WUWT reader “Ari” tips us to this interesting puzzle on the Central England Temperature series and the Met Office temperature series in the UK doing some wonky diversion.
A problem with Temperature
I don’t think people will be able to enlighten me on this one, but probably should be aware of it. It’s a bit technical.
If you work out the CET annual temperature anomalies and then the METO England annual temperature anomalies and deduct one from the other, you should just get a line bouncing around the zero line (unless the climate of one is moving out of phase with the other). If you take 1961-90 as your base period and deduct CET anomaly from England anomaly for all 101 years during which the England series has existed, you get this:
Where the line is above the zero line, that’s a year when the England anomaly is warmer than the CET anomaly and vice-versa. For the 101 years of the METO England series, you get the bouncing around zero that you would expect, then for the last 4 to 8 years, the CET has been growing notably colder than the England anomaly.
So, it seems that one of the series is going astray a bit. I think Essan might chip in here and say the revised CET stations are too cold, but even if you would use the Eden series, you would get something similar (about 0.1c less). Note that this doesn’t mean that it hasn’t got cooler in the last few years. But, according to the METO regional series, it hasn’t cooled by as much.
Here’s the same graph with the red line showing the same thing for the Eden anomaly. The same problem is evident:
I have looked at your explanation and compared the CET anomaly with just the METO series anomaly for the Midlands. You still get the same effect:
BUT, curiouser and curiouser, if we do the same thing with the CET vis-a-vis the METO series for Scotland, Wales and Northern Ireland, we don’t get this problem! Instead, we just get the bouncing around zero that we would expect. Of course, the bounces are of greater magnitude as would be expected as places further away from CET have greater deviations from it on an individual year basis. Here, for example, is the graph for Wales anomaly minus CET anomaly:
So, we now seem to have reached a position where the CET is remaining in relative phase with the METO series for Scotland, Wales and Northern Ireland, but has become colder than the METO series for England (I have also tried the METO series for England North and England South and you get the same thing). Or, rather, the METO series for England (and English regions) is moving out of phase with everything else.
Why this is happening remains the mystery.
You can read the rest of the thread here, ideas and comments welcome. To me, it looks like a case of UHI pollution in one series and not another.