Heavy Rainfall Claims Not Borne Out By The Data
By Paul Homewood
With special thanks to David
In a post last week, I looked at claims that heavy rainfall events were on the increase in the UK, specifically during winter. This claim has now achieved a semi official status, and was repeated by Corinne Le Quere, Director of the Tyndall Centre, last week.
Her claim is based on the UK Climate Climate Projections Report of 2012, itself based on a study by Tim Osborn of the UEA.
For the background to all of this, it is worth reading my original post first, here. Essentially though, I was critical of the Report’s findings because it had only identified the trend from 1961-2006. This is important, as the 1960’s & 70’s were a notably dry period, and an analysis of rainday data showed a clear correlation – in wet years you were more likely to get heavier downpours. In other words, was the selection of the start date creating an artificial bias?
I, therefore, suggested that Osborn’s analysis should be extended back, at least, to 1910. As winter precipitation in the early 20thC was as high or higher than in recent years, it seemed worth checking whether the same trends found in Osborn’s study would still appear.
Back to 1931
Although the Met Office only started publishing rainday data from 1961, readers QV and David both suggested that we might be able to use the UK Regional Precipitation data, which holds daily rainfall data from 1931 and is maintained by the Met Office Hadley Centre.
David kindly downloaded the data and wrote a program to identify the number of days in each month, when precipitation exceeded a preselected amount. The results are interesting.
First, a quick note. The regional data is analysed across nine regions, and David has set up databases for each. To get to the overall UK position, we have applied a crude average of each region, rather than using a geographically weighted average. I will be posting on the regional data in the next day or three.
I have started by using a threshold of 10mm/day. The logic of this is that, very broadly, the number of days over this level would account for roughly 10% of raindays, although different threshold would need to apply to different regions.
The results do show a slight increasing trend, but this appears to be simply a function of the dry period in the middle. Assuming the years from 1901-30 were similar to the following two decades, the trend would actually reverse.
What appears to be more significant is that the number of heavy rainfall days peaked in the 1980’s and 90’s, and has since fallen back to levels that were usual up to the 1960’s. In Figure 3, I also include the last ten years, 2004-13. There has been a slight increase in this most recent period, but the number of days >10mm is still less than 1981-2000, and also less than during the 1950’s. It is also worth noting that the last ten years are close to the 1931-2010 mean.
I have repeated the exercise, using thresholds of 15mm and 25mm, to see if there is any trend at the really heavy end of the rainfall scale.
As Figures 4 & 5 show, the pattern is similar. Numbers of heavy rainfall days during the latest ten year period are close to the long term mean and well below some earlier decades, with little sign of any increasing trend.
1) While it is true to say, as Le Quere does, that heavy precipitation in winter has increased over the past 45 years in all regions of the UK, there is little evidence to suggest that it has increased over the last 80 years or so.
2) The fact that higher levels of heavy precipitation existed during large parts of the 20thC, than we have seen since 2001 surely calls into question her claim that “climate change has raised the risk of flooding”. [She makes clear later on that she means a warmer climate]
3) There are many factors which influence precipitation patterns, but one important one is the North Atlantic Oscillation, or NAO. On the UK Climate Projections Report, mentioned above, there is a section (on the very next page to the precipitation report) which addresses the issue of the NAO. It says:
The average atmospheric pressure over the North Atlantic in winter* is less than that over the Atlantic sub-tropics; this pressure gradient is consistent with the winter climate of the UK being dominated by westerly winds. However, the pressure gradient (usually defined by the sea-level pressure difference between the Azores and Iceland) changes from winter to winter, part of a pattern of atmospheric variability known as the (NAO). In some winters the gradient will be larger (that is, a greater average pressure difference, defined as a positive phase of the NAO), giving westerly winds which are stronger or more persistent than average, leading to northern Europe being warmer and wetter than average and southern Europe colder and drier. In the opposite, negative, phase of the NAO, the pressure gradient is smaller, giving weaker or less persistent westerly winds, northern Europe is colder and drier than average and southern Europe warmer and wetter.
The North Atlantic Oscillation Standardised Index,
The dry winters of the 1960’s are readily explained by the extremely negative state of the NAO, and it is also notable that the NAO was extremely positive in the early years of the 20thC, leading up to 1931, when our record starts.
This reinforces the suspicion that heavy precipitation would have been at least as prevalent then, as in recent years
Quite simply, there is no need at all to invoke global warming as the cause of heavier precipitation since 1960.
Rather than getting bogged on trends over the last 45 years, there seem to me to be two questions that need to be answered.
1) Is global warming/warmer oceans leading to heavy precipitation either becoming more common or more severe in the UK during winter?
The evidence above shows that heavy precipitation was just as common, and often more so, in earlier decades.. This is also supported by the total winter rainfall stats, which show lower winter precipitation in recent years than the early 20thC.
If warmer oceans really do lead to more rainfall, maybe the Atlantic was as warm as now a century ago?
2) What future trends can we expect?
There is little in the long term record to suggest any increase in heavy precipitation. Indeed, the trend in the last decade has been towards drier winters.
There is a third question. None of the above is rocket science – it is all based on readily available data, and the NAO is well understood.
Why, then has Corinne Le Quere ignored all of this, and published misleading claims?
Does she actually understand a lot less about the climate than she pretends?
Has she taken the evidence from the UKCP report at face value, without bothering to ask some very basic questions.
Is it because the trend increase since 1961 fits her theories, and she thinks there is therefore no need to look further?
Or is she aware of everything I have put forward above, but is keeping quiet about it, as it does not suit her agenda?
Whatever the answer, I think we deserve better.
1) All the UK Regional Precipitation is available at the Hadley Centre.