The effects of slope shape and polyacrylamide application on runoff,erosion and nutrient loss from hillslopes under simulated rainfall

Natural hillslopes are mostly composed of complex slope shapes, which significantly affect soil erosion. However, existing studies have mainly focused on uniform slopes to simplify complex hillslopes, and the mechanisms responsible for the influence of slope shape on soil and nutrient losses are still not well understood, especially in the application of soil improvers to reduce soil loss. To investigate the effects of slope shape and polyacrylamide (PAM) application on runoff, soil erosion and nutrient loss, this study conducted artificial field rainfall experiments involving two PAM application rates and nine slope shapes. The results indicate that the average amount of soil loss from convex slopes was 1.5 and 1.3 times greater than that from concave and uniform slopes, respectively, and the average amount of ammonia nitrogen loss and phosphate loss increased by 24.0%–58.6%. Soil and nutrient losses increased as the convexity of the convex slopes increased. For runoff, there was little difference between concave and convex slopes, but the runoff amount for both slopes was greater than that for uniform slopes. After PAM application, the soil loss decreased by more than 90%, and the nutrient loss decreased by 28.2%–68.1%. The application of PAM was most effective in reducing soil erosion and nutrient loss from convex slopes, and it is recommended to appropriately increase the PAM application rate for convex slopes. A strong linear relationship between ammonia nitrogen and phosphate concentrations and sediment concentrations was found in the runoff on slopes with no PAM application. However, this linear relationship weakened for slopes with PAM application. The findings of this study may be valuable for optimizing nonpoint source pollution management in basins.     

A flume design for the study of slope length effects on runoff

Recent publications from field and simulation studies indicate that runoff per unit area decreases as the length of the slope being observed is increased. This scale effect has been observed and documented for slopes with a uniform infiltration capacity as well as for slopes along which infiltration capacity is variable. This paper presents the design and testing of a laboratory flume for the study of the processes that lead to this scale effect, particularly for the case of slopes covered with crops. The features of the flume include reliable experimental control of soil water content prior to rainfall, high intensity rainfall without erosion, uniform crop growth along the slope, and the option of varying the infiltration rate independently of rainfall and soil characteristics. Copyright © 2001 John Wiley & Sons, Ltd.     

Hydrodynamic characteristics of rill flow on steep slopes

Rill erosion is a dominant sediment source on sloping lands. However, the amount of soil loss from rills on steep slopes is vastly more than that on gentle slopes because of differences in rill shape and hydraulic patterns. The aims of this paper are to determine the hydrodynamic characteristics of rills and the friction coefficients in steep slope conditions and to propose modifications of some hydraulic parameters used in soil loss prediction models. A series of inflow experiments was conducted on loess slopes. The results show that the geometric and hydraulic properties of rill on the steep loess slopes, which are characterized by the mean width of cross sections, mean velocity and mean depth of flow, are related to discharge and slope gradient in power functions. However, the related exponents to discharge are 0.26, 0.48 and 0.26, respectively, which are different from the exponents derived in previous studies, which were conducted on gentle slopes. The Manning roughness coefficient ranged from 0.035 to 0.071, with an average of 0.0536, and the Darcy–Weisbach friction coefficients varied from 0.4 to 1.9. The roughness coefficients are closely related to the Reynolds numbers and flow volumes; however, the correlations vary with slope gradient. The roughness coefficients are directly proportional to the Reynolds number and the flow volume on steep slopes, in contrast with the roughness coefficients found on gentle slopes, which decrease as the Reynolds number and flow volume increase. This difference is caused by the interactions among the hydraulics of the flow, the shape of the rills and the sediment concentrations on steep slopes. The results indicate that parameters used in models to predict rill erosion have to be modified according to slope gradient. Copyright © 2015 John Wiley & Sons, Ltd.     

Using high resolution images and elevation data in classifying erosion risks of bare soil areas in the Hatila Valley Natural Protected Area, Turkey

Soil erosion is one of the most important environmental problems. In the case of small scale areas where soil properties and climate have relatively uniform characteristics, vegetation cover and topography (i.e. ground slope) are the main factors that affect the amount of soil erosion. Lack of vegetation cover on bare soil areas, including forest road side slopes, especially in mountainous regions with steep slopes, may significantly increase the erosion rate. Determining and classifying erosion risks in such areas can help preventing environmental impacts. In this study, remotely sensed data and elevation data were used to extract and classify bare soil erosion risk areas for a study area selected from Hatila Valley Natural Protected Area in northeastern Turkey. High resolution IKONOS imagery was used to apply land use classification in ERDAS Imagine 9.0. To generate erosion risk map of the bare soil areas, classified image was superimposed on top of slope map, generated based on a Digital Elevation Model (DEM) in ArcGIS 9.2. The results indicated that 1.43, 5.85, 34.62, 53.16, and 4.94% of the bare soil areas in the study area were under very low, low, medium, high, and very high erosion risks, respectively. The overall classification accuracy of 82.5% indicated the potential of the proposed methodology.     

Rill erosion processes on a constantly saturated slope

Rill erosion processes on saturated soil slopes are important for understanding erosion hydrodynamics and determining the parameters of rill erosion models. Saturated soil slopes were innovatively created to investigate the rill erosion processes. Rill erosion processes on saturated soil slopes were modelled by using the sediment concentrations determined by sediment transport capacities (STCs) measurement and the sediment concentrations at different rill lengths. Laboratory experiments were performed under varying slope gradients (5°, 10°, 15°, and 20°) and unit-width flow rates (0.33, 0.67, and 1.33 × 10⁻³ m³ s⁻¹ m⁻¹) to measure sediment concentrations at different rill lengths (1, 2, 4, and 8 m) on saturated soil slopes. The measured sediment concentrations along saturated rills ranged from 134.54 to 1,064.47 kg/m³, and also increased exponentially with rill length similar to non-saturated rills. The model of the rill erosion process in non-saturated soil rills was applicable to that in saturated soil rills. However, the sediment concentration of the rill flow increased much faster, with the increase in rill length, to considerably higher levels at STCs. The saturated soil rills produced 120–560% more sediments than the non-saturated ones. Moreover, the former eroded remarkably faster in the beginning section of the rills, as compared with that on the non-saturated soil slopes. This dataset serves as the basis for determining the erosion parameters in the process-based erosion models on saturated soil slopes.     

Water flow velocity over frozen and nonfrozen black soil slopes

Ice‐ and snow‐melted water flow over partially thawed frozen soil of cultivated slopes causes serious soil erosion, which results in soil degradation and affects productivity in Northeast China. Water flow velocity over frozen and nonfrozen soil shows importance in understanding meltwater erosion. In this work, a series of laboratory experiments were conducted to measure water flow velocity over frozen and nonfrozen soil slopes. Experiments were performed using the electrolyte trace method under the pulse boundary model, under conditions of 4 slope gradients (5°, 10°, 15°, and 20°), 3 flow rates (1, 2, and 4 L/min), and 7 sensors positioned at 0.1, 1.0, 2.0, 3.0, 4.0, 5.0, and 6.0 m away from the electrolyte injection point. Results showed that velocities over frozen soil slopes increased with flow rate and slope gradient. Flow velocities over nonfrozen soil slopes increased with flow rate and slope gradients from 5° to 15° and stabilized at 15°. Flow velocities over frozen soil slopes were 30%, 54%, 71%, and 91% higher than those over nonfrozen ones at slope gradients of 5°, 10°, 15°, and 20°. Flow velocities over frozen soil slopes under different flow rates of 1, 2, and 4 L/min were approximately 52%, 59%, and 79% higher than those over nonfrozen soil, respectively. This study can help in assessing the erosion of partially thawed frozen soil by meltwater flow.     

Topographic thresholds for plant colonization on semi‐arid eroded slopes

Soil erosion plays an important role in plant colonization of semi‐arid degraded areas. In this study, we aimed at deepening our knowledge of the mechanisms that control plant colonization on semi‐arid eroded slopes in east Spain by (i) determining topographic thresholds for plant colonization, (ii) identifying the soil properties limiting plant establishment and (iii) assessing whether colonizing species have specific plant traits to cope with these limitations. Slope angle and aspect were surrogates of erosion rate and water availability, respectively. Since soil erosion and water availability can limit plant establishment and both can interact in the landscape, we analysed variations in colonization success (vegetation cover and species number) with slope angle on 156 slopes, as a function of slope aspect. After determining slope angle thresholds for plant colonization, soil was sampled near the threshold values for soil analysis [nitrogen, phosphorous, calcium carbonate (CaCO₃), water holding capacity]. Plant traits expressing the plant colonizing capacity were analysed both in the pool of species colonizing the steep slopes just below the threshold and in the pool of species inhabiting gentler slopes and absent from the slopes just below the threshold. Results show that the slope angle threshold for plant colonization decreased from north to south. For the vegetation cover, threshold values were 63°, 50°, 46°, 41° for the north, east, west and south slope aspect classes, respectively, and 65°, 53°, 49° and 45° for the species richness and the same aspect classes. No differences existed in soil properties at slope angle threshold values among slope aspects and between slope positions (just below and above the threshold) within slope aspect classes. This suggests that variations between slope aspect classes in the slope angle threshold result from differences in the colonizing capacity of plants which is controlled by water availability. Long‐distance dispersal and mucilage production were preferably associated with the pool of colonizing species. These results are discussed in the perspective of a more efficient ecological restoration of degraded semi‐arid ecosystems where soil erosion acts as an ecological filter for plant establishment. Copyright © 2009 John Wiley & Sons, Ltd.     

Variability of interrill erosion at low slopes

Numerous models and risk assessments have been developed in order to estimate soil erosion from agricultural land, with some including estimates of nutrient and contaminant transfer. Many of these models have a slope term as a control over particle transfer, with increased transfer associated with increased slopes. This is based on data collected over a wide range of slopes and using relatively small soil flumes and physical principals, i.e. the role of gravity in splash transport and flow. This study uses laboratory rainfall simulation on a large soil flume to investigate interrill soil erosion of a silt loam under a rainfall intensity of 47 mm h^?1 on 3%, 6% and 9% slopes, which are representative of agricultural land in much of northwest Europe. The results show: (1) wide variation in runoff and sediment concentration data from replicate experiments, which indicates the complexities in interrill soil erosion processes; and (2) that at low slopes processes related to surface area connectivity, soil saturation, flow patterns and water depth may dominant over those related to gravity. Consequently, this questions the use of risk assessments and soil erosion models with a dominant slope term when assessing soil erosion from agricultural land at low slopes. Copyright © 2010 John Wiley & Sons, Ltd.     

Response of soil detachment rate to the hydraulic parameters of concentrated flow on steep loessial slopes on the Loess Plateau of China

Using hydraulic parameters is essential for describing soil detachment and developing physically based erosion prediction models. Many hydraulic parameters have been used, but the one that performs the best for describing soil detachment on steep slopes when the lateral expansion (widening) of rills is not limited has not been identified. An indoor concentrated flow scouring experiment was performed on steep loessial slopes to investigate soil detachment rates for different flow rates and slope gradients. The experiments were conducted on a slope‐adjustable plot (5 m length, 1 m width, 0.5 m depth). Sixteen combinations of 4 flow rates (10, 15, 20, and 25 L/min) and 4 slope gradients (17.6%, 26.8%, 36.4%, and 46.6%) were investigated. The individual and combined effects of slope gradient and flow hydraulic parameters on soil detachment rate were analysed. The results indicated that soil detachment rate increased with flow rate and slope gradient. Soil detachment rate varied linearly and exponentially with flow rate and slope gradient, respectively. Multivariate, nonlinear regression analysis indicated that flow depth exerted the greatest influence on the soil detachment rate, followed by unit discharge per unit width, slope gradient, and flow rate in this study. Shear stress and stream power could efficiently describe the soil detachment rate using a power equation. However, the unit stream power and unit energy of the water‐carrying section changed linearly with soil detachment rate. Stream power was an optimal hydraulic parameter for describing soil detachment. These findings improve our understanding of concentrated flow erosion on steep loessial slopes.     

Effects of rainfall and slope on runoff,soil erosion and rill development: an experimental study using two loess soils

Runoff generation and soil loss from slopes have been studied for decades, but the relationships among runoff, soil loss and rill development are still not well understood. In this paper, rainfall simulation experiments were conducted in two neighbouring plots (scale: 1 m by 5 m) with four varying slopes (17.6%, 26.8%, 36.4% and 46.6%) and two rainfall intensities (90 and 120 mm h^?1) using two loess soils. Data on rill development were extracted from the digital elevation models by means of photogrammetry. The effects of rainfall intensity and slope gradient on runoff, soil loss and rill development were different for the two soils. The runoff and soil loss from the Anthrosol surface were generally higher than those from the Calcaric Cambisol surface. Higher rainfall intensity produced less runoff and more sediment for almost each treatment. With increasing slope gradient, the values of cumulative runoff and soil loss peaked, except for the treatments with 90 mm h^?1 rainfall on the slopes with Anthrosol. With rainfall duration, runoff discharge decreased for Anthrosol and increased for Calcaric Cambisol for almost all the treatments. For both soils, sediment concentration was very high at the onset of rainfall and decreased quickly. Almost all the sediment concentrations increased on the 17.6% and 26.8% slopes and peaked on the 36.4% and 46.6% slopes. Sediment concentrations were higher on the Anthrosol slopes than on the Calcaric Cambisol slopes. At 90 mm h^?1 rainfall intensity, increasingly denser rills appeared on the Anthrosol slope as the slope gradient increased, while only steep slopes (36.4% and 46.6%) developed rills for the Calcaric Cambisol soil. The contributions of rill erosion ranged from 36% to 62% of the cumulative soil losses for Anthrosol, while the maximum contribution of rill erosion to the cumulative soil loss was only 37.9% for Calcaric Cambisol. Copyright © 2014 John Wiley & Sons, Ltd.     

Meltwater erosion process of frozen soil as affected by thawed depth under concentrated flow in high altitude and cold regions

Changes in thawed depth of frozen soil caused by diurnal and seasonal temperature fluctuations are commonly found in high altitude and latitude regions of the world. These changes significantly influence hydrologic and erosion processes. Experimental data are necessary to improve the understanding and modeling of the phenomenon. Laboratory experiments were conducted in Beijing to assess the impacts of thawed soil depth, slope gradient, and flow rate on soil erosion by concentrated meltwater flow over an underlying frozen soil layer. Soil samples from watershed were filled in flumes, saturated before being frozen. After the soil was completely frozen, flumes were taken out of storage to thaw the frozen soil from top to the designed depths. Meltwater flow was simulated using a tank filled with water and icecubes at approximately 0°C. The erosion experiments involved four thawed soil depths of 1, 2, 5, and 10 cm; three slope gradients of 5°, 10°, and 15°; and three flow rates of 1, 2, and 4 l/min; and seven rill lengths of 0.5, 1, 2, 3, 4, 5, and 6 m. Sediment‐laden water samples were collected at the lower end of the flume for determination of sediment concentration. The results showed that sediment concentration increased exponentially with rill length to approach a maximum value. The sediment concentrations were closely correlated with thawed soil depth, flow rate, and slope gradient. Shallower thawed depths delivered more sediments than deeper thawed depths. Slope gradient was the primary factor responsible for severe erosion. The effect of flow rate on sediment concentration which decreased with increasing slope gradient, was not as significant as that of slope gradient. Results from these experiments are useful for understanding the effect of thawed soil depth on erosion process in thawed soils subject to freezing and for estimating erosion model parameters. Copyright © 2017 John Wiley & Sons, Ltd.     

Assessment of the impacts of land-use change and slope position on soil loss by magnetic susceptibility-based models

Soil loss is a global environmental problem resulting from the erosion process caused by many factors,including land use and slope position. Estimation of total soil loss from agricultural fields is useful for understanding the consequences of historical and current erosion. The main purposes of the current study are to explore the application of magnetic measurements in the mapping and measuring soil redistribution in cultivated(MZ13) and forested(MZ17) transects in a Moroccan subcatchment, to ...     

黏性土坡地震失稳机制及其在黄土斜坡地震稳定性分析中的应用

基于地震作用下黏性土坡失稳滑动特点,以土体应力状态及其变化分析边坡失稳过程。通过分析地震作用下边坡不同部位土体应力状态和剪应力变化,结合实际地震边坡失稳破坏特征,提出黏性土坡地震三段式滑动失稳机制。在分析该滑动失稳机制与有限元强度折减法之间应力关联的基础上,将两者结合应用于实际黄土地震滑坡动力稳定性分析。依据此考虑得到的动力安全系数相比较其他方法,与极限平衡法得到的结果更为接近。     

Hillslope response to tectonic forcing in threshold landscapes

Hillslopes are thought to poorly record tectonic signals in threshold landscapes. Numerous previous studies of steep landscapes suggest that large changes in long‐term erosion rate lead to little change in mean hillslope angle, measured at coarse resolution. New LiDAR‐derived topography data enables a finer examination of threshold hillslopes. Here we quantify hillslope response to tectonic forcing in a threshold landscape. To do so, we use an extensive cosmogenic beryllium‐10 (¹⁰Be)‐based dataset of catchment‐averaged erosion rates combined with a 500 km² LiDAR‐derived 1 m digital elevation model to exploit a gradient of tectonic forcing and topographic relief in the San Gabriel Mountains, California. We also calibrate a new method of quantifying rock exposure from LiDAR‐derived slope measurements using high‐resolution panoramic photographs. Two distinct trends in hillslope behavior emerge: below catchment‐mean slopes of 30°, modal slopes increase with mean slopes, slope distribution skewness decreases with increasing mean slope, and bedrock exposure is limited; above mean slopes of 30°, our rock exposure index increases strongly with mean slope, and the prevalence of angle‐of‐repose debris wedges keeps modal slopes near 37°, resulting in a positive relationship between slope distribution skewness and mean slope. We find that both mean slopes and rock exposure increase with erosion rate up to 1 mm/a, in contrast to previous work based on coarser topographic data. We also find that as erosion rates increase, the extent of the fluvial network decreases, while colluvial channels extend downstream, keeping the total drainage density similar across the range. Our results reveal important textural details lost in 10 or 30 m resolution digital elevation models of steep landscapes, and highlight the need for process‐based studies of threshold hillslopes and colluvial channels. Copyright © 2012 John Wiley & Sons, Ltd.     

基于树根桩加固的震后公路路基边坡综合修复技术

为实现震后路基边坡的快速修复,提出基于树根桩加固的震后公路路基边坡综合修复技术。根据路基震害及地质环境对路基边坡的影响分析因素,采用树根桩技术及气泡轻质填土技术修复路基边坡;利用三维有限差分法模拟树桩加固气泡轻质填土修复后路基边坡得到参数值,再结合强度折减法计算得到的路基边坡稳定系数,最后得到优化树根桩的设计参数,以实现震后路基边坡综合修复。以上方法通过仿真模拟实验,得到最优的树根桩排距为2.5 dm。     

Modelling sediment trapping in vegetative filter strips on steep slopes

The slope effects on sediment trapping process in vegetative filter strips (VFS) are usually neglected in current modelling practices for VFS operation, which hamper the VFS design and performance evaluation, especially on steep slopes. To fill the knowledge gap, 12 laboratory experiments of sediment trapping in VFS were conducted with three different inflow discharge (80, 100, and 120 ml s⁻¹) and four slope angles (5,10, 15, and 20°). The experimental results show that, on steep slopes (10, 15, and 20°), a part of trapped sediment particles in VFS can be eroded again and then dragged to the downstream as bed load, whilst they do not move on gentle slope (5°). To describe the complex processes, a simple and effective modelling framework was developed for sloped VFS by coupling the slope infiltration, runoff, and modified sediment transport model. The model was tested against the experimental results and good agreements between the modelled and observed results were found in both runoff and sediment transport processes for all cases. On steep slopes, the sediment trapping performance of VFS decreases significantly because the erosion of deposited sediment particles can account for more than 60% of the sediment load in the outflow. The slope effect on sediment trapping efficiency of VFS varies greatly with soil, VFS, and slope properties. The model was compared with previous sediment transport equation and found that both methods can satisfactorily predict the sediment trapping of VFS on gentle slopes, but previous sediment transport equation is likely to overestimate the sediment trapping efficiency in VFS on steep slopes. This model is expected to provide a more realistic and accurate method for predicting runoff and sediment reduction in VFS on sloping surfaces.     

Coupling between landslides and eroding stream channels reconstructed from spruce tree rings (examples from the Carpathians and Sudetes – Central Europe)

The analysis of the positive feedback between landslides and erosion requires determination of the precise temporal and spatial relations between events of colluvium delivery and fluvial erosion. In our study we use decennial datasets on the occurrence of landsliding and erosion achieved through dendrochronological methods. Four sites covering areas of landslide slopes and adjacent valley floors with stream channels were studied. Landsliding on slopes was dated from the tree‐ring eccentricity developed in stems tilted due to bedrock instability. Erosion in channels was dated using the wood anatomy of roots exposed by erosion of the soil cover. Analysis of the temporal relations between dated landsliding, erosion and precipitation record has revealed that two types of repeating sequences can be observed: (1) rainfall → landsliding → erosion; (2) rainfall → erosion → landsliding. These sequences are an indication of the occurrence of slope‐channel positive feedback in the sites studied. In the first type, landsliding triggered by rainfall delivers colluvia into the valley floor and causes its narrowing, which in turn causes increased erosion. In the second type erosion triggered by rainfall disturbs the slope equilibrium and causes landsliding. Landsliding and erosion, once triggered by precipitation, can occur alternately in years with average precipitation and reinforce one another. Bidirectional coupling between landsliding and channel erosion was shown notably through the effects of channel shifting and forced sinuosity and by increased erosion of the slopes opposite the active landslides. Observations also suggest that the repetition of sequences described over longer periods of time can lead to a general widening of the valley floor at the expense of slopes and to a gradual change of the valley cross‐profile from narrow, V‐shaped into a wide flat‐bottomed. Thus landsliding–erosion coupling/positive feedback was recognized as an important factor shaping hillslope–valley topography of the mid‐mountain areas studied. Copyright © 2014 John Wiley & Sons, Ltd.     

Travel distance of failed slopes during 2004 Chuetsu earthquake and its evaluation in terms of energy

A number of slope failures during the 2004 Niigataken Chuetsu earthquake were investigated, revealing that the travel distance becomes longer as the slope gets gentler and the failed soil mass gets larger. An energy-based approach, proposed in previous research to evaluate the travel distance of failed slopes, is modified by adding the model test results and introducing a simple evaluation method. The energy approach is then applied to a number of slopes failed during the earthquake to back-calculate mobilized friction coefficients, revealing their strong dependency on initial slope inclinations. The friction coefficient was found to be smaller than the initial slope inclination for gentler slopes, indicating that the failed soil mass tends to accelerate. In contrast, that for steeper slopes was higher than initial slope inclination. The friction coefficient was found to decrease with increasing volume of failed slope, which is quite consistent with previous case studies including large non-seismic landslides.     

Physical modelling of rainfall‐ and snowmelt‐induced erosion of stony slope underlain by permafrost

Physical modelling experiments have been carried out in a cold room to test on a small scale, the effects of water supply during the thaw of an experimental slope with permafrost. Permafrost was maintained at depth and a thin active layer was frozen and thawed from the surface. Data from the experiments relate to two different conditions, first with moderate rainfall, and second with heavy rainfall during the thaw period. When moderate rainfall is applied during thaw phases, the experimental slope is slightly degraded. At the scale of the experiment, erosion processes involve frost jacking of the coarse blocks, frost creep and gelifluction that induce slow and gradual down slope displacements of the active layer, but also small landslides leading to large but slow mass movements with short displacements. Changes in experimental slope morphology are marked by the initiation of a small‐scale drainage network and the development of a little crest line which shows a progressive upslope migration. With such boundary conditions, there is not enough water supply to evacuate downslope the whole of the eroded material and a topographic smoothing is observed. When heavy rainfall is applied during thaw periods, rapid mass wasting (small mud‐flows and debris flows) become prominent. Slope failures are largely controlled by the water saturation of the active layer and by the occurrence of steeper slopes. At the scale of the experiment, rates of erosion and maximum incision increase by about 100% leading to significant slope degradation with marked and specific scars comparable to gullying. These morphological changes are dependant on both the size and the frequency of catastrophic events. These experiments provide detailed data that could improve the knowledge of the physical parameters that control the initiation, at a small‐scale, of erosion processes on periglacial slopes with a thin active layer and/or with thin cover of mobilizable slope deposits. Copyright © 2010 John Wiley & Sons, Ltd.     

Exploring the interaction of surface roughness and slope gradient in controlling rates of soil loss from sloping farmland on the Loess Plateau of China

Surface roughness and slope gradient are two important factors influencing soil erosion. The objective of this study was to investigate the interaction of surface roughness and slope gradient in controlling soil loss from sloping farmland due to water erosion on the Loess Plateau, China. Following the surface features of sloping farmland in the plateau region, we manually prepared rough surfaces using four tillage practices (contour drilling, artificial digging, manual hoeing, and contour plowing), with a smooth surface as the control measure. Five slope gradients (3°, 5°, 10°, 15°, and 20°) and two rainfall intensities (60 and 90 mm/hr) were considered in the artificial rainfall simulation experiment. The results showed that the runoff volume and sediment yield increased with increasing slope gradient under the same tillage treatment. At gentle slope gradients (e.g., 3° and 5°), the increase in surface roughness prevented the runoff and sediment production, that is, the surface roughness reduced the positive effect of slope gradient on the runoff volume and sediment yield to a certain extent. At steep slope gradients, however, the enhancing effect of slope gradient on soil erosion gradually increased and surpassed the reduction effect of surface roughness. This study reveals the existence of a critical slope gradient that influences the interaction of surface roughness and slope gradient in controlling soil erosion on sloping farmland. If the slope gradient is equal to or less than the critical value, an increase in surface roughness would decrease soil erosion. Otherwise, the increase in surface roughness would be ineffective for preventing soil erosion. The critical slope gradient would be smaller under higher rainfall intensity. These findings are helpful for us to understand the process of soil erosion and relevant for supporting soil and water conservation in the Loess Plateau region of China.