Research Article Physical and Socioeconomic Driving Forces of …downloads.hindawi.com/journals/ddns/2016/8061069.pdf · Research Article Physical and Socioeconomic Driving Forces

Research ArticlePhysical and Socioeconomic Driving Forces of Land-Use andLand-Cover Changes A Case Study of Wuhan City China

Xiangmei Li12 Ying Wang3 Jiangfeng Li3 and Bin Lei4

1School of Environment Resources and International Trade Hubei University of Economics Wuhan 430205 China2Center of Hubei Cooperative Innovation for Emissions Trading System Wuhan 430205 China3School of Public Administration China University of Geosciences Wuhan 430074 China4School of Economics and Management China University of Geosciences Wuhan 430074 China

Correspondence should be addressed to Xiangmei Li xmlihustaliyuncom

Received 3 December 2015 Revised 17 February 2016 Accepted 14 March 2016

Academic Editor Elmetwally Elabbasy

Copyright copy 2016 Xiangmei Li et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

To investigate precise nexus between land-use and land-cover changes (LUCC) and driving factors for rational urbanmanagementwe used remotely sensed images to map land use and land cover (LULC) from 1990 to 2010 for four time periods using Wuhancity China as a case study Partial least squares (PLS) method was applied to analyze the relationships between LUCC and thedriving factors mainly focusing on three types of LULC that is arable land built-up area and water area The results were asfollows (1) during the past two decades the land-use pattern in Wuhan city showed dramatic change Arable land is made up ofthe largest part of the total area The increased built-up land came mainly from the conversion of arable land for the purpose ofeconomic development (2) Based on the Variable Importance in Projection (VIP) the joint effects of socioeconomic and physicalfactors on LUCCwere dominant though annual temperature especially annual precipitation proved to be less significant to LUCCPopulation tertiary industry proportion and gross output value of agriculturewere themost significant factors for threemajor typesof LULCThis study could help us better understand the driving mechanism of urban LUCC and important implications for urbanmanagement

1 Introduction

Land-use and land-cover changes (LUCC) increasingly havebeen regarded as a primary source of global environmentalchange such as emission of greenhouse gases global climatechange loss of biodiversity and loss of soil resources [1ndash3]However the causes of LUCC are complex and change overtime and from region to region [4] In the early 1990s keepingin view the diverse reasons and causes of land-use changeemphasizes the importance of interdisciplinary research toaddress the issue of land-use change with a particular focuson the human dimensions [5] LUCC have been led by a setof socioeconomic driving forces and conditioned by differ-ent natural endowments [6] that determine the trajectoriesof landscape development [7] Understanding the drivingmechanisms of LUCC caused by a variety of driving forces isone of the major goals of global change research in recentdecades [8ndash10]

To understand the human and biophysical processes ofLUCC many researchers focus on the various forces drivingLUCC including socioeconomic [11] demographic [12]political [13] technological [14] biophysical [15] and indus-trial structure [16] to provide effective support for developingurban land planning and management regulations To com-prehensively analyze the driving factorrsquos effects and mitigatethe negative impacts of land-use change Shu et al (2014) [17]investigated the effects of various factors including naturalecoenvironment factors land control policies accessibilityfactors and neighborhood factors on urban land expansionduring various periods in different regions Chen et al(2014) [18] selected industrial structure GDP transportationand policy as the driving factors to study the impacts onurban land expansion and sustainable urban development inShenzhen and Dongguan In addition an integration of bio-physical and human factors was applied in the explanation of

Hindawi Publishing CorporationDiscrete Dynamics in Nature and SocietyVolume 2016 Article ID 8061069 11 pageshttpdxdoiorg10115520168061069

2 Discrete Dynamics in Nature and Society

LUCC dynamics of Mediterranean Europe due to its par-ticular climatic and physical conditions [15] Obviously theoutstanding characteristics of the studied cases such as econ-omy culture climate and policy oftenwere considered as theimportant driving forces in the explanation of LUCC dynam-ics However though many researches on land-use changehave been conducted climate factors were seldom availableon driving analysis of LUCC

Various research methods were employed to explore thenexus between LUCC and their driving forces Multivariateregression was used to model how the major forces drive thephysical expansion of urban land cover at the global level[19] Li et al (2013) [20] applied binary logistic regressionto investigate the effects of the selected driving variableson the probability of urban expansion Analytic hierarchyprocess (AHP) as a subjective method provided rigorousquantitative measures to understand the interactive processof urban growth and factors [21] System dynamics (SD) wascombinedwithCLUE-Smodel to reflect the complexity of theland-use system [22] regarded as a macrolevel ldquotop-downrdquoimplementation process However few attempts have beenmade to investigate precise nexus between LUCC and drivingfactors Partial least squares (PLS) method as a major regres-sion technique for multivariate data may handle highly cor-related noise-corrupted data sets by explicitly assuming thedependency between variables and estimating the underlyingstructures [23] It could effectively reflect the significant PLScomponents using the cross validation technique [24] In thepaper PLS was used to accurately reflect the nexus betweenLUCC and driving factors meanwhile determining the sig-nificant components of the selected driving factors

In the paper our main objectives were to address the pro-cesses of land-use dynamics in Wuhan city and its integrateddriving forces through combining satellite-based efforts atmapping land use and land cover (LULC) and physicalsocioeconomic data Based on RS images in 1990 1995 20002005 and 2010 we analyzed the variant change of each LULCtype during 4 periods (1990ndash1995 1995ndash2000 2000ndash2005and 2005ndash2010) Thirteen variables of physical and socioeco-nomic factors were selected as the potential driving factors ofLUCC Partial least squares (PLS) method is applied to selectthe important driving factors and analyze the relationshipsbetween LUCC and the factors triggering each land-usechange type mainly focusing on three major types of LUCCthat is arable land built-up land and water area Finally wesuggested some possible management measures that are cru-cial for future sustainable utilization and management of itsexisting land resources for example managing urban growthand protecting cultivated land

2 Study Area

Wuhan city as a central hinterland megalopolis of Chinais situated in the east of Jianghan plain and covers over849441 km2 (113∘411015840ndash115∘051015840 E 29∘581015840ndash31∘221015840 N) Terrain isdominated by flat areas with a surface elevation ranging from0 to 100m and the slope is less than 10∘ making up 9578of the total The low hills which have an elevation between200 and 400m and a slope varying from 10 to 25∘ constitute

China

Elevation (m)0ndash404001ndash10010001ndash20020001ndash40040001ndash800

0ndash150

0 10 25 50(km)

151ndash500501ndash10001001ndash25002501ndash4500

Slope (∘)

Wuhan city

Hubei

N

Figure 1 Location of the study area

389 of its total area while the mountainous area (ele-vation 400sim800m slope gt25∘) accounts for only 033(Figure 1) The Yangtze River and its largest tributary HanRiver meet here which divides Wuhan into three partsHankou Hanyang and Wuchang commonly known as ldquothethree towns of Wuhanrdquo The area has the subtropical humidmonsoon climate Its climate feature is obvious characterizedby abundant rainfall summer heat and winter cold with anannual temperature of approximately 158ndash175∘C and meanannual precipitation of 1150ndash1450mm A huge water networkwas formed with the Yangtze River as the backbone and sup-plemented by quantities of lakes or pondsTherefore Wuhancity has a reputation of ldquocity of hundred lakesrdquoThewater areaaccounts for 256 of the city area

After entering the twenty-first century Wuhanrsquos eco-nomic growth speed has accelerated In 2010 per capitaGDP in Wuhan amounted to 66520 Yuan RMB [25] whichwas much higher than the average GDP per capita (22936Yuan RMB) in China The urbanization level increased from562 in 1990 to 647 in 2010 Wuhanrsquos population has beenincreasing by an annual rate of 120 since 1978 With atotal population of about 836 million the population densityin Wuhan city was up to 984 personskm2 in 2010 whichwas much higher than the average population density (141personskm2) of China at the same period [25 26] 323 oftotal population was engaged in farming

However with the population growth and economicdevelopment LULC in Wuhan has changed dramaticallyas evidenced by the rapid increase of built-up land at theexpense of occupying massive high quality arable land inthe urban fringe areas Moreover to gain more construction

Discrete Dynamics in Nature and Society 3

space some land exploitation activities such as deforestationand the filling of lakes for the sake of construction spaceoccurred In addition climate factors have caused certaineffects on LUCC [27] Therefore it is very necessary toidentify the important factors triggering LUCC in order tomitigate the contradiction between human and land

3 Methodology and Data

31 Remotely Sensed Imagery Preprocessing Interpretationand Land-Use Detection The analysis of LUCC in Wuhancity was based on five LULC maps which were derived fromhistorical LandsatMSSTMETM+ images (pathrow 1223912338 12339) To obtain the information of land-use classifi-cation the paper used LandsatMSS images for 1990 and 1995Landsat TM images for 2000 and 2005 and Landsat ETM+images for 2010 The interpretation scheme adopted was thecombination of supervised classification and visual modifi-cation All the procedures were processed using ENVI 50andArcGISDesktop 101 softwareThedetailed interpretationprocess was elaborated in our previous published paper [28]Landscape of the study area was divided into six major land-use types pasture arable land built-up land forest waterarea (including rivers ponds reservoirs and aquaculturewaters) and unused land Moreover in order to solve theinsufficiency of sample data and increase modeling accuracylinear interpolation method was used to generate the time-series data of land use from the year 1990 to 2010 Thus wecould use the time-series data of six major land-use types toexplore the driving forces contributing to land-use changein Wuhan city In addition the six major land-use typesthat is pasture arable land built-up land forest water areaand unused land were labeled as 119910

1 1199102 1199103 1199104 1199105 and 119910

6

respectively A flow chart describing the detection procedureof LUCC was shown in Figure 2

32 Data of Driving Factors LUCC is a result of the complexinteractions between human activity and physical factorsAccording to the related studies [11 13 15 17] four types ofdriving factors of LUCC have been generally identified phys-ical factors socioeconomic factors neighborhood factorsand land-use policies [20] Although similar driving factorscould be found in the related literatures the magnitude ofthe effects which contributed to landscape change differed[29] In this paper we selected thirteen variables representingphysical economic and social factors Because the papermainly focused on a longitudinal time-series effect of thedriving forces on land-use change neighborhood factors asspatial data were considered in the study Meanwhile we didnot include land-use policies because of the lack of long-termdata on these aspects

Physical factors are the fundamental determinants of theextent of land-use change Climate conditions such as precip-itation and temperature affect the potential extent of LULCby restricting the water supply Therefore the annual precip-itation (119909

1) and the annual temperature (119909

2) were included

as the climatic factors for studying the driving factors ofLUCC in Wuhan city These climatic data were collected

from China Meteorological Data Sharing Service Sys-tem (httpwwwesciencegovcnmetdatapageindexhtml)Socioeconomic factors include time-series data on popula-tion (119909

3) urbanization rate (119909

4) regional gross domestic

product (GDP) (1199095) tertiary industry proportion (119909

6) gross

industrial output value (1199097) gross output value of agricul-

ture (1199098) total retail sales of consumer goods (119909

9) gross

imports and exports (11990910) per capita disposable income of

households (11990911) turnover volume of passenger traffic (119909

12)

and turnover volume of freight traffic (11990913) These historical

data were extracted from the statistical yearbook of Wuhan[25] and China statistical yearbook [26] These factors reflectthe dynamics of population and urbanization economicand industrial structure social investment technologicalprogress and external traffic conditions which may havepotential influences on LULC change in Wuhan city Inaddition to achieve comparability in price across the studiedperiod all the factors related to price were converted to theconstant price of 1990 based on the corresponding index ofeach factor and the general consumer price index

33 Partial Least Squares Regression PLS is a well-knownmultivariate statistical technique for the analysis of highdimensional data developed by Wold (1975 1984) [30 31]The function of PLS included multiple linear regression anal-ysis canonical correlation analysis and principal componentanalysis It originated in the social sciences and becamepopular in many fields including first chemometrics [31]sensory evaluation [32] and statistics [33] and ecology [34]The increasing popularity of PLS is partly due to more pow-erful functions of dealing with multicollinearity that existedin the dataset especially in some case of fewer observationsrelative to the number of variables compared to conventionalstatistical method [35]

PLS aims at investigating relationships between the pre-dictor (ie independent or explanatory) and the response(ie dependent or explained) variables The dataset of inde-pendent variables the matrix 119883

119898times119899 consists of 119898 variables

(columns) and 119899 objects (rows) And the dataset of dependentvariables constitutes a response vector 119884

119899 times 1 In this paper

the independent variables were the thirteen driving factorsand dependent variables were different land-use types Thebasic idea of PLS is to identify two sets of components orfactors (also called latent variables) that is 119905

119894and119906119894 extracted

from the independent variables 119883 and dependent variables119884 respectively which can maximize the explained variancein original 119883-values and 119884-values There are two importantplots used to evaluate the applicability of the PLS method11990511199061scatter graph and the 119905

11199052oval graph If there is linear

relationship between 1199051and 119906

1 indicating 119883 significantly

correlated to 119884 it is reasonable to construct PLS linear modelbetween 119883 and 119884 [33] If 119905

11199052of the sample data is all

included in the oval plot these sample data are homogeneousand can be accepted perfectly

It is critical to determine the optimal number of latentvariables when building the PLS model In the paper Leave-One-Out Cross Validation (LOOCV) was used to determinethe number of latent variables (principal components) The


Data acquisition

Atmospheric correction

Geometric correction

Region of interestreduction

Image enhancement

sites

sele

ctio

nRe

pres

enta

tive t

rain

ing

Clas

s sep

arab

ility

eval

uatio

n

Max

imum

-like

lihoo

dcla

ssifi

catio

n

Supervised classification

Land-cover classificationsystem adoption

Interpretation schemeestablishing

Visual modification

assessmentAccuracy

Meet requirement

After treatment

Remote-senseddatapreprocessing

The final classificationresults

LULCmap 2010map 2005

LULCmap 2000

LULCLULCmap 1995

LULCmap 1990

Change detection

Fail

Figure 2 Flow chart of detection procedure of LULC changes

LOOCV systematically omits one plot at a time and performs119899-estimates of the parameters using 119899minus1 plotsThis procedurewas repeated for all plots [36]The LOOCV statistical param-eters including 119877

ℎ

2 119876ℎ

2 119877cm2 and 119876cm

2 were calculatedto determine the confidence in the model performance 119877

ℎ

2

was the percent of variation of the independent variablesexplained by the model which could measure how well themodel fits the data (ℎ represents the number of principal com-ponents) 119876

ℎ

2 was the percent of variation of the dependentvariables predicted by the model according to cross valida-tion which indicated how well the model predicts new data119877cm2 and 119876cm

2 were the cumulative 1198772 and 1198762 over all theselected PLS components When 119877cm

2 and 119876cm2 are greater

than 08 themodel was considered to exhibit good predictiveability

In the PLSmethod theVariable Importance in Projection(VIP) was employed to determine the relative importanceof driving factors with respect to the response variables[35] Terms with large VIP values are the most relevant for

explaining the dependent variables The 119895th predictorrsquos VIPscore to the response variable can be written as

VIP119895= radic

119873

119877119889(119884 1199051 1199052 119905

119898)

119898

sumℎ=1

119877119889(119884 119905ℎ) 1199082ℎ119895 (1)

where119873 is the dimension of input variables119898 is the numberof latent variables119877

119889(119884 1199051 1199052 119905

119898) represents the explana-

tory capacity of components for dependent variables 119908ℎ119895

denotes the loading weight vectors for each component andrepresents the importance of the variable 119895 for component ℎ119895 = 1 2 3 119873 ℎ = 1 2 3 119898

For VIP scores that are larger than a threshold that isVIP119895ge 119878VIP the predictors should be selected as input

variables In general the greater-than-one rule is used as a cri-terion for variable selection In the paper VIP scores mainlyare used to discern the different importance of the driving fac-tors especially between physical factors and socioeconomic


N

Administrative boundaryBuilt-up landUnused landForestWater areaArable landPasture

0 10 25 50(km)

(a) LULC map in 2010

N

Arable land unchangedNewly increased area

From built-up land

To built-up land

From unused land

To unused land

From forest

To forest

From water area

To water area

Reduced areaFrom pasture

To pasture

0 10 25 50(km)

(b) Arable land from 1990 to 2010

N

From water area

To water area

Built-up land unchangedNewly increased area

From unused landFrom forest

From arable land


To unused landTo forest

To arable landTo pasture

0 10 25 50(km)

(c) Built-up land from 1990 to 2010

N

Water area unchangedNewly increased area

From built-up landFrom unused landFrom forestFrom arable land


To built-up landTo unused landTo forestTo arable landTo pasture

0 10 25 50(km)

(d) Water area from 1990 to 2010

Figure 3 LULC map and the spatial change of main LULC types in Wuhan city from 1990 to 2010

factors in the land-use change SIMCA-P 115 Software wasperformed to implement the PLS procedure

4 Results and Analysis

41 Characteristics of LULC Change in Wuhan City Figure 3showed the change tendency of main LULC types in Wuhancity from 1990 to 2010 As observed arable land is made up ofthe largest part of the total area and mainly distributed in theplain area crisscrossed by a network of rivers and lakes Incontrast the distribution of the built-up land was more cen-tralized and mainly situated in the central district of Wuhanbut there were also some rural settlements scattered evenly inthe whole study area During the period 1990sim2010 built-upland showed a noticeable trend of outward expansion alongthe Yangtze River and Han River occupying a large amountof arable land and some water body The water area is alsoan important component of Wuhan and provides importantecological functions The central and southern parts havemore lakes and rivers gathering there when compared withthe northern areas As shown in Figure 3 from 1990 to 2010some waters in the central area were converted to pasture and

built-up land while the newly emerging built-up land wasmainly converted from pasture and arable land The forestof Wuhan was mainly located in the northwestern andnortheastern hilly areas and some forest reserve or urbanforest parks in the south The land-use pattern of the forestwas relatively stable throughout the study periodThe pastureand unused land are distributed in the remote mountainsand waterfront areas Due to the restrains from topographyand geomorphology drainage and irrigation and ecologicalprotection the exploitation potential of the unused land wasquite limited

Figure 4 showed the net change of LULC types duringfour periods (1990ndash1995 1995ndash2000 2000ndash2005 and 2005ndash2010) in Wuhan city As observed LULC has changedgreatly especially during the latter two periods From 1990to 1995 arable land and the forest decreased by 10398 and645 km2 respectively In contrast built-up land water areaand unused land increased by 6590 3562 and 831 km2 inthe same periodThese trends of changes slowed down in theperiod from 1995 to 2000 By the end of 2000 built-up landand water area expanded by 4124 and 2401 km2 Most of thenewly developed areas were from arable land and the unused


1990ndash1995 1995ndash2000 2000ndash2005 2005ndash2010Pasture 060 minus105 minus418 minus461Arable land minus10398 minus4246 minus12391 minus10149Built-up land 6590 4124 12779 12123Forest minus645 306 minus701 minus387Water area 3562 2401 1007 194Unused land 831 minus2480 minus276 minus1320

minus15000

minus10000

minus5000

000

5000

10000

15000N

et ch

ange

of a

rea (

unit

km

2)

Figure 4 Net change of each LULC type inWuhan city (unit km2)

land and their areas decreased by 4246 and 2480 km2respectively The period from 2000 to 2005 witnessed themost significant changes in arable land and built-up landThe former shrunk by 12391 km2 while the latter increasedby 12779 km2 and the areas of the other four LULC typeswere relatively stable within this period The increased built-up land came mainly from the conversion of arable land forthe purpose of developmentThe trends of changes continuedin the last period from 2005 to 2010 Therefore we couldconclude that arable land built-up land and water area werethree main LULC types Therefore in Section 44 we mainlyanalyzed the relationships between the three main LULCtypes and driving factors to reveal the driving mechanism ofLUCC in Wuhan city

42 Parameters of Partial Least Square Regression and ItsAccuracy As shown in the 119905

11199061scatter graph (Figure 5(a))

the 11990511199061relationship of the sample data is nearly linear thus

PLS regression model is suitable for the study of the problemof this paper From the 119905

11199052oval graph (Figure 5(b)) it was

clear that all the sample data collected from 1990 to 2010 wereacceptable except the sample point from the year 2010 becausethis point was not included in the oval In order to obtainbetter model fitting performance the regression model wasrebuilt after eliminating the outlier

As shown in Table 1 two indicators that is 119877ℎ

2 and 119876ℎ

2were used to determine the number of principal componentsWhen three principal components were extracted from vari-ables the values of119876

ℎ

2 lt 0 (for all LULC types) which couldnot meet the decision criterion of 119876

ℎ

2 ⩾ 00975 In additionit is generally held that the regression effect is ideal when119877cm2 and119876cm

2 are all larger than 08 [37]Thus two principalcomponents were the best for the total and most of theLULC types Table 1 shows 949 variation of the dependentvariables can be explained by the model and the accumu-lated cross validation reached 935 suggesting that theregression results are effective

43 Variable Importance Analysis For in-depth analysis ofvariablersquos interpretative ability the importance of a variable

Table 1 Values of parameters under Leave-One-Out Cross Valida-tion regression

LULC types ℎ 119877ℎ

2119877cum2

119876ℎ

2119876cum2

Total1 0897 0897 0886 08862 0052 0949 0429 09353 0005 0953 minus0055 0931

Pasture1 0947 0947 0942 09422 0020 0967 0257 09573 0000 0967 minus0110 0953

Arable land1 0973 0973 0967 09672 0015 0988 0469 09833 0002 0990 minus0059 0982

Built-up land1 0990 0990 0989 09892 0002 0992 minus0021 09893 0001 0993 0009 0989

Forest1 0906 0906 0896 08962 0001 0907 minus0128 08853 0001 0908 minus0118 0874

Water area1 0713 0713 0667 06672 0234 0947 0772 09243 0010 0957 minus0090 0917

Unused land1 0852 0852 0855 08552 0038 0891 0201 08843 0014 0905 0026 0887

was given by the Variable Importance in Projection (VIP) Itis generally held that the independent variable plays a veryimportant role when the VIP value is larger than 10 andrelatively important when the VIP value is larger than 05and smaller than 10 whereas the VIP value smaller than 05indicates weak interpretative ability [38] As shown in theVIPhistogram (Figure 6) it was clear that the annual precipitation(1199091) did not have much influence on the dynamics change

in land system over 20 years from 1990 to 2010 becausethe VIP value was only 0284 On the contrary gross outputvalue of agriculture (119909

8) had the greatest interpretative ability

followed by population (1199093) and urbanization rate (119909

4) per

capita disposable income of households (11990911) regional gross

domestic product (GDP) (1199095) total retail sales of consumer

goods (1199099) turnover volume of passenger traffic (119909

12)

tertiary industry proportion (1199096) gross industrial output

value (1199097) gross imports and exports (119909

10) and turnover

volume of freight traffic (11990913) which proved to be the very

important factors in explaining the changes that occurred inWuhan because their VIP values were larger than or equalto 10 The VIP value of annual temperature (119909

2) was 0714

which indicated that annual temperature was relatively lesssignificant compared with these economic factors Here weobtained the terms with VIP values of independent variablesin total land system which could provide basics for drivingforces of main LULC types in Section 44

With regard to physical factors especially for climaticfactors existing studies have provided enormous evidenceof climatic factors influencing LULC change around theworld [39] For example changing in temperature or rainfall


minus4 minus2 0 2 4minus6

t1

minus6

minus4

minus2

0

2

4

u1

(a)

minus5 0 5 10minus10

t1

minus3

minus2

minus1

0

1

2

3

t2

(b)

Figure 5 (a) 11990511199061scatter graph (b) 119905

11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1

Figure 6 The VIP histogram of each variable

patterns might impact crop yield farm profitability regionalproductivity costs and the number of people at risk ofhunger [40] Over the past two decades in Wuhan city com-pared with the fluctuation trends of the annual precipitation(Figure 7(a)) the annual temperature continually rose inincreasing trend (Figure 7(b))Though the effect of LUCC onregional and global climate change has beenwell documented[41] in turn the temperature changes have reduced the regen-eration ability of forests [42] furtherly affecting the LUCCMeanwhile decreasing rainfall can lead to the need for irri-gation which would increase the land for irrigation facilitiesTherefore the integrated factors including physical factorsthat is temperature and several socioeconomic factors coop-eratively shaped LULC changes An elaborate analysis waslaunched illustrating the relationship of these variables withthree main LULC types in Section 44

44 Relationship between Different Land Types andDriving Factors

441 Arable Land Types and Driving Factors To reveal thedirection and strength of the impact of each independent

variable on arable land we used the PLS method to establishthe regression equation as follows It is worth mentioningthat thirteen variables were considered to establish model inorder to not only be compared with the results based on VIPvalues but also discern different contributions of physical andsocioeconomic factors to LUCC1199101= 0030119909

1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)

The equation showed the importance of every indepen-dent variable when explaining the dynamics of arable landfrom 1990 to 2010 1198772 = 0988 indicated that thirteen factorsinterpreted 988 variation information of arable land and itsuggested that the regression result was effective From thesigns of coefficients ((2) Figure 8) it could be known thatthe annual precipitation positively influenced the variationof arable land while the annual temperature and the othereleven socioeconomic factors negatively did Specifically theinfluences of tertiary industry proportion population andgross output value of agriculture on arable landwere themostobvious and the coefficients were negativeThe related resultsdemonstrated that the area of arable land decreased alongwith the growth of tertiary industry proportion and popu-lation [43] In Wuhan city the share of the tertiary industryboomed from 250 to 514 during 1978ndash2010 [25] Appar-ently the adjustment of industrial structure in Wuhan wasa very significant factor of arable land change and it playedan important role in reducing the quantity of arable landBesides the increase of urban population and the improve-ment of peoplersquos living standards had facilitated the prosper-ousness of the real estate industry which led to the conver-sion of arable land into constructing settlements as well as


Annual precipitationTrendline

1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)

Annual temperatureTrendline

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)

Figure 7 (a) The graphs of annual precipitation (b) The graphs of annual temperature

Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13

Figure 8 The coefficients of arable land

providing various services for entertainment Gross outputvalue of agriculture was also negatively correlated with thechange of arable landThe shares of farming forestry animalhusbandry andfishery in the gross output value of agriculturewere 613 09 246 and 132 respectively in 1990However they were 538 07 263 and 187 respec-tively in 2010 [25]The variation in the proportion illustratedthe direction of agricultural structure adjustment in Wuhanwhich greatly affected the LULC of Wuhan city From theanalysis of LULC patterns from 1990 to 2010 we could detectthat a large proportion of arable land withdrew from cropcultivation for other purposes such as fishponds orchardsor economic forest which consisted in generating higherincome through these production activities In addition theannual temperature posed negative effect on arable land inWuhan city which indicated the area of arable land slowlydecreased along with local climate warming although theircorrelation was not significant The annual precipitation wasthe only one positively correlated with arable land in Wuhancity which indicated that the area of arable land had increasetendency along with the growth of the annual precipitationalthough their correlation was not significant

442 Built-Up Land Types and Driving Factors The regres-sion equation between the variables and built-up land was asfollows

1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)

As shown in Figure 9 and (3) the signs of coefficientswere completely opposite to that of arable land Except forthe annual precipitationwhich negatively influenced the vari-ation of built-up land the other twelve factors positively did1198772 = 0992 indicated that the thirteen variables could inter-pret 988 variation information of built-up land Further-more most of the socioeconomic factors contributed evenlyto the change of built-up land Population gross output valueof agriculture urbanization rate and tertiary industry pro-portion had the highest explanatory ability compared withthe other factors considered in this paperThe area of built-upland would increase 0104 0104 0101 and 0101when these four factors increase by 1 respectively Thesecond group of factors followed were per capita disposableincome of households GDP and total retail sales of consumergoods and they would lead to the increase of built-up landby 0094 0093 and 0092 when they increase by 1These three factors had some similarities because they bothreflected the performance of economic development and theimprovement of peoplersquos life As known increasing economicactivities and economic output led to increasing demandfor built-up land With the robust economic growth localgovernments had the ability to expand public finances fordevelopment purposes including generous investments inurban regeneration key infrastructures universities andschools medical treatment facilities and recreational parks


Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13

Figure 9 The coefficients of built-up land

therefore resulting in rapid expansion of built-up land[44] Meanwhile the increase of the disposable income ofhouseholds stimulated the demand for residential areas andfacilities to improve their quality of life Therefore bothgovernment and residents contributed to the increase ofbuilt-up land A 1 change in turnover volumes of passengerand freight traffic respectively induced 0087 and 0072change in built-up land Though the measure of trafficcondition had relative effect on LULC the construction ofrailway and highway programs directly led to the conversionfrom farmland to construction land In contrast annualprecipitation negatively influenced the variation of built-upland because the increase of precipitation could facilitate theregeneration ability of forests [27] For temperature it hadpositive feedback on the change of built-up land Temperaturewill increase with the increase of urbanized and unexploitedlands while it will decrease with the increase of woodlandgrassland and farmland [27]

443 Water Area and Driving Factors The regression equa-tion between the independent variables and water area was asfollows

1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)

The relationship between these independent variablesand water area was more complex than arable land and built-up land Primarily 1198772 = 0947 indicated that the thirteenvariables could interpret 947 variation information ofwater area From the signs of coefficients ((4) Figure 10) itcould be known that the annual precipitation gross industrialoutput value gross imports and exports and turnover volumeof freight traffic posed rarely negative influence on thevariation of water area while the other nine factors positivelydid Tertiary industry proportion proved to be the mostsignificant factor in driving the change of not only arable landand built-up land but also water area Moreover it was note-worthy that the annual temperature had the second highestexplanatory ability to the dynamics of water area while it wasnot a significant factor in shaping arable land and built-up

Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13

Figure 10 The coefficients of water area

land changes 1 change of the annual temperature wouldlead to 0271 change of water area greater than that of arableland and built-up land by 0106 and 0068 This resultshowed that climate warming would have major impacts onwater resources which mainly contributed to the hydrologiccycle precipitation evaporation patterns the magnitude andtiming of runoff and the intensity and frequency of floodsand droughts thus affecting seasonal availability of watersupply [45] However the magnitude of annual temperatureon the variation of water area still was limited compared withthe joint effects of socioeconomic factors With regard to theinfluences of population and urbanization a largely urban-ized and populated city was bound to push up the higherdemand for water resources in Wuhan reflecting in theamounts of water not only for surviving but also for offeringecological function and recreational function

5 Conclusions

In the paper we investigated the processes of land-usedynamics and the effects of physical and socioeconomic driv-ing forces on LUCC in Wuhan city Hubei province Chinathrough combining RS and PLS The results indicate thatthe land-use pattern inWuhan city showed dramatic changewith different net increase or decrease change speeds Built-up land increased 35616 km2 and water area also enlarged7164 km2 during the period 1990ndash2010 And the largestreducing magnitude for arable land reached 37183 km2 Theincreased built-up land came mainly from the conversion ofarable land for the purpose of economic development BasedonVIP values the joint effects of socioeconomic and physicalfactors on LUCC were dominant in the land system thoughannual temperature especially annual precipitation provedto be less significant to LUCC Population tertiary industryproportion and gross output value of agriculture were themost significant factors for three major types of LUCC 1change of the annual temperature would lead to 0271change of water area

The study revealed a strong influence of integrated driv-ing forces on LULC at regional scale In the paper PLSwas applied with the Variable Importance in Projection asan appropriate technique contributing to eliminating thecodependency among the variables which has a betterprospect than the traditional multiple linear regressions inthis regard However the PLS methodology also has certain


limitations that should be made clear The interaction rela-tionships between the driving forces and land-use changesare dynamics and nonlinear As PLS is a method of lin-ear modeling the regression coefficient cannot fully cap-ture the nonlinear relationship between driving forces andLULC change Hence to explore the detailed information ofdynamic process of land-use change in complex land systemsa hybrid PLS regression and nonlinear models such as backpropagation neural network (BPNN) can be constructedto accommodate possible nonlinearity between the drivingforces and LULC This represents the future research effortsin deriving better dynamic models of LULC

Competing Interests

The authors declare that they have no competing interests

Acknowledgments

This research was funded by the National Natural ScienceFoundation of China (41501183) and the Teaching LaboratoryOpen-End Fund of China University of Geosciences China(Grant no skj2014199)

References

[1] Millennium Ecosystem Assessment Ecosystems and HumanWell-Being Opportunities and Challenges for Business andIndustry World Resources Institute Washington DC USA2005

[2] E F Lambin M D A Rounsevell and H J Geist ldquoAreagricultural land-usemodels able to predict changes in land-useintensityrdquo Agriculture Ecosystems amp Environment vol 82 no1ndash3 pp 321ndash331 2000

[3] M Boissiere D Sheil I BasukiMWan andH Le ldquoCan engag-ing local peoplersquos interests reduce forest degradation in CentralVietnamrdquo Biodiversity and Conservation vol 18 no 10 pp2743ndash2757 2009

[4] M Qasim K Hubacek and M Termansen ldquoUnderlying andproximate driving causes of land use change in district SwatPakistanrdquo Land Use Policy vol 34 pp 146ndash157 2013

[5] B L Turner R H Moss and D Skole ldquoRelating land useand global land-cover change a proposal for an IGBP-HDPcore project A report from the IGBPHDPWorking Group onLand-UseLand-Cover Changerdquo Global Change Report RoyalSwedish Academy of Sciences Stockholm Sweden 1993

[6] A Farina ldquoThe cultural landscape as amodel for the integrationof ecology and economicsrdquo BioScience vol 50 no 4 pp 313ndash320 2000

[7] M Burgi A M Hersperger and N Schneeberger ldquoDrivingforces of landscape changemdashcurrent and new directionsrdquo Land-scape Ecology vol 19 no 8 pp 857ndash868 2005

[8] E F Lambin and H Geist Land-Use and Land-Cover ChangeLocal Processes and Global Impacts Springer Berlin Germany2006

[9] R R Rindfuss S J Walsh B L Turner II J Fox and V MishraldquoDeveloping a science of land change challenges and method-ological issuesrdquo Proceedings of the National Academy of Sciencesof the United States of America vol 101 no 39 pp 13976ndash139812004

[10] R A Pielke Sr ldquoLand use and climate changerdquo Science vol 310no 5754 pp 1625ndash1626 2005

[11] Y Xie Y Mei T Guangjin and X Xuerong ldquoSocio-economicdriving forces of arable land conversion a case study ofWuxianCity Chinardquo Global Environmental Change vol 15 no 3 pp238ndash252 2005

[12] Y Shi R Wang L Fan J Li and D Yang ldquoAnalysis on land-use change and its demographic factors in the original-streamwatershed of Tarim River based on GIS and statisticrdquo ProcediaEnvironmental Sciences vol 2 pp 175ndash184 2010

[13] R Kanianska M Kizekova J Novacek and M Zeman ldquoLand-use and land-cover changes in rural areas during differentpolitical systems a case study of Slovakia from 1782 to 2006rdquoLand Use Policy vol 36 pp 554ndash566 2014

[14] F Hasselmann E Csaplovics I Falconer M Burgi and AM Hersperger ldquoTechnological driving forces of LUCC con-ceptualization quantification and the example of urban powerdistribution networksrdquo Land Use Policy vol 27 no 2 pp 628ndash637 2010

[15] P Serra X Pons andD Saurı ldquoLand-cover and land-use changein aMediterranean landscape a spatial analysis of driving forcesintegrating biophysical and human factorsrdquoApplied Geographyvol 28 no 3 pp 189ndash209 2008

[16] J Xiao Y Shen J Ge et al ldquoEvaluating urban expansion andland use change in Shijiazhuang China by using GIS andremote sensingrdquo Landscape and Urban Planning vol 75 no 1-2pp 69ndash80 2006

[17] B Shu H Zhang Y Li Y Qu and L Chen ldquoSpatiotemporalvariation analysis of driving forces of urban land spatial expan-sion using logistic regression a case study of port towns inTaicang City ChinardquoHabitat International vol 43 pp 181ndash1902014

[18] J Chen K-T Chang D Karacsonyi andX Zhang ldquoComparingurban land expansion and its driving factors in Shenzhen andDongguan Chinardquo Habitat International vol 43 pp 61ndash712014

[19] K C Seto M Fragkias B Guneralp and M K Reilly ldquoA meta-analysis of global urban land expansionrdquo PLoS ONE vol 6 no8 Article ID e23777 2011

[20] X LiW Zhou and Z Ouyang ldquoForty years of urban expansionin Beijing what is the relative importance of physical socioeco-nomic and neighborhood factorsrdquo Applied Geography vol 38no 1 pp 1ndash10 2013

[21] R B Thapa and Y Murayama ldquoDrivers of urban growth in theKathmandu valley Nepal examining the efficacy of the analytichierarchy processrdquo Applied Geography vol 30 no 1 pp 70ndash832010

[22] G Luo C Yin X Chen W Xu and L Lu ldquoCombining systemdynamic model and CLUE-S model to improve land use sce-nario analyses at regional scale a case study of Sangong water-shed in Xinjiang Chinardquo Ecological Complexity vol 7 no 2 pp198ndash207 2010

[23] A Singh A R Jakubowski I Chidister and P A Townsend ldquoAMODIS approach to predicting streamwater quality inWiscon-sinrdquo Remote Sensing of Environment vol 128 pp 74ndash86 2013

[24] SWoldHMartens andHWold ldquoThemultivariate calibrationproblem in chemistry solved by the PLS methodrdquo in MatrixPencils B Kagstrom and A Ruhe Eds vol 973 pp 286ndash293Springer Berlin Germany 1983

[25] Wuhan Statistical Bureau Wuhan Statistical Yearbook ChinaStatistics Press Beijing China 1991ndash2011


[26] China State Statistics Bureau China Statistical Yearbook ChinaStatistics Press Beijing China 1996ndash2005

[27] C Huang M Zhang J Zou et al ldquoChanges in land use climateand the environment during a period of rapid economicdevelopment in Jiangsu Province Chinardquo Science of the TotalEnvironment vol 536 pp 173ndash181 2015

[28] Y Wang X Li and J Li ldquoStudy on the response of ecologicalcapacity to land-usecover change in Wuhan City a remotesensing and GIS based approachrdquo Scientific World Journal vol2014 Article ID 794323 11 pages 2014

[29] P H Verburg J R Ritsema van Eck T CM de Nijs M J Dijstand P Schot ldquoDeterminants of land-use change patterns in theNetherlandsrdquo Environment and Planning B Planning amp Designvol 31 no 1 pp 125ndash150 2004

[30] H Wold ldquoSoft modeling by latent variables the nonlineariterative partial least squares approachrdquo in Perspectives inProbability and Statistics Papers in Honour of MS Bartlett JGani Ed pp 520ndash540 Academic Press London UK 1975

[31] SWold A Ruhe HWold andW J Dunn III ldquoThe collinearityproblem in linear regression The partial least squares (PLS)approach to generalized inverserdquo SIAM Journal on Scientific andStatistical Computing vol 5 no 3 pp 735ndash743 1984

[32] S Wold ldquoPersonal memories of the early PLS developmentrdquoChemometrics and Intelligent Laboratory Systems vol 58 no 2pp 83ndash84 2001

[33] HMartens and T NaesMultivariate Calibration JohnWiley ampSons London UK 1989

[34] L M Carrascal I Galvan and O Gordo ldquoPartial least squaresregression as an alternative to current regression methods usedin ecologyrdquo Oikos vol 118 no 5 pp 681ndash690 2009

[35] P H Garthwaite ldquoAn interpretation of partial least squaresrdquoJournal of the American Statistical Association vol 89 no 425pp 122ndash127 1994

[36] R B Thapa M Watanabe T Motohka and M ShimadaldquoPotential of high-resolutionALOS-PALSARmosaic texture foraboveground forest carbon tracking in tropical regionrdquo RemoteSensing of Environment vol 160 pp 122ndash133 2015

[37] ZWang F Yin Y Zhang and X Zhang ldquoAn empirical researchon the influencing factors of regional CO

2emissions evidence

from Beijing city Chinardquo Applied Energy vol 100 pp 277ndash2842012

[38] M D Ma X J Ma Y Z Xie and T Ma ldquoAnalysis the rela-tionship between ecological footprint (EF) of ningxia and influ-encing factors Partial Least-Squares Regression (PLS)rdquo ActaEcologica Sinica vol 34 no 3 pp 682ndash689 2014 (Chinese)

[39] J Peng Y Xu Y Cai and H Xiao ldquoClimatic and anthropogenicdrivers of land usecover change in fragile karst areas ofsouthwest China since the early 1970s a case study on themaotiaohe watershedrdquo Environmental Earth Sciences vol 64no 8 pp 2107ndash2118 2011

[40] C Rosenzweig and M L Parry ldquoPotential impact of climatechange on world food supplyrdquo Nature vol 367 no 6459 pp133ndash138 1994

[41] Intergovernmental Panel on Climate Change (IPCC) ClimateChange 2014 Mitigation of Climate Change Working GroupIII Contribution to the Fifth Assessment Report of the Inter-governmental Panel on Climate Change IntergovernmentalPanel on Climate Change (IPCC) 2014 httpwwwipccchpdfassessment-reportar5wg3ipcc wg3 ar5 fullpdf

[42] E L Davin and N de Noblet-Ducoudre ldquoClimatic impact ofglobal-scale deforestation radiative versus nonradiative pro-cessesrdquo Journal of Climate vol 23 no 1 pp 97ndash112 2010

[43] S F Zheng B Zhang S Y Zang ZMWang K S Song andDW Liu ldquoImpact of industrial structure on LULC in Daqing cityof Heilongjiang provincerdquo Resources Science vol 29 no 6 pp128ndash132 2007 (Chinese)

[44] K-Y Wu and H Zhang ldquoLand use dynamics built-up landexpansion patterns and driving forces analysis of the fast-growing Hangzhou metropolitan area eastern China (1978ndash2008)rdquo Applied Geography vol 34 pp 137ndash145 2012

[45] K D Frederick and D C Major ldquoClimate change and waterresourcesrdquo Climatic Change vol 37 no 1 pp 7ndash23 1997

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of


Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of


Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of


Mathematical PhysicsAdvances in

Complex AnalysisJournal of


OptimizationJournal of


CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of


Operations ResearchAdvances in

Journal of


Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Algebra

Discrete Dynamics in Nature and Society



Decision SciencesAdvances in

Discrete MathematicsJournal of


Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of


LUCC dynamics of Mediterranean Europe due to its par-ticular climatic and physical conditions [15] Obviously theoutstanding characteristics of the studied cases such as econ-omy culture climate and policy oftenwere considered as theimportant driving forces in the explanation of LUCC dynam-ics However though many researches on land-use changehave been conducted climate factors were seldom availableon driving analysis of LUCC

Various research methods were employed to explore thenexus between LUCC and their driving forces Multivariateregression was used to model how the major forces drive thephysical expansion of urban land cover at the global level[19] Li et al (2013) [20] applied binary logistic regressionto investigate the effects of the selected driving variableson the probability of urban expansion Analytic hierarchyprocess (AHP) as a subjective method provided rigorousquantitative measures to understand the interactive processof urban growth and factors [21] System dynamics (SD) wascombinedwithCLUE-Smodel to reflect the complexity of theland-use system [22] regarded as a macrolevel ldquotop-downrdquoimplementation process However few attempts have beenmade to investigate precise nexus between LUCC and drivingfactors Partial least squares (PLS) method as a major regres-sion technique for multivariate data may handle highly cor-related noise-corrupted data sets by explicitly assuming thedependency between variables and estimating the underlyingstructures [23] It could effectively reflect the significant PLScomponents using the cross validation technique [24] In thepaper PLS was used to accurately reflect the nexus betweenLUCC and driving factors meanwhile determining the sig-nificant components of the selected driving factors

In the paper our main objectives were to address the pro-cesses of land-use dynamics in Wuhan city and its integrateddriving forces through combining satellite-based efforts atmapping land use and land cover (LULC) and physicalsocioeconomic data Based on RS images in 1990 1995 20002005 and 2010 we analyzed the variant change of each LULCtype during 4 periods (1990ndash1995 1995ndash2000 2000ndash2005and 2005ndash2010) Thirteen variables of physical and socioeco-nomic factors were selected as the potential driving factors ofLUCC Partial least squares (PLS) method is applied to selectthe important driving factors and analyze the relationshipsbetween LUCC and the factors triggering each land-usechange type mainly focusing on three major types of LUCCthat is arable land built-up land and water area Finally wesuggested some possible management measures that are cru-cial for future sustainable utilization and management of itsexisting land resources for example managing urban growthand protecting cultivated land

2 Study Area

Wuhan city as a central hinterland megalopolis of Chinais situated in the east of Jianghan plain and covers over849441 km2 (113∘411015840ndash115∘051015840 E 29∘581015840ndash31∘221015840 N) Terrain isdominated by flat areas with a surface elevation ranging from0 to 100m and the slope is less than 10∘ making up 9578of the total The low hills which have an elevation between200 and 400m and a slope varying from 10 to 25∘ constitute

China

Elevation (m)0ndash404001ndash10010001ndash20020001ndash40040001ndash800

0ndash150

0 10 25 50(km)

151ndash500501ndash10001001ndash25002501ndash4500

Slope (∘)

Wuhan city

Hubei

N

Figure 1 Location of the study area

389 of its total area while the mountainous area (ele-vation 400sim800m slope gt25∘) accounts for only 033(Figure 1) The Yangtze River and its largest tributary HanRiver meet here which divides Wuhan into three partsHankou Hanyang and Wuchang commonly known as ldquothethree towns of Wuhanrdquo The area has the subtropical humidmonsoon climate Its climate feature is obvious characterizedby abundant rainfall summer heat and winter cold with anannual temperature of approximately 158ndash175∘C and meanannual precipitation of 1150ndash1450mm A huge water networkwas formed with the Yangtze River as the backbone and sup-plemented by quantities of lakes or pondsTherefore Wuhancity has a reputation of ldquocity of hundred lakesrdquoThewater areaaccounts for 256 of the city area

After entering the twenty-first century Wuhanrsquos eco-nomic growth speed has accelerated In 2010 per capitaGDP in Wuhan amounted to 66520 Yuan RMB [25] whichwas much higher than the average GDP per capita (22936Yuan RMB) in China The urbanization level increased from562 in 1990 to 647 in 2010 Wuhanrsquos population has beenincreasing by an annual rate of 120 since 1978 With atotal population of about 836 million the population densityin Wuhan city was up to 984 personskm2 in 2010 whichwas much higher than the average population density (141personskm2) of China at the same period [25 26] 323 oftotal population was engaged in farming

However with the population growth and economicdevelopment LULC in Wuhan has changed dramaticallyas evidenced by the rapid increase of built-up land at theexpense of occupying massive high quality arable land inthe urban fringe areas Moreover to gain more construction





1 1199102 1199103 1199104 1199105 and 119910

6





2) were included






6) gross



9) gross



12)









119894and119906119894 extracted










Data acquisition




Image enhancement

sites

sele

ctio

nRe

pres

enta

tive t

rain

ing

Clas

s sep

arab

ility

eval

uatio

n

Max

imum

-like

lihoo

dcla

ssifi

catio

n




Visual modification

assessmentAccuracy

Meet requirement

After treatment




LULCmap 2000

LULCLULCmap 1995

LULCmap 1990

Change detection

Fail



ℎ

2 119876ℎ

2 119877cm2 and 119876cm


ℎ

2


ℎ







VIP119895= radic

119873

119877119889(119884 1199051 1199052 119905

119898)

119898

sumℎ=1

119877119889(119884 119905ℎ) 1199082ℎ119895 (1)


119889(119884 1199051 1199052 119905







N


0 10 25 50(km)


N


From built-up land

To built-up land

From unused land

To unused land

From forest

To forest

From water area

To water area


To pasture

0 10 25 50(km)


N

From water area

To water area



From arable land




0 10 25 50(km)


N





0 10 25 50(km)










minus15000

minus10000

minus5000

000

5000

10000

15000N

et ch

ange

of a

rea (

unit

km

2)










2 and 119876ℎ


ℎ


ℎ






2119877cum2

119876ℎ

2119876cum2

Total1 0897 0897 0886 08862 0052 0949 0429 09353 0005 0953 minus0055 0931

Pasture1 0947 0947 0942 09422 0020 0967 0257 09573 0000 0967 minus0110 0953




Water area1 0713 0713 0667 06672 0234 0947 0772 09243 0010 0957 minus0090 0917

Unused land1 0852 0852 0855 08552 0038 0891 0201 08843 0014 0905 0026 0887





4) per




12)



10) and turnover



2) was 0714





t1

minus6

minus4

minus2

0

2

4

u1

(a)


t1

minus3

minus2

minus1

0

1

2

3

t2

(b)


11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1






1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)




1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra













1 1199102 1199103 1199104 1199105 and 119910

6





2) were included






6) gross



9) gross



12)









119894and119906119894 extracted










Data acquisition




Image enhancement

sites

sele

ctio

nRe

pres

enta

tive t

rain

ing

Clas

s sep

arab

ility

eval

uatio

n

Max

imum

-like

lihoo

dcla

ssifi

catio

n




Visual modification

assessmentAccuracy

Meet requirement

After treatment




LULCmap 2000

LULCLULCmap 1995

LULCmap 1990

Change detection

Fail



ℎ

2 119876ℎ

2 119877cm2 and 119876cm


ℎ

2


ℎ







VIP119895= radic

119873

119877119889(119884 1199051 1199052 119905

119898)

119898

sumℎ=1

119877119889(119884 119905ℎ) 1199082ℎ119895 (1)


119889(119884 1199051 1199052 119905







N


0 10 25 50(km)


N


From built-up land

To built-up land

From unused land

To unused land

From forest

To forest

From water area

To water area


To pasture

0 10 25 50(km)


N

From water area

To water area



From arable land




0 10 25 50(km)


N





0 10 25 50(km)










minus15000

minus10000

minus5000

000

5000

10000

15000N

et ch

ange

of a

rea (

unit

km

2)










2 and 119876ℎ


ℎ


ℎ






2119877cum2

119876ℎ

2119876cum2

Total1 0897 0897 0886 08862 0052 0949 0429 09353 0005 0953 minus0055 0931

Pasture1 0947 0947 0942 09422 0020 0967 0257 09573 0000 0967 minus0110 0953




Water area1 0713 0713 0667 06672 0234 0947 0772 09243 0010 0957 minus0090 0917

Unused land1 0852 0852 0855 08552 0038 0891 0201 08843 0014 0905 0026 0887





4) per




12)



10) and turnover



2) was 0714





t1

minus6

minus4

minus2

0

2

4

u1

(a)


t1

minus3

minus2

minus1

0

1

2

3

t2

(b)


11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1






1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)




1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra










Data acquisition




Image enhancement

sites

sele

ctio

nRe

pres

enta

tive t

rain

ing

Clas

s sep

arab

ility

eval

uatio

n

Max

imum

-like

lihoo

dcla

ssifi

catio

n




Visual modification

assessmentAccuracy

Meet requirement

After treatment




LULCmap 2000

LULCLULCmap 1995

LULCmap 1990

Change detection

Fail



ℎ

2 119876ℎ

2 119877cm2 and 119876cm


ℎ

2


ℎ







VIP119895= radic

119873

119877119889(119884 1199051 1199052 119905

119898)

119898

sumℎ=1

119877119889(119884 119905ℎ) 1199082ℎ119895 (1)


119889(119884 1199051 1199052 119905







N


0 10 25 50(km)


N


From built-up land

To built-up land

From unused land

To unused land

From forest

To forest

From water area

To water area


To pasture

0 10 25 50(km)


N

From water area

To water area



From arable land




0 10 25 50(km)


N





0 10 25 50(km)










minus15000

minus10000

minus5000

000

5000

10000

15000N

et ch

ange

of a

rea (

unit

km

2)










2 and 119876ℎ


ℎ


ℎ






2119877cum2

119876ℎ

2119876cum2

Total1 0897 0897 0886 08862 0052 0949 0429 09353 0005 0953 minus0055 0931

Pasture1 0947 0947 0942 09422 0020 0967 0257 09573 0000 0967 minus0110 0953




Water area1 0713 0713 0667 06672 0234 0947 0772 09243 0010 0957 minus0090 0917

Unused land1 0852 0852 0855 08552 0038 0891 0201 08843 0014 0905 0026 0887





4) per




12)



10) and turnover



2) was 0714





t1

minus6

minus4

minus2

0

2

4

u1

(a)


t1

minus3

minus2

minus1

0

1

2

3

t2

(b)


11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1






1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)




1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra










N


0 10 25 50(km)


N


From built-up land

To built-up land

From unused land

To unused land

From forest

To forest

From water area

To water area


To pasture

0 10 25 50(km)


N

From water area

To water area



From arable land




0 10 25 50(km)


N





0 10 25 50(km)










minus15000

minus10000

minus5000

000

5000

10000

15000N

et ch

ange

of a

rea (

unit

km

2)










2 and 119876ℎ


ℎ


ℎ






2119877cum2

119876ℎ

2119876cum2

Total1 0897 0897 0886 08862 0052 0949 0429 09353 0005 0953 minus0055 0931

Pasture1 0947 0947 0942 09422 0020 0967 0257 09573 0000 0967 minus0110 0953




Water area1 0713 0713 0667 06672 0234 0947 0772 09243 0010 0957 minus0090 0917

Unused land1 0852 0852 0855 08552 0038 0891 0201 08843 0014 0905 0026 0887





4) per




12)



10) and turnover



2) was 0714





t1

minus6

minus4

minus2

0

2

4

u1

(a)


t1

minus3

minus2

minus1

0

1

2

3

t2

(b)


11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1






1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)




1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra











minus15000

minus10000

minus5000

000

5000

10000

15000N

et ch

ange

of a

rea (

unit

km

2)










2 and 119876ℎ


ℎ


ℎ






2119877cum2

119876ℎ

2119876cum2

Total1 0897 0897 0886 08862 0052 0949 0429 09353 0005 0953 minus0055 0931

Pasture1 0947 0947 0942 09422 0020 0967 0257 09573 0000 0967 minus0110 0953




Water area1 0713 0713 0667 06672 0234 0947 0772 09243 0010 0957 minus0090 0917

Unused land1 0852 0852 0855 08552 0038 0891 0201 08843 0014 0905 0026 0887





4) per




12)



10) and turnover



2) was 0714





t1

minus6

minus4

minus2

0

2

4

u1

(a)


t1

minus3

minus2

minus1

0

1

2

3

t2

(b)


11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1






1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)




1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra











t1

minus6

minus4

minus2

0

2

4

u1

(a)


t1

minus3

minus2

minus1

0

1

2

3

t2

(b)


11199052oval graph

VIP

00

02

04

06

08

10

12

VIP

val

ues

x8 x3 x4 x11 x5 x9 x12 x6 x7 x10 x13 x2 x1






1minus 0106119909

2minus 0117119909

3minus 0108119909

4

minus 00871199095minus 0141119909

6minus 0066119909

7minus 0117119909

8

minus 00861199099minus 0059119909

10minus 0088119909

11minus 0075119909

12

minus 004611990913(1198772= 0988)

(2)




1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra











1000

1200

1400

1600

1800A

nnua

l pre

cipi

tatio

n

1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

(a)


1992 1994 1996 1998 2000 2002 2004 2006 2008 20101990Year

165

170

175

180

185

Ann

ual t

empe

ratu

re

(b)


Arable land

minus015

minus010

minus005

000

005

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199102= minus0027119909

1+ 0068119909

2+ 0104119909

3+ 0101119909

4

+ 00931199095+ 0101119909

6+ 0081119909

7+ 0104119909

8

+ 00921199099+ 0078119909

10+ 0094119909

11+ 0087119909

12

+ 007211990913(1198772= 0992)

(3)



Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra










Built-up land

minus005

000

005

010

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13




1199103= minus0040119909

1+ 0271119909

2+ 0161119909

3+ 0127119909

4

+ 00501199095+ 0304119909

6minus 0013119909

7+ 0161119909

8

+ 00471199099minus 0038119909

10+ 0053119909

11+ 0009119909

12

minus 007611990913(1198772= 0947)

(4)


Water area

minus01

00

01

02

03

Coe

ffici

ents

x2 x3 x4 x5 x6 x7 x8 x9 x10 x11x1 x12 x13



5 Conclusions





Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra











Competing Interests


Acknowledgments


References







































2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra






















2emissions evidence

















Volume 2014




Journal of











Journal of


Function Spaces






Algebra
















Volume 2014




Journal of











Journal of


Function Spaces






Algebra









Documents

Research Article Physical and Socioeconomic Driving Forces of …downloads.hindawi.com/journals/ddns/2016/8061069.pdf · Research Article Physical and Socioeconomic Driving Forces