Species Under Risk of Extinction (SURE)
For any focal area, the metric indicates the area’s significance for preventing global species extinctions.
Methodology
Leveraging best-possible, high-resolution distribution and population information for tens of thousands of species, the Species Under Risk of Extinction (SURE) dataset informs about the global significance any area of interest holds for the survival of species. It combines formally recognized and inferred threat status with measures of geographic rarity to deliver a single standardized metric for any place, varying from 0 to 100. The metric can be delivered for single or portfolios of sites or landscapes, and disaggregated to species groups or single species, e.g. those that are in rapid decline or those which are legally protected.
The data is built on a foundation of well-documented sources and transparent methodologies. It incorporates clearly defined assumptions and processing steps, with detailed information available on how the data was derived and transformed. Known limitations and potential sources of bias are acknowledged and addressed through careful documentation. The data is structured using consistent, open standards and classifications, supporting both clarity and interoperability.
Our derived datasets conform to internally versioned data schemas with an emphasis on reproducibility and traceability, including extensive metadata and software versioning. Derived and ingested datasets undergo automated validation and filtering to catch non-conforming data and ensure strict data format guarantees and alignment with e.g. taxonomic backbones.
We utilize Cloud-Optimized GeoTIFFs in EPSG:4326 CRS for portable raster data, GeoParquet for vector data, and Parquet for tabular data. We also support GeoPackage formats of our data for easier compatibility with GIS systems. Our dataset metadata is intended to be compatible with the Darwin Core and Humboldt Extension to Darwin Core across our key products.
Data Sources | |
Birds | |
Mammals | Mammal range maps harmonised to the Mammal Diversity Database |
Amphibians | |
Reptiles | |
Suitability | |
Scope and Limitations
The metric has a global extent and is calculated at least once per year. It includes mechanisms to ensure that the data is as complete and accurate as possible, with checks in place to identify and correct inconsistencies or gaps. The methodology accounts for potential biases in data collection and visualization, and these are transparently communicated to users. The dataset resolution and extent is appropriate for local to national to global decision-making. Consideration has also been given to the potential impacts of data use, with safeguards in place to buffer and/or preclude certain data for sensitive species to avoid unintended harm to communities or ecosystems.
Quality Assurance and Review
The metric is maintained through a structured process that includes version control and detailed records of updates. Each release is traceable to its source, with clear documentation of the organizations and individuals involved in its development. The data is reviewed through internal and external processes to ensure its integrity and reliability.
Periodic and triggered dataset updates include automated summaries of input differences, assessment of new results, and comparisons with prior versions. They also include details on the automated and non-automated QA/QC testing for each release. For reproducibility, they also include platform and package environment details of the Conda/Docker environment(s) used and are tied to a specific commit ID of relevant repositories. Metadata and parameters are also recorded. We also support more regular delta updates, which contain only a description of the changes since a prior release.
Metric updates use a combination of materialized views, docker/singularity containers on Azure and AWS instances with Kubernetes, and BigQuery event listeners/SQL triggered actions.
Accessibility and Use
The metric output is designed to be easily accessible and usable by a wide range of users. It is available in formats that support both human readability and machine processing, with minimal restrictions on access.
For each application of use, we produce detailed documentation for each dataset including a detailed field specification, technical white paper detailing the process, and a usage guide. MOL is registered as a DataCite DOI repository and issues DOIs for each versioned release of our datasets. These datasets are self-hosted by MOL via Google CDN in addition to being hosted on a variety of cloud platforms including Google Earth Engine and Esri’s Living Atlas when appropriate.
There are direct channels on the MOL platform for users to provide feedback or ask questions, ensuring that the data remains responsive to user needs.
Creative Commons license types vary based on resolution provided and are clearly communicated in final products.
For any focal area, the metric indicates the area’s significance for preventing global species extinctions.
Methodology
Leveraging best-possible, high-resolution distribution and population information for tens of thousands of species, the Species Under Risk of Extinction (SURE) dataset informs about the global significance any area of interest holds for the survival of species. It combines formally recognized and inferred threat status with measures of geographic rarity to deliver a single standardized metric for any place, varying from 0 to 100. The metric can be delivered for single or portfolios of sites or landscapes, and disaggregated to species groups or single species, e.g. those that are in rapid decline or those which are legally protected.
The data is built on a foundation of well-documented sources and transparent methodologies. It incorporates clearly defined assumptions and processing steps, with detailed information available on how the data was derived and transformed. Known limitations and potential sources of bias are acknowledged and addressed through careful documentation. The data is structured using consistent, open standards and classifications, supporting both clarity and interoperability.
Our derived datasets conform to internally versioned data schemas with an emphasis on reproducibility and traceability, including extensive metadata and software versioning. Derived and ingested datasets undergo automated validation and filtering to catch non-conforming data and ensure strict data format guarantees and alignment with e.g. taxonomic backbones.
We utilize Cloud-Optimized GeoTIFFs in EPSG:4326 CRS for portable raster data, GeoParquet for vector data, and Parquet for tabular data. We also support GeoPackage formats of our data for easier compatibility with GIS systems. Our dataset metadata is intended to be compatible with the Darwin Core and Humboldt Extension to Darwin Core across our key products.
Data Sources | |
Birds | |
Mammals | Mammal range maps harmonised to the Mammal Diversity Database |
Amphibians | |
Reptiles | |
Suitability | |
Scope and Limitations
The metric has a global extent and is calculated at least once per year. It includes mechanisms to ensure that the data is as complete and accurate as possible, with checks in place to identify and correct inconsistencies or gaps. The methodology accounts for potential biases in data collection and visualization, and these are transparently communicated to users. The dataset resolution and extent is appropriate for local to national to global decision-making. Consideration has also been given to the potential impacts of data use, with safeguards in place to buffer and/or preclude certain data for sensitive species to avoid unintended harm to communities or ecosystems.
Quality Assurance and Review
The metric is maintained through a structured process that includes version control and detailed records of updates. Each release is traceable to its source, with clear documentation of the organizations and individuals involved in its development. The data is reviewed through internal and external processes to ensure its integrity and reliability.
Periodic and triggered dataset updates include automated summaries of input differences, assessment of new results, and comparisons with prior versions. They also include details on the automated and non-automated QA/QC testing for each release. For reproducibility, they also include platform and package environment details of the Conda/Docker environment(s) used and are tied to a specific commit ID of relevant repositories. Metadata and parameters are also recorded. We also support more regular delta updates, which contain only a description of the changes since a prior release.
Metric updates use a combination of materialized views, docker/singularity containers on Azure and AWS instances with Kubernetes, and BigQuery event listeners/SQL triggered actions.
Accessibility and Use
The metric output is designed to be easily accessible and usable by a wide range of users. It is available in formats that support both human readability and machine processing, with minimal restrictions on access.
For each application of use, we produce detailed documentation for each dataset including a detailed field specification, technical white paper detailing the process, and a usage guide. MOL is registered as a DataCite DOI repository and issues DOIs for each versioned release of our datasets. These datasets are self-hosted by MOL via Google CDN in addition to being hosted on a variety of cloud platforms including Google Earth Engine and Esri’s Living Atlas when appropriate.
There are direct channels on the MOL platform for users to provide feedback or ask questions, ensuring that the data remains responsive to user needs.
Creative Commons license types vary based on resolution provided and are clearly communicated in final products.
