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

Non-metric multidimensional scaling ...

University of California, Davis.

Linked to FindBook

Google Book

Amazon

博客來

Non-metric multidimensional scaling (NMDS) as a basis for a plant functional group classification and a Bayesian belief network formulation for California oak woodlands.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Non-metric multidimensional scaling (NMDS) as a basis for a plant functional group classification and a Bayesian belief network formulation for California oak woodlands./
Author:	Alonso Valdes, Maximo.
Description:	119 p.
Notes:	Source: Dissertation Abstracts International, Volume: 69-09, Section: B, page: 5157.
Contained By:	Dissertation Abstracts International69-09B.
Subject:	Agriculture, Forestry and Wildlife. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3329589
ISBN:	9780549855385

Non-metric multidimensional scaling (NMDS) as a basis for a plant functional group classification and a Bayesian belief network formulation for California oak woodlands.
Alonso Valdes, Maximo.

Non-metric multidimensional scaling (NMDS) as a basis for a plant functional group classification and a Bayesian belief network formulation for California oak woodlands. - 119 p.

Source: Dissertation Abstracts International, Volume: 69-09, Section: B, page: 5157.

Thesis (Ph.D.)--University of California, Davis, 2008.

One of the main goals of rangeland management is to sustain or obtain a specific plant community composition. Thus, rangeland management requires not only a good characterization of the plant community under management, but also a fairly good prediction of the response of that community to management actions. In this research on California oak woodlands we sampled 455 100-m transects both in the Coast Range and Sierra Nevada Foothill areas during the growing seasons of 2004 and 2005. Each transect site was characterized by its management history, disturbance frequency, soil characteristics and precipitation. Non-metrical multidimensional scaling (NMDS) and cluster analysis were combined to identify and classify the current most common plant communities. These communities and their canonical and spatial distribution were also analyzed using a plant functional group (PFG) approach. Finally, prior knowledge and observational data were combined in a Bayesian belief network to generate predictions about the woody species composition under different levels of precipitation and grazing. Our results indicate that the plant community composition on California oak woodlands is not a random assemblage of species. They also indicate that Quercus douglasii, Q. wislizeni and Pinus sabiniana are homogeneously distributed in the Coast Range and Sierra Nevada Foothills. Q. lobata and Q. kelloggii are mostly, and Q. agrifolia and Q. garryana are only present in the Coast Range. The distribution of the PFG's along precipitation and elevation gradients is not random. Evergreen species are located principally in the northern counties, whereas in the southern counties the predominant species are deciduous and adapted to dry conditions. Precipitation, soil available water capacity, elevation, slope and exposure are the most important variables in explaining the variability in the woody species composition. The probability of occurrence of Q. wislizeni, P. sabiniana, Q. kelloggii, and Q. garryana increases dramatically with precipitation. Grazing increases the probability of occurrence of Q. douglasii and Q. wislizeni; does not affect Q. lobata; and decreases the probability of finding P. sabiniana, Q. agrifolia, Q. garryana, and most of the shrubs. Further studies are needed to transform these patterns of distribution in quantitative rules of plant community assemblages.

ISBN: 9780549855385Subjects--Topical Terms:

783690
Agriculture, Forestry and Wildlife.

Non-metric multidimensional scaling (NMDS) as a basis for a plant functional group classification and a Bayesian belief network formulation for California oak woodlands.
LDR:03332nmm 2200277 a 45 001 867129
005 20100802
008 100802s2008 ||||||||||||||||| ||eng d
020 $a 9780549855385
035 $a (UMI)AAI3329589
035 $a AAI3329589
040 $a UMI $c UMI
100 1 $a Alonso Valdes, Maximo. $3 1035835
245 1 0 $a Non-metric multidimensional scaling (NMDS) as a basis for a plant functional group classification and a Bayesian belief network formulation for California oak woodlands.
300 $a 119 p.
500 $a Source: Dissertation Abstracts International, Volume: 69-09, Section: B, page: 5157.
502 $a Thesis (Ph.D.)--University of California, Davis, 2008.
520 $a One of the main goals of rangeland management is to sustain or obtain a specific plant community composition. Thus, rangeland management requires not only a good characterization of the plant community under management, but also a fairly good prediction of the response of that community to management actions. In this research on California oak woodlands we sampled 455 100-m transects both in the Coast Range and Sierra Nevada Foothill areas during the growing seasons of 2004 and 2005. Each transect site was characterized by its management history, disturbance frequency, soil characteristics and precipitation. Non-metrical multidimensional scaling (NMDS) and cluster analysis were combined to identify and classify the current most common plant communities. These communities and their canonical and spatial distribution were also analyzed using a plant functional group (PFG) approach. Finally, prior knowledge and observational data were combined in a Bayesian belief network to generate predictions about the woody species composition under different levels of precipitation and grazing. Our results indicate that the plant community composition on California oak woodlands is not a random assemblage of species. They also indicate that Quercus douglasii, Q. wislizeni and Pinus sabiniana are homogeneously distributed in the Coast Range and Sierra Nevada Foothills. Q. lobata and Q. kelloggii are mostly, and Q. agrifolia and Q. garryana are only present in the Coast Range. The distribution of the PFG's along precipitation and elevation gradients is not random. Evergreen species are located principally in the northern counties, whereas in the southern counties the predominant species are deciduous and adapted to dry conditions. Precipitation, soil available water capacity, elevation, slope and exposure are the most important variables in explaining the variability in the woody species composition. The probability of occurrence of Q. wislizeni, P. sabiniana, Q. kelloggii, and Q. garryana increases dramatically with precipitation. Grazing increases the probability of occurrence of Q. douglasii and Q. wislizeni; does not affect Q. lobata; and decreases the probability of finding P. sabiniana, Q. agrifolia, Q. garryana, and most of the shrubs. Further studies are needed to transform these patterns of distribution in quantitative rules of plant community assemblages.
590 $a School code: 0029.
650 4 $a Agriculture, Forestry and Wildlife. $3 783690
650 4 $a Agriculture, Range Management. $3 1022885
650 4 $a Artificial Intelligence. $3 769149
650 4 $a Biology, Ecology. $3 1017726
690 $a 0329
690 $a 0478
690 $a 0777
690 $a 0800
710 2 $a University of California, Davis. $3 1018682
773 0 $t Dissertation Abstracts International $g 69-09B.
790 $a 0029
791 $a Ph.D.
792 $a 2008
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3329589