Pollinator decline: what is actually happening

Three separate things called one thing

The drivers, as identified

IPBES identifies several interacting drivers rather than a single cause, and the interaction is part of the finding — a colony coping with poor forage handles a pathogen load worse than a well-fed one.

• Habitat loss and land-use change. The largest driver. Britain lost around 97% of its species-rich grassland during the twentieth century, and that is the flora most specialist bees depended on.
• Agricultural intensification. Large monocultures produce a feast-then-famine forage landscape, remove field margins and hedgerows, and eliminate the bare undisturbed ground many solitary bees need to nest in.
• Pesticides. Insecticides kill directly; sublethal doses impair navigation, foraging and reproduction. Herbicides act indirectly by removing the flowers. Effects are documented most thoroughly for honey bees and least for the solitary species most likely to be affected.
• Pathogens and parasites. Varroa destructor and its viruses for honey bees; pathogen spillover from managed colonies — including commercially reared bumblebees — into wild populations.
• Invasive species, including introduced bees competing with natives and invasive predators such as the yellow-legged hornet.
• Climate change. Range shifts, and mismatches where a bee's emergence and its host plant's flowering move by different amounts. Montane and northern bumblebees are being squeezed from the warm end without gaining ground at the cold end.

What the current metrics say, and what they measure

Everything above is durable: mechanisms and drivers that will read the same in ten years. What follows is not. It is a national statistic with a publication date, and it is presented here as one rather than dissolved into timeless prose — because a number written as prose loses the year it belongs to within one retelling.

The most useful line in the table is the one that splits the total. Wild bees and hoverflies are moving in opposite directions, and averaging them into a single figure hides the finding. Any statement of the form 'pollinators are down 23%' is quoting a composite index of two groups whose components disagree, over a distribution measure, for one country, to 2024. All four of those qualifications belong with the number.

How pollinators are now being counted

The UK indicator above rests on occupancy because, for most of the period it covers, nobody counted pollinators systematically. That is changing. The UK has run a systematic scheme since 2017, and from 16 December 2026 every EU Member State is legally required to monitor wild pollinators by a common method, under the Nature Restoration Regulation and a delegated regulation adopted in September 2025.

The scheme is built for comparability rather than convenience: randomly chosen 2 km squares on the EU's LUCAS land-survey grid, stratified by biogeographical region and by agricultural, forest and other ecosystems, walked monthly along the same 1 km transect for an hour of observation, with light traps for moths that fly at night. Every Member State has a minimum number of sites.

What the evidence does not settle

None of this undermines the finding that wild pollinators are declining where we can measure. It does mean that confident global percentages should be treated with suspicion, and that the honest version of the story includes what is not known.