www.elsevier.com/locate/still

Soil & Tillage Research 93 (2007) 346–355

Soil properties, crop productivity and irrigation

effects on five croplands of Inner Mongolia

Ha-Lin Zhao a,*, Jian-Yuan Cui a, Rui-Lian Zhou a, Tong-Hui Zhang a,Xue-Yong Zhao a, Sam Drake b

a Cold and Arid Regions Environment and Engineering Research Institute,

Chinese Academy of Sciences, 260 Donggang West Road, 730000 Lanzhou, Chinab Office of Arid Lands Studies, University of Arizona, 1955 E. 6th Street, Tucson, AZ 85719, USA

Received 24 September 2005; received in revised form 5 May 2006; accepted 21 May 2006

Abstract

In the Horqin Sand Land, more than half of the original pasture area has been converted to farmland over the last century. A field

experiment was conducted from 2000 to 2001 on five croplands in the Horqin Sand Land of Inner Mongolia to examine differences

in soil properties, crop productivity and irrigation effects across different soils in the region to assess their relative suitability for

cultivation, in the face of continued pressure for conversion of these generally fragile, sandy soils to agriculture.

Two irrigated croplands studied were originally sandy meadow (ISM) and sandy grassland (ISG), and three dry croplands were from

sandy meadow (DSM), sandy grassland (DSG) and fixed sand dunes (DFD). Results showed that most measured properties of soils, and

cropproductivity,differedamong thefivecroplands.Thesilt + clayfraction,bulkdensity,organicmatter content, totalNandP,available

N and P, average soil moisture and temperature, plant height and aboveground biomass were as follows in the DSMjDSGjDFD soils:

51.1%j47.5%j24.3%;1.44 g/cm3j1.49 g/cm3j1.58 g/cm3;6.3 g/kgj4.6 g/kgj3.4 g/kg;0.55 g/kgj0.33 g/kgj0.21 g/kg;0.21 g/kgj0.17 g/

kgj0.13 g/kg; 27.0 mg/kgj13.7 mg/kgj7.7 mg/kg; 2.9 mg/kgj2.9 mg/kgj3.0 mg/kg; 9.4%j7.0%j6.2%; 21.4 8Cj21.7 8Cj22.0 8C;

225 cmj220 cmj181 cm; and 2116 g/m2j1864 g/m2j1338 g/m2. Corresponding values for ISMjISG soils were: 54.3%j47.9%;

1.42 g/cm3j1.49 g/cm3; 8.5 g/kgj6.4 g/kg; 0.58 g/kgj0.42 g/kg; 0.20 g/kgj0.19 g/kg; 29.0 mg/kgj23.3 mg/kg; 4.7 mg/kgj7.9 mg/kg;

13.0%j10.1%; 21.0 8Cj21.1 8C; 266 cmj245 cm; and 2958 g/m2j2702 g/m2.

In general, the ecological origin of a cropland was a stronger determinant of its current characteristics than was irrigation history,

although irrigation was correlated with significantly increased organic matter content, some soil nutrient levels, and aboveground

biomass productivity. Results indicate that fixed sand dunes should not be converted to cropland because of their very sandy and

poorer soil, lower biomass productivity and greater wind-erosion risk. Although both the sandy meadow and sandy grassland may

be reclaimed for farming, the cropland derived from the sandy meadow had higher resistance to wind erosion and higher crop

productivity, so is somewhat more suitable than sandy grassland.

# 2006 Elsevier B.V. All rights reserved.

Keywords: Soil properties; Irrigation; Crop productivity; Inner Mongolia

* Corresponding author. Tel.: +86 931 4967201;

fax: +86 931 4967201.

E-mail addresses: zhaohalin666@hotmail.com,

resdiv@ns.lzb.ac.cn (H.-L. Zhao).

0167-1987/$ – see front matter # 2006 Elsevier B.V. All rights reserved.

doi:10.1016/j.still.2006.05.009

1. Introduction

Indices used to evaluate soil erodibility and potential

productivity differ among farming systems, soil types

and land use types (MacEwan and Carter, 1996; Gomes

et al., 2003). However, research has confirmed that soil

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355 347

erodibility is affected primarily by soil texture (Liu

et al., 2003), and soil potential productivity is affected

by soil fertility as well as soil texture (John et al., 1998).

Commonly, sandy soils are less resistant to erosion by

wind than finer-textured soils (Lopez et al., 2000;

Potter, 1990), and are generally poorer for agriculture,

as their lower soil nutrient levels and lower water-

holding capacity (Murdock and Frye, 1983), result in

lower potential productivity (Zhu and Chen, 1994). The

effect of cultivation on soil erosion by wind and on soil

productivity has been documented in a number of

studies (Gomes et al., 2003; Moreno et al., 2001). It is

well known that irrigation (Hao et al., 2000), minimum

tillage (Hatfield and Stewart, 1994), and organic

fertilizer application (Ouedraogo et al., 2001; Liu and

Zhao, 1996) can reduce soil erosion by wind and

increase crop productivity. It has been recognized that

erosion potential for some soils can be greatly increased

by inappropriate tillage and crop management practices

(Aguilar et al., 1988). Consequences of wind erosion

include a reduction in crop production due to selective

removal of the finest soil particles, rich in nutrients and

organic matter, reduction in soil water-holding capacity

and degradation of soil structure (Lopez et al., 2000). In

particular, newly broken dry land soils become highly

susceptible to erosion by wind (Chenpil et al., 1952).

More than 90% of Inner Mongolia’s land is arid or

semi-arid (Zhu and Chen, 1994), and for much of its

history animal husbandry has been the only significant

industry. In the last hundred years, with increasing

population and demand for food, quite a lot of grassland

in the semi-arid areas has been reclaimed for farming

(Zhang et al., 1998). However, crop production has been

very low and unstable due to the drought-prone climate

and sandy soils, and soil erosion by wind has been very

serious in those areas (Wang, 2000). Data from Liu et al.

(2003) have shown that crop output and soil erosion

intensity differ significantly among different types of

cropland. Crop output was increased and wind erosion

intensity was weakened by irrigation (Zhao et al.,

2003). But thus far there are few studies of the

mechanisms of irrigation effects acting on different soil

properties to impact crop output in this area (Wang,

2000; Zhao et al., 2003).

Horqin Sand Land lies in a semi-arid area of southeast

Inner Mongolia. Due to the long-term influence of heavy

grazing and over-reclamation, Horqin Sand Land has

become one of the areas of most serious land degradation,

and one of the poorest areas in Inner Mongolia. Several

researchers have investigated the desertification types,

and the causes and distribution of sandy desertified land

in this area (Wang, 2000; Xu and Liou, 1997; Zhu and

Chen, 1994). Others have studied characteristics of soil

degradation as affected by wind erosion (Su and Zhao,

2003; Su et al., 2002), and wind erosion effects on crop

production in this area (Li et al., 2004). The objectives of

this paper are to: (1) analyze soil properties and their

effects on crop biomass productivity in croplands derived

from different types of grassland; (2) explore irrigation

effects on soil properties and crop biomass productivity;

(3) discuss the relationship between crop biomass, soil

properties and irrigation; (4) make appropriate proposals

on grassland reclamation and cropland management.

2. Materials and methods

2.1. Study area

The study area is located in Naiman county

(428550N, 1208420E, 345 m a.s.l.) in the eastern part

of Inner Mongolia. Naiman County is located within the

Horqin Sand Land. It has a temperate continental semi-

arid monsoon climate. The mean annual precipitation is

366 mm, the mean annual potential evaporation is

1935 mm, and the mean annual temperature is 6.8 8C.

The annual frost-free period is about 130–150 days. The

average annual wind speed is 3.4 m/s, and the mean

wind speed in the spring is 4.3 m/s. Dunes alternating

with gently undulating lowland and grassland areas

characterize the landscape in this region. Thickness of

the soil layer in the studied cropland is about 30–45 cm,

and the soil consists mainly of coarse sand and silt. Corn

(Zea mays L.) monoculture dominates the cultivated

land. Corn yields differ greatly in different types of

croplands, affected by soil properties and terrain (Li

et al., 2004).

2.2. Experimental design

The study was conducted during 2000 and 2001. Five

study areas, each about 10–20 ha in size, were selected

on five different types of cropland, including two

irrigated croplands and three dry-farmed croplands. All

lie within an area monitored long-term by the Naiman

Desertification Research Station (NDRS), part of the

Chinese Ecosystem Research Network. The two

irrigated croplands were on former sandy meadow

(ISM) and sandy grassland (ISG), while the three dry

croplands were formerly sandy meadow (DSM), sandy

grassland (DSG) and fixed dunes (DFD). All were

reclaimed for cultivation in 1985 and farmed con-

tinuously since then. The ISM and DSM croplands

originated from the same type of sandy meadow and are

assumed to differ only in management practices;

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355348

similarly for the ISG and DSG study areas. However, it

is important to note that the irrigated sites had been

irrigated for 16 years and the dry sites had been dry-

farmed for 16 years, including the study period. The

possible effects of this long-term management are one

subject of the current investigation. During the

experimental period, corn was sown on 14th May,

using the same methods at all sites, including a row

spacing of 40 cm, and was harvested on 21th September

each year in all five croplands. It was managed with

similar practices across all experimental croplands,

except for the addition of three irrigations in the ISM

and ISG croplands during the growing season (at the

jointing stage, heading stage and silk stage).

Six quadrats (2 m � 2 m) were established in each

study area to investigate soil and crop growth properties

in each type of cropland. Quadrat locations were

selected in consultation with NDRS staff for their

representativeness within each 10–20 ha study area.

First, a representative site (20 m � 30 m) was deter-

mined in each given cropland, then each site was

divided into six plots (10 m � 10 m), and one quadrat

was placed in the center of each plot.

2.3. Data collection and analysis

In all the quadrats, soil samples were collected from 0

to 20 cm depth in April 2000 and 2001 (6 quadrats � 2

years = 12 replicate samples for each cropland type).

Data collected in the two different years of the study were

pooled and treated as if they had been collected at the

same time. Each sample was obtained by mixing five sub-

samples collected from five locations in each quadrat.

Soil samples were placed in sealed plastic bags. In the

laboratory, each sample was thoroughly sieved to 2 mm

to remove roots and incorporated litter. Part of each

sieved sample was air-dried for determination of particle

size distribution and selected chemical properties. Soil

particle size distribution was determined by the pipette

method in a sedimentation cylinder, using Na-hexam-

ethaphosphate as the dispersing agent (Day, 1965). Soil

pH and electrolytic conductivity were determined with a

combination pH electrode (Multiline F/SET-3, Germany)

in a 1:1 soil–water slurry and 1:5 soil–water aqueous

extract, respectively. Soil organic matter was measured

by the K2Cr2O7–H2SO4 oxidation method of Walkey and

Black (Nelson and Sommers, 1982), total N by the

Kjeldahl procedure (UDK 140 Automatic Steam Distil-

ling Unit, Automatic Titroline 96, Italy) (ISSCAS, 1978)

and total P by UV-1601 Spectrophotometer (Japan), after

H2SO4–HClO4 digestion (ISSCAS, 1978). Soil available

N was determined by the alkaline diffusion method, and

available P determined by the Bray method (ISSCAS,

1978). In each quadrat, soil temperature (at depths of 5,

10, 20 and 30 cm) and volumetric soil moisture (at depths

of 0–10, 10–20, 20–30, 30–40, 40–50 and 50–60 cm)

were determined during the growing season by

geothermometers (HH82, Exphil Calibration Labs,

Bohemia, NY, USA) and hygrometers (TRIME-FM,

IMKO, GmbH, Ettlingen, Germany), with an observation

interval of about 10 days.

Aboveground biomass in each quadrat (again, 12

replicates per cropland type) was measured with the

clipping method—all green parts above the ground

surface were cut at plant maturity. The biomass samples

were separated based on leaf, stem and seed, and oven-

dried at 85 8C for 24 h before weighing. The height and

basal diameter of plants were measured by tape and

calipers, respectively.

All data were analyzed using the SPSS program for

Windows version 11.5 (Li et al., 2004). Multiple

comparison and one-way analysis of variance

(ANOVA) procedures were used to compare the

differences among the treatments (Sokal and Rohlf,

1995). Least significant difference (LSD) tests were

performed to determine whether treatment means were

significantly different at P < 0.05. Pearson correlation

coefficients were used to evaluate relationships among

the corresponding variables (Su and Zhao, 2003).

3. Results

3.1. Soil particle distribution

There were great differences in soil particle size

distribution among the DSM, DSG and DFD croplands

(Table 1). In order of silt + clay content (particles <0.05 mm) we observed DSM cropland > DSG cro-

pland > DFD cropland, and sand content (>0.05 mm)

was the reverse: DFD cropland > DSG cropland > DSM

cropland. Irrigation did not result in any changes in soil

particle size distribution between the ISG cropland and

the DSG cropland, and only an insignificant difference

(P > 0.05) was observed between the ISM and the DSM

croplands. Soil bulk density was highest in the DFD

cropland and lowest in the DSG cropland. The difference

in soil bulk density was not significant between the

irrigated croplands and the dry croplands (P > 0.05).

While results showed significantly different soil physical

properties in croplands derived from different pastures

(P < 0.05), irrigation effects on soil particle size

distribution and bulk density were not significant

(P > 0.05).

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355 349

Table 1

Comparison of soil particle size distribution and soil bulk density in five croplands in Horqin Sand Land

Cropland types ISM ISG DSM DSG DFD

Silt + clay (<0.01 mm) (%) 54.3 � 4.1 a 47.9 � 2.4 b 51.1 � 2.9 a 47.5 � 3.1 b 24.3 � 1.9 c

Sand (>0.01 mm) (%) 44.1 � 2.7 a 51.3 � 3.1 b 47.9 � 2.3 a 51.1 � 3.0 b 75.0 � 3.8 c

Soil bulk density (g/cm3) 1.42 � 0.09 a 1.49 � 0.10 b 1.44 � 0.10 a 1.49 � 0.10 b 1.58 � 0.09 c

ISM = irrigated sandy meadow; ISG = irrigated sandy grassland; DSM = dry sandy meadow; DSG = dry sandy grassland; DFD = dry fixed dunes.

Values are means (%) � S.D. Values with the same letters within rows are not significantly different at P < 0.05.

Table 2

Comparison of soil chemical properties in the same five croplands shown in Table 1

Items OM (g/kg) Total N (g/kg) Total P (g/kg) Available N (mg/kg) Available P (mg/kg) Available K (mg/kg) pH

ISM 8.5 � 0.7 a 0.58 � 0.02 a 0.20 � 0.02 a 29.0 � 2.7 a 4.7 � 0.9 a 92.0 � 5.6 a 8.25 � 0.05 ab

ISG 6.4 � 0.5 b 0.42 � 0.05 b 0.19 � 0.04 a 23.3 � 2.5 b 7.9 � 0.4 b 73.7 � 3.1 b 8.18 � 0.07 b

DSM 6.3 � 0.7 b 0.55 � 0.04 a 0.21 � 0.04 a 27.0 � 3.5 c 2.9 � 0.2 c 89.7 � 4.6 a 8.22 � 0.08 b

DSG 4.6 � 0.4 c 0.33 � 0.03 c 0.17 � 0.02 ab 13.7 � 3.1 d 2.9 � 0.2 c 74.0 � 3.6 b 8.35 � 0.06 a

DFD 3.4 � 0.2 d 0.21 � 0.03 d 0.13 � 0.02 b 7.7 � 1.2 e 3.0 � 0.2 c 38.0 � 3.5 c 8.49 � 0.06

Values are means � S.D. Values with the same letters within columns are not significantly different at P < 0.05.

3.2. Changes in soil chemical properties

Soil nutrient content showed significant differences

among the three dry croplands (P < 0.05) (Table 2).

Organic matter content, total N and P, and available N, P

and K were 36.9%, 66.7%, 23.5%, 97.1%, 0% and 21.2%

higher in the DSM cropland compared with the DSG

cropland, and 35.3%, 57.1%, 30.8%, 77.9%,�3.3% and

94.7% higher in the DSG cropland than in the DFD

cropland. Ordered by soil pH values, DFD cro-

pland > DSG cropland > DSM cropland; that is, the

dry fixed dune area was most alkaline. Irrigation resulted

in some significant changes in soil nutrients, and the

response to irrigation showed greater differences

between croplands derived from different pastures. For

Table 3

Comparison of soil water content in the same five croplands shown in Tab

Depth (cm)

0–10 10–20 20–30 30

Average soil moisture (%)

ISM 8.7 � 2.0 a 10.5 � 2.4 a 12.6 � 3.1 a 13

ISG 7.8 � 1.3 ac 8.2 � 1.5 b 9.4 � 1.9 b 10

DSM 11.2 � 3.3 b 9.5 � 3.5 ab 8.4 � 3.2 b 8.

DSG 7.6 � 1.7 ac 10.3 � 2.1 a 8.8 � 3.3 b 5.

DFD 6.5 � 2.3 c 6.1 � 2.8 c 5.8 � 2.5 c 6.

Minimum/maximum soil moisture (%)

ISM 5.3/13.9 6.2/15.9 6.7/19.9 7.

ISG 5.1/9.8 4.3/10.5 6.2/13.0 6.

DSM 5.3/16.5 3.7/16.9 4.3/15.5 4.

DSG 5.5/11.4 7.4/13.7 3.4/14.6 3.

DFD 2.4/10.4 2.2/12.0 2.8/11.7 3.

Values are means � S.D. Values with the same letters within columns are

the cropland derived from sandy meadow, irrigation

resulted in a significant increase only in organic matter,

available N and P (P < 0.05). Changes in total N and P

and available K were not significant (P > 0.05). For the

cropland derived from sandy grassland, irrigation

resulted in significant increases in organic matter, total

N and P, and available N and P (P < 0.05), while only

available K had no significant change. Irrigation had no

significant effect on soil pH values (P > 0.05).

3.3. Soil water content

Average soil moisture from 0 to 60 cm depth showed

significant differences among different croplands

(P < 0.05) (Table 3). In the dry croplands, average

le 1

Average

–40 40–50 50–60

.1 � 3.4 a 15.5 � 3.2 a 18.8 � 3.0 a 13.0 � 2.5 a

.9 � 2.6 b 12.7 � 3.9 b 12.8 � 3.3 b 10.1 � 2.1 b

4 � 3.0 c 9.7 � 3.4 c 9.9 � 3.0 c 9.4 � 1.7 b

6 � 2.3 d 5.1 � 2.6 d 4.1 � 1.5 d 7.0 � 1.7 c

0 � 2.6 d 6.5 � 2.9 d 6.0 � 2.2 d 6.2 � 2.3 c

7/20.1 11.9/21.7 14.4/25.4 8.7/18.5

2/16.5 6.5/19.9 7.3/19.0 6.0/13.9

3/13.2 4.9/18.5 7.0/17.9 6.9/12.6

2/12.8 3.1/13.9 2.7/9.3 4.6/11.6

9/14.1 3.6/13.7 3.6/11.3 3.4/12.0

not significantly different at P < 0.05.

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355350

Fig. 1. Comparison of plant height (a) and growth curves (b) of maize

plants in the five cropland types shown in Table 1. Bars represent

means � S.D.

soil moisture was highest in the DSM cropland, and

lowest in the DFD cropland, and was 51.6% higher in

the former compared to the latter. Although average soil

water content showed some difference between the

DSG cropland and the DFD cropland, the difference

was not significant (P > 0.05). The effect of irrigation

on average soil moisture was significant (P < 0.05).

Soil moisture was 38.3% and 44.3% higher in the ISM

cropland and the ISG cropland than in the DSM

cropland and DSG cropland, respectively.

Minimum and maximum soil moisture levels also

showed differences across soil types and with irrigation

(Table 3). The sandy meadow soils had higher minimum

and maximum soil moisture levels than the sandy

grassland soils, and irrigated soils had higher levels than

their non-irrigated counterparts. Maximum and mini-

mum soil moisture increased by 46.8% and 26.1% in the

ISM cropland compared to the DSM cropland, and by

19.8% and 30.4% in the ISG cropland compared to the

DSG cropland.

3.4. Soil temperature status

Soil temperatures from 10 to 30 cm depth did not

differ significantly among soils. At 5 cm depth, the DSG

and DFD soils were slightly warmer than the other soils

(Table 4). Although irrigation resulted in a slight decrease

in soil temperature, this was not significant (P > 0.05).

3.5. Plant heights and crop growth properties

In a comparison among dry croplands, average plant

height was ordered as DSM cropland > DSG cro-

pland > DFD cropland (Fig. 1a), although the only

significant difference was that DFD soils produced

shorter plants than the other dry soils. Plant height was

44 cm (24.3%) higher in the DSM cropland and 39 cm

(21.5%) higher in the DSG cropland than in the DFD

cropland. Irrigation resulted in a significant increase in

crop height (P < 0.05). Plant height was 41 cm (18.2%)

Table 4

Comparison of soil temperatures (8C) in the same five croplands shown in

Depth (cm)

5 10

ISM 22.2 � 3.1 a 21.2 � 3.1 a

ISG 22.2 � 3.3 a 21.4 � 2.7 a

DSM 22.3 � 2.8 a 21.6 � 2.6 a

DSG 24.2 � 3.3 b 21.8 � 2.9 a

DFD 24.8 � 3.5 b 22.1 � 2.8 a

Values are means � S.D. Values with the same letters within columns are

and 25 cm (11.4%) higher in the ISM cropland and the

ISG cropland than in the DSM and DSG cropland,

respectively.

As shown in the growth curves of Fig. 1b, plants

grew fastest in the ISM cropland and slowest in the DFD

cropland. The period of fastest growth started on 4th

June in the former and on 24th June in the latter, and

maximum height was reached on 16th July and 4th

August, respectively. The other growth curves are

generally similar to each other.

3.6. Aboveground biomass and its rate of increase

There were significant differences in aboveground

biomass among the dry croplands (Fig. 2a); biomass

Table 1

Average

20 30

20.5 � 3.0 a 20.3 � 3.0 a 21.0 � 2.9 a

20.7 � 2.6 a 20.0 � 2.6 a 21.1 � 2.4 a

20.9 � 2.6 a 20.8 � 3.2 a 21.4 � 2.5 a

20.6 � 2.7 a 20.2 � 2.7 a 21.7 � 2.6 a

20.8 � 2.4 a 20.4 � 2.5 a 22.0 � 2.3 a

not significantly different at P < 0.05.

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355 351

Fig. 2. Comparison of aboveground biomass (a) and biomass growth

curves (b) of maize plants in the five croplands shown in Table 1. Bars

represent means � S.D.

was highest in the DSM soils (2115.6 g/m2), followed

by the DSG (1863.6 g/m2) and DFD soils (1338.1 g/

m2). This represents 58.1% and 39.3% higher biomass

in the DSM and DSG croplands than in the DFD

cropland. Irrigation resulted in a significant increase of

biomass productivity. Compared with the DSM and

DSG croplands, biomass in the ISM and ISG croplands

was 842 g/m2 (39.8% higher) and 838 g/m2 (44.9%

higher), respectively. The proportionately greater

response to irrigation of DSG soils compared with

DSM soils is most likely due to their generally poorer

plant water availability when unirrigated.

The growth curves for biomass in ISM and ISG

croplands were very similar in the early season, but the

rate of biomass increase later in the season was

significantly greater in the ISM cropland than in the ISG

cropland (Fig. 2b). The period of fastest biomass

increase started on 4th July and ended on 15th August in

the irrigated croplands, and started on 16th July and

ended on 11th September in the dry croplands.

Although this period of rapid growth was longer in

the dry cropland compared to the irrigated cropland, the

average rate of biomass accumulation was significantly

slower in the dry cropland compared to the irrigated

cropland, so total biomass was significantly lower in the

dry cropland compared to the irrigated cropland.

4. Discussion and conclusions

4.1. Comparison of soil properties

Lopez et al. (2000) suggest that soil texture has a

large influence on soil stability, with coarse-textured

soils less resistant to erosion by wind. In the present

study, soil texture differed across cropland types. Sand

content (particles > 0.01 mm) was highest in the DFD

cropland (75%), and lowest in the DSM cropland

(44.1%). In contrast, silt and clay content (parti-

cles < 0.01 mm) was 110.3% higher in the DSM

cropland than that in the DFD cropland. With its

coarse texture and single-grain structure, the DFD soil

exhibits a high risk of erosion by wind in Horqin Sand

Land. This is consistent with the results of Su and Zhao

(2003). Based on particle size distribution, the cropland

derived from sandy meadow had the highest resistance

to wind erosion, followed closely by sandy grassland

soils, because of their higher fine particle content.

Although irrigation did not result in significant

change in soil particle size distribution, irrigation did

result in a significant increase in soil moisture.

Although sand content differed by only 7.9% between

the ISM cropland and DSM cropland, and by 0.4%

between the ISG cropland and DSG cropland, average

soil moisture from 0 to 60 cm depth (Table 3) in the ISM

and ISG croplands was 38.3% and 44.3% higher than in

the DSM and DSG croplands, respectively. Close

inspection of Table 3 shows that soil moisture from 0 to

20 cm in the irrigated soils was actually the same as, or

lower than, that in the dry soils. As these are growing-

season average values unaffected by the timing of

irrigations, they seem somewhat anomalous. In the

absence of data for verification, we speculate that this is

due to the maize crop having a shallower, more

extensive root system under irrigation than under dry

farming conditions. Irrigated maize utilizes soil

moisture closer to the surface, while dry-farmed maize

reaches deeper, seasonally stored water. Hu et al. (1991)

and Marticorena et al. (1997) suggest that an increase in

soil moisture is beneficial because it increases the

threshold shear velocity and makes soils more resistant

to erosion by wind. Thus irrigation can decrease soil

wind erosion by increasing soil moisture, but the effect

may be reversed in surface layers under some

conditions.

Wezel et al. (2000) indicate that in arid and semiarid

desert ecosystems, soil clay and organic matter

concentration is one of the most important factors in

the storage of nutrients and water in nutrient-poor sandy

soils. Saggar et al. (2001) also suggested that soil fine

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355352

particles are often associated with soil nutrients and their

availability, and with water-holding capacity. In our

study, the changing trend in soil nutrients and soil

moisture was consistent with the changing trend in soil

fine particle content. Comparing dry croplands, soil

nutrients and moisture were highest in the DSM cropland

and lowest in the DFD cropland. Organic matter, total N

and P, available N and K, and soil moisture were 46.0%,

61.8%, 38.1%, 71.5%, 57.6% and 34.0% lower in the

DFD cropland compared to the DSM cropland, and

26.9%, 40.0%, 19.0%, 49.3%, 17.5% and 25.5% lower in

the DSG cropland compared to the DSM cropland,

respectively. Compared to the dry croplands, the irrigated

croplands had higher soil nutrient contents in almost all

cases. Organic matter, total N and P, available N, P and K,

and soil moisture were 34.9%, 5.5%, �4.8%, 7.4%,

62.1%, 2.6% and 38.3% higher in the ISM cropland than

in the DSM cropland, and 39.1%, 27.3%, 11.8%, 70.1%,

172.4%, �0.4% and 44.3% higher in the ISG cropland

than in the DSG cropland, respectively. Results show that

cropland derived from sandy (lowland) meadow con-

tained more soil nutrients and soil moisture, and had more

resistance to wind erosion, than the sandy grassland and

particularly the fixed dune croplands. Although irrigation

could not change soil texture, it may be beneficial in

increasing soil nutrient availability and soil moisture

(Zhu and Chen, 1994). This is in agreement with the

results of Wezel et al. (2000), Saggar et al. (2001) and

Zhao et al. (2003).

Table 5

Correlation coefficients of soil physical properties to soil chemical propert

Items Sand Silt + clay Organic

matter

Total N

Dry croplands

OM �0.817** 0.930** 1.000

Total N �0.802** 0.924** 0.987** 1.000

Total P �0.739* 0.860** 0.940** 0.909**

Available N �0.780** 0.895** 0.979** 0.987**

Available P 0.113 0.015 0.150 0.100

pH values 0.812** �0.818** �0.724* �0.777**

Soil temperature 0.200 0.066 0.152 0.086

Soil moisture �0.604* 0.455 0.193 0.110

Irrigated croplands

OM 0.447 0.901** 1.00

Total N 0.444 0.878* 0.983** 1.00

Total P 0.508 0.709 0.468 0.501

Available N �0.240 0.969** 0.966** 0.966**

Available P 0.940** �0.446 �0.727 �0.746*

pH values 0.069 0.988** 0.837* 0.818*

Soil temperature 0.668 0.663 0.367 0.351

Soil moisture �0.605 0.798* 0.965** 0.977**

* Correlation significant at the 0.05 level.** Correlation significant at the 0.01 level (two-tailed).

4.2. Analysis of correlation among soil factors

Correlation analysis of the dry cropland indicated

that there was a significant negative correlation between

sand and organic matter, total N and P and available N

(Table 5), and a significant positive correlation between

fine particles (silt + clay) and these four factors

(P < 0.05). There was a significant positive correlation

between organic matter and total N, total P and

available N, and between available N and total N

(P < 0.05). Correlation was not significant between

available P and soil particle size, organic matter or total

P. There was a significant negative correlation between

soil pH and soil fine particles, organic matter, total N

and available N, and a significant positive correlation

between pH and soil sand content. Soil moisture had no

significant correlation with other soil factors except

sand content. In the dry croplands, soil organic matter,

total N and P, and available N were significantly lower

in soils with higher sand content, and significantly

higher in soils with more abundant fine particles. In

addition, changes (gain or loss) in total N and P are

apparently regulated by organic matter (Wezel et al.,

2000). Average soil temperature from 0 to 30 cm depth

was not influenced by soil texture, while soil

temperature did seem to influence available P content.

Increased soil sand content can result in a significant

decrease in soil moisture. The effect of soil moisture on

soil nutrients was not significant in the dry croplands.

ies

Total P Available N Available P pH values Soil

temperature

1.000

0.928** 1.000

0.426 0.145 1.000

�0.496 0.724* 0.510 1.000

0.425 0.160 0.934** 0.557 1.000

0.259 0.078 0.121 �0.143 �0.053

1.00

0.634 1.00

0.108 0.585 1.00

0.801* 0.934** 0.319 1.00

0.904** 0.550 0.344 0.762* 1.00

0.357 0.902** �0.862* 0.716 0.168

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355 353

Compared to the dry croplands, the irrigated

croplands showed no significant correlation between

sand content and soil nutrients except available P. There

was a significant correlation between fine particles and

total N and available N, but not total P. There was a

significant correlation between organic matter and total

N and available N, but again not total P.

The correlation of pH with other soil factors was

significantly altered in many respects by irrigation.

There was a change from negative to positive

correlation between pH and fine particles, total N and

P, and available N, and a change from significant

correlation to non-significant correlation between pH

and sand, and between soil moisture and sand. The

correlation changed from non-significant to significant

between soil moisture and fine particles, organic matter,

total N, available N and available P. As it changed the

original relationships among different factors in the soil,

irrigation apparently decreased the unfavorable effects

of soil texture and pH on soil nutrients, and intensified

the relationship between soil nutrients and soil moisture

(Zhao et al., 2003).

4.3. Effects of soil properties and irrigation on crop

productivity

Soil texture has a large influence on crop growth and

output (Zhao et al., 2003). Changes in soil properties

were accompanied by extensive changes in crop growth

and crop output (Hennessy and Kies, 1986). In the present

research, crop biomass differed greatly among croplands

with different soil properties. Comparing dry croplands,

aboveground biomass of corn was 13.5% higher in the

DSM cropland than that in the DSG cropland, and 58.1%

and 39.3% higher in the DSM and DSG croplands than in

the DFD cropland, respectively. Crop biomass also

showed a great difference between the irrigated crop-

lands and the dry croplands. Biomass was 39.8% higher

in the ISM cropland than in the DSM cropland, and

Table 6

Correlation coefficients of plant height and aboveground biomass to soil p

Items Sand Silt +

clay

Organic

matter

Total

N

Total

P

Avail

N

Dry croplands

Height �0.949** 0.962** 0.854** 0.821** 0.855** 0.805

Biomass �0.960** 0.949** 0.816** 0.776** 0.767** 0.734

Irrigated croplands

Height �0.310 0.921** 0.970** 0.988** 0.594 0.988

Biomass �0.319 0.921** 0.977** 0.963** 0.614 0.963

* Correlation significant at the 0.05 level.** Correlation significant at the 0.01 level (two-tailed).

45.0% higher in the ISG cropland than in the DSG

cropland. Results showed that both soil properties and

irrigation had significant effects on corn biomass

productivity, and both irrigation and better soil properties

could result in significant increases in crop biomass

productivity. When differences in soil properties were

greater among different croplands, irrigation effects on

crop productivity were greater (Zhao et al., 2004).

To confirm which soil factors were most related to

plant height and crop biomass, we analyzed the

correlations between soil factors and plant height and

aboveground biomass (Table 6). For the dry croplands,

both plant height and biomass had a significant positive

correlation with the soil fine particles, organic matter,

total N and P, available N and K, and soil moisture. A

significant negative correlation was seen with soil sand

content and pH. Ranked by correlation coefficient, the

order of the main factors influencing plant height in

the dry cropland was: sand content > available

K > silt + clay > total P > soil moisture > organic

matter > total N > available N > pH. The factors

influencing aboveground biomass showed a somewhat

different order: sand content > silt + clay > available

K > organic matter > total N > total P > available

N > soil moisture > pH. Results confirmed that all

measured soil physical and chemical factors had

significant effects on crop growth and output, which

is consistent with the results of Su and Zhao (2003) and

Zhao et al. (2003).

The results of correlation analysis on the irrigated

cropland showed that both plant height and aboveground

biomass had a significant positive correlation with soil

fine particles, organic matter, total N and P, available N

and K, pH and soil moisture (P < 0.05). The correlation

was not significant between sand content and plant height

and aboveground biomass. Ranked by correlation

coefficient, the order of the main factors influencing

plant height in the irrigated cropland was: available

N � total N > organic matter � available K > soil

hysical and chemical characteristics

able Available

P

Available

K

PH

values

Soil

temperature

Soil

moisture

** 0.139 0.969** �0.672* 0.003 0.762**

* 0.039 0.947** �0.704* �0.081 0.698*

** �0.639 0.970** 0.878* 0.481 0.933**

** �0.644 0.954** 0.883** 0.477 0.935**

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355354

moisture > silt + clay > pH. The order of the main

factors influencing aboveground biomass was: organic

matter > available N � total N > available K > soil

moisture > pH. Results showed that the effects of

organic matter, soil N and soil moisture on plant height

and biomass output were enhanced in the irrigated

cropland compared with the dry cropland, and soil

texture effects were reduced.

4.4. Conclusion

In the Horqin Sand Land, more than half of the

original pasture area was converted to farmland with the

rapid increase in population and food demand in the last

century. Most of the pasture opened to cropland was

originally meadow, grassland or fixed sand dunes. The

results of the present study showed that soil texture,

nutrient content, crop productivity and the effects of

irrigation differ greatly in the croplands derived from

these different original grassland types. Soil fine

particle content, nutrient content and biomass produc-

tivity were highest in the cropland derived from

meadow, and lowest in the cropland derived from fixed

dunes, with sandy grassland soils intermediate.

Although soil texture was not changed significantly

by irrigation, soil organic matter, some available

nutrients, deep soil moisture and biomass productivity

were significantly higher in the irrigated cropland than

in the non-irrigated cropland. Our results suggest that

fixed sand dunes should not be converted to farmland

because of their low potential biomass output as well as

a higher wind erosion risk in these areas. Although both

the meadow and grassland may be opened to cropland,

special attention must be paid to their agricultural use

and proper management, including irrigation, due to the

fragility and high susceptibility to degradation of soils

in the region.

Acknowledgements

The authors are grateful to the anonymous reviewers

for their critical review and comments on drafts of this

manuscript. This research was funded by the Chinese

Ecosystem Research Network Fund (1731690200015)

and one project of the Chinese National Science Fund

(40471004).

References

Aguilar, R., Kelly, E.F., Heil, R.D., 1988. Effect of cultivation on soils

in Northern Great Plains Rangeland. Soil Sci. Soc. Am. J. 52,

1081–1085.

Chenpil, W.S., Englehorn, C.L., Zingg, A.W., 1952. The effects of

cultivation on erodibility of soils by wind. Soil Sci. Soc. Am. Proc.

16, 19–21.

Day, P.R., 1965. Particle fractionation and particle-size distribution.

In: Black, C.A., Evans, D.D., White, J.L., Clark, F.E. (Eds.),

Methods of Soil Analysis. Agronomy, vol. 9. American Society

of Agronomy, Wisconsin, Madison, pp. 548–549.

Gomes, L., Arrue, J.L., Lopez, M.V., Sterk, G., Richard, D., Gracia,

R., Sabre, M., Gaudichet, A., Frangi, J.P., 2003. Wind erosion in a

semiarid agricultural area of Spain: the WELSONS project.

Catena 52, 235–256.

Hao, X.Y., Chang, C., Francis, J.L., 2000. Effect of minimum tillage

and crop sequence on physical properties of irrigated soil in

southern Alberta. Soil Tillage Res. 57, 53–60.

Hatfield, J.L., Stewart, B.A. (Eds.), 1994. Crop Residue Management

Advances in Soil Science. Lewis Publishers, Boca Raton, FL.

Hennessy, J.T., Kies, B., 1986. Soil sorting by forty-five years of wind

erosion on a Southern New Mexico Range. Soil Sci. Soc. Am. J.

56, 391–394.

Hu, M.C., Liu, Y.Z., Wu, L., 1991. An experiment in wind channel on

soil erosion by wind in Horqin Sand Land. J. Desert Res. 11 (1),

22–29.

Institute of Soil Science, Chinese Academy of Sciences (ISSCAS),

1978. Physical and Chemical Analysis Methods of Soils. Shanghai

Science Technology Press, Shanghai, pp. 7–59.

John, K., Collins, M., Drisch, D., 1998. A review of soil erosion

potential associated with biomass crops. Biomass Bioenergy 14

(4), 351–359.

Li, F.R., Zhao, L.Y., Zhang, H., Zhang, T.H., Shirato, Y., 2004. Wind

erosion and airborne dust deposition in farmland during spring in

the Horqin Sandy Land of eastern Inner Mongolia, China. Soil

Tillage Res. 75, 121–130.

Liu, X.M., Zhao, H.L., 1996. Wind-sand Environment and Vegetation

in Horqin Sand Land. Science Press, Beijing.

Liu, L.Y., Shi, P.J., Zou, X.Y., Gao, S.Y., Erdon, H., Yan, P., Li, X.Y.,

Dong, Z.B., Wang, J.H., 2003. Short-term dynamics of wind

erosion of three newly cultivated grassland soils in Northern

China. Geoderma 115, 55–64.

Lopez, M.V., Gracia, R., Arrue, J.L., 2000. Effects of reduced tillage

on soil surface properties affecting wind erosion in semiarid fallow

lands of Central Aragon. Eur. J. Agron. 12, 191–199.

MacEwan, R.J., Carter, M.R. (Eds.), 1996. Advances in soil quality

for land management: science, practice and policy.Proceedings of

an International Symposium, Centre for Environmental

Management, University of Ballarat, Vic., Australia, April

17th–19th.

Marticorena, B., Bergametti, G., Gillette, D., Belnap, J., 1997.

Factors controlling threshold friction velocity in semiarid and

arid areas of the United States. J. Geophys. Res. 102 (19), 23277–

23287.

Moreno, F., Cabrera, F., Fernandez-Boy, E., Giron, I.F., Fernandez,

J.E., Bellido, B., 2001. Irrigation with saline water in the reclaimed

marsh soils of south-west Spain: impact on soil properties and

cotton and sugar beet crops. Agric. Water Manage. 48, 133–

150.

Murdock, L.W., Frye, W.W., 1983. Erosion: its Effect on Soil Proper-

ties, Productivity and Profit. Publication AGR-102. University of

Kentucky, College of Agriculture, Kentucky.

Nelson, D.W., Sommers, L.E., 1982. Total carbon, organic carbon and

organic matter. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.),

Methods of Soil Analysis. Part 2. 2nd ed. Agronomy, Madison,

WI, pp. 539–577.

H.-L. Zhao et al. / Soil & Tillage Research 93 (2007) 346–355 355

Potter, K.N., 1990. Estimating wind-erodible materials on newly

crusted soils. Soil Sci. 150, 771–776.

Ouedraogo, E., Mando, A., Zombre, N.P., 2001. Use of compost to

improve soil properties and crop productivity under low input

agricultural systems in West Africa. Agric. Ecosyst. Environ. 84,

259–266.

Saggar, S., Yeates, G.W., Sheperd, T.G., 2001. Cultivation effects on

soil biological properties, microfauna and organic matter

dynamics in Eutric Gleysol and Gleyic Luvisol soils in New

Zealand. Soil Tillage Res. 58, 55–68.

Sokal, R.R., Rohlf, F.J., 1995. Biometry. W.H. Freeman, New York.

Su, Y.Z., Zhao, H.L., Zhang, T.H., Li, Y.L., 2002. Processes and

characteristics of soil degradation in rainfed farmland in the

Horqin Sandy Land. J. Soil Water Conserv. 16, 25–28.

Su, Y.Z., Zhao, H.L., 2003. Soil properties and plant species in an age

sequence of Caragana microphylla plantation in the Horqin Sand

Land, north China. Ecol. Eng. 20, 223–235.

Wang, T., 2000. Land use and sandy desertification in north China. J.

Desert Res. 20 (2), 103–113.

Wezel, A., Rajot, J.L., Herbring, C., 2000. Influence of shrubs on soil

characteristics and their function in Sahelian agro-ecosystems in

semiarid Niger. J. Arid Environ. 44, 383–398.

Xu, B., Liou, X.M., 1997. Field study on wind-sand flow affecting

crop seedling and farm soil erosion. J. Desert Res. 17 (1), 78–84.

Zhang, Q., Zhao, X., Zhao, H.L., 1998. Sandy Grassland in China.

China Weather Press, Beijing.

Zhao, H.L., Zhao, X.Y., Zhang, T.H., Wu, W. (Eds.), 2004. Deserti-

fication Processes and its Restoration Mechanisms in the Horqin

Sand Land. Science Press, Beijing.

Zhao, H.L., Zhang, T.H., Zhou, R.L., 2003. Soil environment and

productivity formation process of corn in five types of farmland of

Horqin Sandy Land. Acta Pedologica Sinica 40 (2), 194–199.

Zhu, Z.D., Chen, G.T. (Eds.), 1994. Sandy Desertification in China.

Science Press, Beijing.