By Paul Homewood
The Met Office has just issued this news release:
New innovative research has found that for England and Wales there is a 1 in 3 chance of a new monthly rainfall record in at least one region each winter (Oct-Mar).
In the last few years several rainfall events have caused widespread flooding in the UK. In winter 2013/14 a succession of storms hit the UK leading to record rainfall and flooding in many regions including the south east. December 2015 was similar, and Storm Desmond hit the north-west causing widespread flooding and storm damage.
By their very nature extreme events are rare and a novel research method was needed to quantify the risk of extreme rainfall within the current climate.
Professor Adam Scaife, who leads this area of research at the Met Office said “The new Met Office supercomputer was used to simulate thousands of possible winters, some of them much more extreme than we’ve yet witnessed. This gave many more extreme events than have happened in the real world, helping us work out how severe things could get.”
Analysing these simulated events showed there is a 7% risk of record monthly rainfall in south east England in any given winter. When other regions of England and Wales are also considered this increases to a 34% chance.
Dr Vikki Thompson, lead author of the report, said “Our computer simulations provided one hundred times more data than is available from observed records. Our analysis showed that these events could happen at any time and it’s likely we will see record monthly rainfall in one of our UK regions in the next few years”
The authors have named this novel research method the UNSEEN* method to emphasize that this analysis anticipates possible events that have just not yet been seen. It was also used as part of the recent UK Government National Flood Resilience Review (NFRR)+ when the Met Office was asked to estimate the potential likelihood and severity of record-breaking rainfall over the UK for the next 10 years.
The research has demonstrated that, even with the current climate, it is likely that there will be one or more monthly regional rainfall record events, in the coming decade. This new use for Met Office computer simulations could also be applied to assess other risks such as heatwaves, droughts, and cold spells and could help policy makers, contingency planners and insurers plan for future events.
Curiously they seem to be ignoring real observations, in favour of computer simulations. Possibly this is because the historical observations say that there is nothing unusual about recent precipitation events. But more on this in a bit.
First though, let’s look at the central claim:
Analysing these simulated events showed there is a 7% risk of record monthly rainfall in south east England in any given winter. When other regions of England and Wales are also considered this increases to a 34% chance.
Records for SE England only date back to 1910, so we have 108 years of data now. With six “winter” months, ie October to March, we would expect a record high month every 18 years. That is a probability of 5.6%, so the claim of 7% is barely greater.
There are six regions in England and Wales, so the chance of a record month somewhere there is 33.6%.
So the Met Office is hyping up the chance of record rainfall, when in fact their calculations show it to be little different to normal probability.
Concerns about record winter rainfall seem to arise from the belief that winter rainfall is increasing. But there is no evidence of that in England, and despite two wet winters in 2014 and 2016, little in Wales either.
Let’s take a closer look at SE England.
Below is a chart of the wettest “winter” months, the 35 months with over 150mm. The wettest was January 2014, with 205.2mm. Yet December 1914 was not far behind with 204.9mm.
Other than that, there is no evidence at all that these months are getting wetter, or that extreme wet months are becoming more common.
And what about summer months?
Despite the famously wet summer of 2012, the data strongly suggests that heavy rainfall months are much less common, and extreme, than they were in the past.
I find it astonishing that a supposedly objective, scientific report fails to even mention this fact.
Extending the analysis to England as a whole, we find many more extreme months between 1910 and 1960.
And the England & Wales Precipitation series, which goes back to 1766, also shows much more extreme rainfall in the past.
If the Met Office had looked at the actual data, and not their computer simulations, they would have concluded that UK weather is not getting more extreme.
Xmetman, Bruce, also has his take on the story here.
Footnote
This Met Office analysis was used as part of the UK Government National Flood Resilience Review.
This is pure speculation on my part, but is there an element of backside protection going on here?
The Met Office was badly burnt by its failure to anticipate the wet summer of 2012, and winter of 2013/14. By hyping up the claims of extreme rainfall now, they will able to tell the government “we told you so” next time a flood comes along.
Source
The full Met Office paper is here:
via NOT A LOT OF PEOPLE KNOW THAT
July 26, 2017 at 12:51PM
Iowa Climate Science Education 