Dispersion of particulate matter (PM<sub>2.5</sub>) from wood combustion for residential heating: optimization of mitigation actions based on large-eddy simulations

Abstract

Abstract. Many cities in the world experience significant air pollution from residential wood combustion. Such an advection–diffusion problem as applied to geographically distributed small-scale pollution sources presently does not have a satisfactory theoretical or modeling solution. For example, statistical models do not allow for pollution accumulation in local stagnation zones – a type of phenomena that is commonly observed over complex terrain. This study applies a Parallelized Atmospheric Large-eddy simulation Model (PALM) to investigate dynamical phenomena that control variability and pathways of the atmospheric pollution emitted by wood-burning household stoves. The model PALM runs at spatial resolution of 10 m in an urban-sized modeling domain of 29 km by 35 km with a real spatial distribution of the pollution source and with realistic surface boundary conditions that characterize a medium-sized urban area fragmented by water bodies and hills. Such complex geography is expected to favor local air quality hazards, which makes this study of general interest. The case study here is based on winter conditions in Bergen, Norway. We investigate the turbulent diffusion of a passive scalar associated with small-sized particles (PM2.5) emitted by household stoves. The study considers air pollution effects that could be observed under different policy scenarios of stove replacement; modern woodstoves emit significantly less PM2.5 than the older ones, but replacement of stoves is a costly and challenging process. We found significant accumulation of near-surface pollution in the local stagnation zones. The simulated concentrations were larger than the concentrations obtained only due to the local PM2.5 emission, thus indicating dominant transboundary contribution of pollutants for other districts. We demonstrate how the source of critical pollution can be attributed through model disaggregation of emission from specific districts. The study reveals a decisive role of local air circulations over complex terrain that makes high-resolution modeling indispensable for adequate management of the urban air quality. This modeling study has important policy-related implications. Uneven spatial distribution of the pollutants suggests prioritizing certain limited urban districts in policy scenarios. We show that focused efforts towards stove replacement in specific areas may have a dominant positive effect on the air quality in the whole municipality. The case study identifies urban districts where limited incentives would result in the strongest reduction of the population's exposure to PM2.5.