A Comparison of Existing Models for DBH Estimation from Large-scale Photos
- 1 April 1989
- journal article
- Published by Canadian Institute of Forestry in The Forestry Chronicle
- Vol. 65 (2) , 114-120
- https://doi.org/10.5558/tfc65114-2
Abstract
The performance of 12 diameter prediction models suggested in the literature was studied in a controlled operational test. These models were linear and logarithmic transformations of tree height and/or crown area and were analyzed for white spruce, lodgepole pine, and trembling aspen and balsam poplar combined. Overall, all models were statistically significant, with differences due to variations in species and model form. Although simpler models may be adequate depending on operational objectives, two models emerged as deserving further investigation. It was unclear whether both tree height and crown area were needed as predictors of tree dbh for all species. Tree height was more highly correlated with dbh than crown area for all species except lodgepole pine. Measuring both tree height and crown area in comparison with tree height alone, however, increases measurement cost substantially from $10.29 to $17.50 per plot (1987 dollars).This publication has 3 references indexed in Scilit:
- Potential prediction bias in regression and discriminant analysisCanadian Journal of Forest Research, 1986
- Tree Height: Can Large Scale Photo Measurements Be More Accurate Than Field Measurements?The Forestry Chronicle, 1985
- Use of Logarithmic Regression in the Estimation of Plant BiomassCanadian Journal of Forest Research, 1972