Click here for HadCRUT Data Rank Analysis (I)
Click here for HadCRUT Data Rank Analysis (II)
Click here for HadCRUT Data Rank Analysis (III)
Click here for HadCRUT Data Rank Analysis (IV)
Click here for Results of HadCRUT Data Rank Analysis (V)
This is the fifth and final posting in a series analyzing the information that can be obtained from the available HadCRUT data up to December 2007.
In summary: the world does appear to have warmed (but by the same token, it has cooled considerably during the year 2007). There are strong indications that it has been a very much hemispheric phenomenon, with little seasonality and hence minimal if any contribution from CO2. Likely culprits are therefore hemispheric-wide effects, such as those caused by the Sun and land use.
Note that the HadCRUT data has just been updated to January 2008, with the following anomalies:
Land NH: 0.058 (coldest since 1989; 3rd coldest since 1983)
Land SH: 0.058 (suspiciously identical to NH’s; coldest since 1986)
Land Global: 0.058 (not hard to guess; coldest since 1985; 3rd coldest since 1979)
Sea NH: 0.200 (only 0.003 warmer than the coldest value this century)
Sea SH: 0.027 (coldest since 1986; 3rd coldest since 1980)
Sea Global: 0.114 (3rd coldest since 1991)
TWO ISSUES I WILL NOT DISCUSS HERE: (a) the meaning of using 3 decimal digits; (b) the meaning of obtaining a Global sea-surface temperature value simply as the arithmetic mean of Northern’s and Southern Hemisphere’s
In HadCRUT Data Rank Analysis (I) we have seen that 2007’s yearly temperature averages have been among the top-11 ever but broke no record.
In HadCRUT Data Rank Analysis (II) the 2007 monthly temperature averages are shown as having broken only one record (January’s, for Land/Northern Hemisphere), with rankings getting higher and higher over the year reaching as much as #34 for December/Sea surface/Southern Hemisphere).
In HadCRUT Data Rank Analysis (III) a plot of the yearly temperature averages’ rankings shows a clustering of warm years in the past couple of decades, although most graphs have a “capped” shape suggesting the maximum values have already been reached, at least for now. The steepest gradient in terms of rankings is by the way between 1910 and 1938, again suggesting we may be experiencing just the upper end of a temperature peak. Finally, graphs are much similar intra-then inter-hemispheric (Land NH looks much more like Sea NH than Land SH).
In HadCRUT Data Rank Analysis (IV), the analysis moves a step further by comparing seasons. Correlations can be divided in three groups: Land/Sea, same hemisphere (between 80% and 98.6%, in all seasons); Season-to-following-Season (between 71% and 80.5%, land/sea, all hemispheres); and Season-to-Season (between 64.5% and 78.8%). In other words, there is a much weaker link between the ranking of, say, Northern and Southern Hemisphere Land Spring temperatures (intra-seasonal), than between the ranking of, say, Northern Land and Sea Spring temperatures (intra-hemispheric).
The last result does not look obviously compatible with the theory that world-wide warming has been caused by CO2 and other so-called greenhouse gases. If that were true, the intra-seasonal correlations would be higher, as CO2 concentration has a strong seasonal component: instead, they are generally even lower than the link between, say, Winter temperatures to Spring temperatures.
The main variable to factor is in all evidence the hemisphere. Now, what can have an effect on sea and land temperatures all over a hemisphere, and in a different way all over the other hemisphere? First hypotheses could be the Sun, via some kind of different coupling with the Northern and Southern terrestrial hemispheres; or changes in land use, that do not have the same impact on land-rich NH compared to ocean-rich SH.
Climatology would be a very exciting science indeed, had it not been hijacked by people on a mission to save the world based on inexact data.