東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Predictive modeling of hazardous was...

University of South Carolina., Geography.

Linked to FindBook

Google Book

Amazon

博客來

Predictive modeling of hazardous waste landfill total above-ground biomass using passive optical and LIDAR remotely sensed data.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Predictive modeling of hazardous waste landfill total above-ground biomass using passive optical and LIDAR remotely sensed data./
Author:	Hadley, Brian Christopher.
Description:	171 p.
Notes:	Adviser: John R. Jensen.
Contained By:	Dissertation Abstracts International70-04B.
Subject:	Engineering, Environmental. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3352736
ISBN:	9781109108088

Predictive modeling of hazardous waste landfill total above-ground biomass using passive optical and LIDAR remotely sensed data.
Hadley, Brian Christopher.

Predictive modeling of hazardous waste landfill total above-ground biomass using passive optical and LIDAR remotely sensed data. - 171 p.

Adviser: John R. Jensen.

Thesis (Ph.D.)--University of South Carolina, 2009.

This dissertation assessed remotely sensed data and geospatial modeling technique(s) to map the spatial distribution of total above-ground biomass present on the surface of the Savannah River National Laboratory's (SRNL) Mixed Waste Management Facility (MWMF) hazardous waste landfill. Ordinary least squares (OLS) regression, regression kriging, and tree-structured regression were employed to model the empirical relationship between in-situ measured Bahia (Paspalum notatum Flugge) and Centipede [Eremochloa ophiuroides (Munro) Hack.] grass biomass against an assortment of explanatory variables extracted from fine spatial resolution passive optical and LIDAR remotely sensed data. Explanatory variables included: (1) discrete channels of visible, near-infrared (NIR), and short-wave infrared (SWIR) reflectance, (2) spectral vegetation indices (SVI), (3) spectral mixture analysis (SMA) modeled fractions, (4) narrow-band derivative-based vegetation indices, and (5) LIDAR derived topographic variables (i.e. elevation, slope, and aspect). Results showed that a linear combination of the first- (1DZ_DGVI), second- (2DZ_DGVI), and third-derivative of green vegetation indices (3DZ_DGVI) calculated from hyperspectral data recorded over the 400 -- 960 nm wavelengths of the electromagnetic spectrum explained the largest percentage of statistical variation (R2 = 0.5184) in the total above-ground biomass measurements. In general, the topographic variables did not correlate well with the MWMF biomass data, accounting for less than five percent of the statistical variation. It was concluded that tree-structured regression represented the optimum geospatial modeling technique due to a combination of model performance and efficiency/flexibility factors.

ISBN: 9781109108088Subjects--Topical Terms:

783782
Engineering, Environmental.

Predictive modeling of hazardous waste landfill total above-ground biomass using passive optical and LIDAR remotely sensed data.
LDR:02844nam 2200313 a 45 001 853273
005 20100701
008 100701s2009 ||||||||||||||||| ||eng d
020 $a 9781109108088
035 $a (UMI)AAI3352736
035 $a AAI3352736
040 $a UMI $c UMI
100 1 $a Hadley, Brian Christopher. $3 1019508
245 1 0 $a Predictive modeling of hazardous waste landfill total above-ground biomass using passive optical and LIDAR remotely sensed data.
300 $a 171 p.
500 $a Adviser: John R. Jensen.
500 $a Source: Dissertation Abstracts International, Volume: 70-04, Section: B, page: .
502 $a Thesis (Ph.D.)--University of South Carolina, 2009.
520 $a This dissertation assessed remotely sensed data and geospatial modeling technique(s) to map the spatial distribution of total above-ground biomass present on the surface of the Savannah River National Laboratory's (SRNL) Mixed Waste Management Facility (MWMF) hazardous waste landfill. Ordinary least squares (OLS) regression, regression kriging, and tree-structured regression were employed to model the empirical relationship between in-situ measured Bahia (Paspalum notatum Flugge) and Centipede [Eremochloa ophiuroides (Munro) Hack.] grass biomass against an assortment of explanatory variables extracted from fine spatial resolution passive optical and LIDAR remotely sensed data. Explanatory variables included: (1) discrete channels of visible, near-infrared (NIR), and short-wave infrared (SWIR) reflectance, (2) spectral vegetation indices (SVI), (3) spectral mixture analysis (SMA) modeled fractions, (4) narrow-band derivative-based vegetation indices, and (5) LIDAR derived topographic variables (i.e. elevation, slope, and aspect). Results showed that a linear combination of the first- (1DZ_DGVI), second- (2DZ_DGVI), and third-derivative of green vegetation indices (3DZ_DGVI) calculated from hyperspectral data recorded over the 400 -- 960 nm wavelengths of the electromagnetic spectrum explained the largest percentage of statistical variation (R2 = 0.5184) in the total above-ground biomass measurements. In general, the topographic variables did not correlate well with the MWMF biomass data, accounting for less than five percent of the statistical variation. It was concluded that tree-structured regression represented the optimum geospatial modeling technique due to a combination of model performance and efficiency/flexibility factors.
590 $a School code: 0202.
650 4 $a Engineering, Environmental. $3 783782
650 4 $a Remote Sensing. $3 1018559
690 $a 0775
690 $a 0799
710 2 $a University of South Carolina. $b Geography. $3 1019507
773 0 $t Dissertation Abstracts International $g 70-04B.
790 $a 0202
790 1 0 $a Cowen, David J. $e committee member
790 1 0 $a Edwards, Donald $e committee member
790 1 0 $a Graf, William L. $e committee member
790 1 0 $a Jensen, John R., $e advisor
791 $a Ph.D.
792 $a 2009
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3352736