Multifractal characteristics and spatial variability of soil particle-size distribution in different land use patterns in a small catchment of the Three Gorges Reservoir Region, China
Tai-li Chen,Zhong-lin Shi,An-bang Wen,Dong-chun Yan,Jin Guo,Jia-cun Chen,Yuan Liu,Rui-yin Chen
DOI: https://doi.org/10.1007/s11629-020-6112-5
IF: 2.371
2021-01-01
Journal of Mountain Science
Abstract:<p class="a-plus-plus">Characterizing soil particle-size distribution is a key measure towards soil property. The purpose of this study was to evaluate the multifractal characteristics of soil particle-size distribution among different land-use from a purple soil catchment and to generalize the spatial variation trend of multifractal parameters across the catchment. A total of 84 soil samples were collected from four kinds of land use patterns (dry land, orchard, paddy, and forest) in an agricultural catchment in the Three Gorges Reservoir Region, China. The multifractal analysis method was applied to quantitatively characterize the soil particle size distribution. Six soil particle size distribution (PSD) multifractal parameters (<em class="a-plus-plus">D</em>(0), <em class="a-plus-plus">D</em>(1), <em class="a-plus-plus">D</em>(2), Δ<em class="a-plus-plus">α</em>(<em class="a-plus-plus">q</em>), Δ<em class="a-plus-plus">f</em>[<em class="a-plus-plus">α</em>(<em class="a-plus-plus">q</em>)], <em class="a-plus-plus">α</em>(0)) were computed. Additionally, a geostatistical analysis was employed to reveal the spatial differentiation and map the spatial distribution of these parameters. Evident multifractal characteristics were found. The trend of generalized dimension spectrum of four land use patterns was basically consistent with the range of 0.8 to 2.0. However, orchard showed the largest monotonic decline, while the forest demonstrated the smallest decrease. <em class="a-plus-plus">D</em>(0) of the four land use patterns were ranked as: dry land < orchard < forest < paddy, the order of <em class="a-plus-plus">D</em>(1) was: dry land < paddy < orchard < forest, <em class="a-plus-plus">D</em>(2) presented a rand-size relationship as dry land < forest < paddy < orchard. Furthermore, all land-use patterns presented as Δ<em class="a-plus-plus">f</em>[α(<em class="a-plus-plus">q</em>)] < 0. The rand-size relationship of <em class="a-plus-plus">α</em>(0) was same as <em class="a-plus-plus">D</em>(0). The best-fitting model for <em class="a-plus-plus">D</em>(0), <em class="a-plus-plus">D</em>(1), <em class="a-plus-plus">D</em>(2) and Δ<em class="a-plus-plus">f</em>[<em class="a-plus-plus">α</em>(<em class="a-plus-plus">q</em>)] was spherical model, for Δ<em class="a-plus-plus">α</em>(<em class="a-plus-plus">q</em>) was gaussian model, and for <em class="a-plus-plus">α</em>(0) was exponential model with structure variance ratio was 1.03%, 49.83%, 0.84%, 1.48%, 22.20% and 10.60%, respectively. The results showed that soil particles of each land use pattern were distributed unevenly. The multifractal parameters under different land use have significant differences, except for Δ<em class="a-plus-plus">α</em>(<em class="a-plus-plus">q</em>). Differences in the composition of soil particles lead to differences in the multifractal properties even though they belong to the same soil texture. Farming behavior may refine particles and enhance the heterogeneity of soil particle distribution. Our results provide an effective reference for quantifying the impact of human activities on soil system in the Three Gorges Reservoir region.</p>
environmental sciences