2008 ended with another spotless day bringing the total number of sunspotless days for December to 28 and for the year to 266, clearly enough to make 2008, the second least active solar year since 1901.
See larger image here.
The total number of spotless days this solar minimum as of the end of 2008 stood at 510 days since the last maximum. The earliest the minimum of the sunspot cycles can be is July 2008, which would make the cycle length 12 years 3 months, longest since cycle 9 in 1848. If the sun stays quiet for a few more months we will rival the early 1800s, the Dalton Minimum which fits with the 213 year cycle which begin with the solar minimum in the late 1790s.
Long cycles are cold and short ones like the ones in the 1980s and 1990s are warm as this analysis by Friis-Christensen in 1991 showed.
They noted in subsequent papers that temperatures were rising faster than the solar cycle length would imply perhaps showing the anthropogenic signal, however a lot of that is the bad temperature data. Station dropout in 1990 of 2//3rds of the world stations (most rural), a tenfold increase in missing data, insufficient or no urban adjustment and bad siting, all of which led to exaggerated warming.
The station dropout was not evenly distributed worldwide. There were many more from the Former Soviet Union (Siberia). The temperature bars are not proportionally indicative of world mean temperatures because of the changing distribution but they are an indication of more urban and less rural stations. Also it is clear the distribution of the rural stations appear to be more towards lower latitudes warmer stations (greatest dropout was in Siberia). This was shown in an analysis by Jonathan Drake below.
MSU satellite temperatures rose during the 1979 to 1998 period but at about half the rate of the station based data bases. The enhanced warming in the 1980s and 1990s was related to the predominance of El Ninos thanks to a positive PDO possibly driven by enhanced solar.
The warming was supported by short warm cycles 21 and 22. The length from max to max of 21 to 22 was 9.7 years and cycle 22 length min to min 9.8 years, both very short suggesting warm temperatures in the 1990s.
The interval of cycle 22 max to cycle 23 max centered in the mid 1990s began to increase at 10.7 years and the min to min length of cycle 23 is now at least 12.3 years. With the Wigley suggested lag of sun to temperatures of 5 years and Landscheidt suggested 8 years, a leveling off should have been favored around 2000-2003 and cooling should be showing up now and indeed it is as the following shows.
Looking ahead, put that together with the flip of the PDO in the Pacific to cold and now a cooling of the Atlantic (AMO actually turned negative in November) and you have alarming signals that this cooling of the last 7 years will continue and accelerate. The sum of the AMO and PDO has been shown to be associated with temperatures. When the net is positive, temperatures are warmer than when the net is negative.
The sum has turned sharply negative for the first time since 1994 and especially during the cold 1960s and 1970s.
See this summary which concludes the same using the PDO and AMO.
So with a super long cycle 23 and cooling oceans, we have indications that temperatures are more likely to cool than warm in the years ahead. It will show up first in the satellite data which is uncontaminated.