Hong Kong Traffic Emissions and Volumes, 2003–2023: Road-Link NOx and PM2.5 Data with Hourly Profiles

This dataset provides Hong Kong traffic emission data and traffic volumes for the whole road network from 2003 to 2023. It contains reconstructed road-segment-level nitrogen oxide (NOx) and fine particulate matter (PM2.5) emissions, together with hourly traffic volumes for eight vehicle classes. For each year, the data describe an annual-average 24-hour traffic and emission profile across the modelled road network in Hong Kong. The reconstruction combines historical traffic counts, hourly traffic profiles, camera-derived vehicle-class proportions, and traffic-speed observations. Machine-learning models estimate missing traffic components on unmonitored road links. Hourly traffic volumes for each vehicle class are combined with speed-dependent EMFAC-HK emission factors to calculate emissions. District- and year-specific calibration factors are applied to traffic volumes and emissions to address spatial and temporal prediction biases. The dataset covers motorcycles, private cars, taxis, light goods vehicles, heavy goods vehicles, light-duty buses, heavy-duty buses, and franchised double-decker buses. Data are distributed as 21 annual CSV files, each compressed in a separate ZIP archive. Each file contains road identifiers, road-link geometries, and hourly traffic volumes and emissions by vehicle class. Traffic volumes are expressed as vehicle counts within each hourly interval. Emissions are expressed in grams per kilometre of road link for the corresponding hourly interval. Numerical traffic and emission values are rounded to three decimal places. The hourly profiles represent annual-average conditions for each year and do not provide observations for individual calendar dates. PM2.5 emissions include exhaust sources only and exclude brake wear, tyre wear, and road-dust resuspension. The dataset represents traffic activity and emissions and does not directly provide ambient air pollutant concentrations or personal exposure. These data support research on long-term road transport emissions in Hong Kong, hourly traffic patterns, vehicle-class contributions, spatial emission inventories, and inequalities in traffic-emission burdens. Road-link geometries allow the data to be mapped and aggregated to census tracts, districts, or other spatial units. Please refer to the following paper for a detailed description of the data sources, reconstruction methods, and validation procedures: Niu, C.; Chen, Q.; Wang, A.; Xu, J. Disentangling Near-Road Emission Inequities in Hong Kong through Data-Driven Spatiotemporal Traffic Dynamics. Environ. Sci. Technol. 2026, 60 (11), 8554–8570. https://doi.org/10.1021/acs.est.5c14619