Main content area

Fractal features of sandy soil particle-size distributions during the rangeland desertification process on the eastern Qinghai-Tibet Plateau

Feng, Xiao, Qu, Jianjun, Tan, Lihai, Fan, Qingbin, Niu, Qinghe
Journal of soils and sediments 2020 v.20 no.1 pp. 472-485
analysis of variance, clay, desertification, edaphic factors, fractal dimensions, models, oxidation, particle size, particle size distribution, rangelands, sand, sand fraction, sandy soils, silt, silt fraction, soil density, soil depth, soil organic carbon, soil sampling, soil water, total nitrogen, wind erosion, wind tunnels, China
PURPOSE: Fractal theory has been frequently applied to quantify soil particle-size distributions (PSDs) and evaluate soil degradation. Eolian desertification occurring on the eastern Qinghai-Tibet Plateau is threatening the ecological environment. It is essential to investigate the desertification process by adopting the concept of space as a substitute for time. Here, the fractal features of soil PSDs and their relationships with selected soil properties and wind erosion and deposition were studied. MATERIALS AND METHODS: The tested soil samples were collected from rangelands with four different degrees of desertification (light, medium, severe, and extremely severe). Soil particle sizes were measured by a laser diffraction particle-size analyzer. Soil moisture and soil bulk density were determined by the soil weight difference before and after oven-drying for 8 h. Soil organic carbon was determined by the K₂Cr₂O₇-H₂SO₄ oxidation method and soil total nitrogen was analyzed using the Kjeldahl digestion procedure. Wind tunnel experiments were conducted to simulate wind erosion and deposition during the desertification process. Values of the fractal dimension of soil PSDs were calculated using the volume-based fractal model. ANOVA with an LSD test and Pearson correlation analysis were used to analyze whether different parameters used in the fractal model influenced the results of the fractal dimensions and were reasonable to be applied. RESULTS AND DISCUSSION: Different parameters, including the arithmetic mean size and the upper sieve size of two successive sieve sizes, values of Rₘᵢₙ, and the number of data points used in the fractal model, rarely influenced the results of the fractal dimensions (p > 0.05). The D values showed a significantly positive correlation with the clay and silt contents and a strong negative correlation with the sand contents. Along the degrees of desertification from light to extremely severe, the contents of clay, silt, and very fine sand decreased while the fine sand contents increased, and the D values decreased accordingly. Rangelands with heavier desertification and lower D values were more easily exposed to wind erosion, relating to the further loss of soil organic carbon, soil total nitrogen, and soil moisture and increased soil bulk density. The soil depth could affect soil PSDs and the selected soil properties of rangelands under light and medium degrees of desertification but rarely influenced those of severely and extremely severely desertified lands. CONCLUSIONS: Different parameters, including the arithmetic mean size, the upper sieve size, values of Rₘᵢₙ, and the number of data points used in the fractal model, had no impacts on the results of the fractal dimensions, and were reasonable to be applied. We suggest the fractal dimension as an effective indicator to evaluate soil environment changes induced by desertification.