Tree Conservation Information Service
| Guidelines
for the application of the 1994 IUCN Red List Categories to
trees
IUCN categories of threat have been in use for over thirty years and are widely recognised as a standard for indicating the conservation status of individual species. The 1994 revision represents the first step to make the categories more quantitative, objective and equable over all taxa (except for microorganisms). As from now, a category can only be assigned to a species if one or more of a choice of five criteria apply. The guidelines are intended to provide practical advice and examples of the application of the categories specifically to tree species. Although clear criteria must now be followed to apply a category there is still scope to use a considerable amount of inference and assumption. This is an essential part of the new system as it allows the categorisation of species which are clearly threatened but differ in biology or in the degree to which they have been studied. It is the aim, however, that at least within large groups, such as trees, some standardisation is obtained through the setting down of guidelines. This document gives an idea of useful information sources for applying the categories, highlights areas where assessors frequently fall down and gives advice on how to interpret some of the definitions in the red pamphlet and the criteria to which they apply. Examples from The World List of Threatened Trees are supplied in shaded boxes. A key to the application of categories and criteria and an excerpt from a paper by W. Hawthorne concerning the mass application of categories to species of wet and moist forest types in Ghana are supplied as annexes. Information sources for evaluating tree species In order to qualify for a category of threat, evidence is needed to demonstrate that a species is experiencing, to various degrees, one of the following: Criterion
A. Population is seriously declining or is expected to decline at a specified
rate. Information on species conservation status and threats, although more commonly available than in the past, remains scarce and patchy for many countries. Taxonomists, or their published floras and taxonomic revisions, provide essential baseline data on taxonomy, nomenclature, species distribution, usually habitat type and sometimes additional information. In many cases this is the sole source of information on rare and restricted-range species. It is possible, by examining the geographical and altitudinal ranges, the location and habitat of species, to make preliminary evaluations of their status. The numbers of collections in herbaria and the date they were made can help to indicate species rarity. Species which are known from fewer than five specified locations can technically qualify for a threat category, VU D2. Relevant data may also be obtained from forest inventorying and mapping. Rarely do restricted-range and uncommon species appear in these datasets but these data can indirectly indicate population declines through habitat loss. An excerpt from Will Hawthorne's report in annex 2 gives an example of this type of evaluation. GIS (Geographical Information System) is a powerful tool to store and analyse species distribution information in a standard way. Where species' point localities can be stored in a GIS, data can be very successfully manipulated to assess whether their AOO (Area of Occupancy) or EOO (Extent of Occurrence) are within the limits set by the B criterion and also by the criterion for VU D2. David DuPuy and Jonathan Hughes have evaluated a subset of Madagascan Legumes in this way. Important points which are frequently ignored: . The categories apply only at a global level. . The categories can be applied to any taxonomic level, including infra-specific taxa, microspecies and natural hybrids. . It is not expected that all the criteria will be appropriate for any one species. Different criteria apply to different species according to the biology and habit of the taxon. . Exact population figures or species ranges are not essential to apply categories. The situation known in a major part of range may be extrapolated to give the overall status of the species, if it is unknown elsewhere. Population declines can be inferred from rates of past, present or future habitat loss. . It is essential that the criterion and subcriterion applied are recorded with the category of threat (e.g. VU A1c,d). Lower risk species should be recorded as LRcd, LRnt or LRlc. . The B criterion should only be used when two of the subcriteria are fulfilled (e.g. B1+2c, B1+2a) . A species known from fewer than five localities can qualify as VU D2. . Species in protected areas should receive a threat category, if the criteria are fulfilled. Only when the species is prevented from becoming threatened by protection measures should a category of LRcd be reached. The definitions interpreted for tree species 1. Population ("..the total number of mature individuals"). These estimates should include only mature individuals. For most non-pioneer species population estimates should be a fraction of the total number of individuals. 2. Subpopulation ("..distinct groups in the population between which there is little exchange...one successful migrant..per year"). In the case of tree species we may define a subpopulation as an isolated population which experiences insignificant seed or pollen migration from other populations. Frequently it is unknown to what extent species disperse and subpopulations are more commonly recognised by their geographical isolation. 3. Mature individuals ("..number of individuals known, estimated or inferred to be capable of reproduction"). The capability of reproduction in tree species varies widely and vaguely according to age/size class of individuals (e.g. Bailonella toxisperma first flowers at 50-70 years and doesn't fruit until roughly twenty years later, conversely Sequoiadendron giganteum may produce seed at less than 20 years of age and continue to do so for 3000 years). If little is known about age at fruiting, mature individuals should be counted as those of a typical size; e.g. canopy species should be canopy height etc. In addition, individuals should be reproductively effective. Individuals which are unable to regenerate or those hampered by their isolation or by unbalanced sex ratios may be discounted from population estimates. Clones of apomictic and vegetatively reproducing species count as individuals. 4. Generation ("..average age of parents in the population"). Although generation time should represent a greater value than the age of species at maturity, it is most likely no information is available to specify the age of parents. Alternatively the following estimations may be used: 50 years for most tree species, 10-20 for pioneer species or small trees, 100 years or more for slow-growing trees. 5. Continuing decline ("..a recent, current or projected future decline...liable to continue unless remedial measures are taken"). Declines should be regarded in the context of the species generation time. If dramatic declines in the past have since halted for the duration of a generation then it may not seem appropriate to say the species is suffering from continuing declines, unless, of course, there are strong prospects that further declines will take place in the next generation. The presence of short term measures or recent stability in population numbers should not necessarily cause the species to be considered as Lower Risk. 6. Extreme fluctuations ("..in a number of taxa where population size or distribution area varies widely, rapidly and frequently, typically with a variation greater than one order of magnitude"). Fluctuations in non-pioneer tree species populations are rarely extreme or, at least, are hard to express as extreme within an appropriate time scale. Therefore in order to qualify a species as threatened according to the B criterion the population must generally be fragmented or in ten or fewer locations and also declining or likely to decline. 7. Severely fragmented ("..most individuals within a taxon are found in small and relatively isolated subpopulations"). Most tree species suffer from a fragmented distribution because their habitat is fragmented. The severity of this fragmentation may be judged by the presence of activities, such as farming, that make recolonisation impossible or by the size of the largest populations and whether they are ever likely to be wiped out. 8. Extent of occurrence (EOO) ("..the area contained within the shortest continuous imaginary boundary which...encompasses all the known, inferred or projected sites of present occurrence of a taxon, excluding cases of vagrancy"). In tropical regions, if the total population or the sum of disjunct subpopulations takes up the equivalent of approximately two degree square then this should qualify the species as having an EOO less than 20,000kmē. In more temperate latitudes around 2.5 degree squares can be occupied. Endangered and Critically endangered species should be assessed at a finer scale, perhaps on a 1:1,000,000 map using grid squares of 50km x 50km. Good examples of disjunct distributions are where the species is found in upland vegetation on separate mountains or where species occur on separate islands. More difficult to discern as a disjunct distribution is where a species occurs in a number of isolated forest patches. In the World List of Threatened Trees EOO is frequently estimated from national statistics on forest or vegetation cover or from the areas of national parks and reserves. 9. Area of occupancy (AOO) ("..smallest area essential at any stage to the survival of existing populations"). This is intended to be a more accurate figure for the area to which species are confined. The estimate should take into account the habitat requirements of the species, whether this is a specific canopy height, species composition or microclimatic condition. AOO should be measured on a scale roughly ten times finer than that used to measure EOO. For more endangered species grids of 5 X 5 km and maps of 1:100,000 would be useful. 10. Location ("..geographically or ecologically distinct area in which a single event...will soon affect all individuals"). The location of a species refers to what may frequently be a smaller geographical area than that taken up by a subpopulation. However given the limited data available on gene flow needed to define subpopulation, the concepts of both may inevitably converge and remain somewhat subjective. The criteria applied to tree species Criterion
A How to make estimates of population decline in the last 100+ years. Three generations of a tree species could span the largest part of a millennium. Population declines are rarely supported by direct evidence but are more commonly inferred from rates of habitat loss. The declines can technically have happened during any period within this time span, but it may be inappropriate to include species which are now largely stable but experienced marked declines in previous centuries. Amburana
acreana Should species which are widespread but have experienced a general decline of 20% because of habitat destruction be Vulnerable? Many tropical forest species have experienced a 20% decline in their natural habitat over the past 3 generations. Some of these species, however, do have a wide ecological tolerance (e.g. Baikaea plurijuga). The most likely candidates for this criterion are those which are strictly confined to declining habitat types and are not able to regenerate outside of them, or species which are specifically and heavily exploited. It may be appropriate to apply a ceiling on the acceptable extent of the range of those species to be included. Amanoa
strobilacea Where species are regenerating poorly. If regeneration is known to be poor then the A criterion could be applied by using projected decline. Will individuals replace themselves in the next three generations? If the species is long living or its regeneration depends on stochastic events it will be hard to project a decline. Alternatively, criteria C or D may apply. Adansonia
grandidieri Where species have experienced a genetic decline. In many cases tree species have suffered genetic decline or loss of fitness in the population where the larger and fitter individuals have been harvested. If this is a significant proportion (20% or more) of the mature population then the taxon can be registered as threatened under criterion A. Entandrophragma
cylindricum
The listing of some widespread species under this criterion might seem out of place alongside other species which are reduced to highly restricted areas or a small number of individuals. However, it is very important to be clear that rates of extinction scale with a species' generation time, and not with absolute time. A species with a long generation time, declining at 50% per generation will exhibit almost no detectable decline over a one-year period. Criterion
B Cynometra
filifera Nectandra
herrerae Annona
praetermissa Criteria
C & D Where species are known only from the type locality or fewer than five locations. A species can technically be assigned a category of Vulnerable D2 on this information alone. If there are doubts as to whether the population still exist because of changes or loss of its habitat, a more serious threat category should be assigned by assuming a decline and that the extent of occurrence is small, EN B1+2c, CR B1+2c, or the population size is less than 10,000 mature individuals, EN C1, C2a or b or CR C1, C2a or b. The assessor may choose to assume that the species is more widespread, given that the area is very poorly known and the habitat still intact, and assign a category of DD. It is important that these poorly collected species are highlighted. Aspidosperma
darienense Baphia
pauloi Neoharmsia
madagascariensis Neuburgia
tubiflora Where species are not regenerating. It is embedded in the definition of mature individuals that individuals should be effective at reproducing. Individuals which produce offspring which are not viable should not be counted in the population estimate. Therefore population estimates may be made by taking reasonable estimates of the proportion of the population which are reproducing effectively, the population density and the area of occupancy. Where species are sparsely distributed. Where population density is low or populations do not occur in easily recognisable clumps, from which AOO can be estimated, the only way to denote the species as threatened is through estimating population declines (criterion A) or population sizes (criteria C or D). Extinct
in the wild Byttneria
ivorensis Chrysophyllum
januariense Lower Risk: conservation dependent (LRcd) When should presence in protected areas be used as a means for applying LRcd? In some cases, if protected area boundaries or protection measures were to be taken away the tree species inside would rapidly be wiped out and, where the ensuing declines in species population are significant, the species would register as threatened. If the current status of the species is not threatened, LRcd is the appropriate category to use in these cases. The effectiveness of protected areas in preventing declines in species populations is sometimes questionable or unknown. In these cases, species should be allocated a category of LRnt. All species which can satisfy the criteria for a threat category should be assigned one whether protected or not. Wodyetia
bifurcata Lower Risk:
near threatened (LRnt) Agathis
vitiensis Cola
lizae Commiphora
hodai Lower Risk:
least concern (LRlc) Data Deficient
(DD) Solanum
betaceum Conostegia
grisebachii Annex 1 Key to the application of red list categories to tree species 1
2
3
4
5
6
7
8
9
Annex 2 An excerpt from "Conservation priorities amongst Ghana's forest trees. An assessment of the IUCN red-list guidelines, and a comparison with Ghana's own star-rating Population numbers are next to impossible to estimate for continental tree species in the tropics. Ghana has carried out one of the most extensive surveys of its forests in Africa and yet it is impossible to collect sufficient details about rare species to validate a population estimate. The approach taken by William Hawthorne in assessing the status of tree species in Ghana involves estimating the likely EOO from the distribution of the species habitat and assuming its rate of destruction in the last few decades from published figures. The following is an extract from the working document prepared by William Hawthorne for the project's technical conference in Wageningen in November 1995. Forest
Types are those defined by Hall and Swaine (1981): Wet forest
species: General trends The greatest number of globally rare species in Ghana are wetter evergreen forest species, apparently restricted to these forests by a wetter climate/base-poor soil there, although there is also much speculation about their status as 'refugees' in or near Pleistocene refugia there (see summary discussion and references in Hawthorne, 1995). A sketch of enclaves of Wetter (evergreen) Guinean forest is shown in Parren and de Graaf, 1995. A species restricted to this vegetation in Ghana has a maximum EOO of about 2000 kmē, but the internal variation is substantial and no rare species are found throughout. Many such species favour swamps or riverside forest, for instance. Much of this vegetation type in Ivory Coast is deforested. The total area in the Ghana and Ivory Coast Block is probably < 3000 kmē. A species restricted to this area would qualify as at least endangered (B1,B2), without further consideration. Most species occur outside this block, however, either in the Liberian block of WE forest or in the ME zone in Ghana and elsewhere. In Liberia and Sierra Leone, the remnant WE type zone is larger, possibly about 15000-20000 kmē (from maps in Parren and de Graaf, 1995). Some species are restricted to the WE zone, others extend various degrees into the drier ME zone. In Ghana the area of ME zone forest is about 2.5 times the WE area, but is more widely disturbed. Assuming the same trend for the Liberian block, we can estimate an effective ME area of the order of perhaps 30,000 kmē. Combined with the WE area, this sets a maximum area on Upper Guinean wetter-forest species. White (1983) includes both WE and ME type forest in his mapping unit 1a ("hygrophilous coastal evergreen Guineo-Congolian rain forest"). The extent of this mapping unit in Lower Guinea is about twice the extent of that of Upper Guinea. Most Guinea-wide species (i.e. species in both Upper and Lower Guinea) have been filtered out from consideration; those that remain for consideration here are those like Afrostyrax lepidophyllus which seem restricted within these zones, so these total areas of Guinean Wet forest are much broader than the EOO (esp.AOO) of these species Although the area of this type of forest is limited, it seems that forest loss has not been as bad in such areas (especially the WE type) as in drier areas, because of negligible loss due to fire, lower densities of timber than in semideciduous forests and rather infertile (heavily leached) soils, although mining in the ME belt of Ghana has accounted for significant losses e.g. around Tarkwa. Cocoa farming has not been as great a factor in the WE zone as it as in the ME zone. Mining is a potential threat for deforestation in Ghana, with recent quarrying in Neung and Cape Three Points, and there has been significant forest loss on unreserved land in the WE zone (and in Western Region generally) since Ahn's (1959) summary of land-use patterns. Deforestation of Subri Forest Reserve and elsewhere due to industrial plantation is, being moderately optimistic, unlikely to expand much further. Wet Forest
species: Specific cases Wet forest species: tentative conclusions: VULNERABLE (VU-B1&2c & A1c) Tapura
ivorensis ENDANGERED (B1&2c) Chrysophyllum
azaguieanum (The following species are perhaps of less concern than the above; LRcd "Conservation dependent" (in Forest reserves), might be more appropriate. Crudia
gabonensis Probably well-buffered in lower Guinea. References:
Hall, J.B. and Swaine, M.D., 1981. Distribution and ecology of vascular plants in a tropical rain forest. Forest vegetation in Ghana. Geobotany 1. Junk, The Hague. 383pp. Hawthorne, W.D., 1995. Holes and the sums of parts in Ghanaian forest:regneration, scale and sustainable use. Proc. Roy. Soc. Edinburgh 104b Parren, M.P.E. and de Graaf, N.R. de, 1995. The quest for natural forest management in Ghana, Côte d'Ivoire and Liberia. Tropenbos series 13. Wageningen, The Netherlands. |
For help with this web site contact:
|
UNEP World Conservation
Monitoring Centre |
Information Enquiries |
| Document URL: http://www.unep-wcmc.org
/trees/Background/iucn_cats.htm
|
|
