Spatial Modeling of Daily PM2.5, NO2, and CO Concentrations Measured by a Low-Cost Sensor Network: Comparison of Linear, Machine Learning, and Hybrid Land Use Models.

Previous studies have characterized spatial patterns of pollution with land use regression (LUR) models from distributed passive or filter samplers at low temporal resolution. Large-scale deployment of low-cost sensors (LCS), which typically sample in real time, may enable time-resolved or real-time modeling of concentration surfaces. The aim of this study was to develop spatiotemporal models of PM2.5, NO2, and CO using an LCS network in Pittsburgh, Pennsylvania. We modeled daily average concentrations in August 2016-December 2017 across 50 sites. Land use variables included 13 time-independent (e.g., elevation) and time-dependent (e.g., temperature) predictors. We examined two models: LUR and a machine-learning-enabled land use model (land use random forest, LURF). The LURF models outperformed LUR models, with increase in the average externally cross-validated R2 of 0.10-0.19. Using wavelet decomposition to separate short-lived events from the regional background, we also created time-decomposed LUR and LURF models. Compared to the standard model, this resulted in improvement in R2 of up to 0.14. The time-decomposed models were more influenced by spatial parameters. Mapping our models across Allegheny County, we observed that time-decomposed LURF models created robust PM2.5 predictions, suggesting that this approach may improve our ability to map air pollutants at high spatiotemporal resolution.

• Publication year

  2021

• Venue

  Environmental Science and Technology

• Publication date

  2021-06-16

• Fields of study

  Medicine, Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1021/acs.est.1c02653PMID 34133134
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Contribution of low-cost sensor measurements to the prediction of PM2.5 levels: A case study in Imperial County, California, USA.

  2019cited by this paper
• Machine learning models accurately predict ozone exposure during wildfire events.

  2019cited by this paper
• Demonstration of a Low-Cost Multi-Pollutant Network to Quantify Intra-Urban Spatial Variations in Air Pollutant Source Impacts and to Evaluate Environmental Justice

  2019cited by this paper
• Estimation of daily PM10 and PM2.5 concentrations in Italy, 2013-2015, using a spatiotemporal land-use random-forest model.

  2019cited by this paper
• Random forest as a generic framework for predictive modeling of spatial and spatio-temporal variables

  2018cited by this paper
• Predicting Daily Urban Fine Particulate Matter Concentrations Using a Random Forest Model.

  2018cited by this paper
• Assessing the Potential of Land Use Modification to Mitigate Ambient NO2 and Its Consequences for Respiratory Health

  2017cited by this paper
• Air Pollution and Mortality in the Medicare Population

  2017cited by this paper
• Capturing the sensitivity of land-use regression models to short-term mobile monitoring campaigns using air pollution micro-sensors.

  2017cited by this paper
• Development of land use regression models for PM2.5, SO2, NO2 and O3 in Nanjing, China

  2017cited by this paper
• Exposure assessment models for elemental components of particulate matter in an urban environment: A comparison of regression and random forest approaches.

  2017cited by this paper
• Characterizing the spatial distribution of ambient ultrafine particles in Toronto, Canada: A land use regression model.

  2016cited by this paper
• Transferability and Generalizability of Regression Models of Ultrafine Particles in Urban Neighborhoods in the Boston Area

  2015cited by this paper
• Land use Regression as Method to Model Air Pollution. Previous Results for Gothenburg/Sweden☆

  2015cited by this paper
• Using High-Resolution Satellite Aerosol Optical Depth To Estimate Daily PM2.5 Geographical Distribution in Mexico City.

  2015cited by this paper
• Evaluation of land use regression models for NO2 and particulate matter in 20 European study areas: the ESCAPE project.

  2013cited by this paper
• The changing paradigm of air pollution monitoring.

  2013cited by this paper
• Development of Land Use Regression models for PM(2.5), PM(2.5) absorbance, PM(10) and PM(coarse) in 20 European study areas; results of the ESCAPE project.

  2012cited by this paper
• Near-roadway air quality: synthesizing the findings from real-world data.

  2010cited by this paper
• Modern Applied Statistics with S

  2010cited by this paper
• Intercity transferability of land use regression models for estimating ambient concentrations of nitrogen dioxide

  2009cited by this paper
• A Review of Land-use Regression Models for Characterizing Intraurban Air Pollution Exposure

  2007cited by this paper
• Modeling the Intraurban Variability of Ambient Traffic Pollution in Toronto, Canada

  2007cited by this paper
• Application of land use regression to estimate long-term concentrations of traffic-related nitrogen oxides and fine particulate matter.

  2007cited by this paper
• Spatial Analysis of Air Pollution and Mortality in Los Angeles

  2005cited by this paper
• Application of Breiman's Random Forest to Modeling Structure-Activity Relationships of Pharmaceutical Molecules

  2004cited by this paper

• Data fusion for fine-scale ozone mapping in the New York City metropolitan area using low-cost sensors and model information

  2026cites this paper
• A Machine Learning-Based Calibration Framework for Low-Cost PM2.5 Sensors Integrating Meteorological Predictors

  2025cites this paper
• Spatial Emissions Modeling from Mobile Sources Using Hidden Markov Models and Autoencoders

  2025cites this paper
• Blue spaces and incident dementia: Differences by geospatial and historical contexts

  2025cites this paper
• Assessing Spatiotemporal Characteristics, Health Impacts and Related Economic Losses of PM2.5, PM10 and O3 in Beijing-Tianjin-Hebei Region from 2015 to 2020

  2025cites this paper
• Evolving trends in application of low-cost air quality sensor networks: challenges and future directions

  2025cites this paper
• PM2.5 Concentration Prediction: Ultrahigh Spatiotemporal Resolution Achieved by Combining Machine Learning and Low-Cost Sensors

  2025influential citation
• Refining Air Pollution Exposure Estimates: A Comparison of Citywide and Neighborhood Land Use Regression Models in Toronto.

  2025cites this paper
• Observational Inferences of NOx and CO Emission Factors for Vehicles and Homes in the San Francisco Bay Area

  2025cites this paper
• Impact of sampling frequency on low-cost PM sensor performance including short-term temporal events in high PM environments

  2025cites this paper
• Incorporating Community Knowledge Into Analysis of Air Quality Monitoring Network Data

  2025cites this paper
• Transcending pre-processing of physiological fluids and device-to-device variability in nanosheet networks for point of care (PoC) immunosensors

  2025cites this paper
• Mapping trends and analyzing key themes in low-cost sensors for air quality monitoring

  2025cites this paper
• Nitrogen Dioxide Monitoring by Means of a Low-Cost Autonomous Platform and Sensor Calibration via Machine Learning with Global Data Correlation Enhancement

  2025cites this paper
• Impact of land use characteristics on air pollutant concentrations considering the spatial range of influence

  2025cites this paper
• Cost-Effective Mapping of Hyperlocal Air Pollution Using Large-Scale Mobile Monitoring and Land-Use Machine Learning

  2025cites this paper
• Research trends in spatial modeling of PM2.5 concentration using machine learning: a bibliometric review

  2025cites this paper
• Facing & mitigating common challenges when working with real-world data: The Data Learning Paradigm

  2025cites this paper
• Integrating Doppler LiDAR and machine learning into land-use regression model for assessing contribution of vertical atmospheric processes to urban PM2.5 pollution.

  2024cites this paper
• High-spatiotemporal-resolution mapping of PM2.5 traffic source impacts integrating machine learning and source-specific multipollutant indicator.

  2024cites this paper
• A Multi-Modal Deep-Learning Air Quality Prediction Method Based on Multi-Station Time-Series Data and Remote-Sensing Images: Case Study of Beijing and Tianjin

  2024cites this paper
• Dynamic Traffic Data in Machine-Learning Air Quality Mapping Improves Environmental Justice Assessment.

  2024cites this paper
• On Memory-Based Precise Calibration of Cost-Efficient NO2 Sensor Using Artificial Intelligence and Global Response Correction

  2024cites this paper
• Synergistic control potential of flue gas pollutants under Ultra-Low emission standards in waste incineration plants.

  2024cites this paper
• High spatio-temporal resolution predictions of PM2.5 using low-cost sensor data

  2024cites this paper
• Effect of Three-Dimensional-Printed Thermoplastics Used in Sensor Housings on Common Atmospheric Trace Gasses

  2024cites this paper
• Statistical data pre-processing and time series incorporation for high-efficacy calibration of low-cost NO2 sensor using machine learning

  2024cites this paper
• Emissions and exposure to NOx, CO, CO2 and PM2.5 from a Gas Stove Using Reference and Low-Cost Sensors

  2024cites this paper
• Low-Cost Indoor Sensor Deployment for Predicting PM2.5 Exposure

  2024cites this paper
• Machine-learning-based precise cost-efficient NO2 sensor calibration by means of time series matching and global data pre-processing

  2024cites this paper
• Cost-Efficient measurement platform and machine-learning-based sensor calibration for precise NO2 pollution monitoring

  2024cites this paper
• Source Tracing of PM2.5 in a Metropolitan Area Using a Low-Cost Air Quality Monitoring Network: Case Study of Denver, Colorado, USA

  2024cites this paper
• Investigating the impact of urban-environmental factors on air pollutants: a land use regression model approach and health risk assessment

  2024cites this paper
• Do Machine Learning Methods Improve Prediction of Ambient Air Pollutants with High Spatial Contrast? A Systematic Review.

  2024cites this paper
• National, satellite-based land-use regression models for estimating long-term annual NO2 exposure across India

  2024cites this paper
• Combining Google traffic map with deep learning model to predict street-level traffic-related air pollutants in a complex urban environment.

  2024cites this paper
• High-performance machine-learning-based calibration of low-cost nitrogen dioxide sensor using environmental parameter differentials and global data scaling

  2024cites this paper
• An overview of outdoor low-cost gas-phase air quality sensor deployments: current efforts, trends, and limitations

  2024cites this paper
• Spatiotemporal estimates and health risks of atmospheric trace metals across Hong Kong during 2016–2020

  2024cites this paper
• ViTAE-SL: A vision transformer-based autoencoder and spatial interpolation learner for field reconstruction

  2024cites this paper
• Theoretical Study to Support Proposed Framework for Spatial Modeling of PM2.5 Concentration in Pekanbaru City

  2024cites this paper
• Concentración de PM2.5 y su relación con el flujo vehicular y las condiciones meteorológicas del norte de Ocaña (Colombia)

  2024cites this paper
• Assessing the Spatiotemporal Characteristics, Factor Importance, and Health Impacts of Air Pollution in Seoul by Integrating Machine Learning into Land-Use Regression Modeling at High Spatiotemporal Resolutions.

  2023cites this paper
• Intercomparison of Six National Empirical Models for PM2.5 Air Pollution in the Contiguous US

  2023cites this paper
• Fundamentals of low-cost aerosol sensor design and operation

  2023cites this paper
• The spatial distribution mechanism of PM2.5 and NO2 on the eastern coast of China.

  2023cites this paper
• Assessing the impact of COVID-19 lockdown on fine-scale air quality across a heavy-pollution city using low-cost sensors

  2023cites this paper
• Spatial Distribution of Multiple Atmospheric Pollutants in China from 2015 to 2020

  2023cites this paper
• Robust augmented estimation for hourly PM2.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_{2.5}$$\end{document} usi

  2023cites this paper
• Spatiotemporal informer: A new approach based on spatiotemporal embedding and attention for air quality forecasting.

  2023cites this paper
• Developing high-resolution PM2.5 exposure models by integrating low-cost sensors, automated machine learning, and big human mobility data

  2023cites this paper
• Room Temperature Mixed-Potential Solid-Electrolyte NO2 Sensor for Environmental Monitoring

  2023cites this paper
• Assessing Regional Ecosystem Conditions Using Geospatial Techniques—A Review

  2023cites this paper
• A machine learning-based approach for fusing measurements from standard sites, low-cost sensors, and satellite retrievals: Application to NO2 pollution hotspot identification

  2023cites this paper
• Generating High Spatial Resolution Exposure Estimates from Sparse Regulatory Monitoring Data.

  2023cites this paper
• Intraurban NO2 hotspot detection across multiple air quality products

  2023cites this paper
• Estimating the geographical patterns and health risks associated with PM2.5-bound heavy metals to guide PM2.5 control targets in China based on machine-learning algorithms.

  2023cites this paper
• Predicting Atmospheric Particle Phase State Using an Explainable Machine Learning Approach Based on Particle Rebound Measurements.

  2023cites this paper
• Low‐cost air quality monitoring networks for long‐term field campaigns: A review

  2023cites this paper
• Exposure estimates of PM2.5 using the land-use regression with machine learning and microenvironmental exposure models for elders: Validation and comparison

  2023cites this paper
• Forecasting the concentration of NO2 using statistical and machine learning methods: A case study in the UAE

  2022cites this paper
• Using Low-Cost Sensor Networks: Considerations to Help Reveal Neighborhood-Level Exposure Disparities.

  2022cites this paper
• A Type-2 Fuzzy Time Series Model for Pattern Similarity Analysis: A Case Study on Air Quality Forecasting

  2022cites this paper
• A Simple and Effective Random Forest Refit to Map the Spatial Distribution of NO 2 Concentrations

  2022cites this paper
• A Simple and Effective Random Forest Refit to Map the Spatial Distribution of NO2 Concentrations

  2022cites this paper
• National Land Use Regression Model for NO2 Using Street View Imagery and Satellite Observations.

  2022cites this paper
• Smart systems engineering contributing to an intelligent carbon-neutral future: opportunities, challenges, and prospects

  2022cites this paper
• Towards air quality particulate-matter monitoring using low-cost sensor data and visual exploration techniques: case study of Kisumu, Kenya

  2022cites this paper
• Impacts of pollution heterogeneity on population exposure in dense urban areas using ultra-fine resolution air quality data.

  2022cites this paper
• Data-Driven Machine Learning in Environmental Pollution: Gains and Problems.

  2022cites this paper
• Tutorial: Guidelines for implementing low-cost sensor networks for aerosol monitoring

  2022cites this paper
• Analysis of the temporal and spatial pattern of air pollution and the heterogeneity of its influencing factors in central Inner Mongolia from 2016 to 2018

  2022cites this paper