Black carbon, organic carbon, and nitrogen oxide emission factors for traffic and domestic heating in an urban environment

Urban air quality deterioration has different reasons in warm and cold seasons. During summer, the photochemical production of secondary air pollution causes problems (ozone, secondary particles), while domestic heating significantly increases the primary emission during winter. In addition, the elevated emission is concentrated in a shallow mixing layer that leads to high air pollution levels. The source profile of domestic heating depends on the heating method. In European cities gas heating and wood combustion are the most common heating methods. Although distant heating is also commonly used, its emissions appear at the industrial source and are negligible for local air quality (similarly for less spread electric heating). In this work, we determined the emission ratios (ER) and emission factors (EF) of black carbon (BC), organic carbon (OC), and nitrogen oxide (NO_X) in an urban environment, during the heating season. BC and OC concentrations were measured by the Carbonaceous Aerosol Specification System (CASS, Aerosol d.o.o, Slovenia), while NO_X data was recorded by a nCLD-AL² NO_X analyzer (Eco Physics AG., Switzerland). The monitoring system was complemented by a Carbocap GMP-343 CO₂ sensor (Vaisala, Finland) in order to measure the CO₂ concentration for the EF calculation. The measurement took place in the atmospheric monitoring station of the Aerosol d.o.o in Ljubljana, Slovenia, which can be characterized as an urban background location. The source apportionment was implemented by using the Aethalometer model – multilinear regression combination (AM-MLR) assuming two major sources of urban air pollution: traffic and domestic heating. Fossil fuel-derived BC (BC^FF) concentration was assumed as the tracer of traffic emission, while biomass burning-related BC (BC^BB) was considered as the tracer of domestic heating. The Aethalometer model provides the source-specific (FF- or BB-derived) BC component. During the AM-MLR method, we supposed that the source-specific pollution component is correlated with the corresponding BC component (BC^FF or BC^BB). The slope of the regression line provides the ER, while the ratio of CO₂ to the other pollutants can be converted to EF (g(kg fuel)^-1) using the carbon balance approach. We applied the AM-MLR method to the dataset for the winter of 2021-2022 and determined the ER and EF values. The obtained EFs for traffic-related BC, OC, and NO_X are 0.39, 0.33, and 0.03 g(kg fuel)^-1 respectively, while the heating-related EFs are 0.13, 0.48, and0.01 g(kg fuel)^-1 respectively. This work provided real-world emission factor data of a city that can help to estimate the total BC, OC, and NO_X emissions in a city based on the sold fuel or consumed wood and gas.