Thread by @AndyPurvisNHM

[1] Let me explain it for you in this thread. @IUCNRedList and their many partner organisations have been producing detailed assessments of species extinction risk in different taxonomic groups since 1994 (when the first comprehensive #RedList assessment of birds came out)

https://twitter.com/toadmeister/status/1126063833724989441

[2] Obviously we can't be at all sure that birds are typical, so it would have been wildly unsafe to go from "14% of birds are threatened" to "14% of all species are threatened". Since then, @IUCNRedList assessments have been done for an ever-expanding set of taxonomic groups...

[3] ...with the proportion threatened ranging from ~9% for bony fishes up over 60% for cycads. But the *average* level of threat has started to settle down now, at about 25%, and we now have estimates for lots of different vertebrate, invertebrate and plant groups...

[4] ...and there's also evidence that, as assessment within groups gets more and more complete, the proportion threatened heads towards that average (cell.com/current-biolog…). But, there's much less evidence on levels of threat in insects...

[5] ...because there's so damn many of them (about 5.5 million of the planet's ~8 million animal and plant species). The only major group of insects that's been assessed so far is Odonata (dragonflies etc), and about 15% of them are threatened...

[6] ...but we can't be sure *that's* representative of insects as a whole. So what we did was looked at assessments in Europe (where data are best) of groups within the 3 most diverse insect orders - butterflies (within Lepidoptera), bees (within Hymenoptera) and ...

[7] ...saproxylic beetles (within Coleoptera); these all had from 9%-18% of species threatened within Europe. And Europe has LOW levels of extinction risk for vertebrates compared with the rest of the world - meaning it's unlikely that a lower fraction of insects...

[8] ... are threatened worldwide than in Europe. So it's unlikely that fewer than 10% of insects are threatened worldwide. Then the simple sum, with very conservative rounding to keep it simple and avoid overprecision...

[9] ... The two components are: 10% of 5.5 million insects is ~ 1/2 million threatened insects; and 25% of 2.5 million other animals and plants is ~ 1/2 million threatened non-insects. Adding them together gives the 1 million...

[10] And there's also a wholly independent second line of evidence, using entirely separate data and analysis, detailed in the chapter (which just logistically takes time to get ready for publication, but which), that supports this figure...

[11] Crucially, and absolutely contrary to your premise, ALL of this has been exhaustively peer-reviewed: the #GlobalAssessment received over 20,000 comments from thousands of reviewers, and all of the comments and responses will also be public documents...

[12] You too could have been a reviewer: all you would have had to do was sign up. The journalists you berate prob knew this (@guardian @BBCNews @CNN @nytimes) - they know lots about what they write about. Also, when they don't know something, they ask someone who does...

[13] ...rather than writing a piece in which, say, prejudices and dark suspicions fill in gaps in their knowledge. I trust that you have taken the time to thoroughly research the @IPBES review process? No? Well, I'm glad to have had the chance to bring you up to speed. [ends]

Oops - forgot to tag @toadmeister in my thread.