Author: García, Oscar¹

Source: Forest Science, Volume 51, Number 4, August 2005 , pp. 277-283(7)

Abstract:

Most site index models are built either from repeated measurements on permanent sample plots or by using data obtained by stem analysis. Sometimes both types of data are available. It is then of interest to test them for consistent height-age trends, and to estimate model parameters using all the data. Efficient procedures should take into account the difference in error patterns between the two data sources. A stochastic differential equation model is proposed in which trend and error parameters can take either different or common values for various data groups. All parameters are simultaneously estimated by maximum-likelihood, and computed likelihoods and likelihood-ratio tests can be used for comparing model variants and for hypothesis testing. The methods are applied to Pinus pinaster plantations in northern Spain. FOR. SCI. 51(4):277–283.

Keywords: Height growth; Pinus pinaster; differential equations; Richards; Spain; environmental management; forest; forest management; forest resources; forestry; forestry research; forestry science; natural resources; natural resource management

Document Type: Regular article

Affiliations: 1: University of Northern British Columbia 3333 University Way Prince George British Columbia Canada V2N 4Z9 Phone: (250) 960-5004;, Fax: 960-5539, Email: garcia@unbc.ca