Fractal dimensions of trapped sediment particle size distribution can reveal sediment retention ability of common plants in a dry-hot valley

Aboveground plant parts can trap moving sediment on slopes during concentrated water runoff, resulting in the formation of sediment mounds around plants. Plant morphological and functional traits may influence these sediment trapping patterns. Characterizing the sediment trapping abilities of common plants is of great interest when selecting appropriate plant species for erosion control. Predicting plant sediment trapping abilities using fractal dimensions of sediment particle size distribution could enhance the management and restoration of degraded lands with declining vegetation coverage in dry-hot valleys. Using flume experiments, we studied sediment retention abilities in six common plant species: the tree species Leucaena leucocephala and Melia azedarach; the shrub species Dodonaea viscose and Coriaria sinica; and the graminoids Heteropogon contortus and Eulaliopsis binate. Sediment trapping mass was measured at two ages (three-month-old and one-year-old seedlings). The particle size distribution of trapped sediment was determined using a laser particle size analyzer. The fractal characterizations (singular- and multi-fractal dimensions) of sediment particle size distribution were calculated using the fractal scale theory. Statistical analyses were performed to compare sediment trapping mass and the fractal characterizations in the six species at the two ages. C. sinica had the highest relative trapped sediment at both ages. Trapped sediment mass of one-year-old seedlings were significantly related to the singular fractal dimension (D), capacity dimension (D-0), entropy dimension (D-1), and correlation dimension (D-2) of trapped sediment particle size distribution. However, trapped sediment mass of three-month-old seedlings was only related to D, while that combining the data in both ages had significant correlations with D and D-0. Our results suggest that fractal dimension of sediment particle size distribution can predict sediment mass trapped by plants. Such knowledge may help maximize sediment trapping through appropriate species selection and provide useful information for slope erosion control.