Met Office Temps

http://theweatheroutlook.com/twocommunity/forums/t/23028.aspx

I haven’t a clue how many of you look at the Met Office site or even the global temps they show. Leaving aside the mediocre coverage of Canada, Africa and Eurasia in fact everywhere on the planet, even though they must have the data, I am intrigued as to why they are so shy about posting Low Temperatures. To me that is figures in the double minuses Celsius.

I’ve noticed over the past two years that whenever the temperatures drop below minus 10C they tend to be left off the current temps on these pages.

http://www.metoffice.gov.uk/weather/europe/europelatest.html

and you can see the same for N. American temps as well. As I write the N. American temps have risen so they are shown as of Christmas Eve but only relatively mild temps at say Boston. On the other hand there is no bashfulness about Double Pluses Celsius and temps of say plus 10C to 30C+ are invariably present when they occur. Thus in the N. Hemisphere summer, the temps are invariably shown without any ‘n/a’ (data not available) explanations. To be fair I have noticed occasions when there are n/a explanations but these are rare, unlike in the winter when they are commonplace.

Recently it’s been chilly in North America even by their standards. Kansas, Chicago and Boston have sometimes been below minus 10C yet these figures were not available on the Met Office site at the time. They come back again once the temps have risen above the Minus 10C mark. Of course as anyone here knows you can check elsewhere which is what I started to do some time ago (one and a half years ago) to test my hypothesis that something was not quite right with what was shown on the Met Office site. As another example of this glitch, everyone knows who follows weather knows that Western Russia has had a mild autumn and these temps were invariably reflected on the Met Office site since the summer without any problem as far as I noticed.

If you use this PLEASE DO NOT use my name or email address.

Regards

[snip Then why did you add your name here or in your sign on? Seriously this job is enough work as is ~charles the moderator]

I live in Torquay some 10 mins from the harbour. Now which Pub would that be….

Steven,
You are correct but I was referring to the Hadcet numbers which you can download from here http://hadobs.metoffice.com/hadcet/ these are representative of a roughly triangular area of the United Kingdom enclosed by Lancashire, London and Bristol. I use these numbers because they go back to 1659 and let me look at the really long term trends.

I think there has been marginal level of development in this area since 1659.

It’s was very “basic” research – for a few weeks I made notes of the 3 day forecast for both services and found Accuweather was far more accurate.

No one believes the Met Office 3 day forecast – people just about believe the next day one.

writes

“Note, too, that the long-term mean alluded to by Oldjim includes 200 years’ worth of the Little Ice Age.”

No it doesn’t. The mean referred to is for the period 1971-2000 and covers the UK as a whole (i.e. it includes Scotland, N Ireland & Wales). The figures you posted are from the Central England record which is generally warmer than the rest of the UK. Other posters have made the same mistake.

Don’t get me wrong, I’m not a supporter of CO2 warming but, like it or not, despite all the rain and the lack of sunshine, UK temperatures have been around (and possibly slightly above) average this summer.

Other criticisms of the MET Office are valid.

A nice little slice of data. The significance of which should not be understated.

The important property it illustrates is the inherent failure that will occur in times series forecasts with strongly endogenous explanatory variables.

Imagine the Met was trying to forecast the temperature trend over the forthcoming decade. WHen they do that they don’t get the opportunity to reset their baseline each year, so any errors roll over. If there is serial correlation in said errors you end up with a bias.

So ten years ago the Met would have said that the temperature anomoly was going to increase by 0.6 degree s over the coming decade( sounds pretty familiar). Of course their forecast would only have represented the compounding errors (bias) in their model.

Unfortunately, this doesn’t become completely provable until the 10 years elapse and the data are in. By then of course, the warmeners have “moved on”.

Do you think that scenario sounds awefully like Hansen’s 1988 forecasts???