Nonlinear models of height growth for Douglas-fir in southwestern Oregon Public Deposited

http://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/00000240c

Descriptions

Attribute NameValues
Creator
Abstract or Summary
  • A review is made of methods which assess the bias and non-normality of parameter estimates and predictions obtained with nonlinear regression. Particular emphasis is placed upon curvature measures of nonlinearity, related measures of parameter and prediction bias, and the effects of reparameterizations. Alternate models of individual tree height growth are compared on the basis of mean square error, intrinsic nonlinearity, parameter effects nonlinearity, and estimated bias. While results are specific to the data examined, some general conclusions are made concerning appropriate models for individual tree height growth. Both the Richards and a Weibull-type growth model are found to adequately describe individual tree height growth, with low levels of intrinsic nonlinearity, and acceptable parameter effects nonlinearity following repararneterization. Some evidence is found for a modification of either the Richards or Weibull model to include an asymptotic linear growth rate when modeling the height growth of some western conifers past the age of 200. Stem analysis data on Douglas-fir height growth in mixed confier stands located in southwestern Oregon are used to develop a system of dominant height growth and site index prediction. The Weibull model is used successfully to develop a polymorphic height growth prediction equation. A linear model, estimated with site index as the dependent variable, is used to predict site index. A comparision is made of pooled least squares and random coefficient estimation methods. The random coefficient method is found to more closely model the shape of early height growth, but appears to result in more biased predictions and performs very poorly on older height growth, with both the estimation and validation data. Alternative error assumptions are examined with the pooled data method. The best performance in validation is obtained with assumption of independent errors, heteroscedastic across trees.
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Academic Affiliation
Non-Academic Affiliation
Subject
Rights Statement
Language
Digitization Specifications
  • Master files scanned at 600 ppi (256 Grayscale) using Capture Perfect 3.0 on a Canon DR-9080C in TIF format. PDF derivative scanned at 300 ppi (256 B+W), using Capture Perfect 3.0, on a Canon DR-9080C. CVista PdfCompressor 4.0 was used for pdf compression and textual OCR.
Replaces
Additional Information
  • description.provenance : Submitted by Digital Production (digitalproduc@gmail.com) on 2009-02-03T18:54:24Z No. of bitstreams: 1 Scrivani_John_A_1985.pdf: 667106 bytes, checksum: e60393968d559d9d8f04bf2fe2ea2f24 (MD5)
  • description.provenance : Approved for entry into archive by Linda Kathman(linda.kathman@oregonstate.edu) on 2009-02-04T16:54:47Z (GMT) No. of bitstreams: 1 Scrivani_John_A_1985.pdf: 667106 bytes, checksum: e60393968d559d9d8f04bf2fe2ea2f24 (MD5)
  • description.provenance : Approved for entry into archive by Linda Kathman(linda.kathman@oregonstate.edu) on 2009-02-04T16:53:09Z (GMT) No. of bitstreams: 1 Scrivani_John_A_1985.pdf: 667106 bytes, checksum: e60393968d559d9d8f04bf2fe2ea2f24 (MD5)
  • description.provenance : Made available in DSpace on 2009-02-04T16:54:48Z (GMT). No. of bitstreams: 1 Scrivani_John_A_1985.pdf: 667106 bytes, checksum: e60393968d559d9d8f04bf2fe2ea2f24 (MD5)

Relationships

Parents:

This work has no parents.

Last modified

Downloadable Content

Download PDF

Items