Ecological Engineering 35 (2009) 1719–1726

Contents lists available at ScienceDirect

Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng

Influence of hydrology process on wetland landscape pattern:

A case study in the Yellow River Delta
Sheng-nan Li a,c , Gen-xu Wang a,∗ , Wei Deng a , Yuan-man Hu b , Wei-Wei Hu a,c
a
Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, Sichuan Province, China
b
Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, 110016, Liaoning Province, China
c
Graduate School of the Chinese Academy of Sciences, Beijing 100039, China

a r t i c l e i n f o a b s t r a c t

Article history: A comprehensive study using ecological engineering analysis was conducted on the influence of the
Received 9 November 2008 hydrology process in the Yellow River Delta. It was found that water and sediment played an important
Received in revised form 25 June 2009 role in the formation and maintenance of the estuarine wetland. Based on hydrological data (1950–2005),
Accepted 20 July 2009
meteorological data (1954–2005) and landscape data produced from integrated Landsat TM images of the
Yellow River Delta, the relationship among all the above factors has been analyzed. The results indicated
that runoff, sediment discharge and the area of the reed marsh wetland, meadow wetland and tidal wet-
Keywords:
land had evidently decreased from 1986 to 2001. The runoff and sediment discharge into the Yellow
Water and sediment discharge
Wetland landscape area
River Delta at a rate of 200–300 × 108 m3 and at 5–8 × 108 t, respectively, were probably the most optimal
Dominant factor range for maintaining the stable wetland landscape pattern. Regression analysis and principal component
Regression analysis and principal analysis showed that there was a strong positive correlation between the area of wetland landscape and
component analysis runoff, sediment discharge, whereas there was a negative correlation between the area and the temper-
Yellow River Delta ature, and there was no significant trend with precipitation levels. The water and sediment discharge
are the dominant variable factors of the wetland. Human activities also have an important influence on
transformation of wetland types as well as wetland degradation. Therefore, with economic development,
climate change and sustainable utilization of resources, great attention should be paid to the changes of
natural landscape and their causes.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction wetlands must emphasize riverine inputs of freshwater and sed-

iments.
There is increasing public interest in investigating runoff and As a special ecosystem, the wetland is the outcome of interac-
riverine sediment flux to the ocean and the process of sedi- tions between terrestrial and aquatic ecosystems under integrated
ment migration, partially responding to global climate change, effects of various ecological processes at different scales. It plays an
the explosive growth of the world population and the concur- important role worldwide in nutrient cycling, sediment accretion,
rent economic boom in recent years (Walling and Fang, 2003). It pollution filtration, and erosion control (Qin and Mitsch, 2009).
has been reported that the changes in runoff and sediment dis- Wetland hydrology, which has direct impacts on the biota in the
charge have resulted in the formation and degradation of immense wetland and changes on the landscape ecological pattern, is the
areas of the deltas, and exerted great influences on the envi- most important determinant in establishing and maintaining spe-
ronment around river mouths including those of the Pearl River cific types of wetlands (Mitsch and Gosselink, 2007; Bohemen,
(Zhang et al., 2008), Yangtze (Yang et al., 2002, 2006), Yellow 2007). The estuary area, a junction area of land–ocean and an
River (Wang et al., 2006a,b, 2007; Fan et al., 2006; Saito et al., intersection area of freshwater and salt water, is affected by the
2001), and Mekong River (Ta et al., 2002). Day et al. (2000) have interaction between the land and the ocean. Based on the com-
also indicated that the wetland loss in the Mississippi Delta is plex dynamic mechanism, there exist many unique wetland types
an ongoing complex process involving several interacting fac- and eco-micro-environments that are interrelated in this kind of
tors, and the efforts to create and restore Louisiana’s coastal areas.
The Yellow River is the second largest river in the world in
terms of sediment load over the last few thousand years (Milliman
∗ Corresponding author. Tel.: +86 28 85233420; fax: +86 28 85233420. and Meade, 1983). It shows dominant effects on the formation and
E-mail address: gxwang@lzb.ac.cn (G.-x. Wang). maintenance of the regional hydrology and ecosystem in its estuary

0925-8574/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.ecoleng.2009.07.009
1720 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726

Fig. 1. Location of the Yellow River Delta.

area. Water and sand sedimentation from the Yellow River formed wetland evolution (Allen, 1997; D’Alpaos et al., 2005). Other works
an important base for the extension and development of the Yellow explicitly included the role of wave action in eroding exposed wet-
River Delta (Huang et al., 2005) and greatly affected the wetland lands to modulate the response of wetlands to the sea level rise
landscape patterns of this delta. Because of the large amount of (Ruth and Pieper, 1994; Schwimmer and Pizzuto, 2000). The main
deposition associated with the Yellow River, the delta area had an objective of this study was to investigate the effect of runoff and
accretion by the land extension at a rate of 32.4 km2 /a before 1979, sediment discharge on the wetland landscape pattern in the Yellow
and then with a rate of 2.7 km2 /a because of the channel diversion of River Delta.
the Yellow River and variation of flow and sediment in the estuary,
especially with a long drought period during 1996–2002 (Zhang 2. Methods and materials
and Hu, 2007). The estuary wetland of the Yellow River Delta is not
only the most complete estuary wetland, but also the youngest 2.1. Study area
wetland ecosystem in the warm-temperate zone in China, with
immature, fragile and unstable characteristics. Some research has The Yellow River Delta (117◦ 31 –119◦ 18 E, 36◦ 55 –38◦ 16 N) is
been focused on the Yellow River Delta, including runoff changes situated in the northeast of Dongying City, Shandong Province,
(Wang et al., 2006a,b), sediment transport, deposition and erosion China. It faces the Bohai Sea on the north and borders Laizhou Bay
at the estuary delta (Wang et al., 2007; Chu et al., 2006), evolu- on the east, with an average elevation less than 15 m (Fig. 1). The
tion and variation in land accretion of the delta (Xu, 2007; Ye et al., evolution of the Yellow River Delta is influenced by the river dis-
2007), changes of the land coverage and the landscape pattern (Xu charge, suspended sediment load, and changes of the river channel
et al., 2002; Ye et al., 2004), the formation process, protection of the (Wang et al., 2006a,b). Since the river changed its lower reach from
wetland ecosystem (Chen, 2005) and the subaqueous delta (Liu et Xuhuai route to the Bohai Sea in 1855, a vast delta with an area
al., 2004; Yang and Liu, 2007). Even though extensive studies exist of 6000 km2 has been formed. The fan-delta is composed of many
in the literature, little attention has been paid to the influence of delta lobes and it has a radiate shape resulting from a swing of the
the hydrological-climatic factors on the wetland landscape pattern frequent distributary shifts. Around each channel mouth, a deltaic
in the Yellow River Delta. Most studies have been focused on tidal lobe was formed. The avulsion points or the apices of the deltaic
salt marshes with tidal currents as the major factors controlling the lobes were located near Ninghai before 1934 and have been near
 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726 1721

Fig. 2. The landscape and its changes in the Yellow River Delta from 1986 to 2001.

Yuwa since then (Xu, 2007). The current course of the Yellow River Yellow River delta, and within the 120 km radius from Dongyi-
has resulted from the artificial change from the Diaokou course ing City. Each dataset contains the monthly and annual values
to the Qingshuigou course in 1976, followed by a shift towards of precipitation and temperature from 1954 to 2002. The aver-
the north bank of the Qingshuigou course in 1996. As a result, the age annual and monthly precipitation and temperature at the four
entire modern Yellow River Delta is characterized by complicated meteorological stations are used as the values in the Yellow River
patterns of erosion and accretion together with wetland landscape Delta.
changes (Zhang and Hu, 2007). The satellite images can easily discern the difference in the char-
It has a warm-temperate and continental monsoon climate acteristics between the image of mangrove and other vegetation
with an annual precipitation of 596.9 mm, annual evaporation types. The logistical disadvantages of monitoring long-term veg-
of 1900–2400 mm and annual average temperature of 12.9 ◦ C. etation community changes as well as evaluating an inaccessible
No zonal vegetation type exists in this region. The most impor- area may be overcomed by applying remote sensing techniques
tant factors of the distribution of vegetation in this area include (Lee and Yeh, 2009). Three Landsat TM images of the Yellow River
water, sediment discharge, soil salinity, underground water depth, Delta in the years of 1986, 1996, and 2001 were used to extract
mineral intensity, and geomorphology. According to the spectral the vegetation areas at the study site. Based on the field data and
distribution of vegetation succession in Yellow River Delta (Ye et the landscape data from supervised classification on the basis of
al., 2004), various vegetation types develop perpendicularly to the integrated Landsat TM images, three land-cover maps have been
riverbed of the Yellow River by the sequence of succession, and compiled at the same scale to study the changes of land-cover
wriggle upwards along with the river pathway from the sea to the and the landscape pattern (Fig. 2). According to the Ramsar Con-
land. vention on wetlands, other related classification of wetlands (U.S.
EPA, 2002), and the wetland landscape characteristics of the Yel-
2.2. Datasets and methods low River Delta, the landscape patterns in the delta were divided
into the following 9 types: reed marsh wetland (including Phragmi-
2.2.1. Data sources tas communis Trin. etc.), sparse forest wetland (including Salix spp.,
We obtained hydrological data (1950–2005) of the annual water Robinia pseudoacacia L., etc.), meadow wetland (including Tamarix
discharge and sediment load from four stations in the lower reach chinensis Lour., Suaeda salsa Pall., etc.), tidal wetland (including tidal
of the Yellow River, i.e., Huayuankou, Sunkou, Luokou and Lijin sta- flat, subtidal zone, tidal creek, etc.), water wetland (including river,
tions in a upstream-to-downstream sequence (Fig. 1). Each data lake, reservoir, pond, etc.), artificial wetland (including salt field,
set contains the monthly and annual values from January 1950 to shrimp-crab field, etc.), farmland, residential area and construc-
December 2005. Among the four stations, Lijin hydrological station tion land (including oil well, road and dyke, etc.). Landscape types
is the last gauge for water and sediment entering Bohai Sea in the and their distribution in the study region are shown in Fig. 2. In the
Yellow River Delta, which is about 100 km upstream from the river Yellow River Delta, three main wetland landscape types are reed
mouth. marsh, meadow, and tidal wetland, accounting for 18.55%, 10.79%,
Because meteorological data are unavailable in the Yellow River and 6.17% of the total area in 2001, and they have been decreasing
Delta, four meteorological datasets from meteorological stations by 14.1, 23.7 and 14.4 km2 /a, respectively. However, the artificial
in Zibo, Huimin, Weifang and Yangjiaogou have been processed wetland shows an increasing trend, accounting for 11.88% of the
as the reference. Those meteorological stations are around the total area in 2001.

Table 1
Regression analysis and models of the wetland landscape area and their runoff or sediment discharge.

Wetland type Regression equation Variable explanation Pearson

Reed marsh wetland FR = −4E−0.5 Qr2 + 0.0182Qr + 11.349 FR : area of reed marsh wetland; Qr : runoff; Qs : 0.7509**
FR = −0.032Qs2 + 0.4576Qs + 11.741 sediment discharge; FM : area of meadow 0.8001**
Meadow wetland FM = −6E−0.5 Qr2 + 0.0267Qr + 6.3625 wetland; FC : area of tidal wetland 0.7275**
FM = −0.024Qs2 + 0.5071Qs + 7.0622 0.7869**
Tidal wetland FC = −4E−0.5 Qr2 + 0.018Qr + 3.8463 0.7554**
FC = −0.0289Qs2 + 0.4323Qs + 4.2506 0.8031**
**
Correlation is significant at the 0.01 level (2-tailed).
1722 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726

Fig. 3. Relationship between runoff (a–c) or sediment discharge (1–3) and the area of the main wetland types.

2.2.2. Methods for analysis dimensional reduction of multivariable data with correlated vari-
Because of the asymmetric and nonlinear changes of the wet- ables and can be used to identify and extract the dominant modes
land landscape area from 1986 to 2001 in the Yellow River Delta, of variability from consistently sampled data sets (Preisendorfer,
linearization of different curves was used to establish the piece- 1988; Jackson, 1991).
wise difference and thus to obtain the area-time series of the spatial
distribution pattern of the main wetland landscape types to corre-
3. Results
late the results with the hydrological process. Since the changes
of wetland patterns reflect responses of the wetland system to
3.1. Impacts of runoff and sediment discharge on the wetland
various hydrologic conditions, principal component analysis and
landscape pattern
regression analysis are applied to quantify these responses in the
Yellow River Delta. These techniques have also been utilized to
To synchronously analyze the wetland spatial distribution,
estimate contributions of hydrological controls in the evolution of
runoff, and sediment discharge, we selected the reed marsh,
the coastal geomorphology (Cuadrado and Perillo, 1997; Chen et
meadow and tidal wetland as the representative types in this region
al., 2008). Principal component analysis was originally devised for
to conduct our research. The regression analysis results indicated

Table 2
Regression analysis and models of the wetland landscape area and their temperature or precipitation.

Wetland type Regression equation of the Variable explanation Relationship between Relationship between
area and temperature temperature and the area precipitation and the area

Pearson Sig. Pearson Sig.

Reed marsh wetland FR = 0.1665T2 − 5.3127T + 54.019 T: temperature −0.566* 0.035 0.098 0.738
Meadow wetland FM = −0.1461T2 + 2.4025T + 2.5574 −0.560* 0.037 −0.028 0.923
Tidal wetland FC = 0.1175T2 − 4.0216T + 38.013 −0.570* 0.033 0.077 0.795
*
Correlation is significant at the 0.05 level (2-tailed).
 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726 1723

Table 3
Multiple regression analysis between the wetland landscape area and hydrological-climatic parameters.

Wetland type Multiple regression equation Variable explanation R2

Reed marsh wetland FR = 20.431 + 0.002Qr + 0.184Qs − 0.637T FR , FM , FC , Qr , QS , T: area of reed marsh wetland, 0.7574**
Meadow wetland FM = 17.521 + 0.008Qr + 0.081Qs − 0.783T area of meadow wetland, area of tidal wetland, 0.7792**
Tidal wetland FC = 12.561 + 0.002Qr + 0.158Qs − 0.611T runoff, sediment discharge, temperature 0.7654**
**
Correlation is significant at the 0.01 level (2-tailed).

that the area of reed marsh, meadow and tidal wetland showed Table 4
The contribution percentage of the principal components.
significant multinomial correlations with runoff or sediment dis-
charge in the Yellow River Delta from 1986 to 2001 (Table 1). In Component Total eigenvalue % of variance Cumulative, %
other words, the changes of spatial distribution of the main wetland 1 1.865 62.168 62.168
ecosystem landscape were closely and positively related to runoff 2 0.940 31.326 93.494
and sediment discharge into the delta (p < 0.01). The correlation 3 0.195 6.506 100.000
coefficients between the area of the reed marsh, meadow and tidal Extraction method: principal component analysis.
wetland and annual runoff were 0.7509, 0.7275 and 0.7554, respec-
tively (p < 0.01), and between sediment discharge were 0.8001,
wetlands increased with runoff and sediment discharge entering
0.7869 and 0.8031, respectively (p < 0.01). Although the spatial dis-
into Yellow River Delta, but decreased with the temperature.
tribution of the wetland ecosystem prominently expanded due to
In order to extract the main influence factors on the evolution
runoff and sediment discharge into the river mouth, the wetland
of the wetland pattern, we selected runoff, sediment discharge and
landscape pattern slowly and gradually showed a steady increase
temperature for principal component analysis. The three princi-
when the mean annual runoff and sediment discharge were higher
pal components were showed in eigenvalues from big to small
than 200–300 × 108 m3 and 5–8 × 108 t, respectively, while it had a
in Table 4. The result indicated that the first principal component
trend of sharp decrease when the mean annual runoff and sediment
(eigenvalue > 1) with a contribution rate of 62.168% has a high load
discharge were lower than 150 × 108 m3 and 4 × 108 t, respectively.
of 0.943 and 0.914 at runoff and sediment discharge, respectively
These results to some extent indicated that runoff at a rate of
(Table 5). Therefore, the dominant factors which have great influ-
200–300 × 108 m3 and sediment discharge at a rate of 5–8 × 108 t
ences on the changes of the main wetland landscape pattern are
into the Yellow River Delta might be the most appropriate range in
runoff and sediment discharge in the Yellow River Delta. The first
maintaining the stability of the wetland landscape pattern (Fig. 3).
principal component can be expressed by the linearity function as
follows:
3.2. Impacts of climate conditions on wetland landscape pattern
F1 = 0.943x1 + 0.914x2 − 0.376x3 (1)
The precipitation and temperature from 1954 to 2002 stayed at
where F1 represents the first principal component; x1 , x2 and x3 are
very stable levels in the Yellow River Delta. From 1986 to 2001, the
runoff, sediment discharge and temperature, respectively.
annual mean temperature increased by 0.012 ◦ C with the highest
temperature in July (26.1 ◦ C) and the lowest in January (−2.8 ◦ C),
4. Discussion
whereas the mean annual precipitation exhibited a slight decreas-
ing trend at a rate of 1.41 mm/a.
4.1. Changes of runoff and sediment discharge
The regression analysis showed that there was a negative corre-
lation between the area of wetland landscape and the temperature
Compared with other large rivers in the world, the Yellow River
in the tidal, meadow and reed marsh wetlands (Table 2), with
is characterized by high sediment loads and low water discharges
a correlation coefficient of −0.570, −0.560 and −0.566 (p < 0.05),
(Wang et al., 2007). In the Yellow River Delta, runoff and sediment
respectively. When the temperature gradually increases, the main
loads play a central role in forming and maintaining the wetland.
wetland landscape area gradually decreases. Nevertheless, there
The modern Yellow River Delta, formed slightly more than 100
was no significant trend of the changes of the wetland landscape
year ago, was a highly constructive sedimentary system (Chu et al.,
area with precipitation. This suggested that precipitation had lit-
2006), mostly due to the deposition of a large quantity of sand and
tle impact on the area of the wetland landscape in the Yellow
mud transported by the Yellow River. Since the 1950s, the runoff
River Delta. Therefore, the tidal, meadow and reed marsh wetlands
data measured at Huayuankou, Sunkou, Luokou and Lijin gauging
appeared to be more sensitive to temperature changes.
stations indicated that the average water discharge into the Yel-
low River Delta steadily and synchronously decreased (Fig. 4). The
3.3. Impacts of the combination of hydrological-climatic continually decreasing trend was clearly affected by the upstream
components on the wetland landscape area discharge at Huayuankou station (Fig. 4a). Meanwhile, the runoff
at the Lijin station decreased by 5.48 × 108 m3 /a. There is a sig-
To obtain the quantitative relationship between the wetland nificant positive correlation between annual runoff and sediment
landscape area and hydrological-climatic components, multiple discharge (R2 = 0.7236, n = 56, p < 0.01) (Fig. 4b), clearly demonstrat-
regression analysis with the stepwise selection and elimination
of variables was used to establish regression models of the area
Table 5
and runoff, sediment discharge and temperature in the Yellow Component matrix.
River Delta (Table 3). These results showed a close positive relation
between runoff, sediment discharge and the area of reed marsh, Component 1

meadow and tidal wetlands. One the other hand, it also suggested Runoff .943
a negative relation between temperature and the area, with the R2 Sediment discharge .914
Temperature −.376
values of 0.7574, 0.7792 and 0.7654, respectively (p < 0.01). There-
fore, we concluded that the area of reed marsh, meadow and tidal Extraction method: principal component analysis.
1724 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726

Fig. 4. Runoff and sediment discharge measured from 1950 to 2005.

ing a decreasing trend in both water and sediment discharges, had decreased by 38%, 37% and 17%, respectively, these areas still
which can be approximated through a linear regression analysis remained as the most important ecosystem landscape types that
(Chen et al., 2008). retain the ecological processes in the Yellow River Delta. The reed
Following the same trend, the sediment discharge into the Yel- marsh, tidal and meadow wetland had generally decreased only by
low River Delta since 1950s continuously decreased with a rate at 2%, 7%, and 17.9%, respectively, from 1986 to 1996, and by 15.4%,
0.197 × 108 t/a, and the decreasing rate of sediment discharge was 33.7%, and 23.06%, respectively, from 1996 to 2001. However, the
further elevated after 1970. The annual sediment discharge in the artificial wetland, water wetland and a sparse forest wetland, which
flood seasons from 1950 to 2005 was 6.69 × 108 t, which accounted accounts for only a small proportion, all show increasing trends;
for 84.7% of the average value. Until 1990s, the sediment discharge in particular, the artificial wetland area has increased by 20 times
had been further decreased by 3.45 × 108 t/a. In the last decade, from 35.05 km2 in 1986 to 715.05 km2 in 2001.
the decreasing rate was 1.93 × 108 t/a, which has been the low- Other landscape types, such as the residential area and the con-
est period since 1996. The water–salt balance of the soil had been struction land, also played an increasing role in the whole area
disrupted because the broken river resulted in groundwater miner- except for farmland, which accounted for 39% of the total area from
alization, soil salinity increase, salt-tolerant vegetation community 1986 to 2001 (Fig. 5b). The area of farmland increased substantially
development; vegetation cover reduction and soil sandification. All before 1986, and has been gradually decreased by 13% since 1986.
of these have made the whole wetland ecosystem degraded (Huang The various changes of wetland landscape in the Yellow River Delta
et al., 2005; Chen, 2005). But the strengthening of the regulation of have been investigated by Ye et al. (2004), Yue et al. (2003) and Xu et
water and sediment by emphasizing more environment-friendly al. (2002). However, the driving factors responsible for the changes
practices of the Xiaolangdi Reservoir since July 2002 has allevi- represent complex characteristics. Seiler et al. (2005) have moni-
ated the decreasing trend of runoff and sediment discharge into the tored spatial–temporal patterns of the land use and the land-cover
Yellow River Delta, which has seen a slight increment over time. change in the Niger Inland Delta, and they showed the interactions
among pre-flood, flood and post-flood conditions strongly affected
land-cover patterns inside and around the delta.
4.2. Changes of the wetland landscape pattern

According to the landscape data of the Yellow River Delta wet- 4.3. Anthropogenic impacts
land in the years of 1986, 1996, and 2001, the spatial distribution of
the main wetland types, including reed marsh, tidal and meadow According to remote sensing data in 1986, 1996 and 2001, the
wetland, showed an evident atrophic trend (Fig. 5a). Although the major increases of farmland area in the delta were before 1986,
contributions of the reed marsh, tidal and meadow wetland areas and the farmland area decreased by 12.7% during the past 16 years.

Fig. 5. Wetland ecosystem landscape (a) and other landscapes (b) and their distributions.
 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726 1725

5. Conclusions

The Yellow River Delta is one of the most active regions with
significant land–ocean interactions among the large river deltas in
the world. Large amounts of sand are carried by the Yellow River
and deposited at the river mouth to form new land. However, data
from 1950 to 2005 indicate that runoff and sediment discharge
of the Yellow River have undergone distinct stages of decrease.
As a result, major changes are observed for the spatial distribu-
tion of wetland ecosystem landscape in the Yellow River Delta
under different water and sediment discharge conditions. The most
important types of landscape, including the reed marsh, meadow
and tidal wetland in the Yellow River Delta, decreased by 17%, 37%
Fig. 6. Relationship between the artificial wetland area and runoff. and 38%, respectively, from 1986 to 2001. These results based on the
analysis indicated a close positive correlation between the changes
of spatial distribution of the main wetland landscape types and
the variations of water and sediment discharge to the Yellow River
Moreover, owing to the artificial change of the old Yellow River Delta. The water and sediment discharge are the dominant factors
course from the Diaokou course to Qingshuigou course in 1976, of the variations of the wetland landscape patterns.
the rapidly decreasing and even discontinuous water and sediment The Yellow River Delta, a special and complicated ecosystem, is
discharge resulted in wetland degradation, infiltration of seawa- closely related to the human activity which could play both a pos-
ter, and formation of the secondary saline-alkali area around the itive and a negative role in wetland changes. The restoration zone
Diaokou course in the northern delta. Therefore, with the rapid with an area of 10000 hm2 has been established following the reg-
degradation of natural wetland occurring mainly at the northern ulation of water and sediment discharge in the Yellow River since
and southeast seashore of the delta (Fig. 2), the saline-alkali area 2002. Cui et al. (2008) have shown that because of the salt-wash,
has been explored as the artificial salina (artificial wetland) since by drawing freshwater from the Yellow River to the saline-alkali
1986 to adequately utilize the expanded area. Gu et al. (2007) have lands and the deep water flooding in most sampling sites, soil
shown that since the 1950s, the coastal wetlands around Jiaozhou salinity reached its peak value of about 0.50 g/kg, and the salinity
Bay have been severely impacted by human activities, and more level was generally lower than the minimum limit of the ecologi-
and more natural wetlands have been transformed into artificial cal thresholds of Suaeda salsa. Therefore, with efforts in economic
wetlands, which covered about 33.7% of the total wetlands around development and sustainable utilization of resources, more atten-
Jiaozhou Bay in 2002. Human activity influence on the coastal zone tion must be directed to different natural ecosystem landscapes.
has steadily increased and the coastal wetlands have been decreas- The driving forces of natural landscape can be explored and effec-
ing in surface worldwide (Valdemoro et al., 2007). Even though tive protection strategies can be proposed, such as stopping all
human activities exerted strong influences on wetlands, accord- irrational or unsustainable human activities focused only on eco-
ing to the analysis result, there is a significant negative correlation nomic growth in order to maintain the stability and improve the
between the water/sediment discharge and variations of the artifi- current wetland landscape in Yellow River Delta.
cial wetland area with a correlation coefficient of −0.7356 (n = 16)
(Fig. 6) in the Yellow River Delta, which means that although there
Acknowledgement
are many artificial disturbances to the wetland, the decreasing
water and sediment discharge in the lower reach of the Yellow
This study is funded by the National Key Basic Research
River are the dominant factors of the soil salination and the trans-
and Development Program of China (973 Program, Grant no:
formation among different wetland types.
2006CB403301).
The dominant factor of formation and maintenance of water
system in this region is the Yellow River. Datasets from 1950 to
2005 indicate that runoff and sediment discharge of the Yellow References
River have shown distinct stages of decrease under significant
Allen, J.R.L., 1997. Simulation models of salt-marsh morphodynamics: some implica-
global climate change and extensive human activities (Wang et
tions for high-intertidal sediment couplets related to sea level change. Sediment.
al., 2007). Therefore, the decrease or discontinuation of the water Geol. 113, 211–223.
and sediment discharge from the Yellow River directly affected the Bohemen, H.V., 2007. Book review: Wetlands. W.J. Mitsch, J.G. Gosselink, 4th ed.,
JohnWiley and Sons (2007). Ecol. Eng. 34, 251–252.
freshwater supply in the wetland ecosystem, leading to the imbal-
Chen, X.Y., Chen, S.L., Dong, P., Li, X.Y., 2008. Temporal and spatial evolution of the
ance of water and soil salinity and the degradation of the wetland coastal profiles along the Yellow River Delta over last three decades. GeoJournal
ecosystem (Huang et al., 2005; Chen, 2005). Mikhailov et al. (2003) 71, 185–199.
and Valdemoro et al. (2007) have described that the ensuing evo- Chen, W.F., 2005. Ecological Process in the New-born Wetlands of the Yellow River
Delta. Shandong Agricultural University.
lution is closely related to variations in the sea waves and coastal Chu, Z., Sun, X., Zhai, S., Xu, K., 2006. Changing pattern of accretion/erosion of the
dynamics, except for river water and sediment runoff in the river modern Yellow River (Huanghe) subaerial delta, China: Based on remote sensing
delta evolution. Through comparison of the hydrology and mor- images. Mar. Geol. 227, 13–30.
Cui, B.S., He, Q., Zhao, X.S., 2008. Ecological thresholds of Suaeda salsa to the envi-
phology of Vistula Delta and Red River Delta, Pruszak et al. (2005) ronmental gradients of water table depth and soil salinity. Acta Ecol. Sinica 28
have shown that the rate of development and the dimensions of (4), 1408–1418.
the delta are dependent on the discharge of water and sediment Cuadrado, D.G., Perillo, G.M.E., 1997. Principal component analysis applied to geo-
morphologyic evolution. Estuar. Coast. Shelf Sci. 44, 411–419.
in the river, whereas the evolution of the delta and geometry of Day, J.W.J.R., Shaffer, G.P., Britsch, L.D., Reed, D.J., Hawes, S.R., Cahoon, D., 2000.
the mouth are associated with wave-current motion in the coastal Pattern and process of land loss in the Mississippi Delta: a spatial and temporal
zone. In the case of the Vistula River mouth, little sediment is trans- analysis of wetland habitat change. Estuaries 23, 425–438.
D’Alpaos, A., Lanzoni, s., Marani, M., Fagherazzi, S., Rinaldo, A., 2005. Tidal network
ported to the sea and the fan-delta accumulates very slowly when
ontogeny: channel initiation and early development. J. Geophys. Res. Earth Surf.,
the river discharge is small. doi:10.1029/2004JF000182.
1726 S.-n. Li et al. / Ecological Engineering 35 (2009) 1719–1726

EPA, U.S., 2002. Methods for Evaluating Wetland Condition: Wetlands Classification. Valdemoro, H.I., Sa nchez-Arcilla, A., Jime nez, J.A., 2007. Coastal dynamics and wet-
Office of Water. U.S. Environmental Protection Agency, Washington, DC, EPA- lands stability. The Ebro delta case. Hydrobiologia 577, 17–29.
822-R-02-017. Walling, D., Fang, D., 2003. Recent trends in the suspended sediment loads of the
Fan, H., Huang, H., Zeng, T., Wang, K., 2006. River mouth bar formation, riverbed world’s rivers. Global Planet Change 39, 111–126.
aggradation and channel migration in the modern Huanghe (Yellow) River delta, Wang, H., Yang, Z., Saito, Y., Liu, J., Sun, X., 2006a. Interannual and seasonal vari-
China. Geomorphology 74, 124–136. ation of the Huanghe (Yellow River) water discharge over the past 50 years:
Gu, D.Q., Zhang, Y.Z., Fu, J., Zhang, X.L., 2007. The landscape pattern characteristics of Connections to impacts from ENSO events and dams. Global Planet Change 50,
coastal wetlands in Jiaozhou Bay under the impact of human activities. Environ. 212–225.
Monit. Assess. 124, 361–370. Wang, H., Yang, Z., Saito, Y., Liu, J., Sun, X., Wang, Y., 2007. Stepwise decreases of the
Huang, H.J., Li, F., Pang, J.Z., et al., 2005. Interaction research of land–ocean between Huanghe (Yellow River) sediment load (1950–2005): Impacts of climate change
Yellow sea and Bohai. Science Press, Beijing, pp. 61–64. and human activities. Global Planet Change 57, 331–354.
Jackson, J.E., 1991. A user’s guide to principal components. Wiley, New York. Wang, S., Hassan, M., Xie, X., 2006b. Relationship between suspended sediment load,
Liu, J., Milliman, J., Gao, S., Cheng, P., 2004. Holocene development of the Yellow channel geometry and land area increment in the Yellow River Delta. Catena 65,
River’s subaqueous delta, North Yellow Sea. Mar. Geol. 209, 45–67. 302–314.
Lee, T., Yeh, H., 2009. Applying remote sensing techniques to monitor shifting wet- Xu, J.X., 2007. Response of land accretion of the Yellow River delta to global climate
land vegetation: a case study of Danshui River Estuary mangrove communities. change and human activity. Quaternary Int. 186 (1), 4–11.
Taiwan. Ecol. Eng. 35, 487–496. Xu, X.G., Guo, H.H., Chen, X.l., Lin, H.p., Du, Q.l., 2002. A multi-scale study on land
Mikhailov, V.N., Kravtsova, V.I., Magritskii, D.V., 2003. Hydrological and morpholog- use and land cover quality change: the case of the Yellow River Delta in China.
ical processes in the Kura River Delta. Water Resour. 30 (5), 495–508. GeoJournal 56, 177–183.
Milliman, J., Meade, R., 1983. World-wide delivery of river sediment to the oceans. Yang, S., Zhao, Q., Belkin, I., 2002. Temporal variation in the sediment load of the
J. Geol. 91, 1–21. Yangtze River and the influences of human activities. J. Hydrol. 263, 56–71.
Mitsch, W.J., Gosselink, J.G., 2007. Wetlands, 4th ed. John Wiley, Inc, New York. Yang, Z., Wang, H., Saito, Y., Milliman, J., Xu, K., Qiao, S., Shi, G., 2006. Dam impacts
Preisendorfer, R.W., 1988. Principal component analysis in meteorology and on the Changjiang (Yangtze) River sediment discharge to the sea: the past
oceanography. Elsevier, New York. 55 years and after the Three Gorges Dam. Water Resour. Res. 42, W04407,
Pruszak, Z., Ninh, P.V., Szmytkiewicz, M., Hung, N.M., Ostrowski, R., 2005. Hydrol- doi:10.1029/2005WR003970.
ogy and morphology of two river mouth regions (temperate Vistula Delta and Yang, Z., Liu, J., 2007. A unique Yellow River-derived distal subaqueous delta in the
subtropical Red River Delta). Oceanologia 47 (3), 365–385. Yellow Sea. Mar. Geol. 240, 169–176.
Qin, P., Mitsch, W.J., 2009. Wetland restoration and ecological engineering: Interna- Ye, Q.H., Chen, S.L., Huang, C., Xue, Y.C., Tian, G.L., Chen, S.P., Shi, Y.N., Liu, Q.S., Liu,
tional Conference of Wetland Restoration and Ecological Engineering. Ecol. Eng. G.H., 2007. Characteristic of landscape information Tupu of the Yellow River
35, 437–441. swings and its subdeltas during 1855–2000. Sci. Chin. Ser. D 50 (10), 1566–
Ruth, M., Pieper, F., 1994. Modeling spatial dynamics of sea-level rise in a coastal 1577.
area. Syst. Dyn. Rev. 10, 375–389. Ye, Q.H., Tian, G.L., Liu, G.H., Ye, J.M., Yao, X., Liu, Q.S., Lou, W.G., Wu, S.G., 2004. Tupu
Saito, Y., Yang, Z., Hori, K., 2001. The Huanghe (Yellow River) and Changing (Yangtze methods of spatial–temporal pattern on land use change: a case study in the
River) deltas: a review on their characteristics, evolution and sediment dis- Yellow River Delta. J. Geogr. Sci. 14 (2), 131–142.
charge during the Holocene. Geomorphology 41, 219–231. Yue, T., Liu, J., Jorgensen, S., Ye, Q., 2003. Landscape change detection of the newly
Schwimmer, R.A., Pizzuto, J.E., 2000. A model for the evolution of marsh shorelines. created wetland in Yellow River Delta. Ecol Model. 164, 21–31.
J. Sediment. Res. 70, 1026–1035. Zhang, S., Lu, X., Higgitt, D., Chen, C., Han, J., Sun, H., 2008. Recent changes of water
Seiler, R., Csaplovics, E., Vollmer, E., 2005. Monitoring landcover changes of the Niger discharge and sediment load in the Zhujiang(Pearl River) Basin, China. Global
Inland Delta ecosystem (MALI) by means of ENVISAT-MERIS data. In: Huber, B.A., Planet Change 60, 365–380.
et, al. (Eds.), African Biodiversity, pp. 395–404. Zhang, Z.H., Hu, C.H., 2007. Variation of the processes of flow and sediment and its
Ta, T.K.O., Nguyen, V.L., Tateishi, M., Kobayashi, I., Tanabe, S., Saito, Y., 2002. Holocene effect on epeirogenesis of seacoast in the Yellow River estuary. Adv. Water Sci.
delta evolution and sediment discharge of the Mekong River, southern Vietnam. 18 (3), 336–341.
Quaternary Sci. Rev. 21, 1807–1819.