Height-diameter model for natural black pine stands in Eğirdir region: Quantile regression approach

Abstract

Black pine (Pinus nigra JF Arnold.) is one of the most economically and ecologically important tree species in Turkey. In this context, reliable and accurate information about the current status, growth and yield characteristics of these forests is needed for the sustainable management of black pine forests. One of the most important components of growth and yield models is the height-diameter relationships. Diameter and height variables are also the most important measurement tools in forest inventory studies. Since the height of a standing tree cannot be measured as easily as the breast height diameter, tree height is often estimated as a function of diameter in forest inventory studies. For this reason, accurate and reliable estimation of tree height has a great importance for forestry activity. In this study, a height-diameter model was developed for natural black pine stands in Eğirdir region using Quantile Regression (QR) techniques. The measured data were randomly divided into two equal groups. Chapman-Richards height-diameter model was chosen as the base model for both methods since this model has been provided successful results in previous studies. QR models are developed based on three quantile (3, 5 and 9 quantiles) sets in this study. The most important advantage of QR approach over other estimation methods is that QR allows the calibration of the model with extra sample tree heights. For this purpose, different calibration alternatives were tested using a number of trees ranging from 1 to 10 in each sample plot. As a result of the study, it was seen that the 3QR approach performed better than both 5QR and 9QR approaches in terms of describing the height-diameter relationships. In addition, it has been determined that the calibration with five sample trees in each sample plot is appropriate in terms of both the estimation precision of the models and the sampling costs