Hi

I thought I would just add another chart to the one I created yesterday for ‘Gale index’ from the Objective Lamb Jenkinson data that I downloaded from the Climate Research Unit at the University of East Anglia. This time I thought I would investigate the frequencies of anticyclonicity, and see if I could find any marked anticyclonic spells that occur during Autumn, and see if there was any correlation in them and the many phrases we have to describe them when they do occur.

# Indian Summer (North American in origin and occurring between late September and mid-November)

# Saint Martin’s Summer (associated with St Martin’s day which did fall on October 31 in the Gregorian calendar, but now on November 11) used in Britain, France, Italy, Spain and Portugal.

# Saint Luke’s Summer (18 October)

# Old Wives’ Summer (German in origin and associated with the appearance of Baldachin spiders silken thread) used in the Netherlands, Hungary, Austria, Finland as well as Germany.

Method

Once again I iterated through the 50,000 days of objective Lamb weather types and added up the occurrence of anticyclonic and cyclonic days through the year. I scored one for a pure type – either anticyclonic (A) or cyclonic (C) – and 0.5 for the hybrid types such as ANE through to AN, and CNE through to CN. I then calculated the percentage frequency by dividing by the number of years in the series not forgetting leap years and the 29th of February.

Results

(Charts on my blog at https://xmetman.wordpress.com/2013/0...ndian-summers/)

As you can see from the charts September is probably one of the most anticyclonic months of the year apart from mid-June. The mid-June peak of 43% on the 14th took me by surprise, I always took the first two weeks of June as holiday when at work now I realise I should have made that the second and third weeks! Let me know if you don’t find the same if you parse the data. Getting back to the chart, after the peak of around 33% at the time of the Autumn equinox, the frequency declines to around 25% by the end of September, before rising slightly in the 2nd week of October to around 28% ( a little too early for a St Luke’s Summer). After this It then declines to around 20% by the end of October, with an associated marked rise in Cyclonic frequency to around the same figure. Finally, in the 3rd week of November there is a final pronounced peak of around 28% (St Martins Summer?).

I realise that these phrases are not directly connecting a date with an increased frequency of anticyclonic weather, a so-called ‘singularity’, but obviously down through the centuries people have noticed a spell of weather occurring in Autumn, and labelled it with the nearest Saint’s feast day and it stuck. Nowadays it seems that the more popular term ‘Indian Summer’ has subsumed the older phrases and they have now fallen out of use.

Interestingly, I notice that the latest NWP models are suggesting some form of anticyclonic spell in the next 10 days or so over the UK and western Europe, which I’m sure will keep phrases like ‘Indian Summer’ from falling into disuse.

Bruce.