Biodiversity, Distributions, and Benchmarking    Robert Leverett
   Sep 07, 2002 12:21 PDT 

        Today I spent a couple of hours looking at the forests at the bottom
of Mount Holyoke. I wasn't interested in tall trees. There aren't may of
those, but diversity. Then several questions occurred to me. Given that
species diversity explodes in the tropics and trails off toward the poles,
what is the expected diversity at say 42.5 degrees north latitude and 72.5
degrees west longitude? Yes, it depends on the environment, but what
average and range can we expect? If we work our way north or south, east or
west, how does the diversity of native tree species vary? Somebody has
probably made plots of the numbers, but I've not seen them. One can find a
list of native tree species for say Arcadia Wildlife Sanctuary near the
Oxbow, Skinner State Park on Mt Holyoke, and Mt Tom State Reservation, but
to my knowledge, nobody has put it together in a big picture kind of way for
native tree species. If we randomly choose a spot in central New Jersey, how
would the average and range compare with the location cited above? Can a
diversity of healthy native tree species act as a measure of overall system
health similar to using a large carnivore like the grizzly bear as a
surrogate for hundreds of other species that will survive if the grizzly
survives? How should a benchmarking system work using native tree species?

Re: Distributions and Benchmarking   Lee E. Frelich
  Sep 09, 2002 06:54 PDT 

I don't know of any good studies of tree diversity in eastern North
America. There are lots of anecdotal reports.

We do know that there is a gradient towards fewer trees as the regional
environment becomes harsher. For example, as one goes west from MA, there
are more droughts, more heat waves, more cold spells, and more severe
storms. As one goes north, it simply gets colder. Each species is
individually distributed and has unique sensitivity to extremes. The more
extremes in any one climatic variable there are, the fewer species can deal
with them. That is why species drop out gradually as one moves west. For
example if you take a transect from Green Bay, WI to the priairie forest
border in MN, beech, black spruce, white spruce, balsam fir, hemlock,
yellow birch, white cedar and white ash, successively drop out of the forest.

Certainly, the continued substantial presence of trees native to a region
is one indicator of healthy forests, but I doubt that they can be used to
assess ecosystem health in the same way as the presence of grizzly bears.
Grizzly bears require prey and trees don't.


Re: Distributions and Benchmarking   Neil Pederson
  Sep 09, 2002 08:06 PDT 

Dear ENTS:

There is one decent paper looking at tree species richness in North
America that I know of. This reference is:

Currie, D.J. and V. Paquin. 1987. Large-scale biogeographical
patterns of species richness in trees. Nature 329: 326-327.

It shows the high richness in the Smokies region, of course.
Curiously, it looks like the highest richness looks to be in the
north-central South Carolina area. It is hard to tell since there are
no state boundaries drawn on the figure. And who knows what kind of
error there is on the boundary. Nonetheless, it looks pretty good.
The richness in this contour is said to be 180 species per quadrat. A
quadrat in this study is equal to 2.5 degrees by 2.5 degrees.

For New York and New England, the 140 tree species/quadrat isoline
that cuts through SW NYS and SW New England. Not too bad for so far

In preparing to "sell" the geographic location for my dissertation
to my committee, I calculated the number of important tree species in
the Hudson Valley, NY. Important species here is determined by Elbert
Little's Atlas of US Trees: Vol. 1 Conifers and Important Hardwoods.
Its a decent place to start.

What I actually calculated was the number of eastern US species
found in the Hudson Valley. I defined an E US species as any tree
species found E of the Mississippi River. Arbitrary, but again,
somewhere to start.

I found that 59.7% of E US species can be found in the Hudson
Valley. The number would be higher if I didn't included many palm
trees and other subtropical species found in S FLA.

I would hazard a guess that one might find a similar result
repeating this analysis for the big river valleys of New England and
much of S New England outside of the pitch pine oak forest in eastern

Climate, geography and likely soils all interact and help determine
geographical tree diversity. The big river valleys in the NE allow
warm maritime climate to forge it ways north. Also, precipitation
increases as one moves south from Canada, and like Lee describes, as
one moves east. It may be that this combination of climate supports
increased tree diversity so far north.


RE: Distributions and Benchmarking   Heidi Roddis
  Sep 10, 2002 11:17 PDT 
I wonder whether land use history might have more influence on forest composition in the Northeast than factors such as latitude, soils/geology, or slope aspect. Tree species diversity alone might not reveal as much as diversity of woodland wildflowers, ferns, lichens, and fungi. And non-native species should be catalogued separately and not considered as contributing to an estimate of species diversity except that a large number of nonnatives would be a negative indicator.

Beyond simple species diversity estimates, other measures are also important, such as relative dominance or abundance of various species. For example, some woodland wildflowers and ferns can be found in many second-growth forests yet are far more abundant and form more robust stands in old growth forests where the soils have never been significantly disturbed by people.

Also, consider that you could have a small stand with 20 species of trees in it but if most individuals are all one species and there are only one or two individuals of each of the other species, how does this compare with another stand with the same 20 species each represented in more equal numbers across the site?

Soils, aspect, and moisture can greatly influence plant diversity. There are many rare plants that live only in specialized environments such as acidic peat bogs or the opposite: calcareous areas like some parts of the Berkshires. Some plants have very wide ranges of tolerance for a whole variety of factors including soils, moisture, temperature, etc (white pine is a great example and may be why it has the largest geographic range of any tree species in the eastern US; red maple is another example of a broadly tolerant and adaptable species), whereas other species require a very narrow set of site characteristics. Sites at exactly the same latitude may have greatly varying capacity to support certain species.

Massachusetts has a high amount of biological diversity overall because of its position at the juncture of several distinct "Ecoregions" as defined by the U.S. EPA and The Nature Conservancy, as well as a wide variety of habitats from mountain tops to coastal plains and beaches. Conditions at the same latitude can be very different from the bottom of a valley to the top of a mountain. Consider the Connecticut Valley where several species reach the northerly limit of their range in MA, and are not found naturally elsewhere in the state.

Measuring diversity is a tricky business.

RE: Distributions and Benchmarking    Neil Pederson
   Sep 10, 2002 11:59 PDT 

Forest composition is one thing. Species distributions may be
something different. There is no doubt land-use history has altered
forest composition in the NE US. That is one of the legs that the
Harvard Forest is built upon. However, I'm not sure the same thing
can be said for species distributions at the larger scales. Climate,
latitude, soils/geology, and slope may still be quite important.

Three examples: using varved sediment cores (very high resolution
versus normal sediment coring) Konrad Gajewski has shown that there
was a synchronous change in spruce pollen over the last 1000 years
from Wisconsin to Maine despite distinctly different disturbance
regimes in these regions. He strongly argues for climate being the
important driver rather than disturbance.

I believe the paper on witness trees that Charlie Cogbill is lead
author on coming out this fall will have something to add to species
distributions. In his talk at the NE Natural History meeting last
spring, his results suggest to me that the pre-settlement regional
ecotone between oak/hickory and the northern hardwood forest from
eastern NY State through Maine looks very similar to the one mapped
out by Braun, and the many other forest type maps.

Emily Russell has published a study of sediment core tops in the NE
US covering the period from 1500 to present. She draws a similar
ecotone between the oak/hickory and the northern hardwood forest.

No doubt land-use altered forest composition and species dominance.
But, the last 2 studies suggest that 400 years of European settlement
patterns may not have significantly altered distributions or, at
least, the ecotone between forest types.

There is no clear answer right now. It would take many careful
studies at several scales in an attempt to get to an answer.

Distributions and Benchmarking    Leverett, Robert
   Sep 10, 2002 12:27 PDT 

What I started out to explore (before I sidetracked myself) was the idea of tree species diversity in the East as a function of latitude, longitude, and altitude as surrogates for climate. Lee and now Neil have addressed the topic. Unfortunately, I got off onto a tangent with the forest health debate, or maybe I should say debacle, now that it is a political football.

But since the health issue was broached, a question that immediately occurs to me is: are trees the best organisms to reflect incremental degradation occurring within our environment? Since most tree species are capable of living for several centuries, in them do we not have organisms that are veritable storehouses of annual environmental data? How many kinds of stories do the tree rings tell? This is hardly a new idea and I would imagine that members of our list can site plenty of papers on the use of trees as environmental monitors apart from their disappearance due to introduced diseases. One sour note is the use of trees as environmental barometers regrettably stays wrapped up in economic issues. The widespread use of trees and forests to monitor overall environmental health has to contend with our society's use it or lose it mentality. Uh, oh, I'd better stop here.

Re: Distributions and Benchmarking    Martin, Bill
   Sep 10, 2002 12:29 PDT 
Bob and others,
Carl Monk's paper predates Currie and Paquin (1987) by 20 years and has long
been cited when trends in diversity are discussed:   Monk, C.D. 1967. Tree
species diversity in the eastern deciduous forest with particular reference
to north-central Florida.   American Naturalist 101:173-187.     Using the
Shannon index, Monk established diversity values of 3.00 in the southern
Appalachians as the highest values with values of 1.7-2.0 in the Great Lakes
region and 2.2 or less in your neck of the woods. Since these are log
values, the difference between 2.0 and 3.0 is considerable.

This business of diversity (as in indices) and richness (no. of species ) as
being measures of "ecosystem health" has several pitfalls.     Several years
ago, ecologists (of note) argued that the measure of species diversity
expressed as a single number was THE measure of stability, resilience,
health of ecosystems.   Hundreds of papers were published (mine included) on
measures of species diverstiy in different ecosystems.    Various measures
were used and debated; the clear distinction was drawn between diversity and
richness; benchmarks such as Monk's values were compared.     All of this
hustle,bustle, and navel comtemplation produced nothing of note because (1)
there was wide disagreement on what diversity really means; (2) diversity
does not necessarily mean stability,etc.; (3) which parts of the biota
really represent "diversity"--plants, animals, ??? ; (4) which index should
be used, which is the best, which is the worst, etc.,etc.    Further, some
ecosystems with so-called "low" diversity such as salt marshes and
coniferous forests are (appear?) "stable" and "healthy".   Soooo, after
awhile we apparently decided that diversity was just another quantitative
measure of communities and ecosystems and that the number of species of a
biological group also represents a measure of diversity.     We still tell
the public that diversity means something and I gladly plead guilty to being
a chief promoter of the GSM and the southern Appalachians as THE center of
diversity in the temperate deciduous forests of the eastern U.S., if not the
world.   Are these beautiful, wonderful, diverse forests " healthy" ?    I
love Einstein's pithy comment that "not all that is counted counts, and not
all that counts can be counted."
Bill Martin
RE: Distributions and Benchmarking    Heidi Roddis
   Sep 10, 2002 13:46 PDT 
Yes, an interesting and complex subject connected to other issues. I've seen reports predicting shifts in forest composition due to global climate change. Tree ring analysis has many applications including climate change study.

This is a much better explanation than mine of the difficulties and pitfalls of diversity measurements as a surrogate for natural community health assessment.


Re: Distributions and Benchmarking   Martin, Bill
  Sep 10, 2002 14:15 PDT 
Bob and all,

Ecosystem "health" is not easy to define.    It is not a good scientific
term, but it does help communicate with non-scientists about the condition
of a system , just as human health measures tell about a condition of the
human body.      Certainly evaluating ecosystem health is not easy, because
we usually have not determined what should be measured and monitored.
Some good products have been produced to illustrate the "health" of aquatic
systems, but I have yet to see good illustrations for terrestrial
systems.    Bob, you rightly ask what should be measured (remember
Einstein's admonition in my last email) in forests.     It is my view that,
in general, trees are not going to give us a good , short-term (our lifetime
or career) measure of forest health.   Unlike a thermometer measuring human
temperature, they do not quickly respond to environmental changes.
When one considers the long evolutionary road they have traveled and the
insults of their past,   you can be impressed with the survival capabilities
of these big, woody plants.    They can take a lot of abuse, including
those sensitive to pollutants and change (e.g., white pine).   No, I think
we have to look elsewhere in the forest to the smaller members for their
responses.   Lichens, mosses, select herbs, salamanders, soil organisms
come to mind as better "thermometer" candidates .
Bill Martin