Geospatial artificial intelligence for estimating daytime and nighttime nitrogen dioxide concentration variations in Taiwan: A spatial prediction model

Jennieveive Babaan; Pei-Yi Wong; Pau-Chung Chen; Hsiu-Ling Chen; Shih-Chun Candice Lung; Yu-Cheng Chen; Chih-Da Wu

doi:10.1016/j.jenvman.2024.121198

Geospatial artificial intelligence for estimating daytime and nighttime nitrogen dioxide concentration variations in Taiwan: A spatial prediction model

J Environ Manage. 2024 May 20:360:121198. doi: 10.1016/j.jenvman.2024.121198. Online ahead of print.

Authors

Jennieveive Babaan¹, Pei-Yi Wong², Pau-Chung Chen³, Hsiu-Ling Chen⁴, Shih-Chun Candice Lung⁵, Yu-Cheng Chen⁶, Chih-Da Wu⁷

Affiliations

¹ Department of Geodetic Engineering, University of the Philippines Diliman, Quezon City, Philippines.
² Department of Environmental and Occupational Health, National Cheng Kung University, Tainan City, Taiwan.
³ National Institute of Environmental Health Sciences, National Health Research Institutes, Miaoli County, Taiwan; Institute of Environmental and Occupational Health Sciences, National Taiwan University College of Public Health, Taipei City, Taiwan; Department of Environmental and Occupational Medicine, National Taiwan University Hospital, Taipei City, Taiwan; Department of Public Health, National Taiwan University College of Public Health, Taipei City, Taiwan.
⁴ Department of Food Safety/Hygiene and Risk Management, College of Medicine, National Cheng Kung University, Tainan, Taiwan.
⁵ Research Center for Environmental Changes, Academia Sinica, Taipei City, Taiwan; Department of Atmospheric Sciences, National Taiwan University, Taipei, Taiwan; Institute of Environmental Health, School of Public Health, National Taiwan University, Taipei, Taiwan.
⁶ National Institute of Environmental Health Sciences, National Health Research Institutes, Miaoli County, Taiwan.
⁷ National Institute of Environmental Health Sciences, National Health Research Institutes, Miaoli County, Taiwan; Department of Geomatics, National Cheng Kung University, Tainan City, Taiwan; Innovation and Development Center of Sustainable Agriculture, National Chung Hsing University, Taichung City, Taiwan; Research Center for Precision Environmental Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan. Electronic address: chidawu@mail.ncku.edu.tw.

PMID: 38772239
DOI: 10.1016/j.jenvman.2024.121198

Abstract

Nitrogen dioxide (NO₂) is a major air pollutant primarily emitted from traffic and industrial activities, posing health risks. However, current air pollution models often underestimate exposure risks by neglecting the bimodal pattern of NO₂ levels throughout the day. This study aimed to address this gap by developing ensemble mixed spatial models (EMSM) using geo-artificial intelligence (Geo-AI) to examine the spatial and temporal variations of NO₂ concentrations at a high resolution of 50m. These EMSMs integrated spatial modelling methods, including kriging, land use regression, machine learning, and ensemble learning. The models utilized 26 years of observed NO₂ measurements, meteorological parameters, geospatial layers, and social and season-dependent variables as representative of emission sources. Separate models were developed for daytime and nighttime periods, which achieved high reliability with adjusted R² values of 0.92 and 0.93, respectively. The study revealed that mean NO₂ concentrations were significantly higher at nighttime (9.60 ppb) compared to daytime (5.61 ppb). Additionally, winter exhibited the highest NO₂ levels regardless of time period. The developed EMSMs were utilized to generate maps illustrating NO₂ levels pre and during COVID restrictions in Taiwan. These findings could aid epidemiological research on exposure risks and support policy-making and environmental planning initiatives.

Keywords: Air quality; Ensemble learning; Geo-AI; Land use regression; Machine learning; NO(2).