78
Science in China Ser. D Earth Sciences 2004 Vol.47 Supp.ĉ 78ü88
Study of spatial distribution of sandy desertification in
North China in recent 10 years
WANG Tao1, WU Wei2, XUE Xian1, SUN Qingwei1 & CHEN Guangting1
1. Key Laboratory of Desert and Desertification, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese
Academy of Sciences, Lanzhou 730000, China;
2. Department of Environment Engineering, Peking University, Beijing 100871, China
Correspondence should be addressed to Wang Tao (email: [email protected])
Received November 10, 2003; revised December 16, 2003
Abstract Sandy desertification is a land degradation characterized by wind erosion, mainly
resulted from the excessive human activities in arid, semiarid and part of sub-humid regions in
North China. It is one of main kinds of desertification/land degradation as well as water-soil erosion and salinization in China. Rapid and continuous spread of sandy desertification during last
50 years has created a major environmental and socio-economic problem in North China. Remote sensing monitored results in 2000 showed that the sandy desertified land area has been
38.57h104 km2. The area of potential to slightly sandy desertified land is 13.93h104 km2, moderately land 9.977h104 km2, severely land 7.909h104 km2 and very severely land 6.756h104
km2. Sandy desertification mainly occurs in the semi-arid mixed farming-grazing zone and its
northern rangeland zone, semi-arid dryland rainfed cropping zone and arid oasis-desert margin
zone. The average annually developmental rate of sandy desertified land increased from 2,100
km2ga−1 in 1976ü1988 to 3,600 km2ga−1 in 1988ü2000. The basic status of sandy desertification in North China is “overall deterioration, while local rehabilitation”. Already achieved rehabilitation results and monitoring assessment show that about 60% of desertified land in North China
can be restored under the conditions of rational land-use ways and intensity.
Keywords: sandy desertified land, spatial distribution, remote sensing monitoring, last 10 years, North China.
DOI: 10.1360/04zd0009
Desertification is a very serious environment and
socio-economic problem facing the world today. According to the “United Nations Convention to Combat
Desertification” and China’s actual situation[1,2], desertification can be classified into several major types,
namely sandy desertification, soil erosion and salinization etc. One of the main manifestations of desertification in North China is sandy desertification. Since
the 1950s scientists in China have conducted a series
of researches on natural conditions, resources, desert
environment, sandy desertification and control in
Copyright by Science in China Press 2004
farmlands and grasslands, and rational use of water
and land resources in desert and sandy desertified regions. All these research and technique popularized
works lay a solid basis for launching large-scale stuü
dies of sandy desertification in North China[3 7].
In sandy desertification studies, one of the most
important and rapidly developed themes is the dynamical monitoring of temporal and spatial changes of
sandy desertification. The use of advanced means,
such as remote sensing, GIS, GPS and computer
Study of spatial distribution of sandy desertification in North China in recent 10 years
simulation etc., basically realizes the combination of
theory and practice, qualitative and quantitative studies, and status and prediction, which are fully reflected in the research achievements such as “Handbook on desertification indicators”[8], “Provisional
Methodology of Assessment and Mapping of Desertification (FAO)”[9], “World Atlas of Desertification”[10]
and “Handbook for the Field Assessment of Land
Degradation”[11]. Tucker et al.[12] used NOAA/
AVHRR data to infer the advance and retreat of Sahara
Desert basing on the precipitation calculated by NDVI
(normalized difference vegetation index). Hanna[13]
assessed desertification around deep wells in Sahel
region using NOAN/AVEIRR data. Hutchinson et
al.[14] tried to make famine early warning analysis in
Sahel region using NOAA/AVHRR and meteorological observation data in combination with market food
prices and field investigation data. Establishing the
relation between NDVI and crop yield, Malo et al.[15]
suggested that in the semi-arid Sahel region in West
Africa with an annual precipitation ranging from 150
ü1000 mm there was an obvious linear relation between NDVI and precipitation. From the investigation
in Botswana region during 1982ü1987 Nicholson[16]
found that the precipitation at most of observation stations can be described by NDVI. Palmer et al.[17] made
dynamical monitoring on land-use, vegetation changes
and desertification processes in Kalahari region in
south Africa during 1989ü1994 using TM data. In
Argentina the assessment of land desertification was
also completed by remote sensing technique and image processing system. At first a main image series
corresponding to the phenological phase of grassland
ecosystem was established as a reference different
times of image data using NOAA/AVHRR/LAC (large
area coverage), and then based on field investigation
and high-resolution images the desertification state
types were defined through direct supervision and
non-supervision classification. This method presents a
new evaluation conception and also represents a new
tendency in this respect at present[18]. Many researü
chers in China have also done much work[19 23] in the
dynamical monitoring of land desertification, indicator
system, assessing and monitoring methods, and
achieved encouraged results[24
79
ü28]
.
According to the researches and practices in land
degradation regions in North China over the past 30
years, we defined the sandy desertification is a land
degradation characterized by wind erosion mainly resulted from the excessive human activities in arid,
semiarid and part of sub-humid regions in North China.
It mainly occurred in human historic period, especially
during last century. In this respect, the human impact
and the interaction between the human and nature factors must be stressed, i.e. only the land degradation
resulted from the adverse effects of human activities
and the interaction of wind-dominated external agents
is called sandy desertification. Through remote sensing monitoring, field investigation and statistical
analysis, Zhu et al.[5] found that from 1950s to 1970s
sandy desertification in North China developed at an
average annual rate of 1560 km2ga−1. Following next
10 years the sandy desertified land developed at an
annual rate[29] of 2100 km2ga−1, and by the late 1980s
the area of sandy desertified land has been 33.3h104
km2. In order to further find out the sandy desertification status and developmental trend in North China
from 1987 to 2000, basing on the 1987 monitoring
results we conducted field investigation and dynamical
monitoring of sandy desertification in North China in
2000, and compared monitoring results of 1987 to
analyze the spatial changes and developmental trend
of land desertification in recent 10 years.
1 Monitoring method of spatial changes of sandy
desertification
The monitoring regions selected in 1987 were the
mixed farming-grazing zone and dry-land rainfed
cropping zone in its southern and pasture land in its
northern, extending from Hulun Buir Grassland in the
Northeast to Hedong desert zone of Ningxia in the
southwest and also including several representative
zones in Alaxa Plateau, Qaidam Basin and Tarim Basin to the West of Helan Mountain[29]. The monitoring
regions selected in 2000 were determined according to
previous field investigations, and they rapidly developed desertification zones in different periods, in-
80
Science in China Ser. D Earth Sciences
cluding 177 countries (cities) of 10 Provinces (Regions) in regions suffered from the sandy desertification of North China, with a total area of 256h104 km2.
The monitoring regions mainly contain the mixed
farming-grazing zone, dry-land rainfed cropping zone,
irrigated agricultural oases and grazing zones. In
natural zone they contain sub-humid grassland, semiarid steppe, desert steppe and high-cold meadow. According to the monitoring indicator system in 2000 the
sandy desertified land was divided into four degrees
such as potential to slightly sandy desertified land,
moderate land, severe land and very severely land.
Such a classification is basically consistent with the
desertification types of monitoring indicator system
classified in 1987, which includes potential sandy desertification, on-going sandy desertification, strongly
developing sandy desertification and serious sandy
desertification. On the whole, the present monitoring
results not only represent the real situation but also can
be compared with two previous monitoring results.
For the remote sensing monitoring of sandy desertification the first thing to be done is to establish the
classification indicator system of sandy desertification.
Our studies[30,31] showed that useful indicators for the
desertification processes and environmental conditions
can be divided into three indicators: (1) natural indicators, including the dynamic data of wind-eroded
land, sandy land or sandy dune spread, seasonal and
annual change of precipitation, wind direction, wind
velocity, effective soil thickness, organic matter content, groundwater depth and quality, and surface albedo etc.; (2) biological and agricultural indicators,
and including vegetation cover, biological production,
key plant species distribution, land-use regime (e.g.
farming, grazing, fuel collection, and industrial water
resource use etc.), crop yield, livestock composition
and number, and various economic input; (3) social
Degree
Slight (L)
Moderate (M)
indicators, including inhabitant, population, structure,
variation processes and developmental trend, health
indexes, mandatory policy or stage-specific policy etc.
According to the processes of sandy desertification in
North China, human activity characteristics and existing monitoring results, we summed up the following
directly usable indicators, which can be obtained and
analyzed by remote sensing and computer means: (1)
wind-eroded land area or moving sand area percentage
in a region’s total land area; (2) annual expanded area
percentage of wind-eroded land or moving sand in a
region’s total land area; (3) vegetation cover mainly
referring to grassland and forest land; (4) biological
production. Then we can give out a general classification of sandy desertification degrees and their indicators (table 1).
In fact, under different land-use conditions in
different regions sandy desertification often appears as
different types. We summarized the unified sandy desertification classification used in different periods of
remote sensing monitoring (table 2).
In the remote sensing monitoring, we often use
wind-eroded land area or moving sand area percentage
and vegetation cover as the most important indicators,
others indicators are used as auxiliary types. On the
one hand, wind-eroded land or moving sand area and
vegetation cover are an integrated embodiment of surface features, community structure, biological production, effective soil thickness, structure, organic and
moisture changes. On the other hand, they are clear,
easily judged and representative indicators in the
sandy desertification monitoring in North China.
Sandy desertified land area of 2000 in North
China was obtained by using the GIS such as ARCVIEW and ARC/INFO through the man-computer interpretation of 2000 TM remote sensing data. The
Table 1 Classification and indicators of sandy desertification degrees
Percentage of blown sand
Percentage of annual
Percentage of vegetation
area in total/%
expansion area/%
covera)/%
˘5
˘1
˚60
Percentage of annual
reduction in biomass/%
˘1.5
5ü25
1ü2
60ü30
1.5ü3.5
25ü50
2ü5
30ü10
3.5ü7.5
Severe (S)
Very severe (VS)
˚50
˚5
10ü0
˚7.5
a) Vegetation cover is calculated by projection method, and the vegetation cover of local primary landscape is regarded as 100%.
Study of spatial distribution of sandy desertification in North China in recent 10 years
81
Table 2 Sandy desertification types and classification in North China
Sandy desertification type
Main distribution region
Primary state
1
Potential to slightly
desertified land
Desertification degree
2
Moderately
desertified land
3
Severely desertified
land
4
Very severely
desertified land
A
B
C
D
E
Sandy dune mobilization or moving
sand encroachment
Shrub desertification
Gravel desertification
Wind erosion badland
Dry-farming cropland desertification
Sandy land in East
China desert of
West China, or river
banks in desert
Desert in West
China, sandy land
in East China, middle part of Inner
Mongolia
Gobi edge, middle
and west part of
Inner Mongolia
Lop Nur in Xinjiang, edge of Yardan zone at east
side of Alun Mts;
South Inner Mongolian Plateau (Bashang in Hebei)
Reclamation districts in East China,
grassland, north part
of Loess Plateau
Fixed dunes, oasis
grassland and farmland
Steppe or desert
steppe, steppified
desert
Desert steppe or
steppified desert
Steppe or desert
steppe, steppified
desert
Dry-farming cropland
1(a)
1(b)
1(c)
1(d)
1(e)
Blowouts occurs on
windward slopes of
dunes; area of moving sand spots is 5%
ü25%; primary
vegetation cover˚
90%.
Shrubs grow well;
shrub sand mound
and sand spits of
various shapes
occur.
Gravels are enriched on ground
surface.
Shallow winderoded pits occur
but no obvious
steep bench is
formed.
Sand is accumulated
in furrows in spring,
ridges exhibit obvious wind erosion
evidence.
2(a)
2(b)
2(c)
2(d)
2(e)
Sand dunes exhibit
obvious
wind-eroded slope
and slip face differentiation, area of
moving sand is 25%
ü50%, vegetation
cover is 50%ü90%
of primary state.
Leaved shrubs
cannot entirely
cover sand mounds;
moving sand occurs
on the windward
side of sand
mounds; loose sand
or gravels occur in
inter-ridge flats.
Coarse sand and
gravels cover
ground surface;
there are sparse
plants, grass cover
>25%; surface
exhibits a gravel
pastureland landscape.
Blowouts are
mostly exposed;
small steep benches
occur on ground
surface.
Small patches of
moving sand occur
in loessial farmlands; wind-eroded
cropland has very
low fertility; 50% of
humus layer is
blown away.
3(a)
3(b)
3(c)
3(d)
3(e)
Sandy land is in a
semi-fixed state;
area of moving sand
exceeds 50%;
vegetation cover is
smaller than 50% of
primary state.
Large area of shrubs
begins to die; vegetation cover ˘
25%, area of moving sand exceeds
50%.
Ground surface is
almost entirely
covered by gravels;
sands exist between
gravels; vegetation
cover is 10%ü
25%.
Residual
wind-eroded
benches occur on
ground surface;
grasses are scattered
between low-class
deflation residual
hills; gravelly
farmlands are abandoned.
Humus layer in
wind-eroded farmland is almost entirely blown away;
calcium horizon and
soil parent material
are exposed; moving
sand area of desertified croplands˚25%;
desertified croplands are mostly
abandoned.
4(a)
4(b)
4(c)
4(d)
4(e)
Mobile dune field;
vegetation cover˘
10%
Undulated moving
sand land; vegetation
cover˘10%
Ground surface is
entirely covered by
gravels; vegetation
cover˘10%
Wind-eroded badlands become the
main body of
ground surface.
Flat sand land or
gravel land vegetation cover˘10%
82
technical route for remote sensing monitoring of sandy
desertification is shown in fig. 1.
Fig. 1. Technical route block diagram for remote sensing monitoring
of sandy desertification in 2000.
In the remote sensing monitoring using abovementioned technical route we paid special attention to
the following several problems:
(i) Selection of optimal information source. In
2000 remote sensing monitoring TM data were used,
images could directly display on the computer screen.
Seasonal selection of images can greatly affect the
monitoring effect of sandy desertification. It may enhance or reduce sandy desertification degree classified
by vegetation cover in the interpretation process.
Therefore, selecting optimal time interval is a key to
the successful monitoring. The basic principle for the
selection of high-quality images is that they should
contain large amount of information, i.e. intertype
variance is large and key ground objects to be monitored should have evident color hue. In winter and
spring vegetation is in a poor condition, precipitation
is sparse, and colour hue of ground objects is monotonous. In summer, vegetation on sand land is
flourish, and vegetation cover can be used to classify
sandy desertification degree, but dense crops are easily
confused with forest. Therefore, autumn is the best
season to obtain optimal TM images for the sandy desertification monitoring. The present monitoring almost covered the whole North China. Due to the limitation of objective conditions, not all images in all
regions are obtained in optimal season. Some images
in local places actually reflect the surface regimes in
other seasons, which no doubt affects the interpreta-
Science in China Ser. D Earth Sciences
tion accuracy to a certain degree.
(ii) Influence of image processing technique on
monitoring effect. In the monitoring standard false
color composite TM images (TM band 4, 3 and 2 with
three colors, i.e. red, green, and blue respectively),
such band assemblage can best reflect vegetation
growth situation. TM band 4 can reflect vegetation
features in different degrees of sandy desertified land.
TM band 3 can reflect brightness of sandy soils and
whiteness of salinized soils. TM band 2 is sensitive to
vegetation reflectance and can distinguish forest types
and tree species. Therefore, this kind of assembly images of this band can prominently display vegetation
features, but they have weak response to sand dune
information. Therefore, image processing technique
and band assemblage can affect the monitoring effect
to a great degree.
(iii) Characteristics of sandy desertification images. We established main TM image characteristic
system of different sandy desertification types in different study zones. Interpreters and analysers must be
familiar with interpretation contents, image characteristics and interpretation marks of sandy desertification
of various types in the man-machine dialogue and
visual interpretation processes.
2 Remote sensing monitoring results of sandy desertification in 2000
The monitoring results of sandy desertification in
2000 are presented in table 3.
It can be seen from table 3 that in 2000 potential
to slightly sandy desertified land in North China occupied the largest percentage, about 13.93h104 km2 in
area, accounting for 36.1% of total sandy desertified
land area and 5% of total monitoring land area. Moderately sandy desertified land was about 9.977h104
km2 in area, accounting for 25.9% of total sandy desertified land area and 4% of total monitoring land
area; severely sandy desertified land 7.909h104 km2
in area, accounting for 20.5% and 3.0% respectively;
very severely sandy desertified land 6.756h104 km2 in
area, accounting for 17.5% and 3.1% respectively.
Potential to slight and moderate degrees of sandy de-
Study of spatial distribution of sandy desertification in North China in recent 10 years
Region
Monitoring
area/km2
83
Table 3 Sandy desertification area in North China in 2000
The area of
The area of
The area of The area of very
moderately
potential to
severely deserti- severely desertidesertified
slightly desertified land/km2
fied land/km2
land/km2
fied land/km2
17890.00
852.00
1990.00
161.00
Total area of
desertified
land/km2
Hulun Bair sandy land
83615.0
Songnen sandy land
51588.0
1909.76
1386.25
460.43
8.94
3765.00
7.3
Horqin sandy land
Xilin Gol Meng in Inner
Mongolia
Ulaqab Meng in Inner
Mongolia
Bashang region
10563.8
30669.32
9008.79
5815.42
4673.99
50167.52
47.5
181309.8
20999.21
11300.09
7274.83
5595.37
45169.49
24.9
60967.9
9079.36
3782.65
278.44
41.57
13182.02
21.6
Hobq sandy land
Mu Us sandy land
20893.00
Percentage in
monitoring
area/%
25.0
46013.0
7824.30
3680.73
1302.42
243.87
13051.33
28.4
87158.5
5214.02
13025.43
5705.72
2338.44
26283.59
30.2
97352.0
20509.82
14333.78
7949.56
10679.33
53472.49
54.9
Shiyang River Basin
120172.0
2243.32
3692.43
16704.70
10005.36
32645.81
27.2
Heihe River Basin
202946.0
352.72
1568.20
2852.47
10090.82
14864.2
7.3
Kumtag region
172731.0
2594.94
4341.53
1823.95
325.14
9085.56
5.3
89996.0
7728.00
2586.00
1314.00
1377.00
13005.00
14.5
Qaidam region
446562.7
3008.38
11170.07
2835.49
2606.50
1962.44
4.4
North Xinjiang
272552.0
5715.39
13609.29
12948.15
9046.60
41319.43
15.2
Middle Xinjiang
158843.0
2238.78
2502.04
4397.90
6552.30
15691.02
9.9
Three-river Source area
South Xinjiang
Total monitoring area
386651.0
1290.28
2929.70
5437.88
3813.08
13470.94
3
2564062.0
139267.60
90768.98
78608.85
79091.35
385686.80
15.0
sertified land occupied over 60% of total sandy desertified land area, which showed that in the arid and
semi-arid zones with fragile eco-environment in North
China a large area has entered the developed stage of
sandy desertification and is facing the danger of severity sandy desertification. As primary grassland vegetation occurs degradation, or some initial sandy desertification symptoms such as bare spots, irrational human
activities will aggravate the process and thereby lead
to the occurrence of moving sand or the formation of
sand dunes. When desertification reaches the severe
stage, it is very difficult to restore and control. Compared to the monitoring results in 1987, sandy desertification in some regions has reversed through more
than 10 years of rehabilitation, sandy desertified land
area has decreased and its severity also reduced. But
from an overall point of view, sandy desertification
development is faster than its reverse. Especially over
the past 10 years sandy desertification development
significantly sped up. In 2000 the total area of sandy
desertification increased 46740 km2 and reached at
least 38.57h104 km2, or the area of average annual
increasement was 3595 km2 (fig. 2).
3 Evolutionary feature and trend of sandy desertification in recent 10 years
Basing on our previous studies we can conclude
that the developmental features of sandy desertification in North China over the past 50 years are the expansion in area of sandy desertification and the quickening in developmental rate. As the development rate
of sandy desertification is concerned, from the late
1950s to 1975 its annual development rate in area was
1560 km2ga−1, from 1975 to 1987 was 2100 km2ga−1,
and from 1988 to 2000 about 3600 km2ga−1. The distribution area of desertified land in some representative regions in North China during 1987ü2000 is
presented in table 4.
According to the analytical results of table 4, the
general evolutionary trends of sandy desertification in
North China from 1987 to 2000 are as follows:
(i) In the key monitoring regions sandy desertification shows a continuous developmental trend, for
example in the reclamation district of Bashang Grassland in Hebei Province, reclamation district in Qahar
Grassland and grazing zone, southwest Ordos Grass-
Fig. 2. Sketch map of sandy desertification in North China.
84
Science in China Ser. D Earth Sciences
Study of spatial distribution of sandy desertification in North China in recent 10 years
85
Table 4 Developmental status of sandy desertifiction in representative regions in North China in recent 10 years
1987
Region
East part of reclamation district in
Bashang
West part of reclamation district in
Bashang
Reclamation district and grazing area
in Qahar grassland
Horqin region (Jirem Meng and
Uloanqab Meng)
Houshan grassland grazing area and
reclamation in Ulanqab Meng
Ordos grassland in Ih Ju Meng
Shenmu, Hengshan, Jingbian, and
Dingbian
Southwest part (Yanchi) of Ordos
grassland
Lower reach of Heihe River in west
Alxa Desert
Piedmont Plain of Kunlun Mt. in Qaidam Basin
Margin of Gurbantunggut Desert
Margin of Taklimakan Desert
2000
Monitoring
area/km2
Sandy desertification area/km2
Percentage of
monitoring area/%
Monitoring
area/km2
3471
1336.6
38.5
17715
2213.6
12.5
3272
23.6
13803
4756.8
34.5
5992.9
66.1
28957.7
14148.2
48.9
50198
47.5
13833
9056
Desertification
Percentage of
area/km2
monitoring area/%
61008
57.8
105604
46660
18121.2
38.8
60968
13182.1
21.6
49112.4
45973
93.6
64453
45751.4
80.1
8221.7
34.9
105573
18046.4
8166.9
45.3
23547
6761.2
1845.5
27.3
6744
3495
51.8
16200
5955
36.8
82596.3
11435
13.8
7920
5573
70.4
74360.4
8694
11.7
/
272552
41319.4
15.2
/
386651
13471
land, downstream area of the Heihe River in Alxa Desert and Piedmont Plain Kunlun Mountain in Qaidam
Basin.
3.5
(iii) In some representative regions sandy desertification was severely developed before 1987, but it has
been controlled through the rehabilitation during 1988
ü2000, for example, Horqin region (in Jirem Meng of
Inner Mongolia) and Mu Us Sandy Land. On the
whole, it appears as two situations, i.e. overall reverse
and local reverse.
Horqin sandy land. According to previous and present
study results, the spatial distribution changes of sandy
desertification in Horqin sandy land over the past 50
years are presented in table 5. It can be seen that area
of sandy desertification increased from 42300 km2 in
1959 to 51384 km2 in 1975, 61008 km2 in 1987 and
50198 km2 in 2000. That is to say, it decreased by
10810 km2 or 17.7% in area in recent ten years compared to that of the previous stages. Very severe and
slightly sandy desertified land area increased 393 km2
and 1749 km2 respectively, while severe and moderate
sandy desertified land area showed a decreasing tendency, decreased 488 km2 and 12463 km2 respectively.
It should be pointed out that the decrease in sandy desertified land area occurred in all Horqin sandy land
rather than concentrated in several countries (Qis).
1) Overall reverse. The case of overall reverse is
The overall reverse of sandy desertification in
(ii) In some representative regions sandy desertification exhibited a slow spread trend before 1987, but
in recent ten years it kept a stable state, for example,
Ordos Grassland in Ih Ju Meng of Inner Mongolia and
Shenmu-Hengshan-Jinbian-Dingbian region.
Table 5 Dynamical changes of sandy desertified land in Horqin sandy land
Area of very severe
desertification/km2
Area of severe
desertification/km2
Area of moderate
desertification/km2
Area of slight
desertification/km2
Total
area/km2
1975
2829
7885
22495
18175
51384
+21.47
1987
5162
5422
28950
28950
61008
+18.72
2000
4674
5815
30699
30699
50198
−17.72
Year
1959
Compared to previous
stage/%
42300
86
Science in China Ser. D Earth Sciences
Horqin sandy land attributed to such a factor that local
government, farmers and herdsmen were increasing
aware of the seriousness of sandy desertification and
therefore putted into large amount of man power, material and financial resources to rationally readjust
land-use structure and intensity, cutted down extensively managed cultivated land area, increased intensively managed fertile land area, returned croplands to
forest and grassland, closed large area of pastureland
to livestock grazing and arrest overgrazing. The use of
all these effective measures not only slowed down the
spread rate of sandy desertification but also improved
large area of sandy desertified land.
2) Local reverse. The case of local reverse is Mu
Us sandy land. Table 6 represents the general trend of
the dynamical changes of sandy desertification in Mu
Us sandy land in past 30 years. Sandy desertification
in Yulin, Hengshan, Jingbian and Shenmu significantly
reversed in the past 30 years. Especially in Yulin the
reversing rate of sandy desertification greatly increased after the middle 1980s, the mean annually reversing rate increased from 48.17 km2ga−1 during the
mid-1970s and mid-1980s to 59.21 km2ga−1 during
last decade. Although the monitoring region in Hengshan, Jingbian and Shenmu included loess hills which
led to a significant reduction in desertification percentage compared to previous two stages, the monitoring results of the three stages have higher comparability, because sandy desertified land mainly occurred
in the non-loess hill zones. Sandy desertified land in
Hengshan country decreased from 1596.5 km2 in area
in 1970s to 1292.4 km2 in 1980s and further decreased
to 782.5 km2 in 2000, the annually reversing rate in
first ten years was 30.41 km2ga−1 and in the later 15
years was 39.99 km2ga−1. Unlike Hengshan, sandy
desertification annually reversing rate in Jingbian was
slow, as during the first ten years it was 10.2 km2ga−1
and in the later 15 years was 39.23 km2ga−1, and
in Shenmu 42.71 km2ga−1 and 10.05 km2ga−1 respectively. In one word, except the Shenmu Country
that had a slower reversing rate of sandy desertification after the mid-1980s, the reversing rate of sandy
desertification in other three countries exhibited an
accelerating tendency.
Unlike the situation in the four countries mentioned above, sandy desertification in other countries
of Mu Us sandy land exhibited such an evolutionary
tendency: from the mid-1970s to the mid-1980s sandy
desertification experienced a reverse process but from
the mid-1980s to 2000 it went through a development
process, and the annually increasing rate in Dingbian
country was 37.68 km2, in Yanchi country 109.93 km2
and in Uxin Qi 81.2 km2.
The overall reverse of sandy desertification in
Horqin sandy land and local reverse of sandy desertification in Mu Us sandy land are attributed to local
government and people’s hard work to combat desertification. They increased desertification control input,
readjusted land-use structure and intensity, cut down
Table 6 Dynamical changes of sandy desertification in Mu Us sandy land
Region
Yulin
mid-1970s
mid-1980s
Percentage in
Monitoring Sandy deserti2
2 monitoring
area/km fied area/km
area/%
6551
5729.8
87.5
Percentage in
Monitoring Sandy desertimonitoring
2
2
fied area/km
area/km
area/%
6551
5248.1
80.1
2000
Monitoring
area/km2
Sandy desertified area/km2
6891
4360
782.5
Percentage in
monitoring
area/%
63.3
Hengshan
2584
1596.5
61.8
2584
1292.4
50.0
4219
Dingbian
4493
2093.5
46.6
4493
1729.8
38.5
6847
2295
33.5
18.6
Jingbian
3485
1917.5
55.0
3485
1815.5
52.1
4972
1227
24.7
Shenmu
4463
3772.9
84.5
4463
3345.8
75.0
7509
3195
42.5
Yanchi
6761
1369
20.2
6761
1846
27.3
6744
3495
51.8
Uxin Qi
11645
10164
87.3
11645
9561
82.1
11627
10779
92.7
Otog Qi
5251
4721
89.9
5251
4165
79.3
20245
13103
64.7
Otog Qianqi
7713
6321
82.0
7713
6306
81.8
12321
11426
92.7
Study of spatial distribution of sandy desertification in North China in recent 10 years
cultivated land area, practiced intensive management,
planted trees and grasses, closed large area of pastureland and arrested overgrazing. As a result, considerably large area of sandy desertified land was rehabilitated and restored production potential to a certain
degree. However, the rehabilitated effect was also affected by natural conditions. Horqin sandy land is located in the eastern part of semi-arid zone of China,
mean annual precipitation increases from 340 mm in
the north to 450 mm in the south, mean annual temperature varies between 6.0ü8.5ć. Relatively speaking, the water and heat conditions in Horqin sandy
land area are better than those of Mu Us sandy land
and therefore favorable to the reversing of sandy desertification processes. This is one of the reasons that
the Horqin sandy land achieved a better sandy desertification control effect.
By the year 2000, the area of potential to slight
and moderate degrees of sandy desertified land in
North China has been over 60% of China’s total sandy
desertified land, the area of severe and very severe
sandy desertified land less than 40%. To a certain degree this is a favorable factor, because potential to
slightly and moderately desertified lands have slight
degradation degree and are easy to rehabilitate as long
as human irrational activities are arrested and effective
control measures are adopted. Field investigation and
remote sensing monitoring in Horqin sandy land and
Mu Us sandy land in 2000 demonstrated that local
government and people putted into large amount of
man power, material and financial resources to combat
potential to slightly and moderately desertified lands
and achieved encouraged results. If such experiences
and methods are widely popularized, about 60% sandy
desertified land in North China can be controlled in a
short period.
4
Conclusions
Sandy desertification is land degradation characterized by wind erosion mainly resulted from the excessive human activities in arid, semiarid and part of
sub-humid regions in North China. In this study a
classification system of sandy desertification was perfected and used in the monitoring of sandy desertifica-
87
tion in North China. Remote sensing monitoring results showed that in 2000 sandy desertification in
North China on the whole still exhibited a developmental trend, its area reached 38.57h104 km2. The
area of potential to slightly sandy desertified land was
13.93h104 km2, accounting for 36% of total sandy
desertified land area; moderate sandy desertified land
9.977h104 km2, accounting for 25.9%; severe sandy
desertified land 7.909h104 km2, accounting for 20.5%;
and very severe sandy desertified land 6.756h104 km2,
accounting for 17.5%. Sandy desertification mainly
occurred in the semi-arid mixed farming-grazing zone
and its northern grassland livestock-raising zone,
semi-arid dryland cropping zone and arid oasis-desert
margin zone. All these zones are the priority regions to
implement sandy desertification projects in recent
years. Since the late 1950s sandy desertification in
North China has been exhibiting a rapidly developed
tendency, annual development rate of sandy desertification enhanced from 1560 km2ga−1 in the period
from the 1950s to the mid-1970s to 2100 km2ga−1 in
the period from the mid-1970s to the late 1980s and
further to 3600 km2ga−1 in the period from the late
1980s to 2000. “Overall deterioration while local rehabilitation” is the basic status of sandy desertification
in North China. Existing sandy desertification control
achievements and monitoring assessment show that
about 60% of sandy desertified land in North China
can be controlled under the condition of rational
land-use ways and intensity.
Acknowledgements This work was supported by the “973” Programme of China (Grant No. TG2000048705).
References
1. United Nations, United Nations Convention to Combat Desertification in Those Countries Experiencing Serious Drought and/or
Desertification Particularly in Africa (in Chinese), Beijing: Forestry Press, 1995, 1ü5.
2. Research Group of “Study on Combating Desertification/Land
Degradation in China”, Study on Combating Desertification/Land
Desertification in China, Beijing: China Environmental Science
Press, 1988, 60ü164.
3. Zhu, Z. D., Liu, S., Wu, Z. et al., Introduction to Deserts in China,
Beijing: Science Press, 1980, 20ü29.
4. Zhu, Z. D., Liu, S., Desertification and Its Control in China, Beijing: Science Press, 1989, 1ü122.
88
5. Zhu, Z. D., Wang, T., Theory and practice in the study of desertification in China, Quaternary Research (in Chinese), 1992, 2: 97
ü106.
6. Wang, T., Zhao, H. L., Xiao, H. L., Advance in sandy desertification study in China, Journal of Desert Research (in Chinese), 1999,
19(4): 299ü311.
7. Wang, T., Zhu, Z. D., Several problems on sandy desertification in
North China, Quaternary Research (in Chinese), 2001, 1: 56ü65.
8. Reining, P., Handbook on desertification indicators: Based on the
science, Associations’ Nairobi Seminar on Desertification (ed.
Priscilla, R.), Washington: American Association for the Advancement of Science, 1978, 141.
9. Food and Agriculture Organization and United Nations Environment Programme, Provisional Methodology of Assessment and
Mapping of Desertification, Rome: FAO Press House, 1983, 84.
10. Middleton, N., Thomas, D., World Atlas of Desertification, London: Baltimore Edward Arnold, 1992, 84.
11. Michael, S., Niamb, M., Handbook for the Field Assessment of
Land Degradation, Sterling: Earthscan Publications, 2001, 169.
12. Tucker, C. J., Dregne, H. W., Newcomb, W. W., Expansion and
contraction of Sahara desert from 1980 to 1990, Science, 1990,
253: 299ü301.
13. Hanna, N. R., Assessment of desertification around deep wells in
Sahel using satellite image, Journal of Applied Ecology, 1991, 28:
173ü186.
14. Hutchinson, C. F., Use of satellite data for famine early warning
in sub-Saharan Africa, International Journal of Remote Sensing,
1991, 12(6): 1405ü1421.
15. Malo, A. R., Nicholson, S. E., A study of rainfall and vegetation
dynamics in the Africa Sahel using normalized vegetation index,
Journal of Arid Environment, 1990, 19: 134ü142.
16. Nicholson, S. E., The influence of soil type on the relationship
between NDVI rainfall and soil moisture in semi-arid Botswanan,
NDVI response to rainfall, Remote Sensing Environment, 1994,
50: 107ü120.[DOI]
17. Palmer, A. R., van Rooyen, A. F., Detecting vegetation change in
the Southern Kalahari using Landsat TM data, Journal of Arid
Environments, 1998, 39(2): 143ü154.[DOI]
18. Del Valle, H. F., Elissalde, N. O., Gagliardini, D. A. et al., Status
of desertification in the Patagonian region: Assessment and mapping from satellite imagery, Arid Soil Rehabilitation, 1998(12): 75
ü122.
Science in China Ser. D Earth Sciences
19. Ci, L. J., Wu, B., Classification of desertification climatic type
and establishment of potential occurrence range in China, Journal
of Desert Research (in Chinese), 1997, 17(2): 107ü111.
20. Gao, S. W., Wang, B. F., Indicator system for monitoring and assessment of sandy desertification in China, Forestry Science (in
Chinese), 1998, 34(2): 1ü10.
21. Dong, Y. X., Approach to the indicator system for monitoring land
desertification, Monitoring of Arid Environment (in Chinese),
1992, 6(4): 234ü237.
22. Li, F., Landscape ecological theoretic analysis in desertification
monitoring, Arid Zone Research (in Chinese), 1997, 14(1): 69ü
73.
23. Wang, C. Y., Bu, H., Di, X. C., Role of remote sensing technique
in the study of global environmental changes, Advance in Earth
Science (in Chinese), 1988, 13(3): 278ü284.
24. Wang, X. L., Bao, Y. H., Approach to the study method of landuse
dynamical changes, Advance in Geographical Science (in Chinese), 1999, 18(1): 81ü87.
25. Wang, X. Z., Study of classification code of spatial data of information system for monitoring and evaluation of desertification
disasters, Journal of Desert Research (in Chinese), 1994, 14(1): 41
ü44.
26. Yan, C. Z., Feng, Y. X., Establishment of interpretation marks of
TM images of land resource types in Northwest China, Journal of
Desert Research (in Chinese), 1999, 19(3): 206ü209.
27. Ma, L. P., Han, G. Q., Application of TM images in the investigation of desertification in Hexi region, Journal of Desert Research
(in Chinese), 1996, 16(4): 400ü406.
28. Dong, Y. X., Preliminary study of the judgment of desertification
degrees and zoning in China, Journal of Desert Research (in Chinese), 1995, 15(2): 170ü174.
29. Zhu, Z. D., Wang, T., Analysis of desertification evolution trend
over past 10 year through the investigation in several representative regions in China, Acta Geographic Sinica (in Chinese), 1990,
45(4): 430ü440.
30. Wang, T., Wu, W., Remote sensing monitoring and assessment of
sandy desertification, Quaternary Research (in Chinese), 1998, 2:
108ü118.
31. Wang, T., Zhu, Z. D., Wu, W., Sandy desertification in the North
China, Science in China, Series D, 2002, 45(Supp): 23 ü
24.[Abstract][PDF]