Throughout this publication reference is made to the IUCN Red List Categories and Criteria (IUCN 1994a, 2001). These are intended to be an easily and widely understood system for identifying and classifying species at high risk of global extinction. The general aim of the system is to provide an explicit, objective framework for the classification of the broadest range of species according to their extinction risk. It is important to note that although the Red List system may focus attention on those taxa at highest risk, it is not intended to be the sole means of identifying and setting priorities for conservation action.
Summaries of the Red List Categories (Table A2a.1) and Criteria (Table A2a.2) are presented below. However, readers are referred to the full version of the system available at http://www.iucnredlist.org/info/categories_criteria2001.html . PDF versions in English, French and Spanish can also be downloaded from http://www.iucn.org/themes/ssc/redlists/RLcats2001booklet.html.
Table A2a.1 IUCN Red List Categories (IUCN 2001) Threatened species are listed on one of the three categories printed in RED.
Table A2a.2 Summary of the five criteria (A-E) used to evaluate if a species belongs in a threatened category (Critically Endangered, Endangered or Vulnerable)
During the development of the Red List Categories and Criteria and the subsequent Criteria Review process, a number of difficult issues were encountered that were not fully resolved. As solutions arise, rather than constantly modifying the Red List Criteria, a set of ‘User Guidelines’ have been developed that provide advice on how to deal with some of these issues, and how to apply the criteria under particular circumstances (in other words, they are best practice guidelines). These ‘User Guidelines’ are in effect a living document that is periodically updated; the latest PDF version of which can be downloaded from http://www.iucn.org/themes/ssc/redlists/RedListGuidelines.pdf .
The numbers of described species presented in Table 2.1 are derived from several different sources. There is considerable debate over some of the figures, and those for the invertebrates and the plants have a large degree of uncertainty associated with them (e.g., see Hammond 1992, 1995).
Mammals – From Wilson and Reeder (in press), with deviations based on input from the IUCN/SSC Specialist Groups.
Birds – Provided by BirdLife International from their World Bird Database (see BirdLife International 2004a, 2004b).
Amphibians – Provided by the Global Amphibian Assessment based on Amphibian Species of the World (Frost 2004).
Reptiles – Based on EMBL Reptile Database (Uetz 2004).
Fishes – Based on FishBase (Froese and Pauly 2003).
Invertebrates – From Groombridge and Jenkins (2002), but in turn largely based on Hammond (1992, 1995).
Mosses – Based on Hallingbäck and Hodgetts (2000).
Ferns and allies – Based on Groombridge and Jenkins (2002).
Gyymnosperms – Based on Donaldson (2003); Farjon (2001); and Mabberley (1997).
Dicotyledons and Monocotyledons – Based on Thorne (2002), but see Mabberley (1997); Schmid (1998); Govaerts (2001, 2003); Bramwell (2002); and Scotland and Wortley (2003) for alternative views on the numbers of seed plant species.
The numbers of species evaluated (as shown on Table 2.1) are obtained from the entries in the Red List database (part of the Species Information Service) held at the Red List Office in Cambridge, UK. The origins of this database date back to the production of the 1996 IUCN Red List of Threatened Animals (Baillie and Groombridge 1996) when most of the animal assessments were incorporated in the database. However, a large number of Least Concern assessments were not captured at the time (for example, most of the Least Concern mammals, turtles, butterflies, etc.). The Red List team have been trying to capture the ‘missing’ Least Concern assessments, in between processing new assessments for each annual update. But unfortunately, as a complete record of what was assessed was not kept, this is difficult to do in certain cases, as the taxonomy for some groups has changed considerably since 1996. The task has been completed for the mammals, and all amphibians and birds were completely assessed for the 2004 update of the Red List. However, for all the other taxonomic groups, the numbers evaluated are an under-estimate because the Least Concern assessments have not yet been recorded. To exacerbate matters, for some groups, particularly the plants, Least Concern assessments are seldom submitted to the Red List office. This will remain a work in progress until such time as each taxonomic group is completely assessed.
The IUCN Red List includes assessments for undescribed species under certain special circumstances (namely, that taxonomists are agreed that these are in fact new undescribed species, that their distributions are known, that voucher specimens are deposited in Museums or Herbaria, and most importantly that there is a direct conservation benefit to the listing). There are 143 undescribed species on the 2004 IUCN Red List (83 of which are threatened) and these are all effectively treated in this analysis as ‘described species’.
The percentage of threatened species was calculated for each order and family of mammals, birds and amphibians, and compared with the respective average across all species in each of those classes. The significance (or lack thereof) of the difference between the percentage of threat observed in an order or family in relation to the average value in the respective class was assessed using a binomial one-tailed test ( p given by the fraction of threatened species in the class). Binomial one-tailed tests were also used to produce the bands of significance level in Figures 2.3 to 2.5.
IUCN has not yet agreed criteria for determining whether or not Critically Endangered species are ‘Possibly Extinct’. The information and analyses in Section 3 for Possibly Extinct birds and amphibians are based on somewhat different approaches.
BirdLife International has developed a framework for classifying Critically Endangered species as ‘Possibly Extinct’. This framework has been tested on the 50 or so bird species to which this tag might conceivably apply i.e., those that have not been recorded for a long time, or those whose dwindling populations may have finally disappeared. The framework allows the user to assign such species to one of the following three categories:
CR: species which are likely to be extant, and for which any lack of records probably does not reflect a genuine extinction.
CR(PE): species which are likely to be extinct, but for which there is a small chance that they may still be extant, hence they should not be listed as EX until local or unconfirmed reports have been discounted, and adequate surveys have failed to find the species.
EX: species for which there is no reasonable doubt that the last individual has died (and for which it is not advocated that conservation effort is spent searching for these species).
To determine which category to assign a threatened species, scientists at BirdLife International examined: (1) evidence pertaining to the timing of the last confirmed records; (2) any subsequent unconfirmed records or local reports; (3) the strength of threatening processes currently and historically operating; (4) the adequacy of fieldwork relative to the (presumed) ease of detection of the species; and (5) the extent and quality of remaining suitable habitat (where ‘suitable’ incorporates the absence of introduced predators, pathogens, etc.).
BirdLife International classified species as Possibly Extinct if, on balance, the evidence that they may be extinct outweighs any evidence that they may be still extant (although the latter remains a slim possibility, so they are not yet classified as Extinct).
Such evidence for extinction may include a combination of the following factors:
There have been no confirmed records for a long time (it is difficult to be more prescriptive: the duration will depend on the intensity of fieldwork and the ease of detection).
For species with recent last records, the decline has been well documented.
There are severe threatening processes operating (e.g., extensive habitat loss, introduction of alien predators, intensive hunting).
The species has attributes known to predispose it to extinction, e.g., it was probably naturally rare and/or had a tiny range (as evidenced by paucity of specimens relative to collecting effort), or flightless, etc. In some cases, allospecies or congeners may have gone extinct through similar threatening processes.
Surveys would have detected it (good/recent surveys have been adequate; species is unlikely to be overlooked).
Evidence that the species may remain extant may include a combination of the following factors:
The lack of records is best explained by inadequate fieldwork (any surveys have been insufficiently intensive/extensive, or inappropriately timed; or the species' range is inaccessible, remote, unsafe or inadequately known).
The lack of records is best explained by the fact that the species is difficult to detect (low density, cryptic, inconspicuous, nocturnal, nomadic, silent or call unknown, identification difficult).
There have been reasonably convincing local reports or unconfirmed sightings.
Suitable habitat (free of introduced predators and pathogens if relevant) remains within the species' known range. In some cases, allospecies or congeners may survive despite similar threatening processes.
The balance of the evidence for and against extinction then allows the species to be placed on a continuum from high to low confidence of extinction, on a spectrum from EX to CR(PE) to CR. The position of a species on this continuum will be influenced by the time since the last confirmed record. For species with recently confirmed records to be placed at the EX end of the spectrum, there needs to be greater confidence in the extinction (i.e., greater confidence in the adequacy of surveys, the absence or inadequacy of local/unconfirmed records, the greater understanding and severity of threatening processes, and the greater documentation of, and confidence in, observed population declines). Deciding the strength of evidence for and against extinction is necessarily subjective. However, this framework helps to make these judgements as objective as possible, by obliging the user to set out the evidence, and to weigh this against the time since the last confirmed record.
This approach to defining ‘Possibly Extinct’ species identifies those species for which extinction is likely. For amphibians, the phenomenon of rapid population declines is a new one, and the extent to which it tends to result in complete extinction is still not clear. This lack of information makes it difficult to judge the likelihood that extinction has occurred. The Global Amphibian Assessment adopted a more precautionary approach in identifying Possibly Extinct amphibians. The term was applied to all CR species that can no longer be found, and which appear to have declined dramatically, but which do not qualify for EX because of lack of survey effort. Hence the large number of CR(PE) amphibians includes species that are conceivably extinct as well as those that are likely to be extinct.
The method for calculating Red List Indices (RLIs) has been developed through the Red List Programme and is published in detail in Butchart et al. (in press a and b). RLIs are calculated from the number of species in each IUCN Red List Category in each assessment, and the number of species changing categories as a result of genuine improvement or deterioration in status. Specifically: (1) For species assessed in two consecutive assessments, the total number of species in each Red List Category in the earlier assessment (excluding Data Deficient, Extinct and Possibly Extinct) are multiplied by a category weight (see below), and these are summed to give a total score for the assessment. (2) Over the time period between assessments the net number of genuine changes (losses and gains) in each category is calculated, multiplied by the category weight and summed to give the % change in the total score. (3) The value of the index is set to 100 in the first year of assessment (i.e., 1988 for birds, 1980 for amphibians), and calculated for subsequent assessments by multiplying the previous index value by the % change in the score for the previous time period.
There are a number of potential ways to assign weights to IUCN Red List Categories. We examined two: an ‘equal steps’ approach and an ‘extinction risk’ approach. In the equal steps approach the weights range from 0 for Least Concern, 1 for Near Threatened, 2 for Vulnerable, etc. to reflect the ordinal ranks of the categories, whereby each step from Least Concern towards Extinct indicates that at least one measure of extinction risk has become worse. This approach has the advantage of being simple, and the trends in the resulting index are driven by a relatively large number of species (which produces a more robust and representative index). This is because a species moving from Least Concern to Near Threatened contributes just as much to the changing score as a Critically Endangered species going Extinct, and the numbers of species in each category (and moving in and out of each category) increases disproportionately from Critically Endangered to Least Concern. The main disadvantage is that the weights merely reflect the linear hierarchy of categories. However, the steps between lower categories (e.g., Near Threatened to Vulnerable) translate to smaller increases in extinction risk than steps between higher categories (e.g., Endangered to Critically Endangered). In recognition of this, in the extinction risk approach the weights are based on the relative extinction risk associated with each category, ranging from 0.0005 for Near Threatened and 0.005 for Vulnerable to 1.0 for Extinct.
The most important difference between the equal steps versus the extinction risk approaches is the effect of status changes in less threatened or non-threatened species. With the equal steps approach, the index is heavily influenced by movements of species among the lower categories, such as Near Threatened and Vulnerable. With the extinction risk approach, movements of species in and out of Critically Endangered largely influence the RLI. For example, if a Vulnerable species improves in status and becomes Near Threatened, and at the same time, a Critically Endangered species goes extinct, the RLI based on equal steps weights registers no change, but the index based on extinction risk weights shows a substantial decrease. Downlisting of a Vulnerable species to Near Threatened might represent a very substantial population increase, whereas Extinction of a Critically Endangered species might represent the loss of very few individuals. The latter is arguably more significant in terms of genetic diversity, but the former might be more important as an indicator of wider biodiversity trends. Thus, the extinction risk weights emphasize the loss of biodiversity owing to imminent or potential extinctions of species, whereas the equal steps weights allow the index to capture large changes in the populations of less threatened species.
For the RLI for all species in a taxonomic group and for subsets of species (e.g., in particular realms or ecosystems), we used the equal steps approach. This was because for some disaggregated indices, there were few species in the higher threat categories (those effectively driving trends in the index weighted by extinction risk). For birds, for example, only 23% of all genuine status changes (58 species in total) involved moves in or out of the highest threat categories. However, for examining trends in the species closest to extinction, we used the extinction risk approach.
We calculated, using the following method, the possible range of error associated with the latest (2004) RLI value for birds owing to time-lags before genuine status changes are detected. We estimated how many such undetected category changes there may be for 2000–2004 using the 1994–2000 data (information gathering has improved considerably in recent years, so comparisons with time-lags for the 1988–1994 period are not meaningful). In total, 128 genuine changes for 1994–2000 were identified in 2000, and an additional 17 (13.3%) were identified in the subsequent four years. This suggests that an additional six category changes (13.3% of 45 genuine status changes identified in 2004) may be belatedly detected for 2000–2004. We randomly selected six species from the 9,453 species that did not undergo category changes from 2000 to 2004. We ran 10,000 simulations of six species moving to categories of higher extinction risk, with probabilities for each number of category steps set by the distribution of category changes for 35 species that were uplisted to higher categories of extinction risk in 2004. The maximum value for P (proportional genuine change) from these simulations gave the lower error bar for the 2004 RLI value. Similarly, we ran 10,000 simulations of six species moving to categories of lower extinction risk (with probabilities for each number of category steps set by the distribution of category changes for 10 species downlisted to lower categories in 2004), and took the minimum value for P to give the upper error bar (see Butchart et al. in press a for further details).
The data for birds are derived from four assessments (Collar and Andrew 1988, Collar et al. 1994, BirdLife International 2000, 2004a; also published in Baillie and Groombridge 1996, Hilton-Taylor 2000 and IUCN 2004). For amphibians, the 2004 data come from the Global Amphibian Assessment (IUCN, Conservation International and NatureServe 2004; IUCN 2004). An IUCN Red List Category for each amphibian species in 1980 was retrospectively assigned by considering the present category and information on the spread of disease, habitat degradation and loss, the introduction of alien species and knowledge of population trends. A conservative approach was adopted, and category changes were only recorded as having taken place when the evidence was considered to be strong. In cases of significant uncertainty, it was assumed that no change in category had occurred. Given the uncertainty over these retrospectively assigned categories, the RLI for amphibians is represented with a dotted line in Figures 4.6–4.10.
Extent of occurrence is defined as the area contained within the shortest continuous imaginary boundary that can be drawn to encompass all the known, inferred or projected sites of present occurrence of a taxon, excluding cases of vagrancy. This measure may exclude discontinuities or disjunctions within the overall distributions of taxa (e.g., large areas of obviously unsuitable habitat). Extent of occurrence can often be measured by a minimum convex polygon (the smallest polygon in which no internal angle exceeds 180 degrees and which contains all the sites of occurrence) (taken from IUCN 2001, p. 11).
The term ‘endemic species’ always requires a qualifier as to where the species is endemic to (e.g., a species endemic to Angola is one that occurs in Angola and nowhere else). As such, endemic species can be very widespread (e.g., a species endemic to the Western Hemisphere may occur from northern Canada to southern Argentina) or very restricted in range (e.g., the range of a terrestrial species endemic to São Tomé and Principe is no larger than 1,200 sq. km).
The occurrence of species in biogeographic realms was determined by overlaying species by species Extent of occurrence (EOO) (see above; IUCN 2001) maps onto the biogeographic realm classification of Olson et al. (2001), except for data on total bird species richness, which were provided by J. Lamoreux (unpublished data).
The occurrence of species in biomes was determined by overlaying species by species EOO polygons onto the biome classification of Olson et al. (2001). We considered a species as occurring in a biome if >5% of its EOO overlapped that biome (to minimize commission errors of including species within biomes where they do not occur) or if the species EOO occupied >5% of the biome (whichever is smaller) and considered a species endemic to a biome if >95% of its EOO overlapped that biome (to minimize omission errors in detecting biome-endemic species). Data on Old World bird species richness and endemism were provided by J. Lamoreux (unpublished data).
In some maps (Figures 5.4, 5.9 and 5.10), the mammal data exclude marine species (pinnipeds, cetaceans, sirenians and a few others). Bird polygons include only breeding ranges of species (but including marine species). Turtles include only freshwater species. Only mammals, birds and amphibians have been globally comprehensively assessed, but turtles are relatively well assessed compared to many other taxonomic groups and are included here for comparative purposes. In all cases, introduced and extinct parts of the range are excluded. Maps are based only on current range, not historical ranges. The richness maps are shown at a halfdegree cell, which admittedly does not take into account the species-area relationship.
Standardization is required in order to compare the levels of endemism of different regions. Centres of endemism have traditionally been identified through the overlap of restricted-range species, found using threshold approaches which consider only species with distributions smaller than a given percentile or area (Hall and Moreau 1962). This technique was the first to be applied across an entire higher taxon (Stattersfield et al. 1998). While it faces the problem in that the choice of a given threshold is arbitrary (Peterson and Watson 1998), the choice of weighting for alternative methods, such as summing the reciprocals of the range sizes for all species within a given area, is also arbitrary (Williams 1998).
A locality (Section 5.4) is synonymous with site, and can broadly be defined as having a definable boundary within which the character of habitats, biological communities, and/or management issues have more in common with each other than they do with those in adjacent areas.
In recent years, the threats faced by species that contribute to their IUCN Red List status have been documented by coding the relevant factors according to the IUCN Major Threats Authority File (see http://www.iucnredlist.org/info/major_threats.html, or download the file from http://www.iucn.org/themes/ssc/sis/authority.htm). Threats are coded to provide different levels of details if this information is available: for example, the level 1 coding of “Habitat destruction and degradation” may be further coded to indicate the driving force leading to such destruction, e.g. “Agriculture”, the type of agriculture, e.g. “Crops” and the scale of this threat, e.g. “Agro-industry farming”. Thus, the process of coding the threats for each species is complex, and it is far from complete, so constraining the analyses that can be currently performed. In the 2004 IUCN Red List, all threatened and Near Threatened bird species and all amphibian species are fully coded for their threats, and most threatened mammal species have also been coded. Thus, in Section 6 “The Many Causes of Threat”, it was only possible to perform meaningful threat analyses for threatened birds, mammals and amphibians.
For birds, the timing, scope and severity of the threats is also recorded and these are combined together to provide an indication of the impact of the threat (High, Medium, Low, No/negligible, Past; see below for more details), but this distinction has not yet been made for mammals and amphibians. For comparative purposes, all level 1 threats have been analysed covering all threatened (CR, EN and VU) species and shown according to the % affected. For amphibians, all threats recorded were included, while for birds only those threats identified as having a High, Medium or Low impact were used. Level 1 threats were grouped according to the hierarchy of the Threats Authority File apart from “Disease” which has been shown separately covering both native and alien invasive diseases because of the importance of this threat to amphibian species and because of the difficulty of distinguishing between these two types. In some analyses, all invasive species have also been lumped covering those that affect species directly (e.g., predators which eat the species) and those that affect species indirectly (e.g., herbivores which cause habitat loss).
For each threat, BirdLife assigns the timing (i.e., past, continuing or future), the scope (i.e., the proportion of the total population affected) and the severity (the overall declines caused by the threat) an “impact score” (0–3). The overall impact of the threat is determined by adding these separate impact scores (see table below).
Table A2f.1 Summary of the BirdLife Method for Scoring Threats
Thus all threats are recorded without any initial (unrecorded) pre-judgement, but only those with a High/Medium/Low/Unknown impact are included as relevant in relation to the current IUCN Red List status (i.e., excluding those with Negligible impact or Past impact where the threat is unlikely to return).
The overall impact of the threat is calculated by adding the individual scores for Timing , Scope and Severity as follows:
| High impact | score | 8,9 |
| Medium impact | score | 6,7 |
| Low impact | score | 3,4,5 |
| No/negligible impact | score | 0,1,2 |
| Past impact | special case 1, i.e. if timing = “Only in the past (and unlikely to return)” regardless of impact scores for Scope and Severity | |
| Unknown | special case 2, i.e. if “unknown ” for Timing, Scope or Severity | |
LandScan's human population density modelled for 2002 was used as the measure of current human population density. This dataset gives population density modelled at 1km2 resolution, which was then averaged to give a value within the same quarter-degree grid cell as the species data so that the variables could be compared. The data was also split into three categories ( Low : 0 to 10 people per 1km2, Moderate : 10 to 100 people per 1km2, High : 100 or more people per 1km2). The number of threatened species was calculated by totalling the number of threatened mammals, birds and amphibians whose distributions overlap each quarter-degree grid cell across the world. This total was then divided into three categories ( Low : No threatened species, Moderate : 1 to 10 species, High : 11 or more species).
Population growth was divided into two categories ( Low : less than 1.5% and High : greater than 1.5%). The total number of threatened mammals, birds and amphibians was used as the measure of the threatened status of species in each country. This value is obviously biased towards countries with large land areas and high species diversity. A simple correction for land area however, is not appropriate as the number of species per unit area varies with latitude (see Figure 5.2.1 in Section 5). An alternative measure is the number of species that are threatened compared to the overall diversity in a country (see Section 5). This measure enables a comparison between countries, but does not give an indication of the proportion of the global total of threatened species that each country has responsibility for conserving. Hence for this analysis the total number of threatened mammals, birds and amphibians was used to identify those countries which have a greater share of the responsibility for conserving the world's threatened species, regardless of their land area or total species diversity. This total was divided into four categories ( Low : 0 to 10 species, Moderately-Low : 11 to 30 species, Moderately-High : 31 to 100 species, High : more than 100 species).
In this analysis the Gross National Income (GNI) per capita for each country in 2003 was divided in to two categories ( Low income : $3,035 or less, and High income : $3,036 or more) using pre-defined criteria (following the World Bank Atlas method, World Bank 2004). All those countries for which GNI data were available were used in the analysis. The threatened species data used were the same as for the analysis of population growth shown in Figure 7.2, divided in to the same four categories.
For a detailed description of the methods and data in Table 8.2, see Rodrigues et al. (2004). Data as in Rodrigues et al. (2004) for turtles (Iverson et al. 2003) and mammals (compiled by the IUCN/SSC Global Mammal Assessment), but was updated for birds (corresponding now to the State of the World's Birds 2004 assessment; BirdLife International 2004a) and for amphibians (corresponding to the finalized data obtained from the Global Amphibian Assessment; IUCN, Conservation International and NatureServe 2004). Protected area data was also updated and corresponds now to the 2004 World Database on Protected Areas (WDPA Consortium 2004).
In Table 8.3, species of unknown/unset/fluctuating trends ( n = 1,571 for amphibians; n = 84 for threatened birds) were excluded, while increasing species ( n = 28 for amphibians, n = 39 for threatened birds) were lumped with stable. In all cases, significantly higher percentages of gap species were found to be decreasing than would be expected by chance ( p < 0.001, obtained through bootstrap, 1,000 replicates), while the tendency is for non-gap species to have higher percentages of stable or increasing species ( p < 0.001).