Summer – what summer? As the rain lashes down and the temperature is hovering at a miserable 12 degrees, it hardly seems believable that last week we were basking in Mediterranean sunshine. Perhaps you had convinced yourself that we were all set for a long hot summer of barbecues, beach parties and pink peeling noses. So what are the chances of a fine spell in May being the portent of a scorching summer? Fortunately for us modellers, we have an excellent source of past data to draw on – the grandly named Hadley Central England Temperature records. (see Hadcet) This data is representative of a triangular area of the United Kingdom enclosed by Lancashire, London and Bristol, and is the world’s longest set of temperature records, dating back to 1659 (a cold start to the year; as Samuel Pepys notes in his diary: “we not having one coal of fire in the house, and it being very hard frosty weather.”) Anyway, for our purposes we’ve used the set of daily maximum temperatures which starts in 1878, and our measure of “a fine spell in May” is taken to be “the number of days when the maximum temperature is 21 degrees or higher”. According to this measure, we had 3 fine days in May 2009. Our measure of “a scorching summer” is “the number of days in June to September when the maximum temperature is 24 degrees or more”. So how many scorching days can we expect in the summer of 2009? Sadly, as you can see from the chart below, fine days in May are not a good predictor of a scorching summer. For the statistically minded amongst you, the R squared value is less than 1%. As you can see from the chart, 1976 (I remember it well …) was the best summer ever, with 38 days where the temperature rose above 24°C. It was preceded by a May with three fine days … but so was 1920, which scored zero on our ‘scorching summer’ index! Other scorching summers include 1911, 1933, 1995, 2003 and 2006. But do you remember the major disappointment of 1992? After 13 glorious days in May the weather fizzled out into a miserable summer.