North Wales ( Gwynedd) Winter and Summer Temperatures since 1917 influence of Solar Cycle and prediction of a QBO ( quasi-biennial oscillation) and definition of its average length, by Dr Chris Barnes, Bangor Scientific and Educational Consultants, April 2015.

Abstract

Control and feedback mechanisms in the earth climate system are very briefly discussed. NAO and the QBO are strong indicators of winter temperature in Wales but are rather more irrelevant in summer. Thus a search for a solar link to summer weather is made as an alternative. This suggests Gwynedd’s recent extremes of weather could be due to changes in the solar cycle. Analysis of a century’s worth of UK climate anomaly data suggests that a shorter or longer than average solar cycle gives rise to an increased incidence of both colder than normal winters and hotter than average summers with a very high statistical result for summers. The present work also suggests that Gwynedd will be on average 1.07 C warmer which is closest to the IPCC B1 scenario and at the lower end of their predication scale but that most of the change could be solar induced. It is incredibly instructive to remove the data for the last and very unusual 14 yearlong solar cycle. If this is done then the warming slope changes into a dramatic cooling slope showing a change of -5.7 C in the next 100 years. The data for winter and summer temperature anomaly also allows extraction of a sinusoidal varying QBO like component with a length of approximately 24.9 months if one constrains the solar cycle to its average length.

Introduction

There has been much debate recently regarding the extent to which climate is under solar control and the extent to which anthropogenic change is responsible.

The present author has recently explained how earths’ climate could be geo-magnetically controlled [1] and further has explained how a number of hitherto unsung feedback mechanisms appear to be preventing ‘runaway’ global change [2]. Such mechanisms include ship and aircraft aerosol, increased plant terpene production, and increased lightning NOX etc. The former is only possible if there amplifiers in the ionosphere-stratosphere-troposphere coupling system. Since this system is highly non-linear in all its wave-wave coupling scenarios including; acoustic, gravity wave and electromagnetic wave and in the DC global electric current then the presence of such solar influenced amplification mechanisms is perhaps not so unexpected.

Furthermore the present author has recently shown that both the NAO and the QBO are strong indicators of winter temperature in Wales but are rather more irrelevant in summer [3]. Thus it was decided to explore the influence of the Solar Cycle on winter and summer temperature anomaly between 1917 and the present day.

Data Sets

Only two data sets are required. Firstly, the temperature anomalies and secondly the dates of the solar cycles. In this case the solar cycles being defined by sunspot number rather than Ap value. These are available from standard online sources.

Data Summary

The data have been summarised in an XL file.

Max	End	Length max to max		Coldest winters		Hottest Summers
					yrs+max
1906	1913
1917	?	11		1917	0
1928	1933	11		1929	1	1933	min
1937	1944	9		1940+41	3+4	1934+5	min+1+2
1947	1954	10		1951	4	1947+1949	0+2
1958	1964	11		1963	5	1955+1960	max-3+2
1968	1976	10		1977	9	1975and 6	min and min-1
1979	1986	11	1985	1979+1982	0+3+6	none
1989	1996	10		1991+1996	0+7	1989+90+95	max+1+6
2000	2008	11		none		03+04+06	max+3,4,6
2014	?	14		2010+2011	minus 3+4	2013+2104	max+max-1

Table 1: Main data set

From the above, a dominant mode for the appearance of hot summers would appear to be solar maximum plus an average of circa 2 years and strangely also at solar minimum. A dominant mode for the coldest winters would appear to be on average about 1.5 years before solar minimum. Van Loon and Jeffery C. Rogers (1978) [4] have remarked upon ‘The Seesaw in Winter Temperatures between Greenland and Northern Europe.’ M Lockwood et al 2010 [5], have suggested that as a strictly ‘European’ effect colder winters are associated with Solar Minimum. Kodera (2002) [6] has discussed the ‘seesaw’ in terms of the NAO and AO. QBO may also be relevant, Labitzke (2005) [7] has discussed this. At 1.5 years before solar minimum the extracted QBO signal is just swinging from +ve to negative in phase. For 75% of specific winters examined here QBO is negative. At solar maximum plus 2 years the QBO signal is close to zero. In reality this was also the case with one of Britain’s hottest ever summers, that of 1976.

Further Analysis of Results and Discussion.

Length of solar cycle

It is intuitive to plot the number of occurrences of hot summers against length of solar cycle and number of cold summers against length of solar cycle separately. Considering cold winters first. A linear correlation plot shows no statistical relevance whatsoever. However and more instructively a quadratic correlation shows a weak but interesting effect, see Figure 1 below:

Figure 1: Solar cycle length versus number of Hot Summers in century

It can be clearly seen that the number of harsh winters minimises for the average solar cycle length of 11 years and maximises for cycles which are shorter and longer.

A long solar cycle is of course tantamount to saying there are longer periods with fewer sunspots as happened in the Maunder minimum etc.

Figure 2: Solar cycle length versus number of Hot Summers in century

Perhaps somewhat surprisingly the same general trend is seen for hot summers as for cold winters. The shorter and longer the cycle, the more hot summers are present. The difference is of statistical significance, as here the P Value Results r=.78 DF=10 give a two-tailed P value equals 0.0028and by all conventional criteria, this difference is considered to be very statistically significant.

Whereas with the data for cold winters because the regression coefficient was much lower P could not be defined with any certainty.

The above result is, indeed, highly significant. It tends to suggest that both longer and shorter than average sunspot cycles drive more extremes of weather. The harbingers of climate doom have been telling us for a while now about Britain’s more extreme climate and suggesting that it is down to anthropogenic change. The data acquired here represents 100 years of data tends to suggest that it still very firmly the sun and not human kind which are in control of the Welsh weather. It is interesting to note that the warming effect of short solar cycle length has been observed previously by Friis-Christensen and Lassen ( 1991) [8] but it is believed that this is the first time that such an effect has also been observed for a longer than average solar cycle. Estimates for the length of the next solar cycle vary considerably from as low as 12 years to as high as 17 years. Thus based on solar influence alone the Welsh climate could either return to a more stable regime or continue to be one of wild extremes. Sadly, and presently we are still too close to solar maximum to say with any certainty.

QBO

There has also been much debate recently about the significance of the QBO and if and when and where the QBO signal appears and to what extent it needs to be built into climate models.

Because the Met Office data set for temperature anomaly since 1917 is effectively a colour coded set of GIS like maps, see [9], it has not been possible to evaluate precise figures. However, the present author has taken a cold winter to be one on average equal to or below 1 C less than the figure normally expected and has taken a hot summer to be the reverse i.e. at least 1C greater than the average for the period as expected.

The data have then been treated in the following way. The sunspot cycle has been assumed to be its average value of 11 years long. Further the peak of the sunspot cycle has then been assumed to be central i.e. 5.5 years after the start of previous trough and 5.5 years before the next.

A graph has been constructed representing an idealised solar cycle with abscises 0-11 (years) and ordinates -1 to +1 (Celsius) , for all the available cold winter and hot summer data between 1917 and 2014 and a sinusoidal fit has been applied.

Figure 3: Attempting a QBO extrapolation

Approximately 5 cycles of QBO like behaviour are observed. Yielding a QBO period of between 24-24.9 months. Of course in reality neither the solar cycle nor the QBO is fixed and this accounts for the low correlation factor of some .25. Labitzke (2005) [10] was the first to introduce the constructed annual mean of the solar cycle–QBO relationship. This present work adds considerable weight to his idea.

The latest hypotheses as to the variability in such cycles involves gravitational and magneto-electric solar and inter-planetary beats [11], [12] and seem to the present author highly plausible yet none are taken into account in today’s meteorology.

Is the climate warming?

A unique way to make an overall assessment on climate is to re-plot the data from figure 3 using a linear extrapolation, remembering that it is data acquired across a total number of 11 complete solar cycles.

Figure 4 : Climate warming in last century

By 2099 the above warming slope predicts that Gwynedd will be on average 1.07 C warmer which is closest to the IPCC B1 scenario and at the lower end of their predication scale, see https://www.ipcc.ch/publications_and_data/ar4/wg1/en/spmsspm-projections-of.html [12].

Remembering the strong solar effect, figure 2, it is possible that far more of this warming has been due to the sun alone and far less due to anthropogenic causes than envisaged by the IPCC. It is incredibly instructive to remove the data for the last and very unusual solar cycle. If this is done then the warming slope changes into a dramatic cooling slope showing a change of -5.7 C in the next 100 years. To the present author it is almost as there is divine control at work trying to keep our planetary system stable.

Conclusions

The results of the present work leads to a conclusion highly suggestive that Gwynedd’s recent extremes of weather could be due to changes in the solar cycle. Analysis of a century’s worth of UK climate anomaly data suggests that both a shorter or longer than average solar cycle gives rise to an increased incidence of both colder than normal winters and hotter than average summers particularly with a result of very high statistical significance for summers.

The data for winter and summer temperature anomaly also allows extraction of a sinusoidal varying QBO like component with a length of approximately 24.9 months if one constrains the solar cycle to its average length. This length is very comparable with the standard accepted range for QBO periodicity.

A linear extrapolation of the same data allows an estimate to be made for climate warming which is closest to the IPCC B1 scenario and at the lower end of their predication scale.

It is hoped that the above public domain publication will advance the frontiers of general and meteorological science and the author is willing within reason to attempt to answer questions from most quarters.

References

1. http://www.drchrisbarnes.co.uk/SOLARMAG.htm

2. http://www.drchrisbarnes.co.uk/GEOINDEX.htm

3. http://www.drchrisbarnes.co.uk/CLI.htm

4. http://www.geography.ohio-state.edu/faculty/rogers/pubs.html

5. Environ. Res. Lett. 5 024001 doi:10.1088/1748-9326/5/2/024001 http://iopscience.iop.org/1748-9326/5/2/024001

6. http://onlinelibrary.wiley.com/doi/10.1029/2003GL017290/pdf

7. https://www.terrapub.co.jp/onlineproceedings/ste/CAWSES2007/pdf/CAWSES_257.pdf

8. http://www.sciencemag.org/content/254/5032/698

9. http://www.metoffice.gov.uk/public/weather/climate-anomalies/#?tab=climateAnomalies

10. http://strat-www.met.fu-berlin.de/labitzke/summary/JASTP-Labitzke-2005.pdf

11. http://arxiv.org/pdf/1307.3706.pdf

12. http://www.drchrisbarnes.co.uk/PBMAG1.HTM

13. https://www.ipcc.ch/publications_and_data/ar4/wg1/en/spmsspm-projections-of.html [