IUCN Red List Categories

The IUCN Red List of Threatened ™

1 The Red List Categories IUCN Extinct (EX) + Possibly Extinct Categories (EW) CR(PE) or CR(PEW)

Threatened categories (CR) Adequate data Endangered (EN) Risk Vulnerable (VU)

Evaluated Near Threatened (NT) Least Concern (LC) _

All species (DD)

Not Evaluated (NE)

Can anyone tell me how many IUCN Red List Categories there are?

[TRAINER’S NOTE: let the participants put forward suggestions before revealing the answer]

1. There are nine IUCN Red List Categories. All of the world’s species (except micro-organisms, as I mentioned earlier in the workshop) can be placed into one of these categories:

2. [CLICK] The majority of species currently sit in the Not Evaluated (NE) category; these are not listed on the IUCN Red List website (we don’t have the resources to maintain a full list of all the world’s species in our database). Not Evaluated means that no Red List assessment has been attempted for these species, often because there are insufficient funds or experts available to attempt to even find out if there enough data exist to be able to evaluate its extinction risk. We have no idea whether these species are at a high or a low risk of extinction, or if they are already extinct.

3. [CLICK] All of the species that have had available data compiled for them and have had their extinction risk evaluated will fall into one of the other eight categories. [CLICK] If there are insufficient data available to be able to determine an appropriate Red List Category that reflects the species’ extinction risk, then the Data Deficient (DD) category is used. There may be no information available at all, or lots of the wrong type of information (i.e., no information on population trends, or population size, or range area, or threats), or there may be major taxonomic debates around the species making it impossible to say what the population or range is.

Both the Not Evaluated and Data Deficient categories are very important because they both mean that we don’t know whether the species has a high or low risk of extinction, or if they are even already be extinct. These species are often overlooked by policymakers, so we should all make an effort to promote these species as priorities for future red listing efforts where this is realistically achievable.

4. [CLICK] The remaining seven categories are used for species that have sufficient information available for them; each of these threat categories indicates the species’ relative extinction risk. [CLICK] Least Concern (LC) species have the lowest current risk of becoming extinct. This is usually used for widespread and abundant species, but it may also be used if: • The species has a very restricted range but there are no current or potential threats that realistically could rapidly cause the species to become extinct (e.g., small island endemics with no threats). Or; • Widespread species that are undergoing very slow population declines. In these cases, although current extinction risk is relatively low, some conservation measures may be warranted to prevent them moving into a higher threat category in future.

5. [CLICK] Species that have a higher risk of extinction but are not yet actually threatened are placed in the Near Threatened (NT) category. Usually, these species almost meet the thresholds or requirements for a threatened category, and very little additional pressure is likely to rapidly push them into a threatened category. However, there is another scenario where a species may be listed as Near Threatened: if the species is highly dependent on conservation actions, without which its status would very quickly deteriorate to qualify it for a threatened category, it can be listed as NT.

A classic example of this type of Near Threatened listing is the White Rhinoceros (Ceratotherium simum). This speices is

2 carefully protected throughout its range; without this protection it is extremely likely that poaching would rapidly result in the species becoming threatened.

6. [CLICK] Moving higher up the extinction risk scale, the three categories Vulnerable (VU), Endangered (EN), and Critically Endangered (CR) highlight those species at the highest risk of becoming extinct. [CLICK] Species listed in any of these categories are referred to as “threatened”. The Red List criteria are used to determine which threatened category a species should be listed in: we will cover the Red List Criteria in detail later.

[CLICK] The CR category also has a special flag (the Possibly Extinct flag) that can be attached to identify species that may already be Extinct or Extinct in the Wild, but more information is required to confirm this. Note that this is not a separate Red List category; it is a flag that is specific to the CR category only.

7. [CLICK] Extinct in the Wild (EW) highlights species that are now only known to exist in captivity (e.g., zoos, botanical gardens, etc.) or as populations introduced to areas outside of their natural range for non-conservation purposes. For this category there must be evidence that there have been many repeated surveys within the species’ known range and in suitable habitat and that the species has not been found.

8. [CLICK] When even all of the captive individuals have died, the species becomes Extinct (EX). The same rules for repeated surveys for EW applies to the EX category.

[CLICK] As a species moves through the categories from Least Concern its extinction risk increases until it finally reaches the Extinct category. Unless, of course, we can implement conservation actions to prevent that from happening, which is the main purpose of the Red List.

2 Red List Categories

Changing Red List Category

There are various reasons for a species to change category:

• NON-GENUINE change • New information • Taxonomic changes • Incorrect data used previously • Criteria revision (version 2.3 (1994) versus version 3.1 (2001)) • Knowledge of the criteria

• GENUINE status change

IUCN tries to ensure that species are reassessed at least once every 10 years (resources permitting). Through this reassessment process, species can change Red List category.

There are various reasons why a species would change category: not all category changes reflect a genuine change in status.

[CLICK] Often the change is for non-genuine reasons, which include:

[CLICK] New information becoming available since the previous assessment, allowing the Red List status to be refined;

[CLICK] Taxonomic changes, such as several species being merged together into one species or one species being split into several separate species;

[CLICK] Sometimes the previous assessment used incorrect data (for example, the data may actually have referred to a different species) and the reassessment fixes the error;

[CLICK] For species with very old assessments (pre-2001) being reassessed now, they may change status because the Red List Criteria were revised between 1996 and 2001 and some of the criteria thresholds were slightly modified through that process. So the species may not meet the revised thresholds for a threatened category, but his doesn’t necessarily mean they have improved in status.

[CLICK] The assessors for the previous assessment may have misunderstood the IUCN Red

3 List Criteria and consequently miscalculated one or more of the parameters for their assessment, and the reassessment corrects this error. This scenario is a reminder of how important it is to understand the terms used in the Red List Criteria.

[CLICK] Other changes in Red List reflect a genuine change in status for the species. These species are very important to highlight in the Red List as they are genuine priorities for conservation action.

3 Red List Categories

The Five Year Rule

Red List LC Status Genuine deterioration in NT status: uplist to higher VU threat category immediately EN

CR

Time

For genuine changes, there are certain rules outlined in detail in the Red List Guidelines:

[CLICK] Where a species is genuinely deteriorating in status, it must be uplisted to a higher Red List Category immediately.

4 Genuine Improvements: The Five Year Rule

Downlist to lower threat category only Red List status when the higher category thresholds have not been met for FIVE years NT

VU

EN CR CR

Time (yrs)

First CR thresholds 5 years Reassess and assessment: no longer met Can reassess and update alter status CR documentation, but category must appropriately remain as for first assessment : CR

For species that are showing a genuine improvement in status, it is important to be certain that this improvement is going to continue before downlisting it to a less threatened category. Therefore a five year time lag is implemented in the Red List Categories and Criteria as a precautionary measure.

In the example shown on the screen, a species is initially assessed as CR. Two years later, the species shows a significant improvement and the CR thresholds are no longer met and it now actually qualifies for EN. However, although this species can be reassessed and the supporting information attached to the assessment can be updated, [CLICK] its Red List Category should remain as Critically Endangered for at least 5 years after the point where the species no longer qualifies for the higher category. The reason for keeping the species in the CR category is because one of two things may happen.

[ClLICK] Either the status of the species will continue to improve (shown by the green line in the figure), or

[CLICK] the observed improvement may be temporary and the population may begin to

5 deteriorate again (shown by the red line). For example, the “improvement” may just be a positive fluctuation in population size, or it may not be practical to maintain the implemented conservation measures long enough to ensure the species’ status continues to improve.

[CLICK] Five years after the species is first identified as not meeting the thresholds for the initial assessment (CR in the example), it can be reassessed again and at this point the Red List Category can be changed appropriately.

5 Data Quality & Uncertainty

The IUCN Red List of ™

6 Data quality & uncertainty

Dealing with a lack of high quality data

• The threatened categories use quantitative thresholds • BUT a lack of high quality data should not deter assessors from applying the IUCN criteria.

The Red List criteria can seem daunting to Assessors at first. Although quantitative thresholds are used to determine which threatened category a species can be listed under, the system also allows for uncertainty in the available data, and also for situations where there are very little direct data at all.

It is clear that if we were to rely only on lots of detailed research data that is 100% certain, very few species would ever be assessed beyond Data Deficient. So, a range of data quality is acceptable for use in an IUCN Red List assessment.

7 In the Red List Categories and Criteria you will see the following terms used:

• [CLICK] Observed • [CLICK] Estimated • [CLICK] Projected • [CLICK] Inferred • [CLICK] Suspected

These terms refer to the quality of the information used in the assessment, as defined in the Red List Categories and Criteria. The full definitions for each of these terms appears in the Red List Guidelines document.

8 Data quality & uncertainty

Observed

Observed information is directly based on well-documented observations of all known individuals in the population.

For example: entire global population occurs in only one area and all individuals counted each year

Year 3 Yearpopulation 4 population = 15 = 8 YearYear 2 population1 population = 17= 19 Observed reduction of 58% over 4 years

Observed data is the highest quality information we can obtain. It is directly based on well- documented observations of ALL known individuals in the population.

[CLICK] For example, if the entire global population of a species occurs in one small area and all mature individuals within the population are counted each year [CLICK, CLICK, CLICK] , then the Assessor would know for certain what the rate of decline for that species is.

In the example shown, an observed reduction of 58% over 4 years can be calculated based on data from annual population size counts.

9 Data quality & uncertainty

Estimated

Estimated information is based on calculations that may involve assumptions and/or interpolations in time (in the past).

For example: repeated surveys of sample sites across total range

Population size A Date Site A Site B Site C Site D All estimate across total range B C 2005 105 110 210 59 484 2,000 2006 101 107 70 40 318 1,300 2007 90 100 25 42 257 1,000 2008 63 81 0 33 177 700

D Sampling sites Estimated 65% reduction between 2005 and 2008

Estimated information is generally based on scientific studies.

[CLICK] For example, if regular surveys are carried out [CLICK] at specific sites within the range area of a species, the data gathered from these surveys can be used [CLICK] to estimate a rate of decline across the whole population. [CLICK] In the example shown, this method results in an estimated reduction of 65% between 2005 and 2008.

Estimated information is based on calculations that involve assumptions (e.g., statistical assumptions about sampling methods used, biological assumptions about the relationship between an observed variable (e.g., an index of abundance) to the variable of interest (e.g., number of mature individuals), etc).

All of the assumptions should be stated and justified in the documentation.

10 Data quality & uncertainty

Projected

Projected information is the same as “estimated”, but the variable of interest is extrapolated in time towards the future

For example: repeated surveys of sample sites across total range with knowledge of ongoing causes of population decline

Projected future decline Population based on habitat loss size A continuing at same rate B C as in the past

Estimated past decline based on D collected data

10 yrs now 10 yrs in ago future

Projected information is the same as “estimated”, but the variable of interest is extrapolated in time towards the future.

[CLICK] The example shown here is the same as the one on the previous slide. [CLICK] If additional information is also available on the causes of decline (e.g., habitat loss) and the likelihood of the threats continuing into the future (or even increasing in future), then a population decline model can be created and extended into the future.

[CLICK] In the model, past decline is estimated from the collected data.

[CLICK] Future decline is projected based on the threat continuing as it has in the past.

For projected information, the assessment should include a discussion of the method used (e.g., justification of statistical assumptions or the population model used) and justification for extrapolating current or potential threats into the future, including their rates of change.

11 Data quality & uncertainty

Inferred

Inferred information is based on variables that are indirectly related to the variable of interest, but in the same general type of units (e.g. number of individuals or area or number of subpopulations). Relies on more assumptions than estimated data.

For example: Past and current population sizes are not known, but trade figures for that species have declined over time.

Fresh Fish

Inferred continuing decline in population size based on decline in trade statistics for this species

Inferred information is one of the weaker data quality types. It is based on indirect evidence, or variables that are indirectly related to the variable of interest but that are in the same general type of unit.

[CLICK] For example, we might not have any information at all on past and current population sizes. But, if the species is utilised, there may be trade data that suggest the population has declined over recent years. In this example we are looking for information about the population trend; although there are no direct data on population sizes to give a certain answer about what is happening to the population, trade information does relate to numbers of individuals of that species.

[CLICK] The species may have been commonly sold in the past, but more recently it has become more difficult to find and buy. This may be because the wild populations are indeed declining and it is becoming more difficult to harvest. However, it may also be due to new harvesting techniques being used that target other species and not this one, or perhaps people’s tastes are changing and so supply is shifting to meet demand.

[CLICK] While we don’t have enough data to estimate a population decline, we could infer a continuing decline in the number of mature individuals based on the change in trade figures. We are assuming that the decline in the presence of the species in the market reflects an actual population decline, rather than a change in harvesting efforts targeting that species.

Inferred values rely on more assumptions than estimated values (e.g., inferring reduction from catch statistics requires statistical assumptions (e.g., random sampling), biological assumptions (e.g., relationship of the harvested section of the population to the total

12 population), and assumptions about trends in effort, efficiency, and spatial and temporal distribution of the harvest in relation to the population.

12 Data quality & uncertainty

Suspected

Suspected information is based on circumstantial evidence, or on variables in different types of units. In general, this can be based on any factor related to population abundance or distribution.

For example: Rate of habitat loss is known, but past and current population sizes are unknown.

Population size ??? • Suspected population reduction of e.g., >50% based on 75% of habitat being lost

Population size ?? • Can infer a continuing decline in habitat quality or size of AOO, but suspect a reduction in population size at a specific rate (%)

Suspected information is the weakest type of data quality. It is based on circumstantial evidence (e.g. surveys of local people living within the species’ range, or who utilise the species being assessed), or on variables in units different to the kind of variable we are measuring (e.g. % population reduction based on incidence of a disease).

[CLICK, CLICK, CLICK] In the example shown here the suspected rate of population reduction is based on the rate of habitat loss.

[CLICK] Note that evidence of qualitative habitat loss can be used to infer that there is continuing decline in habitat quality or AOO; but you need evidence of the amount of habitat lost to use for a suspected population reduction at a particular rate.

When translating habitat loss to rate of population reduction, you should have some information about how the species uses its habitat to justify the rate of decline you have used.

There are many more assumptions involved in suspected information and these should also be clearly outlined in the assessment documentation.

14 Data quality & uncertainty

Dealing with data uncertainty Uncertainty in the data itself (different to the lack of data) should also be considered in a Red List assessment

For example: A species has a range of population size estimates from 3 separate studies.

Study A: Population size = 100-200 (Endangered) Study B: Population size = 200-350 (Endangered or Vulnerable) Study C: Population size = 280-410 (Vulnerable)

Data uncertainty can come from:

• Natural variability as species’ life histories and environments change over time and space (e.g., spatial variation in age-at-maturity for marine turtles with a single estimate being required to best represent the naturally occurring range of variables). • Or from vagueness in the terms and definitions used in the criteria (semantic uncertainty leading to lack of consistency in different assessors’ usage of terms). • Or from measurement error (lack of precise information about the quantities used in the criteria).

[CLICK] When there is uncertainty within the data themselves, a plausible range of values may be obtained (e.g., based on confidence intervals, the opinion of a single expert, or a consensus view of a range of experts).

15 Data quality & uncertainty

Dealing with data uncertainty

1. Record the range of possible values based on the available studies: “Based on the studies A, B and C, the current population size is between 100 and 410” 2. State the range of potential Red List Categories that may be used based on the range of data:

Critically Endangered Endangered Vulnerable

3. Select one of these categories using all available information (on population size, trends, habitat status, ongoing threats, etc.) to justify your decision:

Critically Endangered Endangered Vulnerable EN

In such cases, the Assessor should:

• [CLICK] Record the range of possible values in the assessment documentation based on the available data. • [CLICK] State the range of potential categories the species may qualify for. • [CLICK] Use all available information on population, habitat, threats, etc. to select the most plausible category from that range.

When interpreting uncertain data, the Assessors attitude is important. IUCN recommends a low dispute tolerance (representing attitude towards uncertainty) be adopted to include the whole range of uncertainty.

16 Data quality & uncertainty

Dealing with data uncertainty

4. Species with VERY uncertain data (suggesting in a very wide range of potential categories) should be listed as Data Deficient.

CR EN VU NT LC

Data Deficient

When the range of values results in a very wide range of potential categories, [CLICK] the species should be assessed as Data Deficient.

17 Red List Categories and Data Quality exercise (20 minutes)

1. Work in groups of 2-3 people. 2. You have 10 minutes to: • Read through the five multiple-choice questions provided and, as a group, decide which is the correct answer. 3. After 10 minutes the Red List Trainer will go through each question and provide the answers.

18