Estimation of tree growth is generally based on repeated diameter measurements. A major issue in modeling growth of tropical trees is the presence of buttresses in many species. As a tree grows, buttresses gradually encroach up the trunk, and it has become common practice to change the height at which diameter is measured so that buttresses do not lead to biased estimates of inordinately high growth rates. However, trees taper with height, so a diameter taken at a new height will not be comparable to the old diameter. Such height changes in the course of a study can mean that a tree’s growth trajectory is not directly available from its time series of diameters. This important issue is compounded by the facts that diameter measurements often imply negative growth, and individual variation is considerable. These complexities related to data could have a large impact on inference concerning effects of climate change.
Results/Conclusions
Here we introduce a hierarchical state space method that allows formal integration of data on diameter taken at a range of different heights, and that can include individual variation, as well as temporal effects or other covariates. We illustrate our approach using a species from two data-sets. We show how detailed modeling affects inference on climate change