水保所知识产出(1956---)知识库

黄土丘陵区土壤抗侵蚀能力主要依赖于土壤抗冲性能，植被是保持土壤最有
效的方法。针对植物根系如何通过物理固结效应和生物化学效应来提高土壤抗冲
性能这一科学问题，本论文以黄绵土和沙黄土为对象，设置了含根土壤、无根土
壤及模拟根系冲刷试验,引入电容桥方法，定量分析了根系网络串连、根土粘结
和生物化学作用在强化土壤抗冲性能的相对重要性以及在两种土壤类型上的差
异，旨在深化根系内在固土机理研究。同时，分析了植物在生长季、季节性冻融
过程以及不同退耕模式下根系的固土效应，为正确认识植物根系固土机理和固土
效应及植被保土措施提供科学依据。主要结论如下：
1、揭示了沙黄土植物根系固土机理。发现与黄绵土相似，沙黄土中依然存
在植物根系网络串连、根土粘结和生物化学作用。沙黄土根系强化土壤抗冲性中
的物理固结效应贡献值为 66.9–73.7%，平均 70.9%，与黄绵土相比，平均小了
9.1%。随着根系密度的增加，黄绵土和沙黄土中根系物理固结效应在总效应中的
比例分别增大 4.2%和 6.8%。在物理固结效应中，沙黄土根系网络串连作用较为
关键，而黄绵土根土粘结作用越来越重要。植物根系物理固结效应与根系表面积
密 度 在 极 显 著 水 平 （ P < 0.01 ） 上 呈 指 数 递 增 函 数 关 系 ， 即
2
72.87(1 exp( 0.026 )), 0.89**     y x R 和
2
90.77(1 exp( 0.036 )), 0.80**     y x R
因此 ， 根表面积密度可以较好地反映土壤抗冲性变化。
2、植物根系提高土壤抗冲性表现出显著的季节性特征。在生长季，植物根
系能够显著提高土壤结构稳定性和强化土壤抗冲能力。与对照（农地）相比，处
理T（直根系）、F（须根系）、T + F（直根系 + 须根系）和N（农地撂荒）土
壤团聚体分别增加了32.6%、48.6%、64.6%和97.9%。土壤抗剪强度显著增加。
种植草本平均减少泥沙流失量26.5%–57.2%。与种植9周相比，种植21周后，土
壤崩解速率下降了39.0%–58.1%，而植物根系密度和根表面积密度分别增加了
64.0%–104.7%和75.9%–157.1%。处理T + F和N在抗侵蚀性能方面较处理T或F更
强，但处理T + F和N或T和F无统计学显著差异。植物根系在季节性冻融过程中固土效应显著。季节性冻融主要是通过增大土
壤崩解速率，降低土壤抗冲能力，进而增加冲刷泥沙流失量。与冻融前相比，表
层土壤容重、团聚体含量和根系密度在3个处理中均无显著变化，土壤黏聚力略
有下降，而崩解速率在农地对照（CK）、黑麦草传统密度（LD）和加倍密度（HD）
中分别增加了20.6%、18.8%和7.3%。冻融作用延后了3个处理的主要产沙时间，
降低了土壤抗冲能力。与冻融前相比，冻融作用分别增加了处理CK和LD泥沙流
失量19.41%和6.70%，但对处理HD影响较小。植物根系在冻融过程中固土效应显
著，在处理LD和HD上分别减少土壤流失量3.72%和49.39%。
3、黄土丘陵区典型退耕模式（农地撂荒、灌木林、乔木林植被建设）显著
增加了植物根系密度和根土接触面积，增强根系固土效应。与农地相比，撂荒土
壤表层（0–15 cm）和中层（>15–30 cm）容重显著降低，而下层（>30–50 cm）
土壤容重变化较小。土壤团聚体和抗剪强度（黏聚力C和内摩擦角φ）在3个土层
中均呈增加趋势，而土壤崩解速率在各土层均呈下降趋势。随着撂荒年限的增加，
表层土壤抗冲性呈先增加后稳定的趋势，而中层和下层土壤抗冲性呈稳定增加的
趋势，与对照相比，分别增大76.9%和30.7%。根长密度、根表面积密度显著增
加。撂荒地各土层土壤抗冲性与土壤物理性质及根表面积密度的关系均可用线性
方程来表达。土壤团聚体和根表面积密度在3个土层中均是影响土壤抗冲性的关
键指标。
人工灌木林（柠条为例）不同生长阶段，表层土壤（0–20 cm）容重略有减
小，中层（20–40 cm）和下层（40–60 cm）土壤容重无显著变化。土壤崩解速率
在前3个生长阶段内显著下降，而后趋于稳定，而土壤团聚体、抗剪强度和根系
密度在前3个生长阶段内显著增加，而后趋于稳定。土壤抗冲系数在各土层分别
增大9.3、4.1和4.2倍。负指数函数方程能够较好地拟合柠条地土壤流失特征，且
80%的泥沙流失发生在冲刷的前3分钟。针对不同生长阶段，表层土壤抗冲系数
呈先增加，而后趋于稳定的趋势，而中层和下层土壤抗冲系数显著增加。线性方
程能够较好地表达柠条地各土层土壤抗冲性与土壤物理性质及根系密度的关系。
土壤团聚体和根系密度在各土层中均是影响土壤抗冲性的关键指标。
与对照相比，人工乔木林（刺槐为例）地土壤容重在表层（0–20 cm）、中
层（20–40 cm）和下层（40–60 cm）土壤分别平均下降了14.5%、5.7%和3.3%。
土壤团聚体和抗剪强度在各土层上显著增加，而土壤崩解速率显著下降。土壤团
聚体和根系密度对土壤抗冲性的贡献值分别为71.0%和90.8%，是强化土壤抗冲性的关键指标。
4、首次引入电容桥方法，探索了植物根系固土效应的非直接性估测。在沙
黄土和黄绵土中，含根土壤累积泥沙流失量与电阻抗数值可以用线性方程较好地
表达。产流的前3分钟电阻抗均值能够较好地反映根系固土效应大小，与泥沙收
集法相比，电容桥法得到的根系物理固结效应数值偏低，在黄绵土和沙黄土上平
均低了13.5%和14.1%。同时，在泥沙收集方法与电容桥法两组结果中，根系网
络串联、根土粘结和生物化学作用在黄绵土上呈显著线性相关关系，而沙黄土上
两种方法计算出的结果也存在线性相关关系，但未达到统计学显著水平。因此，
电容桥方法在一定程度上能够非直接性估测植物根系固土效应。
关键词：植被根系；土壤抗冲性；电阻抗；黄土丘陵区

Soil erodibility primarily relies on soil anti-scouribility in the concentrated flow
erosion zones of the Loess Plateau. Vegetation is the most effective way in soil
conservation. To clarify the scientific problem of how plant roots improve soil
anti-scouribility through root physical consolidation effects and root biochemical
effects, this paper selected soil samples with roots, pure soil as well as designed roots
for soil anti-scouribility experiment in Sandy Loess and Loess soil. The primary goal
of this study was to evaluate the relative contribution of physical consolidation of soil,
including net-link and soil-root bond functions, and root biochemistry effect, to soil
anti-scouribility. At the same time, this paper analyzed soil anti-scouriblity under the
condition of different patterns of roots (tap or fibrous root) in growing period, and
seasonal freeze-thaw process as well as various abandoned modes. This kind of study
could strengthen the mechanism in root-penetrated soil, and could provide theoretical
basis for the contribution of root to soil anti-scouribility. Main conclusions were
summarized as follows:
(1) This paper reveals the mechanism of root reinforcement in Sandy Loess. Effect
of root physical consolidation is the main form in strengthening soil anti-scouribility,
accounting for 77.7–82.0% and 66.9–73.7%, respectively, in total root effect on soil
anti-scouribility in Loess soil and Sandy soil. As the root density increases, the
physical effects of root consolidation increased. In the root physical consolidation
effect, the network tandem showed an increasing trend with the increasing role of root
density in the Sandy loess, while soil-root bond effect becomes more important in
Loess soil as root density increased. Functions
2
72.87(1 exp( 0.026 )), 0.89**     y x R
and
2
90.77(1 exp( 0.036 )), 0.80**     y x R could fit well the relationship between root
surface area density and root physical consolidation. Therefore, the root surface area
density could effectively reflect the effect of root physical consolidation, thus, could   forecast the changes in soil anti-scouribility.
(2) Root reinforcement shows a significant seasonal characteristic in improving
soil anti-scouribility. During the growing season, grass planting slightly reduced soil
bulk density, while increased soil aggregate content by 32.6%, 48.6%, 64.6%, and
97.9% in the treatments T, F, T + F, and N, respectively. Soil shear strength, including
cohesion (C) and angle of internal friction (φ) significantly increased after grass was
planted. As roots grew, soil C increased by 65.2%–135.5%, whereas soil
disintegration rate decreased by 39.0%–58.1% during the 21st week from the recorded
value during the ninth week. Meanwhile, root density and root surface area density
increased by 64.0%–104.7% and 75.9%–157.1%, respectively. No significant
differences in the soil anti-scouribility were observed between the T and F treatments
or the T + F and N treatments, but T + F and N treatments performed more effectively
than T or F treatment alone in retarding concentrated flow. Compared with treatments
of T, F, T + F, the natural vegetation restoration might be the most appropriate soil
reinforcement.
Root reinforecement is significant in freeze-thaw process. No significant changes
were found in soil bulk density, water-stable aggregate content, and root density after
a cycle of freeze-thaw compared with those before freeze-thaw. Comparatively, soil
cohesion decreased slightly, whereas soil disintegration rate increased by 20.6%,
18.8% and 7.3% in treatments CK, LD and HD as compared with those before
freeze-thaw. In addition, freeze-thaw delayed the occurrence of main sediment
production, reducing soil anti-scouribility, as well as increasing the rate of sediment
loss in the middle scouring time and total sediment. Among the treatments, compared
with those before freeze-thaw, the freeze-thaw increased sediment by 19.41% and
6.7% in treatments CK and LD, while little effect on HD. Combined effect of root and
freeze-thaw in sediment reduction were 3.72% and 49.39% in LD and HD treatments,
respectively.
(2) Typical abandoning modes, including fallow land, shrub land and tree land,
could effectively increase root biomass and the contacting area of root and soil,
enhancing soil anti-scouribility. As the year of abandoned land increased, compared
with control (crop land), soil bulk density in the surface layer (0–15 cm) and middle
layer (> 15–30 cm) were significantly reduced, while little change occurred in the lower
soil layer (> 30–50 cm). The soil water–stable aggregate content and shear strength,
including C and φ were also significantly increased in the three studied soil layers. Soil  disintegration rate reduced in all soil layers, especially for the middle and lower soil
layers, about 4.2 and 1.8 times than those in the surface soil layer. Soil anti-scouribility
in the surface layer increased rapidly before the stage Ⅲ, and kept stable in the
following abandoned stages, while the soil anti-scouribility in the middle and lower soil
layers were increased steadily, approximate 76.9% and 30.7% increments as compared
with those of control. Soil water–stable aggregate content and dry root biomass were
the determining factors in the reinforcement of soil AS in the abandoned land of the
hilly Loess Plateau.
Compared with the control, Caragana planting could effectively reduce soil loss.
Sediment loss over time was well described by a negative exponential function. On
average, about 80% of the soil sediment was lost within the first 3 abrasion minutes.
Compared with control, soil bulk density of surface (0–20 cm) and middle soil layers
(20–40 cm) decreased by 8.9% and 18.0%, respectively, but minimal changes
occurred in the lower soils (40–60 cm). Soil aggregate content and shear strength
increased significantly, whereas soil disintegration rate decreased significantly, with a
maximal reduction of 357.1% in the middle soil layer. Soil erosion resistance
increased 9.3, 4.1 and 4.2 fold in the surface, middle and lower soils, respectively.
Linear regression equations could well fit the relationship between soil erosion
resistance and soil physical properties, with root biomass. Soil aggregate content and
root density were the key factors in the reinforcement of soil erosion resistance for
Caragana plantations on the Loess hilly Region.
Compared with CK, artificial Robinia planting significantly reduced sediment.
Changes in the sediment over scouring time were best described by a negatively
exponential function. Compared with CK, the averaged soil bulk density beneath
Robinia significantly decreased by 14.5% in the surface (0–20 cm) soil layer and
non-significantly by 5.7 and 3.3% in the middle (20–40 cm) and lower (40–60 cm)
soil layers, respectively. Soil aggregate content and shear strength increased while soil
disintegration rate decreased significantly in the three soil layers with Robinia stages.
Mean 6.8, 1.6 and 0.2 times were increased in soil AS. Linear regression equations
between soil anti-scouribility and the soil structural properties were well fitted in the
surface and middle soil layers. Soil aggregate content and root biomass were key
factors, which contributed 71.0 and 90.8% to the reinforcement of soil
anti-scouribility beneath Robinia in the Loess hilly Plateau.
(4) Electronic capacitance method was firstly introduced to estimate the effect of  plant roots on soil reinforcement non-directly. In Sandy Loess and Loess soil, the
relationship between accumulated sediment loss and impedance values can be well
expressed by linear equations. The averaged impedance values within 3 min scouring
time could well reflect root reinforcement. The values of root physical consolidation
effect obtained by the Electronic capacitance method are 13.5% and 14.1% lower than
those through the collected sediment metod. Additionaly, the values obtained by
Electronic capacitance method and collected sediment metod show a significant
relationshp in Loess soil, while non-significant correlation in Sandy Loess even
though a linear correlation existed. Therefore, the Electronic capacitance method, to
some extent, is able to estimate root reinforcement non-directly.
Key words: Vegetation root; Soil anti-scouribility; Electric impedance; Loess Hilly
Region