Assessing the transferability of landuse regression models for ultrafine particles across two Canadian cities

Zalzal, Jad; Alameddine, Ibrahim; Khoury, Celine El; Minet, Laura; Shekarrizfard, Maryam; Weichenthal, Scott; Hatzopoulou, Marianne

doi:10.1016/j.scitotenv.2019.01.123

Cited by 16 publications

(10 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The midpoint of each segment of the road was used as a proxy for all sampling points at that segment location consistent with previous studies ((Shi et al, 2016), (Lim et al, 2019), (Hatzopoulou et al, 2017;Kerckhoffs et al, 2016;Weichenthal et al, 2016a;Weichenthal et al, 2016b;Kerckhoffs et al, 2017b;Zalzal et al, 2019)). The road segment was defined as between two intersections or traffic signals.…”

Section: Mobile Monitoringmentioning

confidence: 91%

Potential for developing independent daytime/nighttime LUR models based on short-term mobile monitoring to improve model performance

Qin

Yang

et al. 2021

Environmental Pollution

View full text Add to dashboard Cite

Land use regression model (LUR) is a widespread method for predicting air pollution exposure. Few studies have explored the performance of independently developed daytime/nighttime LUR models. In this study, fine particulate matter (PM 2.5 ), inhalable particulate matter (PM 10 ), and nitrogen dioxide (NO 2 ) concentrations were measured by mobile monitoring during non-heating and heating seasons in Taiyuan. Pollutant concentrations were higher in the nighttime than the daytime, and higher in the heating season than the non-heating season. Daytime/nighttime and full-day LUR models were developed and validated for each pollutant to examine variations in model performance. Adjusted coefficients of determination (adjusted R 2 ) for the LUR models ranged from 0.53e0.87 (PM 2.5 ), 0.53e0.85 (PM 10 ), and 0.33e0.67 (NO 2 ). The performance of the daytime/nighttime LUR models for PM 2.5 and PM 10 was better than that of the full-day models according to the results of model adjusted R 2 and validation R 2 . Consistent results were confirmed in the non-heating and heating seasons. Effectiveness of developing independent daytime/nighttime models for NO 2 to improve performance was limited. Surfaces based on the daytime/nighttime models revealed variations in concentrations and spatial distribution. In conclusion, the independent development of daytime/nighttime LUR models for PM 2.5 /PM 10 has the potential to replace full-day models for better model performance. The modeling strategy is consistent with the residential activity patterns and contributes to achieving reliable exposure predictions for PM 2.5 and PM 10 . Nighttime could be a critical exposure period, due to high pollutant concentrations.

show abstract

Section: Mobile Monitoringmentioning

confidence: 91%

Potential for developing independent daytime/nighttime LUR models based on short-term mobile monitoring to improve model performance

Qin

Yang

et al. 2021

Environmental Pollution

View full text Add to dashboard Cite

show abstract

“…As far as we are aware, there have been only two other studies that have transferred UFP models to other areas. Spatial UFP LUR models (Zalzal et al, 2019) based on measurements from mobile monitoring (i.e. onvehicle monitoring of UFP along the street network) -23 days in Montreal and 39 days in Toronto -were developed in Toronto (R 2 = 49%) and Montreal (R 2 = 41%).…”

Section: Model Performance and Transferabilitymentioning

confidence: 99%

Development and transferability of ultrafine particle land use regression models in London

Yang

Sterrantino

Fuller

et al. 2020

Science of The Total Environment

View full text Add to dashboard Cite

“…The models that we developed were able to explain roughly half of the observed variance of UFP and BC concentrations in the test set. Our model performance cannot be directly compared to other studies due to the differences in monitoring and analysis methods, but it is encouraging that the proportion of variance explained by our models falls within the range of values reported for UFP and BC LUR models developed for several European and North American cities. ,,,− Of note, our models combining CNN and LUR model predictions consistently outperformed either method on its own, though the improvement was relatively modest. These combined models demonstrated how predictions from CNN models can be used to augment predictions from other modeling approaches.…”

Section: Discussionmentioning

confidence: 65%

“…This general approach to exposure estimation may be particularly promising when combined with traditional and land use information as described above. More generally, the apparent robustness of CNN models to spatial variations in model errors may help to address known issues related to the transferability of LUR models between cities , by capturing additional land use information in images as it relates to ambient air pollution, though this hypothesis would require further research.…”

Section: Discussionmentioning

confidence: 99%

Predicting Within-City Spatial Variations in Outdoor Ultrafine Particle and Black Carbon Concentrations in Bucaramanga, Colombia: A Hybrid Approach Using Open-Source Geographic Data and Digital Images

Lloyd

Carter

Diaz

et al. 2021

Environ. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Outdoor ultrafine particles (UFP, <0.1 μm) and black carbon (BC) vary greatly within cities and may have adverse impacts on human health. In this study, we used a hybrid approach to develop new models to estimate within-city spatial variations in outdoor UFP and BC concentrations across Bucaramanga, Colombia. We conducted a mobile monitoring campaign over 20 days in 2019. Regression models were trained on land use data and combined with predictions from convolutional neural networks (CNN) trained to predict UFP and BC concentrations using satellite and street-level images. The combined UFP model (R 2 = 0.54) outperformed the CNN (R 2 = 0.47) and land use regression (LUR) models (R 2 = 0.47) on their own. Similarly, the combined BC model also outperformed the CNN and LUR BC models (R 2 = 0.51 vs 0.43 and 0.45, respectively). Spatial variations in model performance were more stable for the CNN and combined models compared to the LUR models, suggesting that the combined approach may be less likely to contribute to differential exposure measurement error in epidemiological studies. In general, our findings demonstrated that satellite and street-level images can be combined with a traditional LUR modeling approach to improve predictions of within-city spatial variations in outdoor UFP and BC concentrations.

show abstract

Assessing the transferability of landuse regression models for ultrafine particles across two Canadian cities

Cited by 16 publications

References 45 publications

Potential for developing independent daytime/nighttime LUR models based on short-term mobile monitoring to improve model performance

Potential for developing independent daytime/nighttime LUR models based on short-term mobile monitoring to improve model performance

Development and transferability of ultrafine particle land use regression models in London

Predicting Within-City Spatial Variations in Outdoor Ultrafine Particle and Black Carbon Concentrations in Bucaramanga, Colombia: A Hybrid Approach Using Open-Source Geographic Data and Digital Images

Contact Info

Product

Resources

About