Investigating the impact of urban landscape composition and configuration on PM2.5 concentration under the LCZ scheme: A case study in Nanchang, China

•Integrating LCZ scheme to analyze the effects of landscape composition and configuration on PM2.5.•Built categories demonstrate higher PM2.5 concentrations than natural categories.•High/compact built classes demonstrate higher PM2.5 concentrations than low/open built ones.•HRE and AR are the top two significant landscape configuration metrics affecting PM2.5.•Effects of landscape composition and configuration on PM2.5 is season-independent. Urban landscape pattern alters the distribution of fine particulate matter (PM2.5) by affecting “source and sink” directly and microclimate indirectly. Most previous works investigate their relationship by landscape metrics that are difficult to be applied in sustainable urban planning. This study innovatively linked PM2.5 with the local climate zone (LCZ) which is a visible landscape pattern classification framework at urban scale. Impacts of urban landscape composition and configuration on PM2.5 were evaluated through PM2.5 differences analyzes of inter-LCZ categories and intra-built category and correlation analysis between PM2.5 and landscape metrics adopted in LCZ scheme. Results show that significant PM2.5 differences exist both inter-LCZ categories and intra-built category, implying LCZ scheme can reflect the urban spatial variation of PM2.5. The natural category always has a lower PM2.5 concentration than the built category. PM2.5 concentrations “compact > open” and “high-rise > mid-rise > low-rise” in built category generally exist throughout the year. Height of roughness elements (HRE) and aspect ratio (AR) show the most significant correlation with PM2.5. This study provides a novel perspective investigating the impact of urban landscape pattern on PM2.5 and scientific guidance for sustainable urban planning in regard to air pollution mitigation.