Look Park - Weekend Events   Robert Leverett
  Feb 06, 2006 07:45 PST 
ENTS,

LOOK PARK-CROWN DISPLACEMENT MEASUREMENTS:

     On Saturday and Sunday, I went to Look Park in Northampton
(technically Florence) to begin measuring the horizontal offsets from
the trunk of the high points of a sample of trees. In each case, the
high point was first located and then I lined myself up so that my eye,
the high point, and the trunk were in a straight line. The horizontal,
eye-level distances to the high points and the trunks were measured and
the absolute value of the differences taken. The following table gives
the results. As can be seen, for the sample of 12 trees, the average
offset is 9 feet. This compares favorably with a large 1800-tree sample
I ran from my database many months ago. I think that result yielded an
average of 8.2 or 8.3 feet.

     Many more trees will have to be modeled before we can present
tables of average offsets for various species and form classes of trees.
However, we know what the across species averages are very well. As
proof of that, on Monday, I received a super spreadsheet from Will
Blozan summarizing work done by I presume the Will-Jess team on
crown-point displacement. The spreadsheet includes data on 100 trees of
10 species. The average of the absolute values of the crown
displacements is 8.2 feet!! In terms of knowing the average offset, I
think we are there. Great job Will and Jess. In computing the average
offset, absolute values needed to be averaged so that trees that lean
away from the observer averaged with tree that lean toward the observer
donít cancel each other out when using data gathered from height
measurements.


Species Hgt Cir    Hz-Offset Avg
SY 92.5 9.8 22.5
SY 96 8.5 11.4
SY 96.9 9.2 6.9
SY 99.8 6.8 9
SY 104.3 10.7 7.5
SY 113.1 8.2 9.9
SY 114.3 16.4 11.4
SY 116.6 10 3.3
SY 119.1 12.2 5.7 9.7
--------------------------------------------
WP 121.3 6.6 6
WP 125.8 9.5 6.6
WP 131 9 7.8 6.8
---------------------------------------------
Average 9.0


LOOK PARK RUCKER INDEX:

   On Sunday, I finally completed a Rucker Index for Look Park. It is as
follows:   
       Height Species Circumference
136.0 White Pine     9.6
119.1 Sycamore 12.2
112.1 Shagbark Hic 5.2
110.2 N. Red Oak 10.4
106.3 Red maple 6.5
105.4 Sugar Maple 10.5
101.5 American Elm 8.0
93.1 Black Locust 5.9
92.8 Hemlock         7.5
89.1 White Oak 7.3

Rucker Index   
106.55

    Look Park got hammered in the 1938 hurricane. The park lost 257
trees. Many of the larger, older pines have had their tops blown out.
The loss of crowns is highly visible in the forms of these pines today.
Their recovery is bears testament to the capacity of white pines to
continue life after losing their tops. I do believe that in time Look
Park will produce a few 140-foot pines, though 136 is the tallest now.

    Besides the pines, Look Park has a number of gorgeous, large red
oaks. A sample follows.

Circumference    Height
   
11.1          100.8
10.9             104.4
10.4          110.2
9.3          103.3
       
WHITE PINES OVER 130 FT IN THE CONNECTICUT RIVER VALLEY

   So far, in the Northampton area, Iíve measured 6 white pines to over
130 feet. Look Park has 4 over 130 that Iíve measured. There is one on
Smith College Campus, and Monicaís Pine. There are bound to be others
lurking around. There are no less than a dozen in Mount Tom State
Reservation over 130 and probably a dozen and a half. There are couple
or three in a ravine in Easthampton. However, within the Connecticut
River Valley, so far there has been only one solidly confirmed white
pine that reaches 140 feet. One of the Easthampton pines may make it,
but I get conflicting measurements.

THE MACROSCOPE

  MACROSCOPE 25 is a fixed magnification unit that provides 25X magnification over a field of view measuring approximately 8mm(5/16"). It incorporates a double calibrated interchangeable measuring reticle providing precise measurements in both English and Metric systems. By detaching the base and close-up lens MACROSCOPE 25 can be used for 8x distance viewing, and with formulas included, make distance or height approximations. 


My Macroscope 25 came on Friday. Like Will and Jess say, it is one
sweet instrument. Here are the results of the first simple test.

Dist Mills Diam Diam         D-Tape Diff    RD 1000 Diff
Mtrs            Mtrs    Inches       Inches Inches Inches Inch

21.9 2.825 0.8249 32.47598     32.2    0.28     33.2   1.00
21.9 2.8 0.8176 32.18858     32.2    0.01     32.3   0.10
21.9 2.775 0.8103 31.90119     32.2    0.30     31.4   0.80
Average              0.20      0.63
Pct of act      0.6%            2.0%


      As can be seen, the average difference between three consecutive
readings and the D-Tape reading represents a percentage error of 0.6%
for the Macroscope 25 and 2% for the Dendrometer. I do realize that the
D-Tape assumes circularity, so the above test is flawed from that
perspective.

      A big advantage that the Macroscope 25 has over the RD 1000 is
that you can read the Macroscope 25 scale much easier and to a far finer
resolution. Depending on how good your eyes are, you can read the
reticle to 1 or 2 tenths of a millimeter. There is no question about
reading it to the nearest quarter of a millimeter. In the above table,
the middle row is the most probable reading for the Macroscope 25. For
the RD 1000, it is a toss up between rows one and two. Giving the RD
1000 the benefit of the doubt, I chose to reflect the closer reading to
the D-Tape reading as the best choice. The Macroscope 25 is an
astounding 0.01 inches off and the RD 1000 is an impressive 0.1 of an
inch off.

      A lot more tests for the Macroscope 25 are planned in consultation
with Will and Jess. They are light years ahead of me at this point. But
right now, I couldnít be more pleased. I have a great combination. The
RD 1000 provides for quick modeling when extreme accuracy is not
required, and when it is, then to get ever finer, the Macroscope 25
comes roaring in. Sweet!

...

Robert T. Leverett
Cofounder, Eastern Native Tree Society

Re: Weekend Events   Edward Frank
  Feb 06, 2006 12:56 PST 

Bob,

I have been thinking about your results from look Park. I am not sure that
the results compare that well to your original analysis. I the 1800 tree
set you processed, the tops of the trees were oriented data variety of
angles toward you away from you, angled to the left, angled to the right of
the trunk. These variety of orientations produced an average offset of 8.2
feet. In your experiment you lined up the top with the trunk on the tree -
in effect maximizing the measured offset. Yet you only had an offset of 9.0
feet, compared to that of 8.2 for the larger sample. I would have expected
an offset greater than 8.2 by 30 to 60%. If truly random the actual offset,
what you measured in your experiment, of the l800 tree larger data set would
have been about 13 feet. Yet for your oriented sample set you obtained an
average offset 9 feet. Even if there was some bias in the original data
set, with more of the offsets toward you rather than randomly distributed,
the result still would likely have been higher than you measured 9 feet.
Perhaps a larger sampling would yield different results, perhaps this set of
trees just have less offset than did the larger set. Will and Jess's
samples of 100 trees, if they did not specifically try to line-up the top
with the trunk are consistent with the previous 1800 tree analysis.

Ed

RE: Weekend Events   Robert Leverett
  Feb 06, 2006 13:06 PST 

Ed,

    You are right. I also would have expected a greater average offset
based on the 13-foot derived average offset. From the spreadsheet, Will
sent, it appears that Will's and Jess's data compare with my 1800
measurements. The 8.2 feet is established pretty well where alignment
has not been attempted as a standard procedure.

Bob
RE: Weekend Events   Robert Leverett
  Feb 07, 2006 07:20 PST 

Will,

     I wasn't sure if Jess participated. However, given your separate
test, it is good to have independent verification of the average offset.
Just to reiterate for the membership at large, the 8.2 isn't an indirect
mreasure of the difference between sine and tangent based height
determinations. Of greater impact on the difference is the spot being
chosen as the crown point. Without a laser rangefinder, deciding on what
spot in the crown is the high point to be measured and where that point
is in relation to the base requires a very experienced eye. If you don't
have it, the errors made on tall, broad-crowned trees can be in the tens
of feet, as we've so often confirmed.

     BTW, I ran another comparison test of the macroscope and
dendrometer on a flat target that is exactly 18 inches wide. The
distance to the center of the flat target was 226 feet or 68.9 meters.
The angle to it was -9 degrees. The distance fed to the dendrometer was
222.5 = (226-0.75)cos(-9), since the distance fed to the dendrometer
must be the level distance to the vertical line containing the target.
The dendrometer gave a width of 18.4 inches. The macroscope yielded a
clear width of 5 millimeters on the reticle that convereted to 18.08
inches. The percent errors for the macroscope and dendrometer were
respectively 0.54% and 2.22%. I was quite pleased with the dendrometer
and absolutely ecstatic with the macroscope.

    After a little more practice, with the macroscope, plans to tackle
the Sunderland sycamore aren't far away.   

Bob


  Will Blozan wrote:

     FYI- that list is solely mine...