Skip to content

Analysis Of February CET

May 12, 2016

By Paul Homewood 




I looked at DECC’s statistics on thermal growing seasons earlier, which suggested that seasons were growing longer, principally because of the earlier onset of spring. On average, it appears that the season tends to run from mid Feb to mid Nov.

Remember that the definition of a growing season is:





Earlier analysis of mine suggests that springs have become warmer in the last three decades mainly because the really cold days have become less frequent, rather than all days getting warmer, or there being more extremely warm days.

I wondered therefore whether the same applies to February temperatures, so I have done a couple of tests, using CET means.


First, I have looked at the difference between the high and low daily means for each February. As we can see, the gap between the two has been shrinking, when compared to the years prior to 1960.

This could, of course, mean either that lows are increasing or that highs are reducing (or both). Nevertheless, it is significant that temperatures appear to be less volatile these days. (We are often told that the opposite is the case!)






The second test is to plot the highs and lows for each year. 

This shows that high temperatures have changed little over the years. In contrast, we see that there is a clear upward trend in the lowest temperatures, something that tallies with my findings for March and April.




What has caused this change is another matter. But it is hard to argue that less volatile temperatures, combined with an absence of really cold days, is not beneficial.     

  1. May 12, 2016 5:53 pm

    Less ice in the Arctic probably would correlate with less cold northern winters. Let’s see what happens to Arctic conditions in the next 10 years or so.

  2. May 12, 2016 9:17 pm

    You adopt a very practical method of assessing what climate change means. Its much better than simply adding the maximum and the minimum and dividing by 2. The mean is a statistic of little real interest. A more analytical approach is required if we wish to assess how the weather is affecting the productivity of the local environment, the amount of heating necessary in winter and cooling in summer, or at a simpler level the type of clothing that will be necessary out of doors.

    Of course, the global mean is of no use whatsoever. Is it increasing due to a rise in the maximum or the minimum? Where exactly is change occurring? Is it winter or summer? Which months exhibit the greatest change?

    When you look at variability in temperature according to latitude you will discover that between the Arctic and 30° south its much more extreme in January and in the rest its much more extreme in July. Variability increases either side of the 30°south latitude band and is greatest in Antarctica in winter when the sun does not shine through to October.A similar situation applies in the Arctic. In general, the change that has occurred is beneficial because of the increase in the growing season due to winter warming.

    More here:

  3. Paul2 permalink
    May 12, 2016 10:45 pm

    Gridwatch tonight: Coal 0.0%.

  4. May 12, 2016 10:59 pm

    It is generally true that the problems of CAGW are going to be experienced by “other poor people”. The warmists do not support your observation re quality of life for Brits: that would be a “selfish” thought. The fact that the victimized poor Indians, Chinese, Indonesians and Middle Easterns and Africans don’t consider CAGW to be enough of a threat to curtail CO2 emissions by themselves is awkward but okay from the condescending, paternalistic warmist view: the rich white eco-greens know what is best for their former colonial subjects anyway.

    Despite the general acknowledgment that pluses exist as well as minuses, you won’t find a warmist breakdown as to who wins, only as to who loses. The message would not be Bill Nye acceptable, i.e. not in the “national” or global interest.

    For shame, Paul!

    BTW, this last, warm Alberta winter was fabulous for humans and urban rabbits, but please don’t let Leonardo DiCaprio hear I said it.

  5. May 13, 2016 1:33 am

    Nice work Paul. What’s not to like about milder winters in colder climates?

  6. Ibrahim permalink
    May 13, 2016 2:34 am

    Take a look at the AO (or NAO)

  7. May 13, 2016 7:57 am

    Paul, would it be possible to compliment this with a look at high’s and low’s for November, to see if the growing season is being extended at both ends or is shifting more in one direction than the other. Are the seasons wandering slightly in a cosmic rhythm?
    Nature ignoring man’s need for order would not be a precedent….

  8. Ian_UK permalink
    May 13, 2016 8:31 am

    Isn’t there a problem with assuming higher minimum temperature always = good? I understand that certain crops need frost and also cold weather helps kill off harmful bugs. Be careful what you wish for?

  9. robinedwards36 permalink
    May 15, 2016 4:09 pm

    Hello, Paul and Others,

    If you have the computational facilities I’d advise looking at the Feb Max CETs by regarding the data as belonging to two different regimes. By this I mean considering ony data prior to 1987 as coming from one population, with the later data from another one. Why? Because in the Autumn of 1987 there was an abrupt change in Feb Max temperatures, of the order of 0.8 C. On either side of this date FebMax showed declines, neither “significant” in the statistical sense. In fact, post 1987 the slope of the temperature data is very close to zero.

    Do not be surprised by this. In northern Euro virtually every site time series data that I’ve show exactly the same phenomenon. “The Pause” as far as Europe is concerned began in the Autumn of 1987. If you don’t believe this, simply collect the data – there are plenty of sources, so pick your own. Do a standard regression and note the highly significant parameter estimates, with the probability value for the time coefficient, if the software provides it. Now divide the data into two segments, slitting it as above, and do the regressions on the two segments. You will almost certainly find that post 1986 the slope is slight and probably not significant. Prior to 1987 you may well find a significant time coefficient, probably but not necessarily positive. Try this for places in the Netherlands, France, the UK Switzerland, etc etc.

    “The Pause” began in 1987!

    CET Feb Max and Feb Min behave effectively in an the same way as each other, as does Max minus Min. All show the 1987 step. So does CET itself. Try a few sites in the UK and Ireland if you are feeling industrious. The Met Office has plenty to choose from.


Comments are closed.

%d bloggers like this: