Plot size modulates the effects of vegetation on soil and water conservation at the slope scale on China's Loess Plateau

Zhang, Qin , Xu, Hang , Zhang, Zhiqiang

2025-10-01 CATENA 2025   258(卷), null(期), (null页)

查看原文

  • JCR分区:

    影响因子:

  • Long-term vegetation restoration on China's Loess Plateau has provided multiple ecological services and significantly improved environmental conditions. Many slope scale studies assess the role of vegetation in soil and water conservation but seldom account for plot size, leaving its effects on runoff and sediment yield uncertain. We compiled 533 runoff records from published literature covering nine experimental sites on the Loess Plateau and analyzed them to identify the key factors governing runoff and sediment yield at the slope scale using machine learning methods. Our results demonstrate significant differences in the effectiveness of soil and water conservation between vegetation types (p < 0.05), with forests exhibiting a superior capacity to reduce runoff and sediment. Once vegetation coverage exceeds 60 %, further increases in vegetation cover no longer effectively control sediment yield in forested plots. We found that the area of forest plots has a more significant control effect on their sediment yield, while the sediment yield of grassland plots is more sensitive to changes in vegetation cover. When the plot area surpasses approximately 280 m(2), its influence on sediment yield and sediment concentration tends to stabilize, the contribution of plot area to runoff and sediment yield exhibited significant differences before and after reaching the threshold plot area (p < 0.05). Additionally, the aspect ratio also influences erosion measurements, with narrower plots enhancing runoff scouring and resulting in fluctuations in the observation results. Our findings underscore the critical importance of plot size when evaluating the role of vegetation in soil and water conservation. We recommend the adoption of standardized runoff plots for such studies, emphasizing that the influence of plot size should be explicitly considered when performing cross-plot comparisons.