Urban dynamics, landscape ecological security, and policy implications: A case study from the Wuhan area of central China

Cities 41 (2014) 141–153

Contents lists available at ScienceDirect

Cities journal homepage: www.elsevier.com/locate/cities

Urban dynamics, landscape ecological security, and policy implications: A case study from the Wuhan area of central China Kehao Zhou a,b, Yaolin Liu a,⇑, Ronghui Tan a, Yan Song b a b

School of Resource and Environment Science, Wuhan University, 129 Luoyu Road, Wuhan 430079, Hubei Province, PR China Department of City and Regional Planning, University of North Carolina at Chapel Hill, United States

a r t i c l e

i n f o

Article history: Received 12 January 2014 Received in revised form 16 June 2014 Accepted 23 June 2014 Available online 17 July 2014 Keywords: Urban growth Ecological security Remote sensing Landscape ecological planning Wuhan

a b s t r a c t Monitoring of urban growth and the characterization of its patterns in the Wuhan area of central China from 1988 to 2013 was performed using an integrated approach of remote sensing (RS) and geographic information system (GIS) techniques and statistical methods. We also undertook a qualitative analysis of the impact of urban growth on landscape ecological security. The results showed that the Wuhan area was in a rapid process of urbanization between 2000 and 2011, with an average urban growth rate of 10.666 km2/a and 2.969 km2/a in the surrounding region and city proper, respectively. An aggregated pattern was the primary growth type in the whole study area, while linear and leapfrog patterns mainly occurred in the surrounding region. Rapid urban growth has aggravated the landscape fragmentation and has led to considerable declines in ecosystem services. A Pearson correlation analysis was used to qualitatively explore the relationships between the urban growth patterns and the factors associated with ecological security. It was found that the ecosystem fragmentation and ecosystem services were correlated with the urban growth rate, the three types of urban growth, and the compactness of the urban form. Of the three growth types, the leapfrog growth pattern destroys the integrity of water bodies, thereby preventing the connection of lakes to the main surface water network, and thus results in increase fragmentation and reduction of ecosystem services. Although the land-use policies implemented in Wuhan during the study period have promoted the development of the local economy, they have failed to protect the ecosystem. Urban growth speed should be effectively controlled because natural resource protection is as important as, and even more important, to some extent, than encouraging extensive economic development. This research has highlighted the importance of the joint application of urban growth quantification and the monitoring of the changes in the factors associated with ecological security in landscape planning. Ó 2014 Elsevier Ltd. All rights reserved.

Introduction As the world population has continuously grown and become concentrated in town and city dwellings, urban areas have witnessed an enormous increase in the past 30 years, and this is particularly apparent in developing countries such as China and India (Seto, Fragkias, Güneralp, & Reilly, 2011). Urbanization is a spatial and social process that is related to the transformation of rural areas into urban lands, the movement of people from rural to urban areas as well as the changes in their life styles. Urban growth is a spatial process which refers to the increased area of towns and cities as the population is concentrated in these areas (Bhatta, ⇑ Corresponding author. Tel.: +86 02768778650; fax: +86 02768778893. E-mail addresses: [email protected] (K. Zhou), [email protected] (Y. Liu), [email protected] (R. Tan), [email protected] (Y. Song). http://dx.doi.org/10.1016/j.cities.2014.06.010 0264-2751/Ó 2014 Elsevier Ltd. All rights reserved.

Saraswati, & Bandyopadhyay, 2010). Although urbanization can promote socioeconomic development and improve the quality of life, the irreversible transformation from semi-natural and natural ecosystems into impervious surfaces has resulted in considerable environmental and ecological problems worldwide (Bhatta et al., 2010; Habibi & Asadi, 2011; Su, Jiang, Zhang, & Zhang, 2011). Therefore, understanding the effects of urban growth on the ecosystem and quantifying the relationships between urban dynamics and landscape ecological security is crucial for effective urban planning and environmental protection policy making, in order to support sustainable development. A considerable number of academic studies from all around the world have focused on the driving forces of urban expansion (Li, Zhou, & Ouyang, 2013; Lu, Wu, Shen, & Wang, 2013; Thapa & Murayama, 2010; Wu & Zhang, 2012), and there has been increasing interest in characterizing and quantifying the temporal

142

K. Zhou et al. / Cities 41 (2014) 141–153

dynamics of the spatial patterns of urban growth (Aguilera, Valenzuela, & Botequilha Leitão, 2011; Bhatta et al., 2010; Herold, Scepan, & Clarke, 2002; Ji, Ma, Twibell, & Underhill, 2006; Seto & Fragkias, 2005). However, to date, less attention has been paid to the impact of urban growth on the ecological security of a region. The environmental and ecological problems of the landscape changes resulting from urbanization are significant (Carlson & Traci Arthur, 2000; Wu, Ye, Qi, & Zhang, 2012). It is, however, especially difficult to characterize the relationships between urban growth patterns, changes in ecosystem values, and the consequent loss of ecological capacity in a human-dominated ecosystem (Wu et al., 2012). Rapid urban growth, as the most obvious land transformation, has resulted in significant changes in the structure and functioning of ecosystems (Liu, Li, & Zhang, 2012; Yang, Li, Wang, & Hu, 2011). The study of Su, Xiao, Jiang, and Zhang (2012) reported that landscape fragmentation, configuration, and diversity could significantly impair the provision of ecosystem services. To date, however, to the best of our knowledge, there has not been a study of the influence of different urban growth patterns on the changes in ecosystem structure and ecosystem services. Moreover, as the spatial characteristic of the urban form, and the result of urban growth, urban compactness has been the subject of many discussions (Capello & Camagni, 2000; Liu, Song, & Arp, 2012; Schneider & Woodcock, 2008). Nevertheless, few studies have examined the relationship between urban compactness and landscape structure and ecosystem services. Therefore, there is a need for long-term monitoring of the urbanization process, to evaluate the ecological consequences of urban growth at a landscape scale. For this study, we selected the Wuhan area to examine the different types of urban growth patterns and to monitor the long-term ecosystem changes. The goals of this study were four-fold. Firstly, we attempted to quantify the urban growth in Wuhan and to characterize the patterns of development. Secondly, we aimed to determine the changes in ecosystem structure and ecosystem services, which are fundamental indicators of landscape ecological security, between 1988 and 2013. Thirdly, we attempted to discover the correlations between urban growth and the factors associated with landscape ecological security. Finally, we related the results to changes in land-use policy. Method Study area and data The city of Wuhan, which is the capital of Hubei Province as well as the largest city in central China (Fig. 1), lies in the middle reaches of the Yangtze River. It covers a total area of 8549 km2, of which 39.27% consists of plains and 18.17% is hilly and mountainous regions. The Wuhan area has witnessed rapid urbanization and has experienced significant economic growth. In 2012, it had an urban population of 5.55 million, which was over 67.54% of the total population. Meanwhile, the gross domestic product (GDP) increased almost seven-fold, from 120.7 billion yuan in 2000 to 800.4 billion yuan in 2012 (Wuhan Municipal Bureau of Statistics, 2012). Currently, the municipal territory consists of 13 administrative units, in which seven districts constitute Wuhan city proper, and the other six districts are in the surrounding region. In this study, a multi-spectral Landsat TM/ETM + imagery time series dataset covering six years was downloaded from the US Geological Survey (USGS) and was employed to produce land-use/ cover maps of the Wuhan area. In addition to this, two land-use data maps of 1996 and 2006 were obtained from the Wuhan Land Resources and Planning Bureau and were used as ancillary data for the accuracy assessment of the imagery classification.

Mapping urban dynamics from remote sensing imagery A post-classification comparison method was used to quantify the urban growth extent of 1988, 1995, 2000, 2005, 2011, and 2013 (for a similar case, see Wu & Zhang, 2012). Prior to classification, atmospheric correction was performed. The RS images and the GIS data were then geometrically rectified to a common UTM coordinate system. The 2013 images were first corrected based on 90 ground control points (GCPs) that were evenly distributed in the study area. An image-to-image registration based on the 2013 images was then undertaken to ensure that all the images had the same projection. The root-mean-square error (RMSE) was limited to within 0.5 pixels (15 m). The supervised maximum likelihood classifier (Canty, 2006) in ENVI and a visual interpretation method were employed to classify the images. A classification scheme of five land-use/cover types was used to implement the classification, namely built-up land, forest, cropland, water, and bare land (Were, Dick, & Singh, 2013). The overall accuracies and the Kappa coefficient were greater than 80% and 0.7, respectively, which indicates that the classifications accurately represent the real landscape (Janssen & Van der Wel, 1994; Landis & Koch, 1977). All the aforementioned processing steps were performed in the ENVI 4.8 software environment. Quantifying and characterizing urban growth Urban growth rate and intensity The urban growth rate (UGR) and intensity (UGI) indexes can be used to represent changes in the quantity of an urban area per unit time, and they are key indicators for evaluating the extent and rate of change of urban expansion (Ma & Xu, 2010; Xu & Min, 2013):

ULAi;tþn  ULAi;t 1   100% n ULAi;t ULAi;tþn  ULAi;t UGIi ¼  100% n  TLAi;t

UGRi ¼

ð1Þ ð2Þ

where UGRi is the urban growth rate; UGIi is the urban growth intensity for unit i; ULAi,t+n and ULAi,t are the percentage of urban land area in unit i at time t + n and time t, respectively; TLAi is the total land area of unit i; and n is the time interval between time t and time t + n. The UGR index calculates the growth of an urban area as a percentage of the total growth of the urban areas in the study period. The UGI index denotes the growth of an urban area as a percentage of the total area of the land units in the study period. Urban growth types Three types of urban growth patterns were identified in the Wuhan area: an aggregated pattern, a linear pattern, and a leapfrog pattern. The aggregated pattern corresponds to the clustering of patches to form patches of a larger size (Fig. 2). The linear pattern is related to the elongation process of patches increasing along with road networks. The leapfrog pattern involves an increase in the dispersed or isolated patches distributed in a region. To identify the three growth categories, the urban land-use maps were first converted to polygon maps in the ArcGIS software platform. An individual urban land-use map was then overlaid with its adjacent-year urban map to distinguish the newly developed urban patches. Finally, the distance from the new urban patches to the existing urban patches (d1) and the road networks (d2) of the previous year were determined by the Near tool in ArcGIS 9.3. If the d1 of a new urban patch was equal to 0, we considered it as an aggregated pattern; if the d2 of a new urban patch was within 1 km, it was labeled as a linear pattern; otherwise, the patch did not belong to either of the two aforementioned categories, and it was

K. Zhou et al. / Cities 41 (2014) 141–153

143

Fig. 1. Location of the Wuhan area, and the spatial division of the two main regions.

classified as a leapfrog pattern. We then calculated the quantity of the three urban growth types. Urban growth spatial statistics The compactness index is an effective indicator in urban morphology (Li & Yeh, 2004). The compactness of urban land is estimated according to the average comparison between the

perimeter of each developed cluster and that of a circle which has the same area. The index is defined as:

P pffiffiffiffiffiffiffiffiffiffi 2 Si =p=pi CI ¼ i n2

ð3Þ

where CI is the compactness index, Si and pi are the area and perimeter of the urban patch I, and n is the total number of urban patches.

144

K. Zhou et al. / Cities 41 (2014) 141–153

Fig. 2. The three urban growth patterns in the Wuhan area.

The factors associated with landscape ecological security Indicators of ecosystem structure Landscape structure has two basic components: composition and configuration (Botequilha Leitão & Ahern, 2002). Composition is a non-spatially explicit characteristic, which reflects the richness, evenness, or dominance of a component of the ecosystem. Configuration, in contrast, relates to the spatially explicit characteristics of the land-cover types in a given landscape. Landscape metrics are useful and highly applicable in quantifying landscape structure (Aguilera et al., 2011; Botequilha Leitão & Ahern, 2002; Su et al., 2011). Based on the core set of 10 metrics proposed by Botequilha Leitão and Ahern (2002), we selected five landscapelevel metrics to reflect the characteristics of the ecosystem structure: patch density (PD), mean shape index (SHAPE_MN), area weighted mean patch fractal dimension (AWMPFD), contagion (CONTAG), and Shannon’s diversity (SHDI). These metrics have a low redundancy and are capable of quantitatively describing landscape structure (Botequilha Leitão & Ahern, 2002; Su et al., 2011). Indicators of ecosystem function The ecosystem service value (ESV) quantifies the benefits that the human population gains, directly or indirectly, from ecosystem functions (Costanza et al., 1997). Costanza et al. (1997) identified 17 ecosystem services provided by 16 dominant global biomes at a global scale. Based on the hypothesis that the supply and demand curves of the ESV are nearly vertical, Costanza et al. then estimated the economic monetary value per unit area of each ecosystem

service for each ecosystem type, by the use of the restoration cost (a market-based monetary valuation technique). Finally, they used the land use and land cover (LULC) as a proxy to estimate the total global extent of the ecosystems themselves, and thus to evaluate the total global economic value of the ecosystem services. However, in most cases, the proxy biomes were not perfectly matched with the LULC types. Xie, Lu, Leng, Zheng, and Li (2003) developed an enhanced method for China’s ecosystem services valuation by surveying 200 Chinese ecologists, based on the work of Costanza et al. (1997) (see Table 1). Compared with the method used by Costanza et al., the method used by Xie et al. is considered to be more practicable, and has thus been widely applied in the valuation of China’s ecosystem services (Li, Wang, Hu, & Wei, 2010; Liu, Li, et al., 2012; Wu et al., 2012). We therefore assigned ESVs to the different land-use types, based on the suggestion of Xie et al. (2003):

ESV ¼

m X n X Ai  VC j i¼1 j¼1

where Aj is the area (ha) of land cover I, and VCj is the value coefficient of the ecosystem function for type j (RMB yuan/ha) combined with land-cover type i. Statistical analysis The correlation relationships between urban growth and landscape ecological security were examined using the Pearson correlation method.

145

K. Zhou et al. / Cities 41 (2014) 141–153 Table 1 Ecosystem service value index for each land-use/land-cover type (unit: RMB yuan/ha).

Gas regulation Climate regulation Water reservation Soil formation and protection Waste disposal Biodiversity conservation Food production Raw materials Entertainment and culture Sum

Forest

Farmland

Water

Bare land

Built-up area

3097.0 2389.1 2831.5 3450.9 1159.2 2884.6 88.5 2300.6 1132.6 19334

442.4 787.5 530.9 1291.9 1451.2 628.2 884.9 88.5 8.8 6114.3

0.0 407.0 18033.2 8.8 16086.6 2203.3 88.5 8.8 3840.2 40676.4

0.0 0.0 26.5 17.7 8.8 300.8 8.8 0.0 8.8 371.4

0 0 0 0 0 0 0 0 0 0

Results Spatiotemporal changes in urban growth During the 25 years between 1988 and 2013, the Wuhan area experienced a rapid exponential urban expansion (Figs. 3 and 4). In 1988, there was a total of 206.58 km2 of urban land, which represented only 2.4% of the study area, whereas by 2013, the urbanized area had progressively expanded to 1814.85 km2, which accounted for 21.14% of the study area (Fig. 3a). The annual growth rate increased from 10.6% between 1988 and 1995 to 12.9% between 2005 and 2011, and then dropped to 10.53% between 2011 and 2013 (Fig. 3b). Table 2 shows the general urban growth trends of the 13 sub-cities in the Wuhan area, which indicates the obvious spatial heterogeneity of the urban growth rate and intensity across the city proper and the surrounding region between 1988 and 2013. We calculated the proportion of each growth type in the newly developed urban areas for all five periods (Fig. 5). While the aggregated pattern was detected as the dominant growth pattern, both in the city proper and the surrounding region, the linear pattern and the leapfrog pattern were found to occur in the surrounding districts, especially between 1988 and 2000. After the central government opened up Wuhan as one of the five open cities along the Yangtze River in May 1992, the administrative division of Wuhan was readjusted, and four adjacent counties were converted into districts, facilitating the shaping and urban development of the Wuhan Metropolitan Area (WMA). As a result, the large scale of the newly built infrastructure, industrial zones, and real estate have encroached upon water surfaces and farmland in the suburbs, and resulted in massive leapfrog-pattern urban growth in the

surrounding region in the 1990s. In addition, with the increase in the population, per capita residential areas for both urban and rural residents grew considerably. To accommodate the increasing population, additional settlements in the urban fringe were rapidly developed. As a result, urban areas expanded into rural areas in the surrounding region. Table 3 exhibits the results from the spatial compactness analysis for the different sub-cities. The most concentrated development pattern was observed in Jianghan. Moreover, ongoing increasing trends in the compactness values were identified in Wuchang, Jianghan, and Qingshan from 1988 to 2013, implying that an intensification of the urban land use occurred in these three districts in the city proper. Conversely, the value continually fell in all the sub-cities of the surrounding region over the same period. It was notable that areas around the city center were developed much earlier, and with little land available for sprawl, this compelled the intensive use of urban land under the context of persistent economic growth. For those sub-cities that are far from the city center, there was much more non-construction land available for development, which could accommodate the establishment of industrial parks and college towns. Therefore, the extensive development patterns associated with urban sprawl were inevitable, and have thus led to a reduced degree of compactness. Landscape pattern changes Ecosystem structure can be evaluated by the use of landscape metrics. As shown in Fig. 6, over the study period, the city proper generally displayed more aggregation and less dispersion (CONTAG increased and PD decreased), as a consequence of the new urban patches filling the holes or being aggregated along with the

Fig. 3. Urban area (a) and growth rate (b) in the different periods from 1988 to 2013.

146

K. Zhou et al. / Cities 41 (2014) 141–153

Fig. 4. Urban land expansion and other land-use changes from 1988 to 2013 in the Wuhan area.

existing built-up area. In contrast, the surrounding region exhibited more fragmentation and heterogeneity, evidenced by the decrease in CONTAG and the increase in PD. However, the landscape became less (more) diverse in the city proper (surrounding region), which can be seen from the decline (increase) in SHDI. With increasingly unprotected natural land-use types being transformed to built-up land, the patch shape became more regular and simple in the city proper. Although not obvious, this change can be seen by the slight decrease in SHAPE_AM and AWMPFD. For most districts in the surrounding region, SHAPE_AM initially decreased and later increased after 2005. The reason for this could be that as the local government began to build the Wuhan ‘‘1 + 8’’ City Circle in 2007, the rural-urban fringe became the connection that links Wuhan with the other eight small cities, and it also became a hot spot for township enterprise development. In addition, inter-city highways and railways were also constructed. Therefore, the patch shape of the surrounding region became more irregular and convoluted after 2005. Ecosystem service value changes The results from the ESV calculation are displayed in Fig. 7. In total, the city proper region was characterized by much lower ESV values than the surrounding region from 1988 to 2013, except for Hongshan. It should be noted that the largest downtown lake, East Lake (with an area of 33 km2, which is six times the size of West Lake in Hangzhou), and the East Lake Scenic Spot are located in Hongshan. To a large extent, this has helped Hongshan maintain relatively high ESV values. The ecosystem service function of the city proper region has confronted severe deterioration, in that the average ESV values of the city proper region continuously declined from 2.43  108 RMB to 1.22  108 RMB, which was almost a 50% decrease over the last 25 years. In contrast, the ESV

of the surrounding region decreased sharply between 1988 and 1995, followed by a decade of increasing ESV between 1995 and 2005 (which may have been due to the preservation of green land and the reforestation policy in the late 1990s), and then decreased again after 2005. Relationships between urban growth and landscape ecological security As shown in Table 4, the UGR, AG, LG, and LPFG demonstrated positive correlations with the variations in PD, SHAPE_MN, AWMPFD, and SHDI, but showed negative correlations with the changes in CONTAG and ESV, indicating that whatever the type of urban growth, they could be contributing factors to the fragmented ecosystem structure and reduced ecosystem function. Moreover, for the three growth patterns, the leapfrog growth pattern exhibited the greatest positive correlations with the changes in PD and SHDI, and negative correlations with the changes in CONTAG. This is also the case when it comes to the relationship between the growth patterns and the changes in ESV, which signifies that a scattered growth pattern has a more obvious undermining influence on ecosystem service exchanges between different landscapes than the other two growth patterns (see Table 4). PD, SHAPE_MN, and AWMPFD showed significant negative correlations, while CONTAG showed a positive correlation with the compactness index. These results demonstrate that a more compact urban form can be of benefit to landscape integrity, edge simplicity, and ecosystem recoverability. Discussion Landscape security change in response to urban growth As shown in Section 3.4, all the types of growth pattern are related to the changes in ecosystem structure and services. Of

147

K. Zhou et al. / Cities 41 (2014) 141–153 Table 2 The urban growth rate and intensity of the 13 sub-cities in Wuhan from 1988 to 2013. Region

Sub-cities Urban growth rate (km2/a)

Urban growth intensity (%)

1988–1995

1995–2000

2000–2005

2005–2011

2011–2013

1988– 1995

1995– 2000

2000– 2005

2005– 2011

2011– 2013

City proper

Wuchang Hanyang Jianghan Jiangan Qiaokou Hongshan Qingshan

1.1853(4.31%) 1.3966(7.48%) 1.0633(7.50%) 1.3079(5.25%) 0.6021(3.16%) 5.0483(20.37%) 0.8543(3.16%)

0.8822(2.47%) 2.0299(7.14%) 0.7561(3.50%) 1.7899(5.26%) 0.6772(2.91%) 9.7032(16.14%) 0.9859(2.98%)

0.4780(1.19%) 2.5780(6.68%) 0.3584(1.41%) 2.2845(5.31%) 0.8427(3.16%) 11.7688(10.83%) 0.1487(0.39%)

0.3512(0.82%) 4.6055(8.95%) 0.0501(0.18%) 1.5607(2.87%) 1.0915(3.54%) 15.3276(9.15%) 0.1213(0.31%)

0.1859(0.42%) 3.1241(3.95%) 0.0399(0.15%) 0.8184(1.28%) 0.3931(1.05%) 17.2908(6.66%) 0.1247(0.32%)

1.9318 1.2502 3.7521 1.6264 1.5010 0.8579 1.8485

1.4377 1.8171 2.6680 2.2258 1.6882 1.6490 2.1331

0.7789 2.3077 1.2646 2.8408 2.1006 2.0000 0.3218

0.5723 4.1226 0.1769 1.9408 2.7209 2.6048 0.2625

0.3030 2.7966 0.1408 1.0177 0.9798 2.9384 0.2698

Surrounding districts

Dongxihu Hannan Caidian Jiangxia Huangpi Xinzhou

1.8782(36.10%) 0.3687(25.99%) 1.9196(26.82%) 2.1980(10.57%) 1.9502(24.66%) 2.1558(26.98%)

3.5323(19.25%) 1.4347(35.87%) 4.9003(23.80%) 2.9365(8.12%) 4.0500(18.78%) 2.7144(11.76%)

5.3881(14.96%) 0.4907(4.39%) 9.0946(20.17%) 13.4129(26.37%) 6.1290(14.66%) 7.1886(19.61%)

9.8726(15.68%) 4.0814(29.95%) 16.0183(17.69%) 19.1168(16.21%) 20.9986(28.98%) 16.1988(22.31%)

16.4210(13.44%) 7.9333(20.81%) 22.9924(12.32%) 28.0583(12.06%) 32.5244(16.39%) 27.9118(16.44%)

0.3786 0.1283 0.1754 0.1088 0.0863 0.1470

0.7121 0.4992 0.4479 0.1453 0.1792 0.1851

1.0862 0.1707 0.8312 0.6637 0.2712 0.4902

1.9903 1.4202 1.4640 0.9459 0.9291 1.1046

3.3104 2.7605 2.1015 1.3883 1.4391 1.9033

the three growth patterns, scattered urban growth has the closest relationship with a decrease in ecosystem security. This was largely demonstrated by: (1) The leapfrog pattern increases landscape fragmentation and therefore increases the number of habitat patches and decreases the patch size. In summary, fragmented urban areas, such as densely populated urban areas, roads, and traffic infrastructure, can exert strong, hard, and abrupt physical and biological edge effects (Di Giulio, Holderegger, & Tobias, 2009). These effects are most obvious in biodiversity. (2) All kinds of urban expansion lead to considerable declines in ESVs, due to the encroachment of urban areas on natural or semi-natural resources, although the trade-offs associated with LULC transitions can offset some negative impacts, to some extent (Li et al., 2010; Liu, Li, et al., 2012; Wu et al., 2012). (3) Compared with the linear growth pattern and the aggregated pattern, the leapfrog growth pattern destroys the integrity of water bodies, and can even prevent the connection of lakes to the main surface water network. Du, Ottens, and Sliuzas (2010) reported that in the Wuhan area, nearly 60% of the shallow water bodies have been converted to urban and rural artificial land use, and such conversions are often spatially scattered. In the meantime, the leapfrogging has led to agricultural landscape fragmentation, so as to cause a reduction in the patch size for agricultural landscapes and increasing isolation of patches through the destruction of connecting corridors (Botequilha Leitão & Ahern, 2002; Su et al., 2011). Hence, dispersed growth patterns are considered unsuitable for urban development. Furthermore, it should also be noted that, as shown in Table 4, a more compact spatial urban morphology is always followed by a less-fragmented landscape pattern. Although a more compact development pattern means higher construction costs, it can generate positive effects such as control of arable land loss, improved land-use efficiency, and a reduction in pollution of the local environment. As previous studies have indicated, compact cities need less materials and energy for infrastructure construction per capita, and more multi-family houses share foundations and resources, which is helpful for the exploitation of economies of scale for public services and resources, as well as accessibility to social services. Compact cities therefore play a great role in promoting urban eco-efficiency and resource efficiency (Capello & Camagni, 2000; Liu, Song, et al., 2012). Compact urban growth can therefore be expected to considerably contribute to urban efficiency. However, simultaneously, compact development can reduce landscape diversity and recoverability, as seen from the negative correlation between SHDI and Cpact. This should be taken into consideration in future urban planning. In addition, the rapid urban growth rate in the Wuhan area is certainly responsible for the decline in the ecosystem service values, which can be seen in the negative correlation between UGR and ESV. In this context,

balancing the conflicts between land-use competition, urban development, and ecological conservation is one of the most important tasks for urban planners. Although no specific urban form can be characterized as appropriate for all environmental and ecological contexts (Alberti, 2008), we found that a rapid and scattered urban development pattern has a closer relationship with ecosystem deterioration than a slow and aggregated growth pattern. Thus, we argue that a more compact and intensive urban form can be beneficial to environment/ecosystem protection. Impacts of land-use policy on urban growth China’s urban growth is generally promoted by migration, rapid economic growth, and land-use policy. The former two factors are always facilitated or inhibited by the implementation and enforcement of land-use regulations. Prior to 1978, the land market in China was static and land-use transition was strictly controlled. After the initialization of the ‘‘reform and opening policy’’ of Deng Xiaoping in 1978, market-oriented economic liberalization reforms brought about rapid socioeconomic changes in the coastal regions of China, but inland cities such as Wuhan have lagged far behind the coastal areas, due to their disadvantageous locations and the less preferential political and economic strategy in the 1980s and 1990s (Qiao et al., 1999). Consequently, the urban growth rate in the Wuhan area was relatively low during this period. In 1994, facing severe population growth and the problem of food security, China launched the basic farmland protection policy to control the overheated urban development (Li & Yeh, 2004). This regulation calls for local governments to protect high-quality farmland by designating a basic farmland protection zone in every village and town. In this context, the urban growth rate of Wuhan between 1995 and 2000 was certainly inhibited and was the lowest during the whole study period. However, since the ‘‘go-west’’ and ‘‘promoting the central region’’ strategies were launched in 2000 and 2009, respectively, Wuhan has shown rapid economic growth, due to the support of government and the considerable investment from all over the world. During 2000–2011, there was a great deal of land-use change in Wuhan. However, the massive conversion from agricultural land and water bodies to construction land, which is associated with the rapid urban sprawl, has resulted in deterioration of the ecosystem. In particular, 16.73 km2 of wetlands and lakes were converted into built-up area (Huang, Liu, Liu, & Jiang, 2012). At the same time, although the Chinese central government released a new ‘‘increasing vs. decreasing balance’’ land-use policy in 2005 (aimed at reconciling the demands of urban development and farmland protection by achieving equilibrium in the supply of land in China by balancing increases in urban construction land with decreases in rural

148

K. Zhou et al. / Cities 41 (2014) 141–153

Fig. 5. Structure of the urban growth patterns by sector in Wuhan from 1988 to 2013.

construction land), this top-down policy has failed to curb urban growth, to a certain extent, because it threatens rural lifestyle and culture, as well as farmers’ rights and interests (Long, Li, Liu, Woods, & Zou, 2012). With respect to the regulations or laws designated by local government in the Wuhan area, unlike the top-down approaches set by the central government, which are always aimed at protecting farmland, most of the regulations or laws are intended to stimulate

economic growth through the establishment of high-tech industrial or economic zones. Three large high-tech development zones have been developed in the Wuhan area in the last three decades (Fig. 1). These development zones have in turn encouraged the growth of town and village enterprises (TVEs). Large high-tech development zones have therefore accelerated the aggregated urban growth pattern, while TVEs have increased the scattered urban growth. In addition, with the development of infrastructure

149

K. Zhou et al. / Cities 41 (2014) 141–153 Table 3 The compactness of the urban form in the 13 sub-cities of Wuhan from 1988 to 2013. Sub-cities

1988

1995

2000

2005

2011

2013

Wuchang Hanyang Jianghan Jiangan Qiaokou Hongshan Qingshan Dongxihu Hannan Caidian Jiangxia Huangpi Xinzhou

0.001984 0.001345 0.005033 0.002162 0.004381 0.000311 0.004553 0.000956 0.002612 0.000716 0.000097 0.000480 0.000495

0.004576 0.000789 0.005690 0.001299 0.002926 0.000184 0.008341 0.000238 0.000758 0.000148 0.000065 0.000128 0.000182

0.004780 0.000424 0.034924 0.000891 0.003222 0.000094 0.008886 0.000145 0.000291 0.000084 0.000051 0.000069 0.000115

0.005804 0.000323 0.058478 0.000818 0.004050 0.000078 0.010696 0.000090 0.000212 0.000052 0.000031 0.000039 0.000059

0.005662 0.000650 0.109898 0.001220 0.012918 0.000086 0.015205 0.000065 0.000121 0.000031 0.000028 0.000021 0.000024

0.005857 0.000808 0.146171 0.001165 0.023288 0.000089 0.016416 0.000056 0.000120 0.000029 0.000023 0.000017 0.000019

construction and the flourishing real estate market, the Wuhan local government has had to pay its own debts through selling the land that was ‘‘grabbed’’ from farmers in the past. The local government retains all the profits from such land transactions, and it relies heavily on land transactions to finance ongoing investment in both public service provision and infrastructure (Ding, 2007). Promoting economic growth has been one of the highest priorities of the Chinese government at all levels, and the aforementioned land-use policies have had relatively little effect on the rate of urban spatial expansion (Lichtenberg & Ding, 2009). Urban sprawl has therefore been inevitable. Our study showed that both landscape structure and ESV in the Wuhan area have deteriorated in the last three decades. However, there have been a few policies designated by both the central government and the Wuhan local government that are environment and ecosystem protection oriented. This paper contributes to a literature stream that should be central in this effort.

1998). Moreover, there has been relatively little comparative research between the cities of developed countries and those of less-developed countries (Robinson, 2002, 2011). The reason for this may be that some wealthier cities have been counterpoised with poorer cities too long in urban theorizing, and some scholars think that different urban theories and methodology should be used for these two different kinds of city, because the urban history, urban form, and structure of less-developed cities, or the so-called ‘‘third-world’’ cities, are clearly different from those of developed countries. However, as Dick and Rimmer (1998) and Robinson (2011) showed, since economic and social activities in different cities are now linked together through spatially extensive flows of various kinds, and intense networks of communication, this discrepancy is currently much less than before. Therefore, comparative studies of cities across the globe should be encouraged, and we hope that our research has contributed a little to fill the void.

Comparison with other cities

Methodology discussion

According to Fig. 5, the aggregated pattern accounted for more than 50% of the urban expansion, which indicates that the majority of the urban growth in the Wuhan area was attached to established urban areas. These results are in agreement with previous statements that urban expansion in less-developed countries is primarily an aggregated pattern that attaches to existing urban land, while US and other Western cities are characterized by a dispersed form of expansion and significantly lower population density (Schneider & Woodcock, 2008; Shi, Sun, Zhu, Li, & Mei, 2012). Wuhan has witnessed an increasing fragmentation and patch shape complexity over the last decade, which can also be found in other large and medium-sized second-tier cities in China, such as Guangzhou (Sun, Wu, Lv, Yao, & Wei, 2013), Changsha (Xu & Min, 2013), Nanjing (Xu et al., 2007), Lianyungang (Shi et al., 2012), and Hangzhou (Wu & Zhang, 2012). Unlike Guangzhou and Nanjing, where dispersed urban sprawl occurred in the earlier stage, but with adequate urban planning and regulation, patch infilling and edge-expansion became the dominant growth pattern, and some sub-districts in Wuhan and a great number of small cities in other provinces of China are facing severe, unlimited, and disordered growth patterns. There has been much research into urban growth all over the world, especially in China. The Chinese studies have concentrated on the cities in the Yangtze River Delta and Pearl River Delta, such as Guangzhou, Shenzhen, Nanjing and Beijing. However, Wuhan, the capital of Hubei Province and the largest city of central China, has not, to date, been systematically studied in the existing studies of urban growth patterns. It is also true that many other cities around the world, especially in less-developed countries, have been ignored in the urban studies literature (Dick & Rimmer,

In this study, we used UGR, UGI, urban growth type, and compactness index to quantify the characteristics of urban growth in Wuhan. UGR and UGI provide information regarding the quantitative changes of urban land while urban growth type and compactness index offer insights on the spatial development of urban land. Different urban growth types can reflect the diffusion and coalescent process of urban growth (Shi et al., 2012). Compactness index is an important indicator of urban morphology, which has close relationship with eco-efficiency and resource efficiency (Liu, Song, et al., 2012). Urban growth, which is frequently defined as multi-dimensional, is a complicated process. Thus, it is important that the indicators used to characterize urban growth should not only be capable of capturing the nature of urban growth, but could also be used universally. Moreover, these indicators should be easily interpreted because the end users are city administrators and planners, who generally are not scientists. This study confirmed that the indicators selected are simple and reliable in relation to monitoring and characterizing urban growth. Structure, function, and change are the fundamental characteristics of landscape ecology. We used landscape metrics to quantitatively describe the landscape structure and its changes in Wuhan because these have the ability to provide information on the content of landscape mosaic as well as the shape of landscape component elements. Moreover, these can measure the arrangement of landscape elements in both time and space (Botequilha Leitão & Ahern, 2002). Given that landscape metrics can reflect changes in landscape composition, these are useful and essential tools for monitoring changes in the landscape structure and could considerably assist land planners and managers in ecosystem protection. Therefore, metric analysis offers a useful framework for the

150

K. Zhou et al. / Cities 41 (2014) 141–153

Fig. 6. Temporal spatial metric changes with respect to ecosystem structure from 1988 to 2013. (Note: the AWMPFD and SHDI metrics refer to the second y-axis.).

indirect determination of the ecological impacts of urban growth in ecologically vulnerable areas. However, landscape patterns are scale dependent, both in the spatial scale and temporal scale (Šímová & Gdulová, 2012). Many landscape metrics are also sensitive to grain size (pixel size) and extent of the study area. Therefore, in future research, landscape metrics should be computed at multiple scales to achieve adequate quantification of the landscape structure.

We also used the revised ESV coefficients for China, as presented by Xie et al. (2003), which originally came from Costanza et al. (1997). However, this evaluation method may not be sufficient, to some extent, due to deviations and uncertain factors such as temporal and spatial scale effects, landscape heterogeneity, and the problem of double counting. It does, however, represent the most comprehensive set of first-approximations available for quantifying the changes in the value of services provided by a wide

151

K. Zhou et al. / Cities 41 (2014) 141–153

Fig. 7. Changes in the ESV values in the 13 regions of the Wuhan area from 1988 to 2013.

Table 4 Pearson correlation between urban growth changes and landscape ecological security changes.

DPD

DSHAPE_MN a

UGR Pearson correlation sig. (2-tailed)

0.643 (0.000) 0.051 (0.685) 0.357a (0.004) 0.350a (0.004) 0.455a (0.000) 0.448a (0.000)

UGI Pearson correlation sig. (2-tailed) AG Pearson correlation sig. (2-tailed) LG Pearson correlation sig. (2-tailed) LPFG Pearson correlation sig. (2-tailed)

DCpact Pearson correlation sig. (2-tailed)

a

0.607 (0.000) 0.194 (0.121) 0.262b (0.035) 0.226 (0.070) 0.302b (0.014) 0.296a (0.008)

DAWMPFD a

0.514 (0.000) 0.204 (0.103) 0.265b (0.033) 0.230 (0.065) 0.258b (0.038) 0.554a (0.000)

DCONTAG a

0.489 (0.000) 0.326a (0.008) 0.467a (0.000) 0.479a (0.000) 0.528a (0.000) 0.606a (0.000)

DSHDI a

0.425 (0.000) 0.425a (0.000) 0.476a (0.000) 0.492a (0.000) 0.521a (0.000) 0.632a (0.000)

DESV 0.491a (0.000) 0.112 (0.376) 0.325a (0.008) 0.306b (0.013) 0.393a (0.001) 0.257b (0.023)

Note: D means the variation of the indicators in the five times span, i.e. the index of one later year minus that of the former monitoring year. Abbreviations: Patch density (PD), mean shape index (SHAPE_MN), area weighted mean patch fractal dimension (AWMPFD), contagion (CONTAG), Shannon’s diversity (SHDI), ecosystem service value (ESV), urban growth rate (UGR), urban growth intensity (UGI), aggregated pattern (AG), linear pattern (LG), leapfrog pattern (LPFG), compactness (Cpact). a Correlation is significant at the 0.01 level (2-tailed). b Correlation is significant at the 0.05 level (2-tailed).

Table 5 Percentage change in estimated total ecosystem service value and coefficient sensitivity (CS) resulting from adjustment of the ecosystem valuation coefficients (VC) of wetland and grassland. Change of valuation coefficient

50%

DTotal ESV (%)

CS

DTotal ESV (%)

CS

Wetland VC Grassland VC

2.300 0.059

0.046 0.001

0.767 0.020

0.015 0.000

Note: CS ¼

50%

jðESV j ESV i Þ=ESV i j jðVC j VC i Þ=VC I j .

array of ecosystems (Kreuter, Harris, Matlock, & Lacey, 2001). Although the accuracy of the ecosystem valuation coefficients may have negative influences on the magnitude of the ecosystem values at specific points in time, the estimates of the temporal change of ESVs have been confirmed to be reliable in a time series analysis (Li et al., 2010; Liu, Li, et al., 2012; Su et al., 2011). Thus, accurately calculating coefficients is less critical for time series analyses than for cross-sectional analyses. In addition, we equated wetland with water, and grassland with farmland in the ESV calculation. To verify this processing method, we employed a sensitivity analysis through adjusting their coefficients by ±50%, and we used the wetland and grassland data derived from the 2006 land-use

database for the test (for similar issues, see Liu, Li, et al., 2012). The percentage change in estimated total ESV and its coefficient sensitivity (CS) are shown in Table 5. These results indicate that the estimation in the Wuhan area was relatively reliable, despite uncertainties with the valuation coefficients of wetland and grassland. Conclusion This paper has characterized the spatiotemporal urban growth patterns and has highlighted the relationships between urban dynamics and ecological security in the Wuhan area, which was in a process of rapid urbanization between 1988 and 2013. Quantitative characterization of the urban expansion has provided strong evidence for the spatial heterogeneity of the urban growth within the Wuhan area. The urban growth rate in the surrounding region was higher than in the city proper over the study period, and accelerated after 2005. From the late 1980s to 2013, the Wuhan urban expansion patterns underwent two different trends within the 13 sub-cities, with an aggregated pattern being dominant in the city proper region over the study period, and linear and leapfrog patterns taking up a considerable proportion of the urban growth in the surrounding region. All the types of urban

152

K. Zhou et al. / Cities 41 (2014) 141–153

growth pattern led to a considerable threat to ecological security in the Wuhan area, which was demonstrated by the increased landscape fragmentation and reduced ecosystem service values (ESVs). As shown by detailed analysis, the ecosystem fragmentation and ecosystem services were correlated with the urban growth rate, the three types of urban growth pattern, and the compactness of the urban form. Of the three growth types, the leapfrog growth pattern has much to do with the changes in fragmentation and ecosystem services. The fast urban growth of the Wuhan area is a function of demographic, econometric, and biophysical forces, changes to which can alter both the rate and spatial structure of urban expansion, and thus change the subsequent trends of ecosystem deterioration around the Wuhan area. As shown in this study, scattered growth in the surrounding region of the Wuhan area accounted for approximately 15% of the total growth. Although the area of this extensive growth pattern has declined in several sub-districts, it should not be underestimated in future urban planning. Institutionally, the Chinese government plays the key role in the allocation of land resources, and therefore dominates the decision making for urban planning, public investment, and population migration. Thus, compared with other countries, the Chinese government policies are of crucial importance in balancing economic development and ecosystem protection. We therefore argue that a joint application of quantitative characterization of the urban growth patterns and long-term monitoring of ecological security should be seriously considered by the government. Future research should also focus on the establishment of a strategy to curb urban sprawl, optimize the urban growth patterns, and monitor the long-term changes in the ESVs of each land-use type. Acknowledgements This research was funded by the National Key Technology R&D Program of China (Grant No. 2012BAH28B02). We sincerely thank Editor-in-Chief Ali Modarres, the three anonymous reviewers, and a language editor for their valuable comments, suggestions, and editing, which have greatly helped to improve our manuscript. References Aguilera, F., Valenzuela, L. M., & Botequilha Leitão, A. (2011). Landscape metrics in the analysis of urban land use patterns: A case study in a Spanish metropolitan area. Landscape and Urban Planning, 99, 226–238. Alberti, M. (2008). Advances in urban ecology: integrating humans and ecological processes in urban ecosystems. Springer-Verlag. Bhatta, B., Saraswati, S., & Bandyopadhyay, D. (2010). Quantifying the degree-offreedom, degree-of-sprawl, and degree-of-goodness of urban growth from remote sensing data. Applied Geography, 30, 96–111. Botequilha Leitão, A., & Ahern, J. (2002). Applying landscape ecological concepts and metrics in sustainable landscape planning. Landscape and Urban Planning, 59, 65–93. Canty, M. J. (2006). Image Analysis, Classification and Change Detection in Remote Sensing: With Algorithms for ENVI/IDL. 160 p. Taylor & Francis, CRC Press. Capello, R., & Camagni, R. (2000). Beyond optimal city size: an evaluation of alternative urban growth patterns. Urban Studies, 37, 1479–1496. Carlson, T. N., & Traci Arthur, S. (2000). The impact of land use—land cover changes due to urbanization on surface microclimate and hydrology: a satellite perspective. Global and Planetary Change, 25, 49–65. Costanza, R., d’Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neill, R. V., & Paruelo, J. (1997). The value of the world’s ecosystem services and natural capital. Nature, 387, 253–260. Di Giulio, M., Holderegger, R., & Tobias, S. (2009). Effects of habitat and landscape fragmentation on humans and biodiversity in densely populated landscapes. Journal of Environmental Management, 90, 2959–2968. Dick, H. W., & Rimmer, P. J. (1998). Beyond the third world city: the new urban geography of South-east Asia. Urban Studies, 35, 2303–2321. Ding, C. (2007). Policy and praxis of land acquisition in China. Land Use Policy, 24, 1–13. Du, N., Ottens, H., & Sliuzas, R. (2010). Spatial impact of urban expansion on surface water bodies—A case study of Wuhan, China. Landscape and Urban Planning, 94, 175–185.

Habibi, S., & Asadi, N. (2011). Causes, results and methods of controlling urban sprawl. Procedia Engineering, 21, 133–141. Herold, M., Scepan, J., & Clarke, K. C. (2002). The use of remote sensing and landscape metrics to describe structures and changes in urban land uses. Environment and Planning A, 34, 1443–1458. Huang, P. J., Liu, Y. F., Liu, C., & Jiang, Q. H. (2012). Study on evolution of urban lakes in Wuhan based on RS/GIS. Ecology and Environmental Sciences, 21, 1588–1593 (in Chinese). Janssen, L. L., & Van der Wel, F. J. (1994). Accuracy assessment of satellite derived land-cover data: a review. Photogrammetric Engineering and Remote Sensing, 60, 419–426. Ji, W., Ma, J., Twibell, R. W., & Underhill, K. (2006). Characterizing urban sprawl using multi-stage remote sensing images and landscape metrics. Computers, Environment and Urban Systems, 30, 861–879. Kreuter, U. P., Harris, H. G., Matlock, M. D., & Lacey, R. E. (2001). Change in ecosystem service values in the San Antonio area, Texas. Ecological Economics, 39, 333–346. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159–174. Li, J., Wang, W., Hu, G., & Wei, Z. (2010). Changes in ecosystem service values in Zoige Plateau, China. Agriculture, Ecosystems & Environment, 139, 766–770. Li, X., & Yeh, A. G. O. (2004). Analyzing spatial restructuring of land use patterns in a fast growing region using remote sensing and GIS. Landscape and Urban Planning, 69, 335–354. Li, X., Zhou, W., & Ouyang, Z. (2013). Forty years of urban expansion in Beijing: What is the relative importance of physical, socioeconomic, and neighborhood factors? Applied Geography, 38, 1–10. Lichtenberg, E., & Ding, C. (2009). Local officials as land developers: Urban spatial expansion in China. Journal of Urban Economics, 66, 57–64. Liu, Y., Li, J., & Zhang, H. (2012). An ecosystem service valuation of land use change in Taiyuan City, China. Ecological Modelling, 225, 127–132. Liu, Y., Song, Y., & Arp, H. P. (2012). Examination of the relationship between urban form and urban eco-efficiency in china. Habitat International, 36, 171–177. Long, H., Li, Y., Liu, Y., Woods, M., & Zou, J. (2012). Accelerated restructuring in rural China fueled by ‘increasing vs. decreasing balance’ land-use policy for dealing with hollowed villages. Land Use Policy, 29, 11–22. Lu, C., Wu, Y., Shen, Q., & Wang, H. (2013). Driving force of urban growth and regional planning: A case study of China’s Guangdong Province. Habitat International, 40, 35–41. Ma, Y., & Xu, R. (2010). Remote sensing monitoring and driving force analysis of urban expansion in Guangzhou City, China. Habitat International, 34, 228–235. Qiao, J., Zhang, Z., Wang, B., Cheng, G., Yang, J., & Huang, Z. (1999). Sustainable Development in Hubei Province and Improvement of Wuhan City’s Function as Central City. Earth Science – Journal of China University of Geosciences, 24, 368–373 (in Chinese). Robinson, J. (2002). Global and world cities: A view from off the map. International Journal of Urban and Regional Research, 26, 531–554. Robinson, J. (2011). Cities in a world of cities: the comparative gesture. International Journal of Urban and Regional Research, 35, 1–23. Schneider, A., & Woodcock, C. E. (2008). Compact, dispersed, fragmented, extensive? A comparison of urban growth in twenty-five global cities using remotely sensed data, pattern metrics and census information. Urban Studies, 45, 659–692. Seto, K. C., & Fragkias, M. (2005). Quantifying spatiotemporal patterns of urban land-use change in four cities of China with time series landscape metrics. Landscape Ecology, 20, 871–888. Seto, K. C., Fragkias, M., Güneralp, B., & Reilly, M. K. (2011). A meta-analysis of global urban land expansion. Plos One, 6, e23777. Shi, Y., Sun, X., Zhu, X., Li, Y., & Mei, L. (2012). Characterizing growth types and analyzing growth density distribution in response to urban growth patterns in peri-urban areas of Lianyungang City. Landscape and Urban Planning, 105, 425–433. Šímová, P., & Gdulová, K. (2012). Landscape indices behavior: a review of scale effects. Applied Geography, 34, 385–394. Su, S., Jiang, Z., Zhang, Q., & Zhang, Y. (2011). Transformation of agricultural landscapes under rapid urbanization: a threat to sustainability in Hang-Jia-Hu region, China. Applied Geography, 31, 439–449. Su, S., Xiao, R., Jiang, Z., & Zhang, Y. (2012). Characterizing landscape pattern and ecosystem service value changes for urbanization impacts at an eco-regional scale. Applied Geography, 34, 295–305. Sun, C., Wu, Z.-F., Lv, Z.-Q., Yao, N., & Wei, J.-B. (2013). Quantifying different types of urban growth and the change dynamic in Guangzhou using multi-temporal remote sensing data. International Journal of Applied Earth Observation and Geoinformation, 21, 409–417. Thapa, R. B., & Murayama, Y. (2010). Drivers of urban growth in the Kathmandu Valley, Nepal: examining the efficacy of the analytic hierarchy process. Applied Geography, 30, 70–83. Were, K. O., Dick, Ø. B., & Singh, B. R. (2013). Remotely sensing the spatial and temporal land cover changes in Eastern Mau forest reserve and Lake Nakuru drainage basin, Kenya. Applied Geography, 41, 75–86. Wu, K.-Y., Ye, X.-Y., Qi, Z.-F., & Zhang, H. (2012). Impacts of land use/land cover change and socioeconomic development on regional ecosystem services: the case of fast-growing Hangzhou Metropolitan Area, China. Cities, 31, 276–284. Wu, K.-Y., & Zhang, H. (2012). Land use dynamics, built-up land expansion patterns, and driving forces analysis of the fast-growing Hangzhou metropolitan area, eastern China (1978–2008). Applied Geography, 34, 137–145.

K. Zhou et al. / Cities 41 (2014) 141–153 Wuhan Municipal Bureau of Statistics (2012). National economic and social development statistical bulletin in 2012 of Wuhan city. Available online: . (Accessed on 11.06.13). Xie, G., Lu, C., Leng, Y., Zheng, D., & Li, S. (2003). Ecological assets valuation of the Tibetan Plateau. Journal of Natural Resources, 18, 189–196. Xu, C., Liu, M., Zhang, C., An, S., Yu, W., & Chen, J. M. (2007). The spatiotemporal dynamics of rapid urban growth in the Nanjing metropolitan region of China. Landscape Ecology, 22, 925–937.

153

Xu, X., & Min, X. (2013). Quantifying spatiotemporal patterns of urban expansion in China using remote sensing data. Cities, 35, 104–113. Yang, W., Li, F., Wang, R., & Hu, D. (2011). Ecological benefits assessment and spatial modeling of urban ecosystem for controlling urban sprawl in Eastern Beijing, China. Ecological Complexity, 8, 153–160.