Predicting spatial variations in annual average outdoor ultrafine particle concentrations in Montreal and Toronto, Canada: Integrating land use regression and deep learning models

Traffic-related air pollution backcasting using convolutional neural network and long short-term memory approach

Air pollution backcasting, especially nitrogen dioxide (NO2), is crucial in epidemiological studies, thus enabling the reconstruction of historical exposure levels for assessing long-term health effects. Changes in NO2 concentrations in urban areas are typically influenced by vehicle composition, technology, and traffic volumes. However, the observed NO2 levels at a monitoring site also reflect contributions from other sources, such as industrial and regional backgrounds. This study proposes a model that captures the spatial variability of NO2 concentrations, incorporating temporal trends through traffic-related predictors like nitrogen oxides (NOx) emissions and Annual Average Daily Traffic (AADT). Our approach integrates a Convolutional Neural Network (CNN) for spatial variation and Long Short-Term Memory (LSTM) for long-term temporal dynamics, yielding optimal spatiotemporal predictions for NO2 levels across the City of Toronto, Canada. The model, trained with NO2 measurements collected via the Urban Scanner mobile platform in 2020 and 2021, utilizes a Traffic Emission Prediction scheme (TEPs) to develop NOx and AADT inventories, serving as input to the LSTM model. Our proposed approach successfully estimates traffic-related NO2 levels across Toronto from 2006 to 2020. By intersecting the backcasted levels with census data, we noted that despite an overall decrease in NO2 levels between 2006 and 2020, disparities in exposure grew as more marginalized communities faced environmental injustice.

Air pollution mapping and variability over five European cities

Mapping urban pollution is essential for assessing population exposure and addressing associated health impacts. High urban concentrations are due to the proximity of sources such as traffic or residential heating, and to urban density with the presence of buildings that reduce street ventilation. This urban complexity makes fine-scale mapping challenging, even for regulated pollutants such as NO2 and PM2.5. In this study we apply state-of-the-art empirical and deterministic modeling approaches to produce high-resolution (<100 m) pollution maps across five European cities (Paris, Athens, Birmingham, Rotterdam, Bucharest). These methodologies enable full-city mapping capturing intra-urban gradients of concentrations. Depending on the methodology, regulated pollutants (NO2, PM2.5) and/or emerging pollutants (black carbon (BC) and ultrafine particles (UFP characterized here by particulate number concentration PNC)) are considered. For deterministic modelling, different approaches are presented: a multi-scale Eulerian modelling chain down to the street scale with chemistry/aerosol dynamics at all scales, multi-scale hybrid models with Eulerian regional dispersion and Gaussian subgrid dispersion, and a Gaussian-based model. Empirical land use regression models were developed based upon mobile monitoring. To compare the relative performance of the methodologies and to evaluate their performance and limitations, the modelling results are compared to fixed measurement stations. We introduce a standardized metric to quantify spatial and seasonal variability and assess each method's capacity to reproduce fine-scale urban heterogeneity. We also evaluate how data assimilation affects both concentration accuracy and variability representation-particularly relevant for emerging pollutants where measurement data are sparse. We confirm established seasonal and spatial patterns: spatial variability is more pronounced for PNC, NO2 and BC than PM2.5, and concentrations are higher during the winter periods. We also observe reduced spatial variability in winter for PM2. 5 (linked to residential heating) and for BC in cities with significant wood burning emissions. This study adds unique value by evaluating these patterns using fixed measurement stations, and quantifying them across entire urban areas at very fine spatial resolution (<100 m). Furthermore, important methodological strengths and limitations are pointed out, providing practical guidance for the selection and improvement of urban exposure mapping methods, supporting the implementation of the new EU Air Quality Directive.

Examining the social distributions in neighbourhood black carbon and ultrafine particles in Montreal and Toronto, Canada

BACKGROUND

Socioeconomic inequities in outdoor ultrafine particles (UFP) and black carbon (BC) are understudied in Canada, where metropoles like Montreal and Toronto feature distinct sociodemographic diversity and urban characteristics compared to U.S. cities.

METHODS

We collected vulnerability indicators, including social, economic, household composition, and immigration status, at the dissemination area level for Montreal and Toronto using data from the 2006 and 2021 Canadian Census of Population. Areas were classified as disadvantaged, intermediate, or advantaged following K-means clustering analysis. We aggregated and calculated population-weighted average concentrations of BC and UFP, and UFP size at the dissemination area and cluster levels using high-resolution exposure surfaces, derived from year-long mobile monitoring campaigns conducted in each city during 2020-2021. Final exposure surfaces were generated by integrating predictions from land-use regression models and deep convolutional neural network models.

FINDINGS

We observed high within-city variations in aggregated air pollutant levels, with higher outdoor BC and UFP concentrations and smaller UFP sizes in areas near local sources such as major roads, railways, airports, and densely populated regions. Advantaged areas experienced the lowest median UFP concentrations in both Montreal (10,707 pt/cm3) and Toronto (10,988 pt/cm3), as well as the lowest BC concentrations (650 ng/m3) in Montreal. The highest median UFP concentrations were observed in intermediate areas in Montreal (15,709 pt/cm3) and disadvantaged areas in Toronto (12,228 pt/cm3). Conversely, the highest BC concentrations were observed in disadvantaged and intermediate areas in Montreal (805-811 ng/m3), and disadvantaged and advantaged areas in Toronto (1,228-1,252 ng/m3). Notably, high priority areas for the double burden of vulnerability and high BC and UFP concentrations were located near air pollutants local emission sources.

INTERPRETATION

Our findings highlight the importance of prioritizing exposure mitigation for populations residing near local sources and to understand contextual factors influencing inequities across cities and pollutants.

High-Resolution Geospatial Database: National Criteria-Air-Pollutant Concentrations in the Contiguous U.S., 2016-2020

Concentration estimates for ambient air pollution are used widely in fields such as environmental epidemiology, health impact assessment, urban planning, environmental equity and sustainability. This study builds on previous efforts by developing an updated high-resolution geospatial database of population-weighted annual-average concentrations for six criteria air pollutants (PM2.5, PM10, CO, NO2, SO2, O3) across the contiguous U.S. during a five-year period (2016-2020). We developed Land Use Regression (LUR) models within a partial-least-squares-universal kriging framework by incorporating several land use, geospatial and satellite-based predictor variables. The LUR models were validated using conventional and clustered cross-validation, with the former consistently showing superior performance in capturing the variability of air quality. Most models demonstrated reliable performance (e.g., mean squared error-based R 2 > 0.8, standardised root mean squared error < 0.1). We used the best modelling approach to develop estimates by Census Block, which were then population-weighted averaged at Census Block Group, Census Tract and County geographies. Our database provides valuable insights into the dynamics of air pollution, with utility for environmental risk assessment, public health, policy and urban planning.

Unveiling the Impact of Wildfires on Nanoparticle Characteristics and Exposure Disparities through Mobile and Fixed-Site Monitoring in Toronto, Canada

This study investigates the impacts of wildfires on nanoparticle characteristics and exposure disparities in Toronto, integrating data from a large-scale mobile monitoring campaign and fixed-site measurements during the unprecedented 2023 wildfire season. Our results reveal changes in particle characteristics during wildfire days, with particle number concentrations decreasing by 60% and particle diameter increasing by 30% compared to nonwildfire days. Moreover, the median lung deposited surface area (LDSA) levels rose by 31% during wildfire events. We employed gradient boosting models to estimate near-road LDSA levels on both wildfire and nonwildfire days. The LDSA ratio (wildfire/nonwildfire) exceeded 2.0 in certain areas along highways and in downtown Toronto. Furthermore, our findings show that marginalized communities faced greater LDSA increases than less marginalized ones. Under wildfire conditions, the LDSA ratio difference between the most and least marginalized groups was 16% for recent immigrants and visible minorities and 7% for seniors and children, both statistically significant. This study delivers critical insights into the spatiotemporal variations of nanoparticle characteristics during wildfire and nonwildfire periods, demonstrating the substantial health risks posed by increased LDSA levels and the inequitable distribution of these risks among Toronto's diverse population.

Machine learning-enhanced high-resolution exposure assessment of ultrafine particles

Ultrafine particles (UFPs) under 100 nm pose significant health risks inadequately addressed by traditional mass-based metrics. The WHO emphasizes particle number concentration (PNC) for assessing UFP exposure, but large-scale evaluations remain scarce. In this study, we developed a stacking-based machine learning framework integrating data-driven and physical-chemical models for a national-scale UFP exposure assessment at 1 km spatial and 1-hour temporal resolutions, leveraging long-term standardized PNC measurements in Switzerland. Approximately 20% (1.7 million) of the Swiss population experiences high UFP exposure exceeding an annual mean of 104 particles‧cm-3, with a national average of (9.3 ± 4.7)×103 particles‧cm-3, ranging from (5.5 ± 2.3)×103 (rural) to (1.4 ± 0.5)×104 particles‧cm-3 (urban). A nonlinear relationship is identified between the WHO-recommended 1-hour and 24-hour exposure reference levels, suggesting their non-interchangeability. UFP spatial heterogeneity, quantified by coefficient of variation, ranges from 4.7 ± 4.2 (urban) to 13.8 ± 15.1 (rural) times greater than PM2.5. These findings provide crucial insights for the development of future UFP standards.

Ultrafine Particle Generation from Ozone Oxidation of Cannabis Smoke

Cannabis smoke is a complex aerosol mixture, featuring characteristic monoterpenes and sesquiterpenes which are susceptible to reaction with ozone and other oxidants. These reactions form less-volatile species which can contribute to secondary organic aerosol (SOA) and ultrafine particle (UFP) formation. In this work, the reaction of ozone with cannabis smoke was observed in an environmental chamber. Particle size distribution, and gas-phase and particle-phase composition were monitored in real time. The diameter of primary particles ranged from 10-1 to 1 μm. Ultrafine particle formation occurred when cannabis smoke was exposed to ozone levels greater than 10 ppb, over the entire observed primary particle concentration range (1030-4580 μg m-3). Gas-phase measurements indicate that monoterpene and sesquiterpene levels decayed rapidly upon ozone exposure, while oxygen-containing species were formed during oxidation. On the other hand, measurements of particle composition showed an increase in nitrogen-containing species during oxidation. Although ozone was the only oxidant added to cannabis smoke in the chamber, it is believed that the OH radical plays an important role in the oxidation mechanism, where OH results from the reaction of ozone with terpenes and sesquiterpenes. Overall, smoking cannabis in ozone-rich environments, both indoors and outdoors, will likely lead to UFP formation.

Airborne Nanoparticle Concentrations Are Associated with Increased Mortality Risk in Canada's Two Largest Cities

Rationale: Outdoor fine particulate air pollution (particulate matter with an aerodynamic diameter ⩽2.5 μm; PM2.5) contributes to millions of deaths around the world each year, but much less is known about the long-term health impacts of other particulate air pollutants, including ultrafine particles (a.k.a. nanoparticles), which are in the nanometer-size range (<100 nm), widespread in urban environments, and not currently regulated. Objectives: We sought to estimate the associations between long-term exposure to outdoor ultrafine particles and mortality. Methods: Outdoor air pollution levels were linked to the residential addresses of a large, population-based cohort from 2001 to 2016. Associations between long-term exposure to outdoor ultrafine particles and nonaccidental and cause-specific mortality were estimated using Cox proportional hazards models. Measurements and Main Results: An increase in long-term exposure to outdoor ultrafine particles was associated with an increased risk of nonaccidental mortality (hazard ratio = 1.073; 95% confidence interval = 1.061-1.085) and cause-specific mortality, the strongest of which was respiratory mortality (hazard ratio = 1.174; 95% confidence interval = 1.130-1.220). We estimated the mortality burden for outdoor ultrafine particles in Montreal and Toronto, Canada, to be approximately 1,100 additional nonaccidental deaths every year. Furthermore, we observed possible confounding by particle size, which suggests that previous studies may have underestimated or missed important health risks associated with ultrafine particles. Conclusions: As outdoor ultrafine particles are not currently regulated, there is great potential for future regulatory interventions to improve population health by targeting these common outdoor air pollutants.

Spatial and spatiotemporal modelling of intra-urban ultrafine particles: A comparison of linear, nonlinear, regularized, and machine learning methods

BACKGROUND

Machine learning methods are proposed to improve the predictions of ambient air pollution, yet few studies have compared ultrafine particles (UFP) models across a broad range of statistical and machine learning approaches, and only one compared spatiotemporal models. Most reported marginal differences between methods. This limits our ability to draw conclusions about the best methods to model ambient UFPs.

OBJECTIVE

To compare the performance and predictions of statistical and machine learning methods used to model spatial and spatiotemporal ambient UFPs.

METHODS

Daily and annual models were developed from UFP measurements from a year-long mobile monitoring campaign in Quebec City, Canada, combined with 262 geospatial and six meteorological predictors. Various road segment lengths were considered (100/300/500 m) for UFP data aggregation. Four statistical methods included linear, non-linear, and regularized regressions, whereas eight machine learning regressions utilized tree-based, neural networks, support vector, and kernel ridge algorithms. Nested cross-validation was used for model training, hyperparameter tuning and performance evaluation.

RESULTS

Mean annual UFP concentrations was 13,335 particles/cm3. Machine learning outperformed statistical methods in predicting UFPs. Tree-based methods performed best across temporal scales and segment lengths, with XGBoost producing the overall best performing models (annual R2 = 0.78-0.86, RMSE = 2163-2169 particles/cm3; daily R2 = 0.47-0.48, RMSE = 8651-11,422 particles/cm3). With 100 m segments, other annual models performed similarly well, but their prediction surfaces of annual mean UFP concentrations showed signs of overfitting. Spatial aggregation of monitoring data significantly impacted model performance. Longer segments yielded lower RMSE in all daily models and for annual statistical models, but not for annual machine learning models.

CONCLUSIONS

The use of tree-based methods significantly improved spatiotemporal predictions of UFP concentrations, and to a lesser extent annual concentrations. Segment length and hyperparameter tuning had notable impacts on model performance and should be considered in future studies.

Airborne ultrafine particle concentrations and brain cancer incidence in Canada's two largest cities

BACKGROUND

Malignant brain tumours are rare, but are important to study because survival rates are low and few modifiable risk factors have been identified. Existing evidence suggests that outdoor ultrafine particles (UFPs; particulate matter < 100 nm; sometimes referred to as nanoparticles) can deposit in the brain and could encourage initiation and progression of cancerous tumours, but epidemiological data are limited.

METHODS

High-resolution estimates of outdoor UFP concentrations and size were linked to residential locations of approximately 1.5 million people in Montreal and Toronto, Canada from 2001 to 2015. Cox proportional hazards models were used to estimate associations between annual average outdoor UFPs and malignant brain tumour incidence while adjusting for potential confounding factors including other outdoor air pollutants.

FINDINGS

In total, 1365 incident brain tumour cases occurred during follow-up. Consistent positive associations were observed between long-term exposures to outdoor UFPs and brain tumour incidence with increased risk ranging from 10.5% (95% CI: -1.4, 24.0%) to 15.3% (95% CI: 0.4, 32.5%) per 10,000 particle/cm3 increase. Long-term exposures to oxidant gases, black carbon, or fine particulate matter (PM2.5) were not associated with increased brain tumour incidence.

INTERPRETATION

Our results suggest that long-term exposures to outdoor UFPs are associated with an increased risk of developing malignant brain tumours. On an absolute scale, the magnitude of this risk translates into approximately 24 additional cases per year per 10,000 particle/cm3 increase in annual average outdoor UFPs in a hypothetical city of 3-million people.

FUNDING

Canadian Institutes of Health Research (CIHR) Foundation Grant and The United States Health Effects Institute (HEI).

Combining Google traffic map with deep learning model to predict street-level traffic-related air pollutants in a complex urban environment

BACKGROUND

Traffic-related air pollution (TRAP) is a major contributor to urban pollution and varies sharply at the street level, posing a challenge for air quality modeling. Traditional land use regression models combined with data from fixed monitoring stations may be unable to predict and characterize fine-scale TRAP, especially in complex urban environments influenced by various features. This study aims to estimate fine-scale (50 m) concentrations of nitrogen oxides (NO and NO₂) in Hong Kong using a deep learning (DL) structured model.

METHODS

We collected data from mobile air quality sensors on buses and crowd-sourced Google real-time traffic status as a proxy for real-time traffic emissions. Our DL model was compared with existing machine learning models to assess performance improvements. Using an interpretable machine learning method, we hierarchically evaluated the global, local, and interaction effects for different features.

RESULTS

Our DL model outperformed existing machine learning models, achieving R2 values of 0.72 for NO and 0.69 for NO₂. The incorporation of traffic status as a key predictor improved model performance by 9% to 17%. The interpretable machine learning method revealed the importance of traffic-related features and their pairwise interactions.

CONCLUSION

The results indicate that traffic-related features significantly contribute to TRAP and provide insights and guidance for urban planning. By incorporating crowd-sourced Google traffic information, we assessed traffic abatement scenarios that could inform targeted strategies for improving urban air quality.

Capturing Exposure Disparities with Chemical Transport Models: Evaluating the Suitability of Downscaling Using Land Use Regression

High resolution exposure surfaces are essential to capture disparities in exposure to traffic-related air pollution in urban areas. In this study, we develop an approach to downscale Chemical Transport Model (CTM) simulations to a hyperlocal level (∼100m) in the Greater Toronto Area (GTA) under three scenarios where emissions from cars, trucks and buses are zeroed out, thus capturing the burden of each transportation mode. This proposed approach statistically fuses CTMs with Land-Use Regression using machine learning techniques. With this proposed downscaling approach, changes in air pollutant concentrations under different scenarios are appropriately captured by downscaling factors that are trained to reflect the spatial distribution of emission reductions. Our validation analysis shows that high-resolution models resulted in better performance than coarse models when compared with observations at reference stations. We used this downscaling approach to assess disparities in exposure to nitrogen dioxide (NO2) for populations composed of renters, low-income households, recent immigrants, and visible minorities. Individuals in all four categories were disproportionately exposed to the burden of cars, trucks, and buses. We conducted this analysis at spatial resolutions of 12, 4, 1 km, and 100 m and observed that disparities were significantly underestimated when using coarse spatial resolutions. This reinforces the need for high-spatial resolution exposure surfaces for environmental justice analyses.

Predicting within-city spatiotemporal variations in daily median outdoor ultrafine particle number concentrations and size in Montreal and Toronto, Canada

Background

Epidemiological evidence suggests that long-term exposure to outdoor ultrafine particles (UFPs, <0.1 μm) may have important human health impacts. However, less is known about the acute health impacts of these pollutants as few models are available to estimate daily within-city spatiotemporal variations in outdoor UFPs.

Methods

Several machine learning approaches (i.e., generalized additive models, random forest models, and extreme gradient boosting) were used to predict daily spatiotemporal variations in outdoor UFPs (number concentration and size) across Montreal and Toronto, Canada using a large database of mobile monitoring measurements. Separate models were developed for each city and all models were evaluated using a 10-fold cross-validation procedure.

Results

In total, our models were based on measurements from 12,705 road segments in Montreal and 10,929 road segments in Toronto. Daily median outdoor UFP number concentrations varied substantially across both cities with 1st-99th percentiles ranging from 1389 to 181,672 in Montreal and 2472 to 118,544 in Toronto. Outdoor UFP size tended to be smaller in Montreal (mean [SD]: 34 nm [15]) than in Toronto (mean [SD]: 44 nm [25]). Extreme gradient boosting models performed best and explained the majority of spatiotemporal variations in outdoor UFP number concentrations (Montreal, R 2: 0.727; Toronto, R 2: 0.723) and UFP size (Montreal, R 2: 0.823; Toronto, R 2: 0.898) with slopes close to one and intercepts close to zero for relationships between measured and predicted values.

Conclusion

These new models will be applied in future epidemiological studies examining the acute health impacts of outdoor UFPs in Canada's two largest cities.

High-resolution urban air pollution mapping

Variation in urban air pollution arises because of complex spatial, temporal, and chemical processes, which profoundly affect population exposure, human health, and environmental justice. This Review highlights insights from two popular in situ measurement methods-mobile monitoring and dense sensor networks-that have distinct but complementary strengths in characterizing the dynamics and impacts of the multidimensional urban air quality system. Mobile monitoring can measure many pollutants at fine spatial scales, thereby informing about processes and control strategies. Sensor networks excel at providing temporal resolution at many locations. Increasingly sophisticated studies leveraging both methods can vividly identify spatial and temporal patterns that affect exposures and disparities and offer mechanistic insight toward effective interventions. This Review summarizes the strengths and limitations of these methods and discusses their implications for understanding fine-scale processes and impacts.

High-resolution spatial and spatiotemporal modelling of air pollution using fixed site and mobile monitoring in a Canadian city

The development of high-resolution spatial and spatiotemporal models of air pollutants is essential for exposure science and epidemiological applications. While fixed-site sampling has conventionally provided input data for statistical predictive models, the evolving mobile monitoring method offers improved spatial resolution, ideal for measuring pollutants with high spatial variability such as ultrafine particles (UFP). The Quebec Air Pollution Exposure and Epidemiology (QAPEE) study measured and modelled the spatial and spatiotemporal distributions of understudied pollutants, such as UFPs, black carbon (BC), and brown carbon (BrC), along with fine particulate matter (PM2.5), nitrogen dioxide (NO2), and ozone (O3) in Quebec City, Canada. We conducted a combined fixed-site (NO2 and O3) and mobile monitoring (PM2.5, BC, BrC, and UFPs) campaign over 10-months. Mobile monitoring routes were monitored on a weekly basis between 8am-10am and designed using location/allocation modelling. Seasonal fixed-site sampling campaigns captured continuous 24-h measurements over two-week periods. Generalized Additive Models (GAMs), which combined data on pollution concentrations with spatial, temporal, and spatiotemporal predictor variables were used to model and predict concentration surfaces. Annual models for PM2.5, NO2, O3 as well as seven of the smallest size fractions in the UFP range, had high out of sample predictive accuracy (range r2: 0.54-0.86). Varying spatial patterns were observed across UFP size ranges measured as Particle Number Counts (PNC). The monthly spatiotemporal models for PM2.5 (r2 = 0.49), BC (r2 = 0.27), BrC (r2 = 0.29), and PNC (r2 = 0.49) had moderate or moderate-low out of sample predictive accuracy. We conducted a sensitivity analysis and found that the minimum number of 'n visits' (mobile monitoring sessions) required to model annually representative air pollution concentrations was between 24 and 32 visits dependent on the pollutant. This study provides a single source of exposure models for a comprehensive set of air pollutants in Quebec City, Canada. These exposure models will feed into epidemiological research on the health impacts of ambient UFPs and other pollutants.

Exploring the triple burden of social disadvantage, mobility poverty, and exposure to traffic-related air pollution

Understanding the relationships between ultrafine particle (UFP) exposure, socioeconomic status (SES), and sustainable transportation accessibility in Toronto, Canada is crucial for promoting public health, addressing environmental justice, and ensuring transportation equity. We conducted a large-scale mobile measurement campaign and employed a gradient boost model to generate exposure surfaces using land use, built environment, and meteorological conditions. The Ontario Marginalization Index was used to quantify various indicators of social disadvantage for Toronto's neighborhoods. Our findings reveal that people in socioeconomically disadvantaged areas experience elevated UFP exposures. We highlight significant disparities in accessing sustainable transportation, particularly in areas with higher ethnic concentrations. When factoring in daily mobility, UFP exposure disparities in disadvantaged populations are further exacerbated. Furthermore, individuals who do not generate emissions themselves are consistently exposed to higher UFPs, with active transportation users experiencing the highest UFP exposures both at home and at activity locations. Finally, we proposed a novel index, the Community Prioritization Index (CPI), incorporating three indicators, including air quality, social disadvantage, and sustainable transportation. This index identifies neighborhoods experiencing a triple burden, often situated near major infrastructure hubs with high diesel truck activity and lacking greenspace, marking them as high-priority areas for policy action and targeted interventions.

Harnessing AI to unmask Copenhagen's invisible air pollutants: A study on three ultrafine particle metrics

Ultrafine particles (UFPs) are airborne particles with a diameter of less than 100 nm. They are emitted from various sources, such as traffic, combustion, and industrial processes, and can have adverse effects on human health. Long-term mean ambient average particle size (APS) in the UFP range varies over space within cities, with locations near UFP sources having typically smaller APS. Spatial models for lung deposited surface area (LDSA) within urban areas are limited and currently there is no model for APS in any European city. We collected particle number concentration (PNC), LDSA, and APS data over one-year monitoring campaign from May 2021 to May 2022 across 27 locations and estimated annual mean in Copenhagen, Denmark, and obtained additionally annual mean PNC data from 6 state-owned continuous monitors. We developed 94 predictor variables, and machine learning models (random forest and bagged tree) were developed for PNC, LDSA, and APS. The annual mean PNC, LDSA, and APS were, respectively, 5523 pt/cm3, 12.0 μm2/cm3, and 46.1 nm. The final R2 values by random forest (RF) model were 0.93 for PNC, 0.88 for LDSA, and 0.85 for APS. The 10-fold, repeated 10-times cross-validation R2 values were 0.65, 0.67, and 0.60 for PNC, LDSA, and APS, respectively. The root mean square error for final RF models were 296 pt/cm3, 0.48 μm2/cm3, and 1.60 nm for PNC, LDSA, and APS, respectively. Traffic-related variables, such as length of major roads within buffers 100-150 m and distance to streets with various speed limits were amongst the highly-ranked predictors for our models. Overall, our ML models achieved high R2 values and low errors, providing insights into UFP exposure in a European city where average PNC is quite low. These hyperlocal predictions can be used to study health effects of UFPs in the Danish Capital.