Soil & Tillage Research 105 (2009) 307–312

The effect of plant hedgerows on the spatial distribution of soil erosionand soil fertility on sloping farmland in the purple-soil area of China

Chaowen Lin a,b, Shihua Tu b, Jingjing Huang b, Yibing Chen b,*a College of Animal Science, Sichuan Agricultural University, Sichuan, Ya’an 625014, PR Chinab Soil and Fertilizer Institute, Sichuan Academy of Agricultural Sciences, Sichuan, Chengdu 610066, PR China

A R T I C L E I N F O

Article history:

Received 5 May 2007

Received in revised form 15 January 2009

Accepted 19 January 2009

Keywords:

Hedgerow

Soil erosion

Soil nutrient distribution

Soil texture

A B S T R A C T

Recently the effect of plant hedgerows on controlling soil and water loss has been well recognized, and

this technology has been widely applied in the world. However, there are few studies on hedgerows’

effect on soil fertility of sloping lands. With an 8-year fixed field experiment, we investigated the effect of

two different hedgerows on soil fertility through comparing with the control. Our results showed that

along contour lines across the field, clay particles tended to accumulate above plant hedgerows but to be

eroded downward below hedgerows. Except for potassium (K), all plant nutrients and soil organic matter

showed the same distribution pattern as clay particles. K, however, was evenly distributed in the field

without any noticeable influence from hedgerows. Since the beginning of our field experiment, soil

phosphorus (P) kept accumulating, while soil organic matter and K were in depletion. Taken together,

our results suggest that better nutrient management for the sloping lands should reduce P but increase

farm manure and K. As far as the whole sloping field is concerned, special attention in nutrient

management should be paid to the soil stripes below hedgerows, the portions suffering from more

serious soil erosion.

� 2009 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Soil & Tillage Research

journa l homepage: www.e lsev ier .com/ locate /s t i l l

1. Introduction

Purple soil prevails in the upper reaches of the Yangtze River,which is one of the most important soils for agricultural productionin subtropical areas of China. Due to abundant rainfall and erosivetopographic conditions, tremendous soil loss occurs duringsummer storms. Moreover, intensive cultivation and socioeco-nomic pressure have accelerated soil erosion on sloping lands (Guoet al., 2008). As a result, soil erosion becomes a serious challengefor agricultural development in these areas. Alley-cropping isintroduced to sloping farmlands as a soil-conservation method inhilly areas, which plays a significant role in reducing soil erosion,controlling non-point source pollution, increasing system outputand debasing the input of sloping land (Tu et al., 2005). Thus, thismethod has been widely used in mountain areas of both torrid andtemperate zones. With the efforts of Chen et al. (2002), 11economical hedgerow patterns and their combinations have beendeveloped to fit the slopes, soil types and weather conditions in theupper reaches of the Yangtze River, which has extended to morethan 130,000 ha in southwestern China. Several studies havereported the effect of hedgerows on controlling soil erosion

* Corresponding author. Tel.: +86 28 84504296; fax: +86 28 84796435.

E-mail address: ybchen@mail.sc.cninfo.net (Y. Chen).

0167-1987/$ – see front matter � 2009 Elsevier B.V. All rights reserved.

doi:10.1016/j.still.2009.01.001

(Salvador-Blanes et al., 2006; Cullum et al., 2007; Raffaelle et al.,1997; Xu et al., 1999) and non-point source pollution (Chaubeyet al., 1995) in the last two decades, and particularly some studieshave focused on its influence on micro-topographic features(Dabney et al., 1997) and the shape change of slope (Zheng, 2006;Tian et al., 2003). However, so far there has been no study abouthedgerows’ effect on the spatial variation of soil fertility and theproductivity of lands. In the late 1990s, Zhu et al. (2003) studiedthe redistribution process of soil particles on sloping lands withalley cropping, but failed to estimate the effect of hedgerows withdifferent ages on the redistribution of soil particles for a larger areadue to the small size of trial area (10 m � 2 m) and the limitedperiod (5 years). In this study, we aimed to investigate the spatialredistribution of soil fertility in the plot after hedgerows wereplanted, thereby providing important sights into the hedgerow-management optimization.

2. Study site and methodology

2.1. Background of the study site

Our field experiment was carried out in the upper reaches of theTuo River system of the Yangtze River, 10483401200 to 10483501900Eand 3080501200 to 3080604400N with an altitude of 395 m. This regionis in a subtropical monsoon climate zone with an average annual

Table 2Fertilizer rates used for different crops (kg/ha).

Crops N P2O5 K2O

Corn (I. mays L.) 186.3 99 99

Sweet potato (Dioscorea esculenta) 58.5 58.5 58.5

Wheat (Triticum L.) 135 67.5 37.5

Table 1Some nutrition properties of the soil in this study.

pH 8.0

OM 0.94

Total N (%) 0.056

Total P (%) 0.068

Total K (%) 1.59

Avail. N (mg kg�1) 44

Avail. P (mg kg�1) 3.05

Avail. K (mg kg�1) 91.3

C. Lin et al. / Soil & Tillage Research 105 (2009) 307–312308

precipitation of 966 mm (ranging from 725.2 to 1290.7 mm), ofwhich 70% falls from June to September. The average annualtemperature is 16.8 8C, with an average summer high of 27.4 8C inJuly and an average winter low of 7.4 8C in January. The area isdominated by purple soil, classified as Entisol according to the soiltaxonomy of the U.S.D.A. (Soil Survey Staff, 1999), which is usually50–80 cm in depth with relatively light texture and poor soilfertility (Table 1).

2.2. Materials and methods

The trials started in November 1997, including three runoffplots (7 m � 20 m each) separated by concrete plates. Theexperiment was designed with three treatments: the control(without hedgerows), three strips with vetiver (Vetiveria ziza-

niodes) planted in contour, and three strips with false indigo(Amorpha fruticosa) planted in contour. The hedgerow belt was0.5 m wide, and the distance between two hedgerows was 6.16 m.The hedge crops were planted in rows with a spacing distance of0.2 m, and the distance between tow plants in a row was 0.2 m(Fig. 1). Both hedgerows were controlled to be about 0.5 m high.One year after the hedgerows were planted, the canopy of vetiverfully covered the soil surface of the hedgerow belt; and thedistance between false indigo plants was 0.1 m.

The crop rotation was wheat, corn and sweet potato, accountingfor 40% of the use of arable land in this area (Yang et al., 1997). Foreach year, Xushu 18, a variation of sweet potato, was used andtransplanted around June 16th with a population of 45,000 plantsper ha; Chengdan 18, a corn race, was sown around April 12th witha population of 42,000 plants per ha and harvested around August20th; Chuanmai 22, a wheat variation, was used in the trial andsown around November 7th. Alley cropping was combined withwinter crop (wheat) and summer crops (corn and sweet potato)intercropping. The wheat was seeded by 13.3 cm between seedingholes and 23.3 cm between rows and harvested in middle May.Table 2 lists the fertilizer rates used for different crops. It has been8 years since the hedgerows were first established in 1997, and the

Fig. 1. Treatment layout of

terraces were gradually formed due to the growth of thehedgerows.

After wheat was harvested in 2006, soil samples were taken atfive points at 1.5, 3.5 and 5.5 m from the hedgerows in each plot,and these samples were mixed to represent the upper, middle andlower parts of the plot, respectively (from the bottom to the top ofthe slope, the sample-location distances were 2, 4, 6, 8.5, 10.5, 12.5,15, 17 and 19 m). The total nitrogen (N), phosphorus (P), potassium(K), organic matter (OM) and soil texture were analyzed. The totalN was measured with Kjeldahl method (Lu, 2000a); the total P wasfused by Na2CO3 and analyzed by the colormetric method (Lu,2000b); the total K was fused by NaOH and analyzed by the flamephotometry (Lu, 2000c); OM was analyzed by the K2Cr2O7

volumetric method (Lu, 2000d); and soil texture was analyzedby the specific-gravity method (Lu, 2000e).

The amounts of runoff and sediment from each plot werecollected by tanks at the plot bottom after each rain, measured byV-notch, and recorded daily with a SW40 water level gauge. Therunoff flux was calculated according to the following formula:

Q ¼ 8

15mð2gÞ1=2H5=2 tg

u2

� �;

where Q stands for runoff flux (ml/s), m for runoff coefficient whichcan be obtained by calibration experiment, H for the water level ofV-notch (cm), u for the angle of V-notch in degree (308), g for theacceleration of gravity. The runoff amount was calculatedaccording to the following formula:

M ¼Xn

i¼1

Qiti

1000;

where M stands for runoff amount (l), Qi for runoff flux (ml/s), ti forthe time of runoff overflowing the V-notch keeping in the runoffflux of Qi (s).

After every rain, five runoff samples were taken in the tank witha beaker, and the sediment content was measured by filter paper.The amount of sediment was equal to the runoff volume multipliedby the sediment concentration measured.

The plots, 208 in slope, were divided by the concrete plates of50 cm in height with one half buried into the soil and the other halfprojected above the surface. The slope change of the plots with

this field experiment.

Table 3Annual runoff and soil erosion from different treatments.

Years Item

Control Vetiver hedgerows False indigo hedgerows

Run-off (mm) Sediment (t/ha) Run-off (mm) Sediment (t/ha) Run-off (mm) Sediment (t/ha)

1998 93.2 78.8 27.2 11.3 34.5 14.3

1999 78.3 65.8 15.3 1.3 23.1 3.5

2000 71.6 57.7 12.7 1.0 21.6 2.1

2001 171.7 44.4 33.7 3.1 41.9 8.2

2002 14.8 2.0 3.7 0.1 5.1 0.1

2003 15.3 0.4 5.1 0.1 5.4 0.2

2004 18.6 0.9 8.4 0.3 11.3 0.3

2005 18.2 10.4 6.1 0.4 7.2 0.5

Total 481.7 260.4 112.2 17.6 150.1 29.2

Fig. 3. Effect of false indigo hedgerows on sloping landform.

C. Lin et al. / Soil & Tillage Research 105 (2009) 307–312 309

different treatments was measured with the top of concrete platesas a reference.

3. Results

Due to small land pieces and slope variations in our slopingarea, it was difficult to find a suitable and large-enough field forreplicated runoff plots. So we used large plots (20 m � 7 m) and along study term (8 years), rather than replicated plots. As a result,the statistical analysis of our data cannot be preformed, and ourresults were used to reveal the general trend of differenttreatments on soil fertility qualitatively.

3.1. Influence of hedgerows on soil and water loss

Table 3 shows the runoff and sediment of different treatmentsfrom 1998 to 2005. Clearly, the hedgerows remarkably reducedboth runoff and sediment. The sediment reduction was about231.2–242.8 t/ha, equivalent to 1.8 cm surface soil (bulk density1.3 g/cm3) in 8 years. The 1-year-old hedgerows reduced runoff by63.0–70.8% and sediment by 81.9–85.7%, demonstrating incredibleeffectiveness of hedgerows on controlling soil and water loss.Compared with false indigo hedgerows, vetiver hedgerows couldfurther reduce runoff by 25.2% and sediment by 39.7%, showingthat herbaceous hedges are superior to weedy species on soilconservation. Thus, hedgerows function in blocking sedimentwashed away from the field as well as reducing the runoff velocity.The vetiver hedges were denser than the false indigo hedges, sothey had a better conservation effect, suggesting this is a moreeffective hedge pattern.

Fig. 2. Effect of vetiver hedgerows on sloping landform.

3.2. Influence of hedgerows on micro-landform of sloping lands

As shown in Figs. 2 and 3, we found that the slope of the upperpart of each plot was considerably reduced, since the upper partwas under more erosion. The middle part of the control plot wasobviously eroded because the washing-down runoff was smooth.Therefore, the erosion energy of runoff was increased at thislocation, leading to more severe soil erosion. At the lower part ofthis plot, sediments were blocked by the concrete plates, makingthe slope further reduced (Zheng, 2005). In particular, in thehedgerow plots, sediments were obviously accumulated in front ofthe hedgerow belt, resulting from the slower runoff and theformation of backwater strips above the hedgerows (Sun et al.,1999a). Soil erosion at this part of the slope was also enhanced byplough erosion (Zhang et al., 2001), so the slope steepness wasobviously reduced. Thus, a hedge bund along the hedgerow beltwas formed naturally with time. The bund of vetiver appeared tobe stronger than that of false indigo (Fig. 4), indicating the vetiver

Fig. 4. The difference of soil surface height between the hedgerow treatments and

the control.

Fig. 6. The soil-particle distribution of the vetivar hedgerow treatment.

Fig. 7. The soil-particle distribution of the false indigo treatment.

Table 4Slope changes at different contour positions that were affected by different

hedgerows.

Average slope Plot

Control Vetiver

hedgerows

False indigo

hedgerows

1–6 m (top) 178420 168210 16880

8–13 m (middle) 208310 168100 178340

15–20 m (bottom) 188170 168520 178510

Average 188500 168280 178110

C. Lin et al. / Soil & Tillage Research 105 (2009) 307–312310

hedgerow treatment is more effective than the false indigohedgerow treatment.

Table 4 shows the remarkable effect at the boundary, where theslope was reduced from 208 to 168280 by vetiver and from 208 to178110 by false indigo. Although the average slope of the controlplot was more or less reduced with time, it was induced by soilerosion. But, at the middle part of control plot, the slope was moreincreased; Sun et al. (1999b) proposed that soil erosion under thiscondition is shifted from minor sheet erosion to severe sheeterosion plus rill erosion. The slope reduction in the hedgerowtreatments increased from the upper to the lower part of the plots.This effect was slowed down as sediments gradually piled up at thebottom. The upper part of the plots became a net erosive areabecause the limitation of concrete boundary cut down theincoming sediments. At the middle part of the plots, somesediment was deposited from the top part of the slope, thus furtherreducing the slope. On the other side, the lower part became thenet sediments receiving zone, and the slope turned out to be muchflatter.

3.3. Effect of hedgerows on the spatial distribution of soil particles

Figs. 5–7 show the soil texture with different treatments. Themechanical composition of experimental soil was dominated bysandy silt and silt. But it showed some distribution patterns acrossthe whole slope. In the control treatment, the sand content wasmuch higher at the upper part of the plot (26.7%) than that at themiddle or bottom (19.5%); conversely, clay content was muchlower (24.2%) at the upper part and higher at the bottom. This maybe due to more severe tillage erosion at the upper part (Zhang et al.,2004a,b, 2006). Due to severe soil erosion at the upper part, thedebris of parent rock entered the plough layer. The enrichment oferoded clay particles from the upper part considerably changed themechanical composition at the bottom part.

As the hedgerows grew up and their capability in blocking soilerosion was enhanced, they began to affect the soil-particledistribution on the slope, leading to the silt-particle accumulation

Fig. 5. The soil-particle distribution of the control treatment.

in front of the hedgerows. The average silt content in front ofvetiver hedgerows was 29.6%, which was 6.6% higher than that atthe middle part and 6.5% higher than that at the lower side,respectively. The sand content in front of the hedgerows was lowerthan any other part, which was 17.3, 11.4% lower than at themiddle area and 22.8% lower than at the lower side of thehedgerows, respectively. Just below the hedgerows, clay particleswere reduced and the sand was increased. Thus, we observed asoil-particle redistribution pattern along the slope: the pooling ofclay particles in front of the hedgerows and the accumulation ofsandy particles down below the hedgerows (Figs. 6 and 7).

3.4. Influence of hedgerows on the spatial distribution of OM

In order to accurately control the nutrient application rate, noorganic fertilizer was applied in this experiment. Table 5 showsthat the OM content of all the treatments was lower than the initialOM content. Compared with the initial content, the OM contents inall treatments decreased: 34.3% lower in the control, 25.8% in thevetiver hedgerow treatment and 21.8% in the false indigohedgerow treatment. These results indicate that hedgerows canslow down the mining process of soil OM.

Fig. 8 shows the OM distribution along the slope of the controltreatment followed an order of the upper part < the middle

Table 5Temporal changes of organic matter content in soils with different treatments.

Treatment Average

OM (%)

Compared to the

initial content (�%)

Compared to

CK (�%)

Initial content 0.940 – –

CK 0.618 �34.3 –

Vetiver hedgerow 0.698 �25.8 13.0

False indigo hedgerow 0.735 �21.8 19.0

Fig. 8. Effect of different hedgerows on organic matter content and the distribution

on the slope.Fig. 10. Effect of different hedgerows on soil total P along the slope.

Fig. 11. Effect of different hedgerows on soil total K along on the slope.

C. Lin et al. / Soil & Tillage Research 105 (2009) 307–312 311

part � the bottom part. This is similar to the distribution of soilparticles, probably because OM is usually attached to clay particles.In the hedgerow treatments, the content of soil OM was lower atthe upper part and higher at the bottom of the plots.

3.5. Effect of hedgerows on the N spatial distribution

Fig. 9 shows the spatial distribution of soil N. Our resultsindicated that in terms of soil N, there was virtually no differenceamong the treatments (CK: 0.059%, vetiver hedgerows: 0.062%, andfalse indigo hedgerows: 0.058%); and there was also no significantchange with time, implying that the past fertilizer practice hasmaintained a very good N balance. We observed an increase in soilN from the upper to the lower part of a plot (i.e., from 0.043% at theupper part to 0.068% at the lower part). In the hedgerowtreatments, the transportation of soil N downward the slopewas minimized. At areas near the hedge bunds, N appeared toaccumulate in front of the hedgerows and erode down below thehedgerows.

3.6. Effect of hedgerows on the P spatial distribution

Fig. 10 shows the effect of hedgerows on the P spatialdistribution. The average P content of the control treatment was0.069%, almost the same as the initial P content, which indicates awell balance of soil P in the current farming practice. Similar to thesoil N distribution, the soil P content in the control increased from0.05 to 0.11% from the top to the bottom along the slope. As hedgecrops grew up, downward transportation of P in the slope wasinterdicted. As a result, soil P was accumulated in front but reduced

Fig. 9. Effect of different hedgerows on soil total N along the slope.

below the hedgerows. In the vetiver treatment, the P content was0.105, 54.1% higher than the initial content; while in the falseindigo treatment, the P content was 0.098, 44.1% higher than theinitial content. Taken together, hedgerows can considerablymaintain soil P reserves as well as reduce the P loss resultingfrom soil erosion.

3.7. Effect of hedgerows on the K spatial distribution

Fig. 11 shows the effect of hedgerows on the K spatialdistribution. Although K was relatively rich in the soil we studied,the K content decreased with time regardless of the treatmenttype. This may be due to the easily leachable property of K fromsoil, leading to mining of soil K reserves. As a result, the K spatialdistribution was not coincided with the patterns of soil N, P, or soilerosion.

4. Conclusions

The results from this study were used to reveal the general trendof different treatments on soil fertility qualitatively. Hedgerows cannotably reduce soil erosion and runoff soon after their establish-ment. The ridges built by hedgerows can reduce the slope steepness,and form natural terraces on sloping lands. The herbaceous plantappears to be better than the woody species in controlling soilerosion and water loss, while woody hedgerows can improve thesoil-OM content more effectively than herbaceous hedgerows. Clayparticles were accumulated in front but severely eroded below thehedges. The distributions of soil OM, N and P were similar to that of

C. Lin et al. / Soil & Tillage Research 105 (2009) 307–312312

clay particles. K was uniformly distributed in the field without anydetectable influence from hedgerows. While soil P continued toaccumulate, soil OM and K were depleted with time. These resultssuggest that better nutrient management on sloping lands shouldreduce P but increase farm manure and K. As far as the whole slopingland is concerned, more attention in nutrient management shouldbe paid to soil stripes—the portions below plant hedgerows, sincethey suffer from more serious soil erosion.

Acknowledgements

This study was supported by grants from 973 Foundation ofChina (No. 2006 CB100206), EU 5th framework foundation(Erochinut), the Foundation of International Plant NutritionInstitute and the Foundation of Basic and Applied Foundation ofSichuan Science and Technology Department (No. 08JY-0022).

References

Chaubey, I., Edwards, D.R., Daniel, T.C., MOORE, P.A., Nichols, D.J., 1995. Effective-ness of vegetative filter strips in controlling losses of surface applied poultrylitter constituents. Transactions of the ASAE 38 (6), 1687–1692.

Chen, Y.B., Lin, C.H.W., Zhu, Z.H.L., He, G.Y., 2002. Study on alley cropping systemand its ecological and economical benefits. Journal of Soil and Water Conserva-tion 16 (6), 80–83.

Cullum, R.F., Wilson, G.V., McGregor, K.C., Johnson, J.R., 2007. Runoff and soil lossfrom ultra-narrow row cotton plots with and without stiff-grass hedges. Soil &Tillage Research 93, 56–63.

Dabney, S.M., Meyer, L.D., McGregor, K.C., 1997. Sediment control and landscapemodification with grass hedges. Proceeding of the Conference on Managementof Landscapes Distributed by Channel Incision 1093–1099.

Guo, Z.L., Zhong, C., Cai, C.F., Ding, S.W., Wang, Z.M., 2008. Nitrogen competition incontour hedgerow systems in subtropical China. Nutrient Cycling in Agroeco-system 81, 71–83.

Lu, R.K., 2000a. Analysis Method of Soil and Agrichemistry. China AgricultureScience and Technology Press, Beijing, pp. 147–149.

Lu, R.K., 2000b. Analysis Method of Soil and Agrichemistry. China AgricultureScience and Technology Press, Beijing, pp. 166–168.

Lu, R.K., 2000c. Analysis Method of Soil and Agrichemistry. China AgricultureScience and Technology Press, Beijing, pp. 188–190.

Lu, R.K., 2000d. Analysis Method of Soil and Agrichemistry. China AgricultureScience and Technology Press, Beijing, pp. 107–108.

Lu, R.K., 2000e. Analysis Method of Soil and Agrichemistry. China AgricultureScience and Technology Press, Beijing, pp. 282–288.

Raffaelle Jr., J.B., McGregor, K.C., Foster, G.R., CULLUM, R.F., 1997. Effect of narrowgrass strips on conservation reserve land converted to cropland. Transactions ofthe ASAE 40 (6), 1581–1587.

Salvador-Blanes, S., Cornu, S., Couturier, A., King, D., Macaire, J.J., 2006. Morpholo-gical and geochemical properties of soil accumulated in hedge-induced terracesin the Massif Central, France. Soil & Tillage Research 85, 62–77.

Soil Survey Staff Soil taxonomy, 1999. A Basic System of Soil Classification forMaking and Interpreting Soil Surveys, 2nd ed. USDA, Washington, DC.

Sun, H., Tang, Y., Chen, K.M., He, Y.H., 1999a. Effect of contour hedgerow system ofnitrogen fixing trees on soil fertility improvement of degraded sloping agri-cultural lands. Chinese Journal of Applied and Environmental Biology 5 (5),473–477.

Sun, H., Tang, Y., Chen, K.M., He, Y.H., 1999b. Effect of contour hedgerow system onslope lands erosion control. Bulletin of Soil and Water Conservation 19 (6), 1–5.

Tian, G.M., Wang, F.E., Chen, Y.X., He, Y.F., Fu, Q.L., Kumar, A., Lin, Q., 2003. Effect ofdifferent vegetation systems on soil erosion and soil nutrients in red soil regionof Southeastern China. Pedosphere 13 (2), 121–128.

Tu, S.H.H., Chen, Y.B., Zhu, Q., Guo, Y.Z., Zhu, Z.H.L., 2005. Role and effect of cash crophedgerows on controlling soil and water losses from sloping farmlands in theupper reaches of Yangtze River. Journal of Soil and Water Conservation 19 (6),1–5.

Xu, F., Cai, Q.G., Wu, S.H.A., Zhang, G.Y., 1999. Effect of contour hedgerows on soilnutrient loss in slope land. Journal of Soil Erosion and Soil and Water Con-servation 5 (2), 23–29.

Yang, W.Y., Zhang, Q., Lin, C.W., 1997. Erosivity distribution of rainfall and the crisisperiod of soil erosion in the Sichuan basin. Southwest China Journal of Agri-cultural Sciences (10), 41–47.

Zhang, J.H., Frielinghaus, M., Tian, G., Lobb, D.A., 2004a. Ridge and contour tillageeffects on soil erosion from steep hills lopes in the Sichuan Basin, China. Journalof Soil and Water Conservation 59 (6), 277–284.

Zhang, J.H., Lobb, D.A., Li, Y., Liu, G.C., 2004b. Assessment for tillage translocationand tillage erosion by hoeing on the steep land in hilly areas of Sichuan, China.Soil Tillage and Research 75, 99–107.

Zhang, J.H., Quine, T.A., Ni, S.J., Ge, F.L., 2006. Stocks and dynamics of SOC in relationto soil redistribution by water and tillage erosion. Global Change Biology 12,1834–1841.

Zhang, J.H., Li, Y., David, A., Lobb, D.A., Zhang, J.G., 2001. Quantifying tillagetranslocation and tillage erosion in hilly areas of Sichuan. Journal of Soil andWater Conservation 15 (2), 1–4.

Zheng, F.L., 2006. Effect of vegetation changes on soil erosion on the Loess Plateau.Pedosphere 16 (4), 420–427.

Zheng, F.L., 2005. Effects of accelerated soil erosion on soil nutrient loss afterdeforestation on the Loess Plateau. Pedosphere 15 (6), 707–715.

Zhu, Y.D., Cai, Q.G., Zhang, G.Y., Hu, X., 2003. Impact of hedgerow on the control of soilnutrient loss. Resources and Environment in the Yangtze Basin 12 (4), 345–351.