Scale-dependent associations between urban landscape characteristics and PM2.5 concentrations across contrasting landscape backgrounds in Beijing

Associations between urban landscape characteristics and PM 2.5 concentrations may vary with spatial scale and landscape background, yet these variations remain insufficiently understood. We examined these relationships using annual mean PM 2.5 concentrations from 13 monitoring stations in Beijing during 2021–2025 and high-resolution urban landscape data. We fitted XGBoost models using eight predictors, including five two-dimensional landscape indicators and three three-dimensional morphological indicators, within 1.0, 1.5, and 2.0 km buffers. SHAP analysis was then used to assess the relative importance of these predictors in model predictions. A dominance-ratio approach was used to classify the surroundings of each station into blue-, green-, or grey-space backgrounds. The highest-ranked predictor changed across scales, from frontal area index ( FAI ) at 1.0 km (mean absolute SHAP = 0.824 µg m −2 ) to building height ( BH ) at 1.5 km (0.813 µg m −2 ) and street canyon height-to-width ratio ( HW ) at 2.0 km (0.756 µg m −2 ). At each scale, the single station classified as blue-space had the lowest PM 2.5 concentration, whereas grey-space backgrounds generally showed the highest. Under the 1.0 km classification, the 2023 increase was 1.1 µg m −2 at the blue-space site, compared with mean increases of 3.3 and 3.5 at green- and grey-space stations, respectively.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-15
DOI
https://doi.org/10.1038/s41598-026-80174-0
Primary Topic
Wind and Air Flow Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Scale-dependent associations between urban landscape characteristics and PM2.5 concentrations across contrasting landscape backgrounds in Beijing

Cong Sun, Jing Kang, Gang Wang, Jiong Zhou et al.
Scientific Reports
Wind and Air Flow Studies
article

Scale-dependent associations between urban landscape characteristics and PM2.5 concentrations across contrasting landscape backgrounds in Beijing

Cong Sun, Jing Kang, Gang Wang, Jiong Zhou, Mingsheng Li, Haijiang Liu, Peixin Yu, Zhe Liu, Yang Yu
article en

Abstract

Associations between urban landscape characteristics and PM 2.5 concentrations may vary with spatial scale and landscape background, yet these variations remain insufficiently understood. We examined these relationships using annual mean PM 2.5 concentrations from 13 monitoring stations in Beijing during 2021–2025 and high-resolution urban landscape data. We fitted XGBoost models using eight predictors, including five two-dimensional landscape indicators and three three-dimensional morphological indicators, within 1.0, 1.5, and 2.0 km buffers. SHAP analysis was then used to assess the relative importance of these predictors in model predictions. A dominance-ratio approach was used to classify the surroundings of each station into blue-, green-, or grey-space backgrounds. The highest-ranked predictor changed across scales, from frontal area index ( FAI ) at 1.0 km (mean absolute SHAP = 0.824 µg m −2 ) to building height ( BH ) at 1.5 km (0.813 µg m −2 ) and street canyon height-to-width ratio ( HW ) at 2.0 km (0.756 µg m −2 ). At each scale, the single station classified as blue-space had the lowest PM 2.5 concentration, whereas grey-space backgrounds generally showed the highest. Under the 1.0 km classification, the 2023 increase was 1.1 µg m −2 at the blue-space site, compared with mean increases of 3.3 and 3.5 at green- and grey-space stations, respectively.

Scientific Reports
Chinese Academy of Sciences (CN), China National Environmental Monitoring Center (CN), Institute of Geographic Sciences and Natural Resources Research (CN), University of Chinese Academy of Sciences (CN)
Wind and Air Flow Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon