Stem Loss - Basic Forestry Question   Edward Frank
  Nov 18, 2006 23:59 PST 

A basic forestry question for you. I can't find a satisfactory answer on the internet. After seedling sprout in an opening. They grow for awhile. Dozens of trees seedlings closely packed. Once they grow so much they are competing with each other and most of the seedlings die - this is the stem exclusion phase? - Afterward they continue to grow but are spaced farther apart. In multi-stemmed trees, if each stem is acting like an individual trees, then why don't the "extra" stems die out like nearby separate seedlings do in the stem exclusion phase? Are they acting as separates trees or not?

Ed Frank
RE: Forestry question    Willard Fell
   Nov 19, 2006 05:39 PST 
Don't know the answer to your question.

But I am sure there is some variance by species. I don't recall ever hearing the term "stem exclusion phase" but even among trees in a single genus (genera?) there can be much difference. For example in the pines, Lob and Longleaf eventually exhibit this expression of dominance in thick stands, while slash will sit and stagnate.

The same is evident with multistem trees. Again using the oaks for example, mature multistem Live Oaks and Sand Post Oaks are the norm, yet it is unusual to see other mature oaks (at least on the SE Coast) with other than a single stem. One would have to assume that other oaks are at least somewhat as susceptable to damage while younger. I think the answer lies in the ability of the surviving branches (or sprouts) to express dominance and succeed as a single stemmed tree.
Re: Forestry question
   Nov 19, 2006 08:28 PST 

I understand what you mean about the complexities of seedling mortality,
natural thinning and overall natural stocking and stem number reduction.

What follows are a few observations I have made..

Seedlings seem to rapidly thin out and a lot of the more palatable ones get
cropped by deer and thinned very quickly. Other factors such as: species,
shade tolerance, opening size, drought tolerance, soil type and site quality can
combine to make a difference as well.

Stump sprouts of most species share one thing in common that deer will crop
them. If there are too many deer the sprouts will get eaten so often that
the stump finally dies.

Where there are not too many deer, the sprouts might get browsed lightly
once or twice with several sprouts getting up to four or five feet tall within a
couple of years.

Within five years most sprouts that are likely to persist will be out of
deer range. Most stump sprouts succumb to shortcomings in how they are located
on the stump (too high up, too close to major areas of developing rot or too
close to damaged roots).

Unless the harvest cut that resulted in the stump sprout was severe enough
to open the canopy so that crown closure is at least a decade away a majority
of stump sprouts will get shaded out and die.

One tree that is different is red maple. Deer browse it like candy but it
usually comes back thicker every time so that eventually a couple of sprouts
get above their reach.

Multiple stemmed red maples usually thin out naturally with incidents of
minor crown damage starting the process. The causes can be as simple as being
the first tree to fork to getting less damage during a cicada hatch.

After that things start getting complex.

Re: Forestry question   Jess Riddle
  Nov 19, 2006 16:10 PST 

Hi Ed,

Intersting question. I don't know the answer either, but I'm feeling
speculative, so I'll take a guess.

The independent seedlings compete not only above ground, for light,
but also below ground, for water and nutrients. On the other hand,
the stump sprouts only compete above ground since they all access the
same root system. They might actually still be competing with each
other for resources from that root system, but the already established
root system should be able to supply more than enough water and
nutrients for their early growth. Sprouts that survive long term
usually originate from different sides of the stump so they are
initially more spread out than seedlings would be. Energy reserves in
the roots will often allow the sprouts to far exceed the growth rates
of adjacent independent seedlings, which also have to compete with
well established root system. Hence, while the sprouts may have to
compete with each other on one side for light, they should have only
limited competition on their exterior sides. Even though the mature
stems from sprouts are in much closer proximity than would be expected
for their size, I doubt they experience as much above ground or below
ground competition as independent trees before they reach maturity.
I've only observed parts of this process, so I'd welcome a more
authoritative answer from anyone else.


Re: Forestry question
  Nov 19, 2006 17:08 PST 

There are several things going on here...Oliver and Larson do a go job
explaning your first scenario in Forest Stand Dynamics. As to the tendency of
a tree/plant to be multistemmed or single dominant stemmed, apical dominance
(the term used to describe the control of the shoot tip over axillary bud
outgrowth) and the role of auxins in controlling lateral bud growth would be
good terms to start your search on.
Good analogy though!
Re: Forestry question   Lee Frelich
  Nov 19, 2006 17:28 PST 

Yes, seedlings in a gap can constitute a stem exclusion phase, which will
last until the gaps made when a tree dies is so large that it can't be
filled by sideways expansion of the crowns of other trees in the same
cohort, and smaller trees get a chance to grow up into the gap. Stem
exclusion within a gap is just a spatially smaller version of stem
exclusion at the stand scale after large-scale stand-levelling disturbance.

The extra stems do die in multi-stemmed trees, some of them die very early
when they are very small, and others continue to die as the clump thins,
and often there are 3-5 stems that are so evenly matched that it takes
50-100 years to see any additional deaths, but they will eventually thin
down to one tree unless a derecho knocks the whole thing down before you
see that phase of development. Self thinning takes several to many decades
whether in a clump of trees of an even-aged stand of small trees.
Therefore, they do act as separate trees, and so do the branches of one
tree high up in the crown, which undergo a similar self-thinning process.

Re: Forestry question   Edward Frank
  Nov 21, 2006 02:23 PST 

When you are looking at an area:   Why do you see many large multitrunked trees, while you don't see large single stemmed trees in such a close proximity to each other? What do the other stems on a multiustemmed tree die from? Is it the same mechanism that kills off the other trees in close proximity to each other, or a combination of other factors such as lightning, and so forth that normally cause tree mortality? It strikes me that we are talking about different processes.

Re: Forestry question   Lee E. Frelich
  Nov 21, 2006 07:56 PST 

I don't think the thinning process is any different, just the spatial
arrangement, with the stump sprouted trees attached closely together.
Single stemmed trees do grow at all distances from each other, including
next to each other, it just doesn't happen as often for them because the
location process (and chance of two trees landing next to each other) is
random, whereas the multi-stemmed trees are located next to each other by a
non-random process, leading to a relatively high number of multi-stemmed
trees, at least for basswood and for many species in second growth forests
where stump sprouts are more common.

It takes many decades for self-thinning to complete itself, people see a
relatively high number of multi-stemmed trees as compared to closely
located single-stemmed trees, and make the assumption that multi-stemmed
trees don't thin themselves, when in fact 50-100 years of patience would
show that they do thin themselves just like independent trees.

Forestry question   Edward Frank
  Dec 13, 2006 18:32 PST 


A few weeks ago I posted a question on the discussion list concerning
the apparent disparity between the self-thinning of seedlings and that
of stems on a multistemmed trees. I received a number of replies from
various people which included observations and comments on the subject.
These observations were valuable, and my thinking has been revised to a
degree on the question, however the overall explanations offered are
still not completely satisfactory on a number of issues. I want to
reformulate the questions and offer some observations and suggestions of
my own as a starting point to continue the discussion. I want to look
at some of the ideas expressed individually before trying to tie things

1) One idea expressed was the idea that since seed dispersal was
random, the frequency of seeds sprouting near to each other, mimicking
root-sprouted multitrunked individuals, would be small. As a general
case this isnít true. First if you have a large area and a single seed
source typically the areas closest to the seed source will receive more
seeds than those areas farther away. So there is a gradient across the
entire range. If there are multiple seed sources, there will be
differences in the seed density in different areas dependant on how the
seed dispersal patterns overlap. But for any specific smaller portion of
that range the idea that seed distribution will be random is a fair
approximation. How close the seeds are to each other is a direct
product of average density of coverage - how many seeds have fallen in a
particular area. The more seeds per area the more likely that seeds
will sprout close to each other. In smaller forest canopy openings, the
ground will often be covered with a virtual carpet of seedlings. Any
adjacent pair of them would be similar in spatial terms to multiple
sprouts from a single root set.   So the statement is not true if there
is a relatively high density of seeds present.

What if there are fewer seeds in a given area? It may not be true
there either. Random distribution means that any particular seed may
land anywhere in the area, whether or not another seed has landed in the
same proximity or not. Many of the patterns people see in graphics, or
art, or similar mediums are called random patterns when they really are
not. If the dots are relatively evenly spaced about the area, this is
not a random pattern, but a distribution about a mean spacing. A true
random pattern has clustering. That means some areas have many dots
close together while other areas have fewer dots spread farther apart.
In areas of the clustering the seeds may be very close together. Other
effects include concentration of the seeds in depressions by gravity,
seeds being washed together by rain and water, and seeds being
concentrated by windbreaks. Seeds sprouting close together may not be
the most frequent pattern, but this pattern is not uncommon either.

2) Consider an area of seeds randomly dispersed across the landscape.
There will be areas of greater spacing and areas of tighter spacing.
These seeds will sprout to form seedlings. How will the seedlings be
thinned as they grow, because it is obvious that every seedling does not
grow to become a full sized mature tree. If the thinning is by random
processes it will have a certain pattern. The loss of seedlings will
follow the pattern of the seedling distribution - any one seedling is as
likely to be thinned as any other - therefore areas with higher density
seedlings will be thinned preferentially over those of lower density
simply because there are more seedling representatives in these areas.
The process will lower the density of seedlings in the clusters making
the effect less prominent, while at the same time the overall average
spacing between all seedling will increase. Could be a localized effect
related to scale? No, it is not. consider a gridded area starting out
with 100 seedling in which 25 were lost to thinning. If the size of the
area was again increased to incorporate 100 seedlings, the same pattern
would be retained. Expanding the size of the grid again added seedlings
to increase the numbers, but these would be equally likely to be added
to any grid section. Therefore loss was preferential to denser areas,
while expansion added seedlings equally to all sections. The overall
pattern of decreasing density of clusters and increased average spacing
is maintained.

It strikes me that in a setting in which the spacing is determined by
competition, then the chance of survival for a particular seedling or
sapling will be dependant on its proximity to a dominant tree. A
particular tree may become dominant because of a very small scale
variation in the amount of water, soil quality, light, initial spacing
of the seed, or perhaps even juxtaposition with poorer quality
seedlings. If the proximity to a dominant specimen is the determining
factor, then the distance between trees ordered by this process should
be a regular spacing related to the size of the dominant trees. Trees
within the sphere of influence of a dominant tree would be at a
disadvantage with respect to trees just outside that area. By this
method the spacing could be spread across the landscape in almost a
domino effect. I donít know if this is the case, but I would anticipate
that the spacing of trees at a given stage is a distribution about a
mean average directly related to the size of the trees involved.

How would this pattern be distinguishable from a random pattern? I
havenít quite figured this out yet. The areas with higher densities
should thin at a faster rate than random thinning, and lower density
areas would thin at a slower rate than random thinning. I am not sure
how exactly they would differ, but I would bet the two patterns would be
obviously distinguishable, and not just something imperceptible but
statistically significant.

My idea is that this thinning process would continue until the trees
reached the canopy level. At this point, I donít think competition will
continue to cause individual trees to die out. They may not be the
biggest or best of the trees in the canopy, but so long as they have a
piece of the light, I donít see competition being the proximate cause of
further thinning. Other factors such as insects, beavers, wind,
lightning, disease, and nearby tree falls would initiate or choose which
trees were the next to die. This process certainly would be enhanced by
competition from nearby trees, but the selection would be because of
these other factors. I see these as being random. There should be a
change in the pattern of thinning once the trees change from competition
to other random factors as being the primary selective character for
further thinning.

Lee talked about competition between branches on a tree as it grew. And
further said this was the same process. I must disagree with this
assessment. From an evolutionary perspective, organisms select for
characteristics that give them a competitive advantage over others,
select against characteristics that are disadvantageous, and for
practical purposes ignore things that are neither beneficial or harmful.
What is beneficial or harmful are more tightly defined under stressful
conditions. Having one branch compete against another does not make any
sense. What advantage is it to an organism to have competition among
parts of itself? The notion is Ö   Lower branches are lost as a tree
grows and are replaced by higher branches. Tightly spaced branches are
thinned to larger more widely spaced branches. But this is not
competition except in the loosest sense. The early branches are
designed by nature to provide energy to the tree at that stage and then
to be shed. There is no use in spending energy on a branch that isnít
going to contribute to the overall survival or reproduction of that
individual. They are designed to be ephemeral. Likewise the tree sends
out numerous branches to find which one provides the most benefit,
gathers the most light, then the others are shed as their portion of the
light collection declines. It is not competition but selection of
certain branches over others to achieve an end. Most of the branches
are not supposed to survive. They are designed to see if this is a good
spot and then to die if it not. The entire tree is not in competition
but part of an effort to optimize the light gathering potential (and
related bio processes) of the entire organism. That is why leaves at
the top are smaller and thinner, and leaves at the bottom of the tree
are larger and thicker. They are not competing, but optimizing with
pieces designed to be ephemeral or expendable.

Multitrunked treesÖ It seems to me that the tree treats these multiple
trunks much like they do other major branches of the tree. They may be
lost over time, but they are not ephemeral or expendable. They are lost
not through the thinning processes of seedlings, but through random
processes that affect adult trees as a group. This idea certainly fits
the suggested model better than the concept that it is the same process
only slower.

Multitrunked trees must grow that way for a reason. Many species sprout
multiple trunks easily when the tree is cut down or otherwise damaged.
Other species form multiple trunks less frequently. Shrubs by many
definitions are woody species with multiple upright trunks. What would
be the ecologic advantage of this process? First I can see that sprouts
from existing root systems could have an advantage in already having a
root system, and immediate access to soil nutrients and water that other
seedlings do not have. The fact that so many species exhibit this
multiple-trunk formation says to me this is a special evolutionary
development that should not be lumped together with single trunked
trees by calling each stem a separate individual. Nor should both stems
be considered as one tree, but I believe as a separate category

Anyway these are some of my ideasÖ

Ed Frank

Re: Forestry question (self competition)   Jess Riddle
  Dec 18, 2006 17:03 PST 


I agree, different parts of an organism competing against each other
initially appears counter-intuitive and maladaptive. However, I can
also see some ways that self-competition might be useful. Also, to
some extent self-competition is an inevitability following from basic
laws of physics, so evolution would select for the individuals that
can make the best use of self-competition.

If one leaf shades another, it makes no difference to the shaded leaf
whether the shade is cast by a leaf on the same plant or from a leaf
on another plant. If that leaf is from the same plant, then the plant
is competing with itself for light. If that is not competition for
light, then how can plants be said to compete with each other for
light? However, this self-shading benefits the plant in that it
provides a mechanism for the plant to produce the most efficient light
gathering form. If the plant could somehow ignore shade cast by
itself, then leaves would persist in sub-optimal positions and consume
resources at a high rate for the amount of energy they capture.

What I see lacking from the argument against the existence of
self-competition is a mechanism that would prevent an individual from
competing with itself. You say branches are "selected" as a tree
grows, but how is that selection accomplished? Each shoot starts out
the same, as a single bud. Even if the species development dictates
that almost all the shoots are eventually lost, the response to
physical resources determines which shoots persist. Any shoot that
maintains a positive energy balance will be maintained. Shoots can
negatively impact each others' energy balance through shading, so
competition does influence which shoots are selected. They may also
compete with each other through the consumption of the limited
supplies of nutrients and water. An actively growing branch may act
as a stronger sink for resources and thus deprive a slower growing
branch of those resources. I don't know enough about plant physiology
to explain how that would happen though. Similar differential
resource use throughout the plant could help the individual ensure
optimal resource utilization by sending the resources to the shoots
and roots that use them most efficiently and quickly.

I still have questions about how that competition between parts would
occur, but I can't simply dismiss it as maladaptive.

Jess Riddle
Re: Forestry question (self competition)   Edward Frank
  Dec 18, 2006 18:31 PST 


Thanks for the thoughtful response. You wrote:

  "If one leaf shades another, it makes no difference to the shaded leaf
whether the shade is cast by a leaf on the same plant or from a leaf on
another plant."

Certainly shading from the same tree or from other trees affect which
branches thrive and which do not. The competition between trees is to see
which tree captures the most light. It does make a difference whether the
shade is from the same tree or another. Trees compete by growing upward and
outward. If shading is from a different tree, then that tree is getting the
energy from the light. The ultimate goal is to capture the most energy and
thereby produce the most seeds. Lower leaves and branches are often lost
because of shading. If a branch is lost by the effects of shading from the
same tree, it isn't competition because there is no net gain in energy
received by the tree from the process, and no net loss in the process. It
doesn't matter to the organism overall which branch the energy is coming
from, so long at it gets it. If it from the same tree, then the energy is
staying within the tree, just not that particular leaf or branch.   It is
the organism as a whole's success or failure in collecting energy that
determines its reproductive success.

I do not disagree with your basic description of the mechanism of the
branch selection process. You wrote:

  "Even if the species development dictates that almost all the shoots are
eventually lost, the response to physical resources determines which shoots
persist. Any shoot that maintains a positive energy balance will be
maintained. Shoots can negatively impact each others' energy balance
through shading, so COMPETITION does influence which shoots are selected.
They may also COMPETE with each other through the consumption of the
limited supplies of nutrients and water. An actively growing branch may
act as a stronger sink for resources and thus deprive a slower growing
branch of those resources. I don't know enough about plant physiology to
explain how that would happen though. Similar differential resource use
throughout the plant could help the individual ensure optimal resource
utilization by sending the resources to the shoots and roots that use them
most efficiently and quickly."

I do disagree with your characterization as competition between branches or
leaves. It is a subtle, but important distinction between the two. The
branches that produce the most energy are the ones that grow, the branches
with negative energy production are the ones that are lost. I would not
know how to test it, but it may be possible that branches with a low, but
still positive energy production may also be lost. It is the purpose of
having so many different buds form so many different branches - to find the
branch that gathers the most energy in the most efficient manner. I think
the different shapes, size, and thickness of leaves at different heights in
a tree are an adaptation to gather the most light. These are found in both
forest grown trees and open grown trees. If it were a case of competition
between the upper branches and lower branches, and somehow this mechanism
were differentially passed on from one generation of tree to another, then I
would think the upper branches would be all that would exist.


Re: Forestry question (self competition)    William Morse
   Dec 19, 2006 05:57 PST 
this is my first post and i may have joined in the middle of a discussion,
but what i have read in this thread is very interesting. Along the lines of
competition, many trees will abort a portion of their fruit, seeds, and
flowers, perform allometric self-thinning, reduce geitonogamy by flower
abscission, and therefore hang onto a select few for investment. It appears
that the fertilized ovules are in competition for resources and their
ability to gain resources will cause abortions/abscissions. Depending on how
you look at the tree, it could be competition or it could simply be
self-organization. I would argue that the adaptive success of a tree is due
to this self-organization, rather than "self-competition". The tree, working
as a unit should not be able to invest energies against itself. As even-aged
stands grow, biomass increases while population decreases. Ecological
energetics would pose a tree against it's neighbors, not itself. Competition
usually involves the allocation of limiting resources to non-reproductive
functions, whereby natural selection is expected to favor mechanisms that
increase competition with non-self neighbors and limit
*wasteful*competition with itself. Mortality preferentially affects
those trees that
have fallen behind in growth (i.e. those trees engaged in wasteful behavior
should be eliminated from the gene pool). This is true for the whole tree.
In fact, root growth generally increases in the presence of roots of a
different plant. Plants will develop more and longer lateral roots towards
neighboring roots of different plants than towards other roots of the same
plant. There is an altruistic phenomena that allows certain plants to
support vegetative offspring during establishment and to share resources
among established ramets in patchy environments (e.g. staghorn sumac,
quaking aspen). These plants are shown to be generous with their own genes,
not selfish (competitive). The end goal of competition is to pass on genetic
info. As stated, self-competition does sound maladaptive and would not be
selected for which leads me to believe there is more to the argument. Trees
interact with their biotic and abiotic environments using a large variety of
often species-specific mechanisms, far beyond the traditional view that
plants interact mainly through resource depletion. Regardless, the argument
creates a great experimental challenge.

Travis Morse

RE: Forestry question (self competition)   Edward Frank
  Dec 26, 2006 19:12 PST 


I had been hoping to get some comments on my ideas about patterning in
the process of stem loss when an area is reseeded. 1) I think the
initial selection of which seedlings first dominate may be related to
random factors related to the availability of water, light, soils etc.
2) After some dominant trees are established, I argued that these
seedlings/saplings controlled the spacing of other young trees in the
area through competition. 3) After a long period of structured
selection the forest reaches a steady state of sorts, where the trees
remaining all have access to at least adequate supplies of water,
nutrients and light. Further stem loss at this point is selected by
random factors, perhaps exacerbated by competition, but selected

I am trying to figure out the place of multistemmed trees in this
process. Initially many stump sprouts are lost, perhaps through their
sprouting in poor positions on the trunk, others perhaps by competition.
Later with fewer to two stems these multiple seem to grow with few
losses due to what could be reasonably described as competition with the
other stems. Loss at the mature stage seems to me to be a result of
random selection factors, rather than competition from other stems.
They also do not seem to be dieing from optimization, self-organization
(whichever is the better terminology) or other factors that occur within
a single stemmed tree.

By no means am I sure about these ideas, but the merit of the idea is
that they could be tested by looking at areas repopulated after forest
fires that took place over a series of different years - the Boundary
Waters Area of Northern Minnesota would be such a place with numerous
overlapping or contiguous fire events taking place over a long history
of time frames with the same general repopulation path.

I am trying also to figure out the distinctions between multiple single
stem trees, multiple stemmed trees (stump or root sprouts typically),
coppices of trees formed by root sprouts, to large scale forests of
cloned trees - like occur in aspens. There is a continuum, but are
there distinct separations between subcategories or not?

Ed Frank

P.S.: I look forward to your participation in our ENTS discussions.
RE: Forestry question (self competition)   Lee E. Frelich
  Dec 27, 2006 08:58 PST 


We have studied the seed rain and seedling establishment process in detail
for white pine in northern MN, and 3 or 4 papers were published by my
former Ph.D. student Martin Dovciak, and M.S. student Scott Weyenberg.

Basically, the seed rain from each mother tree is in the form of a
Weibull-shaped donut (basically a donut with a normal hole in the middle,
but gets thinner and thinner towards the outer edge, rather than a round
outer edge). Seed don't fall close to the tree because of the height at
which they are released to begin their ballistic trajectory, but they start
to fall 5-10 m from the tree, peak 15-20 m away and then gradually tail off
so that 90% have fallen by 50 m away, but a few seeds make 100-150 m (these
distance may be longer in the east because trees there are taller).

If you project these distributions for all the mother trees in the stand,
you get a seed rain density map. Germination is in proportion to seed rain
density, but early survival of the seedlings is not. Survival is heavily
concentrated on certain microsites such as mineral soils, moss beds, or
rotting logs. Since there are more microsites far from the trees (for any
series of concentric donuts, the outer donuts have more area due to their
larger radius), more seedlings survive further from the mother tree than
close to it, although there is a balance between the decreasing number of
seeds and increased surface area of appropriate microsites with distance
from the mother tree, and if mother trees are regularly or randomly spaced
throughout the forest, then all microsites my be flooded with seeds and new
seedlings. In any case seedlings are clustered on good microsites, many of
them are close enough together to eventually fuse with each other if two
seedlings happen to be about the same size and stay the same size until
they are saplings. If they are not the same size, then one will quickly
overtop the other and it will die from shading, so that each clump
undergoes self-thinning resulting a much less dense cluster of saplings.

The saplings then face competition from mother trees, other species of
trees, and tall shrubs such as hazel. If the cluster is in an area not
overtopped by a mother tree, then the seedlings near the edge buffer the
middle one or perhaps a pair, from competition from shrubs, so that
individuals in the middle of a cluster are more likely to move to the pole
size class (4 inches dbh). If under a mother tree or some other large
trees, then they usually die without attaining pole size class.

The surviving pole size trees may then compete with each other or not,
depending on whether the microsites on which they originally germinated
were close enough to each other so that they run into each other. If so,
then self thinning continues and the spatial distribution of trees remains
random, but starts a trend towards a more uniform spatial pattern. By the
time trees are mature, wind storms, lightning strikes and other forms of
density-independent mortality interfere, so that a truly uniform pattern is
never attained. If the original germination pattern was extremely
dispersed, so that the pole sized trees don't compete with each other, then
their spatial pattern continues to be that of the favorable microsites on
which they germinated, and one tree, or a fused pair of trees survives on
each microsite.

For white pines invading an abandoned farm field, things work a little
differently. Along any transect from forest edge to center of the field,
seed rain density declines as described by the same Weibull function as for
in forest trees. However, the density of mother trees will likely vary
along the edge of the field, so that total seed rain will vary among
different transects. In this case, the soils are made somewhat uniform by
having been plowed, and the main deterrent to survival of new seedlings is
the harsh environment (intense sunlight and drying). Therefore, the highest
survival is where the maximum hours of shade from the surrounding forest
overlaps a transect with high seed rain density. Clumps of seedlings
establish in those locations during relatively wet summers (especially
1992, the year after Mount Pinatubo erupted, the 'Class of Pinatubo' white
pine seedlings can be found throughout MN and WI, now reaching pole size).
Those clumps self thin, and attain pole size, extending the shaded region
further out into the field, thus allowing younger seedling to take another
step out into the field. At some point a critical density is reached, and
the whole field suddenly becomes good habitat, and the new forest fills in

Regarding self competition within one tree, I don't think that one branch
has any way of telling whether other branches are from the same tree or
other trees, so I think competition among branches of the same tree is just
as intense as it would be with other trees. It can be called
self-organization rather than competition, but that is semantics. If you
define competition as a process that happens among different individuals,
then there is no competition among branches on a tree. It may make you feel
good to define things that way (in which case go ahead, it won't bother
me), but the thinning process still works the same way whether it is among
branches of one tree or among different trees.


Back to Lee on seed dispersal   Robert Leverett
  Dec 27, 2006 10:32 PST 

Lee, Ed, et al,

   Great discussion. Very interesting. It is fascinating to read of all
the research that has gone into understanding seed dispersal patterns
for different species and the survival response to more subtle factors,
such as the march of white pine seedlings out into a field as a
consequence of making use of shade as they go. Which brings me to a
question. In the case of Stafford Meadow in MTSF, the orientation is
southeast to northwest, with the western side receiving shade from the
bulk of Todd Mountain and a rather sharply defined forest boundary as a
result of mowing every other season. The eastern boundary has a border
of fairly tall pines. Shading would be in the morning. So were Mohawk's
Stafford Meadow left to be re-populated by trees, any guess on the rates
of spread from the opposing sides, other factors being equal? I would
love to pose such questions to groups that accompany me into the meadow.
What factors control repopulation of the meadow by trees, if left alone
to succeed? There are ample seed sources for pine, birch, maple, and ash
surrounding the meadow.   

   A second question is how well does the Weibull function work
(obvioulsy, it is being used) to explain seed dispersal patterns for
white pine?       

Re: Back to Lee on seed dispersal   Lee E. Frelich
  Dec 27, 2006 11:29 PST 


My somewhat educated guess is that a thin fringe of dense white pine
saplings would establish on the east side of the meadow and advance out
into the field as a slow moving wave, while scattered white pine seedlings
and saplings would develop along the west side in a much wider zone, and
then gradually fill in solid. Pine would likely have a numerical advantage
over maple, birch and ash, although a few of those would be in the mix. If
you plowed the field and then let it go, birch would probably be the most
abundant invader, and pine would be able to invade under the birches over a
period of several decades.

Re: Forestry question (self competition)   William Morse
  Dec 29, 2006 04:26 PST 
Hi Ed, Lee,

my thoughts are that the initial selection among seedlings is species
specific rather than limited resources. Initial seedling survival depends on
that species energy provision to the offspring or the ability for a quick
harness of the sun's energy (e.g. beech vs. aspen). The short-term answer
would depend on what species were reseeded. Without active mgt., clear cuts,
clearings, etc. that are reseeded turn into a tangle of "beanpole
trees", which slows down natural succession. I think Lee filled in more than
I could about your 2nd notion (in separate thread). The 3rd idea is
generally what is taught in forest ecology under the "all things equal"
hypothesis (ceteris parabus). However, even in climax or steady state
conditions, structured selection still occurs, though usually in occilations
(e.g. red oak and white pine associates).

Part of Lee's response, in a separate thread, stated that the
self-organization vs. competition argument was primarily semantics. I do not
agree. Each limb of a tree represents an economic probe that reacts to
opportunity (branching, reaching, thickening). The individual limbs cannot
be seperated from the parent, which I think is the heart of the competition
argument. That is, the limbs are being viewed as individuals. Those limbs
encountering shade are given a cost/benefit analysis by the tree and if
there is no gain the tree will reduce or stop investment to those limbs.
This is self-pruning (self-organization). The tree is not competing against
itself. The great author, Bernd Heinreich argued that without self-pruning
trees would not grow tall nor could they survive competition. The tree is
not losing energies by cutting off energy to individual branches, nor is the
tree stealing resources from itself.

Re: Forestry question (self competition)   Lee E. Frelich
  Dec 29, 2006 05:41 PST 

The tree may not be competing with itself when branches are self-pruned,
but branches are competing with each other.