Machine learning based estimation of urban on-road CO2 concentration in Seoul

The urban on-road CO2 emissions will continue to increase, it is therefore essential to manage urban on-road CO2 concentrations for effective urban CO2 mitigation. However, limited observations of on-road CO2 concentrations prevents a full understanding of its variation. Therefore, in this study, a...

Full description

Saved in:
Bibliographic Details
Published in:Environmental research 2023-08, Vol.231, p.116256-116256, Article 116256
Main Authors: Park, Chaerin, Jeong, Sujong, Kim, Chongmin, Shin, Jaewon, Joo, Jaewon
Format: Article
Language:English
Subjects:
CO2
Emissions
Machine learning
Mobile observation
Traffic
Urban
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The urban on-road CO2 emissions will continue to increase, it is therefore essential to manage urban on-road CO2 concentrations for effective urban CO2 mitigation. However, limited observations of on-road CO2 concentrations prevents a full understanding of its variation. Therefore, in this study, a machine learning-based model that predicts on-road CO2 concentration (CO2traffic) was developed for Seoul, South Korea. This model predicts hourly CO2traffic with high precision (R2 = 0.8 and RMSE = 22.9 ppm) by utilizing CO2 observations, traffic volume, traffic speed, and wind speed as the main factors. High spatiotemporal inhomogeneity of hourly CO2traffic over Seoul, with 14.3 ppm by time-of-day and 345.1 ppm by road, was apparent in the CO2traffic data predicted by the model. The large spatiotemporal variability of CO2traffic was related to different road types (major arterial roads, minor arterial roads, and urban highways) and land-use types (residential, commercial, bare ground, and urban vegetation). The cause of the increase in CO2traffic differed by road type, and the diurnal variation of CO2traffic differed according to land-use type. Our results demonstrate that high spatiotemporal on-road CO2 monitoring is required to manage urban on-road CO2 concentrations with high variability. In addition, this study demonstrated that a model using machine learning techniques can be an alternative for monitoring CO2 concentrations on all roads without conducting observations. Applying the machine learning techniques developed in this study to cities around the world with limited observation infrastructure will enable effective urban on-road CO2 emissions management. •A machine learning based model predicting on-road CO2 concentration was developed.•The model estimates hourly on-road CO2 concentrations with high precision.•On-road CO2 in Seoul was highly variable (14.3 ppm by time and 345.1 ppm by road).•The reason for the increase in on-road CO2 concentration varies by road type.•Speed (arterial roads) and volume (urban highways) explained increased CO2.
ISSN:0013-9351
1096-0953
DOI:10.1016/j.envres.2023.116256