n Southern African Forestry Journal - Predicting fir trees stem diameters using Artificial Neural Network models : research note

Volume 2005, Issue 205
  • ISSN : 0038-2167



The financial exploitation of forests constitutes an important part of human activity. This effort is being made in order to conserve the sustainable exploitation while avoiding the degradation of the environment. One of the most important operations in forest mensuration is the estimation of the volume of sample trees. Such estimation may be used in order to conduct a large stand inventory, construct volume tables, develop tree competition and growth models, predict tree mortality from diameter growth, and implement a number of other activities. Efficiency, the trade-off between cost and precision, is of paramount importance in the estimation of volume. <BR>The aim of this paper is to examine the applicability of Artificial Neural Network models (ANNs), in the prediction of fir trees stem over bark diameters at 5.3, 9.3, 13.3, 17.3, 21.3, 25.3, 29.3 and 33.3 meters above ground. The values of these diameters are necessary for an efficient estimation of a single tree volume using the well-known Smalian's sectional method. The system proposed in this paper can be applied in forest inventory making an accurate estimate of the volume of a sample tree based on only two diameter measurements (stump diameter, &lt;I&gt;d&lt;/I&gt;&lt;SUB&gt;0.3&lt;/SUB&gt; and diameter at breast height, <I>d</I><SUB>1.3</SUB>) and an estimate of the total tree height <I>(h)</I>. Training of the ANNs was achieved through the cascade correlation algorithm, which is a feed-forward and supervised algorithm. Kalman's learning rule was used to modify the artificial neural networks weights. These networks are designed by putting weights between neurons, through the use of the hyperbolic-tangent function of training. The estimation system proposed is capable of replacing many standard forestry mensuration procedures due to its efficiency and accuracy. The neural network models were found to be appropriate and accurate for the prediction of all diameters.

Loading full text...

Full text loading...


Article metrics loading...


This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error