Analysis: CRU tosses valid 5 sigma climate data

Am I missing something? I do not any labels for the lines in three of Mosher’s graphs.

Steve Mosher: Please add title blocks to your graphs that specify what the graphs represent and the color coding of the datasets. I have no idea which dataset is which.
Title blocks are time consuming but they give the viewers an idea of what you’re illustrating. The more info the better.

Five sigma from what, a thirty year average?
[REPLY – HadCRUt goes back to around 1850. ~ Evan]

More BS from the CRU. The “consensus” continues to crumble.

You know, it would be nice if we had some sort of profession where people gathered data, tried to ensure it was accurate, and recorded and disseminated such data along with trends and patterns. As opposed to fudging, deleting, massaging, and re-imagining data to fit an agenda.
I wonder what we would call such a profession?

You can see the code for this in the MET’s released version of CRUTEM in the station_gridder.perl
# Round anomalies to nearest 0.1C - but skip them if too far from normal
if ( defined( $data{normals}[$i] )
&& $data{normals}[$i] > -90
&& defined( $data{sds}[$i] )
&& $data{sds}[$i] > -90
&& abs( $data{temperatures}{$key} - $data{normals}[$i] ) <=
( $data{sds}[$i] * 5 ) )
{
$data{anomalies}{$key} = sprintf "%5.1f",
$data{temperatures}{$key} - $data{normals}[$i];
}

It is well documented that the winter season has the highest variability.

It is immediately obvious from Table 3 that the year-to-year variability is much greater in winter months (especially January) than in the summer.

Jones, et al, 1982, Variations in Surface Air Temperatures: Part 1. Northern Hemisphere, 1881-1980

In the Northern Hemisphere, Figure 1 shows that year-to-year variability is clearly greater in winter than in summer.

Jones, et al, 1999, Surface Air Temperature and Its Changes Over the Past 150 Years
As to documenting the standard deviation drops …

To assess outliers we have also calculated monthly standard deviations for all stations with at least 15 years of data during the 1921–90 period. All outliers in excess of five standard deviations from the 1961–90 mean were compared with neighbors and accepted, corrected, or set to the missing code. Correction was possible in many cases because the sign of the temperature was wrong or the temperature was clearly exactly 10deg or 20degC too warm or cold. After this step the 1961–90 normals and the 1921–90 standard deviations were recalculated. Exactly how many values were corrected or set to missing is impossible to determine, as we do not know how many were corrected in the component datasets. We made changes to about 500 monthly averages, with approximately 80% corrected and 20% set to missing. In terms of the total number of monthly data values in the dataset this is a very small amount (<0.01%).

Jones and Moberg, 1992, Hemispheric and Large-Scale Surface Air Temperature Variations: An Extensive
Revision and an Update to 2001
Jones notes in an earlier paper that the corrections were prevalent for Greenland. These wx record coding errors remind me of the sort of thing that Anthony himself was pointing out here.
Does Jones inappropriately drop or adjust Feb 1936 records? Someone should go look. Most of the CRU data sets are available here:
http://www.metoffice.gov.uk/climatechange/science/monitoring/subsets.html
———–
Just a side note, Mosher’s stuff was posted prematurely with unlabeled graphs and abbreviated discussion (for instance – which paper is ‘the paper’ reread by Mosher). Given how careful he was with the response to McKitrick, I wonder if he personally approved posting these notes in draft form?

For the love of God, please put comprehensive labels on the axes, and include keys when there are multiple graphs in one plot. You could be reaching a MUCH larger audience with just a little more work. I’m a Ph.D. research scientist, and if I can’t discern what’s in a graph, what hope is there for the average first-time visitor?

What do the various colors signify on the graphs?

Although this paper is not adequately illustrated and summarized, I get enough to see an interesting problem with tossing 5 sig for the future! If we are to have runaway global warming, soon the temp anomalies will be more frequently over 5 sig – are they going to toss these, too? You can bet your life they wont. This is a temporary thing to ensure the trend until they don’t need it anymore to maintain the present hotter than before, then, like GISS, they will re-adjust the earlier numbers.

Sorry Bob,
The Plots are just made on the fly as I walk through the data, For the chart in question, Red is the MAXIMUM departure from normal, blue is the minium. orange is +- 1.96sd, black is the mean. “index” is month from 1-12. For me this is just a curiousity that other can go take a look at as I’m working on debugging other stuff.
The procees of throwing out 5 sigma events has a seasonal bias

That’s a lot like what’s happening now. Not quite as bad yet.
What was so significant about this cold wave was that the 1930s had some of the mildest winters in the US history. 1936 was also one of the coldest years in the 1930s. And the winter was followed one of the warmest summers on record which brought on the 1936 North American heat wave.
That’s classic weather perturbed by volcanoes, and there were a lot of them in the ’30s.
http://www.volcano.si.edu/world/find_eruptions.cfm