Abstract: Permeability reflects the ability of sand and soil bodies' internal pores to transmit fluids, and the permeability coefficient serves as a crucial indicator to assess the permeability of soil bodies. Current research on permeability predominantly focuses on aspects such as porosity, coefficient of uniformity, coefficient of curvature and particle size. Calcareous sand, characterized by its special biological origin, consists of varying proportions of coarse and fine particles. The coarse particles form the skeletal structure, while the fine particles alter the pore distribution. Additionally, the particle shapes are highly irregular. To investigate the influence of coarse particle content and particle shape on the permeability of calcareous sand, we employed the PartAn 3D particle image analyzer to obtain shape parameters of the samples. Furthermore, a constant head permeability test was conducted to explore the correlation between the physical property parameters of mixed sand and the measured permeability coefficients. The research findings indicate that the permeability coefficient of mixed sand does not solely increase with higher coarse particle content; instead, it exhibits a trend of decreasing first and then increasing. Among the quantified shape parameters, concavity (C_a), convexity (C_v), and maximum Feret length (F_L) have a greater impact on permeability. Based on the correlation analysis, a preliminary predictive model for the permeability coefficient, incorporating the shape parameters of calcareous sand particles, has been established. This model can provide a scientific basis for calculating and analyzing the seepage field of artificially filled islands and reefs in the South China Sea.

Key words: permeability coefficient, coarse particle content, particle shape, correlation analysis, predictive model