And finally, as the last of the Weather Eye solar sunspot cycle season, let me tell you about a discovery which suggests that sunspots may in some way be connected to our weather after all.
Sunspots, as we have noted, are dark areas on the solar disc that increase and decrease in number in a rhythmic cycle of about 11 years, and we are now, according to current predictions, approaching the peak of cycle 23.
It is also known that when there are many sunspots to be seen, slightly more energy than usual is emitted by the sun. There has long been an ambition to relate our weather to these indicators of variation in solar activity, but all attempts to find an 11-year cycle in the weather have been unconvincing.
A few years ago, however, two Danish scientists approached the problem from a different angle. Rather than try to relate the weather to sunspot numbers, Eigal Friis-Christensen and Knud Lassen concentrated instead on the length of the sunspot cycle.
No one had ever thought of doing that before. The sunspot cycle, you see, is not exactly 11 years; it varies between 10 and 12, and sometimes even more widely. The Danes hypothesised that a long cycle might be indicative of a "weak" sun and that a short cycle might coincide with high values of the solar constant.
Lassen and Friis-Christensen noted the varying lengths of about 12 solar cycles over 130 years or so, and when they matched these figures to the average global temperature, they found that there was very close agreement.
The cycle was almost 12 years long in the early 1890s when global temperatures were low; it gradually shortened to 10 years during the first four decades of the 20th century, while the Earth was getting warmer; and since then, the pattern of solar cycle length has coincided very closely with that of global temperature. The implications of this startling finding are still being analysed around the world.
More recently, temperature readings at Armagh Observatory were subjected to a similar analysis, and the results were very interesting. First, the correspondence between the annual average temperature at Armagh and sunspot cycle length was as obvious and dramatic as it had been in the case of the Danish study; as the cycle shortened over the years from 11.5 to 10.5 years, the average temperature increased by almost one degree, and as the cycle lengthened again, the temperature decreased.
Moreover, temperature values for Armagh were available for 20 years further back than those available to the Danes, and it was found that the same relationship extended back to 1844.