Growth and Survival of Eucalyptus grandis - a study based on modelling lifetime distributions

Meike Dickel, Heyns Kotze, Klaus von Gadow, Walter Zucchini

Abstract


This study presents a new approach to estimating density-dependent survival and growth in four experimental plots of Eucalyptus grandis.

planted at different densities on a homogeneous site in Zululand/South Africa. Estimates of future basal area could be improved considerably by first estimating the number of surviving trees and including this estimate in the basal area predictions.

We estimate the probability of survival of a single tree using a Weibull distribution. Because we only know the number of trees that died between two observations points and not the exact time points of their death, the maximum likelihood estimation was adjusted to deal with these interval-censored data.

The estimated shape and scale parameters show an exponential relationship with the initial number of planted trees.

The relationship proved to be useful for estimating the shape and scale parameters of the lifetime distribution for any given initial number of planted trees.

The number of surviving trees $(N2)$ at time $t2$, given that there were N1 trees at the initial time point, is binomially distributed.

The binomial probability represents the weight for the calculation of the unconditional distribution of stand basal areas.

This procedure is useful for obtaining a better understanding of the distribution of future stand basal areas.

It was found that the basal area expectation values differ over time, but the variance of the basal areas remains almost constant for all ages and planting densities.

Our approach of estimating future basal area, based on modeling lifetime distributions, proved to be superior to convential methods.  MCFNS 2(2):86-96.


Keywords


Planting density, spacing experiment, Weibull distribution; basal area growth

Full Text:

PDF

Refbacks

  • There are currently no refbacks.


   

© 2008 Mathematical and Computational Forestry & Natural-Resource Sciences