Timber yield projections for tropical tree species: The influence of fast juvenile growth on timber volume recovery

Most growth and yield models for tropical tree species use diameter growth data obtained from permanent sample plots. A potential disadvantage of this data source is that slow-growing, suppressed juvenile trees are included of which only a small fraction will attain harvestable size. If this is the case, the average growth rate of extant juvenile trees will be lower than the historical, juvenile growth rate of trees of harvestable size. Thus, if juvenile growth rates are obtained from permanent plots, future timber yield may be underestimated. To determine the magnitude of this effect we simulated tree growth based on two types of diameter growth data: long-term tree-ring data from harvestable trees (lifetime growth data') and growth data of the last 10 years from trees of all sizes ('plot-type growth data'). The latter data type is a proxy for growth data from permanent sample plots. Second, we evaluated which percentage of harvestable timber volume at initial harvest is available at second harvest using lifetime growth data. We obtained tree-ring data from 89 to 98 individuals of three Bolivian timber species over their entire size range. Based on these data tree growth simulations were performed for two scenarios: a second harvest in 20, and in 40 years. A realistic degree of growth autocorrelation was incorporated in the growth projections, for both the lifetime and the plot-type growth data. Observed ages at the minimum cutting diameter in tree-ring data of harvestable trees were high: average ages varied from 63 to 179 years among species. Commercial ingrowth was 23-46% larger when using lifetime growth data than for plot-type growth data for two of the three species. Thus, the faster juvenile growth of trees that reached harvestable size indeed resulted in higher projected timber yield. In spite of the positive effect of higher juvenile growth on projected timber yield, our simulations showed that the recuperation of timber volume was low. Only 18-33% of the timber volume logged at first harvest could be obtained at second harvest after 20 years. For a second harvest after 40 years this was 26-49%. Based on our results we recommend the use of above-average growth rates in timber yield projections that apply permanent plot data. To determine which percentage of faster growers should be used in such simulations, it is important that similar analyses are conducted for more species. (C) 2010 Elsevier B.V. All rights reserved.