This paper presents a soil-derived dust emission scheme designed to model the atmospheric dust cycle. The scheme incorporates two key factors that influence erodible surfaces: (1) the size distribution of erodible particles, which affects the erosion threshold and emission strength, and (2) surface roughness, which influences the wind friction velocity. These parameters are integrated into a formulation of the threshold wind friction velocity, adapting a size-dependent parameterization from Iversen and White (1982) and applying a drag partition scheme from Arya (1975). This parameterization is included in a horizontal flux equation proposed by White (1979), allowing for the assignment of specific production rates to each soil size range. The dust flux F is considered as a fraction of the total horizontal flux G, with the ratio F/G determined by the soil's clay content. The computed mass fluxes depend on soil size distribution, roughness lengths, and wind friction velocity. The scheme has been validated against experimental data, showing satisfactory agreement and reducing uncertainties in dust flux simulations. The paper discusses the importance of soil-derived dust as a major component of natural atmospheric aerosols, produced by aeolian erosion in arid and semiarid areas. It highlights the sensitivity of dust emission to climatic parameters such as wind velocity and precipitation. The scheme aims to improve the simulation of the desert dust cycle by accounting for variability in source erodibility through a parameterization of the erosion threshold based on soil size distribution and surface roughness. This parameterization is included in dust flux expressions that have been experimentally tested. The paper also discusses the physical processes of dust emission, including the effects of particle size, surface roughness, and wind shear stress on erosion thresholds. The threshold friction velocity is parameterized based on particle size and the effect of roughness elements, which control wind shear stress on the surface. The paper also addresses the drag partition between roughness elements and erodible surfaces, which affects the erosion threshold. The scheme is validated against experimental data, showing good agreement and reducing uncertainties in dust flux simulations. The paper concludes that the proposed scheme provides a more accurate representation of the dust cycle, taking into account the complex interactions between soil properties, wind conditions, and erosion processes.This paper presents a soil-derived dust emission scheme designed to model the atmospheric dust cycle. The scheme incorporates two key factors that influence erodible surfaces: (1) the size distribution of erodible particles, which affects the erosion threshold and emission strength, and (2) surface roughness, which influences the wind friction velocity. These parameters are integrated into a formulation of the threshold wind friction velocity, adapting a size-dependent parameterization from Iversen and White (1982) and applying a drag partition scheme from Arya (1975). This parameterization is included in a horizontal flux equation proposed by White (1979), allowing for the assignment of specific production rates to each soil size range. The dust flux F is considered as a fraction of the total horizontal flux G, with the ratio F/G determined by the soil's clay content. The computed mass fluxes depend on soil size distribution, roughness lengths, and wind friction velocity. The scheme has been validated against experimental data, showing satisfactory agreement and reducing uncertainties in dust flux simulations. The paper discusses the importance of soil-derived dust as a major component of natural atmospheric aerosols, produced by aeolian erosion in arid and semiarid areas. It highlights the sensitivity of dust emission to climatic parameters such as wind velocity and precipitation. The scheme aims to improve the simulation of the desert dust cycle by accounting for variability in source erodibility through a parameterization of the erosion threshold based on soil size distribution and surface roughness. This parameterization is included in dust flux expressions that have been experimentally tested. The paper also discusses the physical processes of dust emission, including the effects of particle size, surface roughness, and wind shear stress on erosion thresholds. The threshold friction velocity is parameterized based on particle size and the effect of roughness elements, which control wind shear stress on the surface. The paper also addresses the drag partition between roughness elements and erodible surfaces, which affects the erosion threshold. The scheme is validated against experimental data, showing good agreement and reducing uncertainties in dust flux simulations. The paper concludes that the proposed scheme provides a more accurate representation of the dust cycle, taking into account the complex interactions between soil properties, wind conditions, and erosion processes.