Skip to content

Sunshine Hours Increasing In Holland As Well

April 6, 2015

By Paul Homewood

 

h/t Sunshine Hours

 

th

 

Recently, I have been looking at the increase in sunshine hours in the UK over the past couple of decades or so. Philip Eden certainly believes this phenomenon has been caused, at least in part, by reduced levels of pollution, not least because the biggest increases have been recorded in urban areas.

In addition, there was a Met Office paper in 2006, Climate Memorandum 21, which also found a pattern of increased sunshine, and speculated that there could be a connection with cleaner air.

My attention has now been drawn to another study published in 2012, this time in the Netherlands, which finds a definite connection between cleaner air, more sunshine and higher temperatures.

 

image

image

image

http://www.staff.science.uu.nl/~delde102/CleanerAirBetterViewsMoreSunshine.pdf

 

These graphs from the report show clear trends in pollution, visibility and sunshine.

 

 

image

image

image

 

The authors believe they can rule out changes in wind patterns as the cause of these changes:

Though yearly variations of the wind regime are fairly large and the data shows some long-term variations, hardly any trend was observed since 1955 or during the dimming and brightening periods. Therefore, it seems unlikely that changes in the wind regime have been an important contributor to the increase in visibility since 1985.

 

Interestingly, they find the biggest increases in sunshine hours is during the summer. This is in contrast to the UK, where the big changes are in spring and autumn. (There is evidence that UK summers have become wetter since the 1980’s. As sunshine hours have remained flat, this suggests that changing weather patterns have cancelled out the effect of cleaner air).

 

The paper concludes:

 

image

image

 

The evidence is piling up that there has been a significant increase in sunshine hours since the 1980’s in Europe, and that this is closely linked to cleaner air. Furthermore, this effect is likely to be responsible for a significant part of the increase in temperatures observed.

Put another way, we are getting the sort of temperatures now that we might have had half a century or more ago, if the air had been as clean then as it is now. And, of course, air pollution in Europe was getting worse from the 19thC, and arguably restricted the temperature increase coming out of the Little Ice Age. (The chart below is from a 2004 paper – since then, it is estimated that European sulphur dioxide emissions have fallen by at least another 30%, thus bringing them down to close to 19thC levels. It is also estimated that, since 2005, reductions in US and European emissions have more than compensated for increases in Asia, thus lowering global emissions).

 

image

http://www.pnl.gov/main/publications/external/technical_reports/PNNL-14537.pdf

19 Comments
  1. robinedwards36 permalink
    April 6, 2015 6:08 pm

    I notice that for yearly average sunshine there seems to have been an abrupt change of behaviour in the second half of the 1980s. I am looking for some sort of explanation for the widespread temperature step change that I detect in 1987. Could this be oit?

  2. April 7, 2015 2:32 am

    I partially disagree.

    The consistent arbiter is burning a line on paper where slight variations in sun strength has little effect, primarily sunny or not. In recent years the equipment has changed but jolly well ought to act that same. (eg. WMO reject electronic anemometers for this reason, not equivalent)

    An effect might be a reduction in fog / mist which anecdotally I’ve noticed. This might be temperature related.

    If it is smoke the effect in England would have been clear and starting much earlier.

    If you can find data from an isolated island, is the same effect there in a ocean environment?

    I don’t know the answer.

    I do have a lot of sunshine data somewhere from a different work. Not sure enough is long record.

  3. April 7, 2015 8:07 am

    Andy55 says
    What happens if you use 1975, 1983, 1992, 1997 for example.?

    Henry says
    That would be a different exercise but not necessary. Namely seeing that R2=1 you can accept that the given formula for the curve is correct. I can therefore predict the EXACT speed of the warming/cooling in K/annum at any point 1973-2015, as far as the rate of change in minimum temperatures is concerned, by calculating it from the formula for the curve that is shown.
    It would be great if Paul would consider publishing my results of my tables then I can elaborate more on how I obtained them. [I had asked him but he did not give me a reply]

    • AndyG55 permalink
      April 7, 2015 9:17 am

      Might be pure luck. It only takes 3 points to define a parabola…

      … and why do you think a parabola is ever relevant in anything to do with climate ?

      Why did you choose the years you are using?

      They are not even evenly spaced,…. please leave cherry picking to the alarmista !!

      Step each year forward 1, then 2, then 3 , them 4 years etc.

      See if your analysis has any robustness at all.. I somehow doubt it.

      • April 7, 2015 10:00 am

        there is no luck in statistics, provided you followed a random sampling technique, and followed a sturdy scientific process, to get to your sampling procedure.
        For the drop in maximum temps. I obtained r2=0996
        For the drop in means I obtained I obtained r2=0.970
        For the drop in minimum temps. I obtained r2=1.000
        Each graph has 4 measuring points.
        3 x lucky? on a random sample of 27 weather stations NH and 27 stations SH?
        impossible….

    • AndyG55 permalink
      April 7, 2015 9:18 am

      “Namely seeing that R2=1 you can accept that the given formula for the curve is correct”

      Sorry, but No !!!

      • April 7, 2015 10:07 am

        some people are 100% lucky because they know what they are doing.
        If you measure anything, and you have 4 final measuring points out of a sample of 54 (x 365d x [2015-x] yrs) and it gives you a perfect relationship r2=1 then you can accept that any statistical test for r2 will show relevance.
        This is no rocket science. This is just basic statistics. .

      • robinedwards36 permalink
        April 7, 2015 10:34 am

        If you are examining observational or experimental data by regression and obtain r-squared = 1.0000 you can be absolutely sure that the data were either fiddled or imprecisely measured and/or recorded. It matters not how many data points you have used. Fitting a parabola with four data points is perfectly OK, but beware of making inferences from it unless you /really/ understand what you are doing. You will find that the critical 95% t value for a parabola computed from four points is 12.7. In other words the coefficient has to 12.7 times greater than its standard error (do you know how to compute that?) for it to be “significant” at that level. Your have only one degree of freedom for the error term of the regression – clearly insufficient for a robust estimate of the fitting/experimental error estimate. If you compute a parabola from three data points you have absolutely no estimate of the errors, and any predicted value is totally useless. R-squared will be indentically 1 no matter what your data are. My software warns those who try such a silly thing that their computations are worthless.

  4. April 7, 2015 10:51 am

    @Andy
    Anyway, not only do I trust my own results, I can relate them exactly to what is happening on the sun. You can easily imagine a mirror of the graph

    happening in the future, as part of an a-c wave (which is what the Gleissberg cycle is, really).

    So, technically speaking, looking at energy-in, coming from the sun, [=not earth’s output], global cooling already started in 1995 and [we hope] it will turn up again some time 2016. However, global cooling will still carry on until around 2038.

    • April 7, 2015 10:55 am

      happening in the future, as part of an a-c wave (which is what the Gleissberg cycle is, really).

      should read:
      happening in the future, hyperbolic, rather than parabolic, as part of an a-c wave (which is what the Gleissberg cycle is, really).

  5. April 7, 2015 11:01 am

    @robin

    You are most welcome to repeat my test with another random sample of 50 or so weather stations. I am convinced you will get the same results.

  6. April 7, 2015 2:29 pm

    here are my results for southern Africa, change in minimum temperatures only

    1) Note with me that apparently we never had any warming here in southern Africa.
    2) Observe with me that every place on earth is apparently on its own particular curve, depending on the composition far up high, TOA.

Comments are closed.

%d bloggers like this: