Introduction to Forest Succession
Doug Bashford
� Area: Sci.Environm
���������������������������������������������������������
Msg#: 62 Date: 03-12-94 17:21
From: Alan Mcgowen Read: Yes Replied:
No
To: All Mark:
Subj: Ecocentral 18
������������������������������������������������������������������������������
From: Alan McGowen <al...@igc.apc.org>
ECO CENTRAL 18
Introduction to Succession
The study of succession was initiated in 1863 by the American
naturalist and writer Henry David Thoreau in his essay "The
succession of forest trees". The causes and outcome of succession
have been the subject of much debate in twentieth century
ecology. Here we are not concerned with the underlying mechanisms
of succession nor with details of the more complicated sorts of
climax that can arise. Instead, our focus in the next few posts
is mainly on the simplest kind of succession (to a monoclimax)
and the energetic and nutritive changes which accompany it. These
are changes of ecosystem functions, and lay the foundation for a
discussion of nutrient cycling.
Succession Concepts
*Succession* refers to the fact that over time different
biological communities are found at the same location. *Primary
succession* occurs when a series of communities develop on an
initially abiotic substrate. *Secondary succession* occurs when
the substrate has previously supported a community, and is not
abiotic. In the classic *monoclimax* view of succession, a
primary succession begins with a *pioneer* community which
establishes itself on the abiotic substrate. It proceeds through
a series of intermediate community types to a relatively stable
*climax* community which continues to replace itself; the climax
community is determined largely by climate and the parent rock
type from which the soil has been built in the course of
succession. A complete sequence, from pioneer through climax
communities, is called a *sere*, and the stages are *seral
communities*. Secondary succession occurs when a community later
in the sere (usually the climax) is returned to a state earlier
in the sere by some disturbance. Secondary succession is usually
faster than primary succession because the soil does not have to
be rebuilt (unless of course one effect of the disturbance is to
cause severe soil erosion, returning the conditions to those of
an abiotic substrate).
The monoclimax theory holds that climax is determined principally
by climate. While there is a large amount of evidence for this,
many exceptions can also be found. The *polyclimax* theory holds
that in addition to a climatic climax, other sorts of climax are
possible when ongoing influences prevent convergence on the
climatic climax. Fire, e.g. can lead to a *pyral climax*, the
effects of grazing animals suppressing the invasion of woody
plants and thus maintaining a grassland give a *biotic climax*.
Local variations in soil composition produce various *edaphic
(soil) climaxes*. In the polyclimax view, the environment is a
mosaic of communities at different stages of succession, some of
which reach climatic climax rapidly, others more slowly --
sometimes so slowly that the climate may change in the
intervening time, effectively eliminating the applicability of
the climatic climax concept, and still others which are held back
from climatic climax by other factors which locally dominate the
effects of climate. [Kimmins, 1987]
Still more complex views of climax community composition have
been proposed. In the *climax pattern hypothesis*, the polyclimax
theory's mosaic of communities is replaced by an integrated
collection of communities, in which the abundance of each
individual species varies throughout the whole area as a result
of environmental gradients and the niche requirements of the
individual species. The climax pattern varies in time, either as
a result of fluctuation (in which case we have a *regeneration
pattern*) or, possibly in some cases, cyclically (a *climax
cycle*).
Shifts of Biotic Climax
In the polyclimax and climax pattern views, as well as in
mathematical and simulation models of succession, chance biotic
effects can produce different climaxes at the same location after
disturbance. This sort of effect may be more common in marine
ecosystems, which are dominated by long trophic chains of animals
rather than by a vegetation community: on land, biotic climaxes
are effected principally by herbivores altering the composition
of the plant community, while in the ocean more than one
relatively stable climax trophic web might sometimes develop from
the same primary production base through different outcomes of
competitive exclusions or or other random factors. It is possible
that such varying biotic climaxes after disturbance could be
responsible for the nonrecovery of some fisheries collapses: a
different biotic climax may become established than originally
existed, one in which the commercial species are less abundant.
As an example, the study of scale deposits in anoxic basins
suggests that the California sardine, Sardinops caerulea,
undergoes huge population cycles ranging from 500 to 1700 years
in length. [Nybakken]. During lows of the sardine, its niche is
taken by anchovies. The cause of the cycle is unknown, but it may
be an example of a biotic climax cycle in an animal community. It
is possible that overfishing during the 1930s and 1940s coincided
with or triggered a shift of biotic climax; the rise in anchovy
numbers in recent years suggests that a new climax community is
developing. Perhaps in 500 years or more the sardines will
return. In many cases of fisheries collapse, however, stocks will
return to normal levels much more quickly if exploitation is
reduced, and if other stresses such as pollution or loss of
coastal breeding habitat is not a factor.
It should be stressed that the applicability of the different
theories of succession depends upon the particular ecosystem
considered and the level of detail at which it is considered. For
many systems and purposes, the monoclimax theory is sufficient.
Examples of Succession
A classic study of primary succession was conducted by the
botanist W.S. Cooper (1923) on moraines at Glacier Bay, Alaska.
[Moraines are debris left by retreating glaciers]. As glaciers
retreat, they leave large areas that are colonized by pioneer
species such as mosses and the perennial herb Epilobium
latifolium, growing on a thin layer of soil on top of polished
rock. The next colonists are horsetails and a nitrogen-fixing
herb, Dryas. These are quickly followed by prostrate willows with
thick stems that lie along the ground, anchored to the surface by
roots growing from the stems. The mat formed by Dryas and the
willows traps humus and builds a nitrogen-rich soil.
The prostrate willows are followed by erect willows, which are
joined by alders forming thickets. By this time the soil pH has
dropped from its original value of about 8.0 to about 5.0. Many
herbs and shrubs form the understory in the thickets:
strawberries, bearberries, yarrows, and arnicas. Sitka spruce
begin to "overtop" (grow above) the willows and alders. The
alders die as they are shaded out by the spruce. A solid stand of
spruce has developed after about 170 years. Hemlocks also grow in
the stands and increase in numbers until a spruce-hemlock
community develops.
On strongly sloped land, this seral community remains - it is the
climax. But on flatter land, new kinds of water-absorbant mosses
invade the floor and begin to develop a sphagnum layer, which
slowly becomes thick and waterlogged, depriving the tree roots of
oxygen. The trees begin to die, and a succession occurs in the
sphagnum mosses, with species that had required the shade
provided by the trees giving way to species that are adapted to
living in the open. The forest gradually turns to a bog, some 50%
of which is ponds, and which begins to form peat- a *muskeg*. It
is not known whether the muskeg is a climax which will persist
until further geological changes occur. [Ehrlich, 1987]
A secondary succession occurs after chaparral vegetation is
burned in southern California, and it is much more rapid than the
above example of primary succession. Most plants are dormant in
the dry season. After a fire, some species from the prior
community (esp. chamise and scrub oak) sprout from burned stumps.
Many seeds remain dormant until the fall rains arrive. The result
is a combination of new "invasive" species and seedlings of the
previous dominants. Some of the "invaders" have sprouted from
seeds that have lain dormant in the soil since the last fire, and
can only germinate after a fire.
In low-altitude chaparral the chamise and scrub oak begin to be
shaded out and a mature shrub community develops within 30-50
years. If another fire does not occur, as the mature shrubs die,
sage begins to fill the gaps.
Beyond 60 years the community becomes "senile": species diversity
declines, and lichen grows on the shrubs, rocks, and leaf litter.
Possibly some toxic substances produced by the chamise interfere
with decomposition and germination of grass seedlings. Increasing
amounts of dry, flammable materials build up and eventually
another fire occurs. [This is an example of a possible *climax
cycle*]
Succession also occurs in marine communities, where it is a
sequence of animal communities. Patch reefs form around single
colonies of the coral Montastrea annularis, which begins as
small, rounded head about 1m in diameter. This grows faster at
the top and begins to overhang itself. The undersurface of the
overhang erodes, and is invaded by other organisms, becoming a
patch reef. Eventually the overhangs become so large that they
collapse under their own weight and the effects of wave action.
Colonies of other corals spread over the collapsed platform,
producing an even more complex patch reef. This sequence may take
500-1000 years or longer.
The diversity of reef fishes grows as the colony provides more
refuges and diverse microhabitats resulting from its collapse.
Species that hover in caves during the day (e.g. cardinalfishes)
shift towards species that live in sponges or abandoned worm
tubes (gobies, blennies) as the caves collapse. A patch reef
climax is not very permanent, as the reefs are eventually
destroyed in storms or slowly break up and disappear. [Ehrlich,
1987]
In boulder fields dominated by algae in the low intertidal zone
off California, succession is created by wave action overturning
boulders. Newly exposed bare surfaces are colonized first by
algae: the first pioneer is sea lettuce (Ulva), followed by a
sequence of red algae culminating in Gigartina canaliculata. But
the earlier algae in the sequence are not replaced unless they
are removed by the grazing crab Pachygrapsus crassipes. Earlier
successional species do not "prepare" the way for later ones --
they prevent the later ones from colonizing until they are
removed. The forms that are easily devoured by the crab are
gradually replaced by tougher species -- an example of *biotic
climax*.
Succession in Pacific Northwest Forest
Moisture-laden air from the Pacific Ocean deposits some 70-100
inches of rain annually on the Westside area of the Cascade
ranges of Washington, Oregon, and northern California, making
this region the wettest of the contiguous 48 states. The forests
which grow here are *temperate rainforest*, a rare ecosystem type
found only in a few places on earth. Because of the high moisture
content, fires are less frequent in temperate rainforest than in
other western forests, and as a result the climax seral
communities persist for much longer periods of time on average.
The entire succession of these forests after a fire (or a
clearcut) is unusually lengthy, and can take the better part of a
millennium. Some ancient forest communities in extremely wet
locations are over 3,000 years old; many are more than 1,000
years old.
Within a year after a fire a pioneer community of herbs and
seedlings has sprouted from underground seeds and root systems,
and from seeds transported into the area in dropping of sparrows,
foxes and bears. An extensive fungal community develops in the
dead phytomass, and begins to transfer the nutrients locked up in
wood into the soil. Within 10 years, red alder and Douglas-fir
saplings are present and there is an abundance of forage
available for black-tailed deer because of the large amount of
light reaching the floor: a shrub community has formed. Animal
production peaks sharply during this period, then falls off
dramatically as the second-growth canopy develops. By 50 years,
deer forage has plummeted to almost nothing and the alders are
shaded out by Douglas fir.
By 120 years floor is dark and very quiet -- animal production is
at a minimum. The floor is also very empty of plants. By 250
years, the largest Douglas firs are producing large amounts of
heartwood, many others have fallen and lie decaying on the forest
floor. Hemlock seedlings grow slowly in shaded spots on the
decaying logs, and hemlock saplings are growing more vigorously
in light gaps, also nourished by the logs.
By 500 years, huge broken-topped Douglas firs tower above an
midstory of large western hemlocks, western redcedars, and
smaller understory trees. There is a great deal of downed wood on
the floor, covered with other plants. Flying squirrels, spotted
owls, and Roosevelt elk have moved in. By 1,200 years, the
Douglas fir is gone, and hemlocks and redcedars form a climax
community of great diversity. [Norse]
Refs
Ehrlich, Paul R. and Johnathan Roughgarden _The Science of
Ecology_, MacMillan, New York, 1987.
Kimmins, J. P. _Forest Ecology_, MacMillan, New York, 1987.
Norse, Elliot _Ancient Forests of the Pacific Northwest_, Island
Press, Washington, D.C. copyright 1990 by The Wilderness Society.
Nybakken, James W. _Marine Biology: an Ecological Approach_,
Harper and Row, New York, 1988.
-!-
! Origin: ONE WORLD Usenet<->Fidonet (1:102/129.1)
=================
Date: 03-12-94 17:26
From: Alan Mcgowen Read: Yes Replied:
No
To: All Mark:
Subj: Ecocentral 19
������������������������������������������������������������������������������
From: Alan McGowen <al...@igc.apc.org>
ECO CENTRAL 19
Changes of Ecosystem Function
During Succession
After invading a site, a species can survive only if it is
adapted to, or at least can tolerate, the physical and biotic
conditions there. However, when a species occupies a site, it
inevitably changes it, sometimes in ways which are not favorable
to the continued presence of that species. These changes can
reduce the competitive abilities of the resident species or
increase those of invaders. For example, shade-intolerant species
may create so much shade that their own seedlings cannot develop,
while seedlings of more shade-tolerant invaders can, leading to
their own replacement by the subsequent seral community. Changes
in soil pH accompanying accumulation of litter and development of
a forest floor favors the nutrition of climax species over the
early or midseral species.
In many primary successions such alteration of the environment is
an absolute requirement for later seral stages to develop. In
some cases, however, and especially in secondary succession, it
can be a contributing factor but is not the main driving force of
succession. In the *floristic relay* successional pattern, each
floral community "prepares" the site for the next sere. In the
*initial floristic composition* pattern, many of the species
dominant in later seres are present quite early in succession (at
least as seeds), but may take many years to germinate or to
outcompete the early dominants; in this pattern, no "preparation"
of the site from stage to stage need occur. Both types of pattern
are probably present, to varying degrees, in all successions,
with floristic relay playing a generally greater role in many
primary successions, and initial floristic composition playing an
important and sometimes predominate role in many secondary
successions.
The climatic environments within which succession occurs are
distinguished according to moisture by the terms *xeric* (dry),
*mesic* (intermediate) and *hydric* (wet). Corresponding
successional sequences are called *xeroseres*, *mesoseres*, and
*hydroseres*.
Energetic Changes During Succession
At the beginning of each seral stage, productivity generally
begins to increase until the middle of the stage or somewhat
later, at which it levels off, and then declines somewhat as the
transition to the next stage is approached. Many factors operate
simultaneously to produce this ecosystem-level effect: incomplete
occupancy of the site by invading species, remnants of the
previous community and occupancy of the site by senescent
individuals of the previous community all act to prevent maximum
primary productivity during the early part of a seral stage and
during the transition to a new stage. As the new community fully
occupies the site and develops towards maturity, net productivity
peaks, and then begins to decline as the ratio of
photosynthesizing to respiring biomass decreases, as competition
increases, and as individuals become senescent.
Net productivity during an entire mesosere rises rapidly from the
pioneer community to peak during a midseral stage, and then may
decline somewhat as the climax community develops. The peak of
productivity may occur in a shrub stage, in an early hardwood
tree stage (such as red alder in coastal British Columbia), or in
an early conifer stage (such as Douglas-fir or pine). Later
successional species are selected more strongly for survival than
for rapid growth, so production in later stages may decline.
However, there is more variation in mesoseres than in hydroseres
or xeroseres; in some mesoseres biomass accumulation continues
even in the climax stage as tree stems are incorporated into the
forest floor in a largely undecomposed condition.
In xeroseres gross production, net production, and biomass are
very small during the pioneer lichen stage, increase slightly
during the moss stage, and again in the shrub stage. Increases
continue throughout the conifer stage and peak when convergence
to the climatic climax community occurs; if convergence to the
climatic climax is precluded by one of the means discussed in ECO
CENTRAL 18, production peaks in the terminal stage.
In hydroseres (e.g. lakes) production increases from the
oligotrophic to the eutrophic stages but little biomass
accumulates. Productivity may drop in the transition from aquatic
to wet terrestrial (bog) stages, though some marshes have very
high productivity. From the bog stage productivity may either
increase or decrease depending on nutrient status of the
semiterrestrial stage, and then will increase further as the
terrestrial climax is approached.
An index of the successional maturity of an ecosystem is given by
the ratio of total production P to respiration R, P/R; this
approaches 1 (no net biomass accumulation) as most climaxes are
approached. A high value of P/R indicates an earlier seral
condition. A value less than 1 indicates senescence. However,
considerable variation in P/R is expected within each seral stage
from juvenility to senescence, and there are other exceptions to
the P/R --> 1 rule of thumb. In Pacific Northwest forests, P/R
may exceed 1 in 1,000 year-old stands, largely because of the
very long decomposition times for dead trees. It is possible that
P/R >1 can be sustained for several thousands of years in some of
these increasingly rare communities.
Nutritive Changes During Succession
The energetic patterns discussed above represent nutrients in a
summarized way; other changes occur during succession when
nutrient flows are broken down in more detail. The three main
pathways of nutrient flow in ecosystems are *geochemical
cycling*, *biogeochemical cycling*, and *biochemical cycling*.
1) Geochemical cycle
At the start of the pioneer stage of a primary succession, there
is no net nutrient conservation: addition of nutrients to the
ecosystem in dust, precipitation, seepage, biotic imports,
biological fixation of gases, and weathering of parent rock are
generally balanced by outputs in biotic exports, water and wind
erosion, and soil leachates.
But as succession proceeds, unconsolidated layers of surface
material are occupied by roots and their mycorrhizal associates
(see ECO CENTRAL 17) forming a soil organic layer which greatly
increases the ability of the abiotic environment to conserve
nutrients. Nutrient outputs from the ecosystem are reduced by the
increasing uptake of nutrients and their incorporation into
plant, animal, and microbial biomass. Organic acids produced by
decomposing organic matter, roots, and soil fungi accelerate the
weathering of primary minerals and rocks, increasing the rate of
mineral nutrient input. The rate of nitrogen input also changes
during succession as a result of the activity of nitrogen-fixing
bacteria in symbiotic association with certain seral dominants
such as alder.
During periods of high biomass accumulation, outputs of essential
nutrients are less than inputs. As net biomass accumulation
decreases in later seral stages, nutrient outputs again equal
inputs. When a catastrophe such as a fire devegetates an area and
speeds decomposition, a sudden increase in nutrient output
occurs; if the surrounding ecosystem area is still functioning,
this increase is reversed and a net biomass and nutrient
accumulation begins. This tendency for an ecosystem to respond to
disturbance and subsequent loss of nutrients by a structural
shift to a community which conserves and reconcentrates nutrients
in the area of loss is a particularly important ecosystem-level
function, realized by many different sets of species in different
seres and bioregions.
Some anthropogenic effects, such as acid rain, influence the
geochemical cycles and input/output balance of ecosystems in ways
which are incompletely understood but may be significant. In some
cases, soil acidification depresses mycorrhizal action, leading
to increased nutrient losses by leaching and decreased efficiency
of uptake by plants.
2) Biogeochemical cycle
The biogeochemical cycle is nutrient flow via decomposer chains
back to the soil. The volume of this cycle is controlled by
biomass turnover (litterfall) from plants and leaching losses
from plants. High turnover of relatively nutrient-rich biomass in
pioneer communities such as shrub or hardwood can exceed that in
late successional stages (e.g. conifer), so that in mesoseres the
biogeochemical cycle often peaks in early seral stages. In
xeroseres and hydroseres, the peak is in early climax or later
subclimax stages, since the organic production is higher in these
stages than in earlier ones. At the mature climax stage, a
nutrient "hoarding" adaptation sometimes develops, in which
nutrients flow rapidly from tree --> floor --> tree, with
relatively little nutrients stored long-term in the soil; this is
common in climax moist tropical forest, for example. This may be
a community-level or dominant species adaptation for invasion
resistance, since (plant) invaders find little stored nutrition
in the soil to utilize.
3) Biochemical cycle
Biochemical cycling of nutrients within plants may also change
during succession, with later successional species being more
efficient at internal storage than species of earlier seres. For
example, nutrients can be withdrawn from old leaves before they
are dropped, and this ability should be favored under conditions
of competition for nutrients, as deposition in the litter
potentially makes them available to other species. However,
exceptions to this trend towards increasing internal storage
ability during succession are known, and it is a subject of
ongoing study.
Stability Changes During Succession
A great deal of research effort has been devoted to the stability
of ecosystems. An early hypothesis that increasing diversity
automatically produces increases of stability has not held up,
but a great deal has been learned about ecosystem stability in
the process of disproving it.
The term "stability" is used in several different ways, of which
the following are especially important:
1) Constancy. The lack of change in one or more parameters of an
ecosystem (e.g. number of species, the type of the community, or
some physical features of the environment).
2) Persistence. The length of time over which the ecosystem is
constant or maintains some condition within specified bounds.
3) Inertia. The ability of the ecosystem to remain constant or to
persist in the face of disturbances (e.g. wind, fire, disease,
herbivore outbreaks, etc.)
4) Resilience or elastic stability. The speed with which the
ecosystem returns to its original condition following
disturbance.
5) Amplitude. The extent to which an ecosystem can be changed and
still return to its original condition.
6) Cyclic stability. The ability of an ecosystem to change
through a sequence of conditions that brings it back to the
original condition (e.g. a regeneration complex or climax cycle).
7) Trajectory stability. The tendency of an ecosystem to return
to a single final condition after disturbances of different
kinds or magnitudes. Ecological convergence to a climatic climax
is an example of this.
Constancy and persistence refer to the stability of individual
seral stages. Cyclic and trajectory stability refer to the
overall successional pattern. Inertial, elasticity and amplitude
stability refer to the degree to which successional retrogression
occurs following a disturbance and the subsequent rates of
succession back to the initial condition.
Ecosystems may be stable according to one of these definitions
but not others. For example, tropical forests may have high
inertial stability but low elastic and amplitude stability. Many
temperate forests have lower inertial stability, but higher
elastic and amplitude stability. Ecosystem stability varies
between different kinds of sere, between different seral stages,
and between primary and secondary successions. The inertial
stability of a mesosere is greater than that of a hydrosere or
xerosere (it takes a greater degree of disturbance to produce a
major change). They are also more elastic (they return to climax
more rapidly), and tend to have greater amplitude stability (they
can be pushed farther back in succession before the return to
climax is greatly delayed). Early stages of primary succession
have lower values of several of these stability indices than
early stages of secondary succession do.
There does not seem to be any pattern for later seral stages to
be either more or less stable than earlier ones according to some
of these definitions of stability. A low intensity surface fire
may destroy an early seral stage but have little effect on a
climax forest, or the opposite may be true, depending on the
species involved. A windstorm may destroy a climax forest but
have no effect on a shrub community. The only stability indices
which climax communities often exhibit are constancy of species
composition (by definition of climax) and persistence (except on
time scales long enough for climatic or geographic changes, or
major evolutionary changes). Early seral stages have low constancy
and persistence, a fact that must constantly be struggled against
in agriculture, which utilizes the high productivity of earlier
successional stages. Susceptibility to erosion by wind and flood
is invariably greater in agricultural ecosystems than in the
regional climax ecosystems in all but the driest areas -- a
property they share with many natural early seres.
Much interest in ecosystem stability in the past has been devoted
to stability of species composition. More attention is being paid
now to stability of ecosystem functions. Ecosystems that have a
large live biomass (i.e. large, long-lived organisms) tend to
have high inertial stability. Resilience stability -- the ability
to recover from change -- is often associated with a small
biomass with a high turnover (short-lived organisms). However, a
large capital of organic matter provides a reservoir or energy
and nutrients that drives the heterotrophic and autotrophic
activities which return the ecosystem to its initial state.
Consequently, ecosystems rich in organic matter and nutrients
tend to have both higher inertial stability and higher resilience
stability than ecosystems lacking such reserves. Disturbances
that deplete organic matter and nutrient reserves are
potentially far more disturbing to long-term ecosystem function
than disturbances that merely rearrange organic matter and
-!-
! Origin: ONE WORLD Usenet<->Fidonet (1:102/129.1)
� Area: Sci.Environm
���������������������������������������������������������
Msg#: 64 Date: 03-12-94 17:26
From: Alan Mcgowen Read: Yes Replied:
No
To: All Mark:
Subj: 02:ecocentral 19
������������������������������������������������������������������������������
nutrient pools within the system.
Different components of ecosystems may provide different types of
stability. In forests, an understory of shorter-lived plants can
confer resilience while a long-lived overstory confers inertial
stability. Spatial and genetic diversity of the biomass may also
contribute to the inertial and resilience stability.
Refs.
Paul R. Ehrlich and Jonathan Roughgarden 1987 _The Science of
Ecology_, MacMillan, New York.
Marius Jacobs 1988 _The Tropical Rain Forest_, Springer-Verlag,
New York.
J.P. Kimmins 1987. _Forest Ecology_, MacMillan, New York.
Robert M. May, 1973. _Stability and Complexity in Model
Ecosystems_, Princeton Monographs in Population Biology #6,
Princeton University Press, Princeton, New Jersey.
-!!!!!!!!!!-
Alan McGowen
- If you scratch a cynic,
- you'll find a defeated idealist.