Long-term Air Quality Research Articles

Numerical analysis of factors causing long-term trends and annual variations of sulfur and nitrogen deposition amount in Japan from 2000 to 2020

The deposition of sulfur and nitrogen from the atmosphere to the ground surface is harmful to ecosystems. This study performed long-term air quality simulations to quantify the influences of factors, including anthropogenic emissions in Japan, meteorological fields, transboundary transport, and volcanic emissions, on the long-term trends and annual variations in sulfur and nitrogen deposition in Japan from 2000 to 2020. The air quality simulations performed well in reproducing the long-term trends and annual variations in the wet deposition amount, whereas the simulated dry deposition amount may contain larger uncertainties. The decreasing trends in sulfur deposition were statistically significant during the entire study period (2000–2020) in most of Japan. They were caused by the reduction of anthropogenic SO2 emissions in Japan and China, which was accomplished by the implementation of stringent emission controls, as well as a gradual decrease in SO2 emissions from the Miyakejima volcano, which erupted in 2000. No significant decreasing trends were found in nitrogen deposition in Japan during the first half of the study period (2000–2010). Decreases caused by the reduction in anthropogenic NOx emissions in Japan were compensated for by increases caused by increasing NOx emissions in China and changes in the gas-aerosol partitioning of nitrates instead of sulfates. The decreasing trend in nitrogen deposition in Japan became statistically significant during the second half of the study period (2010–2020) after anthropogenic NOx emissions started to decline in China. Meteorological fields primarily influenced annual variations in the amount of nitrogen deposition. This study reveals that long-term air quality simulations are useful for quantifying the influences of various factors on long-term trends and annual variations in sulfur and nitrogen deposition.Graphical

Improved daily PM2.5 estimates in India reveal inequalities in recent enhancement of air quality.

Poor ambient air quality poses a substantial global health threat. However, accurate measurement remains challenging, particularly in countries such as India where ground monitors are scarce despite high expected exposure and health burdens. This lack of precise measurements impedes understanding of changes in pollution exposure over time and across populations. Here, we develop open-source daily fine particulate matter (PM2.5) datasets at a 10-kilometer resolution for India from 2005 to 2023 using a two-stage machine learning model validated on held-out monitor data. Analyzing long-term air quality trends, we find that PM2.5 concentrations increased across most of the country until around 2016 and then declined partly due to favorable meteorology in southern India. Recent reductions in PM2.5 were substantially larger in wealthier areas, highlighting the urgency of air quality control policies addressing all socioeconomic communities. To advance equitable air quality monitoring, we propose additional monitor locations in India and examine the adaptability of our method to other countries with scarce monitoring data.

GeoAI Dataset for Urbanized Area Segmentation from Landsat 8/9 Satellite Imagery and GEMS

In South Korea, air pollution has emerged as a pressing social issue, necessitating data-driven approaches to monitor sources of air pollutants. This study constructed a GEO AI dataset for detecting air pollution sources in urbanized areas, utilizing Landsat 8/9 satellite imagery, Geostationary Environment Monitoring Spectrometer geostationary satellite data, and air quality monitoring network data. The dataset is optimized for semantic segmentation tasks, including labeled data for urban area segmentation, and is designed to enable precise detection of pollution sources within urban regions by integrating satellite imagery and air quality information. Using this dataset, we applied a modified U-Net model to classify pollutant sources in urbanized areas, achieving high performance with an mIoU of 0.8592 and pixel accuracy of 0.9433. These results demonstrate the effectiveness of the GEO AI dataset as a tool for identifying and managing major pollution sources, providing foundational data for air quality monitoring and policy development across South Korea and East Asia. With further integration of additional air pollution data, this dataset is expected to contribute to long-term air quality management and the mitigation of health impacts associated with pollution.

AIR QUALITY PREDICTION IN HANOI USING A DEEP LEARNING APPROACH

Air pollution is becoming a serious global crisis, threatening human health, disrupting the balance of the environment, negatively affecting ecosystems, and contributing to climate change. Accurate long-term air quality prediction plays a key role in building early warning systems to mitigate these negative impacts. Efforts to forecast air quality through the combination of knowledge from environmental science, statistics, and computer science have attracted much attention. Among them, deep learning and advanced machine learning have demonstrated an outstanding ability to detect complex non-linear patterns from environmental data. However, the application of deep learning to air quality prediction is still quite new. This paper proposes a deep-learning model using the LSTM (Long Short-Term Memory) network to predict air quality in Hanoi. The research results demonstrate that the proposed model is capable of predicting the air quality index with high accuracy, close to actual values ​​from monitoring data.

The Swiss National Air Pollution Monitoring Network (NABEL) – Bridging Science and Environmental Policy

Awareness of atmospheric air quality in Switzerland became a concern in the 1960s, as a result of which the Swiss National Air Pollution Monitoring Network (Nationales Beobachtungsnetz für Luftfremdstoffe - NABEL) was created in the 1970s. This paper describes the establishment and evolution of NABEL, emphasizing its important role in monitoring air quality in Switzerland, and its contribution to international observation networks and research. The network’s history, legal framework, and measurement program are described, and exemplary time-series of air quality parameters are given. NABEL is an excellent example for reliable, long-term air quality monitoring and demonstrates the importance of such monitoring for air pollution control at both national and international levels.

Long-term urban air quality prediction with hierarchical attention loop network

Air pollution poses severe threats to human health, socioeconomic development, and natural environment, making it one of the most serious environmental issues. Accurate long-term regional air quality prediction plays a significant role in mitigating the severity of pollution events and effectively suppressing the intensification of air pollution, benefiting both the environment and societyAccurate regional air quality prediction contributes to the governance of atmospheric pollution, benefiting both the environment and society In this paper, we propose a novel Hierarchical Attention Loop Network (HALN) comprising four key components for long-term prediction of PM2.5 concentrations across regional multiple target stations, achieving effective prediction horizons of up to 120 h. Instead of using all monitoring stations within a region, the MIC-H selector selects station data exhibiting strong spatiotemporal correlations at each hierarchical level. HALN then injects additional spatial information through annular grid positional encoding. The hierarchical attention feature match layer refines the model’s focus on data with a more substantial predictive impact. As the core integrative component of HALN, attention loop block reinforces the influence of historical outputs, enhancing the model’s ability to capture long-term dependencies.Instead of using all monitoring stations within a cluster, the hierarchical MIC selector utilizes the maximal information coefficient (MIC) to select station data exhibiting strong spatiotemporal correlations at each hierarchical level. HALN then details the relative distances and azimuths between stations through annular grid positional encoding to inject additional spatial information into the selected input data. Subsequently, the hierarchical attention feature match layer employs multi-head attention to independently adapt the input weights at each level, refining the model’s focus on station data with a more substantial predictive impact. As the core integrative component of HALN, attention loop block employs gated recurrent units (GRUs) and a feedback attention mechanism to reinforce the influence of historical outputs, thereby enhancing the model’s ability to capture long-term dependencies. Extensive real-world experimental results demonstrate that the proposed model exhibits exceptional robustness in regional predictions and achieves high accuracy in long-term forecasting. The coefficient of determination (R2) reaches 0.793 and 0.636 for 24-hour and 120-hour predictions, respectively.

Air quality forecasting using a spatiotemporal hybrid deep learning model based on VMD–GAT–BiLSTM

The precise forecasting of air quality is of great significance as an integral component of early warning systems. This remains a formidable challenge owing to the limited information of emission source and the considerable uncertainties inherent in dynamic processes. To improve the accuracy of air quality forecasting, this work proposes a new spatiotemporal hybrid deep learning model based on variational mode decomposition (VMD), graph attention networks (GAT) and bi-directional long short-term memory (BiLSTM), referred to as VMD–GAT–BiLSTM, for air quality forecasting. The proposed model initially employ a VMD to decompose original PM2.5 data into a series of relatively stable sub-sequences, thus reducing the influence of unknown factors on model prediction capabilities. For each sub-sequence, a GAT is then designed to explore deep spatial relationships among different monitoring stations. Next, a BiLSTM is utilized to learn the temporal features of each decomposed sub-sequence. Finally, the forecasting results of each decomposed sub-sequence are aggregated and summed as the final air quality prediction results. Experiment results on the collected Beijing air quality dataset show that the proposed model presents superior performance to other used methods on both short-term and long-term air quality forecasting tasks.

Blockchain and IoT integration for secure short-term and long-term air quality monitoring system using optimized neural network.

Accurate air pollution prediction is vital for residents' well-being. This research introduces a secure air quality monitoring system using neural networks and blockchain for robust analysis, precise predictions, and early pollution detection. Blockchain guarantees data integrity, security, and transparency. Goals include real-time air quality data, secure blockchain recording, and enhanced safety through informed decisions. The research integrates blockchain and IoT for short- and long-term air quality monitoring, utilizing an optimized neural network. IoT sensors collect PM2.5, PM10, CO, NO2, and SO2, processed through noise removal and normalization, with feature extraction using N-tuple contrastive learning. Predictions utilize Graph attention-based deep Residual shrinkage Network and Bidirectional long short Term Memory (GRNBTM) categorized into five levels. An adaptive bowerbird algorithm optimizes parameters, reducing computational complexity. Blockchain integration ensures secure, tamper-proof data storage with a lightweight consensus-based algorithm. The GRNBTM model's air quality monitoring performance is extensively simulated and analyzed at 30-min, 2-h, 1-day, and 1-month intervals, demonstrating superior performance over existing techniques.

Leveraging physics-based and explainable machine learning approaches to quantify the relative contributions of rain and air pollutants to wet deposition

A quantitative understanding of the roles of rainfall and pollutant concentrations in wet deposition is important because they critically influence terrestrial and aquatic ecosystems. However, their relative contributions to wet deposition, which vary across regions, have not yet been identified. We propose two methods that quantitatively separate the contributions of rain and pollutant concentrations to wet deposition: one is based on simplified equations describing the wet scavenging of pollutants and the other is based on random forest models employing SHapley Additive exPlanations. Three-dimensional long-term air quality simulations from 2003 to 2019 are used as inputs for both the physics-based and machine learning models. Remarkably, the results drawn from the explainable machine learning model are consistent with those from the physics-based approach: overall, rain is a more important limiting factor than pollutant concentrations and the relative contribution of rain is larger than that of pollutants by up to a factor of 3–4 in polluted regions. In polluted regions, pollutant concentrations can remain relatively high even in the presence of precipitation owing to continuous and intense emissions; therefore, wet deposition is limited by rainfall. The contribution of rainfall is larger by 1.5–2.5 than that of pollutant concentrations in regions even with low emissions and this considerably large role of rain suggests that regional or transboundary pollutant transport plays a key role in modulating wet deposition. However, in very remote regions, once the rainfall amount exceeds a certain value, rainfall no longer contributes to increasing wet deposition because atmospheric pollutants are readily removed by rain. So, the contributions of the two factors are comparable in pristine regions. Our results can serve as a basis for explaining interannual variations in wet deposition and for future projections of wet deposition under emission control plans and climate change scenarios across regions.

Urban heat and pollution island in the Moscow megacity: Urban environmental compartments and their interactions

Cities are highly interconnected systems where specific interactions between various urban environments occur due to the Urban Heat Island (UHI) and Urban Pollution Island (UPI) effects. Four compartments of the environment (atmospheric air, road dust, streamflow, and people) are discussed for Moscow city. Long-term meteorological, radiative, air quality, and precipitation measurements, the non-hydrostatic regional numerical COSMO model, and extensive hydrological and geochemical sampling were used. To characterize mortality and UPI interaction, a family of distributed lag non-linear models (DLNM) was applied. The study reveals increased aerosols concentrations which reduce the incoming solar radiation and increase the atmospheric longwave radiation. UHI strengthens the low-troposphere convergence due to urban breeze circulation and atmospheric circulation due to elevated surface roughness, the effect which leads to 11.6% heaviest precipitation increase compared to background values. Increased precipitation doubles streamflow rates and enhances the contribution of rain floods to annual flow. Similar geochemical associations with Sb, W, Zn, Cd, Pb, and Cu were found in aerosol PM10, indicating transport and road dust impact. Finally, associations between high temperatures and human mortality which are generally stronger at high levels of air pollution for both PM10 and NO2, and for lag 1 day and 2–6 days are discussed.

Air Quality Class Prediction Using Machine Learning Methods Based on Monitoring Data and Secondary Modeling

Addressing the constraints inherent in traditional primary Air Quality Index (AQI) forecasting models and the shortcomings in the exploitation of meteorological data, this research introduces a novel air quality prediction methodology leveraging machine learning and the enhanced modeling of secondary data. The dataset employed encompasses forecast data on primary pollutant concentrations and primary meteorological conditions, alongside actual meteorological observations and pollutant concentration measurements, spanning from 23 July 2020 to 13 July 2021, sourced from long-term air quality projections at various monitoring stations within Jinan, China. Initially, through a rigorous correlation analysis, ten meteorological factors were selected, comprising both measured and forecasted data across five categories each. Subsequently, the significance of these ten factors was assessed and ranked based on their impact on different pollutant concentrations, utilizing a combination of univariate and multivariate significance analyses alongside a random forest approach. Seasonal characteristic analysis highlighted the distinct seasonal impacts of temperature, humidity, air pressure, and general atmospheric conditions on the concentrations of six key air pollutants. The performance evaluation of various machine learning-based classification prediction models revealed the Light Gradient Boosting Machine (LightGBM) classifier as the most effective, achieving an accuracy rate of 97.5% and an F1 score of 93.3%. Furthermore, experimental results for AQI prediction indicated the Long Short-Term Memory (LSTM) model as superior, demonstrating a goodness-of-fit of 91.37% for AQI predictions, 90.46% for O3 predictions, and a perfect fit for the primary pollutant test set. Collectively, these findings affirm the reliability and efficacy of the employed machine learning models in air quality forecasting.

Ambient Air Quality Assessment in the Limestone Mining Area of Ponjong District, Gunung Kidul, Indonesia

Limestone, a non-metallic industrial mineral, is abundantly found across Indonesia. The extraction process involves overburden removal, limestone breaking using rock breakers, loading onto dump trucks with excavators, stockpiling, and subsequent processing. These activities can negatively impact the environment, particularly air quality. To evaluate the ambient air quality in the mining area, measurements were taken focusing on hazardous substances such as SO2, CO, NO2, O3, and Total Suspended Particles (TSP). This study aims to assess the ambient air quality within the limestone mining area. Results indicate that the ambient air quality in the mining area is relatively good, with no parameters exceeding established quality standards. Specifically, SO2 levels were 9.39 µg/m³, NO2 levels were 4.19 µg/m³, CO levels were 69.74 µg/m³, O3 levels were 24.90 µg/m³, and TSP levels were 133.26 µg/m³. For a more comprehensive assessment, it is recommended to conduct extended monitoring of these parameters to ensure the long-term air quality status for both the environment and the mine workers

Role of short-term campaigns and long-term mechanisms for air pollution control: lessons learned from the "2 + 26" city cluster in China.

Protecting people from air pollution is an important task for developing countries. Over the past several decades, different types of policies have been enacted in developing countries to improve air quality, with policy impact analysis being conducted as well. As for China, while there is some existing literature on the impact of short-term action on air quality, there is much less research to investigate the roles of long-term air quality control mechanism on air quality change. This paper uses the "2 + 26" city cluster in China as a case study area to particularly analyze the roles played by short-term campaigns that compose the Winter Action Plan and a long-term mechanism referred to as the environmental accountability system in pollution control. Based on econometric and descriptive statistical analyses, it is found that the implementation of the Winter Action Plan has significantly improved the air quality, and the environmental accountability system seems to have exerted an impact, as cities with better air quality appear to establish a more stringent environmental accountability system. The paper provides the following insights for other developing countries: short-term campaigns can quickly reduce pollution and protect human health, while long-term mechanisms are needed to ensure sustainable environmental protection.

Long-Term Air Quality Evaluation System Prediction In China Based On Multinomial Logistic Regression Method

The aim of this article evaluate the long-term air quality in China based on the air quality index (AQI) and the air quality composite index (AQCI) though the multinomial logistic regression method. The two developed models employ different dependent variables, AQI and AQCI, while maintaining the same controlled variables gross domestic product (GDP), and a primary pollutant. Explicitly, the primary impurity is associated with one or more contaminants among six pollutant factors: O3, PM2.5, PM10, NO2, SO2, and CO. Model quality verification is an integral part of our analysis. The results are illustrate d using real air quality data from China. The developed models were applied to predict AQI and ACQI for the 31 capital cities in China from 2013 to 2019 annually. All calculations and tests are conducted using R-studio. In summary, both models are able to predict China’s long-term air quality. A comparison of the AQI and AQCI models using the ROC curve reveals that the AQCI model exhibits greater significance than the AQI model.

The Berlin-Brandenburg Air Study-A Methodological Study Paper of a Natural Experiment Investigating Health Effects Related to Changes in Airport-Related Exposures.

Objectives: This paper presents the study design of the Berlin-Brandenburg Air study (BEAR-study). We measure air quality in Berlin and Brandenburg before and after the relocation of aircraft (AC) traffic from Tegel (TXL) airport to the new Berlin-Brandenburg airport (BER) and investigate the association of AC-related ultrafine particles (UFP) with health outcomes in schoolchildren. Methods: The BEAR-study is a natural experiment examining schoolchildren attending schools near TXL and BER airports, and in control areas (CA) away from both airports and associated air corridors. Each child undergoes repeated school-based health-examinations. Total particle number concentration (PNC) and meteorological parameters are continuously monitored. Submicrometer particle number size distribution, equivalent black carbon, and gas-phase pollutants are collected from long-term air quality monitoring stations. Daily source-specific UFP concentrations are modeled. We will analyze short-term effects of UFP on respiratory, cardiovascular, and neurocognitive outcomes, as well as medium and long-term effects on lung growth and cognitive development. Results: We examined 1,070 children (as of 30 November 2022) from 16 schools in Berlin and Brandenburg. Conclusion: The BEAR study increases the understanding of how AC-related UFP affect children's health.

Data reliability and fault diagnostic for air quality monitoring station based on low cost sensors and active redundancy

Air pollution is both an environmental and societal issue. Conventional air quality monitoring involves costly measuring stations that demand specialized personnel. A more affordable and accessible alternative approach consists in the use of low-cost sensor networks (LCS). However, LCS have drawbacks such as variable component quality, unreliable measurements, and short lifespan. Previous research has addressed these problems through redundant LCS and expand measurement capabilities. Nevertheless, challenges remain regarding reliability. Specifically, the use of median aggregations tends to hide errors and failures, especially when most sensors are faulty. This paper proposes a new method to make data reliable and generate diagnostic through three indicators: Synthesis, Detection, and Indexes. These indicators enable the prediction of the station’s health status, identify faulty sensors, and assess the quality of the Synthesis with a confidence index. This approach ensures consistent data collection despite disturbances and malfunctions in the LCS and can be used for informed decisions and actions to mitigate air pollution. The initial implementation of this approach involves deploying the first station in the Communauté de Communes Pyrénées Vallée des Gaves (CCPVG) territory. The obtained results showed the adaptability of the proposed approach for future LCS implementations and long-term air quality monitoring.

Long-Term Assessment of PurpleAir Low-Cost Sensor for PM2.5 in California, USA

Regulatory monitoring networks are often too sparse to support community-scale PM2.5 exposure assessment, while emerging low-cost sensors have the potential to fill in the gaps. Recent advances in air quality monitoring have produced portable, easy-to-use, low-cost, sensor-based monitors which have given a new dimension to air pollutant monitoring and have democratized the air quality monitoring process by making monitors and results directly available at the community level. This study used PurpleAir © sensors for PM2.5 assessment in California, USA. The evaluation of PM2.5 from sensors included Quality Assurance and quality control (QA/QC) procedures, assessment concerning reference-monitored PM2.5 concentrations, and the formulation of a decision support system integrating these observations using geostatistical techniques. The hourly and daily average observed PM2.5 concentrations from PurpleAir monitors followed the trends of observed PM2.5 at regulatory monitors. PurpleAir monitors also captured the peak PM2.5 concentrations due to incidents such as forest fires. In comparison with reference-monitored PM2.5 levels, it was found that PurpleAir PM2.5 concentrations were mostly higher. The most important reason for PurpleAir’s higher PM2.5 concentrations was the inclusion of moisture or water vapor as an aerosol in contrast to measurements of PM2.5 excluding water content in FEM/FRM and non-FEM/FRM monitors. Long-term assessment (2016–2023) revealed that R2 values were between 0.54 and 0.86 for selected collocated PurpleAir sensors and regulatory monitors for hourly PM2.5 concentrations. Past research studies that were conducted for mostly shorter periods resulted in higher R2 values between 0.80 and 0.98. This study aims to provide reasonable estimations of PM2.5 concentrations with high spatiotemporal resolutions based on statistical models using PurpleAir measurements. The methods of Kriging and IDW, geostatistical interpolation techniques, showed similar spatio-temporal patterns. Overall, this study revealed that low-cost, sensor-based PurpleAir sensors could be effective and reliable tools for episodic and long-term ambient air quality monitoring and developing mitigation strategies.

A New Global Air Quality Health Index Based on the WHO Air Quality Guideline Values With Application in Cape Town.

Objectives: This study developed an Air Quality Health Index (AQHI) based on global scientific evidence and applied it to data from Cape Town, South Africa. Methods: Effect estimates from two global systematic reviews and meta-analyses were used to derive the excess risk (ER) for PM2.5, PM10, NO2, SO2 and O3. Single pollutant AQHIs were developed and scaled using the ERs at the WHO 2021 long-term Air Quality Guideline (AQG) values to define the upper level of the "low risk" range. An overall daily AQHI was defined as weighted average of the single AQHIs. Results: Between 2006 and 2015, 87% of the days posed "moderate to high risk" to Cape Town's population, mainly due to PM10 and NO2 levels. The seasonal pattern of air quality shows "high risk" occurring mostly during the colder months of July-September. Conclusion: The AQHI, with its reference to the WHO 2021 long-term AQG provides a global application and can assist countries in communicating risks in relation to their daily air quality.

Dataset from long-term air quality monitoring in the World Natural Heritage Geirangerfjord, Western Norway (AQM-G)

Here, we present a datapaper containing observational air quality and meteorological data related to our long-term air quality monitoring program at the UNESCO Natural World Heritage Site Geirangerfjord, Western Norway. The dataset will be updated with future data.

Revealing Long-Term Indoor Air Quality Prediction: An Intelligent Informer-Based Approach.

Indoor air pollution is an urgent issue, posing a significant threat to the health of indoor workers and residents. Individuals engaged in indoor occupations typically spend an average of around 21 h per day in enclosed spaces, while residents spend approximately 13 h indoors on average. Accurately predicting indoor air quality is crucial for the well-being of indoor workers and frequent home dwellers. Despite the development of numerous methods for indoor air quality prediction, the task remains challenging, especially under constraints of limited air quality data collection points. To address this issue, we propose a neural network capable of capturing time dependencies and correlations among data indicators, which integrates the informer model with a data-correlation feature extractor based on MLP. In the experiments of this study, we employ the Informer model to predict indoor air quality in an industrial park in Changsha, Hunan Province, China. The model utilizes indoor and outdoor temperature, humidity, and outdoor particulate matter (PM) values to forecast future indoor particle levels. Experimental results demonstrate the superiority of the Informer model over other methods for both long-term and short-term indoor air quality predictions. The model we propose holds significant implications for safeguarding personal health and well-being, as well as advancing indoor air quality management practices.