Key Takeaways
- The Biodiversity Intactness Index (BII) is a measure of the average remaining proportion of indigenous faunal and floral populations in an area relative to an intact reference state.
- The BII requires three sets of information: a map of human impacts, the richness of indigenous species, and intactness scores for these species under different human impact conditions.
- The index can accommodate data scarcity by using intactness scores for groups of species that respond similarly to human impacts.
- The BII can be calculated for all species or a subset, such as vertebrates or plants, and can be averaged across different areas, such as continents, biomes, or countries.
- The BII has been estimated for sub-Saharan Africa using a dataset of intactness scores co-produced by 200 experts in African flora and fauna.
Introduction to the Biodiversity Intactness Index
The Biodiversity Intactness Index (BII) is a measure of the average remaining proportion of indigenous faunal and floral populations in an area relative to an intact reference state, which is typically defined as pre-colonial and pre-industrial conditions. The BII requires three sets of information: a map of human impacts across the area, the richness of indigenous species that occur in the area, and intactness scores for these species under different human impact conditions. The intactness scores are estimates of the remaining proportion of intact reference populations of these indigenous species, on a scale from 0 (no remaining individuals) to 1 (same abundance as the reference), and in rare cases, to 2 (two or more times the reference population).
Calculating the BII
The BII for a unit of land (pixel) is calculated by averaging, across the richness of indigenous species that should occur in that pixel, the intactness score of species given the human impacts (that is, land use) in that pixel. The index can accommodate data scarcity by using intactness scores for groups of species that respond similarly to human impacts, known as functional response groups. The BII can be calculated for all species or a subset, such as vertebrates or plants, and can be averaged across different areas, such as continents, biomes, or countries.
Intactness Scores Co-Produced by Experts
Determining intactness scores for indigenous species in sub-Saharan Africa based on field-collected data of population abundances across different land uses is limited by a lack of appropriate data for most species. Instead, a published dataset of intactness scores was used, which was estimated as part of the Biodiversity Intactness Index for Africa project through a structured expert elicitation process involving 200 experts in African flora and fauna. The experts were identified based on their experience of how sub-Saharan species are affected by human land uses, and they were asked to estimate the intactness of different groups of species in nine land uses characteristic of the region.
Species Richness Across Ecoregions
For vertebrates, each species in the IUCN Red List with a sub-Saharan African range was allocated to a species functional response group in the dataset by lead experts. The list was coupled with a list of species per ecoregion to determine the number of species in each functional response group in each ecoregion. For vascular plants, the proportion of plant species per biome that occur in each functional response group was estimated based on the RAINBIO dataset of tropical African vascular plant species distributions.
Mapping Land Uses and Intensities
A land-use map was created that reflected the six broad land-use classes and the spectrum of intensities that occur in four of these classes. The map was produced using an established decision-tree algorithm built on standardized thresholds of human population density and/or land cover. The land-use variables were obtained from pre-existing map products that spanned the full region to ensure consistency of inputs.
Calculating and Mapping BII
BII scores for each pixel in the land-use map were estimated using the average intactness scores for functional response groups in land-use class k (adjusted for intensity in four of the classes) and the ecoregion (which influences species richness per functional response group). A BII score was determined for several categories, including terrestrial vertebrates and plants, vertebrates, plants, and each of the four vertebrate classes and three broad plant groups.
Analyses and Validation
The BII was calculated for sub-Saharan Africa by averaging BII scores across all pixels in the 1 × 1 km and 8 × 8 km maps. The BII per country, ecoregion, biome, and land use was quantified by averaging scores across all relevant pixels. The uncertainty around these BII values was quantified by averaging upper-limit-BII scores and lower-limit-BII scores across all relevant pixels.
Caveats and Limitations
The BII assessment has several limitations, including the potential for cognitive biases in expert-elicited data, the challenge of identifying who is an expert, and the overrepresentation of certain geographies, nationalities, taxonomic groups, and professions among the experts. The approach of contextualized generalization required several epistemological and methodological compromises, including the use of a pre-defined, relative, and bounded notion of biodiversity intactness and the assessment of BII at the level of functional response groups and land-use categories. Despite these limitations, the resulting BII corroborated other assessments of human pressure, and trends in the BII estimates for individual vertebrate species across their particular ranges were robust when compared with their IUCN threat status.