005…UK Climate and Control Charts

From the Met Office 19 June 2013
The UK Met Office recently held an emergency meeting:
“Weather and climate experts from across the UK came together at the Met Office’s HQ in Exeter today for a workshop to discuss the recent run of unusual seasons in Europe.”

“Today’s included sessions which looked at the weather patterns and their potential causes in three recent seasons – the cold winter of 2010/11, the wet summer of 2012, and this year’s cold spring.”

From personal experience and a casual glance at the Met Office temperature figures it struck me that the variations seemed pretty much what we expect from our UK weather.

As a manufacturing engineer I was introduced to the wonderful world of control charts so I decided to construct these for the cold winter of 2010/11, the wet summer of 2012, and this year’s cold spring.

For those not familiar with a control chart an example is deciding if variations in a machine’s output is just normal or if the machine needs adjustment.

To decide if everything is OK the Mean and Standard Deviations of the data are computed and the upper and lower control limits are calculated and plotted on the chart of the data. The control limits are the mean plus and minus three times the standard deviation. These limits represent the edges of the normal distribution curve, so anything outside these limits is considered abnormal.

As you can see below the variations are for the most part well within the control limits, suggesting that he weather process is normal and we don’t have anything to worry about.

When looking for an out of control process here are some simple tests

[These are taken from ASQ.org, other organisations have similar rules ]

1.  A single point outside the control limits.
2. Two out of three successive points are on the same side of the centerline and farther than 2 σ from it.
3.  Four out of five successive points are on the same side of the centerline and farther than 1 σ from it.
4.  A run of eight in a row are on the same side of the centerline.
5. Obvious consistent or persistent patterns that suggest something unusual about your data and your process.

If we apply these rules to the charts below what can we infer?

1.  Yes, naughty Mother Nature is out of control at least once.
2.  No
3.  No
4.  No
5.  I would hazard a guess that from what we see the temperatures and rainfall can sometimes hit extremes but the pattern is normal and chaotic and that the recent experiences are just normal

Interestingly for rainfall it seems a high is followed by a low then a high then a low and this pattern repeats with only a few cases where a high is followed by another high though there can be considerable differences between the highs and lows.

So why did the Met Office call the meeting to discuss the unusual weather? For recent Spring temperatures we had a gradual rise and then one low instance, and again with the Winter temperatures the recent low only followed a gradual rise and the history suggests it is not unusual.

The simple statistics suggest we have nothing to worry about and the World Meteorological Organisation (WMO) requires the calculation of averages for consecutive periods of 30 years, so why did they panic with just a few years data?

1 pic2 pic3 pic