June 2016 was the 11th wettest on record for the UK as a whole (records from 1910) and the third in the last decade to see >100mm of rain (2007 and 2012 being the other two, notably the top 2 wettest in the record). There hasn’t been an ‘outstanding’ June in this period either, the warmest being 2010 at around 1°C warmer than average.
British summers are on average, not fantastic – the long term average England and Wales summer rainfall (228.7mm) isn’t significantly different (95%) to what is seen in winter (235.5mm), with the much drier (184.0mm) spring season probably quite confusing for public perception.
Table 1: Some statistical analyses of the EWP record (MAM 1766 onwards). Values in mm.
Therefore, it’s not as though we’re expecting something amazing given the historical precedent. But, the following question arises:
- Given the 3 very wet Junes in the last decade, is the month getting ‘worse’?
I’ve been asked this rather often. It’s of particular interest this year, because it’s 40 years since the record-breaking “long, hot summer” of 1976 which is heavily ingrained in the public perception of British summers. It was the warmest summer on record for the UK and is the only record-holder for a UK season from outside the current climate “normal” of 1981-2010 (DJF 1988-89 being the oldest), whilst June 1976 was the warmest on record for the UK (mean temperature) and also holds the record for highest average maximum temperature on the CET running from 1878.
Let’s look at some temperature statistics. I’ve chosen maximum temperatures as they are probably what most people pay attention to in summer, as opposed to overnight minima in winter.
Figure 1: UK average maximum temperature time series for June, 1910-2016.
Fig.1 shows no statistically significant long-term trend, and there’s not much visible by eye either suggesting not much has changed. However, by closer inspection it appears the ‘volatility’ of recent Junes has decreased – in the earlier part of the record there seem to be wild swings from warm to cold extremes from one year to the next, something apparently absent since the 1990s. A quick statistical measure is standard deviation – from 1910 through 1994, this was 1.22°C, whilst since then it has been 0.95°C. It’s a little unfair to pick different length time periods, but I chose 1995 as the divider as it marked the change of the AMO to a warm phase, which has been shown to impact UK summer weather (Sutton and Dong, 2012).
Figure 2: Variation in decadal running standard deviation for UK June avg. max temp.
Fig. 2 does the standard deviation analysis a bit better, and it’s apparent that the current variation is at a low point for the dataset, and follows a highly volatile period in the mid 20th century. There is no significant trend through the full period (though it’s obvious we’d have one if we went from peak to trough!).
Recap: June’s maximum temperatures haven’t done anything special in the last few decades, but have been rather clustered year-on-year, without extremes.
Performing the same analysis for rainfall, we see that there’s a large number of wet Junes recently (Fig. 3) and an increase in the decadal running standard deviation (Fig. 4) reaching the maximum value in the entire record.
Figure 3: UK June rainfall time series, 1910-2016.
Figure 4: As for Figure 2 but for the data from Figure 3.
Therefore, it seems June has moved into a period of high rainfall variability but low temperature variability, with a marked reduction in cooler months. These two can be obviously linked – the wetter Junes are warmer due to excessive cloud cover (and often mT/cT air).
We can wrap this up with a look at sunshine statistics (which began a little later in 1929) to give a full picture. Figure 5 shows this, and it turns out to be the most exciting!
Figure 5: UK June sunshine hours time series. Linear trend added (p-value 0.03).
Satisfyingly, I’ve got something statistically significant to say! The slope on the line is -0.28 ± 0.13 hrs/year, and passes at the 95% confidence level. This ties in nicely with the reduced temperature volatility, which would be expected in cloudier, wetter conditions.
Thus, I have an answer: June has become a less sunny month, with extremes of rainfall becoming more frequent. Though the month has seen a reduction in the coolest extremes, it hasn’t seen an increase in the opposite extreme, so yes – it’s got a little ‘poorer’.
As for the why? Well, this could turn into an entire research project at this point – and there are papers exploring it – but the AMO link (and global warming exaggeration…and then don’t forget the Arctic dipole anomaly…) can’t be ignored!
I don’t have an equivalent to the grosswetterlagen record, but the lack of extremes, reduction in sunshine and increase in rainfall suggests a reduction in dominating anticyclones – but that’s for another day.
Long term climate records from the Met Office. Charts and analyses produced using RStudio.