High Concentrations Of Particulate Matter Research Articles

Investigating the Correlation between Air Pollution and Housing Prices in Seoul, South Korea: Application of Explainable Artificial Intelligence in Random Forest Machine Learning

South Korea’s Particulate Matter (PM) concentration is among the highest among Organization for Economic Cooperation and Development (OECD) member countries. However, many studies in South Korea primarily focus on housing characteristics and the physical built environment when estimating apartment prices, often neglecting environmental factors. This study investigated factors influencing apartment prices using transaction data for Seoul apartments provided by the Ministry of Land, Infrastructure, and Transport (MOLIT) in 2019. For this purpose, the study compared and analyzed a traditional hedonic price model with a machine learning-based random forest model. The main findings are as follows: First, the evaluation results of the traditional hedonic price model and the machine learning-based random forest model indicated that the random forest model was found to be more suitable for predicting apartment prices. Second, an importance analysis using Explainable Artificial Intelligence (XAI) showed that PM is more important in determining apartment prices than access to education and bus stops, which were considered in this study. Finally, the study found that areas with higher concentrations of PM tend to have higher apartment prices. Therefore, when proposing policies to stabilize apartment prices, it is essential to consider environmental factors. Furthermore, it is necessary to devise measures such as assigning PM labels to apartments during the home purchasing process, enabling buyers to consider PM and obtain relevant information accordingly.

A Quantitative Assessment of Air Particulate Matter Concentration and Its Correlation with the Climatic Factor of Dinajpur City, Bangladesh

Particulate matter (PM) is the air pollutant that poses the most significant risk to human health compared to other measured criteria pollutants. One of our main goals was to assess the quantities of particulate matter (0.3-10µm) in the Dinajpur municipality's sensitive zones, roads, and residential and commercial buildings. Light scattering particle counters, the Extech VPC-300, were used to measure particulate matter and climatic data at 26 distinct locations in Dinajpur. R 4.2.2 and the data analysis tool in Microsoft Excel were utilized for the statistical analysis. The highest values for PM 0.3, 0.5, 1, 2.5, 5, and 10 µm are 335195, 138118, 37965, 12995, 2087, and 2507 and medium value is accordingly 221304, 71873, 16681, 3393, 361 and 454, respectively. The PM 0.5 μm particle range, which includes 12 samples, is highly concerned and getting closer to the hazardous red polluted zone. Most places of PM 2.5 µm exceed the 2000 particles, which are hazardous for people. The link between climatic variables and particulate matter (PM 0.5-10 µm) significantly correlated with temperature. Humidity and PM (0.3-10 µm) measurements in the study area show a negative correlation. According to observations, factors contributing to the high PM concentration include excessive human movement, combustion byproducts of black smoke, road dust, waste, garbage in every possible place, and rice mill industry emissions.

Long term trend of particulate matter in Philadelphia, Pennsylvania and its association with introduction of environmental policies

Since the 1970s, air quality has improved at the national level in the United States, coincident with the introduction of the Clean Air Act and other air pollution regulations at a greater frequency. We present a case study from Philadelphia, Pennsylvania—the sixth most populous city in the United States. The main objectives of this study are to analyze long-term trends of particulate matter (PM) from 1986 to 2021 in Philadelphia and to examine their association with the introduction of environmental policies relevant to air pollution at the federal, state, and local levels. We find that annual PM10 concentration decreased by 47% from 1986 to 2021 and annual PM2.5 concentration decreased by 31% from 2000 to 2021 in Philadelphia. We find that carbonaceous content (both elemental and organic carbon) has declined over the same period of 2000 to 2021, demonstrating its contribution to overall PM2.5 reduction in Philadelphia. In Philadelphia, high PM concentrations occur in the summer months; however, seasonal patterns have changed for PM10 in the last decade (2011–2020). Overall, PM reductions occurred over all seasons, with the greatest reductions occurring for PM2.5 during summer months and for PM10 during winter months. The Clean Air Act contributed to the creation of many regulatory policies that address unhealthy levels of PM. The introduction of various environmental policies that target the transportation sector has contributed to the reduction of PM levels in Philadelphia. Air quality would continue being improved by implementing such environmental policies specific to the emissions sectors.

Microbial dynamics and climatic interactions in pig sheds: Insights into airborne microbes and particulate matter concentrations.

Emissions of airborne pollutants from livestock buildings affect indoor air quality, the health and well-being of farmers, animals and the environment. This study aimed to evaluate the microbial count within pig sheds and its relationship with meteorological variables (temperature, relative humidity and air velocity) and particulate matter (PM10 and PM2.5) and microbial diversity. Sampling was conducted both inside and outside of two pig sheds over three seasons (summer, rainy and winter), with regular monitoring at fortnightly intervals. Results showed that the bacterial and fungal counts ranged from 0.07 to 3.98 x 103 cfu/m3 inside the sheds and 0.01 to 1.82 x 103 cfu/m3 outside. Seasonal variations were observed, with higher concentrations of particulate matter detected during the winter season, followed by summer. Climatic variables such as temperature, air velocity and relative humidity demonstrated significant impacts on the abundance of Enterobacteriaceae and fungi, while air velocity specifically influenced the presence of mesophilic bacteria and staphylococci. Importantly, no significant disparities were found between microbial counts and particulate matter levels. Staphylococcaceae emerged as the predominant bacterial family, while Aspergillus and Cladosporium spp. were the dominant fungal species within the pig sheds. The average levels of airborne bacteria and fungi in pig sheds were found to be within the recommended range, which can be attributed to the loose housing design and lower animal population on the farms.

Deep learning models for air quality forecasting based on spatiotemporal characteristics of data

Deep learning models for air quality forecasting based on spatiotemporal characteristics of data

Environmental factors as drivers of the spatial distribution of the copepods Eurytemora affinis affinis and Eurytemora velox in the Scheldt tributaries.

The main tributaries of the Scheldt estuary offer an interesting setting to study the response of organisms to various environmental conditions. The Bovenscheldt and the Dender are closed to tidal entrance by locks, and are hence essentially fresh water systems with low turbulence. The Durme and the Rupel, with its two sub-tributaries (the Beneden-Nete and the Dijle), receive estuarine water at each tide and as such, present a fresh-brackish water gradient according to the geographic location of their confluence with the estuary and the tidal phase. This paper aims (1) to determine the spatial-temporal distribution of two calanoid copepod species: Eurytemora affinis affinis (Poppe, 1880) and E. velox (Lilljeborg, 1853) in these tributaries and (2) the environmental conditions favoring each species development. Species distribution and abundance data, collected in the four tributaries between 2018 and 2021 are analyzed by Generalized Linear Mixed Models (GLMM) and the results are interpreted in the context of ecosystem functioning. A spatial structuring between E. affinis affinis and E. velox was revealed. Whilst E. affinis affinis develops best in the tidal tributaries such as the Durme and Rupel, E. velox is more typical of the non-tidal ones: the Bovenscheldt and the Dender. The tidal tributaries have high Particulate Organic Matter (POM) concentrations due to tidal turbulence, and hence generally lower oxygen concentrations than the non-tidal ones (< 10-11 mg L-1). Factors favoring E. affinis affinis development in the tidal tributaries are higher chlorinity, pH and diatom abundance than in the non-tidal ones. In the non-tidal tributaries, with lower POM concentrations, phytoplankton activity more strongly influences the overall oxygen balance, with oxygen concentrations often rising above 10-11 mg L-1 and sometimes showing high variability. The results suggest that E. affinis affinis might suffer from hyperoxia-induced oxidative stress under these high oxygen conditions and advocate the necessity to include this aspect in further research. E. velox, which is most abundant in those tributaries, seems to be more tolerant to hyperoxia and strong fluctuations in O2 concentrations than E. affinis affinis.

Local incomplete combustion emissions define the PM2.5 oxidative potential in Northern India

The oxidative potential (OP) of particulate matter (PM) is a major driver of PM-associated health effects. In India, the emission sources defining PM-OP, and their local/regional nature, are yet to be established. Here, to address this gap we determine the geographical origin, sources of PM, and its OP at five Indo-Gangetic Plain sites inside and outside Delhi. Our findings reveal that although uniformly high PM concentrations are recorded across the entire region, local emission sources and formation processes dominate PM pollution. Specifically, ammonium chloride, and organic aerosols (OA) from traffic exhaust, residential heating, and oxidation of unsaturated vapors from fossil fuels are the dominant PM sources inside Delhi. Ammonium sulfate and nitrate, and secondary OA from biomass burning vapors, are produced outside Delhi. Nevertheless, PM-OP is overwhelmingly driven by OA from incomplete combustion of biomass and fossil fuels, including traffic. These findings suggest that addressing local inefficient combustion processes can effectively mitigate PM health exposure in northern India.

Improvement of the anthropogenic emission rate estimate in Ulaanbaatar, Mongolia, for 2020–21 winter

Ulaanbaatar (UB), the fast-growing capital of Mongolia, is known for its world's worst level of particulate matter (PM) concentrations in winter. However, current anthropogenic emission inventories over the UB are based on data from more than fifteen years ago, and satellite observations are scarce because UB is in high latitudes. During the winter of 2020–21, the first period of the Fine Particle Research Initiative in East Asia considering the National Differences (FRIEND), several times higher concentrations of PM in UB compared to other urban sites in East Asia were observed but not reproduced with a chemical transport model mainly due to the underestimated anthropogenic emissions. Therefore, we devised a method for sequentially adjusting emissions based on the reactivity of PM precursors using ground observations. We scaled emission rates for the inert species (CO, elemental carbon (EC), and organic carbon (OC)) to reproduce their observed ambient concentrations, followed by SO2 to reproduce the concentration of SO42−, which was examined to have the least uncertainty based on the abundance of observed NH3, and finally NO and NH3 for NO3−, and NH4+. This improved estimation is compared to regional inventories for Asia and suggests more than an order of magnitude increase in anthropogenic emissions in UB. Using the improved emission inventory, we were able to successfully reproduce independent observation data on PM2.5 concentrations in UB in December 2021 from the U.S. Embassy. During the campaign period, we found more than 50% of the SO42−, NO3−, and NH4+ increased in UB due to the improvement could travel to Beijing, China (BJ), and about 20% of the SO42− could travel to Noto, Japan (NT), more than 3000 km away. Also, the anthropogenic emissions in UB can effectively increase OC, NO3−, and NH4+ concentrations in BJ when Gobi dust storms occur.

Exploring role-playing as a tool for involving citizens in air pollution mitigation urban policies

The design of air pollution control policies is a complex decision-making process. Public participation can enhance the legitimacy and effectiveness of environmental urban regulations but does not always occur naturally. This article presents the experience of the use of role-playing in six mid-sized cities in southern Chile to evaluate how residents affected by high concentrations of fine particulate matter perceive the problem and debate possible solutions. For this, the participants were asked to assume the role of advisors. On that premise, they had to prioritize between a series of mitigation measures and reach a consensus with other advisors. The results show the potential of role-playing games, on the one hand, to promote critical thinking and argument skills among the general public and, on the other, to help policymakers in the design of more operative and fair plans of action.

Existence of typical winter atmospheric circulation patterns leading to high PM2.5 concentration days in East Asia

Understanding the atmospheric circulation patterns responsible for severe air pollution events in East Asia is important because East Asia is one of the most polluted regions in the world, particularly during the boreal winter (December-January-February). Here, by conducting GEOS-Chem simulation with fixed anthropogenic emission sources, we found that there exist three typical atmospheric circulation patterns conducive to leading to high concentrations of particulate matter with a diameter less than or equal to 2.5 μm (PM2.5) in East Asia. These atmospheric circulation patterns are characterized by weakened horizontal winds, which allows PM2.5 to accumulate, and by enhanced relative humidity, which can favor secondary formation of PM2.5. The occurrence of these atmospheric circulation patterns is associated with increased sea ice cover over the Barents Sea and heavy precipitation over the tropical western Indian Ocean. The existence of these atmospheric circulation patterns among typical atmospheric circulation patterns indicates high PM2.5 days in East Asia are unavoidable given current level of anthropogenic emissions in the region. This conclusion indicates that sustained efforts to reduce anthropogenic emission sources in East Asia should be warranted to avoid high PM2.5 days.

Numerical analysis and improvement strategy of particulate matter concentration in underground subway stations

Particulate matter (PM) is an important air pollutant in subway systems. Exposure to high PM concentrations poses a threat to human health and raises concerns. Piston effect can provide the outdoor fresh air to subway stations through station exits, but it also brings PM from tunnels to platforms when the train enters the station. In this paper, a transient numerical model is developed according to an actual subway station and validated by field tests. The model is used to investigate PM10 concentration at platform induced by train movement. The cumulative PM10 concentration at platform in one day is calculated according to the train schedule. Platform screen door (PSD) with adjustable vents is studied to control the PM10 level at platform and energy saving in 27 cities of five climatic zones in China are discussed. The results show that adjustable vent height should be less than 0.70 m to ensure that PM10 concentrations at platform can be maintained at an acceptable level and the requirement of fresh air can be met. Moreover, energy consumption can be reduced by more than 70 % in different cities compared with the conventional mechanical ventilation system. The optimized adjustable vent can provide an important reference for improving indoor air quality (IAQ) and saving energy consumption of ventilation systems in underground facilities.

Overview of the Alaskan Layered Pollution and Chemical Analysis (ALPACA) Field Experiment.

The Alaskan Layered Pollution And Chemical Analysis (ALPACA) field experiment was a collaborative study designed to improve understanding of pollution sources and chemical processes during winter (cold climate and low-photochemical activity), to investigate indoor pollution, and to study dispersion of pollution as affected by frequent temperature inversions. A number of the research goals were motivated by questions raised by residents of Fairbanks, Alaska, where the study was held. This paper describes the measurement strategies and the conditions encountered during the January and February 2022 field experiment, and reports early examples of how the measurements addressed research goals, particularly those of interest to the residents. Outdoor air measurements showed high concentrations of particulate matter and pollutant gases including volatile organic carbon species. During pollution events, low winds and extremely stable atmospheric conditions trapped pollution below 73 m, an extremely shallow vertical scale. Tethered-balloon-based measurements intercepted plumes aloft, which were associated with power plant point sources through transport modeling. Because cold climate residents spend much of their time indoors, the study included an indoor air quality component, where measurements were made inside and outside a house to study infiltration and indoor sources. In the absence of indoor activities such as cooking and/or heating with a pellet stove, indoor particulate matter concentrations were lower than outdoors; however, cooking and pellet stove burns often caused higher indoor particulate matter concentrations than outdoors. The mass-normalized particulate matter oxidative potential, a health-relevant property measured here by the reactivity with dithiothreiol, of indoor particles varied by source, with cooking particles having less oxidative potential per mass than pellet stove particles.

Air pollution seasons in urban moderate climate areas through big data analytics

High particulate matter (PM) concentrations have a negative impact on the overall quality of life and health. The annual trends of PM can vary greatly depending on factors such as a country’s energy mix, development level, and climatic zone. In this study, we aimed to understand the annual cycle of PM concentrations in a moderate climate zone using a dense grid of low-cost sensors located in central Europe (Krakow). Over one million unique records of PM, temperature, humidity, pressure and wind speed observations were analyzed to gain a detailed, high-resolution understanding of yearly fluctuations. The comprehensive big-data workflow was presented with the statistical analysis of the meteorological factors. A big data-driven approach revealed the existence of two main PM seasons (warm and cold) in Europe’s moderate climate zone, which do not correspond directly with the traditional four main seasons (Autumn, Winter, Spring, and Summer) with two side periods (early spring and early winter). Our findings also highlighted the importance of high-resolution time and space data for sustainable spatial planning. The observations allowed for distinguishing whether the source of air pollution is related to coal burning for heating in cold period or to agricultural lands burning during the warm period.

Comparison Between Smoked Tobacco and Medical Cannabis Cigarettes Concerning Particulate Matter.

Introduction: Cannabis is a widely used drug like tobacco and alcohol. In the meantime, it is also prescribed for medical treatment in some countries. Tobacco smoke contains chemical carcinogens and particulate matter (PM) that are both harmful to health. Method: In this study, we investigated PM levels in second-hand smoke (SHS) of hand-tamped cannabis cigarettes compared to cigarettes with tubing tobacco and the 3R4F reference cigarette. Results: It could be demonstrated that the largest proportion of the particle mass is attributable to particles with a diameter of less than 1μm and that every tested cigarette emitted more PM than the 3R4F reference cigarette. In addition, our data clearly revealed that cannabis smoke contains higher PM levels in SHS than tobacco cigarettes. Compared to the reference cigarette, the PM1 emissions of cannabis were 105% higher. Also, the cannabis mixed cigarettes had higher PM levels than the 3R4F cigarettes. For instance, the PM10 emissions were 93% higher. Also, the Gauloises Mélange tubing tobacco also reached higher PM concentrations than the 3R4F cigarette. Discussion: Regardless of negative health effects, cannabis is seen as a harmless drug in the public eye. We found strong indications for potential health risks by PM from cannabis products and, therefore, the public should be educated about a potential harm.

Fresh air in the city: the impact of air pollution on the pricing of real estate.

This study investigates the impact of ambient air pollution on housing prices in Warsaw, Poland, by examining spatial dependencies. The high concentration of particulate matter (PM10, PM2.5 and PM1) is expected to reduce real estate values. Using a hedonic model with approximately 15,000 observations and a spatial error model, we did not find evidence of this impact. Standard and premium housing submarkets differ in price determinants, but both are insensitive to environmental issues. This could be explained by the lack of comprehensive intra-urban historical information on air pollution, which limits investors' rationality and their ability to properly value real estate based on environmental issues. Additionally, measurement and aggregation issues, along with low pollution variability within the city, may contribute to the insignificance of this information in real estate sales prices. Our empirical research confirms a strong link between air pollution and weather conditions within the city, where low temperatures and low-speed southern winds worsen contamination levels, while high temperatures and westerly winds improve air quality. Furthermore, we find that incorporating pollution data using PM yearly mean concentration works better in modelling than the PCA-reduced air pollution index.

Effect modification of diabetic status on the association between exposure to particulate matter and cardiac arrhythmias in a general population: A systematic review and meta-analysis.

Particulate matter (PM) has various health effects, including cardiovascular diseases. Exposure to PM and a diagnosis of diabetes mellitus (DM) have been associated with an increased risk of cardiac arrhythmias. However, no comprehensive synthesis has been conducted to examine the modifying effect of DM on the association between PM and arrhythmia events. Thus, the objectives of this review were to investigate whether the association of PM is linked to cardiac arrhythmias and whether DM status modifies its effect in the general population. The search was conducted on PubMed/MEDLINE and Embase until January 18, 2023. We included cohort and case-crossover studies reporting the effect of PM exposure on cardiac arrhythmias and examining the role of diabetes as an effect modifier. We used the DerSimonian and Laird random-effects model to calculate the pooled estimates. A total of 217 studies were found and subsequently screened. Nine studies met the inclusion criteria, and five of them were included in the meta-analysis. The participants numbered 4,431,452, with 2,556 having DM. Exposure to PM of any size showed a significant effect on arrhythmias in the overall population (OR 1.10, 95% CI 1.04-1.16). However, the effect modification of DM was not significant (OR 1.18 (95% CI 1.01-1.38) for DM; OR 1.08 (95% CI 1.02-1.14) for non-DM; p-value of subgroup difference = 0.304). Exposure to higher PM concentrations significantly increases cardiac arrhythmias requiring hospital or emergency visits. Although the impact on diabetic individuals is not significant, diabetic patients should still be considered at risk. Further studies with larger sample sizes and low bias are needed.

Particulate matter air pollution in the Republic of Slovenia and its national spatial emissions release

In this work, an impact of particulate matter (PM) on air pollution and its emissions released is investigated. Concentration levels of PM10 and PM2.5 in the Republic of Slovenia (RS) were analyzed on daily, weekly, monthly, quartile, and annual data and then compared. The study was conducted from January 1, 2020 to December 31, 2020 at two monitoring stations, so-called urban traffic and urban industrial. Obtained results showed that the highest concentrations of PM in air were during the cold months of the year when the temperatures are lower (&lt; 6%). Results from heating seasons (from October 1, 2019 and March 31, 2020 and from October 1, 2020 to March 31, 2021) showed the highest concentration levels of PM10 between 9:00 and 14:00 and between 18:00 and 22:00, with highest concentrations recorded during the winter months (January and February). Additionally, emission building inventory was built based on PM10 and PM2.5 emissions released. The study includes yearly spatial building emissions release model considering currently installed high PM emitted stoves. The building model for the Republic of Slovenia has been developed by using 500 x 500 m model grid. Obtained results showed that changing currently low efficient appliance with advanced appliance would result in significant lower PM emissions released from building sector.

SchoolAIR: A Citizen Science IoT Framework Using Low-Cost Sensing for Indoor Air Quality Management.

Indoor air quality (IAQ) problems in school environments are very common and have significant impacts on students' performance, development and health. Indoor air conditions depend on the adopted ventilation practices, which in Mediterranean countries are essentially based on natural ventilation controlled through manual window opening. Citizen science projects directed to school communities are effective strategies to promote awareness and knowledge acquirement on IAQ and adequate ventilation management. Our multidisciplinary research team has developed a framework-SchoolAIR-based on low-cost sensors and a scalable IoT system architecture to support the improvement of IAQ in schools. The SchoolAIR framework is based on do-it-yourself sensors that continuously monitor air temperature, relative humidity, concentrations of carbon dioxide and particulate matter in school environments. The framework was tested in the classrooms of University Fernando Pessoa, and its deployment and proof of concept took place in a high school in the north of Portugal. The results obtained reveal that CO2 concentrations frequently exceed reference values during classes, and that higher concentrations of particulate matter in the outdoor air affect IAQ. These results highlight the importance of real-time monitoring of IAQ and outdoor air pollution levels to support decision-making in ventilation management and assure adequate IAQ. The proposed approach encourages the transfer of scientific knowledge from universities to society in a dynamic and active process of social responsibility based on a citizen science approach, promoting scientific literacy of the younger generation and enhancing healthier, resilient and sustainable indoor environments.

Indoor air quality improvement with filtration and UV-C on mitigation of particulate matter and airborne bacteria: Monitoring and modeling

Indoor air, especially with suspended particulate matter (PM), can be a carrier of airborne infectious pathogens. Without sufficient ventilation, airborne infectious diseases can be transmitted from one person to another. Indoor air quality (IAQ) significantly impacts people's daily lives as people spend 90% of their time indoors. An industrial-grade air cleaner prototype (filtration + ultraviolet light) was previously upgraded to clean indoor air to improve IAQ on two metrics: particulate matter (PM) and viable airborne bacteria. Previous experiments were conducted to test its removal efficiency on PM and airborne bacteria between the inlet and treated air. However, the longer-term improvement on IAQ would be more informative. Therefore, this research focused on quantifying longer-term improvement in a testing environment (poultry facility) loaded with high and variable PM and airborne bacteria concentrations. A 25-day experiment was conducted to treat indoor air using an air cleaner prototype with intermittent ON and OFF days in which PM and viable airborne bacteria were measured to quantify the treatment effect. The results showed an average of 55% reduction of total suspended particulate (TSP) concentration between OFF days (110 μg/m3) and ON days (49 μg/m3). An average of 47% reduction of total airborne viable bacteria concentrations was achieved between OFF days (∼3200 CFU/m3) and ON days (∼2000 CFU/m3). A cross-validation (CV) model was established to predict PM concentrations with five input variables, including the status of the air cleaner, time (h), ambient temperature, indoor relative humidity, and day of the week to help simulate the air-cleaning effect of this prototype. The model can approximately predict the air quality trend, and future improvements may be made to improve its accuracy.

Establishing a Real-Time Prediction System for Fine Particulate Matter Concentration Using Machine-Learning Models

With the rapid urbanization and industrialization in Taiwan, pollutants generated from industrial processes, coal combustion, and vehicle emissions have led to severe air pollution issues. This study focuses on predicting the fine particulate matter (PM2.5) concentration. This enables individuals to be aware of their immediate surroundings in advance, reducing their exposure to high concentrations of fine particulate matter. The research area includes Keelung City and Xizhi District in New Taipei City, located in northern Taiwan. This study establishes five fine prediction models based on machine-learning algorithms, namely, the deep neural network (DNN), M5’ decision tree algorithm (M5P), M5’ rules decision tree algorithm (M5Rules), alternating model tree (AMT), and multiple linear regression (MLR). Based on the predictive results from these five models, the study evaluates the optimal model for forecast horizons and proposes a real-time PM2.5 concentration prediction system by integrating various models. The results demonstrate that the prediction errors vary across different models at different forecast horizons, with no single model consistently outperforming the others. Therefore, the establishment of a hybrid prediction system proves to be more accurate in predicting future PM2.5 concentration compared to a single model. To assess the practicality of the system, the study process involved simulating data, with a particular focus on the winter season when high PM2.5 concentrations are prevalent. The predictive system generated excellent results, even though errors increased in long-term predictions. The system can promptly adjust its predictions over time, effectively forecasting the PM2.5 concentration for the next 12 h.