Research on the fractal characteristics of soil particle distribution and their correlation with soil properties in various forest types: insights from eastern mountainous area of Liaoning Province in Northern China

Abstract Soil particle-size distribution (PSD) is one of the most important physical attributes due to its great influence on soil properties related to water movement, soil management, and degradation. Thus, characterizing variations in the PSDs of soil are a major issue in environmental research. To date, the fractal model could well characterize PSD. Furthermore, scientific understanding and evaluation of forest soil quality is the basis for guiding ecological restoration, planting and improvement of forest soil quality of degraded stands and select suitable tree species for afforestation. Therefore, in this research the typical forest types: Pinus koraiensis, Pinus sylvestris var. mongholica, Quercus mongolica, Juglans mandshurica and mixed conifer-broadleaf (Pinus koraiensis × Quercus mongolica) forests in the mountains of eastern Liaoning were taken as the study objects. The topsoil (0–20 cm) and sub-topsoil (20–40 cm) samples, and litter were collected, and the relationship between the soil physiochemical properties and particle size characteristics under natural cultivation measures were evaluated and compared. The results indicated that the soil layer composition of forest land was mainly sand, followed by silt and clay. The particles size characteristics showed well sorted, positive skewness and narrow kurtosis state, and the singular fractal dimension (D) of soil was between 1.82 and 2.75. The mean particle size, D, litter and soil properties in forested land were higher than those in non-forest cover control plots, and the Ds showed an increasing trend from conifer to broadleaf forests and from pure forest of single species to mixed conifer-broadleaf forests, and the recovery effect of topsoil soil was better. Meanwhile, the multifractal dimensions are displayed as capacity dimension > correlation dimension > information dimension, indicating that the PSD is not completely ideal and uniform, thus it is still necessary to use the D to evaluate soil quality in combination with multifractal analysis. In conclusion, we concluded that D is a sensitive and useful index because it quantifies changes in soil properties and it is highly recommended that broadleaf and mixed conifer-broadleaf forests are suitable for local afforestation for soil degradation restoration purpose.