www.ukbutterflies.co.uk

You won't notice this unless I mention it and so .... I've spent the last two weeks poring over various data sources with a view to updating the phenology charts, and the result can now be seen on all of the species pages and the phonologies summary:

http://www.ukbutterflies.co.uk/species_phenologies.php

The latter now includes the UKBMS trends, nicely aligned (horizontally) to make it easier to compare like with like. You'll see this same alignment on the species pages. It's been quite an interesting exercise, since some of the authoritative works out there seem to be at odds with the data! But at least we're now back in line

Onwards and upwards

[Edit] As ever, all comments/observations welcome (we need to get this right!).

Cheers,

- Pete

Well done, Pete!

Very interesting. We know that most, if not all, species are generally emerging earlier due to climate change so when we compare last year's data against the UKBMS data since 1976 one would think 2016 was an average year . But of course we know that last year was generally cool and dull early on and summer didn't really arrive until August.

ATB

Peter

Thanks Peter!

Yes, ironing out year-specific abnormalities against a general trend has absorbed a lot of time. Having said that, some species have really shifted considerably, the Lulworth Skipper showing a flight period that is way beyond anything I would ever have imagined 10 years ago!

Cheers,

- Pete

Thank you for all this updating work, Pete. With all the current concerns about climate change and the apparent importance of butterflies aas 'indicator' species, these charts are increasingly relevant.

I'm puzzled by the vertical axis labelling. I read on the UKBMS site that "This chart shows the average number of butterflies seen on transects between Arpil and October across all sites (fitted values from a Generalised Additive Model)." Presumably, many of the sites do not record any sightings of many species which, I assume, explains peak figures like 0.5 for Clouded Yellow. Without knowing how many transects there are and where they are located, the absolute values don't seem to mean very much or am I missing something? The graphs themselves all appear to be normalised to the highest recorded value (on either curve) so would it be more sensible simply to set the vertical axis relative to this point? (either '1' or 100%)

Mike

Thanks for highlighting this, Pete. I've never come across these pages before but they are extremely interesting and definitely make you take into account the shifting of the graph in many of the species.