Advertisement

Landscape and Ecological Engineering

, Volume 14, Issue 2, pp 257–267 | Cite as

Quantifying spatiotemporal patterns concerning land change in Changsha, China

  • Bin Quan
  • Hongge Ren
  • Robert Gilmore PontiusJr.
  • Peilin Liu
Original Paper
  • 216 Downloads

Abstract

Changsha has undergone speedy socio-economic development, rapid modification of industrial structure, and acceleration of urbanization, which has influenced land cover change during the most recent three decades. Policies have aimed to conserve total agricultural area, but it is not clear how successful these policies have been. Our purpose is to characterize and interpret spatiotemporal patterns of land change with respect to the policy to maintain agricultural area in Changsha, China. Maps at 1990, 2000, and 2010 show four land categories: Built, Forest, Crop and Other. We compute change components and apply Intensity Analysis to compare the land changes during two time intervals: 1990–2000 and 2000–2010. We also compare the central region to the peripheral region during 1990–2010. The maps show that Changsha’s land change accelerated from 1990–2000 to 2000–2010. Change was more intensive in the central region than in the peripheral region. Crop and Forest experienced net decreases while Built experienced net increase during both time intervals and in both regions. Built’s gain targeted Crop and avoided Forest during both time intervals and in both regions. The central region’s largest change component is quantity change, due to Built’s net gain. The peripheral region’s largest change component is exchange, due to simultaneous transitions from Forest to Crop and from Crop to Forest. According to these data, policies have not maintained the quantity of Crop, as the peripheral region has not gained Crop sufficiently to compensate for Crop’s loss from the central region.

Keywords

Change components Intensity Analysis Urbanization Socio-economic development Land cover Agricultural area 

Introduction

Land use changes have occurred rapidly since the economic restructuring in China (Liu et al. 2010). Chinese cities tripled in size during 1978–2010 (Schneider and Mertes 2014). A series of development strategies including “Rising of Central China” and “The Belt and Road Initiative” has attempted to influence land changes, especially in central China, in provinces such as Hunan Province. Hunan’s capital city is Changsha (27°51′–28°40′N, 111°53′–114°5′E), which has an area of about 12,000 km2 covering five national development zones and nine provincial industrial parks. Changsha has experienced rapid industrialization and urbanization for more than two decades, which makes it representative of cities in central China. Changsha has recently established the “XiangJiang New Area,” which strives to be the major area in the Yangtze River Economic Zone and the core growth pole of the “The Belt and Road” region. This region includes the transition belt of the east coastal area and the mid-west area, along with part of the Yangtze economic belt and coastal economic belt. The XiangJiang New Area is the core of the national pilot zone for overall reform of a resource-conserving economy and an environment-friendly society. Changsha has undergone speedy socio-economic development, rapid modification of industrial structure, and acceleration of industrialization and urbanization. Changsha’s population has grown from 5.5 million in 1990 to 6.5 million in 2010. Changsha’s urban area has grown from 367 km2 in 1990 to 959 km2 in 2010. Changsha’s percentage of annual gross domestic product has shifted from 40% industrial in 1990 to 54% industrial in 2010. If an urban center is surrounded with cropland, as Changsha’s urban center is, then cropland will lose when built land expands. If a spatial extent is to maintain its agricultural output, then cropland must gain elsewhere within the spatial extent to compensate for cropland’s loss near urban centers. Furthermore, if the best quality cropland is lost, then a larger size of poor quality cropland must gain in order to maintain the same level of agricultural production. It is not clear whether this type of compensation has been occurring in Changsha. Changsha has adopted policies to conserve resources, especially cropland. The farmland requisition-compensation balance policy aims to protect cropland by assuring that agricultural loss at some locations is compensated by agricultural gain at other locations. The goal of our article is to quantify the patterns of land change in Changsha during the temporal extent from 1990 to 2010, and to compare the center region to the periphery region, to see whether the policies have been able to maintain agricultural area in the spatial extent.

Several scholars have investigated land change in Changsha. Zhang et al. (2010) simulated the interaction among various agents based on a multi-agent system, to research the urban expansion process in Changsha City. They found that urbanization is extending from suburbs to townships with convenient transportation and that land availability in the downtown area of Changsha city will likely become more problematic in the future. Deng et al. (2012) established an index system to understand spatial–temporal change of cultivated land in Changsha City. They showed that cultivated land advanced in a wave-like form in municipal districts and Liuyang City from 2001 to 2009 while it continued to increase steadily in Wangcheng and Ningxiang Counties. Zhang et al. (2014) used entropy and fractal dimension to study land change in Changsha City. Their results demonstrated that entropy rose during 2005–2008, fell during 2008–2009 then rose again during 2009–2012. They divided counties and cities into areas with high, middle and low entropy. Luo and Li (2014) analyzed land change in Changsha during 10 years and forecasted the trend for each type of change. They forecasted expansion of artificial surfaces due to transitions from farmland, wetlands and unused land. Chen et al. (2015) analyzed the relationship between land change and urbanization in response to industrial evolution in the city of Changsha. They concluded that the increase of urbanization will result in the reduction of agricultural land and the increase of built land. These articles use various methods, e.g. geographical information science, remote sensing and landscape metrics, to analyze the interactions among land, society, economy and ecology. None of the articles have used the recently developed quantitative techniques known as change components (Pontius and Santacruz 2014) and Intensity Analysis (Aldwaik and Pontius 2012, 2013). Our manuscript applies these methods to describe and to interpret the patterns of land change in Changsha.

Materials and methods

Materials

The Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences (RESDC) provided vector maps for 1990 and 2000 (http://www.resdc.cn). RESDC has six primary categories: arable land, woodland, grassland, water, construction land and unused land. The data sources of the classified vector maps were the Landsat Thematic Mapper (TM) and Landsat Enhanced Thematic Mapper images taken in 1989/1990 and 1999/2000, henceforth referred to as “1990” and “2000.” The data from bands 2, 4, and 5 were used. Visual interpretation aided the land cover classification. The Landsat images were enhanced using linear contrast stretching and histogram equalization to identify ground control points to a common ALBERS coordinate system based on 1:100,000 topographic maps of China. After geometric correction, the average georeferencing error in the Landsat images was less than 50 m, meaning about two pixels. Field surveys and sample checks showed that the overall classification errors were 7.08% for 1990 and 2.55% for 2000 (Liu et al. 2005).

Lei (2014) used the maps for 1990 and 2000 from RESDC to help create the classification for 2010. The primary data source of the classified vector maps for 2010 is the Landsat TM images taken in March, June, September and November of 2010 with a spatial resolution of 30 m and the WGS-84 coordinate system, ALBERS projection, central meridian 105°E and double standard parallels 25°N and 47°N, and the Krasovsky spheroid. The classified vector map at 2000 from RESDC and 1:100,000 topographic maps of China served as reference images. The 30 control points in Changsha produced root mean squared error of less than one pixel (Lei 2014).

We categorized the cultivated land, woodland and construction land, respectively as Crop, Forest and Built. Then we aggregated water, grassland, and unused land to form a category called Other. Figure 1 shows the raster maps, which have a spatial resolution of 30 m. The central region includes Changsha’s central urban area, Wangcheng County and Changsha County. The peripheral region is Ningxiang County and Liuyang County.
Fig. 1

Maps of land categories and changes in Changsha

We detected categorical transitions by overlaying pairs of maps from sequential time points. Each pair of time points generates a square contingency table, which we dissect using the mathematical methods in “Methods.” Tables 1 and 2 show transition matrices during 1990–2000 and 2000–2010 for Changsha. Tables 3 and 4 show transition matrices during 1990–2010 for the periphery and center regions.
Table 1

Transition matrix from 1990 to 2000 as percentage of Changsha

 

To 2000

Total

Loss

Built

Forest

Crop

Other

From 1990

 Built

2.20

0.01

0.00

0.00

2.21

0.01

 Forest

0.20

62.32

0.01

0.03

62.57

0.25

 Crop

0.45

0.11

31.29

0.03

31.88

0.59

 Other

0.03

0.01

0.16

3.14

3.34

0.21

 Total

2.89

62.45

31.46

3.20

100.00

1.06

 Gain

0.68

0.13

0.18

0.07

1.06

 
Table 2

Transition matrix from 2000 to 2010 as percentage of Changsha

 

To 2010

Total

Loss

Built

Forest

Crop

Other

From 2000

 Built

2.16

0.16

0.53

0.04

2.89

0.73

 Forest

1.37

60.12

0.78

0.18

62.45

2.33

 Crop

1.56

0.62

28.72

0.57

31.46

2.75

 Other

0.12

0.24

0.32

2.52

3.20

0.68

 Total

5.20

61.14

30.35

3.31

100.00

6.48

 Gain

3.05

1.02

1.63

0.79

6.48

 
Table 3

Transition matrix from 1990 to 2010 as percentage of periphery

 

To 2010

Total

Loss

Built

Forest

Crop

Other

From 1990

 Built

0.68

0.08

0.40

0.02

1.18

0.50

 Forest

0.66

68.05

0.59

0.20

69.49

1.45

 Crop

0.81

0.64

25.09

0.23

26.77

1.68

 Other

0.02

0.18

0.14

2.21

2.56

0.35

 Total

2.16

68.95

26.23

2.66

100.00

3.98

 Gain

1.49

0.91

1.14

0.45

3.98

 
Table 4

Transition matrix from 1990 to 2010 as percentage of center

 

To 2010

Total

Loss

Built

Forest

Crop

Other

From 1990

 Built

3.56

0.16

0.51

0.06

4.28

0.72

 Forest

3.20

44.17

1.12

0.16

48.65

4.48

 Crop

4.12

0.79

36.34

0.90

42.15

5.82

 Other

0.43

0.32

0.67

3.51

4.92

1.41

 Total

11.31

45.43

38.64

4.62

100.00

12.43

 Gain

7.75

1.26

2.30

1.11

12.43

 

Methods

Change components

We budgeted the difference during each time interval into three components called “quantity,” “exchange,” and “shift” (Pontius and Santacruz 2014). Table 5 gives the mathematical notation that the equations below use to compute those components.
Table 5

Mathematical notation

Symbol

Meaning

C tij

Number of pixels that transition from category i to category j during interval t

C tji

Number of pixels that transition from category j to category i during interval t

d tj

Annual difference for category j during interval t

D t

Annual difference overall during interval t

ε tij

Annual exchange between categories i and j during interval t

e tj

Annual exchange component for category j during interval t

E t

Annual exchange component overall during interval t

G tj

Intensity of annual gain of category j during interval t relative to size of category j at Yt+1

i

Index for a category

j

Index for a category

J

Number of categories

L ti

Intensity of annual loss of category i during interval t relative to size of category i at Yt

q tj

Annual quantity component for category j during interval t

Q t

Annual quantity component overall during interval t

R tin

Intensity of annual transition from category i to category n during interval t relative to size of category i at Yt

s tj

Annual shift component for category j during interval t

S t

Annual shift component overall during interval t

t

Index for time interval

W tn

Uniform intensity of annual transition from all non-n categories to category n during interval t relative to size of all non-n categories at Yt

Y t

Year at start of time interval t

Y t+1

Year at end of time interval t

Equations 14 express change in terms of annual percentages of the spatial extent, thus each denominator is the product of the temporal duration and the size of the spatial extent. Equation 1 gives the difference during time interval t for category j. Equation 2 gives the quantity component during interval t for category j. The quantity component is the absolute change in size of the category. Equation 3 gives the exchange between categories i and j during time interval t. Categories i and j form exchange when some locations transition from category i to category j while other locations transition from j to i. Equation 4 sums the exchanges for category j to give the exchange component for category j. Equation 5 gives the shift component for category j during interval t. The shift component is the difference minus the quantity component minus the exchange component. Equations 68 sum the components for each category to give the components overall during time interval t. Division by 2 is necessary in Eqs. 68 because each location of temporal difference involves two categories, i.e. the losing category and the gaining category. Equation 9 shows how the difference overall is the sum of the three components.
$$ d_{tj} = \frac{{\left\{ {\left[ {\mathop \sum \nolimits_{i = 1}^{J} \left( {C_{tij} + C_{tji} } \right)} \right] - 2 \times C_{tjj} } \right\} 100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{i = 1}^{J} \mathop \sum \nolimits_{j = 1}^{J} C_{tij} }}, $$
(1)
$$ q_{tj} = \frac{{\left| {\mathop \sum \nolimits_{i = 1}^{J} \left( {C_{tij} - C_{tji} } \right) 100\% } \right|}}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{i = 1}^{J} \mathop \sum \nolimits_{j = 1}^{J} C_{tij} }}, $$
(2)
$$ \varepsilon_{tij} = \frac{{2\;{\text{MINIMUM}}\left( {C_{tij} ,C_{tji} } \right) 100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{i = 1}^{J} \mathop \sum \nolimits_{j = 1}^{J} C_{tij} }}\quad {\text{for}} \; i > j \; {\text{and }}\varepsilon_{tij} = 0 \; {\text{for }}i \le j, $$
(3)
$$ e_{tj} = \mathop \sum \limits_{i = 1}^{J} \left( {\varepsilon_{tij} + \varepsilon_{tji} } \right) = \frac{{2\left\{ {\left[ {\mathop \sum \nolimits_{i = 1}^{J} {\text{MINIMUM}}\left( {C_{tij} ,C_{tji} } \right)} \right] - C_{tjj} } \right\}100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{i = 1}^{J} \mathop \sum \nolimits_{j = 1}^{J} C_{tij} }}, $$
(4)
$$ s_{tj} = d_{tj} - q_{tj} - e_{tj} , $$
(5)
$$ Q_{t} = \frac{{\mathop \sum \nolimits_{j = 1}^{J} q_{tj} }}{2}, $$
(6)
$$ E_{t} = \frac{{\mathop \sum \nolimits_{j = 1}^{J} e_{tj} }}{2}, $$
(7)
$$ S_{t} = \frac{{\mathop \sum \nolimits_{j = 1}^{J} S_{tj} }}{2}, $$
(8)
$$ D_{t} = \frac{{\mathop \sum \nolimits_{j = 1}^{J} d_{tj} }}{2} = Q_{t} + E_{t} + S_{t} . $$
(9)

Intensity Analysis

Intensity Analysis is a mathematical framework that compares uniform intensities to observed intensities of temporal changes among categories (Aldwaik and Pontius 2012, 2013). We compute the category level and transition level intensities for Changsha City during 1990–2000 and 2000–2010 and in both the central and peripheral regions during 1990–2010.

Intensity Analysis’ category level compares a uniform intensity of change to observed intensities of loss and gain for each category during each time interval. Equation 9 gives the uniform change during interval t at the category level. Equation 10 gives the observed loss intensity for each category i during interval t. Equation 11 gives the observed gain intensity for each category j during interval t. If the observed intensity is less than the uniform intensity, then we say the observed intensity is dormant. If the observed intensity is greater than the uniform intensity, then we say the observed intensity is active. Intensity at the category level identifies the categories for which change is intensive relative to the intensity of change overall in the spatial extent.
$$ L_{ti} = \frac{{\left[ {\left( {\mathop \sum \nolimits_{j = 1}^{J} C_{tij} } \right) - C_{tii} } \right] 100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{j = 1}^{J} C_{tij} }}, $$
(10)
$$ G_{tj} = \frac{{\left[ {\left( {\mathop \sum \nolimits_{i = 1}^{J} C_{tij} } \right) - C_{tjj} } \right] 100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{i = 1}^{J} C_{tij} }}, $$
(11)
Intensity Analysis’ transition level examines how the gain of a particular category transitions from losing categories during each time interval. We compare the uniform transition intensity to observed intensities during each time interval. Equation 12 gives the uniform transition intensity for the gain of category n during time interval t. Equation 13 gives the observed transition intensity from each category i, given the gain of category n during interval t. If the observed transition intensity from i is less than the uniform transition intensity, then we say the gain of n avoids category i. If the observed transition intensity from i is greater than the uniform transition intensity, then the gain of n targets category i.
$$ W_{tn} = \frac{{\left[ {\left( {\mathop \sum \nolimits_{i = 1}^{J} C_{tin} } \right) - C_{tnn} } \right] 100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{j = 1}^{J} \left[ {\left( {\mathop \sum \nolimits_{i = 1}^{J} C_{tij} } \right) - C_{tnj} } \right]}}, $$
(12)
$$ R_{tin} = \frac{{C_{tin} 100\% }}{{\left( {Y_{t + 1} - Y_{t} } \right)\mathop \sum \nolimits_{j = 1}^{J} C_{tij} }}. $$
(13)

Results

Changsha during 1990–2000 and 2000–2010

Change components

Figure 2 shows the results from Eqs. 69 for change overall in Changsha. Change accelerated by a factor of six from 1990–2000 to 2000–2010. The quantity component accounts for most of the change during the first time interval, while the exchange component accounts for most of the change during the second time interval.
Fig. 2

Total change in Changsha during 1990–2000 and 2000–2010

Figure 3 shows the results from Eqs. 1 to 5, which describe how the individual land categories contribute to each component during each time interval. The plus symbol above the quantity component indicates that the quantity change for the particular category is a net gain, whereas the minus symbol indicates net loss. The net gain of Built is the largest component during both time intervals.
Fig. 3

Category change in Changsha during 1990–2000 and 2000–2010 where + indicates net increase and indicates net decrease

Intensity Analysis

Figure 4 shows the results from Intensity Analysis’ category level during the two time intervals. Each category has a pair of bars, where the top bar shows loss and the bottom bar shows gain. The bars that extend to the left from the middle axis are the annual change areas. The bars that extend to the right from the middle axis are the annual change intensities, which are Lti and Gtj from Eqs. 10 and 11. The dashed vertical line is the uniform intensity of annual change for the entire spatial extent, which is Dt from Eq. 9. If a category’s bar ends before the uniform line, then the change is dormant for that category; if a category’s bar extends beyond the uniform line, then the change is active for that category. Crop experiences the largest loss while Built experiences the largest gain during both time intervals. The intensities are active for Crop’s loss and for Built’s gain. Crop loses more than it gains during both time intervals.
Fig. 4a, b

Intensity Analysis for category level in Changsha during 1990–2000 and 2000–2010

Figure 5 shows results from Intensity Analysis’ transition level for Built’s gain. The bars to the left of the middle axis show the size of each annual transition. The left side of Fig. 5 shows that Built’s gain derives mainly from Crop during both time intervals. The bars to the right of the middle axis show the intensity of each annual transition. This intensity Rtin is the size of the annual transition divided by the size of the losing category at the interval’s initial time. If Built’s gain were spread uniformly across the non-Built categories at the initial time, then all the intensity bars would end at the uniform line Wtn. The right side of Fig. 5 shows that Built’s gain targets Crop and avoids Forest during both time intervals. Other has a large intensity, which is due to Other’s small size in the denominator of the intensity.
Fig. 5a, b

Intensity Analysis for transition to Built in Changsha during 1990–2000 and 2000–2010

Periphery and center regions during 1990–2010

Change components

Figure 6 shows the change percentage of Changsha’s periphery and center regions from 1990 to 2010. The concentration of change in the peripheral region is approximately one-third the concentration in the central region. Exchange is the largest component in the peripheral region, whereas quantity is the largest component in the central region.
Fig. 6

Total change in periphery and center regions during 1990–2010

Figure 7 shows the behavior of the individual categories in each region during 1990–2010. Crop is most involved in the changes in the peripheral region, while Built is most involved in the changes in the central region. The periphery’s largest components are exchanges, in particular the simultaneous transitions from Crop to Forest and from Forest to Crop. The center’s largest components are the net gain of Built and net losses of Forest and Crop.
Fig. 7

Category change in periphery and center regions during 1990–2010 where + indicates net increase and indicates net decrease

Intensity Analysis

Figure 8 shows the results from Intensity Analysis’ category level in the periphery and center regions. Crop shows the largest loss and Built shows the largest gain in both regions. The intensities of Crop’s loss and Built’s gain are active. Forest’s losses and gains are both dormant, due mainly to Forest’s large size. Crop loses more than it gains in both regions.
Fig. 8a, b

Intensity Analysis for category level of periphery and center regions during 1990–2010

Figure 9 shows results from Intensity Analysis’ transition level for Built’s gain. The bars to the left of the middle axis show the size of each annual transition. The bars to the right of the middle axis show the intensity of each annual transition. Built gains primarily from Crop in terms of both size and intensity in both regions. Built’s gain targets Crop and avoids Forest in both regions.
Fig. 9a, b

Intensity Analysis for transition to Built in periphery and center regions during 1990–2010

Discussion

Land policies

China has a goal to maintain the quantity of Crop area; therefore, China has policies concerning the transition from Crop to Built. Our results show that this concern is legitimate. Intensity Analysis shows that Crop lost and Built gained actively, as Built’s gain targeted Crop during both time intervals and in both regions. If the policy is to succeed in its goal to maintain the quantity of Crop area, then Crop’s gain must be at least as large as Crop’s loss within the boundaries of the policy. Therefore, the success of this policy depends in part on the spatial boundaries of the policy.

Changsha is one possible boundary for the policy. If one region of Changsha experiences net decrease of Crop while another region experiences at least the same quantity of net increase of Crop, then the policy’s goal would be attained. Our results show that Crop experienced net decrease in the Center region. Therefore, if Changsha were to maintain Crop area, then the periphery region would have to compensate by demonstrating net increase of Crop. However, our results show that the periphery also experienced net decrease of Crop. Consequently, Crop area in Changsha shrank during 1990–2000 and 2000–2010, because the periphery did not compensate sufficiently for the net decrease of Crop in the center. However, this does not imply that compensation was zero. Crop had a positive exchange component within each region, as Crop lost to a particular category at some locations while Crop gained from the same category at other locations within a single region. Furthermore, Crop had a positive shift component in both regions. Shift for Crop occurred when Crop lost to Built more than Crop gained from Built, while simultaneously Crop gained from a non-Built category more than Crop lost to the non-Built category. Exchange and shift are forms of compensation. The sum of exchange and shift for Crop was larger than the quantity component for Crop in each region. This indicates that substantial compensation existed, but the compensation was not sufficient to maintain the quantity of Crop in Changsha.

The entirety of China is another possible boundary for the policy. If one part of China experiences net decrease of Crop while another part experiences at least the same quantity of net increase of Crop, then the policy’s goals would be attained. Our results show that Changsha experienced net loss of Crop during both 1990–2000 and 2000–2010. Therefore, Changsha placed a burden on other parts of China to compensate for Changsha’s net decrease of Crop.

Comparison with other reports

Our findings are consistent with reports of other studies concerning land change in Changsha. Zhou and He (2007) report that growth of urban area accelerated in Changsha steadily through four time intervals from 1949 to 2004. Chen et al. (2015) found that arable land decreased while built land increased steadily from 2003 to 2013. These changes have occurred in a context of policies to manage the desires for industrial development, food security and environmental health. For example, the National Development and Reform Commission of China designated in 2007 the National Demonstration Area as the metropolitan districts of Changsha City, Zhuzhou City, and Xiangtan City. The goal was to harmonize economic development with resource-saving and environment-friendly land use practices (Quan et al. 2013). The government has policies and regulations to protect cropland (Wang and Zhang 2013). However, our results show that those efforts have not reversed the trend of shrinking cropland. Newly built land continues to replace high-quality cropland, while new cropland is developed on poorer quality land (Liu 2008).

The land change patterns in Changsha are similar to patterns elsewhere, particularly in China. Yang and Li (2000) found an overall decrease in arable land from 1978 to 1996 in China, but with substantial regional variation. Huang et al. (2012) used Intensity Analysis to characterize changes among Built, Agriculture and Natural land categories in a coastal watershed in Southeast China between 1986, 1996, 2002 and 2007. Their results showed that land change accelerated. Gains and losses of Built were active during all time intervals, while Agriculture showed net loss during the temporal extent. Built’s gain targeted Agriculture during all time intervals, similar to in Changsha. Zhou et al. (2014) studied the same watershed but with six categories at five time points from 1986 to 2010. Their results revealed that land change accelerated. The Built-up category was active in both gains and losses during all time intervals, as Built-up’s gain consistently targeted Agriculture, Orchard and Water. Akinyemi et al. (2017) used Intensity Analysis to find similar acceleration of land change in Kigali, Rwanda.

Conclusion

We computed three components of change and performed Intensity Analysis to reveal the patterns of change among four land categories across two time intervals, within two regions of Changsha, China. Our goals were to characterize land change and to interpret the results with respect to policies to maintain agricultural area. Results show that change accelerated from 1990–2000 to 2000–2010. Change was more intensive in the center urban region than in the peripheral region. Crop and Forest accounted for the largest net decreases while Built accounted for the largest net increases during both time intervals and in both regions. Built’s gain targeted Crop during both time intervals and in both regions, which confirms that the policies focus on legitimate concerns. Results show some compensation of agricultural area, as both regions of Changsha experienced Crop loss at some locations and simultaneous Crop gain at other locations. However, both regions of Changsha experienced net decrease in quantity of Crop, thus the policy did not maintain Crop area within Changsha. If China is to maintain agricultural area at the national level, then China must rely on regions outside Changsha to compensate for the agricultural area that Changsha has lost.

Notes

Acknowledgements

This work was supported by the National Natural Science Foundation of China (grant 41271167), the Key Project of Hunan Provincial Department of Education of China (grant 17A067) and the Recruitment Program of High-end Foreign Experts of the State Administration of Foreign Experts Affairs of China (grant GWD201543000243). The Talent Introduction Project of Hengyang Normal University also supported this work via grant 17D03. The US National Science Foundation supported this work through the Long-Term Ecological Research Network via grant OCE-1637630. Any opinions, findings, conclusions, or recommendations expressed in this article are those of the authors and do not necessarily reflect those of the funders. Clark Labs facilitated this work by creating the software TerrSet. Anonymous reviewers supplied constructive feedback that helped to improve this paper. The research complies with the laws of the country in which the authors performed the research.

Compliance with ethical standards

Conflict of interest

The authors declare they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

References

  1. Akinyemi FO, Pontius RG Jr, Braimoh AK (2017) Land change dynamics: insights from intensity analysis applied to an African emerging city. J Spat Sci 62(1):69–83Google Scholar
  2. Aldwaik SZ, Pontius RG Jr (2012) Intensity analysis to unify measurements of size and stationarity of land changes by interval, category, and transition. Landsc Urban Plan 106(1):103–114CrossRefGoogle Scholar
  3. Aldwaik SZ, Pontius RG Jr (2013) Map errors that could account for deviations from a uniform intensity of land change. Int J Geogr Inf Sci 27(9):1717–1739CrossRefGoogle Scholar
  4. Chen Y, Xie B, Li X, Deng C, Zhu Y (2015) The preliminary research on relationship between the change of land use and urbanization in Changsha from 2003 to 2013. Econ Geogr 35(1):149–154Google Scholar
  5. Deng C, Xie B, Li X, He Q, Zhu D (2012) Analysis on spatial-temporal change of cultivated land intensive use in Changsha city. Trans Chin Soc Agric Eng 28(1):230–237Google Scholar
  6. Huang J, Pontius RG Jr, Li Q, Zhang Y (2012) Use of intensity analysis to link patterns with processes of land change from 1986 to 2007 in a coastal watershed of southeast China. Appl Geogr 34:371–384CrossRefGoogle Scholar
  7. Lei S (2014) Comparison of land use changes between case areas from the developed east and less developed central China: a case study of the Changsha and Quanzhou city. Dissertation. Hunan University of Science and Technology, XiangtanGoogle Scholar
  8. Liu G (2008) Thinking about the cultivated land accounts for the dynamic balance in Changsha. Land Resour Herald 5(6):7–9Google Scholar
  9. Liu J, Zhan J, Deng X (2005) Spatio-temporal patterns and driving forces of urban land expansion in China during the economic reform era. Ambio 34(6):450–455CrossRefPubMedGoogle Scholar
  10. Liu J, Zhang Z, Xu X, Kuang W, Zhou W, Zhang S, Li R, Yan C, Yu D, Wu S, Jiang N (2010) Spatial patterns and driving forces of land use change in China during the early 21st century. J Geogr Sci 20(4):483–494CrossRefGoogle Scholar
  11. Luo K, Li R (2014) Land-use change process in recent ten years and trends analysis in Changsha in future. J Univ Chin Acad Sci 31(5):632–639Google Scholar
  12. Pontius RG Jr, Santacruz A (2014) Quantity, exchange, and shift components of difference in a square contingency table. Int J Remote Sens 35(21):7543–7554CrossRefGoogle Scholar
  13. Quan B, Xiao Z, Römkens MJM, Bai Y, Lei S (2013) Spatiotemporal urban land use changes in the Changshutan Region of Hunan Province in China. J Geogr Inf Syst 5:136–147Google Scholar
  14. Schneider A, Mertes CM (2014) Expansion and growth in Chinese cities, 1978–2010. Environ Res Lett 9(024008):1–11Google Scholar
  15. Wang W, Zhang Y (2013) Land consolidation and sustainable development. China Popul Resour Dev 14(1):13–18Google Scholar
  16. Yang H, Li X (2000) Cultivated land and food supply in China. Land Use Policy 17(2):73–88CrossRefGoogle Scholar
  17. Zhang H, Zeng Y, Bian L, Yu X (2010) Modelling urban expansion using a multi agent-based model in the city of Changsha. J Geogr Sci 20(4):540–556CrossRefGoogle Scholar
  18. Zhang L, Li M, Zhou L (2014) Analysis of land use structure change in changsha city based on information entropy and fractal dimension. J Nat Sci Hunan Normal Univ 37(4):1–5, 15Google Scholar
  19. Zhou G, He Y (2007) The influencing factors of urban land expansion in Changsha. J Geogr Sci 17(4):487–499CrossRefGoogle Scholar
  20. Zhou P, Huang J, Pontius RG Jr, Hong H (2014) Land classification and change intensity analysis in a coastal watershed of southeast China. Sensors 14(7):11640–11658CrossRefPubMedGoogle Scholar

Copyright information

© International Consortium of Landscape and Ecological Engineering and Springer Japan KK, part of Springer Nature 2018

Authors and Affiliations

  1. 1.College of City and Tourism and HIST Hengyang Base of UNESCOHengyang Normal UniversityHengyangPeople’s Republic of China
  2. 2.School of Resource, Environment and Safety EngineeringHunan University of Science and TechnologyXiangtanPeople’s Republic of China
  3. 3.School of GeographyClark UniversityWorcesterUSA

Personalised recommendations