Metadata Factsheet

3. Goals And Targets Addressed

3a. Goal

Goal A (headline indicator). The integrity, connectivity and resilience of all ecosystems are maintained, enhanced, or restored, substantially increasing the area of natural ecosystems by 2050; Human induced extinction of known threatened species is halted, and, by 2050, the extinction rate and risk of all species are reduced tenfold and the abundance of native wild species is increased to healthy and resilient levels; The genetic diversity within populations of wild and domesticated species, is maintained, safeguarding their adaptive potential.

3b. Target

Target 1 (headline indicator) Ensure that all areas are under participatory, integrated and biodiversity inclusive spatial planning and/or effective management processes addressing land- and sea-use change, to bring the loss of areas of high biodiversity importance, including ecosystems of high ecological integrity, close to zero by 2030, while respecting the rights of indigenous peoples and local communities.

Target 2 (complementary indicator), Target 3 (component indicator) & Target 7 (complementary indicator)

5. Definitions Concepts And Classifications

5a. Definition

Indicator definition: The Red List of Ecosystems framework assesses the relative risk of ecosystem collapse of an ecosystem type. The indicator ‘Red List Index of Ecosystems (RLIe)’ measures the average risk of ecosystem collapse of a group of ecosystems and allows for tracking change over time, based on genuine change in the risk category of each ecosystem.

The RLIe can be calculated for any set of ecosystem types for which there are Red List of Ecosystems assessments. It can thus be calculated at the country level or at the global level, or for broad ecosystem groups (such as forests).

Key concepts that underpin the indicator include the following:

• Ecosystems are made up of living components (biotic complexes and assemblages of species), the abiotic environment, the processes and interactions within and between the biotic and abiotic components, and the physical space in which these operate (Keith et al. 2013).
• Ecosystem types are differentiated from one another by a degree of uniqueness in composition, structure, and ecological processes and function. Ecosystem types present a useful model or abstraction of the complexities of the natural world. Similar definitions are used for other, often synonymous, terms such as ecological communities, habitats, biotopes, and vegetation types (Keith et al. 2013, Nicholson et al. 2021). Ecosystem types can be described, classified, and identified using the IUCN Global Ecosystem Typology (Keith et al. 2022).
• Ecosystem collapse is the endpoint of ecosystem decline, when an ecosystem loses its defining features (i.e., species, assemblages, structure, and functions) and is replaced by a different, often depauperate, ecosystem type. Collapse can be irreversible, but some ecosystems may recover, over long timeframes or with restoration. The risk of ecosystem collapse is the likelihood that an ecosystem will collapse over a specified timeframe (Keith et al. 2013).
• Risk categories: The risk of ecosystem collapse is based on the risk categories each ecosystem is assigned through assessment under the Red List of Ecosystems framework. The risk categories include, in order of increasing risk of collapse: Least Concern, Near Threatened, Vulnerable, Endangered, Critically Endangered, and Collapsed. If there are insufficient data to assign a risk category, a criterion or ecosystem type it is considered Data Deficient, or Not Evaluated if not assessed.

Guidelines for the application of the Red List of Ecosystems can be found on the IUCN website (Keith et al., 2013; Bland et al; 2017).

5b. Method Of Computation

The Red List of Ecosystems is the global standard for assessing risk of ecosystem collapse and biodiversity loss to all marine, freshwater, and terrestrial ecosystem types. Red List of Ecosystems assessments collate standardised knowledge, maps and data about ecosystems, and apply quantitative criteria to estimate relative risks of ecosystem collapse to identify threatened ecosystem types. The five criteria are: (A) change in ecosystem area; (B) restricted ecosystem distribution; (C) change in the abiotic environment (e.g., hydrological processes); (D) change in biotic processes and components (e.g., species interactions); and (E) the probability of collapse estimated using dynamic ecosystem models (where such models are available). Change in area and integrity (Criteria A, C and D) is assessed over a 50-year timeframe (past and/or future), and/or since the onset of industrialised change (1750 at the earliest). Change in integrity (Criteria C and D) is measured using ecosystem-specific metrics, to capture different ways in which ecosystems respond to drivers of biodiversity loss. For example, integrity can be tracked in forests using the proportion of old-growth (Burns et al 2015), in coral reefs using coral cover and fish abundance (Obura et al 2022), and in rivers using hydrological flow (Ghoraba et al. 2019). Through assessment against one or more criteria, ecosystem types are assigned to risk categories, shown in Figure 3.

Detailed guidelines are available to support the assessment of each criteria (Bland et al. 2017; Keith et al. 2013). Ideally, as many criteria as possible should be assessed, but the scope can be tailored to the resources and data available. Guidelines to support rapid assessments using one or a few criteria have been developed (eg Holness & Botts 2022) and used to in national assessments across Africa (e.g. NEMA 2020).

Figure 3. The risk categories of the Red List of Ecosystems: Collapsed, Critically Endangered, Endangered, Vulnerable, Near Threatened, Least Concern; if there are insufficient data to assign a risk category, a criterion or ecosystem type is considered Data Deficient, or Not Evaluated if not assessed. The risk categories highlighted in grey are considered threatened, and ecosystem types in those categories can collectively be referred to as “threatened”.

Red List Index of Ecosystems (RLIe): The RLIe measures trends in ecosystem collapse risk based on the proportion of ecosystem types in each risk category (for details see Rowland et al. 2020). The RLIe is the weighted mean of ordinal ranks assigned to each risk category:

where Wc(i,t) is the risk category rank for ecosystem type i in year t (Collapsed=5, Critically Endangered=4, Endangered=3, Vulnerable=2, Near Threatened=1, Least Concern=0; following Butchart et al. 2007), WCO is the maximum category rank (Collapsed=5), and n is the total number of ecosystem types excluding Data Deficient or Not Evaluated ecosystem types. The RLIe ranges from 0 (all ecosystems Collapsed) to 1 (all Least Concern). The RLIe can be reported at national or global levels, and disaggregated by ecosystem type, using the IUCN Global Ecosystem Typology. It can be presented in multiple ways, including as a current snapshot of risk (Figure 2), or a time-series where available (Figure 4). The RLIe can calculated for the overall risk category, or separately for each criterion, for example to examine the impacts of recent rather than long term change in ecosystem extent or integrity.

Figure 4. The Red List Index of ecosystems (RLIe) presented as a time-series for selected ecosystem types in Norway (Krikjeeide et al. 2021), comparing past change (from 2011 and 2018) with projected impact of policy pathways on ecosystem risk status.

Indicator testing: A key question for indicators is how sensitive they are to biodiversity change. Several studies have tested aspects of the Red List of Ecosystems framework in its capacity to detect meaningful change in ecosystems. For example, Murray et al (2017) tested metrics for restricted range size (Criterion B) for their capacity as predictors of ecosystem collapse in landscapes subject to stochastic threats. They found that the methods currently used in Red List of Ecosystems assessments for measuring range size are the best spatial metrics for estimating risks from stochastic threats. Analyses from Norway found that the RLIe could provide time-series to reliably compare alternative policy scenarios (Kyrkjeeide et al. 2021 – Figure 2). The RLIe has been tested for sensitivity and responsiveness using an ecosystem simulation model of a coral reef (Rowland et al. 2020b), showing that the RLIe can differentiate between low and high threat levels, responds to both increases in threats (e.g., climate change) and decreases (e.g., effective conservation policy), and detects change in area and integrity.

5c. Data Collection Method

Data for Red List of Ecosystems assessments come from a diverse range of sources. These are best examined via published national assessments and individual studies/ecosystem assessments, although some reviews exist (e.g., Rowland et al 2018, Murray et al 2018). Ecosystem classifications and maps used in these assessments typically come from national ecosystem inventories (e.g. forest types), local experts (e.g. within universities of environment institutes), and government agencies. Global classifications and maps of ecosystem types (e.g., the Global Ecosystem Typology, https://global-ecosystems.org/) can also be tailored and downscaled for local use. Data on change in ecosystem area (for Criterion A) typically comes from similar data, although an increasing number of global datasets are also available – many of these are listed as data sources for Headline Indicator A.2 (extent of natural ecosystems) and as complementary indicators in the Global Biodiversity Framework monitoring framework (e.g., tree cover loss, wetland extent trends index, and trends in mangrove extent).

Assessing Criteria C and D requires ecosystem-specific variables, which may come from a range of data sources, including scientific literature, reports, experts, historical accounts, and existing indicators (including some listed as complementary indicators in the monitoring framework, e.g., live coral cover). These data may be field-based empirical data, remotely sensed (e.g., satellite imagery, see Murray et al 2018), modelled (extrapolating from field and/or remotely sensed data) or a combination. The Red List of Ecosystems guidelines (Bland et al 2017) provide advice on the types of data needed, and how it should be analysed. Further guidance and reviews are currently being developed to provide further support for assessors.

5d. Accessibility Of Methodology

Application of the Red List of Ecosystems framework for undertaking Red List of Ecosystems assessments is supported by a range of resources, all accessible via the Red List of Ecosystems website https://iucnrle.org/:

• Formal guidelines published by IUCN (Bland et al 2017): https://portals.iucn.org/library/sites/library/fil....

• Multiple peer-reviewed scientific papers have been published to describe the assessment methods in detail (e.g. Keith et al 2013, and others – see reference list below)
• Free online training material via FutureLearn (in partnership with Deakin University and IUCN) and IUCN.
• Tools to support assessment (available at https://iucnrle.org/rle-material-and-tools)
• A growing database of assessments https://iucnrle.org/rle-database

The Red List Index of Ecosystems can be calculated for any set of ecosystem types for which Red List of Ecosystems assessments have been undertaken. The method for calculating the RLIe was published in an open-access peer-reviewed paper (Rowland et al. 2020). Scripts to calculate the indicator using the program RStudio are publicly available via the Red List of Ecosystems GitHub site (https://github.com/red-list-ecosystem/rle_indices). The script provides the code to calculate the indicator and includes examples of the indicator outputs using sample data from the continental assessment of 136 temperate and tropical forests across 51 countries/territories in the Caribbean and Americas (Ferrer-Paris et al. 2019). The sample data are provided to demonstrate the structure of the data required to calculate the indicator.

5e. Data Sources

Red List of Ecosystems assessments are typically published in technical reports, and/or peer-reviewed publications. An increasing number of assessments are available in a publicly available centralised database of assessments (https://assessments.iucnrle.org). This database will become a key source of data at national, regional and global scales in the medium term. In the short-term, Red List of Ecosystems assessments may be accessed from relevant national agencies or other data holders/providers. The information on risk categories required to calculate the RLIe can typically be obtained from these sources.

5f. Availability And Release Calendar

To date, over 4000 ecosystem types have been assessed using the Red List of Ecosystems framework (Figure 5) in over 100 countries (Bland et al 2019). The majority of these assessments were undertaken as part of national assessments. Red List of Ecosystems assessments for all terrestrial ecosystem types are available for 63 countries, approximately one third of the CBD’s 196 Parties, allowing both summary statistics (eg proportion of threatened ecosystems) and RLIe to be estimated. Subsets of terrestrial ecosystem types are available for another 25 countries (e.g., all forest ecosystems across the Americas, in Ferrer-Paris et al 2019). Comprehensive assessments of all freshwater ecosystems are available for 41 countries (with subsets for a further 43, e.g. all wetlands), and of all marine ecosystems (including coastal or marine transitional ecosystems) for 32 countries (with subsets for a further 46, e.g. all mangroves or coral reefs).

For countries where assessments are not available, global terrestrial assessments are anticipated to be available for key ecosystem types by 2025. Complete global coverage of Red List of Ecosystems assessments for terrestrial ecosystem types is expected to be available by 2025, with anticipated updates every 5 years. Some countries have already undertaken repeat assessments (e.g., South Africa, Norway, and Finland) providing time-series. At a national level, the release will vary by country.

The RLIe can already be calculated for all countries that have completed Red List of Ecosystems assessments (Figure 5), and a time-series for those countries with repeat assessments (eg Norway, Figure 4). Countries can use available code or seek assistance to calculate RLIe values from their National Red List of Ecosystems assessments.

Figure 5. Current availability of Red List of Ecosystems assessments by country. Countries where all terrestrial ecosystems have been assessed are shown in red (63 countries), and those with subsets of terrestrial ecosystem assessed are shown in pink (25). Regions where all marine ecosystems have been assessed are shown in dark blue (32), those with subsets of marine ecosystems assessed in pale blue (46). Black dots show individual ecosystems that have been assessed, separate from comprehensive, wall-to-wall assessments. Coverage of freshwater ecosystems is not shown, but typically mirrors terrestrial assessments (41 countries, with subsets in 43). Further assessments are underway (e.g. in Australia, Namibia).

5g. Time Series

The Red List of Ecosystems uses data on ecosystem trends to assess risk of collapse (Criteria A, C and D), and is therefore inherently trend-based. A first assessment is available based on multiple regional and national assessments (see coverage in Figure 5). As noted, some countries have already undertaken repeat assessments (e.g., South Africa, Norway, and Finland) providing time-series of change in risk. The IUCN has committed to support the ongoing development and application of the Red List of Ecosystems, and therefore calculation of the RLIe, with a goal of assessing all terrestrial ecosystems globally by 2025.

5h. Data Providers

Red List of Ecosystems assessments can be done at a range of scales, most commonly national or global. Data to conduct Red List of Ecosystems assessment and thus generate risk categories to calculate the RLIe, are gathered from a range of sources, including scientific literature, reports, experts, historical accounts, maps, and satellite imagery. Assessment teams can include government, non-governmental organisations (NGOs), and/or academic institutions.

Currently some of the approximately 4000 assessments are in the Red list of Ecosystems database, which can be found via the website: https://assessments.iucnrle.org This includes global/regional, national and subnational assessments. The data available via the database will grow as more existing assessments are entered, and more assessments are completed. 

National: Systematic assessments of all ecosystem types or a subset of ecosystem types within a country can be done for individual countries (see Figure 4). These assessments are typically conducted by governments, often in association with NGOs and/or academic institutions. These assessments may be available in the Red List of Ecosystems Database, national databases, reports and/or in the scientific literature.

Other: Strategic assessments of one of a few ecosystem types may be done by individual researchers or research teams (e.g., mountain ash forests in Australia, Burns et al. 2015; fringe mangroves in the Philippines, Marshall et al. 2018). These are typically then entered into the Red List of Ecosystems Database or may be available to countries as they compile national assessments.

5i. Data Compilers

Data to calculate the indicator are currently compiled independently by the assessment teams for available national or global assessments. In the near future, a global level indicator will be coordinated by IUCN and the IUCN Commission on Ecosystem Management.

5j. Gaps In Data Coverage

To date, over 4000 ecosystems have been assessed in over 100 countries (https://iucnrle.org/rle-in-progress ). The current goal is for all terrestrial ecosystems to be assessed by 2025. More countries havecomprehensive assessments of terrestrial ecosystems available (63) than freshwater (41) or marine (32). South America has much greater spatial coverage than Northern Africa, South Asia or Eastern Europe. Spatial gaps will be closed though a combination by targeted global-level projects across broad thematic ecosystem groups (e.g. forests, mangroves), and national assessments.

5k. Treatment Of Missing Values

Typically, Red List of Ecosystems assessments are done for all ecosystem types within a jurisdiction. This can result in some poorly known ecosystem types being listed Data Deficient, or Not Evaluated if there are insufficient resources to evaluate all criteria or ecosystem types. Due to differences in data availability for spatial, environmental, and biological variables, many assessments only apply criteria relating to distribution (Criteria A and B) (Rowland et al. 2018). Missing data may stem from incomplete assessments, or coarse scale assessments that exclude finer scale variation in ecosystems. Uncertainty in Red list of Ecosystems assessment can be dealt with through a range of methods, including bounds in estimates of risk category – please see the Red List of Ecosystems Guidelines for more information (Bland et al 2017).

6. Scale

6a. Scale Of Use

Scale of application: Global, Regional, national

Scale of data disaggregation/aggregation:

Global/ regional scale indicator can be disaggregated to national level: Yes

National data is collated to form global indicator: Yes

The indicator can be used at national, regional, and global levels, depending on data availability. Global/regional values can be disaggregated to national scales (eg Ferrer-Paris et al 2019, Obura et al 2022). In principle, national data can be aggregated to form regional/global assessments, though this is yet to be applied and will require more testing.

6b. National Regional Indicator Production

6c. Sources Of Differences Between Global And National Figures

The procedure for applying the Red List of Ecosystems framework is the same across national to global scales. The outcomes may vary between national and global levels where:

1. National assessments have more detailed ecosystem units than global assessments; for example, a comparison of ecosystem classifications in South Africa found that more ecosystem types were listed as threatened when using more finely defined units (Payet et al 2013).
2. Ecosystems extend beyond national boundary, particularly for smaller countries. This can also be dealt with to some degree by aligning and aggregating similar ecosystems with the Global Typology, and learning from species red listing (e.g. national assessment guidelines that account for rescue effects).
3. There are inconsistencies among nations in data and indicators used in assessments, and in the criteria assessed. This is will require tools to be developed, learning from experiences in economics where the methods have been developed to account for ways in which national accounts, GDP and other economic indicators vary per country.

6d. Regional And Global Estimates And Data Collection For Global Monitoring

6d.1 Description Of The Methodology

The methods of conducting Red List of Ecosystems assessments are well established, and the same across national, regional and global scales. The methods of calculating the RLIe are also the same across national, regional and global scales. At present, indicators at these different scales are calculated based on assessments at the corresponding scales; country values are not yet aggregated to calculate regional or global assessments. Where ecosystem types extend beyond national boundaries, there is the possibility of aggregating data using the IUCN Global Ecosystem Typology – approaches for doing so are currently being trialled.

6d.2 Additional Methodological Details

6d.3 Description Of The Mechanism For Collecting Data From Countries

7. Other MEA And Processes And Organisations

7a. Other MEA And Processes

N/A

7b. Biodiversity Indicator Partnership

Yes

10. Data Reporter

10a. Organisation

International Union for Conservation of Nature (IUCN) CEM

10b. Contact Person

Emily Nicholson (e.nicholson@deakin.edu.au)

Marcos Valderrabano (marcos.valderrabano@iucn.org)