The Official Iceland Temperature Series

Posted: February 20, 2015 by tallbloke in solar system dynamics


By Paul Homewood

With special thanks to Trausti Jonsson, Senior Meteorologist with the Iceland Met Office.


Now that GHCN and GISS have managed to emasculate the Icelandic temperature record, it is time to present the official version.

The Iceland Met Office, (IMO), have kept meticulous temperature records since the 19thC, including station metadata as well as just the temperature data. In fact the Icelanders took their climate very seriously in those early days, unsurprisingly since they are so vulnerable to climate shifts. There is an interesting overview of the work of people such as Thoroddsen and Nansen in the early 20thC here.

In particular, they have full records since 1931 for seven stations.

It needs to be emphasised that these are not raw temperatures, but have been carefully homogenised and adjusted where necessary, to account for station moves and equipment changes.

Trausti, who has done much of the work…

View original post 431 more words

  1. A C Osborn says:

    As usual Nick Stokes tries to defend the indefensible over at Paul’s forum.

  2. It looks like to me the correct temperature data for ICELAND, correlates with the phase of the AMO/PDO quite strongly.

  3. manicbeancounter says:

    I have looked into the adjustments in a bit more detail. It adds to what the local expert at the Icelandic Met Office is saying – the GHCN and GISS adjustments are unjustified. In particular, the massive adjustments do not tie in with the frequent station relocations. However, the adjustments do change the all-important story from the data. The raw (and properly-adjusted) data shows early twentieth century warming pretty much the same in magnitude and peak value as the warming since the 1980s. In between there is considerable cooling. The adjusted data squashes the earlier warming along with the cooling period, creating the impression of unprecedented recent warming.
    It seems to be a case of adjusting reality to the theory, rather than making corrections to offset known and quantifiable biases in the raw data to make it nearer to the data that would have been provided by perfect instruments.
    I created a graph of average anomalies per decade for the raw data, the GHCN adjusted data, and the GISS homogenised data used to construct GISTEMP.

    Further details at

  4. manicbeancounter says:

    Uploaded the wrong graph. 🙂
    The one in the comment above shows the size of the adjustments from GHCNv3 and GISS homgenised. GHCN drastically cools the past, and does some crazy adjustments for 1939-42, throwing those years totally out of line with the years before and after.
    The graph I meant to link to is below.