Dear Reader

This is the first post in a series which will present global temperatures in a different, yet correct way.

If you’ve been following the climate debate, you’ve no doubt seen the hockey stick graph, the iconic harbinger of dramatic changes in the global climate.

One line shooting upwards; clear, concise and beyond doubt: the science behind it is settled, and the time for debate is over. CO2 emissions are to blame for an accelerating rise in global temperature, which will wreak great havoc around the world.

The hockey stick graph has been extremely effective because of its simplicity. But is it correct ?

That should be easy to verify. Raw temperature measurements, going back to 1700, are available from the National Oceanic and Atmospheric Administration on a public FTP site. These data files contain up to half a million yearly measurements, month by month from thousands of weather monitoring stations. Each yearly measurement is one line in a data file, like this line from v2_mean, the file with unadjusted mean temperatures:

1016035500001984  125  104  125  150-9999  205  248  237  223  188  173-9999

The first column contains a weather station identifier and the year. The next twelve columns contain the temperature in tenths of degrees Celcius. -9999 signifies missing data. So far so good.

Now, how to get a grip on the data ?

A first approach was to import it into a spreadsheet and create a scatter plot, with the year on the X axis and the temperature value on the Y axis. Easy enough.  Alas, the poor spreadsheet program was overwhelmed and the lack of sensible presentation options made a mess out of what should have looked like a nice sports tool.

A few hours with a favorite application development environment resulted in more pleasing results. Here are the scatter plots with the measured temperatures for the months January, April, July and October from 1700 to 2009.

But where is the Hockey Stick ?

January Mean Temperatures

April Mean Temperatures

July Mean Temperatures

October Mean Temperatures


About this entry