David M wrote:This shows the dangers of making judgements based on grid representation. Green Veined White may be more widespread than other species, but it is not more common. I'd say Meadow Brown was probably the UK butterfly with the greatest annual numbers.
As for 'rare', that doesn't necessarily mean that restricted range = scarcity. Some species are present in relatively few grid squares but can be abundant in those they DO occur in.
I agree that this is an imperfect measure - although Ian asked for "the UK's 10 rarest butterflies, using whatever criteria", and I gave an answer based on a defined methodology. Certainly it's not a great measure of the commonest species, and I presented the Green-veined White snippet for interest only - I could have added that it places Small Tortoiseshell second, having been recorded in exactly one hectad fewer than GVW, and clearly Small Tort is not our second most abundant species!
For what it's worth, I think you may be right that Meadow Brown has the
highest peak abundance of any species, but I am not convinced that it has the
greatest annual numbers, because it has only a single generation. I suspect that over the course of two, three or even four annual generations, something like Speckled Wood, Large White or even Small Heath might build up sufficient numbers to compete for Meadow Brown's title.
However, I think distribution size performs rather better as a measure of the rarest species, for a couple of reasons:
From a butterfly-spotter's perspective, a smaller range equals a greater chance of a person needing to travel in order to see the species. For example, I cannot see any of my revised top 10 within 2.5 hours' drive of home.
From a conservation perspective, a restricted range might be a signal of vulnerability to environmental change, even if abundance within that range is currently healthy (for example, a large colony of Small Blue could conceivably be wiped out at a stroke by a single housing project).
As a final thought, I think it would be near-impossible to produce a statistically-grounded rarity ranking like this based on abundance instead of distribution, because of the way in which abundance data are collected on UKBMS transects. As you'll know, these transects aren't distributed evenly throughout the country (many fewer in the North) or amongst land-uses (many more on protected sites). Finally, almost every site that contains a population of a species of conservation concern will have a transect specifically established to monitor that species. Therefore, the abundance of rare species (especially those of chalk grassland) is likely to be overestimated compared to that of species that are more widely present in agricultural systems. UKBMS transects could plausibly count
every single Heath Fritillary that flies in the UK in a given year, but probably >0.001 % of Speckled Woods. Add in to this the issue of variable detectability - what percentage of UKBMS transects pass under an oak tree containing Purple Hairstreak but never record the species? Does that mean Purple Hairstreak is rare (probably not)?