Forest Ecology, Pages 41-51, Timothy J Fahey.pdf

42



FOREST ECOLOGY



competitive advantage of being tall. For example, in

the Great Plains of North America grass fires were usually too frequent to allow trees to become established

and outcompete the grasses for sunlight.

Ecologists have classified the forests and other vegetation of the world into biological regions, or biomes,

on the basis of the physiognomy (outward form) and

species composition of the dominant plants. The biome

classification represents ecological patterns at a very

large scale. However, the patterns that ecologists endeavor to explain occur at a wide range of spatial scales,

from global to local. Global scale maps of the world’s

biomes have been constructed to illustrate the broad

distributional patterns of the biota. Ecological patterns

at smaller scales are represented by subdividing the

various biomes to provide more detailed classifications

and maps. In essence, the forest ecologist regards the

global ecosystem as consisting of a nested series of

ecological associations, at smaller and smaller spatial

scales, down to the level of the relatively uniform forest stand.

The focus of ecological classification systems on the

dominant plants is not just a convenience; the dominant

plants in terrestrial ecosystems (e.g., the trees in the

forest) play a crucial role in setting the stage for all the

other organisms because they convert solar energy into

food for the food web and they form the three-dimensional structure that constitutes the habitat of most

associated biota. Hence, it is of primary importance

in studying the ecology of forests to understand what

controls the distribution and abundance of trees. However, this is not to say that the other organisms are not

important to the functioning of the ecosystem; in fact,

many of the less prominent organisms play important

roles in regulating the distribution and abundance of

the trees.

Underlying these ecological classification systems

and maps at the broad scale is the principle that climate

is the environmental feature that exerts primary control

over the distribution of organisms on the earth’s surface.

This principle was realized long ago by European geographers who observed during their explorations the

coincidence between the broad patterns of climate and

vegetation physiognomy on earth. The key components

of climate, temperature and precipitation, exert this

control through their effects on the growth of plants.

Different plant traits, expressed in part as whole plant

physiognomy, prove to be most suitable for growth and

survival under different combinations of temperature

and precipitation. Hence, climatic patterns largely coincide at the broad scale with the patterns of distribution

of forest biomes. Expressed most simply, the effect of



climate on vegetation distribution can be plotted in

terms of mean annual temperature and precipitation

(Fig. 1); forest vegetation is restricted to relatively moist

climates. However, annual averages cannot account for

the important effects of climatic seasonality—variations

in temperature and precipitation through the year—in

determining biome distributions. More complicated

systems of expressing the influence of climate account

for the important effects of seasonal drought and of

subfreezing temperatures on plant growth and activity.

The principle that climate exerts primary control

over the distribution of organisms fails to account for

the observation that some equivalent climates support

vegetation with differing composition and physiognomy. For example, the temperate rain forests of the

Pacific Northwest of North America are dominated by

evergreen needleleaf trees whereas in some equivalent

climates of the southern hemisphere (New Zealand and

South America) evergreen broadleaf forests may dominate. These differences reflect the fact that the flora and

fauna that have developed by biological evolution in

the various regions of the world are quite distinct because they have been geographically isolated from one

another. Several biogeographic zones have been identified by botanists and zoologists, and these zones reflect

geologic history, especially continental drift. For exam-



FIGURE 1 The affects of mean annual precipitation and temperature

on vegetation. (Modified from Whittaker, 1975, with permission.)



FOREST ECOLOGY



ple, plant geographers typically recognize four plant

domains whose floras are distinct from one another:

(a) paleotropical (Old World Tropics), (b) neotropical

(New World Tropics), (c) north temperate, and (d)

south temperate. To expand on our earlier principle,

then, climate together with the available flora and fauna

exert primary control over the distribution and abundance of organisms on earth. This principle points toward the likelihood that human effects on climate and

the introduction of exotic species across the globe will

have profound consequences for ecology and biodiversity.

At a more localized scale, species distribution and

abundance are also affected by secondary environmental factors, topography and soils. Topography exerts

its influence in part by locally modifying climate; for

example, in the northern hemisphere south-facing

slopes are warmer and drier than north-facing slopes.

Also, topography influences the environment through

the action of gravity moving matter downhill (e.g., water, soil particles). Soil supplies the essential environmental resources, water and mineral nutrients, for tree

growth. Soil properties reflect the combined effects of

geologic, climatic, and biologic forces, emphasizing the

complex web of cause and effect that underlies ecological patterns and processes.

Geologists have divided the earth’s landscape into

physiographic provinces that reflect the effects of geological processes on the earth’s surface features. Within

each physiographic province, relatively orderly and recurring patterns of topography and soils are observed

that differ fundamentally from those of neighboring

provinces. For example, in the coastal plain province

of the eastern United States, sandy sediments have

emerged from the receding ocean to leave a gently rolling landscape of porous, sandy soils. Forest patterns

reflect subtle variations in topography and drainage.

In contrast, directly to the west, the ridge-and-valley

province is marked by recurring combinations of bold

narrow ridges of resistant rocks and intervening valleys

underlain by softer substrates like limestone and shale.

The forest patterns in this province directly mirror the

striking gradients in environment that result from the

combination of topography and soils determined by the

bedrock geology of the province. An understanding

of the ecology of these two regions begins with the

recognition of the underlying differences in geological

forces that have shaped their physiography.

One additional factor that strongly regulates the

composition of the biotic community is the legacy of

disturbance events. Disturbance is a natural phenomenon in all ecosystems and is defined as any event that



43



results in a change in environmental conditions and

resource availability, usually as a consequence of death

of the dominant plants. Among the most prominent

natural disturbance agents in forest ecosystems are fire,

windstorms, and irruptions of pests and pathogens.

When disturbances occur at a spatial scale that is much

larger than the area occupied by individual dominant

plants (i.e., large-scale disturbances), they so profoundly alter the environment that the suite of plants

that subsequently colonizes the disturbed site may be

quite different from the original community. These

large-scale disturbances initiate the ecological process

of succession—that is, successive changes in the composition of the biotic community occur as a result of

progressive changes in environmental factors. Hence,

the actual composition of the biotic community at any

particular time and place depends not only on climate,

flora and fauna, topography and soils, but also on the

time interval since the last large-scale disturbance that

initiated successional change—as well as the nature and

intensity of that disturbance event. That disturbance has

played a major role in shaping ecological patterns and

processes is evident from the many traits of the flora

and fauna that reflect the selective force associated

with disturbance.

The natural disturbance regime that characterizes

any particular ecosystem depends on exogenous factors

that act as disturbance agents (e.g., the combination of

drought, lightning, and wind that favors the occurrence

of fire; or frequent exposure to hurricanes or tornadoes

in certain geographic regions) as well as endogenous

factors, such as the traits of the plants themselves, that

influence the frequency or intensity of the disturbance.

For example, pine forests are more prone to fire disturbance than deciduous broadleaf forests in part because

the fuels produced by pine trees are more flammable.

And forests on sandy soils are more prone to fire because

coarse soils dry out more rapidly than fine-textured

soils. Thus, the influences of disturbances, environment, and biota may be mutually reinforcing in shaping

ecological patterns and processes.

The prevalence of disturbance and succession in natural forest ecosystems complicates the task of classifying and mapping forest distributions, particularly at

smaller scales, because forest composition is continually

changing. Recognizing this problem, ecologists have

defined the climax forest association as the assemblage

of species that would persist under any particular combination of environmental factors in the absence of

large-scale disturbance. In many regions, however, the

recurrence of large-scale disturbances is naturally so

frequent that climax forest associations rarely develop,



44



FOREST ECOLOGY



and the practical value of such a strict climax concept

is somewhat limited. However, in some regions where

large-scale disturbances are infrequent, particularly under humid climates (where fires are rare) and areas not

often exposed to intensive windstorms (i.e., the humid

tropics and some temperate areas), climax forest associations probably were common prior to the advent of

anthropogenic influences. Natural disturbance regimes

in these forests consisted of small-scale events resulting

from the death of individuals or small groups of trees

in a patchwork mosaic. Although the composition and

structure of various patches would differ, at a larger

scale the average composition of the forest ecosystem

would remain relatively steady (i.e., it would exhibit a

shifting-mosaic steady-state). These observations illustrate the importance of the scale of observation to our

understanding of ecological patterns and processes.



II. BIODIVERSITY IN

FOREST ECOSYSTEMS

A. General Patterns

The number of species observed in any particular forest

varies markedly across the earth’s forest biomes. Biodiversity can be conceived as consisting of three distinct

elements, termed the gamma, alpha, and beta diversity.

The total diversity in a large area, the gamma diversity,

can be partitioned into two components, the local

(alpha) diversity in a single habitat or forest stand,

and the turnover of species between stands, the beta

diversity. High gamma diversity could be associated

either with high alpha diversity, high beta diversity, or

both. Ideally, considerations of the patterns of biodiversity across forest regions need to account for the contributions of these different elements of biodiversity. Unfortunately, studies of forest biodiversity patterns

generally have not provided samples of sufficient detail

to resolve these elements, and more systematic approaches are needed. Nevertheless, many valuable insights into the patterns and causes of biodiversity variation across forest regions have emerged from research

to date.

The most striking pattern of forest biodiversity is

the latitudinal gradient: much higher biodiversity is

observed both locally and regionally in tropical than in

temperate forests. Because observations are most comprehensive for tree species, the following description

focuses on these taxa.

The general relationship between latitude and the

alpha diversity of trees is illustrated by a plot of the



number of species in 0.1 ha samples taken from around

the world (Fig. 2). At one extreme are lowland tropical

rain forests in the upper Amazon basin where on average every second tree in a sample belongs to a different

species; at the other extreme, a variety of natural, monospecific forests are common in many temperate and

boreal regions. Within the tropical region considerable

variation in alpha diversity has been observed (Fig. 2).

The two most prominent features correlated with this

variation are annual precipitation and biogeographic

province or region. For example, the average alpha

diversity (number of tree species in 0.1 ha samples) of

neotropical lowland wet and moist forests is 152 species, whereas for seasonally dry forests the average is

65. And the alpha diversity of trees in Africa is generally

much lower than for eastern Asia and America (Fig.

3). The extremely high alpha diversity of lowland tropical rain forest is not paired with equally high beta diversity. For example, the 307 tree species identified on a

single 1-ha plot in the Ecuadorian Amazon constituted

about 16% of the total tree flora (trees Ն 5 cm diameter)

of Amazonian Ecuador and a single 50-ha plot at Pasoh,

Malaysia contained 830 species, 20 to 30% of the total

tree flora of this country.

By comparison, tree species diversity in the temperate latitudes is much lower: the enormous area of temperate zone forests harbors only 1166 tree species, not

much more than the 50 ha plot at Pasoh! As for the

tropics, however, striking geographic differences in the

diversity of tree taxa are observed in the temperate zone.

The highest gamma diversity is in east-central Asia,



FIGURE 2 Species richness of 0.1 ha samples of lowland (Ͻ1000

m) forest as a function of latitude. Dashed line separates dry forest

(bottom) from moist and wet forest (top) with intermediate sites

(moist forest physiognomy despite relatively strong dry season) indicated by alternate lines. X, the anomalous Coloso and Loma de los

Colorados sites in northern Colombia. Reprinted, with permission,

from Gentry (1995).



FOREST ECOLOGY



45



with intermediate levels in eastern North America and

lowest in Europe (Table I).

A comprehensive explanation of these global patterns in tree species diversity remains elusive, but a

growing consensus on the role of historical or biogeographic factors and local physical habitat factors is

emerging. The overall lower diversity of temperate than

tropical regions probably is explained in part by the

physiological constraints on colonization imposed by

the need to tolerate subfreezing conditions and by the

smaller contiguous area in temperate regions. Regional

differences between biogeographic provinces in the

temperate zone appear to owe in large part to higher

extinction rates in Europe and North America during

the Pleistocene glacial epochs as well as greater access

of the east Asian region to dispersal routes from the

tropics. Similarly, the combination of geographic area

and climatic differences probably explains the contrasts

in diversity between tropical Africa versus America and

east Asia and between wet and dry forests. As noted by

Latham and Ricklefs (1993), ‘‘further resolution of the

causes of diversity patterns will require new paleontological, biogeographical and taxonomic data and synthesis.’’ (p. 310)



B. Forest Structure and Pattern and

Disturbance Regimes



FIGURE 3 Number of species shown among trees of at least 0.1 m

in diameter on small plots in tropical lowland rain forest. ᭹, America;

x, Eastern tropics; ᭿, Africa. Lines connect sample plots that lie close

together. Reprinted, with permission, from Whitmore (1998).



The local biodiversity in particular forests depends on

the complex suite of factors that characterize the habitats of individual species. These factors include such

components as the species composition, phenological

timing, structural complexity, and horizontal patterning of the vegetation, which in turn depend on

environment and the legacy of disturbances. The dominant plants (i.e., trees in forests) play a pivotal role

in defining the habitats for associated organisms by

providing food and shelter and by regulating the local

microenvironment. For some organisms (e.g., many insects) host-specific interactions with particular dominant plants result in strong correlations in their distributions, so that forest composition is the key factor

influencing the composition of associated species. For

other organisms, the structural and horizontal patterning of the dominant vegetation may be more important than composition alone.

The importance of the three-dimensional spatial arrangement of the branches and leaves of the plants in

defining animal habitats was realized by pioneers in

the study of ecology and evolution. MacArthur and

MacArthur (1961) demonstrated that the diversity of

birds in forests of the eastern United States could be



46



FOREST ECOLOGY

TABLE I

Summary by Taxonomic Level and Region of Moist Temperate Forest Trees in the Northern Hernisphere

Number of tree taxa characteristic of moist temperate forests in:



Taxonomic level

Subclasses

Orders

Families

Genera

Species

Families excluding those of predominantly tropical

distribution(% of total)

Genera excluding those of predominantly tropical

distribution (% of total)

Species exclusive of predominantly tropical genera

(% of total)



Northern,

central, and

eastern

Europe

5

16

21

43

124

18

(86%)

41

(95%)

122

(98%)



East-central

Asia

9

37

67

177

729

37

(55%)

121

(68%)

570

(78%)



Pacific

slope of

North America



Eastern

North America



Northern

Hemisphere

(total)



6

14



9

26



10

39



19

37

68

18

(95%)

35

(95%)

66

(97%)



46

90

253

29

(63%)

77

(86%)

236

(93%)



74

213

1,166

41

(55%)

149

(70%)

987

(85%)



Data from Latham and Ricklefs, 1993.



predicted by the structural complexity of the vegetation

(Fig. 4): in habitats with high foliage-height diversity

(FHD; defined by the formula FHD ϭ %i pi ln pi, where

pi is the proportion of total foliage area in the ith layer),

bird diversity was much higher than in habitats with low

FHD. Moreover, this relationship largely transcended

differences in plant species diversity. Similarly, for

ground-dwelling organisms, the complexity of habitat

at the soil surface—including vegetation cover, litter,



FIGURE 4 Relationship between foliage height and bird species diversity in areas of eastern North American deciduous forest. (Modified

from MacArthur and MacArthur, 1961.)



rocks, fallen logs, and moisture—is strongly correlated

with local biodiversity. By extension the belowground

structural complexity might influence diversity of subterranean organisms, but little work on this topic has

been accomplished.

The horizontal pattern of forest vegetation, particularly the arrangement of relatively uniform forest stand

units across the landscape, influences biodiversity of

the region and is the subject of a newly emerging branch

of ecology—landscape ecology. The edges or ecotones

(zones of rapid change in plant species composition)

between forest stand units or forest associations often

provide qualitatively different habitat than the interiors

of those units; hence, the size, shape, and spatial arrangement of forest stands in the landscape influence

the populations of associated species. Edges and sharp

ecotones between stands may arise because of environmental discontinuities (e.g., topographic depressions

where soil water accumulates) or because of the legacy

of past disturbances. Thus, the local biodiversity of

forests reflects both the compositional and structural

diversity of the plants as well as the arrangement of

units of relatively uniform composition and structure

that we define as forest stands.

The structure of any particular forest stand traditionally has been defined as the distribution among age or

size classes of the trees in a forest stand. Even-aged

stands arise as the result of large-scale disturbances in

which all or most of the large trees in an area are

killed by a natural disturbance agent (e.g., crown fire,

hurricane, fungal pathogen) or by human activity. The



FOREST ECOLOGY



forest that arises on such a site consists of trees of

roughly the same age, from new seedlings that colonize

the site or from advanced regeneration (preexisting

seedlings and saplings that escaped the disturbance).

The process of stand development following large-scale

disturbance results in gradual changes in the structure

of the forest, as the growth, mortality, and recruitment

of new individuals proceeds. These changes in forest

structure include not only age structure, but also the

spatial arrangement of the stems, branches, foliage, and

roots of the plants that define the habitat of other biota

in the forest. Moreover, changes in species composition

of the forest usually accompany stand development, as

species capable of growing in the shaded understory

replace the pioneer species that colonize the disturbed site.

A typical sequence of forest stand development following large-scale disturbance has been characterized

by Oliver and Larson (1990) as consisting of four stages:

(stage 1) stand initiation, (stage 2) stem exclusion,

(stage 3) understory reinitiation, and (stage 4) old

growth. In the stand initiation stage, trees colonize the

disturbed area. The time interval of this stage varies

markedly depending on the severity of the disturbance,

environmental factors at the site (e.g., climate and

soils), and often herbivory. This stage concludes when

the forest canopy becomes closed or when some soil

resource (often water) becomes limiting to further increases in the leaf area of the forest. During the stem

exclusion stage resources like light and soil water are

so limited that suppressed understory trees die and

regeneration is severely restricted. In this stage there is

usually a continual reduction in the density (number

of stems/area) of the original cohort of trees. The structure of the canopy is exceptionally simple at this stage

as the individual trees grow in height to co-opt the light

resource from neighbors.

In the understory reinitiation stage the overstory

begins to break up as canopy trees die and the differential height growth of various species or individuals results in more complex arrangements of the foliage. Increased light reaching the understory favors the

establishment and growth of new cohorts of species

most capable of surviving in the highly competitive

understory environment. The old growth or late successional forest stage is attained as overstory trees age

and the canopy develops even greater complexity of

structure. Gaps form in the canopy as a result of injury

or death of the large, mature individuals, and the previously suppressed individuals are released from severe

competition and grow in height. Decaying, coarse,

woody debris accumulates on the ground and dead tree

snags also provide new habitats for animals. Obviously,



47



this idealized model of forest stand dynamics exhibits

myriad local variations depending on the nature of the

disturbance, environmental factors, and the tree species

that dominate the area.

The changes in forest structure (e.g., foliage-height

distributions) that accompany stand development following large-scale disturbance result in consequent

gradual shifts in the quality of the habitat for different

animals and plants. In the example cited earlier, MacArthur and MacArthur (1961) observed that maintenance

of high numbers of bird species in eastern deciduous

forests of the United States depended on the adequate

provision of three layers of foliage, corresponding

roughly to ground vegetation (0–2 ft), shrubs and small

trees (2–25 ft), and overstory trees (Ͼ25 ft). If nearly

all the foliage of the forest is in just one layer, as in

the stand initiation and stem exclusion stages of stand

development, bird diversity is much lower than in the

old-growth stage, when canopy stratification becomes

prominent.

The nature and degree of vertical stratification differ

among forests. For example, temperate deciduous forests in the mature stage often exhibit the three strata

just identified: an overstory stratum occupied by the

canopy trees, an intermediate stratum represented by

the crowns of saplings and understory species like dogwoods and hornbeams, and a ground stratum of low

shrubs and forbs. In contrast, in the lowland tropical

rain forest, very tall ‘‘emergent’’ trees with broad shallow

crowns overtop the main canopy, which may be subdivided into two or more additional strata above the understory layers. And in the boreal forest the low stature

and conical crowns of the conifers often preclude the

formation of strong vertical stratification. These structural differences contribute to the contrasts in the number of distinct habitats provided for associated biota.

As noted earlier, many forest regions are only rarely

affected by large-scale disturbances because they are

both too moist to carry fires and not subjected to catastrophic windstorms. In these regions the steady-state

forest is characterized as a shifting-mosaic landscape of

small patches of different sizes and shapes, each patch

reflecting the legacy of disturbance caused by the death

of individuals or small groups of overstory trees. Each

of the patches may follow a sequence of development

analogous to that outlined for large-scale disturbances,

but the overall structure of the forest is dependent on

the arrangement of the tapestry of patches that comprise

the larger forest stand. The edges between these patches

and the vertical distribution of structural elements represent important dimensions of the habitat variability

that permits species coexistence in forests.

Recognizing the importance of forest structure and



48



FOREST ECOLOGY



pattern for biodiversity, ecologists are developing new,

more sophisticated approaches for quantifying these

parameters. Because the structural features that are important in determining animal and plant habitats differ

among taxonomic groups and forest types, it is unlikely

that any single approach will provide a universal standard by which forest structure and biodiversity can be

related. Current efforts are utilizing new tools in the

areas of spatial statistics, computer modeling, and remote sensing to provide suitable protocols for evaluating the connections between forest structure and pattern, management activities, and biodiversity of various

groups of biota. These efforts will provide a better basis

for understanding how forest ecology and biodiversity

are related.



C. Food Webs and Community

Organization in Forests

A food web is a set of species that live together and a

specification of which species ‘‘eat’’ which other species.

Plants provide the base of the food web by converting

solar energy into biomass; herbivores and detritivores

utilize living and dead biomass, respectively, to build

their own tissues; and these organisms are in turn consumed by predatory species. Although a connection

between the structure of food webs and biodiversity

seems axiomatic, the exact nature of this connection

is extremely complex. The biodiversity of a particular

community is expressed in three different elements of

food web structure: (a) the food chain length—the

number of trophic links in the food web (producer1Њconsumer-2Њconsumer-3Њconsumer etc.); (b) the

number of distinct trophospecies—the set of all species

that share some particular set of predators and prey—

within a trophic level; and (c) the diversity of species

constituting each trophospecies.

In forests, as distinct from most other biomes, the

first link in the food web is dominated energetically by

the detritivores rather than the herbivores. That is, most

of the biomass in forests is consumed after the plant

tissues die and are added to the soil as detritus. Thus,

the detrital food web dominates the energetics of forest

ecosystems and much of the complexity of the detrital

food web in forests remains to be explored. These energetic considerations are not translated in a simple way

to biodiversity in the respective food webs. The variety

of different food sources available to herbivores in forests (leaves, stems, fruits, seeds, flowers, and roots of

different plant species) provide numerous niches for

their diversification, whereas most of the biomass energy available to detritivores is in the form of woody



tissues which are structurally and biochemically so similar from species to species of trees that the diversification of wood decay organisms is somewhat limited. For

example, one taxonomic group of insects, the higher

termites (family Termitidae), overwhelmingly dominates in the comminution and decomposition of plant

biomass in many tropical and warm temperate forest

biomes. Nevertheless, in comparison with herbivoredominated biomes (e.g., grasslands, aquatic ecosystems), the role of detritivores in forest biodiversity is

probably relatively high.

The interactions of species in the forest community

are not entirely competitive and predator-prey in nature. Some of the most fascinating interactions provide

selective benefits to both of the interacting individuals

or species populations—mutualistic and symbiotic relationships. Mutualism refers generally to a relationship

in which two interacting species enhance their survival,

growth, or reproduction, while symbiosis refers more

specifically to two such organisms living together in

close association. The most important symbiotic mutualism in forests is the mycorrhiza, an association between the mycelia of fungi and the roots of trees. The

tree roots provide a supply of food to the fungus, which

in turn increases the capability of the plant to acquire

soil nutrients and water. All forest trees are mycorrhizal,

and each tree species may harbor dozens of different

fungal species in this mutually beneficial relationship.

Two distinct types of mycorrhizae are common in forest

trees—the ectomycorrhizae and endomycorrhizae.

These types differ taxonomically, anatomically, and

physiologically. The possible role of these mycorrhizal

associations in regulating the diversity of forests was

pointed out by Connell and Lowman (1989). They observed that pockets of low-diversity forest are found

within the matrix of high-diversity tropical rain forest

in all tropical regions. These low-diversity forests are

composed of trees with ectomycorrhizal associations,

whereas most of trees in the high-diversity forest are

endomycorrhizal. Functional differences between the

mycorrhizal types in soil nutrient acquisition or transfer

could maintain or reinforce competitive interactions

between individuals in these distinct forest types.

Most mutualisms between plants and animals have

developed around the successful completion of the reproductive cycle of the plant—pollination and seed

dispersal. Although many trees simply disperse their

pollen to the wind, this method of pollination is unreliable when individuals are widely scattered in the forest,

as in the species-rich tropical rain forest. Insects, nectivorous birds, and bats visit the plants to exploit them

as a source of food and in the process carry pollen from



49



FOREST ECOLOGY



one flower to another. Similarly, plants with seeds too

heavy to be dispersed by the wind rely on animals to

carry the seeds away from the mother plant. Although

in most cases these mutualistic interactions are nonobligate and facultative (at least on one side), many remarkable examples of highly intricate, obligatory interactions have evolved, especially in the tropics. These

interactions promote specialization and increased biodiversity.



III. HUMAN ACTIVITY AND

FOREST BIODIVERSITY

Few forests have escaped the effects of human activity.

Ancient civilizations decimated forests locally as a

source of fuel and fiber and as sites for intensive agricultural production. In the modern era the pervasive influence of industrial civilization on forests has expanded

to the regional and global scale through the additional

effects of species introductions, air pollution, and likely

climatic change. Insights from forest ecology provide

a basis to evaluate the implications of these human

influences on biodiversity.



A. Introduction of Alien Species

Alien insects and pathogens typically wreak havoc on

host trees because these hosts have not developed adequate defenses through the process of evolution. The

result is widespread decline of the host trees throughout

their range even to near the point of extinction. The

consequences of great reductions in the abundance of

such declining species for the wider biotic community

are not well understood and undoubtedly vary depending on the characteristics of the declining species

(discussed later). Similarly, introductions of other species in different taxonomic groups or at other positions

in the food web will have consequences for forest biodiversity that depend on their particular role in the

community. Invasive trees may displace congeneric species from the forest community; and introduced herbivores that lack natural controls on their populations

may decimate populations of their favored food plant

species. Because the pace of species introductions has

increased very rapidly in recent years, the ultimate consequence for forest ecosystems and biodiversity will be

played out over the coming century.



B. Forest Harvest

The consequences of tree harvesting for forest biodiversity depend on particular features of the forest and the



methods of harvest. Because all forests are regularly

subjected to disturbance, if forest harvest practices

mimic the natural disturbance regime, then consequences for biodiversity should be minimal. However,

the exigencies of the financial bottom line result in

harvest practices that do not mimic natural disturbances, and the consequences of actual forest harvest

practices for biodiversity may be substantial. Most serious are (a) logging practices that result in the failure

of the cut-over site to regenerate (e.g., because of severe

damage to soils); (b) the coincident harvest of extensive

areas, so that most of the landscape is in a single stage

of forest stand development; and (c) recurring harvest

on short rotation intervals. Also, forest harvest differs

fundamentally from natural disturbance in that wood

products are removed from the site; any species that

depends on decaying wood for its habitat will be harmed

by harvest practices that do not recognize this dependency. Finally, some species are believed to be oldgrowth obligates (i.e., they depend on old forests to

complete their life cycles). These species are threatened

when the great majority of natural, old forests in a

region enters the harvest-regrowth system of industrial

forestry, leaving little old-growth habitat.



C. Forest Conversion and Fragmentation

Permanent or semipermanent conversion of forested

areas to other land uses has more severe consequences

for biodiversity than forest harvest. Many forests occur

where climate and soils are suitable for permanent agriculture and where the expansion of urban communities

gobbles up native vegetation. Maintenance of biodiversity in such regions depends on having protected forest

areas large enough to harbor the native flora and fauna.

However, general rules to guide forest preserve planning for biodiversity protection are complicated by variations in the habitat requirements of different species.

Forests in most agricultural regions occur as small fragments dispersed across the landscape, and how effectively these fragments can maintain viable populations

of forest species is a topic of great concern (Schelhas

and Greenberg, 1996).



D. Pollution

Local declines of forests has been associated conclusively with point-source releases of air pollutants, especially sulfur dioxide, fluoride, and toxic metals from

smelters. Broad-scale, regional effects of air pollution

on forests have been more difficult to demonstrate.

Regional pollution—by ozone smog in the southwest-



50



FOREST ECOLOGY



ern United States and by acidic deposition in the eastern

United States and Europe—probably has contributed

to documented forest declines. Although improvements

in emission controls and regulations in these regions

are likely to reduce the chances of further damage, rapid

industrialization without adequate emission controls

in other regions of the world threatens forest health

and biodiversity.



E. Rapid Climatic Change

The Pleistocene epoch was marked by dramatic climatic

shifts that profoundly affected forests and biodiversity.

The rapid rise in greenhouse gas concentrations is likely

to bring about similar climatic shifts in coming decades

or centuries. Of course, the consequences of rapid climate change for forests and biodiversity will depend

on a combination of species’ natural responses (e.g.,

dispersal, colonization, natural selection) and human

mitigation efforts. In many forest regions certain species, like the dominant trees and particular wildlife

populations, are likely to be controlled by management

efforts because of their relatively high value to humans.

For relatively low-valued species and forest regions, the

maintenance of biodiversity may depend on natural

mechanisms or heroic human efforts that recognize

nonmarket values of species. The current level of understanding of the physiological and population ecology

of many forest-dwelling taxa is insufficient to predict

the effects of rapid climatic change, but species with

limited capacity for dispersal and colonization (e.g., soil

invertebrates, perennial herbs) may be most sensitive.

The implication of the loss of these species from forest

communities will vary depending on the role they play

in forest ecosystem function.



IV. BIODIVERSITY AND FOREST

ECOSYSTEM FUNCTION

Forests regulate energy flow and cycling of materials

in the landscape, collectively known as ecosystem functions. The effects of biodiversity, expressed in terms of

species richness, on forest ecosystem functions are not

yet clear and apparently not very straightforward. There

is great interest in these possible effects as ecologists

probe the implications of loss of species diversity for

the integrity of ecosystems. Will species extinctions

result in destabilization of ecosystem functions and possible feedbacks in the form of undesirable shifts in dominant vegetation types?



Whereas the role of species richness per se in regulating forest ecosystem functions remains unclear, it is

well known that loss of particular species from the biota

of a community can have important ramifications for

energy flow, material cycling, and maintenance of stable

biotic composition. That is, all species in the ecosystem

are not equal in terms of their quantitative influence

on ecosystem function. In particular, for some species

there appears to be little or no redundancy with respect

to their role in the ecosystem; if such a species performs

some crucial activity, its loss from a forest can create

havoc for the normal functioning of the ecosystem.

These species are known as keystones.

In forests keystone species are represented among

many different taxonomic groups or food web positions.

For example, at the primary producer level, a nitrogenfixing tree like Alnus rubra in N-poor conifer forests is

a keystone species; elephants appear to be keystone

herbivores in semiarid Africa; beavers are keystone

‘‘ecosystem engineers’’ in northern forests; and jaguars

that prey on seed predators in neotropical forests may

be keystone carnivores. The loss of these species has

consequences for ecosystem structure, function, and

composition that are out of proportion from their individual abundances. Important efforts to conserve forest

ecosystem functions in the face of biodiversity loss are

focused on the identification of keystone species and

ways of maintaining stable populations of keystone

species.

Although the broader effects of overall reductions

in biodiversity on forest ecosystem function remain

more obscure, a specific example will illustrate that this

is also a cause for concern. During the past several

decades, excessive inputs of nitrogen (from air pollution) to forests in northern Europe have resulted in

striking reductions in the abundance of mushroom species. These mushrooms are the fruiting bodies of ectomycorrhizal fungi discussed earlier. Lilleskov et al.

(2000) have shown that a single host tree, white spruce

(Picea glauca), maintains associations with about 90

different mycorrhizal fungi in natural, N-poor forests

in Alaska, whereas in adjacent N-polluted forests only

about five fungal associates are found. The high diversity of this mycoflora in the natural forest certainly

represents some degree of functional redundancy. However, if dozens of species are lost from the mycorrhizal

fungal flora in temporarily N-polluted regions, some

long-term effects on forest ecosystem function are

likely. The fungi in the N-rich forests appear to utilize

only the mineralized nitrogen sources (NH4ϩ and

NO3Ϫ), which are abundant there; with a return to normal, low–mineral N availability, in the absence of



FOREST ECOLOGY



mycorrhiza that access organic N forms, the productivity and nutrient cycling in the forests could be altered

profoundly. Analogous situations probably apply in

other aspects of both the detrital and grazing food webs

of forest ecosystems around the world.



See Also the Following Articles

BOREAL FOREST ECOSYSTEMS • DEFORESTATION AND LAND

CLEARING • DISTURBANCE, MECHANISMS OF • FIRES,

ECOLOGICAL EFFECTS OF • FOOD WEBS • KEYSTONE

SPECIES • LOGGED FORESTS • RAINFOREST LOSS AND

CHANGE • REFORESTATION • SUCCESSION,

PHENOMENON OF



Bibliography

Barnes, B. V., Zak, D. R., Denton, S. R., and Spurr, S. H. (1998).

Forest Ecology, 4th ed. John Wiley & Sons, New York.



51



Connell, J. H., and Lowman, M. D. (1989). Low diversity tropical

rainforests: Some possible mechanisms for their existence. American Naturalist 134, 88–119.

Gentry, A. H. (1995). Diversity and floristic composition of neotropical dry forests. In Seasonally Dry Tropical Forests (S. H. Bullock,

H. A. Mooney, and E. Medina, eds.), pp. 146–194. Cambridge

University Press, Cambridge, England.

Latham, R. E., and Ricklefs, R. E. (1993). Continental comparisons

of temperate-zone tree species diversity. In Species Diversity in

Ecological Communities (R. E. Ricklefs and D. Schluter, ed.), pp.

294–314. University of Chicago Press, Chicago.

Lilleskov, E., Fahey, T. J., and Lovett, G. M. (2000). Ectomycorrhizal

fungal community change over an atmospheric nitrogen deposition gradient in Alaska. Ecological Applications, in press.

MacArthur, R. H., and MacArthur, J. W. (1961). On bird species

diversity. Ecology 42, 594–598.

Oliver, C. D., and Larson, B. C. (1990). Forest Stand Dynamics.

McGraw Hill, New York.

Schelhas, J., and Greenberg, R. (1996). Forest Patches in Tropical

Landscapes. Island Press: Washington, DC.

Whitmore, T. C. (1998). An Introduction to Tropical Rain Forests, 2nd

ed. Oxford University Press, Oxford, England.