Tải bản đầy đủ - 0 (trang)
2 Case Study: Urban Growth and Socioeconomic Development in the Zhujiang Delta, China

2 Case Study: Urban Growth and Socioeconomic Development in the Zhujiang Delta, China

Tải bản đầy đủ - 0trang

Urban and Regional Development

influenced every inch of the land. The possible human forces driving

LULC change can be grouped into five categories: population size,

level of affluence, technology, political and economic institutions,

and cultural attitudes and values. Accordingly, the following variables were selected for modeling (based on the availability of data):

(1) population density, (2) per-capita gross value of industrial output

(GVIO), (3) per-capita gross value of agricultural output (GVAO),

(4) urban population percentage, (5) per-capita value of gross domestic production, and (6) per-capita grain production. No independent

variables were selected from the category of technology because it is

difficult to find a surrogate for it. None were from the categories of

political and economic institutions, as well as cultural attitudes and

values, because there is little possibility of conducting an investigation of the statistical relationships of these three sets of driving forces

with the LULC change. Furthermore, the delta is a small region, has a

highly similar (if not identical) cultural background, and is under the

same political regime.

The relationship between urban growth and the independent

variables was first investigated by correlation analysis because it can

provide a general support for the explanatory variables under evaluation. However, the conceptually and empirically interrelated nature

of the variables suggests that the analysis should be conducted in a

multivariate framework. Using urban growth rate in each county/city

as a dependent variable and the preceding six independent variables,

a multiple linear regression model was applied to analyze the driving

forces of urban growth.



12.2.3



Urban LULC Modeling



In an attempt to account for the spatial pattern of land use in a logical,

consistent way, economic geographers and regional scientists have

long been exploring the field. The earliest work by Von Thunen delineated the concentric zonation pattern of agricultural land uses around

a city. His analysis has been extended in recent decades from the

countryside into the heart of a town. Regional scientists suggest that

a pattern of concentric zonation of land use also can be found within

a city because of land users’ consideration of maximum accessibility. If the assumptions employed in these analyses are correct, a city

should be surrounded by concentric zones of different types of agriculture and forest lands. The characteristics of urban encroachment

over agricultural and other natural uses around an expanding city

should be sufficiently consistent in reference to the distance from

the center of the city. To what extent do the cities in the Zhujiang

Delta and their environs conform to the patterns of concentric zonation, and what is the relationship between urban expansion and

accessibility? The techniques of GIS in conjunction with spatial modeling can facilitate these analyses.



351



352



Chapter Twelve

Many methods have been developed for measuring the spatial

pattern, the degree of spatial concentration and dispersion, of a geographic

phenomenon or process. The well-known measures include variance,

standard deviation, mean deviation, Gini coefficient, Hoover’s concentration index, and Theil’s entropy index. Each has some advantages and

disadvantages (Mulligan, 1991). The measure employed in this study is a

version of the Theil index of inequality derived from the notion of entropy

in information theory (Theil, 1967). The idea behind information entropy

is that unlikely events receive more weight than those that conform to

expectations (Nissan and Carter, 1994). This measure is expressed as

N



H ( y ) = ∑ yi log (1/yi )



(12.1)



i=1



where yi is the share of a geographic variable in the ith region in a total

of N regions, and it is required to be nonnegative and to add to 1:

N



∑ yi = 1



yi ≥ 0i = 1,..., N



(12.2)



i=1



The entropy index is always between 0 and 1. If one share is 1 and

all others are 0, then H(y) = 0 (the minimum entropy value), and we

have the maximum degree of concentration. If all shares are equal

and hence equal to 1/N, we have H(y) = log N, which is the maximum

entropy value and also the minimum degree of concentration. Obviously, the larger the entropy index, the lower is the degree of concentration. Therefore, this index can be regarded as an inverse measure

of concentration, that is, a measure of dispersion.

In this study, the entropy index is applied to measure the spatial

pattern of urban expansion in the Zhujiang Delta during the period of

1989 to 1997. Using the buffer function in GIS, two buffer images were

generated showing the proximity to the city centers and the major

roads. Ten buffer zones were created around a city center with a

width of 1000 m, and 10 buffer zones were created around a major

road with a width of 500 m. Local conditions have been taken into

account in selecting these buffer widths. Each of the two buffer images

was overlaid with the urban expansion image obtained by digital

image processing to calculate the amount of urban expansion in each

zone. The density of urban expansion then was calculated by dividing the amount of urban expansion by the total amount of land in

each buffer zone. These values of density are the shares for computing the entropy index for each city and are used to construct distancedecay functions of urban expansion.



12.2.4



Urban Growth in the Zhujiang Delta, 1989–1997



The remote sensing–GIS analysis indicates that urban or built-up

land has expanded by 47.68 percent (65,690 ha) in the Zhujiang Delta



Urban and Regional Development



Scale

50



0



Kilometers



Urban expansion area



FIGURE 12.3 Urban expansion detected by Landsat imagery in the Zhujiang Delta,

1989–1997. (Adapted from Weng, 2001.)



during the period from 1989 to 1997. Overlaying the 1989 and 1997

LULC maps reveals that most of the increase in urban or built-up

land comes from cropland (37.92 percent) and horticulture farms

(16.05 percent). Figure 12.3 shows the areal extent and spatial occurrence of the urban expansion. The overlay of this map with the city/

county boundaries reveals the spatial occurrence of urban expansion

within administrative regions. In absolute terms, the greatest urban

expansion occurred in Dongguan (23478.90 ha), Baoan (14941.08 ha),

Nanhai (8004.1 ha), and Zhuhai (5869.71 ha). However, in percentage terms, the largest increase in urban or built-up land occurred in

Zhuhai (1100.00 percent), followed by Shenzhen (306.65 percent),

Baoan (233.33 percent), and Dongguan (125.71 percent). Massive

urban sprawl in these areas can be ascribed to rural urbanization,

which is a common phenomenon in postreform China. In contrast,

the old cities, such as Guangzhou and Foshan, do not show a rapid

increase in urban or built-up land because they have no land on

which to expand further (because they already expanded fully in the

past) and because of the concentration of urban enterprises in the city



353



354



Chapter Twelve

proper. Shenzhen and Zhuhai were designated as special economic

zones at the same time, but the pace of urbanization in the two cities

is quite different. Urban development in Shenzhen was mostly complete in the 1980s, whereas Zhuhai’s urban expansion appears primarily during the period of 1989 to 1997 (5869.71 ha).

Spatial patterns of urban expansion can be examined by investigating the pattern and rate of change of urban LULC over space and time.

Because proximity to a certain object, such as a city center or a road, has

important implications in urban land development, urban expansion

processes often show an intimate relationship with distance from these

geographic objects. The density of urban expansion in the Zhujiang

Delta between 1989 and 1997 is plotted against the average distance

from a city’s geometric center in Fig. 12.4. The figure shows that as the

distance increases away from a city center, the density of urban expansion increases first to a peak and then slowly decreases. The break line

has an average distance of 3500 m from a city center. A further inquiry

into the density distribution curve reveals that the majority of urban

expansion (72.85 percent) occurs in the zone from 2500 to 5500 m from

a city center. Using a best-fit technique, a mathematical relationship

between the density of urban expansion Y and the distance from a city

center X can be established:

Y = − 0 . 00575 + 0 . 0000885X



X ≤ 3500 m



Y = 18 . 975676e −0 . 0008 8X



X > 3500 m



(12.3)



Urban expansion density (%)



Urban expansion processes in the Zhujiang Delta during the period

from 1989 to 1997 are further examined by plotting a distance decay



30

27.5

25

22.5

20

17.5

15

12.5

10

7.5

5

2.5

0

0



1000



2000



3000



4000



5000



6000



7000



8000



9000 10,000



Distance from a city center (meters)



FIGURE 12.4 Urban expansion density as a function of distance from the geometric

center of a city in the Zhujiang Delta. (Adapted from Weng and Lo, 2001.)



Urban expansion density (%)



Urban and Regional Development

20

17.5

15

12.5

10

7.5

5

2.5

0

0



500



1000



1500



2000



2500



3000



3500



4000



4500



5000



Distance from a major road (meters)



FIGURE 12.5 A distance decay curve of urban expansion in the Zhujiang Delta.

(Adapted from Weng and Lo, 2001.)



curve from a major road and establishing a mathematical equation.

Figure 12.5 indicates that the density of urban expansion decreases as

the distance increases away from a major road. Most urban expansion (66 percent) can be observed within a distance of 2000 m from a

major road. This rapid urban expansion pattern is vividly illustrated

along the superhighway from Guangzhou to Hong Kong, as seen in

Fig. 12.3, where Hong Kong investors seek sites for constructing factories and housing. The relationship between the density of urban

expansion and the distance from a major road can be expressed

mathematically as

Y = 0.2237267e



12.2.5



–0.00046X



(12.4)



Urban Growth and Socioeconomic Development



The relationship between urban growth and social and economic factors was explored initially by correlation analysis. Table 12.1 shows

the results of correlation analysis between urban expansion rate and

all possible explanatory variables. It was found that urban expansion

rate is closely related to per-capita gross value of industrial and agricultural output in 1997 (r = +0.6367), per-capita gross value of industrial output in 1997 (r = +0.6365), and urban population percentage in

1997 (r = +0.5024). These values of multiple r’s are higher than those

in 1989. The increases in correlation coefficient indicate that the urban

expansion process in the Zhujiang Delta has become more and more

related to industrial development and urban population growth. The

correlation between urban land development and urban population

percentage suggests that there probably was an improvement in

people’s living space and per-capita share of urban infrastructure

over the study period. In addition, a weak correlation was observed

between urban land development and all agricultural variables.



355



356



Chapter Twelve



Correlation

Coefficient for

1989 Data*



Correlation

Coefficient for

1997 Data*



Per-capita gross value of

industrial and agricultural

output



0.4173



0.6367



Per-capita gross value of

agricultural output



0.0436



0.3910



Per-capita gross value of

industrial output



0.4103



0.6365



Per-capita gross domestic

production



0.4978



0.4082



Per-capita grain production



0.3656



0.3208



Population density



0.2524



0.16250



Urban population percentage



0.3723



0.5024



Variable



∗Correlation coefficients are significant at the 0.05 level.



TABLE 12.1 Summary of Correlation Analysis between Urban Expansion

Rate and Socioeconomic Variables by City/County



The mapped pattern of the urban expansion rate also shows a

positive correlation with that of urban population increase (r = +0.7507),

the rate of gross value of industrial and agricultural output (r = +0.5024),

and the rate of grain production (r = +0.5162) (Table 12.2). These

dynamic relationships suggest that urban population and overall economic performance in the region are the two most critical factors contributing to the urban dispersal. In other words, spatially dispersed

urban land is more or less concentrated in the socioeconomically

advanced region of the delta.

Correlation analysis can only provide a general support for the

explanatory variables under evaluation, whereas the conceptually

and empirically interrelated nature of these variables suggests that

the analysis of the driving forces of LULC change should be conducted in a multivariate framework. The independent variables identified above may not be the most important determinants of LULC

change because empirical and theoretical associations do not necessarily work under certain circumstances. Multiple regression models

have the ability to identify the most significant factors and assess the

relative importance of different variables (Wang, 1986). The dynamic

behavior of land use can be explained by a few driving forces, as

found in many ecological systems (Holling, 1992). Along this line, two

multiple regression models have been developed to derive possible

driving forces behind the LULC change. The first model consists of



Urban and Regional Development



Urban Expansion

Rate 1989–1997*



Cropland Loss

Rate 1989–1997*



The rate of gross value of

industrial and agricultural output



0.5024



0.0601



The rate of gross value of

agricultural output



0.1773



0.0350



The rate of gross value of

industrial output



0.4104



0.0272



The rate of gross domestic

production



0.2547



0.3684



The rate of grain production



0.5162



0.1889



Population-density increment



0.1000



0.5447



The rate of urban population

growth



0.7507



0.3118



Socioeconomic Variables



∗Correlation coefficients are significant at the 0.05 level.



TABLE 12.2



Summary of Correlation Analysis between the Rates of LULC Change

and the Change Rates of Socioeconomic Variables by City/County



regression of the urban expansion rate against the possible driving

forces, and the second model consists of regression of the decline rate

of cropland against the forces. The modeling results are

UREXP_RATE = 100.925 + 588.464UBPOP_RATE

– 0.08659PCGVIAO97 + 0.08676PCGVIO97



(12.5)



This multiple regression is significant at the 0.05 level. The multiple

2

coefficient of determination R is 0.671, indicating that the predictors

(UBPOP_RATE, PCGVIAO97, and PCGVIO97) account for 67.1 percent

of variance in the urban expansion rate. The adjusted squared

multiple R reduces this proportion to 0.562, a level expected when

using this model in a new sample from the same population. The standardized multiple regression coefficient betas are 8.426 for PCGVIO97,

–8.312 for PCGVIAO97, and 0.799 for UBPOP_RATE, respectively. Thus

the order of importance in which socioeconomic factors drive urban

expansion is PCGVIO97, PCGVIAO97, and UBPOP_RATE. This result

suggests that the urban land dispersal in the delta is more related to

industrialization than to urban population growth and that industrial

development is the most likely cause of the urban land development.



12.2.6



Major Types of Urban Expansion



The computation of the entropy index indicates that the cities in the

Zhujiang Delta have an average entropy of 0.64 from a city center,

implying a relatively high degree of urban dispersion. However,



357



Urban expansion density (%)



358



Chapter Twelve

30

27.5

25

22.5

20

17.5

15

12.5

10

7.5

5

2.5

0



Type I cities



0



1000



2000



3000



4000



5000



6000



7000



8000



9000 10,000



Distance from a city center (meters)



Urban expansion density (%)



(a)

40

37.5

35

32.5

30

27.5

25

22.5

20

17.5

15

12.5

10

7.5

5

2.5

0



Type II cities



0



1000



2000



3000



4000



5000



6000



7000



8000



9000 10000



Distance from a city center (meters)

(b)



FIGURE 12.6 Two types of urban expansion patterns in the Zhujiang Delta.



significant differences in the entropy-index value persist among the

cities examined. Two major types of urban expansion can be identified: concentrated and dispersed. The difference in the urban expansion pattern is further illustrated by a distance decay curve (Fig. 12.6).

Guangzhou and Jiangmen belong to the first type, with an average

value of the entropy index of 0.42. Most of the urban expansion in

these cities in the study period is far away from the city centers.

Guangzhou’s urban development occurs primarily 6500 m outside the city center, whereas Jiangmen’s urban development occurs

5500 m away from its center. The second type of urban expansion

pattern has an average value in the entropy index of 0.69, including

Zengcheng, Nanhai, Dongguan, Shunde, Xinhui, Zhongshan, Shenzhen,



Urban and Regional Development

and Zhuhai. Urban development in these cities is found to be more

scattered and spreads over to the suburban and surrounding rural

areas from satellite images. The urban expansion zone for these cities

lies between 2500 and 8500 m from city centers. These differences in

urban expansion pattern reflect, to a large extent, the differences in

the history of urban development. Guangzhou and Jiangmen have

long been designed to function as pure urban centers, supported primarily by secondary and tertiary production. Most of the land near

the urban centers was filled before 1978 when China initiated the economic reform policy. Recent urban development in these cities has

had to seek spare land in the suburban areas, which are far from the

city center. In contrast, cities that exhibited a more dispersed development pattern during the period of 1989 to 1997 were county-seat

towns or even small towns before 1978. Urban development in these

cities therefore has had much more freedom and is subject to the

influence of the economic reform policies. In the Zhujiang Delta, the

highly dispersed urban development pattern can be attributed to

rural urbanization, the influence of Hong Kong and foreign investment, and the lack of urban development planning.



12.2.7



Summary



The use of Landsat TM data to detect LULC changes generally has

been a success. The digital image classification, coupled with GIS, has

demonstrated an ability to provide comprehensive information on

the direction, nature, rate, and location of LULC changes as a result of

rapid industrialization and urbanization. Given that digital LULC

data are available, a further examination of the urban dispersal patterns or changes for other individual categories can be pursued. It has

been demonstrated that the combined use of GIS and spatial modeling is sufficiently powerful to discern urban expansion patterns. The

spatial process of urban expansion in the delta has shown an intimate

relationship with the distance from major roads and from the geometric center of a city.

Satellite-derived LULC data can be linked with socioeconomic

data to study the driving forces of LULC change or other environmental problems. Correlation and regression analyses are the two

useful integrated approaches. The results of the analysis from this

chapter suggest that urban land development in the delta is closely

related to industrial development and urban population growth.



12.3



Discussion: Integration of Remote Sensing

and GIS for Urban Growth Analysis

Methodologically, this chapter has focused on the development of an

integrated approach of remote sensing, GIS, and spatial analysis for

urban growth studies. It has been demonstrated that the integrated



359



360



Chapter Twelve

approach is sufficiently powerful and is needed for evaluating urban

growth and its relationship with its driving forces.

The technique of remote sensing has been used to derive LULC

maps by means of classification of satellite images. GIS then allows for

detection of changes between the two years and extraction of information on changes in urban cover. These raster-based digital data have

been further combined with vector-based socioeconomic data to investigate the driving forces of urban growth. Correlation and regression

analyses are the two approaches for integration. Finally, the spatial patterns of urban expansion are examined by integrated use of remote

sensing, GIS, and spatial analysis techniques. GIS are used to compute

the value of satellite-derived urban expansion density in each buffer

zone, and these data, in turn, become the input to subsequent spatial

analyses, including the derivation of distance decay curves.

Technically, this chapter has used an integration strategy called

loose coupling (Goodchild et al., 1992), in which remote sensing, GIS,

and spatial analysis interplay by exchanging data files. This study

demonstrates that this integration algorithm allows full use of existing analytical techniques and produces satisfactory results without

use of any advanced computer programming language. However,

this approach has limitations. For example, it may be error-prone

in transferring data from one system to another, and is also timeconsuming.

Another knotty problem in the integrated approach is related to the

data model. Remote sensing systems collect data in raster mode, whereas

most GIS datasets are built in vector format. The combined use of

remotely sensed data with cartographic data generated in GIS requires

going through three stages: rasterization/vectorization, registration, and

overlay. A great many problems are produced in the format-conversion

process that requires a lot of time and effort to sort out. This study has

approached the raster-vector overlay problem by means of creation of

raster masks. However, the best solution for this kind of problem can be

achieved only when a closer integration comes into being, which will

enable us to query raster pixels within vector polygons and to combine

image statistics with socioeconomic data within polygons.



References

Campell, J. B. 1983. Mapping the Land: Aerial Imagery for Land Use Information.

Washington, D. C.: Association of American Geographers.

Goodchild, M., Haining, R., Wise, S., et al. 1992. Integrating GIS and spatial

data analysis: Problems and possibilities. International Journal of Geographical

Information Systems 6, 407–423.

Holling, C. S. 1992. Cross-scale morphology, geometry, and dynamics of ecosystems. Ecological Monographs 62, 447–502.

Jacobson, H. K., and M. F. Price. 1991. A Framework for Research on the Human

Dimensions of Global Environmental Change. Geneva: International Social

Science Council Human Dimensions of Global Environmental Change Program

(Report No. 1).



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

2 Case Study: Urban Growth and Socioeconomic Development in the Zhujiang Delta, China

Tải bản đầy đủ ngay(0 tr)

×