Nick H

England are playing Australia at cricket.

The first 10 days of July do indeed show no marked increase in daytime warming - in fact the period mid-June to mid-July doesn't either. These are the average max/mins for the London Weather Centre since 1992 (not including 2009). Jun 1-10 20.6/12.7 Jun 11-20 21.4/13.4 Jun 21-30 21.6/13.6 Jul 1-10 21.9/14.3 Jul 11-20 23.3/15.1 Jul 21-31 24.4/15.9 Aug 1-10 24.2/15.9 Aug 11-20 23.5/15.6 Aug 21-31 21.4/14.3 Incidentally, statistically August 6th is the warmest day of the year since 1992 for both average maxima and minima, with 25.5/16.2 respectively.

I thought 13th June was meant to be the only summer date which had failed to record at least one 30°C?

And of course, more famously, 29°C on May 14 1992 which is arguably more extraordinary than any of the other Edinburgh high temperatures mentioned.

Whilst conditions in London have certainly felt hot this week - I think it's fair to call it a heatwave and it is probably newsworthy to some degree - the way we've been put on "Level 3 alert", one below emergency level, is a joke. 30°C has not been widespread at all and if anything the weather has got less hot as the week has gone on, not hotter: 31°C on Mon and Tues, 30°C on Wed and Thurs. And as for the forecasts this morning, they were talking of 32°C or even 33°C in the Greater London region for this afternoon. As it is only one or two places nudged only just above 30°C, unsurprisingly one of them was Heathrow. At such a short timeframe this was poor forecasting. Just from looking at the morning temps I thought it was too low to reach those heights: Heathrow started at 15°C this morning and the London Weather Centre at 16°C. To reach 32°C/33°C you almost always have to be looking at minima the previous night of at least 18°C to 20°C. I therefore think the forecast was overly optimistic just to justify the absurd "alerts" that had been issued the previous day.

The 33.9°C allegedly recorded at Oadby does not stand up to scrutiny: it isn't an official site to start with and the fact that no station apart from Heathrow recorded higher than 30.4°C (as Benson and Brize Norton both did) makes it even more suspicious. Heavens, if a bloody great concrete oven like Heathrow can't exceed 30.9°C what is the likelihood that some place near Leicester can reach 33.9°C? The nearest stations I can find are for Birmingham, Nottingham and East Midlands Airport - all at 29°C, some 5°C lower than Oadby! This map shows where the heat was. http://www.weatheronline.co.uk/cgi-bin/cgi...tmax&KEY=UK

To be honest I think that's a ghastly picture of the British countryside. I've always found oilseed rape to be an immensely unattractive crop and as for the pylons and wind turbines - the least said the better.

Again we avoid a cold April. We now have to go back to 1989 for the last significantly below average April (6.6°C).

I can't say I remember this particularly but here in the Chilterns we had: 24th 10°C 25th 21°C 26th 8°C

Very nice. Striding Edge is a doddle in fine weather. The width is mostly similar to a large living room. For even more exposure and even greater sheer drops try Crib Goch on Snowdon.

Easter Saturday 1984 was very warm and sunny. Hyde Park was as beautiful and as busy in the spring warmth as I've ever seen it (apart from at the dodgy protests/concerts).

Since the London Weather Centre moved to Clerkenwell in 1992 the average April max and min have been 14.6°C and 7.3°C respectively. The averages for Apr 1-15 are 13.5°C and 6.3°C, and for Apr 16-30 they are 15.6°C and 8.3°C. Totally off-topic but some may be interested to know that the warmest day of the year is Aug 6 - this date has both the highest mean maximum and highest mean minimum for Clerkenwell.

I'd hardly call March unusually mild, just 0.6C above average to the 25th and a downward correction to come in the last 6 days. Only 2001, 2006 and 2008 are likely to have been colder Marches since 1996. One thing that does seem to stand out is the fact that is becoming increasingly rare to get a significantly below average April. The last was in 1989 (6.6°C).

9 times in 2008, believe it or not. It is very rare for a July not to record at least one 28°C.

Yes but it received £63m in funding from the MoD (note 3 of the accounts, page 56). http://www.metoffice.gov.uk/publications/a...ort/ARA0708.pdf

I tend to agree. I get the feeling they would have said that regardless of how cold the winter might have been.

I'm not an expert but the answer is the former. At the end of the 6 o'clock news last night, Alex Deakin said February had been the coldest since 1996. In fact, 2003 and 2006 were colder. What are they checking against and why can't they get a simple fact such as this correct? http://www.bbc.co.uk/programmes/b00hvwnv

The usual bland stuff. What did he say for January? Wrong about the last third of February. How he can make this claim with any pretence of accuracy is beyond me: Most Septembers produce such temperatures. What insight and bravery...

The violins are out....

In the 30 years I've been recording the weather in the Chilterns, we've never had 7 days below 0C, let alone 10. 5 days in Dec 1981, 4 days in Jan 1982, 6 days in Jan 1987 (admittedly followed by a 0.5 max followed by a further 3 sub 0C days), 4 days in Feb 1991. It is totally unrealistic to expect such sustained freezing conditions in this country. If you want 7 days below 0C in this country I suggest you emigrate. Philip Eden suggested yesterday that, taking into account the mild second half of February that is forecast, this winter will rank 22nd in the last 100 years in terms of cold. In a warming phase, that is quite notable.

That would be true if the calendar month were correlated with the data. But it isn't. That's the point we're trying to make. The calendar month is a human construction; the weather is a natural phenomenon.

The last week in May actually shows a cooling, for the south east at least, and is often wet and windy. As for Ms Dunkley's forecast, I wouldn't say it was that accurate. The start of January wasn't especially windy, there was more than "a little sun", it didn't turn even colder, and there wasn't snow after the 20th. November was also inaccurate: it might have been dry (but not exceptionally so), it certainly wasn't abnormally cold and it was rather dull.

Calendar months are no more uncorrelated to the data than any other 31 day period. There is as much likelihood that a given heatwave will coincide within 1 Jul-31 Jul as there is that it will coincide within 9 Jul-Aug 8, 16 Jul-15 Aug or 25 Jul-24 Aug. Of course, there is far less likelihood that a heatwave will coincide within a given 31 day period, but that does not make it any more or less "statistically significant" for doing so. Note that I am using that term in the statistical sense.

That is true. But "statistical significance" is a very precise term that is quite unrelated to whether the Hadley Centre records in 28/29/30/31 day periods.

How can that possibly be true?