Flat line or upward curve?

This is the famous the Hockey Stick chart from the 2001 IPCC report. It claims to show a significant and “unprecedented” rise in global temperatures. The methodology used to construct this chart has been heavily criticised (see here). How well does this chart actually capture the way temperatures have changed at various places around the planet over the last 100 years so?

One place to look is the National Climactic Data Center (NCDC) which is the world’s largest active archive of weather data and is based in the US. Data from the NCDC is used in compiling global climate estimates such as the hockey stick graph above and on its web site is a really great facility where you can quickly generate graphs of average temperatures for the USA going back as far as 1895. You can select time periods, which month to look at and how the data is displayed in the graph. If you click here you can see it for yourself and play around generating your own climate reports – I urge you to do so.

When I started to play with it I was surprised to get a graph like this for US temperatures for the month of October going back 114 years.

I don’t see much warming in this graph at all and it is nothing like the Hockey Stick graph. Note in particular how cold October 2009 has been – one of the coldest for a century. If you look at other months you can see a clearer warming trend and after one of the coldest Octobers in 2009 came one of the warmest Novembers.

Not long after messing about in the NCDC web site I came across this chart comparing average annual summer temperatures in England for the 18th and 20th centuries. The data comes from the Central England Temperature (CET) record which is maintained by the UK Met Office and which was started in 1659 and is the oldest and longest continuos temperature record in the world.

As you can see British summers in the second half of the 20th century were warmer than those in the first half and it could be argued that this was a global warming signal although its nothing like the Hockey Stick. In fact the average CET summer temperature in the 18th century was 15.46 degC while that for the 20th century was 15.35 degC. Far from being warmer comparison of actual temperature data shows that UK summers in the 20th century were cooler than those of two centuries previously.

These sorts of charts don’t prove anything but finding that you can get a climate record as flat as these ones made me very interested in how the more alarming global temperature charts, like the Hockey Stick, had been compiled and how many more flat temperature charts there were out there. Turns out there are quite a lot and here is a sample.

Here is the average annual temperature for Central England (the graph above was just for the summers) and it shows a bit of an upward trend but no Hockey Stick.

Here is the chart for Cape Town – oddly it seems to have got cooler.

Lets look at somewhere already nice and warm like Nassau in the Bahamas – looks like that its colder now than in 1926 and 1886.

Rome – a flat line

Here is Berlin –  a bit warmer now but not as hot as the 1760s and only about as hot now as the 1870s

Here is Stuttgart – no over all increase in temperatures from the 18th Century

Here is one from the other end of the world – Australia where temperatures appear to have fallen

How about somewhere cold – here is Iceland where temperatures seem to have been higher in the 1930′s than they are now and still no upward pointing Hickey Stick curve.

Here is Milan showing quite a drop in temperatures since the 1940s

Here is the Netherlands – cooler summers and very slightly milder winters – no Hockey Stick

Kansas USA – where the temperature chart seems like the landscape – as flat as pancake

Another one from the US – Minneapolis where 1930 was hotter than today

Here is Sweden where at last we have a bit of an upward curve – except it only seems about as big as the temperature rise in in the 1730s and is only fractionally hotter now than in 1940s