Localizing a Diameter Increment Model with a Sequential Bayesian Procedure

Author: Gertner, George Z.

Source: Forest Science, Volume 30, Number 4, 1 December 1984 , pp. 851-864(14)

Publisher: Society of American Foresters

Buy & download fulltext article:

Price: $29.50 plus tax (Refund Policy)

Abstract:

A sequential Bayesian procedure is presented and used for localizing a nonlinear diameter increment model taken from a regional forest growth projection system. With this procedure, diameter growth monitored in a subregion of interest can be used to adjust the parameters of a regional model. The amount the regional parameters are adjusted depends on the precision of the growth estimates from the regional model, as well as on the precision of the estimates based on the sample from the subregion. When adjusting the regional parameters, more weight is given to these estimates when the precision of subregional estimates is relatively high in comparison to the regional estimates. An example is given where the procedure is used to localize a model developed for the Western Oregon region to a Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) stand located in northwestern Oregon. Another example is presented where the adjustment procedure is used to adjust for the effects of fertilizer. Forest Sci. 30:851-864.

Keywords: Bayesian procedure; feedback; growth projection system; sequential estimation

Document Type: Journal article

Affiliations: 1: Assistant Professor of Forest Biometrics, 110 Mumford Hall, Department of Forestry, University of Illinois, 1301 W. Gregory Drive, Urbana, IL 61801

Publication date: 1984-12-01

More about this publication?