IV A hierarchical terminology for more or less natural forests in relation to sustainable management and biodiversity conservation

Erik Buchwald

Prepared September 2002 for Forest Ecology and Management

Abstract

Existing terminologies since the late 1980's are reviewed revealing gaps, inconsistencies and overlapping definitions. Many terms have been used very differently by different organisations with semi-natural forest being the worst example. It has variously been used for any kind of forest except exotic plantation. A new hierarchical terminology is proposed based on the work of many authors, persons and organisations working with this issue during the 1990's. Fourteen mutually exclusive levels of naturalness are delimited, defined, named and grouped, including virgin forest, old-growth, newly untouched forest, exploited natural forest, native plantation, exotic plantation and exotic self-sown forest. Consistent definitions for groups of levels and for qualifying features are also presented, e.g. for natural forest, planted forest, native forest, exotic forest, primary forest, minimum-intervention forest, ancient woodland and grazed forest. The definition of natural forest follows UNCED, FAO, EU and the World Bank. For other terms official sources like FAOs Forest Ressources Assesment has been followed as far as possible.

Key words:Natural forest; Terminology; Management; Grazing; Conservation; Biodiversity; Old-Growth;

1. Introduction - the need for common terms

Especially since the UNCED Rio summit 1992 there has been much interest in protecting the natural qualities of forests including their biodiversity and ecological functions. This has lead to many meetings and papers dealing with issues related to more or less natural forests and their biodiversity.

It has been clear in the political processes derived from UNCED that no commonly accepted terminology exists on forests relating to their management and conservation. This has given rise to lengthy debates about the words and long texts explaining local/regional definitions, most of which have no general acceptance. Some papers have not defined their words, paving the way for misunderstandings.

The lack of a generally accepted terminology is a problem both domestically in many countries and even more so internationally - especially in legal/political negotiations and in scientific work. Both inside and between continents there are problems with misunderstandings (IUFRO 1997 & 2002, COST E4 2000, Lund 2001, FAO 2002).

In this paper I use forest as defined in FAO (2001b), and native tree species as defined in FAO (2001a) for indigenous tree species. When using "predominantly consisting of" about tree species ratios, I mean ratios of vertical crown projections. I focus on the terminologies used during the past ten years, including single works of major importance a few years older.

The main aim of this review is to improve understanding of existing terminologies and to suggest a possible consistent terminology - not to monopolise the terms. The general public often uses some of the words in a broad sense or with a different meaning than any published definition (Peterken 96). This cannot be avoided.

The choice of terms and definitions is affected by the objectives of the setting. If definitions are well-documented, logical and hierarchical it should be possible to translate and relate them to other objectives and settings. First I explore some of the background for why inconsistencies develop, and then set out a new hierarchical approach.

1.1. Earlier attempts at a mutual terminology

Several authors have provided overviews of how the international terminology has been used (Broekmeyer & Vos 1993, Schuck et al. 1994, Peterken 1996, COST E4 2000, Lund 2001).

On the basis of these and other works, international attempts have been made to address the problem of a lacking mutual terminology. Precision is especially important in international research, statistics and political negotiations where definitions have therefore been elaborated (MCPF 1996, UN-ECE/FAO 1997 & 2000, FAO 2001a, IUFRO 2002).

1.2. Problems with the existing terminologies and definitions

Analysing the definitions of the commonly used terms referring to more or less natural forests reveals that one reason for the lack of mutual acceptance could be that most of the terms are not defined logically or precisely enough. Overlaps and gaps are common. It is often impossible from the published definitions to understand where the limits or even the typical range of the features in question are (Peterken 1996, Lund 2001).

A further general problem is that many sets of definitions are vague in places, and leave undefined gaps in others - see appendix 1.

There seems to be a need for a more stringent, logical and precise approach to defining a terminology. This might enable common use of terminology in international work in the future. It might also inspire national organisations to define national terminology more precisely. Improved terminology has for example been recommended in the Lisbon resolutions of the Ministerial Conference on Protection of Forests in Europe in 1998 (MCPF3 1998).

1.2.1. Etymology - are humans part of nature?

One set of problems arises around the concept of naturalness because humans have been an integral part of nature until some time in the past. However most people do not accept the major impacts of modern humans as natural - meaning as an integral part of nature.

Therefore we need to consider a level of impact where the transition happens from humans being a part of nature to being something apart from nature impacting on it. Unfortunately there is no simple logical way of discerning when or why human impact on (forest) ecosystems became too large to be accepted as natural.

The wordpairs nature >< culture and natural >< artificial are antonyms. The words are derived from Latin natura nature (from natus born), cultura cultivation of plants (from colere to till) and arte factum art work (Henderson 1979). These antonyms of general acceptance may help in deciding where to set logical limits compatible with the understanding of most people.

From these words it can be seen that some of the features which set humans definitely apart from nature are the cultivation of plants (tilling, sowing and planting) and the production of art. This is supported by the fact that no animals cultivate plants or produce art. From archaeological evidence the departure from nature can thus be taken as happening in the time period after the first art works and before Neolithic farming started. That would be between 5.000 and 25.000 years ago in much of the World depending on location.

Human activities such as hunting and the use of fire were already widespread ten-thousands of years earlier. This indicates that humans had impacts on nature before ceasing to be part of it.

In reality it is not the artworks/artefacts themselves that disturb our notion of people as part of nature, but the escalated impacts on wildlife and landscape which often occured more-or-less simultaneously. For example evidence from isolated islands and continents shows a close correspondance between prehistoric human colonisation and extinctions of large animals (Stuart 1991, Bradshaw & Mitcell 1999). Extinctions especially hit megaherbivores, which are herbivorous animal species which grow larger than 1000 kg. In several continents these went extinct roughly 10-20.000 years ago. Megaherbivores have major influence on the dynamics of ecosystems and landscapes, and thus their disappearance must have had large impacts (Owen-Smith 1988).

In the light of such archaeological information it can be seen that the idea in the head of many people that forests until a few thousands of years ago were totally unimpacted by humans is a misconception.

This misconception is reflected in numerous published definitions of terms like primeval, pristine and virgin forest. These terms - and in a few cases also natural forest and old-growth (Lund 2001) - have often been defined in a way excluding any human impact ever (WWF/IUCN 1996, MCPF 1996, COST E4 2000). Such definitions are a priori problematical because humans have been present for hundreds of thousands of years with great potential impact already from the start of use of fire more than 200.000 years ago (Andersson & Appelquist 1990).

2. Developing a new hierarchical approach

A hierarchical logic is already partly inherent in many of the features covered by commonly used terms, without being appropriately incorporated or delimited in the published definitions.

The following fundamental forest features can be described in a hierarchical manner:

1. More or less natural origin/genesis >< More or less artificially created/regenerated

2. More or less natural flora, fauna and genepool >< More or less modified by human activities

3. More or less natural processes and structures >< More or less modified by human activities

4. Timespan of continuity of features >< History of change including human influence

5. Type and intensity of management operations and utilisation

In practice, although each of these features is a continuum, breaks in continual distribution tend to be present. The features are also interrelated and can be combined in a multi-dimensional matrix. From such a matrix sections can be grouped and identified to develop a logical hierarchic terminology.

2.1. Logical breaking points in the continual distribution of features

There is a logical breaking point in the continuum of origin/genesis when human action changes the existing /spontaneous vegetation of a site to an artificial stock of trees (by planting or seeding). This is also a breaking point in the continuum of the four other features and is therefore fundamental. The breaking point between natural and artificial regeneration can also occur when planted stands later regenerate naturally.

For flora, fauna and genepool there is a major breaking point in the continuum when native vegetation of a site loses its dominant role to non-native vegetation (e.g. to invasive non-native species). The fauna can be with or without key species of large wide-ranging animals. A usually prehistorical breaking point was at the first human-induced extinctions. It is not deemed feasible to incorporate breaking points relating to lower than species level, even though such impacts on the genepool can be of great importance.

Important breaking points in the continuum of processes and structures are presence/absence of significant native/aboriginal human modification and presence/absence of significant modern/western civilised human modification (e.g. logging). In sites with discontinued modification a breaking point in the reverse direction is when signs of the former human modification can no longer be seen/measured. Other breaking points relate to size (landscape-, forest- or stand-scale) and to seral stages of forest dynamics - which is especially relevant for old-growth versus younger seral stages. A last breaking point refers to presence/absence of a plantation-like structure.

A logical breaking point for continuity is when humans clear forest for other types of land-use, e.g. agriculture. Many hundreds of years of woodland continuity is of prime importance for many elements of flora and fauna (Peterken 1996, Graae 2000). Sites with forest continuity back to the oldest maps often also have continuity much further back in time.

Major logical breaking points in the continuum of management are if sites mainly serve industrial wood production purposes, are left untouched or have special management (e.g. old-time traditional management types, biodiversity protection, hunting estates, grazing, water protection or recreation).

2.2. Levels of naturalness

In order to overview the existing terminology I developed an array of levels of naturalness each with a corresponding definition labelled with a number instead of a name. For each new terminology reviewed, levels were added if necessary to incorporate details of definitions.

When the process was finished, 14 mutually exclusive levels emerged. The most logical and generally accepted parts of existing definitions were reused and incorporated in the definitions of levels. Amendments and adjustments were generally necessary in order to make the definitions more universally useful, consistent and without gaps. Appendix 1 shows the results of the review, while appendix 2 gives the level definitions. Level names have been added to both - see below.

In cases where the references use the words without defining them, there are uncertainties regarding the limits of terms, even though the context often gives a quite clear clue. In other cases definitions are vague or imprecise. Uncertainties are displayed with question marks in appendix 1.

Often it can be difficult to document historical and other features, e.g. whether a 200 years old forest stand is planted or not. I propose different solutions to such documentation problems at different levels, including the documentation topic in each relevant level definition. Any forest stand should be referable to the appropriate level by a field visit supplemented by general historical and biological knowledge of the area.

2.3. Names for the levels

I have tried to match the levels with the most commonly used and directly meaningful terms at hand in order to obtain logical hierarchy and best chances of communication with minimal misunderstandings.

The words primeval and virgin have traditionally been connected to forest of an ultimately natural character - with some differentiation. Primeval bears in the word itself the reference to "the earliest ages (as of the world or human history)" (Lund 2001) making it useful for denoting the ultimately natural situation before human impacts (in most cases a very prehistoric situation).

The term virgin on the other hand denotes a situation which during the generations repeatedly can occur, but where the qualities cannot be restored once lost. It should be acceptable that virgin forest can still develop nowadays under certain ultimately natural conditions. This would be in line with the countries still claiming to have at least some virgin forest. Peterken (1996) supplements with Near-virgin Forest and World Resources Institute (1997) uses Frontier Forest which can be relevant in certain cases - see definitions of these in appendix 2.

For forest with a very high degree of naturalness but not as ultimate as the four abovementioned levels several continents use Old-growth - especially America, Australia and Asia (Russia) (Lund 2001). The term has been the issue of a separate working group to discuss definitions (IUFRO 1997). Many local definitions are very specific, but most can be encapsulated by the more generic definitions used at national scale in Russia, Australia, Canada and the US, which are fairly similar (Lund 2001). The main part of the level n6 definition is unchanged from the United States Forest Service 1989 definition of old-growth (via Lund 2001). An almost identical wording is used and recommended in Canada (Wells et al. 1998).

Alongside the abovementioned terms the following ones are sometimes defined almost identically as synonyms dealing with forests of ultimate to high degree of naturalness: Pristine, primary, antique, climax, ancient and natural forest (COST E4 2000, Lund 2001). These terms are less suited for these levels because they either have a very limited and not generally accepted use and/or do not communicate the correct/precise inherent meaning in them and/or have been used widely with other definitions better relating to the inherent meaning of the words.

This leaves us with the lower levels of naturalness, where it is not possible to find good and broadly accepted names or definitions. The terms Semi-natural Forest, Secondary forest, Modified Forest, Man-made Forest, Planted Forest, Plantation and Plantation Forest are much used, with definitions varying between authors and countries. They are often inconsistent (see Appendix 1). Artificial Forest is also sometimes used.

The range of forest naturalness between levels n6 and n1 is particularly important because large areas in many countries are covered by such forests. They often have a complex balance between conservation and use, and raise a lot of debate about sustainability. I have proposed names for the levels which as far as possible directly reflect their content. See appendix 2.

2.3. Names for groups of levels

On the basis of the analysis of the word natural (etymology and antonyms) and of the logical breaking points, I support the many organisations and authors who have used or defined Natural Forest rather broadly covering many levels of naturalness (e.g. UNCED 1992, EU 1996, Peterken 1996, COST E4 2000, Lund 2001, FAO 2001b, IUFRO SilvaVoc 2002). The minimum level of naturalness in these definitions is mostly at naturally regenerated native forest - regardless of intensity of human use.

Natural Forest should thus be used broadly for relatively to ultimately natural situations, encompassing a broad range of naturalness from level n1 through n10. This matches the international definitions of e.g. UNCED, FAO, IUFRO and World Bank (see appendix 1) and national definitions used in many countries e.g. China, Iran, Vietnam, New Zealand, Brasil (Lund 2001), Sweden (Selander 1969) and Japan (IUFRO 2002). See appendix 3 for definition.

Semi-natural Forest is used in Europe, especially Great Britain, about forests which are more or less natural or resemble such forests enough to make it difficult to tell the difference. Inventory and statistical purposes have been the main reasons for this rather unprecise term, e.g. for sites where it is difficult or impossible to find out if the stand was planted or not - maybe 200 years ago (Peterken 1996). Different works have set up quite different limits for what semi-natural forest is - see appendix 1.

The word semi-natural is more difficult to translate to other languages than most others terms and is not directly meaningful to most people. It has been used for any level of naturalness except p2, Exotic plantation (see appendix 1). I therefore recommend it used as little as possible in international work.

Care must also be taken in translation of the other terms to national tongues. In numerous cases a "direct" translation will give useless and confusing results. This problem has been known for years and will continue, no doubt.

See appendix 3 for other names for groups of levels of naturalness.

2.4. Qualifying features (continuity, grazing, protection status)

The breaking points of the fundamental features cover important divisions, but distinctions along another axis of the matrix can be worthwhile for some purposes. Qualifying features can be used to describe and rank sites in a more detailed manner or as determinants of management, when biodiversity is an important objective. Ellenberg (1988), Peterken (1996) and Vera (2000) can be consulted for details on the features.

I recommend the use of a terminology for continuity, grazing and protection regime in combination with the terminology outlined in appendix 2 and appendix 3, because of the importance and management implications of these qualifying features.

2.5.1 Continuity

Woodland continuity is of such prime importance for biodiversity, that it is worthwhile using it across the whole terminology, discerning between ancient woodland and recent woodland, see box 1. Further specification can be made, e.g. continuity of old trees, dead wood, tree species presence, undisturbed soil or of elements developed by grazing at moderate levels. Regardless of level there can be important differences in biodiversity between ancient and recent woodland. Sometimes the terms primary versus secondary forest are used to indicate if non-forest land-use has occurred, but this leads to confusion with another use of the same wordpair, see appendix 1 and figure 1.

2.5.2 Grazing regime

The original natural grazing pressure by wild animals in the Pleistocene and early Holocene is an issue of much debate (Vera 2000, Svenning 2002). Since those prehistoric times grazing levels have oscillated in complex manners with domesticated herbivores being very important in most continents for hundreds to thousands of years (Bradshaw & Mitchell 1999). Today, breaking points are presence/absence of fencing and of wild/domestic large herbivores (deer, antilope, kangaroo, beaver, buffalo, pigs, oxen, horses etc.) and whether grazing impact is significantly lower, higher or thought to be near the originally natural situation with intact wildlife. Absence of grazing by large herbivores is unnatural from a biodiversity and evolutionary point of view, while intensive domestic livestock grazing is also unnatural and detrimental to forest. See box 2 for proposed terminology.

2.5.3 Protection regime

The protection regime of forest reserves can take on three principally different forms: 1) Minimum-intervention, 2) Traditional Management or 3) Designed Management (NFNA 1994, Peterken 1996 & 2000), see box 3. A code of practice can detail the specific management principles and prescriptions of a reserve or of sub-types of the three forms (e.g. NFNA 1994). Minimum-intervention reserves are often called Strict Forest Reserves, Strictly Protected Forests or Scientific Forest Reserves (COST E4 2000).

Peterken (2000) has elaborated on Minimum-intervention, which he sub-divides into a) Non-intervention and b) Protective Intervention. Typically some human activities are accepted even in Non-intervention reserves, e.g. access, research, grazing regulation, recreation, restoration of natural hydrology or felling trees for security reasons along roads. In Protective Intervention non-native species are controlled also. Minimum-intervention as management prescription is synonomous with untouched forest as prescription (NFNA 1994).

IUCN codings of protected areas are not well suited for stand- or forest-scale assesments because the IUCN class definitions work at protected area scale. Very often at that scale there are multiple habitat types and/or levels of protection regimes, e.g. zones, which cannot be accounted for by the IUCN system (IUCN 1994, EUROPARC & IUCN 1999, COST E27 2001).

3. Discussion

3.1. The role of differing objectives

Many stakeholders use words regarding the naturalness of a forest in relation to their different objectives such as forest policy, statistics, research and biodiversity protection as well as to the general objective of sustainable development. Appendix 1 illustrates the many risks of misunderstandings brought about by different organisations using the same words differently.

Differing objectives can be the reason of some of the variation. But in many cases chance, tradition and historical background seem to be the main reasons, e.g. level n1-n10 being called native forest, natural forest and semi-natural forest in Australia, China and the UK respectively. Similarly, plantation, plantation forest and forest plantation are used synonymously for level p2-p4.

There are tendencies for there to be different "Schools" of terminology. The use of the word semi-natural has spread from the UK to European organisations (MCPF and UN-ECE/FAO), but not in a consistent manner. This may be because semi-natural has been used with differing definitions even inside the UK (Peterken 1996). In North America Canada and the US use more or less the same terminology, and from here the word "old-growth" has spread to other continents.

As can be seen in appendix 1, the World Bank, UNCED, IUFRO and FAO represent another school of terminology. It is of major importance, since it has been used internationally in many contexts both before and after the Rio summit in 1992, and because it is used in more or less legally binding documents of UNCED and EU regarding natural forests and primary forests.

3.2. Documentation problems

The genetic level of biodiversity has not been incorporated in the main terminology, mainly because it is almost never included in the reviewed terminologies, but also because inclusion would give rise to documentation problems. Pollen of trees mix over large and uncontrollable distances in a dynamic way. This makes definitions focused on "clean" local provenances and strains/races almost impossible to use in practice.

Dynamic evolution of species, strains and their distributions continues. This makes it necessary also to reflect on how native range is defined and on when species are deemed exotic. Native range should not be seen statically in relation to political boundaries, but dynamically in relation to soil, climate, dispersal barriers, rate of historical change, potential future changes etc.

Ideally Natural Forest might be defined as forest which has always been naturally regenerated and never artificially regenerated. Such a definition has been attempted (NFNA 1994) but has proven problematical because it is very difficult to know whether a generation preceding the extant stand was planted/sown, e.g. oak and beech forests where planting has been used for hundreds of years at least in Europe.

When a planted stand of native species has later regenerated naturally, the planted origin of the previous generation can usually only be determined by studies of archives or by genetic studies of the trees. Often archives do not exist and they cannot prove whether a pre-archive stand was planted or not. The other method is tedious and cannot give a certain answer either. This seems to be the reason why all other reviewed definitions which call levels n1-n10 natural forest, only employ features of the extant stand.

The abovementioned documentation problems have been attempted solved in the proposed terminology by incorporating different levels of necessary documentation at the various levels. This should be a help when referring a forest stand to a level, e.g. for inventory purposes.

3.3. Organisms as indicators

Ellenberg (1988) attempted to use field layer flora combined with forest management as determinants of a forest terminology. He found that his system works quite well in intensively investigated parts of Central Europe, but not quite so well in southern and eastern parts of Central Europe where investigations of forest vegetation are somewhat fewer. The requirement for detailed knowledge on flora makes the system less useful internationally.

Even so, field layer flora can be of great importance as indicator of naturalness of forests by having certain plant species absent, present or common. Different indicators can be identified for different levels of naturalness. The same goes for other groups having indicator species for more or less natural forests (e.g. lichens, saproxylic fungi and insects). Sets of indicator species are necessarily local/regional, because of limited distributions of indicator species (Hermy et al 1999). Using all species matching specific forest-related traits as indicators instead of using just a few listed species, might be worthwile in order to widen the geographical scope (Graae & Sunde 2000).

Certain sites at a specific level of naturalness can happen to have a more rare or typical biodiversity than some sites of a higher level. This can especially be the case when looking at limited sections of biodiversity and when comparing ancient woodland with recent woodland. For instance soil organisms and saproxylic organisms react very differently to forest management operations which break continuity of old trees and dead wood.

4. Recommendations - Conservation importance

The main use of terminologies on naturalness of forests relates to discussions of priorities relating to conservation of biodiversity. A sustainable balance between use and protection is an important factor in such discussions.

14 levels of forest naturalness have been identified in this review. Some levels, e.g. Primeval, Virgin and Old-growth Forest, are mainly retrospective and the appeal is to protect them. Other levels, e.g. Newly Untouched and Near-virgin Forest, have relevance in nature restoration activities and give hope for the future of biodiversity alongside wise management of the less natural levels. I find that it is of importance to understand all 14 levels when dealing with sustainable development and biodiversity protection, although aggregate levels will be enough for most reporting purposes.

Much work on forest conservation relates to the setting up of networks of protected areas. Site selection for conservation should benefit from using a stringent terminology and analysing which types of sites have high priority for inclusion in a reserve network. Many countries have so little forest with high levels of naturalness left that reserves need to be selected also in disturbed forests in order to represent ecological variation or just to reach adopted area targets. In such selection schemes Ancient Woodland should be given high priority because of the importance of continuity for biodiversity.

Other selection criteria for reserve networks include hydrological regime, which is very important for biodiversity. Sites with restored or intact natural hydrology should be ranked higher than sites with modified hydrology. Old stands and trees should be ranked higher than younger ones. In general sites with a rich, typical, unusual or fairly intact biodiversity should be ranked higher than sites with less well-developed biodiversity.

When sites are selected, protection regimes need to be settled. Continuity of management system can be very important for biodiversity, so that changing Specially Managed Forest, e.g. ancient coppice or wood pasture, to Untouched Forest can be detrimental to important parts of site biodiversity. For other site types, the flora and fauna - especially saproxylic species - can benefit by a change to untouched status. Restored or natural hydrology, natural species composition and wildlife including large herbivores are important for the natural structures and functions of the ecosystem to redevelop optimally (Peterken 1996, Vera 2000).

Recommendations on how much forest should be protected in reserves usually range below 10 - 15 % of the forest area. For the remaining at least 85 % of forest areas Natural Forests with forestry and Man-made Forests (levels n3 through p1) are highly relevant in discussions of sustainable management. All production forests sustain some biodiversity. How much depends on how near-to-natural the management and the forest is. The proposed terminology can hopefully enhance and clarify discussions.

Acknowledgements

This review would have been much more difficult without the major work done by H. Gyde Lund and other referenced authors who have gathered and systemized many definitions and terminologies and made them accessible. I am indebted to Bent Aaby, Sten Asbirk, Richard Bradshaw, Jens Emborg, Hans Kampf, Keith Kirby, Peter Friis Møller, Bent Odgaard, Karsten Thomsen and Frans Vera for fruitful discussions, inspiration, material and for their work on natural forests during the years. The Danish Foreign Minister Per Stig Møller is thanked for ordering the Danish Natural Forest Strategy in 1991 while he was Minister of the Environment and for putting me in charge.

Box 1 Terms on continuity of forest.

Box 2 Terms on grazing regime

Box 3 Terms on protection regime of forest

Appendix 1: 14 levels of forest naturalness with definitions

n10 Ultimate degree of naturalness - Primeval Forest - Forest ecosystems never modified by modern man/civilisation even indirectly, where the degree of impact on the ecosystem by indigenous people has not been significantly higher than the impacts of natural wildfire and of large wild animals (e.g. beaver (Castor spp.) or megaherbivores). The fauna includes a rich host of large animal species and is not significantly affected by human-induced extinctions or changes to animal population densities. Size is landscape-scale. Comm: In Europe and America this type belongs in prehistorical times (before megaherbivore extinctions in stoneage), while examples might still occur in other parts of the world (e.g. perhaps some African National Parks). For prehistoric forests the time-period can be specified, e.g. Pleistocene forest or early-Holocene forest.

n9 Extremely high degree of naturalness - Virgin Forest - Forest ecosystems virtually unmodified by man, and where the degree of former human impact on the forest - including soil and hydrology - has been only slightly more significant than the impacts of wildfire and animals (e.g. beaver (Castor spp.) or megaherbivores), and is no longer obvious. Wildlife inhabits the area with a fairly natural density and species composition including large herbivores and carnivores. Size is forest-scale. Comm: In Europe this term has been used by many authors and lay people for the most extremely natural forests of Europe (Bialowieza and several others, see Peterken 1996) well knowing that they have been impacted somewhat by people in former times (grazing, selective cutting, fires etc).

n8 Very high degree of naturalness - Frontier forest - A frontier forest is an area meeting the following criteria:

It is primarily forested and predominantly consists of indigenous tree species. It is big enough to support viable populations of all indigenous species associated with that forest type -- measured by the forest's ability to support wide-ranging animal species (such as elephants, harpy eagles or brown bears). It is large enough to keep these species' populations viable even in the face of the natural disasters -- such as hurricanes, fires, and pest or disease outbreaks -- that might occur there in a century. It is home to most, if not all, of the other plant and animal species that typically live in this type of forest. Its structure and composition are determined mainly by natural events, though limited human disturbance by traditional activities of the sort that have shaped forests for thousands of years -- such as low-density shifting cultivation -- is acceptable. As such, it remains relatively unmanaged by humans, and natural disturbances (such as fire) are permitted to shape much of the forest. In forests where patches of trees of different ages would naturally occur, the landscape exhibits this type of heterogeneity. (Rearranged/shortened from World Ress. Inst.:http://www.wri.org/ffi/lff-eng/).

n7 Very high degree of naturalness - Near-virgin forest - Forest ecosystems (forest scale) untouched long enough to have attained structures, dynamics and species composition similar to virgin forest, even though they may have been significantly modified, e.g. by clearcutting or agriculture at some time in the past. They are distinguished by a mixture in time and space between different seral stages, e.g. between old-growth stages and younger stages. Human impact on the forest structures is not obvious to see. The time necessary in untouched development before this level can be reached depends on how modified the situation was at the start. It is at least several hundred years if the starting point is a plantation-like forest.

n6 High degree of naturalness - Old-growth forest - Ecosystems (stand scale) distinguished by old trees and related structural attributes. Old growth encompasses the later stages of stand development that typically differ from earlier stages in a variety of characteristics which may include tree size, accumulations of large dead woody material, number of canopy layers, species composition, and ecosystem function. The age at which old-growth develops and the specific structural attributes that characterize old growth will vary widely according to forest type, climate, site conditions, and disturbance regime. For example, old-growth in fire-dependent forest types may not differ from younger forests in the number of canopy layers or accumulation of down woody material. However, old-growth is typically distinguished from younger growth by several of the following attributes: 1) large trees for species and site, 2) wide variation in tree sizes and spacing, 3) accumulations of large-size dead standing and fallen trees that are high relative to earlier stages, 4) decadence in the form of broken or deformed tops or bole and root decay, 5) multiple canopy layers, and 6) canopy gaps and understory patchiness. Old-growth is not necessarily "virgin" or "primeval." Old-growth can develop following human disturbances. (USFS 1989, Wells et al 1998). If the stand is known to be planted/sown or predominantly consists of exotics it is referred to level p4, Partly-natural planted forest or p1/p2, Exotic forests.

n5 Quite high degree of naturalness - Long untouched forest - Relatively intact forest (stand level) that has been essentially unmodified by human activity for the past sixty to eighty years or for an unknown, but relatively long time. Signs of former human impacts may still be visible, but strongly blurred due to the decades without forestry operations. The time limit depends on how modified the forest was at the starting point. If the stand is known to be planted/sown or predominantly consists of exotics it is referred to level p4, Partly-natural planted forest or p1/p2, Exotic forests.

n4 Moderately high degree of naturalness - Newly untouched forest - Forest stands where forestry operations have been discontinued or never occurred since stand establishment, and which are known to have been left untouched for less than sixty to eighty years. Signs of former human management are usually easily visible, becoming more blurred with time. In principle any stand would belong here between forestry operations. If discontinuation of operations is only because of long management intervals, the stand is referred to lower levels. If the stand is known to be planted/sown or predominantly consists of exotics it is referred to level p4, Partly-natural planted forest or p1/p2, Exotic forests.

n3 Medium degree of naturalness - Specially managed forest - Forest stands with low-intensity use (allowing for presence of significant old-growth attributes) or where long continuity of pre-industrial type traditional use has given rise to a more or less characteristic forest structure, often with an interesting biodiversity and also often of cultural value. Examples are coppice, pasture forest, non-industrial selective logging and various stands of low accessibility or with protective or recreational functions. Excludes stands where the characteristic structures and attributes have been lost due to discontinuation or to large changes in the traditional use. If the stand is known to be planted/sown or predominantly consists of exotics it is referred to level p4, Partly-natural planted forest or p1/p2, Exotic forests.

n2 Fairly low degree of naturalness -Exploited natural forest - Forest stands used and modified by humans so that the forest structure and species composition is more or less heavily changed from the originally natural situation, but still predominantly consisting of self-sown native trees, and without a plantation-like structure. Can be exploitation without preceding silviculture (e.g. modifying old-growth forest), or stands with medium-intensity forest management (e.g. near-to-nature forestry). If a significant proportion of the stand is known to be planted/sown it is referred to level p4, Partly-natural planted forest.

n1 Low degree of naturalness - Plantation-like natural forest - Forest stands predominantly consisting of self-sown native trees with high-intensity forest management, so that the forest structure has become plantation-like by being even-aged, having relatively low tree ages, fairly regular tree spacing and only one or two tree species in the canopy layer. If it is not certain that the stand is self-sown it is referred to level p3, Native plantation.

p4 Low degree of naturalness - Partly-natural planted forest - Forest stands predominantly consisting of planted or sown native trees, and having somewhat natural structure by being un-evenaged, with mixed species or having significant ingrowth of self-sown trees (e.g. because multiple species have been planted and not intensively managed for timber and/or because of old age and low intensity of forestry). In certain cases such stands are left untouched long enough to correspond to levels n4, n5 or n6. They may be called newly untouched, long untouched and old-growth planted forest respectively.

p3 Extremely low degree of naturalness - Native plantation - Intensively managed, basically even-aged forest stands predominantly consisting of native trees, established artificially by planting or sowing with regular spacing. Often monocultures, but sometimes two or more species are established together. Usually the genetic variety of the plantation trees differs from the original genetic variety of the locality.

p2 Almost no degree of naturalness - Exotic plantation - Forest predominantly consisting of non-native tree species, where the stand origin is artificial by planting or sowing. In certain cases such stands (or p1) might be left untouched long enough to correspond to levels n4, n5 or n6. Such stands may be called newly untouched, long untouched and old-growth exotic forest respectively.

p1 Variably low degree of naturalness - Exotic self-sown forest - Forest stands predominantly consisting of self-sown non-native tree species. In certain cases this category can spread at an undesirable scale, e.g. to the extent that it has replaced or seriously suppressed the species previously occupying the specific area. Such cases can be called forest of invasive species. If it is not certain that the stand is self-sown it is referred to level p2, Exotic plantation.

Any forest stand should be possible to refer to level by a field visit supplemented by general historical and biological knowledge of the area.

Appendix 2: Definitions of terms covering multiple levels

n1 - n10 = Natural forest - Forest stands predominantly composed of self-sown native trees. They are in other words naturally regenerated and complementary to planted/sown stands and exotic self-sown stands. (After FAO 2001a). Comm: They range from undisturbed by humans (e.g. virgin forest) to heavily utilized for grazing, wood production etc. The degree of naturalness can be described by referring to more detailed levels or subclasses of natural forest. Stands which have natural processes or structures present are taken to be self-sown if no information indicates otherwise. "Natural forest dynamics" refer to the natural dynamics of natural forest undisturbed by humans. They often include a cycle of seral stages of growth.

p3, 4 + n1 - n10 = Native forest - Forest stands predominantly consisting of native trees. These can be planted, sown by man or self-sown. They can be used/managed or not. Comm: Native trees means species which have evolved in the same area, region or biotope where the forest stand is growing and are adapted to the specific ecological conditions predominant at the time of the establishment of the stand (FAO 2001a).

p1, 2 = Exotic forest - Forest stands predominantly consisting of non-native trees. These can be planted, sown by man or self-sown. They can be used/managed or not.

p2 + p3 Forest plantation - Intensively managed forest stands established artificially by planting or seeding and having regular spacing and even-aged structure. Usually only one tree-species is planted, but sometimes two or more are planted together. (After FAO 2001a). Comm: They can be either native or exotic species or mixed. Excludes native stands which were established as plantations but which have lost the specific structure of plantation later, e.g. because of long time without intensive management or because many of the planted trees have been cut and self-sown trees have been allowed to take their place.

p2 - p4 = Planted forest - Forest stands established artificially by planting or seeding. (After ITTO 2002).

p1 + n1 - n10 = Self-sown forest - Forest stands predominantly consisting of trees which have germinated and grown from spontaneous seedfall, either wholly naturally or influenced by various silvicultural activities, e.g. scarification.

p1 - p4 = Man-made forest - Forest stands which are either predominantly consisting of non-native (introduced, exotic) tree species or which have been established artificially by planting or seeding.

n4 - n1= Secondary forest - Forest stands of native species regenerated largely through natural processes after significant human and/or natural disturbance of the original forest vegetation at a single point in time or over an extended period, and displaying a major difference in forest structure and/or canopy species composition with respect to primary forests on similar sites. (After Chokkalingam & Jong 2001). Comm: Typical disturbance types are clearcutting, logging, flood, fire and deforestation in general. Continued modification often takes place, e.g. forestry. Can revert to primary forest if untouched natural development proceeds for a sufficiently long time.

n5 - n10 = Primary forest - Relatively intact forest areas that have always or at least for the past sixty to eighty years been essentially unmodified by human activity. Human impacts in such forest areas have normally been limited to low levels of hunting, fishing and harvesting of forest products, and, in some cases, to historical or pre-historical low intensity agriculture. (After World Bank 1991, EU 1996)

n4, n5, n7, n9, n10, most of n8 and n6 and minor parts of p1, p2, p4. = Untouched forest (since specified year/decade/time) - A common denomination of forests uninfluenced by forestry activities (felling, planting etc.) from a specified time. Is not affected by research, hunting or recreational activities etc. Usually the length of time is minor except in woods designated as untouched (=minimum-intervention) forest by administrative or legal order.

References

Aird, P.L., 1992 (Foreword by J.S. Maini). Conservation for the Sustainable Development of all Types of Forests Worldwide: A Compendium of Concepts and Terms. Preliminary version of Aird (1994) circulated for the UNCED Rio Summit by the Canadian Delegation.

Aird, P.L., 1994. Conservation for the Sustainable Development of Forests Worldwide: A Compendium of Concepts and Terms (in English, French and Spanish). The Forestry Chronicle, 70, 666-674. Revised version of Aird (1992).

http://www.forestry.utoronto.ca/ac_staff/emeritus/My%20Webs/english.htm

Andersson, L., Appelquist, T., 1990. The influence of the pleistocene megafauna on the nemoral and boreonemoral ecosystems. A hypothesis with implications for nature conservation strategy. Svensk Bot. Tidskr. 84, 355-368. Lund.

Bradshaw, R., Mitchell, F.J.G., 1999. The paleoecological approach to reconstructing former grazing-vegetation interactions. Forest Ecology and Management 120, 3-12.

Broekmeyer, M.E.A., Vos, W., 1993. Forest reserves in Europe: A review. In: Broekmeyer, M.E.A., Vos, W. and Koop, H. (eds.), 1993: European forest reserves. Proceedings of the European forest reserves workshop. PUDOC-DLO, Wageningen, 306 pp.

Chokkalingam, U., Jong, W.D., 2001. Secondary forest: a working definition and typology. International Forestry Review 3, 19-26.

COST E27, 2001. Memorandum of understanding for the implementation of a European Concerted Research Action designated as COST Action E27 "Protected Forest Areas in Europe - Analysis and Harmonisation (PROFOR)". European COoperation in the field of Scientific and Technical research (COST). Secretariat, Bruxelles.

COST E4, 2000. Forest Reserves Research Network in Europe. Mission, Goals, Outputs, Linkages, Recommendations and Partners. COST Action E4, Final report. With web glossary of international terms of natural forests and natural forest research at:
http://www.efi.fi/Database_Gateway/FRRN/howto/glossary.html . European Cooperation in the field of Scientific and Technical research (COST). The Finnish Forest Research Institute, Joensuu.

Ellenberg, H., 1988. Vegetation Ecology of Central Europe. 4^th Edition. Cambridge University Press.

EU Commission, 1991. CORINE Biotopes Manual. Habitats of the European Community. Data specifications part 2. EUR 12587/3. Luxembourg.

EU Commission, 1996. Forests in sustainable development. Annex 4.6, Glossary of Terms.
http://europa.eu.int/comm/development/forests/en/en4_6.htm

EU Commission, 1999. Interpretation Manual of European Union Habitats. EUR 15/2. DG Environment.

EUROPARC and IUCN, 1999. Guidelines for Protected Area Management Categories - Interpretation and Application of the Protected Area Management Categories in Europe. EUROPARC & WCPA, Grafenau, Germany. 48 pp.

FAO, 1995. Forest Ressources Assesment 1990. Global synthesis. FAO Forestry Paper 124. Rome.

FAO, 2001a. Global Forest Ressources Assessment 2000. Main report. App. 2, Terms and definitions. FAO Forestry Paper 140, Rome.

FAO, 2001b. State of the Worlds forests 2001. Annex 1, Definitions and acronyms.
http://www.fao.org/forestry/FO/SOFO/sofo-e.stm

FAO, 2002. Expert meeting on harmonizing forest-related definitions for use by various stakeholders. Rome, 23-25 January 2002. Proceedings. Rome.

Forestry Authority, 1994. The management of semi-natural woodlands. Forestry Practice Guides 1 - 8. The Forestry Authority, Edinburgh, UK.

Forest Stewardship Council, 2000. FSC Principles and Criteria. Document 1.2. Revised February 2000.
www.fscoax.org/html/1-2.html

Graae, B.J., Sunde, P., 2000. The impact of forest continuity and management on forest floor vegetation evaluated by species traits. Ecography 23, 720-731.

Henderson, I.F., 1979. Henderson's dictionary of biological terms. Ninth edition, edited by S. Holmes. London.

Hermy, M., Honnay, O., Firbank, L., Laweson, J., 1999. An ecological comparison between ancient and other forest plant species of Europe, and the implications for forest conservation. Biol. Cons. 91, 9-22.

ITTO, 2002. International Tropical Timber Organization - Guidelines for the restoration, management and rehabilitation of degraded and secondary tropical forests. Draft report prepared on behalf of ITTO by an International Expert Panel held in Bern, Switzerland 18-22 February 2002.

IUCN, 1994. Guidelines for Protected Area Management Categories. IUCN Commission on National Parks and Protected Areas with the assistance of WCMC. IUCN, Gland, Switzerland and Cambridge, UK. 261 pp.

IUFRO, 2002. SilvaVoc. A Clearing-house at the service of the forest community. Terminological database (SilvaTerm).
http://iufro.boku.ac.at/silvavoc

IUFRO Unit 6.03.02, 1997. A network for terminology within IUFRO network. Discussion Group on Old Growth - in operation January to June 1997.
http://www.wsl.ch/forest/risks/iufro/discussion.html

Lund, H. Gyde, 2001. Definitions of old growth, pristine, climax, ancient forests, and similar terms. [Online publication], Manassas, VA, USA: Forest Information Services. Misc. pagination.
http://home.att.net/~gklund/pristine.html

MCPF, 1995. Interim report on the Follow-up of the Second Ministerial Conference. Ministerial Conference on the Protection of Forests in Europe, 16-17 June 1993 in Helsinki. Annex III. Ministry of Agriculture and Forestry, Ministerial Conference on the Protection of Forests in Europe Liaison Unit in Helsinki.

MCPF, 1996. Ministerial Conference on the Protection of Forests in Europe, Progress Report 1996, 27 - 54. Ministry of Agriculture, Rural Development and Fisheries. Liaison Unit in Lisbon.

MCPF3, 1998. General declaration and resolutions adopted. Third Ministerial Conference on the Protection of Forests in Europe, Lisbon, June 1998. Liaison Unit in Lisbon. Ministry of Agriculture, Rural Development and Fisheries of Portugal.

NFNA, 1994. Strategy for natural forests and other forest types of high conservation value in Denmark. The National Forest and Nature Agency, Ministry of the Environment. Copenhagen.

Owen-Smith, N. 1988. Megaherbivores. Cambridge University Press.

Pajari, B. JK., Schuck, A. (Eds.) 1994. Terms and Definitions for the Follow-up of the Second Ministerial Conference on the Protection of Forests in Europe. Working Paper 17.6.94, 18 pp.

Peterken, G. F., 1996. Natural woodland. Ecology and Conservation in Northern Temperate Regions. Cambridge Univ. Press.

Peterken, G. F., 2000. Natural reserves in English woodlands. English Nature Research Reports, No. 384. English Nature, Peterborough.

Rosson, J.F., 1995. Forest Plantations in the Midsouth, U.S.A. US Forest Service, Southern Forest Experiment Station. Research Paper SO-290.

Selander, E., 1969. Glossary of forest terms, Swedish-English. Skogsordlista, Svensk-engelsk. Swedish Centre of Technical Terminology Publication No. 43. Lund.

Spencer, J.W., Kirby, K.J., 1992. An inventory of ancient woodland for England and Wales. Biological Conservation 62, 77-93.

Stuart, A.J., 1991. Mammalian extinctions in the late Pleistocene of Northern Eurasia and North America. Biol. Rev. 66, 453-562.

Svenning, J., 2002. A review of natural vegetation openness in north-western Europe. Biological Conservation 104, 133-148.

UNCED 1992. As printed in: Johnson, S., 1993. The Earth Summit: The United Nations Conference on Environment and Development (UNCED). 532 pp. International Environmental Law and Policy Series. Graham & Trotman, London.

UN, 1997. Report of the Ad Hoc Intergovernmental Panel on Forests on its Fourth Session. New York, 11-27 February 1997. E/CN.17/1997/12. New York.

UN-ECE/FAO, 1997. Temperate and Boreal Forest Resources Assesment 2000 (TBFRA 2000). Terms and Definitions. UN New York and Geneva, 13 pp.

UN-ECE/FAO, 2000. Forest Resources of Europe, CIS, North America, Australia, Japan and New Zealand. UN-ECE/FAO contribution to the Forest Resources Assesment 2000. Geneva Timber and Forest Study Papers, no. 17. Main report. App. I - TBFRA-2000 Terms and Definitions.

IUCN/UNEP/WWF, 1991. Caring for the Earth. A Strategy for Sustainable Living. Gland, Switzerland.

Vera, F.V.M., 2000. Grazing Ecology and Forest History. CABI, Wallingford.

Wells, R.W., Lertzmann, K.P., Saunders, S.C., 1998. Old-growth definitions for the forests of British Colombia, Canada. Natural Areas Journal 18, 279-292.

Wenming, L., 1999. Recent changes of forest policy in China and its influences on the forest sector. Chinese Academy of Forestry, Beijing.
www.iges.or.jp/fc/ir99/4-10-lu.pdf.

World Bank, 1991. The Forest Sector: A World Bank Policy Paper. Washington D.C. (Also called "World Bank Forest Strategy").

World Ressources Institute, 1997. Last Frontier Forests: Ecosystems and Economies on the Edge.
www.wri.org/wri/ffi/ .

WWF, 1992. Forests in trouble: A Review of the Status of Temperate Forests Worldwide. World Wide Fund for Nature, Gland.

WWF/IUCN, 1996. Forests for Life. The WWF/IUCN forest policy book. WWF-UK.