Thread by @_PhilMartin

(1/9) Our correspondence piece on our concerns about the Biodiversity Intactness Index is now out in @NatureEcoEvo nature.com/articles/s4155… along with a response by @tnewbold31 and colleagues nature.com/articles/s4155…

Thread to follow...

(2/9) The Biodiversity Intactness Index (BII) aims to measure an area’s average abundance of wild species relative to that in primary vegetation - a value of 50% indicates that the species originally present are on average only half as abundant compared with pristine conditions

(3/9) The BII has been adopted as an indicator for various CBD Aichi targets relating to biodiversity loss and we support the development of further indicators like this, provided that they are robust. However we noticed a few things about the BII that concerned us.

(4/9) The BII as modelled by Newbold et al (science.sciencemag.org/content/353/62…) estimated high intactness of biodiversity in areas with widespread habitat loss - such as SE Asia, Indonesia, and Central America. A UK scale analysis suggested high intactness (>95%) in non-native plantations

(5/9) We used a recently published map of plant biomass intactness (nature.com/articles/natur…) to assess the performance of the BII - expecting that the two would co-vary in space because habitat loss is the major driver of both biomass and biodiversity loss.

(6/9) However, the biodiversity intactness index was higher than biomass intactness for much of the world - such as Europe, China, India and Brazil. Biomass intactness was higher than BII in many arid areas where land use change my not be the major driver of biodiversity loss.

(7/9) When compared to the human footprint index - a metric of human pressure - the biodiversity intactness index is not correlated, but biomass intactness is correlated.

(8/9) Given our results, we think people should be cautious in accepting that biodiversity is as secure as the current biodiversity intactness index suggests. In order to improve its credibility we think that the BII should be further tested and refined.

(9/9) In particular, we think that future versions of the BII should:

(1) correlate with metrics of human pressure
(2) show reasonable alignment with other global-scale metrics of biodiversity loss
(3) enable distinction between ecosystems such as primary forests and plantations

@threadreaderapp unroll please!