You are on page 1of 150

Conservation of forest

ecosystems 7
Patterns of loss of forestry biodiversity – a
global perspective

W.A. Rodgers 1

Loss of forest biodiversity is due both to the total loss of forest cover (deforestation), as well as to
the loss of biodiversity components within forest (degradation). The former is in theory easier to
measure. Global estimates for tropical deforestation are from 0.5% to 1.0% per annum. These values
are however bedeviled by definitions. The lower estimates are for total loss of forest land to other land
uses. The higher figures include forest that is degraded to the point that it loses ecological function –
although a wooded community may remain.
Biodiversity loss within forest (at community, species and genetic levels) is due to a variety of
reasons – purposeful forest management (e.g., conversion to uniform and improved silvicultural
systems); to uncontrolled exploitation, to fragmentation, to loss of canopy, etc. Specific site studies
show that both forest loss and biodiversity loss are widespread, including forests that are recognized
as global “Centres of Biodiversity”, arguably the most important sites for global conservation. However,
data on rates and causes of such loss and degradation are grossly inadequate. Possibly 2-8% of
recognized species will go extinct by the year 2015. However, local extinctions affect local people
adversely, as well as genetic levels of biodiversity.
Conservation now recognizes that direct loss (exploitation, etc.) is in itself caused by deeper root
causes, including population pressure, poverty and non-compatible policies at national and international
levels. Working with local communities is one step towards overcoming past inadequacies.
Little real political will exists to reduce forest and biodiversity loss. Further forest clearing will
continue to take place. Conservationists should work to ensure that future loss is both planned and
efficient in terms of resources and future land use. Maintaining key sites as strict core reserves within
a matrix of used and productive forest will undoubtedly be the future biodiversity strategy.

Keywords: Forest conversion, biodiversity, conservation.


The global debate on loss of forest resources and forest biodiversity

Whilst most nations have adopted policies supporting the conservation of forest resources, actual
forest cover continues to decrease all over the world – forest policies often do not work. The
conservation of forest resources continues to generate controversy both internationally and nationally.
This controversy is polarized (Rosendal 1995) between:

EF/UNDP/FAO Regional Biodiversity Project in East Africa, P.O. Box 2, Dar es Salaam, Tanzania.

• The global interests (of the Developed Northern Nations) versus national/local interests (of the
Less Developed Southern Nations)
• The “Forest Rich” Nations versus the “Forest Poor” Nations.

Central to this debate is the belief held by the northern nations that the present pattern of tropical
forest loss and forest degradation is excessive and of great GLOBAL significance for the long-term
benefit of mankind. Simplistically this is countered by developing nations’ belief that “the west got
rich by cutting down forests, and now they want to stop us making similar conversions in the name of
global benefit”. Note however that there is also concern that the loss of temperate forest is as great a
problem as the loss of tropical forest (Rosendal 1995). Note also that developing nations have land-
based economies and rapidly growing populations. Developed nations have industrial economies
and low population growth rates.
There is evidence that forests and their resources are being lost at increasing rates (see below).
This loss has three sets of consequences:
• Lost biomass and tangible resources of immediate value;
• Impaired ecological processes (catchment, carbon, resilience);
• Loss of biological diversity (communities, species, genes).

Whilst there is acceptance that these forest resources are valuable, there is also awareness that the
land itself, after the forest has been cleared, may be of even greater value, especially short-term
financial value. The land is valued by the local people who farm it, and by governments who derive
export crops (and political gain) from it. This is the crux of the conservation argument – what are the
relative values of forest resources as compared to those of agricultural land; and of value to whom?
(Norton-Griffiths & Southey, 1995).
The perceived values of the inherent biodiversity attributes versus the direct values of the resources
themselves also fuel international and national debate. To what extent should forest be maintained for
production or protection values? Can a forest be managed for both purposes? Mittermeir & Bowles
(1993) stated that most biodiversity values (and so the resources making up the biodiversity) are
LOCAL or NATIONAL in nature, and that there is little that is primarily GLOBAL in nature (Table 1).
This contrasts with other statements on global values (e.g., GEF 1995). However there is increasing
recognition of the crucial role played by biodiversity in underpinning forest “resilience” in itself a
global benefit (Perrings et al. 1995).

Table 1. Biodiversity values: local or global in nature? 1

1. Ecosystem Functions (catchment/soil processes, nutrient cycling) – benefit the host country.

2. International Export Values (timber, beeswax, tourism, medicines) – benefit the exporter.

3. National Market Products (fuel, timber, water).

4. Local Market Products (food, poles, local medicines).

5. Household Products (food, fuel, poles, cultural etc).

6. Global Intangible Values (carbon fixation, science).

Only the last that is truly global, the others provide benefit to nations, communities, and individuals. Ecosystem functions can
have global importance, eg resilience, global warming.
After Mittermeir & Bowles 1993

Whilst data show that there is increasing forest loss, there is uncertainty as to how much of a
problem this represents, and still considerable disagreement as to the optimum solutions to preventing
such loss. Proponents of biodiversity see the solution as being in situ protection of natural forest.
Global warming strategists see the development of industrial and fuelwood plantations as being of
greater immediate benefit (Rosendal 1995). Social conservationists want a much greater role for people
in using and managing forests resources. Whilst the world is expressing concern at the scale of loss
of forest resources including biodiversity, there is a decreased ability in many developing nations to
prevent or restrict this loss. Government Forest Departments have lost management capability in this
past decade (e.g., Howard 1992 on Uganda, Sharma 1992 in general). The manner of “Protection”
requires analysis, with realization that the older command and control policing policies do not work.
More recent conservation efforts are aimed at linkages to local communities (Dompka 1996).
Conservation thinking sees the problem of tropical forest loss as more than lack of capacity within
tropical forest departments. The root causes of forest loss are much more deeply ingrained in the
totality of national and international policies, economics and development plans. Strategies to overcome
such loss must therefore be sought in the larger policy framework of development.

Biological diversity or biological resources?

“Biological Diversity” is but one of several parameters by which a natural resource can be described.
Other parameters would include structure or biological productivity for example. We can thus
characterize forests – one forest has greater biomass than another, a second forest has greater
productivity, a third forest has a greater level of species diversity. These are scientific and factual
statements. But the values attached to these statements may differ. Whilst forest A has a greater
overall diversity than forest B, there may be more strictly local endemic species in forest B than in A.
Evaluating A against B now becomes an emotive human value judgement issue rather than a scientific
decision, despite attempts to give an objective weight to different aspects of biodiversity (e.g.,
Humphries et al. 1992). Consider the different emotive reactions if the endemic under threat were to be
a mountain gorilla or an aphid!
In the last three years the level of conservation activism, the success of the Convention of Biological
Diversity, the publicity of UNCED and its aftermath within Agenda 21, etc., have all generated such
awareness of the term biodiversity that the world is more alive to biodiversity problems than to
biomass problems. Often, however, we tend to use the terms forest biodiversity and forest resources
fairly interchangeably. Strictly when we use the term biodiversity we should be referring to the diversity
of components within the forest,2 and not the components of the forest themselves.
My own experience managing one of the world’s first GEF biodiversity projects is that it is not easy
to put across the concept of biodiversity as opposed to a consideration of the resource itself. It is
exceptionally difficult when talking to non-scientists! People use and harvest forest resources, they
do not harvest biodiversity directly. It is the extent of resource use that impacts on the diversity of the
resource, and on the productivity parameters. If we wish to regulate this impact, then it has to be done
through the factors that affect resource use.

What is biodiversity loss?

Forest loss is relatively easy to see and measure. It is less easy to see biodiversity loss. All forests
have some biodiversity value, although some forests are much richer than others in communities and
species. The loss of any forest leads to lost resources, including the loss of biodiversity value. Our
concern should be twofold :

Biodiversity is defined simply as the level of variety and variation in the ecosystem, species and genetic levels.

• The loss and degradation of forest resources in general. These resources have a multitude of
direct and indirect values: including global values of carbon sequestration, other climatic benefits,
and biodiversity, as well as more direct biomass resource uses of value to local economies.
• The loss of biodiversity values, both as a consequence of forest loss and degradation as described
above, and the loss of biodiversity within forest as a consequence of manipulation for other
benefits (the enhancement of biomass for example, by conversion of natural forest to conifer

We can add a third concern, coming from the realization that increasing population growth needs
increased agricultural production. Some forest loss is therefore inevitable. Our concern must be the
forest loss that is unplanned. We must plan for forest conversion, and ensure that such loss is
followed by productive land use. We must be able to determine how much forest is needed, and
needed where. Foresters should say which forests should go!
Determining “rates of biodiversity loss” depends on knowing how many species there were to start
with. Current thinking uses estimates centring around 10 million, of which > 90% are arthropods (Reid
& Miller 1989, but see May 1988, WCMC 1992). Of these, over 50%, and possibly up to 90% of
terrestrial taxa, are in the tropical forest biomass which occupy only 7% of the earth’s land area (Botkin
& Talbot 1992). Many facts are used to express the scale of tropical forest diversity, a common one is
that 10 ha of forest in Borneo can have 700 tree species, more than all of North America!
Rates of loss require knowledge of extinctions. When is a species lost? It is useful to consider both
global and local extinction. Science loses a species when it is globally extinct. Utilitarian considerations
must emphasize the problems associated with more localized loss of species. The biodiversity literature
has discussed loss rates extensively. Ballpark estimates suggest that 3-10% of the earth’s biota will be
lost totally in the coming two decades (Wilcox 1995, quoting several sources, but stressing the many
unproven assumptions in these figures). There are no reviews of local loss.
This paper reviews the main issues leading to the loss of forest and the subsequent loss of forest
biodiversity. Biodiversity loss and degradation is considered at the three levels of ecosystem loss
(equivalent to the loss of named types of forest cover), species loss, and loss of genetic material.


Information on rates of forest loss and degradation

The world conservation lobby acknowledges that rates of global forest loss are excessive. It
recognizes the need for greater attempts to reduce that loss (Agenda 21). However, the quality of data
that are available to quantify this loss leaves much to be desired. We do not have the information to
monitor forest loss adequately at national level. But it is at local and national levels that loss will be
reduced, aided perhaps by changes in international economic policies.
The most comprehensive data set on deforestation comes from FAO’s Tropical Forest Assessment
(FAO 1991, 1993a). This analyses change in forest cover between 1980 and 1990, adding to an earlier
study preceding 1980 (FAO / UNEP 1982). FAO takes a broad view of what is forest,3 basically land
with a woody cover of over 10% canopy.

FAO defines forest as any land generally subject to wild conditions which has a tree canopy cover of over 10%.
Deforestation is the conversion of such tree cover to a permanent land use outside forestry. The key is that this is
land that is lost from the control of the forest sector. Degradation however is changes within the forest class, i.e.
when a forest cover, and so production capacity, is reduced (FAO 1993).

Sharma (1992) summarizes information on potential distribution of forest and woodland resources,
suggesting that in the period 1850-1980, 15% of the world’s forests and woodlands were cleared, in
Asia by 43%. In the 1980-1985 period forest was lost at an annual average rate of 11.4 million ha or
0.6%. Now it has almost doubled to 18.5 million ha or 1% per annum. It is the rate of acceleration of
forest loss that is critical, are we still losing forests at an ever- increasing rate?
WRI (1990) analysed forest loss from different literature sources. Sisk et al. (1994) point out that
FAO and WRI estimates are similar (differing by only 10-15% when the definitions are made compatible).
Jepma (1995) describes the FAO and WRI data sets as the “forestry sector view”, with forest loss
defined as the complete clearance of tree formations and their replacement by other land uses, i.e.
forest loss is land that is lost to forest sector’s control. Several authors argue that this excludes the
loss of closed forest (which is the real focus of biodiversity interest) degrading down to a more open
woodland, and so losing the function of a forest. These arguments led to the scientific lobby producing
different definitions, where deforestation usually includes severe degradation. Myers (1986, 1991)
produced the broadest data set from this lobby.
Table 2 shows the extent of forest loss as assessed by these reviews. Note that the different
definitions of what is forest cover, and so what is forest loss lead to major discrepancies in amounts
and rates.
These different definitions of deforestation present problems when attempting global overviews of
forest loss. They prevent analysis of causes of problems at national level, and prevent the monitoring
of the efficacy of solutions. Forests are a national resource, and policies which cause or prevent forest
loss are largely national in their design and their implementation. Action therefore has to be around
national databases and national monitoring activities. But there are few factual data at national level!
In Tanzania for example there is a rule of thumb figure of 400 000 ha deforestation per annum, but there
is no basis for this estimate. (FAO data are developed from continental averages for Tanzania, as there
has not been two sets of forest cover mapping.) It is not possible to break down this crude figure as to
where this deforestation takes place? or on what sort of land? or why?
Of fundamental interest is the legal status of the site from where such loss occurs. Are we seeing
loss from open unreserved forest (which is to be expected)? or is there also significant loss within the
Protected Area networks (from where loss is not expected to take place)? Key questions to be answered
for forest conservation are :

Table 2. Forest cover and rate of forest loss in the tropics

Region Forest area1 estimates Annual loss (%)


Tropical Africa 4.6 3.5 0.5 0.5 0.7 1.0

Latin America 7.8 8.2 3.9 1.4 0.6 1.7
Tropical Asia 2.7 2.3 1.9 1.0 1.4 2.2

Ecological zone Area Land (%) Deforest 1980-90 (%)

Wet Forest 7.2 77 0.6
Moist Deciduous 5.9 46 1.0
Dry Forest 2.4 19 0.9
Mountain 2.0 28 1.1
Area Estimates are in million km2 .
Data from : WRI (1990) early 1980s – total of open and closed forest; FAO (1993) for total forest area (>10%
canopy) in 1990; Myers (1991) for total closed forest in 1991. See Jepma (1995) & FAO (1995) for discussion.

• What is the rate of forest loss from closed evergreen forest as opposed to all land with woody
cover? This recognizes the especial value of closed forest for biodiversity.
• What is the rate of forest loss from different categories of land: i.e. open or public land, private
land and protected areas (PAs) such as forest reserves.
• From the PAs, what is the rate of forest loss that is legally sanctioned (ie excisions) and what is
illegal (i.e. encroachment).

But information which distinguishes ecological or legal types of forest are less easy to find. Rarely
can tropical forest be classified from satellite imagery. In Tanzania Rodgers et al. (1985) mapped closed
forest cover distinguishing thicket from forest within three altitude classes and two rainfall zones.
Collins et al. (1992) give area data for the main plant communities of the closed forest of Kenya. In
conclusion, therefore, we have rough ballpark figures which suggest that there is a significant loss of
forest cover, and that this loss is still increasing. FAO (1993a) show the rate of forest loss worldwide
has increased from 0.5% in the 1970s to almost 1% in the 1980s.

Recognition of Forest Loss and Forest Degradation as Problems.

Most world forestry organizations recognize the current loss of forest as a significant problem
(Tarasofsky 1995). This problem is expressed in terms of loss of resources, in loss of ecological
function (e.g., carbon sequestration, hydrological function) and in terms of biodiversity. Worldwide
recognition of deforestation as an environmental issue rose dramatically in the 1980-85 period, when
developing country governments and donor agencies came under pressure to exercise greater control
(Deacon 1994).
As an example of this global concern, Agenda 21 (Para 11.10) states:

Forests worldwide are being threatened by uncontrolled degradation and conversion to other
forms of land uses, influenced by increasing human needs; agricultural expansion; and
environmentally harmful mismanagement, including : lack of forest fire control, anti-poaching
measures, unsustainable commercial logging, overgrazing, airborne pollutants, economic
incentives, and activities of other sectors of the economy. The impacts of loss and degradation
of forests are in the form of soil erosion, loss of biological diversity, damage to wild habitats and
degradation of watershed areas, deterioration of the quality of life, and reduction of the options
for development.

These are fine statements at international level, and even, for international consumption, at national
level. They lose meaning at local community and household level however. At household level the
conversion and degradation of forests is controlled and purposeful. Forest and forest land utilization
is a conscious decision, perhaps not a preferred decision, and perhaps a decision forced by external
circumstances. (“Forests are not usually cleared for trivial reasons, they provide food for expanding
populations,” Repetto and Gillis 1988). ‘Whose quality of life is deteriorating?’ is an important question.
Is it our western way of life that is threatened, when faced with a prospect of global warming or loss of
advertised biological riches? Is it national élites, faced with power and water shortages?
Local people have shorter time-scales. Their quality of life can be improved by converting a forest
patch (and acquiring tenure perhaps!), or by selling timber or charcoal! This improvement is essential
with no other alternative to support life! The issue can be posed as a series of options available to
poor households:

Option 1. Seek employment – temporary or longer term
NO INCOME Option 2. Clear forest for cultivation – maybe gain land title
Option 3. Exploit forest for fuelwood for cash

National policy initiatives can influence the choice of option through employment, cottage industry,
urbanization, land tenure, agriculture subsidies, etc.
It is interesting to compare international concerns (e.g., Agenda 21) with those at national level. In
Tanzania for example, there has been little specific concern for biodiversity in the recent past, (e.g., the
past Director of Forestry “there is too much emphasis on conservation, we need more production”).
Concern is with more tangible resources such as fuel and timber. National resource and sectoral
policies, strategies and plans (eg TFAPs, NEAPs, NCSSDs) all decry the loss of forest resources, and
all espouse conservation and sustainable utilization. There is talk of increasing investment: of extending
tree plantations and reafforestation, and of increasing protection levels. But the reality is the opposite
– forest is lost and degraded at increasing rates. Simplistically, there is a policy failure.

Reasons for forest loss

Why are we losing forests? Most national forest policies state that forest resources should be
conserved, and yet we see increasing levels of degradation. The causes of deforestation are not well
understood. Discussion often mixes the direct causes (e.g., slash-and-burn agriculture) with the
underlying fundamental or root causes of loss (Sharma 1993, Deacon 1994, Jepma 1995, Pearce &
Moran 1994 introduce the issues in economic terms). Direct causes include agricultural expansion,
overgrazing, poor logging practice, etc. Underlying causes include market and policy failures, population
growth, rural poverty which lead to land conversion, etc. Table 3 gives a summary of suggested direct
reasons for forest loss.

Table 3. Analysis of direct reasons for forest loss

A) After Otto (1990)
Burning/clearing for cultivation 63% Conversion to commercial agriculture 16%
Felling for firewood 8% Reclamation for livestock ranching 6%
Commercial exploitation of timber 6% Construction of infrastructure 1%

B) Myers (1991)
Shifting Cultivation 61% Commercial Logging 21%
Cattle Ranching 11% Conversion to plantations, roads, etc. 7%

C) LEEC (1992)
Activity Deforestation Degradation Modification (plantation etc)
(%) (%) (%)
Forestry 2 10 71
Agriculture 83 76 26
Others/Mines 15 13 4

TOTALS 100 100 100

Note A-C are still basically figures for loss – not degradation. Factors such as acid rain are not included.

Deforestation and forest degradation, particularly in developing countries, are influenced by an
array of structural problems related to the international economic regime (Tarasofsky 1995), as well as
the underlying socio-economic features of the country itself.
It is useful to consider two levels of underlying causes:
• The Ultimate factors themselves, about which we can do relatively little in the medium term:
Population growth and resource demand;
Economic dependence on natural resources;
General widespread poverty.
• The Proximate factors, which can be addressed in the medium term:
Inadequate policy regimes;
Lack of stakeholder participation;
Lack of adequate tenure and access rights;
Inadequate investment in the forest sector;
Inappropriate valuation systems;
Inadequate land-use planning capacity and systems.
A typology of how forest land is being lost and degraded is of interest. Three sorts of forest loss
can be conjectured, the key being whose land is it to develop in the first place?
• “Theft” from government forests by local people, by industry and by government itself;
• Settlement expansion onto “open” lands (Brazil, Indonesia, the Miombo woodlands of Tanzania);
• Conversion of privately owned forest land. (Note “theft” here means non-authorized exploitation
of resources, including the land itself as a resource).

Why are forest conservation (meaning rational sustainable utilization) policies failing? Is it the
failure of policies themselves (and most tropical countries have some form of forest policy, it is not a
policy vacuum)? or are the policies poorly implemented? Policies exist at several levels in government.
Most are sectoral in nature, reflecting the separation of responsibility within most government systems.
For example in anglophone Africa most governments have forest, wildlife, energy, environment, land,
population, water, agriculture policies, etc. Usually there is little linkage between them, sometimes
they are non-compatible! These sectoral policies should fall under the general economic and
development policies of the country (at a simple level the degree of “isms” of socialism or capitalism,
the dependence on subsistence agriculture or an increased industrialization, private venture or state
driven, etc.).
As well as sectoral policy there are overlapping policy initiatives looking at patterns of resource
use, research, etc. Sectoral policies will have subsets looking at the ratio of utilization to conservation,
for example (see Hamilton 1988, Rodgers 1993) for comment on changing ratios in East Africa.
Maintaining a productive forest cover though is more than a forest issue! Table 4 shows a
hypothetical matrix of policies which can affect patterns of conversion of forest land (see detail in FAO
A forest sectoral policy which does not interact with the larger sectoral concerns of agricultural
expansion, or patterns of employment, or long-term energy and water requirements, cannot expect to
find success. “Interaction” was a principal tenet of the Tanzanian Forest Action Plan, building links to
forest product users, including water, and fuelwood.

Does high population pressure necessarily lead to forest loss?

Conventional wisdom suggests that growing populations and consumerism lead to forest loss and
degradation (e.g., Barnes 1990). However this view has been questioned, for example Agrawal (1995)
writing about largely drier areas of India, and Tiffen et al. (1994) writing about Machakos District in

Table 4. Sectoral policies as separate units within a larger development policy matrix.
Cultivation spread Reservation Fuelwood demand
Incentive/Prices Sustainable use Alternate energy
Subsidies Access/Tenure Incentives/Prices
Intensification Power to people Water demand
Consolidation Stakeholders? Cottage industry
Treecrop farming Plantations?


Population Urbanization Tenure/Access
Employment Conservation Industrialization

dry Kenya. Here an initial period of degradation led to a phase where land consolidation and improved
tenure led to an increase of tree cover – albeit often exotic species, for shade, fruit, fuel, etc. The central
role of tenure in these cases is highlighted by Juma and Ojwang (1996). These improvements are yet
to be seen in the converted humid forest lands of Africa, although the integrated tree crop gardens of
Indonesia, for example, do maintain high tree cover and biodiversity.
Population growth compounded with other factors, such as landlessness, increasing poverty and
government corruption, does add to the rates of deforestation. This is emotively expressed by Witte
(1992) writing on Zaire. Deacon (1994) provides a detailed analysis of forest loss associated with both
population growth and with a variety of political factors which affect resource tenure and access
rights. Using data from a sample of 120 countries he showed that tenure and access to forest values
are critical pre-requisites to an appreciation of forest value and so reduced forest loss.
Barnes (1990) suggests that rates of forest loss are linked to factors of human population growth
and the area of remaining forest, and that the rate of increase is expected to continue to rise. Whilst
negative feedback systems exist to slow the rate of deforestation, they can fail to operate, as in Côte
d’Ivoire where data show an ever-increasing rate of forest loss (see Box 1, taken from Chatelain et al.


Information on biodiversity and the loss of biodiversity

Biodiversity literature stresses the level of our ignorance as to the extent of community, species
and genetic diversity; both at the component level (how many species are there, and where?) and at
the structural and functional levels (how does this assemblage of species function in terms of carbon
storage etc.?) (Wilcox 1995). To this we must add a new set of questions : “How much and which
biodiversity do we need? Because we cannot have it all”!
Science is concerned with the inability to document the magnitude of biodiversity. Pimm et al.
(1995) state: “How can we be confident in discussing extinction rates, when we know well some 10 5
species out of 106 species, and estimates of the total order of species are some 107 or 108?” Our present
level of knowledge does however allow us to recognize biodiversity hotspots (e.g., WCMC’s Centres
of Biodiversity; postulated Refugia and Centres of Endemism, etc.; Birdlife International’s Important
Bird Areas – IBAs, Bibby et al. 1992). Information allows us to compare ecosystems and communities
for some indicator taxa (higher plants and birds), and larger geographic entities (continents, large

Estimates suggest Côte d’Ivoire is losing forest at 6.5% per annum, the highest in the world. In the last
20 years, 79% of unprotected forest has been cleared, encroachment affecting the Protected Areas. Tai
National Park (of 350 000 ha) is the largest block of forest remaining. The case study is the result of
detailed imagery analysis of 100 x 100 km block adjacent to Tai NP, looking at change over a 34-year time
Three types of forest change are apparent : Total forest loss – conversion to smallholder agriculture
leading to fragmentation; degradation due to selective harvesting; severe degradation due to intensive
harvesting. Three phases of deforestation were:

1956-74 Forest cover actually increased.

1974-84 Deforestation 21% in 10 years. Clearings in the forest, not fragmentation.
1984-90 Deforestation 22% in 6 years. Fragmentation, few patches over 5 ha are left.
PAs now threatened.

These major changes followed road building to give access to loggers. Cultivation followed in the
cleared areas, forest patches were left on poor soil. The process intensified with influx of refugees.
Fragments are very degraded. Clearing continued up to and into the Protected Areas.

countries). Pomeroy (1993), for example, discusses the distribution of centres of biodiversity in Africa;
Gentry (1992) does the same for South America. Key sites for the protection of global biodiversity are
known in general terms.
Definitions of biodiversity from within the tropics emphasize their utilitarian values. Many tropical
countries are totally dependent on biodiversity for their subsistence livelihoods, cash crops and
overall agricultural ecological resilience (WWF 1993). Yet it is those attributes of variety and variability
that contribute to resilience that are so difficult to measure – and prove! The values of biodiversity are
often stated to include the indirect functional benefits of hydrological functions, etc. (e.g., Wilcox
1995). Yet there is evidence that less biodiverse communities can provide these benefits almost as well
as intact closed forest, such as tea plantations, bananas, hardwood plantations, etc. (see Norton-
Griffiths 1994 for general discussion, Bruijnzeel 1990 on forest hydrology).

Patterns of biodiversity loss

There is little hard information on rates, amounts and values of biodiversity lost from the tropical
forest ecosystems. Forest loss obviously includes total loss of biodiversity. Such loss is a relatively
tangible entity, measurable on the ground and from imagery monitoring programmes. In theory spatial
and temporal quantitative data on forest loss could be collected. Such data could be combined with
socio-political data as to why such loss was taking place.
Global information on biodiversity loss is available in a number of reviews at national level, the
most comprehensive being WCMC’s (1992) book: Global biodiversity: Status of the earth’s living
resources. Other important sources are selected national level Biodiversity Country Studies (there are
only some 15-20, without a recent review); UNEP’s Global biodiversity assessment (1995), Whitmore
and Sayer (1992) and the WCMC -IUCN Forest atlases (Collins et al. 1992).
WCMC (1992) document 484 animal species known to have become extinct since 1600. Most (75%)
were on islands. Half had a known cause of extinction – “introductions” accounted for 39% and
habitat loss for 36%, and direct hunting, extermination for 23%. It is revealing to look at the present
pattern of habitat loss and conjecture impact on biodiversity. Reid (1992) looks at habitat loss on all

three tropical continents. He projects present rate of loss plus 50%. Using biogeographic models he
then predicts an eventual faunal decline of 2-8% of all species by 2015.
However, the present pattern of loss is much less than this – for vertebrates at least. Jamaica has
lost 90% forest but only 5 species of vascular plant. Kenya has lost 10% plant species this century. All
bird species known since 1900 are still exstant (some only just!). For a species to go extinct as opposed
to local extinctions, that species must be restricted in range and habitat and/or be of exceptionally low
density. Today we have not seen the expected level of extinction, but all the evidence suggests it will
happen. Species are “committed to extinction” due to loss of habitat, and these extinctions will take
place as communities adjust to new equilibria. Most will be local extinctions, a plant can disappear
from one area but remain in another. Strict endemics are most at risk.
Why does biodiversity loss take place? Total loss of habitat is obvious and immediate. Partial
clearance and so reduction of habitat extent leads to loss of part of the species complement – but also
greater species packing in the remaining area, which becomes of great biodiversity risk. Habitat
modification is of concern. Managed forests are often simplified (e.g., conversion to uniform silvicultural
treatments, removing stranglers, climbers and snags).
Sisk et al. (1994) link concern with biodiversity to two sets of information. These are the level of
biodiversity value (simply species richness and proportion of endemism) and the level of threat
(simply population density and rate of forest loss). Data on mammals and butterflies are presented for
130 countries. The authors draw attention to countries with high diversity and high threat (e.g., India,
Kenya, Colombia and Côte d’Ivoire). However the implications of such listing are not clear. Is it
beneficial for a country to be so listed and attract more donor support – e.g., GEF?

Tropical forests and biodiversity hotspots

Tropical forests have the greatest variety of organisms of all ecosystems (WCMC 1992, UNEP
1995). The introduction to this paper said that not all areas of tropical forest were of equal importance
for biodiversity. Some areas by virtue of their intrinsic variety of communities and species can be
described as Centres of Diversity. Some areas by virtue of their high proportion of endemic species4
(or other taxa) can be called Centres of Endemism. Sometimes the two can overlap, often though they
do not. Centres of diversity and endemism can be for one taxonomic group – e.g ., plants (Centres of
plant diversity, WCMC 1992, Davis et al. 1994, see Box 2 for example), or birds (Important bird areas,
Bibby 1992) or for several groups, such as the East Usambara Mountains (plants, herps, birds, millipedes,
gastropods, some wasps etc), Rodgers and Homewood (1982).
Clearly forest loss in these centres of diversity and endemism will have greater consequences for
biodiversity than in other areas less well endowed. Countries with such hotspots are often recognized
as Mega-diversity nations, and it is suggested that they should receive increasing recognition and
support for their conservation programmes. Here we see the GEF criteria of “Global Value” being used.
all countries have biodiversity of national value. If these sites are accepted as being of global
importance, then do nations devote a greater share of resources to them, so reducing levels of
deforestation and degradation? Evidence from the East Usambara Mountains in Tanzania (arguably
Africa’s most intense site of species endemism, see Rodgers & Homewood 1982, Hamilton and Bensted-
Smith 1992) suggest that despite considerable donor investment, there is not yet the expected change
of status and so protection, for the forest areas.

An endemic species is one of restricted distribution, e.g., this species is restricted to Africa, or to Kenya, or to
the Eastern Arc, or to the Taita Hills Forest Reserve. Such species are not known to occur elsewhere in the world.
Areas with high levels of endemism are of obvious importance for biodiversity conservation.

Centres of Plant Biodiversity (Davis et al. 1994, WCMC 1992) are globally important due to two key
criteria of species richness, and number of endemic species. In addition, CPD sites “contain an important
genepool of species of value to man; contain a diversity of habitat types; contain a significant proportion
of species adapted to edaphic conditions; or the site is subject to major threat of devastation”. Davis et al.
identify 234 sites of global importance:

Africa 84 sites Europe 24 sites

Atlantic Isles 4 Indian Islands 4
SW Asia 21 Americas 75
Others 22

The effectiveness of conservation for these sites comes from a qualitative assessment of degree of
threat to the site (Davis):

Region No of Degree of threat


Africa 30 7 5 7 8 3
Europe 9 2 2 - 5 -
At Islands 4 1 - - 1 1
Indian Oc Is 4 - - 1 1 1
SW Asia 11 - - 4 4 3
India SubCont 12 2 6 - 3 2
S America 45 7 5 3 14 16

TOTAL 234 33 40 41 74 45

A = Reasonably safe, B = Partially safe, C = Vulnerable, D = Threatened, E = Severely threatened. Over

half of our globally important sites are insufficiently protected.

The conservation of genetic diversity in tropical forest ecosystems

Biodiversity conservation is not only species and the ecosystems that contain the species, but
also the genetic variation within these species (e.g., Kemp & Palmberg-Lerche,1994). Genetic diversity
in forest ecosystems – the variation within and between species is :
• The basis for their adaption to environmental stress, including more extreme effects of global
climate change in the future, and emergence of new pests and diseases; and
• The basis for sustainably developing and improving forest resources for human use. Such
improvement is an urgent need as human populations grow in number and demand.

Increased forest productivity will depend on selection of superior genotypes; we must therefore
maintain a broad genetic base. Standard conservation procedures of creating protected areas to
maintain species and ecosystems may not be enough to maintain genetic variation. Several populations
of a species should be maintained, preferably covering the geographical and environmental limits of
the species range. Highest levels of genetic variation often occur at the edges of a species range –

unfortunately these edges are often most vulnerable. Genetic conservation therefore requires a network
of PAs across the species range, within a mosaic of other land uses where examples of target species
may be found. If needed ex situ methods (e.g., seed orchards, etc.) should be used. An example of a
species distribution study which allows the planning of genetic level conservation is that of Juniperus
procera – pencil cedar (Hall 1984). Whilst some authors see genetic impoverishment as the least
damaging form of biodiversity loss (Perrings et al. 1995), this is the loss that is related to resources of
proven value (wild crop relatives, timber tree provenances, etc.). It is access to such genetic variation
that drives the global debate on access to biodiversity.

The conservation of biological diversity and protected areas

Conventional wisdom uses Protected Areas (PAs) as the key strategy for in situ conservation of
biodiversity values (Rodgers & Panwar 1988). New people-friendly strategies are being developed to
bridge the PA – Community gap, but the PA is still the core around which conservation activity takes
place. Many countries have extensive PA networks; unfortunately many in Africa were developed for
larger vertebrates not for forest communities (see Rodgers 1995). It is rare to find a PA network that
represents adequately the extent of forest biodiversity. Iremonger et al. (1997 – this congress) document
the major gaps.
Creating a network is the first step, ensuring that the network is implemented and managed is often
a problem. Many PAs exist on paper (see Kothari et al. 1989, in India; Rodgers 1993 for forests in East
Africa). We still need to develop cost-effective methods to ensure the long-term conservation of these
biodiversity values. Forest biodiversity planners emphasize five main concepts (Whitmore & Sayer
• That a core of well protected reserves or parks is needed. These could be relatively small.
• Today < 4% of the terrestrial ecosystem is so protected. Therefore the great majority of biodiversity
resources and values are outside the PA network.
• As the PAs often are not large enough to maintain the full complement of species within them,
they need to be supported by suitable habitat (e.g., worked forest) outside the PA.
• This matrix of forest habitat maintains the functional attributes of soil/water formation, nutrient
cycling, genetic transfer, etc.
• That people are part of the solution, and that this outer forest framework must be integrated into
a multiple land- use mosaic.

Table 5. Examples of biodiversity loss at different component levels

ECOSYSTEM Site Conversion Forest patches, e.g. Box 1 in Côte d’Ivoire
Degradation Acid Rain in many areas
Virtual Total Loss Dry Evergreen Forest, India
Habitat Destruction Primates in coastal Brazil

SPECIES Introductions Maesopsis in Usambaras

Over-exploitation Debarking Prunus in East Africa
Non-viable Pop’ns FF birds in EA Coastal Forest 1
Range Reduction Wild Coffea arabica in Ethiopia

GENETIC Over-use Lost provenances of Juniperus

FF = totally forest dependent birds

Pimental et al. (1994) argue that 95% of the earth’s surface is agriculture, settlement and managed
forest (50% cultivation, 20% commercial forest, 25% human settlement and infrastructure and 5%
unmanaged wildland). Of the earth’s supposed 10 million species (Reid & Miller 1989) 1 million will be
lost within the coming 20 years. Much of this will be from agricultural lands. Issues such as the
2 500 million kg per annum of toxic pesticides are partially responsible for this pattern of loss. Pimental
et al. quote German studies that only some 35% of 30 000 species are in PAs. This however is at
variance with Sayer & Stuart (1988) who show that for larger vertebrates in tropical forests most
species are within PAs (e.g., for birds: Cameroon 77%, Kenya 85%, Uganda 89%). A further example of
biodiversity loss comes from Kenya, which has a growth rate of over 3% per annum, with < 7% of the
land as full protected area but almost 70% of large ungulate outside the PAs. Pimental predicted a
considerable loss. Norton Griffiths (1996) analysed 19 years of continuous aerial monitoring data and
showed a consistent and significant decrease in the main wildlife districts, of almost 3% per annum, or
over 50% of the animals gone!
Methods of maintaining some biodiversity value in managed lands are being developed and
disseminated, including plantations (e.g., FAO 1992). Mosaics of different land uses, land-use planning
at landscape scales, intercropping, hedgerows, environmentally friendly farming techniques, etc., are
all key components.

The significance of biodiversity loss

UNEP (1995) and Grainger (1992) tabulate the consequences of tropical deforestation for biodiversity
(see Table 6). Each consequence leads to another set. “Deforestation” leads to “loss of species
diversity”. What does that lead to? Two sets of consequences say Perrings et al. (1995) : the loss of
genetic material (like losing books from a genetic code library) and reduced resilience. What are the
consequences of this? How much resilience do we lose if 5% of the species disappear? How can we
reach compromise and trade-offs? These are no longer scientific questions. Answers are political and
socio-economic. The significance of biodiversity loss is related to the degree of value that our society
places on biodiversity, and level of value on converted land. Brussard (1994) writing the President’s
column in Conservation Biology said “... in discussing biodiversity with people I have found a startling
naivety on what it is, why it is important, and why we are concerned. Many people are ignorant of,
indifferent to, or opposed to conservation of biodiversity. That was the USA in 1994, not poor villagers
in Ethiopia or Nepal! We have a long way to go!

Table 6. Consequences of tropical deforestation for biodiversity 1

Reduced diversity of species and genes
Species extinctions Loss of genetic material
Inability to select new crop plants Reduced production of non-timber forest produce
Reduced resilience within communities

Changes to ecosystems
Soil degradation Changed water flows from catchments
Increases sedimentation Impact on local and regional climate

Changes to global system

Reduced carbon storage Increased atmosphere carbon dioxide
Changes in temperature/rainfall patterns
After Grainger 1992 and UNEP 1995

“Recent global and national information sources show significant and still increasing loss of
natural forest cover in most tropical regions. Both forest loss, and forest degradation within still
existing forest cover, lead to a reduction in forest values, including biomass, ecological processes and
biodiversity. Such biodiversity losses are predicted from biogeographic theory, but have been difficult
to show in mainland as opposed to island situations. However the number of species under threat of
extinction is high, and forest fragmentation continues, leading to further reduction in populations
above critical viability levels. Modern parlance says “many species are committed to extinction”.
Whilst we can expect to see further species extinctions, and further loss of genetic diversity, it is
difficult to quantify rates, and to quantify the significance of these losses.
Constantly increasing human populations with greater consumerism means that we have to expect
further loss of forest cover and biodiversity. There is no magic wand to prevent such loss. Many
tropical countries have policies which actively promote such forest clearing. What is of concern is
that this loss is unplanned. Working to choose what areas should remain forested, and to choose what
areas should be converted in the most efficient manner, should be a priority for conservation planners.
It is difficult to see real political will to arrest the present patterns of forest loss, either in the
developing world or in the developed west. Present conservation activity is still largely fire-fighting
(often a losing battle). Only in the last few years has there been an attempt to think beyond the
immediate causes of loss on the ground, and consider the root causes of such loss. Involvement of
local people as stakeholders is a major step forward. Conservation at both field and at policy levels will
require further investment of financial and political capital, resources will be needed from both the
tropical countries themselves and from the west. Economic analyses of wildlife conservation have
shown the importance of such financial investment (Leader-Williams 1990) if conservation is to succeed.
Forest conservation should be no different, except that forestry could produce some investment
internally. Whilst the GEF offers a glimpse of potential methodologies to provide resources, the pot is
not big enough to deliver needed support flows to arrest conversion.
The debate focuses on both the loss of resource-rich forest, and about the loss of forest biodiversity.
The World Bank (1994) in their Africa Forest Strategy state that: “Some conversion of forest land to
agricultural usages is unavoidable and necessary for socio-economic development ...” What is not
said is how much is necessary for agriculture and how much for biodiversity! Norton-Griffiths and
Southey (1995) in a provocative paper on biodiversity conservation in Kenya suggest that, given
good land stewardship relatively little forest land for biodiversity is necessary (e.g., using principles
of complementarity, focus on centres of endemism, etc.), and that it is in national interests to develop
productive agricultural land.
The total amount of natural forest cover required to capture significant biodiversity areas (i.e. sites
with a high proportion of the forest biota) is probably relatively small, but larger than the present
forest Protected Area Network. But such PA sites will not maintain these biodiversity components if
the surrounding land-use cover does not permit the normal functioning of a forest ecosystem. Thus a
large forest area is still essential to maintain biodiversity values.

Bibliography FAO. 1994b. The state of food and agriculture:
Agrawal, A. 1995. Population pressure = forest Forest policy dilemmas. FAO, Rome.
degradation; an oversimplistic equation. FAO. 1995. State of the world’s forests. FAO,
Unasylva. 181 46 : 50-57. Rome.
Barnes, R.W. 1990. Deforestation trends in tropi- GEF. 1995. Operational strategies. GEF Secretar-
cal Africa. African Journal of Ecology. iat, Washington DC.
28 :161-173. Gentry, A.H. 1992. Tropical forest biodiversity:
BaskÚn, Y. 1994. Ecosystem function of biodiver- distributional patterns and their conserva-
sity. Bioscience. 44 (10) 657-660. tion significance. Oikos. 63 : 19-28.
Bibby, C.J. et al. 1992. Putting biodiversity on Grainger, A. 1992. Controlling tropical deforest-
the map: Priorities for conservation. ICBP, ation. Earthscan, London.
Cambridge. Hall, J.B. 1984. Juniperus excelsa in Africa: A bi-
Botkin, D.B. & Talbot, L.M. 1992 Biological diver- ogeographical study of an Afromontane tree.
sity and forests, pp 47-74 in N. P. Sharma Journal of Biogeography. 11 : 47-61.
(Ed): Managing the world’s forests. Iowa, Hamilton, A.C. & Bensted-Smith, R. 1988. The for-
Kendall/Hunt. est cover of the East Usambara Mountains.
Bruijnzeel, L.A. 1990. Hydrology of tropical moist IUCN Forest Series. Gland, Switzerland.
forests and effects of conversion: A state of Hamilton, A.C. 1984. Deforestation in Uganda.
knowledge review. UNESCO Paris. Nairobi, Oxford University Press.
Brussard, P.F. 1994 Why do we want to conserve Howard, P. 1992. Nature conservation in Ugan-
biodiversity anyway? President’s Column. da’s tropical forest reserves. IUCN
Conservation Biology Newsletter 1 (4): 1. Switzerland.
Chatelaine, C., Gautier, L. & Spichiger, R. 1996. A Humphries, C.J., Vane-Wright, R.I. & Williams,
recent history of forest fragmentation in P.H. 1992. Biodiversity reserves: Setting new
south-western Ivory Coast. Biodiversity and priorities for the conservation of wildlife.
Conservation 5 : 37-53. Natural History Museum, London.
Collins, M., Sayer, J. & Harcourt, C. 1992. Rain Jepma C.J. 1995. Tropical deforestation – A so-
forest atlas of Africa. MacMillan, UK. cio-economic approach. Earthscan Books.
Davis, S.D., Heywood, V.H. & Hamilton, A.C. 1994. London.
Centres of plant diversity: A guide and strat- Iremonger, S., et al. 1997. A global overview of
egy for their conservation. WWF, IUCN, forest conservation. Paper for World
Switzerland. Forestry Congress. 1997.
Deacon, R.T. 1994. Deforestation and the rule of Juma, C. & Ojwang, J. 1986. In land we trust. Zed
law in a cross-section of countries. Land Books, London.
Economics. 70 (4): 414-430. Kemp, R.H. & Palmberg-Lerche, C. 1994. Conser-
Dompka, V. Ed. 1996. Human population, biodi- vation of genetic resources in forest ecosys-
versity and protected areas: science and tems. In: readings in sustainable forest man-
policy issues. American Association for the agement. FAO Forestry Paper, Rome.
Advancement of Science. Washington DC. Leader-Williams, N. 1990. Black rhinos and ele-
FAO. 1992. Mixed and pure forest plantations in phants: lessons for conservation. Oryx
the tropics. Forestry Paper 103, FAO. Rome. 24 : 23-29.
FAO/UNEP. 1983. Tropical forest resources assess- Lugo, A.E., Parrota, J.A. & Brown, S. 1993. Loss
ment project. Forestry Paper 30, FAO. Rome. in species caused by tropical deforestation
FAO. 1993. Forest resources assessment 1990: and their recovery through management.
Tropics. Forestry Paper 112, FAO. Rome. Ambi., 22 (2-3) 106-109.
FAO. 1994a. Forest resources assessment 1990: May, R.M. 1988. How many species are there on
Country briefs. FAO Rome. earth? Science 241: 1441-1449.

Mittermeir, R. & Bowles, I. 1993. The global envi- Mountain Forests, Tanzania. Biological Jour-
ronment facility. Washington, DC. nal of the Linnaen Society 18: 197-242.
Myers, N. 1986. Tropical forests: patterns of de- Rodgers, W.A., Shishira, T. & Mziray, W. 1985.
pletion. In: Tropical rain forests and the The closed forest cover of Tanzania. Insti-
world atmosphere. Ed G.T.Prance. Westview tute of Resource Assessment, University of
Press, USA. Dar es Salaam. Research Paper New Series.
Myers, N. 1991. Deforestation rates in tropical Rodgers, W.A. & Panwar, H. 1988. Planning a
forests and their climatic implications. protected area network for India. Wildlife
Friends of the Earth, London. Institute of India. Dehra Dun. India.
Norton-Griffiths, M. & Southey, C.1995. The op- Rosendal, G.K. 1995. The forest issue in post
portunity costs of biodiversity in Kenya. UNCED international negotiations: Conflict-
Ecological Economics. 12 : 125-139. ing interests and fora for reconciliation. Bio-
Norton-Griffiths, M. 1996. The need for a conser- diversity and conservation. 4 : 91-107.
vation policy process in Kenya. Paper. Nai- Sayer, J. A. & Stuart, S. 1988 Biological diversity
robi. and tropical forests. Guest Comment. Envi-
Pearce, D. & Moran, D. 1994. The economic val- ronmental Conservation. 15 (3) : 193-194.
ue of biodiversity. London, Earthscan Books. Sharma, N.P. (Ed), 1993. Managing the world’s
Perrings, C., Maler, K.G., Folke, C., Holling, C.S. & forests. Iowa, Kendall/Hunt.
Jansson, B.O. (Eds). 1995. Biodiversity loss: Sisk, T.D., Launer, A.E., Switky, K.R. and Ehrlich,
Economic and ecological issues. Cambridge, P.R. 1994. Identifying extinction threats. Bio-
Cambridge University Press. science 44 9 : 592-604.
Pimental, D. et al. 1992. Conserving biological Tarasofsky, R.G. 1995 The international forest
diversity in agricultural systems. Bioscience regime legal and policy issues. IUCN The
42 354-362. World Conservation Union. Gland
Pimm, S.L. et al. 1995. The future of biodiversity. Tiffen, M., Mortimore, M. & Gichuki, F. 1994. More
Science: 269 : 347-350. people – less erosion. ACTS Press, Nairobi.
Pomeroy, D. 1993. Centres of high biodiversity in UNEP 1995. Global biodiversity assessment.
Africa. Conservation Biology 7 (4) 901-907. Nairobi.
Reid, W.V. & Miller, K.R. 1989. Keeping options WCMC, 1992. Global biodiversity: Status of the
alive: The scientific basis for conserving earth’s living resources, WCMC Cambridge,
biodiversity. Washington DC; World Re- UK.
sources Institute. Whitmore, T.C and Sayer, J.A. 1992. Tropical de-
Repetto, R. & Gillis, M. 1988. Public policies and forestation and species extinction. IUCN,
misuse of forest resources. Cambridge. CUP. Gland.
Rodgers, W.A. 1993. The conservation of the for- Wilcox, B.A. 1995. Tropical forest resources and
est resources of eastern Africa: Past influ- biodiversity: The risks of forest loss and deg-
ences, present practices and future needs, radation. Unasylva 181 (vol 46) : 43-49.
pp 283-331; in: J. Lovett & S. Wasser; Eds. Witte, J. 1992. Deforestation in Zaire: Logging and
Biogeography and ecology of the rain for- landlessness. The Ecologist 22 (2) 58-64.
ests of eastern Africa. CUP. Cambridge, UK. World Bank, 1994 Strategy for the forest sector
Rodgers, W.A. 1995. Conservation of biodiversi- in Sub-Saharan Africa. World Bank, Wash-
ty: The approaches of the forest and wildlife ington DC
sectors compared. In Biodiversity conserva- World Resource Institute. 1991. World resources
tion in East Africa. (Ed Benun, L.) NMK, 1900-1991. WRI Washington DC.
Nairobi. WWF (US) 1993. African biodiversity: Founda-
Rodgers, W.A. & Homewood, K.M. 1982. Spe- tion for the future. WWF Biodiversity Sup-
cies richness and endemism in the Usambara port Programme, Washington DC.

An estimation of opportunity cost for
sustainable ecosystems

James L. Howard 1

The objective of this study was to provide a method for estimating the opportunity cost of meeting
selected sustainable ecosystem objectives on the Olympic National Forest in Washington State. The
Forest Vegetation Simulator (FVS) was used to simulate several silvicultural management regimes. Soil
expectation values (SEVs) were calculated using the FVS system and CHEAPO II, a silvicultural and
economic analysis system. The SEVs were used first to evaluate different FVS structured silvicultural
alternatives for a 200-year rotation period and then to determine the opportunity cost of managing a
sustainable ecosystem. This approach resulted in a shift of emphasis from the traditional approach of
managing the forest for timber production to managing the forest ecosystem. The findings indicate
reduced revenue from harvesting; the opportunity cost of maintaining certain ecosystem features was
US$ 43 023/ha. This estimate would be lower after deducting the difference in harvesting costs between
alternatives. This work is the first attempt to use and adapt the FVS system to long-rotation forestry.
The methodology used here can be used to estimate sustainable ecosystem objectives on any
forested land.

Keywords: Sustainable ecosystem, opportunity cost, simulation.

Sustaining ecosystems is a key objective in the ecosystem approach to management that the
United States Department of Agriculture (USDA) Forest Service has embraced for National Forest
System lands. This approach shifts the emphasis from managing the forest for timber production to
managing the entire forest ecosystem. The traditional approach altered many habitats, and consequently
many species have been endangered (2). Ecosystem management is, in part, an attempt to save for
future generations plant and animal species that are declining as a result of ecosystem modification.
The primary objective of ecosystem management is to sustain the integrity of ecosystems while
providing goods and services to an increasingly diverse set of public interests.
On 13 April 1994, President Clinton convened a Forest Conference in Portland, Oregon to address
the human and environmental needs served by the Federal forests of the Pacific Northwest and
northern California. The President directed his Cabinet to craft a balanced, comprehensive, and long-
term policy for the management of 10 million ha of public lands. On 1 July 1994, President Clinton
announced his proposed “Forest Plan for a Sustainable Economy and a Sustainable Environment.”
This Forest Plan emphasizes the management of various land and resource allocations to achieve
multiple-use objectives in an economically efficient and environmentally sound manner. The decision

U.S. Department of Agriculture Forest Service, Forest Products Laboratory, Madison, WI, USA.

stipulated the use of a sustainable forest management approach. The planning area was defined as the
Federally administered lands within the range of the northern spotted owl (USDA Forest Service 1990;
USDA Forest Service, Department of Interior Bureau of Land Management 1994). There are 10 million ha
of Forest Service, Bureau of Land Management (BLM), and other Federally administered lands within
the range of the northern spotted owl. The land was allocated to seven different land designations
(Figure 1), each with its own land management objectives.
This paper evaluates certain costs associated with achieving sustainable ecosystem management
on the Olympic National Forest, using silvicultural projection modelling as an analytical tool. The
ecosystem management objective was maintenance of habitat for three species: the northern spotted
owl, pileated woodpecker, and marbled murrelet. The ecological processes that support certain
ecosystem features or endangered species include natural changes that develop and maintain late-
successional and old-growth forest ecosystems (Franklin et al. 1981). To provide discounted returns
of specific intrinsic value demands of the public, the management approach being considered for the
Olympic National Forest is harvest scheduling using the Forest Vegetation Simulator (FVS), a system
of silvicultural and economic modelling (USDA Forest Service 1993). There is growing consensus that
sustainable ecosystem management requires the synthesis of both ecological and economic objectives.
There is also recognition that long-term economic health depends upon environmental health and
vice versa.

Figure 1. Land structure

There are 9.7 million of federal land in the planning area of the decision
(defined as the range of the Northern spotted owl)

Late-successional Matrix
reserves 16% Adaptative
30% management areas

Administratively Riparian reserves

withdrawn areas 11%
Managed late-
Set aside by Act of
successional areas

Source: USDA Forest Service 1994.

The purpose of this study was to provide a method for evaluating the opportunity cost of meeting
selected sustainable ecosystem management objectives. The approach was twofold: (1) determination
of the opportunity cost of maintaining certain ecosystem features to achieve sustainability; and
(2) determination of the capability of the FVS system to estimate opportunity costs.
Opportunity cost of a management choice is defined as the additional value that could be obtained
with the most highly valued of the rejected alternatives or opportunities. In the economic analysis of
forest planning, opportunity costs can be measured by differences in the soil expectation value (SEV)
associated with various alternatives. Alternatives with lower SEVs may be required by legal
interpretations of existing laws such as the National Forest Management Act to manage for an
endangered species, like the marbled murrelet on the Olympic National Forest. The SEV is a measure
of present net value of land devoted to timber production in perpetuity. Opportunity costs are measured
by the change in SEV associated with changes in the output of priced resources, such as timber, and
they can be used to assess the relative value traded off to produce particular unpriced outputs.
Simulations of 200-year rotations were used to determine whether the FVS system can estimate
opportunity costs. SEVs were estimated for different sustainable ecosystem management regimes or
harvest scheduling plans on the Olympic National Forest and were compared to SEV estimates from
other studies.

One of the major objectives of forest management continues to be to produce wood. In fact, this is
often the dominant goal for commercial forests. On public forests, an adequate balance between
timber, recreation, water, and wildlife is almost always required. The non-timber goals of sustainability
and ecosystem management constrain timber production by reducing harvest levels. In many cases,
such as in the western United States, there are legal constraints to harvesting, which have often
resulted from organized attempts to preserve ecosystems and associated wildlife. However, within
these constraints, it is still desirable to organize timber production in the best possible way. Silvicultural
simulation models have been helpful with this task in the past. The research question that is being
explored here is: What would be the value or opportunity cost of sustaining certain features of
ecosystems if sustaining these features requires that the rotation schedules shift from 100 to 200 years
and that some timber values be foregone?

The many value concepts recognized by economists include both market (priced) and non-market
(unpriced) values. Non-market values fall into several categories: consumptive, non-consumptive,
existence, bequest, and option values. Over the years, economists have attempted to find methodologies
for identifying society’s preference for unpriced values. Non-market values are associated with issues
as wide-ranging as recreation, water and air quality, old-growth forest, hunting, and preservation of
endangered species. Numerous methods have been developed to derive monetary estimates of non-
market values. One such method is the opportunity cost method. Growing concerns about nature and
the environment have spurred a number of research initiatives in recent years. One is the attempt to
assign monetary values to a wide range of environmental goods and services, which would allow for
an economic basis for decisions governing resource use to include both priced and unpriced values.
Economic analysis is used to help identify what society might have to be prepared to give up to
maintain a particular form of intrinsic value, which is defined as a value dealing with the inherent right
of other life forms to exist.

The analytical method used in this analysis was that of simulation. It can be applied to many
different problems, some of which have nothing to do with forestry or even with management science;
nevertheless, simulation was designed (and is used primarily) to solve managerial problems. In fact,
simulation was one of the first practical tools devised to tackle complex decision-making problems
common to industry, agriculture, and government. The FVS was adapted to perform the task of
calculating the SEV of wood products from the Olympic National Forest for long rotation schedules.
This was done by taking the 100-year old portion of the Olympic National Forest, specifying a year
zero of 1994, and projecting to the year 2194. In 2194, either a clear-cut or regeneration harvest was
done. The thinnings and harvest activities were specified within the silvicultural component of the
FVS, and SEVs were calculated by CHEAPO II, the economic component of the FVS system. The
thinnings or regeneration harvest were planned to occur primarily on Adaptive Management areas.
These are areas designed to develop and test new management approaches to integrate and achieve
ecological, economic, and other social and community objectives (Figure 1). Nine timber stands were
evaluated in this study; the SEV estimates are averages of those stands.
For the purpose of analysis, two management scenarios were developed for the Olympic National
Forest. The Base scenario does not attempt to preserve habitat for species requiring old-growth
forest. This scenario assumes a clear-cut harvest with commercial thinnings every 30 years from below
or above in advance of final harvest. The per-hectare harvest for the Base scenario is 4 000 ft3, 405 ha/
year, and the rotation period is 100 years. The Base scenario serves as an approximation of recent
forestry practices on National Forests in the Pacific Northwest. The alternate scenario maintains old-
growth habitat. It requires 200-year rotations, which conclude in year 200 (i.e., 2194), with commercial
thinnings or commercial thinnings plus a regeneration harvest at 200 years rather than a clear-cut. This
scenario requires thinnings every 30 years, with an annual harvest of 1.6 million ft3, 1 600 ft3/ha,
405 ha/year. This represents a 60% decline of the annual harvest level from the Base scenario. The
alternate scenario maintains the habitat for the northern spotted owl, pileated woodpecker, and marbled
Harvest scheduling algorithms and associated computer programmes have grown in capability and
complexity to the point that available documented programmes are costly to execute and require
highly trained personnel to formulate problems and interpret results. The FVS was developed to solve
simple to moderately complex problems in harvest scheduling. It is primarily designed to simulate
harvest on a forest of any age.

The FVS is an individual-tree, distance-independent growth and yield model. The basic FVS model
structure contains modules for growing trees, predicting mortality, establishing regeneration, performing
management activities, calculating volumes, and producing reports. The FVS provides the user with
the ability to calculate estimates of forest stand structure and species composition over time. It also
allows the user to quantify this information as necessary in order to ask better questions, and it
simplifies complex concepts of forest vegetation into user-defined indices and attributes for describing
current and future conditions. One strength of the FVS system is its ability to incorporate local
growth-rate data directly into the simulation results. Perhaps the most useful component of the system
is the Event Monitor, a stand treatment module. The Event Monitor is used to develop custom
variables associated with forest species composition, stand structure, wildlife habitat, and other
The FVS extensions include insect and disease models to simulate growth reductions, damage, and
mortality caused by bark beetles, defoliators, dwarf mistletoe, and root disease; a regeneration and

establishment model to incorporate regeneration into a projection; a parallel processor for landscape
analysis where multiple stands are projected simultaneously through time incorporating interstand
effects; a cover model for simulating the development of understorey vegetation; and an event monitor,
which allows management activities to be conditionally scheduled on changing stand conditions.

The analytical results reflect the Base and alternate scenarios. The Base scenario, which does not
preserve habitat for species requiring old-growth forest, reflects harvest levels before the President’s
Forest Plan. These harvest levels are higher than harvest levels in the alternate scenario because there
are no environmental constraints. Historical annual harvest levels for the Olympic National Forest
were 40 million ft3 before the Plan. These harvest levels were based on old timber management plans,
with a harvest base of about 202 000 ha. The annual harvest levels since 1994 have been 1.6 million ft 3.
The reduction in harvest levels is primarily due to three factors: (1) allocations for spotted owl habitat,
(2) greater emphasis on recreational and environmental amenities, and (3) increased allocations to
late-successional and riparian reserves. These factors reduced the harvestable land base from 142 000
ha in 1990 to 20 200 ha in 1994. The economic analysis summary for SEV ($/hectare) showed an average
SEV/ha of US$ 68 626, without deducting logging costs (Table 1). When a range of logging costs (US$
426–US$ $1 409/103 ft3) was included, the SEV ranged from US$ 22 682 to US$ -106 843/ha (Table 3).
This reflects the higher harvest level before the President’s Plan. The SEV estimates were calculated
using average stumpage values expressed in terms of dollars per thousand cubic feet per hectare. The
sensitivity analysis was done for the discount rate as well as logging costs. Intuitively, the results
show the model to give predicted results. The SEV estimates reflect the average of high, medium, and
low site values.
The alternate scenario, which maintains old-growth habitat, reflects harvest levels after the Forest
Plan that were constrained as a result of environmental restrictions; less timber was harvested. After
the President’s Plan, the average SEV (US$/ha) was US$ 25 602 (Table 1). The opportunity cost before
adjusting for harvesting costs (US$ 43 023/ha) was derived by taking the difference between the two
scenarios. The estimate would be lower after deducting the difference in logging costs between the
scenarios. This amount indicates what dollar value per acre is given up when the forest is managed for
endangered species. When clear-cuts or regeneration harvest were applied at the end of the rotation,
little value was added to the per acre value of the opportunity cost (Table 2). When discounting over
200 years, value-added in year 200 contributed very little to the SEV.
The SEVs reflect that the FVS can simulate 200-year rotations adequately. Table 4 compares SEVs
to estimates of the value of endangered species made in other studies.

The Forest Service is responsible for determining how to best manage National Forest lands, based
on consideration of both public desires and production capabilities. Arriving at this determination is
an extremely complex process, more so than ever before because of the legal constraints to management.
These constraints have been brought about by various environmental groups through conservative
interpretation of the law. Through such legal interpretations, groups have sought to curtail logging on
the National Forest System in favour of old growth and protection of animal species. A wide range of
public desires and concerns must be interfaced with limited resource production capabilities to develop
possible answers to land management questions. Public desires have become widely varied and often
compete with each other. For example, one segment of the public may wish to have maximum timber
harvest opportunity, while another may desire that old-growth timber stands be preserved as wildlife
habitat. Production capabilities are also complicated and are often mutually exclusive. A given land

Table 1. Summary of soil expectation values (SEVs) ($/ha) 1
Year Before President’s After President’s Opportunity
forest plan forest plan cost
2194 68 626 25 602 43 023
Discount rate = 4%.

Table 2. Clear-cut SEVs ($/ha) 1

Clear-cut without Clear-cut with
periodic thinnings periodic thinnings
Year 2024 2194 2024 2194
Average 119 761 425 27 772 29 549
Discount rate = 4%
Note: Data may not reflect accurate stocking.

Table 3. SEV sensitivity analysis of individual stand in Base scenario ($/ha)

Year Before President’s Discount rate Logging cost
forest plan
(SEV 4%) 2% 8% Low range High range
($426) ($1,409)
2194 $85 104 $167 860 $57 548 $22 682 -$106 843

Table 4. Comparison of existence and opportunity cost values for northern spotted owl habitat
Study Resource valued Value Comment
Rubin et al. (1991) Northern spotted owl $1 500 million/year Value to U.S. households of protecting
habitat protection northern spotted owl habitat

Hagen et al. (1991) Northern spotted owl $48-$190/household 81% of U.S. households
habitat protection in Pacific Northwest favor protection of old growth and
northern spotted owl

Loomis & White (1996) Northern spotted owl $70/household Willingness to pay estimates per
in Pacific Northwest household for protecting northern
spotted owl

Howard (in preparation) Northern spotted owl $43 023/ha Opportunity cost estimate before
adjusting for logging costs

area can provide timber for harvest or old-growth habitat, but not simultaneously. Because of the
complexity of both desires and production possibilities, the number of solutions to land management
problems is virtually infinite.
The principal public desires relating to management of the Olympic National Forest are embodied
in the issues and concerns of the Northwest Forest Plan; i.e., how much timber can be produced while
protecting old-growth timber and not endangering animal species? There are other concerns such as

scenic quality and unroaded recreational opportunity. Although there is a need to examine a wide
range of output mixes, this report is concerned only with harvesting of old-growth timber and
maintenance of endangered species. Because of legal restrictions, harvesting of old-growth timber on
the Olympic National Forest has been curtailed. The opportunity cost reflects the level of timber
values foregone to manage for old-growth and endangered species.
The soil expectation values (SEVs) reflect that the Forest Vegetation Simulator (FVS) can simulate
200-year rotations adequately. It is important to note that the SEV estimates are based on projecting
what would happen over an idealized hectare, starting with 100-year-old timber stands, with no interaction
with what might happen on the surrounding hectares.
The variety of public desires and the complexity of resource production interrelationships combine
to make the problem of developing a comprehensive management strategy exceedingly complicated.
Fortunately, the means for reducing the problem to manageable proportions are available. Advances
in mathematical modelling, such as the FVS system, have made it possible to represent extremely
complex problems in a relatively straightforward and understandable form.


Franklin, Jerry F. et al. 1981. Ecological charac-

teristics of old-growth Douglas fir forest. management plan, Olympic National For-
General Technical Report PNW 118. U.S. est. U.S. Department of Agriculture, Forest
Department of Agriculture, Forest Service, Service, Pacific Northwest Region.
Pacific Northwest Forest and Range Experi- USDA Forest Service. 1993. Forest Vegetation
ment Station, Portland, OR. simulator. U.S. Department of Agriculture,
Hagen, D., Vincent, J., & Welle, P. 1991. The ben- Forest Service, Washington Office Timber
efits of preserving old-growth forest and the Management Unit, Fort Collins, Colorado.
northern spotted owl. Contemporary Policy USDA Forest Service, Department of Interior
Issues 10:13–26. Bureau of Land Management. 1994. Record
Loomis, J. & White, D. 1996. Ecological econom- of decision, standards and guidelines. U.S.
ics. Journal of International Society for Department of Agriculture Forest Service,
Ecological Economics, pp. 197–206. Department of the Interior Bureau of Land
Rubin, J., Helfand, G., & Loomis, J. 1991. A bene- Management, Washington, DC.
fit cost analysis of the northern spotted owl.
Journal of Forestry 89:25–30.
USDA Forest Service. 1990. Land and resource

Study on method for regional ecosystem
biodiversity assessment

Shi Zuomin, Cheng Ruimei and Jiang Youzu 1

In order to provide objective criteria for ecosystem conservation at a regional level, six principles of
assessment and a hierarchical index system were put forward. The index system includes diversity,
peculiarity, representativeness, rarity, stability, naturalness, conservation value, accessibility, human
interference, natural calamity and current conservation conditions. By quantifying indices, a relatively
practical and controllable method of quantitative assessment has been established.

Keywords: Region, biodiversity, ecosystem, priority of conservation, quantitative assessment

Biodiversity conservation is one of the most important ecological issues. Projects on biodiversity
conservation have been carried out in some countries by international organizations (Jiang Youxu and
Liu Shirong 1993). Chinese biodiversity is very rich, but it is seriously threatened (Chen Lingzhi); so
it is necessary and important to protect biodiversity in China. Biodiversity conservation is a social
exercise and and requires much funding. China is a developing country so conservation in key regions
of species with limited funds is a serious problem. An effective way to solve this problem is to assess
the degree of endangerment and conservation scientifically. Currently, many studies concentrate on
the species level (Wei Hongtu & Jin Nianci 1994; Xu Zaifu & Tao Guoda 1987) while methods for
regional ecosystem assessment are few. However, the conservation of the ecosystem is more important
than the conservation of single species (Jiang Youxu & Liu Shirong 1993; Gordon 1993; Naveh 1994;
Jerry 1992). Because the natural characteristics of ecosystems are different and cannot be compared
easily, this paper establishes a relatively practical and controllable method of quantitatively assessing
ecosystems of the same category in different regions.

Principles of assessment are the bases of indices selection, and they scientifically guide the
establishment of indices.. The principles considered are as follows. (1) Index selection must be based
on ecological theory. In other words, ecosystem diversity assessment is an ecological one, so the
indices should have certain ecological meanings; their selection must be based on ecological theory,
especially the theory of ecosystem ecology and community ecology. (2) Indices should be comparable.
Because the objects of assessment are ecosystems of the same category in different regions, the
selection of comparable indices is the technical basis for their comparison and assessment. (3) Indices

Institute of Forest Ecology and Environmental Science, Chinese Academy of Forestry, Beijing, P.R.China.
Facsimile: (86-10) 258 4229

should be practical and controllable. The purpose of the assessment is to conserve the assessment
objectives effectively. The selection of practical and controllable indices can provide maximum
information on the conservation objectives and conservation strategies, which can be used for
improving the function of conservation measures. (4) Indices should be representative. In order to
carry out the assessment scientifically, the indices must represent the natural and social characteristics
and conditions of interference and conservation objectives. (5) Indices should be obtained and
quantified easily. Because assessment work is very complex and necessary, in order to complete the
work quickly and obtain good-quality results, indices should be easily accessible. Since the assessment
is a quantitative one, the quantification of indices is an important step of the assessment procedure as
it is the basis for the selection of quantitative indices. (6) Indices should also consider certain social
aspects. Many indices of assessment of ecosystem diversity take into account natural and ecological
characteristics but, since the procedure of assessment also has relative social characteristics, some
indices must also cover this aspect.


In accordance with the above principles, an index system that includes diversity, peculiarity,
representativeness, rarity, stability, naturalness, conservation value, accessibility, human interference,
natural calamity and conservation conditions was established. The hierarchical structure of the index
system is as follows:


Calculation of the weight of indices and their grading

For the purpose of ecosystem conservation, the importance of each index is different. A key step
to assess each one correctly is to identify and qualify their weights. Analysis of Hierarchy Process
(AHP) is a kind of mathematical method to calculate weight of index, and it is generally accepted and
widely applied in many fields (Zhao Huanchen et al. 1986; Jussi 1994). The weight of each index using
this method can be calculated by AHP. In order to assess the objects quantitatively, we grade the
minimum level indices. All of them can be divided into five gradings. Indices and the bases of gradings
are shown in Table 1.
The gradings and their quantitative value can be obtained through surveys by experts. In order to
make the quantitative value of indices consistent with the priority of conservation, negative values of
resistance stability, resilience stability and conservation were used.

Assessment method
On the basis of the weight value of each index and its quantitative value, the assessment values of
the objectives considered were calculated with the method of weighted average. The conservation
level was identified according to the assessment value of each objective. The higher the assessment
value, the higher the priority degree of conservation.
Assessment value can be calculated with the formula below:

Figure 1. Hierarchical structure of the indices system

habitat diversity A1
structure diversity A2
Diversity A species abundance A31

species diversity A3
species relative abundance A32

habitat peculiarity B1
Peculiarity B community peculiarity B2
species peculiarity B3

habitat representativeness C1
Representativeness C community representativeness C2
species representativeness C3
habitat rarity D1

Ecosystem Rarity D community rarity D2

species rarity D3

Stability E resistance stability E1

resilience stability E2

Naturalness F

scientific value G1
Conservation value G social value G2
economic value G3

Accessibility H

Human interference I direct interference I1

indirect interference I2

Natural calamity J
Conservation condition K

Assessment value = aWa +bWb +cWc +dWd +eWe +fWf +gWg +hWh +iWi +jWj +kWk

3 3 3 3
In the formula: a= Σ(anWan) b= Σ(bnWbn) c= Σ(CnWcn) d= Σ(dnWdn)
n=1 n=1 n=1 n=1

2 3 2 2
e= Σ(enWen) g= Σ(gn Wgn) i= Σ(inWin) a3 = Σ(a3nWa3n)
n=1 n=1 n=1 n=1

Wa ,Wb ,Wc , Wd , We , Wf , Wg , Wh , Wi , Wj , Wk ,Wen , Win , Wa3n (n=1 2); Wan , Wbn , Wcn , Wdn , Wgn
(n=1...3) are the weight values of indices; f, h, j, k; bn , cn , dn , gn (n=1 3); an , en , in , a3n (n =1...2) are
quantitative values of indices gradings.

Table 1. Indices and bases of gradings

Indices Bases of gradings

habitat diversity A1 number and area of habitats and the heterogeneous degree of them
structure diversity A2 complex degree of spatial and trophic structure
species abundance A31 type and number of species in different regional scale
species relative abundance A32 type and relative number of species in different regional scale
habitat peculiarity B1 peculiarity degree of habitat in different regional scale
community peculiarity B2 peculiarity degree of community in different regional scale
species peculiarity B3 peculiarity degree of species in different regional scale
habitat C1 representativeness degree of habitat in different regional scale
community representativeness C2 representativeness degree of community in different regional scale
species representativeness C3 representativeness degree of species in different regional scale
habitat rarity D1 rarity degree of habitat in different regional scale
community rarity D2 community in different regional scale
species rarity D3 rarity degree of species in different regional scale
resistance stability E1 resistance ability of interference
resilience stability E2 resilience ability after interference
naturalness F degree of human interference and invasion of exotic species
scientific value G1 scientific value of evolution, succession, climate change et al.
social value G2 function of education and human entertainment
economic value G3 direct and indirect utility value of ecosystem and its component
accessibility H distance from city and village, convenient degree of communication
direct interference I1 utility degree of natural resource in ecosystem
indirect interference I2 utility degree and environmental pollution of ecosystem boundary
natural calamity J intensity, frequency, continued period of natural calamity
conservation K conservation measure, conservation fund and research level

There are many problems in biodiversity conservation research, and there are few reports about
assessment of an endangered ecosystem and its conservation. It is very complex and comprehensive
work since indices selection and their quantification are applied to different assessment objectives.
Some indices use quantitative methods, such as species diversity, structure diversity and habitat
diversity (Ma Keping 1994; MacArthur et al. 1961; Elton et al. 1954), but these methods are not
consistent. In order to solve this problem, we first describe the indices qualitatively, then grade and
quantify them on the basis of analysis of their objective characteristics, although there may be problems
in grading them, such as for their degree of naturalness (Marglus et al. 1981). Therefore, grading of
indices is a key step in this method and it needs comprehensive investigation by an expert. In
conclusion, this research is of an exploratory nature only since problems still exist in indices selection,
grading methods and relative theory. However, we expect that it will be of interest to researchers in this
field and be further developed in future practice.

Bibliography Ma Keping. 1994. Measurement of biodiversity,
Chen Lingzhi. 1993. China biodiversity: Current in principles and methods of biodiversity
situation and conservation strategies, Bei- conservation research, Beijing, Chinese sci-
jing, Science Press, 1-10. ence and Technology Press, 141-165.
Elton, C.S. & Miller, R. S.1954. The ecological sur- Marglus, C.R. & Usher, M.B.1981. Criteria used in
vey of animal communities: with a practical assessing wildlife conservation potential: A
system of classifying habitats by structural review. Biology Conservatio, 21 :79-109.
characteristics. Journal of Ecology, Naveh, Z., 1994. From biodiversity to ecodiversi-
42:460-496. ty: A landscape ecology approach to con-
Orians, Gordon H. 1993, Endangered at what lev- servation and restoration, Restoration Ecol-
el? Ecological Applications, 3(2):206208. ogy, 2(3):180-189.
Franklin, Jerry F. 1992. Preserving biodiversity: Wei Hongtu and Jin Nianci. 1994. Quantitative
Species, ecosystem, or landscapes? Ecolog- analysis of endangered size of Shanioden-
ical Applications, 3(2):202-205. dron subaequale M.B. Deng, H.T. Wei & X.Q.
Jiang Youxu and Liu Shirong. 1993. Some prob- Wang, Journal of Plant Resource and Envi-
lems about conservation research of region- ronment, 3(3):1-8.
al biodiversity, Journal of Natural Resource, Xu Zaifu & Tao Guoda. 1987, Discussion on the
8(4):289-298. method of systemic assessment to regional
Kunsipalo, Jussi. 1994. Managing biodiversity in threatened plants and their prior conserva-
a forestry environment, Conservation Biol- tion. Journal of Yunnan Plant Reseorch,
ogy, 8(2):450-460. 9(2):193-202.
MacArthur, R.H. & MacArthur, J.W. 1961. On bird Zhao Huanchen. 1986. Hierarchy analysis meth-
species diversity, Ecology, 42:594-598. od. Beijing, Science Press, 1-170.

Conserving the ecological heritage – sacred
groves of Tamil Nadu

Nanditha Krishna 1 and V. Bhavani Shankar 1

The sacred groves of Tamil Nadu are the home of the local flora and fauna and represent a mini
biosphere reserve, making them an essential part of the conservation process. The rich plant life helps
to retain subsoil water and, during the hot summer months, the pond in the grove is often the only
source of drinking water. The groves are a unique form of biodiversity conservation, and are living
examples of the Indian tradition of conserving the ecology as a natural heritage.
In 1993, the C.P.R. Environmental Education Centre commenced a survey of the sacred trees and
sacred groves of Tamil Nadu, including the restoration of 13 sites to their former biological potential.
The sacred groves of Tamil Nadu are a part of the local religion and are protected by strict rules and
The sacred groves of Tamil Nadu are small in size, ranging from 1/2 ha to 20 ha, except in those
places where they are a part of the Western Ghat mountain chain, where they may extend to several
hundred hectares. The climax vegetation is Albizzia amara community.
The paper describes in detail the conservation of the sacred groves by the local people and also
traces the role of terracotta, representing Mother Earth, in the worship of the deity within the grove.

Keywords: Sacred groves, terracotta, restoration, biodiversity, Albizzia amara.

The ecological heritage of the southern Indian state of Tamil Nadu has two living traditions: the
veneration of the sthala vriksha or the sacred tree, to be found in every temple, and the kovil
kaadugal or sacred groves, to be found in every village. In 1993, the C.P.R. Environmental Education
Centre commenced a survey of the sthala vrikshas and sacred groves of Tamil Nadu, including the
restoration of a few of the latter.
The sthala vriksha involves the worship of a single tree which represents a species which has or
once had an important economic status locally, thereby ensuring its linkage to the local deity, while the
sacred grove is a protected area in the village. Sanskrit and Tamil literature are full of references to the
groves where wise and holy men lived. But the tradition probably goes further back in time, to food
gathering societies who venerated nature and the natural resources on which they depended for their
The sacred groves of Tamil Nadu are a part of the local folklore and religion. Every village has a
grove, a protected area associated with local folk deities of obscure origin. In the middle of the grove
there is generally a shrine of Amman or the Mother Goddess in one of her many forms – Kaali, Maari,
Pidaari, Ellai, etc. In front of the shrine is a pond, either natural or artificial. Surrounding the shrine and
the body of water are the male consorts of Amman, generally regarded as spirits or deities –

C.P.R. Environmental Education Centre, 1 Eldams Road, Madras 100 018, India. Facsimile: (91 44) 450656

Karuppusamy, Muniyaandi, Muneeswaran, Madurai Veeran – and their bodyguards or warriors. The
people of the village make votive offerings of terracotta horses, bulls or elephants to Ayyanaar, the
mythical watchman of the village whose statue is also consecrated in the grove, in fulfilment of
prayers for a good harvest or good health. The priest at the village shrine is the potter, who represents
the cyclic powers of mother earth.
The sacred grove, in Tamil Nadu, is an essential part of the village. Every village has at least an acre
of land dedicated to the grove, the shrine of the mother goddess and Ayyanaar, where people make
their offerings of terracotta horses. The horses range from 12 inches to 20 or more feet in height,
depending on the district, local practice and financial situation of the devotees. The gods of the grove
do not conform to the classical Hindu pantheon, although they are equally revered by all hindus.
The sacred groves are the home of the local flora and fauna and represent a mini biosphere reserve,
making them an essential part of the conservation process. The rich plant life helps to retain subsoil
water and, during the hot summer months, the pond in the grove is often the only source of drinking
water. The groves are a unique form of biodiversity conservation, and are living examples of the Indian
tradition of conserving the ecology as a natural heritage.
The sacred groves are protected by strong local taboos which protect the entire ecosystem. Most
are surrounded by thorny bushes or rocks and are insulated from the activities of the village (King,
Narasimhan, Viji 1996). It is interesting to note that many sacred groves are also important archaeological
sites with evidence of palaeolithic or neolithic cultures. Sittannavasal in Pudukottai district, for example,
combines 3000 year old neolithic dolmens, ancient sacred groves, an ancient water tank, 1500 year old
caves of Jaina monks and 1300 year old caves with beautiful paintings. Such situations are found
frequently all over Tamil Nadu.
The conservation of the sacred grove by the local people involves strict rules and taboos:
• Tree felling is strictly prohibited.
• No animals, small or large, can be hunted within the grove.
• Local people are permitted to remove leaves, fruits or roots only in cases of sickness and ill health.
Even children are taught not to pluck the fruits or berries.
• The area is regarded as sacred and visitors must remove their footwear before entering the grove.
• On a festival day or during the dedication of a clay horse, villagers are permitted to come here and
cook pongal (a mixture of rice, lentils and jaggery) as an offering to the deities of the grove. The
food is then shared by all the villagers. Except on festival days or in fulfilment of a vow, villagers
do not enter the grove.
• Transgression of the above taboos may cause crop failure, sickness in the family or diseases to
the livestock.

The role of terracotta is very important. The figures must be made of clay, which represents the
powers of renewal inherent in the earth - from the birth of new plants to animal and human offspring.
It is carried over to the Hindu philosophy of birth, death and rebirth - the soul takes a new life after the
decay of the body just as a new plant is born after the death of the old. This is also the cyclic role of
the clay - it represents the horse, etc., for a certain time: as it slowly disintegrates and goes back to
mother earth, it is time for the creation of a new figure.
In fact, the new figure is often made from a handful of clay from the old figure to which more clay is
added. The main figures of the mother goddess and the male deities must be “renewed” every one or
two years, hence they were never, traditionally, made of any other material besides clay. The entire
phenomenon is closely associated with prayers for fertility.
The votive offerings - the horses, bulls, elephants and ram - are always made of clay and left in the
open to go back to the mud they came from. It is interesting to note that only images of domestic

animals are given as votive offerings and never images of the wildlife which visit the groves. Sometimes,
the worship of the deities within the grove may involve the sacrifice of a goat or rooster.
The potter is the priest at the sacred grove. The potter performs both the ritual of making the
terracottas and the ritual of worship at the temple, before the clay figures are offered to Ayyanaar
(Krishna 1992). He belongs to the caste known as Vishvakarma - “creator of the world”. His tools are
few - the potter’s wheel and his own hand. For figurines he uses a mixture of sand, husk and clay,
unlike the mixture of sand and clay used for pots (Inglis 1980). But the offering must be installed in a
grove, under a tree. In time, the grove gets cluttered with clay images of gods, goddesses and animals,
particularly horses.
Why the horse? Because, say the people, it is considered next in importance only to man. Perhaps
its importance dates back to the period of the Rig Vedic ashvamedha (2500 BC), where the territory
covered by the horse as it roamed for a year was claimed by the tribe (Shah 1985). The making of a
terracotta horse is a matter of great significance and the dedication of the horse is a major event
involving the whole village. Sometimes the horse is replaced by a bull or elephant, the last being the
favourite offering of fishing communities.
The sacred groves of Tamil Nadu are small in size, ranging from 1/2 hectare to 20 hectares, except in
those places where they are a part of the Western Ghat mountain chain, where they may extend to
several hundred hectares. The largest identified so far is in Tirukkurungudi, Nanguneri Taluk, Nellai
Kattabomman District, which is as large as 494 ha. But the average sacred grove in the plains is about
2 ha. Most of these groves are in fairly intact condition.
However, with the increasing pressure on land and gradual encroachments, some have diminished
in size over the years. The maximum number of sacred groves are to be found in Trichy District
followed by Pudukottai and Nellai Kattabomman districts.
Meher-Homji (1986) is of the opinion that the classification proposed by Prof. G. Champion in his
book A Preliminary Survey of the Forest Types of India on scrub jungles of the Coromandel Coast as
Tropical Dry Evergreen Forest may be reclassified as Albizzia amara community. According to
Meher-Homji, the Coromandel Coast was not typically tropical but disymetric, the climate of the
region was not so dry. He also states that the word “forest” is also not appropriate as these groves
have short-sized stature of trees and are few in number. They are not evergreen in nature, as more than
50% of them are deciduous.
The entire state of Tamil Nadu receives rain during the north east monsoon, with little or no rainfall
in the months of May and June. The climax vegetation species in the sacred groves in the drier regions
of the plains are predominantly Memecylon umbulatum (ironwood tree), Chloroxylon swietenia (east
Indian satinwood), Albizzia amara (siris), Glycosmis cochinchinensis, Capparis divaricata (capparis
bush), Gmelina asiatica (small Cashmere tree), etc. However, species like Atlantia monophylla (Indian
wild lime), Tenminalia glabra (hardwood tree), Zizyphus nummularia, Terminalia arjana (arjun),
Hardwickia binnata, Santalum album, (sandalwood) are found in the groves of Tirunelveli, Kamarajar
and Dindigul Anna districts, especially in higher altitudes adjoining reserve forests. Aerva tomentosa
is found only in the Tirunelveli and Kamarajar districts. This plant is endemic only to the plains of the
Tirunelveli District. Flacourtia ramontchi is found in the sacred groves at higher altitudes.
As part of the C.P.R. Environmental Education Centre’s environmental education programme, the
importance of protecting and conserving the existing sacred groves has been recognized and the
message of conservation has been taken to the masses through various folk media and modern
communication methods, apart from actual restoration endeavours. The film Vanadevatai made by the
Centre is a part of this effort. The Centre has also set up a field office in Pudukottai to carry out this
work. It is a matter of satisfaction to report that whenever a grove has been taken up for restoration,
the residents of adjoining villages have come forward with requests to restore their village groves,

often involving only the planting of saplings and technical advice. The spill-over effect of the Centre’s
Conservation and Restoration of Sacred Groves project has been very encouraging. The limited size
of the area to be restored has made it manageable by small local communities as water, watering,
bio-fencing and watch-and-ward requirements are minimal. The saplings chosen by the villagers have
been traditional species. It is also interesting to observe that none of these restoration programmes
have been destroyed by grazing cattle or goats, as the villagers zealously prevent the animals from
entering the groves and have volunteered to fine those who violate this norm. Table 1 gives the list of
sites taken up for restoration.
However, it is felt that the sacred groves should not be brought under the ownership of the
government - state or central - or the forest departments, as it would alienate the people who are
currently preserving these groves and the present community conservation and restoration efforts
may vanish, as in the case of our tribals who have been alienated from the forests which they once
cherished and preserved. Rather, through appropriate education and the strengthening of local
communities, the last remnants of localized rural biodiversity found in the sacred groves can be
preserved for posterity. The loss of such vital natural vegetation would lead to the loss of genetic
resources which are of extreme economic importance.
Sacred groves have been conserved out of religious sentiment by local people. This natural bounty
has helped their survival through the ages. People and nature have coexisted without disturbing the
environment in the past. Such traditional practices have to be strengthened with appropriate scientific
inputs for conservation, to help people live harmoniously with nature in spite of the present age of
rapid and unsustainable development.

The authors would like to acknowledge the contribution of Mr. M. Amritalingam, Mr. R. Selvapandian,
Mr. G. Ravikumar and Mr. S.R.. Durai Murugan for assistance in the Survey and Restoration of Sacred
Groves in Tamil Nadu.

Inglis, Stephen. 1980. A village art of South In- Tamil Nadu. Mapin Publishing Pvt. Ltd., New
dia. Madurai Kamaraj University, p.36. York, p.90.
Israel, Oliver et al. 1996. Socio-biological aspects Meher, Homji V.N. 1986. Puttupet: A acred
of sacred groves of different ecological termite-mound protects a forest. Blackbuck
zones of tamil Nadu. Conserving Our Eco- Madras, Vol.I., No.4., p.l-4.
logical Heritage - The Tamil Tradition. Sem- Shah, Haku. 1985. Form and many forms of moth-
inar Papers, Madras, p.4. er clay. National Crafts Museum, New Delhi,
Krishna, Nanditha. 1992. The arts and crafts of p.66.

Table 1. List of sites taken up for restoration of Sacred Groves in Tamil Nadu
Site Name of the Deity In Hectares (Approx.)
District: Chengai MGR
Iyanery Kannamadai Ayyanar 1.01
Nenmeli Marriamman 1.61
Veeranathur Shiva 0.60

District: Madurai
Papappatty Occhandamman 2.22

District: Nellai Kattabomman

Perunkottur Thirukoti Ayyanar 1.21

District: North Arcot Ambedkar

Sirunamalli Alavattamman 0.80

District: Pasumpon Muthuramalinga Thevar

Kollangudi Vettiudaya Ayyanar/ 1.21
Vettiudaya Kaliamman

District: Pudukkottai
Annavasal Urali Karuppar 0.80
Chitthannavasal Karuppuswamy 1.21
Illayavayal Ayyanar 2.42

Karambakkudy Sri Muthu 1.61

Karuppaiya Swamy

Vadavalam Punugu Karuppar 1.61

& Porpanai Muneeswarar

District: Tiruvannamalai Sambuvarayar

Pavupattu Kannamadai Ayyanar 1.01

Forest biodiversity conservation in India and
Germany: a comparative analysis

S.S. Negi 1, B. Stimm 2

Biodiversity or biological diversity pertains to the diversity of biological organisms, both animals
and plants in a region. In 1992, the Earth Summit at Rio (Brazil) laid stress on the biodiversity over the
globe and the need to preserve it for posterity. Biodiversity conservation has drawn the attention of
governments all over the world. Both developed and developing countries are engaged in conserving
their biodiversity. This paper attempts to draw comparisons between the biodiversity conservation
measures in India (a developing) and Germany (a developed country).


Biological Diversity
India has a rich and varied assemblage of living organisms. There are over 40 000 plant species in
India of which nearly 2 500 are trees (Negi 1994). This accounts for about 12% of the global plant
wealth. Amongst the 21 000 species of flowering plants found in India, 5 000 are woody plants. Almost
a third of plant species of India are endemic and not found anywhere else in the world (Tewari 1992).
The faunal wealth of India is also equally rich. There are over 75 000 species of animals of which about
60 000 are insects; 1 693 fishes; 3 000 birds and 372 are mammals. In addition to this, the marine life in
the shelf zone of over 45 million ha is very rich and varied.

Forest Wealth
According to the Forest Survey of India report (Anon 1993), India has a total natural forest area of
640 107 km2 which is 19.47% of the total geographic area of the country. India’s natural forests can
broadly be grouped into the following :

Forest group/type % of forest area

1. Tropical wet evergreen forest 8.0
2. Tropical semi evergreen forest 4.1
3. Tropical moist deciduous forest 37.0
4. Tropical littoral and swamp forest 0.6
5. Tropical dry deciduous forest 28.6
6. Tropical thorn forest 2.6

Indian Council of Forestry Research & Education, PO New Forest, Dehra Dun 248006, India. Facsimile: (91 135)
623258; E-mail: in/
Forestry Faculty, University of Munich, Freising Munich, Germany.

(Cont’d) Forest group/type % of forest area
7. Tropical dry evergreen forest 0.02
8. Subtropical broadleaved hill forest 0.4
9. Subtropical pine forest 6.6
10. Subtropical dry evergreen forest 2.5
11. Montane wet temperate forest 3.6
12. Himalayan moist temperate forests 3.4
13. Himalayan dry temperate forests negligible
14. Subalpine and alpine forest 2.9

Loss of biodiversity
Causes : The following are the causes of biodiversity loss in India over the past several centuries:
1. Heavy removals of fuelwood and fodder;
2. Clearance of forest areas for agriculture, human settlements and flooding by reservoirs;
3. Overexploitation of forest resources;
a) According to the Red Data book of IUCN, 27 mammals of India are rare and threatened with
b) Over 800 plant species of India are either extinct or threatened with extinction (BSI report).

Biodiversity conservation
India has 77 national parks and 480 wildlife sanctuaries in which biodiversity conservation is the
main activity. These areas cover about 14.03 million ha or 4.2% of the total geographical area of the
country. There are 309 preservation plots covering about 8 500 ha; 1 905 sample plots in different
forest types and 537 protected trees for in situ conservation of plants. However, all forest ecosystems
have not been covered. Ex situ conservation is being done in 55 botanical gardens and 109 zoological

Introduced biodiversity
Extensive plantations, mainly of exotic species have been raised in India. These include eucalyptus,
exotic poplars and Australian acacias. However, they form an insignificant proportion of the total
forest area.

What needs to be done

In spite of the efforts in recent decades, much needs to be done for conserving India’s biodiversity.
Some measures are:
1. Creating a network of special reserves for conserving at least two representative areas of different
2. Generating public awareness and interest in biodiversity conservation;
3. A fool-proof system for monitoring biodiversity conservation;
4. Sustainable use of components of biodiversity;
5. Impact assessment and minimising adverse impacts;
6. No forest working in representative ecosystems;
7. Further development of the network of national parks, sanctuaries and biosphere reserves.

In comparison with India, Germany is characterized by a sparse assemblage of living organisms.
There are about 2, 00 species of flowering plants (including approximately 50 tree species). This is only
1% of the global wealth on flowering plants. Faunistic accounts give an amount of 76 mammal species,
237 bird species, 32 species of reptiles and amphibians, and approximately 30 000 insect species
(WCMC 1992).

Forest data
The total forest area of Germany is 104 330 km2 which is 29.2% of the total geographic area (StBA
1995). The forests of Germany belong to the forest biome type of cool temperate deciduous forests.
This characterization pretends a monotonous appearance. Despite this the potential natural forest
types reconstructed by vegetation scientists show the variety as follows:
1. Broadleaved (beech) hill forest 42.2%
2. Broadleaved (beech) montane forest 23.5%
3. Broadleaved (oak) forest 24.8%
4. Riparian and swamp forest 8.0%
5. Dry temperate pine forest 1.0%
6. Subalpine (spruce) forest 0.5%
In view of the wide range of definitions and (detailed) classifications on “forest types” (e.g., “forest
vegetation types”), which can lead to some confusions, the account mentioned above is very simplistic.
Nevertheless it facilitates the intercomparison of data between the two countries. The data on potential
natural forests indicate a predominance of broadleaved deciduous tree species, especially of beech.
The overall contribution of broadleaved species to natural forests has been estimated at 90%, the
contribution of coniferous species being 10%.
But, in fact, forests in Germany are the results of human impacts over thousands of years, especially
from the Middle Ages onward, when large areas of natural forest ecosystems had been exploited,
converted into agricultural land and forest plantations. Today forests exhibit a share of more than 70%
of coniferous species in the composition of forests. This gives a small impression on the extent of
human activities, including forestry, during the past centuries.
Sustainable forestry was developed during a very long period in middle Europe (mainly in German-
speaking regions). The uncontrolled (over)exploitation of forests was already counteracted by
regulations in the Middle Ages. In the beginning of the 18th century the term “sustainability” made its
way as a tenet in the management of forests. At first “sustainability” referred only to the sustainability
of wood production, but in the middle of the 19th century “sustainability” encompassed other functions,
too, like the protection of soils, watersheds and other forest resources. Today sustainability means
the preservation of forests as a natural system. Therefore sustainable forestry includes a conservation
use of nature and integrating the conservation related objectives in German forestry is of growing
importance. Thus, sustainable forestry seems to be on the way to a perfect management system.

Loss of biodiversity
1. Conversion and fragmentation of large forested areas caused by human settlements, agriculture,
and industrial facilities;
2. Impact of industrial pollution on forest ecosystems;
3. Extension of the distribution of conifer forests during the past three centuries, managed as high
forest and characterized by even-aged and more or less uniform stands.

According to WCMC (1992) 2 mammals, 17 bird species and 49 plant species are threatened with
extinction. However, it is worth mentioning that not all of them are linked with forest ecosystems.

Biodiversity conservation
Germany has 12 national parks and 5 171 nature reserves. These areas cover 13 762 km2 or 3.8% of
the total geographical area of the country. There are also 12 biosphere reserves (> 11 700 km2, 3.2%),
85 natural parks (> 56 000 km2, i.e. > 16%) with different levels of conservation intensity and 6 700 km2
of marshlands of international significance (StBA 1995).
There are about 580 natural forest preservation plots covering more than 16 500 ha for in situ
conservation of plants and animals (StBA 1995), but still not all forest ecosystems (forest vegetation
types) are represented with sufficient coverage.
Ex situ conservation is being done in 73 botanical gardens and 40 zoological parks. Furthermore ex
situ conservation of forest genetic resources takes place by means of seed orchards, clone collections
and conservation of seeds, pollen, plants, and tissues in gene banks.

Introduced biodiversity
About 20 exotic tree species were introduced to forestry in Germany since the middle of the last
century. With the exception of Douglas fir, which covers about 1.5% of the total forest area, most of
them are of minor or negligible importance in forestry.
What needs to be done: some proposals for solutions in forestry:
1. An important goal of nature conservation is to protect various forest ecosystems. Biodiversity
conservation in large scale could be realized in national parks with no management operations.
But, only national parks could not avoid the loss of species and habitats due to their small areas.
2. Other areas are needed to conserve forest habitats. This could be done by preservation of
nature forest reserves (special unmanaged areas) with a sufficient amount of representative
areas of different ecosystems (forest vegetation types).
3. Sustainable use of forests outside protected areas has many advantages and should be able to
maintain species richness and diversity. Nature oriented management systems, which are already
practised to some extent by forest landowners and managers on several thousands of hectares
turn away from homogenous, mostly coniferous, forest forms to site adapted mixed broadleaved
forests, which are assumed to exhibit advantages in ecological stability under the ongoing
changes in environmental and climatic conditions (e.g. global change). Those and related
management concepts are characterized mostly by permanent forest forms without intensive
forest operations at a large spatial scale, where natural regeneration is preferred and supported.
Clearcuts, if still practised at all, are not larger than a few hectares. Artificial regeneration favours
autochthonous tree species and adapted provenances, and an adapted tending takes place
(support of species mixtures and variety in physical structure therein, leaving decaying and
dead trees as microhabitats, single tree harvesting).
Improved methods for inventory of diversity on different levels should be developed and give the
basis for ecologically sound management plans.
4. All the measures must take into account conservation of forest genetic resources, with preference
for in situ conservation for the protection of adaptability. Identification and declaration of gene
resources by means of forest genetic research must be enforced. In 1989 the German Working
Group “Conservation of Forest Genetic Resources” presented a conceptual framework for the
conservation of forest genetic resources in Germany (BLAG 1989) which is continuously put
into action.

5. These and related concepts are applicable not only in state forests, but also in community and
private forests, for the latter this progress should serve as a model for financial support or
6. However, pluralism as a innovative factor in forest management is to be retained. This is ensured
to some extent by several hundred thousands of forest landowners with differing ideas on, and
practices of, forest utilization.

Biodiversity is a complex issue. Life is organized in different hierarchy levels, therefore diversity is
represented on all these levels, e.g. genetic diversity (diversity within and between populations),
species diversity (species richness and evenness), and habitat diversity (ecosystem diversity, landscape
diversity). There are interactions between these levels. Biodiversity is not static, but is realized in
dynamic systems with considerable variation in space and time.
The Ministerial Conference on the Protection of Forests in Europe, held in Helsinki, Finland, in
1993, proved that the European countries have the will to act in accordance with the decisions taken
at the Earth Summit in Rio. Besides guidelines for the sustainable management of forests in Europe,
inter alia, the Conference encouraged the conservation of biodiversity of European forests. In its
resolution H2 the Conference (MAF 1993) emphasised the conservation and appropriate enhancement
of biodiversity as an essential operational element in sustainable forest management, which should be
adequately addressed in forest policy, operational guidelines and legislation. The need of appropriate
biodiversity appraisal systems was stressed as well as of methods for evaluating the impact on
biodiversity of forest development and management techniques. Forest management should conserve
and manage the diversity of habitats and a variety of structure within stands should be favoured.
Future actions focus on the conservation of forest genetic resources, the protection of threatened
forest species and ecosystems.
Inter alia, the Signatory States and the EC agreed to the establishment of a coherent ecological
network of representative climax, primary and other special forests. An educational and public awareness
programme will be established as well as surveys and research programmes (with special attention on
the improvement of methods for assessing biodiversity in forests). Furthermore the participating
states (including Germany) and the EC agreed in developing national or regional guidelines to obtain
sufficient knowledge about the ecosystem functions and services derived from European forests and
of the status and requirements for management of threatened, rare or representative biotic elements.
The impact of different silvicultural techniques on biodiversity will be investigated (MAF 1993).
As one of the signatory states of the Helsinki resolutions, Germany is willing to put the agreements,
also the commitments made in Rio, into practical work. The German Bundestag’s (the German parliament)
Enquete Commission “Protecting the Earth’s Atmosphere” (Deutschland/Enquete Kommission Schutz
der Erdatmosphaere 1994) gives further recommendations for the conservation of temperate forest
ecosystems. European forest ecosystems are subject to atmospheric pollution, which threatens their
survival. Therefore the reduction of industrial emissions below critical loads is one of the major
challenges for the future. In addition, proposals have been made for ecologically sound forestry.
Nature-oriented management concepts are favoured. India too is a signatory to the Rio Conference
and has begun the process of biodiversity conservation as per its guidelines.
Some general conclusions from these and related viewpoints could be summarized as follows:
Based on current knowledge and on ongoing research, the properties and dynamics of forest
ecosystems with a view to their protection, stability and sustainable use, integrate the main objectives
of today’s forestry, e.g. the different demands of various lobbies on forests and the dangers to which
they are exposed by modern industrial society. Forestry must cover expenses in order to ensure the

future existence of forest landowners, enterprises and employees. Wisely managed forests are an
infinitely renewable resource and supply us with our most important regrowing natural resource –
wood. Management directed towards the growth and supply of timber and non-timber forest products
must be aimed at the intelligent and efficient tending, harvesting and regeneration of forests. All the
practices must simultaneously include the preservation of natural goods and the maintenance of site
capacity. This presupposes good knowledge and continuous research, e.g. on the impacts of climate
change on forest ecosystems and on the importance of woodlands and forests in the global carbon
budget as well as to people, forestry and wood industry. Ecosystem management and associated
concepts have increasingly dominated natural resource management discussions. Sustainable forestry
and ecological soundness became keywords and although many people may support this, these terms
are not well defined or are understood in different ways (with a view on, for example, genetic
sustainability, ecological or economical sustainability).
Some general conclusions for both countries :
1. The major challenge is that international or national conventions need to be implemented into a
practical approach. This means bridging the gap between policy, science, practice, as well as
public, and presupposes a participation of the different lobbies (forest landowners, forest
managers, nature conservationists, policy-makers, etc.) and an unprejudiced debate and
willingness to negotiate.
2. Public awareness in biodiversity conservation needs to be generated and improved by means of
continued eduction.
3. There is a strong need of ongoing scientific research and deepening our knowledge on ecosystems
and their inherent natural (dynamic) processes (e.g., monitoring and measuring, as well as
appraisal of biodiversity).
4. Biodiversity is a result of historic and evolutionary processes, the possibility for artificial
enhancement of biodiversity, creation or reconstruction needs to be assessed.
5. Incorporation of biodiversity into sound ecological, well-adapted management concepts in
forestry is an essential prerequisite for the sustainable conservation, stability and elasticity of
one of the most area-extensive habitats on earth.
6. Frequent consultations and joint bio-diversity programmes between neighbouring countries.

Bibliography Negi, S.S. 1994. Biodiversity and its conserva-

Anon. 1993. State of the forests report. Forest tion in India. IPC. New Delhi.
survey of India. Dehra Dun. StBA. 1995. Statistisches Jahrbuch 1995 für die
BLAG. 1989. Konzept zur Erhaltung forstlicher Bundesrepublik Deutschland und für das
Genressourcen in der Bundesrepublik Deut- Ausland. Statistisches Bundesamt.
schland. Forst und Holz 44, pp. 379-404 Wiesbaden.
BSI. Red data book of rare and threatened Tewari, D.N. 1992. Tropical forestry in India. IBD.
plants. Calcutta Dehra Dun.
Deutschland / Enquete Kommission Schutz der Er- WCMC.1992. Global biodiversity status of the
datmosphaere. 1994. Report submitted by the earth’s living resources. World Conserva-
12th German Bundestag’s Enquete Commis- tion Monitoring Centre, London. In WRI
sion Protecting the earth’s atmosphere. Bonn. 1995 World resources 1994-95. A guide to
MAF. 1993. Ministerial Conference on the Protec- the global environment. World Resources
tion of Forests in Europe, 16-17 June 1993, Institute, Washington (publ. in German: Vogl
Helsinki. Conference Proceedings. Finnish et al. (eds.) 1995 – Handbuch des Umwelts-
Ministry of Agriculture and Forestry. chutzes. V. Welt-Ressourcen – 1994/95.
Helsinki. Landsberg.)

Conservation of biodiversity and endangered
ecosystem in Pakistan

Mirza Hakim Khan 1

Pakistan has given serious thought to protecting its biodiversity and endangered ecosystems
through different conservation measures. The different priority areas include Himalayan moist temperate
forests, the juniperus forests of Balochistan, and the mangrove forest of Sindh. Hunting of such
species as the houbara bustard and the trapping of hawks and falcons threaten biological diversity as
does over-use of high mountain pasture.

Keywords: Biodiversity, endangered ecosystems, Pakistan priority areas, conservation measures.

Biodiversity and natural capital encompass all of the country’s wilderness areas and scenic
landscapes, together with their associated fauna and flora. These in turn can be grouped into nine
major ecological zones and 21 out of 22 of the recognized Asian wetland types (Robert 1991).
The contribution of these resources is recognized at three levels: species, genera and communities
(habitats and ecosystems). Both collectively and within each level, the range and variety of the
resources are referred to as biological diversity.
Biological diversity is the total variety of life on earth. No one knows, even the nearest order of
magnitude, how many life forms humanity shares the planet with; roughly 1.4 million species have
been identified, but scientists now believe the total number is between 10 million and 80 million. Most
of these are small animals, such as insects and molluscs in little explored environments, such as the
tropical forest canopy or the ocean floor. But nature retains its mystery in familiar places as well. Even
a handful of soil is likely to contain many species unknown to science.
Biologists estimate that at a minimum 50 000 invertebrate species per year, nearly 140 each day, are
condemned to extinction by the destruction of their tropical rain forest habitat. Large creatures, as well
as small, are vanishing, deforestation condemns at least one species of bird, mammal or plant to
extinction daily. Realizing the facts, the first ever United Nation sponsored Conference on Environment
Conservation in Rio de Janeiro on 3 June 1992 emphasized the most critical issue, as how to preserve
the biodiversity of the earth.
A large number of animals and plant species roughly some 6 000 taxa are found in Pakistan and
Kashmir (Stewart 1972). Out of these, a large number of animals and plant species are endangered,
threatened or at least in a very vulnerable position, because of biotic pressure, in the form of hunting,
poaching and habitat destruction. In 1971, the Government of Pakistan listed 31 species of mammals,
birds and reptiles as being endangered in the country and it is without doubt that this list could now

Deputy Director (Technical), Pakistan Forest Institute, Peshawar, N.W.F.P., Pakistan.

be extended. Wild animals have been hunted to extinction, among them, lion, tiger, cheetah, one
horned rhino, and chau singa. Today, ibex, snow leopard, wild ass and houbara bustard all face
extinction from hunting pressure.
Pakistan has given serious thought to protect its biodiversity and endangered ecosystems. In this
regard two national parks have been used to reintroduce previously lost species into former habitats,
notably the one-horned rhino and the black buck in Lalsuhanra National Park, as well as in the Kaghan
Pakistan is a signatory of international agreements like the Convention on International Trade in
Endangered Species of Wild Flora and Fauna (CITES), the Convention on Wetlands of International
Importance (Ramsar), the World Heritage Convention and the Convention on the Conservation of
Migratory Species of Wild Animal (Bonn Convention). In addition, Pakistan is a member of the World
Conservation Union (IUCN), the International Waterfowl and Wetland Research Bureau (WRB) the
World Wildlife Fund for Nature (WWF)
The present position of biodiversity in Pakistan is shown in Table 1.

Table 1. Biodiversity in Pakistan

Flora/Fauna Number of Species

In the World In Pakistan Endemic

Plants 5,700 species (5,000 wild) 372 species

Mammals 18 orders 10 orders *Indus Dolphin

*Chiltan Markhor
*Suleman Markhor
4 100 species 188 species of which
63 rodents
39 carnivores
38 bats
25 hoofed animals
11 insectivores
9 aquatic animals
3 primates
1 pholidota

Birds 8 600 species 666 migratory species

Reptiles 6 500 species 174 species of which:

88 lizards
72 snakes
10 turtles (2 marine)
2 tortoise
1 crocodile
1 Gavial

Amphibians/Fishes 2 600 species 14 species of which:

400 marine fish
125 freshwater species

Insects/Invertebrates 20 000 species

700 marine
Source: NCS Sector Paper on Natural Capital by Abdul Latif Rao & Abeedullah Jan.

The following ecosystems are designated on top priority for inclusion within national park

1. The Himalayan moist temperate forest: It has biological diversity, and a complement of faunal
species, many of which are endangered or under threat. A suitable representative and relatively
undisturbed area of this ecotype is situated in the Machiara catchment in the Neelum Valley,
Azad Kashmir. It is one of the few sites in which a breeding population of the western tragopan
pheasant, Tragopan melanocephalus, exists.

2. The juniper forests of Balochistan: A representative area to be demarcated which would include
the present Ziarat Wildlife Sanctuary and also include adjacent areas of Zizri, Burwul, Zargat,
Shaidan, Gohar, Suit, and parts of Malikat, Kato and Mazri Shor. This latter area, whilst described
more as arid scrub forest than pure juniper forest, nevertheless is an important ecotone and has
a population of fauna and flora different from the juniper forest or adjacent open rangeland. It is
a natural adjacent to the recommended juniper forest reservation.

3. The mangrove forests of Sindh: The semi-arid mangrove forests of the Indus Delta are threatened
by a number of factors. Changes in the hydrology of the delta through upstream irrigation and
river control schemes, pollution, overexploitation for charcoal production, tan bark, fodder and
firewood all have their toll on this unique and highly productive ecosystem. There are at least
150 000 ha of relatively dense stands of mangrove forest stretching immediately south of Karachi
from the Korangi creek area as far as south as the Hajamro creek, and under the control of the
port Qasim Authority and Sindh Forest Department. Not only would such an area preserve
some of the best semi-arid mangrove forests left, but would also protect diverse terrestrial and
aquatic fauna including three species of shrimps, dolphin and enormous populations of migratory
waders and waterfowl which are to be seen during winter months. The IUCN Coastal Ecosystem
Unit in Karachi should be consulted and involved in any planning for such reserves..

4. Hunting of the houbara bustard: The houbara bustard is a wintering migrant to Pakistan from
former USSR, where it breeds in the Central Asia republics, mainly in the Kizil Kum desert region
south-east of Aral Sea. It also over-winters in part of Iran, Afghanistan and parts of India
(Cramp and Simmon 1980). There are two biologically separate populations of houbara bustard
in Pakistan. Chlamydotis undulata is described by (Robert 1991) as the North African race and
being smaller and darker in plumage than the Asian race or sub species Chlamydotis undulata
macqueenii. Both races are migratory, but there is substantial evidence (Main 1988) that a small
breeding population of C. undulata exists in Balochistan. Whether these are totally resident
birds or young birds which miss out on the first immigration to north is unclear.

Balochistan, Punjab and Sindh are the main wintering areas of the bird, although small number
of birds are reported from NWFP in Dera Ismail Khan District (Malik 1985).

In Balochistan, the houbara bustard occurs throughout the wider open valleys of the Balochistan
plateau, in the major part of Chagai, Kharan District, the Mastung Valley and Kachni District. It
is reported as being most numerous around Yakmuch (Chagai District), the Gishki Valley (Makran)
and Jhalwan (Kharan) (Shams 1985).

In Sindh, (Suhario 1983) reports that the main wintering areas are the arid barani plains like
Rahnn of Kutch and the desert areas of Tharparkar, Badin, Thatta, Sanghar, Khaipur and Sukkur
Deserts, Kirthar ranges, and Dadu, Larkana and Jacobabad Districts.

In Punjab, the preferred habitat is the steppes of the deserts in Central Thal, southern Cholistan
and south western Dera Ghazi Khan (Chaudhry 1991). For conservation studies, a small area in
the Nag Valley of Balochistan is the ideal habitat. It is known that a small breeding population of
the birds occurs here. Within the restricted area, there is an excellent opportunity to study the
breeding biology and status of species. The area is relatively small approximately 40 x 17 miles,
has a low density of human population, the habitat is not seriously degraded and the local
people are likely to be cooperative.

5. High mountain pasture utilization: The principal problems which influence the depletion of the
alpine grasslands are government agency land administration methods, and over-grazing. Most
of the high alpine grasslands are classified as being under the control of provincial forest
departments. However, the forest departments do not have the interest, funds or technical
ability to manage these grasslands and hence they are ignored in terms of nature conservation,
watershed and commercial management inputs. In practical terms the local landowners have in
most cases taken over control and lease out grazing and forest produce collection rights to a
wide range of lessees. Nomadic tribes with large flocks of domestic stock move into these areas
as soon as spring snow melts and will remain there all summer before descending to lower
pasture areas in the autumn. This constant grazing pressure is leading to extreme depletion, with
the more palatable herbs and grasses being grazed out, and vegetation cover of the highly
erodible slopes steadily diminishing. There is an urgent need for agricultural extension work to
be implemented by personnel with appropriate technical skills (not foresters) to illustrate the
methods being used in such projects as the Swiss/Pak Kalam Integrated Development Project in
the upper Swat Valley. Here in grasslands about 3 000 m high controlled grazing has often meant
complete recovery of depleted grassland within three years.

6. Trapping of hawks and falcons: There is a well-established rural industry in Pakistan involving
the trapping of wild hawks and falcons, and their subsequent trading and export to Arab nations
for falconry purposes. The trade involves millions of rupees each season. It is difficult to
establish the scale of the trade in precise terms but one reliable source estimated a conservative
figure of 3 000 falcons being caught each year not necessarily all caught in Pakistan but certainly
traded through Peshawar. There is no real data available to support or discredit this figure.

It is the saker falcon, Falco cherrug, which is the main species targeted. This is a winter visitor
to Pakistan; it breeds in Central USSR and is often trapped on its migratory route to Pakistan in
the Chitral area. (Robert 1991) notes that a good female Saker falcon would fetch as much as
US$ 50 000 in 1983. The conservation status of this bird is scarce becoming rare. It would be
noted that only female of the species are considered of value for falconry. Presumably, the male
which are caught are either released after capture, perhaps injured or destroyed.

In view of this, it is recommended that a carefully designed investigation is carried out in order
to establish the essential parameters of the trade including the number of birds to be caught and
traded each year, identification of traders and trappers, etc. This could be done with international
funding and in cooperation with such institutions as IUCN-WWF-Pakistan.

According to (Sheikh 1992), some of the following animal species are endangered (Table 2):
Table 2. Endangered species
I) FAUNA 9. Elaegnus augustifolia
10. Salvadora persica
Mammals 11. Acer pentapomicum
1. Balochistan Black Bear (Selenarctos thibetanus 12. Populus alba
gederosianus) 13. P. ciliata
2. Smoth Indian Otter (Lutra perspicillata) 14. Debregeasia Hypoleuca
3. Fishing Cat (Felis viverrina) 15. Nyctanthis arbortristis
4. Caracal (Felis caracal) 16. Carpinus viminea
5. Sand Cat (Felis margacita) 17. Corylus celorna
6. Wild Ass (Equus hemionus) 18. Betula utilis
7. Barking deer or Muntjac (Muntiacus muntjak) 19. Berchemia floribunda
8. Hog Deer (Axix porcinus) 20. Helinus lanceolatus
9. Black Buck (Antilope cevicarpa) 21. Rhamnus pentapomica
10. Chinkara (Gazella gazella) 22. Forkhlea tenacissima
11. Coitred Gazelle (Gazella subgutturosa) 23. Trema politoria
12. Nilgai (Boselaphus tragocamelus) 24. Villeburnea frutescens
13. Punjab Urila (Ovis orientalis punjabiensis) 25. Zizyphus sativa
14. Marco Polo’s sheep (Ovis ammon Polii) 26. Dioscorea deltoides
15. Suleman Markhor (Capra falconeri jerdoni) 27. Saussurea lappa
28. Valariana willichii
Reptiles 29. Hyoscyamus muticus
1. Marsh Crocodile (Crocodiles palustris) 30. Podophyllum emodi
2. Coharia (Garialis gangeticus) 31. Butea frondosa
3. Monitor lizards (Baranus species) 32. Ficus carica
4. Rock Python (Python molurus) 33. Rhizophora mucronata
34. Avicinna alba
Birds 35. Prunus padus
1. Marbled Teal (Anas angustirostris) 36. Cornus macrophylla
2. Spotbill Duck (Anas poecilorhyncha) 37. Quercus incana
3. White-headed Duck (Oxyura leucocephala) 38. Q. baloot
4. Cotton Teal (Nettapus coromandelicus) 39. Q. glauca
5. Peafowl (Pavo cristatus) 40. Q. semicarpifolia
6. Common Crane (Grus grus) 41. Fraxinus excelsior
7. Siberian Crane (Grus leucogeralus) 42. Aesculus indica
8. Sarus Crane (Grus antigone) 43. Nannorrhops ritohieana
9. Demoiselle Crane (Anthropoides virgo) 44. Juglans regia
10. Great Indian Bustard (Choriotis nigriceps) 45. Ephedra procera
11. Houbara Bustard (Chlamydotis undulata) 46. Terminalia belerica
12. Lesser Florican (Sypheotides Indica) 47. Tecoma undulata
13. Large Pintailed Sandgrous (Pteroclesalchata) 48. Phyllanthus emblica
49. Cardiospermum helicacabum
II) FLORA 1 50. Colchicum luteum
1. Juniperus marcropoda 51. Asparagus racemosus
2. Taxus baccata 52. Acormus calamus
3. Pinus gerardiana 53. Cassia angustifolia
4. Punica sp. 54. Acacia catechu
5. Ulmus nitida 55. Prosopis cineraria
6. Pistacia integerrima 56. Populus euphratica
7. Monotheca boxifolia 57. Capparis decidua
8. Celtis australis
(Khan 1992) lists the following plants as threatened.


Table 3. Protected areas in Pakistan

Area (ha)
National parks 11 954 246
Wildlife sanctuaries 86 2 749 054
Game reserves 84 3 535 287

Total protected area 7 238 584

Some recommendations are given below for conservation of biodiversity and endangered
• The need to preserve representative areas of all major ecotypes in Pakistan by protection from
grazing, exploitation and fire in order that their syneaological structure, composition and ecological
succession is preserved for all time. Such reserves should be planned on a scientific basis with
respect to location, size, shape and their ability to represent ecological continuum and ecotones.
• Identify and establish new national parks, game reserves and wildlife sanctuaries on a priority
basis and bring all such areas under planned management. Select sites for six additional national
parks or other conservation areas in the next 25 years as recommended by the National
Conservation Strategy.
• That all species (both plant and animal) which are considered threatened or endangered should
be given highest priority for conservation, starting with preservation of habitat and inventories
of populations under threat.
• As a precursor to formal reservation of areas as national parks, wildlife sanctuaries, etc., the
welfare and interests of the local people should be considered of the highest importance if
conservation measures are to be successful. This will, in many cases, entail rural support
programmes which will be significant undertakings in their own right.
• Nature conservation and provisions for wildlife preservation should be an integral part of all
forest management working plans.
• Pakistan as a signatory of several international conservation conventions should honour all
consequent responsibilities, particularly in respect of trading and wildlife, and conservation of
migratory species.
• Recognition of an urgent need for specialist training at a post graduate and sub-professional
level in nature conservation.
• An endangered species monitoring group consisting of plant ecologists, wildlife biologists,
forests and entomologists under the direction and control of National Council for Conservation
of Wildlife be established. The group will assess the present status of all flora and fauna species,
which are currently or likely to be at risk, threatened or endangered.
• Efforts be made to minimize the hunting of houbara bustard and a Houbara Bustard Study Group
be established to study the breeding biology of the species. In the meantime, an investigation on
the organization, methods, dealer and economic returns of the falcon/hawk trade should be
initiated with the cooperation of national and international funding.
• People’s awareness and consciousness regarding the importance of forestry, environmental
conservation, biodiversity and endangered ecosystems be started through press and other
information media. Articles, brochures in national and local languages be written.

• NGOs and people’s participation in biodiversity conservation projects be encouraged, so that the
common man can fully participate in all sorts of conservation activities.
• Organize more seminars, workshops and courses for training and awareness raising.
• Enforcement of strict laws in the form of severe punishment for illicit cuttings of forests and
ecosystem conservation be recommended.
• Suitable projects mentioned in the Master Plan reports for biodiversity and ecosystem conservation
be taken up immediately and through adequate planning (Wilkinson & Khan 1991).
• Pakistan Forest Institute be determined as the focal point for research, training and education at
post-graduate level on Biodiversity, Environment and Ecosystem Conservation. A separate
division for biodiversity, ecology and environmental research and education be established.

Cramp, S. & Simmons, K.E.L. (eds). VT7. The birds Robert, T.J. 1991. The birds of Pakistan. Vol.1.
of the Western Palaeartic. Oxford Universi- Oxford University Press Oxford, New York,
ty Press, Vol.1 Ostrich to duck. 1979. Vol.2. Delhi, Karachi.
Hawks to bustards. 1982. Vol.3. Wader to Sheikh, M.I. 1992. Personal Communication. For-
Gulls. estry Planning and Development Project,
Aleem, Ch. A. 1991. Houbara bustard in the Pun- Margalla Road, F-7/2 lslamabad.
jab. Natura. W.W.F. Pakistan, Lahore: 22-25. Shams, K.M. 1985. Occurrence and distribution
Khan, Mirza Hazim,.1992. Biodiversity note sub- of bustards in Balochistan. Proceedings of
mitted to MINFA, Islamabad. the International Symposium on Bustards,
Malik, M. Mumtaz. 1985. The distribution and Peshawar, 4-6 October 1983: 51-52.
conservation of houbara bustards in north- Stewart, R.R. 1972. Flora of West Pakistan. An
west Frontier Province. Proceedings of annotated catalogue of the Vascular Plan
International Symposium on Bustard, of W. Pakistan and Kashmir.
Peshawar, 4-6 October 1983; 81~5. Suhario, M.I. 1983. Houbara bustard in Sindh.
Mian, Afsar,..1988. Biology of houbara bustard Sindh Wildlife Management Board,
(Chlamydotis Undulata macqueenii) in Karachi.
North West Kyzyl Kum (in Russian). Zoology Wilkinson, G.B. & Khan, Mirza Hakim . 1991. Con-
Zhurnal Vol.9 No.8,:1263-C. servation of ecosystems. Report prepared for
National Conservation Strategy. 1992. Environ- the Government of Pakistan under Asian
ment and Urban Affairs Division, Govt. of Development Bank Tech. Assistance. T.A.
Pakistan and the World Conservation Union No.1170-Pak. United Nationals Development
(lUCN). Pakistan: 48-52. Programme PAK/88/019.

The ecological stewardship project:
a public-private partnership to develop a
common reference for ecosystem
management in the USA

W.T. Sexton 1, N.C. Johnson2, R.C. Szaro 3

Research and policy to support the use of ecosystem approaches to natural resources management
are rapidly evolving in the United States. Still, the use of ecosystem approaches faces a number of
challenges. One of these challenges is enabling natural resource managers, policy-makers, and the
public to keep abreast of the rapidly growing body of research and management experience with
ecosystem approaches. This paper summarizes an ambitious project to collect and synthesize the
latest scientific understanding and management experience on a wide range of topics related to
ecosystem management in the United States. The Ecological Stewardship Project’s principal objectives
are: 1) to develop an information framework to help federal agencies implement ecosystem approaches
to natural resource management on public lands; 2) to provide a comprehensive, peer-reviewed reference
text covering thirty topics in science, economics, social systems, and information management that
are relevant to ecosystem management, and 3) to catalyze and support other efforts to accelerate the
implementation of effective ecosystem approaches on federal lands and waters. Nearly 400 researchers
and natural resource managers from government, industry, universities and non-governmental
organizations are involved in 60 author teams that are documenting the current state of scientific
knowledge and management experience related to ecosystem management. When completed in late
1997, information produced by the Ecological Stewardship Project will be disseminated in both published
and electronic formats to ensure potential users in the United States and internationally have ready

Keywords: Ecological stewardship, ecosystem management, information framework.

Deputy Director, Ecosystem Management, U.S.D.A. Forest Service, PO Box 96090, Washington, DC 20090-
6090, USA.
Senior Associate, Biological Resources Program, World Resources Institute, 1709 New York Ave. NW,
Washington, DC 20006, USA.
Program Director for Developing Countries, International Union of Forestry Research Organizations, Vienna,

Just as sustained-yield and multiple-use concepts emerged as frameworks to manage natural
resources earlier in this century, the end of the 20th century has become an era of change in the use of
knowledge, tools, and strategies to manage lands and waters in the United States. The shift now is to
ecological stewardship approaches that enable public and private resource managers to use science
plus social guidance to help sustain productivity while maintaining biodiversity and other environmental
services. This conceptual shift, as measured by the emergence of new concepts of resource management,
an expanding body of scientific information relevant to management practices, a growing number of
“ecosystem management” efforts around the country, and the development of new policies, is rapidly
evolving. But if the conceptual shift is occurring rapidly, the use of ecological stewardship approaches
in day-to-day management, research, and policy development lags behind, in part for want of practical
assessment and summary of the growing scientific, social, and management experience that already
exists. This paper describes an approach being used in the United States to synthesize both existing
scientific understanding and management experience on a wide range of topics relevant to ecosystem
The concept of integrating ecological knowledge into natural resources management – often referred
to as “ecosystem management” – is not altogether new in the United States. For example, Aldo
Leopold was advocating the use of ecological knowledge in natural resources management as early as
the 1940s and by the 1970s ecological concepts were finding their way into a growing number of
publications on wildlife and natural resources management (e.g., Thomas 1979). More recently, the
concept of “sustainable development” – often defined as managing natural resources to meet present
human needs without compromising the ability of future generations to meet their needs – has led to
greater consideration of the long term impacts of natural resource management decisions. Meanwhile,
the cumulative impacts of numerous local management decisions have led many scientists and resource
managers to conclude that biodiversity, water quality, and other natural resources can only be conserved
through cooperative efforts across large landscapes – landscapes that often cross ownership
boundaries (e.g., PCSD 1996; Aplet et al. 1993).
What distinguishes ecosystem management approaches from earlier single-use or multiple-use
approaches to natural resources management is the integration of environmental, social, and economic
knowledge to make resource management decisions at multiple geographic scales. Shifting from single
resource or single species management to managing an ecosystem for a variety of resources, including
maintenance of its biodiversity, requires using the best available scientific, social, and economic
information (PCSD, 1996). Scientific information is needed to identify ecosystem processes essential
to the productivity of a wide variety of natural resources. Social and economic information can be
used to determine which strategies will best meet public demands and landowner objectives. Voluntary
and cooperative efforts at managing natural resources across ownership boundaries characterize
many existing ecosystem management efforts (Keystone Center 1996).
In 1992, the U.S. Forest Service and the Bureau of Land Management established policies adopting
ecosystem approaches to guide their natural resource management decisions for over 300 million
acres of federal lands. Although ecosystem management is sometimes perceived largely as a federal
effort dominated by research scientists, the rapidly growing number of ecosystem management efforts
around the country indicates otherwise. For example, a study at the University of Michigan identified
over 600 projects around the country that claim to be using ecosystem management approaches
(Yaffee et al. 1996), while the Keystone Center’s national policy dialogue on ecosystem management
inventoried over 150 such efforts (Keystone Center 1996), many of them initiated at the local level by
conservation groups, state and county governments, private landowners, and forest products
companies. In its consensus report, the Keystone National Policy Dialogue on Ecosystem Management

(a diverse group composed of private landowners, conservationists, government resource managers,
and extractive industries) concluded that ecosystem management4 “is a process that can be of benefit
to private landowners, environmentalists, government agencies, developers, and other groups
concerned with natural resource and environmental management.”
Still, the use of ecosystem approaches faces a number of challenges. Most efforts are relatively
recent (nearly all the examples surveyed by the Keystone Dialogue started after 1990) making it
difficult to assess the success of ecosystem management efforts. Another problem is that the concept,
science, and policy of ecosystem management have outpaced the capacity of managers to keep up
with developments and integrate the new knowledge into their routine management practices. Much
of the rapidly growing literature on the subject is presented in idealized form, only limited assessments
of on-the-ground experiences have been conducted, and ecosystem management concepts are
continuing to evolve. This leaves natural resource managers with little guidance on how to apply
ecosystem management concepts in specific field situations.
The development of a practical implementation reference guide can help to bridge this information
gap – a gap that has become one of the major barriers to the wider and more effective use of ecological
stewardship approaches in the United States. This gap, however, cannot be spanned merely by
orchestrating a current literature review and distributing it to interested user groups. Engaging
professional resource managers (public and private), scientists, and environmental interests in a
project to synthesize a comprehensive reference is a necessary and logical next step to ensure the
rapid and broad based acceptance and use of ecosystem stewardship approaches.

In April of 1994, Jack Ward Thomas, Chief of the U.S. Forest Service, met with a number of private
philanthropic foundations5 active in environmental issues to discuss the potential role of the foundations
in supporting natural resource conservation efforts that would complement or build upon U.S. Forest
Service programmes and activities. As part of this discussion, Chief Thomas challenged the group to
find proactive ways in which to help accelerate the implementation of ecological stewardship in the
United States. Over the next several months, the group continued a dialogue on how the Forest
Service and private foundations could collaborate. It was agreed that a jointly sponsored project to
develop a comprehensive reference guide on ecological stewardship could form the basis of a
partnership, and a Memorandum of Agreement was signed between the Forest Service and the
foundations in November 1994. The partners agreed that such a reference would have to be useful to
resource managers, researchers, and policy-makers alike and that it would require a highly collaborative
process between scientists and resource managers – with experience in a broad range of topics – to
develop such a reference. This agreement launched the Ecological Stewardship Project.
Three principal objectives were defined to guide the development of the Ecological Stewardship
Project. First, the project should develop an information framework that can assist federal agencies to
implement an ecosystem approach to natural resources management on federal lands. Such a framework
should not provide prescriptive solutions for management issues at individual sites but should help
agency personnel to develop implementation plans and strategies. Second, the project should publish

The Keystone Dialogue Group defined ecosystem management as: “A collaborative process that strives to
reconcile the promotion of economic opportunities and liveable communities with the conservation of ecological
integrity and biological diversity.”
These included the Ford Foundation, Bullit Foundation, Liz Claiborne & Art Ortenberg Foundation, Moriah Fund,
National Fish and Wildlife Foundation, Henry P. Kendall Foundation, Pew Charitable Trusts, Tides Foundation and
W. Alton Jones Foundation.

a reference text available to resource managers and researchers that: 1) documents the scientific
foundations for ecosystem approaches to natural resource management, and; 2) assesses experience
with various ecosystem management options and alternatives through case studies, literature surveys,
and analysis by resource managers. Third, the project should catalyze or support other efforts to
strengthen and accelerate the implementation of ecosystem approaches on federal lands and waters.
Once the basic vision and objectives for the project were defined, the U.S. Forest Service actively
invited other federal agencies, non-governmental organizations, private sector resource managers,
and professional societies to join the Ecological Stewardship Project as co-sponsors and partners.
Among the organizations that joined were the American Fisheries Society, American Forests, Boise
Cascade Corporation, Consultative Group on Biodiversity, Hispanic Association of Colleges and
Universities, National Forest Foundation, National Parks and Conservation Association, Pacific Rivers
Council, Pinchot Institute for Conservation, The Nature Conservancy, Hancock Timber Resources
Group, University of Arizona, Seneca Jones Timber Company, and a number of federal agencies
including the National Oceanic and Atmospheric Administration in the Department of Commerce, and
the Bureau of Land Management, the U.S. Fish and Wildlife Service, the U.S. Geologic Survey, the
National Biological Service, and the National Park Service in the Department of Interior.

Broad participation characterizes many of the project’s activities. For example, during early planning
for the project in 1995, publicly announced meetings were held every two weeks to discuss and
determine process steps, timing, and identification of key topics to be included in a scientific and
social framework for ecologically-based stewardship of federal lands. During this planning phase,
over 100 participants from non-governmental organizations, private sector companies, universities,
and federal and state agencies helped to shape the structure of the project.
By the end of the planning phase in early fall 1995, a core team of federal land management
personnel and private foundation representatives together with the participants in the public meetings
had defined a several step process for developing the comprehensive reference guide for ecological
stewardship. The first step was to establish a core organizing group with representatives from federal
agencies, non-governmental organizations, private sector companies, and the foundations. Second,
this organizing group would identify key topics relevant to implementing ecosystem approaches to
natural resource management that would be covered in the reference. Third, a science/research team
and a resource management team would be established for each topic – the science team to synthesize
the current state of scientific/research knowledge on the topic and the management team to summarize
and assess current management experience. Fourth, the core organizing group would identify and
invite leading researchers and resource managers in the United States with first hand experience in the
selected topic areas to join the author teams. Fifth, author teams would meet together in an extended
workshop to develop outlines for research and management papers on each topic. Sixth, the U.S.
Forest Service would commission papers on the final set of topics agreed to at the workshop. Finally,
the U.S. Forest Service would select an independent research institution to design and implement a
rigorous peer review process and help identify options for packaging and disseminating the information
in the peer reviewed papers.
For two weeks in December 1995, over 400 researchers and natural resource managers gathered in
Tucson, Arizona to begin a year-long process to document the knowledge base and management
challenges for implementing ecological approaches in natural resource management. The challenge to
workshop participants – who came from federal agencies, non-governmental organizations, universities,
professional societies and industry – was to develop a written “dialogue” between the scientific,

social, and economic foundations for ecological stewardship approaches and the management
challenges of translating those concepts and principles into practice.
Workshop participants were invited on the basis of their research and management experience in
one or more of 30 topics relevant to ecological approaches to natural resources management. The
topics cover a wide range of biophysical, social, institutional, and economic issues (see Box 1). A
science and a management team was organized for each topic. At least one scientist/researcher was
included on the management topic team and at least one resource manager was included on the
science topic team. At the workshop, the teams prepared draft outlines on how they would approach
their topic. For each topic, management and science teams shared drafts and discussed science and
management linkages before developing a final outline at the end of the workshop. Individuals and
teams were strongly encouraged to make the broadest possible range of contacts and had full latitude
to add authors to their teams as needed to fully cover their topics and key questions. An interim report
(Sexton et al. 1996) summarizing the workshop proceedings was published in February 1996 and
widely distributed to inform interested parties who were not able to participate in the workshop.


From the beginning, organizers of the Ecological Stewardship Project viewed an independent and
rigorous peer review process as essential to the project’s credibility. To achieve this, the U.S. Forest
Service and the World Resources Institute (WRI) entered into a cooperative agreement in which WRI
would design and manage an independent peer review process and help the U.S. Forest Service to
define options for packaging and disseminating the information contained in the peer reviewed papers.
WRI, in consultation with the U.S. Forest Service and a steering group composed of representatives
of federal and state agencies, non-governmental organizations, and industry, designed a three-stage
peer review process. The first review stage, completed in August 1996, provided general oversight on
the coverage of key questions in each topic, linkages between science and management papers, and
comments on how first draft papers could be improved for a wider, more detailed peer review of second
draft papers.
Once the second draft papers were received in November 1996, WRI created a dedicated World
Wide Web site to facilitate public access to the draft papers. Notices were placed in a number of
professional journals (e.g., Conservation Biology, Society and Natural Resources, Journal of Forestry,
Journal of Range Management, Journal of Ecological Economics) and computer list serves on
related topics to invite broad peer review. To ensure that all papers were reviewed by appropriate
experts, each paper was also sent directly to four experts who agreed to review the paper. After a six-
week review period which ended in late January 1997, the peer review comments for each paper were
summarized – with emphasis on the most important suggested revisions – by thematic group
“coordinators” who were selected on the basis of their wide ranging knowledge of topics within their
thematic group (see Box 1).
Once the second stage reviews were received, collated, and summarized, a peer review panel was
convened for each thematic group. The peer review panels were chaired by the thematic group
coordinator and included the lead author for each paper in the thematic group, leaders in research and
management (not involved as authors in the project) with broad experience in the thematic area, and
WRI and USFS staff coordinating the project. The two principal goals of the peer review panels were
to: a) ensure that the authors received clear guidance on revising their final drafts (including resolution
of overlaps and gaps between papers); and b) develop concise summaries for each paper that captured
the information and findings most relevant to the needs and concerns of natural resource managers in
the field. Following the peer review panels (which were held in February and March 1997), author

teams had approximately two months to develop final drafts which were then forwarded to a technical

One of the greatest challenges the Ecological Stewardship Project has faced is how to package and
disseminate the materials it has produced so they are widely available and useful to natural resource
managers, researchers, policy makers, and the public. To meet this challenge, the U.S. Forest Service
and WRI agreed that the papers should not only be published in their entirety, but that the papers
should be made available in electronic form and other summary materials based on the papers should
be prepared and made widely available in both print and electronic form. The strategies for disseminating
the materials from the Ecological Stewardship Project include:
• A comprehensive reference text of approximately 1 200 pages containing the complete set of peer
reviewed and edited papers that will be commercially published (possibly in several volumes) in
late 1997.
• A much shorter publication of approximately 150 pages that is a lively, well-written, and graphically
illustrated practical summary and “guide” to the larger reference text. This document will be
published by WRI in late 1997.



Public expectations and shifting values Genetic and species diversity
Processes for achieving consensus Ecosystem/landscape diversity
Regional cooperation Population viability
Evolving public agency beliefs and behaviours Ecological functions and processes
Legal perspectives Role of disturbance and temporal dynamics
Scale phenomena
2. SOCIAL AND CULTURAL Ecological classification
Cultural/social diversity and resource use 5. ECONOMIC DIMENSIONS
Cultural heritage management Shifting human use/demands for natural
Social/cultural classification resources
Social system functions and processes Economic interactions at local/regional
The human ecosystem as an organizing national scales
concept Ecological economics
Integrating social and economic considerations Uncertainty and risk assessment
Human role in evolution of N. American Decision support systems/models and
ecosystems analysis
Land use over time Adaptive management
Ecosystem sustainability and condition Monitoring and evaluation
Ecological restoration Data collection, management, and inventory
Producing and using natural resources Assessment methods

• A dedicated World Wide Web Site that will contain the complete set of papers, a specialized
search engine to locate specific information, and provide links to related World Wide Web sites
and electronic databases maintained by federal agencies, universities, non-governmental
organizations, and research institutions both in the United States and internationally.

Developing a national reference framework to support the implementation of ecological stewardship
approaches is an ambitious undertaking beyond the scope of any single natural resources management
organization in the United States. To successfully synthesize and integrate existing scientific knowledge
and management experience has required a willingness on the part of wide variety of government
agencies and interest groups to share knowledge, experiences, time, and financial resources. We
believe a number of distinctive features of the Ecological Stewardship Project has encouraged the
development of a shared vision about the information framework, and just as importantly, attracted
wide participation in making the vision a reality. These features included:
• A public-private partnership throughout the process from early planning, fund-raising, workshop
participation, analysis, and writing to peer review, product development, and outreach;
• Comprehensive and up-to-date treatment of scientific and management issues relevant to
ecosystem approaches to natural resource management;
• Linkage of conceptual issues, research and technical information to management concerns and
realities faced by natural resource managers, including federal, state and local agencies, private
landowners, community groups, and conservation organizations;
• A neutral but peer-reviewed forum that encouraged discussion and documentation of competing
views, philosophies and facts;
• Practical presentation of tools, options, and strategies for implementing ecosystem approaches
useful in virtually any natural resource management setting, while avoiding site-specific goals,
objectives and prescriptive recommendations;
• A conscious effort to make the Ecological Stewardship materials available to a wide audience
through the Internet, a published companion text, and targeted notices, articles, and workshops
to make potential user groups aware of the materials;
• Innovative use of electronic information technologies to link the Ecological Stewardship materials
to a variety of ongoing federal training, policy analysis, and planning efforts in natural resource
management and to tie in with other electronic information management systems;
• Information organized in formats that made it accessible and readily usable by field level resource
As organizations around the world increasingly turn to ecosystem management, bioregional
management, and other similar integrated approaches to manage complex resource sustainability
problems (Miller 1996), we hope they will find the process and information generated by the Ecological
Stewardship Project in the United States useful in their own context.

Bibliography Leopold, A. 1947. A Sand County Almanac.

Aplet, G. et al. 1993. Defining sustainable Alfred Knopf Publishers, New York.
forestry. Island Press, Washington, DC. Miller, K.R. 1996. Balancing the scales: Guide-
Keystone Center. 1996. The Keystone national lines for increasing biodiversity’s chances
policy dialogue on ecosystem management: through bioregional management. World
Final report:. The Keystone Center, Key- Resources Institute, Washington, DC.
stone, CO. PCSD. 1996. Sustainable America: A new

consensus for prosperity, opportunity, and Sexton, W.T. In press. Ecosystem management:
a healthy environment. The President’s expanding the resource management “tool
Council on Sustainable Development Final kit” for landscape and urban planning.
Report. President’s Council on Sustainable Thomas, J.W. 1979. Wildlife habitats in managed
Development, Washington, DC. forests of the Blue Mountains of Oregon and
Sexton, W.T., Stangel, P. and Szaro, R.C., 1996. Washington. USDA Forest Service Hand-
Toward a scientific and social framework book No. 533. USDA Forest Service, Port-
for ecologically based stewardship of fed- land, OR.
eral lands and waters. Interim report of the Yaffee, S. L. et al. 1996. Ecosystem management
Ecological Stewardship Workshop held in in the United States: An assessment of cur-
Tucson, Arizona, Dec. 4-14, 1996. USDA rent experience. Island Press, Washington,
Forest Service, Washington, D.C. DC.

Forest management and silviculture: essential
elements of a conservation strategy for
tropical forest ecosystems in French Guyana

Jean Douhéret 1

Humid tropical forest covers 96% of the land area in French Guyana. Virtually all of this forested
area is owned by the state, which entrusts the Office National des Forêts (ONF) with the task of
managing the areas with no human settlement, about 7.5 million ha.
Currently, forest survival is only threatened along a coastal strip by urban and agricultural
development together with rapid population growth. Deforestation has been restricted for the time
being by appropriate land planning, which has also helped conserve the hard cores of managed
forests in this coastal strip.
The main threat to forests is wood production, which in time can alter forest composition over large
surface areas. Forest management and silviculture form the basis of the strategy adopted to limit the
risks of harmful development:
(a) Management measures include confining timber production to managed forests, each of which
receives a management document regulating exploitation and operations over a ten-year period.
The Office National des Forêts provides service roads to the massifs and puts the wood on
sale after inventory in clearly demarcated plots (300 ha). A protection ranking is implemented
in each managed forest to preserve evidence of each ecosystem intact.
(b) Silvicultural treatment involves thinning to help future plants (identified previously) flourish.
This is done in each plot five to seven years after exploitation, with the aim of enriching the
stand in commercial species for the next cut, and also of shortening rotation time (40 years
instead of 60) by accelerating growth of these commercial species, thereby economizing an
important surface area of primary forest. To date, this treatment has only been carried out in
two pilot forests, and should start up in normal managed forests in 1997.

Such a conservation strategy takes account of human activity, without restricting itself to the strict
protection of a species or a reserve.

Keywords: Guyana, forest ecosystems, conservation, forest management, silviculture.

In French Guyana, humid tropical forests cover 8 060 000 ha out of a total surface area of 8 350 000 ha,

Chief Forest Engineer of Rural Engineering, Water, and Forests, Regional Director of the Office National des
Forêts, P.O. 7002, 97300 Cayenne, French Guyana, France.

more than 96% of the total land area. Most of this forested area belongs to the state, which entrusts
their management to one body, the Office National des Forêts (apart from several areas where various
local settlers enjoy acknowledged user rights).
This background – sole ownership, one manager, and no resident population in managed areas
(which cover 90% of the land area of Guyana) – presents obvious advantages for a strategy to
conserve forest ecosystems. It also means that managers assume particular responsibility within this
Guyana’s forests are subject to several serious constraints which effectively guide management
decisions regarding conservation. These include the following: rapid demographic growth, with an
annual population increase of 5 to 6%; urban and agricultural development, which is limited to coastal
areas for the time being; the production and processing of forestry products which, with few exceptions,
is restricted to sawing production for local consumption and for the Antilles.


To keep things very simple, a conservation strategy aims at rectifying the impact of a series of
human activities on an ecosystem (or part of an ecosystem). Of course, this definition does not
exclude preventive strategies to avoid damage, or at least to contain damage within acceptable limits
rather than correcting it afterwards.
In Guyana, the activities concerned are the constraints mentioned earlier, which can be distinguished
as follows:
(a) Activities which contribute to deforestation over time: agricultural development, urban
expansion, creation of various infrastructure (roads, dams, electrical lines, etc.).
(b) Activities which maintain forests, but which can alter the composition and structure: forest
exploitation, shifting cultivation, heavy hunting. The impact of “silvicultural” peoples
(Amerindians, Bushi Negue) has been marginal to date and should remain weak for several
years to come. Nevertheless, it should be monitored.
Clearly, management action regarding conservation has to be adapted to these threats, but the
effectiveness of the means depends on the type of threat.


Forest managers only have a spectator’s role when it comes to population growth; if it exists,
immigration and family planning policy is not part of their responsibility.
However, the consequences of population growth, for example the demand for new land for
agricultural and urban development, and for various infrastructural projects, concern forest managers
directly, simply because any new installation can only be set up in, and at the expense of, forests.
The Office National des Forêts has adopted a positive approach to this demand. Deforestation is
accepted, mainly along the coastal strip where there is most demand, and a series of managed forests
have been demarcated in the same strip, which are inalienable and wholly dedicated to wood production.
This measure was introduced within a regional planning scheme, which was voted by the Conseil
régional of Guyana and is currently being approved by the government, representing a consensual
basis for land use over the next twenty to thirty years.
This development has not yet led to any large-scale deforestation, but the situation in the north-
west is starting to give rise to some concern. Here, a fairly rare ecosystem, the “forests on white
sands’, is threatened by the advance of a “pioneer wave” of Suriname, Haitian, and among immigrants.
As this particular case study is outside the scope of this paper, it is not dealt with here.

Gold exploitation is a particular and recurrent problem. Gold prospecting in the major river beds and
its repetition in the same zone could lead to the disappearance of forest cover in land areas of several
tens of hectares. All the same, despite their number, exploitation is naturally limited to gold areas,
which cover only a limited surface area.


In Guyana, despite the modest volumes harvested due to the low extraction rate per hectare, timber
production affects forests over the largest surface area. In 1995, an extraction rate per hectare of less
than 6 m3 meant that 12 000 ha were necessary to harvest 65 000 m3.
The conservation strategy has two objectives: to maintain biodiversity and productive capacity in
the humid tropical forest after forest exploitation; and to minimize the surface area exploited in the long
term (in order to maximize unexploited areas). Globally, the strategy for conserving forest ecosystems
elaborated by the Office National des Forêts is articulated through the following activities:

3.1. Demarcation of an exclusive forest estate with the main vocation of wood
The aim is to confine wood production to the coastal strip, near communication routes and the main
consumer centres, more particularly in a series of massifs called “managed forests”.
The total surface area (S ha) to be dedicated to production is expressed by S = T*P/H, where P m3
represents a reasonable target for annual wood harvesting, H(m3/ha) is the stated harvest per hectare,
and T (years) represents cut rotation.
In practice, the calculation is carried out in successive stages to take account of how these parameters
evolve. For example, when the operation began in 1993, little was known about actual extraction per
hectare; this was overestimated, meaning that the surface area S required was underestimated.
Knowledge only improved after the system was implemented and plot by plot checks were carried out
to determine the actual extraction rate.
To begin with, 54 forests were delineated, representing a surface area of 574 000 ha. But it is clear
now that a total surface area of 725 000 ha will be necessary to meet all requirements (this figure will be
explained later on in this paper).
Outside this managed estate, the only type of exploitation allowed is by river peoples (Amerindians,
and above all the Bushi-Nengue) for their own use. In this way, managers have spared the vast
majority of Guyana’s forests from the main danger: wood exploitation.

3.2. Drafting of forest management documents by massif

First and foremost, the drafting of management documents by forest addresses the need to plan
forest harvests and operations in space and time. Furthermore, it means that forest exploitation can be
checked more effectively, and that data collections on each plot will be available in the future. This is
the basic tool for sustainable forest management.
The first generation of management plans for forest massifs destined for wood production started
to see the light of day in 1993. The aim is to finish drafting the documents by 1998 for the 574 000 ha
mentioned previously. By 15 August 1996, the managed surface area was 333 732 ha, representing
29 forest massifs.
On average, the forest massifs cover 10 000 ha, split up into plots of about 300 ha each, representing
geographical exploitation units with either natural or road borders. Within a ten-year time frame, the
document forecasts how the plots will be used – those which will be cut, those to receive silvicultural
treatment, and those which will be put into reserve or laid to wait.

By necessity, preliminary management will be simplified due to the absence of thorough knowledge
of the terrain and lack of data on the plots. Successive management should benefit from the knowledge
gathered over time in each plot, which will be listed together in a database entitled “forest base”.

3.3. ONF setting up of a coherent and complete network of service roads to the
forest massifs
Management setting up of a service road network addresses two concerns: on the one hand, to
control traffic on newly created roads, and on the other, to avoid the mistakes, duplication and waste
observed during their construction by forest operators, miners or others.
The projects envisage main roads that serve a “supply basin” (set of forests), secondary roads that
serve one forest, and lower category trails serving groups of plots or individual plots. Only logging
trails remain at the discretion (and expense) of producers.
The main roads open in primary production forest areas are normally closed to public traffic to
avoid excessive hunting pressure, as well as the disorderly incursions of potential candidates for
illegal agricultural activity.

3.4. Putting up for sale of timber from limited surface area plots (about 300 ha),
after inventory
In order to check forest exploitation better and ensure it is planned properly by managers, timber is
sold by plots of a limited size (about 300 ha). Beforehand each plot is subject to an inventory by
theoretical sampling of 5% in order to obtain a good idea of species composition (listing includes all
known species above 20 cm in diameter at breast height or above the buttress).
For the moment, timber is sold in standing volume and trees are not marked. Harvesters can cut all
wood above a minimum diameter established by the sale. All harvested wood must be presented to the
agent in charge of supervising the cut, and can only be loaded onto lorries and taken to sawmills after
double mensuration (either partial or total).
This arrangement marks a big step forward from the previous situation, in which the woodcutter
was a dealer in charge of a vast area (several thousand, and often several tens of thousands of
hectares), so that it was impossible to know precisely the amount of timber extraction zone by zone.
At the beginning of 1994, the aim was to carry out all sales in this way from 1 January 1999, with the
intermediate period to facilitate the progression from one method to another. In fact, the system has
evolved faster than expected. Today, apart from one outstanding small “forestry licence” of 9 000 ha,
all sales take place in managed forests or in areas earmarked for future agricultural development (and
thus with no future forestry use).

3.5. Implementing silviculture after exploitation

In public forests in metropolitan France, any cut of wood is not only a harvest but also a silvicultural
activity. Given the current situation, this is not yet the case in humid tropical forests in Guyana – the
cut is only a harvest. Fortunately, the modest harvests per hectare mean forests are only slightly
damaged; only 10% of the surface area of each plot is affected, according to a ONF study.
But such an alteration can become a drawback as it holds back growth of the remaining plants and
in time can lead to an impoverishment of the stand in commercial species unless some intervention is
Silvicultural treatment envisaged involves thinning about 5 to 7 years after the cut and at one time,
uprooting the intrusive plants to the benefit of future plants that were previously identified
(about 30/ha).

Various modalities for these thinning techniques have been tested in two pilot forests; these tests
have already been carried out over a surface area of nearly 3 000 ha. Thinning in “normal’ forests
should begin in 1997.
These thinnings fulfil two objectives:
(1) favouring the enrichment of the stand in commercial species for the next cut: with no treatment,
the stand will be weakened;
(2) shortening rotation time (40 years instead of 60) by accelerating the growth of commercial
species, thereby also economizing considerable surface area of primary forest. For example, with
no silvicultural intervention, an annual harvest of 80 000 m3 with an extraction rate of 5.5 m3/ha
would require exploiting a surface area of 872 700 ha (60-year rotation); however, with silvicultural
treatment, only 581 800 ha would be necessary (40-year rotation), a saving of 190 900 ha.

Thus silviculture would seem to be an essential part of forest ecosystem conservation in Guyana.

3.6. Reservation of evidence zones in all the managed massifs

To maintain evidence of unexploited primary forests, a protection “series” was delineated in each
managed forest, made up of plots or groups of plots deliberately unserved by paths and kept
These reserves stretch over nearly 20% of the surface area of managed forest, and are selected as
evidence of every ecosystem identified in the forest (in practice: humid areas, areas unexploitable for
their sharp relief, and plots representing each geological substrata).
In this way, a realistic network of small reserves has been progressively set up in the whole
northern area of Guyana, where most harmful activity to primary forests takes place.
As these reserves are inside managed forests, the hypothetical example in the previous section
would require a total surface area of about 725 000 ha to meet both wood demand and reservation

3.7. Maintaining research and experimental efforts

The technical and scientific bases for the ONF’s work are the efforts of tropical forestry researchers,
more particularly work carried out by institutes in the field. Currently the most useful work is carried
out be a grouping of institutes called Silvolab, especially in the experimental station of Paracou, where
the impact of varying thinning pressures on forest ecosystem dynamics are studied. The large-scale
silvicultural trials undertaken by the ONF since 1989 in the pilot forests of Risquetout and Organabo,
complemented by Silvolab trials at Counamai, should deliver essential information in future years on
the relevance of the management system implemented and on possible modifications. In particular,
management should develop in a much more rational and effective way by elaborating fairly easy pre-
management and exploitation methods.
The continuation of this work, as well as other initiatives related to the feasibility of forest station
typology (valid descriptive and analytical tool for forest environments), is essential for the forest
manager’s daily tasks.


4.1. Substantial information and awareness raising efforts

In order to promote solidarity among the main local decision-makers and actors, as well as among
young people, a considerable information and awareness raising campaign has been undertaken.
Measures include the following:

• A three-monthly information sheet is jointly published by the ONF and the Conseil général of
Guyana, addressed to the main elected representatives, local administrations either directly or
indirectly concerned by forests, professionals from the wood industry, associations and schools.
• Forest visits and meetings are held for the main elected representatives and administrations, if
• Technical support is offered to forest harvesters in the form of availability of specialist help for
logging and cutting. The aim is to help train personnel to limit wood wastage from the application
of inappropriate techniques.

4.2. Improving regulations

Existing regulations are already out-of-date and should be revised to adapt to the sustainable
management regime; the Commission régionale de la Forêt et des produits forestiers, which gathers
together all ONF partners, was due to discuss this from autumn 1996 onwards. Concurrently, drafting
was due to begin on the “Regional forestry directions”, which will constitute the forest management
charter for Guyana over the next 10 to 15 years, and which will require approval from elected
representatives, professional, and relevant administrations and associations (including nature

A dynamic strategy aimed at improving global interaction between man and the environment.
In Guyana, forest production affects biodiversity most directly and over the widest surface area
(about 10 000 ha annually). Therefore the relationship between forest managers and forest production
represents the key to conserving biodiversity in local humid tropical forests. The strategy applied
involves managers completely in conservation and regulates forestry activity, including production,
with the aim of maintaining diversity and productivity.
Thus the conservation strategy adopted in French Guyana is not confined to just protecting a
reserve or a species without taking into account human activity and social needs. It is implemented so
that the main economic actors involved in how the natural environment evolves, participate as fully as
possible in defining and implementing conservation policy.
Besides, the dynamic nature of the strategy allows for the future incorporation of improved
knowledge on forest ecosystems and their capacity to react to the various types of damage they might

Incorporation of forest dynamics in the
framework of multiple-use management of
woodland areas

Bernard Rey 1

Dynamics express how ecosystem composition evolves in time and space under the influence of
both man-made and natural factors.
In forest stands, variation in any of the following elements originates from dynamic expression at
spatial level: floristic association, number of individuals, the stratal structure of the population, and
the division of individuals into age categories; they also indicate the geographical continuity of each
ecosystem. Similarly, the different ages phases of a stand, from youth to senescence, express both
temporal dynamics and the temporal continuity of the ecosystem.
These dynamics – brought on by climatic, soil, biological, man-made factors and sometimes
accidental phenomena – are analysed and integrated into forest management decisions and the multiple-
use management of French public forests.

Keywords: Forest ecosystem, forestry dynamics, silviculture, forest management, multiple-use


Like any ecosystem, a forest ecosystem lives and evolves according to variations in the different
factors it is subject to.
The most sudden alterations to ecosystems stem from general climatic factors, such as macroclimatic
variations at geological scale, natural accidents like avalanches or land slides, large-scale fires
irrespective of origin, and several man-made factors like deforestation of land earmarked for agriculture
or more generally for other economic activity.
Besides these non-forestry aspects, silviculture can also introduce substantial variability into the
“natural” evolution of forest ecosystems. In particular, and with good reason, some types of mono-
specific reforestation have been sharply criticized, fuelling sometimes bitter debate about the relative
contribution of a forest’s production function and its less visible contribution to maintaining general
After examining the different dynamic elements of forest ecosystems in forests that have been
managed for a long time, and listing the different parameters and limiting factors, the management
choices and tools used in French public forests will be presented.

Chief Engineer of rural engineering, water and forests, Office National des Forêts, Département Environnement,
2 Avenue de Saint Mandé, 75012 Paris, France.

Ever since Darwin, it has become commonplace to say that competition between species and
groups of plant species (equally true in the animal kingdom) constitutes an authentic fight for survival,
both between and within species, for water and energy sources (in this case, light). The end result is
a delicate balance encompassing all living organisms that compose the ecosystem.
As mentioned in the introduction, this ecosystem tends to evolve more or less brutally like any
object held in equilibrium. This evolution forms the basis and substance of forest dynamics in the
cases reviewed here. Thus we might hazard a rapid definition of dynamics as an expression of the
spatio-temporal evolution, under the influence of natural and man-made factors and parameters, of the
intra-specific and interspecific composition of elements forming a wooded area.


A forest ecosystem observed at a given time t0, i.e., the equilibrium analysed at that moment, is
characterized by a space occupied by:
• a more or less abundant floristic association;
• a number of individuals of each species;
• a stratal stand structure; and
• the division of individuals of each species into age categories.

Any alteration of one or several of these elements leads to a new equilibrium at t1, slightly shifted
from t0. Forest dynamics have been expressed between t0 and t1.
Therefore time is both the tool and ageing device (through the death of individual plants) of
dynamics. In management, it is essential to take this into account and to have thorough knowledge of
an ecosystem’s features at any particular time. These two elements enable us to continue our analysis
by successive examining the effects of space and time.


Previously, we saw that space could be conditioned by four elements. Thus the question is to find
out how these elements can be anticipated, quantified and integrated into forest management practices.

3.1. Floristic association

Foresters urgently require knowledge of the species present and potential species. Such information
helps determine which species will be chosen for preferential silvicultural treatment. This floristic
association is made up of several species which form a group characteristic of the ecosystem (in
natural forests, this group belongs to a plant association and may be taxonomic if rare enough).
Plant succession (a sociological plant association) is conditioned by the adaption of the various
species to a fairly wide range of soil and climatic conditions (the species’s ubiquitous character).
Succession is stable if climate-soil and plant are in equilibrium. Any variation in one of these climatic
or soil factors breaks this balance and means that any species outside of its adaptability range
perishes. Dynamics are brought into play. Management practices can therefore anticipate floristic
association by selecting and deciding species composition, quantified by the number of plants of
each species.

3.2. The number of plants present

As well as adapting to its surroundings, each species is also self propagating. Foresters should
take account of the fact that many-seeded species can potentially occupy more space. Thus in a forest
where oaks (Quercus robur and/or sessiliflora) and hornbeams (Carpinus betulus) compete, the

fecundity of the latter produces a stand where the economically valuable species is rapidly overtaken
by its competitor. Silvicultural intervention corrects this inconvenience regarding the multiple-use of
the wooded area without destroying the hornbeam, as often occurred in the past.

3.3. The stratal structure of the population

Forests always comprise at least two strata, generally more. The forest floor is composed of grasses,
moss, ferns, and mushrooms; the remaining volume is formed of shrub and tree layers which compete
for light; CO2 is apparently less and less of a restricting factor in photosynthesis.
In the soil, water and other mineral elements constitute the sought after elements. Accordingly, one
of the forester’s most important roles is to determine the amount of light reaching the ground, which
is instrumental in shaping forest structures. It is also the key to the water regime as it affects plant
transpiration and evaporation.

3.4. Division of individuals into age classes

This is the key to the spatial component of forest ecosystems. The different age of the tree and
shrub species generate a forest landscape for the silvicultural worker to operate in.
Young plants compete directly with the grass strata, causing species to regress when foresters
intervene solely to their benefit. Nevertheless, a balanced division of age groups at t0 rectifies this
disturbance to general dynamics. One could always object that certain species have problems migrating;
I shall come back to this point further on.
Intermediate and adult age groups provide very strong cover, bringing about a vertical division of
light that is different from younger stages of growth. All the same, a priori, the ecosystem is unaffected
but just at a different stage of its normal biological cycle. Light is given back by the ageing groups, and
the ecosystem’s biological cycle comes round full circle.
Obviously, all these elements assume constant soil and climatic conditions. From then on, a balanced
division of different age groups guarantees a division (itself balanced between species development
stages from birth to death) of all ecosystem species. As well as spatial dynamics, this ensures the
spatial continuity of all the various ecosystems present in a particular massif.


A priori, this aspect of forest dynamics may seem very obvious. It relates directly to the age of
individuals in a population. In forest management, a tree can be considered to go through four main
development stages:
a) A period of youth from seedling to low pole, characterized by strong individual sensitivity to
intra and interspecific competition. During this period, it is generally necessary to complement
tree development with clearing and cleaning.
b) An adult stage of development from high pole to young forest. Each tree has asserted itself in
its respective strata, i.e., ensuring a vital space for energy (light, mineral salts) and water. In
primary forest, competition obviously still takes place, particularly when climatic conditions
prove difficult, bringing about a difference in vigour between individuals of the same species
(thus in relative diameter). In planted forest, clearing and improvement operations enhance this
effect to the benefit of the target species.
c) A period of maturity during which the tree reaches its optimal production diameter and maximum
technological quality, hence true economic value. Trees are generally harvested at the end of
this period.
d) A period of senescence in which the overmature tree retains strong economic value for a long
time (often increasing in value), before showing signs of ageing and then dying. Currently, only

the later stages of primary forest and proper reservations enable these final stages to be reached
on any scale.

At these different ages in a stand’s life, each period is different and exclusive in terms of floristic
association, numbers of individuals and species, and stratal composition. For example, several animal
and plant species are subservient to young stands (shall we say light-seeking species in the strictest
sense); on the other hand, other species can only survive when the tree is ageing, or even dead
(species that feed on decaying or dead matter).
Therefore, a different water and light regime characterizes one development stage of the stand from
another, bringing about temporal dynamics. These dynamics are manifested by the appearance and
disappearance of certain species at a given time t along this temporal phylum.
With constant climatic and soil conditions, the regular reappearance at a given age of the stand of
the same species strictly associated with that stage reaffirms evidence of the temporal dynamics of
forest ecosystems and their continuity through time.


At each moment t in the life of an ecosystem, there is an equilibrium between soil and climatic
conditions and the requirements of the species that compose the ecosystem. Dynamics aid the transition
from the state of equilibrium at t0 to t1 (temporal dynamics).
Nevertheless, a forested space in equilibrium simultaneously provides equilibria matching t0 as
well as t1 in different places (spatial dynamics). In forestry, precise knowledge of limiting parameters
and factors is required to help these equilibria survive and last (especially for economic necessity).

5.1. Climatic parameters and factors

Both animal and plant life is conditioned by energy and water. As light provides plant energy, the
following are important: the amount of light received over the seasons; and the intensity of light
radiation received at each stage of the plant’s development. These two factors are respectively translated
by the duration of the annual vegetative cycle and the degree to which each species seeks light.
Water factors are the useable amount of water in the soil (useful reserve) and water division over
the annual plant cycle. These factors are respectively associated with the species’s hygroscopic or
xerophilic nature, and the specie’s evaporation-transpiration regime.

5.2. Soil factors

The second set of limiting factors are the mother rock, the soil (product of its decomposition or
settled through migration), and humus, which is the product of decomposed leaf litter. Constraints
• the depth of the mother rock;
• soil composition, i.e., ratios of clay, sand and silt and pH;
• the decomposition rate of leaf litter.

The depth of the mother rock conditions root characteristics (commonly called line or revolving)
Soil composition determines water retention capacity and the richness of the clay-humus system.
Litter decomposition rates affect seed germination and surface pH, leading to a succession of species
ranging from very neutral to very acidic.

5.3. Biological factors
These can be summarized briefly as follows:
• species longevity;
• prolific nature – number of seeds and their dispersal capacity;
• the dallelopathic nature of certain species.

These three factors influence intra and interspecific competition.

5.4. Natural and man-made accidents

Properly speaking, these are not limiting factors or parameters, but can break an ecosystem’s
temporal continuum. However, several of them, like fires, can play an active part in renewing tree
populations, allowing forest dynamics to be re-established.

5.5. Man-made factors

Like accidental phenomena, these factors can have a brutal and even devastating effect on forest
ecosystems. Agriculture, urbanization and the building of communication roads often lead to the
irreversible disappearance of forest ecosystems. Even where agricultural land is restored to forest, the
impact of cropping methods is noticeable for a long time afterwards.
Man can also have considerable impact on the replenishment of forest stands. A bad choice of
target species, or monospecific silviculture, can have disastrous consequences for the future of the
Nevertheless, forestry action should not be inhibited by factors and parameters which can alter or
restrict natural dynamics. On the contrary, an ever subtle incorporation of these elements could mark
a step forward in the quality of technical strategies applied in silviculture and the choices set out in
management documents (management plans, forest management).


A forest management plan, a medium-term (15 to 20 years approximately) outlook and management
document, incorporates the main forest functions on a suitably local scale: production of a renewable
and carbon-storing primary material, respect for faunal, floral and historical property, and growing
social demand both for recreation and living areas.
In all evidence, respect for floral and faunal patrimony implies thorough knowledge of ecosystem
dynamics. However, pressure for production of wood, living areas (local people) and recreation (social
and urban, even cultural pressure) can greatly upset the equilibria and dynamics whose mechanisms
we have been examining hitherto.
Within this management framework for French public forests, the following arrangements are
set out:
• Floristic and faunal association: attempting to maintain all species present through specific
protection measures for several species (for example, carrying out forestry work outside nesting
periods, maintaining forest grass by taking away material).
• Structuring the populations into strata: attempting to shape and maintain a sub-canopy throughout
the lifespan of a stand.
• Division of individuals into age classes: attempting to achieve a balance in age categories in
management; analysis of the landscaping impact of stand renewal; pruning basic elements adapted
to the biological, social and landscape context; reasonable utilization of irregular stand structure
in line with natural and climatic factors.

• Clearing and cleaning works: maintaining a maximum level of biodiversity, with the intensity and
timing of operations governed by species temperament, limited recourse to agropharmaceutical
• Thinnings and stand improvement:
– widespread thinning to enable better light penetration, and
– work on codominant species in the stand.
• Stand renewal: resort to natural regeneration as often as possible. In plantations, adoption of
wide spacing (1 000 to 2 500 plants/ha according to species).
• Period of senescence: progressive installation of a network of clusters of ageing trees in large
massifs. These clusters are spread over the whole massif and grow to very old ages, well beyond
the exploitable age of the target species (for example, for an oak forest harvested at an age of 150
to 200 years, clusters will be maintained up to an age of 350 to 400 years). According to the
biological quality of surrounding areas, these clusters can cover a surface area of between 1%
and 5% of the forest.
• At the end of their lifespan, several standing specimens will be kept until they collapse naturally
(maintenance of dead trees). In all regenerated plots, one or two dead trees per hectare will be
• Light management: through widespread thinning (see above).
• Water management: by maintaining forest cover in catchment areas, and progressive replenishment
through wise resort to irregular structures from tree bases to forest floors.
• Soil protection: limited resort to full-scale soil preparation so as not to alter the soil horizon and to
reduce subsidence.
• Incorporating biological factors: this aspect of forest dynamics is essentially dealt with through
the allocation of the number of juvenile individuals of each species (target and accompanying
species). In particular, silvicultural rules teach us not to destroy a species during clearing, thinning,
beating and improvement work, as the main purpose is to optimize light penetration for all the
strata present at a given time in the ecosystem.
• To these biological engineering aspects, we could add civil engineering works in mountain areas
(rectification of torrent flow, anti-avalance works, expanding avalance basins), in dune
environments (initial fixing work , biological engineering), in addition to a set of administrative
and regulatory arrangements.
• Administrative and regulatory arrangements: mainly aimed at reducing man-made impacts on
forest ecosystems by:

The dynamics of forest ecosystems thus seem to be the generator of forest management. Spatial
and temporal incorporation of dynamics means that forests can conserve their beauty and patrimony
in terms of both animal and plant species within the framework of sustainable wood production and
active response to modern social demand. Beyond structural arguments (should the area be regular or
irregular, or tree by tree) which will unfortunately persist for some time yet, here I have tried to outline
the main orientation of our management.
Nevertheless, forestry is a living activity which only conforms to standards in 80% of cases.
Moreover, forestry is undertaken by man (“homo”), “foresticetum” alliance, “rusticae” association,
who themselves only conform to standards in ...%.

Organization of forest system monitoring in
France – the RENECOFOR network

Erwin Ulrich 1

The RENECOFOR (Réseau de suivi à long terme des Ecosystèmes forestiers) Network was set up
by the Office National des Forêts (ONF) in 1992 to supplement France’s forest health surveillance
system. The network constitutes the French portion of an organic series of permanent plots for
monitoring forest ecosystems, the series in question also extending to 34 European countries. The
main purpose of the network is that of detecting long-term changes in the functioning of a wide variety
of ecosystems and to determine what lies behind such changes. The network is made up of
102 permanent plots which are to be monitored for at least 30 years. Each plot is 2 ha in size and has a
fenced-in central or core section measuring half a hectare. RENECOFOR is concerned with the following
matters: (i) dendrometric inventories of 36 trees marked for observation purposes, and 16 trees marked
for foliar analysis; (ii) annual entomological and pathological observations; (iii) dendrochronology
measurements based on cores taken from 30 trees, for the purpose of reconstituting forest history;
(iv) yearly foliar analysis; (v) annual necromass production estimates; (vi) description of two soil
profiles per plot; (vii) soil fertility determination every ten years through intensive sampling; (viii) regular
phytoecological inventorying in specially demarcated areas; (automatic meteorological measurements
outside the forest but close to the plot; and (ix) measurements of atmospheric deposits at open field
points and in the forest at 27 plots since 1993 and of soil solutions at 20cm and 70cm depths at 17 plots.
All data so obtained are recorded at a central database. From 1996 to 1998 all measurements taken
between 1992 and 1995 are to be analysed and published in a series of reports. The reports will present
the current state of understanding at the time the network is being started and will provide the basis
for analysing the evolution of the ecosystem under study in the future.

Keywords: Forest ecosystem; monitoring.

Three complementary networks have been set up in France for the purpose of monitoring forest
ecosystem health in the broadest sense.
The first – nationwide – network was set up in 1989 and consists of corresponding observers
(Barthod 1994). It is a network of persons rather than of precise monitoring points. It is made up of
some 230 persons holding diplomas or otherwise working in forestry who are employed in one or other
of forest management and administration bodies, having been trained specially in forest pathology
and entomology. These persons operate under the Forest Health Department (DSF) of the Ministry of

Office National des Forêts, Département des Recherches Techniques, Boulevard de Constance, 77300,
Fontainbleau, France. Facsimile: (33 1) 64224973.

Agriculture, Fisheries and Food. Their work consists in locating and identifying plant health problems
in the respective Regions of the country and informing DSF of what they have found so that the latter
can monitor the evolution of pest populations and abiotic factors as well.
The second network, as its name implies and in fact, is Europe-wide in scope. It is complete for
France (since 1989) and is based on monitoring defoliation and any abnormal colouring at the permanent
plots (Barthod 1994). The network uses a basic 16 x 16 km grid covering all of Europe, each point in the
grid coinciding with a forest stand forms a link in the network. Almost all countries in Europe are taking
part in the monitoring work. In France observers specially trained for this purpose take note each year
between July and August at some 543 points of the condition of the tree crowns. It is not the primary
purpose to elucidate the causes of plant health problems, even though certain elements towards an
explanation of these are to be had in the course of these activities.
The third network, again Europe-wide in extension though showing differences as between
countries, has been given the name in France of Réseau de suivi à long terme des Ecosystèmes
forestiers – RENECOFOR (Ulrich 1995a; ONF, DFT 1996). Of the three networks this is the one where
monitoring is at its most intensive. Also, despite having a smaller number of plots than the other
networks, its operating costs are the highest among the three. Inaugurated in 1992 by ONF, the
network answers to the need for intensive monitoring of the country’s forests. Its task is to note any
changes taking place in a series of long-term key factors and, where appropriate, to assess the rate at
which these changes occur and their impact on the health condition of the forest. For this reason, the
network has been conceived for a term of at least 30 years. Even though it is the most intensive of the
three systems, RENECOFOR does not record everything. It is concerned, rather, with a selection of
symptoms and their possible causes. It also functions as an alarm for setting in train a number of
complementary investigations on a series of problem situations.

The Ministerial Conference on the Protection of Forests in Europe, held at Strasbourg in 1990
(Ministry of Agriculture and Forests, 1991), declared the following official four main objectives, thus
implementing Resolution No. 1 concerning the creation of a European network for intensive monitoring
of forest ecosystems, namely to:
• acquire comprehensive information on the evolution of certain forest ecosystems in Europe,
noting the historical dimension of the evolution and variations in the vitality of forests, conditions
at the various stations, and climatic events;
• establish correlations among the variations in environmental factors and the reaction of
ecosystems thereto;
• determine the critical pollutant load likely to destabilize a given type of forest ecosystem; and
• facilitate interpretation of the findings furnished by the systematic networks of the European
network type.

RENECOFOR, which came into being after the Strasbourg Conference, is integrated into a wider
European complex. The European Union attaches great importance to protecting the continent’s
forests, and between 1985 and 1995 issued six regulations laying down parameters for the intensive
monitoring envisaged. Together with RENECOFOR, France’s ONF has complied with not only this
mandatory series of “specifications” but has taken on several additional tasks.
The Strasbourg Conference was followed in 1993 by the Helsinki Conference (von Weissenber et
al. 1993; Loisleslosky et al. 1995).
RENECOFOR is financed by the European Union, ONF and the Ministries of Agriculture, Fisheries
and Food, and of the Environment and by the Agency for the Environment and Energy. ONF is both

commissioning body and project manager for the network and maintains intensive collaboration with
many public and private institutions.

RENECOFOR was set up in its entirety between 1992 and 1995. It consists of 102 permanent
monitoring plots, each approximately 2 ha in size. Most of the basic operations allowing of an appreciation
of these plots were complete by late 1995, and represent the starting point for understanding the
conditions obtaining in the respective ecosystems. The first observations, measurements and analyses
were interpreted in 1995 and 1996, at first subject matter by subject matter and then in a multidisciplinary
approach. The conclusions will serve as a basis for subsequent measurements to be conducted at
intervals that will vary depending on the subject matter. The purpose of later work will be to establish
whether any operational changes have supervened.
The work is directed by the Network Coordination Centre (with its two graduate and two diploma-
holding staff) based at Fontainebleau. The eleven graduates and 17 diploma-holding members in the
eight Inter-regional Technical Sections of ONF’s Technical Research Department, together with the
188 monitors and their stand-ins for the various plots, perform most of the ongoing measurement and
sampling tasks there. Subcontracting arrangements are made with public bodies and private firms (e.g.
analysis laboratories). Work is supervised by the ONF’s Science Council and a European-level scientific
Monitoring data are interpreted, where France is concerned, by ONF and in specific areas by
research partners – at the Institute National de la Recherche Agronomique (INRA), the Centre National
de la Recherche Scientifique (CNRS) and the universities. The associate expert group scheme puts
forward proposals for additional research supplementing network activities, in which case the permanent
plots are also used for specific types of research. In this way interested researchers can access an
already very well stocked information base. At the European level, findings are evaluated by a consultant
whose services are funded by General Directorate VI of the Commission of the European Community
in Brussels. Data for France and those for other countries are forwarded to him at regular intervals.

Before any monitoring operations or measurements or sampling or analytical work are put in hand,
methods are defined in collaboration with researchers having competence in the respective subject
matter fields and are brought together in the form of reference handbooks, where, once again, experts
are called in for the purpose.

4.1. General aspect of a permanent plot

Most plots are to be found in regular (single-species/even-aged) stands – this in order to have
sufficient homogeneity both in a stand itself and in the measures it is proposed to carry out there. The
species monitored in this way are the oaks (Quercus petraea, Q. robur), beech (Fagus sylvatica),
Norway pine (Pinus sylvestris), sea pine (Pinus pinaster), laricio pine (Pinus nigra subsp. laricio),
spruce (Picea abies), silver fir (Abies alba), Douglas fir (Pseudotsuga menziesii) and larch (Larix
decidua). Each plot consists of a core area half a hectare in size, which is fenced off and given a
“neutral” zone around it about 30 m wide. The various items of equipment are placed in the core area
to afford better protection against theft and vandalism.

4.2. The operation and some of the findings

There follow details concerning the work conducted and where, even now, some preliminary findings
are available.

4.2.1 Dendrometric inventory
The first complete dendrometric inventories were conducted between 1991 and early 1993. In the
winter of 1995/96 the inventorying process was repeated in order to synchronize measurement work in
France with that being done in the other European countries. A first collation of the findings is
expected for 1996/97. This will show any marked differences in evolution in growth or crown size.

4.2.2 Dendrochronology
This work is done by INRA at its François Lebourgeois Forest Research Centre’s Phytoecology
Laboratory, Nancy, in collaboration with ONF. The purpose here is to retrace the history of forests by
observing the radial growth of the trees. The ages of the stands making up the network vary from
about 20 to 120 years, and it is therefore essential to find out about the life of these stands up to the
time the network was created. If earlier crises can be shown to have occurred, it will be possible to
understand the present behaviour of the stands and their sensitivity to climatic extremes such as
drought or frost. The work here is expected to terminate in 1996 and will furnish an identity card, as it
were, in terms of radial growth for the respective stands.

4.2.3 Evaluation of the state of health of stands

At each plot, 36 dominant or codominant trees, which have been duly marked in the core area, will
be used for ongoing sampling for the purpose of year-on-year recording of the extent of defoliation
and any abnormal discolouring of the crowns. The same trees are monitored for signs of pathology or
of insect attack. The observations in question are carried out by teams of monitors of the European
Network and corresponding observers of DSF. From the first two years’ monitoring of the state of
health of the stands (1994 and 1995) it has been possible to discern, as is presently the case, and
despite the fact that the stands initially selected were healthy, a number of plots showing varying
degrees – slight to medium – of defoliation (Ulrich et al. 1995 and Ulrich and Lanier 1996). The signs
noted in the RENECOFOR system agree with those encountered in the European Network.

4.2.4 Foliar diagnostics

Here the task is to monitor each year the concentrations of micro- and macro-elements that are
essential in tree nutrition. The first samplings were carried out in 1993; and the first analyses (Ulrich
and Bonneau 1994) showed a mineral nutrition above the critical thresholds for nitrogen, potassium,
calcium, manganese and zinc in most cases. Where phosphorus is concerned, the level is below the
critical value in 48% of the plots. The deficit affects all species; and the findings warrant the conclusion
that this element is insufficient in many forest soils. Magnesium and iron are below the critical threshold
in 20% of the plots for oak, beech, fir, spruce, douglas fir and pine. In 33% of the plots concentrations
of copper are below the threshold. However, the definition of mineral-deficient for the generality soils
taken for reference at the present time are somewhat theoretical, and with RENECOFOR it may perhaps
be possible to redefine these.

4.2.5 Estimates of necromass from leaf, litter, etc.

Once necromass and the mineral content of forest litter are known, it is possible to a large extent to
estimate, from their rate of decomposition, the amount of nutrient elements entering into the internal
nutrition cycle. The quality and quantity of the primary production can considerably influence the
nutrition status of a given stand and, by that token, its health status. Litter is harvested three to five
times in the course of the year, ten persons being assigned to the task. Leaves, branches and fruits of
the main tree species present are sorted out, and the rest of all secondary species. The dry weight for
these four groups is noted and extrapolated in per hectare terms. Chemical analysis will be done on

these samplings beginning in 1997, provided funds are forthcoming. This is a highly onerous operation
and will take some years to carry through, but it is necessary in order to have an idea of the magnitude
of year-on-year variation in the necromass. In due course, it should be possible to replace these direct
measurements with indirect estimates, for example by means of foliar analysis, of nutrient production
from necromass.

4.2.6 Pedology
In order to place each of the stands studied in 1994/95 in its edaphic context two highly detailed soil
profiles have been produced with particular attention to the description of humus forms. It is the
intention that between that date and late 1996 the findings will be brought together in a report. This
will serve as a basis for interpreting how soils function and the study of the repercussions that this
functioning has on forest stands. Pedological descriptions are the result of this work of a group of
twelve experts reporting to the head of ONF’s Pedology Division, Alain Brêthes.

4.2.7 Soil fertility analysis

Soil fertility is among the factors determining stand health. In keeping track of developments here
one has to allow for the fact that chemical concentrations may vary even within short distances from
one another in the ground. Similar remarks apply to the vertical and horizontal proportions in holorganic
profiles and in fine-textured soils whence the trees derive their nourishment. For this purpose a gross
sampling method is applied to the half-hectare core area (Ulrich 1994a). Every ten years the sampling
is repeated in order to determine any evolution in fertility. All samples are both analysed and archived
in a soils collection. The first sampling campaign was conducted between 1993 and 1995. It is to be
followed in 1996 and 1997 by an analysis of the findings and by the production of a report presenting
a description of the original condition of the soils in question.

4.2.8 Phytoecological inventory

Scientific studies are continually revealing changes taking place over time in floristic composition
(abundance/dominance of the respective species), the latter being determined in relation to eight
carefully identified strips 50 m long by 2 m wide, half of their number being situated in the core area and
half outside. When monitoring campaigns are being undertaken, the actual observations are done in
spring, summer or autumn. The frequency of such campaigns is determined by reference to a research
plain. Apart from this, supplementary observations are to be undertaken prior to any clear felling and,
again, two years afterwards – this in order to study the reactions of the flora whenever a stand is
opened. The first of these campaigns was conducted by twelve experts in 1994 and 1995. Data from
over 2300 situations surveyed are processed and interpreted the University of Savoy’s Dynamics of
Ecosystems at High Elevations Laboratory. A summary report of the findings is to be published by
ONF in late 1996.

4.2.9 Meteorology sub-network

Meteorological phenomena have an influence on most troubles in forestry. For this reason it was
thought to be essential that the large number of measurements carried out should be accompanied by
automatized meteorological measurements. In 1994 and 1995, some 26 automatic stations monitoring
three (or six) parameters were accordingly set up at out-of-forest points but close to plots making up
the CATAENAT (see 4.2.10) sub-network (Ulrich 1995b, Ulrich et al. 1995a). Data are collected every
three or four days and transmitted by telephone or by smart cards every fortnight. Four objectives
were set at the time the network was established (others can be introduced in the future, though
without altering their technical configuration), namely:

• understanding forest evolution and linkages among meteorological parameters and the
physiological parameters of trees in risk-free meteorological situations. The purpose here is to
understand natural variability in certain phenomena;
• understanding effects of extremes (i.e. in excess of a given threshold) situations on stands;
• understanding cumulative effects, i.e. in long-term abnormal situations even those abnormalities
that may not be extreme; and
• evidencing long-term evolution (where this obtains) in the local climate and interpreting any
influence that this might have on other factors that are observed. If any increased greenhouse
effect really brings about an alteration in the climate, it is unlikely to find expression in the same
way everywhere in the world. Local expression of linkages is thus what needs to be observed.

The technical management of the sub-network also extends to a strict quality control programme.
A preliminary report taking stock of the first series of measurements and suggesting a large number of
approaches for monitoring meteorological indicators is now available (Ponette et al. 1996).

4.2.10 The CATAENAT system

Accretions of nutrient and acidifying or fertilizing elements from rainfall and from dust are recognized
as being among the factors that may modify the functioning of forest ecosystems.
Since early 1993, the routine measurements carried out in the 27 plots of the CATAENAT network
(acronym representation total acid load of atmospheric origin in dryland natural ecosystems), the main
purpose accordingly being to add to our knowledge about the impact of atmospheric deposits on the
ecosystem in question. The yearly accretion of chemical elements deriving from dry, wet and hidden
deposits is estimated, the additions thus entering into the internal nutrition cycle. At 17 of the 27 plots,
annual concentrations of solutions in the subsoil for 1993 and 1994 point, inter alia, to high rates of
addition exceeding 10/kg/ha/a of nitrogen and sulphur in several stands (Ulrich and Lanier 1994;
Ulrich et al. 1995c, d). Sodium and chloride additions are particularly high in plots near the Atlantic
coast or otherwise directly under marine influence, even if tens of kilometres from the coast.

4.2.11 Forest history

The past history of forests may be very different from one region to another. Past uses have their
repercussions even today, and this is why a study of all the forests where permanent monitoring plots
have been set up has been put in hand. In this way it is hoped that information can be had on past
factors that continue to have an effect on present and future functioning or malfunctioning of forests.

4.2.12 Computer acquisition, storage and processing of data

A vital desideratum in the operation of a network bringing together such a wide range of disciplines
and observations is that data management shall be expeditious, rationalized and reliable. Furthermore,
it must be possible to expedite these data to those whose task is to collate and interpret them. This is
why the Coordination Centre in 1993 set up a readily accessible data handling system (Paradox for
Windows). By late 1995 some 2.4 million raw data had been entered in this database, which now
contains 130 tables and 1 800 fields. The base will continue to evolve in order to respond to changes
and to future needs for adapting and improving the network. Important work has been done and a
considerable amount of data have been made available for organized collation and for replying to the
many enquiries that arrive daily from those running the network and from other users.

The first five years of the RENECOFOR network, some three to four of which have been given over
mainly to the installation process, have served to show the reliability of this intensive monitoring
system for France’s forest ecosystems. Reliability is reflected not only in the serious concern for
quality, and in the operations conducted, but also in the fact that it has found wide acceptance in
French and European scientific quarters. Acceptance, moreover, is linked to the fact that it has been
possible to build up a mass of information that was not available before previously regarding the
current functioning of forest ecosystems. Between 1990 and 1995 the network is estimated to have
cost in the aggregate FF 28.5 million. The results obtained should lead on to the emergence of other
questions and to novel avenues of research.

Bibliography Ulrich, E. 1995a. Le réseau RENECOFOR: objec-

Barthod, C. 1994. Le système de surveillance de tifs et réalisation, Revue Forestière
l’état sanitaire de la forêt en France, Revue Française, 47, 2: 107-124.
Forestière Française, 46, 5 :564-571. Ulrich, E. 1995b. Le réseau météorologique fores-
Loiskekoski, M., (Eds.), Granholm, H., Starr, M., tier: une aide ˆ l’interprétation du change-
Halko, L., Oinonen, U., (Eds). 1995. Interim ment de l’état sanitaire des forêts et des tend-
report on the follow-up of the second Minis- ances ˆ long terme du climat local. Actes du
terial Conference on the Protection of For- colloque “Avenir des réseaux clima-
ests in Europe, 16-17 June 1993 in Helsinki, tologiques”, 7-8 juin 1995, Toulouse.
255 p. Ulrich, E. & Bonneau M., 1994. Etat nutritionnel
Ministère de l’Agriculture et de la Forêt (France) des peuplements du réseau RENECOFOR:
(Eds.), 1991. Conférence Ministérielle pour brève synthèse de la première année d’échan-
la Protection des Forêts en Europe organ- tillonnage et d’analyse (1993), La Santé des
isé par la Finlande et la France avec la collab- Forêts (France) en 1993, Ministère de
oration du Conseil de l’Europe, Strasbourg, l’Agriculture et de la Pêche (DERF-
décembre 1990, 258 p. Département de la Santé des Forêts):
Office National des Forêts, Département des Re- 51-56.
cherches Techniques, 1996. Notice de Ulrich E. & Lanier M. 1994. Les dépôts atmos-
présentation du Réseau National des suivi phériques en 1993: résultats du sous-réseau
à long termes des Ecosystèmes Forestiers. CATAENAT du réseau RENECOFOR, La
ISBN 2-84207-021-6, 38 p. Santé des Forêts (France) en 1993, Min-
Ponette, Q. et al. 1996 : RENECOFOR – Sous- istère de l’Agriculture et de la Pêche (DERF-
réseau météorologique forestier. Bilan de Département de la Santé des Forêts): 56-61.
la phase test (1994/1995) et perspectives. Ulrich E. et al. 1995a. Quel temps fait-il en forêt?,
Editeur : Office National des Forêts, Dépar- Arborescences, 59: 25-28.
tement des Recherches Techniques, ISBN 2- Ulrich E., Lanier, M. & Schneider, A. 1995b. RE-
84207-2, 102 p. NECOFOR: un premier aperçu de l’État sani-
Ulrich, E. 1994a . Le suivi à long terme de la fertil- taire des peuplements du réseau. La Santé
ité des sols forestiers dans le cadre du Ré- des Forêts (France) en 1994, Ministère de
seau National de suivi à long terme des l’Agriculture, de la Pêche et de l’Alimentation
Ecosystèmes Forestiers. Actes du colloque (DERF-Département de la Santé des Forêts):
“L’eau et le sol” de l’Association 54-56.
Internationale des Entretiens Eco- Ulrich E. et al. 1995c. Interception des pluies en
logiques, 14-16 mars 1994, Dijon, France: forêt: facteurs déterminants. Interprétation
117-124. des mesures réalisées dans le sous-réseau

CATAENAT de RENECOFOR, ONF – Bulle- sanitaire des peuplements du réseau en
tin Technique, 30:33-44. 1994 et 1995. Editeur : Office National des
Ulrich, E., Lanier, M. & Schneider, A. 1995d : Forêts, Département des Recherches Tech-
Dépôts atmosphériques et concentrations niques, ISBN 2-84207-047-X, 40 p.
des solutions du sol (sous-réseau CAT- von Weissenberg, K., Halko, L., Loiskekoski, M.,
AENAT), rapport scientifique sur les années Norros, K., Rizk, N., Veltheim, T., (Eds.), 1993 :
1993 et 1994. Office National des Forêts, Report on the follow-up of the Strasbourg
Département des Recherches Techniques, resolutions, Ministerial Conference on the
ISBN 2 – 84207-011-9, 165 p. Protection of Forests in Europe, 16-17 June
Ulrich, E. & Lanier, M. 1996. RENECOFOR – Etat 1993, Helsinki, 203 p.

A global overview of forest conservation

S. Iremonger 1,2, V. Kapos 1, J. Rhind 1, R. Luxmoore 1

Mounting global concern over the conservation status of the world’s biodiversity, especially at
ecosystem and species levels, has led to calls for increasing the extent of protected areas and for
identifying priority areas for conservation. Although most decisions to establish protected areas are
made at the national level, international perspectives are necessary both to assess the status of
ecosystems occurring in more than one country and to target the use of international resources.
Species and ecosystems are not contained by political boundaries, and international cooperation is
essential to ensure their preservation.
One means of establishing priorities for conservation is analysis of the degree to which existing
networks of protected areas are representative of the full range of ecosystems and species. One
pragmatic but crude approach to performing such an analysis on an international scale is to use very
broad ecological classification to define ecological zones (EZ), and to assume that vegetation occurring
in different EZs does indeed belong to distinct ecological types. In this paper we present the results
of a global analysis of existing forest cover and its protection in each EZ.
The analysis is based on WCMC’s global forest cover data set, which has been compiled over the
last ten years and kept as current as possible by replacing old national data sets with new ones. The
analysis also used WCMC’s protected areas databases, which are similarly continuously updated as
new data become available. The global data sets were divided into twelve geographical regions
according to both biogeographical and political criteria, and an appropriate ecological zone classification
was applied to each region.
With the exception of Insular South-East Asia, all of the geographical regions have substantial
numbers of EZs with less than 10% of the extant forest cover protected. The EZ with highest percentage
of its forests protected globally was the Polar desert of Australasia, but the total area was very small
(25 km2 protected), as this is not an EZ which generally supports forest growth (the percentage of this
zone that was forested was only 1.15%). The EZ with the least of its forests protected was the
Subtropical broadleaved semi-evergreen forests zone of the Middle East (27 km2 protected). In terms
of absolute area, the greatest amount of protected forest was found in the Tropical moist broadleaf
forests of South America (712 000 km2 protected), the next largest in the Boreal forest/taiga zone of
North America (348 863 km2). The ten major global regions defined for this study varied from having
16.61% of their forests protected (Insular SE Asia) to 1.84% (Russia).
From this analysis it is possible to highlight ecological zones with poorly protected forests for
future conservation efforts. In many cases, these efforts will require international cooperation as EZs
transcend national boundaries. Further work is required to differentiate the forest types within the
broad EZs and carry out an additional analysis, so that recommendations for protection can be more
specific. This work is already under way.

World Conservation Monitoring Centre, Cambridge, U.K.
Center for International Forestry Research, Bogor, Indonesia.

“EZ” is used throughout this paper to denote Ecological Zone, Ecoregion, Major Habitat Type,
Landscape Category and Ecofloristic Zone, according to whichever regional system was used.
“Region” is used with a capital within text where referring to the major world regions used in the
analyses, e.g., North America.

Mounting global concern over the conservation status of the world’s biodiversity, especially at
ecosystem and species levels, has led to calls for increasing the extent of protected areas and for
identifying priority areas for conservation. Although most decisions to establish protected areas are
made at the national level, international perspectives are necessary both to assess the status of
ecosystems occurring in more than one country and to target the use of international resources.
Species and ecosystems are not contained by political boundaries, and international cooperation is
essential to ensure their preservation.
One means of establishing priorities for conservation is analysis of the degree to which existing
networks of protected areas are representative of the full range of ecosystems and species. At the
national level, detailed ecosystem or vegetation classifications can provide a basis for assessing the
representativeness of the existing protected areas network. However, such an analysis is far more
problematic at regional and global levels. Although ideally it should be possible to harmonize classes
of existing vegetation across regions, thus permitting analysis of detailed ecological units as they
exist on the ground, such harmonization is an arduous process which has not yet been accomplished.
Another, pragmatic but cruder approach is to use very broad-scale ecological classification to define
ecological zones (EZ), and to assume that vegetation occurring in different EZs does indeed belong to
different ecological types.
Several studies have highlighted the status of forest protection and decline for particular regions
using some version of ecological zones (e.g. Lysenko et al. 1995, Mackinnon 1996). Although FAO
have compiled data on forest resources globally (e.g. FAO 1995), the methodology used has differed
between developed and developing countries. A more uniform approach to the forests of the different
regions of the world is called for (Päivinen 1996). The World Conservation Monitoring Centre (WCMC)
recently produced an analysis of protection of EZs in the tropics (Murray et al. 1996), using the
system of ecofloristic zones developed for FAO.
In the present paper we advance on this work by carrying out a global analysis; the extent of
current forest cover in each EZ is determined and the protection of that existing forest cover is
assessed. Because this analysis examines current forest cover, it depends upon the use of WCMC’s
global forest cover data set. This data set was compiled by WCMC over the last ten years and has
been kept as current as possible by replacing old national data sets with new ones. The analysis also
depends on WCMC’s protected areas databases, which are also continuously updated as new data
become available . Previously, these two data sets were combined to produce an initial analysis of the
protection of global forest cover in five very broad classes (WCMC 1996). The present study analyses
forest cover in much more detailed EZs derived from different classification systems for each of the
major regions of the world (see below).


The forest coverage

The data for the forest coverage of the globe was put together from many national and regional
data sets, with varying resolutions (see Appendix 1 for sources). Some of these were provided to

WCMC only in hard copy, in which case the forest polygons were digitized from this. The entire global
data set was 400 Mb in size. This data set was used in a previous analysis showing the amount of
forest under protection globally in five broad classes (see above). Finer EZs were not used in that
study. The present investigation attempts to illustrate in more detail the conservation status of more
well-defined forest types. A number of ecological zoning systems are now available in digital form,
and it was these that were used to differentiate the forest types. The data sets were first divided into
geographical Regions, according to a mixture of biogeographical and political criteria, taking into
account the EZ systems that were available for categorizing the forests in each Region. The resulting
regional sets and the corresponding ecological zonation schemes used are shown in Table 1. The
countries included in each Region are listed in Appendix 2.

The ecological zones coverages

The EZs defined for each region varied because of the different systems that were used (see
Table 1). In all seven different schemes were used, the two American ones were carried out by the
same group using the same criteria for defining zones, so they are directly compatible and comparable.
The best scheme for Europe was a near-final draft which was designed for use only in Europe, using
CORINE land cover classification units (Bohn & Katenina 1994). The schemes for Africa and South
and South-East Asia were designed according to criteria drawn up by Blasco and Legris (1982), and
finalized by Sharma (1986, 1988). In SE Asia and in Africa the EZs coverages encompassed large areas
of ocean (e.g., from Papua New Guinea to Micronesia). These were clipped to the Digital Chart of the
World coast boundaries (ESRI, 1993). A similar EZ scheme for South America was also drawn up
according to the same criteria (Lavenu et al., 1988), but this has been superseded by the WWF-US
classification (Ricketts et al. 1996) and so was not used. The remaining areas of the world were Russia,
the Middle East and the Far East, and Australasia. Schemes which were considered for these areas
included Bailey (1989), Olson and Watts (1982), Holdridge (1962) and Milanova and Kushlin (1993).
Olson and Watts’ scheme was considered to be unsuitable for our purposes because it was developed
for estimating global carbon flux and included land use as part of the basic land cover data set. Bailey’s
first major division into “Domains” did not fit well with the first major division in the other data sets
being used: Holdridge and Milanova and Kushlin were more compatible. However, the latter did not
seem to define the Australasian EZs as well as Holdridge, so Holdridge was used in this southern
hemisphere Region. Milanova and Kushlin, however, was found to be best in the remaining areas,
which would seem logical as the authors were based in Moscow, the former USSR, and would therefore
be most familiar with these states.
Table 1. Divisions of the world into regions, and the ecological zones system used in each
region for the analysis.
World Region Ecological Zone Scheme
North America Olson & Dinerstein, 1996
Mesoamerica Dinerstein et al., 1995
Caribbean Dinerstein et al., 1995
South America Dinerstein et al., 1995
Africa Sharma, 1988
Australasia Holdridge, 1962
Europe Bohn & Katenina 1993
Middle East Milanova & Kushlin 1993
Russia Milanova & Kushlin 1993
Far East Milanova & Kushlin 1993
South & South-East Asia FAO, 1989
Insular South-East Asia FAO, 1989

All of the zonation schemes used were more or less hierarchical in that they present the ecological
land classifications in a nested fashion. For example the Latin America scheme first divides this
continent into Mesoamerica (which includes Central America and Mexico), the Caribbean and South
America. Within each of these three geographical units there were five main subdivisions, and each of
these subdivisions was again subdivided as outlined below for Mesoamerican tropical broadleaf

Tropical broadleaf forests
Tropical moist broadleaf forests
Tropical dry broadleaf forests

It was this level in the hierarchy that was used for the analysis. There were further subdivisions in
these two last categories, but these were not used for this analysis.
The use of EZs as an indicator of different ecological forest types is not without its problems. An
ecological zone that generally supports a herbaceous ecosystem, e.g., the Russian steppes, will be
found to have some forest cover. This will be due to the production of conditions allowing forest
development by local variations in such environmental parameters as altitude, aspect, slope, soil type
and drainage regime. The coarse scales at which the EZ coverages were developed also cause forest
to be found in essentially non-forest EZs. This phenomenon will be discussed later with the
interpretation of the results.

The protected areas coverage

WCMC has been compiling and analysing data on the world’s protected areas for more than a
decade. The WCMC Protected Areas Database holds data on over 40 000 sites around the world
designated for nature conservation purposes, together with details of sites designated under
international conventions and programmes, such as the Ramsar and World Heritage Conventions,
and the UNESCO-MAB Biosphere Reserves Programme. The data are used to produce informational
items such as the United Nations List of National Parks and Protected Areas.
Information about protected areas is obtained from official sources (those government agencies
responsible for administering protected areas), and elsewhere, through a global network of contacts
ranging in profession from policy-makers and administrators to land managers and scientists. The
information in the databases includes such items as the site name, the IUCN management category
(see Table 2), the size of the area and where possible the exact shape and location of the area in the
spatial databases. For the current analysis all protected areas in management categories from I to VI
were included.
The spatial database includes two types of locational information. In many cases there are polygons
inserted for the site, where this information is known. Where this is not available a point has been
inserted, representing the centre of the site. In order to carry out the overlay analysis the data had to
be processed so that all sites were polygonized. Polygons were made for sites which only had point
locations by using the area information in the textual databases and drawing a circular polygon of the
relevant area around the point location of the site. In some places there was a significant amount of
overlap of protected areas. This was mainly in instances where a site with stringent protection was
located inside a site with a lesser degree of protection, or else where the circles drawn were not the
exact locations of the sites, which in reality might not actually overlap because of their shape. To avoid
massive over-representation of the amount of forest under protection due to the overlap factor, the
protected areas layer was dissolved so that overlapping sites just dissolved into one polygon. This

Table 2. IUCN Protected areas management categories and their definitions (IUCN 1994) 1
Category Name and definition

CATEGORY I Strict Nature Reserve / Wilderness Area: protected area managed mainly for science or wilderness

CATEGORY Ia Strict Nature Reserve: protected area managed mainly for science

Definition: Area of land and/or sea possessing some outstanding or representative ecosystems, geological
or physiological features and/or species, available primarily for scientific research and/or environmental

CATEGORY Ib Wilderness Area: protected area managed mainly for wilderness protection
Definition: Large area of unmodified or slightly modified land, and/or sea, retaining its natural character and
influence, without permanent or significant habitation, which is protected and managed so as to preserve its
natural condition.
CATEGORY II National Park: protected area managed mainly for ecosystem protection and recreation

Definition: Natural area of land and/or sea, designated to (a) protect the ecological integrity of one or more
ecosystems for present and future generations, (b) exclude exploitation or occupation inimical to the
purposes of designation of the area and (c) provide a foundation for spiritual, scientific, educational, recreational
and visitor opportunities, all of which must be environmentally and culturally compatible.

CATEGORY III Natural Monument: protected area managed mainly for conservation of specific natural features
Definition: Area containing one, or more, specific natural or natural/cultural feature which is of outstanding
or unique value because of its inherent rarity, representative or aesthetic qualities or cultural significance.

CATEGORY IV Habitat/Species Management Area: protected area managed mainly for conservation through management

Definition: Area of land and/or sea subject to active intervention for management purposes so as to ensure
the maintenance of habitats and/or to meet the requirements of specific species.
CATEGORY V Protected Landscape/Seascape: protected area managed mainly for landscape/seascape conservation and

Definition: Area of land, with coast and sea as appropriate, where the interaction of people and nature over
time has produced an area of distinct character with significant aesthetic, ecological and/or cultural value,
and often with high biological diversity. Safeguarding the integrity of this traditional interaction is vital to the
protection, maintenance and evolution of such an area.
CATEGORY VI Managed Resource Protected Area: protected area managed mainly for the sustainable use of natural

Definition: Area containing predominantly unmodified natural systems, managed to ensure long term
protection and maintenance of biological diversity, while providing at the same time a sustainable flow of
natural products and services to meet community needs.

UA Unassigned
No management category is appropriate, or the site does not meet the internationally recognised definition of
a protected area.
A protected area is defined in the new Guidelines for Protected Area Management Categories as An area of land
and/or sea especially dedicated to the protection and maintenance of biological diversity, and of natural and
associated cultural resources, and managed through legal or other effective means. IUCN

process may result in a slight under-representation of the amount of forest under protection, but the
error would be much less than the overestimate that would have occurred had the dissolve not been
carried out. More accurate results can only be obtained by building up and refining the spatial
information on protected areas in the databases, which is an ongoing process.
The dissolved protected areas layer was then geographically divided for the analysis into Regions
which corresponded exactly to the forest coverage divisions.

The process
The programme used for the analysis was ARC/INFO 7.0.4 on Sun Ultra Unix workstations hardware,
and the analysis required about 1 Gigabyte of disk space. Two sequential analyses were performed on
each regional three-layer data set (forests, EZs, protected areas). Firstly the forest coverage was
overlaid with the ecological zonation scheme chosen for the Region to yield percent of the EZ which
is covered by forest. This area (forest per EZ) was overlaid on the protected areas data set to show
percent of forest under protection by EZ.

The only nations that are represented in more than one of the regional analyses are Malaysia and
Russia, the former because part of it is insular and the other continental, and in the latter case the
isolated section of Russia surrounded by Poland and Lithuania has been analysed along with these
surrounding nations and the rest of Europe.
Results were generated by Region. Each analysis produced both total area numbers and percentage
land cover, indicating the amount of forest in each EZ and the amount of forest in the zone that is under
protection (see Appendix 3).
The average percentage of forest under protection is shown in Figure 1 for all Regions. These
figures seem relatively high, seven out of the ten Regions showing percentages over 10. However, the
breakdown of the results of each Region to show the amount of forest protected by EZ shows that
within the Regions with high average percentage (Figure 1), some forests are minimally protected. The
total amount of forest in each Region varied from 8.3 and 8.1 million km2 in North America and Russia
respectively, to 0.1 million km2 in the Middle East (see Appendix 3). The number of EZs with forest in
each Region varied from seven in Insular SE Asia to 23 in the Far East (Table 3), the ranges being more
or less indicative of breadth of latitude. The actual EZ system used for each Region also influenced
this figure, because Milanova and Kushlin (1994) identified “interzonal landscapes” in addition to the
zonal ones: in all systems the zones were determined primarily by climate, but in these latter landscape
types crossed through major climatic types, e.g., “River valley landscapes”, “Mountain glacier

Table 3. Number of ecological zones with forest in each region.

Region Number of Ecological zones
Africa 12
Europe 10
Russia 15
Middle East 13
Far East 24
Continental SE Asia 9
Insular SE Asia 7
Australasia 18
North America 9
Latin America and Caribbean 21

Forest protection by Region is illustrated in Figures 2-11. In all Regions except the SE Asian
(Figures 2 and 3) ones and Australasia there was a figure generated by the analysis which showed an
amount of forest which was not assigned to any EZ. The proportion of the forest area that this figure
represented is shown in the captions to Figures 2-11, and the absolute amounts are given in Appendix
3. This was the result of mismatches between (a) the Region edges as defined by the coverages
(different origins of data sets, different resolutions, different projections) and (b) holes in the coverages,
mainly due to the presence of water bodies, that were not in exactly the same position or the same
shape in each data set.
As stated in the Methods section above, the EZs were used to distinguish major different forest
types in the Regions. Some EZs were naturally more conducive to forest growth, but pockets of forest
may also be found in EZs that are mainly herbaceous because the climate is generally too dry or cold
for forest growth, the forests occurring due to some local condition of soil or topography for example.
Within Insular SE Asia the most extensive forest area was in the Lowland tropical very moist EZ
(Figure 2a). The forest covered > 60% of the whole EZ. However not a great deal of this is under
protection (Figure 2a, Appendix 3). The percentage of the EZs that were covered in forest varied from
< 10% in the Lowland tropical moist with long dry season to > 70% in the Montane moist category.
Although the Alpine EZ was over 60% forested (Figure 2b), Figure 2a shows that very little area was
involved (see Appendix 3).
In Continental SE Asia by far the most extensive forest cover was in lowland tropical sub-dry,
Lowland tropical moist and Lowland tropical moist with long dry season (Figure 3a). Only two EZs
were over 30% forested, Lowland tropical very moist and Premontane tropical moist (Figure 3a). The
Lowland tropical moist with long dry season, Lowland tropical sub-dry and Premontane tropical dry
EZs had between 14 and 26% forested, the remaining four zones, which probably have low forest
cover naturally because of low rainfall or low temperature, all were under 4% forested. However
although the Alpine zone had only 4% forest cover, it had the highest percentage protection in the
Region at 30% (Figure 3a and Appendix 3). Two of the other low forest categories had no protected
forest: Lowland tropical arid desertic and Montane tropical dry, and the rest of the zones had between
10 and 5% of the forest under protection.
The Far East (Figure 4a and 4b) had 23 EZs, more than any other Region. The most extensive forests
were the Boreal taiga and the forest cover in the Temperate desert zone. This seems anomalous but
Figure 4b shows that the latter zone was only 9% forested, whereas the former was nearly 40%
forested. The greatest percentage forest cover was found in the Temperate mixed deciduous-coniferous
forests, at 64%, and 16% of these were under protection. The EZ which had by far the largest percentage
of its forest under protection was the Subtropical semi desert (62%). All other zones (22) had less than
14% of their forests protected, and 15 EZs had less than 5%.
The Middle East included temperate, subtropical and tropical EZs, but no boreal areas. All EZs
were low in forest cover, both in the absolute figures (Figure 5a) and in percentage terms. Only one of
the 13 had over 10% of its land area forested, Temperate forest-steppes and prairies, and only one
other more than 6%, Subtropical broadleaved-coniferous evergreen forests. Percentage protection
was also low in every zone, the highest being 9% in Subtropical deserts, the zone with the second
most extensive forests. Forests covered 2% of the total area of this zone though, the absolute figure
only being large because the EZ was very extensive. Seven of the 13 EZs had less than 5% of the forest
cover under protection.
Forested EZs in Russia included boreal, temperate, subtropical and interzonal landscapes (Bog and
marsh, River valley). Boreal taiga forest was by far the most extensive forest type (Figure 6a). Three of
the 14 EZs had more than 50% forest cover, one of which was Subtropical semi-deserts. This (latter)
also had by far the highest percentage under protection (62%). The Boreal open woodlands zone was

96% forested, with very little protection – only 2% (see Appendix 3). All other zones also had less than
10% of their forest cover under protection. Forests in the Temperate forest-steppes and prairies and
Subtropical broadleaved semi-evergreen forests EZs were totally unprotected. The distribution of the
forests and protected areas is shown in Figure 6c. The extent of the Boreal taiga forest stands out.
In North America there were nine EZs with forest (Table 3), the Boreal taiga forest being most
extensive, as in Russia (Figure 7a). The two zones with the next most extensive cover were Temperate
coniferous and Temperate broadleaf mixed. In terms of percentage forest cover of the EZ, these three
EZs were relatively similar, ranging from 61 to 82%. These three were all forest EZs (as opposed to
grassland or desert), being Temperate broadleaf and mixed forests, Temperate coniferous forests and
Boreal forest/taiga. There was a great contrast between the percentage of EZ forested and the percentage
protection figure for the three high forest cover zones – although the forest cover was high, the
protection was low, under 11% for each zone (Appendix 3). The other six EZs had less than 23% forest
cover. Five of these were however non-forest zones, the exception being Tropical moist broadleaf
forests. These latter had the highest percentage protected (70%), the zone with the next greatest
percentage being Flooded grasslands.
The EZ with the highest figures for both absolute and percentage forest cover in Europe was
Boreal and mesophytic coniferous forests (62%) (Figure 8a and b). Only 6% of these were under
protection (Appendix 3). The Temperate broadleaf and mixed forests EZ was the next most extensive
(Figure 8a). This, the Mires and heaths and the Wetland forests all had 27-33% forest cover, and each
had between 10-15% under protection. The zone with the highest percentage of forest under protection
was Polar and alpine deserts and tundra at 28%, however only 15% of the EZ was forested, and
protected forest was therefore not extensive (Appendix 3).
The chart showing forest extent in each EZ in Australasia demonstrated the most gentle of J-
shaped curves of all the Regions (Figure 9a). This would indicate that there is a more even distribution
of forest area among all the EZs, although five EZs had significantly more forest than the other 13. The
minimal forests in the Polar desert zone (2%) were all under protection. Forests in the Boreal rain forest,
Boreal wet forest and Polar rain tundra were all over 50% protected, but the forests were not extensive
(Figure 9a and b). No EZ had more than 36% forest, but there were eight zones with between 21 and
36%. Six of the zones were less than 5% forested.
Of the twelve forested EZs in Africa none were truly temperate, except perhaps the Mediterranean
zone, which had a small amount of forest (4%, Figure 10a,b). The two EZs with by far the most
extensive forests were Lowland tropical moist with long dry season and Lowland subtropical dry
(Figure 10a). Both of these EZs however were not highly forested at < 3% (Figure 10b). The highest
percentage of forest was in the Lowland tropical very moist category (80%), but these are not extensive.
The next highest percentage forest cover figures are less than half of that at about 33% for Lowland
tropical moist with short dry season and Lowland tropical wet respectively. The total extent of forest
in all montane EZs was very similar (Figure 10a), and all of these were protected to a certain extent. The
highest percentage of forest under protection was in the Alpine EZ (74%), all other EZs having less
than 26% of their forests under protection. All the forested EZs in the study had at least some forest
under protection (Figure 10a).
The Caribbean had five forested EZs, Mesoamerica seven and South America eight, and a Mangrove
zone was classified in an undifferentiated manner for the whole of Latin America (Figure 11a, 11b). In
total forests covered 47% of the Mangrove EZ, although their total land coverage was quite small. By
far the most extensive forests in Latin America were the South American Tropical moist broadleaf
forests (Figure 11a). These cover 70% of their large EZ, and are more than 13 times as extensive as the
next most extensive, the forests of the South American Grasslands, savannas and shrubs EZ (Figure
11a). The most forested EZ of the Caribbean was the Flooded grasslands EZ, but the EZ covers a very

small area, and the least of these were least well protected of all EZs. Forests in the other EZs were
between 12 and 21% protected (Appendix 3). Higher percentage forest cover and percentage protection
occurred in Mesoamerica and South America. Figure 11c shows the distribution of forest and protected
areas in Mesoamerica. The most forested EZ in Mesoamerica was the Montane grasslands at 66%,
which were also well represented in the protected areas systems at 50%. These covered the smallest
amount of land of any EZ in Latin America (Figure 11a). The EZ which had the greatest percentage of
its forest protected in Mesoamerica was Flooded grasslands zone at 61%. Apart from these two
grassland classes, the forests in other Mesoamerican classes had relatively little protection, the next
highest being Tropical moist broadleaf forests at 25%, and the rest less than 12% (Appendix 3,
Figure 11c).
In global terms the EZ with highest percentage of its forests protected globally was the Polar desert
of Australasia, but the total area was very small (25 km2 protected), as this is not an EZ which generally
supports forest growth (the percentage of this zone that was forested was only 1.15%). The EZ with
the least of its forests protected was the Subtropical broadleaved semi-evergreen forests zone of the
Middle East (27 km2 protected). In terms of absolute area, the greatest amount of protected forest was
found in the Tropical moist broadleaf forests of South America (712 000 km2 protected), the next
largest in the Boreal forest/taiga zone of North America (348 863 km2). The Region with the highest
percentage of its forests protected was Insular SE Asia (16.61%), with Australasia a close second
(15.02%), and the Region with the poorest protection was Russia (1.84%) (Appendix 3).

To draw some conclusions from this study for application in global conservation, our aim is to
highlight those forest types which (a) are rarer and (b) are not currently well protected. Because we are
using broad EZs as surrogates for major forest types, it will be clear that the forest types with lowest
cover will be those that are found in EZs that are not generally particularly suited to forest growth. To
get around this problem, which would automatically eliminate most of the forests in the forest EZs
from the possibility of being highlighted for conservation, we divided the EZs for each Region into
“forest EZs” and “non-forest EZs”, and looked at them separately.
IUCN recently (1993) suggested target for conservation of having 10% of each world biome under
protection. Carrying this further, each ecosystem should have an ecologically sustainable extent
preserved, and an adequate number of these units preserved so that if one is destroyed there are
others that can maintain all components of the system for posterity. This reasoning is based on that
used for assigning threat ranks to species by CITES, IUCN and the Nature Conservancy.
Overall, as shown in Tables 4 and 5, there were very few zones, either forest or non-forest, in which
the forests were at least minimally protected (for this paper we are using the surrogate of > 10% of the
forest in any EZ protected as being “at least minimally”, as opposed to “negligibly” which refers to
forests more than zero percent but less than 10% protected). Insular SE Asia was the absolute exception
having no EZ listed in either table. Continental SE Asia showed very low protection for Montane dry
and Lowland arid/desertic, but the others were at least minimally protected. Because of the different
zoning systems being used for the different regions of the world it may be that the SE Asian system is
somehow more forgiving when it comes to indicating forests with negligible protection, but this seems
In the Far East the forests in eight of 13 forest EZs were negligibly protected, and eight of the 10
non-forest zones. Clearly the forests in the Far East are in need of much more protection. In the Middle
East all of the EZs, both non-forest and forest, were negligibly protected. For a number of reasons
protection in this region has perhaps not been pursued, not least of all the political unrest. Some
former Soviet Union states are in this Region, and these have not in the past emphasized forest

Table 4. Forest ecological zones that have < 10% of the forest cover protected
Region Ecological zone

Insular SE Asia None Russia (cont’d) Temperate broadleaved forests

Boreal open woodlands
Continental Montane dry River valley landscapes
SE Asia Temperate forest-steppes and prairies
Subtropical broadleaved semi-evergreen
Far East Boreal open woodlands forests
Boreal taiga forests
Temperate broadleaf forests North America Temperate broadleaf and mixed forests
Temperate forest-steppes and prairies Temperate coniferous forests
Subtropical broadleaved-coniferous Boreal forest / taiga
evergreen forests
Subtropical coniferous-broadleaved semi Europe Riverine floodplain and lacustrine
evergreen forests vegetation
Tropical semi-evergreen and evergreen Parklands
forests Mediterranean type forest and scrub
River valley landscapes Boreal and mesophytic coniferous forests

Middle East Temperate forest-steppes and prairies Africa Montane tropical dry
Subtropical broadleaved-coniferous Premontane tropical moist
evergreen forests Lowland tropical moist with short dry
Subtropical coniferous open woodlands season
Subtropical broadleaved semi evergreen Lowland tropical very moist
forests Lowland tropical wet
Mediterranean hardleaved evergreen
forests, open woodlands and shrubs Latin America Mesoamerican tropical dry broadleaf forest
River valley landscapes Mesoamerican tropical and subtropical
coniferous forest
Russia Subpolar forest-tundras and open
woodlands Australia Tropical very dry forest
Boreal taiga forests Subtropical moist forest
Temperate mixed deciduous-coniferous

conservation. Russia itself has at least minimal forest protection in only one of its EZs, Subtropical
Only one of North America’s forest EZs was at least minimally conserved, Tropical moist broadleaf
forests (see Appendix 3), and three of its five non-forest EZs had negligible protection. Four of
Europe’s six forest EZs had negligible protection, and one of its four non-forest zones (Steppes and
other dry grasslands). Six of the eight forest EZs in Australasia were at least minimally protected, and
six of the ten non-forest zones. Five of Africa’s eight forest EZs were negligibly protected, the three at
least minimally ones being Montane moist, Premontane dry and Lowland moist with long dry season.
Two of its four non-forest EZs were negligibly protected. In Latin America two of the ten forest EZs
were negligibly protected, and four of the 11 non-forest zones. The two forest EZs were in Mesoamerica,
Mesoamerican tropical dry broadleaf forest and Mesoamerican tropical and subtropical coniferous
According to this study then all Regions still need more effort to conserve the forest cover properly
with Insular SE Asia appearing to have the best forest protection of its forests. This is however
subject to an examination of the categories of the protected areas that the forests are in this Region –

Table 5. Non-forest ecological zones that have < 10% of the forest cover protected
Region Non-forest ecological zone

Insular SE Asia None North America Temperate grassland/savanna/shrub

Mediterranean scrub and savanna
Continental Lowland arid / desertic Xeric shrublands
SE Asia
Europe Steppes and other dry grasslands
Far East Temperate steppes
Temperate semi-deserts Australasia Tropical thorn steppe
Subtropical deserts Tropical desert bush
Subtropical short-grass steppes Subtropical desert bush
Tropical semi-deserts Subtropical desert
Tropical steppes Warm temperate desert bush
Bog and marsh landscapes Cool temperate desert bush
Mountain glacier landscapes
Africa Mediterranean
Middle East Temperate steppes Lowland very dry/sub arid
Temperate semi-deserts
Subtropical semi-deserts Latin America Caribbean flooded grasslands
Subtropical deserts Mesoamerican grasslands, savannas and
Subtropical short-grass steppes shrublands
Mountain glacier landscapes South American deserts and xeric
Russia Bog and marsh landscapes South American restingas
Arctic tundra
Boreal maritime meadows
Temperate steppes
Temperate semi-deserts

they may be in forest reserves where logging concessions are granted, and that would change the
picture dramatically. This study indicates which EZs should be the focus of forest protection efforts
in each Region. In many cases this may need international cooperation as the EZs cover more than one
country. This analysis has also highlighted problem areas in the form of information gaps that exist,
particularly in the interpreting of different EZ classifications. Additionally, what do these classifications
tell the reader about the nature of the forests themselves? If too coarse a classification is used then
there are difficulties interpreting the importance of the forests for conservation (e.g., forests in the
Flooded grasslands EZ of the Latin American classification). On a global scale it is difficult to use a
more detailed classification because of the number of classes involved, but there is the possibility of
examining each Region separately to highlight the problem areas and just using those then for the final
global synthesis.

Bohn, U. & Katenina, G.D. 1994. Map of natural association with The World Wildlife Fund.
vegetation. Scale 1:2 500 000. Komarov Bo- The World Bank, Washington, DC.
tanical Institute, St Petersburg. ESRI. 1993. Digital Chart of the World for ARC/
Dinerstein, E. et al. 1995. A conservation assess- INFO. ESRI, Redlands, USA.
ment of the terrestrial ecoregions of Latin Holdridge, L.R. 1967. Life zone ecology. Tropical
America and the Caribbean. Published in Science Center, San Jose, Costa Rica. 206 pp.

Lavenu, F., Bellan, M.F. & Meste, C. 1988. Carte ence Foundation, by Environmental Scienc-
ecoflorestique de l’Amérique du Sud. FAO, es Division, Oak Ridge National Laboratory.
Rome. Oak Ridge, Tennessee, USA.
Lysenko, I.G. et al. 1995. Feasibility study : GAP Päivinen, R et al. 1996. Proceedings of the work-
analysis of forest protected areas in Europe. shop on remote sensing for the Global For-
WCMC, Cambridge, U.K., unpublished re- est Resource Assessment (FRA 2000-Remote
port. Sensing). Washington DC: US Department
MacKinnon, J. (Unpublished) 1996. Review of the of Agriculture, Forest Service, Forest Inven-
Indomalayan Protected Areas Asia Bureau tory, Economics and Recreation Research.
for Conservation, UK. Ricketts, T. et al. 1996. A conservation assess-
Milanova, E.V. & Kushlin, A.V. (Eds) 1993. World ment of terrestrial ecoregions of North Amer-
map of present-day landscapes. An explan- ica. World Wildlife Fund, Conservation Sci-
atory note. Prepared by Moscow State Uni- ence Program, Washington D.C., USA.
versity and the United Nations Environment Sharma, M.K. 1986. Eco-floristic zone and vege-
Programme. tation maps of tropical continental Asia.
Murray, M.G. et al. 1995. Biodiversity conserva- FAO, Rome.
tion in the tropics: Gaps in habitat protec- Sharma, M.K. 1988. Eco-floristic zone map of Af-
tion and funding priorities. WCMC, Cam- rica. FAO, Rome.
bridge, U.K. Unpublished report. WCMC. 1996. In: WWF. 1996. The WWF world
Olson, J.S. & Watts, J.A. 1982. Major world eco- forest map. WWF-International (poster).
system complexes. Prepared for the Carbon White, F. (1983. The vegetation of Africa.
Dioxide Research Division of the United UNESCO, Paris.
States Department of Energy and the Eco- WWF-International. 1996. The WWF world forest
system Studies Program of the National Sci- map. WWF-International (poster).


Australia Carnahan, J.A. and AUSLIG (1989) Cameroon Letouzey, R./Institut de la Carte Inter-
Australia Present Vegetation – 1:5 million; nationale de la Végétation (1985) Carte Phy-
based on standard false colour transparen- togéographique du Cameroun; 1:500 000.
cies of 1980-1985 LANDSAT imagery at Canada Manitoba Resource Centre, Manitoba
1:1 million scale. Department of Natural Resources;
Bangladesh World Bank (1981) Bangladesh-Gen- 1:2 000 000, based on 1988-1991 NOAA
eral Vegetation; 1:500 000, derived from 1977 AVHRR imagery. Provided by the State of
Landsat imagery. the Environment Directorate, Environment
Belize Land Information Centre, Ministry of Nat- Canada.
ural Resources, Belmopan provided digital Central African Republic Boulvert, Y./ORSTOM
data on broad forest categories; derived from (1985) Carte Phytogéographique de la
system maps of the Land Resource Assess- République Centrafricaine; 1:1 000 000.
ments compiled by Natural Resources Insti- Chile Conservation International/Ecotrust (1992)
tute, UK; 1:500 000. Mangroves: Zisman, S. Original Extent of Coastal Temperate Rain
(1992) Mangroves in Belize: their charac- Forests for Southern Chile; 1:10 000 000.
teristics, use and conservation. China MacKinnon, J., Cheung, C., Mengsha, Ju
Benin FAO (1979) Ecological Map of the Vege- Xiang and Melville, D. (1992) Biodiversity
tation Cover of Benin; 1:500 000, prepared Review of China; map based on mid 1970s
by the Pilot Project on Tropical Forest Cover Landsat imagery.
Monitoring from 1973-1976 Landsat images. Colombia Ministerio de Hacienda and Instituto
Bolivia Centro de Datos para la Conservación- Geográfico “Agustin Codazzi” (1984) Repú-
Bolivia (1992) Bolivia-Bosques Húmedos blica de Colombia: Mapa de Bosques;
Densos; based on Brockmann (1978) Mapa 1:500 000.
de Cobertura y Uso Actual de la Tierra, 1973- Congo Hand drawn by P. Hecketsweiler; based
1976 Landsat images. on data gathered from 1989 and 1990 field-
Botswana Soil Mapping and Advisory Services work and from the literature.
Project/Government of Botswana (1991) Veg- Costa Rica Ministerio de Recursos Naturales,
etation Map of the Republic of Botswana; Energía y Minas, Dirección General Forestal
1:2 000 000. (1988) Mapa de Cobertura Boscosa de Cos-
Brazil IBAMA/IBGE (1988 reprinted and updat- ta Rica (+60% densidad); 1:200 000; includ-
ed in 1993) Vegetação do Brasil; 1:5 000 000, ing additional data from Mapa Ecológico -
based on the RADAMBRAZIL, a 1975-83 República de Costa Rica based on Hold-
radar and aerial photography survey. ridge Life Zones.
Brunei Darussalam Anderson and Marsden/ Cuba Academia de Ciencias de Cuba (1989) Nuevo
Brunei Forest Department (1988) Brunei For- Atlas Nacional de Cuba: X Flora y Veg-
est Resources and Strategic Planning Study. etación: 1 Vegetación Actual; 1:1 000 000.
Burkina Faso Ministère de L’environnement et Dominican Republic Based on map from Schu-
du tourisme service de l’aménagement fores- bert, A. (1993) Conservation of biological
tier (nd) Burkina Faso – Formations diversity in the Dominican Republic, Oryx
Végétales; 1:1 000 000. 27(2); based on 1984 aerial photographs.
Cambodia Legris, P. and Blasco, F. (1971) Carte Ecuador Centro de Levantamientos Integrados de
Internationale du Tapis Végétal et des Con- Recursos Naturales por Sensores Remotos/
ditions Ecologiques, Cambodge; Dirección Nacional Forestal (1991) República
1:1 000 000. del Ecuador – Mapa Forestal; 1:1 000 000.

El Salvador Ministerio de Agricultura y Ganad- tions by Satellite); derived from 1 km NOAA/
ería (1981) Mapa de Vegetación Arbórea de AVHRR imagery.
el Salvador; 1:200 000. Haiti Ehrlich, M. et al. (1985) Haiti: Country En-
Equatorial Guinea Servicio Geográfico del Ejér- vironmental Profile, a field study.
cito (1960) (no title); 1:100 000. Honduras COHDEFOR (1992) Mapa de Cobertu-
Europe, North Africa and the Confederation of ra Forestal y Deforestación (1965-1990);
Independent States Based on European 1:500 000 scale; in addition COHDEFOR (no
Space Agency (1992) Remote Sensing For- date) Mapa de Recursos Costeros;
est Map of Europe; 1: 6 000 000, derived from 1:1 000 000 scale.
1989-1992 NOAA-AVHRR data and 1987- India Forest Survey of India (1986) National For-
1990 Landsat-MSS imagery with additional est Vegetation Map (1986); 1:1 000 000 scale;
floristic information provided by the Stock- in addition Atlas of Forest Resources of In-
holm Environment Institute. dia (Das Gupta, 1976) was used to delimit
Fiji Ministry of Forests, Fiji, 1985 survey; forest types.
1:500 000. Indonesia The Regional Physical Planning Pro-
French Guiana Atlas des Départements D’Outre gramme for Transmigration (RePPProT) and
Mer, Centre d’Etudes de Géographie Tropi- Ministry of Transmigration and the National
cale – Office de la Recherche Scientifique et Centre for Coordination of Surveys and Map-
Echnique Outre-Mer (1979) Végétation – La ping (BAKOSURTANAL) began systemati-
Guyane: Planche 12; 1:1 000 000. cally mapping land resources in 1984 – work
Gabon Institut Géographique National-France was completed in 1990; 1:250 000, based on
(Paris) /Institut National de Cartographie (Li- existing reports, air photography, Landsat,
breville) (1987) Gabon; 1:1 000 000. SPOT and radar data.
Gambia US Geological Survey, National Mapping Jamaica Muchoney, D., Iremonger, S. and Wright,
Division, EROS Data Center (1985) Range R., eds/Conservation Data Centre-Jamaica
and Forest Resources of Senegal; compiled a land cover map based on 1988-89
1:1 000 000. Landsat TM imagery, supplemented by soils,
Guatemala Plan de Acción Forestal de Guatema- geology and elevation data.
la (nd) Cubierta Forestal de la República Japan Generalised from: Environment Agency
de Guatemala – Plan de Acción Forestal de (1988) Actual Vegetation Map; 1:3 000 000,
Guatemala; 1:500 000, based on an earlier surveyed in 1979 and 1983-1986, based on
map Mapa Preliminar de la Cubierta Fore- actual vegetation at 1:50 000.
stal de Guatemala at a scale of 1:250 000 Kenya Kenya Rangeland Ecological Monitoring
(1992). Unit, Ministry of Environment and Natural
Guinea CTFT/BDPA-SCET AGRI (1989) Potenti- Resources (1983) Land Use Map of Kenya;
alités et Possibilités de Relance de l’Activité 1:1 000 000, based on 1972-1980 remote
Forestière; Synthèse Régionale et Nation- sensed data.
ale; 1:700 000 derived from 1979-80 aerial pho- Lao People’s Democratic Republic Lao PDR For-
tography and Landsat MSS 1984-1986 estry Department (1987); 1:1 000 000, based
imagery. on 1980-1 aerial photography, and a
Guinea-Bissau Hand drawn by Scott Jones 1990 1:1 000 000 land use map prepared by the
based on his personal experience of the re- Mekong Secretariat from 1972-3 Landsat im-
gion; 1:1 000 000, based on an earlier 1:500 000 agery.
Instituto Geográfico Nacional (1981) land- Madagascar Faramalala Miadana Harisoa – I.C.I.V
use chart Guiné Bissau. (nd) Carte des Formations Végétales de
Guyana EU-Joint Research Centre TREES project Madagascar; 1:1 000 000; and from general-
(Tropical Ecosystem Environment Observa- ised 1985 satellite imagery which accompa-

nies Green and Sussman (1990) Deforesta- République Rwandaise, Carte Administra-
tion history of the eastern rain forests of tive et Routière; 1:250 000.
Madagascar from satellite images, Science Sao Tomé and Principe Based on a photocopy of
248: 212-215. a map prepared by the Bureau pour la Dével-
Malaysia Forest Department in Kuala Lumpur oppement de la Production Agricole (1985).
(1989) The Forest Area, 1:1 000 000; Sabah Senegal US Geological Survey, EROS Data Cent-
Forestry Department (1989) Sabah Malay- er (1985) Range and Forest Resources of
sia, Natural and Plantation Forests, Senegal; 1:1 000 000.
1:270 000; and Sarawak Forest Department Solomon Islands Based on a hand-coloured map
(1979) Forest Distribution and Land Use prepared by the Forestry Division (late
Map, 1:1 000 000 scale. 1980s); 1:1 000 000. Additional information
Mexico D. Evans and S.I. McIntosh produced a was added by G. Chaplin, based on his per-
digital map from composite AVHRR images sonal experience of the region.
for 1990 and 1991, including supporting in- South Africa Digital data compiled by Forestek,
formation, vegetation maps (1: 1 000 000), CSIR (1993); based on Landsat TM data.
Landsat TM images and aerial photographs. Sri Lanka Survey Department of Sri Lanka (1988)
Myanmar Britto, N.B. (1987) National forest man- 1983 Sri Lanka: Chena Cultivation in the
agement and inventory of Burma. Report on Dry Zone and Dense Natural Forest;
cartographic consultancy – Forest Depart- l:500 000.
ment of Burma/FAO; 1:1 000 000 Landsat Surinam National Planning Office of Surinam,
MSS and RBV 1979-81 imagery updated us- Regional Development and Physical Planning
ing aerial photography. Department (1988) Surinam Planatlas Plate
New Zealand Wards, I. (ed) (1976) New Zealand B7 Vegetation.
Atlas: Contemporary Forest Cover; Tanzania Rodgers et al., (1985) Forest Cover in
1:3 200 000 Tanzania; 1:2 000 000 based on 1973-79 sat-
Nicaragua Instituto Nicaraguense de Recursos ellite imagery.
Naturales y del Ambiente, Dirección de Ad- Thailand Royal Forest Department (1985) Forest
ministración de Bosques Nacionales (1991) Types Map; 1:1 000 000 based on 1972-7 aer-
Estado Actual de la Vegetación Forestal de ial photographs and 1985 Landsat imagery.
Nicaragua; 1: 1 000 000, based on a 1983 land- Trinidad only (no data were found for Tobago)
use map. Government of Trinidad and Tobago/Insti-
Panama Atlas Nacional de Panamá: 8.1 Veg- tutional Consultants (International) Ltd./
etación Actual (1980); 1:1 000 000. CIDA (1980) Inventory of the Indigenous
Papua New Guinea Paijmans, K. (1975) Vegeta- Forest of Trinidad – Forest Resource Inven-
tion of Papua New Guinea; 1:1 000 000. tory and Management Section.
Paraguay Ministerio de Agricultura y Ganadería Uganda Langdale-Brown, I., Osmaston, H.A. and
and Misión Forestal Alemana/GTZ (1986) J.G. Wilson (1972) Uganda Vegetation;
República del Paraguay, Región Oriental – 1:500 000.
Uso Actual de la Tierra; 1:500 000, based on USA USDA Forest Service (1993) Forest Type
1984-85 Landsat imagery. Groups of the United States; 1:7 500 000,
Peru FAO project GCP/RLA/081/JPN (nd) Mapa based on satellite imagery.
Integrado de Ecosistemas Forestales del USSR (former) Isaev, A.C. (1990) Forests of the
Peru Colombia y Venezuela; 1:5 000 000. USSR, State Committee of USSR for Forests;
Philippines Forest Management Bureau/GTZ 1:2 500 000.
(1988) 1988 Forest Cover Map of the Phil- Venezuela Huber O. and Alarcón, C. (1988) Mapa
ippine Islands; 1:2 000 000. de Vegetación de Venezuela; 1:2 000 000.
Rwanda Service de Cartographie, Kigali (nd) Viet Nam Cac Loai Thuc Vât bi de Doa Dien

Hinh va Môt Vung Tâp Trung (1987); Others World Forestry Atlas (1971) comprising
1:1 000 000, forest inventory data. country maps at a number of different scales
West Africa (Côte d’Ivoire, Ghana, Liberia, Ni- and dates: Argentina (1962; 1:8 000 000), east-
geria, Sierra Leone, Togo) Paivinen, R. and ern Iran (1965; 1:8 000 000), Afghanistan
Witt, R. (1989) The Methodology Develop- (1971; 1:8 000 000), Korea (1956; 1:5 000 000);
ment Project for Tropical Forest Cover As- Mongolia (1971; 1:2 000 000); and Angola
sessment in West Africa; 1989-1990 UNEP/ (1960; 1:10 000 000).
GRID data, derived from 1 km resolution Others Asia-Pakistan, Nepal, Bhutan MacKin-
NOAA/AVHRR data. non, J. (in prep) Review of the Protected Ar-
Zaire 1 km resolution NOAA/AVHRR 1988 data eas System of the Indomalayan Realm;
compiled by NASA/Goddard Space Flight 1:1 000 000 based on late 1980s-early 1990s
Center/University of Maryland. AVHRR and Landsat data, and existing re-
Zambia and Mozambique Wild, H. and Grandvaux ports and maps.
Barbosa, L.A. (1967) Flora Zambesiaca –
Vegetation Map of the Flora Zambesiaca
Area; 1:2 500 000.


North America Egypt Germany Guatemala

Canada Equatorial Guinea Greece Guyana
Greenland Eritrea Iceland Haiti
United States of Ethiopia Ireland Honduras
America except Hawaii Gabon Italy Jamaica
The Gambia Latvia Martinique
Russia Ghana Luxembourg Mexico
Russia except Guinea Macedonia Netherland Antilles
Kaliningrad Oblast Guinea-Bissau Moldova Nicaragua
Kenya Netherlands Panama
Far East Lesotho Norway Paraguay
China Liberia Poland Peru
Japan Libya Portugal Puerto Rico
Mongolia Madagascar Romania St. Lucia
Taiwan Malawi Russia (Kaliningrad St. Kitts and Nevis
Mali Oblast only) St. Vincent and the
Middle East Mauritania Slovakia Grenadines
Afghanistan Morocco Slovenia Surinam
Bahrain Mozambique Spain Tobago
Iran Namibia Sweden Trinidad
Iraq Niger Switzerland Uruguay
Israel Nigeria Ukraine Venezuela
Jordan Rwanda United Kingdom
Kazakhstan Senegal Yugoslavia Continental SE Asia
Kuwait Sierra Leone Bangladesh
Kyrgystan Somalia Australasia Bhutan
Lebanon South Africa Australia Cambodia
Oman Sudan New Zealand India
Qatar Swaziland Laos
Saudi Arabia Tanzania Latin America Malaysia (peninsular)
Syria Togo Antigua and Barbuda Myanmar
Tajikistan Tunisia Argentina Nepal
Turkey Uganda The Bahamas Pakistan
Turkmenistan Western Sahara Barbados Sri Lanka
Uzbekistan Zaire Belize Thailand
Yemen Zambia Bolivia Vietnam
Zimbabwe Brazil
Africa Cayman Islands Insular SE Asia
Algeria Europe Chile Indonesia
Angola Albania Colombia Malaysia (Sabah and
Benin Austria Costa Rica Sarawak)
Botswana Belarus Cuba Papua New Guinea
Burkina Belgium Dominica Philippines
Burundi Bosnia-Herzegovina Dominican Republic Brunei Darussalam
Cameroon Bulgaria Ecuador
Central African Republic Croatia El Salvador
Chad Cyprus Falkland Islands
The Congo Czech Republic French Guiana
Cote D’Ivoire Estonia Grenada
Djibouti Finland Guadeloupe

Forest cover Protected forest Protected
(km 2 ) cover (km 2 ) forest cover (%)
European Major Ecological Zones
Riverine, flood-plain and lacustrine vegetation 19 538 1 446 7.40
Coastal and other halophytic vegetation 1 606 168 10.49
Steppes and other dry grasslands 4 568 142 3.13
Mires and heaths 25 619 3 223 12.58
Wetland forests 11 173 1 733 15.52
Parklands (grassland with patches of forest) 4 791 62 1.30
Mediterranean type forests and scrubs 93 526 6 714 7.18
Temperate broadleaf and mixed forests 780 589 86 149 11.04
Boreal and mesophytic coniferous forests 867 719 40 573 4.68
Temperate deserts and sub-deserts 0 NA NA
Polar and alpine deserts and tundra 34 435 9 483 27.54
Excluded forests 50 274 499 0.99
Total 1893843 150 297 7.94

African Major Ecological Zones

Lowland tropical wet 34 795 2 133 6.13
Lowland tropical very moist 916 323 747 23 8.15
Lowland tropical moist with short dry season 941 725 49 555 5.26
Lowland tropical moist with long dry season 70 963 8 852 12.48
Lowland tropical sub-dry 92 939 23 321 25.09
Lowland tropical very dry / sub-arid 8 484 387 4.57
Lowland tropical arid / desertic 0 NA NA
Premontane tropical moist 97 371 7 717 7.93
Premontane tropical dry 13 604 1 852 13.62
Montane tropical moist 72 772 12 044 16.55
Montane tropical dry 6 226 220 3.54
Alpine 9 720 7 213 74.21
Mediterranean 15 963 920 5.77
Excluded forests 958 12 1.24
Total 2 281 951 188 955 8.28

Continental South-East Asia Major Ecological Zones

Lowland tropical wet 0 NA NA
Lowland tropical very moist 332 867 41 724 12.53
Lowland tropical moist with short dry season 0 NA NA
Lowland tropical moist with long dry season 315 246 36 775 11.63
Lowland tropical sub-dry 435 815 58 237 13.36
Lowland tropical very dry / sub-arid 3 276 361 11.04
Lowland tropical arid / desertic 2 899 0 0
Premontane tropical moist 135 389 13 725 10.14


“Excluded forests”, indicated at the end of each regional ecological zone list, was the figure for forest in the
region that was excluded from the analysis because of mismatched between the forest data sets and the ecological
zone data sets caused by scale and resolution differences, as well as differences associated in particular with placing
of inland waters and coasts. The only regions this did not occur in were those in SE Asia and in Australasia, for
reasons explained in the text. NA is used throughout for “not applicable”.

Appendix 3 (continued)
Forest cover Protected forest Protected
(km 2 ) cover (km 2 ) forest cover (%)
Continental South-East Asia Major Ecological Zones (cont’d)
Premontane tropical dry 51 301 6 806 13.27
Montane tropical moist 0 NA NA
Montane tropical dry 279 0 0
Alpine 7 323 2 008 27.42
Mediterranean 0 NA NA
Total 1 285 400 159 639 12.42

Insular South-East Asia Major Ecological Zones

Lowland tropical wet 0 NA NA
Lowland tropical very moist 1 448 781 215 260 14.86
Lowland tropical moist with short dry season 39 604 6 540 16.51
Lowland tropical moist with long dry season 5 472 1 105 20.20
Lowland tropical sub-dry 4 036 491 12.17
Lowland tropical very dry / sub-arid 0 NA NA
Lowland tropical arid / desertic 0 NA NA
Premontane tropical moist 146 125 42 188 28.87
Premontane tropical dry 0 NA NA
Montane tropical moist 48 123 15 165 31.51
Montane tropical dry 0 NA NA
Alpine 2 720 739 27.19
Mediterranean 0 NA NA
Total 1 694 865 281 491 16.61

Latin American Major Ecological Zones

(in three parts, part 1 Mesoamerica)
Tropical moist broadleaf forests 250 815 62 999 25.12
Tropical dry broadleaf forests 229 412 2 532 1.10
Tropical and subtropical coniferous forests 323 236 11 173 3.46
Grasslands, savannas and shrublands 3 721 5 0.15
Flooded grasslands 2 509 1 524 60.74
Montane grasslands 116 58 50.35
Mediterranean scrub 0 NA NA
Deserts and xeric shrublands 91 814 10 734 11.69
Mangroves (all Latin America) 5 149 704 13.68
Total (see end of Caribbean section)

Latin American Major Ecological Zones

(in three parts, part 2, South America)
Tropical moist broadleaf forests 5 670 644 712 026 12.56
Tropical dry broadleaf forests 16 5047 27 197 16.48
Tropical and subtropical coniferous forests NA NA NA
Temperate forests 158 699 59 723 37.63
Grasslands, savannas and shrublands 412 896 40 964 9.92
Flooded grasslands 59 197 6 847 11.57
Montane grasslands 80 307 19 201 23.91
Mediterranean scrub 0 NA NA
Deserts and xeric shrublands 86 201 7 812 9.06
Restingas 907 Neg. Neg.
Mangroves (all Latin America, see results
for Mesoamerica, above)
Total (see end of Caribbean section)

Appendix 3 (continued)
Forest cover Protected forest Protected
(km 2 ) cover (km 2 ) forest cover (%)
Latin American Major Ecological Zones
(in three parts, part 3, the Caribbean)
Tropical moist broadleaf forests 18 096 3 884 21.46
Tropical dry broadleaf forests 15 582 1 946 12.49
Tropical and subtropical coniferous forests 6 121 927 15.16
Grasslands, savannas and shrublands 0 NA NA
Flooded grasslands 2 253 145 6.46
Montane grasslands 0 NA NA
Mediterranean scrub 0 NA NA
Deserts and xeric shrublands 741 144 19.46
Mangroves (all Latin America, see results
for Mesoamerica, above)
Excluded forests (for all Latin America) 42 963 6 538 15.22
Total (for all Latin America) 1 161 713 7 583 476 12.88

Subtropical deserts 0 NA NA
Subtropical semi-deserts 4 635 2 883 62.21
Subtropical broadleaved semi-evergreen forests 120 0 0
Mediterranean hardleaved evergreen forests, open 0 NA NA
woodlands and shrubs
Temperate deserts 0 NA NA
Temperate semi-deserts 6 869 42 0.61
Temperate steppes 64 997 2 783 4.28
Temperate forest-steppes and prairies 318 850 13 646 4.28
Temperate broadleaved forests 106 717 2 499 2.34
Temperate mixed deciduous-coniferous forests 493 491 15 552 3.15
Bog and marsh landscapes 174 665 2 285 1.31
River valley landscapes 327 200 1 573 0.48
Boreal taiga forest 5 618 677 77 257 1.38
Boreal open woodlands 61 538 969 1.58
Boreal maritime meadows 592 557 23 422 3.95
Subpolar forest-tundras and open woodlands 246 645 1 061 0.43
Subpolar tundras 55 989 1 581 2.82
Arctic tundras 97 0 0
Mountain glacier landscapes 0 NA NA
Solonchak landscapes 0 NA NA
Polar ice and stony deserts 0 NA NA
Excluded forests 59 440 4 225 7.11
Total 8 132 493 149 782 1.84

Middle East
Tropical open woodlands, shrubs and savannas 0 NA NA
Tropical semi-deserts 0 NA NA
Tropical deserts 0 NA NA
Subtropical tall-grass steppes (prairies) 0 NA NA
Subtropical short-grass steppes 16 968 36 0.21
Subtropical open woodlands and shrubs 8 0 0
Subtropical deserts 26 219 2 267 8.65
Subtropical semi-deserts 14 456 0 0
Subtropical coniferous open woodlands 268 0 0

Appendix 3 (continued)
Forest cover Protected forest Protected
(km 2 ) cover (km 2 ) forest cover (%)
Middle East (cont’d)
Subtropical broadleaved semi-evergreen forests 14 456 27 0.19
Subtropical coniferous-broadleaved semi-evergreen forests 0 NA NA
Subtropical broadleaved-coniferous evergreen forests 33 170 204 0.62
Mediterranean hardleaved evergreen forests, open
woodlands and shrubs 22 467 1 223 5.44
Temperate deserts 17 302 511 2.96
Temperate semi-deserts 4 046 278 6.88
Temperate steppes 15 015 1 211 8.07
Temperate forest-steppes and prairies 9 517 557 5.85
Bog and marsh landscapes 0 NA NA
River valley landscapes 471 32 6.85
Mountain glacier landscapes 3 0 0
Solonchak landscapes 0 NA NA
Excluded forests 1 458 35 2.41
Total 161389 6 384 3.96

Far East
Subequatorial semi-evergreen forests 3 760 42 1.13
Tropical semi-evergreen and evergreen forests 65 568 3 273 4.99
Tropical open woodlands, shrubs and savannas 2 377 238 10.02
Tropical steppes 17 832 219 1.23
Tropical semi-deserts 76 857 3 452 4.49
Subtropical short-grass steppes 104 770 0 0
Subtropical deserts 4 780 51 1.09
Subtropical semi-deserts 13 174 8 137 61.77
Subtropical coniferous open woodlands 23 716 3 296 13.90
Subtropical broadleaved semi-evergreen forests 16 246 399 2.46
Subtropical coniferous-broadleaved semi-evergreen forests 104 770 5 620 5.36
Subtropical broadleaved-coniferous evergreen forests 50 485 691 1.37
Temperate deserts 248 491 9 313 3.75
Temperate semi-deserts 42 051 2 613 6.22
Temperate steppes 120 624 926 0.77
Temperate forest-steppes and prairies 173 735 4 847 2.79
Temperate broadleaved forests 66 812 629 0.94
Temperate mixed deciduous-coniferous forests 137 531 21 504 15.64
Bog and marsh landscapes 8 802 173 1.97
River valley landscapes 64 456 1 237 1.92
Boreal taiga forest 251 324 11 522 4.58
Boreal open woodlands 214 397 434 2.02
Boreal maritime meadows 1 118 0 NA
Mountain glacier landscapes 449 0 NA
Solonchak landscapes 0 NA NA
Excluded forests 18 944 3 432 18.12
Total 1 539 615 82 058 5.33

North America
Tropical moist broadleaf forests 262 184 70.36
Temperate broadleaf and mixed forests 1 720 610 100 430 5.84
Temperate coniferous forests 1 739 905 178 140 10.24
Temperate grassland/savanna/shrub 233 117 5 595 2.51

Appendix 3 (continued)
Forest cover Protected forest Protected
(km 2 ) cover (km 2 ) forest cover (%)
North America (cont’d)
Flooded grasslands 2 454 1 212 49.38
Mediterranean scrub and savanna 24 495 1 397 5.70
Xeric shrublands/deserts 146407 7 203 4.92
Boreal forest/taiga 4 132 477 348 863 8.44
Tundra 327 590 53 131 16.22
Excluded forests 45 989 3 052 6.64
Total 8 363 311 699 211 8.36

Tropical dry forest 84 162 8 179 9.72
Tropical very dry forest 19 772 59 0.30
Tropical thorn steppe 1 477 0 NA
Subtropical wet forest 950 30 7.57
Subtropical moist forest 49 979 5 719 11.44
Subtropical dry forest 69 709 6 305 9.06
Subtropical thorn steppe 8 168 18 0.23
Subtropical desert bush 1 807 0 NA
Subtropical desert 0 NA NA
Warm temperate moist forest 26 463 3 578 13.52
Warm temperate dry forest 62 191 8 474 13.63
Warm temperate thorn steppe 3 158 111 3.54
Warm temperate desert bush 0 NA NA
Cool temperate wet forest 18 965 6 023 31.76
Cool temperate moist forest 107 586 21173 19.68
Cool temperate steppe 606 0 0
Boreal rain forest 4 367 2 211 50.64
Boreal wet forest 7 686 3 893 50.66
Polar rain tundra 10 395 5 866 56.43
Polar desert 25 25 99.76
Total 477 476 71 724 15.02

Figure 1. Average percentage of forest under protection (calculation included only forested
ecological zones) for each Region








Middle East

Insular South
South East

Latin America

Far East

North America


East Asia

Figure 2a. Areas of forest cover (solid bar) and forest under protection (open bar) in each
ecological zone in insular SE Asia (see Appendix 2 for countries included in this Region). Zones
without forest have not been included in the chart.

Insular South East Asia




Area km2





Low land Premontane Montane Low land Low land Low land Alpine
very moist moist moist moist w ith moist w ith sub-dry
short dry long dry
season season

Figure 2b. Percentage cover of forest in each ecological zone in Insular SE Asia. Ecological
zones without forest have been omitted. The percentage of forest cover which was not included in
any ecological zone by the analysis was nil (see text).

Insular South East Asia


Low land Low land Low land Low land Premontane Montane Alpine
very moist moist w ith moist w ith sub-dry moist moist
short dry long dry
season season

Figure 3a. Areas of forest (solid bar) and forest under protection (open bar) in each ecological
zone in Continental SE Asia (see Appendix 2 for countries included in this Region). Zones without
forest have not been included in the chart.

Continental South East Asia




Area (km 2 )







Premontane dry
Lowland moist


Montane dry
Lowland arid /
Lowland sub-dry

dry / sub arid

Lowland very

Lowland very
with long dry



Figure 3b. Percentage cover of forest in each ecological zone in Continental SE Asia. Ecological
zones without forest have been omitted. The percentage of forest cover which was not included in
any ecological zone by the analysis was nil (see text).

Continental South East Asia










Premontane dry

Montane dry
Lowland moist

Lowland arid /
dry / sub arid
Lowland sub-dry
Lowland very

Lowland very
with long dry


for countries included in this Region). Zones without forest have not been included in the chart.
Figure 4a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in the Far East (see Appendix 2
Area (km 2)





Boreal taiga forest a

Temperate deserts

Temperate forest-steppes and prairies

Temperate mixed deciduous-coniferous forests

Temperate steppes e

Subtropical coniferous broadleaved semi-evergreen forests

Tropical semi-deserts -

Temperate broadleaved forests

Tropical semi-evergreen and evergreen forests d

River valley landscapes

Subtropical broadleaved-coniferous evergreen forests -

Far East
Temperate semi-deserts
Subtropical coniferous open woodlands

Boreal open woodlands

Tropical steppes l

Subtropical broadleaved semi-evergreen forests

Subtropical semi-deserts

Bog and marsh landscapes

Subtropical deserts l

Subtropical short-grass steppes

Subequatorial semi-evergreen forests n

Tropical woodlands, shrubs and savannas

Boreal maritime meadows

Mountain glacier landscapes

percentage of forest cover which was not included in any ecological zone by the analysis was 1.23% (see text).
Figure 4b. Percentage cover of forest in each ecological zone in the Far East. Ecological zones without forest have been omitted. The







Boreal taiga forest

Temperate mixed deciduous-coniferous forests

Temperate broadleaved forests

Temperate forest-steppes and prairies

Temperate steppes

Temperate semi-deserts

Temperate deserts

Subtropical broadleaved-coniferous evergreen forests

Subtropical coniferous broadleaved semi-evergreen forests

Subtropical broadleaved semi-evergreen forests

Far East
Subtropical coniferous open woodlands

Subtropical semi-deserts

Subtropical deserts

Subtropical short-grass steppes

Tropical semi-deserts

Tropical steppes

Tropical open woodlands, shrubs and savannas

Tropical semi-evergreen and evergreen forests

Subequatorial semi-evergreen forests

River valley landscapes

Bog and marsh landscapes

Mountain glacier landscapes

dix 2 for countries included in this Region). Zones without forest have not been included in the chart.
Figure 5a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in the Middle East (see Appen-
Area (km 2 )






Subtropical broadleaved-confierous
evergreen forests

Subtropical deserts

Mediterranean hardleaved evergreen
forests, open woodlands and shrubs

Temperate deserts

Subtropical short-grass steppes

Temperate steppes

Subtropical broadleaved semi-

evergreen forests Middle East

Temperate forest-steppes and prairies

Temperate semi-deserts

River valley landscapes

Subtropical coniferous open


Subtropical semi-deserts

Mountain glacier landscapes r



Subtropical short-grass steppes

Subtropical deserts

Subtropical semi-deserts

Subtropical coniferous open woodlands

Subtropical broadleaved semi-evergreen forests

Subtropical broadleaved-confierous evergreen forests

Mediterranean hardleaved evergreen forests, open woodlands and shrubs

Middle East
any ecological zone by the analysis was 0.9% (see text).

Temperate deserts

Temperate semi-deserts

Temperate steppes

Temperate forest-steppes and prairies

River valley landscapes

Figure 5b. Percentage cover of forest in each ecological zone in the Middle East . Ecological

Mountain glacier landscapes

zones without forest have been omitted. The percentage of forest cover which was not included in

countries included in this Region). Zones without forest have not been included in the chart.
Figure 6a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in Russia (see Appendix 2 for
Area (km 2 )






Boreal taiga forest t

Boreal martime meadows s

Temperate mixed deciduous-coniferous forests

River valley landscapes

Temperate forest-steppes and prairies

Subpolar forest-tundras and open woodlands

Bog and marsh landscapes


Temperate broadleaved forests

Temperate steppes

Boreal open woodlands

Subpolar tundras

Temperate semi-deserts

Subtropical semi-deserts

Subtropical broadleaved semi-evergreen forests

Arctic tundras

age of forest cover which was not included in any ecological zone by the analysis was 0.7% (see text).
Figure 6b. Percentage cover of forest in each ecological zone in Russia. Ecological zones without forest have been omitted. The percent-









Subpolar forest-tundras and
open woodlands

Bog and marsh landscapes

Arctic tundras

Boreal taiga forest

Boreal martime meadowse

Temperate mixed deciduous-

coniferous forests

Boreal open woodlands

Temperate broadleaved forests

Temperate steppes

River valley landscapes

Temperate semi-deserts -

Subtropical semi-deserts

Temperate forest-steppes and


Subtropical broadleaved semi-

evergreen forests

Figure 6c. Forest by ecoregion and protection areas in Russia

Figure 7a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in North America (see Appendix
2 for countries included in this Region). Zones without forest have not been included in the chart.
North America
Area (km 2 )



broadleaf and

Boreal forest /

mixed forests

Xeric shrublands

savanna / shrub

broadleaf forests


Tropical moist

scrub and
grassland /

/ deserts
Figure 7b. Percentage cover of forest in each ecological zone in North America. Ecological zones without forest have been omitted. The
percentage of forest cover which was not included in any ecological zone by the analysis was 0.5% (see text).
North America


broadleaf and

Xeric shrublands /
savanna / shrub

Boreal forest /
mixed forests
broadleaf forests

coniferous forests

Tropical moist

scrub and
grassland /



countries included in this Region). Zones without forest have not been included in the chart.
Figure 8a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in Europe (see Appendix 2 for
Area (km 2)





Boreal and mesophytic coniferous forests

Temperate broadleaf and mixed forests

Mediterranean type forests and scrubs

Polar and alpine deserts and tundra

Mires and heaths

Riverine, flood-plain and lacustrine vegetation

Wetland forests

Parklands (grasslands with patches of forest)

Steppes and other dry grasslands

Coastal and other halophytic vegetation

age of forest cover which was not included in any ecological zone by the analysis was 2.7% (see text).
Figure 8b. Percentage cover of forest in each ecological zone in Europe. Ecological zones without forest have been omitted. The percent-








Riverine, flood-plain and
lacustrine vegetation

Coastal and other

halophytic vegetation

Steppes and other dry


Mires and heaths

Wetland forests

Parklands (grasslands
with patches of forest)

Mediterranean type
forests and scrubs

Temperate broadleaf and

mixed forests

Boreal and mesophytic

coniferous forests

Polar and alpine deserts

and tundra

for countries included in this Region). Zones without forest have not been included in the chart.
Figure 9a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in Australasia (see Appendix 2
Area (km 2 )





Warm temperate dry forest

Polar rain tundra

Tropical thorn steppe

Cool temperate moist forest

Tropical dry forest

Polar desert

Subtropical thorn steppe

Subtropical moist forest Australasia

Subropical dry forest

Boreal rain forest

Boreal wet forest

Subtropical desert bush

Cool temperate steppe

Tropical very dry forest

Warm temperate thorn steppe

Cool temperate wet forest

Warm temperate moist forest

Subtropical wet forest

percentage of forest cover which was not included in any ecological zone by the analysis was nil (see text).
Figure 9b. Percentage cover of forest in each ecological zone in Australasia. Ecological zones without forest have been omitted. The








Tropical very dry forest

Tropical thorn steppe

Subtropical wet forest

Subtropical moist forest

Subropical dry forest

Subtropical thorn steppe

Subtropical desert bush

Warm temperate moist forest

Warm temperate dry forest

Warm temperate thorn steppe

Cool temperate wet forest

Cool temperate moist forest

Cool temperate steppe

Boreal rain forest

Boreal wet forest

Polar rain tundra

Polar desert

Figure 10a. Areas of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in Africa (see Appendix 2 for
countries included in this Region). Zones without forest have not been included in the chart.
Area (km 2 )


Lowland very dry / sub arid

Lowland sub dry
Lowland moist with long dry season

Premontane dry

Lowland wet


Premontane moist

Lowland moist with short dry season

Montane moist

Montane dry

Lowland very moist

Figure 10b. Percentage cover of forest in each ecological zone in Africa. Ecological zones without forest have been omitted. The percent-
age of forest cover which was not included in any ecological zone by the analysis was 0.04% (see text).



Montane dryy

Lowland very dry / sub arid

Lowland sub dry


Premontane moist

Premontane dry

Lowland very moist

Lowland wet
Lowland moist with long dry season

Lowland moist with short dry season

Montane moist

Mesoamerican. CA = Caribbean. Zones without forest have not been included in the chart.
2 for countries included in this Region). The abbreviations in the ecological zone names denote the following: SA = South American, MA =
Figure 11a. Area of forest cover (solid bar) and forest under protection (open bar) in each ecological zone in Latin America (see Appendix
Area (km 2 )

20 000 000,00

30 000 000,00

40 000 000,00

50 000 000,00

60 000 000,00
10 000 000,00
SA Tropical moist broadleaf forests

SA Grasslands, savannas and shrublands

MA Tropical and subtropical coniferous forests

MA Tropical moist broadleaf forests

MA Tropical dry broadleaf forests

SA Tropical dry broadleaf forests

SA Temperate forests

MA Deserts and xeric shrublands

SA Deserts and xeric shrublands

Latin America
SA Montane grasslands

SA Flooded grasslands

CB Tropical moist broadleaf forests

CB Tropical dry broadleaf forests

CB Tropical and subtropical coniferous forests

Undifferentiated mangrove

MA Grasslands, savannas and shrublands

MA Flooded grasslands

CB Flooded grasslands

SA Restingas

CB Deserts and xeric shrublands

MA Montane grasslands

percentage of forest cover which was not included in any ecological zone by the analysis was 2.2% (see text).
abbreviations in the ecological zone names denote the following: SA = South American, MA = Mesoamerican. CA = Caribbean. The
Figure 11b. Percentage cover of forest in each ecological zone in Latin America. Ecological zones without forest have been omitted. The








Undifferentiated mangrove
CB Tropical moist broadleaf forests

CB Tropical dry broadleaf forests

CB Tropical and subtropical

coniferous forests

CB Flooded grasslands

CB Deserts and xeric shrublands

MA Tropical moist broadleaf forests

MA Tropical dry broadleaf forests

MA Tropical and subtropical

coniferous forests

Latin America
MA Grasslands, savannas and

MA Flooded grasslands

MA Montane grasslands

MA Deserts and xeric shrublands

SA Tropical moist broadleaf forests

SA Tropical dry broadleaf forests

SA Temperate forests

SA Grasslands, savannas and


SA Flooded grasslands

SA Montane grasslands

SA Deserts and xeric shrublands

SA Restingas

Figure 11b. Forest by ecoregion and protected areas in Mesoamerica

Towards priorities of biodiversity research in
support of policy and management of tropical
rain forests

Erik M. Lammerts van Bueren 1, Joost F. Duivenvoorden 2

This paper presents a procedure to identify research priorities on biological and ecological aspects
of biodiversity in order to provide crucial information to policy-makers and forest managers for
conservation and wise utilization of tropical rain forests. The procedure is intended to be used for
assessing the relevance and cost-effectiveness of submitted project proposals, and for taking initiatives
in the development and organisation of new research activities by the Tropenbos Foundation in The
Netherlands, and by other research institutions and organisations. The paper distinguishes seven
policy and management objectives concerning the conservation and wise utilization of tropical rain
forests for which information on biodiversity is indispensable. Next, the paper elucidates the set-up of
an information matrix, which shows the biodiversity information needs for each objective, the relevance
of a research proposal for policy and/or management, and the desirability of methodological
improvements. Additional criteria for final priority setting are formulated.
The Tropenbos Foundation is an internationally oriented organization conducting interdisciplinary
research and training programmes in support of biodiversity conservation and wise use of tropical
rain forests. Tropenbos has five permanent research sites in Colombia, Guyana, Cameroon, Côte
d’Ivoire and Indonesia.

Keywords: Biodiversity research, tropical rain forests, conservation, sustainable forest use,
interdisciplinary research programmes.

The Tropenbos Foundation in The Netherlands is an independent, internationally oriented
organization, conducting research and development activities to support conservation of biodiversity
and sustainable forest use. Tropenbos initiates and finances interdisciplinary research and training
programmes which are being carried out at permanent research sites in Colombia, Guyana, Cameroon,
Côte d’Ivoire and Indonesia. To strengthen the coherence between the sites and to enhance the
effectiveness of the Tropenbos programme, research strategies have been developed on themes such
as biological diversity and non-timber forest products. This paper summarizes the research strategy
for tropical rain forest biodiversity.

The Tropenbos Foundation, P.O. Box 232, 6700 AE Wageningen, The Netherlands.
Hugo de Vries Laboratory, University of Amsterdam, Kruislaan 318, 1098 SM Amsterdam, The Netherlands

In view of the dimension of the work required to obtain the necessary knowledge to halt the rapid
degradation of tropical rain forests and their biodiversity, highest priority should be given to research
that fulfils the demands for information coming from those who are responsible for the design and
implementation of policy and management concerning sustainable forest use and biodiversity
conservation. The purpose of this paper is to present the results of the development and application
of a procedure to identify research priorities on biological and ecological aspects of biodiversity in
order to provide crucial information to policy-makers and forest managers for conservation and wise
utilization of tropical rain forests. The procedure is intended to be used for assessing the relevance
and cost-effectiveness of submitted project proposals, and for taking initiatives in the development
and organisation of new research activities by the Tropenbos Foundation and by other research
institutions and organisations. The document on which this paper is based (Lammerts van Bueren and
Duivenvoorden 1996) delivers two products: a procedure for setting research priorities and a partial
application of this procedure. The initiative for writing this paper was triggered by the observation
that it is very difficult to assess research proposals and to stimulate additional research without a well-
structured approach and clear reference levels.


Biodiversity is defined as the variability among living organisms from all sources, inter alia,
terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part;
this includes diversity within species, between species and of ecosystems. For practical uses this
concept of biodiversity is subdivided into three dimensions: the spatial dimension, the temporal
dimension, and a functional dimension representing the functional relationships within and between
species and ecosystems.


A procedure is presented as a tool by which Tropenbos seeks to prioritize policy and management
research on tropical rain forest biodiversity. This procedure starts with the identification of the most
common and compelling policy and management objectives concerning the conservation and wise
utilization of tropical rain forests, for which information on biodiversity is indispensable (see analytical
process in Box 1). The realisation of these policy and management objectives demands specific
actions which need to be translated into possible activities. For the execution of these activities
information on biodiversity is required. The extent to which this information is available determines
the research to be undertaken.
The analysis of the information and research needed to execute the activities associated with the
seven policy and management objectives yields a first sorting of priority research concerning tropical
rain forest biodiversity. In order to fine-tune the prioritization, a categorization is applied, which
explains more precisely the kind of biodiversity information required, and the nature of the research to
be carried out. Elements for this categorization are the three biodiversity dimensions (spatial, temporal
and functional) and the distinction of generic versus site-specific types of information.
The final priority setting ought to be based on the following additional criteria:
• Sensitivity of the activity, action, or even the policy and management objective to the information
to be produced by the proposed research;
• Cost-effectiveness of the research proposal, by assessing cost-output ratios;
• The anticipated generic value of the output of a research exercise (“general purpose” research)
• The degree of integration of a single research activity with elements of related fields of research,
based on the assumption that integrated results yield relatively high investment-output ratios;
• Justification of the selection of site, ecosystem, or species.






Information input Information need

on biodiversity on biodiversity

Available Information State of knowledge

Research on Missing information


Availability of cost-
Adequate methods effective research

Development of Inadequate methods

research methods


• Land-use planning for conservation and sustainable use of forest land

• Designation and management of totally protected natural forest areas
• Protection and reintroduction of endangered species
• Support, adoption and documentation of indigenous management of tropical rain forests
• Use and valuation of non-timber forest products
• Sustainable timber production
• Rehabilitation and productive use of secondary forests


The procedure has been applied in a preliminary and partial way (Lammerts van Bueren and
Duivenvoorden 1996). Taking into account the international policy initiatives and the observations
made through consulting its own network, Tropenbos has distinguished a total of seven interrelated
policy and management objectives (see Box 2). The important overall objective is adequate land use
planning, which provides a framework for conservation and sustainable use of forest land. The other
six objectives are more specific and have weaker or stronger links with the overall land use issue.
In view of the actions and activities associated with the seven policy and management objectives,
it is attempted to describe the information needs on biodiversity. The assessment of the actual
availability of adequate information that is needed for policy and management in individual cases has
to take place case by case, and is not elaborated upon in this paper. Tropenbos consulted a selection
of scientists in the field of tropical rain forest ecology and management in order to provide a guide to
the current availability of cost-effective methods with regard to the recognised policy and management
oriented research issues on biodiversity of tropical rain forests. The outcome of this consultation
process may be found in Lammerts van Bueren and Duivenvoorden (1996).


The policy and management oriented information needs on biodiversity identified are categorized
according to the generic and site-specific types of biodiversity information and subsequently listed in
an “information matrix’“. This matrix also shows the principal affinity of the information to one or more
of the three dimensions of biodiversity (spatial, temporal and functional) and the results of the
consultations on method availability. As a whole, the matrix provides a quick overview of the kind of
information on tropical rain forest biodiversity that is needed for the development and implementation
of the relevant policy and management objectives. It also shows whether there is a need for the
development of cost-effective research methods in cases where the information is not sufficiently
available, and could consequently be generated through research.
The information matrix can be used in various ways as a basis for selecting the most urgently
needed research. First, the matrix facilitates the identification of repeated needs for similar types of
generic information and of research areas where methodological improvements are most needed. The
matrix also allows for a critical review of the relevance for policy and management of ongoing or
planned biodiversity research. For instance, matching Tropenbos site activities with the information
items in the matrix may indicate whether relevant policy and management objectives pursued at the

sites are being adequately addressed. This matching may also reveal if research at the sites is primarily
of local or generic interest, and may help to classify research activities in terms of the three dimensions
of biodiversity. In this way, the direction of ongoing or projected research activities may be tested and,
when necessary, reconsidered in order to obtain maximum orientation towards policy and management
objectives using the most cost-effective research methods.
Tropenbos will follow the procedure as it has been outlined above to further develop the biodiversity
component in its own research programme. The policy and management focus of each Tropenbos site
will be matched with the policy and management objectives described in this paper. Research proposals
will be assessed on the basis of a comparison with the information matrix.

Figure 1. Basic set-up of the information matrix as presented in Lammerts van Bueren and
Duivenvoorden (1996). In the left column the generic and site-specific information needs are
listed with regard to the seven policy and management objectives. The right section of the matrix
shows the principal affinity of the information to one or more of the three dimensions of biodi-
versity (spatial, temporal and functional). In the full matrix, this affinity is marked along with an
indication of the availability of research methods.

Policy and management Biodiversity dimensions

guided information
Spatial Temporal Functional
patterns patterns relationships

Short Medium-long

Generic information

information needed

Lammerts van Bueren, E.M. and Duivenvoorden, J. F. (1996). Towards priorities of biodiversity research
in support of policy and management of tropical rain forests. A contribution to the conservation and
wise use of tropical rain forests. The Tropenbos Foundation, Wageningen, The Netherlands.

Summaries of voluntary papers
(also published in Spanish, French and Turkish)



James L. Kiyiapi 1

This paper examines current trends and argues that success in sustainable multiple-use forest
management hinges on: (1) the level of natural resource management and institutional integration;
(2) extent of stakeholder interest and sectoral convergence, and (3) mode and degree of implementation
on the ground. A multiplicity of issues, largely conceptual in nature, with socio-economic, policy and
legal implications are sketched and pitfalls in unclarified conservation strategies pointed out. A focused
definition (or at least recognition) of the complexity of issues is urged. To achieve natural resource
conservation objectives, real community empowerment and active participation are central, and
therefore there is a greater need to pay more attention to the devolution process itself. The Maasai
Mara National Reserve, Kenya, with unique assemblage of large mammal populations which utilize
extensive forest and rangeland areas adjoining the park, and its socio-economic and cultural setting,
is used to illustrate the principles discussed.

Keywords: Conservation and multiple-use forest management; integration, stakeholder interests,

community empowerment and participation.

Department of Forestry, Moi University, P. O. Box 1125, Eldoret, Kenya. Facsimile: (254 321) 63206; E-mail:


L. Laalam, M. El Aichouni, A. Zaki 1

Owing to its particular geographical situation between the Mediterranean to the north, Atlantic to
the west and the Sahara to the south, as well as to the country’s general relief map, Morocco is
characterized by wide climatic variations and considerable bio-ecological diversity.
Ecosystems are mainly made up of forest and preforest ecosystems which are exposed to numerous
constraints, leading to various types of degradation. The main reasons for degradation are overgrazing,
fuelwood gathering and the cultivation of forestry land.
A national priority, the conservation of forest ecosystems is implemented within the framework of
a nationwide strategy for environmental protection, with action programmes that involve the
participation of both local people and NGOs.

Keywords: Biodiversity, ecosystem, environment, degradation, strategy.

Administration des Eaux et Forêts et de la Conservation des Sols, Rabat-Challah, Morocco.



Abdelazim M. Ibrahim 1

In this study, the results of three field tests established along the rainfall gradient within the Central
Caly Plains of the Sudan and that of an experiment conducted under controlled environmental conditions
in a growth chamber were reported. The study involved nine Faidherbia albida provenances. The
aim was to assess the possibility of introducing the species into new habitats in general and clay soil
environments in particular. The results suggest that the species has great potential for extending its
range far beyond its natural range, provided adequate cultural practices are applied. The results also
demonstrated the presence of two distinct genotypes within the species’ gene resources that could
be of use in contrasting environments.

Keywords: Faidherbia albida; genetic variation, utilization, new habitats.

Forestry Research Centre, P.O. Box 7089, Khartoum, Sudan.


Shirong Liu, Zuomin Shi, Li Chen, Juan Ma 1

Baotianman Nature Reserve is situated in Neixiang county, southwest of Henan province in China.
It is a transitional zone from subtropics to temperate. The total area is 5 400 ha, with a forest coverage
of 95.4%. The vertical distribution pattern of forests is pronounced along altitude gradient, viz. lowland
deciduous broadleaved forests mixed with some evergreen components below 1 200 m;
theropencedrymion between 1 200-1 600m; mixed broadleaved/coniferous forests between 1 600-1 900
m; and alpine coniferous forests, elfin-wood and alpine scrubs above 1 690 m, respectively. The
preliminary survey indicates that there are 2 174 species of carpophytes, in which 29 are protective
species identified by the Chinese Government, and 52 are rare and precious species including Picea
neovcitchii, Tapiscia siciensis, Tatracentron siciensis, Cinnamomum japonicum, and Phoebe zhennan,
etc. There are 313 valuable commercial tree species, e.g. Ulmus elongata and Manchurian walnut and
Seplemlobate kelopanax. In addition, there are 201 giant vertebrate animal species of which 23 are
protected species, i.e. leopard, antelope and giant salamander. The number of insects is estimated to
be more than 3 000 species which include 160 butterfly species. With the increasing local population
and ecotourism development, the biodiversity resource conservation is being increasingly threatened
resulting from plant and animal poaching, farmland expansion, and medical and edible plant collection.
The main approaches to solving those problems are: strengthening protected area management,
formulating general management plans, controlling population development compatible with biological
resource capacity, encouraging communication, participation and co-management with local

Keywords: Biodiversity, conservation strategy, protected area.

Research Institute of Forest Ecology and Environment, Chinese Academy of Forestry, Beijing 100091, China.
Facsimile: (86-10) 258 4229


Shahbaz Ahmad 1

Every medicinal plant has some curative properties which may be called its bio-medicinal values. A
number of plants have unique curative properties very peculiar to themselves. With the loss in bio-
diversity many plant species have been lost and many are threatened. We are still not very aware of
the bio-medicinal values which have been lost, but we must become aware of those that are threatened.
This awareness has to be coupled with reforms in therapeutic systems so that drug therapy is in
accordance with accepted national and international standards and that stress on natural ecosystems
is minimal.

Conservator of Forests, Principal, Forest Rangers College, Balaghat (M.P.) 481001, India. Facsimile: (91 7632)


Satishkumar Narkhede

Complexity and diversity are the new frontiers of science, and documenting variations exhibited by
nature is essential for better management of environmental resources. Today, however, on account of
increasing population, the diversity of the ecosystem is under threat of extinction. The National
Forest Policy of India (1988) places major emphasis on the maintenance and preservation of biological
diversity, so as to ensure the conservation of a vast genetic pool on sustainable basis needed for the
survival of future generations of mankind. The effective conservation strategies needed are in situ
and ex situ conservation, and strengthening of preservation plots and biosphere reserves. The existing
centres of biodiversity, viz. national parks and wildlife sanctuaries, which support a vast array of flora
and fauna, need to be managed intensively, so as to maintain community diversity, which is much
sought after for sustainable development.

Department of Silviculture and Agroforestry, Dr. Y S Parmar University of Horticulture and Forestry, Nauni,
Solan 173 230 (HP), India.



Wiratno 1, Suyatno Sukandar 1, Trio Santoso 1

Development of National Parks (NP) in Indonesia is intended to conserve the national biodiversity
which is currently the world’s concern. Under existing policy frameworks, the NP system is planned to
be a model where conservation efforts may be carried out complementary with development activities
with the ultimate goal of generating the welfare of the people. Mid-term evaluation of the implementation
of several Integrated Conservation and Development Projects (ICDPs) in NPs indicate promising
results. However, several challenges and constraints still remain, such as inadequate budgets, lack of
biodiversity information, conflict of interest development sectors, and lack of skilled staff.
To overcome these problems, international cooperation and partnership, particularly in the areas of
technical assistance, expertise and funding, are needed. Among the priority items for cooperation with
international agencies are biodiversity research, economic valuation of biodiversity utilization, and
human resource development.

Keywords: National parks, Integrated Conservation and Development Project, international


Directorate General of Forest Protection and Nature Conservation, Ministry of Forestry, Manggala Wanabakti
Building Block I, 8th Floor, Jakarta, Indonesia. Facsimile: (62 21) 5720229


Agathe Cimon 1

The Quebec Government subscribes to the overall principles of sustainable development. In line
with this, in 1996, the Government adopted a strategy for implementing the international Convention
on Biodiversity. The Ministère des Ressources naturelles is closely involved in applying this strategy
and amendments made to the Loi sur les forêts constitute a formal commitment to sustainable forest
management, a commitment defined in terms of six criteria including the conservation of biodiversity.
The ministry has also published a status report of biodiversity in forest environments, which charts a
profile of the current situation of Quebec’s forests and singles out the main action areas to pursue in
order to ensure that forest management is compatible with local biodiversity.

Keywords: Biodiversity, Quebec, sustainable forest management.

Biologist, M.Sc., Direction de l’environnement forestier, Ministère des ressources naturelles du Québec, 880
chemin Sainte-Foy, local 5.50, Québec, Canada, GIS 4x4. Facsimile: (1 418) 6435651


Ken Vance-Borland 1

A conservation planning methodology integrating biodiversity mapping, area-dependent species

analysis, and geoclimatic diversity assessment to maximize sustainability of species, communities,
and ecosystems is described. Biodiversity mapping identifies sites or areas known to support
biologically important features such as rare species, critical habitat, or endangered ecosystems.
Area-dependent species analysis identifies areas required for viable populations of species (large or
medium-sized predators) having large home ranges and low population densities (“umbrella species”).
Geoclimatic diversity assessment identifies areas representing the full range of regional geologic,
climatic, or edaphic variability. Reserve networks are designed to include the areas identified by the
three approaches, and management practices, including restoration and monitoring, are proposed.
The methodology is being applied in the Klamath Mountains region, an area of high conifer diversity.
Biodiversity mapping has identified locations of rare species and ecosystems, roadless areas, and
important watersheds for threatened anadromous fish runs. Area-dependent species analysis has
identified habitat requirements of the Pacific fisher (Martes pennanti), a rare predator associated
with older forests. Geoclimatic diversity assessment, based on soils and climate, indicates the present
reserve system fails to represent environmental variability in the region. A genetic algorithm computer
programme is being developed to identify potential reserve areas fully representing regional geoclimatic

Keywords: Biodiversity, conservation, reserve design, restoration, Klamath Mountains.

Departments of Fisheries and Wildlife and Forest Science, Oregon State University, Corvallis, OR 97331, USA.


Giselda Durigan 1

“Cerrado” is one of the most important types of tropical vegetation in Brazil. Originally, it covered
about 22% of the country, but recently, in spite of its poor and acid soils, “cerrado” domain is being
rapidly occupied by high technology agriculture. With the aim of saving its biodiversity, researches
have been made on preserving and restoring “cerrado” vegetation in Assis, São Paulo State, Brazil.
Some different technologies have been tested, for application in different site conditions, including:
natural regeneration, enrichment planting and reforestation. “Cerrado” has a high potential of regrowth
from roots, mainly in some degraded sites, which have been used as eucalyptus plantations or pasture.
For reforestation, native species, adapted to the acidity and high aluminium content of the soil, have
a better chance of succeeding.

Estação Experimental de Assis, Instituto Florestal, CX Postal 104, Assis, SP, Brazil, 19800-000. E-mail:


Omar Aurelio Melo Cruz 1

The study was carried out in a primary neotropical rain forest (pristine conditions) and secondary
forests (anthropogenic intervention for this race) of between 12 and 20 years of age; situated in the
lower part of the Calima river basin (Buenaventura, Colombia). In each ecosystem, a permanent plot of
0.5 ha was established for which the parameters of vertical, horizontal and total structure were measured.
In the same way, we evaluated the alpha and beta diversity levels in terms of the flora of the woody
plants with a normal diameter (d.a.p.) of over 5 cm.
The profile of the primary forest showed multilayered vegetation with clearings at different stages
of recovery, while the profile of the secondary forest, which was only 12 years old, had only two
layers. The most representative plants in the primary forest had random spatial distribution patterns,
while the plants typical of the secondary and younger forest were distributed in patterns of a gregarious
type or in patches. The 20-year old secondary forest had patterns showing an intermediary level of
completion (three layers of distinguishable trees).
The alpha diversity was measured according to the Margoulef and Menhinick indexes for the
abundance of the species, which gave, for the primary forest and the 12 and 20-year old secondary
forests, values of 15.69, 4.78, 9.43, 2.86 and 12.19, 4.13 respectively. The species abundance indexes of
Shannon (3.352, 2.648, 3.185), Simpson (52.12, 17.5, 32.6) and Berger Parker (28.87, 7.52, 11.95) determined
for the primary and the secondary forests gave the same results.
The beta value – diversity obtained from the percentage of dissimilarity between the primary forest
and the 12-year old secondary forest was 71.4%, while for the two secondary forests it was 63.24%.
This suggests that knowledge of the diversity and the structures of the tropical forest and its behaviour
over time can become tools of great importance for forest recovery and management.

Keywords: Bajo Calima, biodiversity, structure, neotropical.

Forestry Engineer M.Sc., Departamento de Ciencias Forestales, Facultad de Ingeniería Forestal, Universidad del
Tolima, A.A. 546 Ibague, Colombia. Facsimile: (57 982) 644869


J. Lejoly 1, B. Sonke 2

The Dja Fauna Reserve is situated about 200 km from Yaounde in the south-east of Cameroon. With
about 526 000 ha, it is one of the largest reserves in Cameroon. It is covered with heterogeneous
primary forests (Letouzey 1968) and is bordered by the Dja river. The northern part of the reserve is
under intensive human activities on account of the villages located between Somalomo and Ekom.
In order to achieve a good integration of these populations in the reserve, a monitoring and
development programme is carried out in the area. This programme uses line transects to collect data.
The studies carried out on the transects concern the survey of mammals, birds and especially the
survey of trees with dbh (diameter at breast height) over 10 cm.

Laboratoire de Botanique Systématique et de Phytosociologie, Université Libre de Bruxelles, CP 169, Av. F.
Roosevelt 50, 1050 Bruxelles, Belgium. E-mail:
Département des Sciences Biologiques, Ecole Normale Supérieure de Yaoundé, Université de Yaoundé, B.P. 047
Yaoundé, Cameroon.



Emilio Amorini 1, Andrea Cutini 1, Gianfranco Fabbio 1

Reduction in human pressure, e.g. lower demand for fuelwood, has resulted in significant changes
in the forest ecology of the study area which reflect comparable changes in other regions in the
northern and southern Mediterranean basin. Nine permanent sample plots each of 1600 m2 were
established in 1994 in stands aged 45 years and dominated by holm oak. Mensurational data and data
on litter production were collected from these plots which have been subjected to different thinning
regimes. A second survey was carried out after thinning. On the basis of the data obtained
recommendations are made for the future management of these ecosystems.

Istituto Sperimentale per la Silvicoltura, Viale S. Margherita 80, 52100 Arezzo, Italy. Facsimile: (39 575)
353490; E-mail:


O. Heronim, N. Arkadiusz, M. Jerzy, B. Krzysztof, K. Wodimierz 1

The scientific research and development projects dealing with conservation of forest ecosystems
are conducted in Poland in the so-called forest promotional complexes (LKP). Inventory of the
resources and management of multifunctional forestry have to take advantage of new techniques and
technologies, e.g. as those presented at the 18th International Congress of Photogrammetry and
Remote Sensing in Vienna. In Poland the main trends are as follows:
• a new approach to management of forests, based on geographical information systems (GIS),
• use of global positioning system (GPS) for the purposes of updating geographical databases,
• use of image maps and hybrid data models,
• formal and non-formal education in the fields of forestry and ecology, based on multimedia
This paper describes some of the present research projects that are dealing with the afore-mentioned

Keywords: GIS, multimedia, video, education, remote sensing.

Department of Forest Management and Forest Geodesy, Warsaw Agricultural University, Poland. Facsimile: (48
22) 491375; E-mail:


Kâmil œengönül 1, Hüseyin DÚrÚk 2

Being one of the main types of vegetation covers of the world, shrubs called maqui, chaparral, or
with different local names, have great importance in soil and water conservation, grazing, recreational
uses, biomass production and landscaping.
While these ecosystems are mentioned as the buffer zones for soil conservation, in some countries
they have been converted into forest by clearcutting.
Maqui and maqui-like shrubland ecosystems have many different structural properties. Fire is also
one of the main driving factors in these ecosystems. Soils are poor and shallow and plant species are
resistant to drought with deep root systems. This woody cover varies from region to region creating
communities in which some species are dominant.
Utilization of these areas needs new management techniques with a view to conserving biological
diversity and natural balance.

Keywords: Shrubland ecosystems, fire type vegetation, sclerophyll woody plants, potential of

Professor of Watershed Management, University of Istanbul, Faculty of Forestry, 80895 Bahçeköy-Istanbul,
Associate Professor of Silviculture, University of Istanbul.

The following summaries are published only in the original language


Ranil Senanayake 1

The history of modern forestry demonstrates the promulgation of a philosophical attitude which
fostered the growth of plantations that excluded human habitation. The design of such plantations
moved with production-driven criteria to exclude biodiversity. This separation is unfortunate, as
biodiversity is the integral representation of a forest. Forestry, in the sense of extracting timber or
establishing tree plantations, has ignored this critical and fundamental fact.
The state of the world’s forests and plantations attest to this observation. The Convention on
Biological Diversity is the instrument to address this lack in recognizing the true nature of a forest; in
fact, over 99% of its biodiversity is in non timber species. There is a real need for the redefinition of
boundaries and goals in all negotiations that concern forests. The axemen have sheltered long enough
in its shade.

Kenya. Facsimile: 54-2-562175; E-mail:



R.D. Sharma 1

The study of food habits of various wild animals are of paramount importance in wildlife management
practices. Such studies are also helpful in enforcing the provisions of the law.
This paper presents keys for identification of plant remains in faceal matter of ungulates in the Chila
Sanctuary (now included in the Rajaji National Park, U.P.), study area. These keys are useful in the
identification of indigestible or undigested food parts which are eliminated out from the animal body
along with faecal matter.
The main key is based on trichomes and, for cross-checking the main key, supplementary keys
have been based on features like crystals, epidermal cell characteristics, stomatal-characteristics,
sclerides, structural peculiarities in fruit and seed tissues that are present in faecal matter of ungulates.
The studies presented in the paper are also helpful to those interested in the field of plant anatomy,
besides serving the purpose for which they were intended.

Superintendent, Pench Sanctuary, Seoni, India.


Soekotjo 1, Setijono Sastrosoemarto 2, Burhanuddin Sarbini 3

Indonesia has initiated the Indonesian Forest Health Monitoring system (INDO FHM) with the
objective to determine the current condition of Indonesian Forest Ecosystems with respect to
sustainability and biodiversity, and subsequently to tract changes and trends. This initial state is
supported by the International Tropical Timber Organization and Government of Indonesia and also
is a collaborative research activity with U.S. Forest Health Monitoring Program.
A training workshop to discuss selected indicators to provide an assessment of conditions in
tropical rain forests and technology transfer by a U.S. Forest Health Monitoring Team on Forest
Health Monitoring Plot System has been conducted from September 6 to 11 1996. As follow up to this
training workshop, a Forest Health Monitoring Reference Plot was established in Pulau Laut, South
Kalimantan Province. The trainees will become indicator leads and trainers for future Indonesian
Forest Health Monitoring; the activities consist of (1) testing and selecting indicators in Demo Plots,
(2) estimating the conditions and trends of Indonesian forests in detection monitoring, and (3) to train
Indonesian regular crews.
A population of Demonstration Plots of Forest Health Monitoring (Demo Plot) will be established
in Pulau Laut and Jambi. The Demo Plots will be used to test selected indicators as discussed during
training workshop. In addition to Demo Plots, a population-based Forest Health Monitoring plot
system will be established for detection/monitoring purposes. This detection/monitoring is ground-
based, long-term plot monitoring system that will provide population based estimate of the condition
and trends of Indonesian tropical rain forests by monitoring the proportions of the forest population
that are in poor, subnominal, nominal or optimal condition for each indicator.

Keywords: Forest health monitoring, reference plot, demonstration plot, detection monitoring,
indicator lead.

Professor in Forestry, Gadjah Mada University, Yogyakarta, Indonesia
Associate Professor in Forestry, Gadjah Mada University.
Staff of Directorate-General of Forest Inventory and Land Forest Inventory and Land Forest Planning, Ministry
of Forestry, Government of Indonesia.


Farzana Panhwar 1

The Province of Sindh lies in the southern part of Pakistan and extends from 23o-35 to 28o-30 and
66 -42 to 71o-10 E, with an area of approximately 143 000 km2 of which about 5.5 million ha or 55 000 km2

are irrigated from the River Indus. The population of Sindh as of today is about 30 million.
Throughout Sindh, we have the same culture but with slight variations due to either different
occupations and different economies due to local conditions. However, land use varies from place to
place and natural resources make food and life-style slightly different; climate also varies and so does
cropping patterns which in turn are governed by the availability of water, soil type and groundwater
tables. The yields in general are poor, being one-third or even less than those in the developed

President, Sindh Rural Women’s Uplift Group, 154-C., Unit No. 2, Latifabad, Hyderabad (Sindh), Pakistan.
Facsimile: (92 221) 860410/5830826


K. Karoles 1

The paper provides a comprehensive statistical description of the national forest estate of Estonia,
which has significantly increased in area during the past 50 years and now covers 48% of the land
surface. A brief history of forest conservation in Estonia is given and areas currently conserved in
national parks and other protected areas are described. Management regimes for protected areas and
commercial forests are outlined.

Estonian State Centre of Forest Protection and Silviculture, Estonia.


A. Çelik 1, A. Güvensen 2, Ö. Seçmen 2, M. Öztürk 2

Liquidambar orientalis forests occupy a special place among the Turkish forest ecosystems. The
forests of this tertiary relict endemic taxon are found in Southwest Anatolia mainly Marmaris, KöyceXiz
and Çine-Bucak lying between Antalya and Isparta. A big population covering nearly 100 ha is found
locally along the stream bed between Umurlu and Kök in AydÚn. This forest embodies other endemic
species as well, such as Quercus aucheri. However, the forest is facee with great threats due to cutting
and felling for field openings. There is a great need for protection of this area because L. orientalis
holds an important position in the biodiversity of Turkey. The best way to protect this area is to create
awareness among local inhabitants and involve them in protective measures. The paper presents
ecological characteristics and conservation strategies for this forest.

Pamukkale Univ., Sci.Arts Faculty, Biology Dept., KÚnÚklÚ-Denizli, Turkey. Facsimile: (90 232) 3881036
Ege Univ., Sci. Fac., Botany Dept., Bornova-Izmir, Turkey.



Yahya AyalÚgil 1, Adnan Uzun 1

The Tertiary relict Liquidambar orientalis Mill., (Oriental sweet gum tree) occurs as pure stands
mainly in the flood plains of South-west Turkey. One of the largest stands is located in the “Specially
Protected Area of Dalyan-KöyceXiz”. To protect those stands, it was decided by the Authority for the
Protection of Special Areas (APSA) to set aside an area of 286 ha as a nature reserve and arboretum.
The present paper describes the main characteristics and floristic peculiarities of the reserve area and
suggests conservation strategies.

Faculty of Forestry, University of Istanbul, Department of Landscape Architecture, Bahçeköy, 80895 Istanbul,


Saime Eski Baaran

The paper is based on a master’s thesis entitled The floristic composition of Termessus National
Park. The park lies at 1050 m above sea level 30 km northwest of Antalya and embraces 52 plant
families the most important of which are the Composiae (17), Leguminosae (18), Labiatae (10),
Caryophyllaceae (9), Rosaceae (8) and Liliaceae (8).

Zonguldak Karaelmas, University Faculty of Forestry, BartÚn, Turkey.



Battal ÇÚplak1, Fatma Banu Aslan 2

About 35 species of Eupholidoptera Ramme (Orthoptera: Tettigoniidae) have been described.

These species are distributed through the north Mediterranean region where maqui formation is
dominant. The relationship between the distribution of Eupholidoptera and maqui formation appears
to be due to such defensive behaviours as evasive movements, camouflage, bark and twig mimicry.
Especially rapid evasive movement is a common defensive adaptation in Eupholidoptera to escape
from their potential predators. Rapid evasive movement, as well as other defensive adaptations, seems
to strictly depend on maqui plant community. Thus maqui formation plays a protective role on these

Keywords: Orthoptera, Eupholidoptera, defensive adaptations

Akdeniz University, Department of Biology, Antalya, Turkey.|
Research Assistant, Akdeniz University, Department of Biology, Antalya, Turkey.


Nurettin Elbir 1

The large, ancient monument (specimen) trees, documented in this paper, are found scattered
throughout Turkey and are often found in gardens of mosques, madrasahs, near large tombs, in
villages or occurring in isolated areas in nature. Of the various tree species recorded, both conifer and
broad leaved, the plane tree (Plantanus orientalis) is the most common. It is estimated that the oldest
specimen tree in Turkey, a plane tree, which is in the cemetery of KÚzÚksa in the region of BalÚkesir, is
2100 years old. It is recommended that a formal inventory of such specimen trees be made in Turkey.

Forest Engineer, Member of Turkish Association for Conservation of Nature and Natural Resources. Turkey.



A. Gezer 1, e. Özdemir 2, S. Gülcü 2

The lakes district is located in one of the most important floristic regions which is called
“Mediterranean Flora Region”. A lot of primitive plants, such as fungi and fern and hundreds of shrub
and tree species, occur in the district. The natural taxa are not only rich in numeral quantity, but also
rich in genetic diversity in themselves. The main reasons for these are the prevailing climate
differentiation and extraordinary diversity of topographic and soil characteristics in the district.
However, the district is at a stage which is so sensitive that it could rapidly lose its natural balance. For
this reason, and its extraordinary characteristics, the district deserves an exceptional status in Turkey’s
forestry research and application activities.
The vegetation cover of the district has for centuries extended many considerable economical,
socio-political and collective/cultural functions and services to the people of Turkey. However, so far
it has been degraded by extensive intentional or unintentional intervention such as land clearing,
overgrazing, illegal cutting and forest fire. In this way, a considerable number of gene sources are
faced with genetic erosion and reduced in circumstances. The solution to the problems may be
realized by protection, improvement activities and multiple scientific approaches to the matter. This is
an extraordinarily good opportunity to extend a healthy environment and bright future to the people
of Turkey and the world.

Keywords: Genetic resources and diversity, protection of resources and environment.

Faculty of Forestry, Ankara University, ÇankÚnÚ. Turkey. Facsimile: (90 376) 2136983
Faculty of Forestry, Süleyman Demirel University, Isparta, Turkey.


Fatma Güne1, Sami Y. Ölter 2

In order to protect biological diversity, attempts are being made to expand areas of nature conservation
since 1958.
According to the current Turkish laws, national parks, nature reserve areas, nature parks, natural
monument and wildlife protection areas have been established, planned and protected by General
Directorate of National Parks, Game and Wildlife.
Although the majority of these areas are found in forest lands, some are also in steppe and wetland
Thirty-one national parks, 32 nature reserve areas, 12 nature parks, 53 natural monuments, 113
wildlife protection areas and 40 wild animal breeding stations have been set up since 1958. Their total
areas reach more than 4.5 million ha.

Keywords: National parks, wildlife conservation in Turkey.

Biologist, Head of Section, General Directorate of National Parks and Game-Wildlife. Turkey.
Forest Engineer, Deputy General Director, General Directorate of National Parks and Game-Wildlife.



Hayrettin Karaca 1

Biodiversity refers to variety and abundance of species, their genetic composition and the ecosystems
where they live in. It also includes ecological processes taking place among them. It may occur in a
local area, in a region or in the whole biosphere. Biodiversity provides stability, health and productivity
to a forest ecosystem. Turkey exhibits unique features in its biodiversity, for it is located on the cross-
roads of three continents. Relatively large portions of plant and animal species are endemic, and they
are available only in this land as genetic resources. Yet, forest ecosystems in Turkey have been
exploited under different civilizations within the last few millennia, which resulted in undesirable
impacts on biodiversity. Today, only 11.4% of the land is covered with “normal” forests. Various
technical, social and legal measures were suggested to improve and conserve biodiversity in forest

Keywords: Biodiversity, forest ecosystems, Turkish forestry, national parks.

T.E.M.A. Foundation, Istanbul, Turkey.


I. KÚzÚroXlu

In this study, the bird species which are becoming extinct in Turkey are examined. Natural scientists
prepare “Red Data Books” for taking the necessary preventive measures and for determining the
dangers with which the bird, plant and animal species are confronted. The books prepared have an
attribute of international importance since they concern the people from all sectors who are interested
in protecting nature. The reasons why the bird species disappear are analysed in two main groups: 1.
As a result of the consumer pressures of our people that directly pave the way to extinction. 2. The
indirect influence on bird species which leads to extinction. The functions of the “Red Data Book”
which are prepared for the endangered bird species are also mentioned in the study. For the many bird
species living in Turkey which are faced with extinction, censuses, observations, findings and degree
of endangerment are determined from the studies that reflect the size of population of those species.
In addition to this, the objectives of the Red Data Books are scrutinized. The degrees of extinction
degrees of the non-passerine groups of bird species in Anatolia are given.

Keywords: Red Data Books, birds, Turkey.

Professor, Hacettepe University, EXitim Fak., Turkey.


OXuz KurdoXlu

Old-growth forests (primary forests, little touched by man, with a high proportion of old trees
reaching natural maturity) represent one of the most important and threatened of habitats in north-
east Turkey. Such forests are of exceptional biodiversity and aesthetic value, as well as being important
for erosion control, tourism, etc.
The mountains of north-east Turkey are amongst the most important for nature conservation in
Turkey. A combination of varied geology, great altitudinal range and distinctive climate has resulted
in a vegetation that is of exceptional importance to nature conservation.
The fauna is also of outstanding importance. Large mammals (e.g. wolf, chamois, brown bear), birds
(e.g. Caucasian black grouse, Caspian snowcock, woodpeckers, raptors) and reptiles (e.g. vipers) are
well represented.
It is clear that of the total forest cover of over 913 000 ha (excluding protected areas), only 108 571 ha
are of an age that can be regarded as old-growth forest, nearly 12% of the total forest cover.

Keywords: Old-growth forest, biodiversity.

The Society for the Protection of Nature, P.K. 18, Bebek-Istanbul, Turkey. Facsimile: (90 462) 3260020


Gülsün ÖmeroXlu

In spite of various efforts which have been made in the last years for the prevention of the loss of
biological diversity, the loss of the species and their habitat caused by fires, pollution and poor
management of the forest resources is still a serious problem.
New management models should be developed and in situ protection model should be adopted.
The efficiency of existing management models should be increased and the deficiencies of existing
management legislation should be determined.
During the protection strategies, the local people, especially women and young people, should be
selected as target groups and their participation in protection studies should be ensured. On the other
hand, the contribution of local authorities to these protection studies should be encouraged. The
experiences of non-governmental organizations (NGOs) in providing public access to information and
public participation in the name of natural protection should be taken into consideration in the
development of the protection management models.

Keywords: Protection, utilization, in situ, participation, women.

General Directorate of Environment Protection, Eskiehiryolu 8. km, 06530 Ankara, Turkey. Facsimile: (90
312) 2862271


Zehra Özpay 1

The Western Blacksea Region with its rainy and humid climate has conserved its natural state and
is home to a large number of plants. Thus, Bolu (is placed in this region) is a lucky city. Because it has
wide woodlands and varied species, it seems that biological diversity is very rich. Of course this
situation is very important, but it is not enough. First, we have to understand that preservation of this
biological diversity, and the environmental conservation of species in their natural habitat, is the most
important thing.
According to some studies, management plans and our observations, during the last 20 years,
some species are in a critical position, for example, Ulmus sp. (elm), Castanea sp. (chestnut), etc.
Although some species are protected in their habitat, as Corylus colurna L. (hazelnut, in Kale-Bolu
region). Almost all Turkish chestnut trees are critically ill. Otherwise, some species have wide lands, as
Bornmullerian fir. But, this species is very sensible to air pollutants. So, in the near future there is no
guarantee that this species will not die. In past years, productive and high-quality Oriental beech
forests have been reduced due to heavy cutting. On the other hand, varied species (their seeds, bulbs,
etc.), some of which are endemic, are transferred abroad.

The Western Blacksea Forest Research Institute, Bolu, Turkey.