Background: Residing in greener areas has several health benefits, but no study to date has examined the effects of greenness on metabolic syndrome (MetS). We aimed to assess associations between residential greenness and MetS prevalence in China, and to explore whether air pollution and physical activity mediated any observed associations.Methods: We analyzed data from 15,477 adults who participated in the 33 Communities Chinese Health Study during 2009. We defined MetS according to standard guidelines for Chinese populations. Residential greenness was estimated using the Normalized Difference Vegetation Index (NDVI), the Soil Adjusted Vegetation Index (SAVI), and the Vegetation Continuous Field (VCF). We used generalized linear mixed models to assess the associations between greenness and MetS, and mediation analyses to explore potential mechanisms underlying the associations.Results: Higher greenness levels were associated with lower odds of MetS [e.g., for every interquartile range increase of NDVI500-m, SAVI(500-m), and VCF500-m the adjusted odds ratio of MetS was 0.81 (95% confidence interval: 0.70-0.93), 0.80 (95% confidence interval: 0.69-0.93), and 0.91 (95% confidence interval: 0.83-1.00), respectively]. The direction and the magnitude of the associations persisted in several sensitivity analyses. Stratified analyses showed that age and household income modified the associations, with greater effect estimates observed in participants younger than 65 years old or those with higher household income. Particulate matter with an aerodynamic diameter <= 10 mu m nitrogen dioxide, and ozone mediated 2.1-20.3% of the associations between greenness and MetS; no evidence of mediation was observed for physical activity.Conclusions: Our findings suggest a beneficial association for residential greenness and MetS in Chinese urban dwellers, especially for participants younger than 65 years old and those with higher household income. Particulate matter with an aerodynamic diameter <= 10 mu m nitrogen dioxide and ozone, but not physical activity, may only partially mediate the association.