國立高雄大學圖資館 |

登入

回首頁

A digital spatial predictive model of land-use change using economic and environmental inputs and a statistical tree classification approach :Thailand, 1970s--1990s.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	A digital spatial predictive model of land-use change using economic and environmental inputs and a statistical tree classification approach :
其他題名:	Thailand, 1970s--1990s.
作者:	Felkner, John Sames.
面頁冊數:	447 p.
附註:	Adviser: Carl Steinitz.
附註:	Source: Dissertation Abstracts International, Volume: 61-09, Section: A, page: 3795.
Contained By:	Dissertation Abstracts International61-09A.
標題:	Urban and Regional Planning.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9988860
ISBN:	0599957875

A digital spatial predictive model of land-use change using economic and environmental inputs and a statistical tree classification approach :Thailand, 1970s--1990s.
Felkner, John Sames.

A digital spatial predictive model of land-use change using economic and environmental inputs and a statistical tree classification approach :Thailand, 1970s--1990s. [electronic resource] - 447 p.

Adviser: Carl Steinitz.

Thesis (D.Des.)--Harvard University, 2000.

For this research, a spatial statistical predictive model of land use change was created and run in two provinces of Thailand. The model utilized an extensive spatial database, and used a classification tree approach for explanatory model creation and future land use (Breiman, Friedman, Olshen, & Stone, 1984). Eight input variables were used, and the trees were run on a dependent variable of land use change measured from 1979 to 1989 using classified satellite imagery. The derived tree models were used to create probability of change surfaces, and these were then used to create predicted land cover maps for 1999. These predicted 1999 maps were compared with actual 1999 landcover derived from 1999 Landsat 7 imagery.

ISBN: 0599957875Subjects--Topical Terms:

212416
Urban and Regional Planning.

A digital spatial predictive model of land-use change using economic and environmental inputs and a statistical tree classification approach :Thailand, 1970s--1990s.
LDR:03437nmm 2200301 450 001 155176
005 20030117152118.5
008 230606s2000 eng d
020 $a 0599957875
035 $a 00087701
035 $a 155176
040 $a UnM $c UnM
100 0 $a Felkner, John Sames. $3 212881
245 1 0 $a A digital spatial predictive model of land-use change using economic and environmental inputs and a statistical tree classification approach : $b Thailand, 1970s--1990s. $h [electronic resource]
300 $a 447 p.
500 $a Adviser: Carl Steinitz.
500 $a Source: Dissertation Abstracts International, Volume: 61-09, Section: A, page: 3795.
502 $a Thesis (D.Des.)--Harvard University, 2000.
520 # $a For this research, a spatial statistical predictive model of land use change was created and run in two provinces of Thailand. The model utilized an extensive spatial database, and used a classification tree approach for explanatory model creation and future land use (Breiman, Friedman, Olshen, & Stone, 1984). Eight input variables were used, and the trees were run on a dependent variable of land use change measured from 1979 to 1989 using classified satellite imagery. The derived tree models were used to create probability of change surfaces, and these were then used to create predicted land cover maps for 1999. These predicted 1999 maps were compared with actual 1999 landcover derived from 1999 Landsat 7 imagery.
520 # $a The model was able to predict future land use, given a series of assumptions, with 90 percent overall accuracies. The model can be used in other developing or developed country locations for future land use prediction, determination of future threatened areas, or to derive “rules” or principles driving land use change.
520 # $a The primary research hypothesis was that an explanatory model using both economic and environmental input variables would better predict future land use change than would either a model using only economic variables or a model using only environmental. Thus, the eight input variables included four economic and four environmental variables.
520 # $a The results indicated a very slight superiority of the full models to predict future agricultural change and future deforestation, but a slight superiority of the economic models to predict future built change. However, the margins of superiority were too small to be statistically significant. The resulting tree structures were used, however, to derive a series of principles or “rules” governing land use change in both provinces.
520 # $a The scale and extent of global land use change is massive, and has potentially powerful effects on the global climate and global atmospheric composition (Turner & Meyer, 1994). Because of this tremendous change and impact, there is an urgent need for quantitative, empirical models of land use change, especially predictive models with an ability to capture the <italic>trajectories </italic> of change (Agarwal, Green, Grove, Evans, & Schweik, 2000; Lambin et al., 1999).
590 $a School code: 0084.
650 # 0 $a Urban and Regional Planning. $3 212416
650 # 0 $a Geography. $3 174760
650 # 0 $a Landscape Architecture. $3 212405
650 # 0 $a Remote Sensing. $3 212882
650 # 0 $a Environmental Sciences. $3 212393
650 # 0 $a Economics, General. $3 212429
710 0 # $a Harvard University. $3 212445
773 0 # $g 61-09A. $t Dissertation Abstracts International
790 $a 0084
790 1 0 $a Steinitz, Carl, $e advisor
791 $a D.Des.
792 $a 2000
856 4 0 $u http://libsw.nuk.edu.tw/login?url=http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9988860 $z http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9988860