Modelling Forest Growth and Yield: Applications to Mixed Tropical Forests

Modelling Forest Growth and Yield: Applications to Mixed Tropical Forests

Jerome K. Vanclay

Language: English

Pages: 329

ISBN: 2:00134682

Format: PDF / Kindle (mobi) / ePub


This book provides an introduction to growth modeling in mixed forests, with emphasis on the tropics. It is not intended as a "how-to" manual with step-by-step instructions, as there is no simple best way to model such forests. Rather it reviews different approaches, highlighting their strengths and limitations. It emphasizes empirical-statistical models rather than physiological-process type models, because of the proven utility of the former in forest management. Each chapter includes exercises which can be completed manually or on PC and spreadsheet. The book will serve as a reference manual for practitioners and as a text for advanced level courses in forest modeling

 

 

 

 

 

 

 

 

 

 

 

 

 

 

plot variation and thus reduce between plot variance. Cost considerations usually dictate that temporary inventory plots (or point samples) are most efficient for resource inventory. Specialized techniques for timber cruising offer great efficiencies (see e.g. Schreuder et al. 1993), but may not provide data suitable for input to yield forecasting systems. Continuous Forest Inventory for yield control: Some systems of yield regulation monitor the forest growth and harvesting by remeasuring a

investigate silvicultural and management alternatives, the database must include experimental data with paired treatment and control plots, both with adequate isolation. In contrast to continuous forest inventory plots, it is not necessary for the permanent plots to be representative or numerically proportional to forest type areas, but it is essential that they sample the full range of stand conditions. Long Term Monitoring of Environmental Change: Several researchers (e.g. Adlard 1990, Dawkins

unforeseen topic. There may be good reasons to edit or omit data from some specific analysis, but the main database should never be altered. 96 M odelling Forest G rowth and Yield The quality and cost of data available for analysis may be improved substantially through the use of electronic data recorders (e.g. Fins and Rust 1987, Wood 1990). Electronic hand-held devices enable basic checks of the input data to eliminate simple errors (e.g. transposition) at their source, comparisons with

Fig. 6.6. 115 Interpreting residual plots: (A) an outlier, (B) non-constant variance, (C) transformation required, and (D) variable X j should be included in the model. constant variance (Fig. 6.6, B) and outliers which warrant further investigation (A). It is useful to plot residuals against explanatory variables to look for transformations that may be required (C), and to check for additional variables that should be included (D). Residuals may be standardized and plotted against standard

p. 72). The effect of site productivity on survival is unclear. There is empirical evidence that in plantations, density-dependent mortality expresses itself earlier on better sites, and if mortality is expressed as a function of age, it appears that mortality increases with increasing site productivity. However, if mortality is expressed with respect to top height or stand density, a different picture emerges. Better sites should be able to sustain a higher basal area, and all other things being

Download sample

Download