15 Years Ago on SCI.ENVIRONMENT -- how things have changed....remb Alan Mcgowen?
Doug Bashford
I was sniffing around on my disk and found an old text file from sci.env
cryptically called T.TXT, at 213kB.
Here's a typical post:
� Area: Sci.Environm
���������������������������������������������������������
Msg#: 54 Date: 03-12-94 16:48
From: Alan Mcgowen Read: Yes Replied:
No
To: All Mark:
Subj: Ecocentral 11
������������������������������������������������������������������������������
From: Alan McGowen <alanm igc.apc.org>
/* Written 9:49 pm Apr 13, 1992 by alanm in cdp:sci.environmen */
/* -!!!!!!!!- "ECO CENTRAL 11" -!!!!!!!!- */
ECO CENTRAL 11
Tolerance, Limits, and the Niche
Tolerance
Living systems are homeostatic dynamical systems which maintain
themselves in the face of a stochastic universe. As such, they
have regions of stability, within which they survive and outside
of which they perish: so long as the environment does not stray
beyond the range of conditions which the system can tolerate,
homeostasis can be maintained.
We have seen that this is true for a multispecies community of
populations with a codynamics governed by a community matrix A. If
lambda = -max [eigenvalues of A]
and if sigma^2 measures the stochasticity of the environment,
then so long as lambda >> sigma^2 the community will be quite
stable: lambda measures the amount of environmental fluctuation
which this system can tolerate.
The physiology of an organism is also a homeostatic system, with
a range of conditions which it can tolerate. For any measure of
the performance of an organism -- i.e. fitness, foraging
efficiency, survivorship etc. -- and a relevant environmental
variable, there is a range of values of the variable outside
which death or nonperformance ensues (the measure of performance
is zero), but within which performance is possible (the measure
of performance is positive.) When such a measure is graphed as a
function of the variable, it is found to be a bell-shaped curve
within the region of tolerated values of the variable: there is
some optimum value at which performance is at its peak, and
performance falls off as we move away from the optimum, becoming
zero at the limits of the range of tolerance. The *principle of
tolerance* is the observation that organismic responses --
indeed, the responses of all living systems from cells to
ecosystems -- are typified by such tolerance curves. As we have
seen, this very general principle actually arises from the fact
that living systems are homeostatic dynamical systems with
compact regions of stability.
The niche
This leads to a definition of the niche as the hypervolume
enclosed by all an organism's tolerance curves for a complete set
of independent environmental variables: the complete "region of
stability" of the organism in the space of environmental
variables. This is known as the hypervolume, or Hutchinson,
definition of the niche. The niches of individuals forming a
population can be averaged, as can those of an entire species;
often "the niche" under discussion is such an average. In
evolutionary models, mean fitness, averaged over an interbreeding
population, is the measure of performance of greatest interest,
and the niche is defined as the hypervolume of mean fitness
curves over environmental variables.
Limits to growth: density dependence
A resource is something such that when one organism uses it
another organism cannot. By this simple definition, many things
are resources: any oxygen I consume cannot be consumed by you, so
according to this definition, oxygen is a resource. However,
oxygen is usually so plentiful that it is not a scarce, or
*limiting* resource. But it can be limiting, under the right
circumstances, as anyone who has ever maintained an aquarium is
aware.
Limits arise from tolerances coupled with scarcity. Growth often
proceeds until enough of a resource is in use that the remainder
has become scarce: the physiological cost of obtaining an
additional increment of the resource becomes an increasing
function of the number of organisms exploiting it. This is the
condition that brings about density dependence. This can happen
either because the organisms are potentially able to exploit more
of the resource than exists (exploitative competition) or because
they interfere with one another's access to the resource
(interference competition). In either case, as the number of
organisms increases, so does either the physiological cost of
obtaining each increment of the resource, or the mortality rate
(as the probability of obtaining the minimum required increment
of the finite resource decreases). In effect, the resource
increment has an increasing marginal cost per organism added to
the population -- that is the effect of competition.
When the number of organisms is small compared to the total
quantity of each resource they need, the increment of each
resource per organism added to the population has no such
increasing marginal cost, and the rate of intrinsic growth of the
population does not decrease with increasing density; then we
have exponential growth:
dN/dt = r(N)N
r(N) = r0 = constant (exponential growth)
r(N) is a decreasing function of N (density dependence).
As we have seen in earlier ECO CENTRALs, the simplest model of
density dependent growth is the logistic equation, in which r(N)
decreases linearly with N:
r(N) = r0(1 - N/K) (logistic growth)
Logistic growth amounts to assuming a linear increase in the
physiological cost of some limiting resource as the number of
organisms exploiting it rises, but more general functions than
the linear could be used, for example a function r(N) which first
rises with N up to some maximum and then falls: this sort of
growth would typify many social species, which derive fitness
from one another's company up to -- but not beyond -- the point
of overcrowding.
An example of more complex density dependence
Cooperative hunters, e.g. wolves, derive significant advantages
from hunting in groups as the physiological costs of obtaining,
say, one caribou, are shared by the pack (also, the chances of
bringing down a caribou will increase with N). Thus the cost per
pack member per hunt declines (and probability of success
increases) with increasing N up to some optimal pack size, but
net payback decreases above it.
The decrease of net payback is due to inability to coordinate
larger groups, so that a pack of 2N wolves will not in general
kill twice as many caribou as one of N wolves will. Increasing N
then reduces the average physiological cost of the kill somewhat
but also reduces the average payback in quantity of food, and the
balance of the two dictates that there must be an optimal pack
size: the food caloric payback must exceed the caloric cost of
the hunt. The N which maximizes net payback, (payback - cost), is
the optimal pack size. For cooperative hunters, we would expect
r(N) to increase up to this N and decrease above it, to 0 at the
point where (payback - cost) = 0. [Of course, for real
cooperative hunters such as wolves the computation will be
complicated by other factors, e.g. in the social structure of a
wolf pack, food is not shared equally. Also, many smaller prey
are caught and eaten by individuals, so that the food obtained
from hunting in packs is not the whole diet.]
Saturation
The tolerance principle allows us to say that there is some
minimum increment of the limiting resource which is necessary to
the survival of any organism added to the population. There is
also some maximum physiological outlay which can be expended for
the increment: when the physiological cost of the minimum
tolerable increment of resource is itself too great for
tolerance, the limits to growth have been reached, no more
organisms can be added, and the environment is *saturated* with
the organisms. In the logistic population growth model, the N at
which saturation occurs is the carrying capacity K, and moreover
this is a stable equilibrium of the system (if environmental
fluctuation is not too strong). But in other models, the
saturation point need not be a stable equilibrium -- it may be a
maximum from which a decline, or *dieback* commences. Thus in
general a long-term sustainable K need not be the same as the
maximum saturation which is possible for some period of time,
unless we can be sure that the growth is logistic.
The tolerance principle, together with the fact that any organism
has only a finite amount of physiological capital to expend while
obtaining resources, allows us to understand why all growth must
eventually become density dependent -- all growth must eventually
saturate if it proceeds far enough. Indeed, even if no external
resources are limiting, the waste products of the organisms
themselves eventually increase the physiological cost of survival
to the point where saturation occurs. Such "waste limited"
communities occur in wine vats, for example.
More common types of limitation are: food or water, predation,
and space (e.g. nesting sites, space for attachment to the
substrate in littoral communities, territorial limitations, space
providing access to light for plants). Populations can be limited
by more than one factor at the same time, e.g. by both food and
predation.
Some extremes: r and K selection
Some organisms (r-selected organisms) rarely or never reach the
point of saturation before their populations have become extinct.
Such species survive by propagating whole new populations of
themselves with an adequate probability of success even though the
fate of any particular population is rapid extinction. These
species live in an adaptive regime where growth is density-
independent, but mortality so high, and life so short, predators
so numerous and the environment so prone to fatal fluctuations,
that they never come close to saturation. For such species,
success means high r and they are selected for the maximum
reproductive potential.
Long-lived species whose populations persist for long times must
evolve mechanisms for dealing with saturation -- for dealing with
limits. They become more efficient at the use of limiting
resources, for example. This is K-selection: such species are
selected for their ability to cope with scarcity, for their
ability to survive a broad range of adversities, for lower
mortality of offspring. They invest more in survival, and in the
survival of each offspring. Their r is small, so they must use it
carefully. Such species become choosy about mates, and sexual
selection can become important.
Niche overlap, limiting difference, character displacement
If two organisms' niches overlap, the organisms are potential
competitors for limiting resources. In the case where the niches
of two species overlap, if the overlap is large enough, one of
the species may exclude the other.
We must distinguish between the *fundamental niche*, which is set
only by an species' tolerances, and its *realized niche* which is
constrained by competition for scarce resources, predation,
sociality, different ecological roles at different life stages,
or of different sexes, etc. The realized niche is often smaller
than the fundamental niche. This phenomenon, that a niche shrinks
in the presence of competition, is known as *niche compression*.
The converse process, of a niche expanding when competitors or
other limiting factors are removed, is called *ecological
release*.
The minimum distance apart of the centers of two niches which are
able to coexist is called the *limiting difference* of the two
species: if they resembled each other more closely than this, one
would exclude the other. Competitive interactions tend to evolve
so that characters are displaced from each other by at least the
limiting difference: i.e. the competing niches are "pushed apart"
enough to reduce competition to the point where coexistence is
possible. For example, in food limited *guilds* (groups of
species with similar niches that live together) the size of the
food item is frequently the character which is displaced: guilds
of seed-eating birds are commonly marked by a fairly regular
spacing of bill sizes and preferred seed size when the species are
ordered by these characters. In principle, the limiting
separation controls the density with which species can be packed
into the environment, i.e. it controls the maximum possible
species diversity.
-!!!!!!!!!-
Alan McGowen
-- Transfer complete, hit <RETURN> to continue --
-!-
! Origin: ONE WORLD Usenet<->Fidonet (1:102/129.1)
- If you scratch a cynic,
- you'll find a defeated idealist.
Joe
Joe
"Doug Bashford" <pla...@work.edu> wrote in message
news:8--dnTy3da6kkU_X...@pghconnect.com...
Larry Harrell
post about "natural succession" only applies to forests which have no
humans affecting it. Our western forests were mostly shaped by Indian
burning. Unfortunately, the powers that be want to turn our forests
into what they were before man crossed the Bering land bridge. Obama
continues to let the forest disaster roll on like a tsunami washing
across our western forests. Apparently, inaction in the face of
disaster is the Presidential Way. Sadly, Obama has placed public
safety, water and air quality, endangered species, massive erosion,
etc on the back burner in favor
of .........ummmmm......ummmmmm....."free range" fire and forest
destruction??
mhagen
> Doug, what's up with this stuff?
>
> Joe
>snip major text
Looks like somebody regurgitating a few freshman lectures - maybe the
Prof. If he thinks he'll get a better audience here, I sympathize! Really!
(and I know its been too quiet here but don't bait the trolls!)
Mike
Doug Bashford
Joe said
> Doug, what's up with this stuff?
Not much.
--Doug
Larry Harrell
He obviously doesn't want to discuss the current state of our forests,
due to preservationist policies. In fact, discussion about forests
have dropped off to nothing on the National scene and even Obama's
cabinet people, like Tom Vilsack, are being demonized. Vilsack's
speech about the sorry state of our forests has fallen on deaf ears. I
predict we'll see lawsuit against the current Administration over
forest management.
Addressing the habitat issues, what do you think will happen to
endangered species habitat when forests burn to the ground??