitable for harmonization of the results of different local emission inventories with national emission reporting. Full article

Harsh winters, aging infrastructure, and the demand for modern amenities are major factors contributing to the deteriorating air quality in Bishkek. The city meets its winter heating energy needs through coal combustion at the central heating plant, heat-only boilers, and in situ heating equipment, while diesel and petrol fuel its transportation. Additional pollution sources include 30 km² of industrial area, 16 large open combustion brick kilns, a vehicle fleet with an average age of more than 10 years, 7.5 km² of quarries, and a landfill. The annual PM_2.5 emission load for the airshed is approximately 5500 tons, resulting in an annual average concentration of 48 μg/m³. Wintertime daily averages range from 200 to 300 μg/m³. The meteorological and pollution modeling was conducted using a WRF–CAMx system to evaluate PM_2.5 source contributions and to support scenario analysis. Proposed emissions management policies include shifting to clean fuels like gas and electricity for heating, restricting secondhand vehicle imports while promoting newer standard vehicles, enhancing public transport with newer buses, doubling waste collection efficiency, improving landfill management, encouraging greening, and maintaining road infrastructure to control dust emissions. Implementing these measures is expected to reduce PM_2.5 levels by 50–70% in the mid- to long-term. A comprehensive plan for Bishkek should expand the ambient monitoring network with reference-grade and low-cost sensors to track air quality management progress and enhance public awareness. Full article