A new way of computing diameters

[dbhguru]

NTS,

     One of our most important measuring challenges is trunk and limb modeling. Heretofore, serious modeling efforts have involved climbs. The recent ones have included the use of a sophisticated, labor intensive measuring protocol. Data from a climb are fed to a sophisticated computer model developed by Steve Sillet and BVP.

    Ground-based methods are still being employed, but cannot match the Sillet-BVP modeling method. Still, ground measurements can be useful. The problem is that the ground-based methods are also labor intensive. Our standard method is to measure the apparent diameter of the trunk using a reticle device such as a Macroscope 25 or 45. However, visibility from the ground is often a problem. Macroscope optics are minimal and the parallax effect that occures when the object fills most of the reticle is hard to eliminate. So, although, the reticle is useful, it is far from optimal. Enter Laser Technologies TruPulse 360. I'm embarrassed that it has taken me so long to exploit that instrument's full capabilities.

    Over the last several days, I've been working on a modeling method using the TruPulse. I was initially not pleased with the performance of the built in compass to compute horizontal angles, but I now realize that wasn't re-calibrating the compass often enough. In addition, I wasn't using a tripod to hold the TP 360 still. When both steps are taken, the compass proves much more accurate than advertised.

   The attachment shows the results of an experiment to measure diameter consisting of 32 trials. For each trial, I placed a marker on the trunk of a tree. At the level of the marker, I measured the apparent diameter using a sophisticated set of calipers. So, I got the width of the tree at the level of the marker that I would then be seeing from a distance. I set up my tripod and shot the distance to the marker using the TruPulse at the point of change-over to the next half foot. The TP 360 is accurate to about 1/10th of a foot at change-over. Next, I shot the azimuth of the left and right sides of the trunk at the level of the marker. The excellent optics of the 360 made seeing the sides of the tree far easier than looking through the Macroscope. I completed the process by using the routine in the attached spreadsheet to calculate the diameter at the point of the marker and compare the result to the measurement made with the calipers. The results speak for themselves. The average difference is 0.8 inches. I think a new day has dawned for ground-based modeling.

  I have a lot more to do, but a big hurdled has been surmounted. I can see much better with the TruPulse, and with the one instrument, I can get distance to the center of the trunk and azimuth to the edges. All three measurements are easy to take. I can work up the trunk, using the VD mode of the TruPulse. I plan to do a complete trunk tomorrow, and will report the results.

Bob

[edfrank]

Bob,

Looks interesting and worth developing.  My big concern is that with measuring smaller diameters the amount of error would become a much higher percentage of the measured diameter.  The potential error in the measurement of azimuth, I would think, would be relatively constant regardless of the diameter being measured.  So it would measure larger girths well, but would do less well for smaller diameter measurements.  Something like the Macroscope I think would do better overall, even though the same class of estimate errors would still occur affecting smaller diameter measures more than larger diameter ones.

Ed

[DougBidlack]

Bob,

this does look very exciting.  I'm really interested in seeing a volume comparison between your new method vs the Macroscope vs a large white pine that has been climbed, measured and modelled by Will.

Doug

[dbhguru]

Ed, Doug,

  Ed your point is well taken. The diameter error for large and small trunks/limbs alike averages 0.8 inches, based on my tests. This amount certainly represents a larger percent error for small trunks/limbs. However, the chance of large error with the reticle is greater at long distances due to optics, and for larger trunks/limbs from closer range due to the parallax effect. I suppose the combination of the two instruments could resolve the percent error issue. I'll plan an experiment to test the results of the reticle against the TruPulse for small targets at close range and then from a distance.

  For actual modeling exercises, I have no problem using both instruments if it improves the results. I have both the Macroscope 25 and the Macroscope 45. I'll test them both against the TruPulse 360.

Bob

[M.W.Taylor]

Bob,

Bob, it is my opinion that your new method for estimating trunk widths at various height intervals using the TP360 and "missing line" routine should in theory be more precise, faster and easier than using the Macroscope25 with laser.   The Macroscopes are handy little items for estimating trunk width, but as you stated earlier, there is distortion when you fill the entire screen up on the Macroscope. There is just no way around this unless you can get trunk views that are far away which is difficult in a cluttered forest. I don't know how calculate this distortion error so I always try to get as far back as possible with the Macroscope25.

Also, just as significant may be the Macroscope25's 50 tick mark stadia scale which is cruder when compared to the TP360's .1 degree azimuth resolution. The accuracy relative to true North on the TP360 (which seems to drift a little from site to site and day to day) is not relavent to the accuracy of each trunk azimuth estimate from edge to edge.  If I did my calculations correctly, each tick mark on the Macroscope is equivalent to .7625 degrees when using 75 as the Macroscope constant.  This is lousy resolution when compared to the TP360.  The key is for the Macroscope user to estimate in-between ticks. Can be done with practice but to .1 resolution like the TP360 ?  Seems easier said than done.

Now the price tag of $1,600  for TP360 vs. $150 for Macroscope25 is a consideration for most people.  A home-built horizontal transit is one cheaper alternative.

Wendell Flint measured the volumes of the giant sequoias for his book "To Find The Larges Tree", using a transit and horizontal sweep angles to determine trunk widths at various height intervals. His transit/tape-line derived trunk diameters have been proven very close to tape wraps and certainly more accurate than a Macroscope25 based measurement.

Bob, your new trunk diameter measuring technique is very practical and I would say accurate too. Thanks for posting your method on the NTS blog. I plan to use your method as soon as I get my hands on a TP360. I will use the remote trigger + tripod to get the best possible results from the TP360.

I may in fact ditch the Macroscope25 if the TP360 works better for me. Your results are very encouraging.

Michael Taylor
WNTS VP
California AFA Big Trees Coordinator
http://www.landmarktrees.net

[dbhguru]

Michael,

  Last night I performed a simple test on the TP 360. I took an ordinary waste basket, one foot deep and with a top diameter of 13 inches and a bottom diameter of 11 inches.  Its form is a perfect frustum of a cone. Its dimensions yielded a volume of 0.787 cubic feet. I then measured its dimensions using the TP 360 from a distance of 21.5 feet. The results led to a volume of 0.736 cubic feet. The difference is 0.051 cubic feet or 6.5% of the actual volume. That percentage is likely to drop significantly for larger trees, since the angle and distance errors don't go up in absolute value as the diameter increases.

   I have developed an Excel workbook that utilizes the TP 360's features. I'll soon send it to you for review.

Bob