I list here some sources of data about climate change. Sadly ALL these sources are hampered by their inability to release data which has been privately shared with them by National Meteorological Services. As a citizen of the world, this is slightly shocking and sad, but it seems to be the way it is and it is likely to persist for the foreseeable future. The issue was referred to repeatedly at the Surface Temperature Workshop last week but I had not appreciated how profoundly it represented an ultimate block to openness.
Data have been collected by the Hadley Research Centre at the UK Met Office and the Climate Research Unit (CRU) at East Anglia University. These organisations have in general worked together to develop historical data sets with the prefix HadCRU followed by T for temperature, H for humidity etc.
- The data for the global anomaly series (monthly and annual data) are available here. This just allows you to plot your own versions of the well known graphs.
- The Land Surface Climate Station Records themselves are available here. These are homogenised data (see previous post) because the Met Office generally don’t have a licence to reproduce the data supplied by other National Meteorology Organisations. The page includes a rather exhaustive set of FAQs which make clear how hard it is to compile these data records. The current and previous versions of the data set are both available (7.9 Mb). When unzipped, the data consists of 35 Mb of ASCII files in 99 folders, each folder containing sub-folders containing a single station’s data series.
- Also available on the same page are two perl scripts to analyse the data and produce gridded averages. The perl language itself is downloadable from www.perl.org. Please note that the use of an open source language and distribution of the source code is at the same time (a) as open as anyone can possibly be and (b) a bit tricky/completely bewildering for the beginner. The perl scripts are only around 100 lines long and are modestly readable.
- You may also find useful code on the pages of the Open Climate Foundation whose preferred language of analysis is python.
- The main CRU data download page is here and the temperature data page is here. This contains a detailed FAQ page. The ‘products’ available here are generally the results of the analysis (some pre-gridded) and the original data are not available for same reason as discussed above.
There are two key US data sets
- A global Historical Climate Record (GHCN) is available from US National Climatic Data Center (NCDC) part of the US National Oceanic and Atmospheric Administration (NOAA) here. The dataset will shortly be upgraded to version 3 which includes more data sources. When I checked (12 September 2010) several pages are off line (including the main ftp download page) but hopefully that is a temporary problem.
- Update September 24th 2010: An interesting accompaniment to the GHCN site is SurfaceStations.org which is an attempt to build a photographic database of every weather station in the USA used for the GHCN database. The fact that this database is (a) not yet complete, and (b) a ‘crowd source’ project is interesting. It is a measure of how disorganised the temperature record is that this kind of data is not already available. It is also a tacit criticism of NOAA that despite their funding they have not been able to get around to all the in the USA stations and simply take a picture of them! As the site makes clear, most of the stations have poor uncertainty of measurement (> 2°C).
- NOAA also maintain a repository of unprocessed climate records from foreign (i.e. non-US) countries here. This consists of scanned pages from Climate Logs and shows how much data would be available if the resource existed to process it.
- The Goddard Institute for Space Studies (part of NASA) have also compiled and separate dataset (prefixGISSTEMP) available here along with several derived data products. There are several superb linked pages, most notably the reference page which has many journal articles available for free download, and a special comment on the concept of ‘absolute’ measurement of air temperature in meteorology (but not metrology!). The analysis programs used to process the data are also available if your system runs FORTRAN on UNIX, which is not common outside of Universities and Research Institutions.
- GISS have also compiled a Common Sense Climate Index which combines several variables to indicate those regions of the world in which Climate Change would be immediately noticeable if you lived there. Its a great idea but I don’t know much about it,
And there is one key site in Europe (not including the UK ;-))
- The European Climate Assessment Centre also have data available though you have to register. The amount of data available is massive because it contains daily data. It has exhaustive documentation of the source of all data files:very impressive, but utterly overwhelming!
That is enough for now. I will add to this page as I learn of more sites.