Tsuga Search Measurement Protocols

Will Blozan and Jess Riddle, September 2006




All mensuration techniques within the Tsuga Search will adhere to the following protocols. They are based on tested methods and have proven to be accurate and repeatable.


 Diameter measurements

Tree diameter is obtained by measuring the girth of the stem at 4.5 feet (1.37 m- “breast height, BH”) above average ground (midslope) and dividing by pi. Girth is always measured perpendicular to the long axis of the stem.



















Trees having multiple stems at BH but single pith at ground level (as determined by a visual “pith trace”, see below) are measured as individuals with the narrowest point recorded as the diameter. Fused trees forking below breast height but with independent piths will be measured as individuals at 4.5 feet.


Text Box: Measuring multi-trunked trees



Tree height measurements

Tree heights are measured with a laser rangefinder and clinometer. The rangefinder provides the distance to the tree top and the clinometer provides the corresponding angle with respect to eye level. The sine of the angle times the distance gives the elevation difference between the measurer’s eye and the tree top. The same procedure is repeated with the base of the tree in place of the tree top. Adding the two elevation differences gives the total vertical height of the tree. Using the rangefinder to measure the distance to multiple tops may help identify the highest point, especially when the true high point appears lower on the horizon than other tops. All tree heights are vertical distance, not trunk length.

Total tree height= H1+H2, where H1= SIN (A1)*D1 and H2= SIN (A2)*D2


Tree volume measurements

Volume measurements can be achieved via ground-based or aerial methods. Ground–based measurements are obtained by the use of a reticled monocular, laser rangefinder, and a clinometer. Aerial measurements are direct tape measures obtained by a climber in the tree.



A reticled monocular can be used to accurately measure diameters from great distances. The distance from the measured section of trunk multiplied by the reticle reading and divided by an optical factor results in the diameter of the target. Lasered distances are estimated to the nearest 1/10th whole unit by finding “click-over” to the next unit relative to the rear of the monocular. This device coupled with a clinometer for heights and section lengths allows for a volume determination to be made without climbing the tree (see below).



As illustrated in the above diagram, the scale is oriented by fine adjustments of the tripod to line up with and perpendicular to the edge of the trunk at the “0” point on the scale. The optical intercept of the opposite side is read against the scale and estimated to the nearest 1/100th unit. The section of tree above would be recorded as intercepting 1.41 on the reticle scale. To calculate the diameter the following formula would be used:


Diameter= (Reticle scale) X (distance to target) ÷ (optical factor*)


If the section above were 90 feet (27.4 m) away the diameter would be:


Diameter= (1.41) X (90) ÷ 75; which is 1.69 feet (0.52 m)


(*Note: the optical factor is supplied by the manufacturer, and specific to the monocular model.)


Trees with limited visibility of the trunk that obscure laser bounces are measured by using a single baseline measurement to the base of the tree and a single laser shot distance to the highest portion of the trunk that is clearly visible. This allows us to still measure a portion of the trunk visible to the monocular but not directly measurable by the laser. After a point is located with a good view of the entire tree a tripod is set up with the monocular attached. The baseline distance to the center (side) of the base of the tree is measured and the distance to the center (side) of the highest visible point of the trunk is measured with the rangefinder.

This procedure creates a triangle encompassing the measurable portion of the trunk, with vertices at the observer’s eye and the bottom and top of the measurable section of trunk. The position of and distances to measured points along the trunk can then be interpolated based on the clinometer angle.


If the tree deviates significantly (+/- 2 feet; .6 m) from a straight bole the distance to the midpoint of the trunk must be measured for every sighting. Widths of the measured sections are then calculated and the height of the measurement points and lengths of the resulting sections are calculated by the angles obtained from the clinometer.


Tree climbs

Accurate aerial trunk measurements can be obtained by a tree climber. All points of measurement are referenced for height above ground to a fixed tape in the tree which is initiated at the highest point and terminated at midslope. If the top of the tree is not safely reachable a pole or stick is used to place the end of the tape at the highest point. Measurement intervals are subjectively chosen based on changes in trunk taper. An area where a change in profile is observed (in or out) is measured with a diameter tape perpendicular to the lean. Clear sections of trunk are selected so as to not include branch collars, burls, etc. Typically, around fifteen measurements are taken on single-trunked trees in addition to those obtained for the base (see below). Generally, measurements are no more than ten feet (3 m) apart. Trunk reiterations are measured and added to the final trunk volume. Reiterations are identified by an upturned branch that has gained apical dominance and formed a new vertical trunk. A bifurcation is a split in the trunk that forms two or more similarly sized ascending trunks. Bifurcations often form a fused section that cannot be accurately measured with a tape. Small bifurcation lengths are terminated at estimated pith origination. Reiteration lengths are terminated at the point of trunk contact.


Significantly large [>2.5 feet (75 cm) diameter] fused sections are measured with a frame mapping technique (see photo below). With two climbers, each on opposite sides of the tree, an area of fusion is selected to be measured. Two, six foot (~2 m) poles are connected by a thin rope threaded through opposite ends so they are adjustable (we used non-stretch arborist throw line and garden stakes). The poles are temporarily tensioned and the distance between the ends measured. Adjustments are made until they are parallel and perpendicular to the axis of the trunk. The slight tension between the poles holds them steady against the trunk. The climbers sight across the poles and agree on a “0” point from which to begin measurements. The “0” point is the “point of contact” at one end of the frame when a retracting steel carpenter’s tape is stretched across at 90 degrees to the poles. The “0” point is marked on both sides as the common reference. This reference is the “0” on the X axis for each pole, and the tape is used to measure in to the trunk across the entire intercept with the pole. Thus, the trunk profile can be plotted as the distance from the X axis reference point and the distance in to the trunk as the Y axis. Points where the tree contacts the pole is recorded as whatever the X distance is and a “0” for the Y. We measured at changes in the trunk profile and to the nearest 1/8th inch (.32 cm). Graphing of the data illustrates the cross-sectional representation of the fused trunk. See example below.








The data are then entered into a trapezoidal area function in an Excelspreadsheet and converted into area so as to calculate the equivalent circular area to use in the volume formula.


Volume calculations

Cumulative trunk volume is calculated by adding the measured sections of the tree together. No limb volume is measured or estimated. The formula for the frustum of a cone is used for all trunk sections.


Volume= H*π/3*(r12+r22+r1*r2)


Where H is the height of the frustum and r1 and r2 are the radii of the top and bottom of the frustum. The aerial or ground-based measurements are added to a basal section that is measured in more detail than those above due to a typically less columnar trunk profile. For the purpose of the Tsuga Search, the wedge of wood below high side ground formed by a tree growing on a slope is dealt with in the following way: The midslope point is carefully established by using clinometers to transpose the highest side of the ground contact (below duff) against the lowest side. The elevation positioned in-between these is considered the base of the tree and is used in both the basal volume calculations and the absolute height determination. The lowest measurable point (LMP) above root flare is taken as the lowest measurement, even if several feet above the midslope point. The LMP is then extended down to midslope as a column.


This technique, though crude, should satisfactorily approximate the volume of the flared and fissured section (fissures are not illustrated above). In cases where the trunk is extremely incised we may choose a section higher up the trunk (and thus smaller) to represent the basal column. Detailed footprint maps would be very time consuming and not result in significant gains in accuracy relative to the rest of the tree.


Relative canopy height

The heights of canopy trees surrounding the target tree will be measured as described above. One additional set of measurements will be taken to ascertain the surrounding canopy height relative to the target tree. This is achieved by measuring the relative elevation difference between the tree and the target tree. A rangefinder and clinometer is used to obtain the vertical offset, and the distance between the trees is obtained during the canopy tree mapping described below.


Canopy tree mapping

All trees over four inches (10cm) diameter whose trunk center is within 82 feet (25 m) of the center of the subject tree will be mapped. Distance is obtained with a laser rangefinder and recorded to the nearest whole meter. Measurements are taken from the basal midpoint of each tree and no slope correction is used. Species, diameter and live/dead are recorded. A compass is used to obtain the azimuth from the target tree to the mapped tree and recorded to the nearest whole degree. The compass is set to “0” degrees declination. The result is then graphed as seen in the example below.