Assessing Future Vegetation Trends and Restoration Prospects in the Karst Regions of Southwest China

Citation
Assessing Future Vegetation Trends and Restoration Prospects in the Karst Regions of Southwest China

Material Information

Title:
Assessing Future Vegetation Trends and Restoration Prospects in the Karst Regions of Southwest China
Series Title:
Remote Sensing
Creator:
Tong, Xiaowei
Wang, Kelin
Brandt, Martin
Yue, Yuemin
Liao, Chujie
Fensholt, Rasmus
Publisher:
MDPI
Publication Date:
Language:
English
Physical Description:
1 online resource

Subjects

Subjects / Keywords:
Karst ( lcsh )
Restoration ecology ( lcsh )
Genre:
serial ( sobekcm )
Location:
Texas

Notes

Abstract:
To alleviate the severe rocky desertification and improve the ecological conditions in Southwest China, the national and local Chinese governments have implemented a series of Ecological Restoration Projects since the late 1990s. In this context, remote sensing can be a valuable tool for conservation management by monitoring vegetation dynamics, projecting the persistence of vegetation trends and identifying areas of interest for upcoming restoration measures. In this study, we use MODIS satellite time series (2001–2013) and the Hurst exponent to classify the study area (Guizhou and Guangxi Provinces) according to the persistence of future vegetation trends (positive, anti-persistent positive, negative, anti-persistent negative, stable or uncertain). The persistence of trends is interrelated with terrain conditions (elevation and slope angle) and results in an index providing information on the restoration prospects and associated uncertainty of different terrain classes found in the study area. The results show that 69% of the observed trends are persistent beyond 2013, with 57% being stable, 10% positive, 5% anti-persistent positive, 3% negative, 1% anti-persistent negative and 24% uncertain. Most negative development is found in areas of high anthropogenic influence (low elevation and slope), as compared to areas of rough terrain. We further show that the uncertainty increases with the elevation and slope angle, and areas characterized by both high elevation and slope angle need special attention to prevent degradation. Whereas areas with a low elevation and slope angle appear to be less susceptible and relevant for restoration efforts (also having a high uncertainty), we identify large areas of medium elevation and slope where positive future trends are likely to happen if adequate measures are utilized. The proposed framework of this analysis has been proven to work well for assessing restoration prospects in the study area, and due to the generic design, the method is expected to be applicable for other areas of complex landscapes in the world to explore future trends of vegetation.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-05052 ( USFLDC DOI )
k26.5052 ( USFLDC handle )

USFLDC Membership

Aggregations:
Added automatically
Karst Information Portal

Postcard Information

Format:
serial

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

Article AssessingFutureVegetationTrendsandRestoration ProspectsintheKarstRegionsofSouthwestChina XiaoweiTong 1,2,3 ,KelinWang 1,2, *,MartinBrandt 4 ,YueminYue 1,2 ,ChujieLiao 1,2,3 and RasmusFensholt 4 1 KeyLaboratoryforAgro-ecologicalProcessesinSubtropicalRegion,InstituteofSubtropicalAgriculture, ChineseAcademyofSciences,Changsha410125,China;tongxiaowei1996@163.comX.T.; ynwessnso@163.comY.Y.;hnyym829@163.comC.L. 2HuanjiangObservationandResearchStationforKarstEcosystem,ChineseAcademyofSciences,Huanjiang,Hechi547100,China 3 UniversityofChineseAcademyofSciences,Beijng100049,China 4 DepartmentofGeosciencesandNaturalResourceManagement,UniversityofCopenhagen, Copenhagen1350,Denmark;martin.brandt@mailbox.orgM.B.;rf@ign.ku.dkR.F. Correspondence:kelin@isa.ac.cn;Tel.:+86-731-8461-5201 AcademicEditors:ParthSarathiRoyandPrasadS.Thenkabail Received:14March2016;Accepted:17April2016;Published:27April2016 Abstract:ToalleviatethesevererockydeserticationandimprovetheecologicalconditionsinSouthwestChina,thenationalandlocalChinesegovernmentshaveimplementedaseriesofEcologicalRestorationProjectssincethelate1990s.Inthiscontext,remotesensingcanbeavaluabletoolforconservationmanagementbymonitoringvegetationdynamics,projectingthepersistenceofvegetationtrendsandidentifyingareasofinterestforupcomingrestorationmeasures.Inthisstudy,weuseMODISsatellitetimeseriesandtheHurstexponenttoclassifythestudyareaGuizhouandGuangxiProvincesaccordingtothepersistenceoffuturevegetationtrendspositive,anti-persistentpositive,negative,anti-persistentnegative,stableoruncertain.Thepersistenceoftrendsisinterrelatedwithterrainconditionselevationandslopeangleandresultsinanindexprovidinginformationontherestorationprospectsandassociateduncertaintyofdifferentterrainclassesfoundinthestudyarea.Theresultsshowthat69%oftheobservedtrendsarepersistentbeyond2013,with57%beingstable,10%positive,5%anti-persistentpositive,3%negative,1%anti-persistentnegativeand24%uncertain.Mostnegativedevelopmentisfoundinareasofhighanthropogenicinuencelowelevationandslope,ascomparedtoareasofroughterrain.Wefurthershowthattheuncertaintyincreaseswiththeelevationandslopeangle,andareascharacterizedbybothhighelevationandslopeangleneedspecialattentiontopreventdegradation.Whereasareaswithalowelevationandslopeangleappeartobelesssusceptibleandrelevantforrestorationeffortsalsohavingahighuncertainty,weidentifylargeareasofmediumelevationandslopewherepositivefuturetrendsarelikelytohappenifadequatemeasuresareutilized.Theproposedframeworkofthisanalysishasbeenproventoworkwellforassessingrestorationprospectsinthestudyarea,andduetothegenericdesign,themethodisexpectedtobeapplicableforotherareasofcomplexlandscapesintheworldtoexplorefuturetrendsofvegetation. Keywords:growingseasonNDVIGSN;persistenttrends;futurevegetationtrends;Hurstexponent;terrainnicheindexTNI 1.IntroductionVegetationhasconsiderableimpactsonalmostalllandsurfaceenergyexchangeprocessesactingasaninterfacebetweenlandandatmosphere.Itaffectslocalandregionalclimate[1,2]andhydrologicbalanceofthelandsurface[3].ThedynamicsofvegetationhavebeenrecognizedasbeingofprimaryRemoteSens. 2016 8 ,357;doi:10.3390/rs8050357www.mdpi.com/journal/remotesensing

PAGE 2

RemoteSens. 2016 8 ,357 2of17importanceinglobalchangeofterrestrialecosystems[4,5].KarstregionsinSouthwestChinaaresomeofthelargestexposedcarbonaterockareasintheworld,coveringabout540,000km2,withinwhichmorethan30,000,000peopleunderpoverty[6].Underintensiveanthropogenicdisturbances,thesefragilekarstregionshavebeenreportedtoundergosevererockydesertication.Thistypeoflanddegradationisaprocesswhereakarstareacoveredbyvegetationandsoiltransformsintoarockybarrenlandscape,whichisconsideredtobeoneofthemostdangerousecologicalandenvironmentalproblemsinChina[79].Vegetationplaysanimportantroleinecologicalconservationandrestoration[10].ToalleviatethesevererockydeserticationinSouthwestChina,nationalandlocalChinesegovernmentshaveimplementedaseriesofEcologicalRestorationProjectsERPstoimprovevegetationandecosystemconditionssincethelate1990s.ExamplesaretheNaturalForestProtectionProject,theGraintoGreenProgram,thePublicWelfareForestProtectionandtheKarstRockyDeserticationComprehensiveControlandRestorationProject.Arapidandarea-widemappingofthedynamicsandspatialpatternsofvegetationchangesaswellasthepredictionoffuturevegetationtrendsarethusessentialforfurtherecologicalmeasuresandsustainabledevelopment.Remotesensingdatahavebeenincreasinglyusedforecologicalstudies[11].TheNormalizedDifferenceVegetationIndexNDVI,aproxyforphotosyntheticallyactivevegetation,isstronglycorrelatedwithvariableslikeLeafAreaIndex,abovegroundbiomass,andpercentagevegetationcover.NDVIisalsowidelyusedforphenologicalchangestudies[1215].Thus,trendsanddynamicsinNDVIcanbeusedtomonitorchangesinvegetationcover,productivity,andhealthstatusatbothlargespatialandlongtemporalscales[1619].ModerateResolutionImagingSpectroradiometerMODISdatahavebeenconsideredstate-of-the-artsincethelaunchoftheTerraplatformin1999anddensetemporalresolutionNDVIproductsarewidelyusedforvegetationdynamicresearch[20,21].SeveralstudieshavebeenconductedinSouthwestChinausingNDVItimeseries[2226].However,fewstudiesanalyzedthepersistenceofvegetationtrends,whichcanprovideinformationaboutthelong-termmemoryofvegetationtrendsbeyondtheperiodofanalysisanddataavailability.Thelong-termmemoryeffectscanbecapturedbytheauto-covariancefunctionthatdecaysexponentially,withaspectraldensitythattendstoinnity[27,28].TheHurstexponent,basedonauto-covariance,estimatedviaRescaledRangeAnalysisR/Sanalysis,isaparameterwhichcanbeappliedtoquantitativelydetectthepersistenceoftimeseriesdataofnaturalphenomena.TheHurstexponenthasbeenwidelyusedintheeldsofhydrology[29],climatology[30],geology[31]andeconomics[32].Onlyrecentlyhasitbeenappliedinthetimeseriesdetectionofvegetationvariations[3335],thoughnotyetinacomplexterrainregionlikethekarstregionsofSouthwestChina.Vegetationdynamicscouldbecontrolledbyenvironmentalvariables,suchaswateravailability,temperature,andincidentradiation[18,36].Topographicheterogeneityimposesenvironmentalconstraintsonvegetationdynamicsbyproducingcomplexsubstrateswithvariablestructure,hydrology,andchemistry[37].Inparticular,elevationdeterminesthealtitudinalzonalityofsoil[38,39]andslopecontrolsthevelocityofsurfaceows[40].Theaspectaffectsthedirectionofows,insolationandtheintensityofevaporation[41].Thesurfacecurvatureinuenceswatermigrationandaccumulationinlandscapesbygravity[42].Acombinationofthesetopographicfactorselevation,slope,aspectandsurfacecurvaturedeterminestheconditionsforthegrowthandthedistributionofvegetation[43,44].Totheauthors'bestknowledge,previousresearchdidnotincludethecombinedeffectscausedbytopographicvariabilityinvegetationdynamicstudies,andalsothespatialpatternoffuturevegetationdynamicsalongatopographicgradienthasnotyetbeenexplored.InthecontextofmonitoringecologicalconservationprojectsinSouthwestChina,thispaperaimstoprovideinformationonthepersistenceofvegetationtrendsinkarstterrainsofvaryingcomplexity.Thespecicobjectivesofthisstudyare,toanalyzethespatio-temporalvegetationtrendsduring2001andclassifythestudyareaaccordingtopredictedfuturetrends;andidentifyterraintypessuitedforrestorationbutalsoquantifytheuncertaintyoffuturetrendsunderdifferentterrainconditions.Thendingsofthisstudyarerelevantfordecision-makingprocessesinvegetation

PAGE 3

RemoteSens. 2016 8 ,357 3of17restorationprogramsandthegenericdesignoftheframeworkallowsforapplicationsinotherregionsofcomplexlandscapesaroundtheworld. 2.MaterialsandMethodsMODISNDVIdataareutilizedtogenerategrowingseasonNDVIGSNforeachyearfrom2001to2013toidentifyregionswiththebestprospects/highuncertaintyforfuturevegetationrestorationinSouthwestChinaFigure1.LineartrendanalysisandHurstexponentareusedtoproducemapsofvegetationtrendsandtheirpersistence.Basedonthis,afuturevegetationtrendisgenerated.ElevationandslopeareextractedfromadigitalelevationmodelDEMandaterrainnicheindexTNImapiscalculated,whichisclassiedinto7separateclasses.Twodistributionindexes,namelyFutureRecoveryDistributionIndexFRDIandFutureUncertaintyDistributionIndexFUDI,areproposedtoexplorethedistributionoffuturevegetationdynamicsandtoidentifyareaswherevegetationhasthebestprospectsforrestorationorwherefuturetrendsareuncertain. Figure1. Conceptualapproachofthisstudy. 2.1.StudyAreaThestudyareaistheGuizhouandGuangxiprovincesinSouthwestChina541131Nand103361041E.Theareahasatotalcoverageof4.13105km2andapopulationof87.84millionpeople.Theelevationrangesfrom8to2855m.a.s.l.,withmaximumheightsinthenorthwestandtheminimuminthesoutheastFigure2b.Theclimateisasubtropicalandtropicalmonsoonclimate,with1465mmfallingduringthegrowingseasonfromApriltoNovember.TheprecipitationfromApriltoSeptemberaccountsfor70%%oftheannualtotalprecipitation[45]andthemeangrowingseasontemperatureis25C.ATotalof45%and15%ofthebedrocksarepurecarbonateandimpurecarbonatekarstregion,respectively.Thefragilesoilsofthekarstareasarepronetorockydegradationifover-utilizede.g.,bylivestockfarming,Figure2a.Thebedrockfortherestoftheregionconsistsofclasticrocksnon-karstregion[46].Governmentalrockydeserticationcontrolprojectsimplementedinkarstregionscanbedividedintosixmainprojectregions[47]showninFigure2b.Themainvegetationcovertypesarecultivatedvegetation%,scrub%,grassland%,needle-leafforest%,andbroad-leafforest%Figure2c.Althoughtheseasonalcycleofthemainvegetationtypesdiffers,thegeneralpatternisuniform,reectingagrowingseasonfromApriltoNovemberFigure2d.

PAGE 4

RemoteSens. 2016 8 ,357 4of17 Figure2.RockykarstlandscapeinSouthwestChinausedforlivestockgrazingphototakeninAugust2014,Tong.aLocationofthestudyareaandthedistributionofprojectregionsandcountyseatsoverlayingadigitalelevationmodelDEM;Ipeakclusterdepression;IIpeakforestplain;IIIkarstplateau;IVkarstgorge;Vkarsttroughvalley;VIkarstbasinandVIInon-projectregionsrespectivelyb;Thespatialdistributionofvegetationcovertypesisshowninc;TheseasonalvariationsofthemainvegetationtypesbasedonthemeanmonthlyNDVIfrom2001to2013 d 2.2.DataandProcessingNDVIdatausedinthisstudyareMODISMOD13Q1from2001to2013withatemporalresolutionof16days.Theimageshaveaspatialresolutionof250mandareretrievedfromdaily,atmosphericallycorrectedsurfacereectance.Theyhavebeencompositedusingmethodsbasedonproductqualityassurancemetricstoremovelowqualitypixelsandthehighestqualitypixelswerechosenforeach16daycomposite[48].Forthisstudy,weusedthemaximumvaluecompositeMVCmethodtoselectthehighervalueoftwo16-daycompositeimagesandobtainthemonthlyNDVI.Thisfurtherminimizesatmosphericandscan/angleeffects,cloudcontaminationandhighsolarzenithangleeffects[49].ThentheNDVIvaluesfromApriltoNovemberwereaveragedtoobtaintheGSNforeachyearfrom2001to2013seeFigure2d.

PAGE 5

RemoteSens. 2016 8 ,357 5of17ASRTMDEMwith90mresolutionwasdownloadedfromtheInternationalScienticDataServicePlatform[50]andusedasthesourceforelevationandslope.TheDEMwasresampledto250mspatialresolutionusinganearestneighborresamplingalgorithmtomatchtheMODISNDVIdata.Monthlytemperatureandrainfalldatafor53weatherstationsfrom2001to2013wereobtainedfromtheChinaMeteorologicalDataSharingServiceSystem[51].Thehumanpopulationdatain2005and2010wereprovidedbytheDataCenterforResourcesandEnvironmentalSciences,ChineseAcademyofSciencesRESDC[52]. 2.3.MethodsAlinearregressionbetweenannualGSNandtimewasappliedtoderivetheslopeasanindicationofthedirectionandmagnitudeoflineartrendsinthetimeseries.Whereasapositiveslopevalueindicatesapositivetrendinvegetatione.g.,successfulconservationprojects,anegativeslopevaluereferstonegativevegetationtrendswhichcanbeanindicatorofdegradatione.g.,erosionduetoanthropogenicoveruse.Thetrendsareconsideredstatisticallysignicantifthepvalueoftheregressionissmallerthan0.05.Ifthepvalueislarger0.05,nosignicantchangeisdetectedandthepixelisconsideredtobestable.WethuscategorizeGSNtrendsasthreetypes:increasingslope>0and p <0.05,decreasingslope<0and p <0.05andstable p 0.05.TherescaledrangeR/SanalysisdevelopedbyHurstisamethodusedtoestimateauto-correlationpropertiesoftimeseries[53].WeapplytheHurstexponentH,estimatedbytheR/Sanalysis,asameasureofthelong-termmemoryinourGSNtimeseriesannualdata.Themaincalculationstepsare[54]: 1.Todividethetimeseriest x p t qut=1,2...,nintotsubseriesx p t q,andforeachsubseriest =1,..., t 2.Todenethemeansequenceofthetimeseries, x t 1 t t t 1 x p t q t 1,2..., n 3.Tocalculatethecumulativedeviation, x p t t q t u 1 p x p u q x t q ,1 t t 4.Tocreatetherangesequence, R p t q max 1 t t X p t t q min 1 t t X p t t q t 1,2..., n 5.Tocreatethestandarddeviationsequence, S p t q c 1 t t t 1 p x p t q x t q 2 t 1,2..., n 6.Torescaletherange, R p t q S p t q p C t q HThevaluesoftheHrangebetween0and1.Avalueof0.5indicatesatruerandomwalk.Inarandomwalkthereisnocorrelationbetweenanyelementandafutureelement.AnHvaluebetween0.5and1H,0.5
PAGE 6

RemoteSens. 2016 8 ,357 6of17BasedontheGSNtrendsandtheHvalue,wecategorizedthesupposedfuturevegetationtrendsintosixtypesTable1.WhenapixelshowedapositivetrendanditsHvaluewashigherthan0.5i.e.,thepositivetrendislikelytocontinue,thevegetationinthispixelissupposedtohaveanoverallpositivedevelopmentPDafterthetimeseriesendsin2013.IfthetrendispositivebutitsHvaluewaslessthan0.5thepositivetrendisnotlikelytocontinue,thevegetationinthispixelhasananti-persistentpositivedevelopmentAPDafter2013.WhenapixelshowedanegativetrendanditsHvaluewashigherthan0.5thenegativetrendislikelytocontinue,thevegetationinthispixelissupposedtoexperienceanegativedevelopmentNDafter2013.IfthetrendisnegativeanditsHvaluewaslessthan0.5,i.e.,thenegativetrendisnotlikelytocontinue,thevegetationinthispixelischaracterizedbyananti-persistentnegativedevelopmentANDafter2013.WhenapixelwasdetectedtohavenosignicanttrendanditsHvaluewashigherthan0.5,thevegetationinthispixelissupposedtohaveastableandsteadydevelopmentSSDafter2013.IfitsHvaluewaslessthan0.5,thefuturevegetationtrendinthispixelisuncertain,whichmeansthetrendafter2013isanundetermineddevelopmentUD. Table1. Futurevegetationtrendsbasedonvegetationdynamicsandthe H value. Trends Hurst Persistent.5< H <1Anti-Persistent< H <0.5 IncreasingPositiveDevelopmentPDAnti-persistentPositiveDevelopmentAPD StableSustainedandSteadyDevelopmentSSDUndeterminedDevelopmentUD DecreasingNegativeDevelopmentNDAnti-persistentNegativeDevelopmentAND TheterrainnicheindexTNIwasutilizedtocharacterizethetopographicvariationinSouthwestChina[56].Itwascalculatedas: TNI lg e E )]TJ/F174 9.9626 Tf 9.684 0 Td [(1 s S )]TJ/F174 9.9626 Tf 9.684 0 Td [(1 whereeandEaretheelevationofthepixelandtheaverageelevationofthestudyarearespectively,whereassandSsignifytheslopeofthepixelandtheaverageslopeofthestudyarea.LargeTNIvaluescorrespondtohigherelevationandlargerslopeangles,beingtypicalforkarstplateausandgorges.Incontrast,smallerTNIvaluesindicatelowerelevationandsmallerslopeangles.MediumTNIvaluesarefoundinareasofahigherelevationbutsmallslopeangle,orlowelevationbutwithlargerslopeangles,ormoderateelevationandslopeangle.InthisstudytheTNIvalueswerecategorizedinto7classeswithanintervalof0.2.Tocharacterizethestatusofthevegetationforeachterrainintervalbeyond2013,thefuturevegetationtrendclassesshowninTable1wereusedasreference.Allpixelswithineachofthe7terrainintervalswereclassiedaccordingtooneofthefuturetrendclassesandtheoverallsumoftheclasseswascalculatedwhereNDhas1,othershavethevalue1foreachterraininterval.Toaccountfordifferentareasizesoftheterrainintervalsandcapturesupposedfuturechangeforeachspecicinterval,weproposetwodistributionindexes,FRDIidentifyingareaswherevegetationhasthebestprospectsforrestorationandFUDIshowingthedistributionoftheuncertainty.Thesetwocomprehensiveindexesarecalculatedasfollows: FRDI i p e i 1 )]TJ/F168 9.9626 Tf 9.808 0 Td [(e i 2 )]TJ/F168 9.9626 Tf 9.808 0 Td [(e i 3 a i q{p E 1 )]TJ/F168 9.9626 Tf 9.958 0 Td [(E 2 )]TJ/F168 9.9626 Tf 9.958 0 Td [(E 3 A q FUDI i p e i 4 )]TJ/F168 9.9626 Tf 9.808 0 Td [(e i 5 )]TJ/F168 9.9626 Tf 9.809 0 Td [(e i 6 a i q{p E 4 )]TJ/F168 9.9626 Tf 9.958 0 Td [(E 5 )]TJ/F168 9.9626 Tf 9.958 0 Td [(E 6 A qwhereireferstothe7terrainintervals;e ijandE jj=1PD,2SSD,3ND,4UT,5APDand6ANDaretheareaofthesixfuturevegetationtrendsTable1ineachterrainintervaliandthestudyarea;anda iandArefertotheareaoftheterrainintervaliandthestudyarea,respectively.

PAGE 7

RemoteSens. 2016 8 ,357 7of17 FRDI iandFUDI iarestandardizedanddimensionlessindexes.IfFRDI i>1,thecorrespondingterrainintervaliissupposedtobemorefavorableforvegetationrestorationinthefuturethanotherterrainintervals.ThelargertheFRDI iis,thebetterthefutureprospectchanges.IftheFRDI iislowerthan1,theterrainintervaliissupposedtohavelessfavorableprospects.IfFUDI i>1,trendpredictionsintheterrainintervalihaveahigheruncertaintythanotherterrainintervals.AlowerFUDI ireducespredictionuncertaintyandaFUDI i below1impliesmorecertainprospects. 3.Results 3.1.VegetationPatternsandDynamicsduring2001ThegrowingseasonvegetationinthestudyareahasanaverageGSNof0.72fortheperiod2001withahighspatialheterogeneitybetweentheland-coverclassesFigure3a.FortheGuangxiProvince,theaverageGSNis0.74,whichisslightlyhigherthanintheGuizhouProvince.69.Thevegetationtypeoftheneedle-leafandbroad-leafmixedforestclasshasthehighestmeanGSN.78,followedbybroad-leafforest.77,needle-leafforest.74,grassland.73,scrub.72,andcultivatedvegetation.68.TheaverageGSNinthenon-projectregion.74isslightlyhigherthanintheprojectregions.71.TheregionswithaGSNhigherthan0.8aremainlylocatedinGuangxiProvince,suchasNinming,Baise,Tianlin,Zhaoping,Cangwu,Tiane,YongfuCounty,dominatedbybroad-leafforestandneedle-leafforest.TheareaswithaGSNlessthan0.5areassociatedwithscrubs,cultivatedvegetationandbuilt-upareas.Theseregionsareclosetoroadsandrailwaysandcorrespondtocountyseats,especiallyinlargecities,suchasNanning,LiuzhouandGuiyang.Atthecountyscale,wefoundtheaverageGSNofthepast13yearshadasignicantpositiverelationshipwiththedistancetoroadsp<0.01andasignicantnegativerelationshipwiththepopulationdensitychangefrom2005to2010p=0.016.Moreover,areasoflowGSNarefoundinmountainousregionsdominatedbyethnicminorities,suchasWeiningCountypopulatedbytheYinationalityandtheFuchuanCountyinhabitedbytheYaonationality.Atthecountyscale,wefoundsignicantpositivecorrelationsp<0.01betweenGSNandgrowingseasonclimaticfactorstemperatureandprecipitation.GSNwasmorestronglyandpositivelycorrelatedwithtemperature R 2 =0.25thanwithprecipitation R 2 =0.11during2001. Figure3.ThevegetationpatternsaanddynamicsbinSouthwestChinabasedonMODISNDVIfrom2001to2013.Thenumbersinbrepresentcounties:Weining;Dafang;Qianxi;Xingren;Tiane;Tianlin;Baise;Debao;Jingxi;Liucheng;Liuzhou;Yongning;Lingshan;Pubei;Qinzhou;Hepu.Figure3bshowsthetrendsinannualGSNwithdistinctspatialdifferences.Anuptrendslope>0isfoundin67%ofthestudyarea,with23%ofthesetrendsbeingsignicantp<0.05.Thesepixels

PAGE 8

RemoteSens. 2016 8 ,357 8of17aremainlyinQinzhouandBeihaiofGuangxiandBijieofGuizhou,wherethedominantvegetationtypesarescrubsandcultivatedvegetationi.e.,agro-forestry,suchaseucalyptusplantationandoil-teacamelliaplantation.Furthermore,33%ofthepixelsaredetectedtohaveadowntrendslope<0,with10%beingsignicant.ThesepixelsaremainlylocatedinBaiseandLiuzhou,Guangxi.Inaddition,thedominatingvegetationtypesoftheseareasarecultivatedvegetation,scrubsandgrasslands.Consequently,threevegetationtrendtypes,increasing,decreasingandstablenotsignicant;p0.05accountedfor15%,3%and82%,respectively. 3.2.FutureVegetationTrendsafter2013ValuesoftheHrangefrom0.02to0.86Figure4a.ThemeanHvalueofthewholestudyareais0.55,indicatinganoverallmoderatelypersistentvegetationtrendinthetwoprovincesinSouthwestChina,suggestingthatfuturevegetationtrendsarelikelytobesimilartothoseofthepast13years.Inparticular,theHvaluesbetweenmostofthevegetationcovertypesarerelativelysimilar:broad-leafforesthasameanHvalue.552,followedbyneedle-leafforest.551,grassland.548,scrub.547,andcultivatedvegetation.546.Onlytheclassneedle-leafandbroad-leafmixedforestisbelow0.5.473.In69%ofthestudyareatheHvalueisabove0.5.Areaswithalowpersistencewerelocatedinthewestandthesouthofthestudyarea.Inthewest,theregionswithlowHvaluesweremostlycorrespondingtotheroadnetwork,whileinthesouth,theregionswithlowpersistenceareclosetocountyseatsFigure4a.ThevegetationdynamicsandtheHvalueswerecombinedintodifferentclassesTable1toindicatefuturevegetationtrendsFigure4b.Thefuturevegetationtrendsinmostpartsofthestudyarea%arestableandsteady,andparticularlyvegetationbelongingtoscrubandcultivatedvegetationFigure1cisdetectedtohavenosignicanttrendinthepast13yearsandthisstatewillbepersistentH>0.5.However,10%ofthestudyareaisexpectedtohaveapositivedevelopmentinthefuturevegetationhadapersistentpositivetrend.TheseareasconsistprimarilyofscrubsandaremainlydistributedinthenorthandnorthwestGuizhouandinsouthernGuangxi.Areaswithanegativedevelopmentwherevegetationwasdetectedtohaveapersistentdecreasingtrend,cover3%ofthestudyareaandthemainlandcoveroftheseareasiscultivatedvegetation.Vegetationshowingnotrendandnopersistenceoccupy24%ofthestudyarea,wherethemainlandcovertypesarescrubsandcultivatedvegetation.Regionswithananti-persistentnegativedevelopmentcover1%ofthestudyarea.Themainlandcoveroftheseregionsiscultivatedvegetation.Areaswithananti-persistentpositivedevelopmentcover5%ofthestudyarea,withscrubbeingthemainlandcoverclass. Figure4.MapsshowingtheHurstexponenta;andthedistributionoffuturevegetationtrends b basedonobservedvegetationtrendsinSouthwestChina.

PAGE 9

RemoteSens. 2016 8 ,357 9of17 3.3.VegetationTrendsandFutureVegetationTrendsinDifferentProjectRegionsSomedifferencesinvegetationtrendsandfuturevegetationtrendscanbeobservedFigure5indifferentregionsFigure2b.ApartfromthekarstgorgeregionIVandnon-projectregionVII,vegetationin80%ofotherprojectregionsisratherstable.Vegetationin32%ofthekarstgorgeregionIVhavingthelowestaverageGSNwith0.62showsasignicantincreasingtrend,indicatinganimprovementofthevegetationinthisregion.Overall,onlyasmallpercentageofthepixelsintheprojectregionshaveadecreasingtrend.With5%inthepeakforestplainregionII,4%inthepeakclusterdepressionregionIand3%inthenon-projectregionVII,theseregionsaretheonlyoneswithanoticeableshareofpixelscharacterizedbysignicantdecreasingtrends. Figure5.Thearearatiopercentagecoverofdifferentvegetationtrendsto2013leftbarandfuturevegetationtrendsrightbarindifferentprojectregionsseeFigure2b.Thefuturevegetationtrendswithastableandsteadydevelopmentafter2013cover58%and54%ofallprojectregionandnon-projectregionpixels,respectively.Inparticular,thekarsttroughvalleyregionVandthepeakforestplainregionIIaccountfor66%and63%respectively,havingthehighestshareofsteadytrendsamongtheprojectregions.Overall,19%ofthekarstgorgeregionIVand11%ofthekarstplateauregionIIIarepredictedtohaveapositivedevelopmentafter2013.However,37%oftheoverallpositivetrendsareanti-persistent.Inparticular,thekarstgorgeregionIVneedsattention,with13%ofthevegetationshowingapositivebutanti-persistentdevelopment.Atotalof25%ofthepixelsintheprojectregionshaveanundeterminedtrend.Morespecically,in34%ofthekarstbasinregionVIand28%ofthekarstgorgeregionIV,thefuturevegetationtrendscannotbedetermined.ThehighestshareofnegativedevelopmentwasfoundinthepeakforestplainregionII%.Ingeneral,theshareofanti-persistentnegativedevelopmentissmallforallprojectregions.5%,indicatingsomeeffortsinpreventingdegradation. 3.4.DistributionofFutureVegetationTrendsforDifferentTerrainConditionsTheTNIvaluesinthestudyarearangingfrom0.0006to1.4showapatternofdecreasefromthenorthwesttothesoutheastFigure6a,whichingeneralisconsistentwiththedistributionofelevationandslope.ThekarstgorgeregionIVhasthehighestTNI.85,followedbythekarstbasinregionVI.84,thekarstplateauregionIII.70,thekarsttroughvalleyregionV.67,thepeakclusterdepressionregionI.54,andthepeakforestplainregionII.40.ThedistributionoffuturevegetationtrendsvariesbetweenTNIgroupsi.e.,withtheterrainFigure6b.Thelargestsharesofnegativeandpositivedevelopmentarefoundinregionswithlowelevationandslope,whicharealsotheareascharacterizedbythehighestanthropogenicinuence.ThenegativedevelopmentdecreasesandtheundetermineddevelopmentincreasesastheTNIincreases.Thisindicatesadeclineofanthropogenicimpactinroughterrainandamoredifcultpredictionoffuturetrends.

PAGE 10

RemoteSens. 2016 8 ,357 10of17ThedistributionoftheFRDIandFUDIalongtheseventerrainintervalsFigure6cshowsthatFRDIrepresentingtheareaswithbestfutureprospectsislowinareaswithalowTNIbutislargerthan1whentheTNIisintherangebetween0.4and1areaswithamoderateslopeandelevation.TheuncertaintyforfuturepredictionsFUDIishighinareasoflowTNI,butlowinareaswithaTNIbetween0.4and1,beinginlinewiththeFRDI.Thus,theseterrainnicheintervalsareidentiedasregionswiththebestprospectsforvegetationrestorationinthefuture.TheFUDIincreasesagainwithhigherTNI,indicatingthatregionswithasmallslopeandlowelevationoralargeslopeandhighelevationcontainthehighestuncertaintyinfutureprospects.MostpixelswithTNIbetween0.4and1arefoundinthepeakclusterdepressionregionI%,thekarstplateauregionIII%andthekarsttroughvalleyregionV%Figure7a.Therefore,thefuturevegetationrestorationhasthebestprospectsintheseprojectregions.TheareaswithaTNIgreaterthan0.8aremainlydistributedinthepeakclusterdepressionregionI%,thekarstplateauregionIII%andthekarstgorgeregionIV%.Thefuturevegetationtrendsintheseprojectregionshaveahighuncertainty.However,basedontheratioofeachterrainintervalineveryprojectregionFigure7b,wefoundthelargestshareofTNIbetween0.4and1inthekarstplateauregionIII%andthekarsttroughvalleyregionV%.Thus,theshareofpixelsprospectingfromfuturevegetationrestorationinthekarstplateauregionIIIandthekarsttroughvalleyregionVwillbegreaterthaninotherregions.ThelargestsharesofTNIgreaterthan0.8arefoundinthekarstgorgeregionIV%andthekarstbasinregionVI%,whilethelargestsharesofTNIsmallerthan0.2arefoundinthenon-projectregionVII%andthepeakforestplainregionII%.BothhighandlowTNIvaluesindicateahighfuturevegetationuncertaintyintheseregions. Figure6.TheterrainnicheindexTNIinSouthwestChinabasedonDEMdataa;thearearatiooffuturevegetationtrendsfordifferentTNIintervalsb;andthedistributionofFRDIandFUDIindifferentterrainnicheindexesTNI's c .

PAGE 11

RemoteSens. 2016 8 ,357 11of17 Figure7.TheratioofpixelsperTNIintervalforthedifferentproject/non-projectregionsaandtheratioofpixelsperregionforthedifferentTNIintervals b 3.5.ValidationoftheMethodThetimeserieswassplitintotwopartsand2008andpixelswithasignicantp<0.05trendforbothperiodswerechosenforvalidatingthefuturetrendpredictionmethodologyTable2.Inspiteoftheshorttimeperiods,thepredictionperformswellformostclassesonlythepredictionfornegativepersistenttrendsdoesnotapply. Table2.Forvalidation,thetimeserieswassplitintotwoperiodsandpixelswithsignicantp<0.05trendsinbothperiodswereanalyzed.Duetotheshortperiods,thenumbersofpixelsmeetingtheseassumptionsarelimitedshowninbracketsintherstcolumn.ThemeanslopesGSNyear 1forthesepixelsbothperiodsareshown.Thepredictionsarebasedontherstperiodto2007. PredictedGSNTrendafter2007 Basedon2001Period ObservedGSNTrend ObservedGSNTrend PositiveDevelopment+0.01+0.006 Anti-persistentPositiveDevelopment+0.01 0.001 NegativeDevelopment,336 0.02+0.01 Anti-persistentNegativeDevelopment 0.01+0.01 4.Discussion 4.1.DriversofVegetationConditionsandDynamicsManystudieshavesuggestedthatclimatechangeisoneofthemaindriversoftrendsinvegetationproductivityinChina[57,58].However,inourstudytheroleofaveragegrowingseasontemperatureandprecipitationislimitedsincethecoefcientsofdeterminationofbothrelationshipsareratherlowyetstatisticallysignicant,conrmingtheresultsofHouetal.[59].Furthermore,therelationshipbetweenGSNandprecipitationisweakerthanfortemperature,whichisconsistentwiththestudybyWangetal.[60].ThismightbeduetoSouthwestChinabeinglocatedinthesubtropicalmonsoonclimatezone,withrichprecipitationandmoderatetemperature,andtherebyslightchangesinprecipitationwillnotheavilyaffectvegetationgrowth.Ontheotherhand,theaboveandbelow-groundhydrologicalstructuresofthekarstregionleadstoasignicantdrainandrunoff[61],whichsuggeststhatprecipitationcannotbeeffectivelyutilizedforvegetationgrowth.Anthropogenicinuences,suchasecologicalprojects,populationgrowth,andurbanization,werefoundtobemoreimportantfactorsthataffectvegetationchange[6265].Afewpreviousstudieshighlightedthepositiveimpactsofecologicalrestorationprojectsonvegetationdynamics[6669].

PAGE 12

RemoteSens. 2016 8 ,357 12of17TheaccumulativeafforestedareabyaerialseedingandforestplantationinSouthwestChinahadincreasedquicklyfrom65,744hain2000to1,709,366hain2011.Furthermore,theoveralldeterioratedtrendsofrockydeserticationhadbeentransformedfromnetincreasetonetdecreasesince2011inkarstregions,SouthwestChina[70].WiththeimplementationoftheKarstRockyDeserticationComprehensiveControlandRestorationProject,thesoilfertilityandsurfacevegetationcoverrateinkarstregionshavebeenimproved[7173].Partlyduetotheecologicalprojectsinthepast15years,mostvegetationcommunitiesinthepresentstudyareahaveachievedastablegrowingstagewitharelativelyhighGSNandshowthusnosignicanttrendFigure3b.However,someregionswithasignicantdecreaseinvegetationhavebeendetected.Theseareasareusuallyaroundcountyseatsorinproximitytoroads,identifyingrapidurbanizationandroadconstructiontoleadtothedegradationofvegetation.Generally,theaveragecoverandsuccessionalrateofvegetationarelowerinkarstregionsascomparedtonon-karstregions[74].Inourstudy,theaverageGSNinprojectregionskarstregionisstilllowerthaninnon-projectregionnon-karstregion.However,theannualincreaseofGSNfrom2001to2013inprojectregions.0024GSNyear 1ishigherthaninnon-projectregion.0023GSNyear 1,indicatingthatvegetationcovergrowsmorequicklyundertheimplementationofERPs. 4.2.InuenceofTerrainConditionsonVegetationDistributionAframeworkforgovernmentaldecisionmakersisproposedtohighlightregionsthatneedmoreattentionwhenimplementingecologicalprojects.TheFRDIandFUDIindexes,basedonthefuturevegetationtrendsandtheTNI,wereutilizedtodetectspecicareaswherevegetationhadthebestprospectstoberestoredorthehighestuncertaintyinthefuture.TheresultsshowedthatregionswithaTNIbetween0.4and1werethemostadvantageousintervalsforvegetationrestorationinthefuture.Inthepast13years,theaveragegrowingseasontemperatureandprecipitationdroppedastheTNIincreased.InregionswithTNIlessthan0.4,theaveragegrowingseasontemperatureandprecipitationare28Cand1721mmrespectively,whereasinregionswithTNIbetween0.4and1,averagesare25Cand1355mm,andinregionswithTNIgreaterthan1,20Cand1218mm.Intheregionstudied,areaswithalowTNIlessthan0.4havethebestcombinedconditionsofairtemperatureandwaterfavoringvegetationgrowth.Inaddition,theReturningFarmlandtoForestProgramhasbeenimplementedinregionswithslopegreaterthan25,whichsupportsvegetationrestoration.Therefore,vegetationintheseregionsislikelytohavereachedastablecondition,andtheroomforfurtherimprovementislessthaninotherregions.TheconditionsofairtemperatureandwaterinregionswithaTNIbetween0.4and1arelessidealcomparedtoareaswithalowerTNIbutbetterthanforareaswithahigherTNI.Therefore,vegetationintheseregionshasahigherchanceofimprovementundertheimplementationofprojectsthaninregionswithahigherTNI.AccordingtothedistributionofFUDIinareaswithdifferentTNIs,vegetationdevelopmentinregionswithahighTNIgreaterthan0.8hasahigheruncertainty.Theseregionswithhighelevationandlargeslopeanglescaninducesoilerosionunderheavyrainfall.Besides,vegetationintheseareassuffersmoreeasilyunderdroughtssincethesteepterrainconditionisnotsuitableforwaterconservation[75].Thus,vegetationintheseregionsislikelytobemorestronglyinuencedbyclimaticconditionsthanotherregionsanductuatealongwithclimatevariations.AreaswithTNIsmaller0.2arefoundinthenon-projectregionVIIandthepeakforestplainregionII.Theseareasarewellsuitedforvegetationgrowthandarecharacterizedbyhighanthropogenicinuence.TheanthropogenicdisturbancesleadtoagenerallyuncertainreectedinahighFUDIandnegativevegetationdevelopment. 4.3.LimitationsandFutureResearchDirectionsForvalidation,thetimeserieswasdividedintotwoperiods.Resultsshowedthatthepredictionperformswellformostclassesexceptthepredictionfornegativepersistenttrends.Thiscouldbeinterpretedasanindicationofthesuccessfulimplementationofanthropogenicprojectspreventingfurtherdegradation.However,bothperiodsarerathershortduetothelengthofthefulltimeseries

PAGE 13

RemoteSens. 2016 8 ,357 13of17basedontheuseofMODISdata.Thus,thevalidationisbasedontrendsandpredictionscharacterizedbyhighuncertaintyandalongertimeseriesisneededforamorerobustvalidation.Moreover,theTNIwasutilizedtocharacterizetheterrainconditionsandthetwoindexesFRDIandFUDIwereaneffectiveapproachforidentifyingthemostadvantageousintervalsforvegetationrestorationinthefutureandthecorrespondingpredictionuncertainty.However,wewerenotabletoquantifythespecicelevationandslopeoftheseregions.Thus,moreworkisneededtoanalyzetherelationshipbetweenthevegetationvariationsandtopographicalconditionsandtoexploretheexactrangesofelevationandslopesuitedforspecicpurposes.Thevariationsofthevegetationtrendswereaffectedbyclimateandhumanactivities,especiallynational/localpolicies.Weexploredthedrivingforcesassociatedwithvegetationvariationsbutdidnotstatisticallydistinguishtheeffectscausedbyclimateandhumanactivitiesonvegetationdynamics,whichfacilitatesaquanticationoftheefciencyofhuman-inducedeffortonvegetationrestoration. 5.SummaryandConclusions Overall,SouthwestChinawasshowntobecharacterizedbygoodvegetationconditionduring2001and2013withoutsignsofwidespreadvegetationdegradation.Theaveragetemperatureandprecipitationofagrowingseasonwereshowntohavelimitedimpactonvegetationchange.RoadconstructionandincreasingurbanizationhaveledtovegetationwithlowergrowingseasonNormalizedDifferenceVegetationIndex,suchasscrubsandcultivatedvegetation. Mostofthevegetationshowsnosignicanttrendinthepast13years.Thepercentageofpixelswithanincreasingtrendisgreaterthanthatofdecreasingtrends.Thekarstgorgeregionhasthelargestshareofincreasingtrendsamongstthe6projectregions.Thepeakforestplainregionandthepeakclusterdepressionregionarefoundtohaveanoticeableshareofpixelshowingasignicantdecreasingtrend. ThevegetationtrendsinSouthwestChinaareratherpersistent.Forthewholestudyarea,thefuturevegetationwithastableandsteadydevelopmentisdominant,especiallyinthekarsttroughvalleyandthepeakforestplain.Thepercentageofvegetationwithpositivedevelopmentisgreaterthanthatofnegativedevelopment.Thekarstgorgehasthehighestshareofpixelsshowingpositivedevelopmentandanti-persistentdevelopment.Thekarstpeakforestplainregionhasthehighestshareofpixelswithnegativedevelopment.Thekarstbasinhasthehighestshareofpixelscharacterizedbyanundeterminedtrend. Mostnegativefuturedevelopmentisfoundinareasofhighanthropogenicinuence,ascomparedtoareasofroughterrain. TheregionswithterrainnicheindexTNIbetween0.4and1havethebestprospectsforvegetationrestorationinthefutureespeciallyinthekarstplateauregionandthekarsttroughvalleyregion.Moreover,regionswithTNIgreaterthan1showthehighestuncertaintyforfutureprospectsespeciallyinthekarstgorgeregionandthekarstbasinregion.Itisheresuggestedthatgovernmentalrestorationprojectsshoulddulypayattentiontotheseregionstopreventvegetationdegradation. Wehaveshownthattheframeworkofthepresentanalysisworkswellforassessingrestorationprospectsinthestudyarea.Duetoitsgenericdesign,themethodusedisexpectedtobeapplicableforstudieslocatedinotherareasofcomplexlandscapesintheworld,exploringfuturevegetationtrends. Acknowledgments:ThisworkwassupportedbytheNationalNaturalScienceFoundationofChinaGrantNo.41471445,41371418,ScienceandTechnologyServiceNetworkInitiativeNo.KFJ-EW-STS-092,ChineseAcademyofSciencesLightofWestChinaProgram,andEuropeanUnion'sHorizon2020researchandinnovationprogrammeundertheMarieSklodowska-CurieGrantNo.656564.TheauthorswishtothankMODISNDVIGroupforproducingandsharingtheNDVIdataset.NASAandNIMAarethankedforsharingDEMdata.Theteambehindtheweatherdataisthankedforsharingtheclimaticdata.TheauthorswanttothankYanfangXuforhelpingwithdataprocessing,andChunhuaZhangandMingyangZhangfortheirhelpfulcommentsonthe

PAGE 14

RemoteSens. 2016 8 ,357 14of17manuscript.Wealsothankthejournaleditorandtheanonymousreviewersfortheirusefulcommentsandgreateffortsonthispaper. AuthorContributions:Allauthorscontributedsignicantlytothismanuscript.Tobespecic,KelinWangandYueminYuedesignedthisstudy.XiaoweiTongandChujieLiaowereresponsibleforthedataprocessingandanalysis.XiaoweiTongwrotethedraftwithsupportofMartinBrandtandRasmusFensholt. ConictsofInterest: Theauthorsdeclarenoconictofinterest. References 1.Arora,V.Modelingvegetationasadynamiccomponentinsoil-vegetation-atmospheretransferschemesandhydrologicalmodels. Rev.Geophys. 2002 40 .[CrossRef] 2.Douville,H.;Planton,S.;Royer,J.-F.;Stephenson,D.B.;Tyteca,S.;Kergoat,L.;Lafont,S.;Betts,R.A.Importanceofvegetationfeedbacksindoubled-CO2climateexperiments.J.Geophys.Res.Atmos.2000,105,14841.[CrossRef] 3.Eugster,W.;Rouse,W.R.;Pielke,S.R.A.;Mcfadden,J.P.;Baldocchi,D.D.;Kittel,T.G.F.;Chapin,F.S.;Liston,G.E.;Vidale,P.L.;Vaganov,E.;etal.Land-atmosphereenergyexchangeinArctictundraandborealforest:Availabledataandfeedbackstoclimate. Glob.Chang.Biol. 2000 6 ,84.[CrossRef] 4.Suzuki,R.;Masuda,K.;Dye,D.G.InterannualcovariabilitybetweenactualevapotranspirationandPALandGIMMSNDVIsofnorthernAsia. RemoteSens.Environ. 2007 106 ,387.[CrossRef] 5.Kelly,M.;Tuxen,K.A.;Stralberg,D.Mappingchangestovegetationpatterninarestoringwetland:Findingpatternmetricsthatareconsistentacrossspatialscaleandtime. Ecol.Indic. 2011 11 ,263.[CrossRef] 6.Jiang,Z.C.;Lian,Y.Q.;Qin,X.Q.RockydeserticationinSouthwestChina:Impacts,causes,andrestoration.EarthSci.Rev. 2014 132 ,1.[CrossRef] 7.Daoxian,Y.RockDeserticationintheSubtropicalKarstofSouthChina;GerbderBorntraeger:Stuttgart,Germany,1997. 8.ChineseAcademyofSciences.SeveralsuggestionsforthecomprehensivetamingtokarstmountainareasinSouthwestChina. Bull.Chin.Acad.Sci. 2003 3 ,169. 9.Yue,Y.M.;Zhang,B.;Wang,K.L.;Liu,B.;Li,R.;Jiao,Q.J.;Yang,Q.Q.;Zhang,M.Y.Spectralindicesforestimatingecologicalindicatorsofkarstrockydesertication.Int.J.RemoteSens.2010,31,2115.[CrossRef] 10.Huang,Q.H.;Cai,Y.L.SpatialpatternofkarstrockdeserticationinthemiddleofGuizhouProvince,SouthwesternChina. Environ.Geol. 2006 52 ,1325.[CrossRef] 11.Pettorelli,N.;Vik,J.O.;Mysterud,A.;Gaillard,J.-M.;Tucker,C.J.;Stenseth,N.C.Usingthesatellite-derivedNDVItoassessecologicalresponsestoenvironmentalchange.TrendsEcol.Evol.2005,20,503.[CrossRef][PubMed] 12.Tucker,C.J.;Vanpraet,C.L.;Sharman,M.J.;VanIttersum,G.SatelliteremotesensingoftotalherbaceousbiomassproductionintheSenegaleseSahel:1980.RemoteSens.Environ.1985,17,233.[CrossRef]13.Badeck,F.W.;Bondeau,A.;Bttcher,K.;Doktor,D.;Lucht,W.;Schaber,J.;Sitch,S.Responsesofspringphenologytoclimatechange. NewPhytol. 2004 162 ,295.[CrossRef] 14.Levin,N.;Shmida,A.;Levanoni,O.;Tamari,H.;Kark,S.Predictingmountainplantrichnessandrarityfromspaceusingsatellite-derivedvegetationindices. Divers.Distrib. 2007 13 ,692.[CrossRef] 15.Ma,W.;Fang,J.;Yang,Y.;Mohammat,A.BiomasscarbonstocksandtheirchangesinnorthernChina'sgrasslandsduring1982. Sci.ChinaLifeSci. 2010 53 ,841.[CrossRef][PubMed] 16.Myneni,R.B.;Keeling,C.D.;Tucker,C.J.;Asrar,G.;Nemani,R.R.Increasedplantgrowthinthenorthernhighlatitudesfrom1981to1991. Nature 1997 386 ,698.[CrossRef] 17.Tucker,C.J.;Slayback,D.A.;Pinzon,J.E.;Los,S.O.;Myneni,R.B.;Taylor,M.G.Highernorthernlatitudenormalizeddifferencevegetationindexandgrowingseasontrendsfrom1982to1999.Int.J.Biometeorol.2001 45 ,184.[CrossRef][PubMed] 18.Nemani,R.R.;Keeling,C.D.;Hashimoto,H.;Jolly,W.M.;Piper,S.C.;Tucker,C.J.;Myneni,R.B.;Running,S.W.Climate-drivenincreasesinglobalterrestrialnetprimaryproductionfrom1982to1999.Science2003,300,1560.[CrossRef][PubMed] 19.Cleland,E.E.;Chuine,I.;Menzel,A.;Mooney,H.A.;Schwartz,M.D.Shiftingplantphenologyinresponsetoglobalchange. TrendsEcol.Evol. 2007 22 ,357.[CrossRef][PubMed]

PAGE 15

RemoteSens. 2016 8 ,357 15of17 20.Fensholt,R.;Proud,S.R.EvaluationofEarthObservationbasedgloballongtermvegetationtrendsComparingGIMMSandMODISglobalNDVItimeseries.RemoteSens.Environ.2012,119,131.[CrossRef] 21.Wu,Z.T.;Wu,J.J.;Liu,J.H.;He,B.;Lei,T.J.;Wang,Q.F.IncreasingterrestrialvegetationactivityofecologicalrestorationprogramintheBeijingTianjinsandsourceregionofChina.Ecol.Eng.2013,52,37.[CrossRef]22.Zheng,Y.F.;Liu,H.J.;Wu,R.J.;Wu,Z.P.;Niu,N.Y.CorrelationbetweenNDVIandprincipalclimatefactorsinGuizhouProvince. J.Ecol.RuralEnviron. 2009 25 ,12.InChinese 23.Zhang,Y.D.;Zhang,X.H.;Liu,S.R.CorrelationanalysisonnormalizeddifferencevegetationindexNDVIofdifferentvegetationsandclimaticfactorsinSouthwestChina.Chin.J.Appl.Ecol.2011,22,323.InChinese 24.Li,H.;Cai,Y.L.;Chen,R.S.;Chen,Q.;Yan,X.EffectassessmentoftheprojectofgrainforgreeninthekarstregioninSouthwesternChina:AcasestudyofBijiePrefecture.ActaEcol.Sin.2011,31,3255.InChinese 25.Yang,S.F.;An,Y.L.Spatialandtemporalvariationsofvegetationcoverageinkarstareasunderthebackgroundofecologicalrecovery:AcasestudyinthecentralareaofGuizhouProvince.EarthEnviron.2014 42 ,404.InChinese[CrossRef] 26.Tong,X.W.;Wang,K.L.;Yue,Y.M.;Liao,C.J.;Xu,Y.F.;Zhu,H.T.TrendsinvegetationandtheirresponsestoclimateandtopographyinnorthwestGuangxi. ActaEcol.Sin. 2014 34 ,1.InChinese 27.Mantegna,R.N.;Stanley,H.E.AnIntroductiontoEconophysics:CorrelationsandComplexityinFinance;CambridgeUniversityPress:Cambridge,UK,2000. 28.Grau-Carles,P.Empiricalevidenceoflong-rangecorrelationsinstockreturns.Phys.Stat.Mech.Appl.2000,287 ,396.[CrossRef] 29.Mandelbrot,B.B.;Richard,L.H. TheMisBehaviorofMarkets ;BasicBooks:NewYork,NY,USA,2004. 30.Rangarajan,G.;Sant,D.A.FractaldimensionalanalysisofIndianclimaticdynamics.ChaosSolitonsFractals2004 19 ,285.[CrossRef] 31.Hirabayashi,T.;Ito,K.;Yoshii,T.Multifractalanalysisofearthquakes.PureAppl.Geophys.1992,138,591.[CrossRef] 32.Carbone,A.;Castelli,G.;Stanley,H.E.Time-dependentHurstexponentinnancialtimeseries.Phys.Stat.Mech.Appl. 2004 344 ,267.[CrossRef] 33.Peng,J.;Liu,Z.H.;Liu,Y.H.;Wu,J.S.;Han,Y.N.TrendanalysisofvegetationdynamicsinQinghaiTibetPlateauusingHurstExponent. Ecol.Indic. 2012 14 ,28.[CrossRef] 34.Li,S.S.;Yan,J.P.;Liu,X.Y.;Wan,J.ResponseofvegetationrestorationtoclimatechangeandhumanactivitiesinShaanxi-Gansu-NingxiaRegion. J.Geogr.Sci. 2013 23 ,98.[CrossRef] 35.Miao,L.J.;Jiang,C.;Xue,B.L.;Liu,Q.;He,B.;Nath,R.;Cui,X.Vegetationdynamicsandfactoranalysisinaridandsemi-aridInnerMongolia. Environ.EarthSci. 2014 73 ,2343.[CrossRef] 36.Seddon,A.W.R.;Macias-Fauria,M.;Long,P.R.;Benz,D.;Willis,K.J.Sensitivityofglobalterrestrialecosystemstoclimatevariability. Nature 2016 531 ,229.[CrossRef][PubMed] 37.Bledsoe,B.P.;Shear,T.H.VegetationalonghydrologicandedaphicgradientsinaNorthCarolinacoastalplaincreekbottomandimplicationsforrestoration. Wetlands 2000 20 ,126.[CrossRef] 38.Pabst,H.;Khnel,A.;Kuzyakov,Y.Effectofland-useandelevationonmicrobialbiomassandwaterextractablecarboninsoilsofMt.Kilimanjaroecosystems. Appl.SoilEcol. 2013 67 ,10.[CrossRef] 39.Tsui,C.C.;Tsai,C.C.;Chen,Z.S.SoilorganiccarbonstocksinrelationtoelevationgradientsinvolcanicashsoilsofTaiwan. Geoderma 2013 209 ,119.[CrossRef] 40.Liu,Q.J.;An,J.;Wang,L.Z.;Wu,Y.Z.;Zhang,H.Y.Inuenceofridgeheight,rowgrade,andeldslopeonsoilerosionincontourridgingsystemsunderseepageconditions. SoilTillageRes. 2015 147 ,50.[CrossRef] 41.Moore,I.D.;Gessler,P.E.;Nielsen,G.A.;Peterson,G.A.Soilattributepredictionusingterrainanalysis.SoilSci.Soc.Am.J. 1993 57 ,443.[CrossRef] 42.Florinsky,I.V.;Kuryakova,G.A.Inuenceoftopographyonsomevegetationcoverproperties.CATENA1996 27 ,123.[CrossRef] 43.Ohsawa,T.;Saito,Y.;Sawada,H.;Ide,Y.ImpactofaltitudeandtopographyonthegeneticdiversityofQuercusserratapopulationsintheChichibuMountains,centralJapan.FloraMorphol.Distrib.Funct.Ecol.Plants 2008 203 ,187.[CrossRef]

PAGE 16

RemoteSens. 2016 8 ,357 16of17 44.Laamrani,A.;Valeria,O.;Bergeron,Y.;Fenton,N.;Cheng,L.Z.;Anyomi,K.EffectsoftopographyandthicknessoforganiclayeronproductivityofblackspruceborealforestsoftheCanadianClayBeltregion.For.Ecol.Manag. 2014 330 ,144.[CrossRef] 45.Lian,Y.Q.;You,G.J.Y.;Lin,K.R.;Jiang,Z.C.;Zhang,C.;Qin,X.Q.CharacteristicsofclimatechangeinsouthwestChinakarstregionandtheirpotentialenvironmentalimpacts.Environ.EarthSci.2014,74,937.[CrossRef] 46.Tong,L.Q.;Liu,C.L.;Nie,H.F.RemoteSensingInvestigationandStudyontheEvolutionofChinaSouthKarstRockyDeserticationAreas ;SciencePress:Beijing,China,2009. 47.Yuan,D.X.TheResearchandCountermeasuresofMajorEnvironmentalGeologicalProblemsinKarstAreasofSouthwestChina ;SciencePress:Beijing,China,2014. 48.Vermote,E.F.;Saleous,N.Z.;Justice,C.O.AtmosphericcorrectionofMODISdatainthevisibletomiddleinfrared:Firstresults. RemoteSens.Environ. 2002 83 ,97.[CrossRef] 49.Holben,B.N.Characteristicsofmaximum-valuecompositeimagesfromtemporalAVHRRdata.Int.J.RemoteSens. 1986 7 ,1417.[CrossRef] 50.InternationalScienticDataServicePlatform.Availableonline:http://datamirror.csdb.cn/accessedon25April2016. 51.ChinaMeteorologicalDataSharingServiceSystem.Availableonline:http://cdc.cma.gov.cnaccessedon25April2016. 52.DataCenterforResourcesandEnvironmentalSciences,ChineseAcademyofSciencesRESDC.Availableonline:http://www.resdc.cnaccessedon25April2016. 53.Hurst,H.Longtermstoragecapacityofreservoirs. Trans.Am.Soc.Civ.Eng. 1951 6 ,770. 54.SncheGranero,M.A.;TrinidadSegovia,J.E.;GarcaPrez,J.SomecommentsonHurstexponentandthelongmemoryprocessesoncapitalmarkets. Phys.Stat.Mech.Appl. 2008 387 ,5543.[CrossRef] 55.Mandelbrot,B.B.;Wallis,J.R.RobustnessoftherescaledrangeR/Sinthemeasurementofnoncycliclongrunstatisticaldependence. WaterResour.Res. 1969 5 ,967.[CrossRef] 56.Yu,H.;Zeng,H.;Jiang,Z.Y.Studyondistributioncharacteristicsoflandscapeelementsalongtheterraingradient. Sci.Geogr.Sin. 2001 21 ,64. 57.Zhou,L.;Tucker,C.J.;Kaufmann,R.K.;Slayback,D.;Shabanov,N.V.;Myneni,R.B.Variationsinnorthernvegetationactivityinferredfromsatellitedataofvegetationindexduring1981to1999.J.Geophys.Res.Atmos.2001 106 ,20069.[CrossRef] 58.Piao,S.L.;Mohammat,A.;Fang,J.Y.;Cai,Q.;Feng,J.M.NDVI-basedincreaseingrowthoftemperategrasslandsanditsresponsestoclimatechangesinChina.Glob.Environ.Chang.2006,16,340.[CrossRef]59.Hou,W.J.;Gao,J.B.;Wu,S.H.;Dai,E.F.Interannualvariationsingrowing-seasonNDVIanditscorrelationwithclimatevariablesinthesouthwesternkarstregionofChina.RemoteSens.2015,7,11105.[CrossRef] 60.Wang,J.;Meng,J.J.;Cai,Y.L.AssessingvegetationdynamicsimpactedbyclimatechangeinthesouthwesternkarstregionofChinawithAVHRRNDVIandAVHRRNPPtime-series.Environ.Geol.2007,54,1185.[CrossRef] 61.Du,X.L.;Wang,S.J.Space-timedistributionofsoilwaterinakarstarea:AcasestudyoftheWangjiazhaicatchment,Qinzhen,GuizhouProvince. EarthEnviron. 2008 36 ,193.InChinese 62.Zhu,H.Y.Underlyingmotivationforlandusechange:AcasestudyonthevariationofagriculturalfactorproductivityinXinjiang,China. J.Geogr.Sci. 2013 23 ,1041.[CrossRef] 63.Liu,J.X.;Li,Z.G.;Zhang,X.P.;Li,R.;Liu,X.C.;Zhang,H.Y.ResponsesofvegetationcovertothegrainforgreenprogramandtheirdrivingforcesintheHe-LongregionofthemiddlereachesoftheYellowRiver.J.AridLand 2013 5 ,511.[CrossRef] 64.Zhou,S.;Huang,Y.F.;Yu,B.F.;Wang,G.Q.Effectsofhumanactivitiesontheeco-environmentinthemiddleHeiheRiverBasinbasedonanextendedenvironmentalKuznetscurvemodel.Ecol.Eng.2015,76,14.[CrossRef] 65.Yu,F.K.;Huang,X.H.;Yuan,H.;Ye,J.X.;Yao,P.;Qi,D.H.;Yang,G.Y.;Ma,J.G.;Shao,H.B.;Xiong,H.Q.Spatial-temporalvegetationsuccessioninYao'anCounty,YunnanProvince,SouthwestChinaduring1976:AcasesurveybasedonRStechnologyformountainseco-engineering.Ecol.Eng.2014,73,9.[CrossRef]

PAGE 17

RemoteSens. 2016 8 ,357 17of17 66.Deng,L.;Shangguan,Z.P.;Li,R.Effectsofthegrain-for-greenprogramonsoilerosioninChina.Int.J.SedimentRes. 2012 27 ,120.[CrossRef] 67.Sun,Q.Z.;Liu,R.L.;Chen,J.Y.;Zhang,Y.P.EffectofplantinggrassonsoilerodioninkarstdemonstrationareasofrockydeserticationintegratedrehabilitationinGuizhouprovince.SoilWaterConserv.2013,27,67.InChinese 68.Xiao,L.;Yang,X.H.;Chen,S.X.;Cai,H.Y.AnassessmentoferosivitydistributionanditsinuenceontheeffectivenessoflanduseconversionforreducingsoilerosioninJiangxi,China.CATENA2015,125,50.[CrossRef] 69.Wang,J.;Wang,K.L.;Zhang,M.Y.;Zhang,C.H.ImpactsofclimatechangeandhumanactivitiesonvegetationcoverinhillysouthernChina. Ecol.Eng. 2015 81 ,451.[CrossRef] 70.TheStateForestryAdministrationofthePeople'sRepublicofChina.TheBulletinofRockyDeserticationinChina;TheStateForestryAdministrationofthePeople'sRepublicofChina:Beijing,China,2012.InChinese.71.Zhang,J.PlanningforcomprehensivedeserticationcontrolinkarstareaofGuangxiZhuangAutonomousregion. Partacult.Sci. 2008 25 ,93.InChinese 72.Yang,Q.Y.;Jiang,Z.C.;Yuan,D.X.;Ma,Z.L.;Xie,Y.Q.TemporalandspatialchangesofkarstrockydeserticationinecologicalreconstructionregionofSouthwestChina.Environ.EarthSci.2014,72,4483.[CrossRef] 73.Cai,H.Y.;Yang,X.H.;Wang,K.J.;Xiao,L.L.IsforestrestorationintheSouthwestChinakarstpromotedmainlybyclimatechangeorhuman-inducedfactors? RemoteSens. 2014 6 ,9895.[CrossRef] 74.He,Y.B.;Zhang,X.B.;Wang,A.B.DiscussiononkarstsoilerosionmechanisminkarstmountainareainsouthwestChina. Ecol.Environ.Sci. 2009 18 ,2393. 75.Wang,S.Y.;Su,W.C.;Fan,X.R.;Li,C.;Shi,X.T.InuencefactorsofsoilmoistureinkarstrockydeserticationregionAcasestudyofPudingCounty,GuizhouProvince. Res.SoilWaterConserv. 2010 17 ,171. 2016bytheauthors;licenseeMDPI,Basel,Switzerland.ThisarticleisanopenaccessarticledistributedunderthetermsandconditionsoftheCreativeCommonsAttributionCC-BYlicensehttp://creativecommons.org/licenses/by/4.0/.


printinsert_linkshareget_appmore_horiz

Download Options

close
No images are available for this item.
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.