An index of intactness

The global community is committed to reducing the rate of loss of biodiversity, but how can progress be measured? A novel system to tackle the problem may also identify key factors behind the changes.

Setting targets has become an increasingly common part of working life, and one that sometimes seems an unnecessary extra burden. But setting the target is just the beginning: gauging progress can be a major undertaking, and all this work will be in vain if the means to achieve the targets are not in place. In the case of biodiversity, measuring the ways in which different ecosystems are changing has proved a challenge1, but on page 45 of this issue, Scholes and Biggs2 unveil an innovative and practical approach that may also turn out to promote good management.

In 2002, the 188 countries that are signatories to the Convention on Biological Diversity committed themselves to “achieve by 2010 a significant reduction of the current rate of biodiversity loss at the global, regional and national level”3. Unfortunately, this laudable target is very vague as regards practicalities. It presents both a challenge and an opportunity for biodiversity scientists4: a challenge because biodiversity is not a simple concept, and coming up with measures that encompass all its aspects will be difficult; an opportunity because when such measures are in place, it will be possible to guide and manage biodiversity better, and so make progress towards a more sustainable world.

Scientists use the term ‘biodiversity’ to reflect almost every aspect of the living world, applying it across a range of spatial and temporal scales to encompass variability within and between genes, genomes, individuals, communities, traits and ecosystems, and including all organisms. Most policy-makers, in contrast, are used to seeing it represented simply as the changing number of species on a species list.

Evaluations of which aspects of biodiversity contribute to the health of an ecosystem clearly indicate that considering variability alone is not enough5,6,7. Biodiversity assessments need to move away from a reliance on species lists and species extinction rates, because often the existence and proximity of local populations matters more.

Variability — the number or diversity of species in an area, say, or the number of genetic varieties of a crop strain in production — is necessary, but it is not sufficient to support the components of biodiversity that underlie key functions and benefits of an ecosystem. It is not hard to list circumstances where the quantity of a single component is crucial (for example, the biomass of forest for timber, or the area of mangrove offering coastal protection), or where a species' distribution in space and time is critical (for instance, pollinators need to be near their host species, and plant cover must be on valley sides to prevent erosion) (Fig. 1).

Figure 1: Measures of biodiversity.

Across a range of levels at which biodiversity can be assessed, variability is not sufficient to capture the essential features that underpin the functioning and benefits of an ecosystem. Measures of both quantity and distribution are important too. The biodiversity intactness index devised by Scholes and Biggs2 attempts to take such measures into account.

A systematic assessment of the dimensions of biodiversity — the different types (the number of different species, say), quantities and distributions at various ecological levels — will give a set of measurements. But it soon becomes clear that they are not all equal. Depending on the context and perspective, some are more significant than others (Fig. 1), and any meaningful evaluation of biodiversity will have to take account of this.

The development of appropriate global indicators for the 2010 target is progressing on a number of fronts. Existing data sets have been exploited to provide measures of forest area, protected area coverage, and trends in the abundance of certain species4,8,9. Innovations and new data sets are revealing trends in the status of threatened species10,11, and the geographical extent of additional ecosystems12. But data to assess the full range of measures (Fig. 1) are extremely sketchy and unrepresentative because of the large gaps in our knowledge and the fact that there is little systematic monitoring. Genetic measures across spatial scales are almost entirely missing. We have named and described fewer than 2 million of the 5 million to 30 million species expected to exist on Earth. Long-term monitoring covers only a tiny proportion of these, and is certainly unrepresentative. Even in relatively well-studied areas of the world, the number of biodiversity measures for which long-term trends can be assessed is remarkably limited. Clearly, new approaches are required if we are to make progress.

Scholes and Biggs' biodiversity intactness index (BII)2 makes a start in satisfying the many requirements, and provides a robust, sensitive but meaningful indicator. The index is built up from relative abundances of populations of species belonging to different taxonomic groups in different ecosystems, and facing different land-use management practices. It can be calculated for any political or geographical unit, and will give an indication of the overall condition of a region relative to a ‘pristine’ state. This state is defined by Scholes and Biggs as the unaltered, pre-industrial state, for which they use the current condition in protected areas as a surrogate measure.

Several features set their method apart from other available methods. The BII allows trends over time and space to be monitored readily. Also, and most usefully, because of the way it is constructed, the index can be separated out to provide comparative information across taxonomic groups, ecosystems or land-use management practices. Hence, unlike other methods that contribute to one measure of biodiversity (that is, one cell in Fig. 1), the BII can contribute to several at once. It can also assist in diagnosing the causes underlying an observed decline: changes can be traced back to reveal which taxonomic groups or ecosystems are losing populations of species the fastest, and whether the overall deterioration is due to many declining populations, a few localized extinctions, or a combination of the two.

The problem of data availability has been sidestepped rather than solved: Scholes and Biggs' calculation is based on expert opinion about how various species fare under different land use in each ecosystem. Clearly, real data would be preferable. But this method might also help to encourage the collection of data, because sampling systems established against this framework would be both achievable and useful, and might therefore be more likely to be implemented.

In addition, because land-use change is incorporated into the index, the results suggest where best to direct efforts to mitigate loss of biodiversity. For example, Scholes and Biggs' BIIs for different taxa (Fig. 1 on page 47) show the relative sensitivity of birds, mammals and amphibians to a change in land use from moderate to degraded — that is, use at a rate exceeding replenishment and causing widespread disturbance. Thus, this method has already moved beyond the stage of designing measures to suggesting actions to achieve the target.


  1. 1

    The Royal Society Measuring Biodiversity for Conservation (The Royal Society, London, 2003).

  2. 2

    Scholes, R. J. & Biggs, R. Nature 434, 45–49 (2005).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Convention on Biological Diversity (2002).

  4. 4

    Balmford, A. et al. Science 307, 212–213 (2005).

    CAS  Article  Google Scholar 

  5. 5

    Mace, G. M. et al. in Ecosystems and Human Well-being: A Framework for Assessment Vol. 1 (Millennium Ecosystem Assessment Ser., Island Press, Washington DC, 2005).

    Google Scholar 

  6. 6

    Luck, W. G., Daily, G. C. & Ehrlich, P. R. Trends Ecol. Evol. 18, 331–336 (2003).

    Article  Google Scholar 

  7. 7

    Balmford, A., Green, R. E. & Jenkins, M. Trends Ecol. Evol. 18, 326–330 (2003).

    Article  Google Scholar 

  8. 8

    Balmford, A., Crane, P., Dobson, A., Green, R. E. & Mace, G. M. Phil. Trans. R. Soc. Lond. B (in the press).

  9. 9

    Loh, J. & Wackermagel, M. Living Planet (WWF Int., Gland, Switzerland, 2004).

    Google Scholar 

  10. 10

    Butchart, S. H. M. et al. PLoS Biol. 2, e383 (2004).

    Article  Google Scholar 

  11. 11

    Brooks, T. & Kennedy, E. Nature 431, 1046–1047 (2004).

    ADS  CAS  Article  Google Scholar 

  12. 12

    Convention on Biological Diversity (2004).

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Mace, G. An index of intactness. Nature 434, 32–33 (2005).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.