Related Experiment Video
Updated: May 3, 2026

Assessing the Particulate Matter Removal Abilities of Tree Leaves
Published on: October 7, 2018
Unveiling the spatial and temporal effects of urban vegetation coverage on atmospheric formaldehyde pollution in
Yuanyun Gao1, Chengxin Zhang2, Yuwen Yang3
1Key Laboratory of Environmental Optics & Technology, Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, 230031, China; Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China, Hefei, 230026, China; Nanjing Institute of Environmental Sciences, Ministry of Ecology and Environment of China, 8 Jiang Wang Miao St., Nanjing, 210042, China.
Abstract:
Urban vegetation coverage is widely promoted to improve livability and reduce pollutant exposure, yet its air-quality effects remain debated. Formaldehyde (HCHO), formed largely via oxidation of anthropogenic and biogenic VOCs, poses significant health risks and is a precursor of O3 and PM2.5. Here we examined how urban vegetation coverage was associated with HCHO across urbanization stages in Chinese megacities during 2005-2017. We integrated monthly OMI/Aura HCHO columns, MODIS-based Normalized Difference Vegetation Index (NDVI), and Landsat-derived landscape metrics, harmonized on a common 0.1 grid. Cities were grouped into five urbanization classes via k-means on z-scored indicators (GDP per capita, population density, built-up ratio, and nightlights), with k = 5 supported by elbow, silhouette, and Calinski-Harabasz criteria. Associations between NDVI and HCHO were assessed using SRCC and spatial heterogeneity was evaluated with Geographically Weighted Regression (GWR). Across stages, the NDVI-HCHO association shifted from weakly negative in Class V (SRCC = -0.09, p < 0.001) to strongly positive in Class III (SRCC = 0.81, p < 0.001), then weakened in Classes II and I (0.63 and 0.72). Regionally, BTH and YRD showed concurrent increases in NDVI (3.29 % and 5.94 %) and HCHO (22.53 % and 8.7 %), whereas CY experienced a 10.18 % NDVI rise but a 9.74 % HCHO decline. GWR highlighted LPI as a robust structural predictor of HCHO variability across stages. These patterns are consistent with context-dependent interactions among BVOC potential, anthropogenic VOC/NOx, and meteorology. Our results caution against one-size-fits-all greening: the air-quality outcome of urban vegetation coverage depends on urbanization stage and landscape configuration. Differentiated planning and controlling anthropogenic precursors in more urbanized settings and optimizing vegetation type/patch structure around the Class III "peak association" can better align greening with HCHO mitigation.
Related Concept Videos
Global Climate Change
Threats to Biodiversity

