Five Instruments for Measuring Tree Height: An Evaluation

Five instruments were tested for reliability in measuring tree heights under realistic conditions. Four linear models were used to determine if tree height can be measured unbiasedly over all tree sizes and if any of the instruments were more efficient in estimating tree height. The laser height finder was the only instrument o produce unbiased estimates of the true height for all four linear models. An accuracy test showed the clinometer and enbeeco instruments produced biased results in the 0-33 and > 66 fi height classes respectively. The results for the laser may be misleading due to the limited amount of data collected with this instrument. The results (subjectively) confirm that trees up to 40 fi are measured very accurately with no exceptions for the clinometer, relaskop, and enbeeco. The tele-relaskop generally produced results that were poorer than the other instruments. South. J. Appl. For. 18(2):76-82. When measuring the height of atree, where the base and top are well defined and clearly visible, the existing instruments for measuring the height of standing trees are adequate for most applications (Hunt 1958, Warren 1959, Rennie 1979). However, the variance and bias of the height estimates could be large for tall trees in dense stands or for trees which do not have well-defined tops. An opportunity to evaluate some of the currently available height measurement instruments (including a laser-driven instrument) presented itself in the summer of 1991. Tree climbers, who were employed to collect foliage samples for chemical analysis, measured the true height of standing trees. Readings from a number of different height measuring instruments were compared to the values obtained by the tree climbers. These data were used to determine if any one instrument was superior to the other instruments under real world conditions. Data Description and Collection Methods The data set consists of 100 hardwood and softwood trees with 21 different species. The predominant species are loblolly (Pinus taeda L.), slash (Pinus elliottii Engelm. var. elliottii), shortleaf (Pinus echinata Mill.) and longleaf pine (Pinus No•: Mention of trade names or commercial products does not constitute endorsement or recommendation of use. Palustris Mill.). Seventeen additional species including a variety of oaks also appear in the data set. All data collected are from Forest Health Monitoring plots in Georgia. True heights for all trees were taken while crews collected foliage samples from the tree crowns. One crew member climbed as high as possible up each tree. From that point poles were used to measure the remaining distance to the top of the tree. To determine when the pole was at the top of the tree, sightings were taken from the ground by two observers and by the crew member in the tree. When all crew members were in agreement, the total height from the ground to the top of the pole was calculated. While some measurement error exists in this method, no alternative method could be implemented which met time and cost constraints and stall represented realistic measurement situations, such as varying terrain, canopy, tree height distributions, and species mix Readings from a laser height finder (Jasumback 1991), Suunto clinometer (Husch et al. 1982), Speigel relaskop, Enbeeco clinometer 1, Speigel tele-relaskop (Bitterlich 1978, Husch et al. 1982) were recorded from the same location. All measurements were taken by an experienced field crew member. A tripod was used to steady each instrument. Ttus required adapting a tripod mount for the Suunto clinometer, 1 H. Steward Ltd, Enbeeco House, Carlton Park, Saxmundham, Suffolk IP17 2NL, U.K.