Improving Fugitive Dust Emission Inventory from Construction Sector Using UAV Images Recognition

Over rapidly developing and urbanizing countries, frequent construction activities are the primary drivers behind the substantial emissions and contributors of fugitive dust. In this study, an innovative method was developed to compile a high-resolution spatiotemporal emission inventory from construction sector, utilizing unmanned aerial vehicle (UAV) images. This methodology offered detailed activity level information by distinguishing various types of construction lands and equipment. Focusing on the Shunde District of Guangdong in China, the new emission inventory derived from this method highlighted that travel, topsoil excavation, and loading collectively contributed up to 90% of particulate matter (PM) emissions during the earthwork phase. Moreover, this new inventory rectified the tendency of traditional methods to underestimate PM_(10) emissions and overestimate PM_(2.5) emissions, while revealing the non-linear relationship between PM emissions and construction area. This improved PM emission inventory appeared to precisely identify major emission hotspots and enhanced performance of the Community Multi-scale Air Quality (CMAQ) model, and the correlation coefficient (R-value) is 0.08 ± 0.02 higher than that of the traditional emission inventory. Post integration of monitoring data through the Software for the Modeled Attainment Test - Community Edition (SMAT-CE), the contributions of construction dust to local PM_(10) and PM_(2.5) concentrations were estimated at 3.27 ± 0.8 µg m^(-3) and 1.11 ± 0.27 µg m^(-3), respectively, with more pronounced impacts observed in the central, northwestern, and south-central zones of the study region. This study provides valuable insight for improving construction dust and PM emission inventories, which should be benefiting the development of air pollution prevention and control strategies over this study area as well as other rapidly growing urban areas.