Numerical study on dust distribution in an ultra-large-scale computational domain based on an optimised dust injection condition

Globally, numerous residential communities have been developed close to mining operations, resulting in air quality levels that often fall below the World Health Organisation (WHO) standards due to Fine particulate matter (PM10 and PM2.5), which poses significant health risks, particularly to vulnerable populations such as children and pregnant individuals. Therefore, it is important to investigate dust dispersion within realistic urban building layouts. In this field Discrete Phase Model (DPM), widely used in Computational Fluid Dynamics (CFD), has limited applicability in large-scale studies due to its high computational cost. Alternatively, the Species Transport Model (STM) offers a more efficient approach but suffers from unclear dust emission, often compromising accuracy. This study introduces a novel dust emission function developed using Python and Hermite Interpolation, integrated into an STM-based simulation. Findings reveal that high dust concentrations tend to accumulate in low-wind-speed zones dominated by vortex effects. Furthermore, the effects of varying wind speeds and directions on dust dispersion were systematically analysed. The results indicate that high wind velocities cause dust accumulation on the windward side. Based on fractal dimension analysis, an evaluation framework for dust distribution was established. The findings demonstrate that the complexity of dust dispersion boundaries is governed by wind direction and that more irregular transport boundaries lead to a broader spatial extent of dust distribution. Overall, the new dust injection method combined with STM enables efficient large-scale urban dust dispersion simulation. This approach provides a practical reference for researchers and policymakers aiming to design dust mitigation strategies in mining-affected communities.