| Tree
Height Patterns |
John
Eichholz |
| Nov
14, 2006 09:57 PST |
Bob, ENTS
Here I am, but it is getting hard to keep up with Dale and Cook
Forest.
How can he and his team come up with a whole point while we
weren't
looking? Great work, Dale.
But seriously, if we are going to produce further improvements
in the
data, we ought to structure it to allow analysis like Lee
Frelich is
proposing.
As our corps of tree measurers are aware, tree heights are
maximized in
a very localized pattern. Overall, a forest can have many primo
sites,
and in the vicinity of any tall tree are often several others of
the
same species and sometimes of different species. Move away from
that
tree a couple hundred meters and everything changes. In some
cases, you
can move a few hundred meters more, into another high growth
area, and
the pattern repeats. We gain a sense for where this will happen,
but
not only are we sometimes wrong, but we don't know enough about
what
makes a site support high tree growth.
One idea I am working with is to consider the local site
conditions at
each area that contributes a tree to the Rucker index. In order
to
support this research, I am delimiting the topo map for MTSF
into
even-sized sections, using the UTM grids on the map. Those are
1km by
1km, there are 15 to 20 of them in the forest, and they could
usefully
be further divided into 250m by 250m quadrants. On a first pass,
most
of our contiguous high growth areas would each fit into one or
two
quadrants. Many quadrants have high growth areas, but fewer
quadrants
have Rucker iteration members in them. An initial project would
be to
map as many trees from the top 25 iterations onto the grid
system as we
can, then look for patterns. We can revisit an area to
supplement other
data we have about a quadrant identified in this search. We can
make
standard characterizations about the quadrant, and perhaps
contrast tree
populations from other quadrants with similar characteristics.
This process of picking areas to go back to is one way I use to
increase
the Rucker index. Right now I am looking at a cluster of data
showing
high concentrations of tall hemlocks in the area of Black
Brook/Cold
River confluence. The tallest hemlocks are located in or next to
streams, within 200' to 300' elevation above the main river. The
tallest trees are located where slopes are 35 degrees or less.
While 40
to 45 degree slopes exist adjacent to these areas, mostly at
higher
elevations, there is a 20' reduction in maximum observed heights
in those
areas. Also present near the tall hemlocks are a significant
population
of really nice Yellow Birch, including most of the known
examples that
exceed 100' in height. Sugar Maple and to a lesser extent,
American
Beech grow quite tall in these areas, but not to record heights.
White
Ash does intermix with hemlock, but outside the hemlock areas it
grows
much taller. Across Cold River but within sight, is the famous
Ash
Flats, with champion White Ash and Bitternut Hickories.
Of the top 25 hemlocks found, 9 occur in this area. 10 more
occur in
the Trout Brook basin, and 4 near the entrance to the park. At
least
80% of the tallest Hemlocks found at MTSF occur south of the
Cold River,
and would probably lie within the boundaries of 9 quadrants (out
of
60-80 total in MTSF.)
Another idea we are working on is time series measurements of
selected
trees, to establish growth rates. We have some time series data
for
MTSF, but it seems like only the beginning.
John
|
| RE:
Tree
Height Patterns |
foresto-@npgcable.com |
| Nov
14, 2006 12:57 PST |
John/Bob-
It just
occurred to me that the one thing in common with the corps of
tree
measurers (excluding climbers) is that the characterizations of
the environment
they occur in, are for the most part ground-based. One of the
more influential
determinants for tree height, I'd think, wouldn't be
ground-based, but "top-
down", i.e., aspect exposures/topographic
exposures/elevation exposures to the
airborne elements (wind events, acid rain/precipitative events,
winter climate
extremes, airborne insect/pathogen exposure, etc.).
Some of these influences fall into the realm of Gary's
GIS...analysis of
topographic shapes common to existing tall tree areas (eg,
northern cove
exposures, lower third of slopes in valleys, etc.), and aspects
(varies by
species, I suspect, could provide at least a coarse filter to
narrow down
searches...?
-Don
|
| RE:
Tree
Height Patterns |
Gary
A. Beluzo |
| Nov
15, 2006 00:51 PST |
John, Bob, Don, and all the ENTS,
Yes, GIS is very well suited for applying multivariate search
models
(terrain, hydrological, soil,etc)to quickly eliminate large
areas that
probably will (and probably won't) support tall trees. The
software that I
run on my Dell Inspiron XPS laptop (ARCGIS 9.2) is capable of
very
sophisticated 3D as well as 2D visualization, animation,
modeling,
geostatistical analysis, geospatial analysis, and---- basically
any analysis
that an ENT can dream up, we can write into a query algorithm,
and run in
ARCGIS. GIS is great for developing derivative map layers,
interpolating
topology, creating masks for map algebra, and superbly presently
the results
to the public.
In terms of doing geoanalysis of TREES- the rate limiting step
of course is
getting good positional data for a population. This would
involve
establishing two or more good GPS benchmarks, determining the
relative X,Y,Z
position of each tree to the benchmarks, and then applying a
simple
algorithm to determine the absolute coordinates (lat/long or
better yet
State Plane Coordinates). Sounds doable but to my knowledge no
one in our
group has taken the time to do it yet.
I met with the College IT folks last week and they have made the
commitment
to upgrade my ENVGIS lab at the College. We are going to
purchase 15 power
PCs, color laser printer, and beefy server. We already have a
full license
to ARCGIS 9.2, a 5-foot large bed inkjet printer, digitizer,
etc. I have a
sabbatical proposal in for the next academic year so I would
have the time.
Bob has suggested mapping the invasives at Robinson State Park
as an initial
project. Once we have the invasives mapped we can do a host of
geospatial
analyses. If snow limits hiking this winter, I'll be looking for
other
project ideas..
Gary
Gary A. Beluzo
Professor of Environmental Science
Science, Engineering, and Mathematics Division
Holyoke Community College
|
|