The source code used to validate the Met Office data and produce all the visualizations on this blog is here. Enjoy!
This code does the following:
1. Reads and validates the Met Office Land Surface Temperature data file.
2. Spits out errors if it finds problems with the data.
3. Produces a Google Maps/Google Earth compatible KML file with pin marks of the locations of the weather stations used in step 4.
4. Performs gridding of temperature data on a 5 degree square grid.
5. Produces a Google Maps/Google Earth compatible KML file with gridded data for October 2009 to match the latest visualization from the Met Office. This gridding size can be adjusted and any month/year chosen.
6. Dumps CSV files for every weather station in 4 showing month by month anomaly data.
7. Dumps gnuplot compatible .dat files for northern and southern hemisphere and global temperature anomalies since 1850. These are used to produce these visualizations. Note that this does not do cosine-weighting of the grid temperatures (as in the the published literature). Adding it would be simple, but even without it the trend is clear.
8. Produces an experimental animated KML file showing the change in anomaly for October for every year since 1850 to 2009. My machine isn't powerful enough to run this through Google Earth very successfully so I can't guarantee that this is perfect.
If you do use it please refer back to me and this blog.