A

lowaisaipb

K

1

abbrgSuoae2t

1

d

Available online at www.sciencedirect.com

Particuology 6 (2008) 157–166

Effects of land use patterns on soil aggregate stability inSichuan Basin, China

Zhen Zhang a, Chaofu Wei a,∗, Deti Xie a, Ming Gao a, Xibai Zeng b

a College of Resources and Environment, Southwest University, Chongqing 400716, Chinab lnstitute of Environment and Sustainable Development for Agriculture, Chinese Academy of

Agricultural Sciences, Beijing 100081, China

Received 6 November 2007; accepted 7 March 2008

bstract

Soil aggregate stability as a key indicator of soil structure, is a product of interactions between soil environment, management practices, andand use patterns. The objective of this study was to analyze the impact of various land use patterns on soil aggregate stability in Sichuan Basinf southwestern China. The dry- and water-stable aggregate size distributions were determined by manual dry sieving procedure and Yoder’set sieving procedure, respectively, while microaggregates and its mechanical and chemical stabilities by Kachisky’s method, oscillator method,

nd citrate-dithionate (C-D) reagent method, separately. The results indicated that fractal dimension and surface fractal dimension were usefulndicators to reflect soil aggregate distribution. Land use patterns have an obvious influence on soil aggregate stability. In the study area, watertability, mechanical stability, and chemical stability followed the sequence, Barren land > forestland > orchard > cropland, and the original stabilitynd collapse velocity were sensitive to soil properties and soil structure. The difference of aggregate stability under different land use patterns

s mainly due to the intensity of human disturbance and cultivation. Improper land use patterns will lead to breakdown of unstable aggregates,roducing finer and more-easily transportable particles and microaggregates. In the future, inappropriate cultivation and land use patterns shoulde changed to protect soil structure, to improve soil aggregate stability and soil fertility in Sichuan Basin.

© 2008 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. Allrights reserved.

ion; L

uhgvc2marTms

eywords: Aggregate size distribution; Soil aggregate stability; Fractal dimens

. Introduction

Soil structure consists of an aggregate formed by therrangement of soil particles, and depends on the interactionsetween primary particles and organic constituents to form sta-le aggregates (Caravaca, Lax, & Albaladejo, 2004). Thus, theecognition of soil aggregate size distribution and soil aggre-ate stability is important to properly interpret soil structure.oil aggregation is a complex phenomenon, which is a prod-ct of interactions of the soil microbial community, mineral andrganic compositions and is influenced by many factors suchs soil environment, management practices, land use patterns,

tc. (Seybold & Herrick, 2001; Wei, Ni, Gao, Xie, & Hasegawa,006). Recent studies mainly concentrate on analyzing the rela-ionship between soil organic matter and soil aggregate stability

∗ Corresponding author.E-mail address: weicf@swu.edu.cn (C. Wei).

sTsoci

674-2001/$ – see inside back cover © 2008 Chinese Society of Particuology and Institute of Process

oi:10.1016/j.partic.2008.03.001

and use pattern; Purple soil; Sichuan Basin; China

nder different cultivations. Land use and landscape positionave an interactive effect on water-stable aggregates and aggre-ate carbon concentration (Natalia & Nicholas, 2005), and highariability of aggregate stability and soil C has important impli-ations for C sequestration (Bird, Herrick, Wander, & Wright,002). Meanwhile, long-term soil cultivation increases organicatter turnover due to differences in the amount of aggregation

nd aggregate turnover, and aggregates in tilled soils cycle moreapidly in a cultivation loop (Six, Elliott, & Paustian, 2000).hese studies focus on soil water-stability aggregates and soilaterials, but little attention is paid to mechanics and chemical

tability.Purple soils are mainly distributed in Sichuan Basin of

outhwestern China, and are classified as Regosols in FAOaxonomy or Entisols in USDA Taxonomy (He, 2003). Purple

oils, which typically developed by the physical weatheringf red or purple parent materials or rocks, inherit many of theharacteristics of parent materials or rocks. But other changesn soil properties have taken place as a result of land use change,

Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.

1 cuology 6 (2008) 157–166

aXwds

2

2

oaaaass

pllpadv(dt(at2aori(mu

2

2

(1gTwosn0mwt

prop

ertie

sof

the

soils

inth

isst

udy

Roc

kty

peO

rgan

icm

atte

r(g

/kg)

pHC

lay

(g/k

g)So

ilpa

rtic

leco

mpo

sitio

n(g

/kg)

1–0.

25m

m0.

25–0

.05

mm

0.05

–0.0

1m

m0.

01–0

.005

mm

0.00

5–0.

001

mm

<0.

001

mm

T1f

18.9

37.

4842

.132

8.2

192.

810

5.3

184.

214

7.4

42.1

J 3p

11.1

58.

1041

.533

8.8

112.

312

7.3

169.

819

7.1

54.7

J 3s

12.2

87.

8831

.733

1.6

73.6

148.

717

4.4

239.

931

.8J 2

s12

.33

6.72

146.

531

022

.914

8.6

194.

929

8.0

25.6

T1f

22.3

97.

1425

4.7

22.6

210.

133

1.8

165.

919

8.1

71.5

J 3p

14.0

17.

8062

.321

.414

3.4

313.

218

7.0

252.

382

.7J 3

s19

.96

7.88

63.4

10.8

173.

034

4.9

176.

422

1.7

73.2

J 2s

25.9

96.

8712

6.5

16.3

206.

731

5.0

136.

525

1.5

74.0

T1f

29.8

7.63

201.

853

.714

2.8

338.

314

8.0

235.

581

.7J 3

p23

.77

7.79

63.2

37.0

202.

427

4.7

190.

220

2.3

93.4

J 3s

32.7

97.

7714

1.2

27.8

233.

524

9.9

119.

522

8.1

141.

2J 2

s33

.70

6.22

200.

451

.024

3.3

299.

411

7.6

203.

285

.5

T1f

27.1

27.

1495

.618

.823

2.6

292.

918

3.7

115.

515

6.5

J 3p

22.9

18.

0084

.022

.627

1.2

284.

612

6.5

158.

613

6.5

J 3s

31.4

77.

8885

.529

.717

2.3

295.

312

3.8

252.

512

6.4

J 2s

21.2

87.

5962

.535

.423

5.4

305.

013

5.0

166.

712

2.5

58 Z. Zhang et al. / Parti

gricultural practice, or eco-environment disturbances (Wei,ie, & Che, 1989; He, 2003). The objectives of the present workere (a) to analyze the effects of land use on the soil aggregateistribution, and (b) to discuss the effects of land use on watertability, mechanics and chemical stability of soil aggregates.

. Materials and methods

.1. Sites, soils and sampling

Sichuan Basin is located in southwestern China with an areaf 165,000 km2 and an elevation varying from 200 to 500 mbove sea level. The climate is subtropical humid monsoon withn average annual precipitation of 1000–1200 mm and the aver-ge annual temperature is 14–19 ◦C. This basin is well knowns the “Red Basin” due to being covered by a red or purple rockeries of the Trias-Cretaceous system, from which the purpleoils are developed and formed (He, 2003).

A stratified sampling design, based on land use patterns andrincipal rock types, was used to collect the soil samples. Fourand use patterns, namely cropland, orchard, forestland, barrenand were selected. Cropland means land covered with tem-orary crops followed by a bare soil period after harvesting,nd lands dominated by fruit trees were orchard. Barren landenoted lands with exposed rock and never had more than 5%egetated cover and forestland was dominated by woody plantsLiu, Zhuang, Luo, & Xiao, 2003). Rock types studied includeark purple shale of Feixianguan Formation of the Trias sys-em (T1f), brown purple sandy mudstone of Penglai FormationJ3p), red brown purple mudstone of Suining Formation (J3s),nd gray brown purple sandy mudstone of Shaximiao Forma-ion (J2s) of the Jurassic system (Wei, Gao, Shao, Xie, & Pan,006). In each sampling area, soil samples were taken randomlyt 0–20 cm depth with five replicates in July 2004, and eachf the replicates was 5 m apart from each other. The land use,ock types, and basic properties of those tested soils are givenn Table 1. Soil pH was determined by using a glass electrode1:2.5, soil:water) and soil particle composition by the pipetteethod (Dane & Topp, 2002). Organic matter was measured

sing the Walkley–Black method (PCARRD, 1980).

.2. Methods

.2.1. Soil macroaggregate size distribution and stabilitySoil aggregates were broadly classified as macroaggregates

>0.25 mm) and microaggregates (<0.25 mm) (Oades & Water,991). A common method used for describing soil macroag-regate distribution was manual dry sieving method (Dane &opp, 2002). After air-drying of soil samples, stones and litterith size more than 2 mm were removed. Air-dried soil samplesf 500 g were placed on the uppermost of a set of graduatedieves 20 cm in diameter and 5 cm in height, on the top of aest of seven sieves with openings of 10, 5, 3, 2, 1, 0.5, and

.25 mm. The sieves were then oscillated vertically and rhyth-ically by hand for 5 min. Aggregates remaining in each sieveere collected and weighed to get macroaggregate size distribu-

ion, i.e. aggregates of 10–5, 5–3, 3–2, 2–1, 1–0.5, 0.5–0.25, and Tabl

e1

Sele

cted

basi

c

Lan

dus

e

Cro

plan

d

Orc

hard

Fore

stla

nd

Bar

ren

land

cuolo

<dorbAd

N

wkdufTe

A

wuMas

s

V

wsabaif

W

wδ

fdb

l

pd

W

wpwaf

l

m

K

(

P

2

a(mbwosa1m0s&dstpstt

A

Z. Zhang et al. / Parti

0.25 mm in diameter. Fractal dimension D and surface fractalimension P were then obtained to describe the characteristicsf macroaggregate size distribution. Higher value of D or P cor-esponds to more fragmentation of soil aggregate and it coulde explained with intensity of human action on soil structure.s suggested by Mandelbrot (1982), the fractal dimension isefined as

(r) = kr−D, (1)

here, N(r) is the number of the elements of length equal to r,is the number of initiators of unit length, and D is the fractalimension. The value of D depends on the shape of individ-al objects within the distribution, and the extent of aggregateragmentation (Pirmoradian, Sepaskhah, & Hajabbasi, 2005).he surface fractal dimension, P, is determined from thequation

(r) = Cr2−P, (2)

here A(r) is the surface area which would be measuredsing a gauge r, C is a constant (Xu, Sun, & Yao, 2004;andelbrot, 1982). For a regular and smooth surface, P = 2, withhigher P value indicating a more irregular and space-filling

urface.As suggested by Katz and Thompson (1985), the volume of

oil with particle diameter > di(di > di+1, i = 1,2,. . .) is defined as

(δ > di) = A[1 − (di/k)3−D], (3)

here δ is yard measure, k and A are the constants representingize and shape, respectively, and D is the fractal dimension. Forgiven soil, di represents the average value of soil particles

etween di and di+1. Generally, variations of particle density ρ

mong different soil particles could be ignored. And hence ρ

s a constant. Therefore, another expression (Eq. (4)) is derivedrom Eq. (3):

(δ > di) = V (δ > di)ρ = ρA[1 − (di/k)3−D]. (4)

here W(δ > di) is the cumulative mass of particles with sizes> di, and W0 is the total mass of any sizes of soil particles. The

ractal equation, reflecting the relationship between the massistributions of soil particles and average particle diameter, cane obtained as follows:

W(δ > di)

W0= 1 −

(di

dmax

)3−D

or

(di

dmax

)3−D

= W(δ < di)

W0. (5)

Then after regression analysis between log(di/dmax) andog((W(δ < di))/W0), the fractal dimension D can be calculated.

According to integral method, the relationship among soil

article fraction quantity, particle diameter and surface fractalimension is defined as

(≤ R) = C × R(3P2−13P+14/P2−5P+4)+1, (6)

2

p

gy 6 (2008) 157–166 159

here R is soil particle diameter, W is the cumulative mass ofarticles with sizes d ≤ R, and P is the surface fractal dimensionhich ranges from 2 to 3. As suggested by Zhang, Deng, Zhou,

nd Wu (2002), the Eq. (4) can be calculated after logarithm asollows:

n W = ln C +(

3P2 − 13P + 14

P2 − 5P + 4+ 1

)ln R. (7)

Using the measured soil particle dimension and relevantass, we can get the slope K through linear regression analysis,

= 3P2 − 13P + 14

P2 − 5P + 4+ 1. (8)

Finally, the surface fractal dimension, P, is derived from Eq.8),

= 18 − 5K +√

(18 − 5K)2 − 4 × (4 − K) × (18 − 4K)

2 × (4 − K).

(9)

.2.2. Soil microaggregate size distribution and stabilityThe size distribution of microaggregates was determined

ccording to Kachisky (ISSAS, 1978). Fifty grams of <1.00 mmsieved), air-dry soil samples were weighed into 500-ml volu-etric flasks. Soil samples in the flasks were moistened slowly

y adding deionized water up to saturation along the flasksall and kept for 10 min. Two hundred and fifty millilitersf deionized water was once again added into the flasks. Theoil samples were dispersed using a CSF-1A ultrasonic gener-tor with 21.5 kHz at 300 mA for 30 min. The macroaggregate.0–0.25 mm in diameter was isolated by wet sieving, and theicroaggregates <0.25 mm in diameter: 0.25–0.05, 0.05–0.01,

.01–0.005, 0.005–0.001, and <0.001 mm, were separated byedimentation. Aggregate degree and dispersion coefficient (Xu

Long, 2005) were used to describe soil microaggregate sizeistribution. Soil aggregate degree was an index to describe theoil particles changed into soil aggregates owing to amalgama-ion. While dispersion coefficient was another index to reflect theercentage of single soil particles in the soil, and it was in oppo-ition to aggregate degree. The larger the dispersion coefficient,he greater the dispersion characteristic of the soil aggregates;hat is to say, the higher the soil aggregate stability.

ggregate degree =

content of > 0.05 mm soil aggregate

−content of > 0.05 mm soil particle

content of > 0.05 mm soil particle

Dispersion coefficient

= content of < 0.001 mm soil aggregate

content of < 0.001 mm soil particle

.2.3. Water stability of soil aggregateWater stability of soil aggregate is the reflection of nondis-

ersed characteristic under water, and the Yoder’s wet sieving

1 cuolo

moiidcddqsw1stidwawsdt2

.25 m

greg

2

emTwsr3Ykabs

l

wacaswbt

2

t

etw1btaciBdug

2

tolwr

3

3

gsbldcduttfwcb(bag

60 Z. Zhang et al. / Parti

ethod (Dane & Topp, 2002) was used for the determinationf water-stable aggregate size distribution. Air-dried compos-te soil samples of 250 g were weighed and carefully placedn a 1000-mL measuring cylinder. Soil samples in the cylin-ers were wetted slowly by adding deionized water along theylinder wall up to saturation and kept for 10 min so as torive entrapped air from the aggregates. After 10 min, the cylin-ers were then filled with deionized water and allowed to situietly for 10 min. The cylinders were plugged with rubbertoppers and set up and down 10 times. The soil samplesere then transferred into the top sieve of a nest of sieves,0 cm in diameter and 5 cm height, in water, with a sieve meshequence of 5, 3, 2, 1, 0.5, and 0.25 mm from top to bot-om, and immersed in water to simulate flooding. Sieves weremmersed in water and moved up and down through a verticalistance of 3.5 cm with 50 repetitions in 10 min, so that wateras made to flow up and down through the screens and the

ssemblage of aggregates. The aggregates retained on the sievesere removed, oven-dried at 70 ◦C and weighed to obtain six

ize fractions: 5–3, 3–2, 2–1, 1–0.5, 0.5–0.25, and <0.25 mm iniameter. Meanwhile, aggregate deterioration rate was definedo reflect water stability of soil aggregates (He, Jia, Jin, & Qin,007):

Aggregate deterioration rate (%)

= content of > 0.25 mm air-dry aggregate-content of > 0

content of > 0.25 mm air-dry ag

.2.4. Mechanical stability of soil aggregateMechanical stability of soil aggregate belongs to steady prop-

rties, which are immune from outside force and water. Thus,echanical stability could stand for soil erosion resistance.o analyze this characteristic, a subsample (100 g, >0.25 mmater-stable soil aggregate) was incubated for 30 min with a

oil to water ratio of 1:5, and shaken on an oscillator at theate of 150 cycle/min for different shaking times (2, 4, 8, 16,2 min) (Russell & Feng, 1947), and then, sifted according tooder’s wet sieving method (Dane & Topp, 2002). The followinginetic equation was obtained by using contents of water-stableggregates at different surge times to calculate original sta-ility and collapse velocity which stand for the mechanicaltability:

og w = a − b log T,

here w is the weight of water-stable soil aggregates in gram,nd T is the elapsed time of oscillation in min. a and b, the inter-ept and slope of the fitted line, are termed “original stability”nd “collapse velocity”, respectively. Soils with high originaltability or low collapse velocity are more preferable than thoseith low original stability or high collapse velocity. Increase inis associated with a shift of the size distribution curve towards

he smaller size classes.

.2.5. Chemical stability of soil aggregateChemical stability of soil aggregate was the presenta-

ion of chemical structure immobility to measure the change

asea

gy 6 (2008) 157–166

m water-stable aggregate

ate× 100

xtent caused by certain reagent, and also reflected the resis-ance to microbiological factors. Soil samples were dealtith citrate-dithionate (C-D) reagent. Each subsample (50 g),00 mL of the sodium citrate (0.3N), 2.5 mL of sodium bicar-onate (1N) and sodium dithionate (1 g) were mixed, andhose mixtures were dipped in water at 80 ◦C for 30 minnd surged for 10 min (150 cycle/min) to make the ironements soluble (Zhong et al., 1998), and then sifted accord-ng to Yoder’s wet sieving method (Dane & Topp, 2002).y comparing untreated and treated with C-D reagent, theescending extent of water-stable aggregate under different landse patterns could represent chemical stability of soil aggre-ate.

.2.6. Statistical treatmentsStatistical analysis was carried out by using the SPSS (Sta-

istical Package for the Social Sciences) package. Significancef differences between treatment means was examined by aeast significant difference (LSD) range test procedure. Dataere analyzed by correlation and regression analysis to evaluate

elationships between different soil parameters.

. Results

.1. Soil macroaggregate size distribution

The studied results are presented in Table 2. Soil aggre-ates >3 mm under different land use patterns followed aequence from forestland > orchard > barren land > cropland,ut those >0.25 mm followed a sequence from barrenand > forestland > orchard > cropland. Meanwhile, The fractalimensions D under land use patterns followed a sequence fromropland > orchard > forestland > barren land. Surface fractalimensions P under orchard were higher, but they were lowernder forestland and barren land conditions, which indicatedhere were a few large soil particles and a worse particle dis-ribution under orchard. Particle size of soil aggregates underorestland condition had good uniformity and pedality, whichas in opposition to cropland, and higher values of D and P

orrespond to more fragmentation of soil aggregate and it coulde explainable with intensity of human action on soil structureLi & Zhao, 2005). They were consistent with those reportedy other researchers. The low values of D and P characterizedsize distribution dominated by a small number of macroag-

regates (Perfect, Kay, & Rasiah, 1994; Ding & Ding, 2007),

nd indicated that soil structure was more friable and perviouso that soil organism activities were strengthened and soil nutri-nts were improved (Liu, Hong, & Wu, 2002). Jastrow, Miller,nd Lussenhop (1998) confirmed the importance of roots and

Z. Zhang et al. / ParticuoloTa

ble

2So

ilag

greg

ate

frac

talc

hara

cter

istic

sun

der

diff

eren

tlan

dus

epa

ttern

s

Lan

dus

eR

ock

type

Soil

aggr

egat

esi

zedi

stri

butio

n(%

)D

P

10–5

mm

5–3

mm

3–2

mm

2–1

mm

1–0.

5m

m0.

5–0.

25m

m>

3m

m>

0.25

mm

Cro

plan

d

T1f

32.2

625

.49

14.6

17.

538.

031.

04

67.0

0±

7.10

92.8

4±

3.39

2.29

±0.

172.

57±

0.11

J 3p

44.5

720

.65.

639.

1111

.48

2.88

J 3s

54.5

617

.29

6.79

7.2

8.62

2.29

J 2s

61.3

311

.94.

455.

695.

902.

10

Orc

hard

T1f

50.4

320

.09

9.22

3.49

6.13

3.42

71.1

0±

4.04

94.5

6±

1.44

2.26

±0.

102.

63±

0.08

J 3p

56.7

18.6

17.

955.

985.

201.

67J 3

s44

.99

20.8

8.26

7.47

10.1

33.

57J 2

s56

.07

16.6

95.

775.

936.

912.

70

Fore

stla

nd

T1f

49.1

817

.38

8.9

9.00

6.73

2.78

74.0

7±

6.31

96.2

0±

1.56

2.14

±0.

082.

52±

0.08

J 3p

58.3

117

.96

7.39

5.43

6.24

1.81

J 3s

64.4

117

.02

5.76

4.20

4.71

1.31

J 2s

46.4

925

.54

8.61

7.42

6.57

1.66

Bar

ren

land

T1f

45.1

414

.35

10.3

28.

8610

.56

6.98

68.9

1±

9.86

96.4

8±

0.63

2.11

±0.

072.

49±

0.09

J 3p

54.0

816

.10

7.60

6.34

9.02

2.86

J 3s

67.4

214

.73

3.79

4.74

5.47

1.25

J 2s

37.8

026

.00

10.8

10.8

8.81

2.10

mbrf

3

eaotblsCciocaabstfiaw1cCao

4

4

aTwrdmfrJwJTcca

sa

gy 6 (2008) 157–166 161

ycorrhizal hyphae as driving factors for macroaggregate sta-ilization in a system recovering from disturbance. Thus, thoseesults could explain the biological processes in soil structureormation and human activities.

.2. Soil microaggregate size distribution

The characteristics of soil microaggregates under differ-nt land use varied due to inter-aggregate agglomerationnd cohesive action. From Table 3, we found that contentsf microaggregates >0.05 mm were lower under croplandhan barren land. Meanwhile, aggregate degree was affectedy land use patterns, following a sequence from barrenand > forestland > orchard > cropland, but the trend of disper-ion coefficient was just opposite to that of aggregate degree.ultivation and improper land use patterns destroyed surfacerusting and make soils harden, and were not good for the form-ng of water-stable granular structure due to impoverishmentf organic adhesive material. Aggregate degree and dispersionoefficient varied mainly due to the difference of surface oxidend composing, leading to the difference of cohesive actionmong soil particles. The results in this research were supportedy Pulleman, Six, Uyl, Marinissen, and Jongmans (2005), whouggested that significantly more microaggregates were found inhe field under permanent pasture than under conventional arableeld for all macroaggregate functions. The previous studieslso indicated that high proportions of microaggregate fractionshen native grasslands were cultivated (Cambardella & Elliott,993), and higher proportion of macroaggregates under zeroentigrade than conventional tillage practices (Beare, Hendrix,oleman, & Cabrera, 1994). The difference of microaggregatesmong four land use patterns might trigger the activity of soilrganisms such as earthworm activity (Pulleman et al., 2005).

. Discussion

.1. Water stability of soil aggregates

It was found that aggregate deterioration rates of croplandnd barren land were similar, following the same sequence of1f < J3s < J3p < J2s, while the trend of contents of >0.25 mmater-stable aggregate were opposite to that of aggregate dete-

ioration rates under cropland and barren land (Table 4). Theifferences of soil aggregate characteristic among rock typesay result from different sedimentary times and sedimentary

acies of rock types. The rocks of T1f were derived from ter-igenous deposit under dry and high-pressure conditions, while2s were from rivers under humid environment. The rocks of J3sere rooted in the landflood of alluvial lake, but the rocks of

3p were from intermittent lake and rivers under dry condition.he rock types, which stemmed from different hygrothermalonditions, influenced their weathering velocity and mineralomposition directly. And then the quality of soil aggregate was

ffected.

Average values were got based on Table 4 to compare watertability of soil aggregates under different land use patterns,nd listed in Table 5. The water-stable aggregate size distri-

162 Z. Zhang et al. / Particuology 6 (2008) 157–166

Tabl

e3

Soil

mic

roag

greg

ate

char

acte

rist

ics

unde

rdi

ffer

entl

and

use

patte

rns

Lan

dus

eR

ock

type

Soil

mic

roag

greg

ate

size

dist

ribu

tion

(g/k

g)A

ggre

gate

degr

ee(%

)D

ispe

rsio

nco

effic

ient

(%)

1–0.

25m

m0.

25–0

.05

mm

<0.

05m

m0.

05–0

.01

mm

<0.

01m

m0.

01–0

.005

mm

0.00

5–0.

001

mm

<0.

001

mm

>0.

05m

m<

0.00

1m

m

Cro

plan

d

T1f

234.

9934

8.49

416.

5210

6.35

310.

1721

2.32

57.7

340

.12

371.

79±

142.

6238

.9±

8.90

28.6

8±

19.7

553

.18

±28

.71

J 3p

50.7

227

8.64

670.

6421

1.78

458.

8637

6.73

48.4

633

.67

J 3s

120.

1117

3.61

706.

2827

7.17

429.

1125

1.36

146.

8730

.88

J 2s

122.

1915

8.42

719.

3929

1.21

428.

1830

3.95

73.3

050

.93

Orc

hard

T1f

344.

4821

3.15

442.

3714

8.32

294.

0521

3.36

47.6

133

.08

387.

65±

127.

1837

.58

±3.

5239

.94

±22

.93

44.4

3±

13.0

4J 3

p59

.57

194.

7374

5.70

267.

1947

8.51

381.

3157

.35

39.8

5J 3

s20

0.73

192.

8660

6.41

239.

5636

6.85

277.

6852

.61

36.5

6J 2

s17

2.38

172.

6865

4.94

278.

4437

6.50

276.

958

.76

40.8

4

Fore

stla

nd

T1f

347.

4516

7.26

485.

2920

6.14

279.

1524

5.89

19.6

213

.64

403.

52±

99.9

528

.94

±10

.24

49.6

1±

20.2

334

.98

±10

.74

J 3p

118.

9817

0.85

710.

1731

1.43

398.

7431

7.42

47.9

833

.34

J 3s

222.

6123

0.51

546.

8825

2.07

294.

8120

8.51

50.9

235

.38

J 2s

208.

4714

7.93

643.

6034

7.81

295.

7921

4.38

48.0

333

.38

Bar

ren

land

T1f

395.

2817

0.92

433.

8023

5.18

198.

6217

3.86

14.6

110

.15

453.

41±

81.9

121

.45

±7.

9762

.03

±13

.55

30.5

5±

6.89

J 3p

177.

5421

4.20

608.

2624

1.32

366.

9431

3.73

31.3

921

.82

J 3s

272.

3918

9.33

538.

2824

9.74

288.

5422

5.71

37.0

725

.76

J 2s

237.

5415

6.44

606.

0231

1.31

294.

7122

6.23

40.4

028

.08

bflrWtmeaatm&s>s(s(aoci&m(

4

(dOttTfarcM

Fig. 1. Content of >0.25 mm water-stable aggregates vs. organic matters.

utions (aggregates >3, 1 and 0.25 mm) varied greatly amongour land use patterns, following the same sequence of barrenand > forestland > orchard > cropland, while aggregate deterio-ation rates were opposite to contents of water-stable aggregate.

e could draw a conclusion that aggregations were the result ofhe interaction of many factors including the environment, soil

anagement, plant root growth, and soil properties (Pullemant al., 2005). Meanwhile, cultivation reduced the water-stableggregates of soils to smaller diameters and smaller particles orggregates were preferentially removed by erosion to underscorehe need for sustainable soil management practices that would

inimize nutrient loss and maintain good structure (AndersonMcBratney, 2002). In addition, we analyzed the relation-

hips between soil organic matters in Table 1 and content of0.25 mm water-stable aggregates in Table 4 to establish regres-ion equation. It was shown (Fig. 1) that soil organic mattersx) were related positively to water-stable aggregates of purpleoils (y), and the regression equation was y = 29.871 + 1.664xn = 16, r = 0.647**). Different land use patterns and soil man-gements affected the accumulation and conservation of soilrganic matter, especially soil aggregates quality. Cultivationan increase the rate of organic matter loss in soils primar-ly by accelerated microbial decomposition (Angers, Pesant,

Vigneux, 1992), and reduce the proportion of water-stableacroaggregates and increase the proportion of microaggregates

Monreal & Kodama, 1997).

.2. Mechanical stability of soil aggregates

Original stability of soil, which rooted in J3s, was the bestTable 6). Original stability of soil was the best under dry con-itions and sediment rocks of inundation from lakes to rivers.riginal stability of soils which developed from the former rock

ypes were different under different land use patterns, followinghe sequence of barren land > forestland > orchard > cropland.he distinction of collapse velocity among soils developed

rom three rock types followed the sequence of J3s < J3p < J2s,

nd those under land use patterns followed the order of bar-en land < forestland < orchard < cropland. Erosion resistance ofropland was the worst, while that of barren land was the best.eanwhile, anthropogenic process and disturbance were weak-

Z. Zhang et al. / Particuology 6 (2008) 157–166 163

Table 4Aggregate composition of soil aggregate and aggregate deterioration rate

Land use Rock type Water-stable aggregate (%) Aggregatedeterioration rate(%)

5–3 mm 3–2 mm 2–1 mm 1–0.5 mm 0.5–0.25 mm >0.25 mm

Cropland

T1f 13.49 11.51 22.76 18.74 7.27 73.77 17.08J3p 13.23 1.94 5.81 10.88 11.96 43.82 53.51J3s 10.89 1.62 5.65 14.44 15.85 48.45 49.92J2s 5.12 0.78 1.84 15.20 12.12 35.06 61.64

Orchard

T1f 19.79 7.89 21.45 24.02 8.45 81.6 12.05J3p 7.92 2.48 7.99 13.63 14.2 46.22 51.91J3s 37.44 6.04 14.08 16.23 7.40 81.19 14.73J2s 7.62 0.92 5.07 16.83 17.41 47.85 49.13

Forestland

T1f 17.07 10.58 22.53 26.25 8.71 85.14 9.40J3p 21.36 6.07 11.48 18.26 10.24 67.41 30.60J3s 47.38 8.01 15.25 14.48 4.88 90.00 7.61J2s 6.77 1.94 6.90 20.97 17.66 54.24 43.67

Barrenland

T1f 22.3 6.41 22.17 25.15 14.93 90.96 5.46J3p 45.78 4.38 9.98 12.88 6.69 79.71 16.98J3s 53.95 7.08 13.44 11.45 4.03 89.95 7.67J2s 22.73 6.09 13.29 11.86 7.27 61.24 36.41

Table 5Effect of land use patterns on content of soil aggregates and structure deterioration

Land use Contents of water-stable aggregates (%) Aggregate deterioration rate (%)

>3 mm >1 mm >0.25 mm

Cropland 10.68 ± 3.89 23.66 ± 17.05 50.28 ± 16.62 45.54 ± 19.60OFB

eeTinCmr

htsrp

rchard 18.19 ± 14.03 34.67 ± 21.92orestland 23.15 ± 17.28 43.84 ± 22.95arren land 36.19 ± 16.14 56.90 ± 13.83

ned gradually from cropland to barren land, but stability androsion resistance were strengthened. Original stability (a) inable 6 was highly positively related to soil organic matters (x)

n Table 1 (see Fig. 2), but content of soil clay (c) in Table 1 wasot related to original stability (a) in Table 6 (r = 0.0921, n = 12).

ollapse velocity (b) in Table 6 was not related to soil organicatters in Table 1 (r = −0.196, n = 12), while that was positively

elated to soil clay (c) in Table 1 (see Fig. 3). Soil organic matters

Fig. 2. Original stability vs. organic matters.

foo

64.22 ± 19.85 31.96 ± 21.4974.20 ± 16.47 22.82 ± 17.3980.47 ± 13.79 16.62 ± 14.10

ad indirect influence on mechanical stability of soil aggregatehrough original stability, and soil clay had influence on ero-ion resistance of soil aggregate through collapse velocity. Theesults agreed with Zhong et al. (1998), who found, in com-arison with the soil under zonal natural vegetation, hilly Udic

erralisols suffered more structure degradation under secondaryr man-made vegetation and it was characterized by loweringf aggregates stability and decreasing in bigger aggregates.

Fig. 3. Collapse velocity vs. soil clay.

164Z

.Zhang

etal./Particuology6

(2008)157–166

Table 6Values of original stability and collapse velocity in different periods

Land use Rock type T (min) W (g) Originalstability

Collapsevelocity

Correlationcoefficient

Landuse

Rocktype

T (min) W (mg) Originalstability

Collapsevelocity

Correlationcoefficient

Cropland

J3p

2 38.23

1.633 0.342 −0.953**

Forestland

J3p

2 63.66

1.919 0.317 −0.942**4 31.71 4 55.838 24.14 8 51.05

16 16.74 16 29.0332 11.07 32 28.61

J3s

2 64.9

1.753 0.247 −0.962** J3s

2 89.66

2.047 0.226 −0.942**4 40.29 4 73.058 36.79 8 70.99

16 29.21 16 59.6932 21.94 32 53.28

J2s

2 32.15

1.717 564 −0.966** J2s

2 56.87

1.865 0.411 −0.986**4 28.3 4 38.958 15.58 8 33.6

16 13.48 16 21.3432 6.14 32 18.72

Orchard

J3p

2 40.38

1.753 0.335 −0.956**

Barrenland

J3p

2 86.62

2.007 0.297 −0.980**4 38.66 4 75.548 28.2 8 51.06

16 25.64 16 43.6132 15.89 32 36.65

J3s

2 76.33

1.95 0.232 −0.97** J3s

2 93.39

2.07 0.212 −0.969**4 63.61 4 85.968 58.83 8 75.56

16 42.52 16 63.9332 41.76 32 58.69

J2s

2 45.05

1.759 0.523 −0.984** J2s

2 71.27

1.912 0.405 −0.986**4 26.13 4 44.698 16.96 8 36.38

16 14.73 16 25.2532 9.56 32 20.44

** Significant at 1% level (p < 0.01).

Z. Zhang et al. / Particuology 6 (2008) 157–166 165

Table 7Changes of water-stable aggregate of soils after treatment with C-D reagent

Land use Rock type Water-stable aggregate distribution (%) Decrement Mean value

5–3 mm 3–2 mm 2–1 mm 1–0.5 mm 0.5–0.25 mm >0.25 mm

CroplandJ3p 1.84 1.69 2.68 5.76 7.45 19.42 24.40

26.26 ± 1.66J3s 5.54 0.66 1.68 6.12 7.66 21.66 26.79J2s 2.09 0.42 0.95 2.20 1.81 7.47 27.59

OrchardJ3p 1.95 0.83 3.24 7.02 7.84 20.87 25.34

28.07 ± 6.39J3s 5.59 8.88 15.31 12.93 3.11 45.82 35.37J2s 6.71 4.91 1.2 6.77 4.77 24.36 23.49

ForestlandJ3p 24.95 4.91 11.1 10.52 5.22 56.71 10.71

19.24 ± 9.01J3s 28.19 8.42 9.79 10.06 4.87 61.33 28.67J2s 8.47 6.55 1.73 12.73 6.42 35.92 18.34

Bl

11.7415.1811.94

4

ruo(rcCcasiesTectaib

Ft

dt2atbNbiSoh

5

gdp

arrenand

J3p 31.47 4.23 9.00J3s 30.28 5.83 10.84J2s 6.05 5.35 6.72

.3. Chemical stability of soil aggregates

The change degree of soils developed from J3s under orchardeduced least, and soils decreased least under other three landse patterns were originated from J3p, by comparing contentsf >0.25 mm water-stable aggregates before (Table 4) and afterTable 7) treatment with C-D reagent. Water-stable aggregates,anging from 0.25 to 5 mm, decreased to some extent, and thehange degree was the largest under barren land and forestland.ontent of soil organic matter (x) was highly positively related toontent of water-stable aggregate (y) ranging from 0.25 to 5 mmfter treating (see Fig. 4). After treating with C-D reagent, thecope of the straight line changed from 1.664 to 1.7462 and thentercept varied from 29.871 to 0.2592, which indicated that afterliminating the cementing agent of iron and aluminum oxide,oil organic matters enhanced water stability of soil aggregate.he descending extent of water-stable aggregate under differ-nt land use patterns was found to be in the following order:ropland > orchard > forestland > barren land. Soil organic mat-

er played a key role in the formation and stabilization of soilggregates (Plante & McGill, 2002) and Caravaca et al. (2004)ndicated the differences that observed in the percentage of sta-le aggregates according to land use probably resulted from the

ig. 4. Content of water-stable aggregates after C-D treatment vs. organic mat-ers.

detrbub

A

Si

R

a

A

5.87 62.32 17.4021.50 ± 3.725.41 67.54 22.41

6.50 36.57 24.68

ifferences in the quality and quantity of organic matter betweenhe cultivated and forested soils (a Birang, Hauser, & Amougou,003; Caravaca et al., 2004). Organic matter was critical forggregate formation and stabilization, but at the same time a por-ion of the soil organic matter was protected from decompositiony its incorporation into aggregates (Six et al., 2002; Natalia &icholas, 2005). The low soil aggregate stability might haveeen caused partly by low organic and low clay and silt contentn desert soils of Hexi Corridor region in China (Li, Li, Rengel,ingh, & Wang, 2006), and the highest concentrations of totalrganic C and total carbohydrates were recorded in the soils withighest aggregate stability.

. Conclusions

Fractal theory was an effective method to reflect soil aggre-ate distribution, and the fractal dimension and surface fractalimension were sensitive to land use changes. Barren lands canrevent the soil from coarsening, and give it a larger fractalimension. In contrast, cultivated cropland soil had the low-st fractal dimension among all land use patterns, indicatinghat human activities induced soil degradation. The results thateflected water stability, mechanical stability and chemical sta-ility indicated the difference of aggregate stability under landse patterns were mainly due to the intensity of human distur-ance and cultivation.

cknowledgements

The project was supported by the Science and Technologyupporting Plan of China (no. 2006BAD05B01-02) and Min-

stry of Education of People’s Republic China (no. 2000062505).

eferences

Birang, M., Hauser, S., & Amougou, D. L. (2003). Farmers’ perceptions ofthe effects of earthworms on soil fertility and crop performance in southernCameroon. Pedobiologia, 47(5–6), 819–824.

nderson, A. N., & McBratney, A. B. (2002). Fractal dimensions. In J. H. Dane& G. C. Topp (Eds.), Methods of soil analysis. Part 4. Physical methods, Soil

1 cuolo

A

B

B

C

C

D

D

HH

I

J

K

L

L

L

L

M

M

N

O

P

P

P

P

P

R

S

S

S

W

W

W

X

X

Z

66 Z. Zhang et al. / Parti

Science Society of America, Book Series No. 5 (pp. 159–171). Wisconsin:Inc. Madison.

ngers, D. A., Pesant, A., & Vigneux, J. (1992). Early cropping induced changesin soil aggregation, organic matter and microbial biomass. Soil ScienceSociety of America Journal, 56, 115–119.

eare, M. H., Hendrix, P. F., Coleman, D. C., & Cabrera, M. L. (1994).Aggregate-protected and unprotected organic matter pools in conventionaland no-tillage soils. Soil Science Society of America Journal, 58, 787–795.

ird, S. B., Herrick, J. E., Wander, M. M., & Wright, S. F. (2002). Spatialheterogeneity of aggregate stability and soil carbon in semi-arid rangeland.Environmental Pollution, 116(3), 445–455.

ambardella, C. A., & Elliott, E. T. (1993). Carbon and nitrogen distribution inaggregates from cultivated and nature grassland soils. Soil Science Societyof America Journal, 57, 1071–1076.

aravaca, F., Lax, A., & Albaladejo, J. (2004). Aggregate stability and carboncharacteristics of particle-size fractions in cultivated and forested soils ofsemiarid Spain. Soil & Tillage Research, 78(1), 83–90.

ane, J. H., & Topp, G. C. (2002). Methods of soil analysis. Part 4. Physicalmethods. Soil Science Society of America, Book Series No. 5. Madison,Wisconsin, USA: Soil Science Society of America, Inc.

ing, Q. S., & Ding, W. M. (2007). Comparing stress wavelets with fragmentfractals for soil structure quantification. Soil & Tillage Research, 93(2),316–323.

e, Y. R. (2003). Purple soils in China. Beijing: Science Press. (in Chinese).e, B., Jia, L. M., Jin, D. G., & Qin, W. M. (2007). Studies on soil fertility

change in Acacia mangium plantation in Nanning, Guangxi. Scientia SilvaeSinicae, 43(5), 10–16 (in Chinese).

SSAS (Institute of Soil Science, Chinese Academy of Sciences). (1978). Meth-ods for soil physical and chemical analysis. Shanghai: Shanghai Science andTechnology Press. (in Chinese).

astrow, J. D., Miller, R. M., & Lussenhop, J. (1998). Contributions of interactingbiological mechanisms to soil aggregate stabilization in restored prairie. SoilBiology and Biochemistry, 30(7), 905–916.

atz, A. J., & Thompson, A. H. (1985). Fractal sandstone pores: Implica-tions for conductivity and pore formation. Physical Review Letters, 54(12),1325–1328.

i, J. F., & Zhao, X. H. (2005). Application of fractal theory in studying soilaggregates under different skidding models. Chinese Journal of AppliedEcology, 16(9), 1795–1797 (in Chinese).

i, X. G., Li, F. M., Rengel, Z., Singh, B., & Wang, Z. F. (2006). Cultiva-tion effects on temporal changes of organic carbon and aggregate stabilityin desert soils of Hexi Corridor region in China. Soil & Tillage Research,91(1–2), 22–29.

iu, J. F., Hong, W., & Wu, C. Z. (2002). Fractal features of soil clusters undersome precious hardwood stands in the central subtropical region, China.Acta Ecologica Sinica, 22(2), 197–205 (in Chinese).

iu, J. Y., Zhuang, D. F., Luo, D., & Xiao, X. (2003). Land-cover classificationof China: Integrated analysis of AVHRR imagery and geophysical data.International Journal of Remote Sensing, 24(12), 2485–2500.

andelbrot, B. B. (1982). The fractal geometry of nature. San Francisco: W.H.Freeman.

onreal, C. M., & Kodama, H. (1997). Influence of aggregate architecture andminerals on living habitats and soil organic matter. Canadian Journal of SoilScience, 77(3), 367–377.

Z

gy 6 (2008) 157–166

atalia, H., & Nicholas, B. C. (2005). Land use and landscape effects on aggre-gate stability and total carbon of Andisols from the Colombian Andes.Geoderma, 129(3–4), 268–278.

ades, J. M., & Water, A. G. (1991). Aggregate hierarchy in soil. AustralianSoil Research, 29, 815–828.

CARRD. (1980). Manual: Standard methods for analysis of soil, plant, tis-sue, water and fertilizer. Los Banos, Philippines: Philippines Council forAgriculture and Forestry Research.

erfect, E., Kay, B. D., & Rasiah, V. (1994). Unbiased estimation of the frac-tal dimension of soil aggregate size distributions. Soil & Tillage Research,31(2–3), 187–198.

irmoradian, N., Sepaskhah, A. R., & Hajabbasi, M. A. (2005). Applicationof fractal theory to quantify soil aggregate stability as influenced by tillagetreatments. Biosystems Engineering, 90(2), 227–234.

lante, A. F., & McGill, W. B. (2002). Soil aggregate dynamics and the retentionof organic matter in laboratory-incubated soil with differing simulated tillagefrequencies. Soil & Tillage Research, 66(1), 79–92.

ulleman, M. M., Six, J., Uyl, A., Marinissen, J. C. Y., & Jongmans, A. G.(2005). Earthworms and management affect organic matter incorporationand microaggregate formation in agricultural soils. Applied Soil Ecology,29(1), 1–15.

ussell, M. B., & Feng, C. L. (1947). Characterization of the stability of soilaggregates. Soil Science, 63, 299–304.

eybold, C. A., & Herrick, J. E. (2001). Aggregate stability kit for soil qualityassessments. Catena, 44(1), 37–45.

ix, J., Elliott, E. T., & Paustian, K. (2000). Soil macroaggregateturnover and microaggregate formation: A mechanism for C sequestra-tion under no-tillage agriculture. Soil Biology and Biochemistry, 32(14),2099–2103.

ix, J., Feller, C., Denef, K., Ogle, S. M., Sa, J. C. D. M., & Albrecht, A.(2002). Soil organic matter, biota and aggregation in temperate and tropicalsoils—Effects of no-tillage. Agronomie, 22, 755–775.

ei, C. F., Xie, D. T., & Che, F. C. (1989). Study on purple soil particlesand chemical component in south Sichuan Basin. Journal of SouthwestAgricultural University, 11(5), 425–428 (in Chinese).

ei, C. F., Gao, M., Shao, J. A., Xie, D. T., & Pan, G. X. (2006). Soil aggregateand its response to land management practices. China Particuology, 4(5),211–219.

ei, C. F., Ni, J. P., Gao, M., Xie, D. T., & Hasegawa, S. C. (2006). Anthropicpedogenesis of purple rock fragments in Sichuan Basin. China Catena, 68(1),51–58.

u, Y., & Long, J. (2005). Effect of soil physical properties on soil erosion inGuizhou Karst mountainous region. Journal of Soil and Water Conservation,19(1), 157–159 (in Chinese).

u, Y. F., Sun, D. A, & Yao, Y. P. (2004). Surface fractal dimension of bentoniteand its application to determination of swelling properties. Chaos, Solitons& Fractals, 19(2), 347–356.

hang, S. R., Deng, L. J., Zhou, Q., & Wu, G. F. (2002). Fractal dimensions ofparticle surface in the plowed layers and their relationships with main soil

properties. Acta Pedologica Sinica, 22(2), 197–205 (in Chinese).

hong, J. H., Tan, J., Guo, Q. R., Luo, B. S., Zhuo, M. N., & Huang, X. L.(1998). Comparative study on structure characteristics of hilly latored soilsunder different vegetation in south subtropics. Chinese Journal of AppliedEcology, 9(4), 359–364 (in Chinese).