Tom Quirk has analysed variations within and between these data sets and concludes there is 1. Substantial general agreement between the data sets, 2. Substantial short-term variation in global temperature in all data sets and 3. No data set shows a significant measurable rise in global temperature over the twelve year period since 1997.
Global Temperature Revisited
by Tom Quirk
One of the most vexing things about climate change is the endless debate about temperatures. Did they rise, did they fall or were they pushed? At times it seems like a Monty Python sketch of either the Dead Parrot or the 5 or 10 Minute Argument.
However it is possible to see some of the issues by looking at the four temperature series that are advanced from:
GISS – Goddard Institute for Space Studies and home of James Hansen,
Hadley Centre – British Meteorological Office research centre
UAH – The University of Alabama, Huntsville, home of Roy Spencer with his colleagues including John Christy of NASA and
RSS – Remote Sensing Systems in Santa Rosa, California, a company supported by NASA for the analysis of satellite data.
The first two groups use ground based data where possible with a degree of commonality. However since 70% of the surface of the earth is ocean and it is not monitored in a detailed manner, various procedures with possibly heroic assumptions and computer modelling, are followed to fill the ocean gap.
The last two groups use satellite data to probe the atmosphere and with the exception of the Polar Regions which are less than 10% of the globe, they get comprehensive coverage.
One question is of course are the two groups measuring the same temperature? After all the satellite looks down through the atmosphere, while the ground stations are exactly that.
There is an important distinction to be made between measuring the temperature and measuring the change in the temperature. Since the interest is in changing temperatures then what is called the global temperature anomaly is the starting point. The issue of measuring absolute temperatures should be put to one side.
Data from 1997 to 2009 was drawn from the four group websites on the 28 April 2009. When data for 1997 to early 2008 was compared to data acquired in early 2008 differences were found as shown in the first table.
This is evidence of substantial reprocessing and re-evaluation of data. This is not unusual with complicated analysis systems but there is so much interest in the results that adjustments are regarded with great suspicion. This is the fault of those publishing the temperature data as they fail to make the point that monthly and even yearly measurements are about weather and not climate.
The latest series of temperature anomalies are shown in the graph where the monthly data has been averaged into quarters. All statistical analysis that follows is on the monthly data unless stated otherwise.
From inspection, there is substantial agreement over the years 1997 to 2008. This can be statistically measured through correlations. This is a measure of how closely related the series may be. A value of 0 implies independent series while a value of 1 implies complete agreement. The correlation in turn indicates the degree of commonality in the comparison.
It is important to note that the two satellite analysis groups draw measurements from the same satellites. So the differences in temperatures are a result of analysis procedures that are not simple. In fact corrections to the data have been the subject of exchanges between the two groups.
The ground based measurements also have a common data base but it is clear and acknowledged that the two groups have different analysis procedures. While the satellite analysis procedures have converged to reduce their differences over the last thirty years, this has not been the case for the ground based procedures.
It is also clear looking at the measurements that there are substantial short-term, say less than 2 years, variations over the period 1997 to 2009. In fact, while the overall monthly variations show a scatter with standard deviation of 0.20C, the month to month variations are 0.10C. This is a measure of features that are clear in the data. The short run sequences of temperature movement are a reflection of variability in the atmosphere from events such as El Ninos (1997-98) and La Ninas.
Looking for a simple trend by fitting curves through a highly variable series is both a problem and a courageous exercise. The results on an annual rather than a monthly basis are given in the third table. The problem of dealing with real short term variations was resolved by ignoring them.
So for twelve years there has been a rise 0.10C with a 140% error, in other words, no significant measureable temperature rise. You can play with the data. If you omit 1998 then you can double the change. But 1998 was an El Nino year followed in 1999 by a La Nina. If we omit both years then the results are unchanged.
However the lesson from this is to look at the detail.
There is so much variability within the 12 year period that seeking a trend that might raise the temperature by 20C over 100 years would not be detectable. On the other hand there are clearly fluctuations on a monthly and yearly scale that will have nothing to do with the predicted effects of anthropogenic CO2.
The twelve year temperature changes from the data of the four analysis centres reveal some possible differences. Since there is a high degree of commonality amongst the results, any differences may be systematic. Both the GISS and Hadley series show a larger temperature increase then the satellite measurements. This may be due to urban heat island effects.
Finally, if you are looking for temperature increases from CO2 in the atmosphere, then you should choose the satellite approach of measuring temperatures in the atmosphere!
Short term, less than thirty years, temperature series are not the place to seek evidence of human induced global warming.
Tom Quirk lives in Melbourne, Australia.
To read more from Dr Quirk click here http://jennifermarohasy.com/blog/author/tom-quirk/
The photograph is from Anthony Watt’s website that details his program of photographically surveying every one of the 1221 USHCN weather stations in the USA which are used as a “high quality network” to determine near surface temperature trends in the USA, read more here http://wattsupwiththat.com/test/ .