Concentration Levels Of Air Pollutants Research Articles

Enhancing air pollution prediction: A neural transfer learning approach across different air pollutants

Air pollution stands out as one of the most alarming environmental challenges. It poses significant risks to human health and the environment. Accurate forecasting of air pollutant concentration levels is crucial for effective air quality management and timely implementation of mitigation strategies. In this paper, the transfer learning technique is investigated using the artificial neural network (ANN), also called multi-layer perception (MLP), to transfer knowledge across different air pollutants forecasting, and therefore, to generalize over a large set of air pollutants in the same air monitoring station. By leveraging the knowledge learned from a source forecasting task, transfer learning allows us to reduce the data requirements, speed up the training of the models, and enhance the predictive performance for different air pollutants for the target forecasting task. We present a comprehensive analysis of the transfer learning across different air pollutants in the same air monitoring station on a large dataset of air quality measurements. Our results demonstrate that transfer learning significantly improves forecasting accuracy with fewer fine-tuning data, particularly when limited labeled data is available for the target task. The findings of this study contribute to the advancement of air pollution forecasting methodologies, facilitating better decision-making processes and proactive air quality management.

Evaluation of Deep Learning Models for Predicting the Concentration of Air Pollutants in Urban Environments

Air pollution is an issue of great concern globally due to the risks to the health of humanity, animals, and ecosystems. On the one hand, air quality monitoring systems allow for determining the concentration level of air pollutants and health risks through an air quality index (AQI). On the other hand, accurate future predictions of air pollutant concentration levels can provide valuable information for data-driven decision-making to reduce health risks from short- and long-term exposure when indicators exceed permissible limits. In this paper, five deep learning architectures are evaluated to predict the concentration of particulate matter pollutants (in their fractions PM2.5 and PM10) and carbon monoxide (CO) in consecutive hours. The proposed prediction models are based on recurrent neural networks (RNNs), long short-term memory (LSTM), vanilla LSTM, Stacked LSTM, Bi-LSTM, and encoder–decoder LSTM networks. Moreover, a methodology is presented to guide the construction of the prediction model, encompassing raw data processing, model design and optimization, and neural network training, testing, and evaluation. The results underscore the precision and reliability of the Stacked LSTM model in predicting the hourly concentration level for PM2.5, with an RMSE of 3.4538 μg/m3. Similarly, the encoder–decoder LSTM model accurately predicts the concentration level for PM10 and CO, with an RMSE of 3.2606 μg/m3 and 2.1510 ppm, respectively. These evaluations, with their minimal differences in error metrics and coefficient of determination, validate the effectiveness and superiority of the deep learning models over other reference models, instilling confidence in their potential.

Impact of indoor air pollutant concentration levels on the health of press operators of printing facilities in Zaria Metropolis, Nigeria

Impact of indoor air pollutant concentration levels on the health of press operators of printing facilities in Zaria Metropolis, Nigeria

A Short-Term Air Pollutant Concentration Forecasting Method Based on a Hybrid Neural Network and Metaheuristic Optimization Algorithms

In the contemporary era, global air quality has been adversely affected by technological progress, urban development, population expansion, and the proliferation of industries and power plants. Recognizing the urgency of addressing air pollution consequences, the prediction of the concentration levels of air pollutants has become crucial. This study focuses on the short-term prediction of nitrogen dioxide (NO2) and sulfur dioxide (SO2), prominent air pollutants emitted by the Kerman Combined Cycle Power Plant, from May to September 2019. The proposed method utilizes a new two-step feature selection (FS) process, a hybrid neural network (HNN), and the Coot optimization algorithm (COOT). This combination of FS and COOT selects the most relevant input features while eliminating redundant ones, leading to improved prediction accuracy. The application of HNN for training further enhances the accuracy significantly. To assess the model’s performance, two datasets, including real data from two different parts of Combined Cycle Power Plant in Kerman, Iran, from 1 May 2019 to 30 September 2019 (namely dataset A and B), are utilized. Subsequently, mean square error (MSE), mean absolute error (MAE), root mean square deviation (RMSE), and mean absolute percentage error (MAPE) were employed to obtain the accuracy of FS-HNN-COOT. Experimental results showed MSE of FS-HNN-COOT for NO2 ranged from 0.002 to 0.005, MAE from 0.016 to 0.0492, RMSE from 0.0142 to 0.0736, and MAEP from 4.21% to 8.69%. Also, MSE, MAE, RMSE, and MAPE ranged from 0.0001 to 0.0137, 0.0108 to 0.0908, 0.0137 to 0.1173, and 9.03% to 15.93%, respectively, for SO2.

Variability in molecular composition and optical absorption of atmospheric brown carbon aerosols in two contrasting urban areas of China

Atmospheric brown carbon (BrC) aerosols were investigated at two urban sites in southern (Hefei) and northern (Shijiazhuang) China during summer and winter of 2019–2020 to explore regional variability in their compositional and optical properties. Organic matter in ambient PM2.5 samples were characterized at molecular level using ultrahigh performance liquid chromatography coupled with a diode array detector and an Orbitrap mass spectrometer. Although the molecular composition of organic aerosols varied substantially over different ambient environments, they were mainly composed by CHO and CHON species in positive ionization mode while CHO and CHOS species in negative mode. The mass absorption coefficients of BrC aerosols at wavelength range 250–450 nm were relatively higher for winter samples in both cities and for Shijiazhuang samples in both seasons, partly attributed to the higher concentration levels of anthropogenic air pollutants in these environments. The absorption Ångström exponents further revealed that BrC aerosols in winter seasons and in Shijiazhuang had a greater capacity of absorption at shorter wavelengths. A total of 26 BrC species with strong absorption were unambiguously identified from different environments, which mainly consisted of CHO, CHON, and CHN species and had higher degrees of unsaturation and lower degrees of oxidation. The presence and abundance of these BrC species varied dynamically across the seasons and cities, with a greater number of species presented in the winter of Shijiazhuang. The BrC species together contributed 12–26 % in the total absorbance of light-absorbing organic components at 250–450 nm. This study highlights the regional differences in BrC properties influenced by the sources and atmospheric processes, which should be taken into account to assess their climate impacts.

Ambient Air Quality in Upper Silesia Region Pre-During, and Post-COVID-19 Periods

Abstract Governments worldwide have established lockdowns to limit the spread of COVID-19 during the pandemic. The restrictions on travel and reduction of economic activity have led to a temporary improvement in air quality in several countries, especially in urban areas. This study investigates the changes in concentration levels of air pollutants (PM10, PM2.5, SO2, NO2, and bacterial aerosol) in the Upper Silesia Region of Southern Poland, during three periods: pre (March 2018 and 2019), during (March 2020, and 2021) and post-COVID-19 lockdown period (March 2022 and 2023). Our findings indicate that COVID-19 restrictions had a moderate impact on PM10 levels in comparison to pre- and post-COVID-19 periods. PM2.5 during lockdown was significantly lower than in the pre-COVID period and not significantly higher after COVID. PM10 and PM2.5 average concentrations decreased during COVID-19 restrictions by 27.8% and 12.7%, respectively. Compared with the results from the pre-COVID-19 phase, the reductions in NO2 and SO2 during the lockdown were 9.5% and 34.0%. Among other pollutants, bacterial aerosol (BA) concentrations also decreased during the lockdown by 23.0%, compared to the results from the pre-COVID-19 period.

Correlation between the Prevalence of Sick-Building Syndrome and Safe Indoor Air Quality Concept in Private Residential Housing in Jordan

Indoor air quality (IAQ) and related health problems have witnessed remarkable attention recently. The prevalence of sick-building syndrome (SBS) is considered the most common health issue. This study conducted in the Al-Dahrieh neighborhood in Jordan showed for the first time how indoor air quality (IAQ) factors affect the prevalence of sick-building syndrome among occupants in residential buildings. The study investigated the concentration levels of air pollutants and comfort parameters. Architectural and urban design configurations were collected through site observation. In addition, daily activities for occupants were gathered through an online questionnaire. All statistical and descriptive analyses of the data collected for this study were carried out by Spearman’s rho correlation test (SPSS) and Excel 2016. It was done using two-tailed (2-tailed) tests and a 1% statistical significance level ( p < 0.01); interestingly, all expected parameters checked using SPSS are acceptable according to the significant factor of p < 0.05. The research explored low air quality in the selected case studies and suggested simple mitigation strategies to reduce pollutants concentration in the buildings, such as natural ventilation and control of pollution from internal sources. Moreover, architects may take these findings to enhance neighborhood and building design to achieve the goal of constructing healthier buildings.

Evaluation of the prediction performance of air quality numerical forecast models in Shenzhen

As an economically intensive city within the greater bay area (GBA), Shenzhen has suffered from severe ozone pollution in recent years. Numerical modeling is an important tool in daily air quality prediction. However, a combination of factors, such as outdated emission inventories and the uncertainty of weather and numeric models, pose challenges for accurately forecasting pollution periods and levels. Therefore, systematic evaluation of the numerical model performance is crucial for better manual revised prediction of the daily MAD8h and air quality index (AQI). In this paper, a multimodel air quality forecast system, which was built in our public-owned monitoring center station, was evaluated to understand the possibilities of improvements. Based on the 2021 observed data of the national air monitoring station, we utilized a series of metrics to access the predicted air quality index (AQI) and pollution concentration levels from models (including WRF-CMAQ, NAQPMS, and WRF-CHEM) and compared them to the manual prediction. The results indicated that WRF-CMAQ performs outstandingly in comprehensive analysis, with a yearly 24 h (hour) forecast accuracy (A) exceeding 91%. The NAQPMS 24 h forecast presented the lowest accuracy (A = 71%) with a high overprediction rate. The forecast accuracy of WRF-CHEM and NAQPMS in the winter period was 100%. However, frequently changing weather patterns interfered with spring forecasting. Forecasters can accurately predict mild pollution cases (A = 81.8% for 24 h, A = 72.7% for 48 h), whereas numerical models (NAQPMS and WRF-CHEM) showed better sensitivity to moderate pollution cases. For fine particulate evaluation, CMAQ's low biases occurred in winter, and the manually revised air quality forecast (MRAQR) normally gave higher prediction values except in pollution cases. Model underestimation could be explained by missing the long-range dust transport contribution from the northern part of China. The ozone 24 h forecast part revealed that WRF-CMAQ exhibited the highest correlation coefficient (R = 0.70, 19.74% < RMSE < 42.7%), whereas NAQPMS showed relatively poor results with the highest ME (43 μg/m3) and MFE (41.92%). Close examination of synoptic information revealed that ozone pollution cases were subjected to the western Pacific subtropical high (WPSH) system, tropical cycle, and regional strong pollution transport. The NAQPMS ozone overprediction could be explained by the overestimation of troposphere transmission from the surroundings during nonconductive weather patterns. The study investigates a theoretical method for model optimization, which provides solid support for precise emission reduction and targeted pollution control.

Neighborhood-Level Particle Pollution Assessment during the COVID-19 Pandemic via a Novel IoT Solution

In recent years, the concentration levels of various air pollutants have been constantly increasing, primarily due to the high vehicle flow. In 2020, however, severe lockdowns in Greece were imposed to reduce the spread of the COVID-19 pandemic, which led to a rapid reduction in the concentration levels of air pollutants such as PM2.5 and PM10 in the atmosphere. Initially, this paper seeks to identify the correlation between the concentration levels of PM10 and the traffic flow by acquiring data from low-cost IoT devices which were placed in Thessaloniki, Greece, from March to August 2020. The correlation and the linearity between the two parameters were further investigated by applying descriptive analytics, regression techniques, Pearson correlation, and independent T-testing. The obtained results indicate that the concentration levels of PM10 are strongly correlated to the vehicle flow. Therefore, the results hint that the decrease in the vehicle flow could result in improving the quality of environmental air. Finally, the acquired results point out that the temperature and humidity are weakly correlated with the concentration levels of PM10 present in the atmosphere.

Clean air in Europe for all: A call for more ambitious action.

Clean air in Europe for all: A call for more ambitious action.

Air Pollution Patterns Mapping of SO2, NO2, and CO Derived from TROPOMI over Central-East Europe

The analysis of changes in the level of air pollution concentration allows for the control of air quality and its compliance with the normative requirements. Currently, every country in Europe implements air quality monitoring. However, during emergencies in areas that are often difficult to monitor, the only source of information is geospatial data obtained by means of Earth observation techniques. The aims of this study were to estimate the amounts of pollutant concentrations and develop a pattern of spatiotemporal changes in Central and Eastern Europe in Poland and Ukraine. Due to the ongoing military operations in Ukraine, it is an area that is difficult to access. Pollution from industrial facilities, fires, collapsed buildings, and the use of explosive weapons poses a threat to air quality. Additionally, the impact of war on air pollution concentration levels remains unclear. This work characterized the changes in the distribution of sulfur dioxide, nitrogen dioxide and carbon monoxide concentrations in 2018–2022 in local zones in both countries. Publicly available TROPOMI-S5 satellite data were used for this study, which were compared with measurements from ground stations in Poland. It has been estimated that the concentration of NO2 (+0.67 ± 0.47 µmol/m2) in Poland has increased and the level of SO2 and CO have decreased in both studied areas: in Poland (−161.67 ± 5.48 µmol/m2, −470.85 ± 82.81 µmol/m2) and in Ukraine (−32.56 ± 23.51 µmol/m2, −438.04 ± 80.76 µmol/m2). The concentration of NO2 in Ukraine has decreased by −0.28 ± 0.21 µmol/m2.

Impacts of Covid-19 Lockdown on Concentration Levels of Traffic-Related Air Pollutants in Ibadan -a West African City

Trends and sources of air pollution at twenty-five traffic Intersections (TIs) before and during covid-19 lockdown were investigated in Ibadan, Nigeria. The relationships among climatic parameters, vehicular counts and ten air pollutants which includes particulate matter (PM1, PM2.5, PM10 &amp; Total Suspended Particles-TSP) and gaseous pollutants (CO, NO2, SO2, NH3, total volatile organic compounds-TVOCs, and ground level O3) measured simultaneously at TIs were analysed. Results indicated significant decrease in mean concentrations of all pollutants studied except NO2 with 212% increase during the study period. Concentrations of gaseous pollutants CO, SO2, NH3, TVOCs and ground level O3 reduced by 7.92%, 24.80%, 1.58%, 44.08% and 4.28%, respectively while particulates concentrations of PM1, PM2.5, PM10 and TSP concentrations decreased by 49.64%, 60.79%, 81.21% and 84.17%, respectively during lockdown. An integrated source apportionment approach using Pearson’s correlation, Airflow backward trajectories arriving in the study area and Principal component analysis (PCA) identified vehicular emission as the primary source of studied air pollutants at TIs before and during lockdown in Ibadan. Emission from residences, roadside fuel combustion and local air transport of pollutants from nearby upwind areas with industries and farming activities were identified as secondary sources of air pollution affecting the study area.

Using a Digital Twin to Study the Influence of Climatic Changes on High Ozone Levels in Bulgaria and Europe

High concentration levels of air pollutants may cause damage to plants, animals, and the health of some groups of human beings. Therefore, it is important to investigate different topics related to the high air pollution levels and to find reliable answers to the questions about the possible damages, which might take place when these levels exceed some limits. A few of the numerous questions, the answers of which are highly desirable, are listed below: (a) When are the air pollution levels dangerous? (b) What is the reason for the increased air pollution levels? (c) How can the air pollution levels be decreased? (d) Will the future climate changes result in higher and more dangerous air pollution levels? It is necessary to study carefully many issues connected with the distribution of air pollutants in a given region and with the reasons for the increases of the concentrations to high levels, which might be damaging. In order to do this, it is necessary to develop a Digital Twin of all relevant physical processes in the atmosphere and to use after that this tool in different applications. Such a tool, its name is DIGITAL AIR, has been created. Digital Twins are becoming more and more popular). Many complex problems, arising taking place in very complicated surroundings, can be handled and resolved successfully by applying Digital Twins. The preparation of such a digital tool as well as its practical implementation in the treatment of a special problem, the increase of some potentially dangerous ozone levels, will be discussed and tested in this paper. The Unified Danish Eulerian Model (UNI-DEM) is a very important part of DIGITAL AIR. This mathematical model, UNI-DEM, can be applied in many different studies related to damaging effects caused by high air pollution levels. We shall use it in this paper to get a reliable answer to a very special but extremely important question: will the future climatic changes lead to an increase in the ozone pollution levels in Bulgaria and Europe, which can potentially become dangerous for human health?

Estimation of COVID-19 patient numbers using artificial neural networks based on air pollutant concentration levels.

The dilemma between health concerns and the economy is apparent in the context of strategic decision making during the pandemic. In particular, estimating the patient numbers and achieving an informed management of the dilemma are crucial in terms of the strategic decisions to be taken. The Covid-19 pandemic presents an important case in this context. Sustaining the efforts to cope with and to put an end to this pandemic requires investigation of the spread and infection mechanisms of the disease, and the factors which facilitate its spread. Covid-19 symptoms culminating in respiratory failure are known to cause death. Since air quality is one of the most significant factors in the progression of lung and respiratory diseases, it is aimed to estimate the number of Covid-19 patients corresponding to the pollutant parameters (PM10, PM2.5, SO2, NOX, NO2, CO, O3) after determining the relationship between air pollutant parameters and Covid-19 patient numbers in Turkey. For this purpose, artificial neural network was used to estimate the number of Covid-19 patients corresponding to air pollutant parameters in Turkey. To obtain highest accuracy levels in terms of network architecture structure, various network structures were tested. The optimal performance level was developed with 15 neurons combined with one hidden layer, which achieved a network performance level as high as 0.97342. It was concluded that Covid-19 disease is affected from air pollutant parameters and the number of patients can be estimated depending on these parameters by this study. Since it is known that the struggle against the pandemic should be handled in all aspects, the result of the study will contribute to the establishment of environmental decisions and precautions.

Urban ambient air pollution and substance use disorder

There is growing evidence that air pollutants might affect human behavior. This study assesses the associations between air pollution concentrations and emergency department (ED) visits for abuse of psychoactive substances. 28,745 such ED visits were identified and retrieved from a health database containing diagnosed visits from five hospitals in Edmonton (Canada) over 10 years. The ED visits were analyzed as daily counts. Conditional Poisson regression models were used to estimate the associations between the number of ED visits and concentration levels of gaseous air pollutants (carbon monoxide (CO), nitrogen dioxide (NO2), sulphur dioxide (SO2), ozone (O3)) and particulate matters (PM2.5 and PM10, fine and coarse, respectively). Air pollutants and weather factors in the realized statistical models were lagged by the same number of days, from 0 to 5 days. The associations were estimated in the form of concentration-response functions. The results show relative risks and their 95% confidence intervals. Positive and statistically significant associations were obtained for CO for all patients (lags from 0 to 5), males (lags 1 and 3–5), and females (lag 4). For NO2, exposure lagged by 1 and 2 days has a positive statistically significant association for all and male patients. PM10 shows the same type of associations lagged by 2 and 3 days. PM2.5 (lag 2) is associated only in females. The results indicate that urban air pollution may have an impact on the abuse of psychoactive substances.

Improvement of Air Quality during the COVID-19 Lockdowns in the Republic of Slovenia and its Connection with Meteorology

Due to COVID-19 countermeasures, air quality temporarily improved in several countries around the world, especially in urban areas. This study investigates predominantly the changes in concentration levels of air pollutants (PM10 and PM2.5) in the Republic of Slovenia during the year 2020, which was marked by COVID-19 lockdowns. In this study, the data for the year 2020 were divided into four periods, i.e. the period before COVID-19 (1 January–11 March 2020), the first lockdown (12 March–31 May 2020), after the first lockdown (1 June–17 October 2020), and the second lockdown (18 October–31 December 2020). The data were obtained from 25 ground-based nationwide stations, subdivided further by traffic and background sites. For comparison, data from 2018 and 2019 were evaluated as well. Our findings indicate that COVID-19 restrictions had a moderate and indirect impact on PM10 and PM2.5 concentrations, which were more evident at PM10 monitoring sites near traffic locations. The results were additionally supported by providing t- and F- statistical tests. The impact of meteorological conditions (wind speed, temperature, relative humidity, and precipitation) on the concentration levels was also studied. The results show that, the most significant decrease of PM concentration in 2020 compared to 2018 and 2019 was found in the period after the first lockdown, while precipitation was not significantly different between the years during this time.

Exposure to Ambient NO2 Increases the Risk of Dry Eye Syndrome in Females: An 11-Year Population-Based Study.

Previous studies have indicated that women suffer from dry eye syndrome (DES) more significantly than men. Therefore, we specifically explore the associations between air pollutant levels and the risks of DES for women. The study obtained 27,605 participants from the 29 recruitment centers of the Taiwan Biobank, which was established in October 2012. A large scale cross-sectional study involving DES sufferers and age- and education-matched control groups without DES was designed. Based on the municipality of residence, the predicted concentration levels of various air pollutants, including PM2.5, sulfur dioxide (SO2), ozone (O3), and nitrogen dioxide (NO2) were estimated by using hybrid kriging/LUR model. Multiple logistic regressions were applied to estimate the prevalence ratios (PR) of DES and 95% confidence interval. Hormone supplementations, DBP, allergies, and arthritis were considered as important comorbidities for increased PR risk of DES. In addition, with each standard deviation (SD) increment of PM2.5 and temperature, women had significant increases in PRs of DES of 1.09- and 1.07-fold, respectively; conversely, each SD increment of relative humidity (RH) had a protective effect against the risk of DES. After considering hormone supplementation, arthritis, and allergy, the SD increment of NO2 and temperature were associated with the PRs of DES. In conclusion, significant associations of ambient NO2 concentration, RH and temperature with DES indicated the importance of increased environmental protection in the female population. Female exposure to high levels of NO2 when receiving hormone supplementation, or suffering with allergies or arthritis, had significantly increased risk of DES.

Short-term exposure to air pollution and hospital admission for heart failure among older adults in metropolitan cities: a time-series study.

We aimed to investigate the association between air pollution concentration levels and hospital admissions for heart failure (HF) among older adults in metropolitan cities in South Korea. We used hospital admission data of 1.8 million older adults in seven metropolitan cities from 2008 to 2016, derived from the National Health Insurance Service of South Korea. Daily HF admission data were linked to air pollutants concentrations for the respective dates, including particulate matter less than 2.5μm in size (PM2.5), 10 μm (PM10), sulfur dioxide (SO2), nitrogen dioxide (NO2), carbon monoxide (CO), and ozone. We estimated the association between air pollutants and daily HF admissions using quasi-Poisson generalized additive models for each city. During the study period, 142,490 hospital admissions for HF were noted. Increases of 10μg/m3 of PM2.5 and PM10, and 10ppb of SO2, NO2, and CO were associated with an increased risk of HF admission by 0.93% ([95% confidence intervals 0.51-1.36], 0.55% [0.31-0.80], 6.04% [2.15-10.08], 1.10% [0.38-1.82], and 0.05% [0.01-0.09]), respectively, on the same day. Increases in mean exposure to PM2.5, PM10, and SO2 for 8days from the concurrent day were also significantly associated with HF admissions. During the warm season, the risk of HF admissions increased shortly after an increase in PM2.5, whereas prolonged effects were observed during the cold season. Our study suggests the adverse effects of air pollution on HF. Moreover, the evidence of seasonality may help tailor protection guidelines for older adults.

Community Risk Factors in the COVID-19 Incidence and Mortality in Catalonia (Spain). A Population-Based Study.

The heterogenous distribution of both COVID-19 incidence and mortality in Catalonia (Spain) during the firsts moths of the pandemic suggests that differences in baseline risk factors across regions might play a relevant role in modulating the outcome of the pandemic. This paper investigates the associations between both COVID-19 incidence and mortality and air pollutant concentration levels, and screens the potential effect of the type of agri-food industry and the overall land use and cover (LULC) at area level. We used a main model with demographic, socioeconomic and comorbidity covariates highlighted in previous research as important predictors. This allowed us to take a glimpse of the independent effect of the explanatory variables when controlled for the main model covariates. Our findings are aligned with previous research showing that the baseline features of the regions in terms of general health status, pollutant concentration levels (here NO2 and PM10), type of agri-food industry, and type of land use and land cover have modulated the impact of COVID-19 at a regional scale. This study is among the first to explore the associations between COVID-19 and the type of agri-food industry and LULC data using a population-based approach. The results of this paper might serve as the basis to develop new research hypotheses using a more comprehensive approach, highlighting the inequalities of regions in terms of risk factors and their response to COVID-19, as well as fostering public policies towards more resilient and safer environments.

Agglomeration and infrastructure effects in land use regression models for air pollution – Specification, estimation, and interpretations

Established land use regression (LUR) techniques such as linear regression utilize extensive selection of predictors and functional form to fit a model for every data set on a given pollutant. In this paper, an alternative to established LUR modeling is employed, which uses additive regression smoothers. Predictors and functional form are selected in a data-driven way and ambiguities resulting from specification search are mitigated. The approach is illustrated with nitrogen dioxide (NO2) data from German monitoring sites using the spatial predictors longitude, latitude, altitude and structural predictors; the latter include population density, land use classes, and road traffic intensity measures. The statistical performance of LUR modeling via additive regression smoothers is contrasted with LUR modeling based on parametric polynomials. Model evaluation is based on goodness of fit, predictive performance, and a diagnostic test for remaining spatial autocorrelation in the error terms. Additionally, interpretation and counterfactual analysis for LUR modeling based on additive regression smoothers are discussed.Our results have three main implications for modeling air pollutant concentration levels: First, modeling via additive regression smoothers is supported by a specification test and exhibits superior in- and out-of-sample performance compared to modeling based on parametric polynomials. Second, different levels of prediction errors indicate that NO2 concentration levels observed at background and traffic/industrial monitoring sites stem from different processes. Third, accounting for agglomeration and infrastructure effects is important: NO2 concentration levels tend to increase around major cities, surrounding agglomeration areas, and their connecting road traffic network.