Spatial distribution characteristics of PM2.5 and PM10 in Xi'an City predicted by land use regression models

Interpreting hourly mass concentrations of PM2.5 chemical components with an optimal deep-learning model

PM2.5 constitutes a complex and diverse mixture that significantly impacts the environment, human health, and climate change. However, existing observation and numerical simulation techniques have limitations, such as a lack of data, high acquisition costs, and multiple uncertainties. These limitations hinder the acquisition of comprehensive information on PM2.5 chemical composition and effectively implement refined air pollution protection and control strategies. In this study, we developed an optimal deep learning model to acquire hourly mass concentrations of key PM2.5 chemical components without complex chemical analysis. The model was trained using a randomly partitioned multivariate dataset arranged in chronological order, including atmospheric state indicators, which previous studies did not consider. Our results showed that the correlation coefficients of key chemical components were no less than 0.96, and the root mean square errors ranged from 0.20 to 2.11 µg/m3 for the entire process (training and testing combined). The model accurately captured the temporal characteristics of key chemical components, outperforming typical machine-learning models, previous studies, and global reanalysis datasets (such as Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) and Copernicus Atmosphere Monitoring Service ReAnalysis (CAMSRA)). We also quantified the feature importance using the random forest model, which showed that PM2.5, PM1, visibility, and temperature were the most influential variables for key chemical components. In conclusion, this study presents a practical approach to accurately obtain chemical composition information that can contribute to filling missing data, improved air pollution monitoring and source identification. This approach has the potential to enhance air pollution control strategies and promote public health and environmental sustainability.

Socioeconomic Inequalities in PM2.5 Exposure and Local Source Contributions at Community Scales Using Hyper-Localized Taxi-Based Mobile Monitoring in Xi'an, China

The relationship between the socioeconomic status (SES) and PM2.5 exposure is rather inconclusive. We employed taxi-based measurements with 30 m resolution to characterize PM2.5 exposure with local source contribution (PM2.5 adjusted concentration) discerned for 2019 winter and 2020 summer, in Xi'an. A big data set comprising ∼6 × 106 hourly PM2.5 measurements and SES data from ∼5000 communities was utilized to examine the socioeconomic inequalities in community-level PM2.5 exposure. Our results indicate that the inhabitants with lower SES are more likely to be disproportionately exposed compared to those with higher SES. At least 92% of disproportionately exposed inhabitants in rural regions reside in low SES areas, whereas a relatively smaller proportion (69-78%) reside in urban regions. The local source has a more profound impact on PM2.5 exposure during summer than winter. The inhabitants in polluted areas and low PM2.5 adjusted concentration areas accounted for 22% and 26% of total PM2.5 exposure during the winter. However, inhabitants residing in low-concentration areas contributed only 12% of total exposure during summer while those polluted areas contributed 30%. These findings provide valuable insights into the relationship between community-level PM2.5 exposure and SES, highlighting the need for more sophisticated air quality policies to alleviate socioeconomic inequalities in PM2.5 exposure.

Innovative Geo-AI model: An enhance of outdoor PM estimations based on land use and outdoor environmental factors in a highly polluted area

Particulate matter (PM) is a critical component of overall pollutant exposure, but monitoring at the individual level remains impractical for large cohorts. This study aimed to identify PM sources in a highly polluted area in Taiwan and develop generalizable predictive models. We collected daily average PM data from Environmental Protection Administration (EPA) air quality monitoring stations, AirBox sensors, and EPA micro-stations in highly polluted area of Taiwan, recorded between 2018 and 2020. Predictors were derived from various datasets, including EPA environmental resources, meteorological data, land use, road traffic facilities, social information, geospatial data, and landmark databases. Employing ensemble techniques, such as land-use regression (LUR), inverse distance weighting, and three machine learning algorithms (support vector machine, random forest, and multilayer perceptron), we predicted PM2.5 and PM10 levels. The selection of important variables involved Spearman's and Kendall's Tau correlation analyses, along with stepwise regression. The optimal outdoor predictive model developed herein was an ensemble with R2 values of 0.89 for PM2.5 and 0.87 for PM10. Such models may be effective for estimating individual PM exposure in epidemiological studies and serve as a framework for other countries. Notably, our study pioneers the application of LUR models in Southern Taiwan, enriching the general prediction of atmospheric pollutant distributions. This research provides a scientific basis for urban planning, air pollution management, public health policy, and potential early warning strategies.

Towards cleaner air in urban areas: The dual influence of urban built environment factors and regional transport

Exposure to air pollution significantly elevates the risk of disease among urban populations. Improving city air quality requires not only traditional emission reduction strategies but also a focus on the intricate impacts of the urban built environment and meteorological elements. The complexity and diversity of factors within the urban built environment pose significant challenges to pollution control. This study employs machine learning to predict the spatial distribution of inhalable particulate matter (PM10) and fine particulate matter (PM2.5), integrating the clustering of pollutant-emitting enterprises and prevailing wind direction to trace pollutant sources. The results indicate that, compared to the multiple linear regression model, the R2 of the PM10 random forest prediction model improved from 0.64 to 0.88, while the RMSE decreased from 48.63 to 27.34. Similarly, the R2 of the PM2.5 increased from 0.70 to 0.92, and the RMSE decreased from 30.85 to 15.31. High concentrations of PM10 and PM2.5 in Xi'an are primarily concentrated in the northeast and southwest of the central urban area. By integrating a kernel density analysis of polluting enterprises with the analysis of prevailing wind patterns, it is evident that particulate matter in Xi'an is substantially influenced by regional urban transport. Therefore, pollution control efforts must be enhanced through coordinated regional governance. According to the analysis results of the partial dependence plot, reducing winter temperature proves beneficial in reducing PM10 and PM2.5 levels. Effective measures encompass sprinkling and humidifying, reducing traffic emissions, and controlling various dust sources to lower PM10. Enhancing ventilation, increasing green spaces, and regulating vehicle and industrial emissions effectively reduce PM2.5. The study's findings offer a scientific foundation for administrative authorities to craft pollution reduction management policies and create adaptable territorial spatial planning. Moreover, they contribute to diminishing public exposure to pollution and improving the quality of public environmental health.

High-resolution carbon emission mapping and spatial-temporal analysis based on multi-source geographic data: A case study in Xi'an City, China

Cities, contributing over 70% of global emissions, are key areas for climate change mitigation. Heterogeneity within cities determines the need for spatialized urban emissions reduction policies. However, few studies have attempted to characterize the spatial distribution of carbon emissions at the urban scale. To address this issue, a novel mapping method was proposed, using Xi'an as an example to explore the spatial distribution of carbon emissions at the city scale. Firstly, multiple geospatial open-source data, such as point of interest (POI), road networks, and land use, were utilized to identify the locations of emission sources. High-resolution carbon emission distributions were then mapped by allocating emissions based on the Intergovernmental Panel on Climate Change (IPCC) methodology. The study employed Global Moran's I to analyze the changes in spatial heterogeneity at different resolutions. Additionally, the Local Indicators of Spatial Association index (LISA) and Standard Deviation Ellipses (SDE) were adopted to examine the spatiotemporal characteristics of carbon emissions in Xi'an. The results show that carbon emissions at Xi'an City rises from 45.112 million tons to 72.701 million tons between 2010 and 2021. The construction of multi-scale carbon emissions spatial distributions, with a resolution of up to 30 m, allowed for a more detailed characterization of carbon emissions, especially in urban fringe areas. In addition, the results indicate that urban carbon emissions exhibit the strongest spatial autocorrelation at a resolution of 350 m. The study can provide a reference for the development of regional carbon emission reduction policies and spatial planning. In addition, the proposed spatialized method of city carbon emissions depends on open-source data, which allows it to have the potential for application in other cities.

Air quality prediction and control systems using machine learning and adaptive neuro-fuzzy inference system

Accurately predicting air quality concentrations is a challenging task due to the complex interactions of pollutants and their reliance on nonlinear processes. This study introduces an innovative approach in environmental engineering, employing artificial intelligence techniques to forecast air quality in Semnan, Iran. Comprehensive data on seven different pollutants was initially collected and analyzed. Then, several machine learning (ML) models were rigorously evaluated for their performance, and a detailed analysis was conducted. By incorporating these advanced technologies, the study aims to create a reliable framework for air quality prediction, with a particular focus on the case study in Iran. The results indicated that the adaptive neuro-fuzzy inference system (ANFIS) was the most effective method for predicting air quality across different seasons, showing high reliability across all datasets.

Air quality in different urban hotspots in a metropolitan city in India and the environmental implication

This research study investigates hourly data on concentrations of five major air pollutants such as particulate matter (PM10, PM2.5) and gaseous pollutants (SO2, NO2, CO) measured during 2022 at four hotspot sites (industrial site, traffic site, commercial site, harbour, and one residential site) in Chennai, India. The analysis encompasses temporal variations spanning annual, seasonal, and diurnal variations in the pollutants. Notably, PM10 and CO emerge as the predominant pollutants, with the highest concentrations at industrial and traffic sites (PM10: 67.64 ± 40.77 µg/m3, CO: 1.41 ± 0.84 mg/m3; traffic site: PM10: 58.67 ± 20.05 µg/m3, CO: 0.99 ± 0.57 mg/m3). Seasonal dynamics reveal prominent winter spikes in particulate matter (PM10, PM2.5) and carbon monoxide (CO) concentrations, while nitrogen dioxide (NO2) and sulphur dioxide (SO2) levels peak during the summer season, particularly in the harbour area. The proximity to roadways exerts a discernible influence on diurnal patterns, with traffic sites showcasing broader rush hour peaks compared to sharper spikes observed at other sites. Furthermore, distinct bimodal patterns are evident for PM10 and PM2.5 concentrations in residential and harbour areas. A common lognormal distribution pattern is identified across the studied sites, suggesting consistent air quality trends despite contrasting locations. The conditional probability function (CPF) is used in conjunction with local meteorological conditions for identifying key pollution sources in each location. The implementation of polar plots emphasizes industries as principal local sources of pollution, at industrial sites significantly contributing to PM10, SO2, and NO2 concentrations under specific wind conditions. The main objective of the present study is to facilitate a good understanding of pollutant dynamics, pollution sources, and their intricate interplay with meteorological factors, thereby contributing to the formulation and implementation of effective air pollution control and mitigation strategies.

A new attention-based CNN_GRU model for spatial-temporal PM2.5 prediction

Accurately predicting the spatial-temporal distribution of PM2.5 is challenging due to missing data and selecting an appropriate modeling method. Effective imputation of missing data must consider the relationships between variables while preserving their inherent variability and uncertainty. In this study, we employed machine learning techniques to impute missing data by analyzing the relationships between meteorological variables and other pollutants. Subsequently, we introduced an innovative spatiotemporal hybrid model, AC_GRU, which integrates a one-dimensional convolutional neural network (CNN), GRU, and an attention-based network to predict PM2.5 concentrations in urban areas. The AC_GRU model utilizes meteorological variables, PM2.5 concentrations from nearby air quality monitoring stations, and concentrations of other pollutants as inputs. This approach allows the model to learn spatiotemporal correlations within the time-series data, enhancing the accuracy of PM2.5 predictions. Additionally, the attention mechanism improves prediction accuracy by automatically weighting the past input variables based on their importance for future PM2.5 predictions. The experimental results demonstrate that our AC_GRU model outperforms state-of-the-art methods, making it a valuable tool for urban air quality management and public health protection.

PM10 exposure induces bronchial hyperresponsiveness by upreguating acetylcholine muscarinic 3 receptor

Exposure to particulate matter (PM10) can induce respiratory diseases that are closely related to bronchial hyperresponsiveness. However, the involved mechanism remains to be fully elucidated. This study aimed to demonstrate the effects of PM10 on the acetylcholine muscarinic 3 receptor (CHRM3) expression and the role of the ERK1/2 pathway in rat bronchial smooth muscle. A whole-body PM10 exposure system was used to stimulate bronchial hyperresponsiveness in rats for 2 and 4 months, accompanied by MEK1/2 inhibitor U0126 injection. The whole-body plethysmography system and myography were used to detect the pulmonary and bronchoconstrictor function, respectively. The mRNA and protein levels were determined by Western blotting, qPCR, and immunofluorescence. Enzyme-linked immunosorbent assay was used to detect the inflammatory cytokines. Compared with the filtered air group, 4 months of PM10 exposure significantly increased CHRM3-mediated pulmonary function and bronchial constriction, elevated CHRM3 mRNA and protein expression levels on bronchial smooth muscle, then induced bronchial hyperreactivity. Additionally, 4 months of PM10 exposure caused an increase in ERK1/2 phosphorylation and increased the secretion of inflammatory factors in bronchoalveolar lavage fluid. Treatment with the MEK1/2 inhibitor, U0126 inhibited the PM10 exposure-induced phosphorylation of the ERK1/2 pathway, thereby reducing the PM10 exposure-induced upregulation of CHRM3 in bronchial smooth muscle and CHRM3-mediated bronchoconstriction. U0126 could rescue PM10 exposure-induced pathological changes in the bronchus. In conclusion, PM10 exposure can induce bronchial hyperresponsiveness in rats by upregulating CHRM3, and the ERK1/2 pathway may be involved in this process. These findings could reveal a potential therapeutic target for air pollution induced respiratory diseases.

Impact of urban space on PM2.5 distribution: A multiscale and seasonal study in the Yangtze River Delta urban agglomeration

Despite concerted efforts in emission control, air pollution control remains challenging. Urban planning has emerged as a crucial strategy for mitigating PM2.5 pollution. What remains unclear is the impact of urban form and their interactions with seasonal changes. In this study, base on the air quality monitoring stations in the Yangtze River Delta urban agglomeration, the relationship between urban spatial indicators (building morphology and land use) and PM2.5 concentrations was investigated using full subset regression and variance partitioning analysis, and seasonal differences were further analysed. Our findings reveal that PM2.5 pollution exhibits different sensitivities to spatial scales, with higher sensitivity to the local microclimate formed by the three-dimensional structure of buildings at the local scale, while land use exerts greater influence at larger scales. Specifically, land use indicators contributed sustantially more to the PM2.5 prediction model as buffer zone expand (from an average of 2.41% at 100 m range to 47.30% at 5000 m range), whereas building morphology indicators display an inverse trend (from an average of 13.84% at 100 m range to 1.88% at 5000 m range). These results enderscore the importance of considering building morphology in local-scale urban planning, where the increasing building height can significantly enhance the disperion of PM2.5 pollution. Conversely, large-scale urban planning should prioritize the mixed use of green spaces and construction lands to mitigate PM2.5 pollution. Moreover, the significant seasonal differences in the ralationship between urban spatical indicatiors and PM2.5 pollution were observed. Particularly moteworthy is the heightened association between forest, water indicators and PM2.5 concentrations in summer, indicating the urban forests may facilitate the formation of volatile compunds, exacerbating the PM2.5 pollution. Our study provides a theoretical basis for addressing scale-related challenges in urban spatial planning, thereby forstering the sustainable development of cities.

The comprehensive impact of thermal-PM2.5 interaction on subjective evaluation of urban outdoor space: A pilot study in a cold region of China

Urban outdoor space has a very important impact on the quality of people's outdoor activities, which has influenced people's health and moods. Its influence is the result of the combined action of various factors. Thermal and air quality environment are important factors affecting the overall comfort of the urban outdoor space. At present, there are few research on interaction with thermal and air quality environment. Therefore, a meteorological measurement and questionnaire survey have been conducted in a representative open space in a campus in Xi'an, China. The following are the research results:(1) Mean physiological equivalent temperature (MPET) is a significant factor affecting thermal sensation vote (TSV) and thermal comfort vote (TCV). PM2.5 has no significant effect on thermal comfort vote (TCV), but it is a considerable factor affecting thermal sensation vote (TSV) when 10.2°C ≤ MPET<21°C (P = 0.023 *). (2) PM2.5 is a significant factor affecting air quality vote (AQV) and breathing comfort vote (BCV).Mean physiological equivalent temperature (MPET) has no significant impact on air quality vote (AQV), but it is a considerable factor affecting breathing comfort vote (BCV) when 10.2°C ≤ MPET<21°C (P = 0.01 **). (3) Mean physiological equivalent temperature (MPET) is a significant factor affecting overall comfort vote (OCV), but PM2.5 is not. In general, When 10.2°C ≤ MPET<21°C (-0.5 < -0.37 ≤ TCV ≤ 0.12 <0.5), the interaction between thermal and PM2.5 environment is significant on thermal sensation vote (TSV) and breathing comfort vote (BCV). This study can provide experimental support for the field of multi-factor interaction, which has shown that improving the thermal environment can better breathing comfort, while reducing PM2.5 concentration can promote thermal comfort. And can also provide reference for the study of human subjective comfort in urban outdoor space in the same latitude of the world.

A comprehensive review of the development of land use regression approaches for modeling spatiotemporal variations of ambient air pollution: A perspective from 2011 to 2023

Land use regression (LUR) models are widely used in epidemiological and environmental studies to estimate humans' exposure to air pollution within urban areas. However, the early models, developed using linear regressions and data from fixed monitoring stations and passive sampling, were primarily designed to model traditional and criteria air pollutants and had limitations in capturing high-resolution spatiotemporal variations of air pollution. Over the past decade, there has been a notable development of multi-source observations from low-cost monitors, mobile monitoring, and satellites, in conjunction with the integration of advanced statistical methods and spatially and temporally dynamic predictors, which have facilitated significant expansion and advancement of LUR approaches. This paper reviews and synthesizes the recent advances in LUR approaches from the perspectives of the changes in air quality data acquisition, novel predictor variables, advances in model-developing approaches, improvements in validation methods, model transferability, and modeling software as reported in 155 LUR studies published between 2011 and 2023. We demonstrate that these developments have enabled LUR models to be developed for larger study areas and encompass a wider range of criteria and unregulated air pollutants. LUR models in the conventional spatial structure have been complemented by more complex spatiotemporal structures. Compared with linear models, advanced statistical methods yield better predictions when handling data with complex relationships and interactions. Finally, this study explores new developments, identifies potential pathways for further breakthroughs in LUR methodologies, and proposes future research directions. In this context, LUR approaches have the potential to make a significant contribution to future efforts to model the patterns of long- and short-term exposure of urban populations to air pollution.

LCZ method is more effective than traditional LUCC method in interpreting the relationship between urban landscape and atmospheric particles

Landscape classification methods significantly impact the exploration of the mechanism of the relationship between landscapes and atmospheric particulate matter. This study compared the local climate zones (LCZs) and traditional land use/cover change (LUCC) landscape classification methods in explaining spatial differences in concentrations of atmospheric particulate matter (PM_2.5 and PM₁₀) and explored the mechanisms involved in how landscape elements affect atmospheric particulate matter. This was done by establishing a PM_2.5 and PM₁₀ land use regression (LUR) model of LCZ and LUCC landscapes under low, typical, and high particle concentration gradients in urban and suburban areas. The results show that under an LCZ classification system, the number of patches in the urban area of Shanghai was 548 times higher than that of a LUCC system. Moreover, LCZs were successfully established for LUR models in 12 scenarios, while only five models were established for LUCC, all of which were suburban models. The R² of the LUR model based on the LCZ landscape and atmospheric particulate matter was mostly higher than that of the LUCC. For unnatural landscapes, the LUCC demonstrated that an urbanized environment positively affects the concentration of atmospheric particles. However, the LCZ analysis found that areas with high-density buildings have a positive effect on atmospheric particles, while most areas with low-density buildings significantly reduced the number of atmospheric particles present. Generally, compared with the traditional LUCC landscape classification method, LCZ integrates Shanghai's physical structure and classifies the urban landscape more accurately, which is closely related to the urban atmospheric particulate matter, especially in the urban area. Because the low-density building area has the same effect on the particulate matter as the natural landscape, the use of low-density buildings is recommended when planning new towns.

The seasonal variation, characteristics and secondary generation of PM2.5 in Xi'an, China, especially during pollution events

As an important central city in western China, Xi'an has the worst atmospheric pollution record in China and many measures have been taken to improve the air quality in the past few years. In this study, PM_2.5 samples were collected across four seasons from 2017 to 2018 in Xi'an. Organic carbon and elemental carbon, water soluble ions, and elements were monitored to assess the air quality. The average annual PM_2.5 concentration was (134.9 ± 48.1 μg/m³), with the highest concentration in winter (188.8 ± 93.2 μg/m³), and lowest concentration in summer (71.2 ± 12.1 μg/m³). The secondary generation of sulfate (SO₄^2-) and nitrate (NO₃^-) was strong in spring, and secondary organic carbon (SOC) was formed in all seasons. The compositions of PM_2.5 changed greatly during a sandstorm occurred and the Spring Festival. The sandstorm played a positive role in removing local pollutant NO₃^-, but also increased the concentration of SO₄^2-, however both the concentration of SO₄^2- and NO₃^- greatly increased by secondary generation during Spring Festival. Potential source analysis showed that during the sandstorm, pollutants were transported over a long distance from the northwest of China, whereas it was mainly from the local and surrounded emissions during the Spring Festival. Except Ca²⁺ and geological dust (GM), the other components in PM_2.5 increased significantly on the day of the Spring Festival. During sampling time in Xi'an, the positive matrix factorization (PMF) model analysis showed that PM_2.5 mainly came from vehicle emission, coal combustion, and biomass burning.

Greenwashing in the US metal industry? A novel approach combining SO2 concentrations from satellite data, a plant-level firm database and web text mining

This study deals with the issue of greenwashing, i.e. the false portrayal of companies as environmentally friendly. The analysis focuses on the US metal industry, which is a major emission source of sulfur dioxide (SO₂), one of the most harmful air pollutants. One way to monitor the distribution of atmospheric SO₂ concentrations is through satellite data from the Sentinel-5P programme, which represents a major advance due to its unprecedented spatial resolution. In this paper, Sentinel-5P remote sensing data was combined with a plant-level firm database to investigate the relationship between the US metal industry and SO₂ concentrations using a spatial regression analysis. Additionally, this study considered web text data, classifying companies based on their websites in order to depict their self-portrayal on the topic of sustainability. In doing so, we investigated the topic of greenwashing, i.e. whether or not a positive self-portrayal regarding sustainability is related to lower local SO₂ concentrations. Our results indicated a general, positive correlation between the number of employees in the metal industry and local SO₂ concentrations. The web-based analysis showed that only 8% of companies in the metal industry could be classified as engaged in sustainability based on their websites. The regression analyses indicated that these self-reported "sustainable" companies had a weaker effect on local SO₂ concentrations compared to their "non-sustainable" counterparts, which we interpreted as an indication of the absence of general greenwashing in the US metal industry. However, the large share of firms without a website and lack of specificity of the text classification model were limitations to our methodology.

Supporting Design to Develop Rural Revitalization through Investigating Village Microclimate Environments: A Case Study of Typical Villages in Northwest China

China has the largest number of villages in the world, and research on rural microclimate will contribute to global climate knowledge. A three-by-three grid method was developed to explore village microclimates through field measurement and ENVI-met simulation. A regression model was used to explore the mechanistic relationship between microclimate and spatial morphology, and predicted mean vote (PMV) was selected to evaluate outdoor thermal comfort. The results showed that ENVI-met was able to evaluate village microclimate, as Pearson’s correlation coefficient was greater than 0.8 and mean absolute percentage error (MAPE) was from 2.16% to 3.79%. Moreover, the air temperature of west–east road was slightly higher than that of south–north, especially in the morning. The height-to-width ratio (H/W) was the most significant factor to affect air temperature compared to percentage of building coverage (PBC) and wind speed. In addition, H/W and air temperature had a relatively strong negative correlation when H/W was between 0.52 and 0.93. PMV indicated that the downwind edge area of prevailing wind in villages was relatively comfortable. This study provides data support and a reference for optimizing village land use, mediating the living environment, and promoting rural revitalization.

Do We Need More Urban Green Space to Alleviate PM2.5 Pollution? A Case Study in Wuhan, China

Urban green space can help to reduce PM2.5 concentration by absorption and deposition processes. However, few studies have focused on the historical influence of green space on PM2.5 at a fine grid scale. Taking the central city of Wuhan as an example, this study has analyzed the spatiotemporal trend and the relationship between green space and PM2.5 in the last two decades. The results have shown that: (1) PM2.5 concentration reached a maximum value (139 μg/m3) in 2010 and decreased thereafter. Moran’s I index values of PM2.5 were in a downward trend, which indicates a sparser distribution; (2) from 2000 to 2019, the total area of green space decreased by 25.83%. The reduction in larger patches, increment in land cover diversity, and less connectivity led to fragmented spatial patterns of green space; and (3) the regression results showed that large patches of green space significantly correlated with PM2.5 concentration. The land use/cover diversity negatively correlated with the PM2.5 concentration in the ordinary linear regression. In conclusion, preserving large native natural habitats can be a supplemental measure to enlarge the air purification function of the green space. For cities in the process of PM2.5 reduction, enhancing the landscape patterns of green space provides a win-win solution to handle air pollution and raise human well-being.

Spatiotemporal estimation of the PM2.5 concentration and human health risks combining the three-dimensional landscape pattern index and machine learning methods to optimize land use regression modeling in Shaanxi, China

PM_2.5 pollution endangers human health and urban sustainable development. Land use regression (LUR) is one of the most important methods to reveal the temporal and spatial heterogeneity of PM_2.5, and the introduction of characteristic variables of geographical factors and the improvement of model construction methods are important research directions for its optimization. However, the complex non-linear correlation between PM_2.5 and influencing indicators is always unrecognized by the traditional regression model. The two-dimensional landscape pattern index is difficult to reflect the real information of the surface, and the research accuracy cannot meet the requirements. As such, a novel integrated three-dimensional landscape pattern index (TDLPI) and machine learning extreme gradient boosting (XGBOOST) improved LUR model (LTX) are developed to estimate the spatiotemporal heterogeneity in the fine particle concentration in Shaanxi, China, and health risks of exposure and inhalation of PM_2.5 were explored. The LTX model performed well with R² = 0.88, RMSE of 8.73 μg/m³ and MAE of 5.85 μg/m³. Our findings suggest that integrated three-dimensional landscape pattern information and XGBOOST approaches can accurately estimate annual and seasonal variations of PM_2.5 pollution The Guanzhong Plain and northern Shaanxi always feature high PM_2.5 values, which exhibit similar distribution trends to those of the observed PM_2.5 pollution. This study demonstrated the outstanding performance of the LTX model, which outperforms most models in past researches. On the whole, LTX approach is reliable and can improve the accuracy of pollutant concentration prediction. The health risks of human exposure to fine particles are relatively high in winter. Central part is a high health risk area, while northern area is low. Our study provides a new method for atmospheric pollutants assessing, which is important for LUR model optimization, high-precision PM_2.5 pollution prediction and landscape pattern planning. These results can also contribute to human health exposure risks and future epidemiological studies of air pollution.

Influence of Land Use and Meteorological Factors on PM2.5 and PM10 Concentrations in Bangkok, Thailand

Particulate matter (PM) is regarded a major problem worldwide because of the harm it causes to human health. Concentrations of PM with particle diameter less than 2.5 µm (PM2.5) and with particle diameter less than 10 µm (PM10) are based on various emission sources as well as meteorological factors. In Bangkok, where the PM2.5 and PM10 monitoring stations are few, the ability to estimate concentrations at any location based on its environment will benefit healthcare policymakers. This research aimed to study the influence of land use, traffic load, and meteorological factors on the PM2.5 and PM10 concentrations in Bangkok using a land-use regression (LUR) approach. The backward stepwise selection method was applied to select the significant variables to be included in the resultant models. Results showed that the adjusted coefficient of determination of the PM2.5 and PM10 LUR models were 0.58 and 0.57, respectively, which are in the same range as reported in the previous studies. The meteorological variables included in both models were rainfall and air pressure; wind speed contributed to only the PM2.5 LUR model. Further, the land-use types selected in the PM2.5 LUR model were industrial and transportation areas. The PM10 LUR model included residential, commercial, industrial, and agricultural areas. Traffic load was excluded from both models. The root mean squared error obtained by 10-fold cross validation was 9.77 and 16.95 for the PM2.5 and PM10 LUR models, respectively.

Architectural Simulations on Spatio-Temporal Changes of Settlement Outdoor Thermal Environment in Guanzhong Area, China

This paper aims to provide data support for rural sustainable development through analyzing the spatio-temporal characteristics of the interactions of the outdoor thermal environment. The ordinary and representative rural settlements in the Guanzhong area were selected to analyze the dynamic process of the rural thermal environment through field measurements and numerical simulations. RMSE (root mean square error) and MAPE (mean absolute percentage) were used to verify the numerical simulation model, and physiological equivalent temperature (PET) was used to evaluate the outdoor thermal environment. Results show that the ENVI-met model reliably predicts the thermal environment of a rural settlement, as the air temperature and relative humidity values range of the RMSE and MAPE were 0.85–1.79 and 2.04–5.11%, respectively. Moreover, the air temperature rose by 3.08% and relative humidity dropped by 4.42% from 2003 to 2018 as the amount of artificial surfaces increased by 35.4% and the PET index gradually increased by 27.43% at daytime and 34.03% at nighttime. Furthermore, trees could improve the outdoor thermal environment significantly, mainly because the average air temperature decreased by 3.6% and relative humidity increased by 8%, and the PET index decreased by 12.4% and 13.1%, respectively, for daytime and nighttime. This case study is representative of rural settlements in the Guanzhong plain, and thus is an appeal to rural planners to pay attention to the thermal environment issues caused by increased artificial underlay surfaces and to focus on trees in rural areas.

Land Use Quantile Regression Modeling of Fine Particulate Matter in Australia

Small data samples are still a critical challenge for spatial predictions. Land use regression (LUR) is a widely used model for spatial predictions with observations at a limited number of locations. Studies have demonstrated that LUR models can overcome the limitation exhibited by other spatial prediction models which usually require greater spatial densities of observations. However, the prediction accuracy and robustness of LUR models still need to be improved due to the linear regression within the LUR model. To improve LUR models, this study develops a land use quantile regression (LUQR) model for more accurate spatial predictions for small data samples. The LUQR is an integration of the LUR and quantile regression, which both have advantages in predictions with a small data set of samples. In this study, the LUQR model is applied in predicting spatial distributions of annual mean PM2.5concentrations across the Greater Sydney Region, New South Wales, Australia, with observations at 19 valid monitoring stations in 2020. Cross validation shows that the goodness-of-fit can be improved by 25.6–32.1% by LUQR models when compared with LUR, and prediction root mean squared error (RMSE) and mean absolute error (MAE) can be reduced by 10.6–13.4% and 19.4–24.7% by LUQR models, respectively. This study also indicates that LUQR is a more robust model for the spatial prediction with small data samples than LUR. Thus, LUQR has great potentials to be widely applied in spatial issues with a limited number of observations.

Analysis of the spatial differentiation and scale effects of the three-dimensional architectural landscape in Xi'an, China

Three-dimensional landscape patterns are an effective means to study the relationship between landscape pattern evolution and eco-environmental effects. This paper selects six districts in Xi'an as the study area to examine the spatial distribution characteristics of the three-dimensional architectural landscape in the city's main urban area using three-dimensional information on the buildings in 2020 with the support of GIS. In this study, two new architectural landscape indices-landscape height variable coefficient and building rugosity index-were employed in landscape pattern analysis, whilst a system of rigorous and comprehensive three-dimensional architectural landscape metrics was established using principal component analysis. A mathematical model of weighted change of landscape metrics based on the objective weighting method was applied to carry out scale analysis of the landscape patterns. Spatial statistical analysis and spatial autocorrelation analysis were conducted to comprehensively study the differentiation of three-dimensional architectural landscape spatial patterns. The results show that the characteristic scale of the three-dimensional landscape pattern in Xi'an's main urban area is around 8 km. Moreover, the three-dimensional landscape of the buildings in this area is spatially positively correlated, exhibiting a high degree of spatial autocorrelation whilst only showing small spatial differences. The layout of the architectural landscape pattern is disorderly and chaotic within the second ring, whilst the clustering of patch types occurs near the third ring. Moreover, the building density in the Beilin, Lianhu, and Xincheng districts is large, the building height types are rich, and the roughness of the underlying surface is high, such that these are key areas to be improved through urban renewal. The height, volume, density, morphological heterogeneity, and vertical roughness of the architectural landscape vary amongst functional areas within the study area. This paper is the first to apply the study of spatial heterogeneity of three-dimensional landscape patterns to Xi'an. It does so in order to provide a quantitative basis for urban landscape ecological design for urban renewal and the rational planning of built-up areas, which will promote the sustainable development of the city's urban environment.

Estimating PM2.5 concentration using the machine learning GA-SVM method to improve the land use regression model in Shaanxi, China

With rapid economic growth, urbanization and industrialization, fine particulate matter with aerodynamic diameters ≤ 2.5 µm (PM_2.5) has become a major pollutant and shows adverse effects on both human health and the atmospheric environment. Many studies on estimating PM_2.5 concentrations have been performed using statistical regression models and satellite remote sensing. However, the accuracy of PM_2.5 concentration estimates is limited by traditional regression models; machine learning methods have high predictive power, but fewer studies have been performed on the complementary advantages of different approaches. This study estimates PM_2.5 concentrations from satellite remote sensing-derived aerosol optical depth (AOD) products, meteorological data, terrain data and other predictors in 2015 in Shaanxi, China, using a combined genetic algorithm-support vector machine (GA-SVM) method, after which the spatial clustering pattern was explored at the season and year levels. The results indicated that temperature (r = -0.684), precipitation (r = -0.602) and normalized difference vegetation index (NDVI) (r = -0.523) were significantly negatively correlated with the PM_2.5 concentration, while AOD (r = 0.337) was significantly positively correlated with the PM_2.5 concentration. Compared to conventional land use regression (LUR) and SVM models and previous related studies, the GA-SVM method demonstrated a significantly better prediction accuracy of PM_2.5 concentration, with a higher 10-fold cross-validation coefficient of determination (R²) of 0.84 and lower root mean square error (RMSE) and mean absolute error (MAE) of 12.1 μg/m³ and 10.07 μg/m³, respectively. Y-scrambling test shows that the models have no chance correlation. The central and southern parts of Shaanxi have high PM_2.5 concentrations, which are mainly due to the pollutant emissions and meteorological and topographical conditions in those areas. There was a positive spatial agglomeration characteristic of regional PM_2.5 pollution, and the spatial spillover effect of PM_2.5 pollution for seasonal and annual variations does exist. In general, the GA-SVM method is robust and accurately estimates PM_2.5 concentrations via a novel modeling framework application and high-quality spatiotemporal information. It also has great significance for the exploration of PM_2.5 pollution estimation and high-precision mapping methods, especially early warning in high-risk areas. Finally, the prevention and control of atmospheric pollution should take pollution control measures from major cities and surrounding cities, and focus on the joint pollution control measures for plain cities.

PM2.5 Pollutant Concentrations in Greenspaces of Nanjing Are High but Can Be Lowered with Environmental Planning

Small-scale greenspaces in high-density central urban districts serve as important outdoor activity spaces for the surrounding residents, especially the elderly. This study selects six small-scale, popular greenspaces with distinct characteristics that are jointly situated along the same main urban artery in a high-density central urban district. Field investigations and questionnaires are conducted and combined with statistical analyses, to explore the spatial-temporal distribution and influencing factors of PM2.5 concentrations in these greenspaces. The study finds that the air quality conditions in the sites are non-ideal, and this has potential negative impacts on the health of the elderly visitors. Moreover, the difference values of PM2.5 concentrations' spatial-temporal distributions are significantly affected by vehicle-related emissions, which have significant temporal characteristics. PM2.5 concentration is strongly correlated with percentage of green coverage (R = 0.82, p < 0.05), degree of airflow (R = -0.83, p < 0.05), humidity and comfort level (R = 0.54, p < 0.01 and R = -0.40, p < 0.01 respectively). Meanwhile, the sites' "sky view factor" is strongly correlated with degree of airflow (R = 0.82, p < 0.05), and the comfort level plays an indirect role in the process of PM2.5 affecting crowd activities. Based on this analysis, an optimal set of index ranges for greenspace elements which are correlated with the best reduction in PM2.5 concentrations is derived. As such, this research reveals the technical methods to best reduce their concentrations and provides a basis and reference for improving the quality of small-scale greenspaces in high-density urban districts for the benefit of healthy aging.

Effect of urban underlying surface on PM2.5 vertical distribution based on UAV in Xi'an, China

Fine particulate matter (PM_2.5) has become a significant issue of ecological environment. However, few studies have explored the vertical distribution of PM_2.5 in cities. The objectives of this paper are to reveal the vertical distribution regular pattern of PM_2.5 over urban underlying surfaces near the ground with a hexacopter-type unmanned aerial vehicle (UAV) in winter. Results showed that the maximum vertical gradient of PM_2.5 near the ground was typically the greatest in the morning as the stable atmospheric conditions. Moreover, regression model illustrated that relative humidity had the greatest impact on the vertical profile of PM_2.5 compared to air temperature and altitude as hygroscopic of PM_2.5 aerosols. Curve model shown that vertical profile of PM_2.5 over the surfaces of water and green space first increased slowly and then declined, besides, the highest concentration inflection of PM_2.5 above the water body (23.7 m) is higher than the green space (14.3 m). Thus, suggesting residents living vertical of 10-30 m from the ground around large water bodies and green spaces should not open windows for ventilation in the morning. Therefore, this study provides insights into the vertical distributions of PM_2.5 over different underlying surfaces and should be of reference value to urban planners for designing urban spaces to optimize atmosphere environment to provide a healthy living environment.

A Novel Recursive Model Based on a Convolutional Long Short-Term Memory Neural Network for Air Pollution Prediction

Deep learning provides a promising approach for air pollution prediction. The existing deep learning-based predicted models generally consider either the temporal correlations of air quality monitoring stations or the nonlinear relationship between the PM2.5 (particulate matter with an aerodynamic diameter of less than 2.5 μm) concentrations and explanatory variables. Spatial correlation has not been effectively incorporated into prediction models, therefore exhibiting poor performance in PM2.5 prediction tasks. Additionally, determining the manner by which to expand longer-term prediction tasks is still challenging. In this paper, to allow for spatiotemporal correlations, a spatiotemporal convolutional recursive long short-term memory (CR-LSTM) neural network model is proposed for predicting the PM2.5 concentrations in long-term prediction tasks by combining a convolutional long short-term memory (ConvLSTM) neural network and a recursive strategy. Herein, the ConvLSTM network was used to capture the complex spatiotemporal correlations and to predict the future PM2.5 concentrations; the recursive strategy was used for expanding the long-term prediction tasks. The CR-LSTM model was used to realize the prediction of the future 24 h of PM2.5 concentrations for 12 air quality monitoring stations in Beijing by configuring both the appropriate time lag derived from the temporal correlations and the spatial neighborhood, including the hourly historical PM2.5 concentrations, the daily mean meteorological data, and the annual nighttime light and normalized difference vegetation index (NDVI). The results showed that the proposed CR-LSTM model achieved better performance (coefficient of determination (R2) = 0.74; root mean square error (RMSE) = 18.96 μg/m3) than other common models, such as multiple linear regression (MLR), support vector regression (SVR), the conventional LSTM model, the LSTM extended (LSTME) model, and the temporal sliding LSTM extended (TS-LSTME) model. The proposed CR-LSTM model, implementing a combination of geographical rules, recursive strategy, and deep learning, shows improved performance in longer-term prediction tasks.

Air Pollution Measurements and Land-Use Regression in Urban Sub-Saharan Africa Using Low-Cost Sensors—Possibilities and Pitfalls

Air pollution is recognized as the most important environmental factor that adversely affects human and societal wellbeing. Due to rapid urbanization, air pollution levels are increasing in the Sub-Saharan region, but there is a shortage of air pollution monitoring. Hence, exposure data to use as a base for exposure modelling and health effect assessments is also lacking. In this study, low-cost sensors were used to assess PM2.5 (particulate matter) levels in the city of Adama, Ethiopia. The measurements were conducted during two separate 1-week periods. The measurements were used to develop a land-use regression (LUR) model. The developed LUR model explained 33.4% of the variance in the concentrations of PM2.5. Two predictor variables were included in the final model, of which both were related to emissions from traffic sources. Some concern regarding influential observations remained in the final model. Long-term PM2.5 and wind direction data were obtained from the city’s meteorological station, which should be used to validate the representativeness of our sensor measurements. The PM2.5 long-term data were however not reliable. Means of obtaining good reference data combined with longer sensor measurements would be a good way forward to develop a stronger LUR model which, together with improved knowledge, can be applied towards improving the quality of health. A health impact assessment, based on the mean level of PM2.5 (23 µg/m3), presented the attributable burden of disease and showed the importance of addressing causes of these high ambient levels in the area.