Pushing the envelope   Robert Leverett
  Nov 14, 2003 08:28 PST 

ENTS:

    The flurry of e-mails discussing how we fine-tune our tree
measurements to take into consideration the limits of our instruments
might seem like overkill to those with only a casual interest in tree
dimensions, but our obsession is no more nor less than that of the
members of any group who seek to push the envelope for their occupation
or sport. Consider how far technical climbing has evolved to allow rock
faces to be routinely scaled that once appeared unclimbable. We're
pushing the envelope and we know it, but we don't expect those with a
less intense interest in tree measuring to adopt our methods or feel
pressured into seeking ultra precision. There's ample room for lesser
intensity in keeping with the objectives of the champion tree programs -
which we believe are important.

    We will continue to make our expertise available to American
Forests, the state champion tree coordinators, and are quite pleased to
help one of our own - Will Fell develop a measuring guide. More
measuring workshops will be scheduled. However, through it all, we
promise to remain sensitive to those gentle spirits who just want to
admire trees for their beauty.

    As we gain in reputation and accomplishment, our pugnacity will
diminish. We needed to stand our ground early on came to offset the lack
of understanding on the part of others as to what ENTS was all about and
to draw attention to entrenched tree-measuring methods in need of
change. As Russ Richardson stated, we're taking tree measuring to a
higher level, albeit to nowhere near the level that Bob Van Pelt takes
it to in his research. I do sense real progress on both fronts, so the
need to be openly combative/iconoclastic is subsiding. However, we will
still encounter reactionaries on occasion and have to administer a good
verbal spanking. We'll do it with increased sensitivity, though. It'll
be a nice spanking.

    Our site documentation methodology is taking on a life of its own
thanks to the inner group. However, we now need to expand the Rucker
Index concept to encompass both height and circumference indices. We
could even add a third index devoted to circumference x height. The
three indices, taken together, would reveal much more about a site.
Given the large girth cottonwoods out Lee Frelich's way, it makes sense
to add both the Rucker Circumference Index and the Rucker Point Index to
paint a more complete picture of the distinctions between eastern and
mid-western forests of the same species. Differences in forest structure
that can be reflected in the indices that are attributable to the
frequency and intensity of storms and other climate features should be
documented.

Well, enough rambling. Someone else's turn.

RE: Pushing the envelope   Will Blozan
  Nov 14, 2003 10:29 PST 

I like the idea of the expanded Rucker Indices. Could be very compelling.
But what about Baxter Creek were you have a mix of ages and disturbances.
The tall trees are second-growth mainly, but the relic old-growth trees are
fat and stumpy (but not all). An index of same tree cbh x height would
eliminate inflation, which is what I am sure you meant.

Will
Re: Pushing the envelope   Colby Rucker
  Nov 14, 2003 13:38 PST 

Will,

About a year ago I noticed that the tallest trees at some sites were much
heavier older growth that at others, so I determined the girth x height
value of each of the ten trees making up the height index, and then found
the average. The Cataloochee District had many big old specimens, but also
a downright wispy sycamore with a cbh of 3' 3.0". Of course, many of the
trees have since been replaced by taller specimens, but here are some 2002
numbers:

Cataloochee (148.27)     1778.4
Beall Woods (121.52)    1417.7
Carters Grove (122.02) 1233.2
Cook Forest (132.42)    1190.1
Belt Woods (130.97)     1134.4
Corcoran   (119.75)        1131.1
Chase Creek (130.19)   1004.4
Mohawk Trail (132.55)    950.4

Despite the differences in the height index, the numbers largely reflect
ranking by cbh, with the exception of Corcoran, which had the fourth largest
cbh. The rankings seemed to show the role of old trees versus younger
second growth and also varied terrain versus relatively flat habitat. This
would create four groups:

1. Cataloochee
2. Beall, Carters and Corcoran
3. Cook and Belt
4. Chase and Mohawk

Is this close to what you and Bob were thinking?

Colby
RE: Pushing the envelope   Will Blozan
  Nov 14, 2003 14:37 PST 

Colby, Bob,

A problem with the combined index is that it is limited by the girth of the
tallest tree, and may not adequately reflect the structural characteristic
of the site (Although, a "n" value of 10 is not adequate anyway...). If the
GxH index is to reflect anything useful, it may need to be based on the
largest girth and tallest of a species regardless of whether the fiqures are
from the same tree or not. Maybe it does not make sense to combine the
indices, and may better serve our needs as a separate value. Perhaps the two
separate values could then be multiplied. ?????

Many of the Cataloochee height records are "small" trees, but MUCH larger
trees of the same species grow there, but are not the tallest. I would bet
that the "Girth index" would be "shortchanged" by 40% or so if limited to
the girths of the tallest trees. In fact, many old-growth sites may be
indistinguishable from second growth, since the younger, smaller trees are
often the tallest representatives of a species. In Beall woods I focused on
the largest trees, not necessarily the tallest (tho I doubt the Rucker will
rise by much with more searching).

Will
Re: Pushing the envelope   Colby Rucker
  Nov 14, 2003 15:11 PST 

Will,

All good points. The height index works well because heights cluster
regardless of girth, and the matter of a foot or so is significant, as we
see with tuliptree, hophornbeam, black birch and many others. The combined
index is more erratic because the tallest specimen could well have a cbh of
6, 12, or 15 feet, producing massive changes in the rankings.

I think the 2002 numbers give an example of how rankings can encourage new
insights regarding forest structure, but the formula's not a dependable way
to get there.

So, back to the drawing board...

Colby
RE: Pushing the envelope   dbhg-@comcast.net
  Nov 14, 2003 16:19 PST 

Will:

Yes, absolutely. The third Rucker index based of the product of circumference and height would take data from the same trees, i.e. the inputs to the product of circumference x height would be for same tree.

I'm going to try to compute the additional Rucker indices for Ice Glen this Sunday. Ice Glen is a small enough place to permit that.

Bob
RE: Pushing the envelope   dbhg-@comcast.net
  Nov 14, 2003 16:19 PST 

Will, Colby, et al:

   I wasn't confident that combining data from different trees would be received well so I reasoned that separate indices for height, circumference, and "overall bigness" might tell us what we're interested in knowing. The pure height index certainly tells us a good deal about growing conditions. A girth index, kept separate, will reflect growing conditions and age in a slightly different way, although one that is not totally clear in my mind. Then an overall bigness index should add additional information, some redundant, but not necessarily all of it.

   When we look at the 3 indices for a site in adjacent columns and make comparisons to the triplicate for other sites, we may see patterns emerge that latitude and longitude best explain. I don't know, but thought it worth the exercise on a limited scale. I thought of trying it out for MTSF, MSF, Ice Glen, and Mt Tom for starters.

The triplicate may just be a way to capture forest diversity since an abundance of tall trees as one population, large girth trees as another, and a mediocre population of cir x hgt champs might tell us that there are some young, fast growth areas, a scattering of very old trees of a number of species, but a missing population in the range of both significant height and large girth because of the loss of those trees from past clearing operations. Would a forest with high values for all three indices go straight to the front of the class?

Bob
Re: Pushing the envelope even farther   dbhg-@comcast.net
  Nov 15, 2003 05:37 PST 

Will, Colby, John, Dale, Lee, et al:

   Applying the Rucker Index concept to girth for my Mecca, MTSF, I get 13.42. I doubt that I'll be able to increase this index much, if any.

   The current index for Cir X Hgt for MTSF is 1364.36 courtesy of two huge trees, the champion sugar maple and Big Bertha. Otherwise it would be right at 1300.

   So, let's see, the spread of Rucker indices for Mohawk is:


Site     Rucker-Hgt      Rucker-Cir        Rucker-CirXHgt

MTSF       134.45          13.42               1364.36

   It's a start. I have a feeling that throwing in the two new indices will allow the southern sites to shine even more. Congaree and the Smokies have to be through the roof. I'd be very curious as to where Cook Forest is? Dale?

Bob
Re: Pushing the envelope even farther   Colby Rucker
  Nov 15, 2003 08:10 PST 

Bob,

Yes, I'm also curious about the girth index for MTSF. Since Big Bertha's
not one of the ten trees on the height index, I'm with Will in asking if the
ten girth specimens are of the same ten species as the height index, or
simply examples of the ten species of greatest girth at MTSF.

It probably doesn't matter if the two lists are independent. Maximum height
and girth aren't necessarily related. They don't seem to be compatible
attributes of old growth or some sort of exemplary growth. Will wisely
cautions against arbitrary opinions as to what exemplary growth should look
like. You'll recall my overly long essay suggesting that ultimate growth
hardwoods compete by spread, not height, and are, therefore, shorter than
second growth. Second growth is an unavoidable reaction to forest injury,
but is unstable, with continuing windthrow leaving only shorter trees with
crowns broad enough to have strong root systems.

I also suggested that such forest structure is modified by the inherent
height capability of many dominant species, where height is inversely
related to seed weight.

Colby

----- Original Message -----
From: "Will Blozan" 
Sent: Saturday, November 15, 2003 10:16 AM

Was the girth index based on the same trees as the height index?

RE: Pushing the envelope even farther   dbhg-@comcast.net
  Nov 15, 2003 08:31 PST 

Will:

   No,the girth index is based on a different set of trees. It seeks to look at the forest through an entirely different lens. The product index does use the same set of trees.

Bob
RE: Pushing the envelope   Will Blozan
  Nov 15, 2003 07:15 PST 

If the premise in our minds is that for a site to be at the "front of the
class" it needs to have big AND tall trees (as one individual) we may be
missing the entire point. Such a forest development "ideal" gets us back to
the mis-guided attempts we are all to familiar with to put old-growth
forests in a box and define it out of existence. Our ideal attributes may
not exist in Nature in a natural system.

Will
RE: Pushing the envelope   dbhg-@comcast.net
  Nov 15, 2003 08:20 PST 

Will:

   I'm not wedded to any of these approaches and you're right to be dubious of "front of the class" designations. In retrospect, I shouldn't have used that phrase. I think I'm just looking at different indexes to see where they point.

Bob
Re: Pushing the envelope    Michael Davie
   Nov 16, 2003 13:11 PST 

I guess the question is what are we trying to document with these indices?
The regular Rucker index is very useful for encapsulating a sites overall
height (better for diverse sites), and cbh information for those trees is
great comparative information to have on hand. But as many have pointed out,
cbh of the tallest trees in many tall forests is not as indicative of site
potential as much as it might be indicative of age or competition, or even
the species composition. If the purpose is to try to quantify site potential
in productivity or biomass or as Bob put it, "overall bigness" kind of
terms, not simply tall, I think we'd really have to get into the world of
volume measurement (which we know is pretty tough from the ground without
lots of pricey gadgets) and stand densities.
As far as pigeonholing forests into arbitrary definitions, big voluminous
forests that are remaining aren't the only kind of old-growth, or
necessarily better, just on better sites. The ones we find now are only the
most productive remaining, and who knows how they compare to what once was?
But that is one of the reasons we document maximums- to try to realize more
what might have been, or what could be. Baxter Creek is a cranking
second-growth site with great growing conditions. There are almost surely
other sites that could at least approach it. Some of the 70-80 year old
second-growth tulips in my own back yard are about 130 feet. What will these
forests be like in fifty years, or a few hundred years? The Rucker index is
great for what it is. It would be nice to be able to document different
sites with a more complete index, though I think it would have to be more
than just height and girth, and I think we are limited in what we can
reasonably measure accurately. We can subjectively guess, at least.
Re: Pushing the envelope even farther    Lee E. Frelich
   Nov 17, 2003 06:19 PST 

Bob et al.:

I don't see any problem with separate Rucker height and girth indices based
on different trees. In fact the contrast between the two would be very
interesting.

I am not as convinced about an overall bigness index. I would rather see
several basal area readings from a given stand and a diameter distribution
from a random sample of canopy trees, and also have the heights of those
same trees. Then I would compare sites by modeling the volume, whether a
simple model such as height times girth (ENTS index) or a more complex one.

Lee
RE: Pushing the envelope even farther    Robert Leverett
   Nov 17, 2003 07:40 PST 

Lee:

What is your thinking about using the parabolic volume formula of
(C^2*H)/(4*PI)? I think that is it. What do we gain using it over a
simple formula such as C x H? I admit to feeling confused over what a
simple composite index like C x H really tells us, other than the
existence of an abundance or sparsity of individually large tall and
trees. Maybe extended to wider geographical areas, we get more
information. I don't know. What is your thinking?

Incidentally, I personally feel more comfortable with the individual
indices than with the composites, especially when the former are applied
iteratively so that I can see the depth of height and size classes for
each species much better, especially where the site acreage is not too
large. I really sensed that yesterday in Ice Glen, where I can now
picture in my noodle the different topographical regions of the Glen
area, what's growing in each, and how well the trees are doing,
absolutely and relatively.

   When proposing the adoption of the third index (C x H), I suppose I
was trying to identify and distinguish the extremes from the means.

Bob
RE: Pushing the envelope even farther    Lee E. Frelich
   Nov 17, 2003 08:47 PST 

Bob:

If the purpose is to produce an index of tree size, rather than actual
volume, it doesn't matter much which you use. This is especially true for a
Rucker type index, which involves a rank order to determine which species
to include. Both formulas give the same rank order for the same set of
trees.

I would probably prefer the C x H as an index of tree size, and only use
the other formula if wood volume was the variable of interest. Also,
recognize that the parabolic formula will only be accurate for the average
of many trees, and not for estimating volume of individual trees.

Lee
RE: Mount Peak/MTSF   Lee E. Frelich
  Nov 21, 2003 09:07 PST 

Bob:

I agree that having separate height, girth and combined indices and
contrasting their patterns across the landscape will shed the most light on
tree growth form.

Lee
RE: Mount Peak/MTSF    Will Blozan
   Nov 21, 2003 17:23 PST 

One potential problem with combined indices may be when both old-growth and
second-growth trees exist in the same forest being sampled, as in Baxter
Creek. Ya' think?

Will
RE: Mount Peak/MTSF    Will Blozan
   Nov 21, 2003 17:28 PST 

Is there any sense in discarding the tallest tree in a Rucker Index if it is
an emergent, like white pine, that may tower over all other species,
especially when pine is a dominant species and all others fall well short of
the maximum, such as Pine Flats in Cataloochee? (Wm. Cullum Bryant?).

Just a random thought. The Rucker may be skewed by one dominant, and may not
be the best indicator of the site.

Will
Rucker Indices Examined   dbhg-@comcast.net
  Nov 22, 2003 09:05 PST 

Will:

      You’ve opened the door to an airing of the Rucker indices that we’re using. A full discussion for the benefit of new members is in order.

      From my perspective, we compute Rucker indices for 4 separate reasons:

   A. Historical site documentation
   B. Science
   C. Sport
   D. Politics
   
     Historical Documentation: To fulfill the ENTS historical documentation objective, we seek to inspect forest sites through different lenses. You’ll recall that you and I started concentrating on tree height because that measure was being butchered in the champion tree lists and badly needed rescuing. However, thorough historical documentation of a site demands all the supporting data, so when communicating about a site to others, we wouldn’t publish a Rucker height index without accompanying measurements. We’d want to show the constituent species and their individual contributions. In a full report-oriented site documentation, we’d apply the index iteratively so the influence of isolated tall trees or a single species would clearly stand out. We’d want to show how long each species maintained its influence in the index. Thus, our historical documentations focus on individual trees, individual species, and various combinations to provide different forest perspectives. Consequently, in the ENTS historical mission, it is entirely appropriate to point out the impact of a species on the Rucker index. As an example, for MTSF, including the great white, we get 134.45. Without it, we get 130.0. The biggest lesson here is that Mohawk still earns the title of the state’s tallest forest without the big contribution of the great white. That is very significant and serves to put Mohawk as a growing site into a fuller perspective. Yet, because of the overwhelming presence of the pines in Mohawk, from the historical perspective, people need to recognize the role of the species. So, we can list the Rucker height index with and without the white pine. Examining the role of the pine can be extended to other tall tree sites. For instance, the boost of 4.5 feet to Mohawk’s index, as a consequence of the presence of the great white, can be compared with the pine’s contribution to the Ice Glen index. Take away the white pine from Ice Glen, and the Glen’s index falls by 4.6 feet. Interesting.

   Science: For us, science follows and strengthens the historical documentation mission. Lee, Tom, etc. develop scientific protocols for us to employ as we seek answers to explain the patterns that initial, cursory historical documentations may reveal. The science makes the ultimate use of the indices, but readily modifies them as needed. So, in site analysis, if a super canopy species like white pine unduly biases our understanding of overall site fertility, we may compute the index both with and without the super canopy species. Multi-tiered indices make sense to a fuller understanding. However, in science, we stay cognizant of the limitations of indices. For the ENTS indices applied to forest sites, it also works the other way, i.e. they show up the limitations of other measures applied to sites, such all the total volume measures and CFI plot data, which fail to reveal the full character of places like Mohawk. In the end, the ENTS contributions will provide an in-depth look at sites through a family of lenses that serve science and historical documentation better than what has been used heretofore. I think that was the gist of Lee’s last post.

   Sport: The sport objective is sheer enjoyment - the thrill of competition. We push the index up another tenth of a point and feel a surge of satisfaction from improving a site’s performance vis a vis its competition - all in good fun. Ape calls are for sport. The only requirement in pursuing the sport is that we be absolutely scrupulous in our measurements. No fudging.

   Politics: The political objective calls for the judicious presentation of site data to include Rucker indices for the education of the public as to a site’s tall tree-big tree significance/uniqueness. Each of us must follow his/her own course here to play to the specific audience. In presenting data about a site for political purposes, my approach in Massachusetts is geared to the audience. If it is competition/detractors I seek to persuade, I saturate them with numbers. That is basically what we do with the Mohawk Trail State Forest data. However, whether historical documentation, science, sport, or politics, everything presented must be factual. No exaggerations. The recipients of our data must come to expect and respect absolute honesty.

   Will, you’ve taken the lead in suggesting we need to examine the contributions of specific species and take subsets, e.g. all conifers, all hardwoods, specific species, pairs/triplicates of species, etc. That is, in my judgment, the right direction to take. It is, in fact, the direction that Colby has been going for a long time. His private e-mail to me entitled “associations, associations, associations” said it all. After all, we are trying to explain what we see. In ENTS, we are specializing in seeing forest structure vividly. Every species stands out, but not just for what is happening at eye-level. What happens in 3 dimensions over time is our focus and in the Northeast, we a bit more challenged in the vertical dimension than are the southern, west coast, and tropical rainforests. So, John, John, Gary, Howard, and myself have to re-double our efforts. John, think you’ll nudge Mt. Peak’s Rucker index up a tenth or two over the weekend? Whole point?

    Okay, Will, Lee, Colby, John, Tom, Dale, Gary, etc. your turns.

Bob
Re: RE: Mount Peak/MTSF    Jess Riddle
   Nov 22, 2003 08:29 PST 

What about looking at the median of the height of the trees contributing
to the Rucker index. This system would reduce the effects of an emergent
species and exceptional individuals not characteristic of the species of
the site. However, this approach would highlight the question of why use
ten species, and what is really the significance of the fifth or sixth
tallest species at a site. Just some more random thoughts.

Jess Riddle
Jess's Questions    dbhg-@comcast.net
   Nov 23, 2003 07:48 PST 

Jess:

   Thanks for continuing the thread. Ten species as the basis for the Rucker index is just a convenient number. It allows us to quickly calculate what an extra foot of height will contribute to the overall index ( 0.1 to the index per 1 foot of height). Ten also serves as a reasonable measure of site diversity, or at least as a compromise. However, I'd be the first to caution that we should never become wedded to a single site measure, be it a Rucker index or something else.

   From my perspective, the primary value of going through the process of calculating the Rucker index is to focus our full attention on a site. The utility of the Rucker index approach is that it gives us a view of a site that other approaches traditionally haven't provided. We accumulate knowledge about a site by applying the Rucker height (and now girth indices) iteratively. This allows us to follow the behavior of the index and each contributing species through successive iterations. This, in turn, provides us with detailed knowledge as to where the tall and large trees are either concentrated or grow as isolated individuals. It provides us with a picture of which species grow most often in association with one another. The Rucker index iteratively applied gives us special site and species-based measures of performance that we can then use for a variety of purposes; i.e. as historical documentation, for science, sport, and to make political statements.

   Looked at another way, I like to think of the process of computing Rucker indices as following the top-down method of gathering data, it's only partly scientific, but effiecent for some purposes. We cruise a site and selectively measure trees to capture what the site is currently producing and to get an idea of what it is capable of producing. It is true that we reduce the eventually tree measurements to a single number, however, publishing the site index alone without accompaying site details would be of very limited value to others. By itself, the index tells us only if there are tall/large trees growing on the site at the time of measurement. But when we add the list of species along with associated measures, we provide a more complete profile of a site that reveals it as above average, exceptional, or undistinguished in growing potential. If we stopped at that point, we would be providing a valuable community service.

   If we add topographical and climatic variables, we take the process to the next level. We eventually reach the detail required by the study that Lee Frelich heads and enter the realm of pure science. But time will never permit us to take all our sites to the ultimate detailed level. For the sites that are not included in the impending scientific research, we still provide ourselves and others with a better view of site growing potential and perhaps where each site fits, or should fit, in various conservation schemes.   

   One final point. Applying the indexing process to 100-acre parcel, 100,000-acre regions, and to entire states makes sense only if we don't compare disparate regions under the assumption that they should be comparable. But there are reasons to compare sites of different size. Suppose we compute an index for a small area in the Northeast, say Ice Glen at Stockbridge, MA. The area covered by the current index calculations is about 35 acres. Now, we get 126.1 out of the index. How significant is this? For areas designated as forested sites, it is number 2 in the state. But we can go farther in the comparisons. The entire township of Northampton will eventually produce an index of about 119; maybe 120. Northampton's index currently rests at 118.52. What is one to make of the difference between the small 35-acre Ice Glen site and the entire township of Northamption of many thousands of acres? The difference should point to the superb growing conditions of the Ice Glen area, whatever the mix of environmental conditions that produces those growing conditions, which is for science to understand.

   Does the Rucker index comparison make Ice Glen worthy of extra protection? I certainly think so. Ice Glen is a special place for several reasons. The chief one in the eyes of the public has nothing to do with its forests, but with its rock formations. What has revealed the special nature of the Ice Glen forests? Have basal area or aggregate volume measures done it? Have isolated tree measures done it? Not at all. It has been the judicious application of the Rucker indexing process that confirms that the site will stay above 100 for many iterations and above 115 for 4 or 5 and above 120 for 2. The site is a high performer for New England - none of which was known prior to the Rucker indexing process.

Bob
RE: Mount Peak/MTSF    lef
   Nov 23, 2003 09:32 PST 

Will:

Yes, there should be separate indices for stands with different disturbance
histories and different site qualities. For example, I would like to see
the Rucker index for the dwarf stand on Mount Everett, just to see how
Rucker index varies within MA between Everett and MTSF.

You can always pool stands later for regional indices (but only if maximum
height is the main interest), but you can't divide a regional index back
down if you didn't keep the tree data separate to begin with.

Lee
RE: Rucker Indices Examined    Ed Frank
   Nov 23, 2003 15:04 PST 

Hello,

I have commented on the Rucker Index before. The discussion is on the
website under the category of Measuring Big trees. The Rucker index to
quote Bob paraphrasing Colby: "As Colby explains, one of our objectives
in applying the Rucker index is to facilitate site to site comparisons,
individual species comparisons to assess a site's overall growth
potential. To these ends, height is most valuable, but completeness
requires that we other measures including age and
diameter distributions. For more extensive research, we need
environmental, geographical, and topographical data to include latitude,
longitude, altitude, aspect, precipitation distribution, soil type, bed
rock, etc. In pursuit of a scientific objective, we would not inflate
the role of any single statistic, even charismatic ones like the Rucker
index."

I think a Rucker Girth Index has merit. It is a valid of a measurement
of the growth potential of a site as a height index. Height is clearly
a more impressive index, but... Each method would have its strengths
and weaknesses. It would take calculations on a number of sites to asses
patterns for a girth index, like we are starting to do with height, but
it is worthwhile effort. One advantage of a girth index is that the
data could easily be gathered with equipment costing only a few dollars.
Then people could graduate to buying a clinometer and a laser
rangefinder. (I have tapes, and a perfectly good suuto clinometer, but
haven't got a rangefinder yet).

I am not sure that developing an index that multiplied the largest
diameters by the tallest heights would tell us anything of value. If
you are just looking at bigness there is no need to use fancy parabolic
volume calculation, simple multiplication, or cylindrical volume would
suffice. But I don't see that we gain anything by doing it. What is
the point of generating a set of numbers, just because we can, if they
don't tell us something useful??

So people start generating Rucker Girth Indexes from existing data, or
go for a walk and measure some new ones. Lets see what we have. Just
my thoughts.

Ed Frank
RE: Rucker Indices Examined    dbhg-@comcast.net
   Nov 23, 2003 16:12 PST 

Ed:

Thanks for throwing in with us on the topic. The cir x hgt calculation (for the same tree) may provide clues to differences between what otherwise may appear to be comparable sites. As Will and Lee caution, we need to be increasingly cognizant about what we're comparing with what.

A variable, even if treated in a rank sense, that needs to be added to the mix is age, because we see that on good sites trees gain height rapidly and in 100 to 150 years may reach their maximum potential for the conditions. However, in explaining maximum growth and growth potential on any given site, species also gain height rapidly in response to competition from other species. Put white pine or tulip on a central latitude site and watch everything else grow to try to keep pace. If the sites are subject to severe climatic conditions, they may sacrifice a little in height and put on extra girth. Lee Frelich is leading the pack of us in recognizing that kind of species response. His southern Minnesota cottonwoods are phenomenal growers.

Sites where the middle-aged trees reflect significant cir x hgt dimensions may correlate well to certain latitudes. It still isn't clear to me where the northern red oak develops to its greatest dimensions. However, I think there were once true whoppers in West Virginia. With today's black locust height confirmation on the Mill River, it is clear that the species is quite happy in the Northeast - though apparently not native to 42 degrees and farther north. BTW, where the heck is the center of development for the species? Does anyone know? Is it West Virginia?

In developing the Rucker indices, it isn't that we're exploring altogether new territory. It is more that we are pushing farther in certain directions to understand species response to local site conditions, latitude, longitude, competition, etc. Certainly forestry has much to say about species response to sites, but more in an aggregate volume sense than in a maximum development sense. That's our niche. We are focusing on the performance of individual species under specified conditions. That's rather like following individual baseball player performance as opposed to just focusing on team and league performance. There's plenty of people concentration on volume. We're the individual tree and site people.

Bob


RE: Rucker Indices Examined    Ed Frank
   Nov 23, 2003 20:05 PST 

Bob,

I still disagree. Using separate girth and height indexes allows you to
compare differences within a site. Are the trees at this site taller or
fatter? Either index allows you to compare differences between sites?
Which sites or latitudes produce the taller trees? Which sites produce
the fatter trees?

Lumping the height index and girth index into a single faux volume index
adds nothing to the mix. It lumps clearly distinct values, containing
disparate information together, and serves only to muddle the situation.
You classify, contrast, and compare things by noting and evaluating
differences between them, not by lumping distinguishing characteristics
together.

The strength of the Rucker Index is that it allows a simple comparison
between sites with different tree populations. This lack of
definition is what I perceive to be one of the limitations of the Rucker
index as well...a shorthand that is used represent a number of differing
aspects of a site lumped together into a single number, ...not
distinguishing any single feature (such as age)...but an amalgamation.
I can buy this concept. Adding another index for girth having the same
weakness, then merging them into a third index serves exacerbates this
limitation.   

It can be easily done, it is just multiplication. I may be wrong in the
long run, but I just don't see that doing this adds anything useful to
the mix. I do think it further muddies the waters (by adding a
meaningless number into the stew) of what is trying to be examined.

It appears to me that a combined index is much like listing the length,
width, height, and volume of a series of perfect cubes just because you
can easily generate the numbers. Are these numbers any more useful than
any one of them?

So I guess you might someday show me how wrong I am...but

Ed Frank
RE: Rucker Indices Examined    dbhg-@comcast.net
   Nov 24, 2003 04:37 PST 

Ed:

   You're not necessarily wrong, but remember, the three indices usually concentrate on different trees. The tallest tree of a species will often not be the fattest and the tree that produces the highest volume calculation (cir x hgt) will often not be the tree that is either the tallest or fattest. So the three index method is a way of top-down sampling three sets of often distinct populations. If experience shows that the cir x hgt is clearly redundant of the other information, then its lack of utility will be clear. My belief is that older multi-aged forests will develop three distinct populations of trees. The tallest will be younger (though not young) trees. The fattest will be older trees that have grown in substantial openings. The overall largest will be an interesting mix that reflects some of the fattest, some of the tallest, and some neither. The relative proportions may say more about site history than anything else.

    In any case, I think it is important not to just list an index, but provide an accompanying explanation along with the measurements that make up the index. Regardless, I'm not wedded to the cir x hgt index and if one has to go, that would be it.

Bob
Girth and Combined index    Ed Frank
   Nov 24, 2003 20:27 PST 

ENTS

I received a thoughtful post from Colby concerning the potential value
of a combined index. The gist of the argument was that trees grown in
the open tend to be fatter and shorter. Often taller trees may
represent a second growth forest where the trees are much closer to
together, forcing them to grow tall to reach light while remaining
relatively thin. One possible value of the combined index would be that
trees that are both tall and fat might give a better indication of the
more open nature of a true old growth forest. Colby forwarded a workup
of some of Dale's data that I have not yet had a chance to review. It
is a reasonable argument, but we will need to see if this suggestion is
borne out in the data that is collected.

I am in favor of collecting both girth and height measurements where
possible. Clearly a taller tree is more impressive to look at than a
somewhat shorter tree that is fatter. However, I can't convince myself
that height is a more important indicator growth than girth. If the
data is collected for both height and girth, the combined index can be
calculated and we can see if it has a useful probative value. I am not
opposed to it, I am just not ready to jump on the bandwagon. My comments
should not be misinterpreted to mean that I am violently opposed to the
idea.

As a practical matter I have been thinking about the combined index. If
two trees from different sites vary in height by 20%, if the taller tree
was 150 feet tall, then the shorter would be 120 feet tall. This is a
bif difference. Would a 20% difference in girth be as significant? If
one tree had a cbh of 12 feet, then the smaller by 20% would have a
girth of 10 feet, a difference of 3.83 inches in radius between the
trees. Is this difference as meaningful as the difference in height? Is
it less meaningful or more meaningful? In a combined index both would
be treated equally.

[It was pointed out to me in a private email that I made A MATHEMATICAL
ERROR in the post below. The correct numbers are: "20% less than 12 ft.
is 9.6 ft., a difference of 4.58 in. in radius." I am afraid I was
caught up in the point I was trying to make and committed the cardinal
error of faulty math. I will admit I may not be in the league of
mathematicians such as Heisenberg, Albert Einstein, Stephen Hawking, or
Ted Kazinski that many of you may be. But the idea I was trying to
express remains valid, even though the math was wrong.  

I was using an example of a single tree species compared between two
sites. The Rucker Index or proposed Combined Index would be made up of
ten examples not just one. But whatever relationship would affect a a
comparison between single members of both groups, would also affect the
index as a whole. The combination of many trees in each index may serve
to lessen the impact on these concerns, but would not remove their
impact completely. Anyway I wanted to acknowledge the error in my math.]

If the trees were approaching the maximum height for the species in that
area, the only difference in the calculation of a combined index would
be because of variations in the girth. So the question that comes to
mind is the amount of variance in girth between sites the same, less
than, or more than that of the variance in height? If one parameter's
variation is greater than the others, that parameter will dominate the
combined index. If the variation is much greater, that parameter will
overwhelm the other in a combined index.

I wonder if there is a girth maximum per specific latitude as there
seems to be for height?

Both cbh and height data should be collected. What other data should
be, and can be practically collected to help better describe or
characterize these sites?

If any of you want to play with the girth indexes, height indexes, and
combined indexes, many of Dale Luthinger's reports on the website
contain both cbh and height information. I reconverted his spreadsheet
on Other Pennsylvania Trees back from html to xls and added a link to it
on the website. This will facilitate the manipulation and evaluation of
the data sets he has produced. If you resort the data, note that he has
given listing for multiple measurements of the same tree that will be
messed up in a resort if not combined into a single line beforehand.

Ed Frank

RE: Girth and Combined index    dbhg-@comcast.net
   Nov 25, 2003 04:17 PST 

Randy:

   Two kinds of volumes are involved here - individual tree and stand. Real attempts at calculating individual trunk and limb volume take lots of measurements and long periods of time. Using inadequate formulas, including parabolic volume, don't add anything, since the assumptions they make are seldom fulfilled except as an average for large numbers of trees and then only for well-behaved conifers in tree plantations. We've opted to use circumference x height as a crude, easy measure of individual tree "largeness". But none of us who use the measure are pushing it beyond what we think it may contribute.

   When thinking at the stand level, I often do many calculations of stand density and basal area. However, those measures mean little when dealing with large forest sites, unless we're attempting to calculate aggregate volume per forestry. That has not been an ENTS objective since forestry is trained to do that well and does so.

   It is good that we're having these discussions to clear up confusion about what our original and current purposes are for doing the Rucker indices. In a nutshell, we want to assess:

   1. The maximum growing potential of particular species across their full geographical ranges,

   2. Document exemplary trees wherever they may be located,

   3. Document exemplary tall-big tree sites.

   4. Keep track of record trees for various criteria other than for what the champion tree lists maintain. We're not into competition with the lists.

   Our strong focus on tall tree-big tree sites has always had a preservation motive and we have accordingly needed to convey just how special some of the sites are, on occasion to counter misinformation of indifference on the part of the managing entity. New York state offers a number of examples of lack of support and understanding for what are otherwise exemplary forest sites. Zoar Valley is the prime example.

Bob
RE: Girth and Combined index   John Eichholz
  Nov 26, 2003 07:44 PST 

Bob:

The height times circumference measure is not a volume measure. (Volume
is proportional to the square of the circumference: For a cylinder, v
= hgt*pi*r^2 = [hgt*circ^2/(4*pi) .)

As I see it the circumference, which has the greatest variance, is
scaled by the height, which has lesser variance. So hgt*cbh is kind of
like a (linearly) scaled or weighted version of cbh. For example, a
10.3' cbh by 115'h sugar maple is a different beast than one 4.2' cbh by
115'h. The 4.2' cbh tree would have to be 282' tall to get the same
score. A 9.6' sugar maple only needs 123' to make the same score.

I would look at trees differently if I was trying to boost the h*cbh
score.

John
RE: Girth and Combined index   Robert Leverett
  Nov 26, 2003 10:09 PST 

John:

   Points well made. The use of circumference x height was more of our
attempt to derive a simple overall "bigness" index comparable to the
American Forests formula, but staying within the same units. The concept
of tree size is as much psychological as physical. There's a lot of work
we can do on this subject. Basically, crown spread must enter into the
determination, but trees with big crown spreads usually have large
circumferences, so we viewed circumference as a surrogate for crown
spread. Let's keep talking about on this topic. Just think in terms of
overall tree size (whatever that means) instead of trunk volume and see
where the math takes you.

   
Bob
Measurements vs Interpretations   Robert Leverett
  Nov 26, 2003 10:00 PST 


ENTS:

        Those of us who concentrate on the measuring mission of ENTS are
elated at the growing interest in the subject. However, understanding
the proper application for a measurement can be confusing. We need to be
explicit in the assumptions we make, because on our list, it may sound
like we, the high priests and priestesses of measurement, play pretty
loose with our measurements, but those of us who have been at the
measuring business for years are much more conservative than the ENTS
list banter might make it appear. Here are some examples.

1. Despite all the promotional hoopla, we interpret the Rucker height
index for a site merely as an initial indicator of site growing
conditions. We constantly re-interpreted the index using different sets
of criteria that shed light on where a site fits compared to
topographically and climatologically similar and dissimilar sites. We
may move from a purely site-based calculation to one taking in much more
geographical territory such as a watershed or even an entire state. It
is here that things can get confusing unless we explain ourselves.
Comparisons must be given specific contexts and must make sense or
we're know that going to get clobbered. What, compare the 800-square
mile GSMNP to tiny 35-acre Ice Glen? No, we're not really doing that.
However, there are contexts in which comparison of a smaller site such
as Ice Glen with a substantially larger one is highly relevant. Other
would lead to misinterpretation, creating a confusing apples to oranges
comparison that obfuscates instead of clarifies. Ice Glen's 126.1 and
Mohawk's 134.45 Rucker indices serve to illustrate the maximum amounts
of height that can be stuffed into small, highly productive, mult-aged
sites. The more these small and medium-sized Massachusetts sites are
compared with larger geographical areas, the more outstanding they
appear. Significance emerges from disparate comparisons. The scientific
explanations are quite another matter.

2. We keep each tree measure's limits in mind. If we seem to get sloppy
in our discussions, it is because we know the root assumptions made by
our companion high priest/priestess measurers. The pitfall of our
superficial promotion of a measure in these internet discussions is that
we can appear to be reading more into a value than we actually do.
           
3. When we determine that a measure or a data collection process is
inadequate, we put a high priority on developing a better measure or
measurement methodology. Our finickiness over tree height formulas
should be painfully obvious to all from many recent discussions, but
there are other equally good examples of our focus. Perhaps the best is
the trunk and limb volume modeling that we do periodically. Several
years ago, Will Blozan and Michael Davie repeatedly risked life and limb
in climbing a number of huge Smoky Mountain hemlocks to get precise
circumferential measurements at fixed intervals so we could better model
the volumes of those giant trees. The efforts of Will and Mike were the
first of which I am aware in seriously modeling the Smoky Mountain
old-growth giants and it was through their efforts that we confirmed the
stellar status of the hemlocks in the Great Smoky Mountains as a high
volume eastern conifer. We took the modeling process beyond the
traditional measurement methods that focus on simple calculations to
derives volume in the lower trunk. The profile of the lower trunk of an
old growth hemlock giant might follow a slightly parabolic form from
breast height to some point below the change in shape imparted by major
branching. A parabolic volume calculation for the merchantable part of
the tree might be justified, but not for the whole tree. It seemed silly
to Will, Mike, and I to blindly apply a formula that fails to take into
account the changes in shape that our eys reliably see as we move up the
trunk and into the region of limbs. So we took on the challenge of
refining the trunk-limb volume modeling process and Will and Mike
climbed the Smoky Mountain behemoths taking circumferential measurements
at intervals of a yard or meter - very labor intensive.
           
             4. To make the Rucker height index reflect both site
conditions and capture individually outstanding trees (both objectives
are important), iterations are a necessity. Not a luxury - a necessity.
On sites of limited area and exhibiting fairly uniform growing
conditions, the index iteratively applied is a useful scientific
measure. Lee Frelich has decided that. End of story!

            5. There is value in applying the Rucker index to disparate
geographical areas provided we clearly state our assumptions, reasons
for making the comparisons, and interpret the results intelligently. I
know there is a lot of room for mischief, but that's why we function as
a group. Still, applying the index to unequal areas may go against the
grain because the implicit assumption is that we should only engage in
apples to apples comparisons - an apple tree to an apple tree or an
apple orchard to an equally-sized apple orchard, as it were. But what
if we want to compare the apple tree to the orange? What about orchard
to orchard or multiple orchards to multiple orchards comparisons? Are
there contexts in which such comparisons are useful? That's the
fundamental question. Yes there are reasons, often partly for political,
historical, management-preservation reasons. In these cases pure science
must be separated from the other reasons. However, there are ecological
contexts that may flow from disparate comparisons that direct our
attention to legitimate scientific pursuits. What if a single orchard
outperforms a big region with multiple orchards? Should we take notice?
These kinds of comparisons can be meaningfully made if one's assumptions
and expectations are clearly stated.

         Let me say a word or two about differences of opinion on tree
measuring topics that are bound to rise. In the pursuit of excellence,
it is not only appropriate, but expected, that we challenge each other
on and off this list. But, it ain't personal. It's just business. Still,
I admit to a sensitive region. What happens when people of one level of
education challenge those with a higher level? It can get momentarily
tense. But please remember here that I'm just talking about tree
measuring. If those with the higher education levels haven't been
concentrating on tree measuring the way we have, there's no reason why
they should feel that have to come off as super performers. However, if
push comes to shove, we've got Lee Frelich. He can measure trees as
accurately as any of us and he has a PhD. So beware, all ye out there
in tree measuring land, we're loaded for bear (oops, sorry, Lynn
Rogers).

Bob


Robert T. Leverett
Cofounder, Eastern Native Tree Society