This is a linkpost for https://www.sciencedirect.com/science/article/pii/S0006320718313636#f0010
Abstract: "Here, we present a comprehensive review of 73 historical reports of insect declines from across the globe, and systematically assess the underlying drivers..."
...From our compilation of published scientific reports, we estimate the current proportion of insect species in decline (41%) to be twice as high as that of vertebrates, and the pace of local species extinction (10%) eight times higher, confirming previous findings (Dirzo et al., 2014). At present, about a third of all insect species are threatened with extinction in the countries studied (Table 1). Moreover, every year about 1% of all insect species are added to the list, with such biodiversity declines resulting in an annual 2.5% loss of biomass worldwide (Fig. 2)..."
For those concerned about wild animals, such a quick rate of decline could give some reassurance (in addition to the theoretical arguments) that wild insect populations will be small in the long-run. We could have more confidence in the extent of decline if we had a better handle on any publication bias in published papers on conservation and insect populations.
Edit: Note the following comment from below from Gavin Taylor, describing the biased methodology used:
> Good point. I was commenting more on my perception of the conservation field rather than considering biases in the methodology of this study, but they keywords used were:
[insect*] AND [declin*] AND [survey]
Which [is] completely biased to finding studies showing insect declines.
I did overlook addressing the change in biomass value in my original comment but I now see this was the focus of your post. The species in decline % was more striking and the point the authors of the article emphasised. I have now checked the article again to make sure I understand this point.
Unfortunately the article is not particularly clear about the methods it used to get that value, but it the median 2.5% annual rate of biomass loss indicated in Fig. 2 comes from 5 studies measuring biomass loss in specific locations (the introduction lists Germany flying: -2.8% p.a., Puerto Rico: ground foraging -2.7% p.a., canopy dwelling -2.2% p.a., not sure what the other two data points are. They list UK Carabid beetles at -1.05% p.a. but don't include this in the figure).
My interpretation is that 2.5% should be taken to indicate the annual loss of biomass in habitats where many species are in extinction/decline. So although the authors don't state it explicitly, it seems they intend this to represent the gross decline in biomass attributable to species extinction/decline. Yet this should be offset by gains in biomass from habitats where species are increasing in abundance to reach a % for net change. And while the authors don't seem very optimistic about this:
Still, I think the 2.5% loss would be a 'worst-case' scenario. All in all, this value is based on very limited data and I think it should be interpreted cautiously. If more data was available to calculate a net change in insect biomass, I expect this would be much closer to 0%.
Reading this paper carefully actually left me feeling quite skeptical about how species population monitoring is conducted and reported. While I'm not an ecologist or conservationist and may be missing something, it seems there is a strong bias to studying insect groups that are declining vs. those that becoming more abundant (some are mentioned in the text, often generalist species). So the conclusions have to be pessimistic if all the studies you have to review focus on monitoring species with the highest risk of extinction.