Return Pre Home Next




CAO Wenhong, GUO Qingchao and ZENG Qinghua

Institute of Water Resources and Hydropower Research, Beijing 100044, China


In China, the mountainous and hilly area takes two-thirds of the total land with a clear continental monsoon climate and large variation in the interannual rainfall. Poor vegetation and serious land erosion make large quantity of sediment entering rivers. Land erosion, sediment transport, and sedimentation have made a series of problems to hydraulic projects and eco-environment. This compels Chinese scientists and engineers to work hard, and thus fruitful achievements have been accumulated. A brief review of the research progress on land erosion is presented as follows.


3.1.1  Relationship between Land Erosion and Natural Factors

Natural factors affecting land erosion mainly include rainfall, topography, vegetation, and soil constitution. The affecting degree of rainfall on the land erosion is mainly determined by its characteristics, such as its amount, intensity, duration, type, and previous rainfall. As an integration, Wang and Jiao[1] analyzed the relationship among rainfall-land erosion-sediment transport for different spacial scale based on the idea of systematic science. The study shows that the order of affecting degree of rainfall characteristics on land erosion in loess region is intensity, energy, and amount, especially the maximum period intensity. Therefore a new concept, erodible rainfall, is put forward, which is the rainfall amount excess the minimum rainfall being able to produce erosion.

Land slope is the closest factor to affect soil erosion. In the study of soil erosion on the slope, researchers find that land erosion amount increases with the slope, and an exponential relationship between the quantity of land erosion and the slope is established. Further investigation, however, finds that the quantity of land erosion decreases with the increase of slope when the slope is larger than a certain value. Thus there should be a critical slope in reality. To find this critical slope, Horton[2] made a lot analyses in 1940s. In China, Cao[3] theoretically investigated this critical slope in 1990s. In addition, many other researchers[46] gave a lot of concerns on this topic from different view points, such as kinematic wave theory, energy equation, and sediment mechanics. Research results show that the critical slope of land erosion varies with soil size composition, soil type, rainfall factors, and vegetation. The critical slope ranges from 21to 50.

Vegetation plays an important role in alleviating land erosion. Larges quantity researches demonstrated land erosion could be efficiently controlled when the cover degree of vegetation reaches a certain level. Therefore another concept, efficient cover ratio of vegetation, is put forward. It can be described as such cover ratio in which the quantity of land erosion is less than the maximum allowable land erosion in the area covered with trees and grasses. Based on the measured data in the Loessial Plateau, Jiao[7] et al. have analyzed the efficient cover ratio of tree-grass covered land under different rainfall and slopes. The results indicate that, under a comparative stable soil composition and vegetation type, the efficient cover ratio increases with rainfall and slopes. The effect of rainfall and slopes become weak when the cover ratio reaches a certain degree. In the same other conditions, the efficient cover ratio in grass-covered land is bigger than that in tree-covered land for the same soil conservation effect.

3.1.2  Mechanism of Land Erosion Dynamics

Land erosion and sediment yield under hydraulic action is a very complicated process, which includes such steps as the separation of sediment from the soil, transport, and deposition resulted from raindrop splash and erosion by overland runoff.  Raindrop splash

The falling rain drop has certain kinetic energy and therefore results in land erosion of top soil. The splash action of a rain drop has a close relation with its physical characteristics, such as size, shape, velocity hitting land surface, kinetic energy, and eroding force.

Yao[8] gives the following formulas to express the terminal velocity of a rain drop after a comprehensive examination on the falling velocity of a rain drop.

d503 mm

  3 mmd506 mm

where,  is the terminal velocity of a rain drop;  is the median diameter; is the kinematic viscosity coefficient of gas; and is the gravitational acceleration.

Through experiments of raindrop splash to the loess, Zhao[9] investigated the effect of raindrop energy, land slope, soil moisture, and soil size on the incipient time and quantity of soil erosion. The results show that the splash erosion quantity, Ds, is related not only with the raindrop size and kinetic energy, e, but also with soil moisture, q, and median diameter d50. Ds can be expressed as: . The content of medium and coarse particles (>0.1mm) in the splashed sediments is much larger than that in the original soil. The contents of coarse and fine (<0.01mm) particles in the splashed sediment decay quickly with the increase of splash distance, whereas the content of medium size particles (0.010.1mm) increases.  Erosion by slope flow

The processes of erosion by slope flow include the detachment of particles from soil, sediment transport, and deposition, which resulted from raindrop splash and surface runoff. In given soil condition, the eroding- and carrying-capacity of slope flow is determined by its hydraulic characteristics.

Experimental results by Yao[10] et al. show that the slope flow has three types: pseudo-layer flow, transition flow, and turbulence during rainfall. For the pseudo-layer flow, although raindrops make some fluctuation and local mixture to the slope flow, the relation between the drag resistance and Reynolds number still obeys the law of laminar flow in open channels. This phenomenon indicates that viscous force is stronger than the inertia force and the slope flow generally displays as laminar flow.

Based on the experiments of runoff erosion, Li[11] et al. discussed the relation between the soil erosion rate of slope (Dr) and the energy dissipation of runoff (E) by using the principle of energy conservation. Results show the soil erosion could occur only when the energy dissipation of runoff reaches a certain value, which is called as the critical energy dissipation. The critical value is larger than 7.387 Joule.

The sediment-carrying capacity of slope flow is one of control parameters to describe the particle detachment from soil and transport by flow. Generally, the formula to calculate sediment-carrying capacity in rivers is not appropriate for the case of slope flows if the flow depth is very limit. Researches by Abrahams[12], Gary[13], indicate that flow power can reasonably express the sediment-carrying capacity of slope flows. Based on the measured data in the Xindiangou region of Loessial Plateau, Cao[14] established the following expression to describe the carrying capacity of slope flows by using the theory of flow power.


where and are coefficients; I is the rainfall intensity;  is the specific gravity; q is the discharge per unit width; and J is the slope.  Gully erosion

Sediment yield by gully erosion is a process of concentrated energy release in a basin system. Such energy includes the inertia energy and potential energy coming from the runoff.

The mechanism of gully erosion remains unclear because most researches still stay in the stage of qualitative description and empirical relation. The processes of gully erosion generally include: (a) widening of channel due to the collapse caused by incaving bank; (b) deepening of channel bed because of the undercutting; (c) going backwards of the gully head due to the headward erosion, this may cause the mergence of channels under certain condition; and (d) moving the eroded sediment.

3.1.3  Land Erosion and Sediment Yield Models

Land erosion and sediment yield model is a useful method to quantitatively calculate sediment production in basins, also is an efficient tool to guide the allocation of water-soil conservation measures and the utilization of water-soil resources. Since 1950s, Chinese scientists and engineers have developed some land erosion and sediment yield models. A brief introduction to them is given as follows.

Considering the morphologic features in the hilly region of Loessial Plateau, Jiang[15] established an empirical model to describe the sediment production during a rainfall event in the gully area.

where WS is the rainfall erosion amount; a is a coefficient; K is a soil factor coefficient; P is the amount of rainfall; I30 is the maximum 30-minute rainfall intensity; S is the slope; L is the length; GS is the influential coefficient of shallow gully erosion, GS = 1, if there is no shallow gully erosion; V is the influential coefficient of vegetation; and C is the influential coefficient of water-soil conservation measures. V and C are given by 1.0 for bare surface.

Tang[16] et al. established soil erosion dynamics model for small basins based on the principle of sediment yield, transport, and deposition. The model includes two parts, runoff model and sediment transport model. The runoff model can simulate runoff yield and confluence, whereas the sediment model is able to calculate both sediment yield and transport. In the runoff model, the mode of excess flow yield is used to calculate flow production, while the kinetic wave equations are adopted to describe the flow converge of individual unit in the slope and gully. In the sediment model, the amounts of sheet erosion (including rill erosion) and gully erosion (including gravity erosion) on slopes can be calculated by the formula of soil erosion rate. Finally the sediment production at the outlet cross section in a basin can be obtained by summing up the amount of sediment yield in individual unit.

Distributed hydrologic models take an important position in modern hydrology study. However, their application to soil erosion is still at the very beginning stage. Resorting to the idea of distributed model, Wang and Cao[17] divided the basin into many individual units within which the sediment yield is calculated. Then integrating the simulated result in each unit results in the total sediment production for the whole basin. From physical concepts, Tang[18] uses the distributed model to simulate both flow yield/converge and sediment yield/transport, i.e., the 2-d dynamic wave model to calculate the runoff on slopes, and 2-d continuous equation to predict sediment yield.


Effect of human being activities on land erosion can be displayed mainly through changing the underlying surface, such as the change of slope and length due to the artificial terrace, the change of base level of erosion because of the artificial silt arrester, the change of vegetation cover, etc. These reasons could change the processes of runoff formation and converge, flow energy, erosion resistance of surface materials, etc. and thereby result in a big change in the intensity of land erosion[19].

3.2.1  Human Being Activity Accelerating Soil Erosion

Northwest Water & Soil Conservation Institute, Chinese Academy of Sciences, established standard regions and large experimental fields of runoff in Ziwulin of the Loessial Plateau and studied the effect of human being activities on soil erosion through destroying vegetation and cultivation since 1989. The characteristics of soil erosion for different land use, morphology, tree-covered land, and cultivation are investigated by taking the tree-grass region as natural soil erosion, the other ten regions denoting different human being activities[20].

Through the positioning experiments in Ziwulin tree-covered land and field observation, Zha[21] studied the effect of vegetation destroy on the physical and chemical features and the erosion processes of soil. After the cultivation of tree-covered land, the erosion-resisting ability of the soil significantly decreases; meanwhile the rill erosion and shallow gully erosion rapidly develop. Investigation indicates that, after ten-years cultivation in tree-covered land, the organic matter in the cultivation layer drops by 84.3%, disintegration rate and erosion amount are 21.5 and 25 times of those in tree-covered land, respectively.

Located in the middle reach of the Yellow River, Beiluohe Basin with an area of 26900 km2 is the main sediment resource of the Yellow River. The main human being activities, which greatly affect the sediment amount entering Yellow River, include the reclamation by destroying trees, reclamation on slopes, road construction, and mining, etc. Based on the investigation and analyses on related data, Liu[22] estimated the effect of water-sediment characteristics and human being activities in Beiluohe Basin on the treatment of water-soil loss and concluded that the amount of water-soil loss increased by 1.768 million tons during the period of 1990 1996.

3.2.2  Human Being Activity Alleviating Soil Erosion

Since 1950s, the conservation engineering has been conducted. After the 50-year constant effort, so far the total improved area of water-soil loss is about 154000 km2, which takes one-third of total area of water-soil loss in the Loessial Plateau. Among the improved area, the forest of conservation of water and soil is about 8.0106 ha; land covered by planting grass is around 2.3106 ha; sediment training is 3.0105 ha, as well as 850 key engineering projects for improving gullies, 100000 silt arresters, 4 million small engineering projects constructed on gullies to store water and sediment. These measures of water and soil conservation have played a significant effect and made the sediment entering the Yellow River decreased by 300 million tons every year.

Currently, China is conducting great exploitation of its Western Regions and eco-environmental construction. Biological measures are given much more concerns. Bi[23] et al. suggested a comprehensive treatment mode for sustainable development of rill construction, i.e., taking soft dam made of plants as main body of sand-bar engineering, rill-dam land, man-made bottomland, man-made wetland as main compositive part of rill-based agricultural land, combined with backbone engineering, mini-reservoir, to form that small rill intercepts coarse sand, man-made bottomland and rill-dam land store fine sediments, the area between dams form man-made wetland, large scale rills and tree/grass-covered slope accept storms, hydraulic energy can be dissipated and water-sediment can be separated along gullies, man-made bottomland, rill-dam land, and man-made wetland could increase natural filtration, mini-reservoirs can store residual runoff. Through these measurements, the objectives of sustainable development in separating coarse and fine sediments, maintaining water-sediment and ecological balance can be reached.

Due to the active erosion, the Loessial Plateau is cut to innumerable fragments and gullies. Large quantity of sediments in gullies entering the Yellow River through tributaries or directly during storms results in that the sediment concentration in the Yellow River could significantly increase. The serious loss of top soils as well as arid climate causes a very low agricultural produce in this area. Therefore, efficient control of erosion and advancing the development of agricultural production become the two major issues in that region. Practice has proven that silt arrester is an efficient approach to solve problems. This measurement plays an important role in reducing sediment entering the Yellow River, making silt land, and more importantly, the moisture in silt lands guarantees agriculture to get a stable high production. Fang[24] concludes the effect of reducing erosion by silt arrester as follows: (1) raising local erosion base level to decrease gravity erosion and control gully erosion; (2) intercepting flood sediment to alleviate dully erosion. At the initial stage, silt arrester can store flood and sediment, meanwhile curtail flood peak and mitigate erosion downstream; (3) damping surface runoff to increase surface silt. At the late stage, silt arrester can make silt land, change the condition of flow yield and convergence, and therefore reach the purpose to alleviate flood sediment; (4) increasing silt land, advancing unit agricultural production, promoting the steps of returning agriculture on slopes to plants and grassland, and decreasing slope erosion.




Developing a reasonable and usable approach to accurately assess the efficiency of erosion control by water-soil conservation is very important. It can guide comprehensive treatments of basins.

The traditional methods to calculate the effect of water-soil conservation include two types, hydrology and conservation. The hydrological method is firstly to establish the quantitative relationships between rainfall-runoff, rainfall-sediment yield, and runoff-sediment transport through the statistic analyses of field data. The runoff and sediment yields for natural and improved basin can then be calculated respectively. Therefore, the effect of reducing runoff and sediment by water and soil conservation can be obtained. The major merits of hydrological method are simple, easily usable, and a good prediction for the same basin can be expected. However, the reliability of the method greatly depends on the measured flow and sediment data. According to the fact that the conditions of flow and sediment yield are changed due to the change of underlying surface made by human being activities, the conservation method is developed based on the quantity and quality of engineering measurements, indexes of storing flow and sediment, and other more factors, to respectively calculate the reduced runoff and sediment by each measurement. Thus, the comprehensive effect of water and soil conservation can be obtained by summing up individual effect. The reliability of conservation method depends significantly on the accuracy of the quantity, quality and distribution of water and soil conservation measurements and indexes of intercepting runoff and sediment. For a long time, the accuracy of conservation method has been affected by man-investigated and subjective data.

Recently, some researchers are trying to develop a physical conceptual model, which is appropriate to simulate long series of runoff and sediment yield in big or middle size basins, to calculate the effect of reducing runoff and sediment by water and soil conservation in a comparatively big basin. The runoff and sediment yield model for large and middle size basins developed by Tang[25] et al. demonstrated a good simulation in the Dalihe Basin (3906 km2) in the middle region of Yellow River.

With the rapid development of remote sensing and computer technologies, the two technologies have been widely applied to the monitoring of soil erosion and assessment of water and soil conservation. It was the first time in 1980s that the remote sensing technology was applied to investigate national wide land erosion, and the land erosion by hydraulic and wind force was clearly depicted. In 1999 to 2001, the second investigation for national wide land erosion by remote sensing was conducted[26]. Digitalization processes were used by such hi techs as TM information sources, GIS based software, man-computer inter depicting, direct generating and statistic of area of map spots. Working map is 1:100000, the correction rate of reading soil erosion intensity is greater than 90%, and the map spots for quality checking are no less than 5% of the total spots. The intensity of soil erosion is judged according to Standard of Types and Classification of Soil Erosion', and the relative influencing factors, such as vegetation, type of land use, and slope, etc., are also determined. Established TM imagebase, digital map of soil erosion, photobase of typical land for national and provincial wide provide a powerful tool to dynamic monitoring and trend analysis for soil erosion.

Through analyses and management of large quantity spacial data by using GIS, North-West Water & Soil Conservation Institute, China Academy of Sciences, initially established the database for the assessment of water and soil loss in the Loessial Plateau region[27]. The database includes seven types of data: five major factors data, sediment data, and accessorial data. This database settles a good base to do macro- and system assessment of water and soil loss in basins similar to the Loessial Plateau.


[1]      Abrahams D., Gary Li (1998), Effect of saltating sediment on flow resistance and bed roughness in overland flow, Earth Surface Processes and Landforms, 23953-960.

[2]      Bi Cifen (2001), Discussion on Treatment Countermeasure in Bedrock Area on Loess Plateau, Journal of Sediment Research, 41-6 (in Chinese).

[3]      Cao Wenhong (1995), Preliminary study on sediment-carrying capacity of slope flow, Proceedings of the 2nd National Conference of Sediment Fundamentals, China Construction and Material Press House, 227-232 (in Chinese).

[4]      Cao Wenhong (1993), Slope thresholds of soil erosion, Bulletin of Water and Soil Conservation, 13(4)1-5 (in Chinese).

[5]      Ding Wenfeng, Li Zhanbin, and Cui Lingzhou (2001), Experimental study on runoff scouring erosion on Loess slope surface, Journal of Soil and Water Conservation, 15299-101 (in Chinese).

[6]    Fang Xuemin, Wan Zhaohui, and Kuang Shangfu (1998), Mechanism and effect of sand traping by silt arrester in middle Yellow River, Journal of Hydraulic Engineering, No. 10 (in Chinese).

[7]    Gary Li, Abrahams D (1999), Controls of sediment transport capacity in laminar interrill flow on stonecovered surfaces, Water Resources Research, 35(1)305-310.

[8]      HortonR.E.(1945) Erosional development of streams and their drainage basins, Hydrophysical Approach to Quantitative Morphology, Geol. Soc. Amer Ball, 194556(3).

[9]      Hu Liangjun, Zhang Xiaoping et al (2002), Database construction for water & soil loss assessment in Loess Plateau, Journal of Hydraulic Engineering, 181-85 (in Chinese).

[10]     Hu Shixiong, and Jin Changxing (1999), Theoretical analysis and experimental study on the critical slope of erosion, Acta Geographica Sinica,544347-356 (in Chinese).

[11]     Jiang Zhongshan, Wang Zhiqiang, and Liu Zhi (1996), Quantitative study on spacial distribution of soil erosion of small basins in loess hilly land. Journal of Soil Erosion and Water & Soil Conservation, 2(1), 1-10 (in Chinese).

[12]   Jiao JuyingWang Wanzhong, and Li Jing (2000). Effective cover rate of woodland and grassland for soil and water conservation in Loessial Plateau. Acta Phytoecologica Sinica245608-612 (in Chinese).

[13]   Liu Bin, Ran Dachuan et al. (2001), Influence of human activities on control of sSoil and water Loss in Beiluo River Basin, Yellow River, 23215-17 (in Chinese).

[14]   Liu Qingquan, Chen Li and Li Jiachun (2001). Influences of slope gradient on soil erosion, Applied Mathmatics and Mechanics, 225449-457 (in Chinese).

[15]   Tang Keli (1993), Variation of Erosion and Runoff in Yellow River Basin, China Science and Technology Press, Beijing (in Chinese).

[16]   Tang Lihua, and Zhang Sicong (2002), Preliminary study on distributed model of runoff and sediment yield in small watersheds, Journal of Hydroelectric Engineering, 1119-127 (in Chinese).

[17]   Tang Liqun, and Chen Guoxiang (1997), Dynamics model for runoff and sediment yield in small basins, Journal of Hydraulic Dynamics, No. 2, (in Chinese).

[18]   Tang Liqun, and Chen Guoxiang (1997), Modeling of processes of long series runoff and sediment in large and middle size basins, Journal of Hydraulic Engineering, No.6, 19-26 (in Chinese).

[19]   Wang Wanzhong, and Jiao Juying (1996), Sediment Yield by Rainfall Erosion on Loessial Plateau and Sediment Transport in Yellow River, Science Press, Beijing (in Chinese).

[20]   Wang Xiuying, Cao Wenhong et al. (2001), Study on distributed mathematical model for simulating watershed runoff, Journal of Soil and Water Conservation, 15338-41 (in Chinese).

[21]   Xu Jongxin (2001), Study progress of influence of human being activity on erosion, transport, and sedimentation, Appeared in Study Progress of Evolution Processes of Water Resources and Its Recoverable Mechanism, Yellow River Water Resources Press House, Zhenzhou, 186-198 (in Chinese).

[22]   Yao Wenyi and Chen Guoxiang (1993), Formulas of falling and terminal velocity of raindrops, Journal of Hohai University, 21(3) (in Chinese).

[23]   Yao Wenyi (1996), Experiment study on hydraulic resistance laws of overland sheet flow, Journal of Sediment Research, 374-81 (in Chinese).

[24]   Zeng Dalin, and Li Zhiguang (2000), Thoughts on the second national wide remote sense investigation on soil erosion, Journal of China Water and Soil Conservation, No.1, , 28-31 (in Chinese).

[25]   Zha Xuan, and Huang Shaoyan (2001), Effects of vegetation destruction on erosion and soil degradation processes on Loess Plateau, Journal of Mountain Science, 192109-114 (in Chinese).

[26]   Zhao Xiaoguang, and Wu Faqi et al (1999), The critical slope of soil erosion, Research of Soil and Water Conservation, 6242-46 (in Chinese).

Zhao Xiaoguang, and Wu Faqi (2001), Single raindrop Splash law and its selection role on soil particles splashed, Journal of Soil and Water Conservation, 15143-45 (in Chinese).

Return Pre Home Next