Aug 12 2007
More GISSTemp and Surface Station Stuff
If you’ve had enough of the bloghorrea lately about the surface station and GISStemp controversy, read no more. If you think those that tout this as “important” (read: Roger Pielke Sr.) should be smashed into the ground, I offer another arrow to pop their story.
The folks at RealClimate have a nice blog post about this, as does Tim Lambert
On Friday I showed that an independent measurement of the lower troposphere temperature also shows a rise in the global mean temperature. This was from the UAH MSU data, arguably the more “conservative” of the two MSU data products. I haven’t compared the UAH and RSS data since UAH found and corrected an error, but I would guess that they are pretty much the same now. If you weren’t convinced by a totally independent product showing a global warming, then you probably won’t be convinced of anything. However, today I’ve plotted the UAH MSU data (taken from satellites, remember?), and the GISS temperature on the same plot.
|
Now, we can clearly see that they both are rising at the same rate, and that correlated interannual variability is captured in both data sets.
Going back to last posts question, what if the GISSTemp had a significant bias such that the observed warming is not due to any physical change in the system, but is just a consequence of systematic errors? We now see that satellite measurements of the temperature are showing the exact same rate of warming. If GISSTemp systematic errors, what possible reason could there be for the satellite data to have unknown systematic errors that result in the exact same temperature errors? There is none. This surface station BS needs to be squashed.
[Update: August 19]
As promised, a zoomed-in version of the above plots. Not only that, I plotted the average monthly temperatures instead of the average yearly temperatures.
|
The trend of 0.143C/decade is for the black curve which is the MSU data from UAH. The red curve is the GISSTemp data, and it has a different trend, about 0.17C/decade. This represents about a 15-20% difference in the two trends. The offset is actually present in the data, and is because the average temperature was just calculated for different time periods for the two data sets. The difference was left in this plot to make it look prettier.
|
Here we see a scatter plot of the MSU and GISS temperatures. I removed the artificial offset such that the intercept was zero. The red line represents the best fit between the data, and the green line represents a slope of 1. The correlation between the UAH MSU data and the GISStemp was 0.81. I too am not a stats guy, but even without fancy statistical tests, I think it’s easy to see that the slope is not significantly different than 1.
(No Ratings Yet)
Loading ...Related Posts:
4 Responses to “More GISSTemp and Surface Station Stuff”
To reduce spam, comments are automatically closed 30 days after the last comment. If you would like to comment on any closed thread, please use the contact form at the top of this page.
See also Tamino’s thorough post on the same subject. Excellent blog, BTW (found via Michael Tobis). I return now to the vineyards of denialism.
[Response: If the wine is good, send me a few bottles.
]
A. Could you (pretty please) show the graph starting in 1970 (blow up the portion where there is a comparison)?
B. IANASJ (I an not a stats jock), but how about some metric (or three) to indicate the correlation? Rsq? Difference plot?
[Response: Sure, and sure. But I won't have time until tomorrow or Monday.]
Good job man. Now can you train Steve to also make more definite mathematical statements. He seems to be more interested in little pictures and silly blog posts for effect. but it is obvious that he knows the literature, knows how to play with data, and could make meaningful contributions (even from a skeptical side) if someone would just hold his ass to the fire to make definite hypotheses and quantify things mathematically.
[...] the UAH group corresponds to 0.139 degrees Celsius per decade. I’ve previously shown that the surface station data (as analyzed by the team at GISS) show a rate of warming of 0.17 degrees Ce…. If we take it one step further, we can look at the temperature changes observed as a function of [...]