An assessment of the use of airborne LiDAR for estimating growth of Sitka spruce (Picea sitchensis) plantation forestry at Kielder Forest, UK

Abstract

A growing need exists for the collection of accurate and up-to-date information on forest growth rates for management purposes. Recent studies indicate that airborne laser scanning (ALS) offers a quicker and more cost-effective approach than the traditional methods of forest inventorying and may have the potential not only to revolutionise forest management but also provide key data concerning world carbon stocks. This study aims to assess the potential of ALS to estimate forest growth rates of the temperate Sitka spruce plantation forests using canopy height distribution models at Kielder Forest, Northumberland. ALS data from 2003 and 2006 provides an excellent, unique opportunity to contribute to existing work which has so far been limited in focus, looking primarily at individual tree growth m the less densely stocked, slow-growing, cold climate forests of Scandinavia. ALS point cloud data from first and last pulse returns are filtered and classified. Ground returns are used to create digital elevation models (DEM), and first returns used to create digital canopy height models (DCHM). Key ALS variables are then extracted and summarised. Processed ALS data from both years are compared to estimate forest growth. The results are compared with ground truth data. Height correlations are strong and positive. Growth is detected at all plot locations but correlations with ground truth data are weak and mostly negative. Potential explanations for the lack of correlation are presented and discussed, including; data misalignment, inherent error within the ground truth data and the set-up of the LiDAR systems. Further study is necessary to quantify and eliminate systematic and random error within both the LiDAR and ground truth data before ALS may be used routinely for forest management purposes.