Prediction of Timber Quality Parameters of Forest Stands by Means of Small Footprint Airborne Laser Scanner Data

Abstract

The aim of this study was to explore the capability of airborne laser scanner (ALS) data to explain the variation in fieldmeasured variables representing timber quality within square 0.25 ha grid cells in a mature conifer forest in the southeast of Norway. These variables were the mean ratio between stem diameter at six m above ground and the diameter at breast height (R_D6), the volume of saw logs (V_SL), the proportion of saw logs relative to the total volume (P_SL), the ratio between tree height and diameter at breast height (HD), mean basal area diameter (D_g), and crown height (CH). Each of these variables was modeled using a mixed modeling approach. Model fit was expressed by the Pseudo-R², and were 0.85, 0.50, 0.78, 0.57, 0.74, and 0.58 for the respective quality variables. Furthermore, much of the residual error could be attributed to the different forest stands from which the grid cells originated even though we used field-observed tree species proportions as auxiliary information. It was concluded that more auxiliary information is needed to estimate models that are general across stands, but that the relationships between ALS-data and the quality variables considered here seem strong enough to be utilized for example to prioritize between stands in relation to harvest when specific quality distributions are sought.

Keywords:

• Airborne laser scanning
• timber quality
• forest stands
• predictive models