Calculating Forest Biomass with Small Format Aerial Photography, Videography and a Profiling Laser

This paper reviews an aerial sampling technique that could reduce the cost of calculating standing forest biomass in remote regions by collecting a geographically distributed set of large scale transects that include a limited number of permanent plots. The biomass of these plots is calculated from allometric equations and extrapolated to the transects as a function of tree species, crown diameter and height. Two kinds of aerial imagery sampling techniques were tested over a mixed hardwood /conifer forest in Crooksville, Ohio. Thirty-one 28-meter diameter plot sites were established along two transects within the test site and the dbh (diameter at breast height) of trees measured. Each transect was flown at 4,000 feet AGL with a small format photogrammetric camera, then at 1,000 feet AGL with global positioning system (GPS)-logged dual-camera video (wide-angle/zoom), and a profiling laser. The photogrammetric images were used to create 3D models of the terrain from which average tree heights and forest volumes could be calculated. The video / laser coverage was used to visually identify the species of individual trees, measure their crown diameters in a 2D geographic information system (GIS) and estimating their heights from the laser profile. Due partly to an automatic georeference and mosaicking program developed for the video coverage during the project, it proved the more accurate / cost effective approach and this paper concentrates on its results. Average dbh values of one hectare transects calculated from the video are compared to those measured on site, showing a strong positive correlation given the small sample size. A much larger test of 635 plot sites has just been completed in Bolivia, which should allow us to refine the techniques.