Summary
Published in Forestry: An International Journal of Forest Research, vol. 94(5): 611–629.
Manual aerial photo interpretation is used as a cost-effective source of data for forest inventories; however, the process of photo interpretation is inherently subjective and is often undertaken by multiple photo-interpreters for a given forest management area. In contrast, airborne laser scanning (ALS) data enable characterization of forest structures in a systematic fashion with quantifiable levels of accuracy and precision that often exceed required targets and standards. Yet, the gains associated with the use of new technologies for forest inventory are difficult to measure because the quality of existing photo-interpreted inventories have rarely been quantified.
Using ALS data as a reference, the objective of this study was to quantify the precision of photo-interpreted estimates of forest stand height and canopy cover (CC). We examined forest inventories from three study sites in three different forest regions of Canada. Each of the study sites was located within a different provincial jurisdiction with unique photo interpretation standards and forest ecosystems. Stand-level estimates of forest height and cover were compared to reference estimates generated from the ALS data. Overall, our results indicated that precision was greater for photo-interpreted estimates of height. While the relationship between photo-interpreted estimates of height and ALS estimates of height were generally linear and consistent for all study sites, relationships for CC were non-linear. Precision for both stand height and cover varied by dominant species, inventory stand structure, age and ALS canopy complexity. In most cases, the difference between the photo-interpreted estimate and the ALS estimate was statistically significant. Also, the variability in photointerpretation precision as a function of the aforementioned factors was not consistent among our three study sites, indicating that site-specific forest conditions and photo-interpretation procedures influence the precision of photo-interpreted estimates. The influence of local forest conditions and interpretation procedures are therefore important considerations to quantify the potential relative gains in precision, which may be afforded by technologies such as ALS for forest inventory programs.
File
Sector(s):
Forests
Categorie(s):
Scientific Article
Theme(s):
Cartography and Data, Imaging and LiDAR
Departmental author(s):
Author(s)
Piotr TOMPALSKI, Joanne C. WHITE, Nicholas C. COOPS, Michael A. WULDER, Antoine LEBOEUF, Ian SINCLAIR, Christopher R. BUTSON, Marc-Olivier LEMONDE
Year of publication :
2021
Keywords :
airborne laser scanning, ALS, photointerpretation, mapping, precision, lidar, height, canopy cover
