全 文 ：作物学报 ACTA AGRONOMICA SINICA 2011, 37(4): 686−693 http://www.chinacrops.org/zwxb/
ISSN 0496-3490; CODEN TSHPA9 E-mail: xbzw@chinajournal.net.cn

This study was financially supported by Australia Center for International Agricultural Research (CIM-1999-094), Natural Science Foundation of
China (40771132), Key Projects in the National Science & Technology Pillar Program during the Eleventh Five-Year Plan Period (2006BAD15B06),
Education department of Gansu Province (0802-07), Research Fund for the Doctoral Program of Higher Education of China (20106202120004), and
Gansu Provincial Key Laboratory of Arid land Crop Science.
* 通讯作者(Corresponding author): 黄高宝, E-mail: huanggb@gsau.edu.cn, 0931-7632188
Received(收稿日期): 2010-06-18; Accepted(接受日期): 2011-01-06.
DOI: 10.3724/SP.J.1006.2011.00686
Effects of Lucerne Removal Time on Soil Water and Productivity in a Lucerne-
Wheat Rotation on the Western Loess Plateau
LI Ling-Ling1, HUANG Gao-Bao1,*, ZHANG Ren-Zhi2, CAI Li-Qun2, LUO Zhu-Zhu2, JIN Xiao-Jun1,
ZHANG En-He1, BELLOTTI Bill3, and UNKOVICH Murray4
1 Gansu Provincial Key Laboratory of Aridland Crop Science / Faculty of Agronomy, Gansu Agricultural University, Lanzhou 730070, China; 2 Fac-
ulty of Resource and Environment, Gansu Agricultural University, Lanzhou 730070, China; 3 School of Natural Sciences, University of Western Syd-
ney, Locked Bag Penrith NSW 1797, Australia; 4 School of Agriculture, Food and Wine, University of Adelaide, Roseworthy, SA 5371, Australia
Abstract: Rainfed spring wheat (Triticum aestivum L.) is the most important cereal crop on the Western Loess Plateau. Lucerne
(Medicago sativa L.) has been very popular. There are problems associated with both continuous cropping and with perennial
lucerne systems. The key challenge for rain-fed cropping systems is to adopt strategies that make optimal use of water. Develop-
ing lucerne-wheat rotation systems will have significant benefits for agriculture development on the Loess Plateau, nevertheless, it
is very important to terminate lucerne at the right time as it affects soil moisture. However, very little research has been done on
the timing for termination of old lucerne in the semiarid areas of the Western Loess Plateau. Based on field experiments conducted
in a typical semiarid area on the Western Loess Plateau, this paper aimed to investigate the soil water and termination timing of 30
years old lucerne on the productivity of lucerne-wheat rotation. The results showed that the soil profile after long-term lucerne
was very dry down to 3 meters, the three year experiment period was not sufficient to allow soil water recharge, even after a high
rainfall year. Time of 30 years old lucerne removal (in spring or later in the year) had no significant effect on soil water regimes.
As a result, weeds became more competitive, the old lucerne stand showed poor dry matter, yield, had no response to 1 kg ha−1 of
N application, and was overdue for termination. Following spring wheat made no response to 1 kg ha−1 of N fertilizer due to dry
soil profile after 30 years lucerne growing.
Keywords: 30 years lucerne; Lucerne-spring wheat rotation; Soil moisture; Productivity; WUE
西部黄土高原苜蓿终止时间对苜蓿-小麦轮作系统生产力及土壤水分
的影响
李玲玲 1 黄高宝 1,* 张仁陟 2 蔡立群 2 罗珠珠 2 晋小军 1 张恩和 1
BELLOTTI Bill 3 UNKOVICH Murray 4
1甘肃省干旱生境作物学重点实验室 / 甘肃农业大学农学院, 甘肃兰州 730070; 2甘肃农业大学资源与环境学院, 甘肃兰州 730070;
3 School of Natural Sciences, University of Western Sydney, Locked Bag Penrith NSW 1797, Australia; 4 School of Agriculture, Food and
Wine, University of Adelaide, Roseworthy, SA 5371, Australia
摘 要: 旱作春小麦 (Triticum aestivum L.)是西部黄土高原最重要的禾谷类作物, 该区苜蓿(Medicago sativa L.)分布
也非常广泛。持续的作物连作和多年苜蓿种植系统都存在很多问题。雨养农业系统发展的关键是最佳水分利用策略
的应用。发展合理的苜蓿-小麦轮作系统对该区农业的发展有十分重要的意义。由于苜蓿终止时间严重影响土壤水分,
所以在适宜的时间终止苜蓿就显得十分重要。然而, 关于苜蓿-小麦轮作中老苜蓿在一年中适宜终止时间的研究鲜见
报道。本研究利用黄土高原西部典型的半干旱雨养农业区 30年老苜蓿布设田间试验, 旨在探索老苜蓿地土壤水分状
况、苜蓿终止时间和少量氮肥施用对系统生产力及土壤水分的影响。结果表明, 长期种植苜蓿后 0~3 m 土壤水分很
少, 即便遇到丰水年(2003年), 3年的时间都不足以恢复土壤水分。30年苜蓿在一年中春季还是秋季终止对土壤水分
状况无显著影响。种植苜蓿 30年后杂草竞争力增强, 苜蓿干物质和产量水平都相当低, 且对 1 kg hm−2的氮肥使用无
明显响应。由于土壤水分含量太低, 后茬春小麦对 1 kg hm−2的氮肥使用和苜蓿终止时间也无明显响应。因此, 苜蓿
第 4期 李玲玲等: 西部黄土高原苜蓿终止时间对苜蓿-小麦轮作系统生产力及土壤水分的影响 687


持续种植时间太长会耗竭土壤水分, 使后茬春小麦对苜蓿在一年中的终止时间及少量的氮肥使用无响应, 需要 3 年
以上时间才有可能恢复土壤含水量。
关键词: 30年苜蓿; 苜蓿-春小麦轮作; 土壤水分; 生产力; WUE
The Loess Plateau is one of the poorest regions in
China. Severe erosion problems across large areas of the
plateau have made it difficult to develop agriculture
beyond subsistence levels. Despite the Loess Plateau’s
low and variable rainfall, rain fed spring wheat is the
most important crop for people’s food security[1-4].
Traditional wheat dominanted agricultural systems
contrbute significantly to the erosion problems on the
plateau, through extensive tillage and cultivation of
sloping areas. As a result grain yield is also low and
variable[5-8]. On these degraded soils fertilizer require-
ments are increasing steadily and farming systems are
becoming unsustainable[9-11].
Lucerne (Medicago sativa L.), sometimes referred to
as “the King of fodders” is the most nutritious perennial
plant with the highest nutrient value as a forage. Lucerne
has been adapted to a wide range of environments in the
Western Loess Plateau and is grown in many areas, par-
ticularly following the implementation of a central go-
vernment policy aimed at reverting crop land to grass and
forestry[12].
There are problems associated with both continuous
cropping and perenial pasture systems[13]. Continuous
wheat cropping reduces soil nitrogen content thereby
leading to a decline in soil fertility[14-16]. Much research
has shown that perenial pastures reduce soil moisture
content to very low levels, reducing their own
productivity and that of any subsequent crop[17]. Cereal-
legume rotations can overcome these problems by
increasing soil nitrogen availablity under the legume
pasture phase, whilst allowing soil moisture to recharge
under cereal crops[18]. Among the many different possible
rotation systems, a lucerne-wheat rotation has been found
to be best for soil and water conservation[19]. Research by
Pan, et al. on the Loess Plateau has shown that wheat
yield after lucerne can be increased by 34.0%–115.3%.
Organic matter and water stable aggregate at a depth of
0–20 cm were increased by 0.3%–0.5% and 20%–30%
compared with continuious wheat cropping[20]. Therefore,
developing lucerne-wheat rotation systems will have sig-
nificant benefits for agriculture on the Loess Plateau—
increasing crop production, helping develop animal hus-
bandry, reducing erosion and improving the sustainabi-
lity of agricultural systems.
The key challenge for rain-fed cropping systems is to
adopt strategies that make optimal use of water and soil
nutrients[21]. A lucerne-wheat rotation is good for im-
proving water use efficiency and NO3-N use efficiency.
However, it is important to terminate lucerne at the right
time as late termination can reduce soil moisture content
to a level where recharge by rainfall may be insufficient.
This can have negative effects to wheat yield after lu-
cerne[18, 22]. The process and extent of soil water recharge
is affected by environment and soil type[23-24], and has a
significant effect on the performance of the next crop.
Very little research has been done on this issue for old
lucerne in the semiarid areas of the Western Loess Pla-
teau.
This paper is a study of field experiments conducted in
Dingxi City, Gansu Province, a typical semiarid area on
the Western Loess Plateau for 30 years lucerne, aimed to
investigate soil water and its recharging, lucerne produc-
tivity after 30 years growing, and effect of timing of lu-
cerne termination and N application on soil water and
following crop under a lucerne-wheat rotation. Meas-
urements of soil water content, biomass and water use
efficiency (WUE) of lucerne, and biomass, grain yield
and WUE of wheat were taken and discussed.
1 Materials and methods
1.1 Experiment site
The field experiments were conducted in Dingxi,
Gansu province, northwest China. The site is located at
longitude 104º44′E , latitude 35º28′N, at an altitude of
1 971 m above sea level. In this region average radiation
is 591.89 J cm–2 and sunshine hours are 2 476.6 h (REF).
Average annual rainfall over 35 years is 391 mm, about
54% of it occurring between July and September. The
summer is warm and wet, with temperatures of up to
38 , but winter is cold and℃ dry with temperatures
dropping to –22 . Accumulated temperature℃ s above 5 ℃
and 10 are 2℃ 782.5 and ℃ 2 239.1 respectively (Fig. ℃
1 and Fig. 2).

Fig. 1 Monthly rainfall in 2002–2004 compared with averaged
long-term (1970–2004) in Dingxi
688 作 物 学 报 第 37卷


Fig. 2 Averaged long-term daily max and min temperatures
(1970–2004) in Dingxi
Experiment period rainfall data, Fig. 1, were collected
at the research site and compared to long term average
monthly rainfall data (source: Gansu Provincial Mete-
orological Bureau). Fig. 2, long term temperature data,
were also obtained from Gansu Provincial Meteorologi-
cal Bureau.
The soil of the study area is loess soil, locally known
as Huangmian soil. Table 1 shows the physical and
chemical properties of Huangmian soil. It is a uniform
soil with high soil pH down to 3 m. OC and TN are high
at the top of the soil profile, but decline slightly deeper in
the profile. TP is high but available P is low. AK is also
high, sufficient for crop production. Bulk density is con-
sistent (around 1.2) down the profile.
1.2 Experimental design
Eight treatments were designed for this experiment
(Table 2), all the plots were arranged in a split-plot

Table 1 General soil chemical and physical properties of Huangmian soil at Dingxi site (Aug, 2001)
Soil depth pH OC (%)
T-N
(%)
T-P
(%)
A-N
(mg kg−1)
A-P
(P2O5, mg kg−1)
A-K
(K2O, mg kg−1)
B.D
(g cm−3)
0–10 cm 8.29 0.836 0.105 0.157 22.88 6.98 195.44 1.25
10–30 cm 8.35 0.691 0.79 0.156 28.06 5.94 124.31 1.20
30–60 cm 8.42 0.572 0.63 0.153 16.86 2.74 115.88 1.21
60–90 cm 8.44 0.693 0.72 0.161 19.28 2.99 135.44 1.25
90–120 cm 8.44 0.749 0.82 0.154 16.06 3.65 132.94 1.11
120–150 cm 8.48 0.702 0.75 0.149 14.89 2.81 106.38 1.13
150–200 cm 8.43 0.776 0.86 0.158 15.19 2.85 91.81 1.18
200–250 cm 8.36 0.881 0.98 0.166 13.89 3.73 103.56 1.09
250–300 cm 8.27 0.675 0.71 0.155 15.49 2.82 101.75 1.13

Table 2 Treatment description and experiment management details at Dingxi
Treatment and code Lucerne termi-nation time Fallow period Cropping details
N application rate
(kg ha–1)
Continuous lucerne (LC) — — —
Continuous lucerne with nitrogen (LCN) — — — 15
Fallow (LF) May, 2001 Keep fallow since May, 2001 —
Fallow with nitrogen (LFN) May, 2001 Keep fallow since May, 2001 — 15
Lucerne-short fallow-wheat (LFsW) Oct, 2001 5 month
Lucerne-short fallow-wheat with nitrogen (LFsWN) Oct, 2001 5 month 15
Lucerne-long fallow-wheat (LFlW) May, 2001 10 month
Lucerne-long fallow-wheat with nitrogen (LFlWN) May, 2001 10 month
Following Lucerne termi-
nation, spring wheat was
sown in March each year
from 2002 to 2004.
15
Lucerne was established about 30 years before experiment. N was applied at wheat sowing.

design with lucerne-spring wheat rotation as the main
plot, N supplies (0 and 15 kg ha–1) as sub-plots. Each plot
measured 4 m×10 m.
1.3 Initial soil measurements
Before the very beginning of the experiment, mea-
surements of soil chemical and physical properties, and
soil water down to 3 meters were taken across the whole
lucerne paddock. Measurements were taken at nine soil
depths: 0–10 cm, 10–30 cm, 30–60 cm, 60–90 cm,
90–120 cm, 120–150 cm, 150–200 cm, 200–250 cm, and
250–300 cm.
1.4 Regular soil water measurements
Soil water was measured every two weeks using oven
method for 0–10 cm samples and a neutron moisture me-
ter for 10–300 cm samples. Basal sampling layer distri-
bution was used for both soil water and nutrients meas-
urements.
1.5 Crop measurements
1.5.1 Dry matter accumulation To measure dry
matter, wheat samples from each plot were collected in
第 4期 李玲玲等: 西部黄土高原苜蓿终止时间对苜蓿-小麦轮作系统生产力及土壤水分的影响 689


three 1 m rows at three-leaf stage, flowering stage and
maturity. Lucerne and weeds dry matter were measured
using three quadrates with an area of 0.5 m×0.5 m in
each lucerne plot. These measurements were made at
when lucerne was cut each year. Due to the age of the
lucerne and the Dingxi conditions, lucerne was only cut
once per year.
No lucerne data could be collected in 2003 due to the
SARS epidemic in the area during the harvest season.
Restrictions on the area were lifted in time to collect
wheat data.
1.5.2 Wheat yield and biomass At wheat maturity,
0.5 m from the margins of every plot were discarded and
total biomass and grain yield were measured and calcu-
lated from harvested area.
1.6 Water use efficiency (WUE) calculation
Accumulated water use or evapotranspiration (ET)
was calculated using the water balance equation: ET (mm)
= P–ΔS, where P is the in-crop precipitation, ΔS is the
change in soil water storage of the whole soil profile.
Water use efficiency for wheat grain yield (kg ha–1) and
lucerne biomass (kg ha–1) were calculated by the follow-
ing equation: WUE (kg mm–1 ha–1) = grain yield (or lu-
cerne biomass)/ET
1.7 Data analysis
All data collected from the experiments was analysed
using Data Process System.
2 Results
2.1 Soil water regimes and profiles at sowing and
harvest of wheat
The total soil water content of each treatment to a
depth of 3 m are shown in Fig. 3. The lucerne ley had
been standing 30 years before the experiment began, re-
sulting in very low initial total water storage of only 295
mm down to 3 m. This is lower than the threshold value
for wheat of 319 mm.
Despite the removal of lucerne in May and October
2001, no significant difference between treatments
emerged until spring wheat sowing in 2002. Subse-
quently, with rainfall occurring during the April to Au-
gust crop growing season, soil water of the LF treatment
began to recharge, but water depleting continued in the
other treatments due to crop growth. By the August 2002
wheat harvest, the LF treatment had stored 30 mm more
water than the other treatments. Aside from this, seasonal
factors were greater than any difference between treat-
ments.
Very dry conditions prevailed in the second half of
2002, through to sowing in March 2003. As such, all
treatments had similar water storage. Following sowing,

Fig. 3 Total soil water content down to 3 m for each treatment in Dingxi
Sowing and harvest labels in Figure 3 refer to sowing and harvest of wheat, wheat was sown in middle March and harvested in August each
year.

however, 2003 was a relatively wet year, particularly in
August (Fig. 1). The soil water profiles of all treatments
peaked in the middle of October, 2003. LF accumulated
80 mm more water than continuous lucerne (LC) and 25
mm more than the lucerne-wheat rotation treatments
(LFlW and LFsW). Interestingly, there was no difference
in soil water content between LFlW and LFsW treatments.
It is also interesting to note that even after a very wet
season in 2003 the water storage of the LF treatment was
only 52.7% of DUL (823 mm).
2004 was the driest year of the experiment period, es-
pecially during crop growing season, with only 127 mm
rain falling before harvest (Fig. 1). All the treatments’
soil water was depleted very quickly during this period,
especially LC, LFlW and LFsW, all declining to below
280 mm. LF also lost soil water through evaporation
through its bare surface, but lost less than the other
treatments where crops were using soil water. The 2004
harvest period showed the strongest difference between
treatments. LF stored 104 mm more water than LC.
However, again there was still no obvious difference
between LFlW and LFsW.
690 作 物 学 报 第 37卷

Fig. 4 and Fig. 5 show soil water content down the soil
profile of each treatment at sowing and harvest of spring
wheat. Throughout the 2001–2004 experimental period,
soil profiles of all treatments were very dry. Notably, Fig.
4 and Fig. 5 show that there was no change in soil water
content at depths below 1.8 m. This suggests that the
30-year-old lucerne stand has dried out the soil down to
at least 3m and at depths below 1.8 m, soil water may not
be recharged following lucerne, even after a high rainfall
year like 2003 (564 mm). Below 1.8 m the soil water
content of these treatments was considerably below that
of the CLL of spring wheat. Approximately 60 mm more
water would be needed below 1.8 m to bring soil water
content to a level where wheat could extract it for pro-
duction. After three years, water storage of fallow treat-
ments (LF and LFN) were still very low, at only 44.3%
of DUL and only 46 mm more than CLL. Therefore,
when designing a lucerne-crop rotation system, local
scientists should be concerned about the soil water deple-
tion by lucerne at depth and total amount.

Fig. 4 Soil water content down the soil profile of each treatment at sowing of spring wheat (wheat was sown in middle march each year)

Fig. 5 Soil water content down the soil profile of each treatment at harvest of spring wheat (wheat was harvested in August each year)

2.2 Biomass and water use efficiency of lucerne and
weeds
According to Table 3, lucerne dry matter production
was low in this very old lucerne ley, as low as 334.9 kg
ha–1 under the LCN treatment in 2004. The main reasons
for this might be the continuous invasion of weeds
(mainly annuals), age and poor plant density of the stands.
This also led to water use efficiency of lucerne being
quite low, between 4.03 and 6.42 kg ha–1 mm–1. Although
it was very dry in 2004, total biomass of lucerne and
weeds and WUE of lucerne were increased because of
high rainfall in 2003. Therefore, soil water storage is
very important for crop growth in a dry year.
After 30 years there were many weeds in the lucerne
ley. Weeds became more competitive than lucerne in
2004, their total biomass surpassing that of the lucerne
itself. This suggests that while rainfall can increase
growth of weeds, it does not bring about a simultaneous
improvement in lucerne growth. As a result of this both
dry matter of lucerne and the weeds in lucerne paddock
were measured (Table 3).
2.3 Biomass, grain yield, and water use efficiency of
spring wheat
Crop yield, harvest index and water use efficiency are
presented in Table 4. Generally there was no difference
between treatments. Wheat yields were low, especially in
2003 when, although the in crop rainfall (210 mm) was
the same as 2002, and the annual rainfall was the highest
of the experiment period, most of the in crop rain oc-
curred exactly before harvest. It had been particularly dry
in the 15 days around flowering. The late rain had almost
no effect on grain formation and filling, therefore, the
第 4期 李玲玲等: 西部黄土高原苜蓿终止时间对苜蓿-小麦轮作系统生产力及土壤水分的影响 691


harvest index and water use efficiency calculated from
grain yield were abnormally low. Grain yield was highest
in 2004, perhaps due to the late rainfall in 2003, suggest-
ing that soil water storage is important for wheat grain
yield, much as the higher biomass of lucerne and weeds
in 2004, seemed to be due to water stored from 2003.
Crop yield showed no response to the nitrogen treat-
ments, where 15 kg ha–1 of N had been added. The most
likely reason for this is the very dry soil profile following
the long term Lucerne.

Table 3 Biomass and water use efficiency of lucerne and weeds in 30 year old lucerne stands
Biomass (kg ha−1) WUE (kg mm−1 ha−1)
LC LCN Year
Lucerne Weeds Total Lucerne Weeds Total
LC LCN
2002 964.4 9.7 974.1 983.3 12.0 995.3 4.03 4.23
2003 NA NA NA NA NA NA
2004 526.9 700.1 1227 334.9 856.1 1191 6.42 6.17
NA: not available.

Table 4 Biomass, grain yield and water use efficiency of spring wheat sown following the termination of 30 year old lucerne stands in
Dingxi
Treatment code Biomass (kg ha−1)
Grain yield
(kg ha−1) HI
Soil water change
(harvest-sow)
(mm)
WUE
(biomass)
(kg mm−1 ha−1)
WUE
(grain yield)
(kg mm−1 ha−1)
2002 (in crop rain: 211 mm)
LFlW 5000±489 1777±82 0.36±0.05 −47.7 19.32 6.86
LFlWN 4694±417 1797±123 0.38±0.04 −43.8 18.42 7.05
LFsW 4083±436 1618±96 0.40±0.05 −40.4 16.24 6.43
LFsWN 4889±359 1489±77 0.30±0.04 −28.4 20.41 6.22
2003 (in crop rain: 210 mm)
LFlW 4068±192 358±11 0.09±0.00 39.7 23.89 1.92
LFlWN 4382±44 381±56 0.09±0.01 34.7 25.00 2.06
LFsW 3294±655 603±34 0.18±0.03 47.4 20.26 3.18
LFsWN 4453±503 512±43 0.12±0.02 49.4 27.73 2.81
2004 (in crop rain: 127 mm)
LFlW 4933±9 1888±77 0.38±0.01 −91.9 22.57 8.64
LFlWN 5178±92 1921±113 0.37±0.04 −98.9 22.95 8.51
LFsW 4244±169 1903±74 0.45±0.05 −94.7 19.17 8.60
LFsWN 4756±166 1938±2 0.41±0.04 −81.6 22.83 9.31

3 Discussion
Soil moisture content after lucerne is one of the key
factors affecting later crop growth. A great deal of re-
search has been done on this issue[25-26]. Results from this
experiment show that the 30 years lucerne stand has de-
pleted soil water, thus soil profile is very dry down to 3
m. The three year experiment period was not sufficient
for soil water recharge, especially for layers below 1.8 m.
Although significant effects of temination time of lucerne
on soil water had been detected in many research[7, 27], in the
results of this experiment time of lucerne termination (in
spring or later) had no effect on soil water regimes,
probably because the soil was rather dry after 30 years
depleting.
Lucerne is a very important forage plant on the Loess
Plateau due to its high nutrient content and high produc-
tivity. Lots of research has shown that lucerne obtains its
peak productivity between its fifth and eighth years, then
it begin to decrease[11, 28]. Terminating lucerne at the right
time is very important. The 30-year-old lucerne stand
showed poor dry matter yields due to its age, invasion of
weeds and poor plant density, it should be terminated
earlier. In rainfed areas, wheat yield depends on both
rainfall and soil water stored in the soil during fallow[3, 29].
In this research, as soil moisture following long term
lucerne is very low, subsequent spring wheat yielded
poorly.
Lots of researches have been done on wheat yield re-
sponse to N. Both wheat yield and protein content are
sensitive to nitrogen application rate [30-32]. However,
following removal of a long-term lucerne stand, soil
692 作 物 学 报 第 37卷

moisture was very low and spring wheat made no re-
sponse to small amounts of N fertilizer. Clearly, response
to N in cereal crops relies on adequate soil water availa-
bility.
4 Conclusions
If lucerne stands for too long, it will deplete soil water,
resulting in that lucerne yields rather poorly, time of old
lucerne removal within a year has no significant effect on
soil water regimes, following spring wheat has no sig-
nificant response to time of previous lucerne removal and
to small amounts of N application. It will take more than
three years for soil water to be recharged.
References
[1] Fan T, Wang S, Tang X, Luo J, Stewart B A, Gao Y. Grain yield
and water use in a long-term fertilization trial in Northwest China.
Agric Water Manag, 2005, 76: 36−52
[2] Fan T, Stewart B, Yong W, Junjie L, Guangye Z. Long-term ef-
fects on grain yield, water-use efficiency and soil fertility in the
dryland of Loess Plateau, China. Agric Ecosyst Environ, 2005,
106: 313−329
[3] Huang Y, Chen L, Fu B, Huang Z, Gong J. The wheat yields and
water-use efficiency in the Loess Plateau: straw mulch and irriga-
tion effects. Agric Water Manag, 2005, 72: 209−222
[4] Huang G B, Zhang R Z, Zhang G S, Li L L, Zhang E H, Jin X J,
Chan K Y, Heenan D P, Li G D. Conservation tillage effects on
spring wheat and field pea in the western Loess Plateau China. In
International Soil Tillage Research Organization 16th Triennial
Conference, Brisbane, Australia, 2003. pp 560−565
[5] Liu G. Soil conservation and sustainable agriculture on the Loess
Plateau: challenges and prospects. AMBIO, 1999, 28: 663−668
[6] Li L-L(李玲玲), Huang G-B(黄高宝), Zhang R-Z(张仁陟), Jin
X-J(晋小军), Li G D, Chan K Y. Effects of conservation tillage
on soil water regimes in rainfed areas. Acta Ecol Sin (生态学报),
2005, 25(9): 2326−2332 (in Chinese with English abstract)
[7] Shen Y, Li L, Chen W, Robertson M, Unkovich M, Bellotti W,
Probert M. Soil water, soil nitrogen and productivity of lu-
cerne-wheat sequences on deep silt loams in a summer-dominant
rainfall environment. Field Crops Res, 2009, 111: 97−108
[8] Huang G B, Zhang R Z, Li G D, Li L L, Chan K Y, Heenan D P,
Chen W, Unkovich M J, Robertson M J, Cullis B R, Bellotti W D.
Productivity and sustainability of a spring wheat-field pea rota-
tion in a semi-arid environment under conventional and conser-
vation tillage systems. Field Crops Res, 2008, 107: 43−55
[9] Li F-R(李锋瑞), Gao C-Y(高崇岳). Comparison of ecological
efficiency on rotation patterns in Eastern part of Gansu Loess
Plateau. Acta Pratacult Sin (草业学报), 1994, 3(1): 48−55 (in
Chinese with English abstract)
[10] Li F R, Zhao S L, Geballe G T. Water use patterns and agronomic
performance for some cropping systems with and without fallow
crops in a semi-arid environment of northwest China. Agric Eco-
syst Environ, 2000, 79: 129−142
[11] Du S-P(杜世平), Wang L-F(王留芳), Long M-X(龙明秀).
Soil water dynamics and yield response on lucerne in dry land
farming in mountain region, Northern of Ninxia. Pratacult Sci
(草业科学), 1999, 18(1): 12−15 (in Chinese with English ab-
stract)
[12] Feng Z, Yang Y, Zhang Y, Zhang P, Li Y. Grain-for-green policy
and its impacts on grain supply in West China. Land Use Policy,
2003, 22: 301−312
[13] Bee G A, Laslett G. Development of a rainfed lucerne-based
farming system in the Mediterranean climatic region of south-
western Australia. Agric Water Manag, 2002, 53: 111−116
[14] Schultz J E. Effect of rotations on crop yield and soil fertility in
intensive cropping systems. Aust Field Crops Newsl, 1987,
1984−87
[15] Weston E, Doughton J, Dalal R, Strong W, Thomas G, Lehane K,
Cooper J, King A, Holmes C. Managing long-term fertility of
cropping lands with ley pastures in southern Queensland. Tropic
Grasslands, 2000, 34: 169−176
[16] Whitbread A M, Blair G J, Lefroy R D B. Managing legume leys,
residues and fertilisers to enhance the sustainability of wheat
cropping systems in Australia: 2. Soil physical fertility and car-
bon. Soil Till Res, 2000, 54: 77−89
[17] McCallum M, Connor D, O’Leary G. Water use by lucerne and
effect on crops in the Victorian Wimmera. Aust J Agric Res, 2001,
52: 193−201
[18] Liu X-H(刘晓宏), Hao M-D(郝明德). Effects of long-term rota-
tion and fertilization to mineralization of organic-N. Soil Environ
Sci (土壤与环境), 2000, 9(2): 129−131(in Chinese with English
abstract)
[19] Gao C-Y(高崇岳), Zhang X-H(张小虎), Liu Z-Y(刘照耀).
Comprehensive evaluation of 8 rotation systems used in the
Loess Plateau of Qingyang prefecture, Gansu Province. In: Ren
J-Z(任继周). China-Australia Gansu Grassland Agricultural Sys-
tems Research and Development Project (中澳技术合作甘肃草
地农业系统研究与发展项目). Lanzhou: Gansu Science and
Technology Press. 1994. pp 23−33 (in Chinese)
[20] Pan X-L(潘幸来), Sun L-H(孙来虎), Wang Y-J(王永杰), Zhang
G-Y(张贵云), Yao M-P(姚麦萍). Roots’ characteristics of lu-
cerne and wheat on the Loess Plateau. J Triticeae Crops (麦类作
物学报), 1997, 17(1): 32−35 (in Chinese with English abstract)
[21] Connor D J. Designing cropping systems for efficient use of li-
mited water in southern australia. Eur J Agron, 2004, 21: 419−431
[22] Li Y-S(李玉山). Productivity dynamics of alfalfa and its effects
on water-eco-environment. Acta Peodol Sin (土壤学报), 2002,
第 4期 李玲玲等: 西部黄土高原苜蓿终止时间对苜蓿-小麦轮作系统生产力及土壤水分的影响 693


39(3): 404−411 (in Chinese with English abstract)
[23] Zhang X-C(张兴昌), Lu Z-F(卢宗凡). Study on soil moisture
dynamic and water-consuming law in slop-land. Res Soil Water
Conserv (水土保持研究), 1996, 3(2): 46−56 (in Chinese with
English abstract)
[24] Latta R A, Cocks P S, Matthews C. Lucerne pastures to sustain
agricultural production in southwestern Australia. Agric Water
Manag, 2002, 53: 99−109
[25] Ridley A, Christy B, Dunin F, Haines P, Wilson K, Ellington A.
Lucerne in crop rotations on the Riverine Plains 1. The soil water
balance. Aust J Agric Res, 2001, 52: 263−277
[26] Jia Y, Li F, Wang X, Yang S, Aase J K. Soil water and alfalfa
yields as affected by alternating ridges and furrows in rainfall
harvest in a semiarid environment. Field Crops Res, 2006, 97:
167−175
[27] Dunin F, Smith C J, Zegelin S, Leuning R, Denmead T, Poss R.
Water balance changes in a crop sequence with lucerne. Aust J
Agric Res, 2001, 52: 247−261
[28] Clarkson N M, Mears P T, Lowe K F, Lloyd D L. Sustaining
productive pastures in the tropics: 8. Persistence and productivity
of temperate legumes with tropical grasses. Trop Grasslands,
1991, 25: 129−136
[29] Baumhardt R L, Jones O R. Residue management and tillage ef-
fects on soil-water storage and grain yield of dryland wheat and
sorghum for a clay loam in Texas. Soil Till Res, 2002, 68: 71−82
[30] Hocking P, Stapper M. Effects of sowing time and nitrogen fertili-
zer on canola and wheat, and nitrogen fertilizer on Indian mustard:
I. Dry matter production, grain yield, and yield components. Aust
J Agric Res, 2001, 52: 623−634
[31] Sadras V, Baldock J, Cox J, Bellotti W D. Crop rotation effect on
wheat grain yield as mediated by changes in the degree of water
and nitrogen co-limitation. Aust J Agric Res, 2004, 55: 599−607
[32] Stopes C, Millington S, Woodward L. Dry mater and nitrogen
accumulation by three leguminous green manure species and the
yield of a following wheat crop in an organic production system.
Agric Ecosyst Environ, 1996, 57: 1889−1196