CC BY
0GEODERMA
ELSEVIER
Contents lists available at SciVerse ScienceDirect
Geoderma
journal homepage: www.elsevier.com/locate/geoderma
Revegetation as an efficient means of increasing soil aggregate stability on the Loess Plateau (China)☆
Shao-Shan An a,c,*< Frédéric Darboux b, Man Chengc
a State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau, Northwest A&F University, Yangling 712100, PR China b INRA, UR0272, UR Science du sol, Centre de Recherche Val de Loire, CS 40001, F-45075 Orléans Cedex 2, France c Institute of Soil and Water Conservation, CAS&MWR, Yangling 712100, PR China
ARTICLE INFO
ABSTRACT
Article history: Received 23 August 2012 Received in revised form 2 May 2013 Accepted 20 May 2013 Available online 4 July 2013
Keywords:
Soil aggregate stability Le Bissonnais' method Loess Plateau Grain for Green Revegetation Plant communities
Soil aggregate stability influences several aspects of soil physical behavior, such as water infiltration and soil erosion (Amézketa, 1999). We investigated the soil aggregate stability characteristics in the framework of the 'Grain for Green' vegetation rehabilitation project at the Loess Plateau (China) by using the Le Bissonnais (1996) and the modified Yoder (1936) methods. Both non-grazed grassland and afforestation revegetations were considered.
The size distribution mode was always >2 mm for the fast wetting test (FW) in the non-grazed grassland communities. This fraction accounted for approximately 40% to 80% of the total soil weight. The wet stirring (WS) test showed a distribution similar to that determined by the FW test. For the slow wetting (SW) test, 80% or more of the non-grazed grassland soil fragments was > 2 mm.
The mean weighted diameter (MWD) that was determined by the Le Bissonnais' method was different among the tests and land uses. For the FW test, all the plant communities were significantly different from that of the recently abandoned grazing on grassland at the 0-20 and 20-40 cm depths. In the Artemisia sacrorum community, the maximum MWD was approximately 3 mm for the 0-20 cm depth. There were no significant differences among the plant communities after 3 years of afforestation and 7 years of non-grazing of the grassland. The MWDs were lower in the afforestation area than in the non-grazed grassland area. The maximum MWD value from the FW test was approximately 1.8 mm and was significantly lower (<0.5 mm) for cropland. The MWD of the modified Yoder method was positively related to the slow wetting and wet stirring (WS) tests of the Le Bissonnais' method (n = 20 and r = 0.83 and 0.87, respectively).
In the Loess Plateau, revegetations by non-grazed grassland and afforestation are efficient means of increasing aggregate stability and decreasing soil erodibility. The aggregate stability under non-grazed grasslands is higher than that under afforestation. The effect of revegetation is persistent, which makes it a suitable long-term management practice. Compared with the modified Yoder's method, the FW test of the Le Bissonnais' method is better at determining aggregate stability differences among land uses and is recommended for future studies.
© 2013 The Authors. Published by Elsevier B.V. All rights reserved.
1. Introduction
Soil structure is a key soil property that impacts plant and animal life, moderates environmental quality changes through soil organic carbon sequestration, and impacts soil water quality. Soil structure can be
defined in terms of form and stability (Bronick and Lal, 2005). Good soil structure is important for the productivity of sustainable agricultural and for the preservation of environmental quality (Peng et al., 2004). Soil structure depends on the presence of stable aggregates. In fact, aggregate stability influences several soil physical processes, such as
Abbreviations: FW, fast wetting; SW, slow wetting; WS, wet stirring; MWD, mean weighted diameter; St.G., Stipa grandis P. Smirn (56 years); St.B., Stipa bungeana Trin Ledeb (36 years); At.S., Artemisia sacrorum Ledeb (25 years); Th.M., Thymus mongolicus Ronn. (15 years); Hi.O., Hierochloe ordorata Beauv (7 years); Ab.G., recently abandoned grazing on grassland (3 years); C.K.26, Caragana korshinskii Kom. (26 years); C.K.16, Caragana korshinskii Kom. (16 years); A.G.3, abandoned grazing land (3 years); Cr., slope cropland (0 year). ☆ This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
* Corresponding author at: State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau, Northwest A&F University, Yangling 712100, PR China. Tel.: +86 29 87012871; fax: +86 29 87012210.
E-mail address: shan@ms.iswc.ac.cn (S.-S. An).
0016-7061/$ - see front matter © 2013 The Authors. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/! 0.1016/j.geoderma.2013.05.020
Table 1
Geographic and vegetation characteristics of the sampling sites for the revegetation gradient of non-grazed grassland.
Sample site name
Revegetation duration (years)
Elevation (m)
Geographical coordinates
«N), ME)
Slope (°) Dominant species
Accompanying species
Hi.O. Ab.G.
2080 2078
36°15.143' 106°23.204' 10°
36°15.223' 106°22.935' 14°
36°15.751' 106°23.415' 17°
36°11.957' 106°24.592' 12°
36°15.807' 106°23.238' 10°
36°15.807' 106°23.226' 6°
Stipa grandis
Stipa bungeana
Artemisia sacrorum Ledeb Stipa grandis P. Smirn
Thymus mongolicus Ronn. Stipa bungeana Trin Ledeb
Hierochloe ordorata Leymus secalinus Leymus secalinus (Georgi) Tzvel.; Hierochloe ordorata (Linn.) Beauv.
Artemisia frigida Willd.
Potentilla acaulis L.
Medicago ruthenica
Potentilla angustiloba
Artemisia sacrorum
Thymus mongolicus
Leymus secalinus
Potentilla bifurca
Stipa bungeana
Heteropappus altaicus
Thymus mongolicus
Carex rigescens (Franch.) V. Krecz
Medicago ruthenica (Linn.) Trautv
Astragalus scaberrimus Bunge
Artemisia scoparia
Thymus mongolicus
Artemisia scoparia
Thymus mongolicus
Potentilla bifurca
St.G.: Stipa grandis P. Smirn (56 years); StB.: Stipa bungeana Trin Ledeb (36 years); At.S.: Artemisia sacrorum Ledeb (25 years); Th.M.: Thymus mongolicus Ronn. (15 years); Hi.O.: Hierochloe ordorata Beauv (7 years); Ab.G.: recently abandoned grazing on grassland (3 years).
water infiltration and soil erosion (Amezketa, 1999; Le Bissonnais et al., 2007). Specifically, aggregate stability impacts the movement and storage of water in soils, soil aeration, soil erosion, biological activity, and crop growth (Zhang and Miller, 1996). Thus, aggregate stability affects a wide range of physical and biogeochemical processes in natural and agricultural environments. Maintaining high soil aggregate stability is essential for preserving soil productivity and for minimizing soil erosion and environmental pollution that result from soil degradation. Arshad and Cohen (1992) proposed that aggregate stability is one of the physical soil properties that can serve as a soil quality indicator. Hortensius and Welling (1996) included aggregate stability in the international standardization of soil quality measurements. Aggregate stability is also used as an indicator of soil structure (Six et al., 2000).
Human activities have impacted Chinese ecosystems for millennia (Fu et al., 1999, 2002). In the last century, fragmentation and degradation of ecological environments accelerated due to increasing population pressure. To prevent further deterioration of natural ecosystems, the Chinese government has launched a series of nationwide conservation projects. These projects, such as the 'Grain for Green' project, focus on the rehabilitation and recovery of damaged ecosystems (McVicar et al., 2007; Stokes, et al., 2010). The impacts of human activity at the Loess Plateau can be attributed to continuous and widespread stress, such as over-grazing and large-scale monocultures (wheat and maize) (Fu et al., 2000). In this area, soil erosion is a major threat. One of the most urgent tasks for achieving sustainable agricultural development is the recovery of natural vegetation (Li et al., 2005). Among other impacts, revegetation could increase aggregate stability and decrease soil erodibility. Soil water-stable aggregation research has been conducted at the Loess Plateau in relation to soil erosion (for example, Wang etal., 1994; Zhaetal., 1992). According
to this research, soil water-stable aggregation best reflects the ability of soils to resist erosion at the Loess Plateau. However, few papers are available regarding the evolution of soil water-stable aggregation during natural revegetation (e.g., Guo et al., 2010). Hence, the effect of revegetation on soil aggregate stability needs to be confirmed. At the Loess Plateau, two types of revegetation practices have been implemented: non-grazed grassland and afforestation, but no comparison between these revegetation types have been carried out yet.
Numerous research studies regarding soil aggregate stability on the Loess Plateau have been conducted (for example, Gao, 1991; Zhu, 1982). While many methods have been designed to measure aggregate stability, most of the results from the Loess Plateau have been obtained using a method based on Yoder (1936). These studies showed that the fraction of soil aggregates >0.25 mm is strongly related to soil erodibility (Gao, 1991; Zhu, 1982). Wet sieving by Yoder's method involves several mechanisms of soil aggregate breakdown (Le Bissonnais, 1996). Other methods, such as that of Le Bissonnais (1996), have been designed to separate the mechanisms (such as slaking and differential clay swelling) and to provide more information regarding soil aggregation. The Le Bissonnais (1996) method recently became an ISO standard (ISO/DIS, 10930, 2012). Thus, it is important to compare the capabilities of the Le Bissonnais (1996) method with those based on the Yoder (1936) method.
The objectives of this study were to (1) investigate the changes in soil aggregate stability during vegetation rehabilitation in the Loess Plateau of China, (2) assess which revegetation practices improve soil aggregate stability the most, and (3) compare the suitability of the Le Bissonnais' method with that of the modified Yoder's method for the Loess Plateau soils.
Table 2
General situation of selected plots for the revegetation gradient of afforestation.
Plot name Revegetation duration (years) Elevation (m) Slope (°) Topography Dominant species Soil type
C.K.26 26 1670 25 Middle slope C. korshinski Typic loessic orthic primosols
C.K.16 16 1670 18 Middle slope C. korshinski Typic loessic orthic primosols
A.G.3 3 1675 16 Upper slope land Stipa bungeana, Artemisia sacrorum Ledeb, Typic loessic orthic primosols
Thymus mongolicus Ronn.
Cr. 0 1620 <5 Slope farmland Wheat Typic loessic orthic primosols
Table 3
Soil physical and chemical properties for the revegetation gradient of non-grazed grassland.
Plant Depth (cm) Bulk density OM Total N NO3-- N NH+- -N Available P Available K
community (g/cm3) (g • kg-1) (g • kg-1) (mg • kg-1) (mg • kg-1) (mg • kg-1) (mg • kg-1)
StG. 0-20 0.90 32.18 3.44 18.14 4.89 4.06 3.44
20-40 1.12 25.55 2.90 15.68 4.41 3.21 2.90
St.B. 0-20 0.91 27.83 3.13 16.77 3.95 3.53 3.13
20-40 1.14 21.22 2.65 13.70 4.80 2.73 2.65
At.S. 0-20 1.22 23.57 2.61 14.13 6.78 3.65 2.61
20-40 1.16 21.30 2.27 17.20 6.08 3.02 2.27
Th.M. 0-20 0.92 29.69 3.17 14.31 6.91 4.37 3.17
20-40 1.43 24.33 2.74 19.83 5.03 3.25 2.74
Hi.O. 0-20 1.23 20.31 2.19 10.25 5.01 3.37 2.19
20-40 1.16 17.44 2.03 8.86 4.25 2.53 2.03
Ab.G. 0-20 1.13 9.57 1.22 14.40 4.10 2.27 1.22
20-40 1.16 9.19 1.24 11.93 3.65 2.34 1.24
St.G.: Stipa grandis P. Smirn (56 years); St.B.: Stipa bungeana Trin Ledeb (36 years); At.S.: Artemisia sacrorum Ledeb (25 years); Th.M.: Thymus mongolicus Ronn. (15 years); Hi.O.: Hierochloe ordorata Beauv (7 years); Ab.G.: recently abandoned grazing on grassland (3 years).
2. Materials and methods
2.1. The study sites
The study sites were located within the Loess Plateau of China near the city of Yuanzhou (Ningxia province). In this area, 90% of the land area is hilly, 4% is occupied by villages and rivers, and only 6% is considered suitable for intensive agriculture. This area has a sub-arid climate that is characterized by heavy seasonal rainfall with periodic floods and droughts. The average annual rainfall at the observatory is 400 mm (1941-2000; C.V. 18%), and the area has distinct wet and dry seasons. The rainy season starts in July and continues until September. The July rainfall accounts for 24% of the annual rainfall. The mean annual temperature is approximately 7 °C. Most of the land is at an altitude of between 1800 and 2040 m and is frequently dissected by steep and very steep gullies.
Two types of revegetation areas were selected: non-grazed grassland and afforestation. Altogether, ten sites were chosen. The non-grazed grassland area is located within the Yunwu Observatory for Vegetation Protection and Eco-environment (latitude 36° 13' N-36° 19' N and longitude 106° 24' E-106° 28' E). It is the only remaining non-grazed grassland on the Loess Plateau and is a protected area of approximately 1000 ha that is enclosed by fences. This non-grazed grassland area was formally established in 1982 and has not been grazed for 30 years. However, some sections of the non-grazed grassland have been protected from grazing for as many as 56 years. Based on the natural plant succession process in this area (Zou and Guan, 1997), we studied six sites that had been enclosed for different lengths of time to establish natural vegetation after 1982. The non-grazed grassland was used as grazing land prior to enclosure. The enclosure times were determined based on land use records and farmers' indications. Depending on the time of natural succession, the studied plant communities from the longest to shortest enclosure duration were as follows: (a) Stipa grandis P. Smirn (St.G., 56 years), (b) Stipa bungeana Trin Ledeb (St.B., 36 years), (c) Artemisia sacrorum Ledeb (At.S., 25 years), (d) Thymus mongolicus
Ronn. (Th.M., 15 years), (e) Hierochloe ordorata Beauv (Hi.O., 7 years), and (f) recently abandoned grazing on grassland (Ab.G., 3 years). The basic information about these sites are listed in Table 1.
The afforestation area is located at the Guyuan Observatory for Vegetation Protection and Eco-environment (latitude 35° 59' E-36° 02' E and longitude 106° 26' N-106° 30' N), which is approximately 30 km south of the Yunwu Observatory for Vegetation Protection and Eco-environment. The site is a long-term and comprehensively managed field study site that was established in 1982. Several practical revegetation methods were implemented for soil and water protection, including shrub reconstruction and natural enclosure by a steel fence to protect the area from grazing and human activities. Under afforestation, four communities from the neighboring areas were studied, including (a) a 26-year-old Caragana korshinskii Kom. (C.K.26), (b) a 16-year-old C. korshinskii Kom. (C.K.16), (c) a 3-year-old abandoned grazing land (A.G.3), and (d) a slope cropland (Cr.). The basic information about these sites are listed in Table 2. The slope cropland is currently used by farmers for production (with very low fertilizer input) and represents the field conditions just before afforestation. The basic soil physical and chemical properties for the non-grazed grassland and afforested plant communities are listed in Tables 3 and 4, respectively. Based on data from previously-studied sites, the particle size distribution of the upper 0-20 cm of the soil at the non-grazed grassland area was 173-292 g • kg-1 clay (<0.002 mm), 173-267 g • kg-1 silt (0.20.002 mm), and 462-579 g • kg-1 sand (>0.2 mm). At the afforestation area, the particle size distribution of the upper 0-20 cm of soil was 279-357 g • kg-1 clay (<0.002 mm), 584-630 g • kg-1 silt (0.20.002 mm), and 60-91 g • kg-1 sand (>0.2 mm) (An and Huang, 2003; An et al., 2009).
2.2. Soil sampling and preparation
Soil samples were collected at 0-20 cm and 20-40 cm depths in March 2010. An area of 60 x 60 m2 was delineated at each site.
Table 4
Soil physical and chemical properties for the revegetation gradient of afforestation.
Plant Depth Bulk density OM Total N NO3-- N NH+-N Available P Available K
community (cm) (g/cm3) (g • kg-1) (g • kg-1) (mg • kg-1) (mg • kg-1) (mg • kg-1) (mg • kg-1)
C.K.26 0-20 1.04 15.54 1.59 5.96 5.07 4.23 1.59
20-40 1.14 12.14 1.37 7.67 4.37 3.31 1.37
C.K.16 0-20 1.17 13.97 1.45 7.46 6.02 4.89 1.45
20-40 1.30 8.33 0.90 6.40 6.43 3.56 0.90
A.G.3 0-20 1.28 17.74 1.88 6.29 3.96 3.43 1.88
20-40 1.16 11.81 1.29 5.64 3.99 2.71 1.29
Cr. 0-20 1.30 8.12 0.88 13.03 3.73 9.32 0.88
20-40 1.51 4.16 0.67 8.05 3.68 2.54 0.67
100 80 60 40 20 0
100 80 60 40 20 0
E? 100 80
e 60 o
100 80 60 40 20 0
100 80 60 40 20 0
100 80 60 40 20 0
a St. G. 0-20 cm □ St. G. 20-40 cm
en i ^n i iJ~l i i^n i .^n i sn
en.rf1.il,
.-I. x.
nSt. B. 0-20 cm nSt. B. 20-40 cm
Efl. an, SI. en. ¡n, Efl,
H, rfl, fin, rfl. HI, 81.
en jJ -j
a At. S. 0-20 cm nAt. S. 20-40 cm
1-1-1-1- 11 -r
Hi mil ofl „n m rm ffl sfi m an m tl efl i tt .
aTh. M. 0-20 cm nTh. M. 20-40 cm 1-
1 T ____ ____ ___
eH ¡¡H m rfl nn sn ffl .n an oil ral
a Hi. O. 0-20 cm nHi. O. 20-40 cm 1
en nn in nn en W\ Bfl «n m-i «n an m
oAb. G. 0-20 cm nAb. G. 20-40 cm
n Pfll m Ph n HI m * Hi m f7T\ Ffl 1 r?n 7T\
n n n n n
m ti ti , ti , a a m ti a , ti , ti ti m ti ti , ti , a ti
A 1-2 0.5-1 oo OO oo -0 A 1-2 0.5-1 OO oo OO 50.0 -0 A 1-2 0.5-1 oo oo o o -0
FW test WS test SW test
Treatment and Size
Fig. 1. The soil aggregate fraction distributions from the Le Bissonnais' method for the FW, WS and SW tests for the revegetation gradient of non-grazed grassland. (A) Stipa grandis P. Smirn (56 years), (B) Stipa bungeana Trin Ledeb (36 years), (C) Artemisia sacrorum Ledeb (25 years), (D) Thymus mongolicus Ronn. (15 years), (E) Hierochloe ordorata Beauv. (7 years), and (F) recently abandoned grazing on grassland (3 years). FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. Error bars indicate the standard errors (n = 3).
Within this area, three 20 x 20 m2 plots were selected for sampling. Undisturbed soil samples were taken from each plot. Three soil samples were collected at each depth at each site.
Samples were air-dried in the laboratory at room temperature. Undisturbed bulk soil was used for Yoder's method. Each sample was dry-sieved to remove large roots, stones and the macrofauna. For the Le Bissonnais' aggregate stability method, soil aggregates between 5 and 3 mm were used.
2.3. Soil aggregate stability by the modified Yoder's method
Soil aggregate stability was determined by using the modified Yoder's method (Zhu, 1982) with a set of 5, 2, 1, 0.5, and 0.25 mm sieves. The sieves were rapidly immersed in distilled water and were oscillated for 3 min at a displacement of approximately 4 cm at 37 rounds per minute. All fractions were dried at 70 °C prior to weighing. The mean weighted diameter (MWD) was calculated as
100.0 80.0 60.0 40.0 20.0 -y 0.0
100.0 80.0 60.0 40.0 20.0 0.0
100.0 80.0 60.0 40.0 20.0 0.0
100.0 80.0 60.0 40.0 20.0 0.0
a C.K.26 0-20 cm
□ C.K.26 20-40 cm
.H1.ET1
.n. en. gi.il
, 1-1—1 , r-Pn , rfl ■ I i~l .
hC.K.16 0-20 cm nC.K.16 20-40 cm
.rfi.lJ1.il
.En.Bh.fi.
\ I I I I I I I I I I I r
r-n i-n , i-n fffl cd~l
1 I I I I I I
hA.G. 0-20 cm □ A.G. 20-40 cm
BH sTi ufl sH sn il Eil sr, m-, em Si tl
I I I I I I
I I I I I I I
T^ I | | |
a Cr. 0-20 cm
□ Cr. 20-40 cm
FW test
^ ™ in llilf m m Hfi ll I n tl ll Pi n W\
11 n n n n
m ti ti , ti , ti ti m a a , a , a m m ti ti , ti , ti a
A 1-2 0.5-1 O O o o 0. -0 A 1-2 0.5-1 o o d o oo -0 A 1-2 0.5-1 OO oo O O -0
WS test Treatment and Size
SW test
Fig. 2. Soil aggregate fraction distributions from the Le Bissonnais' method FW, WS and SW tests for the revegetation gradient of afforestation. (A) Caragana korshinskii Kom. (26 years), (B) C. korshinskii Kom. (16 years), (C) abandoned grazing land (3 years), and (D) slope cropland (0 year). FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. Error bars indicate the standard errors (n = 3).
XwjXj, where wj is the mean diameter of size fraction i and Xj is the proportion of the size fraction i in relation to the total sample weight. The summation was performed across all size fractions, including the fraction gone through the 0.25 mm sieve. For a given sample, three replicates were used, which led to nine measurements at a given depth at each site.
2.4. Soil aggregate stability by the Le Bissonnais' method
Stability tests were performed with the Le Bissonnais (1996) method, which is the ISO standard #10930 (ISO/DIS, 10930, 2012). Three-gram samples containing aggregates of 3 to 5 mm were air-dried at 40 °C for 24 h. Three tests were conducted, including fast wetting (FW), slow wetting (SW), and wet stirring (WS). For the FW test, the aggregates were quickly immersed in deionized water. The FW test is sensitive to the slaking process. In contrast, the samples were moistened by placing them above humid foam for the SW test. The SW test determines the aggregate sensitivity for differential clay swelling. For the WS test, the aggregates are immersed in ethanol prior to transferring them to a flask of water and subsequent shaking. The use of ethanol inhibits slaking and differential clay swelling. Thus, the WS test is only sensitive to the mechanical breakdown process.
After applying the FW, SW or WS test, the size distribution was determined by sieving (0.05 mm, 0.1 mm, 0.2 mm, 0.5 mm, 1 mm and 2 mm). For a given soil sample, three replicates were analyzed for a total of nine measurements at a given depth at each site. Mean weighted diameters (MWDs) were computed from the fragment size distributions. As for Yoder's method, the MWD is used as a structural stability index. For example, samples that have greater structural stability have a larger MWD (Le Bissonnais, 1996).
2.5. Statistical analyses
All statistical analyses were conducted in SPSS 11.0. A one-way ANOVA followed by the Newman-Keuls test (P < 0.05) was used to compare differences among plant communities and land uses at depths of 0-20 cm and 20-40 cm.
3. Results
3.1. Distribution of soil aggregate fractions for different plant communities
3.1.1. The modified Yoder's method
The results from the wet sieving method for soil aggregate stability at the non-grazed grassland communities are shown in Fig. 5 for the 0-20 and 20-40 cm soil depths. The main aggregate distribution
A 3.5 1 3.02.5-Q 2.0-S 1.51.00.50.0-B 3.5 ,
3.02.5-Q 2.0-S 1.51.00.5-
0.0 C 4.0-,
3.53.02.5-
& 2.0-S
1.51.00.50.0-
Ш FW test 0-20 cm a
□ FW test 20-40 cm
E3 WStest 0-20 cm
A WS test 20-40 cm
IS SW test 0-20 cm
□ SW test 20-40 cm
a A £
St. S.
St. B. At. S. T0. M. Hi. O. Ab. G. Plant Communities
Fig. 3. MWD of the Le Bissonnais' method FW, WS and SW tests for the revegetation gradient of non-grazed grassland. Means with the same letter in different rows are not significantly different at a 0.05 confidence level (Newman-Keuls). Lower case letters (for example, a, b, and c) are used for the 20 cm depth, and upper case letters (for example, A, B, and C) are used for the 20-40 cm depth. MWD: mean weighted diameter; FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. St.G.: Stipa grandis P. Smirn (56 years); St.B.: Stipa bungeana Trin Ledeb (36 years); AtS.: Artemisia sacrorum Ledeb (25 years); Th.M.: Thymus mongolicus Ronn. (15 years); Hi.O.: Hierochloe ordorata Beauv. (7 years); Ab.G.: recently abandoned grazing on grassland (3 years). Error bars indicate the standard errors (n = 3).
3.5 и 3.02.5-Q 2.0-S 1.51.00.50.0-
3.5 -| 3.0 -2.5 -Q 2.0-S 1.51.0 -0.5 -0.0 -
3.5 -| 3.0 -2.5 -Q 2.0-S 1.5 -1.0 -0.5 -0.0 -
a FW test 0-20 cm
□ FW test 20-40 cm
0 WS test 0-20 cm
□ WS test 20-40 cm
a SW test 0-20 cm a
□ SW test 20-40 cm a
C.K.26
C.K. 16
Plant Communities
Fig. 4. MWD of the Le Bissonnais' method FW, WS and SW tests for the revegetation gradient of afforestation. Means with the same letter in different rows are not significantly different at a 0.05 confidence level (Newman-Keuls). Lower case letters (for example, a, b, and c) are used for the 0-20 cm depth, and upper case letters (for example, A, B, and C) are used for the 20-40 cm depth. MWD: mean weighted diameter; FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. C.K.26: Caragana korshinskii Kom. (26 years); C.K.16: C. korshinskii Kom. (16 years); A.G.: abandoned grazing land (3 years); Cr.: slope cropland (0 year). Error bars indicate the standard errors (n = 3).
fractions were always >5 mm and <0.25 mm, and the sum of these fractions accounted for between 50% and 80% of the total weight. The 0.5 mm to 0.25 mm fraction was always the lowest and only accounted for between 3 and 10% of the total weight. Except for Hi.O. and Ab.G., the other fractions decreased in the following order: 5-2 mm > 1-0.5 mm > 2-1 mm > 0.5-0.25 mm. The non-grazed grassland communities had a larger > 5 mm fraction at the 20-40 cm soil depth than at the 0-20 cm soil depth.
The distribution of soil aggregate fractions for the different afforestation types is shown in Fig. 6. The main fractions were >5 mm and <0.25 mm (for example, in the non-grazed grassland communities),
and their sum accounted for between 60 and 80% of the total weight. The size of the lowest fraction was between 0.5 and 0.25 mm in most cases, which accounted for between 2% and 8% of the total weight. Finally, the afforested land uses (for example, C.K.26, C.K.16 and A.G.3) had a higher proportion of > 5 mm aggregates (approximately 30% to 50%) than the cropland soil (lower than 20%).
3.1.2. The Le Bissonnais' Method
The soil aggregate stability results from the Le Bissonnais' method are shown in Fig. 1 for the non-grazed grassland communities. For the FW test, the mode of the size distribution was always larger than
Fig. 5. The soil aggregate fraction distributions from the Modified Yoder's wet sieving method for the revegetation gradient of non-grazed grassland. (A) Stipa grandis P. Smirn (56years), (B) Stipa bungeana Trin Ledeb (36 years), (C) Artemisia sacrorum Ledeb (25 years), (D) Thymus mongolicus Ronn. (15 years), (E) Hierochloe ordorata Beauv. (7 years), and (F) recently abandoned grazing on grassland (3 years). FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. Error bars indicate the standard errors (n = 3). Error bars indicate the standard errors (n = 3).
S 40 a
□ St. G. 0-20 cm □ St. G. 20-40 cm
20----[
> 5 mm 5-2 i
2-1 mm | 1-0.5 mm | 0.5-0.25 mm| 0-0.25 mm mm
Treatment and Size
□ St. B. 0-20cm
□ St. B. 20-40 cm
¡3 40
> 5 mm
5-2 mm
2-1 mm | 1-0.5 mm | 0.5-0.25 mm | 0-0.25 mm Wet sieving Treatment and Size
> 5 mm | 5-2 mm | 2-1 mm | 1-0.5 mm | 0.5-0.25 mm | 0-0.25 mm
Wet sieving Treatment and Size
8 20 S 0
□ Hi. O. 0-20 cm □ Hi. O. 20-40 c
F^r^i V7M~\ mrn
2-1 mm | 1-0.5 mm | 0.5-0.25 mm | 0-0.25 mm
Wet sieving Treatment and Size
□ Ab. G. 0-20 cm □ Ab. G. 20-40 cm
1-0.5 m
Wet sieving Treatment and Size
0.5-0.25 mm
£ 60 sä
1 40 о
^ 20 ад 20 u
ад ад < 0
<3 40 ta
□ C.K.26 0-20 cm □ C.K.26 20-40 cm
УМ>Л T I_,_W7//Á ^ I_,_мина - i
oC.K.16 0-20 cm
□ C.K.16 20-40 cm
У&УЛ T I_,_8ШI " I
3 Cr. 0-20 cm
□ Cr. 20-40 cm
и A.G. 0-20 cm □ A.G. 20-40 cm
iU III Т I "" i Ж
1 1 1 1 1 1
> 5 mm
2-1 mm
Wet sieving Treatment and Size
Fig. 6. Soil aggregate fraction distributions from Modified Yoder's wet sieving method for the revegetation gradient of afforestation. (A) Caragana korshinskii Kom. (26 years), (B) C. korshinskii Kom. (16 years), (C) abandoned grazing land (3 years), and (D) slope cropland (0 year).FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. Error bars indicate the standard errors (n = 3).
2 mm. This fraction accounted for approximately 40% to 80% of the total weight in all but the recently abandoned grazing on grassland (Fig. 1F) (approximately 20%). There were obvious differences between the soils from the recently abandoned grazing on grassland and the other non-grazed grasslands that had been recovered for up to 56 years. The size fractions were evenly distributed in the recently abandoned grazing on grassland soils. The greatest content was approximately 20%, which was similar to the other fractions. The WS test resulted in distributions similar to those found by the FW test. For the SW test, all the soils had a >2 mm fraction that was close to or above 80%. In most cases, the 0-20 cm samples had coarser fragments than the 20-40 cm samples within a given land use.
The soil aggregate fraction distributions for the different types of afforestation are shown in Fig. 2. For the FW test, the main size fraction was >2 mm, which accounted for approximately 40% of the C.K.26, C.K.16 and A.G.3 fragments. The size distribution of the FW test in the Cr. community was close to 0% for the >2 mm fraction and more than 80% for the 0.2 mm fraction (for the 0-20 cm and the 20-40 cm depths). The WS test showed contrasting distributions that depended on the type of afforestation. The A.G.3 had a mode of >2 mm, which accounted for more than 40% of the total sample weight. In contrast, the mode for Cr was <0.05 mm (approximately 30% of the total sample weight). The C.K.26 and C.K.16 types had intermediate size distributions, with the >2 and <0.05 mm fractions accounting for approximately 20% of the total sample weight. For
the SW test, all the afforested soils had high >2 mm fragment contents (approximately 80%) compared with the cropland soils (approximately 20%).
3.2. The MWD for different plant communities
3.2.1. The modified Yoder's method
The MWD values that were determined by the modified Yoder's method are shown in Fig. 7. Among the non-grazed grasslands (Fig. 7A), the highest MWD value at the 0-20 cm soil depth was approximately 3 mm. For the 0-20 cm depth, the MWD values of At.S. and Ab.G. were significantly higher than the other MWD values. The MWD values at the 0-20 cm depth were lower than the corresponding MWD values from the 20-40 cm depth (except for the Hi.O community). At the 20-40 cm depth, the MWD was between 2.0 and 2.8 mm. In addition, except for the recently abandoned grazing on grassland, there were no significant differences between the non-grazed grasslands.
In the afforestation areas (Fig. 7B) at a depth of 0-20 cm, A.G.3 was the most stable (its MWD was the highest), C.K.26 and C.K.16 had statistically identical MWD values, and Cr had significantly lower MWD values than the afforested areas. At the 20-40 cm depth, no significant differences in the MWD values were found among the C.K.26, C.K.16 and A.G.3 communities. However, these plant communities are significantly more stable than the Cr. plant community.
3.2.2. The Le Bissonnais' Method
The MWD by the Le Bissonnais' method showed differences between tests and land uses (Fig. 3). For the FW test, all the plant communities were significantly different from the recently abandoned grazing on grassland (Ab.G.) at the 0-20 and 20-40 cm depths. The maximum MWD was approximately 3 mm for the At. S. community at the 0-20 cm depth, and the minimum MWD was approximately 1 mm for the recently abandoned grazing on grassland at the 20-40 cm depth. In addition, there were differences between the plant communities compared with the recently abandoned grazing on grassland. A similar trend resulted from the WS test. For the SW test, the plant communities were significantly different from the recently abandoned grazing on grassland at the 0-20 and 20-40 cm depths (P < 0.01). The maximum MWD was approximately 3.2 mm for the At.S. community at the 0-20 cm depth, and its minimum was approximately 2.7 mm for the recently abandoned grazing on grassland at the 20-40 cm depth. There was no significant difference between the plant communities and the Ab.G. community.
The MWDs at the afforestation sites were lower than that at the non-grazed grassland sites (Figs. 3 and 4). The maximum MWD value from the FW test was approximately 1.8 mm (Fig. 4A). In addition, no significant differences were found among the C.K.26, C.K.16 and A.G.3 communities at the 0-20 cm depth. At the same depth, the MWD was significantly lower for cropland, which had a MWD of <0.5 mm. For the WS test, the MWD was the highest for the A.G.3 community at the 0-20 cm and 20-40 cm depths. For the SW test, the MWD values
3.5-1 3.02.5-Q 2.0-
s 1.51.00.50.0-
0 W. S. 0-20 cm □ W. S. 20-40 cm
St. G. St. B. At. S. Th. M. Hi. O. Ab. G. Plant Communities
0 W. S. 0-20 cm b
□ W. S. 20-40 cm a
C.K.26
C.K. 16 A.G.
Plant Communities
Fig. 7. MWD of the Modified Yoder's wet sieving method for the revegetation gradient of non-grazed grassland. Means with the same letter in different rows are not significantly different at a 0.05 confidence level (Newman-Keuls). Lower case letters (for example, a, b, and c) are used for the 0-20 cm depth, and upper case letters (for example, A, B, and C) are used for the 20-40 cm depth. MWD: mean weighted diameter; FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. St.G.: Stipa grandis P. Smirn (56 years); St.B.: Stipa bungeana Trin Ledeb (36 years); At.S.: Artemisia sacrorum Ledeb (25 years); Th.M.: Thymus mongolicus Ronn. (15 years); Hi.O.: Hierochloe ordorata Beauv. (7 years); Ab.G.: recently abandoned grazing on grassland (3 years). Error bars indicate the standard errors (n = 3).
were much higher than those for the WS and FW tests. For example, the MWD values were approximately 3 mm for the A.G.3, C.K.16, and C.K.26 communities, which is two times the FW test MWD value for cropland. However, the SW test results were similar to the WS and FW test results. For example, the Cr. land use had a significantly lower MWD than all the other land uses, regardless of depth.
4. Discussion
4.1. Effect of revegetation on soil aggregation and aggregate stability
Cropland on the Loess Plateau is characterized by continuous plowing and low organic and inorganic fertilizer inputs. The conversion of forestland to cropland, grazing land, or settlements has often resulted in soil degradation and nutrient losses (Dinesh et al., 2003). The soil erosion on the Loess Plateau is mainly caused by poor land use practices. Appropriate land use, such as afforestation and natural succession of non-grazed grassland on eroded areas, could be used to mitigate soil erosion. Such practices are reflected in the recent ecosystem reconstruction policy in Northwest China.
The natural soil structure is destroyed by plowing. Moreover, the stabilizing effects of root fibers become insignificant when they are shredded by tillage and decompose after harvesting. Because pore space increases due to mechanical cultivation, the air exchange and oxygen availability increases. This increase in oxygen availability enhances the microbial decomposition of organic matter. When coupled with accelerated soil erosion, enhanced decomposition rapidly depletes the soil organic matter content in the plow layer and reduces the aggregate stability (Zhang and Horn, 2001). Furthermore, the stability of non-cropped soil aggregates may benefit from hydrophobic-ity of soil organic carbon.
In the present study, the cropland soil had the lowest aggregate stability for all the Le Bissonnais' (FW, SW and WS) and Yoder's methods. This finding is consistent with the occurrence of erosion by water and was observed at many of the experimental cropland sites on the Loess Plateau (Zheng, 2006). Aggregate stability measurements on the Loess Plateau consistently indicate that the soil water-stable aggregate contents reflect the ability of the soils to resist erosion (An et al., 2009; Wang et al., 1994; Zhaet al., 1992).
The aggregate stability was much higher in the non-grazed grassland soils than in the cropland soil (Figs. 3, 4 and 7). The MWD values were very high for several plant communities. In addition, the natural succession of non-grazed grasslands had a higher MWD value than the afforestation sites, especially for the FW and SW treatments. The MWD values were generally higher for the non-grazed grasslands than for the afforestation sites. Also, the non-grazed grasslands had the highest fraction of large aggregates (Fig. 1,2,5 and 6). These signify a strong soil structure and a high resistance to water erosion. Fattet et al. (2011) got results similar to ours ("herbaceous vegetation was more efficient than trees in improving aggregate stability."). These traits are especially prevalent in the FW treatment, which measures aggregate stability to slaking. In addition, this trend is consistent with our field observations regarding water and soil erosion: no sediment was produced in this area (unpublished).
For the MWD of SW, there is no significant difference among the 3-to-27-year enclosure durations. But, for the FW, the 3-year non-grazed grassland MWD is quite lower than the other MWDs. Overall the aggregate stabilities of non-grazed grassland increase quickly after enclosure as shown by both Le Bissonnais' and Yoder's methods (Fig. 8A, C). After 7 years of non-grazing, an almost-constant aggregate stability is reached for all kinds of stability tests. In the afforestation area, an almost-constant aggregate stability is reached for all kinds of stability tests 3 years after planting (Fig. 8B, D).
Because of the 'Grain for Green' project that was launched in 1999 in the western part of China, the sloped cropland has been almost completely converted to forest and non-grazed grassland since then.
3.5 32.52-
20 30 40
Duration (years)
10 20 Duration (years)
3.5 32.5-
—J ■
Si 1.5 = ? S
1 -0.50
-■-+■- 0-20 cm —I-20-40 cm
...-(-.. 0-20 cm
—1-20-40 cm
0 10 20 30 40
Duration (years)
0 10 20 Duration (years)
---0-- SW 0-20 cm
SW 20-40 cm
FW 0-20 cm
FW 20-40 cm
■ - -A- ■ WS 0-20 cm
WS - 20-40 cm
Fig. 8. Time evolution of the MWDs. (A) Le Bissonnais' method for non-grazed grassland, (B) Le Bissonnais' method for afforestation, (C) Yoder's method for non-grazed grassland, and (D) Yoder's method afforestation. MWD: mean weighted diameter; FW: Fast wetting; WS: Wet stirring; SW: Slow wetting. Error bars indicate the standard errors (n = 3).
Our results indicate that these land use changes have significantly decreased soil erodibility. Based on the succession of communities, it was estimated that prohibiting grazing and cropping, by stabilizing soil aggregates, will continue to limit erosion.
From a soil aggregate stability enhancement and soil conservation standpoint, we can conclude that non-grazed grassland is more effective than afforestation, also it may take a few more years for aggregate stability to reach its maximum.
According to long-term monitoring and research studies on Loess Plateau soils (Guo et al., 2010; Jiao et al., 2012; Lv et al., 2012; Zheng, 2006), the vegetation structure and species diversity of afforested sites have improved. In our previous studies (An et al., 2009, 2010), soil microbiodiversities and soil nutrients increased for the first 20 years of enclosure, and then they kept a high level. Soil aggregate stability was probably influenced by many other factors such as soil organic carbon, soil inorganic carbon, and soil clay content. More research related to soil aggregate stability should be carried out to identify the mechanisms of soil aggregate stabilization depending on enclosure duration.
4.2. Comparison of the Yoder's and Le Bissonnais' methods
Yoder's method combines two disaggregation mechanisms: slaking during sample immersion and mechanical breakdown during shaking. It makes it impossible to identify the sensitivity of a soil sample to the disaggregation mechanisms. On the contrary, comparing the MWDs between the three tests defined by Le Bissonnais (1996) help identify the sensitivity of soil aggregates to different disaggregation mechanisms: slaking for the FW test, differential clay swelling for the SW test and mechanical breakdown for the WS test. In the present study, the FW test was the most efficient for disrupting soil aggregates; the WS test was intermediate; and the SW test disrupted soil
aggregates the least. Hence the soil aggregates were highly sensitive to slaking (FW test), less sensitive to mechanical breakdown (WS test) and the least sensitive to differential clay swelling (SW test). The summer rains of the studied area are often very intense. During these rains, the aggregates at the soil surface may undergo fast wetting, which causes high aggregate disruption. This finding potentially explains the high soil erosion rates that are observed in the summer following heavy rains.
We hypothesize that the higher organic matter and carbohydrate contents of the surface layer (0-20 cm) contribute to aggregate stabilization by bridging between the mineral particles. Increased aggregate stability in mull A-horizons was reported in forests by Ma et al. (2005). With regard to improving semiarid soil structure, the soil organic matter concentration is the most important aggregate stability factor (Diaz et al., 1994). Organic matter and its biological origin are the main parameters responsible for aggregate stability. These parameters are usually developed in the rhizosphere. Increasing aggregate stability from microorganisms may be of a physical nature (Tisdall and Oades, 1982) or may be due to the formation and excretion of microbial polysaccharides that act as binding agents (Cheshire et al., 1983).
Some MWD by Yoder's method were higher for the 20-40 cm depth than for the 0-20 cm depth (Fig. 7), suggesting a lower aggregate stability in the surface layer. These differences were not consistent with MWD by Le Bissonnais' method for FW, SW and WS (Fig. 3). The cause of these inconsistencies is unclear.
The MWD of the modified Yoder method was positively correlated to the MWD of the Le Bissonnais' method for SW (r = 0.83, n = 20) and WS (r = 0.87, n = 20) but was not related to that of the FW (r = 0.29, n = 20). This result is similar to the one of Guo et al. (2010), obtained on the Loess Plateau of China too. Le Bissonnais et al. (2007) (obtained on Mediterranean vineyards) and Guo et al.
(2010) also found the Le Bissonnais's method more discriminant than the Yoder's method. As such, the Le Bissonnais' method should be preferred over the Yoder's method while working on revegetation in the Loess Plateau of China.
To the credit of Yoder's (1936) early work, it must be reminded that Le Bissonnais' method was developed more recently and is based, among others, on Yoder (1936). Moreover, most authors of Yoder's method are in fact (1) using a modified version, and (2) the changes are not always reported; making difficult to compare studies. Finally, considering the results of the comparisons between Yoder's and Le Bissonnais' method, and considering that Le Bissonnais' method recently became an ISO standard (ISO/DIS, 10930, 2012), Le Bissonnais' method should be considered as a preferred method.
5. Conclusions
At the Loess Plateau, non-grazed grassland and afforestation reveg-etations were efficient measures for increasing aggregate stability and decreasing soil erodibility. High aggregate stabilities were reached after 3 and 7 years under afforestation and non-grazed grassland, respectively. However, aggregate stability in the non-grazed grassland was higher than that under afforestation. This result was likely due to the higher organic matter content in the non-grazed grassland. The effect of revegetation on aggregate stability is persistent, which makes it a suitable long-term management practice.
The Le Bissonnais (1996) method was used to identify the processes responsible for disaggregation. Compared with the modified Yoder (1936) method, the Le Bissonnais (1996) method was better at determining aggregate stability (especially with the fast wetting test) between land uses at the Loess Plateau.
Acknowledgments
This research was supported by the French Minister of Foreign Affairs through a Hubert Curien grant (PFCC 2009-2010 #20919ZC), the National Natural Sciences Foundation of China (41171226, 41030532), and the Foundation for Youth Teachers by Northwest A&F University. In addition, the authors are grateful to Hervé Gaillard (UR0272, Soil Science Lab., INRA) for his technical assistance.
References
Amézketa, E., 1999. Soil aggregate stability: a review. Journal of Sustainable Agriculture 14 (2-3), 83-151.
An, S.S., Huang, Y.M., 2003. Study on the ameliorate benefits of Caragana korshinskii shrubwood to soil properties in loess hilly area. Scientia Silvae Sinicae 42 (1), 70-74.
An, S.S., Huang, Y.M., Zheng, F.L., 2009. Evaluation of soil microbial indices along a revegetation chronosequence in grassland soils on the Loess Plateau, Northwest China. Applied Soil Ecology 41, 286-292. Arshad, M.A., Cohen, G.M., 1992. Characterization of soil quality: physical and chemical
criteria. American Journal of Alternative Agriculture 7, 25-32. Bronick, C.J., Lal, R., 2005. Soil structure and management: a review. Geoderma 124, 3-22.
Cheshire, M.V., Sparling, G., Mundie, C.M., 1983. Effect of periodate treatment of soil on carbohydrate constituents and soil aggregation. Journal of Soil Science 34,105-112. Diaz, E., Roldan, A., Lax, A., Albaladejo,J., 1994. Formation of stable aggregates in degraded
soil by amendment with urban refuse and peat. Geoderma 63 (3-4), 277-288. Dinesh, R., Ghoshal, Chaudhuri S., Ganeshamurthy, A.N., Dey, C., 2003. Changes in soil microbial indices and their relationships following deforestation and cultivation in wet tropical forests. Applied Soil Ecology 24,17-26. Fattet, M., Fu, Y., Ghestem, M., Ma, W., Foulonneau, M., Nespoulous, J., Le Bissonnais, Y., Stokes, A., 2011. Effects of vegetation type on soil resistance to erosion: relationship between aggregate stability and shear strength. Catena 87, 60-69.
Fu, B.J., Cheng, L.D., Mang, K.M., 1999. The effect of land use change on the regional environment in the Yangjuangou catchment in the Loess Plateau of China. Acta Geographica Sinica 54 (3), 241-246.
Fu, B.J., Chen, L.D., Ma, K.M., 2000. The relationship between land use and soil conditions in the hilly area of Loess Plateau in Northern Shanxi. Catena 39, 69-78.
Fu, B.J., Chen, D.X., Qiu, Y., Wang, J., Meng, Q.H., 2002. Land Use Structure and Ecological Processes in the Loess Hilly Area, China. Commercial Press, Beijing, China (in Chinese with English abstract).
Gao, W.S., 1991. Primary study of classification of different indices of soil anti-erosibility and its adaptation in the hill region of the Loess Plateau. Journal of Soil and Water Conservation 5 (2), 60-65.
Guo, M., Zheng, F.L., An, S.S., Liu, Y., Wang, B., Darboux, F., 2010. Application of Le Bissonnais method to study soil aggregate stability in the hilly-gully region. Science of Soil and Water Conservation 8 (2), 68-73 (In Chinese with English abstract).
Hortensius, D., Welling, R., 1996. International standardization of soil quality measurements. Communications in Soil Science and Plant Analysis 27, 387-402.
ISO/DIS 10930, 2012. Soil Quality - Measurement of the Stability of Soil Aggregates Subjected to the Action of Water. International Organization for Standardization, Geneva, Switzerland (12 pp. http://www.iso.org/iso/catalogue_detail.htm?csnumber= 46433).
Jiao, J.Y., Zhang, Z.G., Bai, W.J., Jia, Y.F., Wang, N., 2012. Assessing the ecological success of restoration by afforestation on the Chinese Loess Plateau. Restoration Ecology 20 (2), 240-249.
Le Bissonnais, Y., 1996. Aggregate stability and assessment of soil crustability and erodibility: I. Theory and methodology. European Journal of Soil Science 47,425-431.
Le Bissonnais, Y., Blavet, D., De Noni, G., Laurent, J.Y., Asseline, J., Chenu, C., 2007. Erod-ibility of Mediterranean vineyard soils: relevant aggregate stability methods and significant soil variables. European Journal of Soil Science 58, 188-195.
Li, Y.Y., Shao, M.A., Zheng, J.Y., Zhang, X.C., 2005. Spatial-temporal changes of soil organic carbon during vegetation recovery at Ziwuling, China. Pedosphere 15 (5), 601-610.
Lv, Y.H., Fu, B.J., Feng, X.M., Zeng, Y., Liu, Y., Chang, R.Y., Sun, G., Wu, B.F., 2012. A policy-driven large scale ecological restoration: quantifying ecosystem services changes in the Loess Plateau of China. PLoS One 7 (2), e31782.
Ma, X.H., Jiao, J.Y., Bai, W.J., 2005. The changes of soil water-stable aggregates in abandoned lands during vegetation restoration in hilly and gully regions on the Loess Plateau. Agricultural Research in the Arid Areas 23 (3), 69-74 (in Chinese with English abstract).
McVicar, T.R., Li, L.T., Van Niel, T.G., Zhang, L., Li, R., Yang, Q.K., Zhang, X.P., Mu, X.M., Wen, Z.M., Liu, W.Z., Zhao, Y.A., Liu, Z.H., Gao, P., 2007. Developing a decision support tool for China's re-vegetation program: simulating regional impacts of afforestation on average annual streamflow in the Loess Plateau. Forest Ecology and Management 251, 65-81.
Peng, X.H., Horn, R., Zhang, B., Zhao, Q.G., 2004. Mechanisms of soil vulnerability to compaction of homogenized and recompacted Ultisols. Soil and Tillage Research 76 (2), 125-137.
Six, J., Elliott, E.T., Paustian, K., 2000. Soil macroaggregate turnover and microaggregate formation: a mechanism for C sequestration under no-tillage agriculture. Soil Biology and Biochemistry 32, 2099-2103.
Stokes, A., Sotir, R., Chen, W., Ghestem, M., 2010. Soil bio- and eco-engineering in China: past experience and future priorities. Ecological Engineering 36, 247-257.
Tisdall, J.M., Oades, J.M., 1982. Organic-matter and water-stable aggregates in soils. Journal of Soil Science 33 (2), 141-163.
Wang, Y.M., Guo, P.C., Gao, W.S., 1994. A study on soil antierodibility in Loess Plateau. Journal of Soil and Water Conservation 8 (4), 11 -16 (in Chinese with English abstract).
Yoder, R.E., 1936. A direct method of aggregate analysis of soils and a study of the physical nature of erosion losses. Journal of the American Society of Agronomy 28, 337-351.
Zha, X., Tang, K.L., Zhang, K.L., 1992. The impact of vegetation on soil characteristics and soil erosion. Journal of Soil and Water Conservation 6 (2), 52-58 (in Chinese with English abstract).
Zhang, B., Horn, R., 2001. Mechanisms of aggregate stabilization in Ultisols from subtropical China. Geoderma 99,123-145.
Zhang, X.C., Miller, W.P., 1996. Physical and chemical crusting processes affecting runoff and erosion in furrows. Soil Science Society of America Journal 60, 860-865.
Zheng, F.L., 2006. Effect of vegetation changes on soil erosion on the Loess Plateau. Pedosphere 16 (4), 420-427.
Zhu, X.M., 1982. Soil and Agriculture in Loess Plateau (in Chinese). Agricultural Press, Beijing.
Zou, H.Y., Guan, X.Q., 1997. The explored of the management of grassland natural conservation area in Yunwu mountain. Prataculture Science 14 (1), 3-4 (in Chinese with English abstract).
