Category Archives: weather in the media

London winter forecast 2018-19

Long range modelled forecasts have been all over the place of late and, looking at the underlying signals, it is easy to see why.

When I’ve produced these forecast in the past, in terms of QBO and ENSO data, there’s usually a lot of analogues to compare with. This year, however, seems to be an exception.

Considering QBO first I looked back over data to 1950 and found nothing similar for October. However, looking over the whole series the cyclical nature of this circulation may give some clue.

bestfit

Some 20 months were revealed, ranging from June 1959 to June 2015.  Using NOAA’s  Niño 3.4 region I narrowed this list down to the few that had an ENSO value of around +1 with a rising trend. With NOAA’s forecast of a Modoki El Nino (one that occurs in the central Pacific) this narrowed the list to just 1 period: June 2015. Considering maxima anomalies this would give the following winter.

winter 2018-19 max anomaly.PNG

The above would suggest there being a general cool down through December with a cold spell starting just before Christmas into the new year? And another cold spell end of January into the first week of February?

winter 2018-19 precip anomaly

The above precipitation anomaly chart would suggest a wetter than average December,  January and February, though February by much less so.

It’s been a very busy autumn so I’m keeping this short.

The below figures, particularly January and February, may be different in the event of an SSW occurring. In all then.

The mean:
December: +0.8C
January: -1.2C
February: -1.7C
Overall: -0.3C (broadly average)

Precipitation:
December: 158%
January: 155%
February: 120%
Overall: 134%

 

Advertisements

More hot air on bank holiday records

The media are up to their old tricks of flagging up their usual ‘hottest ever’ line; this time the May Day bank holiday.

Screen Shot 2018-05-05 at 02.53.09
Today’s Guardian

Today’s Guardian informs readers that Monday could be the warmest ever May Day bank holiday, the quite unremarkable record high of 23.9C was set in 1999.

A closer look, however, reveals that dates of the holiday, which only began in 1978, are always moving. Like Easter the date shifts and can fall on any day between the 1st to the 8th. Only six early May bank holidays have occurred on the 7th. It is therefore difficult to compare like with like.

A more correct approach is to compare date records which, for the 7th since 1978, the highest is 27.1C in 2016. The highest temperature for the 7th going back 60 years is 28.5C in 1976.

Screen Shot 2018-05-05 at 12.07.22

Screen Shot 2018-05-06 at 18.37.01

 

 

Past weather to rescue future reporting

“Worst floods ever!” Hyperbolic reporting of weather events is nothing new; whenever the UK is hit by the latest named storm breathless news anchors often try to portray atrocious conditions as ‘unprecedented’. This may soon change thanks to a project that uses citizen science.

Operation Weather Rescue, the citizen science project which last year processed Ben Nevis and Fort William data from between 1883 and 1904, has now recovered thousands of European observations from past Daily Weather Reports compiled from 1900-1903.

Though reanalysis data of the northern hemisphere back to 1851 have been available for some years the maps don’t give enough detail to be able to compare if an old storm has the same pressure characteristics as a modern storm. Evidence of impacts is restricted to fixed observations such as temperature, rainfall and archive press reports.

Weather Rescue: “We want to learn about the frequency of intense storms in the past to compare with now and these rescued observations will significantly improve our understanding.”

The latest release of data features pressure plots from June 1903 that tie in with the record-breaking 59-hour deluge that left much of the London Borough of Redbridge underwater. The following October, nationally, was the wettest month ever recorded in the UK.

Further recovery of these archived data, with the help of thousands of volunteers, will prove invaluable to our understanding of high-impact weather, and if storms are getting worse.

This latest example of Big Data could also offer huge benefits to the insurance and reinsurance industries, as well as planning engineers, as the resolution of climate models steadily improve.

But on a purely weather reporting scale it should also be possible to one day provide the media with a quickly accessible database of weather events going back over a century, enabling them to judge whether the latest storm really is the worst in 100 years.

In this age of clickbait journalism, however, I’m not holding my breath!

rescue