Modeling of Spatial Distributions of Farmland Density and Its Temporal Change Using Geographically Weighted Regression Model-论文.pdf

下载文档 降价啦

10
0
约4.1万字
约 14页
2015-12-30 发布于河南
举报
版权申诉
保障服务

Modeling of Spatial Distributions of Farmland Density and Its Temporal Change Using Geographically Weighted Regression Model-论文.pdf

1、本文档共14页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。

Chin．Geogra．Sci．Vo1．24No．2PP．191——204 垒Springer SciencePress doi：10．1007／s11769—013-0631—8 www．springerlink．corn ／content／1002—0063 M odelingofSpatialDistributionsofFarmlandDensityandItsTempo— ralChangeUsingGe0graphicallyW eightedRegressionM odel ZHANGHaitao，GUO Long，CHEN Jiaying，FU Peihong，GU Jianli，LIA0Guangyu (CollegeofResourcesandEnvironment，HuazhongAgriculturalUniversity，Wuhan430070，China) Abstract：Thisstudyusedspatialautoregression(SAR)modelandgeographicallyweightedregression(GWR)modeltomodelthespa— tialpatternsoffarmlanddensityanditstemporalchangeinGuchengCounty,HubeiProvince，Chinain 1999and2009，anddiscussed thedifferencebetweenglobalandlocalspatialautocorrelationsintermsofspatialheterogeneityandnon—stationarity．Resultsshowed thatstrongspatialpositivecorrelationsexistedinthespatialdistributionsoffarmlanddensity,itstemporalchangeandthedrivingfac— tors，andthecoeffi cientsofspatialautocorrelationsdecreasedasthespatiallagdistanceincreased．SARmodelsrevealedtheglobalspa— tialrelationsbetweendependentandindependentvariables，whiletheGWR modelshowedthespatiallyvaryingfittingdegreeandlocal weightingcoefficientsofdrivingfactorsandfarmlandindices(i．e．，farmlanddensityandtemporalchange)．TheGWRmodelhas smoothprocesswhenconstructingthefarmlandspatialmode1．ThecoefficientsofGWRmodelcanshow theaccurateinfluencedegrees ofdifferentdrivingfactorsonthefamr landatdifferentgeographicallocations．Theperfomr anceindicesofGW R modelshowedthat GW R modelproducedmoreaccuratesimulationresultsthanothermodelsatdifferenttimes．andtheimprovementprecisionofGW R modelwasobvious．Theglobalandlocalfarmlandmodelsusedinthissutdyshoweddifferentcharacteristicsinthespatialdistributions offamr landindicesatdifferentscales，whichmayprovidethetheoreticalbasisforfarmlandprotectionfrom theinfluenceofdifferent drivingfactors． Keywords：spatiallagmodel；spatialerrormodel；geogr